Fatty acid composition of sebum wax esters and urinary androgen level in normal human individuals

Fatty acid composition of sebum wax esters and urinary androgen level in normal human individuals

Journal of Dermatological Science, 1 (1990) 269-216 269 Elsevier DESC 00031 Fatty acid composition of sebum wax esters and urinary androgen level...

725KB Sizes 0 Downloads 18 Views

Journal of Dermatological

Science,

1 (1990) 269-216 269

Elsevier

DESC 00031

Fatty acid composition of sebum wax esters and urinary androgen level in normal human individuals Ayako Yamamoto,

Shigeo Serizawa, Masaaki

Ito and Yoshio Sato

Department of Dermatology, Niigata University School of Medicine, Niigata, Japan

(Received 4 September

Key words: Androgens;

1989; accepted

6 March 1990)

Sebum wax esters; Urine; Fatty acids

Abstract To investigate the effects of androgens on the fatty acid compositions of sebum wax esters, we examined sebum and urinary samples from 36 healthy individuals, aged from 3 to 59 years. The percentages of C,,: , straight chain components in wax esters were correlated positively with the urinary testosterone levels in both sexes, and with the urinary levels of etiocholanolone and total 17-ketosteroids (17-KS) in females. These data suggest that more active sebaceous glands in lipid production excrete sebum with a higher proportion of C,,: , straight chain fatty acid, which is considered to be purely endogenous. It appears, therefore, that the proportion of C,, : , straight chain fatty acid in sebum wax esters may indicate the sebaceous gland activity in both sexes. In comparison of the amounts of various straight and terminally branched fatty acids in sebum with urinary androgen levels, the straight even fatty acids tended to change in a positive correlation with testosterone levels, in contrast to the changes of the iso even fatty acids in both sexes and to those of the iso odd fatty acids in males. The straight odd fatty acids showed a similar change to that of the straight even fatty acids in males, while in females, there was no significant correlation between the amounts of the fatty acids and testosterone levels. Anteiso fatty acids showed no notable change correlated with testosterone levels. This result suggests that the synthesis of iso or anteiso fatty acids may be controlled by complex factors and that there may be a unique source of anteiso fatty acids in human sebaceous glands.

Introduction We previously reported the effect of aging on the fatty acid composition of sebum wax esters [ll- c,,:, straight chain components, which

Correspondence to: Ayako Yamamoto, Department of Dermatology, Niigata University School of Medicine, Niigata 95 1, Japan. Abbreviations: DHEA = dehydroepiandrosterone; ECLs = equivalent chain lengths; FAMES = fatty acid methyl esters; 17-KS = 17-ketosteroids; TLC = thin-layer chromatogra-

phy. 0923-181 l/90/%03.50 0 1990 Elsevier Science Publishers

occupied a large proportion in the fatty acids of sebum wax esters, increased in proportion from infancy toward 20 years and decreased thereafter until 50 years of age. Furthermore, the percentages of Cr6:, straight chain components were correlated positively with the sebaceous gland activity, which was expressed by the ratio of wax esters/[ cholesterol + cholesterol esters] (WE/[C + CE]). Stewart et al [2,3] argued that the factor controlling the sebum fatty acid composition was not age, but rather rates of sebum secretion and, at least in part, genotype. More recently, however, they described that straight chain

B.V. (Biomedical

Division)

270

fatty acids tended to increase with increasing WE/[C + CE], while terminally branched (iso and anteiso) fatty acids tended to decrease in individuals, aged 9-15 [4]. These data are similar to those of our previous report [ 11. On the other hand, the secretion of human androgens from adrenocortical glands are known to increase at adolescence and to decrease at senescence and testosterone is thought to be a most active hormone for human sebaceous glands [ 5,6]. However, Pochi et al [ 71 showed that sebaceous gland activity in prepubertal children increased in a positive correlation with the urinary excretion of 17-KS. These data may suggest an important role of adrenocortical androgens in sebaceous gland activity. In the present study, to elucidate more clearly the possibility that fatty acid composition in human sebum changes with sebaceous gland activity, the various straight and terminally branched fatty acid chain types were analyzed in all ages, and the data were compared with levels of urinary androgen excretion. Wax esters were extracted from the sebum and used for the analysis, because they are recognized as a lipid class that is made exclusively by the sebaceous glands [ 31. The urinary examination included measurements of 24-h urinary excretion levels of total 17-ketosteroids (17-KS), androsterone, etiocholanolone, and dehydroepiandrosterone (DHEA) as well as testosterone. Subjects and Methods Subjects The subjects were 36 healthy individuals,

from 2-59 years and 3 males and 3 females.

