Rubber Substrata Research Articles

Microscopic techniques to evaluate the biofilm formation ability of a marine bacterium Pseudomonas aeruginosa PFL-P1 on different substrata.

Biofilm formation in bacteria is strongly affected by the nature of substrata. Different substrata such as glass, polystyrene, steel, ceramic, and rubber were used to assess the biofilm forming ability of a marine bacterium Pseudomonas aeruginosa PFL-P1 using a scanning electron microscope (SEM), atomic force microscope (AFM), and confocal laser scanning microscope (CLSM). The bacterium formed dense biofilms with varied aggregation on different substrata. SEM study revealed small rod-shaped cells with diverse arrangements within the biofilms on all the substrata under study. The AFM study revealed the highest roughness of 545 nm on the ceramic substratum. The biofilms formed on ceramic substratum were characterized with maximum roughness (742 nm), maximum peak height (1,480 nm), and maximum arithmetic mean height (611 nm), significantly higher than all the other substrata (p < .05). AFM studies confirmed that P. aeruginosa PFL-P1 exhibited biofilm heterogeneity on all the substrata. The CLSM study indicated a higher fraction of nucleic acids to α-polysaccharides ratio in the biofilms. COMSTAT analysis revealed the highest biofilm biomass of ~18 μm3 /μm2 on the ceramic substratum. The maximum biofilm thickness of ~50 μm in the native state on the ceramic substratum was significantly higher than glass (p=.0015), polystyrene (p=.0001), steel (p=.0035), and rubber substrata (p=.0001). The higher surface roughness of ceramic substratum is accountable for more area for colonization, as evident from higher biomass and thickness of the biofilm. This study provides insight into the substratum properties, which modulate the biofilm forming ability in bacteria.

Biofilm formation and changes in bacterial cell surface hydrophobicity during growth in a CAPD model system.

Peritonitis is a frequent complication of continuous ambulatory peritoneal dialysis (CAPD), with patients suffering recurrent attacks. The microorganisms most frequently implicated in the infection are the skin microflora, in particular, the coagulase-negative staphylococci such as Staphylococcus epidermidis. These microorganisms gain access to the peritoneal cavity via the in-dwelling silicone rubber catheter in the abdominal wall and often persist as biofilms on the surface of the catheter. The surface characteristics of S. epidermidis were monitored during growth in a CAPD in-vitro model together with their ability to adhere to silicone rubber substrata. Fresh dialysis fluid exerted an injurious effect on the cells leading to a decrease in cell numbers but during the simulated dialysis period the cells adapted to the applied stresses. Over a 96-h period in the model both a clinical isolate and a skin isolate of S. epidermidis adopted a more hydrophobic phenotype. The data presented here show that the bacteria grown in this in-vivo reflective CAPD model continually adapt to their environment and become more tolerant to the stresses imposed. The adapted cells were seen to colonise silicone rubber substrata.

Physicochemical factors influencing bacterial transfer from contact lenses to surfaces with different roughness and wettability.

The aim of this study was to determine the transfer of Pseudomonas aeruginosa No. 3 and Staphylococcus aureus 835 from contact lenses to surfaces with different hydrophobicity and roughness. Bacteria were allowed to adhere to contact lenses (Surevue, PureVision, or Focus Night & Day) by incubating the lenses in a bacterial suspension for 30 min. The contaminated lenses were put on a glass, poly(methylmethacrylate), or silicone rubber substratum, shaped to mimic the eye. After 2 and 16 h, lenses were separated from the substrata and bacteria were swabbed off from the respective surfaces and resuspended in saline. Appropriate serial dilutions of these suspensions were made, from which aliquots were plated on agar for enumeration. Bacterial transfer varied between 4 and 60%, depending on the combination of strain, contact time, contact lens, and substratum surface. For P. aeruginosa No. 3, transfer was significantly higher after 16 h than after 2 h, whereas less increase with time was seen for S. aureus 835. Bacterial transfer from all tested contact lenses was least to silicone rubber, the most hydrophobic and roughest substratum surface included.

Contraction of cultured human uterine smooth muscle cells after stimulation with endothelin-1.

