Circadian communication between unicells?

�9 1985 by The Humana Press Inc. All rights of any nature whatsoever reserved. 0163-4992/86/0801-0047504.20

Circadian Communication Between Unicells?

Effects on Period by Cell-Conditioning of Medium

HELLMUTH BRODA, t DOUG BRUGGE, KEIICHI HOMMA,

AND J . WOODLAND HASTINGS*

Department of Cellular and Developmental Biology, The Biological Laboratories, Harvard University, Cambridge, ~VIA 02138, USA

Received March 25, 1985; Accepted September 12, 1985

ABSTRACT

Populations of Gonyaulax polyedra, in two different phases, about 11 h apart, were mixed, and the intensity of their spontaneous bioluminescence glow recorded for about 2 wk under conditions of constant dim (35 _+ 3 p.E/m2/s) white light and constant temperature (19.0 + 0.3~ The phases and amplitudes of glow signals recorded from mixed cultures were compared with those obtained from the arithmetic sum of the intensity data from two control vials. Peaks in control cultures generally remained separate, but there was a spontaneous increase in the period beginning 6-11 d after the onset of constant conditions. This did not occur in cultures in which the medium was exchanged with fresh medium every 2 d. In the actual mixes of two cultures there was a merging of the two subpeaks in the signal, which did not occur when the medium was exchanged. The results indicate that conditioning of the medium by cells may affect the period of the circadian rhythm and that this might result in a type of communication.

Index Entries: Gonyaulax polyedra; circadian communication between unicells; cell-conditioning of medium, effects on period; bioluminescence cell communication; circadian rhythms.

tVisiting Scholar, supported by the Deutsche Forschungsgemeinschaft; present address: Ciba-Geigy AG, Basel, Switzerland

*Author to whom all correspondence and reprint requests should be addressed.

Cell Biophysics 4"7 Vol. 8, 1986

48 Broda et ai.

INTRODUCTION

Studies of circadian rhythms in higher organisms, ranging from in- sect (1) to avian and mammalian systems (2,3), have implicated interactions between cells and tissues mediated by humoral agent(s). In con- trast, no such interaction has been reported in unicells or fungi. In a recent report from this laboratory, no difference was found between mixed out-of-phase cultures of Gonyaulax polyedra and the values calculated from the arithmetic summations of the time series data recorded from two individual cultures, but measurements were continued for no more than 5 d after mixing (4). Negative results were also obtained in earlier studies with G. polyedra (5), as well as with Euglena (6,7) and Acetabularia (see ref. 8). In the experiments reported here we were able to follow the behavior of mixed G. polyedra cultures under constant conditions for 14 d and longer. The results provided suggestive, but not definitive, evidence for circadian communication in a unicellular system.

MATERIALS AND METHODS

Measurements of the circadian rhythms were made with equipment described previously (9). The software procedures for data acquisition, storage, and analysis were altered significantly. Key aspects will be described briefly; a detailed description will be published separately (10).

The equipment consists of two similar tracks, each able to hold 30 vials, each vial containing a cell culture maintained at constant temperature by water circulation (19.0 _+ 0.3~ in these experiments) and ex- posed to constant dim light (from cool white fluorescent tubes, at an intensity of 35 _+ 3 ~E/m2/s), necessary to keep the cells alive. The vials are numbered; even numbers assigned to one track, odd numbers to the other. A photomultiplier tube travels along each track, stops at each vial, and measures the bioluminescence intensity. During the time when the photomultiplier is positioned under a given vial, it blocks exposure of three vials to the dim light. Thus, cells nominally in constant light (LL) spend 10% of the time in the dark. How this is distributed in time de- pends on the duration of individual measurements; in Expts. 139 and 152, measurements were made for 60 and 40s, respectively. During this time, measurements of cell bioluminescence [both glow and flashes (11,12)] of a given vial of cells are acquired and stored by the Apple com- puter. Thus, data points on each vial were acquired every 68 and 45 min. for the two experiments. The program distinguishes between glow and flashes; the results reported here are based only on the glow, but the flash data do not appear to differ.

