developmental changes of the proliferative response of...

Development 102, 567-574 (1988)Printed in Great Britain © The Company of Biologists Limited 1988

567

Developmental changes of the proliferative response of mouse

epidermal melanocytes to skin wounding

TOMOHISA HIROBE

Division of Biology, National Institute of Radiological Sciences, Anagawa, Chiba, 260 Japan

Summary

A cut was made on the middorsal skin of mice ofvarious ages of strain C57BL/10J using fine iridec-tomy scissors. Specimens from the wounded skinswere fixed at various days after wounding and weresubjected to the dopa reaction and to the combineddopa-premelanin reaction. When the dorsal skins of1-5-day-old mice were wounded, the melanocyte popu-lation positive to the dopa reaction as well as themelanoblast-melanocyte population positive to thecombined dopa-premelanin reaction increased dra-matically in the epidermis adjacent to a skin wound.Pigment-producing melanocytes in mitosis were fre-quently found in the vicinity of a wound immediatelyafter wounding. When the dorsal skins of 4-5-day-oldmice were wounded, the increase in the melanocyte

and melanoblast-melanocyte populations was smallerthan that of 1-5-day-old mice. The increase in numberof pigment-producing melanocytes in mitosis wasreduced and delayed as compared to 1-5-day-old mice.When the dorsal skins of 8 5-, 20 5-, and 60 5-day-oldmice were wounded, the increase in the melanocyteand melanoblast-melanocyte populations was muchsmaller than the newborn mice. Moreover, pigment-producing melanocytes in mitosis were never found.These results indicate that the proliferative response ofmouse epidermal melanocytes to skin wounding be-comes delayed and diminished with development.

Key words: mouse, melanocyte, proliferation, wounding,epidermis.

Introduction

It has been reported that the number of functioningmelanocytes in the epidermis of mammalian skin isincreased by external stimuli, such as ultravioletirradiation (Quevedo & Smith, 1963; Quevedo et al.1965), carcinogen treatment (Szabo, 1963; Iwata et al.1981) and skin wounding (Staricco, 1961; Snell, 1963;Giacometti & Allegra, 1967; Giacometti et al. 1972).However, whether the increase in the number ofmelanocytes is due to differentiation of precursormelanoblasts or to mitotic division of melanocyteswas not resolved. Recent studies using adult mouseear skin show that the epidermal melanocytes arecontinuously renewed by mitotic division in normalcircumstances (Rosdahl & Lindstrom, 1980; Rosdahl& Bagge, 1981) and that the proliferative activity ofepidermal melanocytes is stimulated by ultravioletirradiation (Rosdahl & Szabo, 1976, 1978; Rosdahl,1978). In the dorsal skin of newborn mice, thepigment-producing melanocytes were shown to

undergo mitotic division during the healing of skinwounds (Hirobe, 1983). Mitotic melanocytes werenever found in normal epidermis. These resultsindicate that the differentiated melanocytes in themouse epidermis can proliferate with or withoutexternal stimuli. However, it is not known whetherthe proliferative activity of differentiated melano-cytes changes with development. The present studywas designed to solve this problem through thewounding of the skin during the postnatal develop-ment of mice. The response of epidermal melano-cytes to skin wounding was analysed by means of lightmicroscopy using techniques of histochemistry and ofconventional histology.

Materials and methods

The animals used in this study were the mouse, Musmusculus, of strain C57BL/10J (substrain C57BL/lOJHir).They were given water, fed ad libitum on a commercial diet(Clea Japan) and maintained at 24±1°C with 40-60%

568 T. Hirobe

relative humidity; 12 h of fluorescent light was provideddaily.