To our knowledge, the problem of how to maintain isolated smooth cells in a "contractile" phenotypic state without deviation after subculturing has yet to be resolved. The present study characterized the in vitro contractile response of human uterine smooth muscle cell to endothelin-1, which induces contractions in isolated uterine strips. Contractile effects were qualitatively investigated using silicone rubber substrata. Endothelin-1 was able to distort and reduce the wrinkles in the silicone surface. Contractions were also quantified by measuring the resulting change in the collagen lattice area. Endothelin-1 significantly increased the contractile response in a dose-dependent manner by selectively activating endothelin A receptors. When myometrial cells were cultured within collagen lattices, a microfilament-disrupting agent, cytochalasin B, abolished contractions, and no change was observed in smooth muscle alpha-actin immunostaining. Taken together, these observations show that the uterine smooth muscle cells are contractile and respond appropriately to a potent uterotonic agent. Based on these findings, a cultured uterine smooth muscle cell model, which could be used to elucidate the mechanisms controlling uterine activity, is proposed.

Interstitial cellular and matrix restoration of cardiac valves after cryopreservation

Objectives: We previously characterized the porcine aortic leaflet interstitial cell phenotype as having both synthetic and contractile characteristics; that is, it is a myofibroblast. In this study we hypothesized (1) that the cryopreservation of aortic valves causes a significant reduction in cell density, (2) that it simultaneously causes alterations in representative components of extracellular matrix, and (3) that both of these processes are reversible. Methods: Seventy-two leaflets from 24 porcine aortic valves were studied. Whole valves were subjected to variable lengths of preharvest ischemia (group 1), ischemia followed by processing analogous to clinical methods (group 2), and ischemia followed processing with an organ culture type of resuscitation (group 3). Vital dye exclusion by cells enzymatically dispersed from leaflets was used to quantify viability. Electron and light microscopy, immunohistochemical assay, and a silicone rubber substratum contractility assay were used both in dispersed cell preparations and in leaflet cross sections to examine structural, ultrastructural, and functional changes across the 3 groups through a range of preharvest ischemic times. Results: Results indicated that harvest ischemic periods between 2 and 24 hours after donor death were not responsible for cell number reductions. During this interval overt dissolution of chondroitin sulfate simultaneous with a relative sparing of fibronectin was evidenced by immunohistochemical staining. Although not reduced in number, ischemic interstitial cells did show significant ultrastructural evidence of injury and suppressed monoclonal binding to vimentin and α–smooth muscle actin. After cryopreservation, viable cell numbers were always markedly reduced at all ischemic intervals and damage to both soluble extracellular matrix components and cell ultrastructure was increased. At all time and processing points, however, some retention of matrix secretory and cellular contractile capabilities was observed among the surviving cells. After the extended periods of preharvest ischemia (2-24 hours) followed by processing, a restitution of functioning cells was accomplished by means of whole-leaflet incubation in 15% fetal bovine serum. Conclusions: After application of the described methods, new cells within restored intact leaflets as well as in single-cell preparations demonstrated normal ultrastructure and contractile and synthetic functions (normal phenotypic expression). If functioning leaflet interstitial cells can contribute to homograft durability, bioengineering methods for pretransplantation cell repopulation could be refined with these techniques and applied to clinical valve transplantation. (J Thorac Cardiovasc Surg 1999; 118:36-49)

Vinculin, talin, integrin α6β1 and laminin can serve as components of attachment complex mediating contraction force transmission from cardiomyocytes to extracellular matrix

Recently, we reported that cardiomyocytes adhere to extracellular matrix at costameres, the striated distribution of vinculin between Z-lines and the sarcolemma, where transmission of contraction forces from myofibrils to the extracellular matrix occurs. To identify other molecules involved in force transmission at costameres, we examined adult rat and embryonic chick cardiomyocytes cultured on coverslips or flexible thin silicone rubber substrata. Immunolocalization of talin showed a costameric, striated distribution, which corresponded to dark contacts with interference reflection microscopy. The molecules involved in substrate adhesion were cross-linked with the non-penetrating cross-linking agent Bis(sulfosuccinimidyl)-suberate and detected by immunohistochemical staining with anti-α6, α3, αv, or β1 integrin antibodies. Both α6 and β1 showed costameric distributions, but α3 and αv did not. The distribution of laminin after cross-linking and extraction also showed a costameric distribution. When anti-integrin β1 antibody was added to live cardiomyocytes grown on the silicone rubber substratum, the transmission of contraction forces was inhibited. These findings suggest that vinculin, talin, integrin α6β1 and laminin system can be involved in transmission of contraction force to the extracellular matrix. Cell Motil. Cytoskeleton 42:1–11, 1999. © 1999 Wiley-Liss, Inc.