A dense (about 12,000 cells/mL) culture of G. polyedra (GP70), grown in a modified seawater medium (13) at 20 + 2~ under conditions of 12-h alternating light and dark cycles (LD, 12:12), was used to inoculate

Cell Biophysics Vol. 8, 1986

Circadian Communication Between Unicells? 49

two flasks, one of which (referred to as culture B) was placed back into the original LD cycle and the other (referred to as culture A) transferred into an LD 12:12 cycle in which the onset of light was 11 h later. The actual phase differences between the two cultures after being transferred to LL were somewhat different from this and also somewhat different in the two Expts., 139 and 152. As determined by extrapolation to d 0 of the regression through the glow peaks, these were 9 h 45 min for Expt. 152 and 11 h 30 min for Expt. 139. There is an apparent phase angle change at the time of transition from LD to LL, and it may be that the magnitude of this differs, depending on whether the light or dark period of LD came last before transition to constant dim light, i.e., DL-LL v LD-LL.

ExpL 139

Three days after dividing the culture, additional fresh medium was added to each culture and 10- and 15-mL aliquots of the cells were pi- peted into scintillation vials. Cell densities at this time were about 5000-6000 and 2000-3000 cells/mL for cultures A and B, respectively. One day later the vials were transferred to the recording apparatus. After 5 d of recording under constant conditions, cells were mixed according to two protocols. In the first (Table 1), two vials containing 15-mL of each of the two cultures were placed on either side of an empty vial in the recording apparatus; the experimental mix was made with 5 mL from each vial, and the two flanking vials remained as controls. The second protocol (Ta- ble 2) utilized sets of four vials (two of each phase, 10-mL each) placed in adjacent compartments on the recording apparatus. Two of these vials from different phases were then mixed and divided to give two identical experimental mixes, and the other two vials remained as controls. Con- trols were subjected to the same pipeting procedures used with experimental mixes. There were four sets of each procedure, and data recording was continued for 12 d after mixing.

ExpL 152

In this protocol, the A and B cultures had the same cell density (about 4000 cells/mL); they were mixed at the time of pipeting into the scintillation vials and then placed into the measuring apparatus. Pipeting was done 5 mL at a time, resulting in the same treatment for mixed (5-mL A + 5-mL B) and control (5-mL A + 5-mL A, or B and B) cultures. After pipeting, some vials were immediately filtered and resuspended in fresh medium. In others, A cells were resuspended in B medium and vice versa. Another set of vials was filtered and resuspended an additional three times at 2-d intervals.

Raw data were plotted directly from the files on a digital plotter (Watanabe digiplot). For the analysis of phase differences between the calculated and experimental time series data, we used a normalizing pro-


50 Broda et al.

TABLE 1 Three-Vial Mixing Protocol"

FRP/' Actual peak times,' Vial no. Cell phase hours _+ SD local time

07 A 23.0 _+ 0.63 1151 09 AB 22.4 _+ 0.34/-- 0834/-- C: 7 + 11 A + B 22.8 _+ 0.59/ 1205/2237

23.1 _+ 0.34 11 B 23.01 _+ 0.01 2142

49 A 23.2 _+ 0.52 1045 51 AB 22.5 _+ 0.35/ 0834/~

22.3 _+ 0.35 C: 49 + 53 A + B 23.0 _ 0.21/ 1128/2314

23.1 +_ 0.54 53 B 23.01 _+ 0.01 2248

12 A 23.0 _ 0.41 0934 10 AB 22.5 + 0.64/-- 0728/--- C: 08 + 12 A + B 22.8 _+ 0.39/-- 0958/-- 08 B 21.9 +_ 0.64 1402

50 A 22.7 +_ 0.37 0834 52 AB 22.3 _+ 0.50/--- 0516/~ C: 50 + 54 A + B 22.7 - 0.68/--- 0828/-- 54 B 22.5 _+ 0.45 1508

"Expt. 139. ~FRP of A phase cells, B phase cells, and mixes thereof (AB), compared with

calculated summations of the two (A + B). Two values are given in which the two components in the mixes could be distinguished.

'Actual times of peak luminescence on d 13 for A phase cells, B phase cells, and mixes thereof (AB), compared with calculated summations of the two (A + B). Two values are given in which the two components in the mixes could be distinguished.

gram to obtain a data set with a constant amplitude of the oscillation. These normalized data were also used for phase plots; data were double plotted. The times at which individual peaks occurred were determined with a program "peakf inder" written to fit a parabola to five data points centered on or about any maximum. A regression line through points on any number of successive days (selectable by operator) was calculated by the least-squares method to give a value for the free-running period (FRP).

