Phototaxis and Membrane Potential in the Photosynthetic Bacterium Rhodospirillum Rubrum

Phototaxis and Membrane Potential in the Photosynthetic Bacterium Rhodospirillum rubrum

SHIGEAKI HARAYAMA* AND TETSUO IINO Laboratory of Genetics, Faculty of Science, University of Tokyo, Hongo, Tokyo 113, Japan Received for publication 23 February 1977

Cells of the photosynthetic bacterium Rhodospirillum rubrum cultivated anaerobically in light show phototaxis. The behavior of individual cells in response to the phenomenon is reversal(s) of the swimming direction when the intensity of the light available to them abruptly decreases. The tactic response was inhibited by antimycin, an inhibitor of the photosynthetic electron transfer system. The inhibitory effect of antimycin was overcome by phenazine methosulfate. Motility of the cells was not impaired by antimycin under aerobic conditions. Valinomycin plus potassium also inhibited their phototactic response; however, valinomycin or potassium alone had no effect. A change in membrane potential of the cells was measured as an absorbance change of carotenoid. Changes in the membrane potential caused by "on-off' light were prevented by antimycin and by valinomycin plus potassium, but not by antimycin plus phenazine methosulfate nor valinomycin or potassium alone. The results indicated that the phototactic response of R. rubrum is mediated by a sudden change in electron flow in the photosynthetic electron transfer system, and that the membrane potential plays an important role in manifestation of the response. Bacterial phototaxis has been observed in systems. The chemotactic behavior of individ- photosynthetic bacteria (6, 22) and in Halobac- ual cells ofE. coli and S. typhimurium has also terium halobium (12). The photoreceptor pig- been examined (1-3, 16, 26, 30). The bacteria ments responsible for positive phototaxis (pho- modulate the frequency of change in direction totaxis towards light) have been identified as when they swim up or down a spatial gradient bacteriochlorophyll and carotenoids in photo- of a chemotactic stimulus; thus, their attraction synthetic bacteria (5, 19) and as bacteriorho- to or migration from it is attained. This modu- dopsin in H. halobium (12). The phototactic lation is also observed when they are exposed to behavior of individual cells of photosynthetic a temporal change in concentration of the stim- bacteria has been extensively examined (10): ulating chemicals. the bacteria show abrupt changes in direction Whereas cellular receptors of the tactic stim- when they swim from a light to a dark field, uli and swimming behaviors induced by the and some of them return to the light field; stimuli have been well documented in bacterial however, they fail to show such a response phototaxis, as well as in bacterial chemotaxis, when they swim from a dark to a light field. As how the stimuli are converted to a tactic signal a consequence of these responses, the number that modulates the frequency of change in their of organisms in a light spot projected onto a swimming direction is unclear. suspension of bacteria increases with time. The In this paper we show evidence that reversals change in swimming direction can be induced of the swimming direction of the photosynthetic by temporal decrease of an actinic light inten- bacterium Rhodospirillum rubrum induced by sity impinging on them. This observation sug- a sudden decrease in the actinic light intensity gested that the phototactic response is the re- are mediated by a sudden decrease of electron sult of sensing temporal, not spatial, differ- flow in the photosynthetic electron transfer sys- ences in light intensity. tem and that membrane potential plays an im- In chemotaxis ofEscherichia coli and of Sal- portant role in the response. monella typhimurium, chemoreceptor mole- cules of several attractants (chemicals that in- MATERIALS AND METHODS duce positive chemotaxis) have been identified Bacterial strains and growth of cells. Wild-type (1, 11, 18). The sugar chemoreceptors are pro- R. rubrum (S-1) was obtained from G. Soe, and the teins that are constituents of active transport carotenoid-less blue-green mutant (G-9) was from Y. 34 VOL. 131, 1977 PHOTOTAXIS IN R. RUBRUM 35 Kobayashi, to whom we are indebted. The bacteria were cultivated in light at 28°C in a 120-ml screw- capped bottle containing a synthetic medium whose composition was as described previously (8), except Li that Casamino Acids was omitted from the