each

decade

aged included

Collection of lipids

The forehead of each subject was swabbed with lipid-free cotton moistened with ether. Then, a lipid-free absorbent paper was attached to the defined area (28 cm’) of the forehead for 3 h to adsorb the sebum, which were then extracted twice with 20 ml of chloroform/methanol (2 : 1).

The extract was evaporated to dryness and stored under nitrogen at - 20 “C until use. Isolation of wax esters

A wax ester fraction was separated from other lipid classes by preparative thin-layer chromatography (TLC) on a 0.5 mm thick layer of silica gel G (20 x 20 cm, Kieselgel 60, Merck, Darmstadt) that had been cleaned by development with chloroform/methanol (2: 1) [8]. The chromatogram was first developed with hexane, and then with benzene to the top of the plate. Bands were detected by spraying the plate with a 0.2% solution of 2’, 7’-dichlorofluorescein in ethanol and viewed under ultraviolet light. The band corresponding to wax esters was scraped from the plate and the lipids were eluted with ether [ 11. Analysis of wax esters

Procedures were the same as those described by our previous report [ 11. Briefly, the dried wax esters were methylated by heating with HCl/ methanol and the fatty acid methyl esters (FAMES) were isolated by preparative TLC. The FAMES were separated by degrees of unsaturation on argentated TLC plates, and the saturated and mono-unsaturated FAMES were analyzed by gas chromatography using a 50 mm x 0.2 mm flexible fused-silica column wall-coated with OV-101 (Shimadzu, Kyoto) at 200 * C. Fatty acids with 2 or more double bonds, which form less than 10% of the total, were not examined in the present study [2]. Peak identification was based on relative retention times which were compared with published values [ 91. Components with chain lengths below isobranched 1Ccarbons or above the straight 18-carbons were not studied, as are any components constituting only a small proportion of the total [ 101. Analysis of urinary 17-KS

and testosterone

Twenty-four hour urine samples were collected from all of the subjects and used for analysis of individual 17-KS and testosterone. Extraction and subsequent analysis were performed ac-

271

cording to the methods of Yakata et al [ 1 l] and Jacobsohn and Lieberman [ 121. The steroids were extracted with ethyl acetate from each urine sample, a 50 ml aliquot of which was used for the extraction of 17-KS and a 500 ml aliquot for that of testosterone. The extract was evaporated, dissolved in ethyl acetate again, and then subjected to solvolysis by adding perchloric acid as a catalyzer and by heating at 50 ‘C for 3 h. After saponification with a 1% solution of KOH in ethanol for 2 h at room temperature, the extract was separated by TLC on a 0.25 mm thick layer of silica gel G that had been preactivated by heating in an oven at 110 ’ C for 2 h. The TLC plate was developed with ethyl acetate/benzene (1 : I), and the bands corresponding to androsterone, DHEA, and etiocholanolone were scraped off and extracted with ethyl acetate. Cholesterol was added to the extract as an internal standard. The resulting mixture was converted to the trimethylsilyl ethers and subjected to gas chromatography on a 2 m glass column packed with 2% XE-60 on chromosorb W (60-80 mesh) at 195 ‘C. Chromatograms obtained by this method showed four dominant peaks, representing androsterone, etiocholanolone, DHEA, and cholesterol (Fig. l), and the areas of these peaks were calculated. Testosterone was extracted and separated by the modified method of Ismail and Harkiness [ 131. The trimethylsilyl ether of

testosterone was analyzed by gas chromatography on a 2 m glass column packed with 2% OV-17 on shimalite W (SO-100 mesh) at 255 “C (Fig. 2). The mean recoveries of all 17-KS and testosterone were 7 1.1 y0 and 79.5 %, respectively. Results

