Cloning and Sequencing of the Gene Encoding an Aldehyde Dehydrogenase That Is Induced by Growing Alteromonas sp. Strain KE10 in a Low Concentration of Organic Nutrients

Abstract

The protein composition of Alteromonas sp. strain KE10 cultured at two different organic-nutrient concentrations was determined by using two-dimensional polyacrylamide gel electrophoresis. The cellular levels of three proteins, OlgA, -B, and -C, were considerably higher in cells grown in a low concentration of organic nutrient medium (LON medium; 0.2 mg of carbon per liter) than cells grown in a high concentration of organic nutrient medium (HON; 200 mg of C liter⁻¹) or cells starved for organic nutrients. In the LON medium, the cellular levels of the Olg proteins were higher at the exponential growth phase than at the stationary growth phase. A sequence of the gene for OlgA revealed that the amino acid sequence had a high degree of similarity to the NAD⁺-dependent aldehyde dehydrogenases of several bacteria. OlgA, expressed in Escherichia coli, catalyzed the dehydrogenation of acetaldehyde, propionaldehyde, and butyraldehyde. The aldehyde dehydrogenase activity of KE10 was higher in cells growing exponentially in LON medium than in HON. OlgA may be involved in the growth under low-nutrient conditions. The physiological role of OlgA is discussed here.

Pelagic seawater environments are characterized by low concentrations of organic nutrients. The levels are generally around 0.5 mg of carbon per liter (C liter⁻¹), that is, far less than conventional bacterial culture media, which range from 20,000 to 2,000 mg of C liter⁻¹. Starvation survival is a possible adaptation mechanism of marine bacteria to such low-nutrient conditions (25, 29). During starvation survival, physiological and morphological changes have been observed, and several characteristic proteins are known to be induced (4, 22, 30).

Aside from starvation survival, there are many reports on bacteria which can grow in low-organic-nutrient concentrations comparable to those found in the natural environment (16–19, 43). Ishida et al. reported that such bacteria were the dominant population in the South China Sea and the West Pacific Ocean (19). Schut et al. isolated a bacterium possibly representing the dominant species in some oligotrophic marine environments (42). The ability to grow in a nutrient-poor condition is an important survival mechanism for these bacteria.

Alteromonas sp. strain KE10 is a heterotrophic marine bacterium isolated from pelagic seawater in the Kumano-nada Sea, Japan (48), using a low-nutrient liquid medium whose organic-nutrient concentration is comparable to those found in pelagic seawater environments (19). In 0.2 mg of C liter⁻¹–peptone media, KE10 grows more rapidly and prolifically than Vibrio harveyi, Cytophaga latercula, or Escherichia coli (27). The cells grown in low-nutrient medium take up leucine more efficiently at low concentrations than do the cells grown under high-nutrient conditions (20, 47, 48). Hence, some change in physiological biochemistry is expected to be induced when this bacterium is grown in low-organic-nutrient conditions. To further investigate this physiological adaptation to low-organic nutrient conditions, we examined the protein composition in the bacterial cells growing in different organic nutrient concentrations. Three proteins, designated as OlgA, -B, and -C, were specifically induced in cells growing exponentially in a low-organic-nutrient concentration. The gene for OlgA was cloned and sequenced.

MATERIALS AND METHODS

Cultivation and starvation of Alteromonas sp. strain KE10.

Alteromonas sp. strain KE10 was isolated from the Kumano-nada Sea, Japan (33°49′N, 136°37′E) (48). High- and low-concentration organic-nutrient media, HON and LON media, respectively, were prepared as described previously (27). Briefly, the HON medium contained 0.5 g of Trypticase peptone (BBL), 0.05 g of Bacto-yeast extract (Difco), and 0.01 g of Na₂HPO₄ in 1 liter of a salt solution (NSS) (28). The NaCl in NSS was baked at 450°C in advance. In preparation, distilled deionized water was further purified using a Nanopure II purifier (Stan-Hansen Co., Ltd.). The organic carbon concentration of HON medium was calculated to be about 200 mg of C liter⁻¹ based on the amino acid composition of the ingredients. The LON medium was a 10³ dilution of the HON medium using NSS. For the starvation experiment, the cells at the mid-exponential growth phase in HON medium were centrifuged, washed twice, and resuspended in NSS. The cultures and cell suspension were incubated at 20°C. The number of bacterial cells were counted with epifluorescent microscopy after staining with 1 μg ml⁻¹ of 4′6-diamidino-2-phenylindole (DAPI) (36). The CFU count was determined on HON medium containing 1.2% agar.