medium. Illumination was provided by one 300-W flood lamp (Toshiba) placed 80 cm from the culture bottle. The II bacteria at the late exponential to stationary growth phase were harvested, centrifuged, washed at least IF c .' two times with 10 mM tris(hydroxymethyl)amino- methane (Tris)-hydrochloride buffer (pH 7.6), sus- * I Ch pended in the appropriate solution, and then used for experiments. In the experiments reported in Ta- ble 3, the bacteria were washed in 10 mM sodium phosphate buffer (pH 6.8). Measurement of phototactic response. To examine quantitatively the phototactic response of the S bacteria, an apparatus for giving successive photo- H MI tactic stimuli to the bacteria was devised (Fig. 1). Two light beams, one for an actinic light and another for an observing light, were projected onto a I 12 L2 phase-contrast microscope condenser through a half IHo mirror. The actinic light beam from a tungsten- halogen lamp (Ushio, JC-24V-150W) was passed RD through a 4-cm cuvette containing water, chopped by a rotary sector for 0.5 s every 2 s, and then projected on a bacterial suspension on a glass slide c7 through a phase-contrast condenser. The intensity of the actinic light, measured with a YSI-Kettering model 65 radiometer, was 500 W/m2. The observing light beam from a 30-W tungsten lamp (Nikon) was filtered by a thermo-cut-off filter (Hoya, HA-50) and a green interference filter (Toshiba, KL-56; peak transmission at 558 nm, with a halfband width of 14 nm) and then continuously projected on the bacterial suspension. The intensity of the observing light FIG. 1. Apparatus for observation of bacterial was 3 W/m2 or below. phototactic response. Abbreviations: Li, tungsten The tactic response of R. rubrum induced by a lamp (Nikon, 30 W); L2, tungsten-halogen lamp sudden decrease in actinic light intensity is simple: (Ushio, JC-24V-150W); IF, interference filter (To- the bacteria reverse their swimming direction after shiba, KL-56) plus thermo-cut-off filter (Hoya, HA- being stimulated (6, 10). Therefore, as an index of 50); 11 and 12, collector lens; H20, 4 cm of water; their phototactic activity, we measured the fraction HM, half-reflecting mirror; Ch, rotary chopper cut- of bacteria that showed reversal(s) of swimming ting actinic light for 0.5 s every 2 s; RD, phase ring direction within 0.5 s after the actinic light was diaphragms; C, condenser lens; S, specimen; and 0, turned off in repeated "on" (1.5 s) and "off" (0.5 s) 40 x phase objective (Nikon, long-working distance). circles, by direct microscopic observations made at an ambient temperature of300C. When more than 50 bacteria were counted, the standard deviation ofthe the measuring beam of the spectrophotometer. The measured values from the same sample did not ex- intensity of the actinic light was 1.3 kW/m2. Be- ceed 20% of the mean value. tween the cuvette and a photomultiplier tube, two Measurement of an absorbance change in bacte- filters (Corning 9782 and HA-50) were placed to rial suspension. Measurement of light-induced ab- protect the photomultiplier tube from the cross-illu- sorbance (A) change was carried out with a Hitachi minating actinic light. Intermittent illumination of 356 double-beam spectrophotometer. The wave- the actinic light (1.5 s on and 0.5 s off) was attained length used for measuring the absorbance change of by insertion or removal of a sector connected to a carotenoid in wild-type cells (S-1) was 525 nm, with rotary solenoid operated by an autotimer. 510 nm as a reference. Those used for measuring Estimation of cellular content of bacteriochloro- that of cytochrome c2 and reaction center bacterio- phyll. Bacteriochlorophyll was extracted with ace- chlorophyll in cells ofthe carotenoid-less mutant (G- tone-methanol (7:2, vol/vol), and then the cellular 9) were 552 and 600 nm, with 540 and 585 nm, content of the pigment was determined spectropho- respectively, as references. Actinic light from a tometrically by using the absorption coefficient tungsten-halogen lamp (Ushio, JC-24V-150W) fil- given by Clayton (5). tered through an R-72 filter (Hoya), allowing trans- Chemicals. Valinomycin and antimycin were mission of light only above 700 nm, was projected purchased from Boehringer Manheim. Other re- onto a cuvette (light path, 1 cm) at a right angle to agents used were of analytical grade. 