- Email: [email protected]
Fine structure of nuptial pad surface of male ranid frogs
rlSSUE AND CELL, 1993 25 (4) 589-598 0 1993 Longman Group UK Ltd.
S. KURABUCHI
FINE STRUCTURE OF NUPTIAL MALE RANID FROGS
PAD SURFACE
OF
Keywords: Nuptial pad. secondary sexual characteristic. anura. amphibia, electron microscopy ABSTRACT, The nuptial pad, a cutaneous secondary sexual characteristic. developed on the ventrolateral aspect of the first digit (the thumb) of the male ranid frogs was observed. Under scanning electron microscopy, numerous small elevations were observed rising above the pad’s surface; they were rounded in Ram breuipoda porosn and R. rugoso, conical in R. nigromaculata, and rather tall and gradually tapering in R. ornarioentris. These elevations were densely crowded, and in some cases several elevations were seen to have fused to form ridges in the R. rugosa pads. Accessory microprocesses completely covered the outermost layer of cells of the elevations. Numerous pile-like microprocesses gradually shortened from the top toward the side of the elevation in R. ornarioenzris and R. nigromaculara. Those on the top were noticeably longer than those on the side in R. rugosa. The bundles of thick processes radically projected above the top in R. breuipoda porosa. In R. breuipodo porosa nuptial pads were observed under transmission electron microscopy, and the outermost monolayer cells of an elevation were usually keratinized, devoid of organelles and containing closely packed. fine filaments within a dense matrix. The accessory processes also contained dense fibrous matrice\
Introduction
Nuptial pads, occasionally termed ‘thumb pads’ or ‘nuptial excrescences’ (Duellman and Trueb, 1986), are one of the male secondary sexual characteristics of anurans and urodeles, appearing prior to the breeding season and disappearing afterward (Duellman and Trueb, 1986; Fox, 1986). They consist mainly of a thickened epidermis with characteristic architecture and dermis with embedded large skin glands. It has already been firmly established that the occurrence of these pads depends upon the blood level of testicular hormones (Greenberg, 1942; Izzo et al., 1982), and that seasonal variation in development is correlated with reproductive activities (Penhos and Cardeza, 1957; Iwasawa and Asai, 1959; Lofts, 1964). In testectomized anuran subadult, Rana nigromaculata, forelimbs without nuptial pads, Department of Histology, School of Dentistry at Tokyo. Nippon Dental University, Chiyoda-ku. Fujimi l-9-20. Tokyo 102. Japan. Received 19 November 199’2. Revised 15 February 1993 58’)
pads were successfully induced with the administration of testosterone (Iwasawa and Kobayashi, 1974). Furthermore, both male non-breeding newts, Notophthalmus, viridescens, and females, which normally never have pads, have been shown to develop nuptial pads after concurrent administration of prolactin and testosterone (Zimmer and Dent, 1981). The function of nuptial pads is to facilitate the male’s grip on the female during amplexus, and the surface architecture provides the friction necessary for clasping the smooth female body. The pads may also be important in male-male combat (Duellman and Trueb, 1986) and in males holding on to mates when other males are trying to dislodge them (Wells, 1977). According to Duellman and Trueb (1986). the nuptial pads are absent or small on frogs that have amplexus on land, but almost universal on those that have amplexus in water. It has been shown that the position of the nuptial pads varies; they are usually on the thumb, occasionally on the second and third digits, and/or on the ventral surface of the forearm, or on the feet (Duellman and Trueb, 1986). Therefore, the nuptial pads
590
KURABUCHI
seem to exhibit structural peculiarities according to species and family, being adapted to particularly courtship behavior, amplexus and circumstances of breeding. Only a few studies have been made of the 3-dimensional structures of the surface architecture of the nuptial pads, and these have only considered limited spaces and families. For examples, studies of conical projections in the red-spotted newt Notophthalmus viridescens (Forbes et al., 1975), spiky projections in the pipid Xenopus laevis (Kurabuchi and Inoue, 1981), several types of elevations in hylid and leptodactylid frogs (Tyler and Lungershausen, 1986) and two Papuan species of the hylid genus Nyctimystes (Zweifel, 1983) have been conducted. In the current study, scanning and transmission electron microscopes were employed to analyze the surface of the thumb pads of four species of ranid frogs. The Rana observed here are distributed in Japan and their amplexus is axillary. Materials and Methods
The four species of mature male frogs from the family Ranidae observed in the current study were: the brown frog Rana ornativentris with a snout-vent length of 40-45 mm, two pond frogs, R. brevipoda porosa with a snout-vent length of 50-55mm and R. nigromaculata with a snout-vent length of 5055mm, and the mud frog R. rugosa with a snout-vent length of 40-45 mm. All were obtained during their breeding seasons. Male R. ornativentris were collected in the mountainous districts of Tokyo in March. Male R. brevipoda porosa and R. rugosa were
collected in May collected obtained three to
in the vicinity of Maebashi City and June. Male R. nigromaculata in Hamamatsu City in June were from an animal dealer. In each case, five individuals were used.
Preparation of tissues for scanning electron microscope (SEM). After being anesthetized
in 0.1-0*5% solutions of MS222, the first digits (the thumbs) or the entire hands with well-developed nuptial pads were excised and fixed with Karnofsky solution at 4°C for 2hr. The solution contained 2.5% glutaraldehyde and 2% paraformaldehyde in a cacodylate buffer at pH7.4. Some of the materials were trimmed with a razor to allow observation of surface architectures from several angles. Materials were then postfixed in cacodylate-buffered 1% osmium tetroxide solution at 4°C for 2 hr. After they were dehydrated in serial ethanol, they were dried using the critical point method with liquid CO;! and coated with gold in a D.C. sputtering system. Finally, materials were examined with a stereoscan scanning electron microscope (HXM-2X and X-560, Hitachi). Preparation of tissues for transmission electron microscope (TEM). As in the above
methodology, the thumb pads were first fixed with a glutaraldehyde and paraformaldehyde mixture, then cut into smaller blocks. The tissues were then post-fixed in 1% Os04, dehydrated in a graded series of ethanols, embedded in epoxy-resin. The semi-thin sections, after deepoxy-resin according to the method of Imai et al. (1968), were stained with Heidenhain’s iron hematoxylin and roughly observed. Then, the ultrathin sec-
Fig. 1. Scanning electron micrograph of the first digit (the thumb) of male R. breuipoda porosa. The ventrolateral base of this digit is greatly swollen. Numerous small, distinctive elevations crowd the gently-sloping surface of the nuptial pad distally from the digit base to the tip. x24. Fig. 2. Scanning electron micrograph of the first digit (thumb) of male R. ornatiuenlris. Numerous large, pointed elevations project outward from the nuptial pad surface. In the joints shown by arrows, there are narrow areas of the pads where the degree of pad swelling is slight and where few elevations are observed. x24. Fig. 3. The central area of the nuptial pad of R. brevipoda porosa. Rounded the surface appear as a harsh cap-like shape at the top. X600.
