- Email: [email protected]
Accelerated maturation of limb mesenchyme by the BrachypodH mouse mutation
Differentiation
Differentiation (1983) 24: 145-148
0 Springer-Verlag
1983
Accelerated maturation of limb mesenchyme by the BrachypodHmouse mutation Elizabeth M. Owens and Michael Solursh Department of Zoology, University of Iowa, Iowa City, IA 52242, USA Abstract. Mesenchyme cell populations prepared from proximal and distal halves of stage 20 mouse forelimb buds are shown to behave under in vitro micromass culture conditions like analogous cell populations obtained from chick embryo limb buds. While the distal cells are spontaneously chondrogenic, the proximal cells make aggregates which are only potentially chondrogenic after treatment with dibutyryl cyclic AMP. In addition, stage 20 mouse whole limb bud cells homozygous for the brachypodismH (bp") mutation are shown to behave similarly to 'normal' proximal cells. Both make fewer aggregates and nodules and both have faster aggregation rates (determined as the rate of disappearance of single cells over time) in rotation cultures than 'normal' distal or whole limb bud cells. These results support the hypothesis that the bpHmutation specifically decreases the proportion of spontaneously chondrogenic mesenchyme cells (that is, distal-like cells) present at certain developmental stages in the limb bud, resulting in a prematurely high proportion of proximal-like cells.
Introduction Brachypodism" (bp") [5]is a single gene, recessive mutation in mice that causes dwarfed limbs, apparently as the result of an abnormality first manifested during early skeletal development [6].The effects of the mutation can first be recognized on day 12 of gestation (stage 20) ([lq for mouse embryonic stages), when the initial formation of the precartilage mesenchymal condensations in bp" forelimbs is apparently delayed [7, 91. The mutation does not affect the chemical composition of cartilage matrix components [4, 141. Instead, it may cause a specific cell subpopulation, responsible for inducing certain early steps in chondrogenic differentiation, to have abnormally reduced inductive capabilities [ll]. Several observations made on bp" limb bud cells in vitro, however, are not readily understood in the light of this hypothesis. For instance, cells obtained from day 12 bpHwhole hindlimb buds have been found to exhibit higher levels of surface adhesion than cells obtained from 'normal' limb buds of the same embryonic stage [2, 31. In this paper, we describe several recent observations which suggest that only some in vitro bpHcell behaviors are directly related related to the effects of bp" gene. Analysis of the early inductive steps involved in limb Supported by NlH grant HD 05505
bud chondrogenesis is complicated by the fact that the proportion of the inducing cell type is high in early stage limb buds (e.g., stage 17) but decreases progressively with increasing developmental stage [lo, 111. At the same time the proportions of other cell types, including non-chondrogenic mesenchyme and muscle cells, increase in the limb buds. Recently, Swalla et al. [16] reported that a cell type capable of serving as an effective chondrogenic inducer under specific in vitro conditions is concentrated in the distal, rather than the proximal, half of stage 23-24 chick wing buds. The proximal cell type was found to have some phenotypic similarities in vitro to the 'protodifferentiated' cells obtained previously from stage 26 chick limb buds [13, 151. It was hypothesized that the two cell types (distal and proximal) represented two stages through which cells undergoing chondrogenic differentiation will pass. If this is true, it is possible to obtain populations of inductive and non-inductive mesenchyme cells for further study. Results presented in this paper suggest that a similar separation of cell types can also be accomplished by the isolation of proximal and distal mouse limb bud pieces. Furthermore, it is found that several of the phenotypes expressed by stage 20 (early day 12) proximal mouse limb bud cells in vitro bear a striking resemblance to those of whole limb bpH cells from the same stage embryos. For example, proximal cells invariably demonstrate higher levels of cell adhesion in rotation culture than distal cells. These findings suggest that the proportion of proximal-like cells is higher in whole stage 20 bpH limb buds than in whole stage 20 'normal' limb buds. Methods Cell suspensions
Stage 20 forelimb buds were obtained from abpHlabpH, a'+ ae (purchased from Jackson Laboratory, Bar Harbor, Maine, and referred to as 'bp"' and ' / ', respectively), and C3H strain mouse embryos. Details concerning the bpH and + / + mice are described by Owens and Solursh [ 113. Cell suspensions were prepared from either whole limb buds, proximal half pieces, or distal half pieces (cut in the manner illustrated in Fig. 1) by one of two methods: (1) dissociation of the limb buds following incubation of the limb bud pieces with trypsin, as described by Owens and Solursh [lo], or (2) dissociation following incubation in calcium-magnesium-free (CMF) Tyrode's solution. In the
