Collagenous ligaments connect the teeth of sea urchins to their jaws. These ligaments are faced with contradictory demands: they have to be stiff to hold the teeth firmly when the sea urchin scrapes rocks and hard surfaces during feeding, but they have to be soft enough to allow shift of the teeth towards the outside as new tooth material is added to the inner (aboral) growth zone to compensate for wear. A similar dilemma is found in rodent teeth. Our morphological studies showed, however, that the collagen remodeling mechanism that solves the problem in rodents (Beertsen and Everts, 1977) is not likely to be the answer for echinoids (Birenheide and Motokawa, 1996). Instead, a possible solution lies in the use of catch connective tissue. This is a connective tissue peculiar to echinoderms and unique in its ability to change its mechanical properties under nervous control (Motokawa, 1984a; Wilkie, 1984). In this study we investigated the mechanical properties of echinoid tooth ligament and their possible nervous control. We present data supporting the idea, that sea urchins employ catch connective tissue to solve their dilemma. This is the first physiological study ever conducted on sea urchin tooth ligaments.
Materials and Methods
Specimens of the sea urchin Diadema setosum Leske were obtained near Misaki Marine Biological Station (University of Tokyo), Sagami Bay, Japan. They were kept in aquaria with circulating artificial seawater for no more than 6 months. Lanterns were dissected and divided into five preparations, each consisting of a jaw with a tooth. Experiments were conducted in artificial seawater (ASW), ASW with elevated potassium concentration (KASW), and ASW containing acetylcholine (ACh). The composition of ASW was NaCl 433.7 mM; KCI 10.0 mM; Ca[Cl.sub.2] 10.1 mM; Mg[Cl.sub.2] 52.5 mM; NaHC[O.sub.3] 2.5 mM (pH 8.0). The composition of KASW was NaCl 343.7 mM; KCl 100.0 mM; Ca[Cl.sub.2] 10.1 mM; Mg[Cl.sub.2] 52.5 mM; NaHC[O.sub.3] 2.5 mM(pH 8.0). Acetylcholine (ACh) was added to ASW to make a final concentration of [10.sup.-3] or [10.sup.-4] M. All experiments were performed at room temperature. The chemicals were washed out by thorough rinse with several changes of ASW. The mechanical properties of fresh ligaments were compared with those of frozen control ligaments. Frozen ligaments had been kept for more than 1 week at -20 [degrees] C, then rethawed in ASW for several hours at room temperature. For mechanical testing, a jaw was glued with cyanoacrylate glue to a platform in a trough containing ASW.
A load of 30 g was applied to the tooth by means of a lever. In pull tests the tooth was pulled in the same direction as it naturally shifts during growth. In these tests the lever was connected to the tooth with a silver wire. In push tests the tooth was pushed in the opposite direction as in pull tests, and the lever was connected with a plastic holder. Displacement of the tooth was detected by a linear eddy current sensor (502-F, EMIC Corp., Tokyo, Japan) and recorded with a microcomputer.
Teeth that had been pulled out of the jaw in mechanical tests, and the corresponding jaws, were examined by scanning electron microscopy. Specimens were fixed in 3.5% glutaraldehyde in 0.1 M cacodylate buffer, postfixed with 1% osmium tetroxide in the same buffer, and dehydrated in an acetone series. After an intermediate step of amyl acetate, they were dried in a critical point drying apparatus and mounted on holders. The specimens were then coated with gold and observed in a JEOL JSM-T220 scanning electron microscope.
The tooth ligament of Diadema setosum is morphologically similar to the ligament of Eucidaris tribuloides described in the preceding paper (Birenheide and Motokawa, 1996). The main difference is in the size of the tooth ligament. In D. setosum the tooth ligament is shorter in relation to the whole length of the jaw. The tooth coelom on the abradial side extends along both sides of the U-shaped tooth and narrows the tooth ligament [ILLUSTRATION FOR FIGURE 1 OMITTED]. The length of the dental slide varied with size of the lantern, although we could not establish a clear allometric relationship. Average length was 11.7 mm (SD = 1.83, n = 104). Thickness and width of tooth ligaments were measured in semithin sections of different samples from average-sized lanterns. Average thickness was 24.8 [[micro]meter] (SD = 11.1, n = 13) with the adradial parts being thinner than the abradial parts. Width showed a rather constant value of 600 [[micro]meter].