elevations
Fig. 4. The central area of the nuptial pad of R. nigromaculafa. Large conical elevations rounded tops rise from the surface. x600.
on
with
592
KURABUCHI
tions were stained with many1 acetate and lead citrate, and examined with electron microscope (JEM-2000EXI1, JEOL). Quantitative analysis. The following measurements were taken while viewing over the nuptial pads in varying positions under SEM micrographs. The elevations on the pads were counted on the photomicrographs at low magnification, with the density of elevation presented as the number per unit surface area (mm2) and the mean -Cstandard deviation calculated among 10 SEM micrographs. The area at the bottom of the elevation was measured in the SEM micrographs viewing just above the elevations, and was also measured in the median plain of the section of one elevation. The mean 2 standard deviation of their measurement was calculated for 20 randomly-selected elevations. Results
Observation with the naked eye showed the nuptial pads to be distributed on the first digits (thumbs) of males’ forelimbs, where the cutaneous swelling extended distally from the ventrolateral base of the thumb to near the digital tip. The degree of swelling of the nuptial pads in R. rugosa was slight compared with other Rana. The pad regions were coloured dark grey in R. ornativentris, greyish-white in R. brevipoda porosa and yellowish-grey in R. nigromaculata and R. rugosa. The skin of R. rugosa was rough to the touch at the nuptial pads, as over the entire body. The skin of the other three Rana was smooth, except for the skin of the nuptial pads.
Scanning electron microscopy (SEM) Rana ornatiuentris. Under
SEM low magnification, the surface of the nuptial pads was easily distinguishable from the rest of the skin surface. Numerous tall elevations were seen in the pads, while the rest of the skin surface was smooth (Fig. 2). Viewed from just above, the elevations were almost circular at the base, but many were gently squarish, because they rose up densely, crowding one another. Observation from one side showed that many elevations tapered upward to a point (Fig. 7). Under higher short microprocesses conmagnification, densed on the free surfaces of the outer layer cells of the elevation, whereas few did so in the cell margins (Fig. 9A). Such microprocesses were generally pile-like, with flat or depressed ends, exce$ for the ones located on the top of the elevation which were club-like with rounded, sometimes bifurcated ends (Fig. 9B). Their length gradually shortened from the top toward the side of the elevation. The density of elevations was calculated as having a mean of 199.5 + 9.3 pieces per mm2. Although the elevations located at the margin of the pads were slightly smaller than others, elevation size was nearly uniform, showing a mean of 4211.3 ? 749.5 pm* in area at the base, and 103.0 * 9.7 pm in height. Rana brevipoda porosa. Numerous rounded elevations crowded the surface of the pads (Fig. 1). Viewed from just above, all the elevations were circular and gently squarish toward the base, because they rose up densely, crowding one another (Fig. 3).
Fig. 5A. Rounded elevations with cap-shaped tops on the nuptial (B) High-magnification view of the top position of an elevation. with depressed tips project from the surface. X 18,000. Inset
Fig. 6. Several ridges, which seem to be formed on the R. rugosa nuptial pad. x600.
pad of R. rugosa. X600. Pile-like microprocesses
by the fusion of several elevations,
Fig. 7. Tall, pointed elevations on the R. ornatiuenfriF nuptial pad. Unique (shown by arrows) of the skin glands are seen among the elevations. x600.
are seen
cylindrical
ducts
Fig. 8. A rounded elevation of the R. breuipoda porosa nuptial pad. Rather long and thick processes project radially in several bundles at the top of the elevation. x3200. Fig. 9A. A pointed elevation of the R. ornarioenrr& nuptial pad. Numerous microprocesses arc visible all over the outer most cells, but not at the cell margins. x2000. Inset (B) High magnification of the top of an elevation. Microprocesses with rather rounded tips, some of which are bifurcated, project outward. ~14,000.
KURABUCHI
594
Every elevation was rough on the top under low SEM magnification, and irregular processes with flat ends and clear edges radially projected outward in bundles on the elevation top, as could be seen under high magnification (Fig. 8). Viewed down toward the lateral side, the lateral surfaces of elevations were covered with faint grain-like microprocesses, among which noodle-like microridges were distributed. Therefore, every elevation was observed to have a harsh caplike shape on its top. The density of elevations was calculated as having a mean of 1315.1 2 131.2 pieces per mm*. The mean area at the base was 535.0 * 129.7 pm*, and the mean height was 33.8 ? 3.8 pm. Rana
nigromaculata. Numerous conical elevations crowded the surface of the pads (Fig. 4). All elevations were nearly circular at the base. The top of each was rather rounded and rough, bearing numerous pilelike microprocesses over all the outer cell surfaces except the cell margins. Such micreprocesses were gradually shortened down toward the side. The density of the elevations was calculated as having a mean of 295.1 -C24.7 pieces per mm*. The mean area at the base was 2688.9 ? 262.5 pm*, and mean height was 25.1 2 2.3 pm. However, elevations gradually became smaller and the space between them widened, near the margin of the pads. Rana rugosa. Numerous
crowded
the surface
rounded elevations of the pads. Many
elevations were nearly circular at the base (Fig. 5a). Some were irregularly squarish, because they rose up densely, crowding one another, and their shape and size were not regular. It was frequently noted that two or more elevations had fused to form ridges in several areas of the pads (Fig. 6). These ridges were well-developed especially in the pads of younger frogs. On the top of each elevation, rather long pile-like microprocesses with depressed ends covered the cell surface (Fig. 5B). As the microprocesses on the lateral side of the elevation were shorter, each elevation was observed to have a caplike shape on its top. Quantitative analysis was done on the pad surfaces, where few fused elevations and ridges were seen. The density of elevations was calculated as having a mean of 490.7 5 77.1 pieces per mm*. The mean area at the base was 611.8 * 230.0 pm2, and the mean height was 25.1 ? 4.6pm. Many tubercules were scattered on the cutaneous surface of the digits other than the pads, similar to the type developed all over the males’ bodies as well as the females’. Transmission electron microscopy (TEM)
The semi-thin sections showed that the skin of the nuptial pads of R. brevipoda porosa consisted of rugged epidermis and thick dermis embedding hypertrophied alveolar glands, of which the secretory granules were stained with Heidenhain’s iron-hematoxylin (Fig. 10). In the median section of the elevation on the pads, 6-7 cell layers were piling up in the epidermis, possessing a stratum
Fig. 10. A semi-thin section, stained with Heidenhain’s iron hematoxyline, of the R. breuipoda porosa nuptial pad. sp: stratum papillae. g: alveolar gland. x 120.