+
++
146
1
proximal
Fig. 1. A diagram of a stage 20 mouse forelimb bud. The dashed line indicates where the limbs were cut to separate the proximal and distal limb halves. A photograph of a longitudinal section through a stage 20 C3H mouse forelimb bud has been published previously [101
latter procedure, the limb pieces were cut into several smaller parts to increase the available surface area, then placed in CMF Tyrode’s solution in a 37” C shaker bath for 10 min. The pieces were then dissociated mechanically to obtain a cloudy cell suspension. These cells were resuspended in CMRL medium (Grand Island Biological Co.) plus antibiotics [l]and immediately filtered through two layers of #20 Nitex mesh. The cell counts of the suspensions were made with a hemacytometer. Micromass cultures
Micromass cultures were prepared by placing single 5 p1 dots of CMRL medium (supplemented with 10% fetal calf serum [Grand Island Biological Co.] and antibiotics) containing 2 x lo5 cells in 35 mm Falcon tissue culture dishes. The cells were allowed two hours to attach before the dishes were flooded with 2 ml of the same medium. The culture medium in each dish was aspirated and replaced with 1 ml of medium after 24 hours and again 24 hours after that. In some cases, cultures were treated with 1 mM N6,0’dibutyryl adenosine 3‘: 5’-monophosphoric acid (dibutyryl cyclic AMP; Sigma), which was included in each of the 1 ml aliquots of medium used to feed the cultures after 24 and 48 hours of culture. All micromass cultures were fixed in Kahle’s fixative [8] after 72 hours. They were stained first with Alcian blue at pH 1 to allow the localization of cartilage nodules, then with Carazzi’s hematoxylin for localization of cell aggregates. The aggregates and nodules in each culture were counted as described previously [lo], while the areas of the cultures were determined as described in Owens and Solursh [ll]. It should be noted that we have not observed any major differences between the basic patterns of behavior of C3H and +/+ (the ‘wild type’ counterpart of the bp” mouse strain used in this paper) forelimb bud cells under the micromass culture conditions used here. The + / + mice, however, have proven significantly more difficult to breed than the C3H strain. Thus, certain direct comparisons between bpH and proximal cultures are made possible by the use of both types of ‘normal’ mouse strains. Rotution cultures
Immediately before use in rotation culture, cell counts were made of the cell suspensions in culture medium without serum. Two x lo6 cells in 1.8 ml of medium (pregassed with
Fig. 2. Seventy-two hour micromass cultures prepared by the trypsin dissociation method and stained only with Alcian blue. Included are cultures prepared from stage 20 proximal C3H forelimb bud halves (a, b); distal C3H forelimb and halves (c); whole C3H forelimb buds (a); whole bpa forelimb buds (e, f), bud whole +/+ forelimb buds (9). The cultures in b and f were treated with dibutyryl cyclic AMP from 2 6 7 2 hours. All of the C3H cultures were prepared at the same time from a single group of well matched stage 20 forelimb buds. x 10
0.5% CO, in air) were placed in 30 ml Nalge screw-top bottles, and the cell count was checked again. At time zero, 0.2ml of fetal calf serum was added to each bottle. The bottles were then gassed and quickly placed in a 37” C water shaker bath at 70 rotations/min (0.65 an radius of gyration). Samples were withdrawn at regular intervals, after which each bottle was gassed and returned to the water bath. The number of single cells in each sample was counted with a hemacytometer and averaged.
147
Table 1. Aggregate and nodule formation' Cell source
Proximal C3H Distal C3H Whole C3H Whole bf Whole +/ +
Aggregates/mm2
Nodules/mm2
Untreated
Dibutyryl cAMP treated'
Untreated
Dibutyryl cAMP treated'
15.36k0.44 26.36f0.24 22.98k0.71 13.02f1.21 21.2 f1.07
22.42 f 0.71 23.70k0.94 24.09f 0.83 18.40f 0.23 26.26k 1.88
8.54&0.99 28.96k0.32 22.09 f 0.43 6.80k0.51 22.53 k 1 .O
19.70k1.58 22.02k 2.43 21.57+0.52 16.88f0.38 25.79 k0.73
a
All cultures were prepared from stage 20 limb buds and all were fixed after 72 h of culture. Each mean is calculated from data obtained from three separate cultures Aggregate and nodule densities are per millimeter2of culture area +SD ' Treated cultures were fed medium containing 1 mM dibutyryl cAMP after the first 24 h of culture and again after 48 h
Results Proximal and distal micromass cultures
