Photosynthetic Apparatus in Roseateles depolymerans 61A Is ...

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Apr. 2002, p. 1665–1673 0099-2240/02/$04.00⫹0 DOI: 10.1128/AEM.68.4.1665–1673.2002 Copyright © 2002, American Society for Microbiology. All Rights Reserved.

Vol. 68, No. 4

Photosynthetic Apparatus in Roseateles depolymerans 61A Is Transcriptionally Induced by Carbon Limitation Tetsushi Suyama,1* Toru Shigematsu,1† Toshihiko Suzuki,1 Yutaka Tokiwa,1 Takahiro Kanagawa,1 Kenji V. P. Nagashima,2 and Satoshi Hanada1 National Institute of Advanced Industrial Science and Technology, Tsukuba Central 6, Tsukuba, Ibaraki 305-8566,1 and Department of Biology, Tokyo Metropolitan University, Hachioji, Tokyo 192-0397,2 Japan Received 24 September 2001/Accepted 8 January 2002

Production of a photosynthetic apparatus in Roseateles depolymerans 61A, a recently discovered freshwater ␤-Proteobacterium showing characteristics of aerobic phototrophic bacteria, was observed when the cells were subjected to a sudden decrease in carbon sources (e.g., when cells grown with 0.1 to 0.4% Casamino Acids were diluted or transferred into medium containing 0.2% O2), and was reduced in the presence of light. Transcription of the R. depolymerans puf operon is considered to be controlled by changes in carbon nutrients in addition to oxygen tension and light intensity. While these observations have been reported for the aerobic phototrophic bacteria belonging to the ␣-Proteobacteria, the physiological significance of a photosynthetic apparatus in the obligatory aerobic ␤-Proteobacterium producing bacteriochlorophyll (BChl) a, i.e., R. depolymerans, has not been reported at all. The photosynthetic apparatus of purple bacteria is composed of a reaction center (RC) complex and light-harvesting (LH) complexes. Two kinds of LH complexes (LH1 and LH2) are present in many species. However, the LH2 complex is absent in some species, e.g., Blastochloris viridis and Rhodospirillum rubrum. R. depolymerans has been shown to have RC and LH1 complexes containing BChl a, but it may lack the LH2 complex (34). The subunits of the LH1 complex and most of the subunits of the RC complex are encoded by a gene cluster called the puf operon. The structures of the puf operon and the puf gene products in aerobic phototrophic bacteria have been studied well only for R. denitrificans (20) and Acidiphilium rubrum (23). It has been reported that the photosynthesis genes in these species are quite similar to those in the other phototrophic species. Expression of the puf operon is known to be controlled by oxygen tension and light intensity in some species of photosynthetic bacteria (4, 24, 25, 43). Little is understood regarding other stimuli controlling production of the photosynthetic apparatus in either phototrophic or aerobic phototrophic bacteria. Moreover, no direct proof of photosynthesis gene regulation at the level of mRNA has been reported in regard to environmental factors besides oxygen and light. However, it has been found that accumulation of BChl a and carotenoid pigments in R. depolymerans cells is largely dependent on the

Purple phototrophic bacteria are taxonomically affiliated with the ␣, ␤, and ␥ subclasses of the class Proteobacteria. Many of the species grow both heterotrophically under aerobic-dark conditions and phototrophically under anaerobic- light conditions. The production of a photosynthetic apparatus in most of these phototrophic species is reduced under aerobic conditions (18). On the other hand, some species, the so-called aerobic phototrophic bacteria (e.g., Erythrobacter longus and Roseobacter denitrificans), do not grow phototrophically under anaerobic-light conditions, whereas they produce a photosynthetic apparatus under aerobic conditions (31, 41). Aerobic phototrophic bacteria had been reported only for the ␣ subclass of the Proteobacteria (31, 41) until the discovery of Roseateles depolymerans, which belongs to the ␤ subclass of the Proteobacteria (34). Although the presence of photochemical activity has been clarified in some species of aerobic phototrophic bacteria, they do not grow on light as a sole energy source (14, 31, 36, 38). However, stimulation of growth in the presence of light has been reported for R. denitrificans, Erythromicrobium hydrolyticum, and Acidisphaera rubrifaciens (13, 15, 42). It has also been suggested for R. denitrificans, Bradyrhizobium strain BTAi 1, and A. rubrifaciens that the preservation of viability is improved by the presence of light in the absence of nutrients (9, 15, 29). * Corresponding author. Mailing address: National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba Central 6, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8566, Japan. Phone: 81-298-616026. Fax: 81-298-61-6587. E-mail: [email protected]. † Present address: Department of Applied Chemistry and Biochemistry, Faculty of Engineering, Kumamoto University, Kumamoto 8608555, Japan. 1665

1666

SUYAMA ET AL.