Preparation of cell extract from KE10.

The bacterial cells grown in HON medium or starved in NSS were harvested by centrifugation, while the cells in LON medium were centrifuged after being concentrated by filtration through a 0.2-μm-pore-size polycarbonate membrane filter (Nuclepore; Corning Costar). The bacterial pellet was suspended in a lysis buffer containing 9 M urea, 2% Nonidet P-40, and 2% β-mercaptoethanol and then sonicated on ice. The lysate was centrifuged at 20,000 × g for 20 min. The concentrations of protein in the supernatant were determined by means of a protein assay kit (Bio-Rad Laboratories, Tokyo, Japan) using bovine serum albumin as a standard.

Two-dimensional polyacrylamide gel electrophoresis.

Two-dimensional polyacrylamide gel electrophoresis was performed on the cell extract according to the method of Görg et al. (11) using a Multiphor II system (Pharmacia LKB Biotechnology). A volume of cell extract equal to 200 μg of protein was loaded on the first-dimension gel after the addition of Pharmalyte (pH 3 to 10) at a final concentration of 1% (vol/vol). In the first dimension, isoelectric focusing electrophoresis was performed with an immobilized pH gradient gel (Immobiline DryPlate) ranging from pH 4.0 to 7.0. The gel was equilibrated for 10 min in buffer A (0.05 M Tris-HCl, 1% [wt/vol] sodium dodecyl sulfate [SDS], 6 M urea, 30% [vol/vol] glycerol; pH 6.8 containing 0.25% (wt/vol) dithiothreitol (DTT) and then immersed in buffer A containing 4.5% (wt/vol) iodoacetamide and 0.01% bromophenol blue for 10 min. The gels were then placed on the second-dimension SDS polyacrylamide gradient gel (Excel Gel SDS) ranging from 8 to 18% of acrylamide. All of the gels were purchased from Pharmacia LKB Biotechnology. The electrophoresis was performed according to the procedure recommended by the manufacturer. After the electrophoresis, the proteins were stained with Coomassie brilliant blue R250 or blotted electrically on polyvinylidene difluoride (PVDF) membrane at 0.8 mA cm⁻² for 1 h using NovaBlot (Pharmacia LKB Biotechnology).

N-terminal amino acid sequencing.

The PVDF membrane containing the target protein was excised and directly applied to a model 470A gas-phase protein sequencer (Applied Biosystems, Inc.).

Southern hybridization analysis.

Southern hybridization was carried out using a standard method (39). The DNA of KE10 was extracted using the phenol-chloroform method (23), digested with restriction enzymes, and then blotted on a Hybond-N⁺ membrane (Amersham International, Plc.). An oligonucleotide DNA probe was designed based on the N-terminal amino acid sequence of OlgA protein and labeled with [γ-³²P]ATP using a 5′-labeling kit (Takara Co., Ltd.). Hybridization was performed at 45°C overnight in a solution containing 6× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate), 10× Denhardt's reagent, 0.5% SDS, 0.05 mg of denatured salmon sperm DNA ml⁻¹, and 0.2 pmol of ³²P-labeled oligonucleotide ml⁻¹ (39). The [γ-³²P]ATP was purchased from Amersham.

Cloning and DNA sequencing.

