BIOLOGY OF REPRODUCTION 62, 806–810 (2000)
Testosterone-Dependent Primer Pheromone Production in the Sebaceous Gland
of Male Goat1
E. Iwata, Y. Wakabayashi, Y. Kakuma, T. Kikusui, Y. Takeuchi, and Y. Mori2
Laboratory of Veterinary Ethology, Department of Veterinary Medical Science, The University of Tokyo, Yayoi,
Bunkyo-ku, Tokyo 113–8657, Japan
ABSTRACT
thereby modulates the pulsatile secretion of LH [7, 8]. A
method has been developed for monitoring the electrophysiological manifestation of this GnRH pulse generator
activity from conscious and loosely restrained goats as the
characteristic increases in the hypothalamic multiple-unit
activity (MUA volleys) [9–12]. This technique provides us
with the practical means for the real-time assessment of
timed application of testing materials and their effects on
the reproductive neuroendocrine system. These characteristics are particularly important for establishing a satisfactorily sensitive as well as specific bioassay system for assessing the primer pheromone activity [13]. In fact, exposure to male goat hair, which had been timed midway between succeeding MUA volleys, resulted in an occurrence
of the MUA volley within a few minutes. On the other
hand, hair samples obtained from castrated goats had no
such stimulatory effect on the GnRH pulse generator, but
testosterone replacement resulted in a restoration of the
pheromone activity [13].
Although attempts have been made to purify the pheromone(s) responsible for the male effect [14, 15], chemical
identification has not yet been successful. To achieve this
objective, we need to know more about where and how the
male pheromone is synthesized and released. This study
was therefore aimed at revealing the time course of morphological change and pheromone-producing activity of the
sebaceous gland, a putative pheromone-synthesizing organ,
of the castrated male goat throughout 4 wk of testosterone
treatment and the following 4-wk waning-off period by utilizing a novel neurophysiological bioassay system for the
primer pheromone.
To test the hypothesis that the primer pheromone responsible
for inducing the ‘‘male effect’’ is produced in the sebaceous
gland androgen dependently, we examined the correlation between morphological changes of sebaceous glands and the pheromone activity in skin samples taken from castrated goats that
had been treated with testosterone. Five castrated goats were
implanted s.c. with testosterone capsules to maintain physiological levels of plasma testosterone for four weeks. Skin samples
were obtained from the head region on Day 0 (the day of testosterone implant), Day 7, Day 14, Day 28 (the day of testosterone removal), Day 36, Day 42, and Day 56. Matched blood
samples were also collected for measurement of testosterone
concentration. The pheromone activity of the ether-extracts of
the upper dermal layer containing sebaceous glands was assessed by its stimulatory effect on the hypothalamic GnRH pulse
generator, which was monitored for changes of specific multiple
unit activity (MUA) in ovariectomized estradiol-primed goats as
described previously. The sebaceous gland enlarged during the
testosterone treatment but reduced in size after testosterone removal. The pheromone activity first appeared in 2 out of 5 goats
on Day 7 and in all the 5 goats by Day 28. Fourteen days after
testosterone removal (Day 42), the pheromone activity was no
longer detectable in any of the 5 goats. In short, the sebaceous
gland size and the pheromone activity shifted almost in parallel.
The present results provide strong support for the view that the
primer pheromone is produced testosterone dependently in the
sebaceous gland of the male goat.
INTRODUCTION
The ‘‘male effect’’ is well known in sheep [1, 2] and
goats [3–5] as a good example of pheromone-induced activation of reproductive function in which seasonally anovulatory females resume ovarian cyclicity in response to
the male pheromone. It has been shown that the olfactory
signal emanating from the male hair stimulates the reproductive neuroendocrine system, and that the frequency of
pulsatile LH secretion increases, leading to the normal sequence of preovulatory endocrine events [5, 6].