culture conditions; i.e., a high concentration of carbon sources (e.g., skim milk, nutrient broth, or Casamino Acids) reduces pigmentation of this organism (34). In addition, culturing cells on solid media in the absence of light is known to be favorable for producing photosynthetic pigments in this organism (34). Therefore, it is possible that this organism controls production of a photosynthetic apparatus responding to stimuli other than oxygen and light, particularly to carbon nutrient conditions. In the present study, the physiological significance of the photosynthetic apparatus, as well as the factors controlling the production of a photosynthetic apparatus and the expression of the puf operon, was investigated for the aerobic phototrophic ␤-Proteobacterium R. depolymerans 61A. MATERIALS AND METHODS Bacterial strain. R. depolymerans 61A (DSM 11813) was maintained on PHC agar plates supplemented with 0.08% (wt/vol) nutrient broth (Difco Laboratories, Detroit, Mich.) at 30°C in the dark (35). Media. C-free medium (carbon source-free medium) contained (per liter) 0.5 g of (NH4)2SO4, 1 ml of vitamin mixture, 10 ml of basal salt solution, and 25 ml of 200 mM potassium phosphate buffer (pH 7.0). The compositions of the basal salt solution and the vitamin mixture were previously described (12). A series of media containing different concentrations of Casamino Acids (0.4, 0.2, 0.1, 0.04, 0.02, and 0.01%, wt/vol), glucose (0.4 and 0.1%, wt/vol), and sodium succinate (0.4 and 0.1%, wt/vol) were prepared by adding each carbon source to the C-free medium. Preculture conditions. R. depolymerans cells were precultured in 250 ml of medium containing 0.4, 0.2, or 0.1% carbon sources as specified for each experiment. The culture was placed in a 1-liter Erlenmeyer flask with shaking at 200 rpm at 30°C in the dark. Growth was monitored by measuring the optical density at 660 nm (OD660) of the culture medium. Impact of dilution. R. depolymerans cells were precultured in medium containing 0.4% Casamino Acids, glucose, or sodium succinate. Cell suspensions at mid-log phase (OD660 of ⬇0.2 to 0.3) were diluted into 10 volumes of C-free medium, sterile water, or sterile water osmoequilibrated with glycerol. The diluted suspensions were incubated for three more days, and then the cells were collected by centrifugation for spectroscopic analysis. Replacement of medium. Experiments with replacement of medium were conducted as follows unless modifications are specified. R. depolymerans cells were precultured in medium containing 0.2% Casamino Acids. The cultured cells at mid-log phase (OD660 of ⬇0.35) were harvested by centrifugation, washed twice with 50 ml of C-free medium, and then suspended into fresh medium (containing different concentrations of substrates) for each experiment. The OD660 of each cell suspension was adjusted to 0.35. The cells in the fresh media were incubated under the conditions described below and then collected for spectroscopic analysis. Incubation under aerobic conditions. Ten-milliliter portions of the cell suspensions were placed in L-shaped tubes (diameter, 18 mm by 120 by 70 mm) and incubated at 30°C with shaking at 45 strokes per min in a Personal-11SD water bath shaker equipped with an MD-1218 Monod-type shaking platform (TAITEC, Koshigaya, Japan). Incubation under semiaerobic conditions. In order to minimize O2 consumption under incubation, C-free medium was used. Ten-milliliter portions of the cell suspensions were placed in 123-ml rubber-capped bottles, and at 24-h intervals the headspaces were replaced with 100, 10, 1, or 0.1% (vol/vol) air (in N2 gas at 1 atm), which correspond to 20, 2, 0.2, or 0.02% oxygen, respectively. The suspensions were incubated with shaking at 200 rpm on a rotary shaker at 30°C. Incubation in the presence of light. The effect of light on the cell suspensions was determined under continuous illumination with a 100-W krypton lamp (KR100/110V100W; Toshiba, Tokyo, Japan) over a distance of 50 cm with 2 to 3 cm of water layer (10 W m⫺2, under conditions without shading). The intensity of illumination was adjusted by shading with aluminum foil. The irradiance at the position of the samples was determined with a model 4090 radiometer (Springfield Jarco Instruments, Yellow Springs, Ohio). Spectroscopic analysis. Cells incubated under the various conditions were harvested by centrifugation. Pigments were extracted from the cells with methanol. Absorption spectra were recorded with a DU640 spectrophotometer (Beckman Instruments, Fullerton, Calif.). The molar absorption coefficient for

APPL. ENVIRON. MICROBIOL. BChl a in methanol, 54.9 mM⫺1 cm⫺1 at 770 nm, was used for quantification. Spectral data were normalized to the dry weight of cells. Effects of Casamino Acids concentration on growth rate. R. depolymerans cells were precultured in medium containing 0.2% Casamino Acids, and the medium was replaced with a series of media containing different concentrations of Casamino Acids (0.4 to 0%). The cell density of the suspensions was adjusted to an OD660 of ⬇0.01. The cell suspensions (10 ml) were placed in L-shaped tubes and were incubated aerobically as described above in the presence and absence of light (10 W m⫺2). Growth was monitored by measuring the OD660. Determination of viability of starved cells. R. depolymerans cells were precultured in medium containing 0.2% Casamino Acids, and a cell suspension (adjusted to an OD660 of ⬇0.01) in C-free medium was prepared. Ten-milliliter portions were placed in L-shaped tubes (tubes 1, 2, 3, and 4). Tubes 1 and 2 were incubated in the dark, and tubes 3 and 4 were incubated in the light (10 W m⫺2), aerobically, as described above. One-hundred-microliter portions of the suspension were collected from the tubes at 4-day intervals, diluted with C-free medium, spread onto plates of CAV agar medium (34), and incubated at 30°C in the dark. The number of colonies formed after 3 days of incubation was counted. Cloning of the puf gene cluster. The probe for detecting the puf gene cluster (probe LM in Fig. 4) was prepared by PCR with the PUFLMF-PUFLMR primers (23) and genomic DNA extract from R. depolymerans 61A (33) as a template. An 8.1-kbp BamHI fragment containing the puf gene cluster was cloned into a ZAP Express vector (Stratagene, La Jolla, Calif.) and sequenced by methods previously reported (30). Sequence analysis. DNA and deduced amino acid sequences were analyzed using a GENETYX-MAC (version 9.0) software package (Software Development Co. Ltd., Tokyo, Japan). A search for homologous proteins was conducted with the BLAST algorithm (3). Multiple alignments and distance matrices were constructed with the CLUSTAL W1.6 program (37). Analysis of mRNA. The cells were precultured, incubated for 72 h, and harvested as described for the spectroscopic analysis. Cell samples were fixed with chloroform-methanol-diethyl pyrocarbonate (2:1:0.03, vol/vol/vol) and stored at ⫺80°C. Total RNA was extracted from the cell samples by the hot-phenol method (32). Northern hybridization experiments were performed with the standard protocols of the DIG DNA Labeling Kit, DIG Easy Hyb, and DIG Luminescent Detection Kit (Boehringer GmbH, Mannheim, Germany). Specific probes for the R. depolymerans puf operon and for neighboring regions (probes BA, C, D, and Z in Fig. 4) were prepared by PCR amplification from the R. depolymerans puf clone (pRB3 plasmid) as the template. Accumulation of the puf transcript in total RNA was measured by hybridization against the specific probe (probe BA). Densitometric analysis was performed using the NIH Image 1.59/ ppc program (National Institutes of Health; http://rsb.info.nih.gov/nih-image/). The 5⬘ ends of R. depolymerans puf transcripts were determined by primer extension analysis using the ABI 377 automated DNA sequencer and GeneScan program (Perkin-Elmer Applied Biosystems). Reverse transcription was performed with RAV-2 reverse transcriptase (Takara) and the 5⬘-labeled primer 5⬘-[6FAM]-CAGAGCCACGGACGCCACATCC (kindly provided by PerkinElmer Applied Biosystems), corresponding to the position at the 5⬘ end of probe BA. A sequence ladder for the 5⬘ end was generated by cycle sequence reaction with the 5⬘-labeled primer in the presence of dideoxynucleoside triphosphates. Nucleotide sequence accession number. The complete sequence of the puf gene cluster of R. depolymerans 61A has been deposited in the DDBJ/GenBank/ EMBL data libraries under accession number AB028938.

