Hormones and Hair Growth: Variations in Androgen
Recet&or Content of Dermal PatSlla Cells Cultured
fromLHuman and Red Deer (C&us EZqks)
Hair Follicles
Valerie Anne Randall, Margaret Julie Thornton, Andrew Guy Messenger,*
Andrew Stewart Irving Loudon,? and Barry Robert Brinklowt
Department of Biomedical Sciences, The University of Bradford; *Department of Dermatology, Royal Hallamshire Hospital, Sheffield
and TInstitute of Zoology, Regents Park, London, U.K.
Many hair follicles produce different types of hair in response
to environmental changes or the mammals age, that are
translated to the follicle by hormones. Androgens cause many
changes, such as transforming vellus follicles producing in-
significant hairs on the face to terminal beard ones at puberty
or the reverse on the scalp. In male red deer the breeding
season rise in androgens causes the annual production of a
mane on the neck that is lost during the spring.
Because the dermal papilla situated at the base of the hair
follicle is important in determining the type of hair ro-
duced, androgens may act via the dermal papilla. There Pore,
primary cell lines of dermal papilla cells from human and red
deer follicles with different res onses to androgens have
f
been established. Specific saturab e androgen receptors were
present in all human papilla cells examined, with higher
H
air follicles often produce different types of hairs at
various stages in a mammal’s life; the hairs formed
may change quite significantly in color, thickness,
and length, as can be readily seen by comparing the
facial hair of a man to those he had as a child. This
transforming ability of the hair follicle allows many mammals to
adjust to seasonal changes in the weather by increasing or decreasing
the thermal insulating properties of their coat or, like some arctic
mammals, altering the color to retain camouflage when the coun-
tryside colors change (reviewed in [l]). It also permits changes in
hair types during the lifecycle of a mammal, which allow rapid
distinction between young animals and adults and between the adult
sexes. This can be readily seen in lions, where the young are speck-
led for camouflage and the adult males have a very obvious mane,
but also occurs in human beings. Children have relatively little
terminal hair (i.e., pigmented, relatively long and thick) except for
that with protective roles on the scalp, eyebrows, and eyelashes,
whereas both sexes develop axillary and pubic hair during adoles-
cence. Sexually mature men are also readily identified by the char-
Part of the research on red deer dermal papilla cells has previously been
presented in 1991 at the British Society for Dermatology, Leicester, U.K.
and the Society for Endocrinology, London, U.K.
Reprint requests to: Dr. Valerie A. Randall, Department of Biomedical
Sciences, University of Bradford, Bradford BD7 lDP, UK.
Abbreviations: MSH, melanocyte-stimulating hormone; [3H]-miboler-
one, 7a, 17cu-dimethyl-1Pnortestosterone.
0022-202X/93/$06.00 Copyright 0 1993 by The Society for Investigative Dermatology, Inc.
114s
Nigel Andrew Hibberts,
levels in cells from androgen-dependent follicles, e.g., beard
than in control, non-balding scalp cells. In preliminary in-
vestigations of red deer, androgen receptors were only
present in cells derived from mane follicles and were unde-
tectable in flank or spring neck follicles.
These similar results from both species support the hy-
pothesis that androgens are acting on hair follicles via the
dermal papilla. They also suggest that dermal papilla cells are
potentially useful models for investigating the mechanism of
androgen action because cultured cells appear to retain dif-
ferences that relate to the androgen responsiveness of their
parent follicle. The red deer seems particularly interesting in
view of the much shorter hair-growth cycle than human
scalp or beard follicles. J Invest Dermatol 202:114S-
1993
12OS,
acteristic hair of the beard, chest, and suprapubic regions. This
major role of human hair growth in social and sexual communica-
tion explains why hair disorders such as hirsutism, androgenetic
alopecia, and alopecia areata, though not in themselves life-threaten-
ing, often create psychologic problems (reviewed in [2]) [3,4].
To enable these changes to occur hair follicles pass through regu-
lar cycles of hair development and growth (anagen) followed by
eriods of resting (telogen) [5,6] (see Fig 1). The hair produced may
! e very similar to the previous one, as seen in many hair follicle
cycles on the eyelids and the young human scalp; it may differ
slightly or even be quite markedly altered, such as those in the neck
region of the adult male red deer (Cervus elaphus), which produces a
mane every year only during the breeding season [i’]. The precise
mechanisms of how alterations occur in the type of hair produced by
a hair follicle are unknown. Further studies of hair follicle biology
are needed to elucidate this problem so that better treatments can be
developed to regulate the hair follicle disorders and increase hair
production by domesticated animals such as sheep. Our approach is
to study the hair follicular responses to hormones in both human
and red deer follicles. In particular, the role of androgens in the
control of hair growth is being investigated (reviewed in [2,8 - lo]),
as androgens dramatically modulate the type of hair produced by
hair follicles in both species.
REGULATION OF HAIR-GROWTH CYCLES
Seasonal coat changes occur twice a year in many mammals to
accommodate alterations in the environment [ 1 l] (reviewed in [ 11).