Vinculin, Talin, Integrin alpha6beta1 and laminin can serve as components of attachment complex mediating contraction force transmission from cardiomyocytes to extracellular matrix.

Recently, we reported that cardiomyocytes adhere to extracellular matrix at costameres, the striated distribution of vinculin between Z-lines and the sarcolemma, where transmission of contraction forces from myofibrils to the extracellular matrix occurs. To identify other molecules involved in force transmission at costameres, we examined adult rat and embryonic chick cardiomyocytes cultured on coverslips or flexible thin silicone rubber substrata. Immunolocalization of talin showed a costameric, striated distribution, which corresponded to dark contacts with interference reflection microscopy. The molecules involved in substrate adhesion were cross-linked with the non-penetrating cross-linking agent Bis(sulfosuccinimidyl)-suberate and detected by immunohistochemical staining with anti-alpha6, alpha3, alphav, or beta1 integrin antibodies. Both alpha6 and beta1 showed costameric distributions, but alpha3 and alpha(v) did not. The distribution of laminin after cross-linking and extraction also showed a costameric distribution. When anti-integrin beta1 antibody was added to live cardiomyocytes grown on the silicone rubber substratum, the transmission of contraction forces was inhibited. These findings suggest that vinculin, talin, integrin alpha6beta1 and laminin system can be involved in transmission of contraction force to the extracellular matrix.

Scanning electron microscopic, transmission electron microscopic, and confocal laser scanning microscopic observation of fibroblasts cultured on microgrooved surfaces of bulk titanium substrata.

During this study, microtechnology and plasma etching were used to produce gratings 1.0 (TiD01), 2.0 (TiD02), 5.0 (TiD05), and 10.0 microns wide (TiD10) into commercially pure titanium wafers. After incubation of rat dermal fibroblast (RDFs) on these surfaces for 3 days, the cells were observed with scanning electron (SEM), transmission electron (TEM), and confocal laser scanning microscopy (CLSM). Results showed that the RDFs as a whole and their stress fibers oriented strictly parallel to the surface pattern on the TiD01 and TiD02 surfaces. On the TiD05 and TiD10 surfaces, this orientation was not observed. In addition, TEM and CLSM demonstrated that the focal adhesion points (FAP) were located mainly on the surface pattern ridges. TEM revealed that FAP were wrapped occasionally around the edges of the ridges. Only the RDFs on both the TiD05 and TiD10 surfaces protruded into the grooves and possessed FAP on the walls of the grooves. Attachment to the groove floor was observed only on the TiD10 textures. Comparison of these results with earlier observations on microtextured silicone rubber substrata suggests that material-specific properties do not influence the orientational effect of the surface texture on the observed RDF cellular behavior. The proliferation rate of the RDFs, however, seems to be much higher on titanium than on silicone rubber substrata.

Application of an artificial neural network in the enumeration of yeasts and bacteria adhering to solid substrata

