Macrocyst Development in Dictyostelium Discoideum. I

J. Cell Set. 55, 341-352 (1982) Printed in Great Britain © Company of Biologists Limited 1982

MACROCYST DEVELOPMENT IN DICTYOSTELIUM DISCOIDEUM. I. INDUCTION OF SYNCHRONOUS DEVELOPMENT BY GIANT CELLS AND BIOCHEMICAL ANALYSIS

Y. SAGA AND K. YANAGISAWA Institute of Biological Sciences, University of Tsukuba, Ibaraki 305, Japan

SUMMARY In Dictyostelium discoideum, cytological and physiological studies on macrocyst formation revealed that this process consists of at least two steps: the production of giant cells, which are believed to be formed from the fusion of cells of two opposite mating types, and the subsequent induction of macrocyst development by the giant cells. The conditions that had been considered formerly to be required for macrocyst formation, such as darkness and the presence of two cells of complementary mating types in heterothallic strains, were actually required only for the production of the giant cells. Once giant cells are produced, the surrounding cells can aggregate and form macrocysts even in the light. Furthermore, it was demonstrated that giant cells can switch the developmental mode of the surrounding cells to macrocyst formation. That is, if a critical number of the isolated giant cells are introduced into a cell population of a single strain of NC4, which normally would produce only fruiting-bodies, macrocysts are formed instead. When in the presence of giant cells, the development of macrocysts may be initiated by starvation. Therefore, if all cells are made to starve simultaneously development begins and proceeds synchronously. Using this technique of synchronous development, the developmental kinetics of enzyme activities were assayed during macrocyst and fruiting-body formation. Considerable differences in the patterns of those enzyme activities were demonstrated between the two developmental modes of D. discoideum.

INTRODUCTION The cellular slime mould Dictyostelium discoideum has two alternative modes of development. One culminates in fruiting-body formation; the other culminates in macrocyst formation and is considered to be a sexual cycle (Clark, Francis & Eisen- berg, 1973; Erdos, Raper & Vogen, 1973; Maclnnes & Francis, 1974). In fruiting-body formation development is initiated by starvation. Accordingly, development can be synchronized easily by subjecting all the cells to starvation simultaneously (White & Sussman, 1961). In contrast, a method of synchronous development has not yet been developed in the case of macrocyst formation. This has made biochemical studies on macrocyst formation difficult. The production of macrocysts in Dictyostelium requires particular conditions, such as the presence of cells of two complementary mating types in heterothallic strains, darkness, humidity, and the appropriate temperature (Blaskovics & Raper, 1957; Clark et al. 1973; Erdos et al. 1973; Nickerson & Raper, 1973; Maclnnes & Francis, 1974; Erdos, Raper & Vogen, 1976). These complicated requirements have made it 342 Y. Saga and K. Yanagisawa difficult to detect precisely a factor(s) that is responsible for the initiation of macrocyst development. Erdos et al. (1976) found that darkness was required only for a certain period in the early stages of macrocyst development. When they shifted a culture with cells of two mating types, initially incubated in the dark, to the light they discovered that macrocyst formation could still take place. Recently, apart from this, O'Day (1979) found that in macrocyst formation cells slightly larger than the surrounding amoebae, called giant cells, appeared in the culture prior to cell aggregation, and that those cells served as centres for aggregation in macrocyst development. We isolated giant cells from the dark-grown mixed cultures of cells of two opposite mating types, and studied their characteristics. We found out why darkness was required only at the beginning of macrocyst development. In this paper we present the results of experiments that demonstrate the function of giant cells in macrocyst development, a method of initiating synchronous development of macrocysts, and the results of a comparative biochemical analysis of macrocyst and fruiting-body formation.

MATERIALS AND METHODS Organisms and culture conditions Two strains of D. discoideum, NC4 and HMi, were used. HMi was derived from Vi2, which is a strain of the opposite mating-type to NC4. HMi was kindly provided by Dr R. R. Kay. The strains were maintained on Klebsiella aerogenes on nutrient agar (Sussman, 1966).