Variationsin thefatty acid composition of wax esters Fig. 3 shows the concentrations of various straight and terminally branched fatty acids in the saturated and mono-unsaturated fractions of wax esters plotted against ages of experimental subjects. The chain lengths included are 14, 16, and 18 for straight-even; 16 and 18 for iso-even; 15 and 17 for straight-odd, iso-odd, and saturated anteiso; 17 for mono-unsaturated anteiso. In both sexes, C6: 1 straight chain fatty acid, which is a main component in sebum wax esters, showed a significant change in relative amount with age, while C,, straight chain fatty acid showed no notable change, as demonstrated in our previous study [ 11. The proportions of straight even fatty acids showed a curve with age similar to that of C,, : 1 straight chain fatty acid, although they were more variable among females than among males. The proportions of iso-even fatty acids followed a reversed course with advancing age, decreasing from infancy through maturity and increasing up until 50s; the changes with age were, however,

Et

Fig. 1. Gas chromatogram of 17-KS standards on a column packed with 2% XE-60 on chromosorb W. An = androsterone; C = cholesterol; DHEA = dehydroepiandrosterone; Et = etiocholanolone.

TlME

MIN

Fig. 2. Gas chromatogram of authentic testosterone on a column packed with 2% OV-17 on a shimalite W. C = cholesterol; Te = testosterone.

212

I_Saturated

00

Monounsaturated

0

0

0

0

.

‘0

.fpeee

0 40

.

0.0’



o .0O ‘00’

l

.

.

