- Email: [email protected]
Improvement of growth stability of photosynthetic bacterium Rhodobacter capsulatus
JOURNAL OF BIOSCIENCE AND BIOENGINEERING Vol. 100, No. 6, 672–677. 2005 DOI: 10.1263/jbb.100.672
© 2005, The Society for Biotechnology, Japan
Improvement of Growth Stability of Photosynthetic Bacterium Rhodobacter capsulatus Reza Yegani,1 Satoshi Yoshimura,1 Kazunori Moriya,1 Tomohisa Katsuda,1 and Shigeo Katoh1* Graduate School of Science and Technology, Kobe University, 1-1 Rokkodai, Nada-ku, Kobe 657-8501, Japan1 Received 15 June 2005/Accepted 12 September 2005
Using a semicontinuous culture method, in which operational parameters such as cell concentration and light intensity distribution were maintained almost constant, instability of the specific growth rate of Rhodobacter capsulatus B-100, a purple bacterium, was observed to be similar to that of R. capsulatus ST-410 when cultivated under high ratios of light intensity on the illuminated side to that of the transmitted light. Such instability was not observed in the cultivation of Chlorella vulgaris, a eukaryotic green alga, even at higher cell concentrations. Under the same conditions, the increase in only the ferrous concentration from 43 µM, the concentration in the original RCV medium, to 172 µM sustained a stable growth, whereas Fe2+ was slightly consumed during the cultivation. Supplemental illumination with a fluorescent lamp on the transmitted side of a flat plate photobioreactor sustained a moderate level of stable growth, while a halogen lamp slightly affected the growth stability. Our results showed that an increase in Fe2+ concentration or supplemental illumination improves the growth stability of R. capsulatus. [Key words: Rhodobacter capsulatus, growth stability, semicontinuous culture, light intensity distribution, iron concentration]
In this study, the effects of the concentrations of Fe2+ and Mg2+ on the growth stability of R. capsulatus, and supplemental illumination using different light sources emitting different spectra from the rear side (transmitted side) of a flat-plate photobioreactor, were investigated. The growth stability of the alga, Chlorella vulgaris, under similar illumination conditions as R. capsulatus was also examined.
Every microorganism must find in its environment all the substances necessary for energy generation and cellular biosynthesis (1). To increase the productivity and/or yield of metabolites when cultivating photosynthetic microorganisms, it is not only necessary to optimize cultivation conditions, particularly the general variables, such as temperature, pH, DO, components of culture media and transport properties, but also light dependent variables in photobioreactors, such as average light intensity, wavelength of illumination and light intensity distribution (2–5). In our previous work (6, 7), we have shown that the distribution of light intensity affected the growth stability of Rhodobacter capsulatus ST-410, one of the purple nonsulfur-photosynthetic bacteria, of which genera are widely used for the production of hydrogen, biopolymer and other secondary metabolites (8, 9). We have also shown that higher average light intensities alone does not cause a decrease in the specific growth rate, and that the ratio of light intensity on the illuminated surface to that of the transmitted light through a flat-plate photobioreactor might affect the specific growth rate stability. However, it is still an unknown phenomenon. From the engineering viewpoint, growth stability under optimal bioreactor configurations and cultivation conditions is crucial, particularly in continuous or cyclic cultivation systems (10). Thus, it is necessary to identify which parameters can cause growth instability and how this instability can be prevented.
MATERIALS AND METHODS Materials R. capsulatus ST-410 is a hydrogenase-deficient mutant derived from R. capsulatus B-100. R. capsulatus ST-410 and R. capsulatus B-100 were anaerobically cultured in RCV medium (pH 6.8) containing 7.5 mM (NH4)2SO4 and 30 mM D,L-malate as sole nitrogen and carbon sources, respectively, and the following basal salts: 43 µM FeSO4, 0.8 mM MgSO4, 0.5 mM CaCl2, 3 µM thiamine–HCl, 61 µM EDTA, 10 mM potassium phosphate buffer (pH 6.8) and trace elements (Mn, Borate, Cu, Zn, and Mo) (11). C. vulgaris was cultured in a medium (pH 6) containing 8 mM KNO3 and 8 mM NaNO3 as nitrogen sources and the following salts: 3 mM NaH2PO4 ⋅H2O, 1 mM MgSO4, 1 mM Na2HPO4 ⋅2H2O, 0.1 mM Ca(NO3)2 ⋅ 4H2O, 10 µM FeEDTA and micronutrients (Borate, Zn, Mn, Cu, and Mo) (12). The chemicals used were of reagent grade. Preculture and semi-continuous cultivation R. capsulatus was inoculated in a screw-capped test tube (25 ml) completely filled with the RCV culture medium and precultured by incubation in a glass-sided water bath at 30°C under illumination with a halogen lamp (90 W, JDR110V90WN/S-K12; Toshiba, Tokyo) with an incident light intensity of approximately 10 mW/cm2. At the logarithmic growth phase, the cells were harvested by centrifugation at
* Corresponding author. e-mail: [email protected] phone/fax: +81-(0)78-803-6193 672
GROWTH STABILITY OF R. CAPSULATUS
VOL. 100, 2005
673
FIG. 1. Experimental setup for semi-continuous cultivation of R. capsulatus.
