![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
-Linolenic Acid in Borage Oil Reverses Epidermal Hyperproliferation in Guinea Pigs1
Department of Medical Nutrition, Graduate School of East-West Medical Science, Kyung Hee University, Seoul, Korea
3To whom correspondence should be addressed. E-mail: choyunhi{at}khu.ac.kr.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
-linolenic acid [GLA; 18:3(n-6)], borage oil (BO; 24–25 g/100 g GLA) and evening primrose oil (PO; 8–10 g/100 g GLA) are efficacious in treating skin disorders. The triglycerol stereospecificity of these oils is distinct, with GLA being concentrated in the sn-2 position of BO and in the sn-3 position of PO. To determine whether the absolute level and/or the triglycerol stereospecificity of GLA in oils affect biological efficacy, epidermal hyperproliferation was induced in guinea pigs by a hydrogenated coconut oil (HCO) diet for 8 wk. Subsequently, guinea pigs were fed diets of PO, BO or a mixture of BO and safflower oil (SO) for 2 wk. The mixture of BO and SO (BS) diet had a similar level of GLA as PO but with sn-2 stereospecificity. As controls, two groups were fed SO and HCO for 10 wk. Epidermal hyperproliferation was reversed by all three oils in the order of BO > BS > PO. However, proliferation scores of group PO were higher than of the normal control group, SO. The accumulations of dihomo--linolenic acid [DGLA; 20:3(n-6)], an elongase product of GLA, into phospholipids and ceramides, of 15-hydroxyeicosatrienoic acid (15-HETrE), the potent antiproliferative metabolite of DGLA, and of ceramides, the major lipid maintaining epidermal barrier, in the epidermis of group BO were greater than of groups BS and PO. Group BS had higher levels of DGLA, 15-HETrE and ceramides than group PO. With primary dependence on absolute levels, our data demonstrate that the antiproliferative efficacy of GLA in the epidermis is preferably exerted from sn-2 stereospecificity of GLA in BO.
KEY WORDS: • -linolenic acid • borage oil • evening primrose oil • stereospecificity • hyperproliferation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
-linolenic acid [GLA4
; 18:3(n-6)] has been reported to be efficacious in treating premenstrual syndrome (1
), atopic eczema (2
,3
), cancer (4
) and acute and chronic inflammation (5
–7
). The mechanism of this action has been associated with the possible generation of antiproliferation and anti-inflammatory metabolites from GLA. Because of the limited activity of the desaturase and a very active elongase in human epidermis (8
), GLA is promptly elongated into dihomo--linolenic acid [DGLA; 20:3(n-6)] followed by its oxidative metabolism either via the cyclooxygenase pathway to series 1 prostaglandins (PGE1) or via the 15-lipoxygeanse into 15-hydroxyeicosatrienoic acid (15-HETrE) (9
,10
). This hydroxy fatty acid has been reported to have antiproliferation and anti-inflammatory effects in vitro (11
).
A high concentration of GLA in plant sources first was reported for evening primrose oil (PO; Oenothera biennis), which contains 8–10 g/100 g of GLA (12
). More recently, higher levels of GLA have been found in borage oil (BO; Borago officinalis) (12
), which has 24–25 g/100 g of GLA, in black currant seed oil (Ribes nigrum) (13
), which has 16–17 g/100 g GLA, and in fungal oils such as Mucor javanicus, which has 16–19 g/100 g GLA (12
). Of these plant seed oils, PO and BO have been commonly used as GLA dietary sources in the management of hyperproliferative and inflammatory disorders of the skin (14
). PO supplementation is efficacious in treating atopic eczema (2
, 3
). BO supplementation also has been reported to improve human skin disorders (15
,16
).
Ingestion of GLA-rich oils results in the accumulation of GLA and DGLA in lipids of various tissues (17
–19
). However, GLA-related biological effects, such as formation of PGE1 or 15-HETrE, may not be solely due to GLA levels in dietary oils. For example, although the GLA concentration in BO is threefold that in PO, intake of BO or PO generates comparable amounts of PGE1 in mouse peritoneal macrophages based on the amount of GLA supplemented in diet (20
). When the same amount of GLA in absolute terms is provided from BO, PO, blackcurrant oil or fungal oils, PO is the most likely oil to generate PGE1 from the mesenteric vascular bed of rats (21
). These oils are distinct in the triglycerol stereospecificity of GLA. It is possible that a particular stereospecificity is more efficacious than others (12
). GLA is concentrated in the sn-3 position of PO and blackcurrant seed oil, the sn-2 position of BO and the sn-2 and sn-3 positions of fungal oil (12
). It has been suggested that the stereospecificity of triglycerols affects fatty acid absorption and metabolism, ultimately affecting the fatty acid-induced biological effects (22
).