Artificial neural networks (ANNs) combined with automated image processing are being used in a growing number of applications, ranging from car license plate identification to speech recognition. ANN analysis is capable of handling complicated images that cannot be dealt with using conventional image filters, such as images originating from yeasts and bacteria adhering on turbid solids (silicone rubber or dental enamel) or metals. In this paper, a feed-forward ANN processing greyvalues from a 9×9 pixel microscope subimage, is presented for the image enhancement and subsequent enumeration of microorganisms adhering to solid substrata. The ANN developed is a simple software routine that can be implemented as a digital image filter in existing software and assists in locating the positions of adhering microorganisms. In order to evaluate the performance of the ANN, adhesion experiments were carried out in a parallel plate flow chamber with Candida albicans, Enterococcus faecalis, Streptococcus thermophilus, Pseudomonas aeruginosa, Staphylococcus epidermidis, and polystyrene latex particles to different solid substrata under ideal and subideal illumination and focus of the microscope setup. The performance of the ANN-based enumerations was compared with the performance of nonneural, conventional image-processing methods and visual enumeration. For high-contrast images, nearly all of the adhering organisms can be detected correctly after conventional image filtering. However, in the lower quality images which were employed in this study, for metal or silicone rubber substrata, image enhancement by the ANN yielded enumeration of the adhering bacteria with an accuracy of 93%–98%, while using conventional image filters, an accuracy of 86–96% was achieved. Adhering yeasts could be enumerated after conventional image filtering with an accuracy of 92%, but after ANN enhancement, 98% of the yeast cells adhering on silicone rubber substrata were enumerated correctly. It can be concluded that for low quality and complicated images, ANNs perform better than conventional filters, while as additional advantages, the training of an ANN requires significantly less expert knowledge than optimal setting of conventional image filters, and an ANN can be used to distinguish between strains in adhesion experiments involving more than one strain.

Streptococcus thermophilus and its biosurfactants inhibit adhesion by Candida spp. on silicone rubber.

The adhesion of yeasts, two Candida albicans and two Candida tropicalis strains isolated from naturally colonized voice prostheses, to silicone rubber with and without a salivary conditioning film in the absence and presence of adhering Streptococcus thermophilus B, a biosurfactant-releasing dairy isolate, was studied. Coverage of 1 to 4% of the surface of silicone rubber substrata with adhering S. thermophilus B gave significant reductions in the initial yeast adhesion regardless of the presence of a conditioning film. Mechanistically, this interference in yeast adhesion by S. thermophilus B was not due to direct physical effects but to biosurfactant release by the adhering bacteria, because experiments with S. thermophilus B cells that had released their biosurfactants prior to adhesion to silicone rubber and competition with yeasts did not show interference with initial yeast adhesion. The amounts of biosurfactants released were highest for mid-exponential- and early-stationary-phase bacteria (37 mg.g of cells-1 [dry weight]), but biosurfactants released by stationary-phase bacteria (14 mg.g of cells-1 [dry weight]) were the most surface active. The crude biosurfactants released were mixtures of various components, with a glycolipid-like component being the most surface active. A lipid-enriched biosurfactant fraction reduced the surface tension of an aqueous solution to about 35 mJ.m-2 at a concentration of only 0.5 mg.ml-1. The amount of biosurfactant released per S. thermophilus B cell was estimated to be sufficient to cover approximately 12 times the area of the cross section of the bacterium, making biosurfactant release a powerful defense weapon in the postadhesion competition of the bacterium with microorganisms such as yeasts. Preadsorption of biosurfactants to the silicone rubber prior to allowing yeasts to adhere was as effective against C. albicans GB 1/2 adhesion as covering 1 to 2% of the silicone rubber surface with adhering S. thermophilus B, but a preadsorbed biosurfactant layer was less effective against C. tropicalis GB 9/9.

Quantitative analysis of fibroblast morphology on microgrooved surfaces with various groove and ridge dimensions

Fibroblasts have been shown to respond to substratum surface roughness. The change in cell size, shape and orientation of rat dermal fibroblasts (RDF) was therefore studied using smooth and microtextured silicone rubber substrata. The microtextured substrata possessed parallel surface microgrooves that ranged in width from 1.0 to 10.0 μm, and were separated by ridges of 1.0 to 10.0 μm. The grooves were either 0.45 or 1.00 μm deep. Prior to incubation, the substrata were cleaned and given a radio frequency glow discharge treatment. After surface evaluation with scanning electron microscopy and confocal laser scanning microscopy, RDF were incubated on these substrata for 5 days. During this period of incubation, the RDF were photographed on days 1, 2, 3, 4, and 5, using phase contrast microscopy. Digital image analysis of these images revealed that on surfaces with a ridge width ≤4.0μm, cells were highly orientated (< 10°) and elongated along the surface grooves. Protrusions contacting the ridges specifically could be seen. If the ridge width was larger than 4.0 μm, cellular orientation was random (≈45°) and the shape of the RDF became more circular. Furthermore, results showed that the ridge width is the most important parameter, since varying the groove width and groove depth did not affect the RDF size, shape, nor the angle of cellular orientation.