Formation and isolation of giant cells Giant cells were obtained by the following method. Growth-phase cells of NC4 and HMi, cultured separately on nutrient agar plates, were harvested and suspended together at a ratio of 1:1 in Bonner's salt solution (Bonner, 1947) at a total concentration of 5 x 10' cells/ml. The bacteria were then added to the cell suspension at a final concentration of 1 x io10 cells/ml. The mixture was agitated on a reciprocal shaker (120 strokes/min), in flasks 5-6 times the volume of the mixture, in the dark at 22 °C. After 20-22 h cultivation, cells were harvested, washed three times with Bonner's salt solution by centrifugation and resuspended in a 5 mM-EDTA, 17 mM-phosphate buffer (pH 6-5). Giant cells were isolated by straining the suspension through a nylon mesh (10 fun pore size).

Fixation and staining procedure Washed cells were fixed with 60 % methanol (4 °C), placed on slides, and air-dried. The slides were then treated with trypsin (o-i % tryps in in 0-85% NaCl solution) for 30-90 s, washed with water, and stained with 10% Giemsa stain in phosphate buffer (pH 6-8).

Synchronous development of macrocysts and fruiting-bodies Development of macrocysts was synchronized by the following method. NC4 cells, which were harvested from nutrient agar plates, were washed and suspended, together with the isolated giant cells, at a ratio of 1000:2 in Bonner's salt solution at a concentration of 5 x io7 cells/ml. The mixture was spread on a non-nutrient agar plate (5x10* cells/cm1) and then incubated at 22 °C in either the dark or the light. Synchronous development of fruiting- bodies was achieved by incubating washed NC4 cells on a non-nutrient agar under exactly the same conditions, except that no giant cells were added. Macrocyst development in D. discoideum 343

Enzyme assays Cells were taken at various times during development, washed and stored as pellets at — 20 °C. Frozen pellets were thawed and suspended in the following solutions: (I) cold water for assays of TV-acetylglucosaminidase, a-D-mannosidase, /9-D-glucosidase and alkaline phosphatase; (II) 20 mM-potassium phosphate buffer, 1 mM-MgSO4 (pH 7-4) for assay of cellular phosphodiesterase; (III) o-i M-tricine buffer (pH 7-4) for assay of UDP-glucose pyrophosphorylase. Cell suspensions containing 1 x io7 cells were sonicated and used for the enzyme assays. iV-acetylglucosaminidase, a-D-mannosidase and alkaline phosphatase were measured by the method of Loomis (1969a, b, 1970). /?-D-glucosidase was measured by the method of Coston & Loomis (1969), phosphodiesterase as described by Malchow, NSgele, Schwarz & Gtrisch (1972), and UDP-glucose pyrophosphorylase by the method of Hames (1976). Proteins were determined by the method of Lowry, Rosebrough, Fair & Randall (1951) using bovine serum albumin as a standard. Specific activities of the enzymes were expressed as nanomoles per min per mg protein. All of the enzyme assays were performed at least three times.

