NAD(P)-Dependent Aldehyde Dehydrogenases Induced during Growth of Ralstonia eutropha Strain Bo on Tetrahydrofurfuryl Alcohol

Abstract

Different aldehyde dehydrogenases (AlDHs) were formed during growth of Ralstonia eutropha Bo on tetrahydrofurfuryl alcohol (THFA). One of these enzymes, AlDH 4, was purified and characterized as a homodimer containing no prosthetic groups, showing a strong substrate inhibition, and having an N-terminal sequence similar to those of various NAD(P)-dependent AlDHs. The conversion rate of THFA by the quinohemoprotein THFA dehydrogenase was increased by AlDH 4.

The cyclic xenobiotic ether tetrahydrofurfuryl alcohol (THFA) is a versatile solvent which is used for various purposes in industry and thus becomes released into the environment. However, information about the biological degradation of this compound is scarce.

We have previously isolated a strain identified as Ralstonia eutropha that is capable of growing on high concentrations of THFA as a sole source of carbon and energy (10). The organism, designated strain Bo, induces a quinohemoprotein, THFA dehydrogenase (THFA-DH), during growth on this substrate, catalyzing the oxidation of the alcohol via the aldehyde to the corresponding carboxylic acid (10). From this observation, it might be concluded that both oxidation steps were catalyzed by THFA-DH. However, in Comamonas testosteroni and Pseudomonas aeruginosa, a gene encoding a protein with a high similarity to NAD(P)-dependent aldehyde dehydrogenases (AlDHs) was discovered adjacent to the gene encoding a pyrroloquinoline quinone (PQQ)-dependent alcohol dehydrogenase (ADH) (5, 9). Thus, the NAD(P)-dependent AlDHs might also be involved in formation of the carboxylic acid. It was shown previously that R. eutropha strain H16 induces different NAD(P)-dependent ADHs and AlDHs in response to the growth conditions used (2, 3, 7, 8).

To obtain information about the NAD(P)-dependent AlDHs expressed by R. eutropha Bo during growth on different alcohols, we investigated the induction patterns of these enzymes. An NAD(P)-dependent AlDH specifically induced during growth on both THFA and n-pentanol was purified to apparent homogeneity and also characterized with respect to its involvement in the formation of carboxylic acids.

Growth of R. eutropha strain Bo (DSM 11098) on THFA, furfuryl alcohol, n-pentanol, ethanol, 1,5-pentandiol, and 2-propanol and preparation of cell extracts were carried out in principle as described previously (10, 11). Extracts used for investigation of the induction patterns of AlDHs were analyzed directly by polyacrylamide gel electrophoresis (PAGE) or previously fractionated on Q-Sepharose.

Sodium dodecyl sulfate (SDS)-PAGE was performed as described previously (10). A gradient from 6 to 16% polyacrylamide in 0.14 M Tris-borate buffer, pH 8.9, was used for preparation of native gradient gels. The gels were run in 30 mM Tris-borate buffer, pH 8.9, at a constant current of 20 mA at 4°C for 6 h. Native nongradient gels containing 7% polyacrylamide were run at 30 mA for 80 min under the same conditions. Staining of gels with Coomassie blue or silver was carried out as reported previously (10). Activity staining of NAD(P)-dependent AlDHs was performed by incubating native gels in 100 mM potassium phosphate buffer (pH 7.6) containing 0.08% NAD(P), 0.04% Nitro Blue Tetrazolium, 0.003% phenazine ethosulfate, and 0.1 to 0.4% of the corresponding aldehyde at room temperature.

Crude extracts prepared from R. eutropha strain Bo grown on different alcohols contained several NAD(P)-reducing enzyme activities. Thus, the induction pattern of AlDHs was analyzed after separation of these activities by chromatography on Q-Sepharose. The extracts from the corresponding cells (80 mg of total protein) were applied to a Q-Sepharose column previously equilibrated with 50 mM Tris-HCl (pH 8.0) containing 1 mM dithiothreitol (DTT) and 2 mM EDTA (buffer A). The column was washed with buffer A, and bound protein was eluted by a stepwise gradient (each step, 50 mM KCl) from buffer A to 0.5 M KCl in buffer A. Fractions containing AlDH activities were pooled, dialyzed against buffer A, and concentrated using an Amicon spin filter. The pools were stored at −20°C, or at −80°C in cases of longer periods.