Positive DNA fragment in the southern hybridization was cloned by a standard method (39) using pBluescript II KS(+) (2) and E. coli DH5α [supE44 ΔlacU169 (φ80 lacZΔM15) hsdR17 recA1 endA1 gyrA96 thi-1 relA1]. E. coli cells were cultivated at 37°C in Luria-Bertani (LB) medium (39). When necessary, ampicillin (50 μg ml⁻¹) was added to the medium. The nucleotide sequence was determined using the Dye Terminator Cycle Sequencing Kit (Perkin-Elmer/Applied Biosystems, Inc.). A series of deletions were made by using the Kilo-Sequence Deletion Kit (Takara Co., Ltd.). Similar sequences were searched for using BLAST (3).

Determination of aldehyde dehydrogenase activity.

A bacterial pellet was suspended in a 10-fold weight of distilled water, sonicated on ice, and centrifuged at 20,000 × g for 20 min. The aldehyde dehydrogenase activity of the supernatants was determined by measuring the initial rate of NADH production at 340 nm at 25°C in a buffer containing 100 mM Tris-HCl (pH 8.0), 100 mM KCl, 1 mM DTT, 5 mM NAD⁺, and 10 mM aldehydes (8). One unit of activity was defined as the amount of enzyme required to produce 1 μmol of NADH per min.

Nucleotide sequence accession number.

The nucleotide sequence data reported here has been deposited in the DDBJ, EMBL, and GenBank nucleotide sequence databases under accession number AB009654.

RESULTS AND DISCUSSION

Growth and starvation survival characteristics of Alteromonas sp. strain KE10.

The growth results of Alteromonas sp. strain KE10 in HON and LON media is shown in Fig. 1. The doubling time at the exponential-growth phase in HON medium was 1.8 h and in LON medium was 2.3 h. The maximum growth yield in HON medium was 1.5 × 10⁹ cells ml⁻¹, while the yield in LON medium was 2.7 × 10⁶ cells ml⁻¹, which could be determined only by using a direct microscopic count. Because no change was observed in the growth rate and the yield through three serial cultures in LON medium, the increase in the number of cells was concluded to be the result of substantial growth and not of fragmentation at the onset of starvation survival as described below. The characteristics of starvation survival with KE10 in an artificial seawater, NSS, are presented in Fig. 2. In the initial phase of starvation, both the direct count and CFU number increased, while the optical density decreased and the cell size also declined (data not shown). An initial increase in cell numbers, without an increase in biomass, has previously been referred to as fragmentation (24, 33) and has been observed with several marine bacteria such as Vibrio sp. strains ANT-300, DW1, and S14 and Pseudomonas sp. strain S9 (24, 28, 33). After 2 days, while the direct count remained at the same level, the CFU level began to decrease. The percent ratio of CFU to maximum CFU (on day 2) decreased to 60% in KE10 after 1 week of starvation. In comparison with other strains, the ratio was 100% in Vibrio sp. strain S14 (34), 45% in E. coli K-12, and 30% in Salmonella enterica serovar Typhimurium LT2 (38).

FIG. 1.

Growth in HON medium (●) and LON medium (■) of Alteromonas sp. strain KE10. The arrows indicate the harvest points.

FIG. 2.

Starvation survival in NSS of Alteromonas sp. strain KE10. Symbols: ●, direct count; ○, CFU; ■, optical density at 660 nm.

Specific proteins to growth in LON medium.

Cells of strain KE10 were collected in the exponential and stationary growth phases in LON media and in the mid-exponential, late-exponential, and stationary growth phases in HON media (harvest points are indicated with arrows in Fig. 1). Starved cells were also collected after starvation for 8 days. Equal amounts of protein from the cell extracts were applied to a two-dimensional polyacrylamide gel electrophoresis.