RESULTS Formation of giant cells and their role in macrocyst development Cells of two strains of complementary mating type, NC4 and HM1, were suspended together with bacteria in Bonner's salt solution. The mixture was then shaken in a flask in the dark. Under these conditions numerous macrocysts were formed after cultivation for a few days. We examined cell growth in the mixed dark-grown culture, and found that the number of cells increased rather rapidly during the early period of cultivation, but then very slowly after 10 h of cultivation (*10) as shown in Fig. 1. In contrast, when cells were cultured in the light, the number of cells increased continuously until the stationary phase was reached (approx. 2 x io7 cells/ml) at t^-t^. No macrocysts were produced in this case. In the dark-grown culture, we found that very large cells began to appear at about tw and the number of these cells increased with time until it reached a maximum of about 13-14 % of the total cells. These large cells were fixed, stained with Giemsa, and observed. We found that they were all multinucleated and some of them had more than 50-60 nuclei, as shown in Fig. 2D. The reduction of the rate of increase in the number of cells observed after 10 h of cultivation in the dark-grown culture (Fig. 1) is conceivably due to the formation of these multinucleated large cells. These large cells seem to be similar to the giant cells described previously by O'Day (1979). According to O'Day, these giant cells, which were binucleated zygotes and slightly larger than the surrounding cells, appeared in the early stage of macrocyst development and served as the centres for cell aggregation, which subsequently form macrocysts. Our large giant cells are multinucleated and differ from O'Day's giant cells in their morphology. It is certain, however, that our giant cells also play an important role in macrocyst formation, as they never appear in either mixed cultures grown in the light (Fig. 2 A, B) or cultures of single strains grown in the dark; that is, under the conditions in which macrocysts fail to be produced, these giant cells are not found. Furthermore, we found that once the giant cells were produced in the dark-grown mixed culture, macrocysts were formed even if the culture was shifted from dark to light conditions or the bacteria were removed. In order to study in more detail the role of giant cells in macrocyst formation, the 344 Y. Saga and K. Yanagisawa

10s 0 2 4 6 8 10 12 14 16 18 20 22 24

Time of incubation (h) Fig. i. Cell growth in mixed cultures of NC4 and HMi suspended together with the bacteria in Bonner's salt solution in the light (O) and in the dark (#). Giant cells (A) appear only in the dark-grown culture. Details are described in the text.

Table 1. Correlation between the number of giant cells added to a population of NC4. cells and the number of macrocysts consequently formed

No. of Mean no. of Formation of giant cells macrocysts fruiting-bodies

IOO 119 (± 327) + + 500 497 (± 835) + + 1000 912 (±1671) + 3000 2068 (±3628) + 5000 2229 (±769-0) — IOOOO 2763 (±428-1) - Details are described in the text. +, formation of fruiting-bodies; —, no formation of fruiting-bodies.

following experiments were performed. Giant cells were isolated from dark-grown mixed cultures using a fine nylon mesh (Fig. 2E), and added at varying ratios to a bacteria-free suspension of NC4 cells (2-5 x io7 cells/ml) in Bonner's salt solution. The mixtures were spread on a non-nutrient agar (25 x io6 cells/cm2), and incubated at 22 °C in the light. Under these conditions normal macrocysts were produced after incubation for 2-3 days. The data demonstrated that the number of macrocysts Macrocyst development in D. discoideum 345

Fig. 2. Photographs showing the formation and isolation of giant cells. A and B, a mixed suspension of NC4 and HMi cells cultured in the light for 20 h; no giant cells appeared, B. Shows Giemsa-stained cells; c and D, NC4 and HMi cultured in the dark for 20 h. Note appearance of giant cells. Giemsa-staining in D reveals the presence of multiple nuclei within giant cells. E and F, separation of giant cells from dark-grown mixed culture using fine nylon mesh; E shows isolated giant cells; F shows remaining smaller cells. Bar, 20 fim. formed was dependent upon the number of giant cells added (Table 1). This relationship was a directly proportional one when the number of giant cells added was relatively small. However, when a greater number of giant cells was added the number of resulting macrocysts was fewer than would be expected. This seems to be due to the fusion of giant cells or cell-aggregates surrounding giant cells. There is evidence that giant cells have a specific affinity for other cytophagic giant cells and ingest them predominantly by phagocytosis (Fukui, 1976). The cells that did not participate in macrocyst development produced fruiting-bodiea instead. In this experiment the 12 CEL JJ 34° Y. Saga and K. Yanagisawa addition of 5000 giant cells to a suspension of 2-5 x io6 NC4 cells resulted in all of the cells participating in macrocyst formation, with no fruiting-bodies being formed. Thus, giant cells act as centres for aggregation and are able to direct macrocyst development even within populations of cells of a single strain. Furthermore, the production of macrocysts by giant cells can take place even in the light. Giant cells were also added to HMi cells on agar plates. In this case, however, a large number of fruiting-bodies was formed. This was assumed to be the result of aggregation for fruiting-bodies that had already started in the HMi cells. In order to prevent fruiting-body formation, HMi cells were suspended in Bonner's salt solution after the addition of giant cells. Three ml of this cell suspension (2-5 x io6 cells/ml) was shaken in a 20 ml flask on a reciprocal shaker (40 strokes/min) and incubated at 22 °C for 5-6 days. When giant cells were added to the HMi cell suspension at a ratio of 2:1000, no macrocysts were formed. However, when the proportion of giant cells was increased, a few macrocysts were produced. Their number increased as the number of giant cells added increased, and when the ratio of giant cells reached 8:1000, almost all HMi cells were found to participate in macrocyst formation. No exact quantitative relationship between giant cells added and macrocysts produced, however, could be determined using this liquid system.

