Cytoplasmic Streaming in Chara Corallina Studied by Laser Light Scattering

J. CM Sci. aa, 633-643 (1976) 633 Printed in Great Britain

CYTOPLASMIC STREAMING IN CHARA CORALLINA STUDIED BY LASER LIGHT SCATTERING

D. B. SATTELLE* AND P. B. BUCHAN A.R.C. Unit of Invertebrate Chemistry and Physiology, Department of Zoology, Dotvning Street, Cambridge, England

SUMMARY An apparatus is described by means of which the power versus frequency spectrum of the photomultiplier current can be obtained for laser light scattered by streaming cytoplasm in the algal cell Chora corallina. A Doppler peak is noted in the spectrum which is abolished when cytoplasmic streaming is arrested by electrical stimulation. For 5 cells of Cliara, this simple laser-Doppler velocimeter gave streaming velocities (467 /im s"1, S.D. ±4-8 at 20 °C) similar to those obtained for the same cells using the light microscope (443 /4m s"1, S.D. ±53 at 20 °C). A narrow distribution of streaming velocities is indicated. The technique described provides a rapid, quantitative assay of the in vivo rheological properties of cytoplasm.

INTRODUCTION Quasi-elastic laser light scattering has provided an accurate and rapid method for determining the velocities of flowing fluids (cf. Yeh & Cummins, 1964; Trolinger, 1972; Chu, 1974; Crosignani & Di Porto, 1974). Light scattered by particles under- going directed motion is Doppler shifted and the size of the frequency shift provides a direct measure of the velocity. The potential of 'laser Doppler velocimetry' for assessing flow properties in biological systems has been demonstrated for cytoplasmic streaming in the alga Nitella (Mustacich & Ware, 1974, 1976; Langley, Piddington, Ross & Sattelle, 1976; Sattelle, Langford& Langley, 1977) and the slime mould Physarum (Newton, Ford, Langley & Sattelle, 1977). The chemotactic swimming behaviour of sperm (Dubois, Jouannet, Berge" & David, 1974) and motile bacteria (Nossal & Chen, 1972, 1973) have also been followed. The technique has in addition enabled measurement of in vivo rheological properties of blood in arteries (Tanaka, Riva & Ben-Sira, 1974) veins (Tanaka & Benedek, 1975) and small vessels in the peripheral circulation (Stern, 1975). For signal processing of the scattered light detected by a photomultiplier, these studies have utilized such instruments as the spectrum analyser and the photocount autocorrelator in order to extract either the power spectrum of the scattered light or its inverse Fourier transform, the autocorrelation function. We describe a simple signal-processing system for spectrum analysis of scattered laser light, based on the • Currently Wellcome Travel Fellow of the Medical Research Council of the United Kingdom. 634 D. B. Sattelle and P. B. Buchan use of 2 phase detectors in quadrature. This technique is applied in a study of cytoplasmic streaming in the alga Chara corallina.

THEORY The optical spectrum of laser light is a single delta function at the laser frequency (w0). If a scattering particle moves across the incident wavefronts at a constant velocity and in a constant direction, light scattered by the flowing particle will be Doppler- shifted to a frequency other than w0 (Fig. i). When laser light scattered from a flowing fluid is mixed with unshifted light on a photomultiplier the beat frequency gives the

4 i i

Fig. i. A. Parallel wavefronts from an incident (laser) light source interact with the scattering centre (sc). When sc is stationary, light is scattered at the incident frequency. When sc is moving along XY, light is scattered at a higher frequency since sc 'sees' more wavefronts per unit time. B. Optical spectrum [intensity (I(w)) versus frequency ((o)] for light scattered by stationary scattering centres. Specrum is a delta function at the optical (laser) frequency w0. c. Optical spectrum [I(w) versus w] for light scattered by a mixture of stationary and flowing scattering centres. Flow is at uniform velocity along XY. Spectrum contains a component at the laser frequency w0 and a flow component which can be attributed to light Doppler-shifted to a frequency