Micromass cultures prepared from stage 20 whole mouse forelimb buds form numerous aggregates, the majority of which differentiate into cartilage nodules by 72 hours of culture [lo, 181. Typical 72 hour cultures prepared by the trypsin dissociation method from whole stage 20 C3H and + / + mouse forelimb buds and stained with Alcian blue are shown in Figs. 2d and g, respectively. These cultures each contain numerous cartilage nodules. If, however, the proximal and distal halves of stage 20 C3H limb buds are cut apart (Fig. 1) and used separately to prepare micromass cultures, the results are very different. Cultures prepared from the proximal halves ('proximal cultures'; Fig. 2a) form substantially fewer aggregates than whole limb bud cultures, only a few of which ever differentiate into cartilage nodules. In a typical example given in Table 1, the proximal cultures contain 33% fewer aggregates/mm2 than whole limb cultures, while only 56% of the aggregates formed ever become nodules, compared to 96% in the whole limb cultures. Cultures prepared from the distal halves ('distal cultures'; Fig. 2c), on the other hand, form numerous aggregates, most of which undergo cartilage differentiation (Table 1). The differences observed between whole, proximal, and distal cultures are the same, if not actually more pronounced, if the limb pieces are dissociated following incubation in CMF Tyrode's (data not shown) rather than trypsin. The in vitro behavior of proximal stage 20 mouse limb bud cells under these micromass culture conditions is clearly reminiscent of the in vitro behavior of cells obtained from whole stage 20 bp" limb buds [2,4, 11, 131. In the example in Table 1, for instance, the stage 20 bp" whole limb cultures form approximately 38% fewer aggregates than control +/+ cultures. Furthermore, only 52% of the bp" aggregates differentiate into nodules by 72 hours, compared to 100% of the +/+ aggregates. A stage 20 bpR whole limb bud culture stained with Alcian blue is shown in Fig. 2e. Response to dibutyryl cyclic AMP
Stage 24 chick proximal limb bud cultures [16] respond to treatment with dibutyryl cyclic AMP with a dramatic
increase in the proportion of aggregates that differentiate into cartilage nodules. A similar effect is observed when mouse proximal cultures are treated with dibutyryl cyclic AMP. Figure 2 b shows a culture prepared from proximal forelimb mouse cells [C3H] that had been treated from 24-72 hours of culture with the drug. The dibutyryl CAMPtreated proximal mouse cultures in Table 1 contained 46% more aggregates/mm2 and 130% more nodules/mm2 than the untreated cultures. Similarly treated distal mouse cultures did not show an increase in the number of nodules formed (Table 1). Stage 20 whole limb bpR cultures respond to treatment with dibutyryl cyclic AMP very much in the same manner as proximal C3H cultures (compare Figs. 2e and 2f with Figs. 2a, b). Dibutyryl cAMP treated stage 20 bp" cultures produced 41 % more aggregates/mm2 and 148% more nodules/mm2 than the untreated bp" cultures. Nearly all of the treated aggregates differentiated into cartilage nodules. Neither the +/+ cultures nor the whole C3H limb bud cultures responded in this extreme manner to dibutyryl CAMP. Aggregation of proximal and distal cells in rotation culture
It has been suggested that the bp" abnormality results in increased cell surface adhesion (2, 31. It was shown, for example, that the rate of loss of single cells in rotation cultures composed of day 12 bp" mouse forelimb cells was significantly greater than in similar cultures composed of 'normal' limb bud cells. Figure 3 shows the results of similar tests carried out using stage 20 proximal and distal limb bud cells from C3H mouse embryos. The proximal and distal cells used in these tests were obtained from the same limb buds, and the tests were carried out simultaneously. While great care was taken to assure that the suspensions contained the same numbers of cells at time zero (see Methods), at all later times there were always more single cells in the distal suspension than the proximal suspension. The results show that the number of single cells in the proximal cell suspension declines more rapidly than in the distal cell suspension. Repetitions of this experiment are highly consistent. When chick stage 24 proximal and distal cells are used in identical tests, the proximal cells also consistently aggregate more rapidly than distal cells (unpublished data).
148
cells obtained from whole stage 20 bp” limb buds may not illuminate further the nature of the primary bp” abnormality. This is because many of the in vitro behaviors manifested by stage 20 bp” limb bud cells would be expressed only as indirect consequences of the mutation. The present data support the idea that the bpH defect may not result directly from the assumption of proximal-like cell properties, but rather from defective distal-like properties. Most likely these include distal cell properties involved in certain early steps required for cell-cell interactions during chondrogenic induction [l 11.