VOL. 101. NO. 1, SUPPLEMENT, JULY 1993
Figure 1. The early stages of anagen (seen in the lower diagram) of the
hair-growth cycle appear to partially recapitulate some of the stages of the
embryogenesis of the hair follicle (represented in the upper diagram). Re-
printed by permission of the New York Academy of Sciences from Randall
VA et al: Androgens and the hair follicle: cultured human dermal papilla cells
as a model system. Ann NY Acad Sci 642: 355 -375,1992. Copyright 1991
by the New York Academy of Sciences.
The hair growth cycles in these species are synchronized and, gener-
ally, waves of new hair growth and molting of the previous season’s
hair type pass over the body (e.g., [12]); however, in some s ecies,
such as the guinea pig, molting is more diffuse and does not gollow
such a sequential pattern [13]. These molts are linked to climatic
variation, particularly fluctuations in the length of daylight, but also
of temperature. In addition, the ty e of new hair produced is af-
f
fected by the availability of the foo B sup ly, presumably because of
[
the large metabolic requirements invo ved in making new hairs
1 l]. Because these parameters are also important for the timing of
seasonal reproduction, there is a correlation between the molting
and reproductive cycles in many mammals, e.g., red deer [7].
Changes in the environment are interpreted for the hair follicle
by the endocrine system. In many species this has been shown to
involve melatonin from the pineal plus the hormones of the hy-
othalmic-pituitary system, which generally act indirectly at the
Peve1 of the hair follicle via the gonadal, thyroid, and, possibly,
corticosteroid hormones. Melanocyte-stimulating hormone (MSH)
and prolactin are pituitary hormones that have been ascribed direct
actions on hair follicles (reviewed in [1,2]).
Hair follicles whose role involves social and sexual communica-
tion usually indicate sexual maturity, e.g., human pubic and axillary
hair or the loss of juvenile coats in many species such as red deer.
They may also distinguish between the adult sexes, like the beard
and greater body hair of men and the manes of lions. The transfor-
mation of these follicles is related to the changes in hormonal levels
at puberty, particularly androgens (reviewed in [14,15]) [1,2,8-lo].
The seasonal development each year of the mane and antlers in the
red deer also parallels the annual rise in androgens in adult males [7].
Human hair growth is generally believed to differ from that of
other mammals, not only because there is relatively little present on
the “naked ape,” but also because after synchronous molting of
follicles before, and just after, birth [16] human hair follicles have
been considered to be asynchronous, except for local groups of three
hairs, the De meijhe trios [17]. Nevertheless, human hair follicles are
influenced by a range of circulating factors, including thyroid hor-
mones [18,19], pregnancy [20] (Randall and Swann, unpublished
ANDROGEN RECEPTORS IN HUMAN AND DEER HAIR CELLS 115s
observations), and nutritional status [21]. Pronounced seasonal varia-
tion has also been detected [22- 241, which indicates some de ree of
synchrony in follicular activity. The single annual peaks o B beard
and thigh hair growth in the summer reported by Randall and
Ebling [24] may well be due to the raised testosterone levels detected
in Euro ean men at this time (reviewed in [24]), like the seasonal
mane o P the red deer [7]; however, this seems unlikely to be the only
answer for the biannual fluctuations of anagen on the thigh that
resemble those of seasonally molting animals. The single cyclic
pattern they reported on the scalp could possibly be regulated by
androgens in men, but this seems less likely, as it also occurs in
women [22].
HORMONES AND THE INTRINSIC RHYTHMS
OF HAIR FOLLICLES
The response of a hair follicle to hormones appears to be intrinsic to
that particular hair follicle. This was shown in classic ex eriments
by Ebling and colleagues where hair follicles trans lame 8 to differ-
ent sites on the same [25] or genetically similar P261 rats retained
their donor site patterns for some time. Eventually, circulating hor-
mones must have overcome the intrinsic rhythms as the homograft
follicles came into phase with their hosts [26] and parabiotic rats
gradually molted together [27].
Human follicles also show clear evidence of intrinsic activity in
their responses to androgens, as discussed in [2]. Des ite exposure to
the same circulating hormones, the reactions of fol Picles depend on
their body site, varying from no effect on the eyelashes, to stimula-
tion in many areas [28], and regression and balding on the scalp in
genetically disposed individuals [29]. The gradual responses of these
follicles to androgens, taking many years to have full effects [30],
parallel the rat experiments with hormones eventually overriding
the internal follicular cycle [26,27], but the situation in human
beings is more complex; not only do human follicles vary in their
reactions to androgens but they retain their endogenous responses
when transplanted, this being the basis of corrective transplants for
androgenetic alopecia [31]. Presumably, this characteristic response
of each hair follicle to androgens is determined by alterations in
gene expression in the cells of individual hair follicles during em-
bryogenesis imposed by as-yet unknown factors. This makes the
hair follicle a potentially interesting tissue to scientists from a range
of disciplines.
In the red deer the neck follicles exhibit an intrinsic ability to
react to male sex hormones, in contrast to other body follicles.