Computing cell tractions from particle displacements on silicone rubber substrata

In order to better understand the distribution of cell traction forces generated by rapidly locomoting cells, we have applied a mathematical analysis to our modified silicone rubber traction assay, based on the plane stress Green’s function of linear elasticity. To achieve this, we made crosslinked silicone rubber films into which we incorporated many more latex beads than previously possible (Figs. 1 and 6), using a modified airbrush. These films could be deformed by fish keratocytes, were virtually drift-free, and showed better than a 90% elastic recovery to micromanipulation (data not shown). Video images of cells locomoting on these films were recorded. From a pair of images representing the undisturbed and stressed states of the film, we recorded the cell’s outline and the associated displacements of bead centroids using Image-1 (Fig. 1). Next, using our own software, a mesh of quadrilaterals was plotted (Fig. 2) to represent the cell outline and to superimpose on the outline a traction density distribution. The net displacement of each bead in the film was calculated from centroid data and displayed with the mesh outline (Fig. 3).

Video-microscopic measurement of cell-substratum traction forces generated by locomoting keratocytes

We are using video-enhanced light microscopy to investigate the pattern and magnitude of forces that fish keratocytes exert on flexible silicone rubber substrata. Our goal is a clearer understanding of the way molecular motors acting through the cytoskeleton co-ordinate their efforts into locomotion at cell velocities up to 1 μm/sec. Cell traction forces were previously observed as wrinkles(Fig.l) in strong silicone rubber films by Harris.(l) These forces are now measureable by two independant means.In the first of these assays, weakly crosslinked films are made, into which latex beads have been embedded.(Fig.2) These films report local cell-mediated traction forces as bead displacements in the plane of the film(Fig.3), which recover when the applied force is released. Calibrated flexible glass microneedles are then used to reproduce the translation of individual beads. We estimate the force required to distort these films to be 0.5 mdyne/μm of bead movement. Video-frame analysis of bead trajectories is providing data on the relative localisation, dissipation and kinetics of traction forces.

Costameres are sites of force transmission to the substratum in adult rat cardiomyocytes.

Costameres, the vinculin-rich, sub-membranous transverse ribs found in many skeletal and cardiac muscle cells (Pardo, J. V., J. D. Siciliano, and S. W. Craig. 1983. Proc. Natl. Acad. Sci. USA. 80:363-367.) are thought to anchor the Z-lines of the myofibrils to the sarcolemma. In addition, it has been postulated that costameres provide mechanical linkage between the cells' internal contractile machinery and the extracellular matrix, but direct evidence for this supposition has been lacking. By combining the flexible silicone rubber substratum technique (Harris, A. K., P. Wild, and D. Stopak. 1980. Science (Wash. DC). 208:177-179.) with the microinjection of fluorescently labeled vinculin and alpha-actinin, we have been able to correlate the distribution of costameres in adult rat cardiac myocytes with the pattern of forces these cells exert on the flexible substratum. In addition, we used interference reflection microscopy to identify areas of the cells which are in close contact to the underlying substratum. Our results indicate that, in older cell cultures, costameres can transmit forces to the extracellular environment. We base this conclusion on the following observations: (a) adult rat heart cells, cultured on the silicone rubber substratum for 8 or more days, produce pleat-like wrinkles during contraction, which diminish or disappear during relaxation; (b) the pleat-like wrinkles form between adjacent alpha-actinin-positive Z-lines; (c) the presence of pleat-like wrinkles is always associated with a periodic, "costameric" distribution of vinculin in the areas where the pleats form; and (d) a banded or periodic pattern of dark gray or close contacts (as determined by interference reflection microscopy) has been observed in many cells which have been in culture for eight or more days, and these close contacts contain vinculin. A surprising finding is that vinculin can be found in a costameric pattern in cells which are contracting, but not producing pleat-like wrinkles in the substratum. This suggests that additional proteins or posttranslational modifications of known costamere proteins are necessary to form a continuous linkage between the myofibrils and the extracellular matrix. These results confirm the hypothesis that costameres mechanically link the myofibrils to the extracellular matrix. We put forth the hypothesis that costameres are composite structures, made up of many protein components; some of these components function primarily to anchor myofibrils to the sarcolemma, while others form transmembrane linkages to the extracellular matrix.