01 0
15 30
60
90
120
150
MINUTES
Flg. 3. Loss of single cells in rotation cultures prepared with dim1 or proximal cells. Both populations of cells were obtained from the same set of stage 20 C3H forelimb buds. Suspensions of single cells (1 x lo6 cells/ml) were placed in a 37” C rotation water shaker bath as described in the Methods. At each time point, three separate samples were withdrawn from each cell suspension and the numbers of single cells were counted. Each point represents the mean of three cell counts, &standard deviations
Discussion The differences between the in vitro behaviors of proximal and distal stage 20 mouse forelimb bud cells appear to be analogous to the differences between proximal and distal cells of stage 23-24 [16] and stage 25 [12] chick embryo limb buds. In addition, similarities are observed between stage 20 bp” whole limb bud cells and ‘normal’ proximal limb bud cells in vitro. Under identical micromass culture conditions, each of these forms lower numbers of aggregates per unit area of culture than cultures prepared from whole ‘normal’ stage 20 limb bud cells. Of the aggregates that form, only a relatively small fraction in both types of culture undergoes spontaneous chondrogenic differentiation, although these fractions can be increased significantly by treatment with dibutyryl cyclic AMP. Finally, proximal forelimb bud cells exhibit a level of cell surface adhesiveness which is greater than that of distal forelimb cells, much in the same way that bpH stage 20 whole hindlimb cells have been shown to have higher adhesiveness than ‘normal’ stage 20 hindlimb cells [2, 31. The similarities between proximal and bp” cells described in this paper suggest an explanation for many of the known phenotypic differences between bpa and ‘normal’ limb mesenchyme. We suggest that the cell alterations brought about by the bpH mutation cause an apparent premature superabundance of proximal-like cells in the limb buds. By stage 20, cells from whole bp” forelimbs express in vitro proximal phenotypes [ l l ] that are not expressed by cells from whole C3H and + / + forelimbs until middle stage 21 [lo, 111. This would occur, for instance, if cells in the distal region of the bpH limb buds tended to differentiate into proximal-like cells at an earlier developmental time than distal limb bud cells in ‘normal’ limb buds. The proximal phenotype is expressed by cells from ‘normal’ limb buds at stage 20 only if the proximal cells are cultured in the absence of distal cells. One implication of these observations is that the detailed examination of individual in vitro behaviors displayed by
References 1. Ahrens PB, Solursh M, Reiter R (1977) Stage-related capacity for limb chondrogenesis in cell culture. Develop Biol60: 69-82 2. Duke J, Elmer WA (1977) Effect of the brachypod mutation on cell adhesion and chondrogenesis in aggregates of mouse limb mesenchyme. J Embryol Exptl Morphol42:209-217 3. Duke J, Elmer WA (1978) Cell adhesion and chondrogenesis in brachypod mouse limb mesenchyme fragment fusion studies. J Embryo1 Exptl Morphol48:161-168 4. Elmer WA, Selleck DK (1975) In vitro chondrogensis of limb mesoderm from normal and brachypod mouse embryos. J Embryol Exptl Morphol33 :371-386 5. Green EL (ed) (1966) Biology of the laboratory mouse, 2nd edn. McGraw-Hill, New York 6. Griineberg H (1963) The pathology of development. A study of inherited skeletal disorders in animals. Blackwell. Oxford 7. Griineberg H, Lee AJ (1973) The anatomy and development of brachypodism in the mouse. J Embryol Exptl Morphol 30:119-141 8. Guyer M F (1953) Animal micrology. University of Chicago Press, Chicago 9. Milaire J (1965) Etude morphogbnbtique de trois malformiitions congknitales de I’autopode chez la souris (syndactylismebrachypodisme-hbmimtlitedominante) par des mbthodes cytochimiques. Mem Acad R Belg Clin Sci 16: 1-20 10. Owens EM, Solursh M (1981) In vitro histogenic capacities of limb mesenchyme from various stage mouse embryos. Develop Biol88 :297-31 1 11. Owens EM, Solursh M (1982) Cellcell interaction by mouse limb cells during in vitro chondrogenesis: analysis of the brachypod mutation. Develop Biol91: 376-388 12. Paulsen DF, Parker CL, Finch RA (1979) Region-dependent capacity for limb chondrogenesis: patterns of chondrogenesis in cultures from different regions of the developing chick wing. Differentiation 14: 159-165 13. Selleck DK, Elmer WA (1973) Phenotypic instability of brachypod limb mesoblast cells in vitro. Am Zoo1 13:1313 14. Shambough J, Elmer WA (1980) Analysis of glycosaminoglycans during chondrogenesis of normal and brachypod mouse limb mesenchyme. J Embryol Exptl Morphol56:22>238 15. Solursh M, Jensen KL, Singley CT, Linsenmayer TE, Reiter RS (1982) Two distinct regulatory steps in cartilage differentiation. Develop Biol94: 31 1-325 16. Swalla BJ, Owens EM, Linsenmayer TF, Solursh M (to be published) Two distinct classes of prechondrogenic cell types in the embryonic limb bud. Develop Biol97 17. Theiler K (1972) The house mouse. Development and normal stages from fertilization to 4 weeks of age. Springer-Verlag. New York 18. Umansky R (1966) The effect of cell population density on the developmental fate of reaggregating mouse limb bud mesenchyme. Develop Biol13 :31-56 Received October 1982 / Accepted in revised form February 1983
Differentiation (1983) 24: 145-148
0 Springer-Verlag
1983
Accelerated maturation of limb mesenchyme by the BrachypodHmouse mutation Elizabeth M. Owens and Michael Solursh Department of Zoology, University of Iowa, Iowa City, IA 52242, USA Abstract. Mesenchyme cell populations prepared from proximal and distal halves of stage 20 mouse forelimb buds are shown to behave under in vitro micromass culture conditions like analogous cell populations obtained from chick embryo limb buds. While the distal cells are spontaneously chondrogenic, the proximal cells make aggregates which are only potentially chondrogenic after treatment with dibutyryl cyclic AMP. In addition, stage 20 mouse whole limb bud cells homozygous for the brachypodismH (bp") mutation are shown to behave similarly to 'normal' proximal cells. Both make fewer aggregates and nodules and both have faster aggregation rates (determined as the rate of disappearance of single cells over time) in rotation cultures than 'normal' distal or whole limb bud cells. These results support the hypothesis that the bpHmutation specifically decreases the proportion of spontaneously chondrogenic mesenchyme cells (that is, distal-like cells) present at certain developmental stages in the limb bud, resulting in a prematurely high proportion of proximal-like cells.