o

0 10

20

20

30

40

50

60

10

20

30

40

50

60

20

30

40

50

60

10

20

30

40

50

60

10

20

30

40

50

60

10

20

30

40

50

60

10

20

30

40

50

60

10

20

30

40

50

60

r

o-

10

40

20

1

0

~~~*w.w.03~ ‘1

. L----. 10

IO

5

20

1.

20

.

08

b

20

30

o”

8O

WJ

40

50

0

0.

.*

O0

l

‘I[

O

l

l*

O

4

60

.z

0

.0.0

10

3o

50

0

8

0

40

O

0

.O.

30

000

O

,

,

10

20

o

l

0

w*

00 0

l

, 8 ,-O , 30

40

0

00

.”

.

60

50

, 60

Age

(years)

Fig. 3. Variations in the concentrations ofvarious straight and terminally branched fatty acid chain types from sebum wax esters For each chain type, its percentage concentration was plotted against ages of subjects. o, female; 0, male.

213

more significantly shown in the mono-unsaturated fatty acids than in the saturated fatty acids. Straight odd, iso-odd, and anteiso-odd fatty acids displayed no significant change in amount in relation to age. Individual differences among anteiso-odd chain types in the mono-unsaturated fatty acids were quite small (Fig. 3). Variations in urinary androgen excretion Age-related changes in 24-h urinary excretion of testosterone and individual 17-KS are shown in Fig. 4. All 17-KS were excreted in larger quantities in girls than in boys under 10 years of age. In males, excretion of either individual or total 17-KS abruptly increased at adolescence, attained a peak for those aged between 20 and 40, and gradually decreased thereafter, whereas that

10

20

30

40

50

60

“r

10

20

30

40

50

60

in females showed a peak between 10 and 30 years of age and then began to decrease earlier than it did in males. The excretion peak was significantly higher in males than in females. Urinary testosterone excretion was very low in both males and females under 10 years of age; then, in males it markedly increased in the teens, reached a peak between 20 and 40 years and was maintained at a high level even between 40-50 years while in females it decreased more rapidly than in males after maturity (Fig. 4). The fatty acid composition of wax esters and urinary androgen levels Tables I and II show the correlation between the proportions of various straight and terminally branched fatty acids in the sebum wax esters and the amounts of urinary androgen excretion from infancy to senescence. In mono-unsaturated fatty acids, the proportion of the C,,: 1 straight chain type showed a fine correlation with testosterone levels in both sexes, and with etiocholanolone and total 17-KS levels in females (Table II). C,, straight chain type showed no notable correlation with testosterone levels (Table I); this may be a contaminated exogenous lipid from circulation [4]. The straight-even fatty acids, a main chain type in sebum wax esters, had a tendency to change in a positive correlation with testosterone levels, but contrastedly the iso-even fatty acids in both sexes and the iso-odd fatty acids in males revealed a negative correlation with testosterone levels (Tables I and II). The amount of the straight-odd fatty acids showed to change in a positive correlation with testosterone levels in males, while in females there was no significant correlation (Tables I and II). Discussion

?

5

$ ;; e z a

0 10

20

30

40

50

60

Fig. 4. Variations in urinary androgen excretion. For each androgen, its 24 h-urinary excretion was plotted against ages of subjects. o, female; 0, male.

Relationship between sebaceous gland activity and androgens During human life, the C,,: , straight chain fatty acid amount of sebum changes with a very fine correlation with testosterone levels in both sexes, as shown in the present study (see Table II

274 TABLE I Correlation between proportions of various straight and terminally branched chain types in saturated fatty acids from the sebum wax esters and urinary androgen excretion Correlation

coefficient

Te

DHEA

M

F

M

C6 straight C,, iso-branched

0.307 - 0.496**

0.293 - 0.540**

straight even straight odd iso-even

0.273 0.469** - 0.360 - 0.488** - 0.224

0.474** 0.090 - 0.468** - 0.3 13 - 0.440

iso-odd anteiso-odd

0.032 - 0.173 0.192 0.241 - 0.167 - 0.445 - 0.100

An = androsterone; DHEA = dehydroepiandrosterone; sterone; F = females; M = males. * (P
Et F 0.07 1 - 0.433 0.405 0.080 - 0.327 0.042 - 0.351

An

M 0.026 -0.200 0.187 0.411 -0.193 - 0.295 - 0.214

F

M

0.015 -0.491** 0.445 0.225 -0.495** 0.100 - 0.399

Et = etiocholanolone;

0.393 -0.192 0.128 0.261 -0.184 - 0.205 - 0.074

17-KS = total

