Abstract
Previous work has shown that when the bacterium Stenotrophomonas maltophilia is grown on polypropylene glycol, different dye-linked polypropylene glycol dehydrogenase (PPG-DH) activities are induced during growth. Here the purification and characterization of the dehydrogenase activity induced in the stationary phase, and present in the periplasmic space, is described. The homogeneous enzyme preparation obtained consists of a homodimeric protein with a molecular mass of about 123 kDa and an isoelectric point of 5.9. The cofactor of the enzyme appeared to be pyrroloquinoline quinone (PQQ), no heme c was present, and holo-enzyme contained two PQQ molecules per enzyme molecule. In these respects, PPG-DH described here is similar to already known quinoprotein alcohol dehydrogenases, but in other respects, it is different. Therefore, it is suggested that PPG-DH could be a new type of quinoprotein alcohol dehydrogenase. Based on its strong preference for polyols, PPG-DH seems well fitted to carry out the first step in the degradation of PPGs, synthetic polymers containing a variety of hydroxyl groups.
1 Introduction
Polypropylene glycols (PPGs) are synthetic polymers that are produced in large quantities as they are used for the manufacturing of surfactants, polyurethane, cosmetics and medicines. PPGs have rather complicated structures since they can have a straight (diol type) or a branched (triol type) chain, and contain structural as well as optical isomers with primary and secondary alcohol groups and ether bonds. However, since these chemically stable compounds do not form an environmental problem, apparently they disappear (from sewage water) by biological degradation, just as in the case of polyethylene glycols (PEGs) [1]. In line with this, it has been found that several bacteria are able to use PPGs as carbon and energy source [2]. Since most PEG-utilizing bacteria do not grow on PPG [3,4], the biological degradation pathways of PEG and PPG could be dissimilar.
It has been found that the first step in the degradation of related synthetic polymers, PEGs and the polyvinyl alcohols (PVAs), is catalyzed by a dye-linked alcohol dehydrogenase. In the case of the PEGs, this is a flavoprotein (FAD-containing) alcohol dehydrogenase [5] and in the case of the PVAs, a quinohemoprotein [pyrroloquinoline quinone (PQQ)- as well as heme c-containing] alcohol dehydrogenase [6–8]. Biological PPG degradation also occurs in an oxidative way but until recently, the first step in the pathway was completely unknown. However, on investigating the bacterium Stenotrophomonas maltophilia for the presence of PPG-oxidizing enzymes, we found that several dye-linked dehydrogenase activities are induced during growth on PPG [9]. Here we describe the purification and characterization of the periplasm-located dehydrogenase activity that is present in stationary phase-grown cells of this PPG-grown bacterium.
2 Materials and methods
2.1 Microorganism and cultivation
S. maltophilia was cultivated aerobically in a fermentor at 30°C for 7 days in a medium (pH 7.2) containing 0.7% PPG 2000 (diol type), 0.2% (NH4)2SO4, 0.2% K2HPO4, 0.1% NaH2PO4, 0.02% MgSO4·7H2O, 0.02% yeast extract, and spore minerals [9]. The cells were harvested by centrifugation (3000×g, 15 min, at 4°C), washed twice with physiological saline, and stored at −20°C until use.
2.2 Enzyme isolation
Preparation of the periplasmic fraction was done as previously described [9], except that 20 mM Tris–HCl buffer containing 5 mM MgCl2 (pH 8.0) was used in place of cold water containing 5 mM MgCl2. All subsequent steps were carried out at 4°C. The fraction was applied to a Q-Sepharose Fast Flow column (1.5 cm×15 cm) equilibrated with 20 mM Tris–HCl buffer, pH 8.0. After washing with the same buffer, the enzyme was eluted with a linear gradient from 0 to 0.3 M NaCl in this buffer. Active fractions were combined and concentrated by ultrafiltration (Centriprep YM-30, Millipore). The concentrate was dialyzed against 5 mM potassium phosphate buffer, pH 6.8. Subsequently, the solution was applied to a hydroxyapatite column (1.0 cm×9.0 cm, Microprep ceramic hydroxyapatite (type I), Biorad) equilibrated with 5 mM potassium phosphate buffer, pH 6.8. After washing, the enzyme was eluted with a linear gradient from 5 to 160 mM of the same buffer. Active fractions were pooled, concentrated by ultrafiltration, and then dialyzed against 20 mM Tris–HCl buffer, pH 8.0.
