INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Pseudomonas lipases have a great potential in the areas of detergent additives, processing of fats or oils, organic synthesis, chiral resolution, and so on (20). P. fluorescens lipase (molecular mass, 50 kDa), in particular, has special applications in the chiral resolution of racemic mixtures (14, 31) and in the synthesis of polyunsaturated fatty acids (24, 38).

Recent understanding of the secretion mechanism in gram-negative bacteria has provided various approaches to the production of the Pseudomonas lipase. Pseudomonas lipases are secreted through two different secretion pathways. First, a general secretion pathway (GSP; type II pathway) is used by the signal sequence-containing lipases of P. glumae (11), P. alcaligenes (13), and P. aeruginosa (35). For correct folding and translocation, these lipases need molecular chaperones, which are located immediately downstream of the lipase structural genes (10, 17, 19, 37). Secondly, the lipase of P. fluorescens is secreted by the ABC pathway (7), which secretes the proteins, including toxins, proteases, and lipases, by a one-step mechanism (type I pathway). This secretion pathway involves only three membrane proteins, i.e., ABC protein, membrane fusion protein, and outer membrane protein, while the GSP (type II pathway) involves over 12 secretory proteins, in addition to sec proteins (36).

Many Pseudomonas lipases have been cloned and expressed in Escherichia coli (10, 19, 21, 30). However, they usually accumulate as an inactive inclusion body in the cell, and active lipase can be obtained only by refolding the inclusion body (1, 32). Lipases are also produced by chaperone-mediated folding and secretion in E. coli (15, 18, 30). On the other hand, lipase can be produced in homologous hosts by inserting a lipase gene into the original host. Pseudomonas lipases, especially those secreted by GSP, are produced by the recombinant Pseudomonas species harboring both lipase and its chaperone genes, as has been reported in the case of P. cepacia (16), Pseudomonas sp. KWI-56 (34), and P. alcaligenes (12).

Previously, we cloned an ABC transporter of the lipase from P. fluorescens SIK W1 and found that the lipase was secreted in recombinant E. coli with the aid of the ABC transporter gene (2). To enhance the production of extracellular lipase, as described in this report, we tried a homologous expression of lipase and ABC transporter genes in Pseudomonas. To our knowledge, there have been no other reports on the production of lipase in recombinant Pseudomonas with the aid of ABC transporter gene. Therefore, we report here the coexpression of lipase (tliA) and ABC transporter (tliDEF) genes in P. fluorescens, the original host. In addition, we compared the expressions of lipase and ABC transporter genes in other Pseudomonas species.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains, plasmids, and growth conditions. Various Pseudomonas species such as P. fluorescens SIK W1 (4), P. putida GM730 (23), P. acidovorans (recently reclassified as Comamonas acidovorans) KCTC1638, P. aeruginosa KCTC1750, P. syringae KCTC1832, and P. fragi KCTC2345 were used for the expression of tliDEF and tliA. E. coli XL1-Blue (Stratagene) and DH5 (33) were used as recipients for plasmid transformation and conjugation. Broad-host-range vectors, pDSK519 (IncQ; Km^r) and pRK415 (IncP; Tc^r) were donated by N. T. Keen (22). Broad-host-range vector pBBR1MCS-4 (IncP-1; Ap^r) was donated by K. M. Peterson (25). Luria-Bertani (LB) medium was used for the growth of E. coli and various Pseudomonas species. P. fluorescens, P. putida, P. syringae, P. acidovorans, and P. fragi were grown at 25°C, and P. aeruginosa was grown at 37°C, unless otherwise described. The LAT plate (LB medium, 1.5% Bacto Agar, 0.5% tributyrin) was used to detect the lipase activity of recombinant E. coli and Pseudomonas. If required, ampicillin (50 µg/ml), chloramphenicol (170 µg/ml), kanamycin (50 µg/ml), and tetracycline (25 µg/ml) were included in the growth media.

Plasmid construction. Figure 1 shows the plasmids used in this study. Plasmids pAJH3 and pAJH4 were constructed by inserting 2.9-kb HindIII-EcoRI fragments into pRK415 and pBBR1MCS-4, respectively. Plasmid pAJH5 was generated by inserting a 4-kb BamHI-EcoRI fragment into pDSK519. Plasmid pAJH6 was constructed by inserting a 3.5-kb BsrBI-HindIII fragment into the HindIII-ClaI (filled-in) site of pAJH4. The 6.3-kb KpnI-SacI fragment from pAJH6 was inserted into pDSK519 and pRK415, resulting in pAJH10 and pAJH11, respectively. Expression of these plasmids was constitutive, rather than influenced by isopropyl--D-thiogalactopyranoside (IPTG) induction, so IPTG was not used for induction.