10,000×g for 10 min and resuspended in a glass rectangular vessel (flat-plate photobioreactor) 9.8 cm high, 9.2 cm wide and 2.0 cm thick (130 ml working volume) containing the RCV medium (pH 6.8). C. vulgaris was inoculated in a 500-ml Erlenmeyer flask containing 50 ml RCV medium capped with a silicone sponge and incubated for 3 d under shaking at 30°C and illuminated using a florescent lamp at the incident light intensity of approximately 0.1 mW/cm2. During this procedure, carbon was supplied by penetration of air containing CO2 through the silicone sponge into the medium. After 3 d, cells were harvested by centrifugation at 5500×g for 15 min and resuspended in 50 ml of fresh RCV medium and precultivated in a 100-ml Erlenmeyer flask. The culture was incubated at 30°C under stirring with a magnet stirrer, and air containing 10% v/v CO2 was bubbled. At logarithmic growth phase, cells were again harvested by centrifugation at 5500×g for 15 min and resuspended in the glass rectangular vessel (height, 9.8 cm; width, 9.2 cm; thickness, 2.0 cm; working volume, 130 ml) containing fresh RCV medium. Figure 1 shows the experimental setup used for semi-continuous cultivation of R. capsulatus. The rectangular vessel was incubated in the glass-sided water bath at 30°C and illuminated on one side or both sides with a halogen lamp. In some cases, rear side illumination was carried out with one or two florescent lamps (FL 20SS, N18, 18 W; Matsushita Electric Industrial, Osaka). The incident light intensities in both cases were from 0.2 to 8.8 mW/cm2 for the halogen lamp and 0.1 to 0.2 mW/cm2 for the fluorescent lamp. The incident light intensities at 9 points on the illuminated side of the vessel were measured with a thermopile photosensor (3), and the averaged value was used as the incident light intensity on the illuminated side. In the semi-continuous cultivation method, a part of the culture broth containing grown cells was repeatedly replaced with fresh medium at a predetermined time interval to keep the cell concentration, the volume of the culture broth and the distribution of the light intensity almost constant to their initial values. On the basis of the specific growth rate of each strain, the predetermined time intervals were set at 40–60 min and 4–5 h for R. capsulatus and C. vulgaris, respectively. In the cultivation of R. capsulatus, before and after the replacement of medium, the culture was mixed by bubbling N2 sterilized through a membrane microfilter. Although the culture broth was mixed only by natural convection during the cultivation cycles,
sedimentation of cells was not observed. In the case of C. vulgaris, the culture broth was continuously mixed with a magnetic stirrer. To prevent the penetration of O2 into the culture broth of R. capsulatus, the reactor and the reservoir of the fresh medium were kept under N2 atmosphere. Cultivation of C. vulgaris was carried out aerobically. Measurements of dry cell weight and ion concentration The dry cell weights of R. capsulatus ST-410 and B-100, and C. vulgaris in the culture broth were determined by drying the cells at 80°C for 50 h after centrifugation and thorough washing of the cells with distilled water; the weights were correlated to the absorbance measured with a UV spectrophotometer (UV-1600; Shimadzu, Kyoto) at 660 and 560 nm, respectively. At the end of each cultivation cycle, the cell concentration was determined from the absorbance of the culture broth. The concentration of Fe2+ was measured using Fe C-Test Wako (Wako Pure Chemical Industries, Osaka). This assay method is based on the selective chelation of Fe2+ into Nitroso-PASP (2-Nitroso-5-[N-n-propyl-N-(3-sulfopropyl)amino]phenol). Because EDTA in the RCV medium competed for chelation and lowered the measured concentration of Fe2+, we inactivated the EDTA with hydrogen peroxide before the assay. Hydrogen peroxide (30 wt%, 0.1 ml) was added into 0.4 ml of the supernatant of the culture broth obtained by centrifugation, and this mixture was incubated for 10 min at 80°C. This treatment also oxidized Fe2+ in the mixture to Fe3+. Therefore, the concentration of iron was determined using the Fe C-Test Wako after the reduction of Fe3+ to Fe2+ by 0.45 M thioglycolic acid in 0.4 M acetate buffer. Measurement of chlorophyll content of cells Chlorophyll is a colored material that absorbs light and uses its energy to synthesize carbohydrates from CO2 and water. The absorption spectrum of R. capsulatus consists of two main peaks at 800 and 850 nm, which correspond to chlorophyll a and b, respectively. The concentration of chlorophyll may reflect cell physiological condition under illuminated conditions. During the semicontinuous cultivation of R. capsulatus, the absorbance of chlorophyll at 800 nm was measured using the UVspectrophotometer and the ratio of this absorbance to the cell concentration was considered as the chlorophyll content per unit cell. Calculation of light intensity distribution The light intensity distribution in the photobioreactor was calculated from the equation given in our previous work (13), assuming that the attenuation of the light intensity of polychromatic light is given as a
674
YEGANI ET AL.
FIG. 2. Time courses of cell concentration in semicontinuous cultivation of R. capsulatus ST-410. Solid circle: I 0 = 8.8 mW/cm2, C = 0.1 mg/ml, Iave = 4.8 mW/cm2, I0/It = 3.6. Solid triangle: I0 = 8.8 mW/cm2, C = 0.55 mg/ml, Iave = 2.0 mW/cm2, I0/It = 40. Open triangle: I0 = 2.1 mW/cm2, C = 0.55 mg/ml, Iave = 0.47 mW/cm2, I0/It = 40.
J. BIOSCI. BIOENG.,
FIG. 3. Specific growth rates of R. capsulatus ST-410 (solid circle and triangle) and B-100 (open circle and triangle). Circle: I0=8.8 mW/cm2, C = 0.1 mg/ml, Iave = 4.8 mW/cm2, I0/It = 3.6. Triangle: I0 = 8.8 mW/cm2, C = 0.55 mg/ml, Iave = 2.0 mW/cm2, I0/It = 40.
summation of the attenuation of each wavelength. Our previous results showed that the predicted distributions showed good agreement with the measured distributions. The average light intensity was calculated by dividing an integrated value of the light intensity along the light path by the light path length of the photobioreactor similar to the method previously reported.
RESULTS AND DISCUSSION Semi-continuous cultivation of other species and family In the semi-continuous cultivation, the specific growth rate was calculated by dividing the increase in the cell concentration by the initial cell concentration and time during one cycle. Figure 2 shows the time courses of the cell concentration in semicontinuous cultivation at various cell concentrations (C, mg/ml), incident light intensities (I0, mW/cm2) and average light intensities (Iave, mW/cm2). The cell concentration at the beginning of each cycle was kept almost constant. When cells grew at a constant specific growth rate (solid circles and open triangles in Fig. 2), the slope of each increment was also constant, as shown in Fig. 2. In our previous papers (6, 7), we have shown the growth instability of R. capsulatus ST-410 under high ratios of light intensity on the illuminated side (I0) to that on the transmitted side (It) caused by high cell concentrations and/or long light paths. We tried to clarify whether this growth instability caused by different illumination conditions is a common phenomenon in photosynthetic microorganisms and also to determine suitable cultivation conditions to prevent this instability. At first, experiments with another Rhodobacter species, R. capsulatus B-100, and also with a eukaryotic green alga C. vulgaris were carried out. The results of cultivation of R. capsulatus B-100 and C. vulgaris are shown in Figs. 3 and 4, respectively. As indicated by the circle in Fig. 3, at a low cell concentration (C = 0.1 mg/ml) both R. capsulatus ST-410 and B-100 showed stable specific growth rates for 16 h. At a cell concentration of 0.55 mg/ml, R. capsulatus B-100 (open triangle in Fig. 3) showed unstable specific growth rate similar to R. capsulatus ST-410 (solid triangle in Fig. 3) under the same incident light intensity (I0 = 8.8
FIG. 4. Specific growth rate of C. vulgaris. Open triangle: I 0 = 8.8 mW/cm2, C = 0.55 mg/ml, Iave = 3.2 mW/cm2, I0/It = 10. Open square: I 0 = 8.8 mW/cm2, C= 1.1 mg/ml, Iave = 2.0 mW/cm2, I0/It = 30. Open diamond: I 0 =8.8 mW/cm2, C= 2.2 mg/ml, Iave = 1.2 mW/cm2, I0/It = 79.