36 HARAYAMA AND IINO J. BACTERIOL. RESULTS its oxidation (decrease in A552-540) by light was Effect of antimycin and PMS on phototac- inhibited, and its reduction after the light was tic response and photosynthetic electron turned off was prevented by the inhibitor. Re- transfer in the bacteria. Cells of R. rubrum dox changes of reaction center bacteriochloro- reverse their swimming direction when the ac- phyll, measured as A605585, were hardly af- tinic light intensity impinging onto them is fected by the inhibitor. Changes in carotenoid abruptly decreased. Since the phototactic phe- absorbance, which reflects changes in mem- nomenon, i.e., accumulation ofthe bacteria in a brane potential (14, 15), were affected by the light spot projected onto a bacterial suspension, inhibitor: an increase in membrane potential is a result of this response (6, 10), we estimated (hyperpolarization) was observed after the light the phototactic activity ofthe bacteria by count- was first turned on, but a decrease in mem- ing the fraction that reversed swimming direc- brane potential (depolarization) after the light tion in a 0.5-s dark period after the actinic light was turned off was prevented by the inhibitor. was turned off. The value of the fraction de- That the observed absorbance change at 525 to pended on decreased intensity of the actinic 510 nm corresponded to that of carotenoid, re- light as well as intensity of the observing light. flecting changes in membrane potential, was Furthermore, sensitivity of the cells to the pho- confirmed by the fact that, in a cell suspension totactic stimulus differed in different cultures. of the carotenoid-less mutant (G-9), no light- However, under the conditions described in induced absorbance change between the two Materials and Methods, 60 to 95% of the cells wavelengths was observed with or without an- responded to the stimulus in control experi- timycin and PMS (Fig. 2). ments. Effect of valinomycin on phototactic re- Antimycin, a specific inhibitor of photosyn- sponse and light-induced absorbance change thetic electron transfer (13, 33), inhibited the of carotenoid. Valinomycin is an ionophore phototactic response of the bacteria (Table 1). that increases conductivity of K+, Rb+, and Cs+ Motility of the bacteria was not affected by the in biological membranes; in the presence of inhibitor under aerobic conditions. The inhibi- valinomycin, these cations move down the elec- tory effect of antimycin on the phototactic re- trochemical potential across the membrane and sponse was abolished when phenazine metho- discharge it (9, 28). We used the ionophore to sulfate (PMS) was subsequently added to the examine the role of membrane potential in the bacterial suspension. PMS had a stimulative bacterial phototactic response because its effect effect on phototactic response when it was on biological membranes was well character- added to the bacterial suspension containing no ized. antimycin. When valinomycin was added to the cell sus- Antimycin affected light-induced redox pension of R. rubrum with or without KCl, it changes of cytochrome c2 measured as A552-540: had no effect on the phototactic response or on the light-induced absorbance change of carotenoid. Since it is known that the outer mem- TABLE 1. Effect of antimycin and PMS on brane of gram-negative bacteria acts as a pene- phototactic response tration barrier for hydrophobic antibiotics hav- Phototaxis ing a molecular weight of 1,100 (20), it is possi- Additiona Nb ble that the insensitivity of intact R. rubrum Ethanol (0.2%, vol/vol) 224 60d cells to the ionophore is due to the existence of Antimycin (20 ,uM) 129 5d an intact outer membrane. It was reported that Ethanol (0.2%, vol/vol) + PMS 180 84e ethylenediaminetetraacetate (EDTA) removed (0.2 mM) the penetration barrier (17), and thus E. coli, a Antimycin (20 gM) + PMS (0.2 163 67e gram-negative bacterium, became sensitive to mM) valinomycin (27). Upon EDTA treatment, cells a Cells ofR. rubrum S-1 were suspended in 10-2 M of R. rubrum also became sensitive to valino- Tris-hydrochloride buffer (pH 7.6) containing the mycin: the phototactic activity of the EDTA- addition. treated cells was strongly inhibited by valino- bNumber of the bacteria examined. mycin. Spontaneous reversal frequency of the ' Fraction of the bacteria that showed reversal(s) cells, under steady light illumination, was also of swimming direction 0.5 s after actinic light was diminished by valinomycin. The concentration turned off. d Phototactic response was observed 1 to 1.5 h of EDTA required to render the cells sensitive after the addition of ethanol or antimycin. to valinomycin differed in different cell cul- e PMS was added 1.5 h after the addition of tures: cells in the stationary growth phase were ethanol or antimycin, and then the phototactic re- more resistant to EDTA than those of the expo- sponse was observed 10 to 20 min afterwards. nential growth phase. Cumulative results on VOL. 131, 1977 PHOTOTAXIS IN R. RUBRUM 37