However, the result differs markedly from the situation in humans
because a quite dramatic enlargement occurs in one cycle and the
follicle then produces a much smaller hair in the next cycle with the
fall in testosterone during the non-breeding season.
HOW DO ANDROGENS REGULATE HAIR FOLLICLES?
Androgens are very obvious regulators of hair growth, as reviewed
in [1,2,8- 10,14,15]. In humans they stimulate the gradual transfor-
mation of small vellus follicles, making non-pigmented, fine, short
hairs, in many areas to larger terminal follicles producing pigmented,
thicker, longer hairs, as illustrated in Figure 2. This ap ars to occur
more rapidly in other species, such as in the manes o p”the deer and
lion, but these follicles are already reducing terminal hairs as part
of their normal coat. In contrast to t l! e androgen-stimulated produc-
tion of terminal follicles, an androgen-potentiated gradual regres-
sion of terminal hair follicles to vellus ones may occur on the human
scalp [8] and has also been described on the scalp of both sexes of the
stump-tailed macque [32]. This appears to involve the opposite re-
sponses (see Fig 2). Because these paradoxical reactions are clearly
end-organ responses to the same hormones, the key to understand-
ing them and, hopefully, therefore further elucidating the control
of the hair growth cycle, must lie within the hair follicles them-
selves.
To cause these modifications, androgens must alter, either di-
rectly or indirectly, the activity of many components of the hair
follicle, in particular the melanocytes, and epithelial and dermal
papilla cells. Randall’s hypothesis for some years now [2,8-
,
116s RANDALL ET AL
FACE
d
short, fine long thick
I
c unpigmented
hair
pigmented -
hair
+ protein X in beard hair follicle cells although inhibiting it in scalp
ANDROGENS
SCALP
unpigmerded _
ANDROGENS
Fi re 2. Dia rammatic representation of the gradual changes that occur
in ??uman hair 4;.olhcles m response to androgens. This occurs both in areas
stimulated to produce longer hairs by androgens, such as a man’s face or the
axillae of both sexes (upperdiagram),and in those on the scalp where follicles
are inhibited in genetically disposed individuals ([ower diagram). Reprinted
by permission of the New York Academy of Sciences from Randall VA et al:
Androgens and the hair follicle: cultured human dermal papilla cells as a
model system. Ann NY Acad Sci 642: 355-375, 1992. Copyright 1991 by
the New York Academy of Sciences.
10,33,34] has been that androgens are acting on the dermal papilla
cells, which in turn secrete paracrine factors that influence the other
follicular components; these factors are currently unknown, but
would probably include soluble mitogenic factors and/or extracel-
lular matrix components. Briefly, this hypothesis, which has been
discussed in more detail elsewhere [2,8 - lo], is based on the concept
that the dermal papilla plays an essential role in inducing and main-
taining epithelial differentiation, as suggested by the exciting ex-
eriments carried out in Dundee, Scotland by the research group led
g y Oliver and Jahoda ([35,36] and this issue, p. 33s). During the
hair-growth cycle the follicle appears to recapitulate in early anagen
some of the ste s involved in the embryogenesis of the hair follicle,
as can be seen ! y comparing these processes in Figure 1. Because
androgens are believed to act on the epithelium via the mesenchy-
ma1 cells in the developing prostate, the classic androgen target
organ [37] and autoradiography of ‘[HI-testosterone had located the
radioactivity only in the dermal papillae but not in the follicular
epithelial cells [38], it seemed logical that androgens could act on
other follicular components indirectly via the dermal papilla.
Androgens, like other steroid hormones, are‘believed to act via
specific intracellular androgen receptors, which are present in all
target cells. After diffusing across the plasma membrane the steroids
THE JOURNAL OF lNVESTIGATIVE DERMATOLOGY
combine with the receptors, which then undergo a change in config-
uration, revealing sites that can bind the specific hormone-response
elements in the chromation. Binding of the hormone-receptor
complexes to these sites stimulates alterations in RNA production,
processing, and translation, leading ultimately to changes in rotein
synthesis and modification that depend on the original speci f!c gene
expression in that cell type (reviewed in [39]). In other words,
androgen-receptor complexes might stimulate the production of
cells. More details of the mechanism of androgen action in hair
follicles have recently been given elsewhere [2,8- lo].
The relevance of the general model for androgen action to andro-
gen-dependent hair growth can readily be seen in people with testic-
ular feminization. These genetic men are phenotypically female
despite relatively normal androgen levels because they lack func-
tional androgen receptors. They do not produce any androgen-de-
pendent hair growth, not even the female axillary and pubic pat-
terns, and do not go bald [40], p roviding very good evidence for the
role of androgens in human hair growth. Recently, androgen recep-
tors have been localized only in the dermal papillae of human hair
follicles using a specific monoclonal antibody to the androgen re-
ceptor [41]. This gives strong support to our hypothesis that andro-
gens act via the dermal papilla.