Doppler shift from which the flow velocity can be determined. Under these conditions, the photocurrent power spectrum reproduces the optical spectrum with the centre frequency shifted from w0 to zero frequency. Photocurrent power spectra for several kinds of motion are illustrated in Fig. 2 (for comprehensive accounts of theoretical considerations see Trolinger, 1972; Chu, 1974; Crosignani & Di Porto, 1974). For the scattering geometry employed in the experiments described in this report (see Fig. 3) the velocity of a flowing particle (v) is given by:

v = —J»- (i) q cos a (where/D is the frequency at which the Doppler peak appears in the photocurrent Laser-Doppler study of cytoplasmic streaming 635 t S(,o) Stationary

Flow at uniform velocity

Flow with distribution of velocities

Brownian motion

Fig. 2. Photocurrent spectra [power (S(fii)) versus frequency (w)] derived from light scattered by various kinds of motion of scattering centres. Following square-law detection the optical spectrum is reproduced with o>0 shifted to zero frequency. power spectrum, q = the scattering vector and a = the angle between the direction of flow and q: equation (1) is from Yeh & Cummins (1974). Since 47m . 0 q = :Lr- sin- (2) K 2 (where n = refractive index of the medium; AQ the wavelength of the incident laser light; 0 = the scattering angle), by substituting in equation (2) we obtain,

v =- (3) zn sin - cos a 2 It is clear from equation (3) that the flowvelocit y can be determined from a knowledge of the scattering geometry of the experiment and the measurement of/D. If the flowing scatterers exhibit a range of velocities, the spectral peak centred at/D is broadened, the width relating to the distribution of velocities - a narrow peak corresponding to a narrow range of velocities. 636 D. B. Sattelle and P. B. Buchan

Fig. 3. Geometry of the laser light-scattering experiment. The directions of the incident light, the scattered light, the streaming cytoplasm and the scattering vector (5) are shown. The scattering angle (6) and the angle between the scattering vector and the direction of streaming (a) are shown. The scattering vector bisects the angle sup- plementary to 6.

MATERIALS AND METHODS Living material Internodal cells of Chara corallina (about 5 cm long) were maintained in a physiological medium containing (mM/1.): NaCl, 10; CaCli, o-i; KC1, ci. This solution was filtered using a Millipore filter (0-45 fim pore diameter). Cells in which cytoplasmic streaming could be clearly seen with a x 50 stereomicroscope were cleaned with a camel-hair brush and mounted hori- zontally on short, rigid platinum stimulating electrodes which were in turn held firmly in a micromanipulator (Leitz). Experiments were performed at 20 °C.

Light-scattering apparatus The apparatus illustrated in Fig. 4 was mounted on a stout wooden table. A 15-cm-thick base-plate (1 m x 1 m) was shock-mounted on a tyre inner tube. A smaller (o-s-cm-thick) steel plate was supported on the base-plate by rubber stoppers and carried the turntable-mounted, light-scattering arm and a T-shaped table which supported the laser and the glass experimental chamber. The turntable used was the stage of a milling machine. A 5-mW He-Ne laser (Spectra Physics 120) served as an incident light source and was provided with an adjustable mount to facilitate alignment. The plant cell was positioned in the chamber over the centre point of the turntable. The light-scattering arm was constructed to enable ready adjustment to a variety of scattering angles. Although the experiments reported here were carried out at a single scattering angle the angular dependence of the spectral width can be useful in the characterization of diffusive and directed motions (Langley et al. 1976). The main features of the light-scattering arm are shown in Fig. 5. The laser light scattered from the preparation entered the arm through an iris aperture and was focused in the plane of a pinhole which was situated immediately in front of the face of a photomultiplier tube (Twentieth Century Electronics Type P4283 TIRA). The pinhole was mounted on a vertical, Laser-Doppler study of cytoplasmic streaming 637 adjustable X-Y stage. To achieve a single coherence area (iris aperture area) the radii of the aperture and the pinhole are related by 1-22 Aoz P = T—, (4) where a = aperture radius; p = pinhole radius; A,, = wavelength of incident laser light; z — distance between lens and pinhole (see Jakeman, 1974). From equation (4), a range of pinhole diameters appropriate to a particular aperture diameter were computed. In the experiments reported here an aperture diameter of 08 mm and a pinhole diameter of 2'0 mm were employed.