As in E. tribuloides, the collagen fibril bundles made up a rough crosstexture. The angles between the fibril bundles and the longitudinal axis of the tooth were 20 [degrees] to 40 [degrees]. If we consider a minimal angle of 20 [degrees] and a tooth ligament thickness of 25 [[micro]meter], the maximal length of a collagen fibril bundle would be no more than about 150 [[micro]meter].
Creep tests in artificial seawater
When a load was applied in pull tests, the tooth slowly shifted until it was pulled out of the jaw. The time span until the tooth was extracted varied between 25 min and 67 h. In all cases the ligament finally failed and the tooth was pulled out. The surface of both tooth and jaw after failure carried a great number of collagen fibril bundles clasping either to the pillar bridges of the tooth or to the trabeculae of the jaw [ILLUSTRATION FOR FIGURE 2 OMITTED!. Most fibrils were oriented parallel to the longitudinal axis of the tooth. At failure of the ligament, the tooth had shifted for 550 [[micro]meter] (SD 230, n = 6), corresponding to about 4.7% of the length of the dental slide.
The creep curves were characterized by three phases [ILLUSTRATION FOR FIGURE 3A OMITTED!: a short initial phase with a high creep rate, a second phase with a slow, constant creep rate, and a third phase with a creep rate that gradually increased until the ligament failed. The average creep rate in the constant second phase was 6.8 [[micro]meter]/h (SD 6.1, n = 36).
Relative viscosity was defined as the reciprocal of creep rate, with the value just before application of a test solution taken as unity.
Controls were push tests and tests with frozen jaws. In push tests we observed no significant differences from pull tests. The creep curves were similar [ILLUSTRATION FOR FIGURE 3A OMITTED]. Experimental time until the tooth was pushed out was 17 to 97 h. The tooth had shifted 460 [[micro]meter] (SD 170, n = 6) until failure. The creep rates during the second phase were of the same order as in pull tests.
Freezing and rethawing of the preparations did not change the shape of the creep curves or the creep rates. Shift until failure in pull tests was 820 [[micro]meter] (SD 410, n = 4) and in push tests was 430 [[micro]meter] (SD 160, n = 7).
Responses to stimuli
Acetylcholine (ACh) caused a decrease in creep rate [ILLUSTRATION FOR FIGURE 4A OMITTED!; i.e., the ligament stiffened. In 17 out of 23 tests, relative viscosity doubled in 15 min [ILLUSTRATION FOR FIGURE 6 OMITTED]. The effect was reversible: the creep rate recovered in about 1 h after washing with artificial seawater. Treatment with artificial seawater containing 100 mM [K.sup.+] (KASW) also stiffened the ligament [ILLUSTRATION FOR FIGURE 5A OMITTED!. Relative viscosity doubled in 15 minutes in 6 out of 8 tests. This effect, too, was reversible.
Push tests showed no difference in response to stimuli. ACh caused stiffening of the ligament. In two of three cases, relative viscosity increased more than 10-fold [ILLUSTRATION FOR FIGURE 6 OMITTED]. KASW caused increase of relative viscosity in all three cases - in two of them more than sixfold [ILLUSTRATION FOR FIGURE 6 OMITTED].
In extreme cases stimulation stopped movement of the tooth; i.e., the ligament did not creep under the load of 30 g. This reaction was observed in four cases after application of ACh in pull tests and in two cases after application of KASW, once in pull tests and once in push tests.