RESULTS Impact of dilution. When R. depolymerans cells were continuously grown in liquid medium containing 0.4% Casamino Acids, glucose, or sodium succinate, the cells were hardly pigmented during the incubation. The changes in pH in the culture media were negligible over the course of culturing. When the cells grown in these media were diluted with C-free medium, however, the cells showed accumulation of BChl a. Accumulation of BChl a was also observed when the culture was diluted with sterile water or water osmoequilibrated with glycerol (this organism does not utilize glycerol [34]). Therefore, this phenomenon is not a response to osmotic stress but may be a response to a sudden decrease in carbon sources to be utilized.

STARVATION-INDUCIBLE puf OPERON OF ROSEATELES

VOL. 68, 2002

1667

TABLE 1. Effect of changes in concentrations of carbon sourcesa Carbon source concn (%) for incubation

0.4 0.1 0.0

BChl a (nmol/g [dry wt] of cells) with preculture with the following carbon source: Casamino Acids 0.4% b

ND ND 270.4

Glucose

Na-succinate

0.1%

0.4%

0.1%

0.4%

0.1%

ND ND 248.0

ND 24.4 159.4

ND ND 103.8

ND ND 239.6

ND ND 190.5

a Cells were precultured in medium containing 0.4 or 0.1% carbon sources as indicated. Cells at the exponentially growth phase (OD660 of ⬇0.1 to 0.2) were harvested and suspended (OD660 of ⬇0.2) in medium containing 0.4, 0.1 or 0% of the same carbon sources. The cells were incubated aerobically in the dark for three more days, and then BChl a was extracted for spectroscopic analysis. b ND, the absorption peak of BChl a was not detected.

Replacement of medium. As shown in Table 1, all of the R. depolymerans cells grown in 0.1 to 0.4% Casamino Acids, glucose, or sodium succinate became pigmented when the culture medium was replaced with fresh C-free medium. Remarkably, the cells precultured in 0.4% glucose also showed accumulation of BChl a under incubation with 0.1% glucose, whereas the cells precultured in 0.1% glucose did not show pigmentation under incubation in the same medium. No BChl a accumulation was observed after replacement with medium containing concentrations of carbon sources the same as or higher than that in the preculturing. Detailed examination of the conditions for producing BChl a, based on the medium replacement experiment using Casamino Acids as a carbon source, was conducted in the following experiments. Time course of pigmentation under aerobic-dark conditions. Figure 1 shows the changes in turbidity of cell suspension and the accumulation of BChl a in R. depolymerans cells after replacement of medium with media containing 0.2, 0.02, and 0% Casamino Acids. The increase in OD660 was suspended in cells incubated with 0.02 and 0% Casamino Acids, while cells

FIG. 1. Time courses of OD660 (a) and BChl a content (b) in R. depolymerans cells incubated aerobically in the dark in medium containing 0.2, 0.02, and 0% Casamino Acids after medium replacement. Cells were precultured (OD660 of ⬇0.35) in medium containing 0.2% Casamino Acids and resuspended (OD660 of ⬇0.35) into each medium.

incubated with 0.2% Casamino Acids showed a continued increase in the OD660 at the beginning of incubation (Fig. 1a). The OD660 values under each condition showed a steady decay after 12 h of incubation (Fig. 1a). The cells incubated with 0.02 and 0% Casamino Acids showed accumulation of BChl a after 36 h of incubation (Fig. 1b). At the beginning of the incubation, BChl a was not detected, and the cells were nonpigmented during 24 h of incubation (Fig. 1b). Accumulation of BChl a was observed to be time dependent and was observed only upon the decrease in OD660 (Fig. 1). The cells incubated with 0.2% Casamino Acids, however, stayed nonpigmented even after the period of decrease in OD660, and no accumulation of BChl a was detected under this condition over a week (Fig. 1). Production of BChl a in R. depolymerans was inversely proportional to the concentration of Casamino Acids in the medium for replacement (Fig. 1b). As shown in Fig. 1b, the presence and absence of the accumulation of BChl a were clearly distinguishable at 72 h after replacement of the media. An incubation period of 72 h was chosen for the following experiments to compare the accumulation of pigments under different conditions. Effects of Casamino Acids concentration on pigmentation under aerobic-dark and -light conditions. The relationship between the Casamino Acids concentrations in the medium for replacement (0.2 to 0.01%) and the accumulation of BChl a was determined in detail. Figure 2a shows the absorption spectra of methanol extracts from R. depolymerans cells incubated for 72 h in the dark with different concentrations of Casamino Acids after medium replacement. It was found that the extents of the absorption bands corresponding to carotenoids (composed mainly of spirilloxanthin, with absorption peaks at 471, 496, and 529 nm [34]) and BChl a (with an absorption peak at 770 nm) were enhanced by incubation of the cells with not more than 0.04% Casamino Acids (Fig. 2b). The absorption peak at 460 nm that was enhanced by incubation with 0.02 to 0.04% Casamino Acids (indicated by an arrowhead in Fig. 2a) originated from a hydrophilic yellow pigment which was soluble in either water or methanol and may not be associated with the photosynthetic apparatus. Accumulation of BChl a was reduced in the presence of light illumination (Fig. 2b [spectra not shown]). Also in the presence of light, accumulation of BChl a was observed in cells incubated with 0.04% or lower concentrations of Casamino Acids (Fig. 2b). Production of BChl a was observed under incubation with ⱕ0.04% Casamino Acids even when the cells were precultured in medium containing higher concentrations of Casamino Acids.

1668

SUYAMA ET AL.

APPL. ENVIRON. MICROBIOL.

