THE JOURNAL OF EXPERIMENTAL ZOOLOGY 267:47-56 (1993)

Fiber Growth Initiation in Hair Follicles of Goats Treated With Melatonin

ALLAN J. NIXON, VERNON J. CHOY, ALTHEA L. PARRY, AND ALLAN J. PEARSON New Zealand Pastoral Agriculture Research Institute, R uakura Agricultural Centre, Hamilton (A.J.N., V.J.C., A.J.P.); and Flock House Agricultural Centre, Bulls (A.L.P.), New Zealand

ABSTRACT The sequence of structural changes in goat hair follicles was investigated using mel- atonin implants to advance and synchronize spring hair growth. Ten pasture fed cashmere wethers each received a controlled release formulation of 70 mg of melatonin on September 1 1989, and showed plasma melatonin elevated above physiological levels over 14 days post-treatment (914 2 154 pg/ml [mean 2 SEMI on day 14). In ten untreated animals, daytime plasma melatonin was 19.9 4.7 pg/ml. Histological examination of skin biopsies taken over the 14 days from the start of the experiment showed that primary hair follicles of goats with manipulated hormone levels had initiated fiber growth (entered proanagen), whereas primary follicles of untreated goats largely remained in the quiescent phase (telogen). A standardized terminology was used to describe the sequence of events during in- duced proanagen. Structural reorganization of follicles began in treated animals between days 6 and 12 post-treatment, and emergent fibers grew by day 24. Advancement of spring fiber growth was associated with a suppression of the normal rise in plasma prolactin concentration. Prolactin levels in untreated goats increased from 7.4 ? 1.8 ng/ml on day 1 to 12.8 f 1.6 nglml on day 14, but declined in treated goats from 6.3 f 2.3 ng/ml to 2.2 k 0.8 nglml over the same period. o 1993 Wiley-Liss, Inc.

Mammalian fiber growth cycles consist of two steady state and two transitional phases (Davis, '62). During the active steady state (anagen), epithelial cells of the germinal matrix of the follicle bulb proliferate rapidly in response to signals from specialized fibrocytes of the adjacent dermal papilla (Oliver, '80; Horne et al., '86). Cells arising from the bulb differentiate in concentric layers, the innermost forming keratinized fiber. After a period of such activity, the follicle enters a comparatively short regressive phase (catagen) (Straile et al., '61) leading to a quiescent state (telogen). During telogen, the germinal matrix is reduced to a sec- ondary hair germ situated adjacent to the fibrocytes of the former dermal papilla at the distal end of the shortened follicle. Regeneration of the second- ary hair germ giving rise to a new fiber was ear- lier regarded as part of anagen (Dry, '26; Chase et al., '51). Chase ('65) applied the term proanagen to the initiation phase, distinguishing it from contin- uing fiber growth from a fully developed follicle (which he termed metanagen). This brief transi- tional phase between quiescence and growth is the focus of the present investigation.

Proanagen appears to be a key period in relation to control of the hair cycle. Cell division is initi- ated in response to as yet undefined proximal stim-

Q 1993 WILEY-LISS. INC.

uli (Moore, '89) and characterization of structural changes may provide a framework for investigation of these controlling factors. Growth during pro- anagen also has many parallels with embryonic follicle development (Lyne, '66). The hormonal mech- anisms controlling seasonal hair growth cycles are poorly understood, but have been shown to involve the pineal-hypothalamo-pituitary neurosecretory axis (Rougeot et al., '84). In many mammals, fiber growth initiation is cued by photoperiod, via the pineal hormone melatonin (Allain et al., '81) and probably also pituitary prolactin (Martinet et al., '84; Badura and Goldman, '92). Melatonin has been shown to regulate seasonal variation in prolactin levels in goats (Maeda et al., '861, and the seasonal increase in prolactin is associated with the initia- tion of pelage cycles in shedding sheep (Lincoln, '90). A more precise description of the behavior of hair follicles at critical times in the seasonal cycle is a prerequisite for understanding the linkage between systemic and local control of fiber growth.

In this study we report induction of early fiber growth by exogenous melatonin at a time of near

Received September 28,1992; revision accepted March 8,1993.