Fig. 11. Transmission electron micrograph of the outside of a median section of an elevation of the R. brevipoda porosa nuptial pad. The epidermal cells vary in shape from cuhoid to flat and the intercellular space gradually narrows, seen in transition from stratum basal (sb), stratum spinosum (ss), stratum granulosum (sg) and stratum corneum (SC).The dermal stratum papilla (sp) is located in the core of elevation epidermis. bm: basement membrane. x3400. Fig. 12. The TEM median section of the top of the elevation of R. breuipoda porosa nuptial pads. In the rather wide space between the stratum granulosum (sg) and stratum corneum (SC), many cytoplasmic digital projections are visible. x.3400. Fig. 13. Desmosomal adhesion (d) of the stratum granulosum and stratum corneum. Several cytoplasmic digitations with tonofilaments project out to adhere by desmosome to the rather flat bottom of the stratum corneum (SC)cells. The digitation’s tip is rather depressed. ~30,000. Fig. 14. TEM micrograph of stratum spinosum cells. Dense ovoidal granules are seen in the upper regions of nuclei. Desmosomes (arrows) are intercalated between the cellular processes. x6OCQ.
KURABUCHI
papillae in a core. Observations of the elevation by TEM (Fig. 11) in the epidermis showed the bowl-shaped outermost stratum corneum to consist of a monolayer of keratinized cells, where few remnants of nucleus and organelles remained. Nearly all of the cytoplasm was occupied by filamentous dark material within a dark matrix (Fig. 12). These adjacent cells were closely jointed with intercellular digitations. Dense materials, which were remnants of desmosome, remained in the narrow intercellular spaces. The accessory processes, as observed by SEM, were projected and seen to contain dense filamentous materials. Towards the bottom was a layer of stratum granulosum, which represented a replacement layer. Numerous dense granules and bundles of filaments were irregularly complicated in a rather light matrix. Such granules seem to correspond to the keratohyaline granules present in mammarian skin epidermis, as considered by Lavker and Matoltsy (1970). Almost all of the bundles consisted of tonofilaments terminating in the dense plaques of the desmosome. The space between the stratum corneum and the stratum granulosum was rather wide, but rather long cytoplasmic processes projected out from the stratum granulosum cells to the flat bottom of the upper cell layer, where desmosomal adhesion occurred between the two layers (Fig. 13). Such processes of the stratum granulosum seem to represent the accessory processes present on the elevation surface following the shedding of the outer cell layer. Beneath these layers, the stratum spinosum consisted of 3-4 layers of cuboidal cells. The cytoplasm of these cells contained a few bundles of tonofilaments and mucoid ovoidal granules located in the upper region of nucleus (Fig. 14). Cellular processes were widely distributed around almost the entire perimeter of each cell, and some of these processes were linked to neighboring cells by desmosome. The stratum basale consisted of one layer of cuboidal or low columnar cells resting upon the basal lamina and underlying dermis. Hemidesmosomes which were not well-developed were found along the membrane adhering to the basal lamina. Discussion
In all the ranid frogs observed here, nuptial
pads developed on the ventrolateral aspects of the forelimb’s first digit (the thumb) as cutaneous swellings. The SEM observation has shown that epidermal surface specializations can be found on the nuptial pads. The structures of the elevations on the pads, although minutely species-specific, can be generally categorized: they were uprightly rounded or conical in shape with unique accessory processes. The TEM observation of R. brevipoda porosa has shown that the surface accessory processes projecting from the outer layer cells of every elevation are well keratinized. Therefore, their accessory processes are rigid and thorny structures. The outermost keratinized cell layer is also present in the smooth skin other than nuptial pads, the surface of which is either covered with short microvilli-like microprocesses and microridges or looked spongey when observed with SEM. So, the surface elevations and accessory processes on the nuptial pads must be due to rough-felt surface. The amplexus is auxillary in these ranid frogs. Such surface architecture on the pads apparently provides the friction for the male’s forelimbs to firmly grasp the female in the axilla. R. nigromaculata and R. brevipoda porosa (=R. porosa porosa, Maeda and Matsui, 1989) are similar in shape, size and bodycolor, and their habitat overlaps in several districts, where natural hybridization seems to be frequent. Their taxonomic status has previously been discussed (Kawamura, 1962; Maeda and Matsui, 1989). From the current observation of the nuptial pads, it is concluded that the two Rana are most likely different species. The elevations in the R. nigromaculata pads were far larger than those in R. brevipoa porosa, and their accessory processes were clearly different; short pilelike processes distributed over the R. nigromaculata elevation surface, but bundles of processes radially projected out strikingly on the R. brevipoda porosa elevation top. In R. rugosa, the entire body of which felt rough, the swelling of the pads is slight and the surface elevations are irregular in shape and rather small. In R. ornativentris, the pads of which are furnished with larger elevations (which probably lend high frictional resistance), breeding is accomplished in early spring and at low temperature, compared with breeding seasons of other three Rana,
NUPTIAL
PAD OF MALE
597
FROGS