To determine whether the absolute level of GLA in oils is the sole determinant of biological efficacy or whether its stereospecificity modulates its effect, EFA deficiency (EFAD)-induced epidermal hyperproliferation was induced in guinea pigs by feeding them a hydrogenated coconut oil (HCO) diet for 8 wk (23
). During the following 2 wk, EFA-deficient guinea pigs were fed PO, BO or a mixture of BO and safflower oil (SO)-containing diets. The mixture of BO and SO (BS) diet had a level of GLA similar to that of PO but with a different stereospecificity. The control group was fed an SO-containing diet and the EFA-deficient group was fed a HCO-containing diet for 10 wk. Compared with groups SO and HCO, the reversal of epidermal hyperproliferation, tissue levels of fatty acids and hydroxy fatty acids, and the synthesis of ceramides were evaluated.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Fourteen-day-old male Hartley guinea pigs were purchased from Samtako Laboratory (Osan, Korea) and assigned to two groups. One group was fed a basal diet (Table 1
) containing 40 g/kg HCO (Dyets, Bethlehem, PA) supplemented with 20 g/kg triolein (Sigma-Aldrich, St. Louis, MO) to induce EFAD as described previously (23
). The other group, which served as the normal control, was fed a basal diet containing 60 g/kg SO (Dyets) (23
). The normal control group (group SO) was pair-fed to the mean of the EFA-deficient group for 8 wk and given water containing 0.5 g/L ascorbic acid to serve as an antioxidant. Guinea pigs were housed individually in an air-conditioned room (22-24°C) with a relative humidity of 55–60%. These conditions were previously established as ideal for guinea pigs (23
). Their weights and food intakes were monitored weekly. The induction of EFAD was evident after 8 wk in guinea pigs fed the HCO diet. EFAD was established biochemically by gas chromatography analysis of the epidermal total fatty acyl groups for the accumulation of eicosatrienoic acid [20:3(n-9)], a marker of EFAD and histologically by evaluation of the hyperproliferative epidermal layer (23
). Animal care and handling conformed to accepted guidelines (24
).
|
), the amounts of GLA supplemented were 690 mg/kg in the PO diet, 1338 mg/kg in the BO diet and 786 mg/kg in the BS diet, respectively. The EFA-deficient guinea pigs were fed the diets for 2 wk to evaluate the efficacy of each dietary treatment on the reversal of epidermal hyperproliferation. The normal control group continued to be fed the basal diet containing 60 g/kg SO (Dyets) for 2 wk. At the end of 10 wk, all guinea pigs were killed by cervical dislocation and epidermal strips were quickly removed by a Castroviejo electrokeratome (customized by Dae Jong Machinery, Ilsan, Korea) as described previously (23
).
|
For morphological evaluation of the epidermis, 4-mm2 biopsies were excised from skin of SO-, HCO-, PO-, BO- and BS-supplemented guinea pigs and placed in neutral buffered formalin. The specimens were fixed, stained with hematoxylin and eosin, sectioned to a thickness of 6 µm and affixed to slides for viewing by light microscopy (x200 magnification). The sections were evaluated for thickening of the epidermal layer (a characteristic feature of epidermal hyperproliferation) by two investigators, including a pathologist, who were unaware of the experimental diets and groups. Proliferation scores ranging from 1 to 5 (1, low; 3, moderate; 5, high) on a 0.5 scale were assigned to each section.
Epidermal DNA synthesis
Mitotic activities in the epidermis of each group was ascertained by [3H]thymidine incorporation (25
). Epidermal biopsies (4 mm2) were placed in 1x DMEM in which [3H]thymidine (3.7 MBq/L) had been dissolved. After incubating for 3 h at 37°C, the epidermis was rapidly frozen in liquid N2 to stop DNA synthesis. Epidermis was thawed, placed in 2 mL of 0.5 mol/L NaOH and heated at 95°C for 30 min to dissolve the tissue and release DNA. Aliquots were placed on cellulose filters (Millipore, Bedford, MA) previously treated with 10 g/L trichloroacetic acid. Filters with absorbed DNA were dried and [3H]thymidine incorporated into DNA was measured by scintillation counting. The rest of the solution after 95°C heating was centrifuged at 800 x g for 5 min to remove tissue debris and the protein concentration was determined by a modified Lowry method (26
) using bovine serum albumin as the standard. The specific [3H]thymidine incorporation was calculated as [3H]thymidine Bq/µg protein.
Lipid analysis
Epidermal strips were taken from the guinea pigs and homogenized with a polytron as reported previously (23
). A portion of the crude epidermal homogenate was used to measure protein concentration by a modified Lowry method (26
) and the rest was extracted with chloroform (CHCl3):methanol (MeOH) (2:1, v/v) to obtain total lipids (27
). The extracted lipids were fractionated into phospholipids, neutral lipids and ceramides by high performance thin layer chromatography (HP-TLC) on 0.20-mm silica gel 60-coated plates (Whatman, Clifton, NJ) by a modified method of Uchida et al. (28
). Specifically, the samples applied to the plates were first developed to 1 cm and then to 3.5 cm in CHCl3:MeOH:acetone (76:20:4, v/v/v). Then it was developed to 7.5 cm in CHCl3:acetone:MeOH (80:10:10, v/v/v) and finally developed to the top in CHCl3:ethylacetate:ether:MeOH (76:20:6:2, v/v/v/v). Each stage of development was carried out after plates were completely air-dried. The fractions containing ceramides, phospholipids and neutral lipids that comigrated with respective standards were visualized under UV light after spraying the plates with 2,7-dichlofluororescein. The gel corresponding to each lipid fraction was scraped off, eluted with CHCl3:MeOH (2:1, v/v) and dried under N2 gas.