Ethanol Antimycin Ethanol Antimycin PMS PMS 1444444444 I A6004585 Anti,rnycI (G 9) fff tftffff

2 7A552540 Iftt fIftit itltfttlIft ftf itftftt f tftfft A52s-5 o tftt tftftt t0.01

fl fttft ft tftttft ftft ttt ftttti tt ffitt 208 FIG. 2. Effect of antimycin on the light-induced absorbance change in bacterial suspension. Cells of R. rubrum strains S-i and G-9 suspended in 10 mM Tris-hydrochloride buffer (pH 7.6) were incubated aerobically on a reciprocating shaker (120 strokes per min) at 28°C for 1 h. Then, 20 pM antimycin dissolved in ethanol (final concentration, 0.2% [vol/vol]) or 0.2% (vollvol) ethanol was added to the bacterial suspensions. They were further incubated aerobically for 1 to 1.5 h, and then light-induced absorbance changes were measured as described in the text. Freshly prepared PMS (0.2 mM) was added 1.5 h after the addition of ethanol or antimycin, and the light-induced absorbance changes were measured within 10 min. Arrows (.I t) indicate light on and light off, respectively. Absorbance changes of bacteriochlorophyll (A,5858 and cytochrome c2 (A552--W) were measured with the carotenoid-less mutant G-9, at a bacteriochrolophyll concentration of 18W. The mutant strain was used for the measurements because, in wild-type cells, these absorbance changes were obscured by associated carotenoid changes at these wavelengths (21). The absorbance change ofcarotenoid was measured with wild-type strain S-l, at a bacteriochlorophyll concentration of 23 pM, at A525 510, where no absorbance change was observed in carotenoid-less mutant G-9. Phototactic activities in the bacteria used in the experiment were 85% (+ethanol), 10% (+antimycin), 98% (+PMS), and 70% (+antimycin and PMS) in G-9 cells; and 73% (+ethanol), 9% (+antimycin), 86% (+PMS), and 73% (+antimycin and PMS) in S-i cells. The motility of G-9 cells has been observed to be impaired under high light intensity in aerobic conditions. This may be a result ofphotodynamic action (7). However, under the present experimental conditions, no impairment of motility was observed.

phototactic response are presented in Table 2. Light-induced absorbance changes of carote- In these experiments, S-1 cells at a growth noid were measured with the samples listed in phase indicated in the table were suspended in Table 2; some of the results are presented in 10 mM Tris-hydrochloride buffer (pH 7.6) con- Fig. 3. In the cells used in experiment 2 ofTable taining 10 mM KCI and were then incubated 2, absorbance changes of carotenoid were nor- aerobically with shaking (120 strokes per min) mal when ethanol was added to the cell suspen- at 280C. The indicated concentrations of EDTA sion, but in the presence of valinomycin they and ethanol (1% vol/vol) or valinomycin (10 were prevented. In cells used in experiment 5 of ,uM) dissolved in ethanol (final concentration, Table 2, even without the addition ofvalinomy- 1%, vol/vol) were added to the cell suspensions, cin, phototactic activity was strongly reduced which were further incubated aerobically for by EDTA. The cells showed only the slight the indicated times. Since the bacteria treated absorbance change of carotenoid. Why EDTA with EDTA had a tendency to adhere to cover inhibited the reaction was unclear. These re- slips and glass slides, the phototactic response sults showed that inhibition of the light-in- of the EDTA-treated cells was observed by us- duced absorbance change ofcarotenoid, i.e., the ing a bacteria counting chamber (20-gm depth) light-induced change in membrane potential, in which their motility was fair. by valinomycin or by EDTA resulted in a con- 38 HARAYAMA AND IINO J. BACTERIOL. TABLE 2. Effect of valinomycin on phototactic response

Expt Growth phase EDTA concn (M) IncubationEDTA (min)time in Additiona Nb Phototaxis(% 1 Stationary 10-3 42-52 E 121 53 V 55 7 2 Stationary 10-3 280-285 E 33 67 V 44 2 3 Stationary 10-3 240-245 E 64 73 V 52 6 4 Late exponential 10-4 35-50 E 51 49 V 35 6 5 Late exponential 10-4 80-83 E 61 10 V 44 0 6 Stationary 10-3 100-105 E 73 62 V 46 2 10-4 200-205 E 92 78 V 64 78 a V, 10-5 M valinomycin; E, 1% (vol/vol) ethanol. b Number of bacteria examined.