Figure 3 shows the two-dimensional gel electrophoretic patterns of the cell extracts. Three spots, designated as OlgA, -B, and -C proteins, were clearly detectable in the cells collected at the exponential growth phase in LON media but were undetectable or much smaller at the stationary growth phase in LON media, through the phases of growth in HON media, and in the starved cells. Some proteins are reported to be enhanced in the stationary growth phase or by starvation survival in Vibrio sp. strains ANT-300 (4) and S14 (22), V. vulnificus (30), E. coli (12, 13, 41), P. putida (10), and Mycobacterium spp. (49). In contrast to these proteins, the cellular levels of Olg proteins were clearly higher at the exponential growth phase under low-nutrient conditions than those at the stationary growth phase or under nutrient starvation conditions. Therefore, it was considered that Olg proteins had some adaptive physiological role for the growth of this bacterium in low-nutrient conditions.

FIG. 3.

Two-dimensional polyacrylamide gel electrophoretic pattern of cell extracts of Alteromonas sp. strain KE10. The cells were collected at the exponential (A) and stationary (B) growth phases in LON medium and at the mid-exponential (C), late-exponential (D), and stationary (E) growth phases in HON medium, and after starvation for 8 days (F). Arrows indicate the positions of the Olg proteins. Bacterial cells were harvested at the same points indicated by the arrows in Fig. 1. The protein was stained with Coomassie brilliant blue R250.

Cloning and sequencing of the gene for OlgA protein.

OlgA was selected among Olg proteins for cloning because the amount of this protein was relatively higher (Fig. 3). The molecular weight and isoelectric point of OlgA were determined by electrophoresis and were approximately 55 kDa and 5.0, respectively (Fig. 3). The spots of OlgA protein on the PVDF membranes were excised and subjected to automated N-terminal amino acid sequencing. The N-terminal region of OlgA had the following sequence: M-I-Y-A-K-P-G-S-E-G-S-V-V-()-F-K-E. The 14th amino acid residue could not be identified. On the basis of the N-terminal amino acid sequence of OlgA, an oligonucleotide probe (5′-ATGAT[A/C/T]TA[C/T]GC[A/C/G/T]AA[A/G]CC[A/C/G/T]GG-3′) was designed and hybridized with the genomic DNA of KE10 digested by six-base-recognizing restriction enzymes (HindIII, EcoRI, SalI, XbaI, XhoI, PstI, or KpnI) and then double digested by EcoRI and SalI. Only one positive band was detected for each enzyme digest (data not shown). Hence, it was concluded that the genome of KE10 contains a single copy of the gene for OlgA.

The positive DNA fragment from a HindIII digestion (approximately 2.7 kbp) was cloned into the corresponding restriction enzyme site in pBluescript II KS(+). Sequencing of the insert fragment in the produced plasmid, pBOI, revealed that it contained a part of the gene for OlgA (Fig. 4). Therefore, another positive fragment produced by double digestion of EcoRI-SalI (ca. 3.9 kbp) was also cloned, and pBOES was produced. The HindIII-SalI fragment in the pBOES was subcloned into the corresponding site in pBluescript II KS(+) to produce pBOHS and then sequenced (Fig. 4). The obtained sequences revealed the presence of one open reading frame (ORF) which could encode a protein of 55.3 kDa and comprised 505 amino acids. The isoelectric point of the protein was calculated to be 4.9. The predicted molecular weight and isoelectric point were comparable to those estimated for OlgA by two-dimensional gel electrophoresis. The deduced N-terminal amino acid sequence was identical to that of OlgA as determined by amino acid sequencing (Fig. 5). Therefore, it was concluded that the ORF was the structural gene for OlgA protein. A putative ribosome-binding site existed ca. 10 bp upstream from the starting codon. Two inverted repeat sequences were found downstream of the ORF which might work as terminators.

FIG. 4.

Map of the region containing the gene encoding OlgA. The plasmids pBOI, pBOES, and pBOHS contained each DNA fragment represented by the bars. A broad arrow indicates the structural gene for OlgA. A triangle indicates the region hybridized to the probe. A broad solid bar indicates the sequenced region.

FIG. 5.