INVESTIGATIONS INTO ANDROGEN ACTION IN
HUMAN AND RED DEER FOLLICLE DERMAL
PAPILLA CELLS
Experimental Design Based on this hypothesis, we are trying to
understand how androgens affect hair follicles by investigating the
mechanism of androgen action in cultured dermal papilla cells from
human and red deer hair follicles. The main thesis behind our ap-
proach is that comparisons of the mode of androgen action in cells
from follicles that are androgen-dependent, such as human beard or
red deer mane, with those from control follicles, such as non-bald-
ing scalp, deer flank or non-breeding-season deer neck follicles,
should reveal important information about the androgenic control
of hair follicles.
Dermal papilla cells seem a particularly useful model system for
studying hair follicles not only because of the important functions
of the dermal papilla and its putative role in androgen action, as
discussed above, but also because they are relatively easy to isolate as
a distinct population and culture and because rat and human cells
have been shown to retain the ability to induce hair growth after
culture (reviewed in [36]).
Characteristics of Cultured Dermal Papilla Cells Dermal
papilla cells are a specialized type of fibroblast with distinct proper-
ties in culture [42,43]. Confluent fibroblasts normally show an
elongated spindle shape and grow in characteristic arrays on plastic
dishes, whereas dermal papilla cells from both species maintain a
more spread-out morphology and a gregate intoiittle groups, fre-
quently lvincr partially on top of eat E other and leaving areas of the
dish s&&e ;&overe’d [9]. &man dermal papilla cellHgrow much
more slowly than dermal fibroblasts and the cultures are normally
much more short-lived [43]. Our preliminary studies on red deer
dermal papilla cells have found that they differ from the human
attern and grow faster than deer dermal fibroblasts$ [l]; this could
g e related to the shorter anagen of red deer follicles or the much
younger age of the red deer at adulthood.
The Growth of Dermal Papilla Cells in the Presence of An-
drogens Because the measurement of cell growth is much easier
than the assay of androgen rece tors, we first investigated the effects
of androgens on the growth o 4 beard and non-balding scalp dermal
papilla cells. An alteration in growth rates would have indicated the
$ Thornton MJ, Brinklow BR, Louden ASI, Randall VA: The culture of
dermal papilla cells from the mane and flank of the red deer: a comparison of
growth and androgen receptor content (abstr). Br J Dermatol 125:485,
1991.
VOL. 101, NO. 1, SUPPLFMENT, JULY 1993 ANDROGEN RECEPTORS IN HUMAN AND DEER HAIR CELLS 117s

activation of both androgen receptors and a range of androgen-re-

sponsive genes. Unfortunately, neither cell type showed significant
changes in growth rates as assessed by [3H]-thymidine uptake over a
range of concentrations of either testosterone or the non-metabo-
lized androgen mibolerone [33]. As the lack of growth in response
to androgens does not mean that the cells are unable to respond to
androgens in a more delicate manner than the gross response of
alterations in cell growth, androgen receptor content was measured
in dermal papilla cells from both species.

MATERIALS AND METHODS

Culture of Dermal Pa illa Cells In both species dermal papilla
cells were derived from f: 11-thickness skin samples by microdissect-
ing the dermal pa illa from the lower part of the follicle. The
dermal papillae o P the androgen-potentiated hair follicles were
larger in both species than those from the control follicles ([8] and
unpublished observations). Dermal papillae explants were trans-
ferred to small, sterile, culture-treated, plastic Petri dishes and cul-
tured in medium El99 sup lemented with 20% fetal bovine serum
as previously described [42 P. All experiments described in this paper Figure 3. Competition for the androgen binding site in cultured dermal
have been carried out on individual primary cell lines at passages papilla cells by various natural and synthetic steroids. Confluent cultures of
between 3 and 6, inclusive. primary cell lines of beard and scalp cells were incubated with 1 nm [sH]-
mibolerone in the presence, or absence, of 100 nM unlabeled steroids for 2 h
Measurement of Androgen Receptor Content The binding at 37°C. The cultures were .nreviouslv incubated with serum-free medium to
of the synthetic non-metabolizable androgen [3H]-mibolerone (7a, reduce the endogenous levels of androgen. Each point represents the
17~dimethyl-19-nortestosterone, specific activity 2.74-3.33 mean f SEM of at least two measurements in each of three scalp or two
TBq/mmol) (Amersham International plc) was measured in dermal beard dermal papilla cell lines.
papilla cells from both species cultured until almost confluent in
IOO-mm Petri dishes in normal medium; they were then grown in
non-balding scalp follicles. These results follow the normal pattern
medium without serum for 24 h prior to the assay to reduce intra-
for competition studies of androgen receptors and demonstrate the
cellular androgens. A saturation analysis was carried out over the
presence of specific androgen receptors.