Fig. 4. Diagram of the light-scattering apparatus. A tyre inner tube (T) provides shock- mounting for the apparatus which is mounted on a steel base-plate (BP). The turntable- mounted, scattering arm carries the photomultiplier (PMT) and the collection optics. The aperture (A) and the X-Y pinhole adjustment screws (P) are shown. The laser light intensity is attenuated by a neutral density filter (NDF). The glass experimental chamber (C) containing the cell is shown.

Signal processing A block diagram of the main units in the circuitry employed in signal processing is illustrated in Fig. 6. The photomultiplier output was fed into a precision a.c. amplifier (Brookdeal 5002). The output impedance from the photomultiplier was 2-0 Mft and the input impedance of the a.c. amplifier was 100 Mfi shunted by 20 pF. This combination of impedances including stray capacitance did not significantly attenuate the signal over the range 1 to 500 Hz. Following amplification, the signal was fed directly into 2 phase-sensitive detectors (Brookdeal 9421A) whose inputs were connected in parallel. A square wave reference signal was fed into one detector and that signal in quadrature (+ 900) was fed into the second detector. The reference signal was taken from a signal source (Brookdeal 9742) which was operated in the external 638 D. B. Sattelk and P. B. Buchan voltage controlled frequency mode. The frequency of the reference signal was in turn controlled by a ramp waveform generated by a ramp generator (Prosser A341). In most experiments the reference signal was swept through a range of either 1 to 110 Hz or 2 to 220 Hz. The time taken to sweep through the spectrum was about 20 s.

Microscope with vertical adjustment Lens Pinhole adjustment

Cell Lens rMirror PMT

Fig. 5. Schematic sectional view of the light-scattering arm of the apparatus. A cell, under saline, is shown in the experimental chamber (C). Light scattered at a given angle is collected via an aperture (diameter = 2a), and focused by the lens on to the pinhole (diameter = zp) situated immediately in front of the photomultiplier (PMT). When the modified microscope is lowered into the position shown, the mirror deflects the light upwards enabling the investigator to examine the illuminated cell.

Phase- fi +90< detector

PMT Amplifier Signal Ramp source generator

Phase- IScattered r sensitive [light \~ detector

LASER 1 /, ident Streaming *>ht cytoplasm X

Fig. 6. Block diagram of the major components in the signal processing system. The ramp generator drives the X-axis and the vector unit (^/(a' + i1)) drives the Y-axis of a recorder (oscilloscope in this case). PMT = Photomultiplier.

The amplitude of the first phase detector output voltage was proportional to the amplitude of the photomultiplier signal at the frequency being swept at any given moment. The output of the second phase detector was the same but in quadrature with the first. The component values of the signal at any moment were therefore present in rectangular form (i.e. a +jb). The final unit in the system (Brookdeal ^(a' + i*)) changes the information from rectangular into polar form (i.e. (a' + b1)^ A0). This signal was then fed into the Y input of a Tektronix 565 storage oscilloscope whose timebase (X input) was driven directly from the ramp generator - Laser-Doppler study of cytoplasmic streaming 639 the same generator that controls the reference signal. The resulting display was a frequency spectrum of the photomultiplier current [»'(w)] which was squared to give the power spectrum.

Experimental procedure and calibration The mounted internodal cell was positioned in the attenuated beam and moved along the beam until the image of the illuminated region appeared centrally located in the microscope of the scattering arm. Final positioning was achieved using the X-Y adjustments of the pinhole to obtain maximum intensity. In some experiments an audio-amplifier was connected to the output of the precision a.c. amplifier and fed into a loudspeaker, thereby facilitating alignment procedures. Throughout all experiments the mounted cell could be observed using a stereomicroscope ( x 50 magnification) mounted above the preparation chamber. It was therefore possible to ascertain whether or not a cell was streaming, and if so to estimate the velocity of the larger cytoplasmic particles.