17-KS F 0.207 -0.258 0.280 0.152 -0.262 0.074 - 0.361

M

F

0.250 -0.250 0.128 0.388 -0.152 - 0.288 - 0.106

17-ketosteroids;

0.153 -0.440 0.424 0.120 -0.338 0.035 - 0.368

Te = testo-

TABLE II Correlation between proportions of various straight and terminally branched chain types in mono-unsaturated the sebum wax esters and urinary androgen excretion Correlation

coefficient DHEA

Te M Ci6: i straight C,,: 1 iso-branched straight even straight odd iso-even iso-odd anteiso-odd For abbreviations

0.685* - 0.555** 0.482** 0.435 - 0.478** - 0.475** - 0.157

fatty acids from

F 0.752* - 0.466 0.407 - 0.372 - 0.396 - 0.115 - 0.240

M 0.416 - 0.388 0.391 0.323 - 0.326 -0.451 - 0.056

An

Et F 0.362 - 0.255 0.235 - 0.338 - 0.276 -0.119 - 0.179

M 0.467 - 0.428 0.421 0.200 - 0.361 - 0.389 - 0.134

F 0.664* - 0.467 0.458 - 0.239 - 0.410 - 0.192 - 0.300

M 0.310 - 0.282 0.286 0.013 - 0.184 - 0.346 - 0.090

17-KS F 0.428 - 0.146 0.162 0.245 - 0.165 - 0.032 - 0.238

M 0.405 -0.363 0.339 0.109 - 0.259 - 0.425 - 0.065

F 0.573* -0.313 0.312 - 0.290 - 0.3 13 - 0.055 - 0.283

see Table I.

and Fig. 5). However, there is a large difference in testosterone levels between males and females in spite of the identical proportions of C,, : 1 straight chain fatty acid in both sexes (Figs 4 and 5). These data suggest that more active sebaceous glands in lipid production excrete sebum with a higher proportion of C,,: , straight chain fatty

acid, which is considered to be purely endogenous [4]. It appears, therefore, that the proportion of C,,: 1 straight chain fatty acid in sebum wax esters may indicate the sebaceous gland activity in both sexes. Recently, some new findings indicating that Ci6: 1 play an important role in comedogenesis have been reported [ 14,151. This may be sup-

215

As .. lr r=0.685 L

,”

pco.01

201 ?OC 01

10

20

Testosterone

30 (pg/

40 day

1

Fig. 5. The relationship between 24 h-urinary excretion levels and concentrations of C,,: , straight chain fatty acids in sebum wax esters. The 24 h-urinary testosterone excretion levels positively correlate with the proportions of C,,: 1 straight chain components in males (0) (r = 0.685, P < 0.01) and females (0) (r = 0.752, P < 0.01).

ported by the present findings that there are high concentrations of C,, : 1 straight chain fatty acids in maturity. In males, the abrupt increase of sebaceous gland activity in teens and the persistence of the elevated activity after 20 years of age, as reported in our previous study [ 11, may be a result from the high levels of testosterone in these ages (Fig. 4). On the other hand, in females, the higher sebum secretion rate under 10 years of age [l] may be a consequence of higher levels of 17-KS compared with those in males; probably, their sebaceous glands are responsive to 17-KS as well as to testosterone in all age groups (Fig. 4). The 17-KS are known to originate from the adrenal cortex, in which DHEA, a precursor of etiocholanolone and androsterone, is synthesized and secreted [ 161. DHEA is considered to be a prehormone that can be converted through testosterone to dihydrotestosterone. Elevated plasma DHEA levels have been found in patients with acne [6,17,18]. Among our subjects, however, DHEA levels did not correlate with the proportions of C,,: 1 straight chain fatty acids in sebum, whereas etiocholanolone as well as total 17-KS positively correlated with those, especially in females. On the other hand, sebaceous gland activity is correlated

positively with testosterone in males and etiocholanolone in both sexes of ages ranged from 6 months to 86 years [ 191. As an explanation for this discrepancy, it is thought that testosterone levels may be more correlated with sebaceous gland activities during life from the teens to the mid 50s than after 60 years of age, especially in females, and that adrenocortical androgens may be the main stimulator of sebum production in females and infant males. Changes in various straight and terminally branched chain fatty acids

Nicolaides and his co-workers [ 9,20-221 reported that normal-even, normal-odd, iso-even, iso-odd, and anteiso-odd chain structures were produced from the independent precursors. The primers of terminally branched chain fatty acids (isobutyrate, isovalerate, and 2-methylbutyrate) are thought to be derived from the breakdown of cell protein [22]. These studies support our concept in previous report; active sebaceous glands might have low concentrations of the terminally branched chain fatty acids compared with the straight chain fatty acids, because of limited availability for the terminal portions of iso or anteiso fatty acids [ 11. In the present study, however, subjects in their 20s who had high proportions of Ci6: 1 straight chain fatty acid probably resulting from high levels of testosterone showed no decrease in levels of anteiso fatty acids. This result suggests that there may be another de novo source of anteiso fatty acids in human sebaceous glands. Nicolaides proposed an alternative route to anteiso fatty acids, specifically the condensation of methylmalonyl CoA with acetate to start the chain [20]. It is also difficult to explain the gap in amount between iso-even and iso-odd fatty acids; namely our female subjects in their 20s had a relatively high amount of iso-odd fatty acids in spite of their low concentrations of iso-even fatty acids. The synthesis of iso or anteiso fatty acids may be controlled by complex factors and the origin and metabolism of these fatty acid types should be elucidated in further studies.

216

References 1 Yamamoto A, Serizawa S, Ito M, Sato Y: Effect of aging on sebaceous gland activity and on the fatty acid composition of wax esters. J Invest Dermatol 89: 507-512, 1987. 2 Stewart ME, Downing DT: Proportions of various straight and branched fatty acid chain types in the sebaceous wax esters of young children. J Invest Dermatol 84: 501-503, 1985. 3 Stewart ME, McDonnell MW, Downing DT: Possible genetic control of the proportions of branched-chain fatty acids in human sebaceous wax esters. J Invest Dermatol 86: 706-708, 1986. 4 Stewart ME, Steele WA, Downing DT: Changes in the relative amounts of endogenous and exogenous fatty acids in sebaceous lipids during early adolescence. J Invest Dermatol 92: 371-378, 1989. 5 Rosentield RL: Pilosebaceous physiology in relation to hirsutism and acne. Clin Endocrinol Metab 15: 341-362, 1986. 6 Agache P, Blanc D: Current status in sebum knowledge. Int J Derm 21: 304-315, 1982. 7 Pochi PE, Strauss JS, Downing DT: Skin surface lipid composition, acne, pubertal development, and urinary excretion of testosterone and 17-ketosteroids in children. J Invest dermatol 69: 485-489, 1977. 8 Sansone-Bazzano G, Cummings B, Seeler AK, Reisner RM: Differences in the lipid constituents of sebum from prepubertal and pubertal subjects. Br J Dermatol 103: 131-137, 1980. 9 Nicolaides N, Apon JMB: The saturated methyl branched fatty acids of adult human skin surface lipid. Biomed Mass Spectrom 4: 337-347, 1977. 10 Green SC, Stewart ME, Downing DT: Variation in sebum fatty acid composition among adult humans. J Invest Dermatol 83: 114-117, 1984. 11 Yakata M, Koami E, Ishida M: Analysis of urinary 17-ketosteroids by gas chromatography. Rinshobyouri 17: 477-481, 1969 (in Japanese).

12 Jacobsohn GM, Lieberman S: Studies on the chemical cleavage of the urinary glucuronosides of the lFketosteroids. J Biol Chem 237: 1469-1475, 1962. 13 Ismail AAA, Harkiness RA: A method for the estimation of urinary testosterone. Biochem J 99: 717-725, 1966. 14 Stewart ME, Grahek MO, Cambier LS, Werts PW, Downing DT: Dilutional effect of increased sebaceous gland activity on the proportion of linoleic acid in sebaceous wax esters and in epidermal acylceramides. J Invest Dermatol 87: 733-736, 1986. 15 Melnik B, Plewig G: Neue lipidbiochemische Aspekte in der Pathogenese der follikuliren Verhornungsstorung bei Acne vulgaris. Z Hautkr 63: 591-596, 1988. 16 Nieschlag E, Loriaux DL, Ruder HJ, Zucker IR, Kirschner MA, Lipsett, MB: The secretion of dehydroepiandrosterone and dehydroepiandrosterone sulphate in man. J Endocr 57: 123-134, 1973. 17 Lucky AM, Mcguire J, Rosenfield RL, Lucky PA, Rich BH: Plasma androgens in women with acne vulgaris. J Invest Dermatol 81: 70-74, 1983. 18 Darley CR, Moore JW, Besser GM, Munro DD, Edwards CRW, Rees LH, and Kirby JD: Androgen status in women with late onset or persistent acne vulgaris. Clin Exp Dermatol 9: 28-35, 1984. 19 Yamamoto A, Serizawa S, Sato Y: Sebum secretion and urinary androgen level in Kligman AM, Takase Y (ed.). Cutaneous aging. Edited by AM Kligman, Y Takase. University of Tokyo Press, Tokyo, 1988, pp 149-158. 20 Nicolaides N: The structures of the branched fatty acids in the wax esters of vemix caseosa. Lipids 6: 901-905, 1971. 21 Nicolaides N, Fu HC, Ansari MNA, Rice GR: The fatty acids of wax esters and sterol esters from vernix caseosa and from human skin surface lipid. Lipids 7: 506-517, 1972. 22 Nicolaides N: Skin lipids: Their biochemical uniqueness. Science 186: 19-26, 1974.