FIG. 2. Production of photosynthetic pigments in R. depolymerans cells incubated under various conditions after medium replacement. (a, c, and e) Absorption spectra of methanol extracts from cells incubated for 72 h aerobically in the dark in medium containing 0.2, 0.1, 0.04, 0.02, and 0.01% Casamino Acids (a); from cells incubated for 72 h in C-free medium in the dark in the presence of 20, 2, 0.2, and 0.02% oxygen (c); and from cells incubated for 72 h aerobically in C-free medium under light illumination (10, 5, 2.5, and 0 W m⫺2). (b, d, and f) BChl a contents per gram (dry weight) of cells, corresponding to panels a, c, and e, respectively. The intensities of spectra in panels a, c, and e were normalized for the dry weight of the sample cells. The arrow in panel a indicates the absorption peak of an unknown hydrophilic component. The BChl a contents in cells incubated in the presence of light (10 W m⫺2) (b and d) and in cells incubated under semiaerobic conditions (2% oxygen) (f) are also plotted as reference data. All cells were precultured (OD660 of ⬇0.35) in medium containing 0.2% Casamino Acids and resuspended (OD660 of ⬇0.35) into each medium.

Effects of oxygen tension on pigmentation in the absence of Casamino Acids. Figure 2c shows the absorption spectra of methanol extracts from R. depolymerans cells incubated in Cfree medium for 72 h in the dark under different oxygen tensions after medium replacement. The oxygen tensions in the headspace were adjusted at 24-h intervals, and the consumptions of oxygen during every 24 h were less than 0.2, 4.6, and 41% of the initial concentrations at 20, 2, and 0.2% oxygen, respectively, determined by gas chromatography. Accumulation of BChl a was observed in the presence of oxygen (Fig. 2d), in contrast to observations for many other purple phototrophic bacteria (18) but similar to those for the aerobic phototrophic bacteria (31, 41). Accumulation of BChl a was also observed in the presence of light but was lower (Fig. 2d). The maximum accumulation of BChl a was observed under semiaerobic conditions with 2% oxygen in both the presence and absence of light illumination (Fig. 2d). The cells incubated with 0.02% oxygen showed no accumulation of BChl a (Fig. 2d), although the incubation with 0.02% oxygen was not lethal for the R. depolymerans cells. The motility and colony-forming

ability of the cells were hardly impaired during the 72 h of incubation even under this condition. Effects of light intensity on pigmentation in the absence of Casamino Acids under aerobic and semiaerobic conditions. Figure 2e shows the absorption spectra of methanol extracts from R. depolymerans cells aerobically incubated in C-free medium for 72 h under light illumination with 10, 5, and 2.5 W m⫺2 and in the dark after medium replacement. BChl a was accumulated optimally with incubation in the dark. The accumulation of BChl a was found to be reduced severely in the presence of dim light (2.5 W m⫺2) (Fig. 2f). The reduction of BChl a accumulation was more evident under semiaerobic conditions (2% oxygen), where the accumulation of BChl a under dark conditions was more prominent (Fig. 2f). Effects of Casamino Acids concentration on maximum growth rate under aerobic-dark and -light conditions. Figure 3 shows the maximum growth rates of R. depolymerans cells, which were observed at the beginning of incubation, when the cells were precultured in 0.2% Casamino Acids and then resuspended into various concentrations of Casamino Acids at a


VOL. 68, 2002

1669

TABLE 2. Colony-forming ability of starved cells of R. depolymerans incubated in the presence and absence of lighta Avg countb (107 CFU ml⫺1) Incubation time (days)

4 8 12 16

Dark

Light

Tube 1

Tube 2

Tube 3

Tube 4

7.08 8.36 3.30 1.21

7.76 7.34 3.21 1.05

7.02 9.56 4.75 4.83

8.32 9.81 5.20 4.36

a Cultured cells at an OD660 of ⬇0.35 were harvested and transferred into C-free medium (final OD660 of ⬇0.01), and then the cell suspension was incubated in the presence or absence of light (10 W m⫺2). b The colony-forming ability of the cells in the suspension before incubation was 8.96 ⫻ 107 CFU ml⫺1. Each value represents the average from duplicate platings with dilution at 10⫺5.

FIG. 2—Continued.

low cell density (OD660 of ⬇0.01). The cells grown in medium with more than 0.1% Casamino Acids uniformly showed the maximal growth rate (⬇0.4 doubling h⫺1). The growth rate of the cells was reduced with lower Casamino Acids concentrations (Fig. 3). Although accumulation of BChl a was not de-

tected at the beginning of incubation showing growth of the cells, the R. depolymerans cells accumulated BChl a after 2 to 3 days of incubation under these conditions (ⱕ0.04% Casamino Acids). As shown in Fig. 3, light illumination made little contribution to the growth rate of R. depolymerans cells. Survival of cells in the absence of carbon sources under aerobic-dark and -light conditions. Similarly to some aerobic phototrophic bacteria belonging to the ␣-Proteobacteria, R. depolymerans cells in the absence of carbon sources showed better maintenance of viability in the presence of light (Table 2). Colony-forming abilities at the beginning of incubation (8.96 ⫻ 107 CFU ml⫺1) were substantially maintained over 8 days of incubation, independently of illumination (tubes 1 to 4). Although the cells incubated in the dark and in the light both thereafter began to lose colony-forming ability, the loss of viability was less intense in the light-incubated cells (tubes 3 and 4) than in the dark-incubated cells (tubes 1 and 2). This experiment was repeated four times, and each time a greater number of colonies was obtained in the group with light illumination.

FIG. 3. Maximum growth rate of R. depolymerans cells in media containing different concentrations of Casamino Acids in the presence and absence of continuous illumination (10 W m⫺2). The cells were precultured (OD660 of ⬇0.35) in medium containing 0.2% Casamino Acids and resuspended (OD660 of ⬇0.01) into fresh medium. Accumulation of BChl a as well as puf mRNA was observed with Casamino Acids concentrations lower than the level shown by the dotted line.

1670

SUYAMA ET AL.


FIG. 4. Genetic maps of the BamHI fragment containing the R. depolymerans puf gene cluster. The restriction map for R. depolymerans genes is indicated at the top. The puf genes and the hypothetical ORFs are indicated by open boxes. At the bottom the positions corresponding to the probes and the region transcribed as mRNA are indicated by bars and the arrows, respectively. The electropherograms obtained from primer extension experiments and the position suggested to be the stable 5⬘ end of puf mRNA are indicated below the maps. Positions with potential stem-loops are indicated.