For further analysis of phases, the method of complex demodulat ion (14) was applied to all mixed cultures and to all corresponding calculated traces from vials of Expts. 139 and 152. This method provides exponen- tial multiple smoothing of the real and complex part of the signal separately; subsequent calculations yield the circadian component in the data



TABLE 2 Four-Vial Mixing Protocol ~

F R P , ~' Actual peak times, Vial no. Cell phase hours _ SD local time

13 A 22.9 _+ 0.82 1045 15 AB 22.6 _+ 0.41/ 0939/---

22.4 _+ 0.45 17 AB 22.8 *- 0.42/ 0939/--

22.4 _+ 0.46 C: 13 + 19 A + B 23.0 _+ 0.12/ 1146/2131

23.0 _+ 0.17 19 B 22.9 _ 0.33 2036

41 A 22.7 _+ 0.43 1045 43 AB 23.1 + 0.67/- 1151/-- 45 AB 22.7 + 0.38/ 0939/--

22.4 _+ 0.34 C: 41 + 47 A + B 22.9 _+ 0.24/ 1013/2357

23.2 _+ 0.43 47 B 23.0 + 0.01 2353

14 A 22.7 _+ 0.40 0939 16 AB 22.5 _+ 0.45/ 0728/--

21.9 _+ 0.38 18 AB 22.4 _+ 0.37/ 0622/--

21.8 _+ 0.64 C: 14 + 20 A + B 22.8 _+ 0.28/ 0946/2028

22.9 _ 0.49 20 B 23.1 _+ 0.36 1931

42 A 22.1 + 0.64 0622 44 AB 22.5 _+ 0.57/ 0622/--

21.9 _+ 0.55 46 AB 22.6 _+ 0.46/-- 0728/--- C: 42 + 48 A + B 22.3 _+ 0.68/ 0605/2021

22.5 _+ 0.64 48 B 22.7 _+ 0.35 1825

"Expt. 139. q:RP of A phase cells, B phase cells, and mixes thereof (AB), compared with

calculated summations of the two (A + B). Two values are given in which the two components in the mixes could be distinguished.

LActual times of peak luminescence on d 13 for A phase cells, B phase cells, and mixes thereof (AB), compared with calculated summations of the two (A + B). Two values are given in which the two components in the mixes could be distinguished.

set, its a m p l i t u d e and phase . This m e t h o d has the a d v a n t a g e of taking into a c c oun t the total t ime series r a the r than jus t the peak of the maxima.

Physical ly, each of the two 30-vial tracks are separa te , bu t t h e y are cons t ruc t e d the same, and in these expe r imen t s the e n v i r o n m e n t a l con-

Cell Biophysics VoL 8, 1986

52 Broda et aL

ditions, notably light intensity and temperature, were intended to be the same, and were within the limits specified. From previous work, it was believed that such small differences would not result in significant differences in the FRP, but they evidently did: In Expt. 152, the control vials (Nos. 11, 19, 21, 37, and 49) in the odd track had a period of 22.56 _+ 0.10 h; in the even track (Nos. 10, 26, 34, 42, and 44) the period was 22.04 +_ 0.04 h. Thus, contrary to design, data from the two instruments, one with odd- and the other with even-numbered vials, were not pooled (see Tables A1 and A2 in Appendix).

If, in a mixed culture (AB), there is no interaction between cells having different phases (the null hypothesis) then the measured light emission from the mixed culture should not differ from the point-by-point addition (referred to as A + B) of the time series data measured from the A and B cultures, separately. The calculated result of mixing two out-of- phase cultures was determined with the aid of an "addition" program written for this purpose. A "subtraction" program, completely analo- gous to the addition program, was then used to obtain difference curves between calculated and experimental mixes or between two controls and identical experimental mixes. Data presentation in a phase plot was used to visualize both phase differences and differences in the FRP, while still maintaining the shape of the oscillations in the original data.