View larger version (29K): [in this window] [in a new window]   FIG. 1.   Restriction endonuclease map of the various plasmids used. The relative positions of genes in P. fluorescens SIK W1 and restriction enzyme sites are depicted at the top. Inserts of subcloned plasmids are represented by the heavy line with pertinent enzyme sites of both ends. The names of subcloned plasmids and cloning vectors used are given above and on the right side of the heavy line, respectively. Plac is lac promoter. Restriction enzymes: B, BamHI; Bs, BsrBI; Cl, ClaI; E, EcoRI; H, HindIII; K, KpnI; X, XhoI; Ev, EcoRV; S, SacI.

Conjugal transfer of broad-host-range vector. Triparental spot matings were performed essentially as described by Miller (29). Broad-host-range plasmids were mobilized by using helper plasmid pRK2013 (9). Late-log cultures of donor (E. coli), helper (E. coli/pRK2013), and recipient (Pseudomonas) cells were centrifuged, washed, and resuspended in one-third volume of 0.9% NaCl. Five microliters of recipient cells was spotted on an LB agar plate and allowed to dry, followed by spotting and drying of the same volume of helper and donor cells. Matings were allowed to proceed for 8 h at the growth temperatures appropriate for the various Pseudomonas species. The cells in the spots were resuspended in 0.9% NaCl and plated on selective LB media containing ampicillin, as well as other appropriate antibiotics. Because some Pseudomonas species did not grow on the Pseudomonas-isolating agar (Difco), selection was done by using ampicillin (50 µg/ml), to which all the Pseudomonas species tested were resistant. P. aeruginosa, P. acidovorans, and P. fragi were resistant to kanamycin (50 µg/ml), so these Pseudomonas species containing pDSK519 derivatives were plated and screened on LB agar containing 400 µg of kanamycin per ml. Plates were incubated at 37 or 25°C, and transconjugants usually appeared within 2 to 3 days.

SDS-PAGE and immunoblotting. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was carried out with a 10% polyacrylamide gel as described by Laemmli (26). One milliliter of culture solution was harvested and centrifuged for 10 min at 13,000 × g. Cell pellets were dissolved in 200 µl of 0.1 M NaCl, 10 to 20 µl of the cell suspension was mixed with SDS sample buffer (5×), and then 15 µl was subjected to SDS-PAGE. The culture supernatant was obtained by centrifugation and mixed with SDS sample buffer (5×), and then 15 to 20 µl was subjected to SDS-PAGE. Proteins were stained with Coomassie brilliant blue R-250.

Analytical methods. Protein concentrations were determined by the Bradford assay (Bio-Rad) using bovine serum albumin as a standard. The concentration of extracellular lipase was estimated from the concentration of total extracellular proteins, and the percentage of extracellular lipase was estimated with the Imaging Densitometer (Model GS-700; Bio-Rad). The amount of cellular protein was estimated by determining the concentration of the cell extract obtained by centrifugation, resuspension, and sonication of cultured cells. The percentage of secreted lipase was calculated by dividing the amount of the secreted lipase by that of the total protein (cellular and extracellular). OD was measured at 600 nm (Ultraspec 2000 UV/visible spectrophotometer; Pharmacia).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Vector construction for the expression of tliDEFA in Pseudomonas Previously, the lipase gene tliA (6) and the ABC transporter gene tliDEF (2) had been cloned. We expressed tliA and tliDEF in E. coli, which secreted lipase up to 5 U/ml · OD, while E. coli harboring only tliA secreted no lipase. To improve low heterologous expression in E. coli, we introduced tliA and tliDEF into P. fluorescens, from which tliA and tliDEF were cloned.

View larger version (33K): [in this window] [in a new window]   FIG. 2.   Secretion of lipase by P. fluorescens harboring various plasmids. (A) The activity estimation of culture supernatant. P. fluorescens SIK W1 organisms harboring different plasmids were grown at 25°C in LB medium. The cells were harvested at the stationary phase when they reached an OD600 of around 3.0. Extracellular lipase activities were estimated by the pH titration method, and the activity is represented as relative value to the activity produced by P. fluorescens (pAJH11 pAJH10) (set as 100%). (B) Immunodetection of the lipase culture supernatant. Culture supernatant of 0.04 OD equivalent (12 µl of culture supernatant) was subjected to SDS-PAGE, transferred to nitrocellulose membrane, and analyzed by immunodetection. One OD equivalent corresponds to 1 ml of culture, OD600 = 1. Lanes: 1, no plasmid; 2, pAJH3; 3, pAJH5; 4, pAJH10; 5, pAJH11; 6, pAJH3 pAJH10; 7, pAJH11 pAJH10.