A bacterium which grows on an alcohol should express a high AlDH activity to convert the potentially toxic aldehyde to the nontoxic carboxylic acid. Although THFA-DH from R. eutropha Bo is able to oxidize several aldehydes, we had indications that the strain expresses additional NAD(P)-dependent AlDHs to cope with the aldehyde generated. Four AlDHs with molecular masses ranging from 240 to 100 kDa were detected by activity staining and designated according to their migration behavior as AlDH 1 to AlDH 4 (Fig. 1). AlDH 1 (240 kDa) was detected in all extracts using acetaldehyde as the assay substrate (Fig. 1A). In contrast, the induction pattern of AlDH 2 (185 kDa) indicated that the expression of the enzyme activity was highly specific only during growth on THFA (Fig. 1B) and that no activity was expressed with acetaldehyde as the substrate (Fig. 1A). Unfortunately, the activity of AlDH 2 was extremely labile and, thus, no detailed characterization of this enzyme could be performed. AlDH 3 (170 kDa) was induced mainly after growth on furfuryl alcohol, ethanol, 2-propanol, and, to a much lower extent, THFA. AlDH 4 (110 kDa) was detected only in cells grown on THFA and n-pentanol and showed activity with both acetaldehyde and pentanal (Fig. 1).

FIG. 1.

Induction of different NAD-dependent AlDH activities after growth of R. eutropha Bo on various alcohols. Activity staining was performed with extracts separated on Q-Sepharose and by native PAGE (7% polyacrylamide). Shown is activity staining with acetaldehyde (0.4%) (A) and pentanal (0.08%) (B) as the substrate. The applied samples (15 μg) were prepared from cells grown on THFA (lane 1), furfuryl alcohol (lane 2), n-pentanol (lane 3), ethanol (lane 4), 1,5-pentandiol (lane 5), 2-propanol (lane 6), and succinate (lane 7).

As far as the degradation of THFA is concerned, AlDH 2 seemed to be the most interesting enzyme because it was expressed only in extracts obtained from cells grown on THFA. However, due to the observed low stability of AlDH 2 activity, the enzyme could not at this point be purified. AlDH 4 was also specifically induced during growth on THFA and n-pentanol; therefore, this enzyme was investigated in more detail.

The standard assay for determination of AlDH 4 activity at 30°C contained 5 mM NAD and 4 mM acetaldehyde in 50 mM MOPS (morpholinepropanesulfonic acid)-NaOH, pH 8.2. The kinetic parameters of AlDH 4 were estimated in principle as described for THFA-DH (11). The purification of AlDH 4 was performed at 4°C. Cell extract from THFA-grown cells of R. eutropha strain Bo (450 mg of total protein) was supplied to a Q-Sepharose column (2.4 by 2.2 cm) equilibrated with 20 mM Tris-HCl (pH 7.5) containing 1 mM DTT and 2 mM EDTA (buffer B). The column was washed with 3 volumes of buffer B, and bound protein was eluted by a linear gradient from 0 to 0.5 M KCl in buffer B. Fractions containing AlDH 4 activity (acetaldehyde as the substrate) were pooled, concentrated, and applied to a Superdex 200 column (1.6 by 60 cm) equilibrated with 50 mM Tris-HCl (pH 7.5), containing 150 mM KCl, 1 mM DTT, and 2 mM EDTA. Active fractions were concentrated by the addition of ammonium sulfate (70%) and dialyzed against buffer B. The obtained preparation was applied to a Cibracon Blue 3GA column (6.9 by 3.6 cm) equilibrated with buffer B, and bound protein was eluted by a linear gradient from 0 to 0.6 M KCl in buffer B. The fractions containing AlDH activity were analyzed by SDS-PAGE (12% polyacrylamide) and pooled, depending on their apparent purity. The obtained pool was dialyzed against buffer B, concentrated, analyzed for AlDH activities by activity staining of native polyacrylamide gels, and stored at −80°C.