Synchronous development of macrocysts When giant cells totalling more than 02 % of the total NC4 cells were added, all of the NC4 cells participated in the formation of macrocysts. This was demonstrated in cell suspensions spread on a non-nutrient agar, as described before. Subsequently, it was found that if a critical number of giant cells was added to a population of washed NC4 cells macrocyst development was initiated synchronously. All the cells simultaneously form loose aggregates at *7-*9. Giant cells are located at the centre of each aggregate. This is apparent from the presence of endocytes (Fig. 3 A, B). The aggregates develop into precysts, which are enclosed by a thin slime sheath called the primary wall, at about tlt-tlt (Fig. 3 c), and the engulfment of peripheral cells by the giant cells proceeds (Fig. 3 D). A thick second wall is then formed at about tM (Fig. 3 E). At /36 almost all peripheral cells are completely engulfed by the giant cells (Fig. 3F). This synchronous development and the final gross morphology of the macrocysts produced by NC4 and giant cells were the same as those seen in normal macrocyst formation as reported previously (O'Day, 1979). During aggregation of cells giant cells acted as centres of aggregation. However, it was noticeable that the surrounding cells did not elongate, as occurs in fruiting-body formation, and the aggregation streams were not clear and distinct as those seen prior to the formation of fruiting-bodies.

Specific activities of enzymes during development of macrocysts and fruiting-bodies The biochemical analysis of macrocyst formation requires a technique by which development may be induced synchronously. Taking advantage of the technique described above, we attempted to study how the developmental kinetics of the specific activity of certain enzymes are altered by the switching of the developmental Macrocyst development in D. discoideum 347

Fig. 3. Photographs showing the synchronous development of macrocysts. Taken at 6 h (A), 9 h (B), 12 h (c), 15 h (D), 24 h (E) and 36 h (F) after the initiation of development. Details are described in the text. The arrows in A and B indicate endocytes. Bar, 40 fim. mode of NC4 cells from fruiting-body formation to macrocyst formation. Six developmentally regulated enzymes: iV-acetylglucosaminidase, a-D-mannosidase, phosphodiesterase, UDP-glucose pyrophosphorylase, alkaline phosphatase and /9-D-glucosidase, identified in fruiting-body formation (Ashworth & Sussman, 1967; Coston & Loomis, 1969; Loomis, 1969a, b, 1970; Malchow et al. 1972), were assayed in both macrocyst and fruiting-body formation. The results are as shown in Fig. 4. In fruiting-body formation iV-acetylglucosaminidase and a-D-mannosidase have been identified as enzymes whose activities begin to increase immediately after the initiation of development. These two enzymes Y. Saga and K. Yanagisawa

200 40

100 20

0 I I I I

_. 15 100

o Q. 10

£ 5 .E 5 I o 40 20

0 3 6 9 12 15 18 21 24 27 0 3 6 9 12 15 18 21 24 27 Time of development (h) Fig. 4. Developmental kinetics of the enzymes; AT-acetylglucosaminidase (A), a-D- mannosidase (B), phosphodieaterase (c), UDP-glucose pyrophosphorylase (D), alkaline phosphataae (E), and /?-D-glucosidase (F), during macrocyat formation (#) and fruiting-body formation (O)- The cells were collected at various times during synchronous development. Methods for assay of the specific activities of the enzymes are described in Materials and Methods. In fruiting-body formation, cells begin to aggregate at about <,, form tight aggregate at tlt, start to culminate at tlt, and form fruiting-bodies at tu-tu. also began to increase after the initiation of development in macrocyst formation. However, the increase in their activities stopped at fB, and remained almost constant thereafter (Fig. 4 A, B). The maximum levels in the activities of ./V-acetylglucosaminidase and a-D-mannosidase during macrocyst development were about two-thirds and one-third, respectively, of those observed in fruiting-body formation. Phosphodiesterase appears only at the aggregation stage of fruiting-body formation, and is known as an enzyme that plays an important role in cell aggregation. It was noted that (as is shown in Fig. 4 c) in macrocyst development the activity of this enzyme showed very little increase during the aggregation stage. Macrocyst development in D. discoideum 349 The pattern of UDP-glucose pyrophosphorylase activity, also, was not identical between the two developmental modes. In macrocyst formation the activity was initiated at t9 and increased continuously until macrocysts were produced. However, in fruiting-body formation, the activity began to increase at a later point, tlt, and subsequently decreased at ti7 (Fig. 4D). The specific activity of alkaline phosphatase during macrocyst formation was very different from that observed during fruiting-body formation. The activity started to increase at the culmination stage, tn, in fruiting-body formation, whereas it began to increase at the aggregation stage, tt, and reached a maximum at the stage of precyst formation, tn, then decreased thereafter in the case of macrocyst development

(Fig. 4E). /?-D-glucosidase has two isozymes, /ff-D-glucosidase 1 and 2 (Coston & Loomis, 1969). In fruiting-body formation the activity of/9-D-glucosidase 1 decreased gradu- ally during the first 15 h, while /?-D-glucosidase 2 increased rapidly at the culmination stage. In contrast, no increase of /?-D-glucosidase 2 activity was observed during macrocyst development (Fig. 4F). Thus, the results demonstrate that the pattern of specific activity of several developmentally regulated enzymes differs significantly between the two developmental modes of fruiting-body formation and macrocyst formation. It should be noted that the differences between the biochemical steps involved in macrocyst and fruiting- body formation are apparent even at the aggregation stage.