In frozen ligaments, ACh had no effect on the creep rate in the first 15 min after stimulation [ILLUSTRATION FOR FIGURE 4B OMITTED]. Creep rate either did not change or increased slightly [ILLUSTRATION FOR FIGURE 6 OMITTED!. The increase is not due to ACh because in all experiments ACh was applied in the third phase (accelerating creep rate) in ASW and because the increase was not reversible when ACh was washed out by ASW. Although the effect was not obvious in 15 rain, KASW caused an increase in creep rate [ILLUSTRATION FOR FIGURE 5B OMITTED!. Creep rate doubled; i.e., relative viscosity halved within 1 h after application of KASW. In some cases the viscosity dropped considerably, even in the first 15 min after the stimulus [ILLUSTRATION FOR FIGURE 6 OMITTED!. The creep rate recovered after washing with artificial seawater of normal potassium concentration. This reversible softening of the ligament was observed in almost all frozen preparations regardless of whether pull tests or push tests were used.
The creep curve allows some assumptions about the general mechanical properties of the tooth ligament. Generally, the similarity of creep curves in push and pull tests in living and frozen samples indicates that the basic properties of the tooth ligament do not depend on the direction of load application or on the existence of living cells. The high creep rate at the beginning indicates that the ligament contains elements with low viscosity that relax quickly after a load is applied. The constant creep rate in the second phase is caused by elements with a comparatively high viscosity that relax slowly. The gradually increasing creep rate in the third phase might be explained by the decreasing area of contact between tooth and jaw due to shift of the tooth. However, the tooth shifted on average only 4.7% at failure; the high increase in creep rate can not be explained by the shift of the tooth alone. It is probable that elements of the ligament lose contact after shifting for a comparatively short distance. Loss of contact results in loss of friction and accelerating creep rate.
The tooth is connected to the jaw by collagen fibrils. The average shift of the tooth (460 [[micro]meter]) is more than the calculated maximal length of the collagen fibril bundles (about 150 [[micro]meter]). Since collagen fibrils cannot elongate more than 10% without rupture (Mason, 1964), the shift of the tooth cannot be explained by strain of the collagen fibrils. Rather, we think that the collagen fibrils or the fibril bundles slowly slide past each other while the tooth is shifting. This conclusion is supported by the observation that collagen is left on both tooth and jaw after the tooth is pulled out. The collagen fibrils probably have no stable bonding, as indicated by our finding that the tooth was extracted in all experiments. The sliding collagen fibrils would lose contact with each other when the tooth has shifted for some distance. This loss of contact results in an accelerating creep rate, as described above.
The constant creep rate in the second phase allows calculation of average shear viscosity of the ligament. With an average creep rate of 6.8 [[micro]meter]/h and average ligament dimensions of 11.7 mm x 600 [[micro]meter] x 24.8 [[micro]meter] (length x width x thickness) the shear viscosity is calculated to be about 550 MPa [center dot] s. The dermis of the sea cucumber Actinopyga echinites has a tensile viscosity of 100 MPa [center dot] s (Motokawa, 1984b). For viscoelastic polymers, tensile viscosity is three times higher than shear viscosity. A shear viscosity of 550 MPa [center dot] s in the tooth ligament thus corresponds to a tensile viscosity of 1650 MPa [center dot] s, which is one order of magnitude higher than the viscosity of sea cucumber dermis. This finding is reasonable because the tooth ligament of sea urchins must provide a firm connection between tooth and jaw, and thus it may well be much stiffer than sea cucumber dermis.
Variability of mechanical properties
The growing end of the sea urchin tooth is covered by an epithelium that contains some muscular elements. The muscle cells are embedded in the epithelium and arranged parallel to the length axis of the tooth (Birenheide and Motokawa, 1996). Contraction of these cells might push the tooth outward and thus increase the creep rate in pull tests and decrease the rate in push tests. Our results, however, showed that KASW or ACh decreased the creep rate in both tests. Thus the observed effects of ACh and KASW cannot be explained by contraction of the epithelial muscles. The ligament contains only a few cell processes that probably cannot produce any force (Birenheide and Motokawa, 1996). We thus conclude that the observed change in mechanical properties depends solely on changes in the mechanical properties of the extracellular matrix of the tooth ligament.