The method of skin wounding was reported previously(Hirobe, 1983). A full-thickness cut 7 mm long was madeanteroposteriorly on the middorsal skin of mice of 1-5, 4-5,8-5, 20-5 and 60-5 days of age using fine iridectomy scissors.The incision extended from the epidermis to the deepestlayer of macrophages under the panniculus carnosusmuscle. Immediately after the incision was made, themargins retracted and the wound cavity widened to about2mm. The wounds were not sutured or dressed. Theanimals were killed at various times thereafter. The entirewound area, including the bed, was then removed from theanimals. Biopsy specimens from the wounded skins andfrom corresponding fields of skin from intact controlanimals were fixed with 16 % formalin in phosphate buffer(pH7-0) for 20-24 h at 2°C. Each age group was rep-resented by three mice and one sample was obtained fromthe wounded skin area of each animal. One wounded skinconsisted of two regenerating epidermes, namely right andleft sides of the wound bed. The experiments were repeatedthree times. The specimens were washed with distilledwater and incubated with 0 1 % L-dopa (3,4-dihydroxy-phenylalanine, Wako) solution in phosphate buffer(pH7-4) for 20-24 h at 37°C. This staining reveals tyrosin-ase-containing differentiated melanocytes (Hirobe, 1982).The specimens were oriented transversely to the woundedge and 10 ^m serial sections were deparaffinized andcounterstained with eosin. For combined dopa-premelaninreaction (combined dopa-ammoniacal silver nitrate stain-ing), deparaffinized sections after the dopa treatment wereincubated with 10 % ammoniacal silver nitrate (Wako)solution for lOmin at 58°C (Mishima, 1960; Hirobe &Takeuchi, 1977). This preferential staining reveals undiffer-entiated melanoblasts that contain unmelanized stage-I and-II melanosomes in addition to tyrosinase-containing differ-entiated melanocytes (Mishima, 1964; Hirobe, 1982). Thespecimens were also counterstained with eosin.

The number of melanocytes (cells positive to the dopareaction) and the number of stage-I and -II melanosome-containing melanoblasts plus melanocytes (cells positive tothe combined dopa-premelanin reaction) were estimatedper 0-1 mm2 of the epidermis of each section of skin, andthe calculations based on ten consecutive sections with thewidth of 1 mm covering the area 04 mm2 of the skin.

In some cases, the specimens from wounded and controlanimals were fixed with Bouin's fixative and sectionedtransversely to the wound edge. Serial sections, 8^m inthickness, were stained with haematoxylin and eosin. Pig-ment-producing melanocytes in resting phase and mitosiswere examined with the light microscope using numeroussections.

Results

Changes in the melanocyte and melanoblast-melanocyte populations in the epidermis afterwounding

When the dorsal skins of 1-5-day-old mice werewounded, the melanocyte population positive to the

dopa reaction as well as the melanoblast-melanocytepopulation positive to the combined dopa-premela-nin reaction increased dramatically in the epidermisadjacent to a skin wound (Fig. 1A,B)- On day 1 afterwounding, the melanocyte and melanoblast-melano-cyte populations in the epidermis within 1 mm of thewound edge significantly (P<0-05) exceeded thecontrols on day 0 and day 1 (Fig. 2). Both populationsshowed maximal number on day 3, then graduallydecreased (Fig. 2). From day 3, both populationswere observed in the roots of hair follicles in additionto the basal layer of epidermis. This suggests that theepidermal melanoblasts or melanocytes migrate intohair follicles. In all stages of wound healing, thenumber of melanocyte population did not differsignificantly from that of melanoblast-melanocytepopulation, suggesting that all melanoblasts differen-tiate into melanocytes in the epidermis adjacent to askin wound. The melanocyte and melanoblast-mela-nocyte populations appeared in the regeneratingwound epidermis on day 3 and increased in number(Fig. 2). Both populations were observed in the rootsof the advancing epidermal sheets from day 3 andlagged behind their forward edges. This suggests thatepidermal melanoblasts or melanocytes increase innumber adjacent to a skin wound and, thereafter,migrate into the regenerating wound epidermis. Bothpopulations showed a maximal number on day 7 anddecreased thereafter. From day 7, both populationswere observed in the roots of hair follicles in additionto the basal layer of the regenerating wound epider-mis, suggesting that the epidermal melanoblasts ormelanocytes migrate into hair follicles. In all stages ofwound healing, the melanocyte and melanoblast-melanocyte populations did not differ significantly innumber, suggesting that all melanoblasts differentiateinto melanocytes in the regenerating wound epider-mis.