* 4 +$* +4+4+

4 +t t + + t+ +

9* + t+ + + + +*+ J0.001 A r*,,,,,44B T~lOs4 FIG. 3. Light-induced absorbance changes ofcarotenoid in EDTA-treated cells with or without the addition ofvalinomycin. Light-induced absorbance changes at 525 nm, with 510 nm as a reference ofthe bacterial suspensions used in experiments ofTable 2, were measured immediately after the measurements ofphototactic response. (A) Bacterial suspension used in experiment 2 of Table 2 (bacteriochrolophyll concentration, 23 pM). (B) Bacterial suspension used in experiment 5 ofTable 2 (bacteriochrolophyll concentration, 13 M). (E) Ethanol (19%, vollvol) added. (V) valinomycin (10 pM) added. comitant inhibition of bacterial phototaxis. only when appropriate monovalent cations The effect of incubation time of the cells in such as K+ and Rb+ are present. To examine EDTA solution on their phototactic response in whether the inhibitory effects ofvalinomycin are the presence or absence of valinomycin was dependent on the presence of these cations, the examined (Fig. 4). In the experiment, 10 mM effects of valinomycin on phototactic response CaCl2 was added to the cell suspension to stop and light-induced absorbance change of carote- the action of EDTA at the indicated time. It noid in the EDTA-treated cells suspended in was shown that sensitivity ofthe cells to valino- buffer free from these cations were examined. mycin increased with increased incubation For this purpose, cells ofR. rubrum S-1 at the time in EDTA solution; however, prolonged in- late exponential growth phase were harvested, cubation resulted in inhibition of phototactic suspended in 10 mM sodium phosphate buffer response even without addition ofvalinomycin. (pH 6.8), and incubated aerobically with shak- Valinomycin discharges membrane potential ing (120 strokes per min) at 28°C for 1 h; 0.3 mM VOL. 131, 1977 PHOTOTAXIS IN R. RUBRUM 39 TABLE 3. Requirement ofK+ or Rb+ for the inhibitory effect of valinomycin on both phototactic response and light-induced absorbance change of carotenoid

Addi- Monovalent cation b Photo- AAB/M tiona added (%)xi BChlc5 R t\ I~~~~9I'Jr- E water (1%, vol/vol) 41 73 3.0 V water (1%, vol/vol) 36 86 2.6 E 10-2 M KCl 44 73 2.3 0 V 10-2 M KCl 100 15 0.8 E 10-2 M RbCl 41 80 2.6 V 10-2 M RbCl 70 16 1.0 a E, Ethanol (1%, vol/vol); V, valinomycin (10-5 M). ' Number of bacteria examined. c Mean of 10 absorbance changes of carotenoid (A525510) Val per micromolar bacteriochrolophyll (BChl) 0.5 s after the actinic light was turned off.