The GnRH pulse generator regulates the intermittent
GnRH discharge into the pituitary portal circulation and
MATERIALS AND METHODS
Experimental Design
Five long-term (. 1 yr) castrated male goats (2–3 yr
old) were obtained from the closed colony at the experimental station of the University of Tokyo. They were isolated from other male goats and housed under conditions
of natural day length and temperature. Six testosterone capsules made of Silastic sheet (5 3 5 cm; Dow Corning Co.,
Ambridge, PA), each containing 1 g of crystalline testosterone (Wako Co., Osaka, Japan) were implanted s.c. for
28 days in each animal [16]. The sampling of skin and
blood was conducted on Day 0 (immediately before the
testosterone implantation), Day 7, Day 14, Day 28 (the day
of testosterone removal), Day 35, Day 42, and Day 56. For
sampling, the goat was anesthetized by ketamine HCl and
xylazine HCl, and a square (1 cm 3 1 cm) of skin was cut
off by scalpel from the head region between the horns,
where the sebaceous glands were shown to be well-developed [17]. The collected skin samples were stored at
2808C. For the testosterone assay, blood was collected on
the same day, prior to the skin sampling. A blood sample
(2 ml) was collected from the jugular vein with syringes
This work was supported by ‘‘Research for the Future’’ Program, The
Japan Society for the Promotion of Science (JSPS-RFTF 97L00904), Grantsin-aid for Scientific research from the Ministry of Education, Science,
Sports and Culture and CREST of the Japan Science and Technology Corporation.
2
Correspondence: Y. Mori, Laboratory of Veterinary Ethology, Department of Veterinary Medical Science, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan. FAX: 81 3 5841 8190; e-mail:
[email protected]
1
Received: 12 July 1999.
First decision: 9 August 1999.
Accepted: 27 October 1999.
Q 2000 by the Society for the Study of Reproduction, Inc.
ISSN: 0006-3363. http://www.biolreprod.org
806
SEBACEOUS GLAND PRODUCES PRIMER PHEROMONE
807
FIG. 1. Changes in MUA in response to the application of the extract
of the upper dermal layer in a representative ovariectomized goat carrying
estradiol implant, shown as intervals (min) between MUA volleys. When
the goat was exposed to the positive sample (vertical arrow) midway between volleys, the subsequent MUA volley was induced within a few
minutes of the exposure.
containing heparin and centrifuged immediately after collection. The plasma was removed and stored at 2208C in
a plastic tube until assayed for testosterone concentration.
Preparation of Skin Samples for Bioassay
Half of each skin sample was trimmed by removing the
epidermal, subcutaneous layer, and the lower part of the
dermal layer containing sweat glands, and only the upper
dermal layer containing sebaceous glands was used for the
bioassay to minimize interference by exogenous chemical
compounds from the environment [18]. The samples were
homogenized by ultrasonic homogenizer (Tomy Co., Tokyo, Japan), extracted by diethyl ether, filtered using filter
paper No. 1 (Advantec TOYO Inc., Tokyo, Japan), and dehydrated using Na2SO4. The extracts were then dried under
a stream of nitrogen gas at 408C and stored at 2208C in a
glass vial with a Teflon (DuPont, Wilmington, DE)-sealed
cap that was filled with nitrogen gas to prevent oxidation.
On the day of bioassay, the sample was dissolved in
diethyl ether again, in which a piece of gauze was soaked
and dried again a few hours before the bioassay. The gauze
was stored at 2208C in a sealed plastic bag filled with
nitrogen until used.
MUA Recording and Bioassay for Pheromone Activity
Arrays of bilateral recording electrodes were stereotaxically placed in the medial basal hypothalamus of 3 longterm (. 2 yr) ovariectomized female goats [19]. The goats,
from which clear MUA volleys had been consistently recorded [11, 20], were used for the bioassay of pheromone
activity according to the procedure previously described by
Hamada et al. [13] with a little modification.
