0040-8166/90/0022-0407/$10.00

TISSUE AND CELL, 1990 22 (4) 407-417 @ 1990 Longman Group UK Ltd

H. PAUL EHRLICH and JOSEPH 6. M. RAJARATNAM

CELL LOCOMOTION FORCES VERSUS CELL CONTRACTION FORCES FOR COLLAGEN LATTICE CONTRACTION: AN IN W-R0 MODEL OF WOUND CONTRACTION Keywords: Myofibroblasts, contraction

tibroblasts, collagen lattices, contraction,

actin filaments, wound

ABSTRACT. Cultured human dermal tibroblasts suspended in a rapidly polymerizing collagen matrix produce a fibroblast-populated collagen lattice. With time, this lattice will undergo a reduction in size referred to as lattice contraction. During this process, two distinct cell populations develop. At the periphery of the lattice, highly oriented sheets of cells, morphologically identifiable as myofibroblasts, show cell-to-cell contacts and thick, actin-rich staining cytoplasmic stress fibers. It is proposed that these cells undergoing cell contraction produce a multicellular contractile unit which reorients the collagen fibrils associated with them. The cells in the central region, referred to as fibroblasts, are randomly oriented, with few cellto-cell contacts and faintly staining actin cytoplasmic filaments. In contrast it is proposed that cells working as single units use cell locomotion forces to reorient the collagen tibrils associated with them. Using this model, we sought to determine which of these two mechanisms, cell contraction or cell locomotion, is responsible for the force that contracts collagen lattices. Our experiments showed that fibroblasts produce this contractile force, and that the mechanism for lattice contraction appears to be related to cell locomotion. This is in contrast to a myofibroblast; where the mechanism for contraction is based upon cell contractions. Fibroblasts attempting to move within the collagen matrix reorganize the surrounding collagen tibrils; when these collagen fibrils can be organized no further and cell-to-cell contacts develop, which occurs at the periphery of the lattice first, these cells can no longer participate in the dynamic aspects of lattice contraction.

Introduction

An open wound may close by wound contraction, a process by which the surrounding, uninjured dermis and epidermis are pulled over the defect. Regeneration does not occur; instead, the centripetal movement of the surrounding skin closes the defect. The healed wound site is covered by normal skin, and minimal scar develops. Although initially wound contraction was thought to be caused by forces generated by newly deposited colWound Healing Laboratory, Shriner Bums Institute, Departments of Pathology and Surgery, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts USA. Correspondence to: Dr. H. Paul Ehrlich, Shriner Burns Institute, 51 Blossom Street, Boston, MA 02114, USA. Received 29 March 1990

lagen fibers (Billingham and Russell, 1956), in 1956 Abercrombie and colleagues showed that this movement was the result of forces generated by cells residing in the central granulation tissue (Abercrombie et al., 1956). It appears that wound contraction is a dynamic process where cells organize their surrounding connective tissue matrix. In 1971, Gabbiani et al. (1971) put forth the myofibroblast as the specialized cell responsible for the forces at work in wound contraction. While a large amount of histological evidence has lent credence to this popular theory, little physiological evidence exists to support myofibroblasts undergoing cell contraction as the mechanism of contraction of ‘granulation tissue’ and the organizer of resident connective tissue fibers. The term ‘myofibroblast’ is used to

