- Email: [email protected]
Structural and mechanical aspects of the skin of Bufo marinus (Anura, Amphibia)
Tissue & Cell, 2001 33 (5) 541±547 ß 2001 Harcourt Publishers Ltd DOI: 10.1054/tice.2001.0208, available online at http://www.idealibrary.com
Tissue&Cell
Structural and mechanical aspects of the skin of Bufo marinus (Anura, Amphibia) G. Schwinger,1 K. Zanger,1 H. Greven2 Abstract. Specific biomechanical characters and some structures possibly related to them were investigated in the skin of the toad Bufo marinus using tensile testing techniques (at constant strain till rupture) as well as morphological methods (histological, immunohistochemical and electronmicroscopical). Mechanical parameters of the native skin varied considerably according to sex, individual variability and/or site of specimen collection. In skin strips of males and females excised from different parts of the body thickness ranged from 0.45 to 0.87 mm, strain (ef) from 96.52 to 211.03, tensile strength (sm) from 5.72 to 9.38 MPa, and stiffness (E-modulus) from 5.76 to 6.73. The dermis of B. marinus is provided with a collagenous stratum compactum of considerable thickness, a stratum spongiosum with loosely arranged fibres and a marked calcified layer (substantia amorpha). Collagen appears to be the main determinant of skin mechanics. However, the slope of the J-shaped static stress-strain curves indicates elastin to be responsible for the high values of strain. Contrary to van Gieson and orcein staining, immunostaining with a monoclonal antibody against elastin revealed very few elastic fibers between collagen bundles and in the vertical fiber tracts (perforating bundles), but a considerable amount in the tela subcutanea. This was partly confirmed at the ultrastructural level by tannic acid staining. ß 2001 Harcourt Publishers Ltd Keywords: biomechanics, tensile test, anuran skin, collagen, elastin
Introduction The anuran skin is a highly composite structure that consists of the squamous epidermis, including basal, spinosous and horny layers and the underlying dermis. The latter contains epidermis-derived alveolar glands and can be differentiated in a stratum spongiosum and a stratum 1 Institut fuÈr Topographische Anatomie und Biomechanik, Medizinische Einrichtungen der Heinrich Heine-UniversitaÈt, UniversitaÈtsstr. 1, D-40225 DuÈsseldorf, Germany, 2Institut fuÈr Zoomorphologie und Zellbiologie der UniversitaÈt, UniversitaÈtsstr. 1, D-40225 DuÈsseldorf, Germany
Received 13 April 2001 Accepted 25 July 2001 Correspondence to: Gerhard Schwinger, Institut fuÈr Topographische Anatomie und Biomechanik, Medizinische Einrichtungen der Heinrich-Heine UniversitaÈt, UniversitaÈtsstr. 1, D-40225 DuÈsseldorf, Germany; Tel.: 49 211 811 2719; Fax: 49 211 811 2806; E-mail: [email protected]
compactum. Both strata are characterized by their collagen bundles, which are loosely arranged in the st. spongiosum and in a criss-cross manner in the st. compactum. In many species a calci®ed substantia amorpha lies between the two dermal layers (for review see Fox 1986). It has been suggested that the st. compactum is the most conservative element in the vertebrate skin (Meyer et al., 1989), and that collagen mainly characterizes the mechanical properties of a skin (Vincent, 1982; Wainwright et al., 1976). As shown by histological methods, electronmicroscopical techniques and typically J-shaped static stress strain curves, the same appeared to hold for the skin of anurans (Greven et al., 1995; Zanger et al., 1995). In these articles, we examine the reasons why anurans living predominantly in an aquatic and terrestrial environment or in both (amphibious species) should have differently structured skins with different mechanical properties. Comparing for instance the mechanical 541
542
SCHWINGER ET AL.
Table 1 Parameters obtained for strips of dorsal, ventral and lateral skin from males and females of Bufo marinus stretched uniaxially along the longitudinal axis. Values are means plus standard deviation; sm maximal stress prior to failure; ef maximal strain at break; E modulus of elasticity Sex
Sample
n
thickness (mm)
Stress sm (MPa)
Strain ef (%lo)
E (MPa)
Male
Ventral Dorsal Lateral
11 20 3
0.50+0.05 0.87+0.13 0.71+0.16
8.70+2.71 6.28+1.88 9.38+5.28
123.80+37.82 106.39+32.54 211.03+18.63
6.65+2.41 6.27+1.21 5.76+2.48
Female
Ventral Dorsal Lateral
10 6 4
0.45+0.04 0.80+0.10 0.54+0.10
7.40+1.31 5.72+2.29 7.39+2.24
96.52+18.21 114.60+51.73 204.54+31.96
6.73+1.89 6.66+1.80 6.46+0.85
parameters of the skin of the fully aquatic Xenopus laevis (Greven et al., 1995) and the semiaquatic Rana esculenta (Zanger et al., 1995), most of the values, in particular the value for extension load, were lower in R. esculenta. In the present paper, we report on the skin of the terrestrial toad Bufo marinus, to cover the whole spectrum of the habitats used by anurans, one of the most widespread and common American anuran species. We determined the same mechanical and structural parameters as in the two previous papers with the exception of toughness to facilitate comparison.