RESULTS Spectrum from streaming cell The power spectrum [S(w)] was derived first of all from light scattered by the platinum electrodes which served as a mount for the cell. The power spectrum was essentially flat over the range 2-200 Hz (Fig. 7 A). A small amount of curvature was detected in the frequency range 2-10 Hz which may have been the result of scatterers present in the physiological medium. Light was next scattered from a cell observed to be streaming. The cell was positioned so that none of the scattered light originated from the electrodes. No specific local oscillator was introduced in such experiments

6r 6r

? 4

™ 3

100 200 100 200

Fig. 7. Photocurrent power spectra [S(w) versus w] derived from laser light scattered by: A, platinum electrodes serving as a mount for the cell; B, cell of Chara corallina with cytoplasm streaming; and c, cell of Chara corallina during temporary arrest of cytoplasmic streaming induced by electrical stimulation. 640 D. B. Sattelle and P. B. Buchan and scattering from the plant cell wall provided unshifted light. The spectrum of light scattered by a streaming cell contained a substantial 'very low frequency' (VLF) component and in addition a clearly defined peak (Fig. 7B). With the cell aligned so that streaming was in the same plane as the scattering vector and with 6 = 6o°; a. = 250 (see Fig. 3, p. 636), this peak was centred at 80 Hz.

Spectrum from non-streaming cell The spectrum of light scattered by the same cell, in the same orientation but this time with streaming arrested by the application of a single depolarizing pulse (5 V, 2 ms duration, square wave) is shown in Fig. 7 c. The VLF component was still present when streaming ceased but the peak at 80 Hz noted in the spectrum during streaming was no longer detectable. From these observations it seems reasonable to conclude that this peak in the spectrum is the result of light that has undergone a frequency (Doppler) shift as a result of being scattered by particles in the streaming cytoplasm.

Streaming velocities from laser light scattering and light microscopy Using data shown in Fig. 7 from a single cell it is possible using equation (3) to 5 calculate the streaming velocity. In this case with/D = 80 Hz, Ao = 6-328 x io~ cm; 6 = 6o°, a = 25°, and n taken to be 1-34, a streaming velocity of 42-0 ps"1 was calculated. For 5 cells of Chara corallina streaming velocities at 20 CC were obtained using both laser light scattering and by timing the movements of individual cytoplasmic particles observed in the stereomicroscope over known distances. An average velocity of 46-7 (s.D. ± 4-8) /tm s"1 was calculated from laser light-scattering data, and an average velocity of 44-3 (s.D. ±5-3) //m s~x was obtained by direct observation of streaming particles. Unless the choice of the value for the refractive index in the analysis of the light-scattering data is seriously in error, it therefore appears that the 2 methods give comparable results when applied to the same material.