Introduction Brachypodism" (bp") [5]is a single gene, recessive mutation in mice that causes dwarfed limbs, apparently as the result of an abnormality first manifested during early skeletal development [6].The effects of the mutation can first be recognized on day 12 of gestation (stage 20) ([lq for mouse embryonic stages), when the initial formation of the precartilage mesenchymal condensations in bp" forelimbs is apparently delayed [7, 91. The mutation does not affect the chemical composition of cartilage matrix components [4, 141. Instead, it may cause a specific cell subpopulation, responsible for inducing certain early steps in chondrogenic differentiation, to have abnormally reduced inductive capabilities [ll]. Several observations made on bp" limb bud cells in vitro, however, are not readily understood in the light of this hypothesis. For instance, cells obtained from day 12 bpHwhole hindlimb buds have been found to exhibit higher levels of surface adhesion than cells obtained from 'normal' limb buds of the same embryonic stage [2, 31. In this paper, we describe several recent observations which suggest that only some in vitro bpHcell behaviors are directly related related to the effects of bp" gene. Analysis of the early inductive steps involved in limb Supported by NlH grant HD 05505
bud chondrogenesis is complicated by the fact that the proportion of the inducing cell type is high in early stage limb buds (e.g., stage 17) but decreases progressively with increasing developmental stage [lo, 111. At the same time the proportions of other cell types, including non-chondrogenic mesenchyme and muscle cells, increase in the limb buds. Recently, Swalla et al. [16] reported that a cell type capable of serving as an effective chondrogenic inducer under specific in vitro conditions is concentrated in the distal, rather than the proximal, half of stage 23-24 chick wing buds. The proximal cell type was found to have some phenotypic similarities in vitro to the 'protodifferentiated' cells obtained previously from stage 26 chick limb buds [13, 151. It was hypothesized that the two cell types (distal and proximal) represented two stages through which cells undergoing chondrogenic differentiation will pass. If this is true, it is possible to obtain populations of inductive and non-inductive mesenchyme cells for further study. Results presented in this paper suggest that a similar separation of cell types can also be accomplished by the isolation of proximal and distal mouse limb bud pieces. Furthermore, it is found that several of the phenotypes expressed by stage 20 (early day 12) proximal mouse limb bud cells in vitro bear a striking resemblance to those of whole limb bpH cells from the same stage embryos. For example, proximal cells invariably demonstrate higher levels of cell adhesion in rotation culture than distal cells. These findings suggest that the proportion of proximal-like cells is higher in whole stage 20 bpH limb buds than in whole stage 20 'normal' limb buds. Methods Cell suspensions
Stage 20 forelimb buds were obtained from abpHlabpH, a'+ ae (purchased from Jackson Laboratory, Bar Harbor, Maine, and referred to as 'bp"' and ' / ', respectively), and C3H strain mouse embryos. Details concerning the bpH and + / + mice are described by Owens and Solursh [ 113. Cell suspensions were prepared from either whole limb buds, proximal half pieces, or distal half pieces (cut in the manner illustrated in Fig. 1) by one of two methods: (1) dissociation of the limb buds following incubation of the limb bud pieces with trypsin, as described by Owens and Solursh [lo], or (2) dissociation following incubation in calcium-magnesium-free (CMF) Tyrode's solution. In the
+
++
146
1
proximal
Fig. 1. A diagram of a stage 20 mouse forelimb bud. The dashed line indicates where the limbs were cut to separate the proximal and distal limb halves. A photograph of a longitudinal section through a stage 20 C3H mouse forelimb bud has been published previously [101
latter procedure, the limb pieces were cut into several smaller parts to increase the available surface area, then placed in CMF Tyrode’s solution in a 37” C shaker bath for 10 min. The pieces were then dissociated mechanically to obtain a cloudy cell suspension. These cells were resuspended in CMRL medium (Grand Island Biological Co.) plus antibiotics [l]and immediately filtered through two layers of #20 Nitex mesh. The cell counts of the suspensions were made with a hemacytometer. Micromass cultures