When the dorsal skins of 4-5-day-old mice werewounded, the increase in the melanocyte (Fig. 1C)and melanoblast-melanocyte populations (Fig. ID)in the epidermis adjacent to a skin wound was smaller

Fig. 1. Vertical sections of the dorsal skins of C57BL/10Jmice during wound healing. A cut was made on themiddorsal skins of 1-5- (A,B), 4-5- (C,D), 8-5- (E,F),20-5- (G,H) and 60-5- (I,J) day-old mice. Cells positive tothe dopa reaction (A,C,E,G,I) as well as to the combineddopa-premelanin reaction (B,D,F,H,J) are shown in thedorsal skins on day 3 after wounding. Epidermalmelanocytes or melanoblasts are seen in the vicinity of awound (arrows). The number of epidermal melanocytesor melanoblasts is greater in younger mice (A-D) than inolder mice (E-J). The right sides of all figures indicatethe wound edge and regenerating wound epidermis.xl20.

Skin wounding and melanocytes 569

570 T. Hirobe

M 280| 260- 240-^. 220= 200.° 180.> 160•| 140

9- 120D. 100-§ 80"3 60'6 40-Z 20

0J

E 280S 260-1"Si 240< 220t 200

180-160140120100806040200

C57BL/10JDay 1-5

••—ff

0 1 2 3 4 5 7 10Days after wound formation

14

enO.O

•O

oo.

I C57BL/10J\ Wound edge Day 1-5

•> \ Regenerating

'•5-

I0 1 2 3 4 5 7 10

Days after wound formation14

Fig. 2. Changes in the number of melanocyte positive tothe dopa reaction (A) and melanoblast-melanocytepositive to the combined dopa-premelanin reaction (B)in the dorsal skin of l-5-day-old C57BL/10J mice per0-1 mm2 of the epidermis within 1 mm of the wound edge( • ) , the control epidermis (D), and the regeneratingwound epidermis (O). Bars indicate S.E.M.

than that of l-5-day-old mice (Fig. 3). In the regener-ating wound epidermis, both populations appearedon day 2, increased in number until day 7, thengradually decreased (Fig. 3).

When the dorsal skins of 8-5- (Figs 1E,F, 4) or20-5- (Figs 1G,H, 5) day-old mice were wounded, themelanocyte and melanoblast-melanocyte popu-lations in the epidermis within 1 mm of the woundedge exceeded that of the control epidermis. How-ever, the increase in both populations was muchsmaller than that of the younger mice. In the regener-ating wound epidermis, the melanocyte and melano-blast-melanocyte populations appeared on day 3,increased in number until day 10, then decreased.Neither populations exceeded the initial density(Figs 4, 5).

When the dorsal skins of 60-5-day-old mice werewounded, no marked increase in the melanocyte andmelanoblast-melanocyte populations in the epider-mis adjacent to a skin wound was observed (Figs 1I,J,

I 100

60

£ 4°1o.o2 20'oo 0'

C57BL/10JDay 4-5

**• T Regenerating

edge * - . . ^

*2 3 5 7 10


E6

4?

jop

a ^

ive

to (

Is p

osit

iof

eel

No.

120

100

80

60

40

20

0

C57BL/10JDay 4-5

\ ,1

2 3 5 7 10Days after wound formation

Fig. 3. Changes in the number of melanocyte andmelanoblast-melanocyte populations in the dorsal skin of4-5-day-old C57BL/10J mice. Data presented as Fig. 2.

6). In the regenerating wound epidermis, the melano-cyte and melanoblast-melanocyte populationsslightly increased in number after wounding. Bothpopulations on day 14 significantly (P<0-05)exceeded the initial density.

Changes in the mitotic indices of melanocytes in theepidermis after woundingWhen the dorsal skins of l-5-day-old mice werewounded, melanocytes in mitotic division were foundin the epidermis within 1 mm of the wound edge fromday 1 to day 4 (Figs7A,B> 8). Mitotic melanocyteswere most frequently found on day 2 (Mitotic in-dex = 5-06%; Fig. 8). In contrast, mitotic melano-cytes were never found in either the regeneratingwound epidermis or control epidermis.

