OUP user menu

The nitrogen-fixing symbiotic bacterium Mesorhizobium loti has and expresses the gene encoding pyridoxine 4-oxidase involved in the degradation of vitamin B6

Baiqiang Yuan, Yu Yoshikane, Nana Yokochi, Kouhei Ohnishi, Toshiharu Yagi
DOI: http://dx.doi.org/10.1111/j.1574-6968.2004.tb09537.x 225-230 First published online: 9 January 2006

Abstract

The gene product of mll6785 of a nitrogen-fixing symbiotic bacterium Mesorhizobium loti MAFF303099 was identified as pyridoxine 4-oxidase, the first enzyme in the vitamin B6-degradation pathway. The gene was cloned and ligated into pET-21a(+). Escherichia coli BL21(DE3) was co-transformed with the constructed plasmid plus pKY206 containing groESL genes encoding chaperonins. The overexpressed protein was purified to homogeneity by the ammonium sulfate fractionation and three chromatography steps. The enzymatic properties of the purified protein, such as Km values for pyridoxine (213 ± 19 μM) and oxygen (78 ± 10 μM), were compared to those of pyridoxine 4-oxidase from two bacteria with known vitamin B6-degradation pathway. M. loti grown in a Rhizobium medium showed the enzyme activity. The results suggest that M. loti also contains the degradation pathway of vitamin B6.

Keywords
  • Vitamin B6
  • Pyridoxine 4-oxidase
  • Degradation pathway

1 Introduction

Two different but related degradation pathways of free forms of vitamin B6, pyridoxine, pyridoxal, and pyridoxamine, exist in bacteria that can use pyridoxine as a sole source of carbon and nitrogen [1]. In pathway I, which has been found in Pseudomonas sp. MA-1 [2] and Microbacterium luteolum YK-1 [3], pyridoxine is first oxidized to pyridoxal by pyridoxine 4-oxidase (EC 1.1.3.12) and then degraded through the following seven steps to succinic semialdehyde, ammonia, and carbon dioxide. In pathway II, which has been found in Pseudomonas sp. IA and Arthrobacter sp. Cr-7 [4], pyridoxine is first oxidized to isopyridoxal by pyridoxine 5-oxidase and then degraded in 4 subsequent steps to 2-hydroxymethyl succinic semialdehyde, acetic acid, ammonia, and carbon dioxide. The primary structures of two enzymes in the two pathways have been reported [3, 5].

Pyridoxine 4-oxidase, the first step enzyme of the pathway I, from M. luteolum YK-1 uses FAD as the coenzyme [3] and belongs to the GMC oxidoreductase family [6]. Interestingly, by searching for genes similar in sequence in other organisms pyridoxine 4-oxidase showed 66% identity in amino acid sequence with an mll6785 gene product in Mesorhizobium loti MAFF303099 [3]. The mll6785 gene is located at coordinates of 5585874–5584312 (a complementary direction) of a chromosomal DNA. Predicted functions of other genes around the mll6785 gene do not seem to be related to pyridoxine degradation, except an mlr6788 gene coordinated at 5587799–5588878 (a direct direction) encoding 2-methyl-3-hydroypyridine-5-carboxylic acid oxygenase (EC 1.14.12.4). This enzyme catalyzes the seventh reaction step in the pyridoxine degradation pathway I. Because pyridoxine 4-oxidase has been found so far only in the microorganisms that can use pyridoxine as carbon and nitrogen sources, why the nitrogen-fixing symbiotic microorganism might contain pyridoxine 4-oxidase is unclear. The biochemical roles of the enzyme both in the soil and in the nodule should be elucidated. As the first step, here, we have examined whether the mll6785 gene encodes pyridoxine 4-oxidase or not. The gene was expressed in Escherichia coli, and the gene product was purified and characterized. The mll6785 gene was demonstrated to encode pyridoxine 4-oxidase.