RESULTS

In the first protocol of Expt. 139, two out-of-phase vials, one A (no. 7) and one B (no. 11) (Fig. la,c) initially contained 15 mL. At the time of mixing, 5 mL was removed from each to create a mixed (AB) vial (no. 9; Fig. lb; solid line). The difference in amplitude of the two peaks reflects the difference in the cell densities of the two cultures. A calculated result (C) (Fig. lb, dotted line; C: 7 + 11) was derived from the time series data from vials 7 and 11. The fact that this differs from the actual mix, and increasingly so after several days, appears from visual inspection. Using the subtraction program with amplitude normalized data, the difference between the experimental mix and the calculated trace generates as oscillation with an increasing amplitude [Fig. ld; C: 9-(7 + 11)]. A similar subtraction of two control cultures (Fig. le) illustrates the reproducibility from vial to vial. The apparent small phase difference in the first days after mixing appears to be a transient phenomenon; in other pairs the differences were less.

Data from this and three additional sets employing this protocol are summarized in Table 1. In all four cases the mixed culture itself exhibited a period shorter than that of the calculated trace, so that 8 d after mixing (d 13) the time at which the luminescence peak of the mixed culture occurred was earlier. The disappearance of the A and B components as distinct peaks also occurred earlier in the mixes, as compared to the cal-



>-

Z UJ l-- Z

0 . .1

(e) T - 1 3

(d) 6.-(7f 1 I)

(c) 11

(b) g ( 7 f l 1 ) - -

[a) 7"

I I I l I i J I I I l I I

6 8 lO 12 14 1 EXP. 139

TIME (Days in LL) Fig. l. Experiment 139: Bio[uminescent glow emission (ordinate) with

time (abscissa; days after placing cells in constant light, LL) from culture A (panel a), culture B (panel c), and a mix of the two (AB; panel b, solid line), compared with the arithmetic sum of the two (A + B; dotted line). The difference between AB and A + B is given in panel d, and the difference between two vials containing the same population is given in panel e (A - A, vials 7 and 13). The vial numbers marked on the graph refer to Table 1. Scaling for panels a through c is identical. The calculated subtractions are scaled for maximum amplitude for each panel.

culated traces; in two of the four calculated cases the two peaks were still resolvable on d 13, whereas this was not so in any of the mixes.

The four sets of data from the companion protocol in Expt. 139 are summarized in Table 2. Each set started with four vials of 10-mL each, two A and two 13 cultures. Two were mixed on d 5 to give two experimental (AB) mixes (e.g., Fig. 2b,c; nos. 17 and 15), whereas the other two (Fig. 2a,d; nos 19 and 13) remained as controls, from which the calculated trace (line with points) was obtained. The deviations of the actual mixes, AB, from the calculated trace, again became evident only after several days, and were plotted as the difference between the two (Fig. 2e). The identity of the two mixes and the reproducibility of measurements from adjacent vials in the recording apparatus are shown by subtraction of the normalized time series data from vials 15 and 17 (Fig. 2f). Similar subtractions were carried out on the other pairs of similar vials with similar results.

In Expt. 152, the protocol involved mixing at the very outset of the run, and included some vials in which the med ium was exchanged with either fresh medium or with med ium from the other culture. The location of the cultures on the measur ing apparatus and their disposition is given


54 Broda et al.

I--

Z LLI I-- Z

O _1

(a) 19 --

(b) 17 1 3 + 1 9

(c) 15 13+19 - -

(d) 13

(e ) 1 5 - ( 1 3 + 1 9 )

( f ) 1 5 - 1 7

I L i i J i l l 2 i ~ L l l 6 6 8 1 0 1 4

EXP. 139

TIME (Days in LL)

Fig. 2. Experiment 139: Culture A, panel a; Culture B, panel d. Experi- ment as in Fig. 1, except that two duplicate mixes were made (panels b and c). The difference between one of these (vial 15) and the arithmetic sum (A + B) is given in panel e; the difference between the two mixed populations (15 and 17) is shown in panel f. Vial numbers refer to Table 2.

in the Appendix in Tables A1 and A2 for odd and even vials, respectively. Traces for all the experimental mixes of the odd vials, compared with traces calculated from controls (A + B) in nearby vials, are also given in the Appendix (Fig. A-l). In this case, the amplitudes of the two peaks were similar, as were the cell densities of the A and B cultures. Most mixes differed considerably from the calculated traces; the two components (A and B) generally appeared to merge in the mixes, whereas double peaks usually persisted in the calculated traces (Fig. A-l). This may be visualized better in phase plots of normalized raw data (Fig. 3a,b) and in plots of the maxima (Fig. 3c,d), the same data as Fig. A-lf. In the mixed cultures, the two peaks corresponding to the two cultures, A and B, have slightly different periods and ultimately merge (Fig. 3a,c). On the other hand, in the calculated traces, based on the separate cultures (Figs. 3b,d), these peaks remain distinct.