In brief, the estradiol-primed goats were loosely tied to
the stanchion in Faraday cage during MUA recording under
controlled temperature (238C), relative humidity (55%), and
photoperiod (12L:12D). For the bioassay, the gauze containing the sample extract to be tested was placed 2–3 cm
in front of the goat’s nose for 3 min to allow olfactory,
visual, and tactile investigation. When an MUA volley followed within a 3-min period of odor application, the sample
was assessed to be pheromone positive. The bioassay was
carried out 2–4 times a day in each of 3 female goats for
3–4 consecutive days, and the exposures to samples were
separated by an interval sufficient in length (usually 2–3 h)
for the frequency of MUA volleys to return to the same
level as before exposure. Each female was tested with the
samples from more than one male and the sequence of sam-
FIG. 2. Proportion of pheromone positive samples obtained from the five
castrated goats treated with testosterone between Day 0 and Day 56 (top).
Plasma concentrations of testosterone between Day 0 and Day 56 (middle). The change of the major axis length of sebaceous glands between
Day 0 and Day 56 (bottom). Each point represents the mean 6 SE (n 5
5). *P , 0.05 and **P , 0.01, by paired t-test, as compared with Day 0.
ples exposed was random, but the last sample was intentionally the predetermined positive control (male goat hair
extract) for confirmation of normality of the bioassay.
Measurement of the Size of the Sebaceous Glands
The remaining halves of the skin samples were fixed in
10% paraformaldehyde and were cut on a freezing microtome at 15-mm thickness perpendicular to the skin surface.
They were then stained with Sudan black B backgrounded
by nuclear fast red. The sections were projected at a magnification of 3100, and the outlines of 50 sebaceous glands
were traced for each sample. The length of the major axis
was measured for each gland, and mean values and standard errors (SE) were calculated for each experimental day.
808
IWATA ET AL.
FIG. 3. Morphological changes of sebaceous glands in castrated goats treated
with testosterone between Day 0 and Day
56. a) Day 0; b) Day 7; c) Day 14; d) Day
28; e) Day 35; f) Day 42; g) Day 56. Bar
5 100 mm. Sebocytes in the sebaceous
glands were small in size and in number
on Day 0 (a, black arrow), developed on
Day 28 (d, arrow head), then atrophied on
Day 56 (g, white arrow).
Analysis of variance and paired t-test were used for detection of significant differences.
assay coefficients of variation were 9.8% and 11.2%, respectively.
Assay of Plasma Concentration of Testosterone
RESULTS
Plasma testosterone concentrations were measured by an
enzyme-immunoassay using anti-testosterone serum (provided by Dr. A. Miyamoto, Obihiro University of Agriculture and Veterinary Medicine, Obihiro City, Japan), and
HRP-testosterone (Medical System Service Co., Kanagawa,
Japan) as described elsewhere [21]. The intra- and inter-
The effect of odor from five castrated goats (#1–5) on the
intervals between MUA volleys in three female goats used
for the bioassay is shown in Table 1. In most cases samples
were applied 40 min (in some cases 35 min) after the onset
of the preceding MUA volley. Representative positive data of
the bioassay is shown in Figure 1, where the application in-
SEBACEOUS GLAND PRODUCES PRIMER PHEROMONE
TABLE 1. Intervals (min) between MUA volleys at the time of application
of upper dermal layer extracts obtained from 5 individual castrated goats
during and after 4 weeks of treatment with testosterone (Day 0–56). 2
Goat #
Day 0 Day 7
Day 14 Day 28 Day 35 Day 42 Day 56
#1
#2
#3
#4
#5
57(40)
67(40)
88(40)
80(40)
72(40)
40(40)*
40(40)*
88(40)
40(40)*
40(40)*
69(40)
71(40)
80(40)
40(40)*
40(40)*
42(40)*
40(40)*
36(35)*
40(40)*
40(40)*
41(40)* 63(35)
81(40)
76(40)
41(40)* 107(40)
82(40)
72(40)
58(35)
70(40)
72(40)
64(40)
61(40)
96(40)
60(35)
a
Numerals in parentheses are the times of the sample application from
the onset of preceding MUA volleys.