Nucleotide sequence of the gene for OlgA. The deduced amino acid sequence of the OlgA protein is shown below the nucleotide sequence. A possible Shine-Dalgarno sequence (SD) and inverted repeat sequences (IR) are underlined. An asterisk indicates a stop codon. The double-underlined sequence is the N-terminal amino acid sequence of OlgA as determined by protein sequencing. These nucleotide sequence data are listed in the DDBJ, EMBL, and GenBank nucleotide sequence databases under accession number AB009654.

The deduced amino acid sequence of OlgA showed similarities to various NAD⁺-dependent aldehyde dehydrogenases of prokaryotes and eukaryotes. The highest similarities were found in several bacteria: an NAD⁺-dependent acetaldehyde dehydrogenase (ExaC) involved in the ethanol oxidation system in P. aeruginosa (75% identity) (40); AldB and AldA, chloroacetaldehyde dehydrogenases in the degradation pathway of 1,2-dichloroethane in Xanthobacter autotrophicus GJ10 (68 and 67% identities, respectively) (7); AcDH-II (AcoD), acetaldehyde dehydrogenase II in the catabolism of acetoin by Alcaligenes eutrophus (65% identity) (37); AldB, an aldehyde dehydrogenase induced at an early stationary growth phase in E. coli (65% identity) (50); ThcA, an aldehyde dehydrogenase concerning degradation of thiocarbamate herbicide in Rhodococcus sp. strain NI86/21 (65% identity) (31); ToxR-regulated AldA of Vibrio cholerae (64% identity) (35); and a hypothetical protein, Rv0458, which was identified in Mycobacterium tuberculosis by its complete genomic sequencing (63% identity) (9). The alignment of the OlgA sequence with these proteins is shown in Fig. 6.

FIG. 6.

Alignment of amino acid sequences of OlgA with bacterial aldehyde dehydrogenases. ExaC-Pae, ExaC of P. aeruginosa (40); AldA-Xau and AldB-Xau, AldA and AldB of X. autotrophicus (7); AcDH2-Aeu, AcDH-II of A. eutrophus (37); ThcA-Rho, ThcA of Rhodococcus sp. strain NI86/21 (31); AldB-Eco, AldB of E. coli (50); AldA-Vch, AldA of V. cholerae (35); and Rv0458-Mtu, a hypothetical protein Rv0458 in M. tuberculosis (9). A possible NAD⁺-binding site (G-X₄-G motif) is double overlined. The PROSITE motifs are single overlined. The G-X-G-X₃-G motif is marked by a broken line. The residues implicated in the catalytic activity are marked by asterisks. Dots represent residues identical with the sequence of OlgA, and dashes indicate gaps for alignment.

A glutamic acid and a cysteine residue have been implicated in the catalytic activity in aldehyde dehydrogenases (1, 14, 45). Two consensus sequences including these residues were proposed by using PROSITE (release 15.0) (5). The consensus sequence with the glutamic acid, [LIVMFGA]-E-[LIMSTAC]-[GS]-G-[KNLM]-[SADN]-[TAPFV], was found in OlgA between residues 261 and 268 (VELGGKSP). Another consensus sequence with the cysteine, [FYLVA]-X₃-G-[QE]-X-C-[LIVMGSTANC]-[AGCN]-X-[GSTADNEKR], was found between residues 294 and 305 (YFNQGEVCTCPS). In these sequences, Glu262 and Cys301 of OlgA probably corresponded to the active site residues. OlgA contained a G-X-G-X₃-G motif between residues 218 and 224 (GFGAEAG) which is found in aldehyde dehydrogenases catalyzing irreversible reactions (15). An NAD⁺-binding motif, [G-X₄-G], which is involved in interactions with the nicotinamide ring (14, 26), was found between residues 240 to 245 (GSTPVG). Glu191 in OlgA probably corresponded to adenine ribose-binding residue (26) (Fig. 6). These features found in OlgA strongly suggested that it is an NAD⁺-dependent aldehyde dehydrogenase.

Expression of the gene for OlgA in E. coli.