2.3 Assays
PPG dehydrogenase (PPG-DH) activity was determined at 30°C by measuring the initial rate of 3-(4,5-dimethylthiazolyl-2-)-2,5-diphenyl tetrazolium bromide (MTT, ɛ=6410 M−1 cm−1) reduction at 570 nm with phenazine methosulfate (PMS) as the primary electron acceptor. The reaction mixture consisted of 0.1 M Tris–HCl buffer (pH 8.0), 1 mM KCN, 2 mM CaCl2, 0.1 mM MTT, 0.1 mM PMS, 10 mM PPG 700 (diol type), and an appropriate amount of enzyme solution in a total volume of 1 ml. One unit of enzyme activity is defined as the amount of enzyme that catalyzes oxidation of 1 µmol of substrate per minute under the applied conditions. The data presented in Table 2 were corrected for the blank value (determined for the mixture with all compounds except the substrate). Protein concentrations were determined with the micro BCA protein assay kit (Pierce Chemical Co.) with bovine serum albumin as the standard.
Purification of PPG-DH from S. maltophilia
| Purification step | Total protein (mg) | Total activity (mU) | Specific activity (mU mg−1) | Recovery (%) |
| Periplasm extracts | 205.8 | 21 300 | 103 | 100 |
| Q-Sepharose FF | 7.1 | 17 800 | 2 507 | 88 |
| Hydroxyapatite | 3.1 | 8 700 | 2 806 | 41 |
| Purification step | Total protein (mg) | Total activity (mU) | Specific activity (mU mg−1) | Recovery (%) |
| Periplasm extracts | 205.8 | 21 300 | 103 | 100 |
| Q-Sepharose FF | 7.1 | 17 800 | 2 507 | 88 |
| Hydroxyapatite | 3.1 | 8 700 | 2 806 | 41 |
Purification of PPG-DH from S. maltophilia
| Purification step | Total protein (mg) | Total activity (mU) | Specific activity (mU mg−1) | Recovery (%) |
| Periplasm extracts | 205.8 | 21 300 | 103 | 100 |
| Q-Sepharose FF | 7.1 | 17 800 | 2 507 | 88 |
| Hydroxyapatite | 3.1 | 8 700 | 2 806 | 41 |
| Purification step | Total protein (mg) | Total activity (mU) | Specific activity (mU mg−1) | Recovery (%) |
| Periplasm extracts | 205.8 | 21 300 | 103 | 100 |
| Q-Sepharose FF | 7.1 | 17 800 | 2 507 | 88 |
| Hydroxyapatite | 3.1 | 8 700 | 2 806 | 41 |
2.4 Analyses
The native molecular mass of the enzyme was estimated by gel filtration (on a Superdex 200 HR 10/30 column, Pharmacia) and native polyacrylamide gel electrophoresis (PAGE) as described in a technical bulletin of Sigma-Aldrich Co. The denatured enzyme molecular mass was determined by gel filtration (in the presence of 6 M guanidine–HCl on a Superdex 200 PC 3.2/30 column, Pharmacia, the eluate monitored at 280 nm) and sodium dodecyl sulfate (SDS)–PAGE. The isoelectric point of the purified enzyme was determined on an isoelectric focusing gel with a gradient of pH 3.5–9.5 using ampholine pH 3.5–9.5 (Amersham Pharmacia Biotech). The marker proteins used were from a broad pI calibration kit (Amersham Pharmacia Biotech).
The cofactor was dissociated from the enzyme with 0.2% SDS as described [10]. The sample was heated at 90°C for 10 min and the detergent was removed by adding KCl to 0.1 M and centrifugation (12 000×g, 10 min, at 4°C). The amount of dissociated PQQ was determined with the apo-form of soluble glucose dehydrogenase [11], using authentic PQQ (Sigma Aldrich) as a reference.
N-terminal amino acid sequence analysis of the enzyme was carried out by blotting the protein band obtained with SDS–PAGE on a polyvinylidene difluoride membrane. Subsequently, automated Edman degradation was applied with an ABI 437A gas/liquid phase protein sequencer [12].
