Proc. Natl. Acad. Sci. USA Vol. 80, pp. 5646-5649, September 1983 Developmental Biology

Thermotaxis of Dictyostelium discoideum amoebae and its possible role in pseudoplasmodial thermotaxis (temperature/sensory transduction/slime mold) CHOO B. HONG*, DONNA R. FONTANAt, AND KENNETH L. POFFt Michigan State University-Department of Energy, Plant Research Laboratory, -Michigan State University, East Lansing, Michigan 48824 Communicated byJ. T. Bonner, June 6, 1983 ABSTRACT Thermotaxis by individual amoebae of Dictyo- bae. For example, the amoebae respond to cyclic AMP and fo- stelium discoideum on a temperature gradient is described. These late, whereas the multicellular pseudoplasmodia do not, and amoebae showpositive thermotaxis attemperatures between 14°C the action spectra for phototaxis by the amoebae substantially and 28WC shortly (3 hr) after food depletion. Increasing time on the differ from those for phototaxis by the pseudoplasmodia, sug- gradient reduces the positive thermotactic response at the lower gesting different photoreceptor pigments regulating the re- temperature gradients (midpoint temperatures of 14, 16, and 18°C), sponses to light. and amoebae show an apparent negative thermotactic response Thermotaxis by the pseudoplasmodia has been knownfor many after 12 hr on the gradient. The thermotaxis response curve for years (8). The characteristics of this response include (i) a very "wild-type" amoebae after 16 hr on the gradient is similar to that high sensitivity (response to less than 0.0005'C across an in- shown for the pseudoplasmodia. Growth of the amoebae at a dif- and a relatively narrow temper- ferent temperature causes a shift in the thermotaxis response curve dividual pseudoplasmodium) for the amoebae. This adaptation is similar to that shown for the ature range across which the response is given, (ii) negative pseudoplasmodia. Two mutants in thermotaxis, H0428 and H0813, thermotaxis at temperatures several degrees below the tem- show changes in amoebal thermotaxis similar to the observed perature at which the amoebae were grown and permitted to changes in pseudoplasmodial thermotaxis. On the basis of the sim- develop into pseudoplasmodia and positive thermotaxis at higher ilarities between these responses, thermotaxis by the amoebae is temperatures, and (iii) adaptation-dependence of the direc- proposed to be the basis for thermotaxis by the multicellular pseu- tion of thermotaxis on the recent thermal history of the or- doplasmodium. ganism. This work was undertaken to determine whether or not in- Amoebae of the cellular slime mold Dictyostelium discoideum dividual amoebae of D. discoideum also exhibit thermotaxis, live and grow in the mulch on the forest floor, where they feed and it was expanded after the affirmative answer to examine on bacteria. Upon depletion of the food supply, the amoebae 'the relationship between thermotaxis by the individual amoe- aggregate, forming a multicellular pseudoplasmodium that bae and thermotaxis by the multicellular pseudoplasmodia. This eventually develops into a sorocarp consisting of a stalk bearing paper reports the evidence supporting the conclusion that ther- a sorus containing spores. Under the proper conditions, the motaxis by the amoebae is the basis for the thermotaxis by the spores may germinate, releasing amoebae, thus repeating the pseudoplasmodia. developmental cycle (1). Much of the work with this organism has been centered upon its development and the fact that one MATERIALS AND METHODS can easily separate growth and cell division from development. Amoebae of D. discoideum, strains HL50, H0428, and H0813 Dictyostelium is also being increasingly recognized as an ideal (14), were grown in association with Klebsiella aerogenes (15) system for the study of sensory transduction in a eukaryotic or- for about 22 hr at 23.5 ± 0.50C. At this time, the bacteria had ganism. Sensory responses thus far described are chemotaxis by been cleared from about 80% of the agar plate. In some ex- the amoebae to cyclic AMP (2) and folate (3), both positive and periments designed to test for thermal adaptation, amoebae of negative phototaxis by the amoebae (4-6), chemotaxis by the strain HL50 were grown for about 22 hr at 27.50C, at which pseudoplasmodia to an endogenous "slug turning factor" (7), time