The period differences and merging of peaks in mixed cultures did not occur when the cells were resuspended in fresh medium every 2 d for the first week. This is illustrated in the plots of Fig. 4 (same data as in Fig. A-li). In mixed cultures that had received fresh medium every 2 d, the peaks of the two components remained separated and their slopes paral- lel (Fig. 4a,c); the same was true for the traces calculated from control


Circadian Communication Between Unicells? .5.5

cultures (A and B), which were also provided with fresh medium every 2 d (Figs. 4b,d). This result is to be expected if the mixed-culture effects are mediated by the culture medium. However, a single addition of fresh medium at the start of the experiment was not sufficient (Fig. A-lh; Ta- bles A1 and A2; vials 31, 33, 35, 20, 22, and 24). Also, exchanging the media of A and B cells did not show a detectable modification of the observed rhythm (Tables A1 and A2; vials 7, 9, 39, 41, 6, 8, 46, and 48).

In cultures not provided with fresh medium, a spontaneous period change usually occurred after 7-11 d in both mixed and control cultures. The day at which this occurred in the different vials is recorded (Appen- dix, Tables A1 and A2, day of break) and its extent is evident from the difference between the early period (before the change) and the late period. Figure 5 shows that this period change did not occur in cells repeat- edly provided with fresh medium.

DISCUSSION

In the present experiments, the rhythmic patterns of the luminescence of mixed out-of-phase cultures have been found to differ from those calculated on the basis of the two separate cultures. The differences were not large, but in most cases there were features indicative of cell interactions. For example, in mixed cultures the peaks corresponding to the two cultures merged, whereas they generally remained distinct in the calculated traces (Fig. A-l). Even more striking was the effect of repeat- edly replacing the culture medium with fresh medium. In these cases, the rhythmic peaks in the mixed cultures paralled those calculated from separate out-of-phase cultures. This suggests the involvement of a substance in the medium on the period of the rhythm. Indeed, a spontaneous increase in the FRP generally occurred in control (nonmixed) cultures after 7-11 d in constant conditions, and this change was also absent in cultures provided with repeated changes of fresh medium.

An effect of medium composition upon the period in Gonyaulax has recently been reported by Sweeney and Folli (15); nitrate deficiency shortened the period. Depletion of nitrate could not, therefore, account for the longer period in older cultures reported here. However, the result illustrates the possible effects upon period of substances in the medium. Feldman and Dunlap (16) discussed experiments that suggest the exist- ence of a "phasing" substance in Neurospora, but the evidence has not been reported.

One can speculate on how the action of a putative phase entraining substance might act, and under what conditions one or another of the two cultures might be dominant. Neither the mode of action (be it production and/or consumption) nor the nature of the substance(s) involved is known, but the effect is related to cell density. Cultures containing


56 Broda et aL

UJ m

Z ) - .r r't

2

4

6

8

10

12

!5"3 i~

~ F - 2 4 H r . - ~

�9 ~ ~"=,-,,j . . 1 5 + 1 9

~ ' ~

N 2 ~ N 2 ~

~111K. JGL ;4~

TIME (Hours) Fig. 3a,b.

152:

,1

4 r r UJ m

Z ,= >. ,:, < a

12

17

12 I I I

I 3

6 8

18 12

14 6

I 6

4

123.3:1:0.23J

I 12

1 :-: ~_~ 4. .~. ~, 36 I I I

1 2 2

123.0•

r ='~~I!i 6

.?.,,." -'F'

�9 ml i ~ "~

I I I I

24 30 36 42 18

TIME (Hours)

Fig. 3c.

4,2

q

Ii 13 15 17

i I--I

19 ,.- ~21 ':,. ";, k.L.