* Tests in which the sample was assessed to be pheromone positive.
duced advancement of the following MUA volley. This stimulatory effect, however, lasted only briefly, and the subsequent
MUA volley interval returned to the preapplication level. The
appearances of pheromone activity in the 5 castrated goats are
shown in Figure 2 (upper). The pheromone activity was positive in 2 out of 5 goats on Day 7, soon after testosterone
implantation. On Day 28, all of 5 goats turned positive; but
after testosterone removal the pheromone activity vanished,
and all individuals became negative again on Day 42. Four
wk after the commencement of testosterone treatment all the
castrated goats (Day 28) presented a male-like appearance,
but they had returned to their pretreatment figures 4 wk after
testosterone withdrawal (Day 56). Plasma concentrations of
testosterone were maintained at physiologically high levels
(0.8–15.8 ng/ml for the intact males in the same colony) during the period of testosterone implantation as shown in Figure
2 (middle).
The sebaceous glands developed gradually during the
course of the testosterone treatment (Days 0–28) and then
atrophied during the follow-up period (Days 35–56) as
shown in Figure 2 (bottom) and Figure 3. On Day 28 the
size of the sebaceous glands eventually attained the normal
size of mature intact males in the same colony (793.3 6
32.26 mm), and this was due to increases in both the number and the volume of sebocytes. Consequently, the length
of major axis of sebaceous gland was larger on Day 7 (P
5 0.0297), Day 14 (P 5 0.0013), Day 28 (P 5 0.0001),
Day 35 (P 5 0.0004), and Day 42 (P 5 0.0019) as compared with Day 0.
DISCUSSION
In this study, primer pheromone activity appeared in upper dermal layer extracts obtained from castrated goats following testosterone treatment. This indicates that the primer
pheromone responsible for the male effect is produced in
the sebaceous gland under the control of testosterone. In
previous attempts to purify this primer pheromone, the hair
of male goats was used as a source of pheromone [13–15].
Now it has been revealed that sebaceous glands can be
another and probably more suitable resource for the isolation of the pheromone.
Kealy et al. [22] isolated human sebaceous glands by
shearing, but in this study we used the upper dermal skin
layer for extraction because the glands at Day 0 were too
small to shear off; nevertheless, we have confirmed that we
could obtain results with the extracts from the sheared sebaceous gland that were similar to results from the skin of
a mature male goat in our pilot study (data not shown).
In this study, primer pheromone activity first appeared
on Day 7 and disappeared on Day 35 in some individuals.
This change appears to be somewhat faster than the wax
809
and wane of glandular development. Downing et al. [23]
suggested that an average interval between synthesis and
excretion of sebum was about 8 days. These results may
suggest that pheromone production would have begun at an
early stage of the sebum production process and that the
size of the sebaceous gland enlarged gradually according
to the flourish of sebum production. However, it is still
unclear how testosterone acts on the sebaceous gland to
induce pheromone production, and so further investigation,
probably including the establishment of a primary culture
of sebaceous gland cells, is needed to elucidate the mechanism of pheromone production as well as to identify pheromone molecules.
ACKNOWLEDGMENTS
We would like to thank Dr. Sawazaki for providing animals from the
closed colony of the University of Tokyo. This study was supported by
‘‘Research for the Future’’ Program, The Japan Society for the Promotion
of Science (JSPS-RFTF 97L00904), Grants-in-aid, for Scientific research
from the Ministry of Education, Science, Sports and Culture and CREST
of the Japan Science and Technology Corporation.
REFERENCES
1. Underwood F, Sheier F, Davenpoert N. Studies in sheep husbandry in
W. A. V. The breeding season in Merino, crossbreed and British Breed
ewes in the agricultural districts. J Agric West Aust 1944; 2:135–143.
2. Martin G, Oldman C, Lindlay D. Increased plasma LH levels in seasonary anovular Merino ewes following the introduction of rams.
Anim Reprod Sci 1980; 3:125–132.
3. Shelton M. The influence of the presence of the male on initiation of
oestrous cycle activity in Angola does. J Anim Sci 1960; 19:368–375.
4. Ott R, Nelson D, Hixon J. Effect of presence of the male on initiation
of oestrus cycle activity of goats. Theriogenology 1980; 13:183–190.