Micromass cultures were prepared by placing single 5 p1 dots of CMRL medium (supplemented with 10% fetal calf serum [Grand Island Biological Co.] and antibiotics) containing 2 x lo5 cells in 35 mm Falcon tissue culture dishes. The cells were allowed two hours to attach before the dishes were flooded with 2 ml of the same medium. The culture medium in each dish was aspirated and replaced with 1 ml of medium after 24 hours and again 24 hours after that. In some cases, cultures were treated with 1 mM N6,0’dibutyryl adenosine 3‘: 5’-monophosphoric acid (dibutyryl cyclic AMP; Sigma), which was included in each of the 1 ml aliquots of medium used to feed the cultures after 24 and 48 hours of culture. All micromass cultures were fixed in Kahle’s fixative [8] after 72 hours. They were stained first with Alcian blue at pH 1 to allow the localization of cartilage nodules, then with Carazzi’s hematoxylin for localization of cell aggregates. The aggregates and nodules in each culture were counted as described previously [lo], while the areas of the cultures were determined as described in Owens and Solursh [ll]. It should be noted that we have not observed any major differences between the basic patterns of behavior of C3H and +/+ (the ‘wild type’ counterpart of the bp” mouse strain used in this paper) forelimb bud cells under the micromass culture conditions used here. The + / + mice, however, have proven significantly more difficult to breed than the C3H strain. Thus, certain direct comparisons between bpH and proximal cultures are made possible by the use of both types of ‘normal’ mouse strains. Rotution cultures
Immediately before use in rotation culture, cell counts were made of the cell suspensions in culture medium without serum. Two x lo6 cells in 1.8 ml of medium (pregassed with
Fig. 2. Seventy-two hour micromass cultures prepared by the trypsin dissociation method and stained only with Alcian blue. Included are cultures prepared from stage 20 proximal C3H forelimb bud halves (a, b); distal C3H forelimb and halves (c); whole C3H forelimb buds (a); whole bpa forelimb buds (e, f), bud whole +/+ forelimb buds (9). The cultures in b and f were treated with dibutyryl cyclic AMP from 2 6 7 2 hours. All of the C3H cultures were prepared at the same time from a single group of well matched stage 20 forelimb buds. x 10
0.5% CO, in air) were placed in 30 ml Nalge screw-top bottles, and the cell count was checked again. At time zero, 0.2ml of fetal calf serum was added to each bottle. The bottles were then gassed and quickly placed in a 37” C water shaker bath at 70 rotations/min (0.65 an radius of gyration). Samples were withdrawn at regular intervals, after which each bottle was gassed and returned to the water bath. The number of single cells in each sample was counted with a hemacytometer and averaged.
147
Table 1. Aggregate and nodule formation' Cell source
Proximal C3H Distal C3H Whole C3H Whole bf Whole +/ +
Aggregates/mm2
Nodules/mm2
Untreated
Dibutyryl cAMP treated'
Untreated
Dibutyryl cAMP treated'
15.36k0.44 26.36f0.24 22.98k0.71 13.02f1.21 21.2 f1.07
22.42 f 0.71 23.70k0.94 24.09f 0.83 18.40f 0.23 26.26k 1.88
8.54&0.99 28.96k0.32 22.09 f 0.43 6.80k0.51 22.53 k 1 .O
19.70k1.58 22.02k 2.43 21.57+0.52 16.88f0.38 25.79 k0.73
a
All cultures were prepared from stage 20 limb buds and all were fixed after 72 h of culture. Each mean is calculated from data obtained from three separate cultures Aggregate and nodule densities are per millimeter2of culture area +SD ' Treated cultures were fed medium containing 1 mM dibutyryl cAMP after the first 24 h of culture and again after 48 h
Results Proximal and distal micromass cultures