408

describe a cell that is morphologically similar to both a fibroblast and a smooth muscle cell (Majno et al., 1971). Myofibroblasts, which form many cell-to-cell contacts, are rich in cytoplasmic microfilaments in the form of cable-like, actin-rich stress fibers. Antibodies directed to filamentous actin is a technique that allowed myofibroblasts to be identified by light microscopy (Hirschel et al., 1971). The cytoplasmic stress fibers characteristic of myofibroblasts stain intensely with fluorescent anti-actin antibodies or fluorescenttagged phalloidin (Hembry et al., 1986). Fibroblasts stain less intensely, because their cytoplasmic actin is contained in finer filamentous structures. Experiments using strips cut from capsules surrounding experimentally induced granulomas, which are rich in myofibroblasts, showed that a smooth muscle contractile agent, such as 5_hydroxytryptamine, produced contraction of the tissue in a doseresponse manner; the length of the capsule tissue strip became shorter (Gabbiani ef al., 1972). In another series of experiments in the same report, these same agents were shown to be ineffective in producing contraction of granulation tissue excised from contracting open wounds from rats. This tissue had to be stretched prior to the addition of agents in order to show a response. One explanation for these differences is that capsule tissue is in a static state and is prevented from contracting because of the underlying granuloma. Capsule tissue has the potential to contract, but the physical barrier conferred by underlying granuloma hinders contraction. The removal of the granuloma by isolating the capsule tissue strip allows contraction of the tissue by a mechanism involving cell contraction. In the case of the contracting wound, the contraction process is in a dynamic state, and the tissue composed of cells and connective tissue is at the maximal contracted state. The induction of cell contraction in this tissue would not produce adequate force for the movement of collagen fibers and the contraction of this tissue. The hypothetical mechanism of wound closure by myofibroblasts is dependent on coordinated cellular contraction (Ryan et al., 1974). As myofibroblasts form cell-to-cell contacts and undergo sequential cellular contractions, the multi-cellular unit rearranges the surrounding connective tissue matrix into

EHRLICH AND RAJARATNAM

a more compact configuration. An experimental model that describes and supports this theory is based on the Hoffman-Berling (1954) report of fibroblast ATP-induced cell contraction. Glycerol-permeabilized, cultured monolayer fibroblast preparations treated with ATP, undergo rapid cell contraction. Because cell contraction involves the shortening of the actin-myosin filaments in cells that have cell-to-cell contacts, a multicellular model is presented. When the individual cells in this multi-cellular tissue sheet contract, the encircling connective tissue becomes rearranged, and the surrounding normal skin is pulled over the defect. This myofibroblast theory of wound contraction suggests that a specialized cell population within the granulation tissue closes wounds by a mechanism involving cell contraction in a multi-cellular unit. The origin of the myofibroblast is the fibroblast rather than the smooth muscle cell, hence fibroblasts develop into myofibroblasts. The proposed mechanism of wound closure by fibroblasts employing cell locomotion forces is proposed. Fibroblasts produce tractional forces which are responsible for cell locomotion (Harris ef al., 1980). Fibroblasts surrounded by collagen fibrils rearrange those fibrils by cell locomotion forces. This rearrangement of collagen fibrils compact the tissue. reducing its size. The association of individual fibroblasts interacting with collagen fibrils is the proposed mechanisms for wound contraction. The introduction of a simple tissue culture model by Bell et al. (1979) has facilitated a more comprehensive study, in a controlled environment, of the cellular mechanism of wound contraction. The fibroblast-populated collagen lattice (FPCL) is manufactured from freed monolayer cultured fibroblasts, culture medium containing serum, and soluble native collagen. When these components are mixed together rapidly at 37 “C, the collagen polymerizes and traps the fibroblasts in the matrix. Under the conditions employed in our laboratory. within hours after manufacture the size of the FPCL is reduced. This process of lattice contraction is assumed to be similar to wound contraction. Using this dynamic contracting tissue model, we wish to investigate whether cell contraction or cell locomotion is responsible for producing the forces that contract this experimental tissue.

CELL LOCOMOTION

FORCES VERSUS CELL CONTRACTION

Methods and materials Cells

Fibrobiasts were derived from foreskin explants (a gift from Dr. David Wyler, Department of Medicine, Tufts New England Medical Center) and grown in Dulbecco’s Modification of Eagle’s Medium with 10% fetal bovine serum. This formulation of culture medium will be referred to as ‘DMEM’ in this paper. Cells were grown to confluency and passed. The cells used in these experiments were used in their tenth to fifteenth passage. Collagen

The native collagen solutions used in these experiments were derived from either 0.5 M acetic acid extraction of rat tail tendon or from limited pepsin extraction of leiomyoma, a benign uterine tumor which has been previously described (Ehrlich & Griswold, 1984). Collagen solutions were maintained at 4 “C in 1 mM HCl at 5 mg/ml. Casting FPCL