5. Chemineau P. Possibilities for using bucks to stimulate ovarian and
oestrus cycles in anovulatory goats—a review. Livest Prod Sci 1987;
17:135–147.
6. Martin G, Oldman C, Cognie Y, Pearce D. The physiological responses of anovulatory ewes to the introduction of rams—a review. Livest
Prod Sci 1986; 15:219–247.
7. Knobil E. The neuroendocrine control of the menstrual cycle. Recent
Prog Horm Res 1980; 36:53–88.
8. Lincoln D, Freser H, Lincoln G, Martin G, MacNeilly A. Hypothalamic pulse generators. Recent Prog Horm Res 1985; 41:369–419.
9. Mori Y, Nishihara M, Tanaka T, Takeuchi Y, Hoshino K. Chronic
recording of electrophysiological manifestation of the hypothalamic
gonadotropin-releasing hormone pulse generator activity in the goat.
Neuroendocrinology 1991; 53:392–395.
10. Nishihara M, Mori Y, Yoo M-J, Takahashi M. In vivo electrophysiological monitoring of the GnRH pulse generator in rats and goats. In:
Levine JE (ed.), Pulsatility in Neuroscience. Volume 20. San Diego:
Academic Press; 1994: 114–126.
11. Mori Y, Tanaka T. Electrophysiological approach to the hypothalamic
GnRH pulse generator. J Reprod Fertil Suppl 1995; 49:231–243.
12. Nishihara M, Takeuchi Y, Tanaka T, Mori Y. Electrophysiological correlates of pulsatile and surge gonadotrophin secretion. Rev Reprod
1999; 4:110–116.
13. Hamada T, Nakajima M, Takeuchi Y, Mori Y. Pheromone-induced
stimulation of hypothalamic gonadotropin-releasing hormone pulse
generator in ovariectomized, estrogen-primed goats. Neuroendocrinology 1996; 64:313–319.
14. Claus R, Over R, Denhard M. Effect of male odour on LH secretion
and the induction of ovulation in seasonary anoestrus goats. Anim
Reprod Sci 1990; 22:27–38.
15. Cohen-Tannoudji J, Einhorn J, Signoret JP. Ram sexual pheromone: first
approach of chemical identification. Physiol Behav 1994; 56:955–961.
16. Kanai Y, Yamasaki J, Takeuchi Y, Koike N, Fujiyama M, Mori Y.
Electrophysiological recording of hypothalamic GnRH pulse generator
activity in castrated male goats. J Reprod Dev 1995; 41:57–62.
17. Jenkinson DME, Blackburn PS, Proudfoot R. Seasonal changes in the
skin glands of the goat. Br Vet J 1967; 123:541–548.
18. Hay JB, Hodgins MB. Distribution of androgen metabolizing enzymes
in isolated tissues of human forehead and axillary skin. J Endocrinol
1978; 79:29–39.
810
IWATA ET AL.
19. Mori Y, Takeuchi Y, Shimada M, Hayashi S, Hoshino K. Stereotaxic
approach to hypothalamic nuclei of the Shiba goat with radiographic
monitoring. Jpn J Vet Sci 1990; 52:339–349.
20. Nishihara M, Hiruma H, Kimura F. Interactions between the noradrenergic and opioid peptidergic systems in controlling the electrical
activity of luteinizing hormone-releasing hormone pulse generator in
ovariectomized rats. Neuroendocrinology 1991; 54:321–326.
21. Miyamoto A, Okuda K, Schweigert FJ, Schams D. Effects of basic
fibroblast growth factor, transforming growth factor-b and nerve
growth factor on the secretory function of the bovine corpus luteum
in vitro. J Endocrinol 1992; 135:103–114.
22. Kealey T, Lee CM, Thody AJ, Coaker T. The isolation of human
sebaceous glands and apocrine sweat glands by shearing. Br J Dermatol 1986; 114:181–188.
23. Downing DT, Strauss JS. On the mechanism of sebaceous secretion.
Arch Dermatol Res 1982; 272:343–349.