Cell adhesion and chondrogenesis in brachypod mouse limb mesenchyme: fragment fusion studies

Transcription

1 /. Embryol. exp. Morph. Vol. 48, pp , 1978 Printed in Great Britain Company of Biologists Limited 1978 Cell adhesion and chondrogenesis in brachypod mouse limb mesenchyme: fragment fusion studies By JACKIE DUKE 1 AND WILLIAM A. ELMER 2 From the Department of Biology, Emory University, Atlanta SUMMARY This study is a continuing investigation of the effect of the brachypod mouse mutation on cell interactions and chondrogenesis during early limb development. In this report, cell adhesiveness was assessed in fused fragments of brachypod and normal limb-bud mesenchyme. Examination of the interface of fused distal postaxial limb fragments show brachypod limb mesenchyme to be more adhesive than normal limb mesenchyme. Chondrogenesis within brachypod fragments is delayed and less extensive than in normal fragments. In addition, chondrogenesis within normal fragments is not affected by the juxtaposition of the brachypod fragment, and vice versa. INTRODUCTION Brachypodism is a limb anomaly of the mouse which severely affects the development of those limb elements derived from the postaxial mesenchyme. It has been suggested that the mutation acts by interfering with the mesodermal condensation process (Griineberg & Lee, 1973). It was reported that the disturbance in the formation of the blastemata could be related to the abnormal behavior observed when mutant cells were grown in monolayer cultures (Elmer & Selleck, 1975). An effect of the mutation on cell-cell interactions was recently demonstrated when 12-day brachypod limb mesenchymal cells were placed in rotation culture. Under these conditions it was shown that the mutant cells aggregate more rapidly than normal cells, suggesting that the brachypod cells are more adhesive (Duke & Elmer, 1977). Although the more rapid loss of single cells in the mutant rotation reaggregation cultures suggested that these cells were more adhesive, other factors that could affect the cell's adhesive properties could not be ruled out conclusively. For example, the apparent differences in adhesiveness between the two cell types could have been attributed to a difference in cell size, differential cell lysis 1 Author's address: Department of Pediatrics, Grady Memorial Hospital, Atlanta, Ga , U.S.A. 2 Author's address (for reprints): Department of Biology, Emory University, Atlanta, Georgia 30322, U.S.A.

2 162 J. DUKE AND W. A. ELMER during culture, or a differential effect of the dissociation treatment on the surface of the cells. All of the evidence available suggests, however, that these factors did not play a role in the above system since (I) no differences in cell size have been observed (Hewitt & Elmer, 1976), (2) there was no observable increase in broken cells in the brachypod cultures, and (3) both the rate of attachment of these cells to culture flasks and the amount of [ 125 I] bound to the surface proteins is the same for both cell types under the dissociation conditions that were described (Hewitt & Elmer, 1978). Nevertheless, it is important to assess adhesiveness by another method. It is well known in fusion studies employing unlike tissues, such as heart and liver, liver and limb-bud (Holtfreter, 1943; Steinberg, 1962; Wiseman, Steinberg & Phillips, 1972), differently treated tissues (Morris, 1976; Phillips, Wiseman & Steinberg, \911 a), or tissues of different embryonic ages (Lesseps, 1973), that one fragment may totally envelope the other. When the interface between two fragments is straight, the tissues are assumed to be equally adhesive (Niederman & Armstrong, 1972; Moyer & Steinberg, 1976). However, if the interface is curved, the fragment exhibiting a convex boundary in the region of apposition is regarded as being more adhesive than the fragment with a concave boundary (Moyer & Steinberg, 1976). The configuration of the interface is therefore a reflexion of and precursor to the equilibrium configuration assumed by combinations of tissue fragments in which the more adhesive tissue becomes enveloped by the less adhesive tissue. This paper reports observations made on the fusion pattern of blocks of tissue removed from the postaxial region of normal and brachypod hindlimbs. The data show that the brachypod fragment exhibits a convex boundary, indicating it is more adhesive than the normal fragment. In addition, differences observed in the pattern of chondrogenesis within the fragments during fusion further confirm the abnormal differences observed in the rotation reaggregation studies (Duke & Elmer, 1977). MATERIALS AND METHODS Twelve-day-stage embryos were obtained from timed matings of homozygous normal ( + / + ) or brachypod (bp H /bp H ) mice. The time of vaginal plug appearance as well as morphological characteristics of the embryos were used for age determination (Krotoski & Elmer, 1973). Distal fragments cut from ectoderm-free postaxial halves of the hindlimbs (Duke & Elmer, 1977) were washed twice in Tyrode's solution and then transferred to 60 mm glass Petri dishes containing 3 ml of BGJ medium (Difco), with 25 % fetal calf serum (Gibco) and 50 /*g/ml Gentocin (Schering). Twenty fragments were cultured per dish and allowed to round up overnight at 37 C in a humidified atmosphere of 95 % air: 5 % CO 2. One-half of the cultures of each genotype was labeled with 10 /tci/dish of methyl- 14 C-thymidine (Amer-