Circadian Communication Between Unicells? .57

f t . I l l

II]

Z

< C~

15+19 152" I:-. t .", �9 " 1 ,-, 1 =-J

! I 1 m

, - !

r

4 4

6

r

m

18- N

" - - ! 1

14

. J

i "

9 II 13 i.=; 17

19 c l

�9 D "3' .- . j

25 I 6

":' 4 L_

1 "2

122.6+0.16 I 1 ~ -

,$~

P2

I'

123.3+o.o5 I,~

i I

�9 = . , _ _ ' . ' | REI ~ 4c I I I /,

2 / 8 .~-I I "2

14 y,,l', 1 ~l

1 :--', C, D ,. rll I~ I]l~ .:,., l 2

I I I 12 18 24 30 36 42

TIME (Hours) Fig. 3d. Experiment 152: Phase plots of normalized raw data from a

mixed AB culture (panel a, vial 17) compared with the summation of vials 15 and 19 (A + B; panel b), showing the apparent merging of peaks in the mixed culture. From plots of the maxima, as determined by peakfinder, values for the free running periods of the different components were calculated (panels c, d). These are noted on the graphs and, along with those for all other vials of Expt. 152, in Tables A1 and A2.

only 500 cells/mL at d 0 exhibited only a single (short period) component in the FRP (data not shown).

Although the time of onset of the period change is variable (Tables A1 and A2), the reproducibility between similar cultures (also, see ref. 17) was impressive, as shown by subtraction (Figs. 1,2) and also by the rather small standard errors for the mean and standard deviations of the peak times on d 13 (Tables A1 and A2; vials 11, 19, 21, 37, and 49, and vials 10, 26, 34, 42, and 44). The corresponding vials with cells resuspended four times in fresh medium had significantly different peak times on d 13. The same was true for mixed cultures.

The fact that the differences between mixes and the calculated values were not so large in Expt. 152 may be because of the production in near equal amounts of the putative entraining substance by both cultures, such that only weak interaction/entrainment occurs. With Expt. 139, the A culture was apparently dominant, and its cell density was two to three

Ce//Biophysics Vol. 8, 1985

58 Broda et al.

rr LU m

Z >. < t~

n- U.I

:3 Z >- < C~

10

12

' 'l ~" ''~' 2f5

2

4

6

8

' ~ - - 24 Hr. - - - ~

'] ~ ~_,:, 23+27

' I ~ [ ~ 24 Hr.

TIME (Hours) Fig. 4a, b.

1 5 2 : 2 5

m

r -

4 -

6

0 I I

IP~,

6 12 I I

1 0

4 , l l

6 7 B 9 8

12 14 i~.

1 8 122.6+o.71

7 '5

I I 6 12

I :1 m ,-, 24

8

.ml

13

,:3

18 24 30

3~3 36 42 I I ".4

[ 22.5+_0.81

oJ

i i 13

. J

i / 19 2

2 1 -" 23 2 4 25 26

�9 D 7 .. r

I I 36 42

TIME (Hours)

Fig. 4c.

Ceil Biophysics VoL 8, 1986


152: 23+27

6 1 2 1 8 ? 4 ~ R R tq 4 :-'

- ' t 0 ~ 4 4 III -:, II~ -

:~

Z ,-, ,-, - -

>- < - 13 a 10-5

12- m

( '

1 "7 . i

. I

I 22.7_+0.8

0

122.6_+0.71

,~_ �9 I ~

d :, 1 7

'-' 19 1 '~ '-' _ ..~

I I I I I I

6 12 18 24 30 36 42

q

I

TIME (Hours)

Fig. 4d. Experiment 152: Plots illustrating the effects of four replacements with fresh medium on the normalized patterns of luminescence (panels a and b) and the free-running periods (panels c and e). In the mixed culture (vial 25; panels a and c), the two components corresponding to the parent cultures remained distinct, as did those in the separate cultures (vials 23 and 27).

times that of B. Some amplitude differences were observed already on the first day after mixing, especially noticeable in the valleys in Expt. 139. Although amplitude effects could have been caused by mixing errors, the differences were sometimes much too large to be explained in that way and may have been caused by the putative substances.