Analysis of the puf gene cluster. The puf gene cluster of R. depolymerans was sequenced and its transcripts were identified in order to clarify the presence of photosynthesis gene regulation. A total of 11 complete open reading frames (ORFs) were found in the 8.1-kbp nucleotide sequence from R. depolymerans, all of which were preceded by presumed Shine-Dalgarno sequences (Fig. 4). Among the ORFs, five were presumed to be pufB, -A, -L, -M, and -C, coding for the ␤ and ␣ subunits of the LH1 complex and the L, M, and cytochrome subunits of the RC complexes, respectively, based on comparison with known puf gene sequences. All of the amino acid residues shown to be ligands for the cofactors in the L, M, and cytochrome subunits of the RC complexes of B. viridis (6, 7) and Rhodobacter sphaeroides (1, 2) and the putative ligands for Mg2⫹ of BChls in the ␣ and ␤ subunits of the LH1 complex (16, 44) were fully conserved in the corresponding products inferred from the nucleotide sequence of R. depolymerans. Immediately upstream and downstream of pufB, two ORFs (ORF31 and ORF74 in Fig. 4) homologous to Rubrivivax gelatinosus ORF1 and ORF2 (22), respectively, were found. The upstream region of the R. depolymerans puf operon was highly homologous to the latter half (239 amino acids) of the bchZ gene, coding for a subunit of chlorin reductase (Fig. 4), which is known to be situated immediately upstream of the puf operon in other purple bacteria (5, 17, 22–24, 39). The nucleotide sequences as well as the deduced amino acid sequences of the puf genes, bchZ, ORF31, and ORF74 were most closely related to those of R. gelatinosus, a phototrophic ␤-Proteobacterium which can grow anaerobically using light as an energy source and which is the closest relative of R. depolymerans,

showing 16S rRNA gene sequence similarity of 96.3% (34). Primer extension experiments indicated that the stable 5⬘ end of the R. depolymerans puf mRNA was 105 bases upstream of the start codon of ORF31 (Fig. 4). Northern hybridization experiments with region-specific probes (probes BA, C, D, and Z in Fig. 4) detected RNA fragments containing the regions of probe BA (1 kb in length) and of probes BA-D (faint and smeared; 7.3 kb in length). No signal was obtained with probe Z in this experiment. The region lying between ORF31 and ORF428 appeared to be transcribed as a unit (Fig. 4). The R. depolymerans puf gene cluster is thought to contain at least three additional ORFs, ORF77, ORF350, and ORF428 (Fig. 4). The presence of ORFs flanked by pufM and -C (ORF77) has not been reported so far. Although the pufX gene, coding for a small peptide which may function in lateral exchange of ubiquinone and ubiquinol across the LH1 ring (10, 28), has been observed downstream of the pufM gene in Rhodobacter capsulatus and R. sphaeroides (5), the deduced product of ORF77 showed no sequence similarities to PufX or to any other proteins reported so far. Comparisons of nucleotide sequences as well as amino acid sequences of the deduced products indicated that ORF350 resembled the crtJ (27) and ppsR (26) genes, coding for repressor proteins functioning in photosynthesis gene regulation. The crtJ and ppsR genes showed DNA sequence similarities of 51.4 and 53.9% (in the continuous 1,080 nucleotides), respectively, to ORF350. Moreover, the amino acid sequences of the CrtJ and PpsR proteins contained 24.85 and 24.92% amino acid identities (in the continuous 329 amino acids), respectively, to the deduced product of ORF350. However, the helix-turn-helix motif conserved in


VOL. 68, 2002

FIG. 5. Northern hybridization with probe BA to total RNA extracted from R. depolymerans cells incubated under various conditions after medium replacement. The cells were incubated for 72 h aerobically in the dark in media containing 0.02% (lane 1), 0.04% (lane 2), 0.1% (lane 3), and 0.2% (lane 4) Casamino Acids; incubated for 72 h in C-free medium in the dark in the presence of 20% (lane 5), 2% (lane 6), 0.2% (lane 7), and 0.02% (lane 8) oxygen; and incubated for 72 h in the presence of 2% oxygen in C-free medium in the dark (lane 9) and with light illumination (10 W m⫺2) (lane 10). All cells were precultured (OD660 of ⬇0.35) in medium containing 0.2% Casamino Acids and resuspended (OD660 of ⬇0.35) into each medium. Three micrograms of total RNA extracted from the cell sample was applied for each lane.

the C-terminal regions of these proteins (26) was missing in the ORF350 product. Transcriptional study of the puf mRNA. Figure 5 shows the Northern hybridization profiles of 1-kb puf transcripts in total RNA samples (3 ␮g for each lane) from R. depolymerans cells incubated for 72 h after medium replacement under conditions of different Casamino Acids concentrations and oxygen tensions and with or without light illumination. Longer mRNA (around 7.3 kb in length) may also be transcribed but was hardly detected by hybridization with only 3 ␮g of total RNAs. The transcripts of the puf operon were detected only in cells incubated under the conditions producing a photosynthetic apparatus (Fig. 2 and 5, lanes 1, 2, 5 to 7, 9, and 10). The expression of puf mRNA was more intense with a lower concentration of Casamino Acids (Fig. 5, lane 1). The R. depolymerans puf mRNA was most markedly accumulated in cells incubated with 0.2% oxygen, and the levels of puf mRNA accumulation were lower with higher oxygen tensions (Fig. 5, lanes 5 to 7). The accumulation of puf mRNA in R. depolymerans cells incubated with 2% O2 was reduced to 15% by the presence of light (Fig. 5, lanes 9 and 10). DISCUSSION Role of photosynthetic apparatus. As shown in a previous report (34), most of the BChl a molecules in R. depolymerans cells are considered to be bound to RC-LH1 complexes. Hence, the accumulation of BChl a (Fig. 1) may reflect the production of a photosynthetic apparatus. Accumulation of BChl a in R. depolymerans cells was shown to occur when the culture medium was diluted or when the cells were transferred into a medium containing a low concentration of carbon sources (Table 1). Neither continuous depletion of carbon sources by means of consumption nor continuous incubation with low levels of carbon sources was shown to trigger this phenomenon. Cells precultured with 0.4% glucose produced BChl a upon incubation in 0.1% glucose, while cells precul-