3 Cell adhesion and chondrogenesis 163 sham Radiochemical Laboratory, specific activity 56 mci/mmol) at the beginning of the culture period. After 24 h of culture the fragments were washed thoroughly with isotope-free medium, then placed in additional isotope-free medium and allowed to stand for 30 min to permit free isotope to diffuse out. After a final wash the fragments were placed in hanging-drop cultures suspended from tops of siliconized 60 mm Petri dishes in the following combinations: bp H /bp H * and +/ +, bp H /bp B * and bp"/bp H, +/ + * and bp H /bp H, + / + * and +/+ (asterisk refers to fragment containing radioactivity). Additional medium was placed in the bottom of the dish to retard evaporation from the drops, and the cultures were incubated at 37 C in a humidified atmosphere of 95 % air: 5 % CO 2. Fused fragments were removed at 24, 30, and 72 h, washed with Tyrode's solution, fixed in 10 % buffered formalin, paraffin embedded and sectioned at 6 /tm. Slides were hydrated, coated with NTB-2 emulsion (diluted 1:1 with water), stored for 2 weeks in light-tight boxes, developed in Dektol (Kodak) and stained with 0-1 % toluidine blue as previously described (Duke & Elmer, 1977). RESULTS A total of 16 genotypically like (bp H /bp H *:bp H /bp H or +/ + *:+/ + ) and 16 genotypically different (bp u /bp H *: +/+ or + / + *: bp H /bp H ) fragments were placed in contact in hanging-drop cultures for a period of 3 days. At the end of three time-periods (24, 30, and 72 h) tissue combinations were removed, fixed, sectioned, and examined in serial autoradiographs to ascertain whether cells of one tissue migrated over the surface of the second tissue to envelop it. It can be seen in Fig. 1A, B, and C and Table 1 that the cells within the brachypod fragments are more adhesive. In 13 of the 16 unlike combinations (81-25 %) the brachypod fragments exhibited a convex boundary (Table 1). In the remaining three combinations (18-75 %), the boundary between the tissues was straight (Table 1). Over the 3-day culture period a reversal of adhesiveness was never observed; that is, none of the genotypically different combinations showed a brachypod fragment enveloping a normal fragment. With all of the 16 genotypically like combinations, a straight interface along the boundary of fusion was observed (Fig. 1D, Table 1). Migration of labeled cells into unlabeled regions was not seen and the interface between fragments remained sharp, in accordance with the results obtained by other investigators (Steinberg, 1962; Weston & Abercrombie, 1965). There were in a few cases slight amounts of silver grains scattered over the unlabeled tissue in both like and unlike combinations. This could possibly be due to the presence of nonincorporated [ 14 C-]thymidine not removed during the extensive wash and/or the incorporation of the labeled precursor released by autolysis of the cultured cells (Saunders, 1966; Niederman & Armstrong, 1972). The chondrogenic differences which were reported in the rotation reaggregation study (Duke & Elmer, 1977) were also observed in the fragment fusion

4 164 J. DUKE AND W. A. ELMER D Fig. 1. (A) Autoradiograph of labeled normal fragment (N) fused for 24 h with unlabeled brachypod fragment (Bp). Condensation can be easily recognized in normal fragment, x 100. Arrow points to interface showing the normal fragment enveloping the mutant fragment. (B) Autoradiograph of labeled mutant fragment (Bp) fused for 30 h with unlabeled normal fragment, x 100. (C) Autroradiograph of labeled mutant fragment (Bp) fused for 72 h with unlabeled normal fragment. Note extensive cartilage formation in normal fragment, x 100. (D) Autoradiograph of labeled mutant fragment (Bp) fused for 30 h with unlabeled mutant fragment (Bp). x 100. Arrow points to straight interface.

