The DDBJ accession numbers for the 16S rDNA sequences of Moritella sp. strain 2D2 and Moritella sp. strain 5710 are AB120660 and AB120661, respectively. The DDBJ accession number for the sequence data of the MDH of Moritella sp. strain 2D2 is AB097559.
The gene encoding malate dehydrogenase (MDH) of the obligately piezophilic deep-sea bacterium Moritella sp. strain 2D2 was cloned and sequenced. There were two positions [close to the active site (Ala-180) and in the subunit interaction site (His-229)] with 2D2-specific substitutions. The MDH genes of strain 2D2 and a psychrophilic bacterium Moritella sp. strain 5710 exhibiting the highest sequence similarity were overexpressed in Escherichia coli. The 2D2 MDH was more heat-stable than the 5710 MDH. The apparent Km value at 62.1 MPa for NADH of the 2D2 MDH was higher than that of the 5710 MDH. The 2D2 MDH in which a His–Gln substitution was introduced at position 229 decreased the thermal stability and Km value at 62.1 MPa. The 5710 MDH that was substituted Gln-229 with His increased the thermal stability and Km value at 62.1 MPa. These results indicate that the His residue at position 229 of the 2D2 MDH may play a role in the thermal stability and the MDH function at high pressure.
Studies on enzymatic adaptation to the deep sea (high pressures and low temperatures) have been carried out mostly on dehydrogenases of deep-sea animals [1,2]. It has been shown that one of the adaptations of these enzymes is characterized by more pressure insensitive Kms, resulting in more stable enzymes to pressure perturbations . However, the molecular basis for the adaptation to high pressure remains still unexplored.
On the other hand, it is widely accepted that the typical properties of enzymes from cold- and 1-atm-adapted organisms are high catalytic efficiency at low temperatures and relative instability at high temperatures [4,5]. Recently, directed evolution experiments demonstrated that low temperature activity and thermal stability could be improved simultaneously . However, it remains a fact that such enzymes are not generally found in nature .
Here, we suggest that such enzymes may be found in the deep-sea environments characterized by high hydrostatic pressures and low temperatures. For instance, as deep-sea bacteria are high-pressure- and cold-adapted organisms, their enzymes may have both low temperature activity and thermal stability. Unfortunately, very few studies have been carried out on the enzymes of piezophilic deep-sea bacteria. There is only one paper on the malate dehydrogenase (MDH) from the facultatively piezophilic deep-sea bacterium Photobacterium sp. strain SS9 . However, the amino acid residues important for protein stability or function at low temperatures or high pressures are still unknown.
In order to elucidate the molecular basis for MDH function at high pressure and low temperature, we have cloned, sequenced, and overexpressed the gene encoding MDHs from the obligately piezophilic deep-sea bacterium Moritella sp. strain 2D2  and the psychrophilic Moritella sp. strain 5710 which was isolated and classified to the genus Vibrio by Hamamoto et al. . The amino acid sequence of the 5710 MDH was submitted to the DDBJ database in 1995 by Ohkuma (Swiss-Prot Accession No. P48364) . Analyses of sequence-function relationships in the MDHs from these congeneric deep-sea bacteria adapted to different pressures were carried out to determine amino acid residues important for the MDH function at high pressure.
2Materials and methods
2.1Bacterial strains and growth conditions
The obligately piezophilic deep-sea bacterium strain 2D2 which was isolated from the intestinal content of the deep-sea fish (Coryphaenoides yaquinae) retrieved from a depth of 6,000 m in the Northwest Pacific Ocean  and the psychrophilic deep-sea bacterium Vibrio sp. strain 5710 which Hamamoto et al.  isolated from the sediment retrieved from a depth of 2,220 m at Suruga Bay were used in this study. Strain 5710 was obtained from Dr. Yuichi Nogi, Japan Marine Science and Technology Center. Strain 2D2 was grown in a retortable pouch containing marine broth (Difco) at the in situ pressure (62.1 MPa) and 4°C for 10 days. Strain 5710 was grown in marine broth at 1 atm (0.1 MPa) and 4°C for 2 days. To confirm the piezophilism of strain 5710, the growth curves of the strain both at 1-atm (0.1 MPa) and at the in situ pressure (20.7 MPa) were made at 4°C for 4 days by the methods described previously .