One milliliter of DMEM was placed in a 35 mm Petri dish, followed by 0.5 ml of DMEM containing freshly trypsinized fibroblasts and finally 0.5 ml of collagen solution. The three components were mixed immediately, and the dish was transferred to a 37 “C incubator with 95% air and 5% COZ. In less than 90sec, the collagen polymerized, trapping the fibroblasts within the rapidly forming matrix. The fibroblasts are initially surrounded by a collagen matrix. A TP-induced cell contraction

The dynamic aggregation of actin-myosin cytoplasmic filaments can be induced in permeabilized cell preparations by the addition of ATP in a buffer with cofactors (Goldman et al,, 1976). Briefly, fibroblasts plated onto coverslips or in FPCL were transferred to 60mm Petri dishes and incubated at room temperature for 30min in a 50% glycerol solution (50% glycerol in 50 mM Tris : HCl, pH 7.6). The 50% glycerol solution was removed and replaced with 25% glycerol in the same buffer for another 30min incubation period. This wash was followed by 30 min incubations of 12% and then 5% glycerol. These treatments produced a permeabilized cellular preparation.

409

FORCES

At room temperature, 1 mM ATP in 20mM phosphate buffer, pH 7.6, 30 mM KCl, 0.1 mM CaCl* and 5 mM MgClz was added to these preparations. Some preparations were then fixed in buffered 4% paraformaldehyde immediately following glycerol treatments or 10 min after ATP addition. In either case, the paraformaldehyde was removed after 5 min, and the preparations were washed 3 times with phosphate-buffered saline (PBS). To stain for rhodamine-tagged actin filaments, Rh-phalloidin (Molecular phalloidin, Probes, Eugene, OR), was added to the fixed preparations. The stain was removed after a 30 min incubation period at room temperature. The preparations were washed 3 times with PBS, then mounted in glycerol:PBS (9 : 1) and viewed with a Zeiss IM 35 microscope with fluorescent optic filters (Ehrlich et al., 1986). Results An FPCL containing 75,000 cells and 2.5 mg of pepsin-extracted collagen in a volume of 2 ml in 48 hr underwent a size reduction from an initial area of 960 mm2 to 122 mm2. When the FPCL was viewed at 24 hr, two populations of cells had evolved. This observation is similar to contracting wounds when myofibroblasts appear after a delay of one week after wounding. The centrally located cells, fibroblasts (Fig. lA), were surrounded by collagen and randomly oriented, with diffusely stained microfilaments and few cellto-cell contacts. The high density cells at the periphery of the FPCL, myofibroblasts (Fig. lB), were highly oriented, with many cellto-cell contacts and prominent a&n-rich cytoplasmic stress fibers. Based upon morphology criteria myofibroblasts appear in FPCL during lattice contraction. We examined the question which of these cell types is responsible for lattice contraction. Donut experiment

To test the hypothesis that myofibroblasts functioning as a multicellular contracting unit are responsible for lattice contraction, the donut experiment was done. FPCL that initially contained 75,000 fibroblasts were cast. They were incubated for 6 hr. To enhance the proportion of myofibroblasts, the central area of 3 lattices was punched out

EHRLICH

AND RAJARATNAM

Fig. 1. Myofibroblasts and fibroblasts in FF’CL. FF’CL were made with 60,000 cells and 1.25 mg/ml of leiomyoma pepsin-extracted collagen in DMEM supplemented with 10% FBS. Lattices were fIxed and stained with Rh-phalloidin at 48 hr. (Magnification 128x). A. Cells in central region of J?PCL fibroblast morphology. B. Cells at periphery of FF’CL have myolibroblast morphology.

wil :h the top edge of a 100 x 13 mm test tube.

If Inyofibroblasts were responsible for lattice co1Itraction, then the contractile force must be generated from the edge of the donut lat tices. The donut FPCL would be expected

to contract faster and to a greater degl -ee than the whole intact FPCL. Comparisons between a donut and an intact FPCL were made at the 48 hr tilme point. As can be seen in Fig. 2, the inti act

CELL LOCOMOTION FORCES VERSUS CELL CONTRACTION FORCES

Fig. 2. The donut experiment. As described in the text, a hole was punched out of the middle of some FPCL to form donut lattices. Other lattices remained whole. A. A pair of lattices immediately following the removal of a central disc. B. The same lattices 48 hr later.