When the dorsal skins of 4-5-day-old mice werewounded, mitotic melanocytes were also found in theepidermis adjacent to a skin wound from day 3 to day7 (Figs 7C,D, 8). The mitotic index was maximal onday 5 (0-59 %). The increase in the mitotic indices of4-5-day-old mice was reduced and delayed as com-pared to l-5-day-old mice (Fig. 8). In contrast, mi-totic melanocytes were never found in either theregenerating wound epidermis or control epidermis.

14

30

1 2 5©

0 20

>•i 15a.

1 10"S

6 5

0

30

g. 20o

I 1 5

I 10

! 5odZ n

C57BL/10JDay 8-5

0 2 3 7 10Days after wound formation

B

14

C57BL/10JDay 8-5

2 3 7 10Days after wound formation

14

Fig. 4. Changes in the number of melanocyte (A) andmelanoblast-melanocyte (B). Populations of the dorsalskin of 8-5-day-old C57BL/10J mice. Data presented asFig. 2.

When the dorsal skins of 8-5-, 20-5- or 60-5-day-oldmice were wounded, mitotic melanocytes were neverfound in the epidermis within 1 mm of the woundedge, the regenerating wound epidermis, and controlepidermis (Fig. 8).

Discussion

The present study demonstrated that the proliferativeresponse of mouse epidermal melanocytes to skinwounding was diminished as developmental age ad-vanced, since the increase in the melanocyte andmelanoblast-melanocyte populations as well as theincrease in the mitotic indices of pigment-producingmelanocytes of older mice was reduced and delayed


16 A

E

6

eel

>

sit

8.

14

12

10

8

6o

13

No


C57BL/10J

3 7 10Days after wound formation

T 14o<" 12-

a. 10o

•5 6

a= 4oo 26

0 J »

14

C57BL/10JDay 20-5

i3 7 10



e 5

6 4

C57BL/10JDay 60-5

}Regenerating

. Io


14


572 T. Hirobe

Fig. 7. Vertical sections of the dorsal skins of C57BL/10J mice on day 2 (A), day 3 (B), day 5 (C) and day 7 (D) afterwounding. A cut was made on the middorsal skin of 1-5- (A,B) and 4-5- (C,D) day-old mice. Melanocytes in metaphaseare observed in the epidermis (arrows) adjacent to a skin wound. All specimens were fixed with Bouin's fixative andstained with haematoxylin and eosin. No dopa or silver treatment. x460.

as compared to newborn mice. It has been reportedthat melanocytes begin to differentiate in the epider-mis around the time of birth and increase in numberuntil 4 days, then gradually decrease and disappear by30 days of age in the dorsal skin of C57BL/10J strainmice (Hirobe & Takeuchi, 1977; Hirobe, 1984).Therefore, it is conceivable that the differentiatingmelanocytes or newly differentiated melanocyteswhich can be stimulated to undergo mitosis by skinwounding diminish in the epidermis after birth. Theyare thought to migrate into hair follicle in early daysafter birth. These potent cells, which are consideredas 'stem cell', may undergo mitosis by external stimuli

Fig. 8. Developmental change of the mitotic indices ofepidermal melanocytes of the dorsal skin of C57BL/10Jmice stimulated by skin wounding. A cut was made onthe middorsal skin of 1-5- (O), 4-5- ( • ) , and 8-5- (D)day-old mice (arrows). Specimens were fixed at variousdays after wounding with Bouin's fixative. Mitotic indicesof melanocytes in the epidermis within 1 mm of thewound edge are shown. When the dorsal skins of 20-5-,and 60-5-day-old mice were wounded, no melanocytes inmitosis were found.

even after migrating into hair bulbs. This hypothesisis supported by the observations of Silver et al. (1969)that the melanocytes in the hair follicle began todivide when the hair growth cycle entered into newactivation stage (Anagen; Dry, 1926).