range 0.05- 10 nM [3H]- ml‘b0 1erone involving nine or ten points in
Cells from beard, moustache, pubis, and scrotum (Kd = 0.22
each of the 12 primary cell lines investigated, non-specific binding
nmol/l; n = 8) and also Shinogi 115 cells had a similar affinity for
was assessed in parallel incubations in the presence of 100 X excess
[3H]-mibolerone, but that of non-balding scalp cells was, rather
5a-dihydrotestosterone to saturate any non-specific binding sites
surprisingly, even greater (Kd 0.081 nmol/l; n = 4) [34]. Androgen
and 1000 X excess triamcinolone acetonide was included in all incu-
receptor content was calculated in relation to cell number, protein,
bations to counter any ossible binding of [3H]-mibolerone to pro-
and DNA content, but gave similar values by all three methods
gesterone receptors. A Pter incubation for 2 h at 37”C, the medium
[9,34]. The receptor concentration in androgen-dependent cells was
was removed and retained for counting to estimate the amount of
significantly higher (p < 0.05) than that of non-balding scalp cells
unbound radioactivity, i.e., “ free” radioactivity for Scatchard analy-
(Fig 4); in six different cell lines of beard and moustache cells the
sis. The cells were then washed 4 X with cold phosphate-buffered
B, was 0.033 fmol/104 cells, 17 fmol/mg protein, and 0.32 fmol/
saline to remove any radioactivity loosely associated with the cell
p DNA, whereas that of four scalp cell lines was only 0.01 fmol/
surfaces and removed by scraping into cold phosphate-buffered sa-
104, 6.0 fmol/mg protein, and 0.052 fmol/pg DNA. Cells from
line before separation by centrifugation. The cell pellets were resus-
female pubis and male scrotum contained even higher levels (0.063
pended in chloroform-methanol (1: 1, v : v) to extract the radioactiv-
fmol/I04 cells, 30.5 fmol/mg protein). The binding affinity [34]
ity and, after centrifuging out the cellular residue, the radioactivity
and receptor content (Fig 4) were related to passage number to see if
in each supernatant was measured on an LKB scintillation spectro-
there were any significant trends such as a fall off in receptor content
photometer with a counting efficiency of 50%. Scatchard plots
with increasing passaging, as has been detected in the inductive
were constructed for each cell line to enable the calculation of the
capacity of rat dermal papilla cells (441. No significant changes were
affinity (Kd) of the receptors for [3H]-mibolerone and the concen-
detected in relation to the passage number with either binding af-
tration of androgen receptors (i.e., B,) [34]. The androgen-re-
finity or receptor content, but the actual range investigated was
sponsive mouse mammary carcinoma cell line, Shinogi 115, was
small, ranging from P3 to P6; earlier passage cells are difficult to
assayed as a positive control [34].
obtain in the numbers necessary for a full androgen receptor assay.
The ability of various steroids to inhibit the binding of 1 nM
PHI-mibolerone to the dermal papilla cells was assessed by incubat- Red Deer Investigations Primary cell lines have been estab-
ing the cells in duplicate or triplicate with a range of unlabeled lished from large autumn breeding season neck, i.e., mane follicles,
steroids at 100 nM using the same methodology. from the smaller spring neck follicles, and from the flank in both
seasons. In our preliminary investigations saturable androgen recep-
RESULTS tors have been detected in cells derived from male mane follicles
Human Studies The androgen receptor assays demonstrated the (n = 2), but not in cells derived from the corresponding non-
presence of specific, high-affinity, low-capacity androgen receptors breeding-season neck follicles (Fig 5). Little or no specific binding
in cultured dermal papilla cells from both androgen-dependent and was seen in cells obtained from flank hair follicles in the breeding
non-balding scalp hair follicles with roperties similar to those of season (Fig 5). These are very interesting observations, which merit
classical androgen target tissues. Bin P mg was inhibited by around further study.
50% by the androgens testosterone, 5a-dihydrotestosterone, and
DISCUSSION
mibolerone, in both beard and scalp cells. The antiandrogen cypro-
terone acetate and estradiol also caused reductions but progesterone, Cultured Dermal Papilla Cells from Human and Deer Hair
triamcinolone acetonide (with progestational and corticosteroid Pollicles: A Model System for Studying Androgen Action in
binding), and cortisol had little effect (Fig 3). The same pattern of Hair Follicles The findings of significant levels of androgen re-
specificity was seen in dermal papilla cells devised from beard or ceptors in cultured dermal papilla cells are in accordance with our
 118s RANDALL ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

. P4 cells at least, requires circulating levels of androgens. In other

words, androgens appear to be inducing the dermal papilla cells to
alter their gene expression and induce androgen receptors so that
they can respond to androgens. Because androgen receptors are
present only in cells from follicles producing large hairs as a result of
androgen stimulation, it seems probable that the alteration of an-
. P3 drogen receptor gene expression has taken place either prior to, or
during, early anagen; very interestingly, these alterations are main-
20
tained during anagen and retained when these cells are cultured. If

c
t
4
androgens have to cause alterations in gene expression before or
10 . P6

::;
! P4
0
scakJ Beard Genital

Figure 4. Differences in the concentration of androgen receptors in cul-

tured human dermal papilla cells derived from scalp, beard, and genital skin
hair follicles. Androgen receptor content was measured by assaying the
binding capacity of confluent IO-mm dishes of primary cell lines over a
range of concentration of [3H]-mibolerone (0.05 - 10 ti, 9 or 10 points)
after a 2-h incubation at 37°C. Non-specific binding was assessed by parallel
incubations in the presence of 100 X excess 5cr-dihydrotestosterone and
1000 X excess of triamcinolone acetonide was added to all dishes to revent
binding to any progesterone receptors. The cells had been incubated Por 24 h
previously in serum-free medium to reduce endogenous androgens. The
specific binding (i.e., the difference between total and non-specific binding)
is that due to binding to the low-capacity, high-affinity receptors; it was
equivalent to about 175 cpm at each point between 1 and 10 nM [3H]-mibo-
lerone. P, assage number. Reprinted by permission of Blackwell Publica-
tions Ltd, Prom Randall VA, Thornton MJ, Messenger AG: Cultured dermal
papilla cells from androgen-dependent human hair follicles (e.g. beard) con-
tain more androgen receptors than those from non-balding areas of scalp.