FPCL contracted from an initial area of 960 mm2 to 95 mm* (11 mm in diameter) in 48 hr. The outer diameter of the donut lattice was also 11 mm-at this time. Both the intact and donut lattices were identical in size. Lattice contraction was not enhanced by increasing the proportion of myofibroblasts. This result supports the single cell unit, fibroblast, mechanism for lattice contraction. Wedge experiment

Another modification was used to explore

of the FPCL model whether lattice con-

traction requires fibroblasts alone, myofibroblasts alone, or a combination of both cells. FPCL were cast with 75,000 fibroblasts and 2.5 mg of acetic acid-extracted rat tail tendon collagen in a total volume of 2 ml. In our experience, lattices composed of rat tail tendon collagen, which is rich in type I collagen, contracted to a lesser degree than lattices composed of leiomyoma pepsin-solubilized collagen, which is rich in type III collagen (Ehrlich, 1988). Lattice contraction proceeded for 24 hr, at which time 2 wedges (a and b) were cut from the FPCL (see diagram,

412

EHRLICH

RAJARATNAM

which was part of the periphery of the FPCL; the three cut edges of wedge ‘b’ contained only fibroblasts. Changes in lattice contraction in these wedges was documented 3 days later. The three edges of wedge ‘b’ contracted equally, while the edge of wedge ‘a’ containing myofibroblasts demonstrated less reduction (Fig. 3). In this experiment, myofibroblasts show a retarded ability to contract FPCL when compared to fibroblasts. Therefore, fibroblasts with fine, diffuse cytoplasmic microfilaments are better at organizing collagen and contracting FPCL.

A

ATP-induced

Fig. 3. The wedge experiment. A. Two cut from individual FF’CL. B. A lattice cut Wedge (a) is at the bottom, and wedge (b) C. These same wedges 3 days later. Wedge bottom of the dish, and wedge (b) is at the

AND

wedges were into wedges. is at the top. (a) is at the top.

Fig. 3). Wedge ‘a’ was cut so that one edge included the periphery of the FPCL; myofibroblasts were prominent in that edge, while the remaining two cut edges were populated by fibroblasts. Wedge ‘b’ was designed to contain myofibroblasts only at one apex,

cell contraction

Cell contraction is a proposed mechanism for producing the forces that contract FPCL. The effect of inducing cell contraction by adding ATP directly was investigated. Cells harvested by trypsinization of normal human fibroblasts were plated onto 22 x 22 mm coverslips. Coverslips were placed in 35 mm Petri dishes and 25,000 cells in 2 ml of medium. Fibroblasts were plated, and ATP induced cell contraction was tested at various time points. Half of the glycerol-permeabilized fibroblast preparations were fixed and stained with Rh-phalloidin; ATP and cofactors were added to the other half. After lOmin, the ATP-buffered solution was removed, and the cell preparations were fixed and stained with Rh-phalloidin. Plated fibroblasts were found to demonstrate cytoplasmic actin-rich stress fibers as early as 6 hr after plating. At 2 and 4 hr after plating, few prominent cytoplasmic actin-containing stress fibers were identifiable. At time points less than 4 hr, no stress fibers were evident. At all time points tested, ATP addition induced both cell contraction and the aggregation of cytoplasmic actin filamentous material (Fig. 4). Apparently, cytoplasmic stress fibers are not required for ATP-induced cell contraction and actin and myosin aggregation. Since ATP caused the contraction of cells containing fine filamentous cytoplasmic microfilaments, these fine filaments contain all the components required to produce cell contraction. Fibroblasts as well as myofibroblasts can be induced to contract and to aggregate actin filamentous material by the addition of ATP and cofactors.

