J Comp Physiol A (1997) 181: 41±46 Ó Springer-Verlag 1997 ORIGINAL PAPER K. Hasegawa á Y. Tsukahara á M. Shimamoto K. Matsumoto á Y. Nakaoka á T. Sato The Paramecium circadian clock: synchrony of changes in motility, membrane potential, cyclic AMP and cyclic GMP Accepted: 23 January 1997 Abstract The behavior of a ciliate protozoan, Paramecium, is known to represent the electrical state of the cell membrane, and regulation of the membrane potential and ciliary motion are known to involve cAMP and cGMP. The present study shows the synchrony of circadian changes in motility, resting membrane potential and cyclic nucleotides in P. multimicronucleatum. Using an automated system for tracking isolated single microorganisms, the isolated Paramecium cells are con®rmed to swim fast and straight during the day (and subjective day) and slowly, with frequent turning, at night (and subjective night). The resting membrane potential is more negative during the day than at night. cAMP and cGMP concentrations oscillate in a manner, such that both cAMP and cGMP are higher during the day (or subjective day) than at night (or subjective night). The ratio of cGMP to cAMP during the light and dark cycle (LD) ¯uctuates, paralleling the ¯uctuation of the resting membrane potential measured during the LD. These results suggest that the Paramecium will provide an excellent model to explore daily and circadian orchestration of second messengers mediating signals from ambient light/dark cycles and circadian pacemaker to ion channels and cilia, directly involved in daily and circadian cellular outputs of resting membrane potential and motility. Key words Paramecium á Circadian periodicity á Motility á Resting membrane potential á Cyclic nucleotides Abbreviations DD constant darkness á LD light and dark cycle á PKA cAMP-dependent protein kinase á TF traverse frequency (cells/h) á 8-Br-cAMP 8 bromo cyclic adenosine mono-phosphate Introduction K. Hasegawa (&) á T. Sato Department of Physiology, School of Medicine, Kitasato University, 1-15-1 Kitasato, Sagamihara, Kanagawa 228, Japan Fax: +81-427/78-8441, e-mail: khase@medcc.kitasato-u.ac.jp M. Shimamoto Department of Psychiatry, School of Medicine, Kitasato University, 1-15-1 Kitasato, Sagamihara, Kanagawa 228, Japan Y. Tsukahara Photodynamic Research Center, The Institute of Physical and Chemical Research (RIKEN), 19-1399, Koeji, Nagamachi, Aoba, Sendai 980, Japan K. Matsumoto Educational Promotion Foundation for Oriental Medical Science and Techniques, 3-26-16 Nishi-Waseda, Shinjuku, Tokyo 169, Japan Y. Nakaoka Department of Biophysical Engineering, Faculty of Engineering Science, Osaka University, Toyonaka, Osaka 560, Japan Populations of the ciliate protozoan Paramecium exhibit a clear circadian rhythm in traverse frequency, TF [cells/h, hourly number of cells interrupting an infrared (IR) test light in an experimental vessel] (Hasegawa et al. 1984; Hasegawa and Tanakadate 1984), which sustains for several cycles in constant darkness (DD). Statistical analyses of swimming velocities and frequencies of changing directions of cells interrupting the IR light predicted that individual cells tend to swim fast and straight during the day (and subjective day) and slowly with frequent direction changes at night (and subjective night), and that these characteristic patterns of individual cells can be attributed to the circadian accumulation and dispersal rhythm of the population. The present study ®rst shows tracking of isolated single cells of P. multimicronucleatum in LD and DD, using an automated system for tracking isolated single microorganisms (``bug-tracker'') (Hasegawa et al. 1988), to con®rm the circadian pattern of individual cells as predicted by the statistical analyses of the population rhythm. 42 Paramecium behavior is described by two factors: the velocity of forward swimming and the frequency of changing direction (Diehn 1979). Electrophysiological studies have shown that these factors are closely related to the membrane potential (Naitoh and Eckert 1969; Hennessey et al. 1985; Saimi and Kung 1987). When the membrane is hyperpolarized, the beating frequency of cilia covering the entire cell body increases, thus cells swim faster forwards. Membrane depolarization causes opposite e€ects: small depolarization leads a slowing of forward swimming, but larger depolarization causes a reverse of ciliary beat and therefore backward swimming of the cell (reviews: Dryl 1974; Saimi and Kung 1987; Hinrichsen and Schultz 1988; Preston 1990). The circadian pattern suggests that the Paramecium resting membrane potential ¯uctuates in a daily and circadian manner. The present study con®rms this suggested ¯uctuation of the resting membrane potential in LD, demonstrating that it is more hyperpolarized during the day than at night. Hyperpolarization of the Paramecium membrane by manipulating the ionic composition of the medium is correlated with an increase in intracellular cAMP concentration (Bonini et al. 1986; Schultz et al. 1987, 1992); and depolarization with an increase in cGMP (Schultz et al. 1986; Schultz and Schade 1989). In addition, cAMP and cGMP are known to act antagonistically for regulating ciliary beating of