5 Cell adhesion and chondrogenesis 165 Table 1. Analyses of fragment fusions between genotypically like and genotypically unlike mouse postaxial hindlimb tissue Relative adhesiveness based on shape of interface Fusion time (h) Bp H /bp H > Experimental combinations* Bp H /Bp H /+ = Bp H /Bp H Control ccombinations*! + /+ = + / A Bp H /Bp H = Bp H /Bp H * Fragments were allowed to round up in culture 24 h prior to fusion either in the presence or absence of [ 14 C-]thymidine (see Methods and Materials), f All combination of genotypically like tissues exhibited a straight interface. study. After 24 h of fusion, normal fragments typically had 4-5 areas of condensation, whereas, brachypod fragments had only 1-2 areas of condensation (Fig. 1 A). Over the 3-day culture period, however, condensations within normal fragments appeared to expand and fuse. In brachypod fragments the extent of expansion and fusion was less and in some cases new condensations were initiated. The initiation of new condensations within fragments is in contrast to the results obtained in the rotation reaggregation cultures. In the latter case, new condensations were not initiated in either genotype after the first 24 h of culture (Duke & Elmer, 1977). This difference is probably related to differences in the two culture systems employed. The delay in differentiation of brachypod limb mesodermal cells seen in vivo (Milaire, 1965) in limb explants (Duke & Elmer, 1976), in monolayer cultures (Elmer & Selleck, 1975), and in aggregates (Duke & Elmer, 1977) was also observed in the tissue fragments. After 24 and 30 h of fusion (48-54 total hours of culture) histotypic regions of cartilage are seen in normal fragments, but not in brachypod fragments. After 3 days of fusion, brachypod fragments contain cartilaginous areas, but the amount of cartilage is less extensive than in normal fragments (Fig. 1C). It was also observed that limb tissue of one genotype failed to influence the differentiation of the limb tissue of the other genotype. This is of interest, since it has been reported that brachypod cartilaginous limb anlagen contain a diffusable proteinaceous inhibitor that influences the growth and differentiation of normal cartilaginous limb anlagen in organ culture (Konyukhov & Ginter, 1966; Konyukhov & Bugrilova, 1968; Pleskova, Rodionov, Bugrilova & Konyukhov, 1974). In the present study no difference was observed in the amount of cartilage present in normal fragments combined with either each other or brachypod tissue. Likewise, brachypod fragments combined with either each other or