Chromosomal DNAs of strains 2D2 and 5710 were extracted by heating their cells in 500 μl sterile distilled water at 100°C for 5 min. The solutions were centrifuged at 13,000g at 4°C for 10 min. Each of the supernatants and a universal eubacterial primer set, corresponding to position 8–27 (forward primer) and 1,491–1,512 (reverse primer) of the Escherichia coli numbering system , were used for PCR amplification of their 16S rRNA genes. The PCR amplifications were carried out with Premix Taq (EX Taq version) (TaKaRa) according to the manufacturer's directions. The PCR products were sequenced by using Thermo Sequenase II Dye Terminator Cycle Sequencing Kit (Amersham Bioscience) as directed by the manufacturer, and reaction mixtures were analyzed with an Applied Biosystems model 373A DNA Sequencer. These sequences were aligned with closely related sequences obtained from the DDBJ database by the FASTA algorithm [11,12]. Phylogenetic analysis based on the nucleotide sequences of 16S rRNA genes was done using the Clustal W Multi-Alignment Program , and a phylogenetic tree was constructed by the neighbor-joining method .
2.3Cloning and sequencing of MDH gene of strain 2D2
Chromosomal DNA of strain 2D2 was extracted and purified by a modified Thomas's procedure [15,16] in which SSC (0.15 M NaCl, 0.015 M sodium citrate) and 27% sucrose were replaced with Tris–HCl buffer and 25% sucrose, respectively. The partial MDH gene was amplified from the chromosomal DNA by PCR using Premix Taq (EX Taq version) (TaKaRa) according to manufacturer's directions as shown in Fig. 1(A). The PCR primers used were primer 238F (5′-GCICGTAARCCNGGNATGGAY-3′) and primer 619R (5′-ACTTCIGTNCCNGCRTTYTG-3′), where I represents inosine, R represents A or G, N represents A, C, G, or T, and Y represents C or T. These degenerate primer sequences were derived from the amino acid sequences of ARKPGMD for 238F and of QNAGTEV for 619R conserved in several Gram-negative bacterial MDHs. The numbers of 238 and 619 are based on the nucleotide sequence of the E. coli gene encoding MDH published by McAlister-Henn et al. , Blattner et al. , and Swiss-Prot Accession No. P06994.
The amplified fragment (0.4 kb) was cloned by TOPO TA Cloning Kits with pCR2.1-TOPO vector and TOP10 chemically competent E. coli (Invitrogen) according to manufacturer's directions. White colonies of transformants obtained by blue/white colony screening were examined by One Shot Insert Check PCR Mix (TaKaRa). The transformants deduced to harbor plasmids with a correct insert were incubated in 5 ml LB broth containing 50 μg/ml kanamycin (Wako). The plasmids were purified by Quantum Prep Plasmid Miniprep Kit (Bio-Rad) from the transformant cells which were harvested from 2 ml aliquot of the cultured LB broth by centrifugation. The purified plasmids were sequenced with M13F and M13R primers by the methods described above (Fig. 1(B)). The upstream and downstream sequences from the cloned fragment were amplified by using TaKaRa LA PCR in vitro cloning Kit (TaKaRa) as shown in Fig. 1(C). To obtain the upstream sequence and to obtain the downstream sequence, two reverse primers (MDHLASR1: CGCTAGATTTTTGATAATGCCAGCATTGATATTGA and MDHLASR2: ATAATGCCAGCATTGATATTGAATAGATCTGAACG) and two forward primers (MDHLASF1: CGTTCAGATCTATTCAATATCAATGCTGGCATTAT and MDHLASF2: TCAATATCAATGCTGGCATTATCAAAAATCTAGCG) were prepared on the basis of the 5′-end nucleotide sequence of the cloned fragment, respectively. The LA PCR products (2.4 kb) amplified with EcoRI cassette by using MDHLASR1, MDHLASR2, and EcoRI cassette primers for the upstream sequence and the LA PCR products (2.0 kb) amplified with SalI cassette by using MDHLASF1, MDHLASF2, and SalI cassette primers for the downstream sequence were cloned and sequenced by the methods described above. Sequencing for the upstream and the downstream were done until a proper methionine codon was found and until the first stop codon was found, respectively (Fig. 1(D)). Finally, the overall sequence of the MDH gene was obtained by joining these sequences (Fig. 1(E)).