DISCUSSION The present series of experiments demonstrates that the process of macrocyst formation consists of at least two steps: the production of giant cells, and the subsequent induction of macrocysts by the giant cells. In the presence of giant cells, even single-strain cells in heterothallic strains can produce normal macrocysts in the light. Thus, giant cells can shift the developmental mode of the surrounding cells from fruiting-body formation to macrocyst formation. O'Day (1979) reported that giant cells, which acted as centres for macrocyst aggregation, were only slightly larger than the surrounding cells, and that the giant cells were zygotes formed by the fusion of two cells of complementary mating type (Chagla, Lewis & O'Day, 1980). However, the giant cells we obtained were, in contrast, very large multinucleated cells. Although we do not have precise knowledge about the fate of these multiple nuclei, preliminary cytological observations show that fusion of a number of nuclei and the formation of a single large nucleus occur during macrocyst formation. We hope to study these points in more detail. The difference between our giant cells and O'Day's may be partly due to different culture conditions and/or differences in the strains used. Actually, when we mixed HM1 and NC4 cells together, with a low density of bacterial suspension in a o-1 % lactose/peptone medium, as described by Chagla et al. (1980), we obtained small giant cells. As giant cells, however, their function during macrocyst formation was apparently the same in spite of their small size (unpublished data). 3so Y. Saga and K. Yanagisawa We found significant differences in the response of NC4 and HMi cells to giant cells during macrocyst formation. When giant cells are added to cultures of NC4 or HMi cells at a ratio of 2:1000, the NC4 cells will aggregate around the giant cells and macrocysts will be formed, whereas HMi cells will tend to aggregate among them- selves. It is only when the ratio of giant cells to total cell number reaches 8:1000 that exclusive macrocyst development is seen in HMi cultures. This fact suggests that the response of HMi and NC4 cells to some factor secreted by giant cells, which induces participation in macrocyst development, may be different and/or that HMi cells have a greater mutual affinity than NC4 cells. The data on certain enzyme activities during development revealed significant differences between the biochemical processes involved in macrocyst and fruiting- body formation. These differences are considered to be caused by the two developmental modes. Although HMi cells are probably contained within giant cells used to induce macrocyst development, the total number of giant cells accounts for only o-2 % of the total cell number, making it highly unlikely that HMi cells would have any significant influence on the overall pattern of enzyme activity, even if their pattern of activity was different. The results of our biochemical studies on macrocyst formation show that the activity of most enzymes, such as iV-acetylglucosaminidase, a-D-mannosidase, UDP- glucose pyrophosphorylase and alkaline phosphatase, changed at the stage during which cells form aggregates, i.e. at to-t9. This suggests that the aggregation stage is the most critical period in terms of biochemical activity in macrocyst development as is the case in fruiting-body development (Alton & Lodish, 1977; Blumberg & Lodish, 1981; Jacquet, Part & Felenbok, 1981). The pattern of phosphodiesterase activity during macrocyst formation is quite interesting. This enzyme is known to play an important role in the process of cell aggregation, and shows a sharp high peak during the aggregation stage in fruiting- body formation (Malchow et al. 1972; Malchow, Fuchila & Nanjundiah, 1975; Darmon, Barra & Brachet, 1978). In macrocyst formation, however, phosphodiesterase activity was very low during cell aggregation. This fact might suggest that the aggregation mechanism of cells in macrocyst formation is different from that in fruiting-body formation. However, to confirm this possibility it would be necessary at least to measure the activities of extracellular phosphodiesterase, phosphodiesterase inhibitor and extracellular cyclic AMP levels during cell aggregation. According to Dimond & Loomis (1976), mutants lacking or having a greatly reduced /?-glucosidase activity do not require the presence of the opposite mating- type to form macrocysts. This fact suggests that the enzyme acts to block selfing in macrocyst formation. Although they do not mention giant cells, it will be interesting to investigate whether these mutants can produce giant cells homothallically or not. O'Day & Lewis (1975) have reported that macrocyst formation in D. discoideum is mediated by a volatile pheromone of low molecular weight, and that NC4 cells secrete this substance, while V12 cells respond to it to form macrocysts. We have also been investigating the possibility of the involvement of such a factor(s) in giant cell formation. Our preliminary data show the existence of the factor(s); however, it Macrocyst development in D. discoideum 351 seems to be different from that reported by O'Day & Lewis. The process of macrocyst formation is probably rather complicated. There may be a number of factors that regulate the sexual cycle of D. discoideum. We are now investigating further the factors that are involved in the production of giant cells.