2 Materials and methods

2.1 Bacterial strains, plasmids, chemical, and cultivation

Escherichia coli strains, JM109 (endA1, gyrA96, thi, hsdR17, supE44, relA1, Δ(lac-proAB), recA1, F′[traD36, proAB+, laclq, lacZΔM15]) and BL21(DE3) (E. coli B, F, dcm, ompT, hsdS (rBmB), gal, λ(DE3)), were purchased from Takara Bio (Shiga, Japan). M. loti MAFF303099 [7] was obtained from MAFF GenBank (Tsukuba, Japan) and was grown in the Rhizobium medium (pH 6.8) consisted of 0.5% (w/v) tryptone, 0.3% (w/v) yeast extract and 0.1% (w/v) CaCl2· 2H2O. The PN synthetic medium was M9 minimal medium without d-glucose and ammonium chloride [8] supplemented with 0.1% (w/v) pyridoxine hydrochloride and 100 ng/ml biotin and 5 ng/ml cobalt chloride. Cloning and expression vectors, p3T (ampicillin resistance, TA cloning vector) and pET-21a(+) (ampicillin resistance, T7 promoter), were obtained from MOBiTec (Goettingen, Germany) and Takara Bio, respectively. Plasmid pKY206 was provided by Ashiuchi (Kochi University) and Kawata (Tottori University) [9]. Horseradish peroxidase, sodium pyruvate and oxaloacetic acid were obtained from Wako Pure Chemical Industries (Osaka, Japan). Malate dehydrogenase and NADH were obtained from Oriental Yeast (Tokyo, Japan). V-8 protease was from Sigma Aldrich Japan (Tokyo, Japan). All other chemicals were of analytical grade.

2.2 Cloning and expressing of the mll6785 gene

The mll6785 gene was PCR-amplified from chromosomal DNA of M. loti MAFF303099 prepared by the reported method [10]. PCR was performed in a 50-μl reaction mixture containing Ex Taq buffer, 0.2 mM dNTPs, 2.5 mM MgCl2, 1 μM each of primers, and 1.25 U Ex Taq DNA polymerase (Takara Bio). The primers used were 5′-CATATGGCAGATGGAGTGAGGATG-3′ (mll6785-F) and 5′-GTCGACTTAGTACTGTCGGGCGA-3′ (mll6785-R). The bold and underlined letters correspond to NdeI and SalI restriction sites, respectively. The primer mll6785-F corresponds to the 5′ end of the mll6785 gene including the initiation codon and mll6785-R anneals to the 3′ end of the complementary strand including the termination codon. PCR conditions were an initial denaturation at 96 °C for 5 min, then 30 cycles of 1 min at 95 °C, 1 min at 55 °C and 3 min at 72 °C, and final 10 min at 72 °C. The amplified 1.5-kb DNA fragment was cloned into the p3T vector, to construct p3T-mll6785. The nucleotide sequence of amplified DNA fragment cloned in p3T-mll6785 was analyzed using an ABI PRISM 3100-Avant genetic analyzer (Applied Biosystems, Foster City, CA). The gene fragment was extracted by digestion with NdeII and SalI and re-cloned into a pET-21a(+), to make pET-mll6785. E.coli BL21(DE3) was co-transformed with the pET-mll6785, and pKY206 carrying the groESL genes. The transformed cells were grown aerobically in 5 ml of LB medium containing ampicillin (50 μg/ml) and tetracycline (10 μg/ml) at 23 °C until the absorbance at 600 nm reached to 1.0. The pre-cultured cells were transferred into the 100 ml of the same medium, and the culture was incubated under the same condition for 60 h. The bacterial cells were harvested by centrifugation at 8000g for 10 min at 4 °C.

2.3 Purification of the recombinant pyridoxine 4-oxidase

All steps were done at 4–10 °C.

  • Step 1: The harvested cells (5 g) were suspended in 25 ml of buffer A (20 mM potassium phosphate buffer (pH 8.0) containing a stabilizing reagent (5 μM FAD, 1% 2-mercaptoethanol, 10% glycerol, 1 mM EDTA, and 0.01% Tween 20)) supplemented with 1 mM phenylmethylsulfonyl fluoride. The suspension was sonicated for 1 min 5 times with a 2-min interval between each step on ice with a model W-220 sonicator (Heat Systems-Ultrasonics, Farmingdale, NY). The cell extract was obtained by centrifugation at 10,000g for 30 min at 4 °C. Proteins precipitated fractionally with 35–75% saturated ammonium sulfate was resuspended in 20 ml of buffer A.

  • Step 2: The suspension was applied to a QA52 column (Whatman International, Maidstone, England, 3.0 × 19.0 cm) equilibrated with Buffer A, and then the column was washed with buffer A until the absorbance of the eluate at 280 nm decreased to 0.1. The enzyme activity was detected at 0.2 M NaCl when proteins were eluted with a linear gradient (0–0.5 M NaCl).

  • Step 3: The eluted enzyme solution (45 ml) was concentrated to 8.6 ml by the 75% saturated ammonium sulfate precipitation, dialyzed thoroughly against buffer B (5 mM potassium phosphate buffer (pH 8.0) containing the stabilizing reagent), and applied to a hydroxylapatite column (Wako Pure Chemical Industries, 1.5 × 5.0 cm) equilibrated with buffer B. The column was washed as described above. The enzyme activity was detected at 20mM potassium phosphate when proteins were eluted with a linear gradient (5–50 mM phosphate).

  • Step 4: The eluted enzyme solution was concentrated as described above. The enzyme solution (1 ml) was applied to an Ultrogel AcA 34 column (Biosepra, Cergy-Saint-Christophe, France, 2 × 100 cm) equilibrated with buffer A containing 0.1 M NaCl.

2.4 Molecular mass measurement

Purity of the enzyme and the subunit molecular mass were analysed on SDS–PAGE by the method of Laemmli [11]. The molecular mass of the native enzyme was estimated by a gel filtration (Hiprep 16/60 Sephacryl S-300 column, Amersham Biosciences, Piscataway, NJ) equilibrated with Buffer A containing 0.15 M NaCl at a flow rate of 1.0 ml min−1 with FPLC at 4 °C. A calibration curve was made from the elution pattern of lactate dehydrogenase (140 kDa), glutamate pyruvate transaminase (115 kDa), malate dehydrogenase (74 kDa), and cytochrome c (12 kDa).

2.5 Amino acid sequencing

Purified enzyme (2.0 μg) was separated on SDS–PAGE and transferred to a PVDF membrane with a semi-dry transfer apparatus (Bio-Rad, Hercules, CA). The N-terminal amino acid sequence was analysed with a Model 492 protein sequencer (Applied Biosystems). For the internal amino acid sequences, purified enzyme (300 μg), which had been dialysed against 100 mM Tris–HCl (pH 7.8) containing 2 M urea for 1 day, was digested at 25 °C with 12 μg of V-8 protease. The digested enzyme fragments were separated by HPLC with 0–80% gradient of acetonitrile containing 0.1% trifluoroacetic acid.

2.6 Enzyme assay and identification of the reaction product

Enzyme activity was measured by two methods.

  • Method 1: The reaction mixture (1 ml) containing 0.1 M Tris–HCl (pH 8.0), 5 mM pyridoxine, 5 μM FAD, and the enzyme was incubated aerobically at 30 °C for 10 min. The absorbance at 415 nm of the Schiff base formed by the reaction between pyridoxal and Tris base was measured.

  • Method 2 was the phenylhydrozine method described previously [3]. Method 1 that is convenient but less accurate was used for routine enzyme assays. Method 2 is a bit complicated but very accurate, so Method 2 was mainly used for determining enzymatic properties. One unit of enzyme was defined as the amount that catalyses the formation of 1 μmol of pyridoxal per min. The substrate specificity was identified with 5–10-fold higher amounts of enzyme than those used in the standard assay. Protein was measured by the protein-dye method [12].

The enzyme reaction was performed for 60 min in the reaction mixture with 0.0075 units of purified enzyme, and the products were analyzed with the reversed-phase isocratic HPLC by the method described previously [13].

3 Results and discussion

3.1 Cloning and expression of the mll6785 gene in E. coli

The mll6785 gene of M. loti MAFF303099 was PCR-amplified, cloned into the pET vector, and expressed in E.coli BL21(DE3). The transformant cells showed the substantial enzyme activity of pyridoxine 4-oxidase (Fig. 1A, lane 1). When the chaperonins, GroEL and GroES, were co-expressed to assist protein folding under the cold stress conditions, the enzyme activity increased by a factor of more than ten (Fig. 1A, lane 2). Without chaperonins, almost all the mll6785 gene product formed the inclusion body, and no soluble forms of gene product were observed. With the aid of the chaperonins, although it remained in the inclusion body, detectable amounts of the mll6785 gene product were found as a soluble form (Fig. 1B). These results showed that GroEL and GroES helped the mll6785 gene product to be folded properly at 23 °C. The co-expression of the chaperonins did not increase the soluble form when the co-transformant cells were cultured at 37 °C. Thus, the recombinant pyridoxine 4-oxidase belongs to the class of proteins [14] that require both of the chaperonins and the cold stress to increase their soluble forms in the host cells.

Figure 1

Expression of the mll6785 gene product. (A) Specific enzyme activities of E. coli transformants containing pET-mll6785 alone (lane 1) and both pET-mll6785 and pKY206 (lane 2). Transformant cells were incubated at 23 °C for 60 h, harvested, and sonicated. The cell extract was used for enzyme assay. The value was an average ± SD of three measurements. (B) The SDS–PAGE profile of the crude cell extract (10 μg) of BL21(DE3)/pET-mll6785/pKY206 (CE). Sd, molecular mass markers (97.4, 66.2, 45, 31, 21.6, and 14.4 kDa, from top to bottom).

During the incubation of these transformant cells, we found that the enzyme activity decreased with time. In order to keep the enzyme activity high, we had to transform cells just before use and use fresh transformants. Although we had not checked the plasmid stability, the plasmid or the mll6785 gene itself might be lost during incubation. As observed unstable gene expression in E. coli[15, 16], the mll6785 gene product could be toxic to E. coli cells.

3.2 Purification and amino acid sequences of the mll6785 gene product

The mll6785 gene product was purified to homogeneity from the cell extract of the co-transformant cells by the ammonium sulfate fractional precipitation followed by three column chromatography steps (Table 1). Proteins from each step were analyzed on SDS–PAGE (Fig. 2). The purified enzyme showed a single protein band with a molecular mass of 53,000 ± 1000 Da (the average and SD of two measurements) on SDS–PAGE (Fig. 2). The molecular mass of the native enzyme was 54,500 ± 700 Da (average and SD of two measurements) by gel filtration. These values were in good agreement with a predicted molecular mass (55,179 Da with 520 amino acids) deduced from the nucleotide sequence. The results showed the mll6785 gene product was a monomeric enzyme.

View this table:
Table 1

Purification of pyridoxine 4-oxidase of M. loti MAFF303099

StepTotal activity (U)Total protein (mg)Specific activity (U/mg)Yield (%)
Cell extract814.0630.01.3100
Ammonium sulfate662.5442.11.581.4
QA52470.043.29.757.7
Hydroxylapatite179.918.715.222.1
Gel filtration74.43.8919.49.1
Figure 2

Patterns of SDS–PAGE of preparations obtained at each purification step. Lane A, the cell extract (16.3 μg); lane B, the ammonium sulfate precipitate (22.1 μg); lane C, the preparation from QA52 (3.7 μg); lane D, hydroxylapatite (1.9 μg); lane E, gel filtration (1.5 μg); Sd, molecular mass makers.

The N-terminal and internal amino acid sequences of the purified enzyme were determined and compared with the deduced amino acid sequences of the mll6785 gene product (Fig. 3). The amino terminal amino acid residue was alanine. Seventy nine out of 520 residues were determined and the sequences coincided with deduced ones, indicating that the purified enzyme was the mll6785 gene product.

Figure 3

N-terminal and internal amino acid sequences of the mll6785 gene product. The purified enzyme was digested with V-8 protease to generate peptide fragments. Amino acid sequences of the purified enzyme and peptide fragments determined were underlined.

3.3 Pyridoxine 4-oxidase activity of the mll6785 gene product and in the crude extract from M. loti, and growth of M. loti in the synthetic medium

The reaction product of pyridoxine was identified as pyridoxal by the reversed-phase isocratic HPLC. In the enzyme reaction, H2O2 was produced in equimolecular amounts to pyridoxal. Thus, mll6785 gene product of M. loti MAFF303099 is a pyridoxine 4-oxidase.

Mesorhizobium loti, as M. luteolum YK1, could grow in the synthetic medium containing pyridoxine as a sole carbon and nitrogen source, suggesting that it contains the degradation pathway (probably pathway I) of pyridoxine.

The pyridoxine 4-oxidase activity in the crude extract from M. loti cells, which were grown in the Rhizobium medium, was 0.63 ± 0.19 mU/mg, showing that the enzyme gene is inherently expressed in M. loti cells.