mW/cm2). However, in the cultivation of C. vulgaris, instability of its specific growth rate was not observed under the same conditions. After 100 h cultivation at the cell concentration of 1.1 mg/ml, the cell concentration was increased to 2.2 mg/ml, and the growth was still stable at a different specific growth rate. Effect of media compositions Although the RCV medium has already been optimized for the cultivation of R. capsulatus particularly for long-term cultivation (14, 15), the instability of its specific growth rate might be caused by deficiency of some components in the medium. Because fresh medium was repeatedly added in the case of semicontinuous cultivation at a predetermined time interval, nutrient deficiency seldom occurred. To confirm this hypothsis, the effects of different concentrations of a key metal ion Fe2+ in RCV medium, which is the central atom in the proto-porphyrin called heme, were examined. Figure 5 shows the results of cultivation of R. capsulatus ST-410 at different Fe2+ concentrations in the RCV medium. By increasing ferrous ion concentration from the original 43 µM to 86 µM, the slope of decrease in the specific growth rate became less steep. At 172 µM (4 times), the specific
VOL. 100, 2005
FIG. 5. Effects of Fe2+ concentration on stability of specific growth rate. I 0 = 8.8 mW/cm2, C = 0.55 mg/ml, Fe2+ = 172 µM (open diamond); 86 µM (solid diamond); 43 µM (solid triangle); 22 µM (solid square).
FIG. 6. Concentration of Fe2+ during cultivation under unstable condition. I 0 =8.8 mW/cm2, C=0.55 mg/ml, Fe2+ =43 µM, Iave =2.0 mW/cm2, I0/It = 40.
growth rate was kept almost constant. Similar growth stability was observed with the addition of 129 µM ferric ion to the original RCV medium (data not shown). When the ferrous ion was 22 µM, the specific growth rate was decreased after 5–6 h similar to the original concentration of ferrous ion in the RCV medium. We speculated that the lack of Fe2+ might have caused the instability of specific growth rate of R. capsulatus. However, as shown in Fig. 6, the Fe2+ concentration under a typical unstable cultivation condition (I0 = 8.8 mW/cm2, C = 0.55 mg/ml, initial Fe2+ = 43 µM) remained almost constant, although the measured values were slightly higher than that calculated from an added amount. Thus, Fe2+ was slightly consumed by cells. The increase in only the Fe2+ concentration of the RCV medium could maintain growth stability. This finding indicates that the decrease in the specific growth rate was not caused by deficiency of any nutrients including Fe2+ in the RCV medium. The increase in the Fe2+ concentration enhanced the transport of Fe2+ into cells, and thus enhanced the availability of ferrous ions that might have induced a change in the physiological conditions in cells to sustain stable growth. The twofold increase in the concentration of Mg2+, which is contained in the chlorophyll molecule, did
GROWTH STABILITY OF R. CAPSULATUS
675
FIG. 7. Effects of Fe2+ addition on specific growth rate during cultivation under unstable condition. Solid triangle: 0–13 h: I 0 =8.8 mW/cm2, C= 0.55 mg/ml, Fe2+ = 43 µM. Open diamond: 13 h~: I 0 = 8.8 mW/cm2, C= 0.55 mg/ml, Fe2+ = 172 µM. Broken line: calculated specific growth rate.
not affect the growth stability (data not shown). Because an increase in Fe2+ and not in Mg2+ changed the growth stability of cells, electron transfer systems that are involved in photosynthesis, such as cytochromes and ferredoxin, might be affected by a higher concentration of Fe2+. Recovery of specific growth rate After 13 h cultivation under an unstable condition (I0 = 8.8 mW/cm2, C = 0.55 mg/ml), the concentration of Fe2+ in the culture broth and added fresh medium was increased 4 times, and the specific growth rate of R. capsulatus ST-410 was measured. The specific growth rate increased gradually after approximately 2 h of lag phase, as indicated by the open diamonds in Fig. 7. The calculated specific growth rate (broken line) was obtained with the assumption that the fraction of viable cells was given by the ratio of the specific growth rate at a particular time to the original specific growth rate under the initial conditions, that is, 0.27 1/h, and that the viable cells could grow at the specific growth rate of 0.27 1/h after the lag phase. The observed recovery of the original specific growth rate was attained at a shorter time than that of the calculated result. This means that the decrease in the specific growth rate was not only caused by the death of the cells, but also caused mainly by the change in and depression of metabolism of viable cells. Thus, the increase in the Fe2+ concentration induced a change in physiological conditions of the cells leading to the recover of the original specific growth rate at a faster rate than the calculated specific growth rate. Supplemental illumination from opposite side In our previous work (6, 7) we have shown that the light intensity distribution, that is, the ratio of the light intensity on the illuminated side of the cultivation vessel to that of the transmitted light, seemed to be one of the factors causing growth instability. Therefore, supplemental illumination on the opposite side of the vessel would decrease this ratio and prevent growth instability caused by the light intensity distribution. The vessel was illuminated on the opposite side of the main halogen lamp with another halogen lamp at a light intensity of 0.2 or 0.4 mW/cm2 or with a fluorescent lamp(s) at a light intensity of 0.1 or 0.2 mW/cm2.
676
J. BIOSCI. BIOENG.,
YEGANI ET AL.
FIG. 8. Effects of supplemental illumination with halogen lamp. Open circles: I0,f = 1.8 mW/cm2, I0,t = 0.4 mW/cm2, C = 0.55 mg/ml, Iave = 0.75 mW/cm2, I0/It = 4.1. Solid squares: I0,f = 8.8 mW/cm2, I0,t = 0.2 mW/cm2, C = 0.55 mg/ml, Iave = 2 mW/cm2, I0/It =22.
FIG. 9. Effects of supplemental illumination with fluorescent lamp(s). Open diamond: I0,f = 8.8 mW/cm2, I0,t = 0.1 mW/cm2, C = 0.55 mg/ml, Iave = 2 mW/cm2, I0/It = 28 (one fluorescent lamp). Solid diamond: I0,f = 8.8 mW/cm2, I0,t = 0.2 mW/cm2, C = 0.55 mg/ml, Iave = 2 mW/cm2, I0/It = 22 (two fluorescent lamps). Solid triangle: I 0 = 8.8 mW/cm2, C = 0.55 mg/ml, Iave = 2.0 mW/cm2, I0/It = 40.