the mating of which is done during late spring and early summer. Three-dimensional architecture on the nuptial pad surface has been reported by SEM in Hylidae and Leptodactylidae (Zwiefel, 1983; Tyler and Lungerhausen, 1986). These general elevations are conical, densely crowded together and are furnished with unique accessory processes, but their amplexus style was not mentioned. Axillary amplexus is common to almost all members belonging to both Hylidae and Leptodactylidae (Duellman and Trueb, 1986). Some of the elevation types resemble those of the current Rana. However, it has been reported of several species without assorting to such cases; large spines in hylid Litoria nannotis and L. nyakalensis, and large thorns in leptodactyl Lechriodus melanopyga (Tyler and Lungershausen, 1986). The oviposition of hylid and leptodactylid frogs varies and may occur in water, in contracting foam nets on water, in terrestrial burrows, or in trees (Duellman and Trueb, 1986). So, the diversity of oviposition will yield the variation of the surface architecture of the nuptial pads. On the other hand, 3-dimensional architecture on the nuptial pads of Xenopus laevis is clearly distinct from that of the current Rana. The nuptial pads of Xenopus laevis developed on the ventral aspect of the forelimb. The amplectic position of Xenopus luevis is inginal. Numerous hook-like or rosespiny projections, with a very strong keratinization of surface cells (Dolder, 1976) is crowded on the Xenopus laevis pad’s surface
(Kurabuchi and Inoue, 1981). Therefore. the conclusion is that there is a correlation between amplexus site and the position of nuptial pads and their surface architecture. According to Duellman and Trueb (1986), nuptial pads are absent or minimally developed on frogs that have amplexus on land, nearly universal on those that have amplexus in water, and are most developed in frogs that breed in streams. So, 3-dimensional examinations of nuptial pads remain to be done on other species and families, and their amplexus site and the habitat in which mating takes place also needs to be researched further. The nuptial pad naturally develops in a definite position of the body, which varies in species and families. In addition, the same architecture sometimes develops in the accessory forelimbs (Meyer-Rochow and Koebke, 1986; Iwasawa and Takasu. 1985) and the regenerating outgrowths of the forelimbs (Kurabuchi, 1991). It is well-documented that the development of nuptial pads is dependent upon the concentration of testicular hormones in the blood (Greenberg, 1942; Iwasawa and Kobayashi, 1974; Zimmer and Dent, 1981; Izzo et al.. 1984). However. a receptor for androgens was found throughout the skin, not only in the nuptial pads, suggesting that the skin of the entire body may be a secondary sexual characteristic (D’Istria et al., 1975). Therefore. the mechanism inducing pad development on a definite position of the forelimbs is still unknown.
References Delrio. G. and D‘lstria, M. 1974. Androgen receptor in the thumb pads of Ram esculenfu. Gen. Camp. Endocnnoi 22, 349-350. D‘lstria, M., Delrio, G. and Chiefft. G. 1975. Receptors for sex hormones in the skin of the amphibia. Gen. (‘onr~~. Endocrinol., 26, 281-283. Dolder, H.. 1976. Ultrastructure et formation des crotchets sexuels chez Xenopus. Rev. Sutise. Zool.. 82, 71671X. Ducllman, W. E. and Trueb. L. 1986. Biology amphibians. McGraw-Hill Inc., New York, St. Louis. Forbes. M. S., Dent, J. N. and Singhas, C. A. 1975. Developmental cytology of the nuptial pad in the red-Fpottcd newt. Dee. Bio[. , 46,5678. Fox. H. 1986. Epidermis. In Biology of the inregment. Vol. 2 Verrebrutes. (eds Bereiter-Hahan. J.. Matoltsy. A (+. and Richards, K. S.), pp 78-110. Springer Verlag, Berlin. Heidelberg. Greenberg. B. 1942. Some effects of testosterone on the sexual pigmentation and other sex characters of the cricket frog (Avis grylh). J. Exp. Zoo/. , 91, 435-446. Imai. Y.. Sue. A. and Yamaguchi, A. 1968. A removing method of the resin from epoxy-embedded sections for light microscopy. .I. Electron. Microsc. (Tokyo), 17, 8C85. Iwasawa, H and Asai. 0. 1959. Histological observations on the seasonal change of the testis and the thumb pad in the frog, Ram nigromaculata. J. Fat. Sci. Niigata Uniu., Ser.. 2, 215-219.
of
598
KURABUCHI
Iwasawa, H. and Kobayashi, M. 1974. Effects of testosterone and estradiol on the development of sexual characters in young Rana nigromaculata. Biol. Reprod., 11, 398-405. Iwasawa, H. and Takasu, T. 1985. Study of thumb pad regions developed by the administration of testosterone in a young female of Rana nigromaculata with a supernumary forelimb. Jap. J. Herpetol.. 11, 5-10. Izzo, I., Di Matteo, L., Minucci, S., lela, L., Di Meglio, M. and Rastogi, R. K. 1982. The control of the frog (Rana esculenta) thumb pad. Experientia, 38, 134-135. Kawamura, T. 1962. On the names of some Japanese frogs. J. Sci. Hiroshima Uniu., (B-l), 20, 181-193. Kurabuchi, S. 1991. Nuptial pad developed in anuran heteromorphic limb regeneration. Zool. Sci., 8, 1074. Kurabuchi, S. and Inoue, S. 1981. Small spiny projections in the epidermis of the mature Xenopus laeuis. Ann. 2001. Jap., 54, 182-190. Lavker, R. M. and Matoltsy, G. 1970. Formation of horny cells. The fate of cell organelles and differentiation products in ruminal epithelium. 1. Cell Biol., 44,501-512. Lofts. B. 1964. Seasonal changes in the functional activity of the interstitial and spermatogenetic tissues of the green frog. Rana esculenta. Gen. Comp. Endocrinol., 4, 55&562. Maeda, N. and Matsui, M. 1989. Frogs and toads @Japan. Bun-ichi Sogo Shupan Co., Tokyo. Meyer-Rochow, V. B. and Koebke. .I. 1986. A study of the extra extremity in a five-legged Rana temporaria frog. Zool. Am., 217, 1-13. Penhos, J. C. and Cardeza, A. E. 1957. Caracteres sexuales secundarios de1 sapo macho castrado tratado con hormonas sexuales y acid folio o aminopterina. Reu. Sot. Argent. Biol., 33, 121-128. Tyler, M. _I. and Lungershausen, K. 1986. The ultrastructure of male nuptial pads in some Australopapuan frogs. Trans. R. Sot. S. Awt.. 110, 37-41. Welles, K. D. 1977. The social behaviour of anuran amphibians. Anim. Behau., 25, 66693. Zimmer, J. A. and Dent, J. N. 1981. Hormonal regulation of nuptial pads and tail fins in the female red-spotted newt. Gen. Comp. Endocrinol., 44,436-443. Zweiefel, R. G. 1983. Two new hylid frogs from Papua New Guinea and a discussion of the Nyctimystes Papua species group. Amer. Mu. Nouit.. 2759, 1-18.