1671

tured with 0.1% glucose did not (Table 1), suggesting that R. depolymerans cells produce BChl a responding not only to the absolute concentration of carbon sources but also to the magnitude of decline in carbon sources as the signal. Discrete changes from high to low concentrations of carbon sources may activate pigmentation of this bacterium. Among the three substrates, however, only glucose was shown to trigger pigmentation upon a change from 0.4 to 0.1% carbon sources (Table 1), probably because the threshold beyond which the bacterium felt starved was relatively high in this substrate or possibly because the relatively slow consumption of glucose in the preculturing medium resulted in sufficient changes in concentration among the three. Accumulation of BChl a was found to occur time dependently not in growing cells but in cells in a nongrowing state showing a decay in turbidity (Fig. 1). Whether or not the accumulation of BChl a occurs is also dependent on oxygen tension (Fig. 2c and d) and light intensity (Fig. 2e and f), and these responses may be partly regulated by the levels of puf mRNA, which codes for apoproteins of LH1 and RC complexes (Fig. 5). Therefore, it appears that R. depolymerans cells are dependent on their photosynthetic apparatus only in quite limited situations due to their own genetic regulation. Light illumination neither supported the phototrophic growth of R. depolymerans, as shown in the previous report (34), nor contributed to the growth of this organism under aerobic conditions, as shown in this study (Fig. 3). It is quite intelligible that this organism does not positively utilize light illumination as an energy source for growth, since this organism does not produce a photosynthetic apparatus at the growing stage. However, it was found that light has an effect on maintenance of viability in starved cells of R. depolymerans (Table 2) similar to that known for some aerobic phototrophic bacteria belonging to the ␣ subclass of the Proteobacteria (9, 15, 29). From the fact that the R. depolymerans cells produce photosynthetic apparatus in the absence of carbon sources, it appears that the capacity for light-dependent survival under starvation conditions is mediated by photosynthesis-like energy conversion using the RC-LH1 complex. R. depolymerans as well as these other bacteria may not have available electron donors for photosynthesis under conditions without carbon sources (9, 15, 29). There is a slight possibility that under such conditions the light-induced cyclic electron flow would maintain the electrochemical proton gradient essential for cellular integrity during carbon starvation. Effect of light and oxygen. Nishimura et al. (24) reported that R. sphaeroides and R. denitrificans reduce the expression of the puf operon in the presence of light and that the response of R. denitrificans to light is more sensitive. They speculated that the light sensitivity of the R. denitrificans puf operon may be a consequence of adaptation to highly aerobic conditions that is advantageous for avoiding photodynamic damage of the cells in the presence of oxygen. As shown in this study, the expression of the puf operon (Fig. 5, lanes 9 and 10) and the accumulation of BChl a (Fig. 2f) in R. depolymerans cells were also greatly reduced in the presence of light. It is possible that the response to light of R. depolymerans may also have a meaning similar to that assumed for R. denitrificans (24). The reduction of the photosynthetic apparatus of these organisms at higher light intensities is potentially related to the inability to grow

1672

SUYAMA ET AL.

phototrophically. R. depolymerans, belonging to the ␤-Proteobacteria, must have evolved into an aerobic phototrophic bacterium separately from R. denitrificans, belonging to the ␣-Proteobacteria. It is quite interesting that these two species from two different taxonomic branches, both of which adapted to aerobic conditions while maintaining photosynthetic apparatus, similarly regulate the puf operon in response to light intensity. We found previously that R. depolymerans grows only in the presence of oxygen (34). However, the effect of oxygen tension on the production of a photosynthetic apparatus has not been estimated. As shown in this study, significant levels of BChl a (Fig. 2d) as well as puf mRNA (Fig. 5, lane 5) accumulated in R. depolymerans cells even in the presence of 20% oxygen. Transcription of the puf operon is greatly reduced in many phototrophic species in the presence of atmospheric levels of oxygen, as has been well studied for R. capsulatus (4, 43) and R. sphaeroides (24). Expression of the R. gelatinosus puf operon was detected only under anaerobic (ⱕ0.02% oxygen)-light condition (data not shown). Regulation of photosynthesis genes in R. depolymerans seems to be clearly distinct from that in these phototrophic species belonging to the ␣-Proteobacteria (R. capsulatus and R. sphaeroides) and ␤-Proteobacteria (R. gelatinosus). R. denitrificans has been shown to express the puf operon independently of changes in oxygen tension (24). However, the levels of accumulation of R. depolymerans puf mRNA (Fig. 5, lanes 5 to 7) as well as photosynthetic pigments (Fig. 2d) were shown to be higher under semiaerobic conditions than under fully aerobic conditions. This finding suggests that the R. depolymerans puf operon is regulated by oxygen tension differently from that of R. denitrificans (24). Two phototrophic species of ␣-Proteobacteria, Rhodovulum sulfidophilum (11, 21) and Rhodospirillum centenum (40), are known to produce a photosynthetic apparatus under aerobic conditions, due to their weak repression of photosystem synthesis. The response to oxygen of R. depolymerans under aerobic and semiaerobic conditions may be understood similarly to those of R. sulfidophilum and R. centenum. However, R. depolymerans does not grow on light anaerobically, and the responses to oxygen of this organism (Fig 2d) and the two phototrophic species are not wholly alike (11, 21, 40). As shown by the response to oxygen, R. depolymerans is a unique organism containing BChl a which is distinguishable from both phototrophic and aerobic phototrophic bacteria studied so far. R. depolymerans and R. gelatinosus represent a tight phylogenetic cluster within the ␤ subclass of the Proteobacteria (34), and their puf operons as well as the deduced products of the puf genes are similar to each other. In particular, the genes homologous to ORF31 and ORF74 (Fig. 4) have been reported only for R. gelatinosus (ORF1 and ORF2, respectively [22]). However, the responses to oxygen turned out to be a major difference between R. depolymerans and phototrophic species, including R. gelatinosus. It would be worthwhile to investigate the differences in the structure, function, and expression mechanisms of the photosynthetic apparatus in these organisms to obtain insights into the evolution of aerobic organisms containing BChls. Regulation by nutrient levels. The induction of R. depolymerans puf mRNA seems to occur when the cells are suddenly