Figure 8 shows the changes in the specific growth rate when the vessel was illuminated from the both sides with halogen lamps. At a low incident light intensity of 1.8 mW/cm2 of the main lamp, the growth rate was kept almost constant. At the light intensity of the main-illuminated side was increased to 8.8 mW/cm2, the specific growth rate again decreased after about 5 h cultivation, and reached another stable specific growth rate, that is, 0.15 1/h, which is higher than that under one side illumination. Under supplemental illumination with a fluorescent lamp, as shown in Fig. 9, a moderately stable specific growth rate, that is, 0.23 1/h, was sustained. By illumination with two fluorescent lamps, however, the sustained specific growth rate decreased slightly. The effective light intensity supplied by one fluorescent lamp was about 0.1 mW/cm2. The main difference in the effectiveness of the halogen lamp and the fluorescent lamp might be caused by the differences in spectra of their emitting light wavelengths. The wavelengths emitted from the halogen lamp were longer (500–800 nm) than those emitted
FIG. 10. Changes in chlorophyll content under various cultivation conditions. Open circle: I 0 = 8.8 mW/cm2, C= 0.1 mg/ml, Fe2+ = 43 µM, Iave = 4.8 mW/cm2, I0/It = 3.6. Open diamond: I 0 = 8.8 mW/cm2, C = 0.55 mg/ml, Fe2+ = 172 µM, Iave = 2 mW/cm2, I0/It = 40. Asterisk: I0 = 8.8 mW/cm2, I0,t = 0.1 mW/cm2, C = 0.55 mg/ml, Fe2+ = 43 µM, Iave = 2 mW/cm2, I0/It = 28 (one fluorescent lamp on transmitted side). Solid square: I0 = 8.8 mW/cm2, I0,t = 0.2 mW/cm2, C = 0.55 mg/ml, Fe2+ = 43 µM, Iave = 2 mW/cm2, I0/It = 22 (hologen lamp from transmitted side). Solid triangle: I 0 = 8.8 mW/cm2, C= 0.55 mg/ml, Fe2+ =43 µM, Iave = 2.0 mW/cm2, I0/It = 40.
from a florescent lamp (400–650 nm). Light with shorter wavelengths might be effective in sustaining a stable specific growth rate in long-term cultivation. Relationship between ratio of chlorophyll concentration and growth instability During the semicontinuous cultivation of R. capsulatus, the absorbance of chlorophyll at 800 nm was measured using a UV-spectrophotometer. The ratio of the absorbance to the cell concentration was considered as the chlorophyll content per unit cell. This chlorophyll content in the semicontinuous cultivation is plotted against the cultivation time as shown in Fig. 10. We examined the chlorophyll content in the cells under the cultivation conditions tested. When 172 µM Fe2+ was added (0.27 1/h in Fig. 5), the decrease in the chlorophyll content was less than 5% after 15 h, while more than 10–15% decrease was observed under the unstable cultivation condition at 43 µM Fe2+. Interestingly, although the growth was sustained at a moderately stable level using one fluorescent lamp (0.23 1/h in Fig. 9), the chlorophyll content was lower than that in the cultivation condition with a high Fe2+ condition. These results suggest that chlorophyll content might be used as a criterion for predicting the growth stability of R. capsulatus during long-term cultivation. REFERENCES 1. Todar, K.: Todar’s online textbook of bacteriology. University of Wisconsin, Madison, WI (2004). 2. Fernández, F. G. A., Camacho, F. G., Pérez, J. A. S., Sevilla, J. M. F., and Grima, E. M.: A model for light distribution and average solar irradiance inside outdoor tubular photobioreactors for the microalgal mass culture. Biotechnol. Bioeng., 55, 701–714 (1997). 3. Katsuda, T., Arimoto, T., Igarashi, K., Azuma, M., Kato, J., Takakuwa, S., and Ooshima, H.: Light intensity distribution in the externally illuminated cylindrical photo-bioreactor and its application of to hydrogen production by Rhodobacter
VOL. 100, 2005
capsulatus. Biochem. Eng. J., 5, 157–164 (2000). 4. Lee, H. Y., Erickson, L. E., and Yang, S. S.: Kinetics and bioenergetics of light-limited photoautotrophic growth of Spirulina platensis. Biotechnol. Bioeng., 29, 832–843 (1987). 5. Suh, I. S. and Lee, S. B.: A light distribution model for an internally radiating photobioreactor. Biotechnol. Bioeng., 82, 180–189 (2003). 6. Katsuda, T., Yegani, R., Fujii, N., Igarashi, K., Yoshimura, S., and Katoh, S.: Effect of light intensity distribution on growth of Rhodobacter capsulatus. Biotechnol Prog., 20, 998– 1000 (2004). 7. Yegani, R., Yoshimura, S., Katsuda, T., and Katoh, S.: Semi-continuous cultivation of photosynthetic cells in flat plate photobioreactor. Iran. J. Chem. Eng. (2005). (in press) 8. Maness, P-C. and Weaver, P. F.: Hydrogen production from a carbon-monoxide oxidation pathway in Rubrivivax gelatinosus. Int. J. Hydrogen Energy, 27, 1407–1411 (2002). 9. Fales, L., Kryszak, L., and Zeilstra-Ryalls, J.: Control of hemA expression in Rhodobacter sphaeroides 2.4.1: effect of a transposon insertion in the hbdA gene. J. Bacteriol., 183, 1568–1576 (2001).
GROWTH STABILITY OF R. CAPSULATUS
677
10. Cornet, J.-F., Favier, L., and Dussap, C.-G.: Modeling stability of photohetrotrophic continuous cultures in photobioreactors. Biotechnol. Prog., 19, 1216–1227 (2003). 11. Weaver, P. F., Wall, J. D., and Gest, H.: Characterization of Rhodopseudomonas capsulata. Arch. Microbiol., 105, 207– 216 (1975). 12. Becker, E. W.: Microalgae: biotechnology and microbiology, p. 13. Cambridge University Press, Cambridge (1994). 13. Katsuda, T., Fujii, N., Takata, N., Ooshima, H., and Katoh, S.: Light attenuation in suspension of the purple bacterium Rhodobacter capsulatus and the green alga Chlamydomonas reinhardtii. J. Chem. Eng. Jpn., 35, 428–435 (2002). 14. Tsygankov, A. A., Laurinavichene, T. V., and Gogotov, I. N.: Laboratory scale photobioreactor. Biotechnol. Tech., 8, 575–578 (1994). 15. Ormerod, J. G., Ormerod, K. S., and Gest, H.: Light-dependent utilization of organic compounds and photoproduction of molecular hydrogen by photosynthetic bacteria; relationships with nitrogen metabolism. Arch. Biochem. Biophys., 94, 449–463 (1961).