Materials and methods Animals We used 4 adult males and 2 females of the toad Bufo marinus outside the breeding season bought from a commercial dealer. Biomechanics Preparation of the samples and the tensiometer used were the same as described previously (Greven et al., 1995; Zanger et al., 1995). Light microscopy Strips were ®xed in 4% formaldehyde or absolute ethanol and embedded in paraplast. 7 mm-sections were stained either with Picro Sirius Red (Junqueira et al., 1979) and examined in a Leitz polarizing microscope or stained according to van Gieson and with resorcin to visualize elastin ®bres (Romeis, 1989). Resorcin staining was modi®ed and counterstained with Kernechtrot (Novotny, personal communication). Orcein staining was also carried out. For photographic documentation, collagenous ®bres were stained with Goldner's trichrome. In order to demonstrate the substantia amorpha, a calci®ed layer in the dermis of many anurans (e.g. Elkan, 1976; Toledo & Jared, 1993), sections were treated by the van Kossa method (ethanol ®xation). For immunohistochemistry, native cryostate sections were treated with 0.1% trypsin (Sigma Chemicals Co.) in PBS for 20 min at 378C. Endogenous peroxidase was blocked with fresh 3% hydrogen peroxide in isopropanol for 30 min. Before incubation sections in aqua dest. were
placed for 15 min in a microwave oven (850 W). To reduce non speci®c staining, they were incubated for 60 min at room temperature with normal rabbit non immune serum (1:100 dilution; Dako 049A), washed in PBS, and then incubated for 2 h with the primary (mouse) antibody (1:100 dilution; monoclonal anti-Elastin; Sigma E-4013), washed again and incubated with HRPconjugated secondary (goat anti-mouse IgG) antibody (1:100 dilution; Sigma A-9917) for 2 h. The sections were then washed in PBS and TRIS and reacted with 3,3'-diaminobenzidine (freshly prepared in 0.05 mol/l TRIS buffer, pH 7.6, containing 0.015 mol/l hydrogen peroxide). Pieces of a rabbit arteriole were used as a positive control. Electron microscopy For SEM, the stratum compactum was mechanically separated using ®ne forceps, and using the same ®xation as for TEM. Then the specimen were dehydrated and critical point dried. Preparations were sputtered with gold and examined in a Jeol JSM 35 CF scanning electron microscope. Collagen ®ber angles within the stratum compactum were measured in low-power SEM-micrographs. For TEM, some samples of skin were ®xed in 2.5% glutaraldehyd in 0.1 mol/l cacodylate buffer, pH 7.2, with and without 1% tannic acid, and post®xed with 1% osmium tetroxide (Imayama & Braverman, 1988). Ultrathin sections were stained with lead citrate and uranyl acetate and examined in a Zeiss EM 9 transmission electron microscope.
Results The thickness of the skin is similar in both sexes in the different body regions examined (Table 1). Measurements with the vernier caliper revealed a thickness of 0.45±0.87 mm, whereas measurement in the histological sections gave values of 0.48±1.34 mm. The dorsal skin is thicker than the ventral, involving both epidermis and dermis. Signi®cant differences between sexes could not be detected in the small samples available. The thickest layer with 0.45±1.28 mm dorsally and 0.51±0.72 mm ventrally is the dermis. It includes dermal glands (derivatives of the epidermis; Figs 1, 3, 5), blood
BIOMECHANICS OF BUFO MARINUS SKIN
ep
ep mg ss
ss
gg sc
1
2
* sc
3
4
ss
sc
5
6
543
544
SCHWINGER ET AL.
Fig. 1 Slightly oblique paraplast section of the dorsal skin stained with trichrome Goldner; ep, epidermis; gg, granular glands; mg, mucous glands; sc,
stratum compactum; ss, stratum spongiosum.450. Fig. 2 Paraplast section of the ventral skin stained with orcein. Note orcein-positive material (arrows) attached to a mucous gland; ep, epidermis; ss, stratum spongiosum.1000. Fig. 3 Paraplast section of the dorsal skin stained according van Kossa's method. Note the prominent calcified substantia amorpha (asterisk) immediately above the stratum compactum (sc).450. Fig. 4 Paraplast section of the ventral skin stained with resorcin and counterstained with Kernechtrot. Resorcin-positive fibers (arrows) are seen in the connective tissue (stratum spongiosum) between adjacent glands.900. Fig. 5 Polarized light micrograph of the dorsal skin stained with Sirius red. Note different course of the abundant collagen fiber bundles in the stratum spongiosum (ss) and st. compactum (sc).450. Fig. 6 Immunohistochemical demonstration of elastin fibers (arrows) in a perforating bundle of the ventral skin (without counterstain).800.
vessels, smooth muscles, chromatophores, nerves, (not shown), and abundant collagen ®bers (Figs 7±9). It is clearly differentiated in an upper stratum spongiosum with loosely arranged ®bers, the stratum compactum with the typical criss-cross pattern of ®bers (Figs 1, 5, 7, 8±10). Between these both strata, the calci®ed substantia amorpha is visible (Figs 1, 3, 8). Within the dermis, several different structures surrounding the glands (Fig. 2), ®bers in the tela subcutanea (not shown) as well as ®bers within the bundles of connective tissue perforating the st. compactum could be stained with resorcin and Kernechtrot. Fibers in the stratum spongiosum between the glands (Fig. 4) as well as ®bers in the tela subcutanea (not shown, compare Fig. 11), and ®bres within the bundles of connective tissue perforating the stratum compactum (not shown, compare Fig. 6) were also stained. Sections stained for immunohistochemistry showed a strong reaction in the tela subcutanea and sporadically a positive reaction in the perforating bundles (Fig. 6). Ultrathin sections of tannin treated samples revealed typical elastic ®bers that are composed of amorphous elastin with a peripheral micro®brillar protein sheath (Fig. 10). In conventional ultrathin sections elastic ®bers appeared as electron-dense ®brous structures (Fig. 11). Dermal components surrounding the glands and stained with orcein or resorcin, did not reveal elastin structure in ultrathin sections. Collagen ®bers were the conspicous components of the st. spongiosum and the st. compactum. They were arranged more or less disordered in the stratum spongiosum and show a plywood pattern in the stratum compactum. These patterns could be shown by polarizing microscopy of Sirius Red stained sections (Fig. 5) and were con®rmed by TEM (Figs 8, 9). The thickness of the st. compactum ranged from about 233 to 720 mm. The angles, in which collagen ®bers were orientated varied from 508 to 808 in both sexes (Fig. 7). The substantia amorpha was localized between the two dermal layers. It contained a great number of small, irregular electron-dense granules (Fig. 8). The st. compactum is perforated by vertical bundles of collagen, which number 3 to 7 per mm of skin. The static stress-strain curves were J-shaped, consisting of an initial toe part and a linear portion (Fig. 12). No differences between males and females were detected. The biomechanical data measured are summarized in Table 1. Maximal tensile strength (sm), maximal strain at break (ef), and Young's modulus E varied considerably, as re¯ected by the high standard deviations.