2.4Overexpression and purification of MDHs of strains 2D2 and 5710
The MDH genes of strains 2D2 and 5710 were expressed in E. coli from Novagen's pET system. Chromosomal DNAs of the strains were extracted and purified by the modified Thomas's procedure as described above. The MDH genes were amplified by PCR using primer MDHFactorXaN (5′-AATGGAGAGCTCCATGAAAGTCGCTGTATTAGGTGCT-3′) to create SacI site and FactorXa recognition site at the upstream region of genes and primer MDHSTOPC (5′-TAATGACTCGAGTTAACCAGCAACGAATTCTT-3′) to create a XhoI site at the downstream region of the gene (restriction sites and FactorXa site are underlined with lines and a dotted line, respectively). The amplified genes were cloned into the pET41b plasmid (Novagen) digested with SacI and XhoI beforehand. The resulting plasmids for strains 2D2 and 5710 were designated p2D2MDH and p5710 MDH, respectively. Then, E. coli BL21(DE3) cells were transformed with p2D2MDH and p5710MDH by using kanamycin selection. The resulting transformants for strains 2D2 and 5710 were designed E. coli BL21(DE3)/p2D2MDH and E. coli BL21(DE3)/p5710MDH, respectively. These transformants were grown at 37°C in 500 ml LB medium containing 30 μg/ml kanamycin until the OD600 reached 0.6. The target proteins were induced by the addition of isopropyl-β-d-thiogalacto-pyranoside (IPTG) (0.4 mM, final concentration). Incubation was then continued at 20°C for 12 h. The induced cells were harvested by centrifugation. Purification of the recombinant glutathione S-transferase (GST) fusion proteins from the induced cells were done by GST · Bind Kits (Novagen). However, several steps of the manufacturer's procedure were modified as follows: the treatments of lysozyme, freezing, and Benzonase were added as described below. The induced cell pellets were suspended in 1 × GST binding/wash buffer (4.3 mM Na2HPO4, 1.47 mM KH2PO4, 137 mM NaCl, 2.7 mM KCl, pH 7.3) 20 ml, added lysozyme to a final concentration of 200 μg/ml, incubated at 30°C for 15 min, frozen at −80°C more than 1 h. The thawed cell lysates were sonicated on ice for 4 min. Since the cell lysates were so viscous, the lysates were treated by Benzonase of BugBuster GST · Bind Purification Kit (Novagen) as directed by the manufacturer. The treated cell lysates were centrifuged at 32,000×g for 20 min at 4°C to remove debris. The supernatants, the soluble fractions of the recombinant GST fusion proteins, were digested by 10 U FactorXa (Novagen) per 1 mg recombinant GST fusion protein at 20°C for 20 h to cut off the GST · Tag. The resulting recombinant proteins without a GST · Tag were purified by GST · Bind Resin columns. The fractions which did not bind to GST · Bind Resin were collected and concentrated by Centricon 30 (Amicon). Glycerol was added to the fractions to a final concentration of 50%. The fractions were stored at −80°C until the MDH activities were assayed. The resulting recombinant MDHs of strains 2D2 and 5710 were designed r2D2MDH and r5710MDH, respectively. Thus, these recombinat MDHs had completely the same sequences of amino acids as those of the native MDHs of strains 2D2 and 5710.
2.5Characterization of MDHs
The activities, the Tm values, the optimal temperatures, and the optimal pHs of the MDHs were measured by the methods of Ohkuma et al. . However, several procedures were modified as described in Table 1.
Table 1. Catalytic properties of r2D2MDH and r5710MDH
Optimum temperatures of the MDHs were determined by incubating the reaction mixtures at various temperatures in 40 mM potassium phosphate buffer (pH 7.0). To determine optimum pHs of the MDHs, the enzyme solutions were incubated at 30°C in reaction mixtures at various pHs (40 mM potassium phosphate buffer between pH 6.0 and 8.5, 100 mM Tris–HCl buffer between pH 7.5 and 9.5, and 100 mM glycine–NaOH buffer between pH 9.0 and 11.0). The Km values were determined by incubating the reaction mixtures containing NADH at various concentrations of 0.02–0.15 mM in 40 mM potassium phosphate buffer (pH 7.0). Apparent Km and kcat values for NADH were calculated from a Lineweaver–Burk plot. The midpoints of denaturarion (Tm) were estimated from thermal denaturation curves that were determined by the residual MDH activities of the enzyme solutions (200 μg/ml) after a 10-min incubation at different temperatures (30–50°C) in 40 mM potassium phosphate buffer (pH 7.0).
aData are shown from two independent experiments.
bData are shown from three independent experiments.
Optimum temperature for activitya
Optimum pH for activitya
Km values of NADH at 30°Cb
1.72 ± 0.15 × 10−5 M
1.35 ± 0.28 × 10−5 M
kcat of oxaloacetate reduction at 30°Cb
4.83 ± 0.41 × 10−3 s−1
3.86 ± 0.61 × 10−3 s−1
2.6Site-directed mutagenesis of MDH genes of strains 2D2 and 5710
Site-directed mutageneses of the MDH genes of strains 2D2 and 5710 were performed by the use of QuikChange XL Site-Directed Mutagenesis Kit (STRATAGENE), using p2D2MDH and p5710MDH as templates, respectively. Two sets of complementary mutagenic DNA oligomers as primer pairs were designed for the substitution of His-229 (codon: CAT) with Gln-229 (codon: CAG) for p2D2MDH and of Gln-229 (codon: CAG) with His-229 (codon: CAT) for p5710MDH. The primer sets for p2D2MDH and p5710MDH were 5′-TCTATGGGTCAGGCCGCTGCACG-3′ and 5′-CGTGCAGCGGCCTGACCCATAGA-3′, and 5′-TCTATGGGTCATGCAGCTGCA-3′ and 5′-TGCAGCTGCATGACCCATAGA-3′, respectively. The nucleotide sequences of the resulting clones were confirmed by sequencing. The resulting mutant MDHs were overexpressed and purified to homogeneity as described above.
2.7Thermal stability measurements
Thermal stabilities were measured by incubating the enzyme solution (200 μg/ml) in 40 mM potassium phosphate (pH 7.0) at 40°C. After incubation, the enzyme activities were assayed immediately by the method of Ohkuma et al. .
2.8Kinetic studies at atmospheric pressure (0.1 MPa) and high pressure (62.1 MPa)
The MDH activities were monitored in a Shimadzu Model UV-160A UV–visible recording spectrophotometer with a high pressure optical cell (PCI-400; Teramecs Co., Ltd, Kyoto, Japan; 2 ml volume). Temperature was regulated (±0.2°C) with a circulating water bath. The activities were followed by monitoring the decrease in absorbance of NADH at 339 nm in 2.0 ml reaction mixture of 100 mM Tris–HCl buffer (pH 7.5), 0.5 mM oxaloacetate, and various concentrations (0.02–0.1 mM) of NADH at 5°C and at two pressures (0.1 and 62.1 MPa). Apparent Km and kcat values for NADH were calculated from a Lineweaver–Burk plot. For the kinetic studies at 0.1 MPa the high pressure cell was not pressurized, but for the kinetic studies at 62.1 MPa the high pressure optical cell was pressurized to 62.1 MPa with a hydraulic pump (Enerpac 40,000 PSI; Enerpac Group, Applied Power, Inc., Butler, WI, USA).
3Results and discussion
3.1Piezophilism of strain 5710
Strain 5710 grew better at 1-atm (0.1 MPa) than the in situ pressure (20.7 MPa) (data not shown). Thus, strain 5710 was found to be piezotolerant.
To investigate the phylogenetic relationship of strains 2D2 and 5710, 16S rRNA genes amplified by PCR were subjected to a sequence analysis. Strain 2D2 exhibited the highest levels of identity (98.0–99.6%) to the five Moritella species which were isolated from the deep-sea sediment, mud, and water samples from depths of 1,200–6,356 m [19–21] and Atlantic salmon [22,23] (Fig. 2). Strain 5710 also exhibited the highest levels of identity (97.4–98.5%) to the same five Moritella species.
A phylogenetic tree based on 16S rRNA gene sequences is shown in Fig. 2. Strains 2D2 and 5710 formed one cluster with the five Moritella species described above and did not cluster with five Shewanella species. Therefore, we placed strains 2D2 and 5710 in the genus Moritella, as Moritella sp. strain 2D2 and Moritella sp. strain 5710, respectively, despite Hamamoto et al. who isolated strain 5710 reported that strain 5710 was classified to the genus Vibrio as mentioned above.
From these results, it is considered that strains 2D2 and 5710 belong to the genus Moritella and that they adapt to low temperatures at the deep-sea floors but to different pressures (strains 2D2 and 5710 adapt to high-pressure and 1-atm, respectively).