J Endocrinology 133: 141- 147, 1992. Reproduced by permission of the 2 4 lb
Journal of Endocrinology Ltd. Cone BH-Mibolerone nM

original hypothesis that androgens are acting via the dermal papilla;
b
the presence of higher levels of androgen receptors in cells derived
from androgen-dependent hair follicles in both human and red deer
strengthens this concept. However, the detection of measurable
amounts of androgen receptors in non-balding scalp dermal papilla
cells, but not in those of control red deer follicles, is an intriguing
difference. Whether or not a scalp hair follicle from “non-balding”
regions can be truly said not to be an androgen target tissue is
perhaps debatable; some people with very well developed androge-
netic alopecia show regression of hair follicles over much of the
scalp.
Nevertheless, we have already detected significant differences in
many aspects of the mechanisms of androgen action in human der- Cone 3H-Mibolerone nM
ma1 papilla cells from beard, pubis, and scalp follicles, as reported
elsewhere [8,9]. This includes a greater ability of beard cells to form
Set-dihydrotestosterone [45], believed to be the active intracellular
androgen in some hair follicles, a larger mitogenic response to me-
dium conditioned by other dermal papilla cells, and the production
only by beard cells of an as-yet unknown factor that stimulates the
growth only of beard dermal papilla cells [46]. All of these differ-
ences support the hypothesis that androgens are acting through the
dermal papilla in hair follicles, though ofcourse they do not exclude
actions on other follicular cell types. The differences in both human
and red deer between dermal papilla cells from androgen-dependent
and androgen-independent follicles suggest that alterations in gene
expression, presumably determined by their location on the body
and related to their androgen responsiveness, are retained in culture. 2 4 s lb
This makes both human and red deer dermal papilla cells appear to Cone 3H-Mibolerone nM
be good model systems for further studies of androgen action. These
advantages should outweigh the quite major drawbacks of obtain- Pigure 5. Variations in androgen receptor content in cultured dermal pa-
pilla cells derived from mane and flank follicles of a red deer stag at different
ing the initial full-depth skin material in both species and the slow-
stages of the year. Androgen receptor measurements were carried out as
growing, short-lived nature of the cells produced. described in the legend of Fig 4. Cells were derived from mane (a) and flank
Despite the current preliminary nature of the results, the changes (b) follicles in the autumnal breeding season and from the neck region of the
in androgen receptor content of cells from deer neck follicles in same animal in the spring when no mane was present (c). Solid circles, total
direct correlation to their androgen responsiveness at the time of binding of [sH]-mibolerone; open &c/es, non-specific binding; dotted line,
culture suggest that the presence of androgen receptors, in these specific binding.
VOL. 101. NO. 1, SUPPLEMENT, JULY 1993 AND1 lOGEN RECEPTORS IN HUMAN AND DEER HAIR CELLS 119s

very early in the hair cycle, this could account for the gradual hair follicles with varying responses to androgens in viva. J Invest
changes seen in human hair follicles in response to andro ens (Fig 2) Dermato198:865-915, 1992
because anagen in many areas, including the scalp and f!ace, is very IO. Randall VA: The mechanism of action of androgens in the human hair
long. Nevertheless, if the seasonal fluctuations in beard and thigh follicle. Proceedings of 1st Terra Symposium: Androgens & Antian-
growth [24] are related to androgens, as seems likely, this means that drogens (in press)
at least growth rate is able to be altered during the cycle. On the Il. Johnson E: Seasonal changes in the skin of mammals. In: Spearman
other hand, the biannual fluctuation in anagen in the thigh appeared RIC (ed.). ComparativeBiologyofSkin. SymposiumoftheZoologi-
independent of whatever was controlling the growth rate. The cal Society of London, Vol. 39. Academic Press, London, 1977, pp
length of anagen has been suggested to play a major role in the 373-404
response of hair follicles to androgen [4i’] and it is possible that this 12. Johnson E: Quantitative studies of hair growth in the albino rat. I.