DISCUSSION A simple apparatus and signal processing system has been devised to enable measurements of spectral changes in cells under different physiological conditions. Novel features of the apparatus include a microscope in the scattering arm which readily enables observation, focusing and alignment of the preparation. In addition a turntable is included which, whilst providing rigid support for the scattering arm, also enables rapid change to a new scattering angle. The signal-processing system which employs 2 phase-sensitive detectors in quadrature to recover the frequency spectrum from the amplified anode current of the photomultiplier provides sufficient resolution to detect spectral differences between streaming and non-streaming states of proto- plasm in the algal cell Chara corallina. The signal-processing system described here is slower to generate the spectrum than a conventional spectrum analyser but its construction in separate modules provides for other uses of the equipment (e.g. signal averaging). Laser-Doppler study of cytoplasmic streaming 641 Using laser-Doppler velocimetry, cytoplasmic streaming in Chara has been detected as a distinct peak in the power spectrum derived from the scattered laser light. This peak is consistently abolished when streaming is arrested by electrical stimulation of the cell. The velocities obtained in this way are comparable with those obtained by direct observations on streaming cytoplasmic particles from the same cells. Streaming speeds reported here for Chara corallina [~ 45 fim s"1 at 20 °C] are similar to those obtained for Charafoetida[\2/im s~1at20°C](Velten, 1876)and Chara braunii [75 /tm s"1] (Hayashi, 1952) using light-microscopical techniques. Velocities of the same order of magnitude have been obtained from cells of Nitella opaca [~ 60 fim s"1] (Langley et al. 1976) and Nitella flexilis [~ 50 fim s^1 at 20 °C] (Mustacich & Ware, 1976). These latter authors have carefully investigated the temperature dependence of streaming and have shown that over the temperature range 4-35 °C, a 1 °C increase in temperature corresponds to a 2-6 fim s~x increase in streaming velocity. The spectral peak attributed to cytoplasmic streaming in Chara is comparable with similarly narrow Doppler peaks associated with cytoplasmic flow in Nitella flexilis (Mustacich & Ware, 1974,1976) and Nitella opaca (Langley et al. 1976). These findings are interpreted in terms of a narrow distribution of velocities for the streaming particles (cf. Fig. 2). Thus a plug-like flow of cytoplasm appears to be a common feature of streaming in Chara and Nitella. It is tempting to postulate that similar rheological properties noted in various algal cells may share common mechanisms of motive force generation. In this context it is of interest to note that microfilaments from both Chara corallina (Williamson, 1974) and Nitella flexilis (Palevitz, Ash & Helper, 1974) react with subfragments of muscle myosin to produce ' arrowhead' filaments indicating that these microfilaments in algal cells are very similar to actin. In addition, it has recently been demonstrated that all actin filaments located in the non-streaming (ectoplasmic) outer layer of cytoplasm have the same polarity (Kersey, 1974). These findings provide an ultrastructural basis for the 'active shearing' mechanism proposed much earlier (for a review see Kamiya, 1959) in which it is envisaged that the interaction of sliding filaments at the interface between the streaming cytoplasmic layer (endoplasm) and the non-streaming layer (ectoplasm) generates the motive force for cytoplasmic streaming. The localization of actin in characean cells does not, however, rule out an alternative view for the generation of motive force which proposes the existence of micro- filament bundles anchored in the ectoplasm but extending into the endoplasm where beating movements generate the flow of cytoplasm (Allen & Allen, 1972; Allen, 1974). Spectral components other than the Doppler peak attributed to cytoplasmic streaming can also be distinguished in the power frequency spectrum derived from light scattered by a Chara cell. The prominent very low frequency (VLF) component is present in both streaming and non-streaming cells. It is suggested that this component can be attributed in part to slow-moving particles located either in the periphery of the flowing endoplasm or in the non-flowing ectoplasm. No significant high-frequency component was detected in these experiments, indicating the absence of particles moving at substantially greater velocities than those reported for cytoplasmic streaming. The recent advances in knowledge of the biochemistry of the machinery of motility in non-muscle cells (Pollard & Weihing, 1974; Pollard, 1975; Goldman, Pollard 42 C K L 22 642 D. B. Sattelle and P. B. Buchan & Rosenbaum, 1976) only emphasizes the need for complementary biophysical studies. Laser light scattering provides an objective, rapid, quantitative assay for cytoplasmic streaming. Further laser light-scattering studies of streaming under various physiological conditions and with the streaming in a variety of orientations may lead to a better understanding of the mechanism of this most readily observable of all intracellular motions.

The authors express particular thanks to Professor K. H. Langley (Department of Physics and Astronomy, University of Massachusetts) for helpful discussions during the course of this work. We thank Mr J. W. Rodford for preparing Fig. 3.