© 2005, The Society for Biotechnology, Japan
Improvement of Growth Stability of Photosynthetic Bacterium Rhodobacter capsulatus Reza Yegani,1 Satoshi Yoshimura,1 Kazunori Moriya,1 Tomohisa Katsuda,1 and Shigeo Katoh1* Graduate School of Science and Technology, Kobe University, 1-1 Rokkodai, Nada-ku, Kobe 657-8501, Japan1 Received 15 June 2005/Accepted 12 September 2005
Using a semicontinuous culture method, in which operational parameters such as cell concentration and light intensity distribution were maintained almost constant, instability of the specific growth rate of Rhodobacter capsulatus B-100, a purple bacterium, was observed to be similar to that of R. capsulatus ST-410 when cultivated under high ratios of light intensity on the illuminated side to that of the transmitted light. Such instability was not observed in the cultivation of Chlorella vulgaris, a eukaryotic green alga, even at higher cell concentrations. Under the same conditions, the increase in only the ferrous concentration from 43 µM, the concentration in the original RCV medium, to 172 µM sustained a stable growth, whereas Fe2+ was slightly consumed during the cultivation. Supplemental illumination with a fluorescent lamp on the transmitted side of a flat plate photobioreactor sustained a moderate level of stable growth, while a halogen lamp slightly affected the growth stability. Our results showed that an increase in Fe2+ concentration or supplemental illumination improves the growth stability of R. capsulatus. [Key words: Rhodobacter capsulatus, growth stability, semicontinuous culture, light intensity distribution, iron concentration]
In this study, the effects of the concentrations of Fe2+ and Mg2+ on the growth stability of R. capsulatus, and supplemental illumination using different light sources emitting different spectra from the rear side (transmitted side) of a flat-plate photobioreactor, were investigated. The growth stability of the alga, Chlorella vulgaris, under similar illumination conditions as R. capsulatus was also examined.
Every microorganism must find in its environment all the substances necessary for energy generation and cellular biosynthesis (1). To increase the productivity and/or yield of metabolites when cultivating photosynthetic microorganisms, it is not only necessary to optimize cultivation conditions, particularly the general variables, such as temperature, pH, DO, components of culture media and transport properties, but also light dependent variables in photobioreactors, such as average light intensity, wavelength of illumination and light intensity distribution (2–5). In our previous work (6, 7), we have shown that the distribution of light intensity affected the growth stability of Rhodobacter capsulatus ST-410, one of the purple nonsulfur-photosynthetic bacteria, of which genera are widely used for the production of hydrogen, biopolymer and other secondary metabolites (8, 9). We have also shown that higher average light intensities alone does not cause a decrease in the specific growth rate, and that the ratio of light intensity on the illuminated surface to that of the transmitted light through a flat-plate photobioreactor might affect the specific growth rate stability. However, it is still an unknown phenomenon. From the engineering viewpoint, growth stability under optimal bioreactor configurations and cultivation conditions is crucial, particularly in continuous or cyclic cultivation systems (10). Thus, it is necessary to identify which parameters can cause growth instability and how this instability can be prevented.
MATERIALS AND METHODS Materials R. capsulatus ST-410 is a hydrogenase-deficient mutant derived from R. capsulatus B-100. R. capsulatus ST-410 and R. capsulatus B-100 were anaerobically cultured in RCV medium (pH 6.8) containing 7.5 mM (NH4)2SO4 and 30 mM D,L-malate as sole nitrogen and carbon sources, respectively, and the following basal salts: 43 µM FeSO4, 0.8 mM MgSO4, 0.5 mM CaCl2, 3 µM thiamine–HCl, 61 µM EDTA, 10 mM potassium phosphate buffer (pH 6.8) and trace elements (Mn, Borate, Cu, Zn, and Mo) (11). C. vulgaris was cultured in a medium (pH 6) containing 8 mM KNO3 and 8 mM NaNO3 as nitrogen sources and the following salts: 3 mM NaH2PO4 ⋅H2O, 1 mM MgSO4, 1 mM Na2HPO4 ⋅2H2O, 0.1 mM Ca(NO3)2 ⋅ 4H2O, 10 µM FeEDTA and micronutrients (Borate, Zn, Mn, Cu, and Mo) (12). The chemicals used were of reagent grade. Preculture and semi-continuous cultivation R. capsulatus was inoculated in a screw-capped test tube (25 ml) completely filled with the RCV culture medium and precultured by incubation in a glass-sided water bath at 30°C under illumination with a halogen lamp (90 W, JDR110V90WN/S-K12; Toshiba, Tokyo) with an incident light intensity of approximately 10 mW/cm2. At the logarithmic growth phase, the cells were harvested by centrifugation at
* Corresponding author. e-mail: [email protected] phone/fax: +81-(0)78-803-6193 672
GROWTH STABILITY OF R. CAPSULATUS
VOL. 100, 2005
673
FIG. 1. Experimental setup for semi-continuous cultivation of R. capsulatus.