may be controlled in a different way from the rate of growth. The Normal males and females. J Endocrinol 16:337-350, 1958
complex actions of hormones, in general, and androgens, in particu- 13. Jackson D, Church RE, Ebling FJ: Hair diameter in female baldness. Br
lar, on hair follicles are still very little understood. J Dermatol 87:361-367, 1972
Further studies on the models discussed here should hopefully 14. Ebling FJG: Hair follicles and associated glands as androgen targets.
yield important information about how androgens affect hair Clin Endocrinol Metab 15:319-339, 1986
growth. Beard or mane, scalp, or spring neck/flank follicles offer 15. Ebling FJG: Hair. In: Rook A, Wilkinson DS, Ebling FJG, Champion
the possibility of comparing androgen-potentiated and relatively RH, Burton JL (eds.). Textbook of Dermatology, 5th ed. Blackwell
Scientific Publications, Oxford, 1992
independent tissues. The red deer is particularly promising, as it
offers androgen-dependent hair follicles with much shorter hair- 16. Pecoraro V, Astore I, Barman JM: The pre-natal and post-natal hair
growth cycles. Our human studies have focussed on androgen-po- cycles in man. In: Baccaredda-Boy A, Moretti G, Frey JR (eds.).
Biopathology of Pattern Alopecia. Karger, Basel, 1968, pp 29 -38
tentiated follicles rather than balding follicles on the basis that posi-
tive changes are probably easier to detect than inhibitory ones; 17. Saitoh M, Uzuka M, Sakamoto M: Human hair cycle. J Invest Derma-
to1 54:65-81, 1970
however, it is probable that information from such studies should be
relevant to androgen inhibition of hair growth and should lead to 18. Eckert J, Church RE, Ebling FJ, Munro DS: Hair loss in women. Br J
Dermatol 79:543-548, 1967
better control of hair disorders in general, whether androgen-re-
lated or not. In addition, the relatively accessible hair follicle has 19. Jackson D, Church RE, Ebling FJ: Hair diameter in female baldness. Br
J Dermatol 87:361-367, 1972
strong parallels to other androgen targets, such as the prostate (dis-
cussed in [2,8,10]); further studies may well provide greater under- 20. Lynlield YL: Effect of pregnancy on the human hair cycle. J Invest
Dermatol 35:323-327, 1960
standing of androgen action in human tissues, which would be
21. Bradfield RB: Protein deprivation: comparative response of hair roots,
valuable in understanding hormone-related cancer.
serum protein and urinary nitrogen. Am J Clin Nutr 24:405 - 510,
1971
22. Orentriech N: Scalp hair replacement in man. In: Montagna W, Dob-
This paper is dedicated to the late Professor F.John G. Ebhg, without whose classic son RL (eds.). Advances in Biology of Skin, Vol IX. Hair Growth.
research andpersonaltraining ofone ofus (VAR) this work may well not have been Pergamon Press, Oxford, 1969, pp 9- 108
carried out. 23. Saitoh M, Uzuka M, Sakamoto M: Human hair cycle. J Invest Derma-
The technical assistance ofKaherine Elliott, Department ofDermatology, Royal to1 54:65-81, 1970
Hallamshire Hospital, Shejeld, and David Thomas of the Institute of Zoology, 24. Randall VA, Ebling FJG: Seasonal changes in human hair growth. BrJ
London, in the obtaining oJsample material is gratefully acknowledged. We also Dermatol 124:146-151, 1991
thank Jean Maleki and Margaret Wrightfor the preparation of the manuscript, and 25. Ebling FJG, Johnson E: Hair growth and its relation to vascular supply
Jenny Braithwaite and Chris Bowersfor Be artwork. in rotated skin arafts and transuosed elaos in the albino rat.
Thepersonal research described in this paper was supported byprojeclgrantsfrom J Embryo1 Exp M&phol7:417-450,195’9 *
the Medical Research Council to Doctors V.A. Randall and A.G. Messenger 26. Ebling FJG, Johnson E: Systemic influence on activity of hair follicles
(G8610976/SB) and the Agriculture and Food Research Council to Doctors V.A. in skin homografts. J Embryo1 Exp Morph01 9:285-293, 1961
Randall, B.R. Brinklow, and A.S.I. Loudon (AG63/529). 27. Ebling FJG, Hervey GR: The activity of hair follicles in parabiotic rats.
J Embryo1 Exp Morph01 12:425-438, 1964
28. Hamilton JB: Age, sex and genetic factors in the regulation of hair
REFERENCES growth in man: a comparison of Caucasian &Japanese populations.
In: Montagna W, Ellis RA (eds.). The Biology of Hair Growth.