The fragment containing the gene for OlgA was inserted into pUC18. The generated plasmid pUDE42 contained positions 21 to 2001 shown in Fig. 5, and it was located downstream of the lac promoter. pUDE42 and pUC18 were introduced into E. coli JM109 (46). The transformants were cultivated in LB medium and isopropyl-β-d-thiogalactopyranoside (IPTG) was added when the absorbance at 600 nm reached 0.6. The crude extracts of transformants were subjected to SDS-polyacrylamide gel electrophoresis. Accumulation of a protein with the expected size was observed in E. coli JM109 carrying pUDE42 but not in strain JM109 carrying pUC18 (data not shown). The aldehyde dehydrogenase activities in the crude extracts were determined with acetaldehyde, propionaldehyde, and butyraldehyde as substrates. As shown in Table 1, the enzyme activity was enhanced by transformation with pUDE42 but not by transformation with pUC18. Therefore, it is concluded that OlgA has dehydrogenase activities against aldehydes.

TABLE 1.

Aldehyde dehydrogenase activities in crude cell extracts from E. coli transformants and Alteromonas sp. strain KE10

Strain and condition	Enzyme activity (U g of protein−1) on substrate:
	Acetaldehyde	Propionaldehyde	Butyraldehyde
E. coli JM109
pUC18	0.3	0.8	0.3
pUDE42	21.6	22.0	16.9
Alteromonas sp. strain KE10
Grown in LON medium	633.6	861.7	566.0
Grown in HON medium	4.9	14.7	9.8

A broad substrate spectrum has been shown for enzymes which show a high degree of similarity to OlgA in their amino acid sequences. For example, AcDH-II of A. eutrophus catalyzes the dehydrogenation of acetaldehyde, formaldehyde, propionaldehyde, butyraldehyde, and glutaraldehyde (21); AldA of X. autotrophicus oxidizes chloroacetaldehyde, propionaldehyde, acetaldehyde, and benzaldehyde (7); and ThcA of Rhodococcus sp. strain NI86/21 oxidizes long-chain aliphatic aldehydes (31). Bacterial aldehyde dehydrogenases are divided into two types according to their specificity to substrates (6). Enzymes such as the malonic semialdehyde dehydrogenase of P. aeruginosa (32) belong to a high-specificity type, and enzymes such as the aldehyde dehydrogenase encoded by ald gene of E. coli (15) belong to a low-specificity type. OlgA should possibly be classified with the latter type according to its broad substrate spectrum. The low-specificity type is engaged in the detoxification of some aldehydes which can be involved in several metabolic pathways (6).

Aldehyde dehydrogenase activity of Alteromonas sp. strain KE10 cultivated in HON and LON media.

The aldehyde dehydrogenase activities of KE10 were determined in cell extracts prepared from cells in the exponential growth phase from both HON and LON media. The activities found in the cells grown in the LON medium ranged from 566.0 to 861.7 U g of protein⁻¹, a level far higher than the activities in the cells of the HON medium, which ranged from 4.9 to 14.7 U g of protein⁻¹ (Table 1). The substrate spectrum for the enzyme found in the cell extract of KE10 grown in LON medium was the same as that for OlgA expressed in E. coli. Even if all of the activity did not depend on OlgA, the cellular level of aldehyde dehydrogenase activity was enhanced when strain KE10 was cultivated in a low concentration of organic nutrients. Some aldehydes would be produced in the cells.

The production during growth under low-organic-nutrient conditions has not been reported for enzymes which show similarity to OlgA. However, it is noteworthy that some enzymes are concerned with limited or poor nutritional conditions. AldB of E. coli is induced during the transition from the exponential to the stationary growth phase (50). AcDH-II of A. eutrophus (37) and AldA and AldB of X. autotrophicus (7) were controlled by RpoN, which is involved in nitrogen limitation. AldA of V. cholerae is a member of ToxR regulon (35), which is also influenced by the global regulator cAMP-CRP under various environmental conditions (44).