Discussion Bufo marinus has the typical anuran skin, i.e. a squamous epidermis, a dermis differentiated in a stratum spongiosum (with alveolar glands) and a stratum compactum and, not present in all anuran skins (see Elkan, 1957; Toledo & Jared, 1993), a prominent calci®ed substantia amorpha. The st. compactum as well as the s. amorpha are perforated by vertical bundles of collagen (see among others Fox, 1986; Dene¯e et al., 1987, 1993). The angle of the collagen ®bres in the st. compactum is in the same order as in other anurans. Due to its composite nature, it is very dif®cult to associate a particular structure with particular biomechanical properties of the skin. These also holds for the collagenous bundles that perforate the s. amorpha and st. compactum (Greven et al., 1995; Zanger et al., 1995). A rough estimation revealed 3 to 7 bundles per mm in the dorsal skin of B. marinus and a similiar number in Xenopus laevis. Also, proteoglycans which bind collagen ®brils to long ®bers, the proportions of collagen ®bers of different thickness examined in B. marinus by Craig et al. (1987), and different types of collagen may play an important role. Compared to other anurans, the thickness of the skin in B. marinus, in particular that of the dorsal skin, is enormous. Values measured by us were two or even three times the values obtained in Rana esculenta and X. laevis skins. Thickness values also varied with the body regions (as in other species, the skin is thinner ventrally than dorsally). According to the techniques employed to native skin it appears thinner than ®xed and sectioned skin, but as discussed previously values re¯ect real differences (Greven et al., 1995; Zanger et al., 1995). Furthermore, thickness may be related to body size ± to our knowledge, this has not been investigated in detail as yet ± and sex. Despite the small number of B. marinus specimens investigated in the present study, we do not believe, however, that a sex difference in the skin thickness of this species exists, at least for nonreproductive terrestrial specimens. A different skin thickness has been described, however, in the fully aquatic non-reproductive X. laevis, that has a noticeable in¯uence on the biomechanical properties of the skin in this species. Thickness differences may even be enhanced during reproduction (Greven et al., 1995). R. esculenta males and females have skins of similar thickness. During reproduction the skin of the terrestrial Bufo species becomes slimy under hormonal in¯uence and more hydrated (Greven, 1987). Thus, a variety of exogenous and endogenous
BIOMECHANICS OF BUFO MARINUS SKIN
545
sa
sc
7
8
sc
9
10
11
Fig. 7 Scanninng electron micrograph of collagen fibers in the stratum compactum. Body long axis (white arrow). Angle ca 558, 1800. Fig. 8 Low power transmission electron micrograph of the stratum compactum (sc). Note criss-cross arrangement of collagen fibers, and the calcified substantia amorpha (sa). 900. Fig. 9 Criss-cross arrangement of collagen bundles in the stratum compactum (sc). 30 000. Fig. 10 Transmission electron micrograph of a small elastic fiber (arrows) between collagen fibers in a perforating bundle as visualized with tannic acid. 30 000. Fig. 11 Transmission electron micrograph of the tela subcutanea with numerous elastin fibers (arrows). 9000.
546
SCHWINGER ET AL.
maximal stress curves of three species 20 Bufo marinus Rans esculenta spec. Xenopus laevis
stress MPa
15
10
5
0 0
50
100
12
150
200
250
strain %
Fig. 12 Comparison of typical stress strain curves of Bufo marinus, Xenopus laevis and Rana esculenta (data for X. laevis and R. esculenta from Greven et al., 1995 and Zanger et al., 1995).