If a "circadian" substance does exist and is cell density dependent , its effect might be greater and detected more promptly if the effect of the medium from a high-density culture is tested upon a responding culture at low density. Populations could be kept separate by a filter that would allow only the media to exchange. Alternatively, a nonluminous mutant might serve as the high-density culture. These experiments might also test the possibility, albeit an unlikely one, that information exchange between cells involves bioluminescence emission and its detection by the second culture (12).


60 152:

t....

4

ul

m 6

z 8 >- <

23

18-

i ' " -

6 I

7 9

18 11

12

I

6

12 I 1

,.,,== s R -.=

4 ,_1

I 12

i - ! = - i

1 ~", "' �9 . 4 ,.X6 36

~;t " '" , ~ l I22.6+0 7 ]

I ~I6s 1 l I 1 8 2 4 3 0 3 6 4 2

Broda et aL

42

6

15

152:21 6 n

1

K.

4

W 6

z 8

1.2

7 8

9 18

i i 12 13 14

I 6

._1

12 !

2

TIME (Hours) Fig. 5a.

1 '-',:, ,_;', 28_ 36 ~2 I I I "JI

C 1 / 2 2 . 4 _ + 0 . 1 ]

/ 8 [23.3_+0.2]

r

4

I I t 1 1 1 2 1 8 2 4 3 0 3 6 4 2

TIME (Hours) Fig, 5b, The effect on the free-running periods of repeated replacements

with fresh medium (vim 23), compared with no medium change (vial 21). Peri ods (Table A1) noted on Fig.


Circadian Communication Between Unicelis? 61

REFERENCES

1. Truman, J. W. (1972), J. Comp. Physiol. 81, 99. 2. Halberg, F. (1983), Am. J. Anat. 168, 543. 3. Takahashi, J. S. , and Menaker, M. (1984), Biological Regulation and Develop-

ment, vol. 313, (Goldberger, R. F., and Yamamoto, K. R., eds.), Plenum, NY, pp. 285--303.

4. Sulzman, F. M., Gooch, V. D., Homma, K., and Hastings, J. W. (1982), Cell Biophys. 4, 97.

5. Hastings, J. W., and Sweeney, B. M. (1958), Biol. Bull. 115, 440. 6. Brinkmann, K. (1966), Planta (Berl.) 70, 344. 7. Edmunds, L. N., Jr. (1971), Biochronometry (Menaker, M, ed.), National

Academy of Sciences, Washington, DC, pp. 594-611. 8. Schweiger, H.-G., and Schweiger, M. (1977), Int. Rev. Cytol. 51, 315. 9. Taylor, W., Wilson, S., Presswood, R., and Hastings, J. W. (1982), ].

Interdisci. Cycle Res. 13, 71. 10. Broda, H., Gooch, V. D., Taylor, W., Aiuto, N., and Hastings, J. W. (1986),

J. Biol. Rhythms (submitted). 11. Krasnow, R., Dunlap, J., Taylor, W., Hastings, J. W., Vetterling, W., and

Gooch, V. D. 1980, J. Comp. Physiol. 138, 19. 12. Krasnow, R., Vetterling, W., Haas, E., Dunlap, J., Taylor, W., and Has-

tings, J. W. (1981), Bioluminescence: Current Perspectives (Nealson, K. H., ed.), Burgess, Minneapolis, pp. 52-63.

13. Guillard, R. R. L., and Ryther, J. H. (1962), Can. J. Microbiol. 8, 229. 14. Broda, H., Dirk, J. and Schweiger, H.-G. (1983), Cell Biophysics 5, 43-59. 15. Sweeney, B. M. and Folli, S. I. (1984), Plant Physiol. 75, 242. 16. Feldman, J. F., and Dunlap, J. C. (1983), Photochemical and Photobiological Re-

views, vol. 7 (Smith, K. C., ed.), Plenum, NY, pp. 319-368. 17. Njus, D., Gooch, V. D. and Hastings, J. W. (1981), Cell Biophysi. 3, 223.