6 166 J. DUKE AND W. A. ELMER normal tissue differentiated at the same rate and contained the same number and size of cartilaginous regions. DISCUSSION Fragment fusion studies employing labeled and unlabeled tissue masses from normal and brachypod 12-day limb-buds confirm that brachypod limb mesoderm is more adhesive than normal limb mesoderm. Complete envelopment of one fragment by another did not occur, indicating that the difference in adhesiveness between the two genotypes is relatively small when compared, for example, with the difference between chick precartilage and embryonic heart. In the fusion studies between fragments of normal and brachypod limbs, it is probable that the partial envelopment observed is indeed an equilibrium situation, since prolonged culture did not lead to complete envelopment, and no reversal of tissue assembly occurred. Reversal of assembly patterns within fused fragments is attributed to changes in cohesiveness during the culture (Niederman & Armstrong, 1972; Wiseman et ah 1972; Phillips et ah 1977a). These changes are thought to be due to in vitro maintenance of tissue, since they do not occur as part of the normal development sequence (Niederman & Armstrong, 1972; Phillips et ah 1977a). The lack of tissue reversal in our system is another indication that brachypod tissue is more adhesive than normal tissue. As the fragments develop in culture, the cells become increasingly immobilized as matrix is laid down (Burdick, 1970) and are unable to slide past one another as required for tissue reorganization (Phillips, Steinberg & Lipton, 1977&). However, matrix production in brachypod fragments is minimal during the early phases of culture and the lack of spreading of brachypod tissues on to normal fragments must be due primarily to the strength of brachypod cell-cell adhesions, not to increasing rigidity of the tissue. Chondrogenic differences apparent in rotation culture were also noted in fragment fusion studies. The 4-5 condensations in normal fragments after 24 h of fusion probably represent portions of digital, tarsal, metatarsal, and fibular precartilage condensations. Brachypod fragments have only 1-2 condensations after 24 h of fusion, although the number of condensations increases over the 3-day culture period. The delay in formation of condensations in the distal region of the brachypod limb was also evident in vivo (Griineberg & Lee, 1973). Precartilage condensations in brachypod aggregates were also smaller than normal, but in the aggregate system, no new condensations were initiated after 24 h of culture. If the formation of precartilage condensations within aggregates is due to a sorting out of precartilage cells within the aggregates, or perhaps a selective reassociation of cells, precartilage cells that were unable to sort out, and are surrounded by non-chondrogenic cells would probably not differentiate into cartilage. Precartilage cells in fragments, however, are not displaced from their contacts with other precartilage cells, and can go on to differentiate.

7 Cell adhesion and chondrogenesis 167 Condensations within normal fragments increased greatly in size, with resultant fusion of adjacent condensations, but in brachypod fragments little expansion of pre-existing condensations occurred and the amount of cartilage formed was less. Lack of condensation expansion was also observed in brachypod aggregates, and in both systems may be due to a lack of chondrogenic cells within brachypod limbs, and/or a lack of recruitment of cells from surrounding mesenchyme. Other disturbances in cell movement and adhesion related to abnormal limb morphogenesis have been observed in the chick mutant, talpid 3 (Ede, 1976). Time-lapse cinematography studies of wing mesenchyme fragments of normal and talpid 3 chick embryos explanted in plastic petri dishes showed the mutant cells moved much more slowly than normal cells (Ede & Flint, 1975). The difference in rate of movement appeared to be related to differences in cell morphology. The mutant cells were more flattened and during movement small spiky microvilli all around the cell periphery were observed, rather than the long cyloplasmic filopodia seen in the elongated normal cells (Ede & Flint, 1975). In a different study, Niederman & Armstrong (1972) reported that they were unable to detect adhesive differences by cell sorting or fragment fusion between limb mesenchyme from normal and taplid 2 chick embryos, a mutant which has a similar phenotype as talpid 3 (Abbott, Talylor & Abplanalp, 1959). Since different techniques may measure different aspects of cell adhesion, it is possible that (1) fragment fusion techniques will not detect cell adhesion differences between the talpid mutants and control mesenchymes and/or (2) the property of cell adhesion is affected differently by the talpid 3 allele than it is by the talpid 2 allele. It is clear that the studies on the talpid and brachypod mutants only serve to underline the complexity of the cell surface in terms of the mechanics of cell adhesion. REFERENCES ABBOTT, U. K., TALYLOR, L. W. & ABPLANALP, H. (1959). A second talpid-like mutation in the fowl. Poultry Sci. 38, BURDICK, M. L. (1970). Cell sorting out according to species in aggregates containing mouse and chick embryonic limb mesoblast cells. /. exp. Zool. 175, DUKE, J. & ELMER, W. A. (1976). Effect of the brachypod mutation on cell interaction during blastema formation in the mouse limb. Am. Zool. 16, 246 (Abstr). DUKE, J. & Elmer, W. A. (1977). Effect of the brachypod mutation on cell adhesion and chondrogenesis in aggregates of mouse limb mesenchyme. J. Embryo!, exp. Morph. 42, EDE, D. A. & AGERBAK, G. S. (1968). Cell adhesion and movement in relation to the developing limb pattern in normal and talpid 3 mutant chick embryos. /. Embryol. exp. Morph. 20, EDE, D. A. & FLINT, O. P. (1975). Intercellular adhesion and formation of aggregates in normal and talpid 3 mutant chick limb mesenchyme. /. Cell Sci. 18, EDE, D. A. (1976). Cell interactions in vertebrate limb development. In The Cell Surface in Animal Embryogenesis and Development (ed. G. Poste & G. L. Nicolson), pp Amsterdam: North-Holland. ELMER, W. A. & SELLECK, D. K. (1975). In vitro chondrogenesis of limb mesoderm from normal and brachypod mouse embryos. /. Embryol. exp. Morph. 33,