48 A.J. NIXON ET AL.

total quiescence so as to achieve a synchronized growth response. We describe the timing of micro- anatomical changes of hair follicles over the short period after melatonin implant but prior to visible effects on pelage, and relate these to the contem- poraneous effects on circulating melatonin and prolactin.

MATERIALS AND METHODS Animals and treatment

Twenty mixed age cashmere-bearing wethers maintained on hill country pasture were used in the experiment which commenced during the spring period of hair follicle inactivity. On September 1, 1989, ten randomly assigned goats each received subcutaneous injections of 70 mg melatonin in a controlled release formulation. This consisted of a polylactide-polygalactide copolymer formed into 25-300 pm diameter microspheres, and suspended immediately prior to use in 4.5 ml normal saline and 2% carboxymethylcellulose. The remaining ten goats served as untreated controls.

Blood collection and hormone assay Ten milliliter blood samples were taken by jugu-

lar venipuncture of all animals between 10 am and noon on days 1, 7, and 14 after melatonin treat- ment. Plasma melatonin and prolactin levels were measured in duplicate by radioimmunoassay (RIA).

Melatonin RIA reagents were Eo-methyL3H]- melatonin, Amersham (TRK 798); sheep anti- melatonin, Guildhay Antisera (Batch S704-8483); standard melatonin, Behring (Q-1300). The assay protocol was that supplied with the antiserum. Sep- aration of antibody-bound label from free label was by dextran-charcoal. Sensitivity was 8 pg/ml. Inter- assay and intra-assay variations at 280 pg/ml were 15.9% and 7.4%, respectively.

Caprine prolactin was assayed using ovine prolactin (NIDDK-oPRL-1-21 for standards and radioiodination, and ovine prolactin antiserum (NIDDK-anti-oPRL-2). Prolactin was iodinated by the Iodogen technique (Pierce, Rockford, IL), using [12511-iodide (New England Nuclear NEZ0033A). The assay method was essentially as prescribed for the NIDDK reagents. Separation of antibody-bound label from free label was by second antibody precipitation using excess sheep antirabbit serum (generated at the Ruakura Agricultural Centre). Sensitivity was 0.6 ng/ml. Inter-assay and intra-assay variations at 32 ng/ml were 14.4% and 12.1%, respectively.

Skin sampling and histology Samples of skin were taken by snip biopsy from

the mid-side of all 20 animals at the time of mela-

tonin treatment and again 14 days later. Four of the treatment group were skin biopsied from within a 8 cm2 mid-side site on days 0, 4, 6, 8, 10, 12, 14, 20, and 24 after melatonin treatment. Skin sam- ples were fixed in phosphate buffered 4% formal- dehyde, processed through graded alcohols, and embedded in paraffin wax.

Skin samples from day 0 and day 14 were sepa- rated into two pieces and embedded for both trans- verse and longitudinal sectioning. Transverse sections of 7 pm thickness were stained by the Sacpic method (modified from Auber, '52) in order to screen both primary and secondary follicle activity. Active follicles were detected in transverse sections by the presence of red staining inner root sheath (Nixon et al., '91).

Serial longitudinal sections of hair follicles in all samples were cut at 8 pm thickness, and stained by the Sacpic method in order to visualize aspects of follicle structure and keratinocyte differentiation. Additional longitudinal sections representing key stages of development were stained with periodic acid-Schiff (PAS) and Gomori's trichrome methods to aid interpretation, and methyl violet (Clarke and Maddocks, '63) to show mitoses. Primary (guard hair) follicles were classed according to phase of the hair cycle. For this purpose a nine point scale de- noting follicle development was constructed, based on the description and nomenclature of Chase et al. ('51) and Chase ('65). The length of those folli- cles cut in near median longitudinal section was measured from base of the bulb to the epidermal surface, using a Visilog 3.6 image analysis system (Noesis, Velizy).

Statistical analysis Hormone data were subjected to analysis of vari-

ance using the GENSTAT statistical package. Dif- ferences in prolactin levels between groups and over time were analysed by examining trends of indi- vidual animals. The incidence of follicle activity states was compared using a chi-square test, and lengths of quiescent and active follicles were com- pared using a t-test.