S. KURABUCHI
FINE STRUCTURE OF NUPTIAL MALE RANID FROGS
PAD SURFACE
OF
Keywords: Nuptial pad. secondary sexual characteristic. anura. amphibia, electron microscopy ABSTRACT, The nuptial pad, a cutaneous secondary sexual characteristic. developed on the ventrolateral aspect of the first digit (the thumb) of the male ranid frogs was observed. Under scanning electron microscopy, numerous small elevations were observed rising above the pad’s surface; they were rounded in Ram breuipoda porosn and R. rugoso, conical in R. nigromaculata, and rather tall and gradually tapering in R. ornarioentris. These elevations were densely crowded, and in some cases several elevations were seen to have fused to form ridges in the R. rugosa pads. Accessory microprocesses completely covered the outermost layer of cells of the elevations. Numerous pile-like microprocesses gradually shortened from the top toward the side of the elevation in R. ornarioenzris and R. nigromaculara. Those on the top were noticeably longer than those on the side in R. rugosa. The bundles of thick processes radically projected above the top in R. breuipoda porosa. In R. breuipodo porosa nuptial pads were observed under transmission electron microscopy, and the outermost monolayer cells of an elevation were usually keratinized, devoid of organelles and containing closely packed. fine filaments within a dense matrix. The accessory processes also contained dense fibrous matrice\
Introduction
Nuptial pads, occasionally termed ‘thumb pads’ or ‘nuptial excrescences’ (Duellman and Trueb, 1986), are one of the male secondary sexual characteristics of anurans and urodeles, appearing prior to the breeding season and disappearing afterward (Duellman and Trueb, 1986; Fox, 1986). They consist mainly of a thickened epidermis with characteristic architecture and dermis with embedded large skin glands. It has already been firmly established that the occurrence of these pads depends upon the blood level of testicular hormones (Greenberg, 1942; Izzo et al., 1982), and that seasonal variation in development is correlated with reproductive activities (Penhos and Cardeza, 1957; Iwasawa and Asai, 1959; Lofts, 1964). In testectomized anuran subadult, Rana nigromaculata, forelimbs without nuptial pads, Department of Histology, School of Dentistry at Tokyo. Nippon Dental University, Chiyoda-ku. Fujimi l-9-20. Tokyo 102. Japan. Received 19 November 199’2. Revised 15 February 1993 58’)
pads were successfully induced with the administration of testosterone (Iwasawa and Kobayashi, 1974). Furthermore, both male non-breeding newts, Notophthalmus, viridescens, and females, which normally never have pads, have been shown to develop nuptial pads after concurrent administration of prolactin and testosterone (Zimmer and Dent, 1981). The function of nuptial pads is to facilitate the male’s grip on the female during amplexus, and the surface architecture provides the friction necessary for clasping the smooth female body. The pads may also be important in male-male combat (Duellman and Trueb, 1986) and in males holding on to mates when other males are trying to dislodge them (Wells, 1977). According to Duellman and Trueb (1986). the nuptial pads are absent or small on frogs that have amplexus on land, but almost universal on those that have amplexus in water. It has been shown that the position of the nuptial pads varies; they are usually on the thumb, occasionally on the second and third digits, and/or on the ventral surface of the forearm, or on the feet (Duellman and Trueb, 1986). Therefore, the nuptial pads
590
KURABUCHI
seem to exhibit structural peculiarities according to species and family, being adapted to particularly courtship behavior, amplexus and circumstances of breeding. Only a few studies have been made of the 3-dimensional structures of the surface architecture of the nuptial pads, and these have only considered limited spaces and families. For examples, studies of conical projections in the red-spotted newt Notophthalmus viridescens (Forbes et al., 1975), spiky projections in the pipid Xenopus laevis (Kurabuchi and Inoue, 1981), several types of elevations in hylid and leptodactylid frogs (Tyler and Lungershausen, 1986) and two Papuan species of the hylid genus Nyctimystes (Zweifel, 1983) have been conducted. In the current study, scanning and transmission electron microscopes were employed to analyze the surface of the thumb pads of four species of ranid frogs. The Rana observed here are distributed in Japan and their amplexus is axillary. Materials and Methods
The four species of mature male frogs from the family Ranidae observed in the current study were: the brown frog Rana ornativentris with a snout-vent length of 40-45 mm, two pond frogs, R. brevipoda porosa with a snout-vent length of 50-55mm and R. nigromaculata with a snout-vent length of 5055mm, and the mud frog R. rugosa with a snout-vent length of 40-45 mm. All were obtained during their breeding seasons. Male R. ornativentris were collected in the mountainous districts of Tokyo in March. Male R. brevipoda porosa and R. rugosa were
collected in May collected obtained three to
in the vicinity of Maebashi City and June. Male R. nigromaculata in Hamamatsu City in June were from an animal dealer. In each case, five individuals were used.
Preparation of tissues for scanning electron microscope (SEM). After being anesthetized
in 0.1-0*5% solutions of MS222, the first digits (the thumbs) or the entire hands with well-developed nuptial pads were excised and fixed with Karnofsky solution at 4°C for 2hr. The solution contained 2.5% glutaraldehyde and 2% paraformaldehyde in a cacodylate buffer at pH7.4. Some of the materials were trimmed with a razor to allow observation of surface architectures from several angles. Materials were then postfixed in cacodylate-buffered 1% osmium tetroxide solution at 4°C for 2 hr. After they were dehydrated in serial ethanol, they were dried using the critical point method with liquid CO;! and coated with gold in a D.C. sputtering system. Finally, materials were examined with a stereoscan scanning electron microscope (HXM-2X and X-560, Hitachi). Preparation of tissues for transmission electron microscope (TEM). As in the above
methodology, the thumb pads were first fixed with a glutaraldehyde and paraformaldehyde mixture, then cut into smaller blocks. The tissues were then post-fixed in 1% Os04, dehydrated in a graded series of ethanols, embedded in epoxy-resin. The semi-thin sections, after deepoxy-resin according to the method of Imai et al. (1968), were stained with Heidenhain’s iron hematoxylin and roughly observed. Then, the ultrathin sec-
Fig. 1. Scanning electron micrograph of the first digit (the thumb) of male R. breuipoda porosa. The ventrolateral base of this digit is greatly swollen. Numerous small, distinctive elevations crowd the gently-sloping surface of the nuptial pad distally from the digit base to the tip. x24. Fig. 2. Scanning electron micrograph of the first digit (thumb) of male R. ornatiuenlris. Numerous large, pointed elevations project outward from the nuptial pad surface. In the joints shown by arrows, there are narrow areas of the pads where the degree of pad swelling is slight and where few elevations are observed. x24. Fig. 3. The central area of the nuptial pad of R. brevipoda porosa. Rounded the surface appear as a harsh cap-like shape at the top. X600.