Paramecium cell model permeabilized with a detergent: cAMP being associated with the orientation of ciliary power stroke to a 7 o'clock position (with 0 o'clock position parallel to the anterior direction of the cell body), causing right-handed helical movement of the cell model; and cGMP equivalent to a 5 o'clock position, causing left-handed helical moving (Bonini and Nelson 1988, 1990). In intact Paramecium cells, the extracellular addition of cAMP increases beat frequency and modi®es beat orientation from the basal 4 or 5 o'clock position toward 6 o'clock, encouraging cells` fast forward swimming, and that of cGMP favor to maintain the ciliary beating position at the basal position (Nakaoka and Ooi 1985; Bonini et al. 1986, 1991). The observed pattern of tracking and ¯uctuation of resting membrane potential therefore suggest daily and circadian ¯uctuations of cellular cAMP and/or cGMP concentrations. The present study con®rms this, which shows ¯uctuations of cellular concentrations of cAMP and cGMP, such that both are relatively high during the day (and subjective day), and low at night (and subjective night). Interestingly the ratio of cGMP to cAMP in LD ¯uctuates to parallel the ¯uctuation of the resting membrane potential measured in LD. Involvement of cyclic nucleotides in the circadian clock mechanism has been demonstrated in many multicellular animals (Eskin and Takahashi 1983; Prosser and Gillette 1989; Zats 1992); the key issue to be addressed concerns the roles of second messengers in conveying photic and/or circadian signals to the nucleus to induce mRNAs in the circadian pacemaker cell (Kornhauser et al. 1990; Ginty et al. 1993; Takahashi et al. 1993). The Paramecium cells used in the present study exclude the possible e€ects of intercellular communications (Hasegawa et al. 1995) which might agitate second messengers. The observed circadian synchrony among the motility, resting membrane potential and cyclic nucleotides argue that P. multimicronucleatum will be a principle cell model for understanding the circadian orchestration of second messengers conducted by intracellular pacemaker and/or direct photic stimulation, contributing to circadian changes in the behavior as well as the cell membrane potential. Materials and methods Cells and culture Stock culture of P. multimicronucleatum, acycler, syngen 2 (Barnet 1966), were grown in hay infusion inoculated with Klebsiella pneumoniae under a cycle of 12 h light and 12 h dark (LD, 12:12; L = 1000 lx from a cool-white ¯uorescent lamp) at 20 °C. Measurement of circadian tracking changes in a single Paramecium cell A cell was isolated from stock culture in a late stationary phase, and transferred to a chamber of restricted height (i.e., virtually a two-dimensional cavity), which was framed by a doughnut-shaped acrylate plate (diameter 58 mm; hole diameter 15 mm; thickness 2 mm) between two ¯at-bottomed laboratory dishes; one 30 mm diameter inside the other 60 mm diameter. The cavity was previously ®lled with cell-free culture medium, prepared by syringe-®ltering about 100 ml of stock cell suspension. The vessel was placed at the center of a container thermo-regulated by circulating water of 20 °C. The x-y coordinates of the cell freely swimming in the cavity were measured using an automatic tracking system (``bugtracker''; Hasegawa et al. 1988) for 11 min at 2-h intervals in LD (12:12, L = about 1000 lx at the surface of the chamber) followed by DD. From the total length of tracking for each 11 min calculated using the x-y coordinates sampled at intervals of 0.25 s, swimming velocity for the scheduled time was obtained. Preliminary results in P. multimicronucleatum have been previously reported elsewhere (Hasegawa et al. 1990), which demonstrate almost the same characteristics as shown in the present study. There were observed a great variety in amplitudes of circadian tracking changes of individual cells showing clearly circadian motility. Since simple averages of their velocities make circadian characteristics obscure, relative swimming velocities were calculated with the maximum velocity of each cell normalized to 1, and averaged to characterize circadian changes in the swimming velocity of the cells. Measurement of circadian changes in resting membrane potential Samples of cells were taken from stock in stationary state at 6-h intervals for seven cycles in LD. Measurements were made in the culture medium with the temperature controlled at 20 °C. Methods for intracellular recording of resting membrane potential were similar to those described previously (Matsuoka and Nakaoka 1988). Electrodes were ®lled with 0.1 mol á l)1 KCl and had resistance of 100±150 MW. The cells were placed in a glass vessel mounted on an inverted microscope and electrodes were inserted from above. External solution was the cell-free culture medium. The resting membrane potential was estimated from the potential di€erence before and after penetrating the electrodes. It took about 20 min to obtain a stable resting membrane potential from a cell, which allowed us to obtain one or two data at each determined time of day. The data were repeatedly measured at predetermined times for 7 days in LD and averaged.