RESULTS Melatonin treatment

In the control group, daytime plasma melatonin concentration (mean & SEM, 19.9 t 4.7 pg/ml) did not differ between sample dates (P = .06). In nine of the ten animals receiving melatonin, levels ex- ceeded 2,000 pg/ml at day 1, then declined to 1,518 * 110 pg/ml by day 7, and further to 914 & 154 pg/ml by day 14 post-treatment. In the remaining treated

FIBER GROWTH INITIATION 49

Stages of this developmental sequence are illustrated by photographs in Figures 2 and 3, and diagram- matically in Figure 4 using the classification of Chase et al. (’51) with modification. In order to avoid confusion, we have termed stages of proanagen “proanagen I” to “proanagen V,” rather than “anagen I” to “anagen v” as originally proposed.

Telogen

Days after melatonin treatment

Fig. 1. Plasma prolactin in untreated control (stippled) and melatonin treated goats (solid) between day 1 (September 2) and day 14 (September 15) after implant. Error bars rep- resent SEM.

animal, elevated circulating melatonin was observed on days 7 and 14, but not on day 1.

Plasma prolactin Trends in plasma prolactin concentrations in con-

trol and melatonin treated goats are compared in Figure 1. Significant differences in prolactin were detected between groups, and within groups over time. The control group showed an increase (from 7.4 -+ 1.8 ng/ml to 12.8 ? 1.6 ng/ml) in circulating prolactin (I‘ < . O l ) over 14 days in September (early spring). In treated animals, this rise was suppressed and prolactin levels declined (from 6.3 & 2.3 ng/ml to 2.2 5 0.8 ng/ml) between day 1 and day 14 post- treatment (P < . O l ) .

Follicle development and fiber growth Examination of hair follicles in transverse sec-

tion was a convenient means of assessing follicle activity at the start of the trial, although accurate detection of the very early stages of proanagen was difficult in transverse sections. Six of the 20 goats had greater than 5% active secondary follicles at the start of the experiment. During the 14 day pe- riod from melatonin implant, secondary follicle ac- tivity commenced in some individuals and ceased in others. Secondary follicles were therefore unsuit- able for proanagen studies in these cases, as sub- sequent growth would be out of phase. By contrast, active primary follicles were rare, and no day 0 sam- ples observed in transverse section contained more than 5% active primary follicles.

Fiber growth was initiated within the skin of mel- atonin treated goats, and a sequence of morpholog- ical change in hair follicles from the quiescent state (telogen) to continuous fiber growth (anagen) was apparent over the 24 day post-treatment period.

The secondary hair germ and fibrocytes of the presumptive dermal papilla formed a compact mass of cells attached by an isthmus of epithelial cells below the club end (Fig. 2A). The dermal and epider- mal cell populations could be clearly distinguished. Dermal cell aggregations were approximately hem- ispherical in shape. A trail of fibrocytes was usu- ally present below the dermal papilla proper. The dermal papilla was PAS positive and stained as for connective tissue. Hair germ cells adjacent to the dermal papilla were generally columnar in form with larger cytoplasmic volume than the overlying epithelial strand. None of these epithelial cells were readily counterstained by either of the trichrome methods, but were often vacuolated and contained melanin granules (Fig. 2A). The hair germ and der- mal papilla were usually oriented toward the ectal aspect of the follicle.

Proanagen I The first stage of a new hair cycle was marked

when cell division had commenced in the hair germ (Fig. 2B). The hair germ cell layer had a less regu- lar appearance and was sometimes slightly curved around the dermal papilla. Gross structure of the follicle was otherwise similar to that of telogen.

Proanagen I1 The hair germ began to grow around the dermal

papilla (Fig. 2 0 . Hair germ cells were often tightly packed. In later proanagen 11, the epithelial strand thickened and the bulb began to form.

Proanagen IIIa The beginning of proanagen I11 was marked by

the formation of a hair cone. Undifferentiated in- ner root sheath cells did not take red (safranin) counterstain, but a hair cone of presumptive kera- tinocytes was apparent with hematoxylin staining (Fig. 2D). The bulb increased in size, and the der- mal papilla cavity became rounded in shape.