elevations
Fig. 4. The central area of the nuptial pad of R. nigromaculafa. Large conical elevations rounded tops rise from the surface. x600.
on
with
592
KURABUCHI
tions were stained with many1 acetate and lead citrate, and examined with electron microscope (JEM-2000EXI1, JEOL). Quantitative analysis. The following measurements were taken while viewing over the nuptial pads in varying positions under SEM micrographs. The elevations on the pads were counted on the photomicrographs at low magnification, with the density of elevation presented as the number per unit surface area (mm2) and the mean -Cstandard deviation calculated among 10 SEM micrographs. The area at the bottom of the elevation was measured in the SEM micrographs viewing just above the elevations, and was also measured in the median plain of the section of one elevation. The mean 2 standard deviation of their measurement was calculated for 20 randomly-selected elevations. Results
Observation with the naked eye showed the nuptial pads to be distributed on the first digits (thumbs) of males’ forelimbs, where the cutaneous swelling extended distally from the ventrolateral base of the thumb to near the digital tip. The degree of swelling of the nuptial pads in R. rugosa was slight compared with other Rana. The pad regions were coloured dark grey in R. ornativentris, greyish-white in R. brevipoda porosa and yellowish-grey in R. nigromaculata and R. rugosa. The skin of R. rugosa was rough to the touch at the nuptial pads, as over the entire body. The skin of the other three Rana was smooth, except for the skin of the nuptial pads.
Scanning electron microscopy (SEM) Rana ornatiuentris. Under
SEM low magnification, the surface of the nuptial pads was easily distinguishable from the rest of the skin surface. Numerous tall elevations were seen in the pads, while the rest of the skin surface was smooth (Fig. 2). Viewed from just above, the elevations were almost circular at the base, but many were gently squarish, because they rose up densely, crowding one another. Observation from one side showed that many elevations tapered upward to a point (Fig. 7). Under higher short microprocesses conmagnification, densed on the free surfaces of the outer layer cells of the elevation, whereas few did so in the cell margins (Fig. 9A). Such microprocesses were generally pile-like, with flat or depressed ends, exce$ for the ones located on the top of the elevation which were club-like with rounded, sometimes bifurcated ends (Fig. 9B). Their length gradually shortened from the top toward the side of the elevation. The density of elevations was calculated as having a mean of 199.5 + 9.3 pieces per mm2. Although the elevations located at the margin of the pads were slightly smaller than others, elevation size was nearly uniform, showing a mean of 4211.3 ? 749.5 pm* in area at the base, and 103.0 * 9.7 pm in height. Rana brevipoda porosa. Numerous rounded elevations crowded the surface of the pads (Fig. 1). Viewed from just above, all the elevations were circular and gently squarish toward the base, because they rose up densely, crowding one another (Fig. 3).
Fig. 5A. Rounded elevations with cap-shaped tops on the nuptial (B) High-magnification view of the top position of an elevation. with depressed tips project from the surface. X 18,000. Inset
Fig. 6. Several ridges, which seem to be formed on the R. rugosa nuptial pad. x600.
pad of R. rugosa. X600. Pile-like microprocesses
by the fusion of several elevations,
Fig. 7. Tall, pointed elevations on the R. ornatiuenfriF nuptial pad. Unique (shown by arrows) of the skin glands are seen among the elevations. x600.
are seen
cylindrical
ducts
Fig. 8. A rounded elevation of the R. breuipoda porosa nuptial pad. Rather long and thick processes project radially in several bundles at the top of the elevation. x3200. Fig. 9A. A pointed elevation of the R. ornarioenrr& nuptial pad. Numerous microprocesses arc visible all over the outer most cells, but not at the cell margins. x2000. Inset (B) High magnification of the top of an elevation. Microprocesses with rather rounded tips, some of which are bifurcated, project outward. ~14,000.
KURABUCHI
594
Every elevation was rough on the top under low SEM magnification, and irregular processes with flat ends and clear edges radially projected outward in bundles on the elevation top, as could be seen under high magnification (Fig. 8). Viewed down toward the lateral side, the lateral surfaces of elevations were covered with faint grain-like microprocesses, among which noodle-like microridges were distributed. Therefore, every elevation was observed to have a harsh caplike shape on its top. The density of elevations was calculated as having a mean of 1315.1 2 131.2 pieces per mm*. The mean area at the base was 535.0 * 129.7 pm*, and the mean height was 33.8 ? 3.8 pm. Rana
nigromaculata. Numerous conical elevations crowded the surface of the pads (Fig. 4). All elevations were nearly circular at the base. The top of each was rather rounded and rough, bearing numerous pilelike microprocesses over all the outer cell surfaces except the cell margins. Such micreprocesses were gradually shortened down toward the side. The density of the elevations was calculated as having a mean of 295.1 -C24.7 pieces per mm*. The mean area at the base was 2688.9 ? 262.5 pm*, and mean height was 25.1 2 2.3 pm. However, elevations gradually became smaller and the space between them widened, near the margin of the pads. Rana rugosa. Numerous
crowded
the surface
rounded elevations of the pads. Many
elevations were nearly circular at the base (Fig. 5a). Some were irregularly squarish, because they rose up densely, crowding one another, and their shape and size were not regular. It was frequently noted that two or more elevations had fused to form ridges in several areas of the pads (Fig. 6). These ridges were well-developed especially in the pads of younger frogs. On the top of each elevation, rather long pile-like microprocesses with depressed ends covered the cell surface (Fig. 5B). As the microprocesses on the lateral side of the elevation were shorter, each elevation was observed to have a caplike shape on its top. Quantitative analysis was done on the pad surfaces, where few fused elevations and ridges were seen. The density of elevations was calculated as having a mean of 490.7 5 77.1 pieces per mm*. The mean area at the base was 611.8 * 230.0 pm2, and the mean height was 25.1 ? 4.6pm. Many tubercules were scattered on the cutaneous surface of the digits other than the pads, similar to the type developed all over the males’ bodies as well as the females’. Transmission electron microscopy (TEM)
The semi-thin sections showed that the skin of the nuptial pads of R. brevipoda porosa consisted of rugged epidermis and thick dermis embedding hypertrophied alveolar glands, of which the secretory granules were stained with Heidenhain’s iron-hematoxylin (Fig. 10). In the median section of the elevation on the pads, 6-7 cell layers were piling up in the epidermis, possessing a stratum
Fig. 10. A semi-thin section, stained with Heidenhain’s iron hematoxyline, of the R. breuipoda porosa nuptial pad. sp: stratum papillae. g: alveolar gland. x 120.