AlDH 4 was purified 242-fold from crude extracts of strain Bo prepared from cells grown on THFA as a sole source of carbon and energy (Fig. 2). The purified protein exhibited a specific activity of 12.9 U mg of protein⁻¹. The unaltered migration behavior of AlDH 4 in cell extracts and after purification indicated that the native structure of the protein was unchanged under the conditions used (Fig. 2). A significant increase in activity after chromatography on Q-Sepharose and especially Superdex 200 might be explained by separation of NADH-oxidizing enzymes and/or by a strong inhibition of AlDH 4 activity in crude extracts by present aldehydes (see below). The activity of the obtained enzyme preparation was stable at −20°C if it was stored in buffer B.

FIG. 2.

(A) SDS-PAGE (12% polyacrylamide) and Coomassie blue staining of purified AlDH 4 from R. eutropha strain Bo. Lane M, marker proteins; lane 1, extract from THFA-grown cells; lane 2, purified AlDH 4. (B) Native PAGE (7% polyacrylamide) and activity staining with acetaldehyde (0.4%) as the substrate. Lane 1, extract from THFA-grown cells; lane 2, purified AlDH 4.

The molecular mass of AlDH 4 as analyzed by matrix-assisted laser desorption ionization mass spectrometry (10) was estimated to be 49.5 kDa, whereas a value of 52 kDa was obtained by SDS-PAGE analysis. Native PAGE and gel filtration experiments revealed a native molecular mass of 100 kDa, indicating a dimeric native structure.

Although AlDH 4 was not induced during growth of R. eutropha strain Bo on ethanol, the highest V_max value in the standard assay was obtained using acetaldehyde as the substrate (Table 1). Increasing the chain length of the aldehyde decreased the enzyme activity (Table 1). This decrease was due to a strong substrate inhibition which increased with the chain length of the aldehyde. The apparent K_m value for acetaldehyde was 1.7 mM, and the corresponding K_i value was determined to be 6.9 mM. Quite different results were obtained if the kinetic parameters were investigated for pentanal. This substrate caused an extremely strong substrate inhibition. Therefore, the exact K_i and K_m values could not be determined. THFA is not commercially available, and thus an investigation into whether AlDH 4 was actually able to convert it could not be carried out. However, THFA was a quite selective inducing substrate; thus, AlDH 4 should convert the corresponding aldehyde. The K_m value for NAD was estimated to be 37 μM with acetaldehyde as the substrate. No significant changes in enzyme activity were obtained if NAD was replaced by NADP. No activity of AlDH 4 was determined with the corresponding alcohols as substrates. A narrow pH optimum was observed for conversion of the aldehydes in the range of pH 7.5 to 8.0.

TABLE 1.

Aldehydes converted by AlDH 4 from R. eutropha Bo and substrate concentrations at which apparent V_max values were determined

Substrate	Vmax (%)a	Concn (mM)
Acetaldehyde	100	4
Propionaldehyde	88	0.04
Butyraldehyde	62	0.004
Pentanal	55	0.004
Furfurylaldehyde	4	0.4

^a100% = 12.9 U mg of protein⁻¹. 

Spectra recorded with the purified enzyme showed an absorption maximum at 279 nm and a shoulder at about 290 nm (data not shown). From these spectral features, we concluded that AlDH 4 did not contain a prosthetic group.