Morphology of donor lens-capsule material studied by SEM

The lens capsule of a 70-year-old male donor with a cataractous lens was carefully prepared for SEM by first washing the capsule with buffer solution to remove lens-fibres and subsequently attaching it to silicon rubber. During the fixation and drying stages of the preparation procedure the capsule stayed attached to the rubber substratum. In the equatorial zone germinating cells were found with knob-shaped microvilli, closely connected to lens-fibres. Large units of pathological capsule epithelial cells were found, only slightly inter connected by a few pseudopodia. In addition, single pathological epithelial cells with pseudopodia, arranged on top of the cell in a rosette-like configuration, were found at certain locations. Both types are probably related to the original lens-cataract.

Effects of electric fields on fibroblast contractility and cytoskeleton

We used silicone rubber substrata and fluorescent staining of cytoskeletal components to study the mechanisms by which electrical voltage gradients cause reorientation of embryonic chick fibroblasts in tissue culture. No evidence was found for a direct stimulation of cell contractility, either parallel or perpendicular to the voltage gradient. Instead, there was a gradual weakening in cell contractility in the axis parallel to this gradient, accompanied by progressive retraction of lamellae oriented along this axis, apparently due to selective weakening of cell-substratum adhesions. The cells then elongated perpendicular to the electric field, and strengthened their contractility in that axis. Fluorescence microscopy showed that cytoplasmic actin stress fibers and microtubules oriented perpendicular to the imposed voltage gradient. Many more cases were observed in which cell morphology had reoriented, but the actin fibers had not, as compared to the converse (cytoskeleton oriented, but no morphology). This disparity further supports the interpretation that the redirection of cell contractility is a consequence of morphological reorientation, rather than its cause. We also studied the effects of reversing the polarity of the electric fields at constant intervals (of as long as 1 minute). Fibroblasts failed to orient in response to such alternating fields, even after long exposure, but these same cells did reorient in response to pulsed currents in a consistent direction separated by "rest periods" (with no current). This combination of results is more consistent with an electrophoretic mechanism than with one depending on voltage-induced changes in membrane permeabilities.

Fibroblast contractility and actin organization are stimulated by microtubule inhibitors

Despite considerable evidence that cytoplasmic microtubules play some role in guiding or controlling the locomotion of tissue cells, the nature of this control is not understood. In particular, little is known about the role of microtubules in the exertion of propulsive 'traction' forces, or about microtubule effects on the organization of the cytoplasmic actin stress fibers. In this study, the silicone rubber substratum technique was used in combination with fluorescence microscopy in order to observe the effects of microtubule-depolymerizing drugs on the contractile strength and organization of cytoplasmic actin networks. Perfusion with a variety of microtubule poisons (either colcemid, nocodazole or vinblastine) was found to cause a rapid and substantial strengthening of fibroblast contractility. This was demonstrated in two established fibroblast cell lines, as well as in primary cultures of rat gingival fibroblasts and embryonic chick heart fibroblasts. Treatment with the drug taxol, which promotes microtubule assembly, was found to prevent the strengthening effects of the microtubule inhibitors. It was also found that the disruption of actin stress fibers by the phorbol ester tumor promoter, TPA, is reversed by microtubule poisons: stress fibers reform within 30 min of the addition of the microtubule drugs, despite the continued presence and activity of the TPA. Several possible mechanisms are considered, including the idea that microtubule assembly normally exerts a pushing force, which counterbalances part of the contractile force exerted by the actin stress fibers. However, the mechanism that seems best to account for the observations is that microtubules modulate, in an inhibitory fashion, the contractility and the state of organization of cytoplasmic actin.