8 168 J. DUKE AND W. A. ELMER GRUNEBERG, H. & LEE, A. J. (1973). The anatomy and development of brachypodism in the mouse. J. Embryol. exp. Morph. 30, HEWITT, A. T. & ELMER, W. A. (1976). Reactivity of normal and brachypod mouse limb mesenchymal cells with Con A. Nature, Lond. 264, HEWITT, A. T. & ELMER, W. A. (1978). Developmental modulation of lectin-binding sites on the surface membranes of normal and brachypod mouse limb mesenchymal cells. Differentiation 10, HOLTFRETER, J. (1943). A study of the mechanics of gastrulation: Part I. J. exp. Zool. 94, KONYUKHOV, B. V. & GINTER, E. (1966). A study of the action of the brachypodism-h gene on development of the long bones of the hind limb in the mouse. Folia biol., Praha 12, KONYUKHOV, B. V. & BUGRILOVA, R. S. (1968). The growth-inhibiting factor in embryos of mutant stock brachypodism-h mice. Folia biol., Praha 14, KROTOSKI, D. M. & ELMER, W. A. (1973). Alkaline phosphatase activity in fetal hind limbs of the mouse mutation brachypodism. Teratology 7, LESSEPS, R. J. (1973). Developmental changes in morphogenetic properties: Embryonic chick heart tissues and cells segregate from other tissues in age-dependent patterns. /. exp. Zool. 185, MILAIRE, J. (1965). Etude morphogenetique de trois malformations congenitales de l'autopode chez la souris (syndactylisme-brachypodisme-hemimelie dominante) par des methodes cytochimiques. Mem. Acad. r. Belg. (Cl. Sci.) 4, 2me ser., 16, 120. MORRIS, J. E. (1976). Cell aggregation rate vs. aggregate size. Devi Biol. 54, MOYER, W. A. & STEINBERG, M. S. (1976). Do rates of intercellular adhesion measure the cell affinities reflected in cell-sorting and tissue-spreading configurations? Devi Biol. 51, NIEDERMAN, R. N. & ARMSTRONG, P. B. (1972). Is abnormal limb bud morphology in the mutant ta/pid 2 chick embryo a result of altered intercellular adhesion? Studies employing cell sorting and fragment fusion. J. exp. Zool. 181, PHILLIPS, H. M., WISEMAN, L. L. & STEINBERG, M. S. (1977a). Self vs. nonself in tissue assembly: Correlated changes in recognition behavior and tissue cohesiveness. Devi Biol. 57, PHILLIPS, H. M., STEINBERG, M. S. & LIPTON, B. H. (19776). Embryonic tissues as elasticoviscous liquids. II. Direct evidence for cell slippage in centrifuged aggregates. Devi Biol. 59, PLESKOVA, M. V., RODIONOV, V. M., BUGRILOVA, R. S. & KONYUKHOV, B. V. (1974). The partial purification of growth-inhibiting factor of the brachypodism-h mouse embryos. Devi Biol. 37, SAUNDERS, J. W. (1966). Death in embryonic systems. Science, N.Y. 154, STEINBERG, M. S. (1962). On the mechanism of tissue reconstruction by dissociated cell. III. Free energy relationships and the reorganization of fused, heteronomic tissue fragments. Proc. natn. Acad. Sci. U.S.A. 49, WESTON, J. A. & ABERCROMBIE, M. (1965). Cell mobility in fused homo- and heteronomic tissue fragments. /. exp. Zool. 164, WISEMAN, L. L., STEINBERG, M. S. & PHILLIPS, H. M. (1972). Experimental modulation of intercellular cohesiveness: reversal of tissue assembly patterns. Devi Biol. 28, (Recieved 25 April 1978, revised 3 July 1978)