Temperature-dependent expression of tliDEFA in P. fluorescens Previously, we reported that E. coli complemented by tliDEFA secretes the lipase only at temperatures below 30°C because tliDEF is functional at this temperature (2). We tested whether P. fluorescens also secreted the lipase in a temperature-dependent manner. E. coli and P. fluorescens harboring pAJH5 or pAJH10 were grown at different temperatures, and then the expression of lipase was determined inside the cell (Fig. 3A) and in the supernatant (Fig. 3B). P. fluorescens harboring tliA or tliDEFA secreted the lipase at temperatures below 30°C, demonstrating that the ABC transporter was also functional at low temperature in its original host. P. fluorescens harboring pAJH10 (lane 4) secreted much more lipase than E. coli harboring pAJH10 (lane 2) at 25°C. P. fluorescens harboring pAJH5 (lane 3) also secreted a small amount of lipase at 25°C by means of the chromosomal ABC transporter gene, although no lipase was secreted by E. coli harboring pAJH5 (lane 1).

View larger version (48K): [in this window] [in a new window]   FIG. 3.   Immunodetection of the lipase in cellular extracts (A) and culture supernatant (B) of E. coli and P. fluorescens carrying different plasmids. E. coli DH5 and P. fluorescens SIK W1 harboring different plasmids were grown at different temperatures in LB medium. P. fluorescens did not grow at 37°C, so there was no result. The cells were harvested at the stationary phase when they reached an OD600 of around 3.0. Cell extract and culture supernatant were prepared as described in Materials and Methods. Cell extract of 0.09 OD equivalent and culture supernatant of 0.04 OD equivalent were subjected to SDS-PAGE, transferred to nitrocellulose membrane, and analyzed by immunodetection. Lanes: 1, E. coli (pAJH5); 2, E. coli (pAJH10); 3, P. fluorescens (pAJH5); 4, P. fluorescens (pAJH10).

Expression of tliDEFA in various Pseudomonas species. To see whether tliA and tliDEF could be expressed in other Pseudomonas species, pAJH5 and pAJH10 were introduced by conjugation into various Pseudomonas species, P. aeruginosa, P. fragi, P. putida, P. syringae, and P. acidovorans ("Comamonas acidovorans"). They were all grown in LB at 25°C for the expression of tliDEF, because tliDEF was functional at temperatures below 30°C. The lipase activities in the culture supernatants of the various Pseudomonas species were measured (Table 1), and the secreted lipase was detected by Western blotting (Fig. 4). All Pseudomonas species carrying no plasmid secreted no lipase (<0.3 U/ml · OD). However, the recombinant P. fluorescens, P. syringae, and P. putida, harboring pAJH5 containing tliA, secreted the lipase, although P. fragi secreted no lipase by complementing only tliA, indicating that these hosts had inherent ABC transporter for P. fluorescens lipase but P. fragi had not. Recombinant P. fluorescens, P. fragi, P. syringae, and P. putida secreted much more lipase by supplementing pAJH10 containing tliDEFA. Recombinant P. fluorescens harboring tliDEFA secreted 70 times more lipase (333 U/ml · OD) than P. fluorescens harboring tliA (4.7 U/ml · OD). P. aeruginosa and P. acidovorans did not secrete the lipase by supplementing pAJH5 or pAJH10. tliA or tliDEF seemed not to be functionally expressed in these hosts.

View larger version (23K): [in this window] [in a new window]   FIG. 4.   Immunodetection of the lipase in culture supernatant of high lipase producers. Culture supernatant of 0.04 OD equivalent was subjected to SDS-PAGE, transferred to nitrocellulose membrane, and analyzed by immunodetection. Lanes: 1, P. fluorescens (pAJH5); 2, P. fluorescens (pAJH10); 3, P. fragi (pAJH5); 4, P. fragi (pAJH10); 5, P. syringae (pAJH5); 6, P. syringae (pAJH10); 7, P. putida (pAJH5); 8, P. putida (pAJH10); 0.9, E. coli (pAJH10). The arrow indicates the expected band of lipase whose molecular mass is 49.9 kDa.

tliDEFA expression in P. fluorescens against cell growth. P. fluorescens harboring pAJH10 secreted a considerable amount of lipase, but this result was reproducible neither in a flask nor in a fermentor but only in a test tube. This result led us to suspect that improved aeration, either in a flask or a fermentor, reduced the extracellular production of lipase, compared to the limited aeration in a test tube. Therefore, we designed several culture conditions to simulate various aeration rates. For this purpose, a 500-ml baffled flask containing 100 ml of LB medium and 500-ml unbaffled flasks containing 100 and 350 ml of LB medium were prepared to investigate the effect of aeration on the growth and lipase production for P. fluorescens harboring pAJH10.