2 15 45 85 tory effects of valinomycin on the phototactic INCUBATION TIME IN EDTA(MIN) response and light-induced absorbance change FIG. 4. Effect of incubation time in EDTA solu- of carotenoid were dependent on the existence tion on phototactic response. S-i cells harvested in of appropriate cationic species such as K+ and the late exponential growth phase were suspended in Rb+, but not Na+. 10-2 M Tris-hydrochloride buffer containing 10 mM KCl and were incubated aerobically at 280C for 1 h. DISCUSSION EDTA (0.1 mM) and ethanol ( %, vollvol) or valino- Pigmented cells of R. rubrum reverse their mycin (10 pM) were added to the bacterial suspen- swimming direction when the intensity of light sions, which were further incubated aerobically at impinging on them is abruptly decreased. Since 28°C. CaCl2 (1 mM) was added to stop the action of EDTA at a time, indicated on the figure, after the the reversal of swimming direction is a key addition ofEDTA, and then the phototactic response movement in attaining phototactic accumula- was observed 15 min afterwards. tion (6, 10), we examined the fraction of bacteria that reverse their swimming direction after a sudden decrease in light intensity to estimate EDTA and ethanol (1%, vol/vol) or valinomycin their phototactic activity. Under the conditions (10 AM) were added to the cell suspension and of the experiments, very few cells showed re- further incubated aerobically for 30 min. CaCl2 peated reversals within 0.5 s after a sudden (2 mM) was added to stop the action ofEDTA on decrease in actinic light intensity; i.e., almost the cells. Thirty minutes after the addition of all of the cells showed one or no reversals after CaCl2, 10 mM KCl, 10 mM RbCl, or water (final the stimulus. concentration, 1% [vol/vol]) was added to the Photoreceptor pigments responsible for the cell suspensions, and phototactic activities of phototactic response of the bacteria have been the cells in these suspensions were measured 20 identified as bacteriochlorophyll and carote- to 35 min and 40 to 55 min afterwards. The noids (4, 19). These are pigments driving photo- mean of two measurements is presented in Ta- synthesis. Thus, the hypothesis that phototaxis ble 3. The absorbance changes of carotenoid is induced by a transient disturbance of photo- within 0.5 s after the actinic light was turned synthetic metabolism was proposed and widely off were measured 60 to 65 min after the addi- accepted (6). A requirement of photosynthetic tion ofKCl, RbCl, or water (Table 3). When the electron transfer for the manifestation ofphoto- bacteria were suspended in the 10 mM sodium taxis has been claimed by Throm (32). He ob- phosphate buffer containing no KCI, addition of served that inhibitors of photosynthetic elec- valinomycin after EDTA treatment of the cells tron transfer affected the maximum level of did not affect significantly their phototactic re- accumulation of cells in a light spot. However, sponse or the light-induced absorbance change the rate and the maximum level of bacterial ofcarotenoid. Further addition of 10 mM KCl or accumulation in a light spot may be affected RbCl to the bacterial suspension inhibited the complexly by changes in the speed of the bacte- phototactic response as well as the light-in- ria or the fraction of motile bacteria, as well as duced absorbance change of carotenoid (Table their phototactic activity. Therefore, it is pref- 3). The results clearly indicated that the inhibi- erable to observe the behavior of individual 40 HARAYAMA AND IINO J. BACTERIOL. cells rather than the mass accumulation of bac- The involvement of membrane potential in teria in a light spot for determination of their bacterial tactic response has been proposed by phototactic response. The conclusion of Throm several authors (1). Recently, Szmelcman and was confirmed by the present experiments in Adler (31) reported that a chemotactic stimulus which antimycin, an inhibitor of photosyn- induces changes in membrane potential of E. thetic electron transfer, prevented phototactic coli cells. They discussed the possible role of ion response. Motility of the bacteria was not af- fluxes in regulating flagellar rotation. The im- fected by the inhibitor under aerobic condi- portant role of ions in chemotactic responses tions, indicating that the energy for