Fig. 11. Transmission electron micrograph of the outside of a median section of an elevation of the R. brevipoda porosa nuptial pad. The epidermal cells vary in shape from cuhoid to flat and the intercellular space gradually narrows, seen in transition from stratum basal (sb), stratum spinosum (ss), stratum granulosum (sg) and stratum corneum (SC).The dermal stratum papilla (sp) is located in the core of elevation epidermis. bm: basement membrane. x3400. Fig. 12. The TEM median section of the top of the elevation of R. breuipoda porosa nuptial pads. In the rather wide space between the stratum granulosum (sg) and stratum corneum (SC), many cytoplasmic digital projections are visible. x.3400. Fig. 13. Desmosomal adhesion (d) of the stratum granulosum and stratum corneum. Several cytoplasmic digitations with tonofilaments project out to adhere by desmosome to the rather flat bottom of the stratum corneum (SC)cells. The digitation’s tip is rather depressed. ~30,000. Fig. 14. TEM micrograph of stratum spinosum cells. Dense ovoidal granules are seen in the upper regions of nuclei. Desmosomes (arrows) are intercalated between the cellular processes. x6OCQ.
KURABUCHI
papillae in a core. Observations of the elevation by TEM (Fig. 11) in the epidermis showed the bowl-shaped outermost stratum corneum to consist of a monolayer of keratinized cells, where few remnants of nucleus and organelles remained. Nearly all of the cytoplasm was occupied by filamentous dark material within a dark matrix (Fig. 12). These adjacent cells were closely jointed with intercellular digitations. Dense materials, which were remnants of desmosome, remained in the narrow intercellular spaces. The accessory processes, as observed by SEM, were projected and seen to contain dense filamentous materials. Towards the bottom was a layer of stratum granulosum, which represented a replacement layer. Numerous dense granules and bundles of filaments were irregularly complicated in a rather light matrix. Such granules seem to correspond to the keratohyaline granules present in mammarian skin epidermis, as considered by Lavker and Matoltsy (1970). Almost all of the bundles consisted of tonofilaments terminating in the dense plaques of the desmosome. The space between the stratum corneum and the stratum granulosum was rather wide, but rather long cytoplasmic processes projected out from the stratum granulosum cells to the flat bottom of the upper cell layer, where desmosomal adhesion occurred between the two layers (Fig. 13). Such processes of the stratum granulosum seem to represent the accessory processes present on the elevation surface following the shedding of the outer cell layer. Beneath these layers, the stratum spinosum consisted of 3-4 layers of cuboidal cells. The cytoplasm of these cells contained a few bundles of tonofilaments and mucoid ovoidal granules located in the upper region of nucleus (Fig. 14). Cellular processes were widely distributed around almost the entire perimeter of each cell, and some of these processes were linked to neighboring cells by desmosome. The stratum basale consisted of one layer of cuboidal or low columnar cells resting upon the basal lamina and underlying dermis. Hemidesmosomes which were not well-developed were found along the membrane adhering to the basal lamina. Discussion
In all the ranid frogs observed here, nuptial
pads developed on the ventrolateral aspects of the forelimb’s first digit (the thumb) as cutaneous swellings. The SEM observation has shown that epidermal surface specializations can be found on the nuptial pads. The structures of the elevations on the pads, although minutely species-specific, can be generally categorized: they were uprightly rounded or conical in shape with unique accessory processes. The TEM observation of R. brevipoda porosa has shown that the surface accessory processes projecting from the outer layer cells of every elevation are well keratinized. Therefore, their accessory processes are rigid and thorny structures. The outermost keratinized cell layer is also present in the smooth skin other than nuptial pads, the surface of which is either covered with short microvilli-like microprocesses and microridges or looked spongey when observed with SEM. So, the surface elevations and accessory processes on the nuptial pads must be due to rough-felt surface. The amplexus is auxillary in these ranid frogs. Such surface architecture on the pads apparently provides the friction for the male’s forelimbs to firmly grasp the female in the axilla. R. nigromaculata and R. brevipoda porosa (=R. porosa porosa, Maeda and Matsui, 1989) are similar in shape, size and bodycolor, and their habitat overlaps in several districts, where natural hybridization seems to be frequent. Their taxonomic status has previously been discussed (Kawamura, 1962; Maeda and Matsui, 1989). From the current observation of the nuptial pads, it is concluded that the two Rana are most likely different species. The elevations in the R. nigromaculata pads were far larger than those in R. brevipoa porosa, and their accessory processes were clearly different; short pilelike processes distributed over the R. nigromaculata elevation surface, but bundles of processes radially projected out strikingly on the R. brevipoda porosa elevation top. In R. rugosa, the entire body of which felt rough, the swelling of the pads is slight and the surface elevations are irregular in shape and rather small. In R. ornativentris, the pads of which are furnished with larger elevations (which probably lend high frictional resistance), breeding is accomplished in early spring and at low temperature, compared with breeding seasons of other three Rana,
NUPTIAL
PAD OF MALE
597
FROGS