In the epidermis of hairy skin, epidermal melano-cytes are found only during the early weeks after birth(Quevedo et al. 1966; Takeuchi, 1968; Hirobe &

2 1

0J

C57BL/10J

Day 1-5

0 1-5 4-5

Day 4-5 D a y 8 . 5

8-5 14-5 18-5Days after birth

22-5


Takeuchi, 1977). However, in the glabrous skin ofear, nose and tail, numerous differentiated melano-cytes are found in the epidermis of adult mice. Recentstudies using the mouse ear skin show that theepidermal melanocyte is normally in continuous re-newal (Rosdahl et al. 1980, 1981). However, it is notknown whether the mitotic activity of epidermalmelanocytes in the mouse skin changes with develop-ment. Also, the epidermal melanocytes of adulthuman skin are reported to undergo mitotic divisionnormally (Jimbow et al. 1975). However, it is notclear whether the mitotic activity of human melano-cytes changes from newborn to adult. The solution ofthese problems is expected to clarify the mechanismof the proliferation of mammalian epidermal melano-cytes during development.

Bucher et al. (1964) showed that in a regeneratingrat liver DNA synthesis became reduced and delayedas developmental age advanced. In the postnataldevelopment of mouse seminal vesicle, the proliferat-ive activity of seminal vesicle cells was shown to bemaximal at both 8 and 30 days of age (Okamoto et al.1982). Shirasawa & Yoshimura (1982) reported thatthe mitotic rate of GH and prolactin cells increasedfrom 5 to 70 days of age in male rat pituitary. Incontrast, ACTH- and TSH-cells decreased with age.On the other hand, Takahashi et al. (1984) reportedthat the mitotic indices of prolactin cells decreasedsteadily from 20 days of age in male rat pituitary,while the mitotic rate in female rat was maximal at 60days of age and then decreased. These studiestogether with the present study show that the differ-entiated cells in mammals can undergo mitosis nor-mally or by external stimuli and that the proliferativeactivity of mammalian cells changes during develop-ment. However, it remains to be solved in a futurestudy whatever common mechanism exists in regulat-ing the proliferative activity of mammalian cells indevelopment.

This work was supported in part by Grant 58740290 fromthe Ministry of Education, Japan.

References

BUCHER, N. L. R., SWAFFIELD, M. N. & DITROIA, J. F.(1964). The influence of age upon the incorporation ofthymidine-2-C14 into the DNA of regenerating rat liver.Cancer Res. 24, 509-512.

DRY, F. W. (1926). The coat of the mouse (Musmusculus). J. Genet. 16, 287-340.

GIACOMETTI, L. & ALLEGRA, F. (1967). The effect ofwounding upon the uptake of 3H-thymidine by guinea-pig epidermal melanocytes. Adv. Biol. Skin 8, 89-95.

GIACOMETTI, L., MONTAGNA, W., BELL, M. & Hu, F.(1972). Epidermal melanocyte proliferation in skinwound healing and after ultraviolet irradiation in

rhesus monkeys. In Pigmentation: Its Genesis andBiologic Control (ed. V. Riley), pp. 451-459. NewYork: Appleton-Century-Crofts.

HIROBE, T. (1982). Origin of melanosome structures andcytochemical localizations of tyrosinase activity indifferentiating epidermal melanocytes of newbornmouse skin. /. exp. Zool. 224, 355-363.

HIROBE, T. (1983). Proliferation of epidermal melanocytesduring the healing of skin wounds in newborn mice.J. exp. Zool. 227, 423-431.

HIROBE, T. (1984). Histochemical survey of thedistribution of the epidermal melanoblasts andmelanocytes in the mouse during fetal and postnatalperiods. Anat. Rec. 208, 589-594.

HIROBE, T. & TAKEUCHI, T. (1977). Induction ofmelanogenesis in the epidermal melanoblasts ofnewborn mouse skin by MSH. J. Embryol. exp.Morph. 37, 79-90.