parameters may in¯uence anuran skin including biomechanical properties, regardless of its thickness. Very probably, the s. amorpha does not contribute to the tensile strength of the skin, since it consists of an aggregation of small electron-dense granules separated from each other (Toledo & Jared, 1993). X. laevis has only traces of a s. amorpha (unpublished), whereas R. esculenta possesses distinct calci®ed granules. In both, the stress-strain curves are steeper than in B. marinus (Fig. 12). Tensile strength of the skin of X. laevis was almost twice that of B. marinus. The comparison of the extension of the skin due to the applied stress varies from species to species, depending on various factors including the composition of the noncollagenous compound, elastin content and the ®bre weave of the collagenous network (Wainwright et al., 1976). The strain range at which failure occurs is relatively small in X. laevis and R. esculenta and considerably higher in B. marinus. The low slope of the stress-strain curve as seen in B. marinus is indicative of a highly elastic component (Oxlund et al., 1988). Conventional histological stainings such as orcein, resorcin and others revealed numerous elastin ®bers in the dermis (Tonkoff, 1900; Greven et al., 1995; Zanger et al., 1995). However, when examined with the electron microscope, the elastin content was reduced to only a few ®bers (Zanger et al., 1995). The same applies to the skin of B. marinus. Most structures in the dermis stained with orcein do not contain elastin as seen by immunohistochemistry and electron microscopy. The highest amounts of orcein and resorcin-positive ®bers were found in the tela subcutanea of R. esculenta by Tonkoff (1900), and this could be con®rmed in the present paper for the skin of B. marinus. Immunostaining and electron microscopy reveal these structures to be elastin and, thus
emphasize the signi®cance of the tela subcutanea for the biomechanical properties. A further layer that determines the main biomechanical properties in anuran skin is the st. compactum that exhibits a special arrangement of collagenous ®bers and large angles of the folding trellis. Estimation of angles between the collagenous ®bres ranged from 50 to 808; especially the larger angles may also contribute to the elongation properties of the skin. However, at the present time the exact amount of high angle crossing is unknown. We hope to get more insight into the structure and biomechanical parameters of this layer by studies of stretched and unstretched strata and hysteresis experiments. Stress-strain curves exhibit the wide-spread J-slope, with an initial toe and a more linear part ending in a sudden break. Comparing this curve with those of X. laevis and R. esculenta, the slope is considerably lower in B. marinus, regardless the site from which the skin samples were obtained. When considering the individual parameters measured, their values are smaller in B. marinus than those obtained for R. esculenta and X. laevis. The most obvious differences are seen in the values for strain. A crucial point is the value of W, the energy per related area or volume until break (Gordon 1978; Bauer et al., 1989). In the present paper we abstained from determining this parameter on several reasons. First, comparison with the few data in the relevant literature is not useful as authors did not show exactly how the values were calculated. Second, they did not distinguish between native, dried and ®xed skin samples (Bauer et al., 1989). Our own investigations showed that the elastic modulus is very different in such skin samples. Third, the machine we used measured the area under the stress-elongation curve rather than the
BIOMECHANICS OF BUFO MARINUS SKIN
area under the stress-strain curve, as the manufacturer had implied. Therefore, the interpretation of the values of W published in our previous papers is open to doubt. Regarding the complex structure of the anuran skin, we suggest that the E-modulus alone appears to be suf®cient to characterize the biomechanical properties for comparative purposes.
ACKNOWLEDGEMENTS We thank Professor G. Novotny BSc, PhD, University of London, for linguistic advice and Dr J. BuÈth, Altena, for technical supervision. REFERENCES Bauer, A.M., Russell, A.P. and Shadwick, R.E. 1989. Mechanical properties and morphological correlates of fragile skin in gekkonid lizards. J. Exp. Biol., 145, 79±102. Craig, A.S., Eikenberry, E.F. and Parry, D.A.D. 1987. Ultrastructural organization of skin: Classification on the basis of mechanical role. Connect. Tiss. Res., 13, 213±223. Denefle, J.P., Zhu, Q.L. and Lechaire, J.P. 1987. Dermal tracts in frog skin: fibronectin pathways for cell migration. Biol. Cell, 59, 219±226. Denefle, J.P., Zhu, Q.L. and Lechair, J.P. 1993. Localisation of fibronectin in the frog skin. Tissue Cell, 25, 87±102. DIN Deutsches Institut fuÈr Normung e.V. (ed.). 1988. Kunststoffe. Mechanische und thermische Eigenschaften. PruÈfnormen (9. Aufl). Benth Verlag, KoÈln. Elkan, E. 1976. Ground substance: anuran defence against desiccation. In: Lofts, B. (ed) Physiology of the Amphibia, Vol. III. Academic Press, New York, pp. 101±110.
547
Fox, H. 1986. The skin of Amphibia. In: J. Bereiter-Hahn, Matoltsy, A.G. and Richards, K.S. (eds) Biology of the Integument, Vol. 2. Springer, Berlin, Heidelberg, 78±135. Gordon, J.E. 1978. Structures, or why things don't fall down. Penguin Books, Harmondsworth. Greven, H. 1987. Skin texture and specific gravity of three amphibian species during their aquatic and terrestrial phase. In: van Gelder, J.J., Strijbosch, H. and Bergers, P.J.M. (eds) Proc Fourth Ord Gen Meet SEH Nijmegen. Faculty of Sciences, Nijmegen, pp. 163±166. Greven, H., Zanger, K. and Schwinger, G. 1995. Mechanical properties of the skin of Xenopus laevis (Anura, Amphibia). J. Morph., 224, 15±22. Imayama, S. and Braverman, I. 1988. Scanning electron microscope study of elastic fibers of the loose connective tissue (superficial fascia) in the rat. Anat. Rec., 222, 115±120. Junqueira, L.C.V., Bignolas, G. and Brentani, R.R. 1979. Picrosirius staining plus polarization microscopy, a specific method for collagen detection in tissue sections. Histochem. J., 11, 447±455. Mayer, W., Bartels, T. and Neurand, K. 1989. Anmerkungen zur Faserarchitektur der Vertebraten-Dermis. Z. Zool. Syst. Evolut.-Forsch., 27, 115±125. Oxlund, H., Manschott, J. and Viidik, A. 1988. The role of elastin in the mechanical properties of skin. J. Biomechanics, 21, 213±218. Romeis, B. 1989. Mikroskopische Technik. Urban und Schwarzenberg Oldenburg, MuÈnchen. Toledo, R.C. and Jared, C. 1993. The calcified dermal layer in anurans. Comp. Biochem. Physiol., 104A, 443±448. Tonkoff, W. 1900. UÈber die elastischen Fasern in der Froschhaut. Arch. mikrosk. anat. Entwick.-gesch., 57, 95±101. Vincent, J.F.V. 1982. Structural Biomaterials. The Macmillan Press, London. Wainwright, S.A., Biggs, W.D., Currey, J.D. and Gosline, J.M. 1976 Mechanical Design in Organisms. Princeton University Press, Princeton. Zanger, K., Schwinger, G. and Greven, H. 1995. Mechanical properties of the skin of Rana esculenta (Anura, Amphibia) with some notes on structures related to them. Ann. Anat., 177, 509±514.