The N-terminal sequence of AlDH 4 from R. eutropha strain Bo determined with the homogeneous protein as described previously (10) was analyzed by comparison to sequences deposited in the EMBL and GenBank databases. The highest sequence similarity with up to 57% identity was obtained for NAD(P)-dependent AlDHs from various sources (data not shown). Interestingly, two AlDHs of eukaryotic origin, mitochondrial AlDH from Leishmania tarentolae (53%) and a putative AlDH from Agaricus bisporus (57%), showed a high degree of identity. A low identity of only 20% was obtained for the N-terminal sequence of the acetaldehyde dehydrogenase II from R. eutropha H16 induced during growth on ethanol and acetoin (4) and for the acetaldehyde dehydrogenase ExaC from P. aeruginosa (data not shown). The latter protein was suggested to be a component of the ethanol oxidation system in this organism (5). In summary, the obtained results clearly indicated that AlDH 4 from R. eutropha Bo is a member of the AlDH superfamily.

The properties of AlDH 4 determined during the present study were in good agreement with the data obtained for corresponding enzymes from other sources (1, 3, 4, 6). The enzymes usually have a dimeric or tetrameric native structure and are composed of subunits of about 50 kDa. Many AlDHs have a broad substrate spectrum and exhibit K_m values for the aldehydes in the micromolar range. They are devoid of prosthetic groups, and in most cases, NAD is the preferred electron acceptor. An unusual feature of AlDH 4 from R. eutropha strain Bo was the very strong substrate inhibition. It might be speculated that AlDH 4 is necessary only to remove traces of aldehydes which were not converted by THFA-DH. Due to the toxicities of aldehydes, this will be an advantage for the organism. Furthermore, it has to be considered that the PQQ-dependent THFA-DH from R. eutropha Bo is probably located in the periplasm (10, 11), and thus, there have to be additional enzymes converting toxic aldehydes occurring in the cytoplasm. In P. aeruginosa and C. testosteroni, the genes encoding the dye-linked PQQ-dependent ADHs are located close to a gene with high similarity to NAD-dependent AlDHs, indicating that both enzymes might be physiologically connected (5, 9). Interestingly, the putative acetaldehyde dehydrogenase from P. aeruginosa showed the highest sequence similarity to acetaldehyde dehydrogenase II from R. eutropha strain H16 (4, 5). The latter enzyme had only a low similarity to AlDH 4 from R. eutropha strain Bo. Unfortunately, for nearly all of the proteins which exhibit a significant sequence similarity to the N-terminal sequence of AlDH 4, no physiological or biochemical data are available.

Since AlDH 2 and AlDH 4 were specifically induced during growth on THFA and THFA or n-pentanol, respectively, it has to be assumed that these enzymes are involved in their degradation under in vivo conditions. We investigated the effect of added purified AlDH 4 on the conversion of THFA and n-pentanol by homogeneous THFA-DH. For this purpose THFA-DH was incubated in the presence or absence of similar amounts of AlDH 4 activity, and the conversion of the substrates (2 mM each) and formation of intermediates and products were analyzed by gas chromatography as reported previously (10). If THFA was used as the substrate, THFA-DH oxidized the alcohol directly to the corresponding carboxylic acid without detectable intermediates (Fig. 3A). However, if AlDH 4 was also added to the reaction mixture, the velocity of alcohol degradation and carboxylic acid formation increased (Fig. 3A). In contrast to THFA, the conversion of n-pentanol by THFA-DH was accompanied by the intermediary formation of the aldehyde (Fig. 3B). The addition of AlDH 4 had no influence on the utilization of the alcohol, but the oxidation of the aldehyde to the corresponding carboxylic acid was accelerated (Fig. 3B). Due to the substrate concentrations used during the in vitro analysis and the very strong substrate inhibition of AlDH 4 by pentanal observed, these experiments might not reflect the real physiological relevance of the enzyme. Under in vivo conditions, the aldehyde might be present in very low concentrations, and thus the influence of AlDH 4 on its conversion should be much more significant. However, the experiments clearly showed that AlDH 4 is also involved in the degradation of both alcohols by R. eutropha strain Bo.

FIG. 3.

Conversion of THFA (A) and n-pentanol (B) by THFA-DH in the presence (open symbols) and absence (filled symbols) of similar amounts of AlDH 4 activity. Shown are concentrations of alcohol (circles), carboxylic acid (triangles), and aldehyde (squares).