Acetylcholine (ACh) is a ubiquitous neurotransmitter in echinoderms (Cobb, 1987). Its stiffening effect on the tooth ligament strongly suggests that the mechanical properties are under nervous control. Freezing and re-thawing of samples presumably destroys all living cells in the ligament. Lack of any reaction to ACh in frozen ligaments indicates that the stiffening reaction depends on the existence of living cells and is not just a chemical reaction of the extracellular matrix to ACh. Seawater with high potassium content (KASW) also had a stiffening effect only on living ligaments. KASW is a potential stimulator of living cells by depolarizing the cell membrane; here it probably exerted its effect on cellular elements of the tooth ligament. As we have demonstrated (Birenheide and Motokawa, 1996), cells that send processes into the tooth ligament are in contact with a jaw nerve. We think that ACh and KASW influence these cells or the axons of the nerve.
The stiffening effect of KASW on intact ligaments was apparent in 15 min. In frozen ligaments, KASW had a much slower and opposite effect: it accelerated the creep rate within 1 h. Because living cells are presumably destroyed in frozen ligaments, the reaction is probably due to a direct effect of potassium ions on the extracellular matrix. It has been suggested that divalent calcium ions provide crosslinks between proteoglycans of the extracellular matrix, thus stiffening the tissue (Motokawa, 1984a). If calcium ions are replaced by monovalent potassium ions, crosslinking between proteoglycans would be weakened and the tissue would soften.
The effects of ACh and KASW on living and frozen ligaments are strikingly similar to those on the dermis of sea cucumbers (Motokawa, 1994). Living dermis stiffens and Triton-extracted dermis softens after application of KASW. ACh causes stiffening of living dermis, but has no effect on Triton-extracted dermis. Sea cucumber dermis is a composite material consisting of an extracellular matrix with interspersed collagen fibrils. Cellular elements spread throughout the extracellular material (Motokawa, 1982; Matsuno and Motokawa, 1992). In mechanical tests, sea cucumber dermis was shown to be a typical catch connective tissue (CCT) (Motokawa, 1981). CCT is unique to echinoderms and can change its mechanical properties under nervous control (Motokawa, 1984a; Wilkie, 1984). During the stiff state, weak bonds are thought to connect the collagen fibrils and thus prevent them from slipping; proteoglycans are probably involved in the bonding. By some unknown mechanism, nervous signals would induce loosening of the bonds and thus allow slippage of the collagen fibrils. Our study indicates that the tooth ligament is another example of CCT. Variability of mechanical properties is further supported by the high variation in experimental time until failure (25 min to 67 h). Probably our samples were in different states of catch at the beginning of the experiments.
In a preliminary study, Birenheide (1990) showed that papain loosens the tooth ligament dramatically. Papain degrades proteoglycans only, without affecting collagen (Junqueira et al., 1980; Hardingham, 1981). This finding demonstrates convincingly that proteoglycans are responsible for the stiffness of the tooth ligament.
Driving force of tooth shift during natural growth
The presence of muscles in the coelomic epithelium of the plumula seems to have no effect on the mechanical properties of the tooth ligament. If the muscles were connected to the tooth by means of basement membrane and connective tissue, their contraction might provide the force that drives the tooth during growth. This idea is unlikely, however, because the coelomic epithelium is too delicate to provide a solid structure for anchoring the muscles. The tooth itself is pliant in the plumula region and would not transmit forces effectively. We thus do not think that the epithelial muscles of the plumula are the source of the driving force for the tooth in growth. Among other possibilities is a force generated by the newly synthesized tooth elements in the plumula or by the tooth ligament. As in the case of the epithelial muscles, a force generated in the plumula would lack a solid structure as counterpart, but a force generated by the tooth ligament is a probable candidate. Such a force has been postulated for the periodontal ligament of rat incisors (Kirkham et al., 1993). Although there are no data about driving forces for tooth growth in sea urchins, the tooth ligament should be considered as a possible source.
Catch mechanism in tooth ligament
Our data indicate that the dilemma of the echinoid tooth ligament - to be stiff for firm connection of the tooth or to be pliant to allow growth of the tooth - is solved by using catch connective tissue. To summarize our findings, we propose the following model. During feeding the tooth ligament is in a stiffened state. Proteoglycans connect stationary collagen on the jaw side and movable collagen on the tooth side. The connection prevents collagen fibrils from slipping. In this state the tooth is firmly fixed to the jaw and the ligament withstands strong forces when the tooth scrapes over rocks. When the sea urchin is not feeding, a nervous signal causes the connections between collagen fibrils and proteoglycans to loosen. The collagen fibrils can slip past each other and allow slow shift of the tooth during growth. When the sea urchin starts feeding, another nervous signal stiffens the ligament again.