CELL LOCOMOTION FORCES VERSUS CELL CONTRACTION FORCES

Cell contraction in collagen lattices

To investigate the possibility that fibroblast contraction causes FPCL contraction, six FPCL containing 75,000 cells in 2.5 mg leiomyoma collagen in a total volume of 2 ml were treated with glycerol 18 hr after manufacture. Three FPCL were immediately fixed after being permeabilized by glycerol; the remaining three were treated at room temperature with ATP and cofactors for 10 minutes and then fixed. All lattices were measured, and no change in area was noted between ATP-treated and non-treated lattices. The lattices were stained with Rh-phalloidin and viewed. In Figure 5, collagen lattice cells treated with glycerol alone show cytoplasmic actin stained and continuous filaments running parallel to the cell. After ATP was added, discontinuous, broken actin filaments ran parallel to the cells. Some cytoplasmic actin filaments ruptured in response to ATP addition, but neither cells nor lattice contracted. ATP-induced actin filament contraction produced forces that broke filaments but it was not adequate to pull cells away from the collagen matrix or to reorient those surrounding collagen fibrils. Acute microfilament or stress fiber contraction did not produce forces that contract collagen lattices or reorient collagen fibrils. Here dynamic contracting FPCL could not be contracted more by the induction of cell contractions.

Discussion

Here we have shown that cells in the middle to a FPCL produce centractile forces. The addition of ATP to these cells in a permeasible state did not produce lattice contraction. It appears that multicellular cellcontraction or single cellular induced cellcontraction does not produce collagen fibril organization or lattice contraction. A direct procedure was used to examine a multicellular mechanism of lattice contraction. A donut-shaped FPCL was produced by removing the centre of an actively contracting lattice. Based upon morphology, fibroblasts predominate in the center of the FPCL, whereas the periphery is composed mainly of myofibroblasts. This excision increased the proportion of myofibroblasts in the FPCL, at the same time the collagen content

413

of the lattice was diminished, which also should have enhanced lattice contraction. This did not occur because the fibroblast population was also diminished. If myofibroblasts are responsible for lattice contraction, then the donut-shaped FPCL in which they were more prevalent should have contracted faster and to a greater degree than the intact lattice. Both lattices showed the same degree of contraction, thus denying the possibility of a coordinated multicellular mechanism producing lattice contraction. For this reason, the mechanism involving a multicellular cell contraction in lattice contraction appears to be minimal. The single cellular unit concept appears more credible. The wedge experiment showed that lattice contraction requires only fibroblasts, which produce the force that contracts the collagen lattice and reorganizes the surrounding collagen matrix. It further recommends that collagen fiber organization is the result of cell locomotion as opposed to a cell contraction mechanism. To study the effect of cell contraction forces directly on lattice contraction, ATPinduced cell contraction of fibroblasts suspended in the collagen matrix was studied. ATP-induced cell contraction produced no perceptible lattice contraction. Histological examination of ATP-treated cells showed aggregation of the actin filaments but no cell contraction. It appears that the acute forces produced by the addition of ATP-induced microfilament sliding and the aggregation of actin-rich filaments were great enough to fracture cytoplasmic filament structures but not to relocate surrounding collagen fibers or produce cell contraction. Since the mechanism of lattice contraction requires the rearrangement of collagen fibers, ATPinduced acute aggregation of cytoplasmic filaments was incapable of producing such forces. The results reported here are similar to the in vitro experiment reported by Gabbiani and coworkers (1972) where the induction of cell contraction did not produce contraction of tissue excised from dynamically contracting healing wounds. On the other hand, our results are unlike those reported by Tomasek and coworkers (1989), where FPCL were made so that they were prevented from contracting for 5 days. Upon release, these matrices underwent a rapid contraction

CELL LOCOMOTION

FORCES VERSUS CELL CONTRACTION FORCES

Fig. 5. ATP induced cell contraction in FPCL. FPCL were treated with glycerol. Half were fixed and stained with Rh-phalloidin. Note the presence of F-actin filaments in the cytoplasm. B. After glycerol treatment, FPCL were incubated with ATP for 10 min, then tixed and stained. Note that the actin filaments appear fractured at numerous points along the cell axis; however, the length of the cell has not changed. ATP-induced actin filament aggregation is present but not cell contraction. (Magnification 128x).

it was shown that the resident cells that were elongated within the matrix upon release underwent a morphological change to a more rounded cell shape. It appears that inhibition of contraction produces a static condition where a potential contraction forces develop. The release of this hindrance produces a very rapid contraction of the tissue by the contraction of resident myofib-

where

roblasts. In vivo this rapid contraction of tissue which is prevented from contracting was demonstrated by granuloma capsules (Gabbiani et al., 1972) and in vitro by splinted FPCL (Tomasek et al., 1989). A flaw with these models and their kinship with wound contraction is that wound contraction is a steady active process that transpires over a relatively long period of time. An in vitro