Proanagen IIIb The hair cone elongated and began to stain with

safranin (Fig. 3A), but the fiber was not yet kera-

Figure 2.

FIBER GROWTH INITIATION 51

tinized. Follicle length was visibly greater than in tel- ogen. Outer root sheath cells began to acquire size and shape characteristics of those in anagen follicles.

Proanagen IV The fiber shaft became keratinized (picrophilic),

and the inner root sheath intensely stained with safranin (Fig. 3B). A cuticle scale structure was vis- ible. Outer root sheath around the bottom of the club end formed a bulge, as the tip of the fiber grew beside the club end on the ectal side of the follicle.

Proanagen V The fiber tip reached the duct of the sebaceous

gland. The papilla was more elongated and had sometimes become acentric. This stage is equiva- lent to “anagen V” of Chase et al. (’51) and “mes- anagen” of Chase (’65).

Anagen Growth of the new emergent fiber continued

alongside the old brush end (Fig. 3C). The old f i - ber subsequently shed, and further increase in fol- licle bulb diameter occurred up until growth of the medullated fiber shaft (Fig. 3D). The is stage “anagen VI” of Chase et al. (’51) and “metanagen” of Chase (’65).

The increase in primary follicle length through- out this process is shown in Figure 5. Follicles in telogen measured 889 & 39 pm, from which there was a significant length increase by proanagen I1 stage (P < .05). By anagen stage at 20 and 24 days after melatonin stimulus, total follicle length had almost doubled to 1684 +- 50 pm. Follicle size in- creased further after this time until the growth of strongly medullated fiber was established.

At the beginning of the experiment (day 0) 95.2% of longitudinally sectioned primary follicles were in telogen. Incidence did not differ between treated and control groups (P > .lo). By 14 days after mel- atonin implant, eight out of ten treated animals possessed proanagen follicles, while the follicles of untreated control animals largely remained in telogen (Fig. 6).

Fig. 2. Longitudinal sections of goat primary hair follicles during early proanagen. A Telogen, at time of melatonin im- plant. The hair germ lies adjacent to an aggregation of dermal cells. B: Proanagen I, 6 days after melatonin implant. Chro- matin of cells in mitosis is stained violet (arrows). C: Proanagen II ,8 days after melatonin implant. D Proanagen IIIa, 14 days after melatonin implant. The hair cone has begun to form. b, brush end of old fiber; c, connective tissue sheath; d, dermal cells; e, epithelial strand; g, germinal epithelium (“hair germ”). A, C , and D are stained by the Sacpic method. B is stained with methyl violet and eosidphloxine. Bars, 50 pm.

Two treated animals failed to respond by day 14. One of these (goat number 76 in Fig. 6) showed the previously noted unusual melatonin release pattern, giving this individual a shorter exposure to mela- tonin than the other nine treated animals. The other treated goat with no proanagen follicles at day 14 (goat number 56 in Fig. 6) showed prolactin con- centration values of 24.9 pg/ml at day 1 and 7.6 pg/ml at day 14, compared with group means (k SEMI of 6.3 2 2.3 pg/ml and 2.2 r 0.8 pg/ml, re- spectively. Spontaneous initiation of fiber growth was evident in one of the untreated animals (goat number 32 in Fig. 6).

The chronological sequence of morphological change during proanagen in a subset of the treated goats is illustrated in Figure 7. The first change to follicle structure (proanagen 11) was evident between days 6 and 12. This was preceded by cell prolifera- tion in the hair germ (Fig. 2B), but follicles in telogen and proanagen I stages could not be dis- tinguished with confidence. Differentiation of in- ner root sheath cells was first apparent in three of the four animals at day 14. Growth of emergent f i - bers occurred in all four animals by day 24. Some between animal variation in response time was ap- parent, with a difference of approximately 7 days between goat 7 and goat 10 (Fig. 7C,D). A small number of anagen follicles were observed early in the experiment. These were assumed to be related to the previous cycle, and not stimulated by the mel- atonin treatment.