The fatty acid profiles of each fraction were determined by gas chromatography (GC) after transmethylation in 60 g/L methanolic HCl (29
). The gas chromatograph (model 5730A; Hewlett-Packard, Palo Alto, CA) was equipped with a SPB-225 fused silica capillary column (30 m x 0.25 mm x 0.15 µm; Supelco, Bellefonte, PA) with an oven temperature of 140°C increasing 4°C/min to 240°C. The detection was performed by a flame ionization detector. Because DGLA and arachidonic acid (AA), which are not major polyunsaturated fatty acids (PUFA) in epidermis, are not discernible, with similar retention times in many occasions of GC analysis, known amounts of DGLA (1 g/L) and AA (0.5 g/L) were spiked to confirm identification and relative contribution of these fatty acids to each lipid fraction. The separated fatty acids were also identified by comparison of retention times to external standards (Supelco).
Hydroxy fatty acids from the epidermal homogenate were extracted with ice-cold CHCl3:MeOH (2:1, v/v) after acidification to pH 3. 15-Hydroxy-11–13-eicosadienoic acid [15-OH-20:2] was used as an internal standard (30
). The levels were determined by reversed-phase high-performance liquid chromatography (HPLC) using a 5-µm octadecyl silica (ODS 18) column (25 cm x 4.6 µm inside diameter; Beckman, Fullerton, CA) as reported previously (31
). The chromatographic system was run isocratically at a flow rate of 1 mL/min on an Agilent 1100 series HPLC system equipped with a G1329A autosampler, a G1311A quaternary pump, a G1315A diode array detector and the Chemstation Controller/Analysis program (Rev.A. 07.01; Hewlett-Packard). The mobile phase solvents were 74% methanol and 26% H2O containing 0.8 g/L acetic acid. The UV absorption spectrum with a photo-diode array detector (PDA 996; Waters, Milford, MA) confirmed the purity of separated hydroxy fatty acids from each sample in the presence of an analytically useful absorption band with Amax = 237 nm (
237 = 27,000/M/cm) (30
). The levels of hydroxy fatty acids in each sample were quantitated with external standards of 13-hydroxyoctadecadienoic acid (13-HODE), 15-HETrE, 15-hydroxyeicosatetraenoic acid (15-HETE), 12-HETE or 5-HETE (Cayman Chemical, Philadelphia, PA). Calibration curves with the various concentrations of external standards revealed good linearity with R2 values exceeding 0.99 (peak area vs concentration). Each detection limit for external standards was 2 mg/L. A representative chromatogram of the hydroxy fatty acid standards as based on reports of major lipoxygenase metabolites from guinea pig epidermis (11
,17
,31
) and as separated by this system is depicted in Figure 1
.
|
Because the process of ceramide biosynthesis is initiated with incorporation of serine (32
), the dietary effect on ceramide synthesis in epidermis was determined by incubating epidermal strips (0.5 g) with the dorsal side face down onto a Petri dish containing 5 mL Krebs bicarbonate buffer (pH 7.5) and [14C]serine (3.7 MBq/L) at 37°C for 16 h as described previously (32
). After incubation, the epidermal strips were washed with PBS and homogenized in buffer. An aliquot was taken for a protein assay according to a modified Lowry method (26
) and the rest was extracted with CHCl3:MeOH (2:1, v/v) to obtain total lipids (27
). The extracted lipids were subjected to HP-TLC for ceramide separation as described in Lipid analysis. The fractions containing total ceramides that comigrated with respective standards were visualized by 2,7-dichlorofluorescein spray and eluted with CHCl3:MeOH (2:1, v/v). The eluants were dried under N2 gas and radioactivity was counted in a Beckman scintillation counter.
Statistical analysis
Data are expressed as means ± SD, and the differences among the SO, HCO, PO, BO and BS groups were determined by one-way ANOVA coupled with the Duncan’s multiple range test (33
) (SAS 6.03; SAS institute, Cary, NC). Differences with P < 0.05 were considered significant. The homogeneity of variance was confirmed using Bartlett’s test. Data that were identified as nonhomogeneous were logarithmically transformed before ANOVA analysis (Table 3
; Figure 6
).
|
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The food intake of all groups was 
53 g/d during the 10-wk feeding period. At the end of 8 wk, group HCO weighed less than group SO (P < 0.05) (Fig. 2
). The weights of groups PO, BO and BS were not different from that of group SO at 10 wk but were greater than that of group HCO (P < 0.05). The weights of groups PO, BO and BS did not differ.