3.3Nucleotide and amino acid sequence of MDH gene from strain 2D2
The complete DNA and deduced amino acid sequences were submitted to the DDBJ database on June of 2003 by Nakayama (DDBJ Accession No. AB097559) (the DNA sequence is not shown but the amino acid sequence is shown in Fig. 3). The MDH gene of strain 2D2 consisted of 939 bp and was coding 312 amino acid residues. The MDH had an estimated molecular weight of 31,940. The MDH exhibited high sequence similarities with MDHs of six 1-atm adapted and a facultatively piezophilic bacterium which belong to the γ-subclass of proteobacteria. The sequence identity was 94.9% for Vibrio sp. strain 5710 (P48364), 79.4% for Shewanella oneidensis (P82177), 78.7% for E. coli (P06994), 77.8% for Photobacterium profundum strain SS9 (P37226), 77.4% for Salmonella typhimurium (P25077), 76.1% for Vibrio cholerae (Q9KUT3), and 72.8% for Haemophilus influenzae (P44427). When the MDH was compared to these seven MDHs exhibiting high similarities, it was surprisingly found there were two positions with 2D2-specific substitutions, a valine to alanine substitution at position 180 and a glutamine to histidine substitution at position 229 (Fig. 3). It is considered that the position 180 is close to the active site His-177, and that the position 229 is a functionally important residue involved in subunit interaction [6,25–28].
The 5710 MDH which exhibited the highest sequence similarity (94.9%) also had a valine at position 180 and a glutamine at position 229 like the other six bacteria that belong to the γ-subclass of proteobacteria (Fig. 3). The 2D2-specific substitutions at position 180 and position 229 are not the genus Moritella-specific substitution. Strain 2D2 adapts to high pressure and strain 5710 adapts to 1-atm like the remaining five strains except for facultatively piezophilic Photobacterium profundum strain SS9 as mentioned above.
3.4Catalytic properties of 2D2 MDH and 5710 MDH
Recombinant MDHs from E. coli BL21(DE3)/p2D2MDH and E. coli BL21(DE3)/p5710MDH were purified to homogeneity as confirmed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE)  (data are not shown). The optimum temperature, the optimum pH, and the Km value (NADH) for catalytic activity of the purified recombinant 5710 MDH (r5710MDH) were almost same values as those of the native 5710 MDH reported by Ohkuma et al. . As shown in Table 1, the optimum temperature for catalytic activity of r2D2MDH was approximately 5°C higher than that of r5710MDH. The midpoint of denaturation (Tm value) of r2D2MDH, which was estimated from the thermal denaturation curves, was approximately 3°C higher than that of r5710MDH. Thus, the difference of heat-stability between the MDHs of the obligately piezophilic strain 2D2 and the psychrophilic strain 5710 was found.
3.5Heat denaturation of wild-type and mutant MDHs at 40°C
As it is considered that the position 229 is a functionally important residue involved in subunit interaction of MDHs [6,25–28] as mentioned above, at first site-directed mutageneses of the MDH genes of strains 2D2 and 5710 were performed at position 229. Then thermal stabilities of resulting wild-type and mutant MDHs were compared. As shown in Fig. 4, the heat resistance of r2D2MDH (wild-type) was higher than that of r5710MDH (wild-type). r2D2MDH was more resistant to heat denaturation than rH229Q2D2MDH (mutant) (Fig. 4(A)). In contrast, r5710MDH was less resistant to heat denaturation than rQ229H5710MDH (mutant) (Fig. 4(B)). These results indicated that at least the His residue at position 229, one of 2D2-specific substitutions, may contribute to the thermal stability.
3.6Kinetic properties of wild-type and mutant MDHs at 0.1 and 62.1 MPa
Although it has been shown that one of the enzyme adaptations to high pressure of deep-sea animals is characterized by more pressure insensitive Kms [1,2], the Km value of r2D2MDH was more pressure sensitive than that of r5710MDH (Fig. 5(A)). As shown in Fig. 5, however, the kinetic parameters at 62.1 MPa, the Km and kcat values for NADH of rH229Q2D2MDH, were lower than those of r2D2MDH. In contrast, the Km and kcat values for NADH of rQ229H5710MDH appeared to be higher than those of r5710MDH.
Based on these results, we suggest that the His residue at position 229 of 2D2MDH which is a functionally important residue involved in subunit interaction may play a role in the thermal stability and the MDH function at high pressure.
An additional study of the Ala at position 180, another 2D2-specific substitution, with site-directed mutagenesis is now in progress. To elucidate the mechanism of adaptation to high hydrostatic pressure in the MDHs, further studies, including kinetic experiments at several high pressures, crystallization experiments, and/or deliberated site-directed mutagenesis, are necessary.
This study was partially supported by a grant (BRP-00-I-A-8) from the Ministry of Agriculture, Forestry, and Fisheries to A.N. and by the Sasakawa Scientific Research Grant (13-190) from The Japan Science Society to R.S. We thank Drs. Yuichi Nogi and Chiaki Kato, Japan Marine Science and Technology Center, and Dr. Moriya Ohkuma, The Institute of Physical and Chemical Research (RIKEN) for providing Vibrio sp. strain 5710 (Moritella sp. strain 5710).