IWATA, K., INUI, N. & TAKEUCHI, T. (1981). Induction ofactive melanocytes in mouse skin by carcinogens: anew method for detection of skin carcinogens.Carcinogenesis 2, 589-593.

JIMBOW, K., ROTH, S. I., FITZPATRICK, T. B. & SZABO, G.

(1975). Mitotic activity in non-neoplastic melanocytesin vivo as determined by histochemical,autoradiographic, and electron microscope studies.J. Cell Biol. 66, 663-670.

MISHIMA, Y. (1960). New technic for comprehensivedemonstration of melanin, premelanin, and tyrosinasesites. Combined dopa-premelanin reaction. J. invest.Derm. 34, 355-360.

MISHIMA, Y. (1964). Electron microscopic cytochemistryof melanosomes and mitochondria. J. Histochem.Cytochem. 12, 784-790.

OKAMOTO, S., TAKATSUKA, D., TATEISHI, K., OGASAWARA,

Y., YAMANE, T., KITAMURA, Y. & MATSUMOTO, K.

(1982). Proliferative pattern of seminal vesicle cells andthe production of testosterone and 5ar-androgens frombirth to adulthood in mice. Endocrinology 111,1230-1234.

QUEVEDO, W. C , JR & SMITH, J. A. (1963). Studies onradiation- induced tanning of skin. Ann N. Y. Acad.Sci. 100, 364-388.

QUEVEDO, W. C , JR, SZABO, G., VIRKS, J. & SINESI, S. J.

(1965). Melanocyte populations in UV-irradiatedhuman skin. J. invest. Derm. 45, 295-298.

QUEVEDO, W. C , JR, YOULE, M. C , ROVEE, D. T. &

BIENIEKI, T. C. (1966). The developmental fate ofmelanocytes in murine skin. In Structure and Control ofthe Melanocyte (ed. G. Delia Porta & O. Miihlbock),pp. 228-241. Berlin: Springer-Verlag.

ROSDAHL, I. K. (1978). Melanocyte mitosis in UVB-irradiated mouse skin. Ada derm. Venereol. 58,217-221.

ROSDAHL, I. & BAGGE, U. (1981). Vital microscopy ofepidermal melanocytes. Ada derm. Venereol. 61,55-58.

574 T. Hirobe

ROSDAHL, I. K. & LINDSTROM, S. (1980). Morphology ofepidermal melanocytes in different stages of mitosis.Ada derm. Venereol. 60, 209-215.

ROSDAHL, I. K. & SZABO, G. (1976). Thymidine labellingof epidermal melanocytes in UV-irradiated skin. Adaderm. Venereol. 56, 159-162.

ROSDAHL, I. K. & SZABO, G. (1978). Mitotic activity ofepidermal melanocytes in UV-irradiated mouse skin.J. invest. Derm. 70, 143-148.

SHIRASAWA, N. & YOSHIMURA, F. (1982).Immunohistochemical and electron microscopicalstudies of mitotic adenohypophysial cells in differentages of rats. Anat. Embryol. 165, 51-61.

SILVER, A. F., CHASE, H. B. & POTTEN, C. S. (1969).Melanocyte precursor cells in the hair follicle germduring the dormant stage (telogen). Experientia 25,299-301.

SNELL, R. S. (1963). A study of the melanocytes andmelanin in a healing deep wound. J. Anat. 97, 243-253.

STARICCO, R. G. (1961). Mechanism of migration of themelanocytes from the hair follicle into the epidermisfollowing dermabrasion. /. invest. Derm. 36, 99-104.

SZAB6, G. (1963). The effect of carcinogens onmelanocytes. Ann. N.Y. Acad. Sci. 100, 269-278.

TAKAHASHI, S., OKAZAKI, K. & KAWASHIMA, S. (1984).Mitotic activity of prolactin cells in the pituitary glandsof male and female rats of different ages. Cell TissueRes. 235, 497-502.

TAKEUCHI, T. (1968). Genetic analysis of a factorregulating melanogenesis in the mouse melanocyte.Jap. J. Genet. 43, 249-256.

(Accepted 25 November 1987)

developmental changes of the proliferative response of...

Documents