|
Evidence of EFAD was found in the guinea pigs fed the HCO diet. Histological examination of the epidermis of group HCO (Fig. 3
Ab) demonstrated characteristic epidermal thickness (acanthosis) and hyperplasia compared with epidermis of the control SO group (Fig. 3
Aa). After feeding the EFA-deficient guinea pigs the PO, BO or BS for 2 wk, the hyperproliferative state was not apparent (Fig. 3
Ac, d, e). On a quantitative basis (Fig. 3
B), group HCO had higher proliferation scores than other groups in relation to the histological appearance (Fig. 3
Ab). Although feeding PO, BO or BS diets reversed the hyperproliferative state of the epidermis in EFA-deficient guinea pigs, groups BO and BS had lower proliferation scores than group PO. The proliferation scores of group PO were higher than those of the normal SO control group (P < 0.05).
|
), suggesting that epidermal hyperproliferation was associated with increased DNA synthesis. In guinea pigs fed BO or BS, DNA synthesis was suppressed more than in those fed SO. In contrast, the DNA synthesis in the epidermis of groups PO and SO did not differ. In EFA-deficient guinea pigs fed BO, the epidermal hyperproliferation was reversed more than in groups PO and BS.
|
In an analysis of fatty acid in epidermal phospholipids, ceramides and neutral lipids (Table 3)
, only data for PUFA, the fatty acids that are available for enzymatic oxygenation (31
), and eicosatrienoic acid [20:3(n-9)], only generated in EFAD (23
), are listed. In guinea pigs fed SO, 74.3 g/100 g total fatty acids of linoleic acid (LA) [18:2(n-6)] was incorporated into epidermal major lipids. The levels of LA and AA [20:4(n-6)] incorporated into epidermal phospholipids of group HCO were lower than that of group SO (P < 0.05). Eicosatrienoic acid [20:3(n-9)], an abnormal fatty acid generated from oleic acid during the development of EFAD (23
), was detected only in the epidermal lipids of group HCO. In groups PO, BO and BS, eicosatrienoic acid [20:3(n-9)] was not detected.
In guinea pigs fed PO, BO or BS, GLA was not detected (<0.1 g/100 g total fatty acids), but DGLA, an elongase product of GLA, was detected in epidermal major lipids. In contrast to the comparable incorporation of DGLA into epidermal neutral lipids of groups PO, BO and BS, the level of DGLA incorporated into epidermal phospholipids and ceramides of group BO was higher than that of groups BS and PO (P < 0.05). When compared between groups PO and BS, in which similar levels of GLA and LA with a different stereospecificity were supplemented in diets, group BS had more DGLA incorporated into epidermal phospholipids and ceramides than group PO (P < 0.05). The level of LA incorporated into epidermal phospholipids of groups PO and BS did not differ, but group BS had a higher incorporation of LA into ceramides than group PO (P < 0.05). Although the amount of LA in BO oil was lower than in BS and PO oils (Table 2)
, the incorporation of LA into epidermal phospholipids and ceramides of group BO was not different from that of group BS. These results suggest that the absolute level of GLA in oils primarily determines the accumulation of DGLA in epidermal phospholipids and ceramides, and sn-2 stereospecificity of GLA in triglycerol structure of oils may preferentially modulate it. Furthermore, sn-1 stereospecificity of LA in BO (12
) or the mixed stereospecificity of LA (sn-1 and sn-2) in BS oil caused by the mixture of BO (LA in the sn-1 position) (12
) and SO (LA in the sn-2 position) seem to facilitate the accumulation of LA in epidermal ceramides of groups BO and BS more than of group PO. The comparable levels of LA incorporated into epidermal phospholipids of groups PO, BO and BS, despite the different level and stereospecificity of LA in oils, need to be further investigated.
Analysis of epidermal hydroxy fatty acids
The contents of 13-HODE and 15-HETrE in epidermis (Fig. 5
) reflected the level of LA and DGLA incorporated into epidermal phospholipids and ceramides (Table 3)
. The most abundant hydroxy fatty acid, 13-HODE, was derived from the most abundant PUFA, LA, in the epidermis of all groups. The level of 13-HODE in control epidermis of group SO was higher than in the hyperproliferative epidermis of group HCO (P < 0.05). The levels of 13-HODE in the epidermis of groups PO, BO and BS were less than of group SO (P < 0.05).
|
B). The level of 15-HETrE in the epidermis of group BO was fourfold that of group PO (P < 0.05). Although a similar level of GLA was supplemented from PO or BS, the level of 15-HETrE in the epidermis of group BS was higher than that of group PO (P < 0.05). The trace level of 15-HETE, 15-lipoxygenase metabolite of AA, was detected but did not differ among all five diet groups (data not shown). Other lipoxygenase metabolites of AA, 12-HETE and 5-HETE, were not detected in any group. Metabolic labeling of ceramides
Ceramide synthesis in the epidermis of group HCO was less than that of the SO control group (P < 0.05) (Fig. 6
), suggesting that epidermal hyperproliferation was associated with a decrease in ceramide synthesis. Ceramide synthesis in the epidermis of groups PO, BO and BS was higher than of group SO in the order of BO > BS > PO (P < 0.05). These results indicate that BO is the most likely oil to increase ceramide synthesis in epidermis, with the BS second best.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
,3
,14
–16
), the comparative effect has not been well examined. Furthermore, the comparative effect of these two oils on reversing epidermal hyperproliferation was even less clear. The present study indicates that BO is more prominent in reversing epidermal hyperproliferation by profoundly affecting the fatty acid composition and levels of hydroxy fatty acids and ceramides in epidermis.