View larger version (19K): [in this window] [in a new window]   FIG. 5.   Lipase secretion against cell growth in different culture condition. (A) Lipase activity in the supernatant at different growth phases. Culture supernatant was prepared by centrifugation of culture solution and used for extracellular lipase activities by the pH-STAT method. (B) Cell growth curve. P. fluorescens harboring pAJH10 was grown in different conditions at 25°C in LB medium. , 100-ml baffled flask; , 100-ml unbaffled flask; , 350-ml unbaffled flask. Arrows indicate the sampling times indicated in the legend to Fig. 6.

View larger version (53K): [in this window] [in a new window]   FIG. 6.   SDS-PAGE for culture supernatant of different culture conditions. The culture supernatants of three different conditions at three different times (23, 47, and 71 h) indicated by arrows in Fig. 5B were seen on SDS-PAGE. Each 16 µl of culture supernatant was subjected to SDS-PAGE. Lanes: SM, size marker; 1, well-aerated (100-ml baffle flask); 2, medium aeration (100-ml unbaffled flask); 3, microaerated conditions (350-ml unbaffled flask).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The ABC pathway is very simple, involving only three secretory components. In contrast, the GSP usually involves sec proteins for transport through the inner membrane and more than 12 secretory proteins for export through the outer membrane. Also, the ABC pathway is also efficient, because it is dedicated to transporting only some specific extracellular proteins, while the GSP secretes not only extracellular proteins but also various envelope proteins (8). In this report, we introduced a series of vectors carrying the ABC transporter and lipase genes into several Pseudomonas species, with the aim of achieving the extracellular production of lipase.

Although the secretion of the lipase was not observed above 30°C, the protein of tliA, organized as an operon behind tliDEF, was detected inside the cell at high temperature. Furthermore, more lipase was detected inside the cell in the presence of tliDEF, irrespective of the temperature. It was reported that substrate binding is required for assembly of three ABC transporter components (28). Thus, it could be speculated that, at high temperature, tliDEF proteins could interact with the lipase, thus protecting the lipase from the proteolytic degradation inside the cell, although they were not functional in lipase secretion. The secretion defect at high temperature was shown, both in the homologous host P. fluorescens and in the heterologous host E. coli, indicating that temperature dependency was derived from tliDEF itself, regardless of the host cells. The reason for this secretory defect at high temperature might be due to the disorder of the targeting and the assembling of the ABC transporter in the membrane at high temperature.

tliA and tliDEF were functionally expressed in all Pseudomonas species tested, except P. aeruginosa and P. acidovorans in liquid media. However, when recombinant P. aeruginosa was tested on LAT agar plate for the lipase activity, P. aeruginosa with pAJH5 containing only tliA showed an activity halo, whereas the original P. aeruginosa showed no halo. Previously, P. aeruginosa was reported to have an ABC transporter specific for alkaline protease, and the lipase of P. fluorescens was reported to be secreted by this ABC transporter (7). The recombinant P. aeruginosa harboring tliA seems to have secreted lipase using the endogenous ABC transporter on the solid medium. P. syringae and P. putida carrying only tliA secreted lipase, suggesting that they have an ABC transporter compatible with tliA, although there has been no report on the lipase secreted by ABC transporter or the presence of ABC transporter itself in these species.

Whereas the lipase was detected as a major single band on SDS-PAGE in the recombinant P. fluorescens under microaeration, it disappeared, and other proteins showed up in the culture supernatant with an increase of aeration. P. fluorescens lipase was strong enough to retain its activity after prolonged incubation or even after heat treatment at 100°C (3, 4), indicating that a decrease in the lipase band did not originate from autolysis. We suspected that proteases leaked during cell lysis might have degraded the secreted lipase under medium aeration and well-aerated conditions. We tested various protease inhibitors, such as EDTA, phenylmethylsulfonyl fluoride, pepstatin, E-64, or a protease inhibitor cocktail, to inhibit the degradation of lipase under well-aerated conditions. However, we could not find a specific inhibitor which blocked the degradation of the lipase. More investigation is needed to elucidate the mechanism for lipase disappearance under medium-aerated and well-aerated conditions.

While a small amount of the lipase was produced in P. fluorescens harboring only tliA, P. fluorescens harboring both tliA and tliDEF secreted up to 28 µg/ml · OD of lipase, which corresponded to about 14.8% of the total protein. Recently, Braun et al. attempted to use Pseudomonas as enzyme factories exploiting GSP for secretory production (5). However, our results strongly demonstrated the potential of Pseudomonas as a promising host for the extracellular production of lipase using the ABC pathway, which is more efficient than the GSP.