APPENDIX

In general, two subpopulations in a mixed culture have a tendency to merge if "rA -- "r~ is positive (where "r is the free-running period) so that the lagging subpopulation, B, will "catch up" with the leading and dominant subpopulation A. In Expt. 139, the two subpopulations clearly tend to merge when mixed, whereas they remain essentially separate in math- ematical calculations of mixes. In Expt. 152, the differences are less, but in most of the mixes (Fig. la) the peaks appear to merge b y comparison with the distinct peaks present in the arithmetic summations. It may be that this is because of the fact that the cell densities of the two populations were approximately equal in this case, whereas in Expt. 139, the cell densities of the two cultures were different (see Discussion).


p=

z uJ ) - z

0 ,,.I (0

49

53

I I

(c) 51

5 5

13

, ~ t ; , . , / d

( f ) ' ~ ,, , ., , . .

17

. : : ; / / cg~ " ' : : f / i

47

;; �9 a ,

(h) 33

(i)

25

, , J , , , , , . , J , 6 8 1 0 1 2 1 4 1 6

EXP. 152 T IME ( D a y s in LL)

Fig. A-1. Experiment 152: Mixing experiments similar to those shown in Fig. 1, but with A and B cultures having similar cell densities at time of mixing. (Vial numbers refer to TableA1). The traces of the luminescent glows of cultures A and B are shown in panels a and b, respectively. Panels c, d, e, f, and g show AB mixes (solid lines) compared with the arithmetic sum of two nearby separate A and B cultures (dotted lines). The cultures shown in panel h (vials 31, 33, and 35) were filtered and resuspended in fresh medium at the time of mixing cultures. Those in panel i (vials 23, 25, and 27) were treated in this way three additional times at 2 d intervals.

Ceil Biophysics Vol. 8, 1986


TABLES A1 AND A2. EXPERIMENT 152

Tables A1 and A2 used similar protocols, but were measured in different vial tracks (numbered odd and even), so the values must be con- sidered separately. Either phase A or B cells or a mixture thereof (AB), were used (column 2). The protocols included some vials in which the medium was changed at the outset of the experiment, as indicated in column 3, with either fresh medium (f) or medium from the culture of the other phase ( = a or b). In some vials, cells were resuspended in fresh medium four times over the course of the first week (f4). In most cases, a spontaneous change in the period occurred after about 1 w (see Fig. 5), so values for both the "early" and "late" free-running periods are given (columns 4 and 5; number of days used in calculating the values shown in parentheses) along with the day on which the "break" occurred (column 6). Where two components in mixes could be distinguished, the values are given in the line below, with the value corresponding to the A-culture component always being given first. The times at which peak luminescence occurred on d 6 and 12 are given in columns 7 and 8, respectively.


64 Broda et al.

D D 0 I

<

o

,-.~ "43

o -,z

~r3 "t3 .o +t

,.~ u'3

.o §

r'~ ~

; > - . ~

hi3

~ r't

=~

+1 +1 +1 +1 +1 +1 +L +1 +1 ~ +1 +1 +1 +l +1

v

c5 4-1

C'4

i t l l ~ l

. . . . . . ~ . . ~ ~ ~ ~ h 0 0 0 0 0 0 0 0 0 0 ~ 0 ~ 0 0 ~

+1 +1 +1 +1 +1 +1 +1 +! +1 +1 +1 +1 +1 +1 +! +1 +1 +1 +1 +l +1

= = < < ~ + = ~ + < < < ~ + = < < <

m

+ ~ +


Circadian Communication Between ~lnicells? 65

d d N N d d d d d d d d d d d " "

�9 < <~ .<

+ + +

Cell Biophysics Vol. 8, ] 986

66 Broda et aL

Z >

I t'x 1 < r,1

<

o

o +l

5

I I ~ I I I I ~ ~ I ~ I I ~

v

go

cc

II lll § +I § + I ~ § +I +I +I +I § +~ +~

I ~ ~

~ v

+1 +l +& +1 § +1 +1 +l +1 +1 § +1 +1 +1 +1 +1 +L +1 +b § § +l

< < <

+ + +

Ceil Biophysics VoL 8, 1986


+I +I +I +I ++ +I +I +I +I +I +I +I +I +I +I +I +I

+1 +t +1 +1 +1 +1 +1 +L +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1

< + < ~ + ~<~ + < < < = = m ~ + < <~ <~ <~ <~

+ + + +

D o

Ceil Biophysics Vol. 8, 1986

Documents

Circadian communication between unicells?