Specifically, our data in Table 3
demonstrate that feeding EFA-deficient guinea pigs the BO-containing diet, in which GLA [18:3(n-6): 24–25 g/100 g] is concentrated in the sn-2 position of the triglycerol structure, resulted in the incorporation of DGLA, an elongase product of GLA, into epidermal phospholipids and ceramides. These findings agree with earlier studies in which dietary oils were shown to exert a profound effect on the fatty acid composition of epidermal phospholipids (17
,18
,34
,35
). Although ceramides are thought to be of particular importance in maintaining the structural integrity of the epidermal barrier in stratum corneum (36
,37
), the incorporation of PUFA, other than LA, into these lipid classes is not clear (38
). In comparing intake of PO and BS in which a similar level of GLA but with a different stereospecificity was supplemented in diets, the higher level of DGLA was incorporated into epidermal phospholipids and ceramides of group BS, which paralleled lower proliferation scores (Fig. 3
B). This result agrees with previous studies of DHA [22:6(n-3)] distribution in milk triglycerols of the rats which demonstrated that supplementation of DHA in the sn-2 position of the triglycerol structure in fish oil is more efficient in the incorporation of DHA into milk triglycerol than with sn-1/sn-3 positions by seal oils (39
). In contrast, this does not agree with another studies of rats fed the same amount of GLA from different dietary sources that reported the elevated outflow of GLA, DGLA, AA, adrenic acid [22:4(n-6)] and docosapentaenoic acid [22:5(n-6)] from the mesenteric vascular bed by PO intake (21
). In contrast to measuring the outflow level of fatty acid released from mesenteric vascular beds (21
), the epidermal level of DGLA incorporated into phospholipid and ceramides were determined in this present study. Because phospholipids and ceramides are major lipid classes maintaining the normal epidermal barrier (40
), determining the epidermal level of DGLA incorporated into these two lipids rather than free fatty acids released from epidermis is more meaningful for comparing the effect of PO and BS on reversing the epidermal hyperproliferation.
The ability of PUFA to reverse epidermal hyperproliferation is largely explained by the relationship between fatty acids incorporated into phospholipids and ceramides, and the epidermal level of 15-lipoxygenase metabolites (hydroxy fatty acids) (31
, 41
,42
). The 15-lipoxygenase metabolite of LA, 13-HODE, appears to be necessary for regulating the proliferative rate of epidermal keratinocytes (41
). Although not a major PUFA in epidermis, GLA is promptly elongated into DGLA followed by 15-lipoxygenase catalyzed conversion into 15-HETrE in the epidermis (10
,17
), which exerts a superior biopotency to 13-HODE for antiproliferation (43
). In contrast, lipoxygenase metabolites of AA are moderately proinflammatory, although the relevance of this in the epidermis is not fully understood (44
). Coinciding with the changes in fatty acid profiles of the epidermal phospholipid and ceramides (Table 3)
, the epidermis of group BO had a higher level of 15-HETrE than group BS (Fig. 5)
, whereas the epidermis of group BS was more effective than group PO in generating 15-HETrE. When the level of the 15-lipoxygenase metabolite is calculated based on the amount (in grams) of GLA supplemented in diets, BO and BS were more effective than PO in generating 15-HETrE in the epidermis. Specifically, BS and BO generated more 15-HETrE (BS, 610.7 pg/g GLA; BO, 523.2 pg/g GLA) than PO (231.9 pg/g GLA). In comparison to the previous studies of mice fed the GLA-enriched diet, which have shown that formation of PGE1 in peritoneal macrophage is comparable for both BO and PO oils based on the amount of GLA supplemented in diets (20
), our results demonstrate that BO is the most likely oil to release 15-HETrE in epidermis, with the BS second best. Previous studies demonstrated the conserved stereospecificity of fatty acid during digestion (39
); our data suggest that phospholipids of groups BO and BS, of which DGLA was preferably incorporated into sn-2 position, seem to be more hydrolyzed by phospholipase A2 (45
), thus providing DGLA for lipoxygenation into 15-HETrE. The relationship between lipoxygenation and incorporation of precursor fatty acids into ceramides remains to be determined.