motility is was suggested by Ordal and Goldman (23-25). provided by respiratory metabolism, which is They reported that uncouplers of oxidative not affected by antimycin (data now shown), as phosphorylation, inhibitors of electron trans- well as by photosynthetic metabolism. The in- port, and permeable anions induce transient hibitory effect of antimycin on both phototactic "tumbling"; that A23187, an ionophore for diva- response and photosynthetic electron transfer lent cations, plus EDTA induces incessant was annulled by PMS (Table 1). The effect of "tumbling"; and that permeable cations inhibit PMS has been interpreted as the formation of a "tumbling" in Bacillus subtilis. Ordal (23) pro- bypass around the blocking site ofantimycin in posed that binding of certain cations to the the electron transfer pathway (29). switch of a hypothetical chemotactic machinery From these observations, it is inferred that inhibits "tumbling." the reversal of swimming direction after a sud- Two possible roles ofthe membrane potential den decrease in actinic light intensity is a con- in the phototactic response of R. rubrum are sequence ofthe decrease in photosynthetic elec- conceivable: (i) the tactic signal inducing rever- tron flow. Accompanying the decrease of elec- sal of swimming direction is the depolarization tron flow is a change in the redox state of ofmembrane potential; (ii) a normal membrane components of the electron transfer system and potential is required for flagellar reversal to a decrease (depolarization) of membrane poten- occur. Since we could not determine the role of tial measured as an absorbance change of the membrane potential in the phototactic re- carotenoid (Fig. 3). Both the reduction of cyto- sponse, we present three alternative possibili- chrome c2 and the decrease in membrane poten- ties for the signal that induces the response: a tial after the light was turned off were inhibited change in the redox state of the components in by antimycin. Thus, by considering only the the photosynthetic electron transfer system; action of antimycin on phototactic response, uptake of H+ (decrease in ApH across the mem- one cannot determine whether a redox change brane); and depolarization of the membrane po- of electron transfer components or another tential. event, such as depolarization of membrane po- Bacterial motility was not paralyzed by vali- tential, is important in the induction of a photo- nomycin plus potassium (or rubidium). The re- tactic response in bacteria. lationship between membrane potential and Valinomycin is an ionophore that increases cell motility will be discussed elsewhere (S. the permeability of biological membranes to Harayama and T. Iino, manuscript in prepara- certain cations (K+, Rb+, and Cs+). In the pres- tion). ence of these cations, the electrochemical gradient across the membrane is affected. Intact ACKNOWLEDGMENTS cells ofR. rubrum were insensitive to the anti- This work was supported by research grants 048068 and biotic, presumably because oftheir intact outer 178097 from the Japanese Ministry of Education, Science membrane. However, when they were and Culture. "properly" treated with EDTA, valinomycin plus 10 mM KCl inhibited both the phototactic LITERATURE CITED response and the light-induced absorbance 1. Adler, J. 1975. Chemotaxis in bacteria. Annu. Rev. change of carotenoid (Table 2, Fig. 4) but not Biochem. 44:341-356. the redox changes of components in the photo- 2. Berg, H. C. 1975. Bacterial behaviour. Nature (London) synthetic electron transfer system (data not 254:389-392. shown). The inhibitory effect of valinomycin on 3. Berg, H. C., and P. M. Tedesco. 1975. Transient response to chemotactic stimuli in Escherichia coli. the bacterial phototactic response is probably Proc. Natl. Acad. Sci. U.S.A. 72:3235-3239. due to depolarization of the membrane, not to 4. Clayton, R. K. 1953. Studies in the phototaxis of Rho- permeability of any particular cation, since the dospirillum rubrum. I. Action spectrum, growth in phototactic response was affected by the iono- green light, and Weber law adherence. Arch. Mikro- biol. 19:107-124. phore only when the membrane was depolar- 5. Clayton, R. K. 1963. Toward the isolation of a photo- ized by the presence ofboth valinomycin and an chemical reaction center in Rhodopseudomonas sphe- interacting cation (Table 3). roides. Biochim. Biophys. Acta 75:312-323. VOL. 131, 1977 PHOTOTAXIS IN R. RUBRUM 41 6. Clayton, R. K. 1964. Phototaxis in microorganisms, p. phobic substances. Biochim. Biophys. Acta 433:118- 51-77. In A. C. Giese (ed.), Photophysiology, vol. 2. 