At 7 weeks after treatment, simultaneous shed- ding of all fiber types left most treated animals largely denuded. Therefore both primary and sec- ondary follicle activity cycles remained in phase for at least the next molting cycle. Control animals un- derwent normal shedding in which a covering coat of guard hair was maintained throughout.

DISCUSSION Molting occurs annually in goats so that there

is a short, growing summer coat, and a longer, denser winter coat (Ryder, ’66). This cycle of growth be- tween early summer and late autumn t~ fully grown winter pelage is under photoperiodic control (Mc- Donald et al., ’87). Amongst New Zealand cashmere- bearing goats, growth in primary (guard hair) follicles commences over a period of about 12 weeks following the spring equinox (Nixon et al., ’91).

The administration of a controlled release formu- lation of melatonin advanced and synchronized the onset of this early summer growth of new fibers in

Figure 3.

FIBER GROWTH INITIATION 53

telogen proanagen I 1 proanagen 11 I

I proanagen Ilia proanagen lllb I pioanagen IV I I

proanagen V

Fig. 4. Changes in hair follicle morphology during fiber growth initiation. Skin epithelial cells are represented by light stippled areas; brush ends are dark stippled; keratinized fiber is shown as black; medullated fiber is shown as cross-hatched; connective tissue is unshaded. This diagram summarizes follicle development observed over 24 days after melatonin implant, from sections such as those represented in Figures 2 and 3.

primary follicles. This result conforms to docu- mented effects of manipulation of photoperiod and melatonin on timing of pelage growth in goats (Mc- Donald et al., '87; Litherland et al., '90; Welch et al., 'go), in which the response has been found to depend critically on time of year or precondition- ing day-length. Previous reports have, however, been based on observation of fiber growth. In the pres- ent study, histological examination of follicle struc-

~~

Fig. 3. Longitudinal sections of goat primary hair follicles during late proanagen. A: Proanagen IIIb 14 days after mela- tonin implant. Inner root sheath cells have begun to differen- tiate. B: Proanagen IV, 20 days after melatonin implant. The cortical cells of the fiber have become picrophilic. C: Early anagen, 24 days after melatonin implant. D Mid anagen. A strongly medullated guard hair is growing. b, brush end of old fiber; c, connective tissue sheath; d, dermal papilla; e, outer root sheath (derived from epithelial strand); g, germinal ma- trix; f, fiber cortex; i, inner root sheath; m, medulla. All sec- tions are stained by the Sacpic method. Bars, 150 pm.

ture prior to the visible growth of new fibers provides a more precise description of growth initiation.

Goat follicles underwent rapid metamorphosis be- tween 6 and 24 days after the melatonin stimulus.

telogen I I llla lllb IV V anagen

Stage of proanangen

Fig. 5. Changes in primary follicle length during fiber growth initiation. Roman numerals represent proanagen stages of the hair cycle as illustrated in Figure 4. Error bars represent SEM.

A.J. NIXON ET AL.

Anagen - : Proanagen v . j Proanagen IV . j Proanagen lllb- j Proanagen ills.

Proanagen (1 .

54

Anagen

0 0

_______________._._________.___-____-- -~-- . - - - - - - - - - - - -~- - - - - - - - 0

1 * A + A o j ; * * * O A X + A X + + 4k A

! Proanagen v j Proanagen IV j Proanagen lllb

i Proanagen Ilia ; Proanagen II

Proanagen I Telogen 1

* _ _ _ _ _ _ - - _ _

A Untreated Animals

0

_______________.________________________------

A

* # * * * * + * + * r n i i i r 1 8 1 1

2 5 8 1 1 12 16 32 66 74 80 Goat number

o f + Proanagen I 1 Telogen 1 '

, 1 1 1 1 1 1 1 1 1

1 3 4 7 10 14 36 54 56 7 6

Goal number

Fig. 6. Primary hair follicle development at 14 days after melatonin implant. Stages of the hair cycle on the vertical axis are as shown in Figure 4. The transitional proanagen stages are enclosed in the box. The "wagon wheel" symbols show num- bers of follicles observed in each category. Each radial bar rep- resents one hair follicle. Single. follicles are represented by a small circle. A: Untreated control animals. B: Melatonin treated animals. Animal numbers 32,56, and 76 are discwsed in the text.