Although ceramides play a structural role in maintaining the epidermal barrier (36
,37
) and also can function as an intracellular mediator of proliferation (32
), little is known about the effect of diet-induced changes in fatty acid content and stereospecificity on ceramide synthesis as related to the regulation of tissue hyperproliferation. Our data demonstrate that feeding GLA-rich oils, PO, BO or BS, increased ceramide synthesis more than feeding a LA-rich oil such as SO. Specifically, BO was the most likely oil in ceramide synthesis and BS was the second best (Fig. 6)
, suggesting that the reversal of epidermal hyperproliferation (Figs. 3
, 4)
was associated with an increase in ceramide synthesis. These results agree with previous studies of stratum corneum lipid abnormalities in atopic dermatitis (AD), which reported a decreased level of ceramides, specifically ceramide 1 subtype, in dried skin of AD patients (46
). Ceramide 1 is the acylceramide esterified with LA (38
). In view of functional importance of LA in regulation of epidermal proliferation, LA esterified in phospholipids of the basal layer appeared to be transferred to acylceramides in stratum corneum followed by the de novo synthesis of ceramides from phospholipid (47
). For maintaining the epidermal barrier, the linoleate moiety of ceramide 1 favors the formation of lipid lamellar packing (48
). From the present study, the incorporation of DGLA into epidermal phospholipid and ceramides of groups BO and BS suggests that DGLA can substitute for LA in the transferring process from phospholipid to ceramides. These findings agree with the previous studies of age and seasonal variation in stratum corneum lipids, which reported that the level of LA in ceramide 1 was depleted, whereas the level of oleic acid in ceramide 1 was increased by age and in winter (49
). Although our data demonstrate first that feeding BO increased ceramide synthesis, which paralleled the suppression of epidermal hyperproliferation, the relationship between fatty acid profiles of epidermal ceramides and the regulation of epidermal proliferation needs to be further investigated.
The biological and clinical efficacy of the oils can not be based on their GLA level alone. The present study indicates that the antiproliferative biopotency of GLA in the epidermis primarily depends on the absolute level of GLA and is preferably exerted from sn-2 stereospecificity in oils. This suggests that the stereospecificity of GLA is of importance in determining the biological and clinical efficacy of the oils. Dietary supplementation of BO containing the highest level of GLA with sn-2 stereospecificity in the triglycerol structure may be an alternative therapy or may serve as an adjunct to conventional therapies used in the treatment of hyperproliferative disorders of skin.
|
FOOTNOTES |
|---|
2 S.C. and S.K. contributed equally to these studies.
4 Abbreviations used: 13-HODE, 13-hydroxyoctadecadienoic acid; 15-HETrE, 15-hydroxyeicosatrienoic acid; AA, arachidonic acid; AD, atopic dermatitis; BO, borage oil; BS, mixture of BO and safflower oil; DGLA, dihomo--linolenic acid; EFAD, EFA deficiency; GLA, -linolenic acid; HCO, hydrogenated coconut oil; HETE, hydroxyeicosatetraenoic acid; HP-TLC, high-performance liquid chromatography; LA, linoleic acid; PGE1, series 1 prostaglandin; PO, evening primrose oil; PUFA, polyunsaturated fatty acids; SO, safflower oil
Manuscript received 22 February 2002. Initial review completed 26 March 2002. Revision accepted 3 July 2002.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Budeiri, D., Li Wan Po, A. & Dornan, J. C. (1996) Is evening primrose oil of value in the treatment of premenstrual syndrome?. Control Clin. Trials 17:60-68.[Medline]
2. Lovell, C. R., Burton, J. L. & Horrobin, D. F. (1981) Treatment of atopic eczema with evening primrose oil. Lancet 1:278.
3. Wright, S. & Burton, J. L. (1982) Oral evening-primrose-seed oil improves atopic eczema. Lancet 2:1120-1122.[Medline]
4. Cai, J., Jiang, W. G. & Mansel, R. E. (1999) Inhibition of angiogenic factor- and tumour-induced angiogenesis by -linolenic acid. Prostaglandins Leukot. Essent. Fatty Acids 60:21-29.[Medline]
5. Tate, G., Mandell, B. F., Laposata, M., Ohliger, D., Baker, D. G., Schumacher, H. R. & Zurier, R. B. (1989) Suppression of acute and chronic inflammation by dietary linolenic acid. J. Rheumatol. 16:729-733.[Medline]
6. Kunkel, S. L., Ogawa, H., Conran, P. B., Ward, P. A. & Zurier, R. B. (1981) Suppression of chronic inflammation by evening primrose oil. Prog. Lipid Res. 20:886-888.
7. Hansen, T. M., Lerche, A., Kassis, V., Lorenzen, I. & Sondergaard, J. (1983) Treatment of rheumatoid arthritis with prostaglandin E1 precursors cis-linoleic acid and -linolenic acid. Scand. J. Rheumatol. 12:85-88.[Medline]
8. Chapkin, R. S., Ziboh, V. A., Marcelo, C. L. & Voorhees, J. J. (1986) Metabolism of essential fatty acids by human epidermal enzyme preparations: evidence of chain elongation. J. Lipid Res. 27:945-954.[Abstract]
9. Johnson, M. M., Swan, D. D., Surette, M. E., Stegner, J., Chilton, T., Fontech, A. N. & Chilton, F. H. (1997) Dietary supplementation with -linolenic acid alters fatty acid content and eicosanoid production in healthy humans. J. Nutr. 127:1435-1444.