Fig. 4. ATP induced cell contraction. Fibroblasts were grown on glass coverslips as described in text. Medium was removed from coverslips at 4 and 24 hr. All coverslips were treated with glycerol to permeabilize cells. Half the permeabilized cell preparations were fixed and stained with Rh-phalloidin; the other halves were incubated with ATP for 10 min and then fixed and stained. A. Cells were plated on coverslips and incubated for 4 hr, then half were fixed and stained. Note the pattern of F-actin staining. B. The other half were fixed and stained after a 10min incubation period with ATP. Note that ATP can induce cell contraction and actin aggregation without requiring cytoplasmic parallel stress fibers. C. Cells were plated for 24 hr on coverslips and half the coverslips were tixed and stained. Note the presence of parallel Factin staining cytoplasmic stress fibres. D. The remaining coverslips were tied and stained after treating with ATP for 10min. Note the cell contraction and the aggregation of actin material. (Magnification 256x)

416

EHRLICH

model where rapid lattice contraction occurs does not correctly replicate the slow steady process of in vivo wound contraction that happens during wound healing. The morphological attributes of myofibroblasts, specifically their prominent stress fibers, make these cells the most logical choice for the production of contractile forces (Rungger-Brandle & Gabbiani, 1983). In the ATP-induced cell contraction model, fibroblasts with only fine, filamentous microfilaments are capable of producing cell contraction; large cytoplasmic stress fibers are not exclusively required. Apparently stress fibers develop as a consequence of frustrated cell locomotion. Because when a fibroblast acquires a myofibroblast morphology it is no longer undergoing cell locomotion, this cell is not contributing to further lattice contraction. The myofibroblast is contributing to holding collagen fibers in place and it produces tension in connective tissue. The in vitro appearance of myofibroblasts on tissue culture surfaces and in FPCL is an end result of the apparent termination of cellular locomotion. When the attachment of the cell membrane to a surface stabilizes and cell locomotion is blocked, the cytoplasmic microfilaments aggregate and form thick, cable-like static filaments. Fine cytoplasmic microfilaments are involved in cell locomotion, and their aggregation into cable like stress fibers denotes cessation of cell locomotion. Fibroblasts at the periphery of an FPCL because of their location become contact-inhibited, that is, they no longer move and thus acquire a myofibroblastic morphology. These static cells at the periphery of a FPCL are poor at organizing collagen. Experimental evidence in support of the myofibroblast hypothesis that this specialized cell is responsible for wound contraction is incomplete. Myofibroblasts are not readily identifiable in healing rat wounds before 7 days, when closure is about 50% complete (Majno et al., 1971), a fact that contradicts the hypothesis. Their presence during the

AND RAJARATNAM

period of delayed wound closure of full excision wounds in the genetic strain of Tight Skin Mice (TSK) has been documented (Hembry et al., 1986). TSK mice have a three-week delay in the onset of wound contraction, at which time high densities of myofibroblasts are observed by the stained cytoplasmic stress fibers of resident cells in the granulation tissue of the wound bed. These myofibroblasts are replaced by fibroblasts from TSK mouse wounds at 3 weeks, when wound contraction commences. The fibroblasts are replaced by mobfibroblasts at 5 weeks, when wound contraction is nearly complete. Fibroblasts are present during active wound contraction. These experiments are based on the morphological assessment of a pathophysiological process. The same type of experimental evidence has been used to support the importance of myofibroblasts in wound contraction (Majno et al., 1971, Gabbiani et al., 1972, Ryan et al.. 1974). It is proposed that fibroblasts promote collagen reorganization by moving collagen fibers. As fibroblasts in the healing wound continue to proliferate and migrate, their density in the healing wound site increases. In addition, a contracting open wound rapidly diminishes the volume at the healing site, which also contributes to increasing cell density. These activities cause many cell-tocell attachments, which in turn inhibit cell locomotion, when fibroblasts have no place to move. The inhibition of the generation of locomotive forces on the cell surfaces of the contact-inhibited cells somehow condenses the microfilaments into large stress fibers. Frustrated locomotion causes the development of cytoplasmic stress fibers and the appearance of myofibroblasts morphology in healing wounds. Acknowledgements

The work was supported by funding from the Shriners of North America and a NIH award, grant GM-32705.