subjected to conditions causing reduction of the growth rate (Fig. 3 and 5, lanes 1 to 4). This correlation of transcriptional control with a reduction of growth rate is quite similar to nutrient limitation responses known in Escherichia coli and some other nonphototrophic bacteria (8). Besides the three substrates examined in this study, high concentrations of other carbon sources, e.g., lactose, casein, tryptone, beef extract, and yeast extract (T. Suyama, unpublished data) and skim milk and nutrient broth (34), also reduce pigmentation of R. depolymerans cells. Therefore, the expression of a photosynthetic apparatus in R. depolymerans is considered to occur transcriptionally, responding to, in general, decreases in levels of nutrients as respiratory substrates, in addition to oxygen tension and light intensity. Expression mechanisms of photosynthesis genes have been studied mainly in terms of responses to light and oxygen (4, 24, 25, 43). It has been reported that Bradyrhizobium strain BTAi 1 accumulated BChls at the end of exponential growth (9). This phenomenon is possibly related to depletion of carbon sources. Recently, Kolber et al. (19) reported that some oceanic aerobic phototrophic bacteria are also capable of controlling the expression of the photosynthetic apparatus in response to the organic condition of seawater. In the present study the production of a photosynthetic apparatus in R. depolymerans cells was shown to be regulated by transcriptional levels (Fig. 5). Our results indicate a possibility that these organisms also regulate the expression of the photosynthetic apparatus partly by the transcriptional control of the puf operon, although the mechanism is not yet understood. REFERENCES 1. Allen, J. P., G. Feher, T. O. Yeates, H. Komiya, and D. C. Rees. 1988. Structure of the reaction center from Rhodobacter sphaeroides R-26: proteincofactor (quinones and Fe2⫹) interactions. Proc. Natl. Acad. Sci. USA 85: 8487–8491. 2. Allen, J. P., G. Feher, T. O. Yeates, H. Komiya, and D. C. Rees. 1987. Structure of the reaction center from Rhodobacter sphaeroides R-26: the protein subunits. Proc. Natl. Acad. Sci. USA 84:6162–6166. 3. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403–410. 4. Bauer, C. E. 1995. Regulation of photosynthesis gene expression, p. 1221– 1234. In R. E. Blankenship, M. T. Madigan, and C. E. Bauer (ed.), Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands. 5. Beatty, J. T. 1995. Organization of photosynthesis gene transcripts, p. 1209– 1219. In R. E. Blankenship, M. T. Madigan, and C. E. Bauer (ed.), Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands. 6. Deisenhofer, J., O. Epp, K. Miki, R. Huber, and H. Michel. 1985. Structure of the protein subunits in the photosynthetic reaction centre of Rhodopseudomonas viridis at 3Å resolution. Nature 318:618–624. 7. Deisenhofer, J., O. Epp, I. Sinning, and H. Michel. 1995. Crystallographic refinement at 2.3 Å resolution and refined model of the photosynthetic reaction centre from Rhodopseudomonas viridis. J. Mol. Biol. 246:429–457. 8. Ferenci, T. 1999. Regulation by nutrient limitation. Curr. Opin. Microbiol. 2:208–213. 9. Fleischman, D. E., W. R. Evans, and I. M. Miller. 1995. Bacteriochlorophyllcontaining Rhizobium species, p. 123–136. In R. E. Blankenship, M. T. Madigan, and C. E. Bauer (ed.), Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands. 10. Francia, F., J. Wang, G. Venturoli, B. A. Melandri, W. P. Barz, and D. Oesterhelt. 1999. The reaction center-LH1 antenna complex of Rhodobacter sphaeroides contains one PufX molecule which is involved in dimerization of this complex. Biochemistry 38:6834–6845. 11. Hagemann, G. E., E. Katsiou, H. Forkl, A. C. Steindorf, and M. H. Tadros. 1997. Gene cloning and regulation of gene expression of the puc operon from Rhodovulum sulfidophilum. Biochim. Biophys. Acta 1351:341–358. 12. Hanada, S., A. Hiraishi, K. Shimada, and K. Matsuura. 1995. Isolation of Chloroflexus sp. and related thermophilic photosynthetic bacteria from hot springs using an improved isolation procedure. J. Gen. Appl. Microbiol. 41:119–130.

VOL. 68, 2002


13. Harashima, K., K. Kawazoe, I. Yoshida, and H. Kamata. 1987. Light-stimulated aerobic growth of Erythrobacter species OCh 114. Plant Cell Physiol. 28:365–374. 14. Harashima, K., M. Nakagawa, and N. Murata. 1982. Photochemical activity of bacteriochlorophyll in aerobically grown cells of aerobic heterotrophs Erythrobacter species (OCh 114) and Erythrobacter longus (OCh 101). Plant Cell Physiol. 23:185–193. 15. Hiraishi, A., Y. Matsuzawa, T. Kanbe, and N. Wakao. 2000. Acidisphaera rubrifaciens gen. nov., sp. nov., an aerobic bacteriochlorophyll-containing bacterium isolated from acidic environments. Int. J. Syst. Evol. Microbiol. 50:1539–1546. 16. Hu, X., and K. Schulten. 1998. Model for the light-harvesting complex I (B875) of Rhodobacter sphaeroides. Biophys. J. 75:683–694. 17. Hunter, C. N., P. McGlynn, M. K. Ashby, J. G. Burgess, and J. D. Olsen. 1991. DNA sequencing and complementation/deletion analysis of the bchApuf operon region of Rhodobacter sphaeroides: in vivo mapping of the oxygenregulated puf promoter. Mol. Microbiol. 5:2649–2661. 18. Imhoff, J. F. 1995. Taxonomy and physiology of phototrophic purple bacteria and green sulfur bacteria, p. 1–15. In R. E. Blankenship, M. T. Madigan, and C. E. Bauer (ed.), Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands. 19. Kolber, Z. S., F. G. Plumley, A. S. Lang, J. T. Beatty, R. E. Blankenship, C. L. VanDover, C. Vetriani, M. Koblizek, C. Rathgeber, and P. G. Falkowski. 2001. Contribution of aerobic photoheterotrophic bacteria to the carbon cycle in the ocean. Science 292:2492–2495. 20. Kortlüke, C., K. Breese, N. Gad’on, A. Labahn, and G. Drews. 1997. Structure of the puf operon of the obligately aerobic, bacteriochlorophyll a-containing bacterium Roseobacter denitrificans OCh114 and its expression in a Rhodobacter capsulatus puf puc deletion mutant. J. Bacteriol. 179:5247–5258. 21. Masuda, S., Y. Matsumoto, K. V. Nagashima, K. Shimada, K. Inoue, C. E. Bauer, and K. Matsuura. 1999. Structural and functional analyses of photosynthetic regulatory genes regA and regB from Rhodovulum sulfidophilum, Roseobacter denitrificans, and Rhodobacter capsulatus. J. Bacteriol. 181:4205– 4215. 22. Nagashima, K. V. P., K. Matsuura, S. Ohyama, and K. Shimada. 1994. Primary structure and transcription of genes encoding B870 and photosynthetic reaction center apoproteins from Rubrivivax gelatinosus. J. Biol. Chem. 269:2477–2484. 23. Nagashima, K. V. P., K. Matsuura, N. Wakao, A. Hiraishi, and K. Shimada. 1997. Nucleotide sequences of genes coding for photosynthetic reaction centers and light-harvesting proteins of Acidiphilium rubrum and related aerobic acidophilic bacteria. Plant Cell Physiol. 38:1249–1258. 24. Nishimura, K., H. Shimada, H. Ohta, T. Masuda, Y. Shioi, and K. Takamiya. 1996. Expression of the puf operon in an aerobic photosynthetic bacterium, Roseobacter denitrificans. Plant Cell Physiol. 37:153–159. 25. Oh, J. I., J. M. Eraso, and S. Kaplan. 2000. Interacting regulatory circuits involved in orderly control of photosynthesis gene expression in Rhodobacter sphaeroides 2.4.1. J. Bacteriol. 182:3081–3087. 26. Penfold, R. J., and J. M. Pemberton. 1994. Sequencing, chromosomal inactivation, and functional expression in Escherichia coli of ppsR, a gene which represses carotenoid and bacteriochlorophyll synthesis in Rhodobacter sphaeroides. J. Bacteriol. 176:2869–2876. 27. Ponnampalam, S. N., J. J. Buggy, and C. E. Bauer. 1995. Characterization of an aerobic repressor that coordinately regulates bacteriochlorophyll, carotenoid, and light harvesting-II expression in Rhodobacter capsulatus. J. Bacteriol. 177:2990–2997. 28. Recchia, P. A., C. M. Davis, T. G. Lilburn, J. T. Beatty, P. S. Parkes-Loach, C. N. Hunter, and P. A. Loach. 1998. Isolation of the PufX protein from Rhodobacter capsulatus and Rhodobacter sphaeroides: evidence for its inter-