The sea urchin tooth ligament is a convincing example of the use of catch connective tissue to connect skeletal ossicles that need firm connection as well as movability.
Mr. Ohki, Tokyo Inst. Technology, Research Association Dept., Electron Microscopic Laboratory, skillfully operated the scanning electron microscope. Supported by a research grant from the Ministry of Education, Science and Culture of Japan (no. 07640897) to T.M.
Beertsen, W., and V. Everts. 1977. The site of remodeling of collagen in the periodontal ligament of the mouse incisor. Anat. Rec. 189: 479-498.
Birenheide, R. 1990. Functional analysis of the tooth support in echinoids. Pp. 203-206 in Echinoderm Research. C. De Ridder, P. Dubois, M. C. Lahaye, and M. Jangoux, eds. Balkema, Rotterdam.
Birenheide, R., and T. Motokawa. 1996. To be stiff or to be soft - the dilemma of echinoid tooth ligament. I. Morphology. Biol. Bull. 190: 218-230.
Cobb, J. L. S. 1987. Neurobiology of the Echinodermata. Pp. 483527 in Nervous Systems in Invertebrates, NATO ASI Ser. A, M. Ali, ed. Plenum Press, New York.
Hardingham, T. E. 1981. Proteoglycans: their structure, interactions and molecular organization in cartilage. Biochem. Soc. Trans. 9: 489-497.
Junqueira, L. C. U., G. S. Montes, P. A. S. Mourao, J. Carneiro, L. M. M. Salles, and S. S. Bonett. 1980. Collagen-proteoglycan interaction during autotomy in the sea-cucumber Stichopus badionotus. Rev. Can. Biol. 39: 157-164.
Kirkham, J., C. Robinson, J. K. Phull, R. C. Shore, B. J. Moxham, and B. K. B. Berkovitz. 1993. The effect of rate of eruption on periodontal ligament glycosaminoglycan content and enamel formation in the rat incisor. Cell Tissue Res. 274:413-419.
Mason, P. 1964. The viscoelasticity and structure of keratin and collagen. Kolloid-Z. Z. Polym. 202: 139-147.
Matsuno, A., and T. Motokawa. 1992. Evidence for calcium translocation in catch connective tissue of the sea cucumber Stichopus chloronotus. Cell Tissue Res. 267: 307-312.
Motokawa, T. 1981. Stiffness change of the holothurian dermis caused by chemical and electrical stimulation. Comp. Biochem. Physiol. 70C: 41-48.
Motokawa, T. 1982. Fine structure of the dermis of the body wall of the sea cucumber, Stichopus chloronotus, a connective tissue which changes its mechanical properties. Galaxea 1: 55-64.
Motokawa, T. 1984a. Connective tissue catch in echinoderms. Biol. Rev. 59: 255-270.
Motokawa, T. 1984b. Viscoelasticity of holothurian body wall. J. Exp. Biol. 109: 63-75.
Motokawa, T. 1994. Effects of ionic environment on viscosity of Triton-extracted catch connective tissue of a sea cucumber body wall. Comp. Biochem. Physiol. 109B: 613-622.
Wilkie, I. C. 1984. Variable tensility in echinoderm collagenous tissues: a review. Mar. Behav. Physiol. 11: 1-34.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||To be Stiff or to be Soft - the Dilemma of the Echinoid Tooth Ligament, part II|
|Author:||Birenheide, Rudiger; Tsuchi, Akifumi; Motokawa, Tatsuo|
|Publication:||The Biological Bulletin|
|Date:||Apr 1, 1996|
|Next Article:||Mutable collagenous structure or not? A comment on the re-interpretation by del Castillo et al. of the catch mechanism in the sea urchin spine...|