the plates were usually about 50% cleared of bacteria. phototaxis by the pseudoplasmodia (8-11), and positive and The amoebae were freed of bacteria by washing three or four negative thermotaxis by the pseudoplasmodia (8, 12, 13). In ad- times with 15 mM potassium phosphate buffer, pH 6.1, with dition to the phenomenology, work is also progressing toward intervening centrifugations of 1 or 2 min at about 500 X g, and an understanding both of the primary steps and of the trans- resuspended in phosphate buffer at 1 x 107 amoebae per ml. duction sequence for each of these responses. Moreover, the Aliquots (1 ml) of the suspension were mixed with 0.3 ml of a use of mutants is permitting a study of the interconnections suspension of washed charcoal in phosphate buffer (2%, wt/vol). between these various sensory responses. The charcoal was added to mark the point of cell spotting. The Because of the ease with which the single-celled and mul- cell/charcoal suspension was agitated and deposited in a line ticelled stages may be separated in Dictyostelium, one would on 2% (wt/vol) water agar which was about 1 mm thick covering also expect this to be an excellent system for studying the a 1 X 25 X 75 mm microscope slide.'The line of amoebae (Fig. expression of a particular sensory transduction system in the 1A) was formed with the aid of two parallel strands of fine plat- two motile forms, the amoebae and the pseudoplasmodia. inum wire, 0.1 mm in diameter, which were attached at their However, neither chemotaxis nor phototaxis appears to use the same system in the pseudoplasmodia that is used in the amoe- * Present address: Chemistry Dept., Texas Tech Univ., Lubbock, TX 79409. The publication costs of this article were defrayed in part by page charge t Present address: Dept. of Biochemistry, Johns Hopkins Univ., Bal- payment. This article must therefore be hereby marked "advertise- timore, MD. ment" in accordance with 18 U.S.C. §1734 solely to indicate this fact. t To whom reprint requests should be addressed.

5646 Downloaded by guest on September 25, 2021 Developmental Biology: Hong et al. Proc. Natl. Acad. Sci. USA 80 (1983) 5647 ends to two metal bars. The bars were attached to a rack and pinion gear so that the wires could be raised and lowered. After B spotting, the agar slides were placed on a temperature gradient A A (0.220C/cm), formed as described by Poff and Skokut (12), and left there for 3, 6, 9, 12, 16, or 20 hr. After the exposure to the thermal gradient, the number of amoebae that moved in each direction away from the original P. : .. .:. . .: .:.. : : ae . . .. 0 line was determined. This was done by examining five ran- P domly chosen microscope fields along the original line of about 3 x 5 mm on each of ten slides. The response (A percentage) for each replicate was calculated as: A percentage = 100(no. of cells moving up gradient - no. of cells moving down gradient)/ :,:*...,s 0@ t total no. of cells that moved. Thus, A percentage is a measure of the directed movement on a thermal gradient. For thermal response curves, relative A percentage was determined for thermal gradients with various midpoint temperatures. c D The experiments on the thermotaxis of pseudoplasmodia were carried out as described by Schneider et al. (14) with a slight modification. Some of the HL5O amoebae, used for the ad- aptation experiments, were grown at 27.50C for 22 hr. How- ever, the inoculum was reduced so that no clearing was visible. This change appeared to delay the formation of fruiting bodies from the pseudoplasmodia. Development in this case also oc- curred at 27.50C. RESULTS Amoebae of D. discoideum strains HL50, H0428, and H0813 regularly showed amoeboid movement for up to 20 hr without any sign of aggregation. In the absence of a thermal gradient, the amoebae moved randomly on the agar surface. Depending FIG. 1. Thermotactic responses of D. discoideum amoebae. Tem- on the midpoint temperature of the thermal gradient used, there peratures are higher at the top. Larger dots represent amoebae and were conditions that resulted in more cells on the warmer side, smaller dots are charcoal. Tracings from photomicrographs showing cooler side, or no directed movement (Fig. 1 C, D, and B, re- original spotting (A), random response (B), positive thermotaxis (C), spectively). Thus individual amoebae are capable of positive and negative thermotaxis (D). and negative thermotaxis. Amoebae of strain at were then ex- HL50, grown 23.50C, grown and allowed to develop into pseudoplasmodia at 27.5'C, posed to thermal gradients for six different time intervals. After also demonstrates both and thermotaxis 3 was seen entire tem- positive negative (Fig. hr, positive thermotaxis throughout the 3B), with the transition between the two responses occurring were perature range (Fig. 2A). When the amoebae removed around 20.40C. this transition is consistent with that seen a Again, from the gradient after 6 hr, the amoebae did not show sig- with amoebae. Strains H0428 and H0813 are mutants in pseu- nificant positive thermotactic response on gradients with mid- doplasmodial thermotaxis derived from strain HL50 (14). Amoe- point temperatures of 14, 16, and 18'C (Fig. 2B). This trend continued, as can be seen from the results for amoebae exposed to the gradients for 9 and 12 hr (Fig. 2 C and D). After 12 hr A 8BXC on thermal gradients with midpoint temperatures of 14 and 160C, significant negative thermotaxis was evident. After 16 hr, the amoebae demonstrated negative thermotaxis on gradients with 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0 midpoint temperatures of 14, 16, and 18'C and positive ther- -8 motaxis when the midpoint temperature was above 20'C (Fig. 0' 2E). The maximal response was at a midpoint temperature of 230C, with the transition from a negative to a positive response -16- occurring at about 19°C. The transition at about 19°C is com- -DI6__E F parable to that observed with pseudoplasmodia of this strain 84 18 222 4182 6 141 22 that have been grown and allowed to develop at 23.5°C (Fig. '18 3B). With these pseudoplasmodia, the transition occurs at 18.7°C. To measure adaptation, amoebae of D. discoideum strain HL50 -8 were grown at 23.5°C or at 27.50C and exposed to the thermal 0-6 gradients for 16 hr, and their responses were compared with those of the pseudoplasmodia of the same strain. HL50 amoe- bae grown at 27.5°C moved negatively on the gradients with 14 18 22 26 14 18 22 26 14 1822 26 midpoint temperatures below about 21°C, randomly on gra- Gradient Midpoint Temperature (OC) dients with midpoints of 21 and 23°C, and positively with gra- dient above 23TC This shift in transition FIG. 2. Temperature-response curves for strain HL50 amoebae midpoints (Fig. 3A). grown at 23.5'C and exposed for different lengths of time on the ther- from that seen with amoebae grown at 23.50C, for which the mal gradient. Exposure was 3 hr (A), 6 hr (B), 9 hr (C), 12 hr (D), 16 hr transition is about 190C, suggests that thermal adaptation oc- (E), or 20 hr (F). Vertical bars represent the 95% confidence interval curs at the level of the amoebae. Strain HL5O, which has been calculated according to Steel and Torrie (ref. 16, p. 63). Downloaded by guest on September 25, 2021 5648 Developmental Biology: Hong et al. Proc. Nad. Acad. Sci. USA 80 (1983)

-A -B 16 16

a) 8 8 8- CL 0 O I 1 ULI I ( 1 a) --,' 4) 0 III -84 -8_ -16 FIG 3)Tmeauersoscuvsfrtain L0 A me -16j1 1 22 -16-~~~~~~~-. 14 18 22 26 14 18 22 26 14 18 22 26 "-14 18 22 26 Gradient Midpoint Temperature (0C) Gradient Midpoint Temperature (0) FIG. 4. Temperature-response curves ofthe mutant strains H0428 FIG. 3. Temperature-response curves for strain HL50. (A) Amoe- (A) and H0813 (B); amoebae were grown at 23.5°C and exposed for 16 bae grown at 27.50C and (B) pseudoplasmodia grown and developed at hr on the thermal gradient. Vertical bars represent the 95% confidence 23.50C (solid line) or 27.50C (broken line). Both amoebae and pseudo- interval. plasmodia were permitted to migrate on the gradientfor 16 hr. Vertical bars inA represent the 95% confidence interval. Vertical bars inB rep- resent ± one standard error. Directness (r) is a measure of oriented The stages of development that occur after starvation but movement; a value of + 1.0 represents movement directly toward the prior to cell-cell contact are (i) acquired chemotactic sensitivity warmer side of the gradient and 0 represents random movement (14). to cyclic AMP, (ii) acquired capability to relay the cyclic AMP signal, (iii) initiation of autonomous cyclic AMP signalling, and (iv) increased cohesiveness, which