References Abercrombie, scorbutic

M., Flint, M. H. and James, D. W. 1956. Wound guinea pigs. J. Embryol. Exp. Morph. 4, 167-175.

contraction

in relation

to collagen

formation

in

CELL

LOCOMOTION

FORCES

Bell, E., Ivarsson, B. and Merrill, human fibroblasts of different

VERSUS

CELL

CONTRACTION

FORCES

417

C. 1979. Production of a tissue-like structure by contraction of collagen lattice by proliferative potential in t&o. Proc. N&l. Acad. Sci. (USA), 76, 1274-1278.

Billingham, R. E. and Russell, P. S. 1956. Studies on wound healing, with special reference to the phenomenon of contracture in experimental wounds in rabbits’ skin. Ann. Surg. 144, 961-981. Ehrlich, H. P. 1988. The modulation of contraction of fibroblast population collagen lattices by types I, II and III collagen. Tiss. CeN. 20, 47-50. Ehrlich, H. P. and Rajaratnam, J. B. M. 1986. ATP-induced cell contraction in dermal fibroblasts: Effects of CAMP and myosin Light chain kinase. 1. Cell. Physiol. 128, 22LL230. Ehrlich, H. P. and Griswold, T. R. 1984. Epidermolysis bullosa dystrophica recessive fibroblasts produce increased concentrations of CAMP within a collagen matrix. J. Iwest. Dermarol. 83, 23&233. Gabbiani, G., Ryan, G. B. and Majno, G. 1971. Presence of modified fibroblast in granulation tissue and their possible role in wound contraction. Experientia 27, 549551. Gabbiani, G., Hirschel, B. J., Ryan, G. B., Stalkov, P. R. and Majno, G. 1972. Granulation tissue as a contractile organ: A study of structure and function. J. Exp. Med. 135, 719-734. Goldman, R. D., Schloss, J. A. and Starger, J. M. 1976. Organizational changes of actin like microfilaments during animal cell movement. In: Cell Motiliry. R. Goldman, T. Pollard and J. Rosenbaum, eds. Cold Spring Harbor Laboratory, New York. pp. 217-245. Harris, A. K., Wild, P. and Stopak, D. 1980. Silicone rubber substrate: A new wrinkle in the study of cell locomotion. Science 280, 177-179. Hembry, R. M., Bernanke, D. H., Hayakashi, K., Trelstad, R. L. and Ehrlich, H. P. 1986. Morphologic examination of mesenchymal cells in healing wounds of normal and tight skin mice. Am. J. Pathol. 125, 81-89. Hirschel, B. J., Gabbiani, G., Ryan, G. B. and Majno, G. 1971. Fibroblasts of granulation tissue: Immunofluorescent staining with anti-smooth muscle serum. Proc. Sm. Exp. Biol. Med. 138, 466469. Hoffman-Berling, H. 1954. Adenosentriphosphat als betriebsstoff von zellbewegumgen. Biochem. Biophys. Acra 14, 182-184. Majno, G., Gabbiani, G., Hirschel, B. J. and Ryan, G. B. 1971. Contraction of granulation tissue in vitro: Similarity to smooth muscle. Science 173, 548-550. Rungger-Berling, E. and Gabbiani, G. 1983. The role of cytoskeletal and contractile elements in pathologic process, Am. J. Pafhol. 110, 359-392. Ryan, G. B., Cliff, W. J., Gabbiani, G., Irle, C., Montandon, D., Statkov, P. R. and Majno, G. 1974. Myofibroblasts in human granulation tissue. Hum. Path& 5, 55-67. Tomasek, J. J., Haaksma, C. J., Eddy, R. T. 1989. Rapid contraction of collagen lattices by myofibroblasts is dependent upon organized actin microfilaments. J. Cell Biol. 170, Abs # 3410.

Cell locomotion forces versus cell contraction forces for collagen lattice contraction: an in vitro model of wound contraction.

Cultured human dermal fibroblasts suspended in a rapidly polymerizing collagen matrix produce a fibroblast-populated collagen lattice. With time, this...
3MB Sizes 0 Downloads 0 Views