3.4 Enzymatic properties of M. loti pyridoxine 4-oxidase

Mesorhizobium loti pyridoxine 4-oxidase showed high hydrogen donor specificity: only three pyridinium compounds were used as the substrate with very low reactivity (Table 2). The enzyme showed no reactivity toward pyridoxal 5′-phosphate, the coenzyme form of vitamin B6. Pyridoxine showed a Michaelis–Menten kinetics, and its apparent Km value was 213 ± 19 μM. As the hydrogen acceptor, O2 and 2,6-dichloroindophenol were good substrates: they also showed the Michaelis–Menten kinetics, and their Km values were 78 ± 10 and 56 ± 7 μM, respectively. The optimum pH was 8.0–8.5 when the enzyme activity was measured in a reaction mixture containing 0.1 M universal buffer (pH 2.6–12.0) [17]. The optimum temperature measured in the standard buffer was 40 °C: the reactivity was decreased to 40% of the maximum at 50 °C. The activation energy calculated from an Arrhenius plot was 7240 ± 330 cal/mol.

View this table:
Table 2

Pyridoxine 4-oxidase activity toward several substrates as a hydrogen donor

SubstratesRelative activity (%)a
Pyridoxine100
PyridoxamineND
Pyridoxal 5′-phosphateND
Pyridoxamine 5′-phosphateND
Pyridoxine 5′-phosphateND
Benzyl alcoholND
2-Hydroxybenzyl alcoholND
3-Hydroxybenzyl alcoholND
2,6-Dihydroxypridine0.040 ± 0.027
3-Pyridinemethanol0.036 ± 0.020
4-Pyridinemethanol0.121 ± 0
  • a The relative enzyme activity was calculated as percentages of the activity toward pyridoxine. ND shows no reactivity.

3.5 Comparison of pyridoxine 4-oxidases

The enzymatic properties of the M. loti enzyme were compared with those of pyridoxine 4-oxidases from M. luteolum[3] and Pseudomonas sp. MA-1 [18] (Table 3). The enzymes of M. loti and M. luteolum are monomeric. Although the optimum pH and temperature were shared among enzymes, the affinities for pyridoxine and 2,6-dichloroindophenol of the M. loti enzyme were 4–5-fold lower than those of the M. luteolum and Pseudomonas enzymes. The M. loti enzyme showed 2.6-fold higher affinity for oxygen than the M. luteolum enzyme. The Km of FAD for the M. loti enzyme could not be determined because its affinity for FAD was so high that the preparation of an apoenzyme was unsuccessful. A tertiary structure of glucose oxidase from Aspergillus niger[6], a representative of the GMC oxidoreductase family, has been determined. While the FAD-binding domains (2α-helixes and 8β-sheets) in the glucose oxidase are conserved in both M. loti and M. luteolum pyridoxine 4-oxidases, minor differences in the amino acid residue(s) between the pyridoxine 4-oxidases are found in the βA1, βA3, βA4, βA5, βB2 and βB3 domains. The high affinity of the M. loti enzyme for FAD may be attributable to the differences. Elucidation of the tertiary structure of M. loti enzyme is under way.

View this table:
Table 3

Comparison of pyridoxine 4-oxidases

M. loti MAFF303099M. luteolum YK-1aPseudomonas sp. MA-1b
Molecular mass (Da)54,500 ± 70053,000 ± 1,000?
Optimum pH8.0–8.57.5–8.07.5–8.0
Optimum temperature (°C)4040?
Activation energy (cal/mol)7,240 ± 3307,890 ± 360?
Km for pyridoxine (μM)213 ± 1954.5 ± 5.743
Km for DCIPc (μM)56 ± 79.7 ± 1.78.3
Km for oxygen (μM)78 ± 10206.6 ± 18.8?
Km for FAD (μM)UM0.16 ± 0.012.7
  • a Data were cited from [3].

  • a Data were cited from [2].

  • c DCIP stands for 2,6-dichloroindophenol. UM shows un-measurable.

3.6 Concluding remarks

Here, it was confirmed that the pyridoxine 4-oxidase-coding gene (mll6785) is contained in the nitrogen-fixing symbiotic bacterium M. loti and expressed in M. loti cells. The existence of the enzyme suggests that M. loti contains the degradation pathway of vitamin B6. Because the bacterium also contains the deduced 2-methyl-3-hydroxypyridine-5-carboxylic acid oxygenase [5] that catalyzes the seventh reaction step of the degradation pathway I, the degradation pathway in the bacterium is probably the pathway I. The presence of the other enzymes in the degradation pathway, biochemical roles and the mechanism of inheritance of this pathway in the symbiotic bacterium should be studied.

Acknowledgements

We acknowledge Dr. Makoto Ashiuchi, Kochi University for helpful discussion and Dr. Yasushi Kawata, Tottori University for a gift of pKY206 vector.

References

  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
View Abstract