1. Ebling FJG, Hale PA, Randall VA: Hormones and hair growth. In: Academic Press, New York, 1958, pp 399-433
Goldsmith L (ed.). Biochemistry and Physiology of the Skin, 2nd Hamilton JB: Male hormone stimulation is a prerequisite and an inci-
29.
ed. Oxford University Press, New York, 1991 tant in common baldness. Am J Anat 71:451-480, 1942
2. Randall VA: Androgens and human hair growth. Clin Endocrinol (in Hamilton JB: A secondary sexual character that develops in men but
30.
press) not in women upon aging of an organ present in both sexes. Anat
3. Moerman DE: The meaning of baldness and implications for treat- Ret 94:466-467,1946
ment. Clin Dermatol 6:89 - 92, 1988 31. Orentreich N, Durr NP: Biology of scalp hair growth. Clin Plast Surg
4. Patzer GL: Psychologic and sociologic dimensions of hair: an aspect of 9:197-205, 1982
the physical attractiveness phenomenon. Clin Dermatol6:93- 101, Uno H: Stumptailed macques as a model of male pattern baldness. In:
32.
1988 Maibach HI, Lowe NJ (eds.). Models in Dermatology. Vol 3.
5. Pinkus H: Embryology of hair. In: Montagna W, Ellis RA (eds.). The Karger, Basel, 1987, pp 159-169
Biology of Hair Growth. Academic Press, New York, 1958, pp Thornton MT. Messenger AG. Elliott K, Randall VA: The effect of
33.
l-32 androgens>in the griwth 02 cultured’human dermal papilla cells
6. Kligman AM: The human hair cycle. J Invest Dermatol 33:307 - 16, derived from beard and scalp hair follicles. J Invest Dermatol
1959 97:345-348, 1991
7. Lincoln GA: The seasonal reproductive changes in the red deer stag 34. Randall VA, Thornton MJ, Messenger AG: Cultured dermal papilla
(Cervus elaphus). J Zoo1 163:105-123, 1971 cells from androeen-deoendent
” 1
human follicles (e.e. beard) contain
8. Randall VA, Thornton MJ, Hamada K, Redfem CPF, Nutbrown M, more androgen receptors than those from non-Galling a&as. J En-
Ebling FJG, Messenger AG: Androgens and the hair follicle: cul- docrinol 133:141- 147, 1992
tured human dermal papilla cells as a model system. Ann NY Acad 35. Oliver RF, Jaboda CAB: The dermal papilla and maintenance of hair
Sci 642:355-375, 1991 growth. In: Rogers GE, Reis PR, Ward KA, Marshall RC (eds.). The
9. Randall VA, Thornton MJ, Hamada K, Messenger AG: Mechanism of Biology of Wool and Hair. Chapman & Hall, London, 1989, pp
androgen action in cultured dermal papilla cells derived from human 51-67
120s RANDALL ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

36. Reynolds AJ, Jahoda CAB: Inductive properties of the hair follicle of hair growth, sebaceous glands and sweat glands, J Endocrinol
cells. Ann NY Acad Sci 642:226 - 242, 1992 133:467-475, 1992
37. Cunha GR, Taguchi 0, Shannon JM, Chung LWK, Bigsby RM: 42. Messenger AG: The culture of dermal papilla cells from human hair
Mesenchymal role in hormone-induced epithelial development. In: follicles. Br J Dermatol 110:685-689, I984
Serio M, Motta M, Zanisi M, Martini L (eds.). Sexual Differentia- 43. Messenger AG, Senior J, Bleihen SS: The in vitro properties of dermal
tion: Basic and Clinical Aspects. Serono Symposia Publications, Vol papilla cell lines established from human hair follicles. Br J Derma-
II. Raven Press, New York, 1984 to1 114:425-430, 1986
38. Stumpf WE, Sar M: Autoradiographic localisation of estrogen, andro- 44. Jahoda CAB, Horne KA, Oliver RF: Induction of hair growth by
gen, progestin and glucocorticosteroid in ‘target tissues’ and ‘non- implantation ofcultured dermal papilla cells. Nature 311:560- 562,
target tissues.’ In: Pasqualini J (ed.). Modern Pharmacology-Toxi- 1984
cology. 8. Mechanism of Action of Steroid Hormones. Marcel
45. Thornton MJ, Laing I, Hamada K, Messenger AG, Randall VA: Der-
Dekker, New York, 1976, pp 41-84
ma1 papilla cells for beard hair follicles produce 5&ihydrotestos-
39. King RJB: Structure and function of steroid receptors. J Endocrinol terone in culture unlike scalp cells. Clin Endocrinol (in press)
114:341-349,1987
46. Randall VA, Thornton MJ. Messenger AG: Androgens stimulate the
40. McPhaul MJ, Wilson JD: Molecular defects in the androgen receptor secretion of mitogenic substance (c) by dermal papilla cells cultured
causing androgen resistance. J Invest Dermatol 98:975 - 998, 1992 from beard but not scalp follicles. Br J Dermatol 125:485, 1991
41. Choudhry R, Hodgins MB, Van der Kwast TH, Brinkmann AD, 47. Seago SV, Ebling FJG: The hair cycle on the human thigh and upper
Boersma WJA: Localisation of androgen receptors in human skin by arm. Br J Dermatol 113:9- 16.1985
immunohistochemistry: implications for the hormonal regulation