10. Cho, Y. & Ziboh, V. A. (1997) A novel 15-hydroxyeicosatrienoic acid-substituted diacylglycerol (15-HETrE-DAG) selectively inhibits epidermal protein kinase C-ß. Biochim. Biophys. Acta 1349:67-71.[Medline]
11. Miller, C. C., McCreedy, C. A., Jones, A. D. & Ziboh, V. A. (1988) Oxidative metabolism of dihomo--linolenic acid by guinea pig epidermis: evidence of generation of anti-inflammatory products. Prostaglandins 35:917-938.[Medline]
12. Lawson, L. L. & Hughes, B. G. (1988) Triacylglycerol structure of plant and fungal oils containing -linolenic acid. Lipids 23:313-317.
13. Traitler, H., Winter, H., Richli, U. & Ingenbleek, Y. (1984) Characterization of -linolenic acid in ribes seed. Lipids 12:923-928.
14. Barre, D. E. (2001) Potential of evening primrose, borage, blackcurrant, and fungal oils in human health. Ann. Nutr. Metab. 45:47-57.[Medline]
15. Bahmer, F. A. (1992) ADASI score: atopic dermatitis area and severity index. Acta Derm. Venereol. Suppl. (Stockh.) 176:32-33.
16. Henz, B. M., Jablonska, S., Van de Kerkhof, P. C., Stingl, G., Blascyk, M., Vandervalk, P. G., Veenhuizen, R., Muggli, R. & Raederstorff, D. (1999) Double-blind multicentre analysis of the efficacy of borage oil in patients with atopic eczema. Br. J. Dermatol. 140:685-688.[Medline]
17. Miller, C. C. & Ziboh, V. A. (1988) -Linolenic acid-enriched diet alters cutaneous eicosanoids. Biochem. Biophys. Res. Commun. 154:967-974.[Medline]
18. Barre, D. E. & Holub, B. J. (1992) The effect of borage oil consumption on the composition of individual phospholipids in human platelets. Lipids 27:315-320.[Medline]
19. Fan, Y. Y. & Chapkin, R. S. (1993) Phospholipid sources of metabolically elongated -linoleic acid: conversion to prostaglandin E1 in stimulated mouse macrophages. J. Nutr. Biochem. 4:602-607.
20. Fan, Y. Y. & Chapkin, R. S. (1992) Mouse peritoneal macrophage prostaglandin E1 synthesis is altered by dietary -linolenic acid. J. Nutr. 122:1600-1606.
21. Jenkins, D. K., Mitchell, J. C., Manku, M. S. & Horrobin, D. F. (1988) Effects of different sources of -linolenic acid on the formation of essential fatty acid and prostanoid metabolites. Med. Sci. Res. 16:525-526.
22. Mattson, F. H., Nolen, G. A. & Webb, M. R. (1979) The absorbability by rats of various triglycerides of stearic and oleic acid and the effect of dietary calcium and magnesium. J. Nutr. 109:1682-1687.
23. Cho, Y. & Ziboh, V. A. (1995) Nutritional modulation of guinea pig skin hyperproliferation by essential fatty acid deficiency is associated with selective down regulation of protein kinase C-ß. J. Nutr. 125:2741-2750.
24. National Research Council (1985) Guide for the Care and Use of Laboratory Guinea Pigs, publication no. 85–23 (rev.) 1985 National Institutes of Health Bethesda, MD. .
25. Halprin, K. M., Taylor, J. R., Levine, V. & Adachi, K. (1979) A combined alkali extraction-ethidium bromide technique for the measurement of DNA in small pieces of tissue. J. Invest. Dermatol. 73:359-363.[Medline]
26. Lowry, O. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951) Protein measurement with the folin phenol reagent. J. Biol. Chem. 193:265-275.
27. Folch, J., Lee, M. & Sloane-Stanley, G. H. (1957) A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226:497-509.
28. Uchida, Y., Hara, M., Nishio, H., Sidransky, E., Inoue, S., Otsuka, F., Suzuki, A., Elias, P. M., Holleran, W. M. & Hamanaka, S. (2000) Epidermal sphingomyelins are precursors for selected stratum corneum ceramides. J. Lipid Res. 41:2071-2082.