29. 30.

31. 32. 33. 34.

35.

36. 37.

38.

39.

40. 41. 42. 43.

44.

1673

action with the ␣-polypeptide of the core light-harvesting complex. Biochemistry. 37:11055–11063. Shiba, T. 1984. Utilization of light energy by the strictly aerobic bacterium Erythrobacter sp. OCh 114. J. Gen. Appl. Microbiol. 30:239–244. Shigematsu, T., S. Hanada, M. Eguchi, Y. Kamagata, T. Kanagawa, and R. Kurane. 1999. Soluble methane monooxygenase gene clusters from trichloroethylene-degrading Methylomonas sp. strains and detection of methanotrophs during in situ bioremediation. Appl. Environ. Microbiol. 65:5198– 5206. Shimada, K. 1995. Aerobic anoxygenic phototrophs, p. 105–122. In R. E. Blankenship, M. T. Madigan, and C. E. Bauer (ed.), Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands. Stanssens, P., E. Remaut, and W. Fiers. 1986. Inefficient translation initiation causes premature transcription termination in the lacZ gene. Cell 44: 711–718. Suyama, T., H. Hosoya, and Y. Tokiwa. 1998. Bacterial isolates degrading aliphatic polycarbonates. FEMS Microbiol. Lett. 161:255–261. Suyama, T., T. Shigematsu, S. Takaichi, Y. Nodasaka, S. Fujikawa, H. Hosoya, Y. Tokiwa, T. Kanagawa, and S. Hanada. 1999. Roseateles depolymerans gen. nov., sp. nov., a new bacteriochlorophyll a-containing obligate aerobe belonging to the ␤-subclass of the Proteobacteria. Int. J. Syst. Bacteriol. 49:449–457. Suyama, T., Y. Tokiwa, P. Ouichanpagdee, T. Kanagawa, and Y. Kamagata. 1998. Phylogenetic affiliation of soil bacteria that degrade aliphatic polyesters available commercially as biodegradable plastics. Appl. Environ. Microbiol. 64:5008–5011. Takamiya, K., and K. Okamura. 1984. Photochemical activities and photosynthetic ATP formation in membrane preparation from a facultative methylotroph, Protaminobacter ruber strain NR-1. Arch. Microbiol. 140:21–26. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673–4680. Wakao, N., N. Yokoi, N. Isoyama, A. Hiraishi, K. Shimada, M. Kobayashi, H. Kise, M. Iwaki, S. Ito, S. Takaichi, and Y. Sakurai. 1996. Discovery of natural photosynthesis using Zn-containing bacteriochlorophyll in an aerobic bacterium Acidiphilium rubrum. Plant Cell Physiol. 37:889–893. Wiessner, C., I. Dunger, and H. Michel. 1990. Structure and transcription of the genes encoding the B1015 light-harvesting complex ␤ and ␣ subunits and the photosynthetic reaction center L, M, and cytochrome c subunits from Rhodopseudomonas viridis. J. Bacteriol. 172:2877–2887. Yildiz, F. H., H. Gest, and C. E. Bauer. 1991. Attenuated effect of oxygen on photopigment synthesis in Rhodospirillum centenum. J. Bacteriol. 173:5502– 5506. Yurkov, V. V., and J. T. Beatty. 1998. Aerobic anoxygenic phototrophic bacteria. Microbiol Mol. Biol. Rev. 62:695–724. Yurkov, V. V., and H. van Gemerden. 1993. Impact of light/dark regimen on growth rate, biomass formation and bacteriochlorophyll synthesis in Erythromicrobium hydrolyticum. Arch. Microbiol. 159:84–89. Zhu, Y. S., D. N. Cook, F. Leach, G. A. Armstrong, M. Alberti, and J. E. Hearst. 1986. Oxygen-regulated mRNAs for light-harvesting and reaction center complexes and for bacteriochlorophyll and carotenoid biosynthesis in Rhodobacter capsulatus during the shift from anaerobic to aerobic growth. J. Bacteriol. 168:1180–1188. Zuber, H., and R. J. Cogdell. 1995. Structure and organization of purple bacterial antenna complexes, p. 315–348. In R. E. Blankenship, M. T. Madigan, and C. E. Bauer (ed.), Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands.