Download Essential fatty acids and acne

Essential fatty acids and acne
Donald T. Downing, Ph.D., Mary Ellen Stewart, Ph.D., Philip W. Wertz, Ph.D.,
and John S. Strauss, M.D. Iowa City, IA
Acne is characterized by hyperkeratosis of the follicular epithelium, leading to
horny impactions that may lie dormant as open or closed comedones or may
cause inflammation of the follicle. Although persons with acne have consistently
been observed to have elevated levels of sebum secretion, no mechanism
relating sebum secretion rates to comedogenesis is known. Acne patients have
also been shown to have low levels of linoleic acid in their skin surface
lipids. To explain this observation, the hypothesis is advanced that the linoleate
concentration in human sebum depends on the quantity of linoleic acid
present in each sebaceous cell at the commencement of its differentiation and
on the extent to which this initial charge is diluted by subsequent endogenous
lipid synthesis in each sebaceous cell. A corollary hypothesis holds that
low concentrations of !inoleate in sebum impose a state of essential fatty acid
deficiency on the cells of the follicular epithelium and induce the
characteristic response of hyperkeratosis. Both hypotheses could hold, without
there being a systemic deficiency of linoleic acid, simply as the result of
elevated lipogenesis in individual sebaceous cells. (J AM ACAD DERMATOL
14:221-225, 1986.)
Acne has long been thought to be influenced by
dietary factors, and several reports have implicated
essential fatty acid deficiency. 1-3 However, no firm
support for this view has developed, and no successful treatment based on this idea has appeared.
Nevertheless, our studies have provided evidence
for a role of linoleic acid deficiency in acne, and
we present here a hypothesis that may explain how
a deficiency localized in the hair follicle could arise
without there being any dietary cause. The hypothesis is based on recent developments concerning the physiology and biochemistry of sebaceous
gland function and the pathogenesis of acne, as
well as on recent data from our studies of sebum
From the Marshall Dennatology Research Laboratories, The University of Iowa College of Medicine.
Supported in part by Grants Nos. AM22083 and AM32374 from the
U.S. Public Health Service.
Accepted for publication Sept. 30, 1985.
Reprint requests to: Dr. Donald T. Downing, 270 Medlabs, University
of Iowa College of Medicine, Iowa City, IA/319-353-5788.
composition and excretion and the role of linoleic
acid in the structure and function of the epidermis.
BACKGROUND
Physiology and biochemistry of sebum
secretion
Sebum is synthesized in differentiating cells that
are generated on the periphery of the sebaceous
gland and move centripetally as they become engorged with lipid. Eventually, all of the subcellular
organelles are degraded and the cells disrupt, allowing their lipid contents to pass through the sebaceous duct into the pilosebaceous canal. In humans the overall process from cell division to final
rupture requires about 14 days.4 During this time
the cells increase up to 150-fold in volume, almost
all of which is occupied by lipid. Obviously an
external carbon source is necessary to provide the
material for this exuberant generation of lipid, but
the circulating substrate is yet to be identified.
Nevertheless, it is quite clear from analyses of
sebum from various species that circulating lipids
221
222
J oumal of the
American Academy of
Dermatology
Downing et at
are not incorporated by the differentiated sebaceous cell. 5 Human sebum consists principally of
triglycerides (57%), wax esters (26%), and squalene (12%), with small proportions of cholesterol
and cholesteryl esters. Many of the component
fatty acids are branched chain compounds, and
more than half are monounsaturated with unusual
double-bond positions not found in circulating lipids. Thus the fatty acid composition of the wax
esters, as well as of the triglycerides, is such that
most of the acids could not have been acquired
from circulating lipids.
Copious amounts of phospholipids are required
to provide the cellular and subcellular membranes
of the expanding sebaceous cells during their differentiation, and labeling experiments indicate that
these lipids, too, are produced within the cell. 6
This is supported by the observation that the phospholipids contain fatty acids that are characteristic
of the sebum of the species. 7,8 In the final stages
of differentiation, when the membranes are degraded, the phospholipids also disappear, and it is
presumed that their fatty acids become esterified
with whatever hydroxylic lipids are available.
These mechanisms explain why sebum does not
contain phospholipids, in spite of the holocrine
character of sebaceous gland function.
Pathogenesis and clinical course of acne
The first pathologic change in acne appears to
be the formation of the comedo by a hyperkeratotic
reaction of the follicular epithelium. 9 In such impacted follicles, large increases in the population
of Propionibacterium acnes can subsequently occur. This organism is thought to be responsible for
most of the hydrolysis of sebaceous triglycerides
to free fatty acids and glycerol. Inflammatory acne
may require rupture of the wall of the follicle and
release of the contents into the dermis, where typical foreign-body reactions and chemotactic responses are elicited. In addition, permeation of
chemotactic molecules through intact follicular epithelium may be involved in producing inflammation. Usually only a few lesions are inflamed
at one time, although comedones may lie dormant
and persist for years.
Inflammatory acne usually occurs only after the
large increase in sebum secretion rate that takes
place at puberty, and it has been shown that all
subjects with pustular acne have a high rate of
sebum secretion. 1O Further support for the role of
high sebum secretion rates in the pathogenesis of
inflammatory acne is the accepted view that any
treatment that reduces sebum secretion rate can
alleviate the disease. This is true of treatment with
estrogens I I or antiandrogens,12 with the prostaglandin synthetase inhibitor 5,8,1l,14-eicosatetraynoic acid, 13 or with the synthetic retinoid 13cis-retinoic acid. 14 •1s Nevertheless, it is clear that
formation of uninflamed comedones often begins
in the early stages of puberty, when sebum secretion rate is just beginning to rise.
Role of linoleic acid in skin
It has long been recognized that a dietary deficiency of linoleic acid can result in scaly skin and
decreased epidermal barrier function. 16 With the
discovery of the prostaglandins and related eicosanoid metabolites of linoleic acid, it seemed that
the basis for essential fatty acid requirement had
been determined. However, it has been demonstrated that linoleic acid itself can alleviate some
of the cutaneous symptoms of essential fatty acid
deficiency, even when its conversion to arachidonic acid, and subsequently to the eicosanoids,
is prevented. 17,18 Reasons for the specific requirement of linoleic acid became clearer following elucidation of the structures of two series of linoleaterich epidermal lipids that appear to playa role in
the formation of the water barrier. 19.20 These series
are based on w-hydroxy acids with very long
chains (C30 to C34). One series, the acylglucosyl
ceramides, is thought to induce assembly of the
lamellar granules in the epidermal granular cells.
Removal of the glucose moieties produces the second series, the acyl ceramides. This process is
accompanied by dispersal of the stacks of lamellar
disks that are extruded from the granular cells,
allowing their rearrangement into the multiple intercellular lipid bilayers that constitute the barrier.
Relation between sebum secretion rate and
linoleate content of sebum
The first clear evidence of a connection between
linoleate and acne was reported in 1976, when it
was shown that patients with acne had a signifi-
Volume 14
Number 2, Part 1
February, 1986
cantly lower level of linoleic acid in their skin
surface lipids than nonnal,21 Subsequent studies
have suggested that this effect results from the
higher sebum secretion rates that are characteristic
of acne, since there is an inverse relationship between secretion rate and the linoleate content of
skin surface wax esters, 22 which are purely of sebaceous origin. Studies have now shown that reduction in the rate of sebum secretion by treatment
with the antiandrogen cyproterone acetate 23 or with
13-cis-retinoic acid (unpublished data) causes a
corresponding increase in the linoleic acid content
of the sebaceous wax esters, as well as in the
triglycerides, free fatty acids, and cholesteryl esters in the skin surface lipids.
We now wish to present a hypothesis that has
been formulated to explain these changes in linoleate content that result from changes in sebum
secretion rate. A corollary addresses the question
of how a low sebum linoleate concentration could
initiate the development of acne.
THE HYPOTHESIS
The linoleate concentration in human sebum depends on the quantity of linoleic acid present in
each sebaceous cell at the commencement of its
differentiation and on the extent to which this initial endowment is diluted by the subsequent endogenous synthesis of lipid in each sebaceous cell.
At the time of cell division, when the sebaceous
cells still have contact with the basement membrane, they still have access to circulating lipids,
including linoleate. However, studies of sebum
composition indicate that once sebum synthesis
begins, no further lipids are accepted from the
circulation, so that the more sebum that is synthesized per cell, the more the initiallinoleate content will be diluted. Most likely, the linoleate initially will be incorporated into phospholipids in
the cell membranes. This linoleate will be released
and incorporated into each of the sebaceous ester
lipids at the time of final cell disruption, in proportion to the relative rates at which these lipids
are being synthesized at the time of cell disintegration. Cholesteryl esters should incorporate the
greatest proportion of this linoleate, since they are
likely to be synthesized principally at the end of
the life of the cells. Triglyceride should also in-
Essential fatty acids and acne
223
corporate much of the linoleate, since diglyceride
intermediates already containing linoleate should
become available through hydrolysis of the phospholipids. However, triglycerides are synthesized
throughout the differentiation of the sebaceous
cell, under conditions in which little linoleate may
be available. Wax esters also are known to be
synthesized continuously during the differentiation
of the sebaceous cells, and hence only a small
proportion would be synthesized at the time of cell
disruption. The result of this scenario could be that
the highest content of linoleate is found in the
cholesteryl esters, the next highest in the triglycerides, and the lowest in the wax esters, which,
in fact, is the case. 23
THE COROLLARY
Low concentrations of linoleate in sebum impose a state of essential fatty acid deficiency on
the cells of the follicular epithelium, thereby inducing the characteristic responses of hyperkeratosis and decreased barrier function.
Like all epidermal cells, those of the follicular
epithelium can be expected to obtain adequate essential fatty acids from the general circulation (albeit by diffusion from the dermis, since the epidermis lacks a vascular circulation). Additional
fatty acids would be available from sebaceous lipids that diffuse in from the follicular lumen. However, when the secretion rate of sebum is high and,
as a result, its linoleate concentration is low, the
cells of the follicular epithelium might thereby be
subjected to lipids deficient in essential fatty acids.
This deficiency might be especially effective in
influencing the cells if the linoleate-deficient lipids
were presented as free fatty acids, which occurs
when the sebaceous triglycerides are hydrolyzed
by follicular bacteria (P. acnes and Malassezia
furjur).24
Support for the corollary has been obtained by
examination of the polar lipids recovered from
comedones,25 the acyl ceramides of which contained only 6% linoleate among the esterified fatty
acids, in comparison with 45% in the acyl ceramides from normal human epidermis. In comparison, epidermal acyl ceramides from essential
fatty acid-deficient rats contained 2% esterified
linoleate, whereas those from normal rats con-
224
Journal of the
American Academy of
Dermatology
Downing et al
tained 38%.26 These data can be taken as a clear
indication of essential fatty acid deficiency in the
comedo-fonning epithelial cells in which the acyl
ceramides were synthesized. Furthennore, the esterified acids from the comedo acyl ceramides included a high proportion of L\6-hexadecenoic acid,
which is available only from sebum, providing
direct evidence that the cells of the follicular epithelium incorporate sebaceous fatty acids.
This evidence of essential fatty acid deficiency
in the follicular epithelium suggests three additional mechanisms that may be relevant to the
pathogenesis of acne. First, as long as the barrier
properties of the follicular epithelium are intact,
availability of water is likely to be a limiting factor
in the growth of follicular microorganisms. Decline in the effectiveness of the water barrier,
which is characteristic of essential fatty acid deficiency, would therefore be likely to promote bacterial growth and the liberation of additional free
fatty acids from sebaceous triglycerides. Second,
penetration of the linoleate-deficient free fatty
acids from the follicular lumen into the viable cells
of the follicular epithelium would also be promoted by decreased barrier function, which would
further compromise the ability of the epithelial
cells to produce the !inoleate-rich lipids that are
thought to be required for barrier function. Third,
chemotactic factors would more readily penetrate
the deficient follicular epithelium and promote inflammation.
IMPLICATIONS
The hypothesis indicates how elevated sebum
production alone could result in essential fatty
acid-deficient sebum, and the corollary postulates
how this effect may result in the follicular hyperkeratinization that leads to acne. Together the hypothesis and the corollary may explain the effectiveness of treatments for acne that reduce sebum
secretion (estrogens, antiandrogens , and 13-cisretinoic acid). However, it is clear that a change
in sebum secretion rate can result either from a
change in the amount of sebum synthesized per
cell or from a change in the sebocyte mitotic rate
(and thereby the number of sebocytes producing
sebum). According to our present hypothesis, it is
only an increase in the amount of sebum per cell
that will reduce the linoleate content of sebum.
Awakening of the pubescent sebaceous gland, under the influence of extragonadal honnones, appears to result in such an increase in the amount
of sebum produced per sebocyte, since linoleate
content decreases dramatically long before the
principal increase in sebum secretion that ensues
from the effects of testicular or ovarian testosterone. Thus our data on the decline in sebum linoleate concentration between 6 and 10 years of age
correspond with the onset of comedonal acne in
this age range.
For the corollary to be valid, it must be the
reduced concentration of linoleate, rather than a
reduced amount, that results in deficiency of essential fatty acids, since the hypothesis predicts
no change in the absolute amount of linoleate secreted by the sebaceous glands when its concentration is diluted by increased sebum output. The
question is dependent on the specificity of the enzyme that incorporates the linoleate into the acyl
ceramides. Ifthe specificity for linoleate were absolute, dilution of the available linoleate might not
be significant, as long as the required absolute
amount remained available. However, it is known
that the specificity is not absolute, since in deficiency of essential fatty acids, oleate can replace
linoleate. Furthermore, in comedones, L\6-hexadecenoic acid is also incorporated into the acyl
ceramides. Therefore dilution of the available linoleate with sebaceous fatty acids affects the acyl
ceramide composition and thereby disturbs the differentiation of the follicular epithelium. This circumstance leaves open the possibility that increased systemic availability of linoleate might
help to counteract the effect of decreased linoleate
concentration in sebum.
REFERENCES
1. Michaelsson G: Diet and acne. Nutr Rev 39:104-106,
1981.
2. Hubler WR: Unsaturated fatty acids in acne. Arch Dermatol 79:644-646, 1959.
3. Hoehn GH: Acne and diet. Cutis 2:389-394, 1966.
4. Downing DT, Strauss JS: On the mechanism of sebum
secretion. Arch Dermatol Res 272:343-349, 1982.
5. Downing DT: Mammalian waxes, in Kolattukudy PE,
editor: Chemistry and biochemistry of natural waxes.
New York, 1976, Elsevier Science Publishing Co., Inc.,
pp. 17-48.
Volume 14
Number 2, Part 1
February, 1986
Essential fatty acids and acne
6. Downing DT, Strauss JS, Norton LA, et al: The time
course of lipid formation in human sebaceous glands. J
Invest Dermatol 69:407-412, 1977.
7. Stewart ME, Downing DT, Pochi PE, Strauss JS: The
fatty acids of human sebaceous gland phosphatidylcholine. Biochim Biophys Acta 529:380-386, 1978.
8. Colton SW VI, Downing DT: The time course of lipid
biosynthesis in horse skin, Biochim Biophys Acta
836:306-311, 1985.
9. Lavker RM, Leyden JJ, McGinley KJ: The relationship
between bacteria and the abnormal follicular keratinization in acne vulgaris. J Invest Dermatol 77:325-330,
1981.
10. Harris HH, Downing DT, Stewart ME, Strauss JS: Sustainable rates of sebum secretion in acne patients and
matched normal control subjects. JAM ACAo DERMATOL
8:200-203, 1983.
11. Pochi PE, Strauss JS: Sebaceous gland suppression with
estradiol and diethylstilbesterol. Arch Dermatoll08:21O215, 1973.
12. Hammerstein J, Meckies J, Leo-Rossberg I, et al: Use
of cyproterone acetate (CPA) in the treatment of acne,
hirsutism and virilism. J Steroid Biochem 6:827-836,
1975.
13. Strauss JS, Pochi PE, Whitman EN: Suppression of sebaceous gland activity with eicosa-5 :8 : 11 : 14-tetraynoic
acid. J Invest Dermatol 48:492-493, 1967.
14. Peck GL, Thomas TO, Yoder FW, et al: Prolonged remissions of cystic and conglobate acne with 13-cis-retinoic acid. N Engl J Med 300:329-333, 1979.
15. Farrell LN, Strauss JS, Stranieri AM: The treatment of
severe cystic acne with 13-cis-retinoic acid: Evaluation
of sebum production and the clinical response in a multiple-dose trial. JAM ACAo DERMATOL2:602-611 , 1980.
16. Holman RT: Essential fatty acid deficiency, in Holman
RT, editor: Progress in the chemistry of fats and other
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
225
lipids. New York, 1968, Pergamon Press, Inc., pp. 275348.
Prottey C: Investigation of functions of essential fatty
acids in the skin. Br J DermatoI97:29-38, 1977.
MacDonald ML, Rogers QR, Morris JG: Role oflinoleate
as an essential fatty acid for the cat independent of arachidonate synthesis. J Nutr 113:1422-1433, 1983,
Wertz PW, Downing DT: Olycolipids in mammalian epidermis: Structure and function in the water barrier. Science 217:1261-1262, 1982.
Abraham W, Wertz PW, Downing DT: Linoleate-rich
acylglucosylceramides of pig epidermis: Structure determination by proton magnetic resonance. J Lipid Res
26:761·765,1985.
Morello AM, Downing DT, Strauss JS: Octadecadienoic
acids in the skin surface lipids of acne patients and normal
subjects. J Invest Dermatol 66:319-323, 1976.
Stewart ME, Wertz PW, Grahek MO, Downing DT: Relationship between sebum secretion rates and the concentration of linoleate in sebum and epidermal lipids.
Clin Res 33:684, 1985. (Abst.)
Stewart ME, Greenwood R, Cunliffe WJ, et al: Effect
of cyproterone acetate-ethinyl estradiol treatment on the
percentages of linoleic and sebaleic acids in various skin
surface lipid classes. Arch Dermatol Res. (Submitted for
publication. )
Marples RR, Downing DT, Kligman AM: Influence of
Pityrosporum species in the generation of free fatty acids
in human surface lipids. J Invest Dermatol 58: 155-159,
1972.
Wertz PW, Miethke MC, Long SA, et al: The composition of the ceramides from human stratum corneum and
from comedones. J Invest Dermatol 84:410-412, 1985.
Wertz PW, Cho ES, Downing DT: Effect of essential
fatty acid deficiency on the epidermal sphingoFpids of
the rat. Biochim Biophys Acta 753:350-355, 1983.
ABSTRACTS
Immunofluorescence in skin specimens from three .
different biopsy sites in patients with scleroderma
Chen ZY, Dobson RL, Ainsworth SK, et al: Clin
Exp Rheumatol 3:11-16, 1985
Biopsies from nail fold, forearm, and buttock were compared with normal skin, sclerodermatous skin, and skin from
patients with systemic lupus erythematosus. Nail fold skin
was most reactive in eighteen scleroderma patients. The site
for biopsy affects the chance of diagnostic findings. Normal
skin was studied and was not reactive.
P. C. Anderson, M.D.
Depth of invasion and tumor thickness in primary
cutaneous malignant melanoma. A study of 2012
cases
Sondergaard K: Acta Pathol Microbiol Immunol
Scand [AJ 93:49-55, 1985
This study of 2,012 patients with melanoma urges tne use
of Clark levels. Melanomas from 0.76 to 1.5 mm thick, at
each specific thickness had the poorer prognosis according to
the depth of invasion by Clark's system.
P. C. Anderson, M.D,