29. Tang, W. & Ziboh, V. A. (1988) Reversal of epidermal hyperproliferation in essential fatty acid deficient guinea pigs is accompanied by rapid generation of inositol triphosphate. Arch. Dermatol. Res. 280:286-292.[Medline]
30. Claeys, M., Kivits, G. A., Christ-Hazelhof, E. & Nugteren, D. H. (1985) Metabolic profile of linoleic acid in porcine leukocytes through the lipoxygenase pathway. Biochim. Biophys. Acta 837:35-51.[Medline]
31. Miller, C. C., Tang, W., Ziboh, V. A. & Fletcher, M. P. (1991) Dietary supplementation with ethyl ester concentrates of fish oil (n-3) and borage oil (n-6) polyunsaturated fatty acids induces epidermal generation of local putative anti-inflammatory metabolites. J. Invest. Dermatol. 96:98-103.[Medline]
32. Tohyama, J., Oya, Y., Ezoe, T., Vanier, M. T., Nakayasu, H., Fujita, N. & Suzuki, K. (1999) Ceramide accumulation is associated with increased apoptotic cell death in cultured fibroblasts of sphingolipid activator protein-deficient mouse but not in fibroblasts of patients with Farber disease. J. Inherit. Metab. Dis. 22:649-662.[Medline]
33. Hair, J. F., Anderson, R. E., Tatham, R. L. & Black, W. C. (1995) Multivariate Data Analysis with Readings 4th ed. 1995 Prentice-Hall Upper Saddle River, NJ. .
34. Miller, C. C., Ziboh, V. A., Wong, T. & Fletcher, M. (1990) Dietary supplementation with oils rich in (n-3) and (n-6) fatty acids influences in vivo levels of epidermal lipoxygenase products in guinea pigs. J. Nutr. 120:36-44.
35. Chapkin, R. S., Ziboh, V. A. & McCullough, J. L. (1987) Dietary influences of evening primrose and fish oil on the skin of essential fatty acid-deficient guinea pigs. J. Nutr. 117:1360-1370.
36. Grubauer, G., Feingold, K. R., Harris, R. M. & Elias, P. M. (1989) Lipid content and lipid type as determinants of the epidermal permeability barrier. J. Lipid Res. 30:89-96.[Abstract]
37. Wertz, P. W., Cho, E. S. & Downing, D. T. (1983) Effect of essential fatty acid deficiency on the epidermal sphingolipids of the rat. Biochim. Biophys. Acta 753:350-355.[Medline]
38. Wertz, P. W. & Swartzendruber, D. C. (1987) Essential fatty acids and epidermal integrity. Arch. Dermatol. 123:1381-1384.[Abstract]
39. Jensen, M. M., Sorensen, H. & Hoy, C. E. (1996) Influence of triglycerol structure and fatty acid profile of dietary fats on milk triacylglycerols in the rat: a two-generation study. Lipids 31:187-192.[Medline]
40. Hedberg, C. L., Wetz, P. W. & Downing, D. T. (1988) The time course of lipid biosynthesis in pig epidermis. J. Invest. Dermatol. 91:169-174.[Medline]
41. Miller, C. C. & Ziboh, V. A. (1990) Induction of epidermal hyperproliferation by topical n-3 polyunsaturated fatty acids on guinea pig skin linked to decreased levels of 13-hydroxyoctadecadienoic acid (13-HODE). J. Invest. Dermatol. 94:353-358.[Medline]
42. Nugteren, D. H., Christ-Hazelhof, E., Vander Beek, A. & Houtsmuller, U. M. (1985) Metabolism of linoleic acid and other essential fatty acids in the epidermis of the rat. Biochim. Biophys. Acta 834:429-436.[Medline]
43. Chapkin, R. S. (1998) Reappraisal of the essential fatty acids. Chow, C. K. eds. Fatty Acids in Food and Their Health Implications 2nd ed. 1998 Marcel Dekker New York, NY. .
44. Hammarstrom, S., Hamberg, M., Samuelsson, B., Duell, E. A., Stawiski, M. & Voorhees, J. J. (1975) Increased concentrations of non-esterified arachidonic acid, 12L-hydroxy-5,8,10,14-eicosatetraenoic acid, prostaglandin E2, and prostaglandin F2
in epidermis of psoriasis. Proc. Natl. Acad. Sci. USA 72:5130-5134.
45. Cho, Y. & Ziboh, V. A. (1994) Incorporation of 13-hydoxyoctadecadienoic acid (13-HODE) into epidermal ceramides and phospholipids: phospholipase C-catalyzed release of novel 13-HODE-containing diacylglycerol. J. Lipid Res. 35:255-262.[Abstract]
46. Yamamoto, A., Serizawa, S., Ito, M. & Sato, Y. (1991) Stratum corneum lipid abnormalities in atopic dermatitis. Arch. Dermatol. Res. 283:219-223.[Medline]
47. Wertz, P. W. & Downing, D. T. (1990) Metabolism of linoleic acid in porcine epidermis. J. Lipid Res. 31:1839-1844.[Abstract]
48. Bouwstra, J. A., Gooris, G. S., Dubbelaar, F. E., Weerheim, A. M., Ijzerman, A. P. & Ponec, M. (1998) Role of ceramide 1 in the molecular organization of the stratum corneum lipids. J. Lipid Res. 39:186-196.
49. Rogers, J., Harding, C., Mayo, A., Banks, J. & Rawlings, A. (1996) Stratum corneum lipids: the effect of ageing and the seasons. Arch. Dermatol. Res. 288:765-770.[Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