is the result of the formation bae of strain H0428 showed positive thermotaxis throughout of contact sites (18). Because the time course of development the entire temperature range (Fig. 4A). Pseudoplasmodia of depends on cell density, the total number of cells, the homo- this strain also showed only a positive response throughout the geneity of the initial population, and culture conditions (19), we entire temperature range (14). At a midpoint temperature of are presently unable to temporally place the development of 180C, amoebae demonstrated their strongest response, and there the capacity for negative thermotaxis in this framework. was a sharp drop in response at midpoint temperatures above Comparisons between the temperature response curve for 2100. This temperature-dependent magnitude of response is HL50 amoebal thermotaxis after 16 hr on the gradient and the consistent with recent stimulus-response results obtained with curve for HL50 pseudoplasmodial thermotaxis reveal remark- pseudoplasmodia of H0428 (17). able similarities. The shapes of both curves, the transition points, Amoebae of strain H0813 demonstrated negative thermo- and the relative strengths of the positive and negative re- taxis on gradients with midpoint temperatures below 210C and sponses are analogous. These results suggested that pseudo- a positive response on gradients with midpoint temperatures of plasmodial thermotaxis could be the product of the directed 24 and 260C (Fig. 4B). This transition from negative to positive movement of individual amoebae within the pseudoplasmodia. thermotaxis is significantly higher than that seen with HL50 To test this idea, the ability of the amoebae to adapt their ther- amoebae grown at 23.50C. Pseudoplasmodia of strain H0813 mal responses when presented with an altered growth tem- showed a transition from negative to positive thermotaxis at perature was tested. The amoebae were grown at 27.50C, as 20.700 (14). This transition is also significantly higher than that opposed to 23.50C, and then exposed to a thermal gradient for seen with pseudoplasmodia of HL50 but it is similar to the tran- 16 hr. This difference in growth temperature raised the tran- sition that H0813 amoebae demonstrate. sition point from 190C to between 20 and 240C. If the pseu- doplasmodial response is a composite of amoebal responses, DISCUSSION the transition point of similarly treated pseudoplasmodia should The low density of amoebae (less than 3 X 103 amoebae per be in this range. The transition point found when these pseu- cm2) used in these experiments permitted us to examine the doplasmodia were examined was 20.50C. This upward shift from effects of development on amoebal thermotaxis. Recently starved 18.7 to 20.50C was consistent with that predicted from other amoebae demonstrate positive thermotaxis throughout a wide thermal adaptation experiments with pseudoplasmodia of other range of temperatures. However, when the amoebae are left strains (13) and with that predicted on the basis of the amoebal on the thermal gradient for 6 hr after they are removed from experiments. bacteria, no positive response can be seen on gradients with Clearly, the best evidence connecting pseudoplasmodial midpoint temperatures of 14, 16, or 1800. This "random" re- thermotaxis and amoebal thermotaxis is provided by the results sponse may be the result of amoebae moving toward the warmer with the mutants. Without exception, the amoebae of these side of the gradient (3-hr time point) and then changing their strains showed alterations similar to those seen with the cor- preferred direction. These results suggest that the ability to re- responding pseudoplasmodia. The easiest interpretation of these spond positively on a thermal gradient is present early in de- data is that pseudoplasmodial thermotaxis is a composite of the velopment and may even be present in growing amoebae. The amoebal thermotactic responses. capacity to respond negatively develops shortly thereafter (be- The one difference found when comparing amoebal and fore 6 hr), and subsequent time points show that it increases as pseudoplasmodial thermotaxis is the thermal sensitivity. The development proceeds. This increase in the ability to respond pseudoplasmodia give a strong response on a thermal gradient negatively is reflected in the upward shift in the transition point of 0. 