This transition through proanagen is slower than that reported for induced and spontaneous cycles in mice (Chase et al., '51) or ferrets (Nixon et al., '92). In mice, the proanagen I1 stage was reached by 2-3 days after plucking and proanagen V by 8 days, a passage of 5 to 6 days compared with 10 to 12 days in goats. Between species differences could be attributable to differences in follicle size and cell turnover time (Fraser, '65) and to photoperiodic pre- conditioning (Pearson and Nixon unpublished data). The response time of goat hair follicles is longer, and between animal variation and between folli- cle variation is greater in goats than in mustelids (Nixon and Pearson, unpublished data). Whilst the response of cells in the skin to a change in hormone levels is dependent on photoperiod, this response can be much shorter than indicated by externally visible pelage changes. Follicle metamorphosis oc- curring within 8 days in goats is preceded by a pe- riod of inter- and intracellular communication and cell cycle progression, probably lasting at least 24 to 48 hours (Saywell and Nixon, '92).

There have been two principal methods used in earlier studies involving synchronized follicle growth. In the first, the skin of young mice or rats is sampled at known stages of waves of hair growth (e.g., Ebling and Johnson, '64). However, the small size of animals imposes limits on skin and blood sampling, and the precise time from the initial growth stimulus is uncertain. In the second method, a new cycle is induced by plucking out fibers (e.g., Straile et al., '61; Slominski et al., '91). Plucking provides a synchronized response, but has been shown to result in abnormal hyperplasia and hy- pertrophy associated with wound repair (Silver et al., '69). Synchronization of primary follicles using melatonin at the time of near total quiescence and photosensitivity appears to be a useful alternative method of obtaining follicles or isolated cell popu- lations at known stages of the hair cycle.

The observed shift in fiber growth initiation is consistent with the widely recognised role of pineal melatonin secretion in mediating the effects of pho- toperiod on pelage cycles (Allain et al., '81; Lin- coln and Ebling, '85). Melatonin regulates seasonal changes in pituitary prolactin secretion (Maeda et al., '861, which in turn has been implicated in the timing of fiber growth in a range of mammals, in- cluding cashmere goats. In this species, prolactin has a diurnal rhythm, with seasonal shifts in base- line and amplitude (Buttle, '74; Prandi et al., '88). Untreated goats of the present study showed low, rising daytime prolactin levels in early spring consistent with these published data. They subse- quently underwent a normal molt at a time corres- ponding to further increase in blood prolactin (Prandi et al., '88). Lynch and Russel ('90) reported that administration of ovine prolactin from mid- winter advanced the spring molt, while suppres- sion of prolactin with bromocriptine delayed molt. This might suggest that a certain level of prolactin is required for hair follicle reactivation. However, in the present experiment, prolactin concentration was only beginning to increase from the winter low, and was reduced in the earlier molting treated animals.

An hypothesis involving direct effects of relatively small changes in the level of prolactin on goat fiber growth must also account for differences between follicles in hair cycle phase and responsiveness which have been observed in this experiment and elsewhere. Low but rising prolactin levels have been associated with primary follicle quiescence. At the same time, secondary follicles either remain inac- tive or undergo a shortened activity cycle leading to vellus hair production (Nixon et al., '91). In con-

55 FIBER GROWTH INITIATION

Anagen - Proanagen V .

Proanagen IV . Proanagen lllb . Proanagen llla .

Proanagen II .

B Goat 4 A Goat 1

0 Proanagen V . Proanagen IV . + Proanagen lllb .

t Proanagen llla .

+ 0

A # * Proanagen II .

I o *

t * 0

0 1 A * A

0 Anagen 7 + Anagen 0 Proanagen V . Proanagen V .

0 Proanagen IV . Proanagen IV . t Proanagen lllb . Proanagen lllb. Proanagen llla . Proanagen Ilia .

Proanagen I I . t A + o Proanagen I I .