Tissue&Cell
Structural and mechanical aspects of the skin of Bufo marinus (Anura, Amphibia) G. Schwinger,1 K. Zanger,1 H. Greven2 Abstract. Specific biomechanical characters and some structures possibly related to them were investigated in the skin of the toad Bufo marinus using tensile testing techniques (at constant strain till rupture) as well as morphological methods (histological, immunohistochemical and electronmicroscopical). Mechanical parameters of the native skin varied considerably according to sex, individual variability and/or site of specimen collection. In skin strips of males and females excised from different parts of the body thickness ranged from 0.45 to 0.87 mm, strain (ef) from 96.52 to 211.03, tensile strength (sm) from 5.72 to 9.38 MPa, and stiffness (E-modulus) from 5.76 to 6.73. The dermis of B. marinus is provided with a collagenous stratum compactum of considerable thickness, a stratum spongiosum with loosely arranged fibres and a marked calcified layer (substantia amorpha). Collagen appears to be the main determinant of skin mechanics. However, the slope of the J-shaped static stress-strain curves indicates elastin to be responsible for the high values of strain. Contrary to van Gieson and orcein staining, immunostaining with a monoclonal antibody against elastin revealed very few elastic fibers between collagen bundles and in the vertical fiber tracts (perforating bundles), but a considerable amount in the tela subcutanea. This was partly confirmed at the ultrastructural level by tannic acid staining. ß 2001 Harcourt Publishers Ltd Keywords: biomechanics, tensile test, anuran skin, collagen, elastin
Introduction The anuran skin is a highly composite structure that consists of the squamous epidermis, including basal, spinosous and horny layers and the underlying dermis. The latter contains epidermis-derived alveolar glands and can be differentiated in a stratum spongiosum and a stratum 1 Institut fuÈr Topographische Anatomie und Biomechanik, Medizinische Einrichtungen der Heinrich Heine-UniversitaÈt, UniversitaÈtsstr. 1, D-40225 DuÈsseldorf, Germany, 2Institut fuÈr Zoomorphologie und Zellbiologie der UniversitaÈt, UniversitaÈtsstr. 1, D-40225 DuÈsseldorf, Germany
Received 13 April 2001 Accepted 25 July 2001 Correspondence to: Gerhard Schwinger, Institut fuÈr Topographische Anatomie und Biomechanik, Medizinische Einrichtungen der Heinrich-Heine UniversitaÈt, UniversitaÈtsstr. 1, D-40225 DuÈsseldorf, Germany; Tel.: 49 211 811 2719; Fax: 49 211 811 2806; E-mail: [email protected]
compactum. Both strata are characterized by their collagen bundles, which are loosely arranged in the st. spongiosum and in a criss-cross manner in the st. compactum. In many species a calci®ed substantia amorpha lies between the two dermal layers (for review see Fox 1986). It has been suggested that the st. compactum is the most conservative element in the vertebrate skin (Meyer et al., 1989), and that collagen mainly characterizes the mechanical properties of a skin (Vincent, 1982; Wainwright et al., 1976). As shown by histological methods, electronmicroscopical techniques and typically J-shaped static stress strain curves, the same appeared to hold for the skin of anurans (Greven et al., 1995; Zanger et al., 1995). In these articles, we examine the reasons why anurans living predominantly in an aquatic and terrestrial environment or in both (amphibious species) should have differently structured skins with different mechanical properties. Comparing for instance the mechanical 541
542
SCHWINGER ET AL.
Table 1 Parameters obtained for strips of dorsal, ventral and lateral skin from males and females of Bufo marinus stretched uniaxially along the longitudinal axis. Values are means plus standard deviation; sm maximal stress prior to failure; ef maximal strain at break; E modulus of elasticity Sex
Sample
n
thickness (mm)
Stress sm (MPa)
Strain ef (%lo)
E (MPa)
Male
Ventral Dorsal Lateral
11 20 3
0.50+0.05 0.87+0.13 0.71+0.16
8.70+2.71 6.28+1.88 9.38+5.28
123.80+37.82 106.39+32.54 211.03+18.63
6.65+2.41 6.27+1.21 5.76+2.48
Female
Ventral Dorsal Lateral
10 6 4
0.45+0.04 0.80+0.10 0.54+0.10
7.40+1.31 5.72+2.29 7.39+2.24
96.52+18.21 114.60+51.73 204.54+31.96
6.73+1.89 6.66+1.80 6.46+0.85
parameters of the skin of the fully aquatic Xenopus laevis (Greven et al., 1995) and the semiaquatic Rana esculenta (Zanger et al., 1995), most of the values, in particular the value for extension load, were lower in R. esculenta. In the present paper, we report on the skin of the terrestrial toad Bufo marinus, to cover the whole spectrum of the habitats used by anurans, one of the most widespread and common American anuran species. We determined the same mechanical and structural parameters as in the two previous papers with the exception of toughness to facilitate comparison.
Materials and methods Animals We used 4 adult males and 2 females of the toad Bufo marinus outside the breeding season bought from a commercial dealer. Biomechanics Preparation of the samples and the tensiometer used were the same as described previously (Greven et al., 1995; Zanger et al., 1995). Light microscopy Strips were ®xed in 4% formaldehyde or absolute ethanol and embedded in paraplast. 7 mm-sections were stained either with Picro Sirius Red (Junqueira et al., 1979) and examined in a Leitz polarizing microscope or stained according to van Gieson and with resorcin to visualize elastin ®bres (Romeis, 1989). Resorcin staining was modi®ed and counterstained with Kernechtrot (Novotny, personal communication). Orcein staining was also carried out. For photographic documentation, collagenous ®bres were stained with Goldner's trichrome. In order to demonstrate the substantia amorpha, a calci®ed layer in the dermis of many anurans (e.g. Elkan, 1976; Toledo & Jared, 1993), sections were treated by the van Kossa method (ethanol ®xation). For immunohistochemistry, native cryostate sections were treated with 0.1% trypsin (Sigma Chemicals Co.) in PBS for 20 min at 378C. Endogenous peroxidase was blocked with fresh 3% hydrogen peroxide in isopropanol for 30 min. Before incubation sections in aqua dest. were
placed for 15 min in a microwave oven (850 W). To reduce non speci®c staining, they were incubated for 60 min at room temperature with normal rabbit non immune serum (1:100 dilution; Dako 049A), washed in PBS, and then incubated for 2 h with the primary (mouse) antibody (1:100 dilution; monoclonal anti-Elastin; Sigma E-4013), washed again and incubated with HRPconjugated secondary (goat anti-mouse IgG) antibody (1:100 dilution; Sigma A-9917) for 2 h. The sections were then washed in PBS and TRIS and reacted with 3,3'-diaminobenzidine (freshly prepared in 0.05 mol/l TRIS buffer, pH 7.6, containing 0.015 mol/l hydrogen peroxide). Pieces of a rabbit arteriole were used as a positive control. Electron microscopy For SEM, the stratum compactum was mechanically separated using ®ne forceps, and using the same ®xation as for TEM. Then the specimen were dehydrated and critical point dried. Preparations were sputtered with gold and examined in a Jeol JSM 35 CF scanning electron microscope. Collagen ®ber angles within the stratum compactum were measured in low-power SEM-micrographs. For TEM, some samples of skin were ®xed in 2.5% glutaraldehyd in 0.1 mol/l cacodylate buffer, pH 7.2, with and without 1% tannic acid, and post®xed with 1% osmium tetroxide (Imayama & Braverman, 1988). Ultrathin sections were stained with lead citrate and uranyl acetate and examined in a Zeiss EM 9 transmission electron microscope.
Results The thickness of the skin is similar in both sexes in the different body regions examined (Table 1). Measurements with the vernier caliper revealed a thickness of 0.45±0.87 mm, whereas measurement in the histological sections gave values of 0.48±1.34 mm. The dorsal skin is thicker than the ventral, involving both epidermis and dermis. Signi®cant differences between sexes could not be detected in the small samples available. The thickest layer with 0.45±1.28 mm dorsally and 0.51±0.72 mm ventrally is the dermis. It includes dermal glands (derivatives of the epidermis; Figs 1, 3, 5), blood
BIOMECHANICS OF BUFO MARINUS SKIN
ep
ep mg ss
ss
gg sc
1
2
* sc
3
4
ss
sc
5
6
543
544
SCHWINGER ET AL.
Fig. 1 Slightly oblique paraplast section of the dorsal skin stained with trichrome Goldner; ep, epidermis; gg, granular glands; mg, mucous glands; sc,
stratum compactum; ss, stratum spongiosum.450. Fig. 2 Paraplast section of the ventral skin stained with orcein. Note orcein-positive material (arrows) attached to a mucous gland; ep, epidermis; ss, stratum spongiosum.1000. Fig. 3 Paraplast section of the dorsal skin stained according van Kossa's method. Note the prominent calcified substantia amorpha (asterisk) immediately above the stratum compactum (sc).450. Fig. 4 Paraplast section of the ventral skin stained with resorcin and counterstained with Kernechtrot. Resorcin-positive fibers (arrows) are seen in the connective tissue (stratum spongiosum) between adjacent glands.900. Fig. 5 Polarized light micrograph of the dorsal skin stained with Sirius red. Note different course of the abundant collagen fiber bundles in the stratum spongiosum (ss) and st. compactum (sc).450. Fig. 6 Immunohistochemical demonstration of elastin fibers (arrows) in a perforating bundle of the ventral skin (without counterstain).800.
vessels, smooth muscles, chromatophores, nerves, (not shown), and abundant collagen ®bers (Figs 7±9). It is clearly differentiated in an upper stratum spongiosum with loosely arranged ®bers, the stratum compactum with the typical criss-cross pattern of ®bers (Figs 1, 5, 7, 8±10). Between these both strata, the calci®ed substantia amorpha is visible (Figs 1, 3, 8). Within the dermis, several different structures surrounding the glands (Fig. 2), ®bers in the tela subcutanea (not shown) as well as ®bers within the bundles of connective tissue perforating the st. compactum could be stained with resorcin and Kernechtrot. Fibers in the stratum spongiosum between the glands (Fig. 4) as well as ®bers in the tela subcutanea (not shown, compare Fig. 11), and ®bres within the bundles of connective tissue perforating the stratum compactum (not shown, compare Fig. 6) were also stained. Sections stained for immunohistochemistry showed a strong reaction in the tela subcutanea and sporadically a positive reaction in the perforating bundles (Fig. 6). Ultrathin sections of tannin treated samples revealed typical elastic ®bers that are composed of amorphous elastin with a peripheral micro®brillar protein sheath (Fig. 10). In conventional ultrathin sections elastic ®bers appeared as electron-dense ®brous structures (Fig. 11). Dermal components surrounding the glands and stained with orcein or resorcin, did not reveal elastin structure in ultrathin sections. Collagen ®bers were the conspicous components of the st. spongiosum and the st. compactum. They were arranged more or less disordered in the stratum spongiosum and show a plywood pattern in the stratum compactum. These patterns could be shown by polarizing microscopy of Sirius Red stained sections (Fig. 5) and were con®rmed by TEM (Figs 8, 9). The thickness of the st. compactum ranged from about 233 to 720 mm. The angles, in which collagen ®bers were orientated varied from 508 to 808 in both sexes (Fig. 7). The substantia amorpha was localized between the two dermal layers. It contained a great number of small, irregular electron-dense granules (Fig. 8). The st. compactum is perforated by vertical bundles of collagen, which number 3 to 7 per mm of skin. The static stress-strain curves were J-shaped, consisting of an initial toe part and a linear portion (Fig. 12). No differences between males and females were detected. The biomechanical data measured are summarized in Table 1. Maximal tensile strength (sm), maximal strain at break (ef), and Young's modulus E varied considerably, as re¯ected by the high standard deviations.
Discussion Bufo marinus has the typical anuran skin, i.e. a squamous epidermis, a dermis differentiated in a stratum spongiosum (with alveolar glands) and a stratum compactum and, not present in all anuran skins (see Elkan, 1957; Toledo & Jared, 1993), a prominent calci®ed substantia amorpha. The st. compactum as well as the s. amorpha are perforated by vertical bundles of collagen (see among others Fox, 1986; Dene¯e et al., 1987, 1993). The angle of the collagen ®bres in the st. compactum is in the same order as in other anurans. Due to its composite nature, it is very dif®cult to associate a particular structure with particular biomechanical properties of the skin. These also holds for the collagenous bundles that perforate the s. amorpha and st. compactum (Greven et al., 1995; Zanger et al., 1995). A rough estimation revealed 3 to 7 bundles per mm in the dorsal skin of B. marinus and a similiar number in Xenopus laevis. Also, proteoglycans which bind collagen ®brils to long ®bers, the proportions of collagen ®bers of different thickness examined in B. marinus by Craig et al. (1987), and different types of collagen may play an important role. Compared to other anurans, the thickness of the skin in B. marinus, in particular that of the dorsal skin, is enormous. Values measured by us were two or even three times the values obtained in Rana esculenta and X. laevis skins. Thickness values also varied with the body regions (as in other species, the skin is thinner ventrally than dorsally). According to the techniques employed to native skin it appears thinner than ®xed and sectioned skin, but as discussed previously values re¯ect real differences (Greven et al., 1995; Zanger et al., 1995). Furthermore, thickness may be related to body size ± to our knowledge, this has not been investigated in detail as yet ± and sex. Despite the small number of B. marinus specimens investigated in the present study, we do not believe, however, that a sex difference in the skin thickness of this species exists, at least for nonreproductive terrestrial specimens. A different skin thickness has been described, however, in the fully aquatic non-reproductive X. laevis, that has a noticeable in¯uence on the biomechanical properties of the skin in this species. Thickness differences may even be enhanced during reproduction (Greven et al., 1995). R. esculenta males and females have skins of similar thickness. During reproduction the skin of the terrestrial Bufo species becomes slimy under hormonal in¯uence and more hydrated (Greven, 1987). Thus, a variety of exogenous and endogenous
BIOMECHANICS OF BUFO MARINUS SKIN
545
sa
sc
7
8
sc
9
10
11
Fig. 7 Scanninng electron micrograph of collagen fibers in the stratum compactum. Body long axis (white arrow). Angle ca 558, 1800. Fig. 8 Low power transmission electron micrograph of the stratum compactum (sc). Note criss-cross arrangement of collagen fibers, and the calcified substantia amorpha (sa). 900. Fig. 9 Criss-cross arrangement of collagen bundles in the stratum compactum (sc). 30 000. Fig. 10 Transmission electron micrograph of a small elastic fiber (arrows) between collagen fibers in a perforating bundle as visualized with tannic acid. 30 000. Fig. 11 Transmission electron micrograph of the tela subcutanea with numerous elastin fibers (arrows). 9000.
546
SCHWINGER ET AL.
maximal stress curves of three species 20 Bufo marinus Rans esculenta spec. Xenopus laevis
stress MPa
15
10
5
0 0
50
100
12
150
200
250
strain %
Fig. 12 Comparison of typical stress strain curves of Bufo marinus, Xenopus laevis and Rana esculenta (data for X. laevis and R. esculenta from Greven et al., 1995 and Zanger et al., 1995).
parameters may in¯uence anuran skin including biomechanical properties, regardless of its thickness. Very probably, the s. amorpha does not contribute to the tensile strength of the skin, since it consists of an aggregation of small electron-dense granules separated from each other (Toledo & Jared, 1993). X. laevis has only traces of a s. amorpha (unpublished), whereas R. esculenta possesses distinct calci®ed granules. In both, the stress-strain curves are steeper than in B. marinus (Fig. 12). Tensile strength of the skin of X. laevis was almost twice that of B. marinus. The comparison of the extension of the skin due to the applied stress varies from species to species, depending on various factors including the composition of the noncollagenous compound, elastin content and the ®bre weave of the collagenous network (Wainwright et al., 1976). The strain range at which failure occurs is relatively small in X. laevis and R. esculenta and considerably higher in B. marinus. The low slope of the stress-strain curve as seen in B. marinus is indicative of a highly elastic component (Oxlund et al., 1988). Conventional histological stainings such as orcein, resorcin and others revealed numerous elastin ®bers in the dermis (Tonkoff, 1900; Greven et al., 1995; Zanger et al., 1995). However, when examined with the electron microscope, the elastin content was reduced to only a few ®bers (Zanger et al., 1995). The same applies to the skin of B. marinus. Most structures in the dermis stained with orcein do not contain elastin as seen by immunohistochemistry and electron microscopy. The highest amounts of orcein and resorcin-positive ®bers were found in the tela subcutanea of R. esculenta by Tonkoff (1900), and this could be con®rmed in the present paper for the skin of B. marinus. Immunostaining and electron microscopy reveal these structures to be elastin and, thus
emphasize the signi®cance of the tela subcutanea for the biomechanical properties. A further layer that determines the main biomechanical properties in anuran skin is the st. compactum that exhibits a special arrangement of collagenous ®bers and large angles of the folding trellis. Estimation of angles between the collagenous ®bres ranged from 50 to 808; especially the larger angles may also contribute to the elongation properties of the skin. However, at the present time the exact amount of high angle crossing is unknown. We hope to get more insight into the structure and biomechanical parameters of this layer by studies of stretched and unstretched strata and hysteresis experiments. Stress-strain curves exhibit the wide-spread J-slope, with an initial toe and a more linear part ending in a sudden break. Comparing this curve with those of X. laevis and R. esculenta, the slope is considerably lower in B. marinus, regardless the site from which the skin samples were obtained. When considering the individual parameters measured, their values are smaller in B. marinus than those obtained for R. esculenta and X. laevis. The most obvious differences are seen in the values for strain. A crucial point is the value of W, the energy per related area or volume until break (Gordon 1978; Bauer et al., 1989). In the present paper we abstained from determining this parameter on several reasons. First, comparison with the few data in the relevant literature is not useful as authors did not show exactly how the values were calculated. Second, they did not distinguish between native, dried and ®xed skin samples (Bauer et al., 1989). Our own investigations showed that the elastic modulus is very different in such skin samples. Third, the machine we used measured the area under the stress-elongation curve rather than the
BIOMECHANICS OF BUFO MARINUS SKIN
area under the stress-strain curve, as the manufacturer had implied. Therefore, the interpretation of the values of W published in our previous papers is open to doubt. Regarding the complex structure of the anuran skin, we suggest that the E-modulus alone appears to be suf®cient to characterize the biomechanical properties for comparative purposes.
ACKNOWLEDGEMENTS We thank Professor G. Novotny BSc, PhD, University of London, for linguistic advice and Dr J. BuÈth, Altena, for technical supervision. REFERENCES Bauer, A.M., Russell, A.P. and Shadwick, R.E. 1989. Mechanical properties and morphological correlates of fragile skin in gekkonid lizards. J. Exp. Biol., 145, 79±102. Craig, A.S., Eikenberry, E.F. and Parry, D.A.D. 1987. Ultrastructural organization of skin: Classification on the basis of mechanical role. Connect. Tiss. Res., 13, 213±223. Denefle, J.P., Zhu, Q.L. and Lechaire, J.P. 1987. Dermal tracts in frog skin: fibronectin pathways for cell migration. Biol. Cell, 59, 219±226. Denefle, J.P., Zhu, Q.L. and Lechair, J.P. 1993. Localisation of fibronectin in the frog skin. Tissue Cell, 25, 87±102. DIN Deutsches Institut fuÈr Normung e.V. (ed.). 1988. Kunststoffe. Mechanische und thermische Eigenschaften. PruÈfnormen (9. Aufl). Benth Verlag, KoÈln. Elkan, E. 1976. Ground substance: anuran defence against desiccation. In: Lofts, B. (ed) Physiology of the Amphibia, Vol. III. Academic Press, New York, pp. 101±110.
547
Fox, H. 1986. The skin of Amphibia. In: J. Bereiter-Hahn, Matoltsy, A.G. and Richards, K.S. (eds) Biology of the Integument, Vol. 2. Springer, Berlin, Heidelberg, 78±135. Gordon, J.E. 1978. Structures, or why things don't fall down. Penguin Books, Harmondsworth. Greven, H. 1987. Skin texture and specific gravity of three amphibian species during their aquatic and terrestrial phase. In: van Gelder, J.J., Strijbosch, H. and Bergers, P.J.M. (eds) Proc Fourth Ord Gen Meet SEH Nijmegen. Faculty of Sciences, Nijmegen, pp. 163±166. Greven, H., Zanger, K. and Schwinger, G. 1995. Mechanical properties of the skin of Xenopus laevis (Anura, Amphibia). J. Morph., 224, 15±22. Imayama, S. and Braverman, I. 1988. Scanning electron microscope study of elastic fibers of the loose connective tissue (superficial fascia) in the rat. Anat. Rec., 222, 115±120. Junqueira, L.C.V., Bignolas, G. and Brentani, R.R. 1979. Picrosirius staining plus polarization microscopy, a specific method for collagen detection in tissue sections. Histochem. J., 11, 447±455. Mayer, W., Bartels, T. and Neurand, K. 1989. Anmerkungen zur Faserarchitektur der Vertebraten-Dermis. Z. Zool. Syst. Evolut.-Forsch., 27, 115±125. Oxlund, H., Manschott, J. and Viidik, A. 1988. The role of elastin in the mechanical properties of skin. J. Biomechanics, 21, 213±218. Romeis, B. 1989. Mikroskopische Technik. Urban und Schwarzenberg Oldenburg, MuÈnchen. Toledo, R.C. and Jared, C. 1993. The calcified dermal layer in anurans. Comp. Biochem. Physiol., 104A, 443±448. Tonkoff, W. 1900. UÈber die elastischen Fasern in der Froschhaut. Arch. mikrosk. anat. Entwick.-gesch., 57, 95±101. Vincent, J.F.V. 1982. Structural Biomaterials. The Macmillan Press, London. Wainwright, S.A., Biggs, W.D., Currey, J.D. and Gosline, J.M. 1976 Mechanical Design in Organisms. Princeton University Press, Princeton. Zanger, K., Schwinger, G. and Greven, H. 1995. Mechanical properties of the skin of Rana esculenta (Anura, Amphibia) with some notes on structures related to them. Ann. Anat., 177, 509±514.