Micromass cultures prepared from stage 20 whole mouse forelimb buds form numerous aggregates, the majority of which differentiate into cartilage nodules by 72 hours of culture [lo, 181. Typical 72 hour cultures prepared by the trypsin dissociation method from whole stage 20 C3H and + / + mouse forelimb buds and stained with Alcian blue are shown in Figs. 2d and g, respectively. These cultures each contain numerous cartilage nodules. If, however, the proximal and distal halves of stage 20 C3H limb buds are cut apart (Fig. 1) and used separately to prepare micromass cultures, the results are very different. Cultures prepared from the proximal halves ('proximal cultures'; Fig. 2a) form substantially fewer aggregates than whole limb bud cultures, only a few of which ever differentiate into cartilage nodules. In a typical example given in Table 1, the proximal cultures contain 33% fewer aggregates/mm2 than whole limb cultures, while only 56% of the aggregates formed ever become nodules, compared to 96% in the whole limb cultures. Cultures prepared from the distal halves ('distal cultures'; Fig. 2c), on the other hand, form numerous aggregates, most of which undergo cartilage differentiation (Table 1). The differences observed between whole, proximal, and distal cultures are the same, if not actually more pronounced, if the limb pieces are dissociated following incubation in CMF Tyrode's (data not shown) rather than trypsin. The in vitro behavior of proximal stage 20 mouse limb bud cells under these micromass culture conditions is clearly reminiscent of the in vitro behavior of cells obtained from whole stage 20 bp" limb buds [2,4, 11, 131. In the example in Table 1, for instance, the stage 20 bp" whole limb cultures form approximately 38% fewer aggregates than control +/+ cultures. Furthermore, only 52% of the bp" aggregates differentiate into nodules by 72 hours, compared to 100% of the +/+ aggregates. A stage 20 bpR whole limb bud culture stained with Alcian blue is shown in Fig. 2e. Response to dibutyryl cyclic AMP
Stage 24 chick proximal limb bud cultures [16] respond to treatment with dibutyryl cyclic AMP with a dramatic
increase in the proportion of aggregates that differentiate into cartilage nodules. A similar effect is observed when mouse proximal cultures are treated with dibutyryl cyclic AMP. Figure 2 b shows a culture prepared from proximal forelimb mouse cells [C3H] that had been treated from 24-72 hours of culture with the drug. The dibutyryl CAMPtreated proximal mouse cultures in Table 1 contained 46% more aggregates/mm2 and 130% more nodules/mm2 than the untreated cultures. Similarly treated distal mouse cultures did not show an increase in the number of nodules formed (Table 1). Stage 20 whole limb bpR cultures respond to treatment with dibutyryl cyclic AMP very much in the same manner as proximal C3H cultures (compare Figs. 2e and 2f with Figs. 2a, b). Dibutyryl cAMP treated stage 20 bp" cultures produced 41 % more aggregates/mm2 and 148% more nodules/mm2 than the untreated bp" cultures. Nearly all of the treated aggregates differentiated into cartilage nodules. Neither the +/+ cultures nor the whole C3H limb bud cultures responded in this extreme manner to dibutyryl CAMP. Aggregation of proximal and distal cells in rotation culture
It has been suggested that the bp" abnormality results in increased cell surface adhesion (2, 31. It was shown, for example, that the rate of loss of single cells in rotation cultures composed of day 12 bp" mouse forelimb cells was significantly greater than in similar cultures composed of 'normal' limb bud cells. Figure 3 shows the results of similar tests carried out using stage 20 proximal and distal limb bud cells from C3H mouse embryos. The proximal and distal cells used in these tests were obtained from the same limb buds, and the tests were carried out simultaneously. While great care was taken to assure that the suspensions contained the same numbers of cells at time zero (see Methods), at all later times there were always more single cells in the distal suspension than the proximal suspension. The results show that the number of single cells in the proximal cell suspension declines more rapidly than in the distal cell suspension. Repetitions of this experiment are highly consistent. When chick stage 24 proximal and distal cells are used in identical tests, the proximal cells also consistently aggregate more rapidly than distal cells (unpublished data).
148
cells obtained from whole stage 20 bp” limb buds may not illuminate further the nature of the primary bp” abnormality. This is because many of the in vitro behaviors manifested by stage 20 bp” limb bud cells would be expressed only as indirect consequences of the mutation. The present data support the idea that the bpH defect may not result directly from the assumption of proximal-like cell properties, but rather from defective distal-like properties. Most likely these include distal cell properties involved in certain early steps required for cell-cell interactions during chondrogenic induction [l 11.