the mating of which is done during late spring and early summer. Three-dimensional architecture on the nuptial pad surface has been reported by SEM in Hylidae and Leptodactylidae (Zwiefel, 1983; Tyler and Lungerhausen, 1986). These general elevations are conical, densely crowded together and are furnished with unique accessory processes, but their amplexus style was not mentioned. Axillary amplexus is common to almost all members belonging to both Hylidae and Leptodactylidae (Duellman and Trueb, 1986). Some of the elevation types resemble those of the current Rana. However, it has been reported of several species without assorting to such cases; large spines in hylid Litoria nannotis and L. nyakalensis, and large thorns in leptodactyl Lechriodus melanopyga (Tyler and Lungershausen, 1986). The oviposition of hylid and leptodactylid frogs varies and may occur in water, in contracting foam nets on water, in terrestrial burrows, or in trees (Duellman and Trueb, 1986). So, the diversity of oviposition will yield the variation of the surface architecture of the nuptial pads. On the other hand, 3-dimensional architecture on the nuptial pads of Xenopus laevis is clearly distinct from that of the current Rana. The nuptial pads of Xenopus laevis developed on the ventral aspect of the forelimb. The amplectic position of Xenopus luevis is inginal. Numerous hook-like or rosespiny projections, with a very strong keratinization of surface cells (Dolder, 1976) is crowded on the Xenopus laevis pad’s surface
(Kurabuchi and Inoue, 1981). Therefore. the conclusion is that there is a correlation between amplexus site and the position of nuptial pads and their surface architecture. According to Duellman and Trueb (1986), nuptial pads are absent or minimally developed on frogs that have amplexus on land, nearly universal on those that have amplexus in water, and are most developed in frogs that breed in streams. So, 3-dimensional examinations of nuptial pads remain to be done on other species and families, and their amplexus site and the habitat in which mating takes place also needs to be researched further. The nuptial pad naturally develops in a definite position of the body, which varies in species and families. In addition, the same architecture sometimes develops in the accessory forelimbs (Meyer-Rochow and Koebke, 1986; Iwasawa and Takasu. 1985) and the regenerating outgrowths of the forelimbs (Kurabuchi, 1991). It is well-documented that the development of nuptial pads is dependent upon the concentration of testicular hormones in the blood (Greenberg, 1942; Iwasawa and Kobayashi, 1974; Zimmer and Dent, 1981; Izzo et al.. 1984). However. a receptor for androgens was found throughout the skin, not only in the nuptial pads, suggesting that the skin of the entire body may be a secondary sexual characteristic (D’Istria et al., 1975). Therefore. the mechanism inducing pad development on a definite position of the forelimbs is still unknown.
References Delrio. G. and D‘lstria, M. 1974. Androgen receptor in the thumb pads of Ram esculenfu. Gen. Camp. Endocnnoi 22, 349-350. D‘lstria, M., Delrio, G. and Chiefft. G. 1975. Receptors for sex hormones in the skin of the amphibia. Gen. (‘onr~~. Endocrinol., 26, 281-283. Dolder, H.. 1976. Ultrastructure et formation des crotchets sexuels chez Xenopus. Rev. Sutise. Zool.. 82, 71671X. Ducllman, W. E. and Trueb. L. 1986. Biology amphibians. McGraw-Hill Inc., New York, St. Louis. Forbes. M. S., Dent, J. N. and Singhas, C. A. 1975. Developmental cytology of the nuptial pad in the red-Fpottcd newt. Dee. Bio[. , 46,5678. Fox. H. 1986. Epidermis. In Biology of the inregment. Vol. 2 Verrebrutes. (eds Bereiter-Hahan. J.. Matoltsy. A (+. and Richards, K. S.), pp 78-110. Springer Verlag, Berlin. Heidelberg. Greenberg. B. 1942. Some effects of testosterone on the sexual pigmentation and other sex characters of the cricket frog (Avis grylh). J. Exp. Zoo/. , 91, 435-446. Imai. Y.. Sue. A. and Yamaguchi, A. 1968. A removing method of the resin from epoxy-embedded sections for light microscopy. .I. Electron. Microsc. (Tokyo), 17, 8C85. Iwasawa, H and Asai. 0. 1959. Histological observations on the seasonal change of the testis and the thumb pad in the frog, Ram nigromaculata. J. Fat. Sci. Niigata Uniu., Ser.. 2, 215-219.
of
598
KURABUCHI
Iwasawa, H. and Kobayashi, M. 1974. Effects of testosterone and estradiol on the development of sexual characters in young Rana nigromaculata. Biol. Reprod., 11, 398-405. Iwasawa, H. and Takasu, T. 1985. Study of thumb pad regions developed by the administration of testosterone in a young female of Rana nigromaculata with a supernumary forelimb. Jap. J. Herpetol.. 11, 5-10. Izzo, I., Di Matteo, L., Minucci, S., lela, L., Di Meglio, M. and Rastogi, R. K. 1982. The control of the frog (Rana esculenta) thumb pad. Experientia, 38, 134-135. Kawamura, T. 1962. On the names of some Japanese frogs. J. Sci. Hiroshima Uniu., (B-l), 20, 181-193. Kurabuchi, S. 1991. Nuptial pad developed in anuran heteromorphic limb regeneration. Zool. Sci., 8, 1074. Kurabuchi, S. and Inoue, S. 1981. Small spiny projections in the epidermis of the mature Xenopus laeuis. Ann. 2001. Jap., 54, 182-190. Lavker, R. M. and Matoltsy, G. 1970. Formation of horny cells. The fate of cell organelles and differentiation products in ruminal epithelium. 1. Cell Biol., 44,501-512. Lofts. B. 1964. Seasonal changes in the functional activity of the interstitial and spermatogenetic tissues of the green frog. Rana esculenta. Gen. Comp. Endocrinol., 4, 55&562. Maeda, N. and Matsui, M. 1989. Frogs and toads @Japan. Bun-ichi Sogo Shupan Co., Tokyo. Meyer-Rochow, V. B. and Koebke. .I. 1986. A study of the extra extremity in a five-legged Rana temporaria frog. Zool. Am., 217, 1-13. Penhos, J. C. and Cardeza, A. E. 1957. Caracteres sexuales secundarios de1 sapo macho castrado tratado con hormonas sexuales y acid folio o aminopterina. Reu. Sot. Argent. Biol., 33, 121-128. Tyler, M. _I. and Lungershausen, K. 1986. The ultrastructure of male nuptial pads in some Australopapuan frogs. Trans. R. Sot. S. Awt.. 110, 37-41. Welles, K. D. 1977. The social behaviour of anuran amphibians. Anim. Behau., 25, 66693. Zimmer, J. A. and Dent, J. N. 1981. Hormonal regulation of nuptial pads and tail fins in the female red-spotted newt. Gen. Comp. Endocrinol., 44,436-443. Zweiefel, R. G. 1983. Two new hylid frogs from Papua New Guinea and a discussion of the Nyctimystes Papua species group. Amer. Mu. Nouit.. 2759, 1-18.