REFERENCES ALTON, T. H. & LODISH, H. F. (1977). Developmental changes in messenger RNAs and protein synthesis in DictyosUlhtm discoideum. Devi Biol. 60, 180—206. ASHWORTH, J. M. & SUSSMAN, M. (1967). The appearance and disappearance of uridine diphosphate glucose pyrophosphorylase activity during differentiation of the cellular slime mold Dictyostelium discoideum. J. biol. Chem. 242, 1696-1700. BLASKOVICS, J. C. & RAPER, K. B. (1957). Encystment stages of Dictyostelium. Biol. Bull. mar. biol. Lab., Woods Hole 113, 58-88. BLUMBERG, D. D. & LODISH, H. F. (1981). Changes in the complexity of nuclear RNA during development of Dictyostelium discoideum. Devi Biol. 81, 74-80. BONNER, J. T. (1947). Evidence for the formation of cell aggregates by chemotaxis in the development of the slime mold Dictyostelium discoideum. J. exp. Zool. 106, 1-26. CHAGLA, A. H., LEWIS, K. E. & O'DAY, D. H. (1980.) Ca1+ and cell fusion during sexual development in liquid cultures of Dictyostelium discoideum. Expl Cell Res. 126, 501-505. CLARK, M. A., FRANCIS, D. & EISENBERG, R. (1973). Mating types in cellular slime molds. Biochem. biophys, Res. Commun. 5a, 672-678. COSTON, M. B. & LOOMIS, W. F. (1969). Isozymes of /?-glucosidase in Dictyostelium discoideum. J. Bact. 100, 1208-1217. DARMON, M., BARRA, J. & BRACKET, P. (1978). The role of phosphodiesterase in aggregation of Dictyostelium discoideum. J. Cell Sci. 31, 233-243. DIMOND, R. L. & LOOMIS, W. F. (1976). Structure and function of y?-glucosidases in Dictyo- stelium discoideum. J. biol. Chem. 351, 2680—2687. ERDOS, G. W., RAPER, K. B. & VOGEN, L. K. (1973). Mating types and macrocyst formation in Dictyostelium discoideum. Proc. natn. Acad. Sci. U.S.A. 70, 1828-1830. ERDOS, G. W., RAPER, K. B. & VOGEN, L. K. (1976). Effects of light and temperature on macrocyst formation in paired mating types of Dictyostelium discoideum. J. Bact. ia8, 495-497. FUKUI, Y. (1976). Enzymatic dissociation of nascent macrocysts and partition of the liberated cytophagic giant cells in Dictyostelium mucoroides. Develop. Growth &f Differ. 18, 145-155. HAMES, B. D. (1976). An improved radiochemical assay for uridine diphosphoglucose pyrophosphorylase. Anal. Biochem. 73, 215-219. JACQUBT, M., PART, D. & FELENBOK, B. (1981). Changes in the polyadenylated messenger RNA population during development of Dictyostelium discoideum. Devi Biol. 81, 155-166. LOOMIS, W. F. (1969a). Developmental regulation of alkaline phosphatase in Dictyostelium discoideum. J. Bact. ioo,*4i7-422. LOOMIS, W. F. (19696). Acetylglucosaminidase, an early enzyme in the development of Dictyostelium discoideum. J. Bact. grj, 1149-1154. LOOMIS, W. F. (1970). Developmental regulation of a-mannosidase in Dictyostelium discoideum. J.Bact. 103, 37S-38i. LOWRY, O. H., ROSEBROUCH, N. J., FARR, A. L. & RANDALL, R. J. (1951). Protein measurement with the folin phenol reagent. J. biol. Chem. 193, 265-275. MACINNES, M. A. & FRANCIS, D. (1974). Meiosis in Dictyostelium mucoroides. Nature, Lond. 251. 32I-323- MALCHOW, D., FUCHILA, J. & NANJUNDIAH, V. (1975). A plausible role fora membrane-bound cyclic AMP phosphodiesterase in cellular slime mold chemotaxis. Biochim. biophys. Ada 386, 421-428. MALCHOW, D., NAGELE, B., SCHWARZ, H. & GERISCH, G. (1972). Membrane-bound cyclic AMP phosphodiesterase in chemotactically responding cells of Dictyostelium discoideum. Eur.J. Biochem. 28, 136-142. 352 Y. Saga and K. Yanagisawa NICKERSON, A. W. & RAPER, K. B. (1973). Macrocysts in the life cycle of the Dictyosteliaceae. I. Formation of the macrocysts. Am. J. Bot. 60, 190-197. O'DAY, D. H. (1979). Aggregation during sexual development in Dictyostelium discoidettm. Can. J. Microbiol. 25, 1416-1426. O'DAY, D. H. & LEWIS, K. E. (1975). Diffusible mating-type factors induce macrocyst development in Dictyostelium discoidettm. Nature, Lond. 254, 431-432. SU8SMAN, M. (1966). Biochemical and genetic methods in the study of cellular slime mold development. In Methods in Cell Physiology, vol. 2 (ed. D. Prescott), pp. 397-410. New York: Academic Press. WHITE, G. J. & SUSSMAN, M. (1961). Metabolism of major cell components during slime mold morphogenesis. Biochim. biophys. Acta 53, 285-293.

(Received 10 September 1981)