132. Academic Press Inc., New York. 21. Nishimura, M., and B. Chance. 1963. Studies on the 7. Griffiths, M., W. R. Sistrom, G. Cohen-Bazire, and R. electron-transfer systems in photosynthetic bacteria. Y. Stanier. 1955. Function ofcarotenoids in photosyn- I. The light-induced absorption-spectrum changes thesis. Nature (London) 176:1211-1214. and the effect of phenylmercuric acetate. Biochim. 8. Harayama, S., and T. Iino. 1976. Phototactic response Biophys. Acta 66:1-16. of aerobically cultivated Rhodospirillum rubrum. J. 22. Nultach, W. 1973. Phototaxis and photokinesis in bacte- Gen. Microbiol. 94:173-179. ria and blue-green algae, p. 70-81. In A. Perez-Mirav- 9. Harold, F. M. 1972. Conservation and transformation of ete (ed.), Behaviour of micro-organisms. Plenum energy by bacterial membranes. Bacteriol. Rev. Press, London. 36:172-230. 23. Ordal, G. W. 1976. Control of tumbling in bacterial 10. Haupt, W. 1966. Phototaxis in plants. Int. Rev. Cytol. chemotaxis by divalent cation. J. Bacteriol. 126:706- 19:267-299. 711. 11. Hazelbauer, G. L. 1975. Maltose chemoreceptor ofEsch- 24. Ordal, G. W., and D. J. Goldman. 1975. Chemotaxis erichia coli. J. Bacteriol. 122:206-214. away from uncouplers ofoxidative phosphorylation in 12. Hildebrand, E., and N. Dencher. 1975. Two photosys- Bacillus subtilis. Science 189:802-805. tems controlling behavioural responses ofHalobacter- 25. Ordal, G. W., and D. J. Goldman. 1976. Chemotactic ium halobium. Nature (London) 257:46-48. repellents of Bacillus subtilis. J. Mol. Biol. 100:103- 13. Izawa, S., and N. E. Good. 1972. Inhibition ofphotosyn- 108. thetic electron transport and photophosphorylation. 26. Parkinson, J. S. 1975. Genetics of chemotactic behavior Methods Enzymol. 24:355-377. in bacteria. Cell 4:183-188. 14. Jackson, J. B., and A. R. Crofts. 1971. The high energy 27. Pavlasov&, E., and F. M. Harold. 1969. Energy cou- state in chromatophores from Rhodopseudomonas pling in the transport of ,3-galactosides by Esche- spheroides. FEBS Lett. 4:185-189. richia coli: effect of proton conductors. J. Bacteriol. 15. Jackson, J. B., and P. L. Dutton. 1973. The kinetic and 98:198-204. redox potentiometric resolution of the carotenoid 28. Pressman, B. C. 1976. Biological applications of iono- shifts in Rhodopseudomonas spheroides chromatophores. Annu. Rev. Biochem. 45:501-530. phores: their relationship to electric field alterations 29. Siebeama, C. 1970. Photosynthetic bacteria, p. 57-93. In in electron transport and energy coupling. Biochim. P. Halldal (ed.), Photobiologie of microorganisms. Biophys. Acta 325:102-113. Wiley-Interscience, London. 16. Koshland, D. E., Jr. 1974. Chemotaxis as a model for 30. Spudich, J. L., and D. E. Koshland, Jr. 1975. Quantita- sensory systems. FEBS Lett. 40:S3-S9. tion of the sensory response in bacterial chemotaxis. 17. Leive, L. 1965. Release of lipopolysaccharide by EDTA Proc. Natl. Acad. Sci. U.S.A. 72:710-713. treatment of E. coli. Biochem. Biophys. Res. Com- 31. Szmelcman, S., and J. Adler. 1976. Change in mem- mun. 21:290-296. brane potential during bacterial chemotaxis. Proc. 18. Lengeler, J. 1975. Mutations affecting transport of the Natl. Acad. Sci. U.S.A. 73:4387-4391. hexitols r.-mannitol, D-glucitol, and galactitol in 32. Throm, G. 1968. Untersuchungen zum Reaktionsme- Escherichia coli K-12: isolation and mapping. J. Bac- chanismus von Phototaxis und Kinesis an Rhodospi- teriol. 124:26-38. rillum rubrum. Arch. Protistenkd. 110:313-371. 19. Milatz, J. M. W., and A. Manten. 1953. The quantita- 33. van Grondelle, R., L. N. M. Duysens, and H. N. van der tive determination of the spectral distribution ofpho- Wal. 1976. Function of three cytochromes in photo- totactic sensitivity in the purple bacterium Rhodospi- synthesis of whole cells ofRhodospirillum rubrum as rillum rubrum. Biochim. Biophys. Acta 11:17-27. studied by flash spectroscopy. Evidence for two types 20. Nikaido, H. 1976. Outer membrane ofSalmonella typhi- of reaction center. Biochim. Biophys. Acta 449:169- murium. Transmembrane diffusion of some hydro- 187.