* * * * # * + I

0

0

Proanagen 1 Telogen 1. Proanagen I *

Telogen 1.

Proanagen I Proanagen I Telogen 1 * * * * * * Telogen 1 * * * * *

7 0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24

0 - * * A

+ * * o * * I

* * * * * *

Days after melatonin treatment Days after melatonin treatment

trast, low but falling levels of prolactin induced by melatonin treatment are accompanied by early and synchronous activation of both follicle populations leading to complete shedding and fiber growth.

In conclusion, initiation of fiber growth in cash- mere goats with melatonin during midspring has provided a means of describing follicle development during proanagen. Key events in this regenerative transition are the start of cell proliferation (pro- anagen I), bulb development (proanagen 111, inner root sheath differentiation (proanagen 111), fiber cor- tical cell keratinization (proanagen IV), and ongo- ing fiber growth (proanagen V and anagen). As an experimental model, proanagen induced with mel- atonin at critical stages of the seasonal growth cy- cle offers 1) a known time frame within which to compare effects on developmental stages, 2) synchro- nization of follicle activity, 3) high contrast in ac- tivity state occurring over a short time, and 4) apparently similar growth processes to those occur- ring in spontaneous hair cycles.

ACKNOWLEDGMENTS We are grateful to Dr. M.P. Gurnsey and Mr. K.

Betteridge for supplying and administering the mel- atonin microcapsules. Skin samples were collected by Mr. G.J. Hamilton and Mr. A. McConachy from animals in a longer term trial conducted by Ms. A.J.

Litherland. Ms. J.E. Wildermoth provided techni- cal assistance with hormone assays. We thank the National Institute of Diabetes and Digestive and Kidney Diseases (Rockville, MD) and the National Hormone and Pituitary Programme (University of Maryland School of Medicine, Baltimore, MD) for prolactin RIA reagents.

LITERATURE CITED Allain, D., L. Martinet, and J. Rougeot (1981) Effects of mela-

tonin implants on changes in the coat, plasma level, and tes- tis cycle in the mink (Mustela uison). In: Photoperiodism and Reproduction in Vertebrates. Colloque de I’INRA 6. R. Ortravant, J. Pelletier, and J.P. Ravault, eds. INRA, Paris, pp. 263-271.

Auber, L. (1952) The anatomy of follicles producing wool-fibres, with special reference to keratinization. Trans. R. SOC. Edin., 62:191-254.

Badura, L.L., and B.D. Goldman (1992) Prolactin-dependent sea- sonal changes in pelage: role of the pineal gland and dopa- mine. J. Exp. Zool., 261:27-33.

Buttle, H. (1974) Seasonal variation of prolactin in plasma of male goats. J. Reprod. Fertil., 37:95-99.

Chase, H.B. (1965) Cycles and waves of hair growth. In: Biol- ogy of the Skin and Hair Growth. A.G. Lyne and B.F. Short, eds. Angus & Robertson, Sydney, pp. 461-466.

Chase, H.B., H. Rauch, and V.W. Smith (1951) Critical stages of hair development and pigmentation in the mouse. Physiol. Zool., 24:l-9.

Clarke, W.H., and I.G. Maddocks (1963) Crystal violet for se- lective staining of mitosis in follicle bulbs of sheep skin. Stain Technol., 38:252-254.

56 A.J. NIXON ET AL.

Davis, B.K. (1962) Phases of the hair-growth cycle. Nature,

Dry, F.W. (1926) The coat of the mouse (Mus musculus). J. Genet.,

Ebling, F.J., and E. Johnson (1964) The control ofhair growth. Symp. Zool. SOC. Lond., 12:97-130.

Fraser, I.E.B. (1965) Cellular proliferation in the wool follicle bulb. In: Biology of the skin and hair growth. A.G. Lyne and B.F. Short, eds. Angus and Robertson, Sydney, pp. 427-446.

Horne, K.A., C.A.B. Jahoda, and R.F. Oliver (1986) Whisker growth induced by implantation of cultured vibrissa dermal papilla cells in the adult rat. J. Embryol. Exp. Morphol.,

Lincoln, G.A. (1990) Correlation with changes in horns and pelage, but not reproduction, of seasonal cycles in prolactin in rams of wild, feral and domesticated breeds of sheep. J. Reprod. Fertil., 90:285-296.