3 Results and discussion
3.1 Purification of PPG-DH
The results of the purification procedure are summarized in Table 1. Using the periplasmic fraction as a reference, the enzyme in the final preparation was obtained in a 41% yield with a 27-fold purification. The final preparation appeared to be homogeneous, as judged from the presence of only one band on protein staining of the gels from native PAGE, SDS–PAGE and isoelectric focusing.
Substrate specificity of PPG-DH
Enzyme activity: 11.4 U mg−1 protein.
The values in brackets present data obtained for substrates dissolved in dichloroethane.
PVA500: final concentration was 1% (w/v).
Substrate specificity of PPG-DH
Enzyme activity: 11.4 U mg−1 protein.
The values in brackets present data obtained for substrates dissolved in dichloroethane.
PVA500: final concentration was 1% (w/v).
3.2 Characterization
The enzyme appears to be an acidic protein since an isoelectric point of 5.9 was found. Estimation of the native enzyme's molecular mass by gel filtration and PAGE gave values of 95 and 123 kDa, respectively. Values for the denatured enzyme determined by gel filtration in 6 M guanidine–HCl and SDS–PAGE gave 63 and 60 kDa with only one peak or band, respectively. In view of the ratio of the molecular masses for the native and denatured protein, and the homogeneous products seen for each step in the N-terminal amino acid sequencing, the purified enzyme must be a homodimeric protein. Quinoprotein methanol dehydrogenases have an α2β2 structure [13], consisting of a 60-kDa and a small subunit. However, it has been reported that quinoprotein alcohol dehydrogenase from Pseudomonas aeruginosa does not contain the small subunit [14].
Comparison of the N-terminal amino acid sequence obtained, AEYKPVTAARLANPEPSNWLLLTK*S, with sequences present in data bases (SWISS PROT, PIR) revealed low similarity with quinoprotein and quinohemoprotein alcohol dehydrogenases but 50.0% identity with a stretch of amino acids derived from the gene believed to encode the soluble quinoprotein sugar polyol dehydrogenase of Pseudogluconobacter saccharoketogenes IFO 14464 [15]. It has been stated that the latter is a monomeric enzyme with a molecular mass of 65 kDa.
On testing several electron acceptors, only PMS (and to a lesser extent DCIP) functioned as electron acceptor but NAD, MTT, NBT, potassium ferricyanide, or horse heart cytochrome c did not. Using 0.1 M Tris–HCl buffer, the pH optimum for PPG-DH activity was 8.5 and the enzyme was stable in the pH range 6.5–9.0. The temperature optimum for PPG-DH activity was found to be 50°C, and the enzyme was stable up to 60°C (both tested in 0.1 M Tris–HCl buffer, pH 8.0). Addition of NH4Cl, ethylamine, EDTA, Ca2+ or Mg2+ to the assay mixture did not affect the specific activity value.
3.3 Cofactor identification
The absorption spectrum of the final preparation (Fig. 1) showed maxima at 280 and 350 nm, and a shoulder around 400 nm, characteristic for quinoprotein alcohol dehydrogenases [16–19] and soluble glucose dehydrogenase [20], enzymes containing the cofactor PQQ. PQQ also occurs in quinohemoprotein alcohol dehydrogenases, enzymes containing PQQ as well as heme c[21,22]. However, since the typical maxima of heme c were absent in the absorption spectrum (Fig. 1), only efforts were made to establish the presence of PQQ in the final preparation.
Absorption spectrum of PPG-DH. The absorption spectrum of the final preparation (3.0 mg protein ml−1) was recorded in 50 mM Tris–HCl buffer, pH 8.0, at 25°C.
Absorption spectrum of PPG-DH. The absorption spectrum of the final preparation (3.0 mg protein ml−1) was recorded in 50 mM Tris–HCl buffer, pH 8.0, at 25°C.
Reversed-phase chromatography of the supernatant obtained after denaturing the protein gave a peak with exactly the same retention time as authentic PQQ. On estimating the amount, assuming a molecular mass of 123 kDa for the enzyme, 1.02 of PQQ molecules per enzyme molecule were present. Since quinoproteins investigated so far contain two PQQ molecules per enzyme molecule, but sometimes loss of PQQ occurs during enzyme purification (e.g. with the quinoprotein alcohol dehydrogenase of P. aeruginosa[23]), PQQ was added to the assay mixture, leading to a two-fold increase in activity. Titration of the final preparation with PQQ revealed that full activity was obtained when 23.5 pmol of PQQ were added to 25.2 pmol of enzyme (Fig. 2). It is concluded, therefore, that the final preparation consists of a mixture of apo- and holo-enzyme, and that the holo-enzyme contains 1 PQQ molecule per subunit molecule.
Reconstitution of PPG-DH with PQQ. The purified PPG-DH was preincubated with different amounts of PQQ (0–500 pmol) and 2 mM of CaCl2 at room temperature for 30 min prior to the enzyme assay.
Reconstitution of PPG-DH with PQQ. The purified PPG-DH was preincubated with different amounts of PQQ (0–500 pmol) and 2 mM of CaCl2 at room temperature for 30 min prior to the enzyme assay.
3.4 Substrate specificity
Since the purified enzyme preparation appeared to be a mixture of apo- and holo-form, substrate specificity was determined by using the holo-form of PPG-DH. The holo-enzyme was prepared by the addition of PQQ to the purified enzyme preparation followed by gel filtration in 20 mM Tris–HCl buffer, pH 8.0, containing 5 mM CaCl2 to remove the excess PQQ. By this treatment the specific activity of the PPG-DH increased from 2.8 to 11.4 U mg−1 protein.
The holo-enzyme oxidized primary alcohols (Table 2) and aldehydes but not secondary alcohols (2-propanol, 2-butanol, 2-pentanol, 2-hexanol were tested) (results not shown). Small alcohols (methanol, ethanol) were not oxidized and the enzyme exhibited a preference for polyols (compare ethanol with ethyleneglycol, and 1-propanol with propanediols and glycerol). The absence of a second free OH group at a short distance from the first one strongly diminishes the activity (compare ethylene glycol, having a vicinal diol group, with diethyleneglycol). In line with this, PVA and PPGs (polymers with a variety of OH groups in the chain) are good substrates for the enzyme but PEGs are not (Table 2). In the case of PPGs, substrate specificity determination is hampered by the fact that the polymers have low solubility in water. Therefore, PPGs were dissolved in dichloroethane before adding to the assay mixture. The data obtained under this condition show that the enzyme has excellent activity for large PPGs like PPG (diol) 2000 and PPG (triol) 3000 (however, it should be noted that since it is unknown how much of the polymer in each case was transferred to the water phase in the assay mixture, no reliable quantitative comparisons can be made so that it is not clear which of the polymers is the optimal substrate).
Certain quinohemoprotein alcohol dehydrogenases oxidize glycerol [24] and PEGs [25], and certain quinoprotein alcohol dehydrogenases oxidize glycerol [15], PEGs [26] and PVAs [26]. Although these enzyme oxidize primary and secondary alcohols and aldehydes, PPG-DH does not oxidize secondary alcohols. In addition, oxidation of PPGs by either quinoprotein or quinohemoprotein alcohol dehydrogenases has never been reported.
4 Conclusion
The enzyme here described is a quinoprotein alcohol dehydrogenase. Although it exhibits several characteristics of this type of enzyme (homodimer with subunits of about 60 kDa, two PQQ molecules per enzyme molecule, typical absorption spectrum, oxidation of alcohols and aldehydes), in other respects it does not (no activation by ammonia or alkylamines, aberrant N-terminal amino acid sequence). Absence of activation has also been observed for quinoprotein glucose dehydrogenases, quinoprotein sugar polyol dehydrogenase, and quinohemoprotein alcohol dehydrogenases. However, in other respects (the N-terminal amino acid sequence or the substrate specificity) these enzymes are different. Therefore, our enzyme seems to be a new type of quinoprotein alcohol dehydrogenase.
The enzyme preferably oxidizes small-sized polyols (the diols of propane and butane, and the triol glycerol) as well as large ones such as PVA and PPGs. Since the enzyme is induced at growth on PPG 2000 (diol type), it seems appropriate to indicate it as PPG-DH. Thus, just as in the case of PEGs and PVAs, the first step in the degradation of PPG is the oxidation of OH groups, carried out by a dye-linked, periplasm-located alcohol dehydrogenase. Whether the other PPG dehydrogenase activities induced in S. maltophilia at growth on PPG derive from PPG-DH or are distinct enzymes remains to be investigated.