10,000×g for 10 min and resuspended in a glass rectangular vessel (flat-plate photobioreactor) 9.8 cm high, 9.2 cm wide and 2.0 cm thick (130 ml working volume) containing the RCV medium (pH 6.8). C. vulgaris was inoculated in a 500-ml Erlenmeyer flask containing 50 ml RCV medium capped with a silicone sponge and incubated for 3 d under shaking at 30°C and illuminated using a florescent lamp at the incident light intensity of approximately 0.1 mW/cm2. During this procedure, carbon was supplied by penetration of air containing CO2 through the silicone sponge into the medium. After 3 d, cells were harvested by centrifugation at 5500×g for 15 min and resuspended in 50 ml of fresh RCV medium and precultivated in a 100-ml Erlenmeyer flask. The culture was incubated at 30°C under stirring with a magnet stirrer, and air containing 10% v/v CO2 was bubbled. At logarithmic growth phase, cells were again harvested by centrifugation at 5500×g for 15 min and resuspended in the glass rectangular vessel (height, 9.8 cm; width, 9.2 cm; thickness, 2.0 cm; working volume, 130 ml) containing fresh RCV medium. Figure 1 shows the experimental setup used for semi-continuous cultivation of R. capsulatus. The rectangular vessel was incubated in the glass-sided water bath at 30°C and illuminated on one side or both sides with a halogen lamp. In some cases, rear side illumination was carried out with one or two florescent lamps (FL 20SS, N18, 18 W; Matsushita Electric Industrial, Osaka). The incident light intensities in both cases were from 0.2 to 8.8 mW/cm2 for the halogen lamp and 0.1 to 0.2 mW/cm2 for the fluorescent lamp. The incident light intensities at 9 points on the illuminated side of the vessel were measured with a thermopile photosensor (3), and the averaged value was used as the incident light intensity on the illuminated side. In the semi-continuous cultivation method, a part of the culture broth containing grown cells was repeatedly replaced with fresh medium at a predetermined time interval to keep the cell concentration, the volume of the culture broth and the distribution of the light intensity almost constant to their initial values. On the basis of the specific growth rate of each strain, the predetermined time intervals were set at 40–60 min and 4–5 h for R. capsulatus and C. vulgaris, respectively. In the cultivation of R. capsulatus, before and after the replacement of medium, the culture was mixed by bubbling N2 sterilized through a membrane microfilter. Although the culture broth was mixed only by natural convection during the cultivation cycles,
sedimentation of cells was not observed. In the case of C. vulgaris, the culture broth was continuously mixed with a magnetic stirrer. To prevent the penetration of O2 into the culture broth of R. capsulatus, the reactor and the reservoir of the fresh medium were kept under N2 atmosphere. Cultivation of C. vulgaris was carried out aerobically. Measurements of dry cell weight and ion concentration The dry cell weights of R. capsulatus ST-410 and B-100, and C. vulgaris in the culture broth were determined by drying the cells at 80°C for 50 h after centrifugation and thorough washing of the cells with distilled water; the weights were correlated to the absorbance measured with a UV spectrophotometer (UV-1600; Shimadzu, Kyoto) at 660 and 560 nm, respectively. At the end of each cultivation cycle, the cell concentration was determined from the absorbance of the culture broth. The concentration of Fe2+ was measured using Fe C-Test Wako (Wako Pure Chemical Industries, Osaka). This assay method is based on the selective chelation of Fe2+ into Nitroso-PASP (2-Nitroso-5-[N-n-propyl-N-(3-sulfopropyl)amino]phenol). Because EDTA in the RCV medium competed for chelation and lowered the measured concentration of Fe2+, we inactivated the EDTA with hydrogen peroxide before the assay. Hydrogen peroxide (30 wt%, 0.1 ml) was added into 0.4 ml of the supernatant of the culture broth obtained by centrifugation, and this mixture was incubated for 10 min at 80°C. This treatment also oxidized Fe2+ in the mixture to Fe3+. Therefore, the concentration of iron was determined using the Fe C-Test Wako after the reduction of Fe3+ to Fe2+ by 0.45 M thioglycolic acid in 0.4 M acetate buffer. Measurement of chlorophyll content of cells Chlorophyll is a colored material that absorbs light and uses its energy to synthesize carbohydrates from CO2 and water. The absorption spectrum of R. capsulatus consists of two main peaks at 800 and 850 nm, which correspond to chlorophyll a and b, respectively. The concentration of chlorophyll may reflect cell physiological condition under illuminated conditions. During the semicontinuous cultivation of R. capsulatus, the absorbance of chlorophyll at 800 nm was measured using the UVspectrophotometer and the ratio of this absorbance to the cell concentration was considered as the chlorophyll content per unit cell. Calculation of light intensity distribution The light intensity distribution in the photobioreactor was calculated from the equation given in our previous work (13), assuming that the attenuation of the light intensity of polychromatic light is given as a
674
YEGANI ET AL.
FIG. 2. Time courses of cell concentration in semicontinuous cultivation of R. capsulatus ST-410. Solid circle: I 0 = 8.8 mW/cm2, C = 0.1 mg/ml, Iave = 4.8 mW/cm2, I0/It = 3.6. Solid triangle: I0 = 8.8 mW/cm2, C = 0.55 mg/ml, Iave = 2.0 mW/cm2, I0/It = 40. Open triangle: I0 = 2.1 mW/cm2, C = 0.55 mg/ml, Iave = 0.47 mW/cm2, I0/It = 40.
J. BIOSCI. BIOENG.,
FIG. 3. Specific growth rates of R. capsulatus ST-410 (solid circle and triangle) and B-100 (open circle and triangle). Circle: I0=8.8 mW/cm2, C = 0.1 mg/ml, Iave = 4.8 mW/cm2, I0/It = 3.6. Triangle: I0 = 8.8 mW/cm2, C = 0.55 mg/ml, Iave = 2.0 mW/cm2, I0/It = 40.
summation of the attenuation of each wavelength. Our previous results showed that the predicted distributions showed good agreement with the measured distributions. The average light intensity was calculated by dividing an integrated value of the light intensity along the light path by the light path length of the photobioreactor similar to the method previously reported.
RESULTS AND DISCUSSION Semi-continuous cultivation of other species and family In the semi-continuous cultivation, the specific growth rate was calculated by dividing the increase in the cell concentration by the initial cell concentration and time during one cycle. Figure 2 shows the time courses of the cell concentration in semicontinuous cultivation at various cell concentrations (C, mg/ml), incident light intensities (I0, mW/cm2) and average light intensities (Iave, mW/cm2). The cell concentration at the beginning of each cycle was kept almost constant. When cells grew at a constant specific growth rate (solid circles and open triangles in Fig. 2), the slope of each increment was also constant, as shown in Fig. 2. In our previous papers (6, 7), we have shown the growth instability of R. capsulatus ST-410 under high ratios of light intensity on the illuminated side (I0) to that on the transmitted side (It) caused by high cell concentrations and/or long light paths. We tried to clarify whether this growth instability caused by different illumination conditions is a common phenomenon in photosynthetic microorganisms and also to determine suitable cultivation conditions to prevent this instability. At first, experiments with another Rhodobacter species, R. capsulatus B-100, and also with a eukaryotic green alga C. vulgaris were carried out. The results of cultivation of R. capsulatus B-100 and C. vulgaris are shown in Figs. 3 and 4, respectively. As indicated by the circle in Fig. 3, at a low cell concentration (C = 0.1 mg/ml) both R. capsulatus ST-410 and B-100 showed stable specific growth rates for 16 h. At a cell concentration of 0.55 mg/ml, R. capsulatus B-100 (open triangle in Fig. 3) showed unstable specific growth rate similar to R. capsulatus ST-410 (solid triangle in Fig. 3) under the same incident light intensity (I0 = 8.8
FIG. 4. Specific growth rate of C. vulgaris. Open triangle: I 0 = 8.8 mW/cm2, C = 0.55 mg/ml, Iave = 3.2 mW/cm2, I0/It = 10. Open square: I 0 = 8.8 mW/cm2, C= 1.1 mg/ml, Iave = 2.0 mW/cm2, I0/It = 30. Open diamond: I 0 =8.8 mW/cm2, C= 2.2 mg/ml, Iave = 1.2 mW/cm2, I0/It = 79.
mW/cm2). However, in the cultivation of C. vulgaris, instability of its specific growth rate was not observed under the same conditions. After 100 h cultivation at the cell concentration of 1.1 mg/ml, the cell concentration was increased to 2.2 mg/ml, and the growth was still stable at a different specific growth rate. Effect of media compositions Although the RCV medium has already been optimized for the cultivation of R. capsulatus particularly for long-term cultivation (14, 15), the instability of its specific growth rate might be caused by deficiency of some components in the medium. Because fresh medium was repeatedly added in the case of semicontinuous cultivation at a predetermined time interval, nutrient deficiency seldom occurred. To confirm this hypothsis, the effects of different concentrations of a key metal ion Fe2+ in RCV medium, which is the central atom in the proto-porphyrin called heme, were examined. Figure 5 shows the results of cultivation of R. capsulatus ST-410 at different Fe2+ concentrations in the RCV medium. By increasing ferrous ion concentration from the original 43 µM to 86 µM, the slope of decrease in the specific growth rate became less steep. At 172 µM (4 times), the specific
VOL. 100, 2005
FIG. 5. Effects of Fe2+ concentration on stability of specific growth rate. I 0 = 8.8 mW/cm2, C = 0.55 mg/ml, Fe2+ = 172 µM (open diamond); 86 µM (solid diamond); 43 µM (solid triangle); 22 µM (solid square).
FIG. 6. Concentration of Fe2+ during cultivation under unstable condition. I 0 =8.8 mW/cm2, C=0.55 mg/ml, Fe2+ =43 µM, Iave =2.0 mW/cm2, I0/It = 40.
growth rate was kept almost constant. Similar growth stability was observed with the addition of 129 µM ferric ion to the original RCV medium (data not shown). When the ferrous ion was 22 µM, the specific growth rate was decreased after 5–6 h similar to the original concentration of ferrous ion in the RCV medium. We speculated that the lack of Fe2+ might have caused the instability of specific growth rate of R. capsulatus. However, as shown in Fig. 6, the Fe2+ concentration under a typical unstable cultivation condition (I0 = 8.8 mW/cm2, C = 0.55 mg/ml, initial Fe2+ = 43 µM) remained almost constant, although the measured values were slightly higher than that calculated from an added amount. Thus, Fe2+ was slightly consumed by cells. The increase in only the Fe2+ concentration of the RCV medium could maintain growth stability. This finding indicates that the decrease in the specific growth rate was not caused by deficiency of any nutrients including Fe2+ in the RCV medium. The increase in the Fe2+ concentration enhanced the transport of Fe2+ into cells, and thus enhanced the availability of ferrous ions that might have induced a change in the physiological conditions in cells to sustain stable growth. The twofold increase in the concentration of Mg2+, which is contained in the chlorophyll molecule, did
GROWTH STABILITY OF R. CAPSULATUS
675
FIG. 7. Effects of Fe2+ addition on specific growth rate during cultivation under unstable condition. Solid triangle: 0–13 h: I 0 =8.8 mW/cm2, C= 0.55 mg/ml, Fe2+ = 43 µM. Open diamond: 13 h~: I 0 = 8.8 mW/cm2, C= 0.55 mg/ml, Fe2+ = 172 µM. Broken line: calculated specific growth rate.
not affect the growth stability (data not shown). Because an increase in Fe2+ and not in Mg2+ changed the growth stability of cells, electron transfer systems that are involved in photosynthesis, such as cytochromes and ferredoxin, might be affected by a higher concentration of Fe2+. Recovery of specific growth rate After 13 h cultivation under an unstable condition (I0 = 8.8 mW/cm2, C = 0.55 mg/ml), the concentration of Fe2+ in the culture broth and added fresh medium was increased 4 times, and the specific growth rate of R. capsulatus ST-410 was measured. The specific growth rate increased gradually after approximately 2 h of lag phase, as indicated by the open diamonds in Fig. 7. The calculated specific growth rate (broken line) was obtained with the assumption that the fraction of viable cells was given by the ratio of the specific growth rate at a particular time to the original specific growth rate under the initial conditions, that is, 0.27 1/h, and that the viable cells could grow at the specific growth rate of 0.27 1/h after the lag phase. The observed recovery of the original specific growth rate was attained at a shorter time than that of the calculated result. This means that the decrease in the specific growth rate was not only caused by the death of the cells, but also caused mainly by the change in and depression of metabolism of viable cells. Thus, the increase in the Fe2+ concentration induced a change in physiological conditions of the cells leading to the recover of the original specific growth rate at a faster rate than the calculated specific growth rate. Supplemental illumination from opposite side In our previous work (6, 7) we have shown that the light intensity distribution, that is, the ratio of the light intensity on the illuminated side of the cultivation vessel to that of the transmitted light, seemed to be one of the factors causing growth instability. Therefore, supplemental illumination on the opposite side of the vessel would decrease this ratio and prevent growth instability caused by the light intensity distribution. The vessel was illuminated on the opposite side of the main halogen lamp with another halogen lamp at a light intensity of 0.2 or 0.4 mW/cm2 or with a fluorescent lamp(s) at a light intensity of 0.1 or 0.2 mW/cm2.
676
J. BIOSCI. BIOENG.,
YEGANI ET AL.
FIG. 8. Effects of supplemental illumination with halogen lamp. Open circles: I0,f = 1.8 mW/cm2, I0,t = 0.4 mW/cm2, C = 0.55 mg/ml, Iave = 0.75 mW/cm2, I0/It = 4.1. Solid squares: I0,f = 8.8 mW/cm2, I0,t = 0.2 mW/cm2, C = 0.55 mg/ml, Iave = 2 mW/cm2, I0/It =22.
FIG. 9. Effects of supplemental illumination with fluorescent lamp(s). Open diamond: I0,f = 8.8 mW/cm2, I0,t = 0.1 mW/cm2, C = 0.55 mg/ml, Iave = 2 mW/cm2, I0/It = 28 (one fluorescent lamp). Solid diamond: I0,f = 8.8 mW/cm2, I0,t = 0.2 mW/cm2, C = 0.55 mg/ml, Iave = 2 mW/cm2, I0/It = 22 (two fluorescent lamps). Solid triangle: I 0 = 8.8 mW/cm2, C = 0.55 mg/ml, Iave = 2.0 mW/cm2, I0/It = 40.