01 0
15 30
60
90
120
150
MINUTES
Flg. 3. Loss of single cells in rotation cultures prepared with dim1 or proximal cells. Both populations of cells were obtained from the same set of stage 20 C3H forelimb buds. Suspensions of single cells (1 x lo6 cells/ml) were placed in a 37” C rotation water shaker bath as described in the Methods. At each time point, three separate samples were withdrawn from each cell suspension and the numbers of single cells were counted. Each point represents the mean of three cell counts, &standard deviations
Discussion The differences between the in vitro behaviors of proximal and distal stage 20 mouse forelimb bud cells appear to be analogous to the differences between proximal and distal cells of stage 23-24 [16] and stage 25 [12] chick embryo limb buds. In addition, similarities are observed between stage 20 bp” whole limb bud cells and ‘normal’ proximal limb bud cells in vitro. Under identical micromass culture conditions, each of these forms lower numbers of aggregates per unit area of culture than cultures prepared from whole ‘normal’ stage 20 limb bud cells. Of the aggregates that form, only a relatively small fraction in both types of culture undergoes spontaneous chondrogenic differentiation, although these fractions can be increased significantly by treatment with dibutyryl cyclic AMP. Finally, proximal forelimb bud cells exhibit a level of cell surface adhesiveness which is greater than that of distal forelimb cells, much in the same way that bpH stage 20 whole hindlimb cells have been shown to have higher adhesiveness than ‘normal’ stage 20 hindlimb cells [2, 31. The similarities between proximal and bp” cells described in this paper suggest an explanation for many of the known phenotypic differences between bpa and ‘normal’ limb mesenchyme. We suggest that the cell alterations brought about by the bpH mutation cause an apparent premature superabundance of proximal-like cells in the limb buds. By stage 20, cells from whole bp” forelimbs express in vitro proximal phenotypes [ l l ] that are not expressed by cells from whole C3H and + / + forelimbs until middle stage 21 [lo, 111. This would occur, for instance, if cells in the distal region of the bpH limb buds tended to differentiate into proximal-like cells at an earlier developmental time than distal limb bud cells in ‘normal’ limb buds. The proximal phenotype is expressed by cells from ‘normal’ limb buds at stage 20 only if the proximal cells are cultured in the absence of distal cells. One implication of these observations is that the detailed examination of individual in vitro behaviors displayed by
References 1. Ahrens PB, Solursh M, Reiter R (1977) Stage-related capacity for limb chondrogenesis in cell culture. Develop Biol60: 69-82 2. Duke J, Elmer WA (1977) Effect of the brachypod mutation on cell adhesion and chondrogenesis in aggregates of mouse limb mesenchyme. J Embryol Exptl Morphol42:209-217 3. Duke J, Elmer WA (1978) Cell adhesion and chondrogenesis in brachypod mouse limb mesenchyme fragment fusion studies. J Embryo1 Exptl Morphol48:161-168 4. Elmer WA, Selleck DK (1975) In vitro chondrogensis of limb mesoderm from normal and brachypod mouse embryos. J Embryol Exptl Morphol33 :371-386 5. Green EL (ed) (1966) Biology of the laboratory mouse, 2nd edn. McGraw-Hill, New York 6. Griineberg H (1963) The pathology of development. A study of inherited skeletal disorders in animals. Blackwell. Oxford 7. Griineberg H, Lee AJ (1973) The anatomy and development of brachypodism in the mouse. J Embryol Exptl Morphol 30:119-141 8. Guyer M F (1953) Animal micrology. University of Chicago Press, Chicago 9. Milaire J (1965) Etude morphogbnbtique de trois malformiitions congknitales de I’autopode chez la souris (syndactylismebrachypodisme-hbmimtlitedominante) par des mbthodes cytochimiques. Mem Acad R Belg Clin Sci 16: 1-20 10. Owens EM, Solursh M (1981) In vitro histogenic capacities of limb mesenchyme from various stage mouse embryos. Develop Biol88 :297-31 1 11. Owens EM, Solursh M (1982) Cellcell interaction by mouse limb cells during in vitro chondrogenesis: analysis of the brachypod mutation. Develop Biol91: 376-388 12. Paulsen DF, Parker CL, Finch RA (1979) Region-dependent capacity for limb chondrogenesis: patterns of chondrogenesis in cultures from different regions of the developing chick wing. Differentiation 14: 159-165 13. Selleck DK, Elmer WA (1973) Phenotypic instability of brachypod limb mesoblast cells in vitro. Am Zoo1 13:1313 14. Shambough J, Elmer WA (1980) Analysis of glycosaminoglycans during chondrogenesis of normal and brachypod mouse limb mesenchyme. J Embryol Exptl Morphol56:22>238 15. Solursh M, Jensen KL, Singley CT, Linsenmayer TE, Reiter RS (1982) Two distinct regulatory steps in cartilage differentiation. Develop Biol94: 31 1-325 16. Swalla BJ, Owens EM, Linsenmayer TF, Solursh M (to be published) Two distinct classes of prechondrogenic cell types in the embryonic limb bud. Develop Biol97 17. Theiler K (1972) The house mouse. Development and normal stages from fertilization to 4 weeks of age. Springer-Verlag. New York 18. Umansky R (1966) The effect of cell population density on the developmental fate of reaggregating mouse limb bud mesenchyme. Develop Biol13 :31-56 Received October 1982 / Accepted in revised form February 1983