REFERENCES ALLEN, N. S. (1974). Endoplasmic filaments generate the motive force for streaming in Nitella. J. Cell Biol. 63, 270-287. ALLEN, N. S. & ALLEN, R. D. (1972). Endoplasmic filaments and rotational streaming in Nitella. J. Cell Biol. 55, 2a. CHU, B. (1974). Laser Light Scattering. New York and London: Academic Press. CROSIGNANI, B. & Di PORTO, P. (1974). Investigation of laminar and turbulent flows by means of light scattering. In Photon Correlation and Light-beating Spectroscopy (ed. H. Z. Cummins & E. R. Pike), pp. 457-473. New York: Plenum Press. DUBOIS, M., JOUANNET, P., BERGE, P. & DAVID, G. (1974). Study of spermatozoa motility in human cervical mucus by inelastic light scattering. Nature, Land. 252, 711-713. GOLDMAN, R. D., POLLARD, T. D. & ROSENBAUM, J. L. (1976). (eds.) Cold Spring Harbor Conferences on Cell Proliferation, vol. 3. Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory (in Press). HAYASHI, T. (1952). Some aspects of behaviour of the protoplasmic streaming in plant cells. Bot. Mag., Tokyo 65, 51-55. JAKEMAN, E. (1974). Photon correlation. In Photon Correlation and Light-beating Spectroscopy (ed. H. Z. Cummins & E. R. Pike), pp. 75-149. New York: Plenum Press. KAMIYA, N. (1959). Protoplasmic streaming. Protoplasmatologia 8, 3a, 1-199. KERSEY, Y. M. (1974). Correlation of polarity of actin filaments with protoplasmic streaming in characean algae. J. Cell Biol. 63, 165 a. LANGLEY, K. H., PIDDINGTON, R. W., ROSS, D. & SATTELLE, D. B. (1976). Photon correlation analysis of cytoplasmic streaming. Biochim. biophys. Acta (in Press). MUSTACICH, R. V. & WARE, B. R. (1974). Observation of protoplasmic streaming by laser-light scattering. Phys. Rev. Lett. 33, 617-620. MUSTACICH, R. V. & WARE, B. R. (1976). A study of protoplasmic streaming in Nitella by laser Doppler spectroscopy. Biophys. J. (in Press). NEWTON, S. A., FORD, N. C. JR., LANGLEY, K. H. & SATTELLE, D. B. (1977). Laser light scattering analysis of protoplasmic streaming in the slime mold Physarum polycephalum. (in preparation). NOSSAL, R. & CHEN, S. H. (1972). Laser measurements of chemotactic response of bacteria. Optics Comm. 5, 117-122. NOSSAL, R. & CHEN, S. H. (1973). Effects of chemoattractants on the motility of Eschtrichia coli. Nature, New Biol. 244, 253-254. PALEVITZ, B. A., ASH, J. F. & HEPLER, P. K. (1974). Actin in the green alga Nitella. Proc. natn. Acad. Sd. U.S.A. 71, 363-366. POLLARD, T. D. (1975). Functional implications of the biochemical and structural properties of cytoplasmic contractile proteins. In Molecules and Cell Movement (ed. S. Inou6 & R. E. Stephen), pp. 259-286. New York: Raven Press. POLLARD, T. D. & WEIHING, R. R. (1974). Actin and myosin and cell movement. CRC Crit. Rev. Biochem. 3, 1. SATTELLE, D. B., LANGFORD, G. M. & LANGLEY, K. H. (1977). The study of intracellular particle motion. In Photon Correlation Spectroscopy and Laser Doppler Velocimetry (ed. E. R. Pike & H. E. Cummins). New York: Plenum Press (in Press). Laser-Doppler study of cytoplasmic streaming 643 STERN, M. D. (1975). In vivo evaluation of microcirculation by coherent light scattering. Nature, Lond. 254, 56-58. TANAKA, T. & BENEDEK, G. B. (1975). Measurement of the velocity of blood flow (in vivo) using a fiber optic catheter and optical mixing spectroscopy. Appl. Optics 14, 189-196. TANAKA, T., RIVA, C. & BEN-SIRA, I. (1974). Blood velocity measurements in human retinal vessels. Science, N.Y. 186, 830-831. TROLINGER, J. D. (1972). Laser Instrumentation for Flow Field Diagnostics (ed. S. M. Bodgonoff & J. A. Smith). NATO Advisory Group for Aerospace Research and Development, AGARDograph No. 186. VELTEN, W. (1876). Die Einwirkung der Temperatur auf die Protoplasmabewegung. Flora, Jena 59, 177-182. WILLIAMSON, R. E. (1975). Cytoplasmic streaming in Chara: a cell model activated by ATP and inhibited by cytochalasin B. J. Cell Set. 17, 655-668. YEH, Y. & CUMMINS, H. Z. (1964). Localized fluid flow measurements with an He-Ne laser spectrometer. Appl. Phys. Lett. 4, 176-178. (Received 18 May 1976)

42-2