Lincoln, G.A., and F.J.P. Ebling (1985) Effect of constant-release implants of melatonin on seasonal cycles in reproduction, pro- lactin secretion and moulting in rams. J. Reprod. Fertil.,

Litherland, A.L., D.J. Paterson, A.L. Parry, H.B. Dick, and L.D. Staples (1990) Melatonin for cashmere production. Proc. N.Z. SOC. Anim. Prod., 50:339-344.

Lynch, P., and A.J.F. Russel (1990) The hormonal manipula- tion of cashmere growth and shedding. Anim. Prod., 50561.

Lyne, A.G. (1966) The development of hair follicles. Aust. J. Sci., 28:370-377.

Maeda, K.-I., Y. Mori, and Y. Kano (1986) Superior cervical ganglionectomy prevents gonadal regression and increased plasma prolactin concentrations induced by long days in goats. J. Endocrinol., 110:137-144.

Martinet, L., D. Allain, and C. Weiner (1984) Role of prolactin in the photoperiodic control of moulting in the mink (Mustela uison). J. Endocrinol., 103:9-15.

McDonald, B.J., W.A. Hoey, and P.S. Hopkins (1987) Cyclical fleece growth in cashmere goats. Aust. J. Agric. Res.,

Moore, G.P.M. (1989) Growth factors, cell-cell and cell-matrix interactions in skin during follicle development and growth.

194:694.

16:287-340.

97:lll-124.

73~241-253.

38:597-609.

In: The Biology of Wool and Hair. G.E. Rogers, F!J. Reis, K.A. Ward, and R.C. Marshall, eds. Chapman and Hall, London,

Nixon, A.J., M.P. Gurnsey, K. Betteridge, R.J. Mitchell, and R.A.S. Welch (1991) Seasonal hair follicle activity and fibre growth in some New Zealand cashmere bearing goats, Capra hircus. J. Zool. Lond., 224.589498.

Nixon, A.J., A.J. Pearson, M.A. Aehby, D.P. Saywell, J.E. Wildermoth, and V.J. Choy (1992) Melatonin induced pro- anagen in ferrets as a model for wool and hair growth. Endocr. SOC. Aust. Proc., 35:NZ30.

Oliver, R.F. (1980) Local interactions in mammalian hair growth. In: The Skin of Vertebrates. R.I.C. Spearman and P.A. Riley, eds. Academic Press, London, pp. 199-210..

Prandi, A., M. Motta, F. Chiesa, and C. Tamanini (1988) Circ- annual rhythm of plasma prolactin concentration in the goat. Anim. Reprod. Sci., 17:85-94.

Rougeot, J . , D. Allain, and L. Martinet (1984) Photoperiodic and hormonal control of seasonal coat changes in mammals with special reference to sheep and mink. Acta Zool. Fenn.,

Ryder, M.L. (1966) Coat structure and seasonal shedding in goats. Anim. Prod., 8289-302.

Saywell, D.P., and A.J. Nixon (1992) Cell proliferation during fibre growth initiation in hair follicles of ferrets. Proc. N.Z. SOC. Anim. Prod., 52:299-302.

Silver, A.F., H.B. Chase, and C.T. Arsenault (1969) Early anagen initiated by plucking compared with early spontaneous anagen. Adv. Biol. Skin 9.265-286.

Slominski, A., R. Paus, and R. Costantino (1991) Differential expression and activity of melanogenesis-related proteins dur- ing induced hair growth in mice. J. Invest. Dermatol.,

Straile, W.E., H.B. Chase, and C. Arsenault (1961) Growth and differentiation of hair follicles between periods of activity and quiescence. J. Exp. Zool., 148:205-221.

Welch, R.A.S., M.P. Gurnsey, K. Betteridge, and R.J. Mitchell (1990) Goat fibre response to melatonin given in spring in two consecutive years. Proc. N.Z. SOC. Anim. Prod., 50;

pp. 351-365.

171:13-18.

96:172-179.

335-338.