Changes in fibroblast contractility, morphology, and adhesion in response to a phorbol ester tumor promoter

We have studied the effects of the phorbol ester tumor promoter 12-O-tetradecanoyl-phorbol-13-acetate (TPA) on the contractility, locomotion, morphology, and adhesion of two mammalian fibroblastic cell lines. Using the silicone rubber substratum technique, we have found that the first observable response to the tumor promoter is a rapid weakening of cell contractility (8–15 min). This is followed by gradual morphological changes, characterized by a hyperextension of the cells' leading lamellae, which stretch out to an unlimited degree, and occasionally even detach from the cell bodies. Treated cells also become able to crawl onto hydrophobic substrata which are insufficiently adhesive to support the spreading of untreated fibroblasts. We suggest that both the hyperextension and the ability to spread on nonadhesive surfaces can be explained as consequences of the reduced contractility, and that this reduced contractility may also help to explain the increased invasiveness and loss of anchorage dependence by transformed cells.

Locomotion of sponges and its physical mechanism.

Active locomotion by individual marine and freshwater sponges across glass, plastic and rubber substrata has been studied in relation to the behavior of the sponges' component cells. Sequential tracing of sponge outlines on aquarium walls shows that sponges can crawl up to 160 microns/hr (4 mm/day). Time-lapse cinemicrography and scanning electron microscopy reveal that moving sponges possess distinctive leading edges composed of motile cells. Sponge locomotion was found to be mechanically similar to the spreading of cell sheets in tissue culture both with respect to exertion of traction (which causes the wrinkling of rubber substrata) and with respect to the patterns of adhesive contacts formed with the substratum (as observed by interference reflection microscopy). Other similarities include the orientation of sponge locomotion along grooves and the preferential extension onto more adhesive substrata. Neither the patterns of wrinkling produced in rubber substrata nor the distributions of adhesive contacts seen by interference reflection microscopy show evidence of periodic, propagating waves of surface contractions, such as would be expected if the sponges' mechanism of locomotion were by peristalsis or locomotory waves. Our observations suggest that the displacement of sponges is achieved by the cumulative crawling locomotion of the cells that compose the sponge's lower surface. This mode of organismal locomotion suggests new explanations for the plasticity of sponge morphology, seems not to have been reported from other metazoans, and has significant ecological implications.

Cell Motility and the Problem of Anatomical Homeostasis

The locomotion of tissue cells need not be invasive or disrupt normal tissue geometry, as occurs in cancer. The normal relationship between anatomical structure and cell locomotion is exactly the reverse, with motility serving to create and maintain the structures of the body. This relationship is most extreme in sponges, where time-lapse films show that the cells move about continually in patterns that restructure these animals' simple anatomy. The cells choose their position according to their differentiated cell type, which is the opposite of what is usually assumed to occur in development and has important implications for the functional significance of histotypic cell sorting. A particular type of cellular force that seems to be important in morphogenesis is the traction that all motile tissue cells exert. This traction can be studied by culturing cells on very thin sheets of silicone rubber, so that the locations and variations in the cellular forces are made visible by the wrinkles they produce in the rubber substratum. One finding has been that the traction forces exerted by untransformed fibroblasts are very much stronger than is needed for their own locomotion, but are well adapted for the function of rearranging and aligning collagen fibres to form structures like ligaments, tendons and muscles. These forces are found to be greatly weakened by neoplastic transformation, however, suggesting that malignant invasiveness results from some sort of deflection of cell traction forces from their proper morphogenetic functions, so as to produce uncontrolled invasion. To explain how motile cells create and maintain structures, as well as how their locomotion sometimes becomes perverted into the form of cancerous invasiveness, what seems to be needed is an extension of the concept of homeostasis to apply to the control of geometric relations between cells. This task may not be easy; one obstacle is the widespread belief that asymptotic stability implies the minimization of free energy. Instead, I suggest that stability results from balance of opposing forces within tissues, and that genes control which anatomical shapes will exist by determining the rules by which the relative strengths of these forces vary as functions of shape: to control shape, one must control the way forces vary with shape.