Figure 8 shows the changes in the specific growth rate when the vessel was illuminated from the both sides with halogen lamps. At a low incident light intensity of 1.8 mW/cm2 of the main lamp, the growth rate was kept almost constant. At the light intensity of the main-illuminated side was increased to 8.8 mW/cm2, the specific growth rate again decreased after about 5 h cultivation, and reached another stable specific growth rate, that is, 0.15 1/h, which is higher than that under one side illumination. Under supplemental illumination with a fluorescent lamp, as shown in Fig. 9, a moderately stable specific growth rate, that is, 0.23 1/h, was sustained. By illumination with two fluorescent lamps, however, the sustained specific growth rate decreased slightly. The effective light intensity supplied by one fluorescent lamp was about 0.1 mW/cm2. The main difference in the effectiveness of the halogen lamp and the fluorescent lamp might be caused by the differences in spectra of their emitting light wavelengths. The wavelengths emitted from the halogen lamp were longer (500–800 nm) than those emitted
FIG. 10. Changes in chlorophyll content under various cultivation conditions. Open circle: I 0 = 8.8 mW/cm2, C= 0.1 mg/ml, Fe2+ = 43 µM, Iave = 4.8 mW/cm2, I0/It = 3.6. Open diamond: I 0 = 8.8 mW/cm2, C = 0.55 mg/ml, Fe2+ = 172 µM, Iave = 2 mW/cm2, I0/It = 40. Asterisk: I0 = 8.8 mW/cm2, I0,t = 0.1 mW/cm2, C = 0.55 mg/ml, Fe2+ = 43 µM, Iave = 2 mW/cm2, I0/It = 28 (one fluorescent lamp on transmitted side). Solid square: I0 = 8.8 mW/cm2, I0,t = 0.2 mW/cm2, C = 0.55 mg/ml, Fe2+ = 43 µM, Iave = 2 mW/cm2, I0/It = 22 (hologen lamp from transmitted side). Solid triangle: I 0 = 8.8 mW/cm2, C= 0.55 mg/ml, Fe2+ =43 µM, Iave = 2.0 mW/cm2, I0/It = 40.
from a florescent lamp (400–650 nm). Light with shorter wavelengths might be effective in sustaining a stable specific growth rate in long-term cultivation. Relationship between ratio of chlorophyll concentration and growth instability During the semicontinuous cultivation of R. capsulatus, the absorbance of chlorophyll at 800 nm was measured using a UV-spectrophotometer. The ratio of the absorbance to the cell concentration was considered as the chlorophyll content per unit cell. This chlorophyll content in the semicontinuous cultivation is plotted against the cultivation time as shown in Fig. 10. We examined the chlorophyll content in the cells under the cultivation conditions tested. When 172 µM Fe2+ was added (0.27 1/h in Fig. 5), the decrease in the chlorophyll content was less than 5% after 15 h, while more than 10–15% decrease was observed under the unstable cultivation condition at 43 µM Fe2+. Interestingly, although the growth was sustained at a moderately stable level using one fluorescent lamp (0.23 1/h in Fig. 9), the chlorophyll content was lower than that in the cultivation condition with a high Fe2+ condition. These results suggest that chlorophyll content might be used as a criterion for predicting the growth stability of R. capsulatus during long-term cultivation. REFERENCES 1. Todar, K.: Todar’s online textbook of bacteriology. University of Wisconsin, Madison, WI (2004). 2. Fernández, F. G. A., Camacho, F. G., Pérez, J. A. S., Sevilla, J. M. F., and Grima, E. M.: A model for light distribution and average solar irradiance inside outdoor tubular photobioreactors for the microalgal mass culture. Biotechnol. Bioeng., 55, 701–714 (1997). 3. Katsuda, T., Arimoto, T., Igarashi, K., Azuma, M., Kato, J., Takakuwa, S., and Ooshima, H.: Light intensity distribution in the externally illuminated cylindrical photo-bioreactor and its application of to hydrogen production by Rhodobacter
VOL. 100, 2005
capsulatus. Biochem. Eng. J., 5, 157–164 (2000). 4. Lee, H. Y., Erickson, L. E., and Yang, S. S.: Kinetics and bioenergetics of light-limited photoautotrophic growth of Spirulina platensis. Biotechnol. Bioeng., 29, 832–843 (1987). 5. Suh, I. S. and Lee, S. B.: A light distribution model for an internally radiating photobioreactor. Biotechnol. Bioeng., 82, 180–189 (2003). 6. Katsuda, T., Yegani, R., Fujii, N., Igarashi, K., Yoshimura, S., and Katoh, S.: Effect of light intensity distribution on growth of Rhodobacter capsulatus. Biotechnol Prog., 20, 998– 1000 (2004). 7. Yegani, R., Yoshimura, S., Katsuda, T., and Katoh, S.: Semi-continuous cultivation of photosynthetic cells in flat plate photobioreactor. Iran. J. Chem. Eng. (2005). (in press) 8. Maness, P-C. and Weaver, P. F.: Hydrogen production from a carbon-monoxide oxidation pathway in Rubrivivax gelatinosus. Int. J. Hydrogen Energy, 27, 1407–1411 (2002). 9. Fales, L., Kryszak, L., and Zeilstra-Ryalls, J.: Control of hemA expression in Rhodobacter sphaeroides 2.4.1: effect of a transposon insertion in the hbdA gene. J. Bacteriol., 183, 1568–1576 (2001).
GROWTH STABILITY OF R. CAPSULATUS
677
10. Cornet, J.-F., Favier, L., and Dussap, C.-G.: Modeling stability of photohetrotrophic continuous cultures in photobioreactors. Biotechnol. Prog., 19, 1216–1227 (2003). 11. Weaver, P. F., Wall, J. D., and Gest, H.: Characterization of Rhodopseudomonas capsulata. Arch. Microbiol., 105, 207– 216 (1975). 12. Becker, E. W.: Microalgae: biotechnology and microbiology, p. 13. Cambridge University Press, Cambridge (1994). 13. Katsuda, T., Fujii, N., Takata, N., Ooshima, H., and Katoh, S.: Light attenuation in suspension of the purple bacterium Rhodobacter capsulatus and the green alga Chlamydomonas reinhardtii. J. Chem. Eng. Jpn., 35, 428–435 (2002). 14. Tsygankov, A. A., Laurinavichene, T. V., and Gogotov, I. N.: Laboratory scale photobioreactor. Biotechnol. Tech., 8, 575–578 (1994). 15. Ormerod, J. G., Ormerod, K. S., and Gest, H.: Light-dependent utilization of organic compounds and photoproduction of molecular hydrogen by photosynthetic bacteria; relationships with nitrogen metabolism. Arch. Biochem. Biophys., 94, 449–463 (1961).