110C/cm, whereas the amoebae demonstrate a weaker re- as development proceeds. The apparent increase in the tran- sponse on a gradient of 0.22°C/cm. This difference is easily sition temperature from 16 hr to 20 hr may not be significant explained when one considers that there are 104-105 amoebae due to the decreased amplitude of the response, possibly as a per slug. It is reasonable to expect that the combined response result of the length of time on the gradient. of many amoebae would be greater than that of one amoeba. Downloaded by guest on September 25, 2021 Developmental Biology: Hong et al. Proc. Natl. Acad. Sci. USA 80 (1983) 5649

This emphasizes the sensitivity of the thermal responses in D. Dr. M. J. Schneider and Dr. D.-P. Hader, and the technical assistance discoideum. The pseudoplasmodium can respond to a gradient of Douglas DeGaeteno and Sandra Davis. This work was supported by of 0.0005TC across itself. It can be calculated that a single amoeba the U.S. Department of Energy under Contract DE-AC02-76ER01338. within the pseudoplasmodium must be exposed to (and re- 1. Raper, K. B. (1940) J. Elisha Mitchell Sci. Soc. 56, 241-282. sponding to) a gradient of 0.00005TC across itself. This calcu- 2. Robertson, A. & Drage, D. J. (1975) Biophys. 1. 15, 765-775. lation is based solely on the relative widths of pseudoplasmodia 3. Pan, P., Hall, E. M. & Bonner, J. T. (1972) Nature (London) New and amoebae and assumes that the temperature measurement Biol. 237, 181-182. is 4. Hader, D.-P. & Poff, K. L. (1979) Exp. Mycol. 3, 121-131. spatial. The best evidence that the measurement is spatial in 5. Hider, D.-P. & Poff, K. L. (1979) Photochem. Photobiol. 29, 1157- pseudoplasmodia is the observation of Poff and Skokut (12) that 1162. the first discernible turn by a pseudoplasmodium is always in 6. Hong, C. B., Hader, M. A., Hader, D.-P. & Poff, K. L. (1981) the "correct" direction. In contrast, the first turn would be ran- Photochem. Photobiol. 33, 373-377. dom for a temporal sensing system. It will now be necessary to 7. Fisher, P. R., Smith, E. & Williams, K. L. (1981) Cell 23, 799- examine amoebae to see if their thermal measurement is spatial 807. 8. Bonner, J. T., Clarke, W. W., Jr., Neeley, C. L., Jr., & Slifkin, or temporal. M. K. (1950) J. Cell. Comp. Physiol. 36, 149-158. The results presented in this paper show that the amoebae 9. Francis, D. W. (1964)J. Cell. Comp. Physiol. 64, 131-138. of D. discoideum demonstrate directed movement when placed 10. Poff, K. L., Butler, W. L. & Loomis, W. F., Jr. (1973) Proc. Natl. on a thermal gradient. In spite of the fact that the amoebae are Acad. Sci. USA 70, 813-816. not permitted to aggregate and form pseudoplasmodia, amoe- 11. Poff, K. L. & Butler, W. L. (1974) Photochem. Photobiol. 20, 241- bae show responses after 16 hr on the thermal gradient that are 244. 12. Poff, K. L. & Skokut, M. (1977) Proc. Natl. Acad. Sci. USA 74, 2007- very similar to the responses of pseudoplasmodia on similar 2010. gradients. These similarities between the amoebal and the 13. Whitaker, B. D. & Poff, K. L. (1980) Exp. Cell Res. 128, 87-93. pseudoplasmodial thermal responses when mutants and ther- 14. Schneider, M. J., Fontana, D. R. & Poff, K. L. (1982) Exp. Cell mal adaptation are examined strongly suggest that pseudoplas- Res. 140, 411-416. modial thermotaxis is a result of the thermotactic responses of 15. Sussman, M. (1966) in Methods in Cell Physiology, ed. Prescott, the composite amoebae. These observations should simplify D. (Academic, New York), Vol. 2, pp. 397-410. the search for biochemical and steps in 16. Steel, R. G. D. & Torrie, J. H. (1980) Principles and Procedures biophysical the ther- of Statistics, A Biometrical Approach (McGraw-Hill, New York), mosensory transduction pathway because they demonstrate that 2nd Ed. thermotaxis is not a function of the multicellular organization. 17. Fontana, D. R. (1982) Dissertation (Michigan State Univ., East Lansing). The authors gratefully acknowledge assistance with the statistical 18. Loomis, W. F., Jr. (1979) Dev. Biol. 70, 1-12. analysis provided by Dr. Charles Cress, the helpful discussions with 19. Gingle, A. & Robertson, A. (1976) J. Cell Sci. 20, 21-27. Downloaded by guest on September 25, 2021