Purification and properties of the extracellular lipase, LipA, of Acinetobacter sp. RAG-1 Erick A. Snellman 1,2 , Elise R. Sullivan 1,3 and Rita R. Colwell 1,4 1 Center of Marine Biotechnology, University of Maryland Biotechnology Institute, MD, USA; 2 HQ USAFA/DFB, USAF Academy, CO, USA; 3 Department of Microbiology, University of New Hampshire, USA; 4 Department of Cell and Molecular Biology, University of Maryland, USA An extracellular lipase, LipA, extracted from Acinetobacter sp. RAG-1 grown on hexadecane was purified and proper- ties of the enzyme investigated. The enzyme is released into the growth medium during the transition to stationary phase. The lipase was harvested from cells grown to sta- tionary phase, and purified with 22% yield and > 10-fold purification. The protein demonstrates little affinity for anion exchange resins, with contaminating proteins removed by passing crude supernatants over a Mono Q column. The lipase was bound to a butyl Sepharose column and eluted in a Triton X-100 gradient. The molecular mass (33 kDa) was determined employing SDS/PAGE. LipA was found to be stable at pH 5.8–9.0, with optimal activity at 9.0. The lipase remained active at temperatures up to 70 °C, with maximal activity observed at 55 °C. LipA is active against a wide range of fatty acid esters of p-nitrophenyl, but preferentially attacks medium length acyl chains (C 6 ,C 8 ). The enzyme demonstrates hydrolytic activity in emulsions of both medium and long chain triglycerides, as demonstrated by zymogram analysis. RAG-1 lipase is stabilized by Ca 2+ , with no loss in activity observed in preparations containing the cation, compared to a 70% loss over 30 h without Ca 2+ . The lipase is strongly inhibited by EDTA, Hg 2+ ,andCu 2+ , but shows no loss in activity after incubation with other metals or inhibitors examined in this study. The protein retains more than 75% of its initial activity after exposure to organic solvents, but is rapidly deactivated by pyridine. RAG-1 lipase offers potential for use as a biocatalyst. Keywords: lipase; LipA; Acinetobacter sp.RAG-1;protein purification; zymogram. Lipases are glycerol ester hydrolases (EC 3.1.1.3) that catalyze the hydrolysis of triacylglycerols to free fatty acids and glycerol. They resemble esterases in catalytic activity, but differ in that their substrates are water-insoluble fats containing medium to long fatty acyl chains [1]. Lipases are further distinguished from esterases in that they are activated at the substrate–water interface [2]. Interest in lipases has increased recently due to their recognition as important virulence factors [3] and their biotechnological potential. Lipases have proven to be versatile enzymes in nonaqueous solvent systems in which they catalyze the synthesis of a variety of acylglycerols and specialized esters via transesterification. Lipases demonstrating high activity under alkaline conditions are used as additives in detergents, one of the largest industrial uses of these enzymes [4]. Lipase-catalyzed synthesis of structured triacylglycerols comprised of both long and medium chain fatty acids has been investigated as a means of providing single substitutes for mixed acylglycerides in dietary applications [5]. In both hydrolysis and synthesis reactions, lipases demonstrate stereo- and regio-selectivity, making them good candidates for production of optically active compounds used in the pharmaceutical and agricultural industries. Previously, we reported on the cloning and sequence of an extracellular lipase (LipA) from Acinetobacter sp. RAG-1, which contained several conserved regions common to bacterial lipases [6]. This strain has also been characterized with respect to its production of a powerful emulsifying agent, termed emulsan [7,8]. Emulsan is a heteropolysaccha- ride complex produced as a capsule in cells grown on hydrocarbons and ethanol as sole carbon sources and is released into the growth medium during transition to stationary phase [8,11]. The bioemulsifier is composed of a polysaccharide backbone with attached fatty acids and noncovalently bound proteins [9,10]. Because LipA is also produced during growth on hydrocarbons [6], despite the fact that alkanes are not lipase substrates, we are currently investigating the potential role of LipA in facilitating emulsification by interaction with emulsan. To this end we report here the purification and characterization of LipA and discuss its potential as a biocatalyst in synthesis reactions and in the context of related biotechnological applications. MATERIALS AND METHODS Media and culture conditions Acinetobacter sp. RAG-1 (ATCC 31012) recovered from frozen stock ()80 °C) was used to inoculate Spirit Blue agar (Difco, Liverpool, Australia) cultures, which were incubated Correspondence to R. R. Colwell, Center of Marine Biotechnology, Suite 236, Columbus Center, 701 East Pratt Street, Baltimore, Maryland, USA, 21202. Fax: + 1 703 292 9232, Tel.: + 1 703 292 9232, E-mail: rcolwell@umbi.umd.edu Abbreviations: LNPS, low nitrogen, phosphorous, sulfur; pNPP, p-nitrophenyl palmitate; pNP, p-nitrophenol; HIC, hydrophobic interaction chromatography. (Received 5 April 2002, revised 23 July 2002, accepted 6 September 2002) Eur. J. Biochem. 269, 5771–5779 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03235.x overnight at 37 °C. Single colonies were selected for inocu- lation of 100 mL low nitrogen, phosphorous, sulfur (LNPS) medium consisting of (per litre): KH 2 PO 4 ,3.3g;Na 2 HPO 4 , 2.2 g; Na 2 SO 4 ,1.0g;NH 4 NO 3 ,1.0g;NaCl,5.0g;MgSO 4 , 0.29 g; CaCl 2 ,0.05g;FeSO 4 , 1 mg (pH 7.0). Hexadecane (10 m M ) was employed as carbon source. Inocula were grown overnight at 30 °C, with shaking at 200 r.p.m. in a rotary incubator/shaker (New Brunswick Scientific, Edison, NJ). Aliquots from overnight cultures were transferred to freshLNPSamendedwithhexadecaneandusedforgrowth studies and lipase production. Cultures were grown (as above) for 48 h prior to harvest. Extracellular lipase activity in these cultures ranged from 0.2 to 0.35 units (U) mL )1 . Lipase assay Lipase activity was measured by hydrolysis of p-nitrophenyl palmitate (pNPP) in deoxycholate buffer, as described elsewhere [6,12]. All assay reagents were purchased from Sigma (St. Louis, MO). Samples (20 lL)500 lL) were added to prewarmed (30 °C) phosphate buffer (50 m M , pH 8) containing 0.2% (w/v) sodium deoxycholate and 0.1% (w/v) gum arabic, final volume 3.0 mL. The mixture was incubated for 5 min at 30 °C. pNPP (0.30 m M final concentration) was added and the mixture shaken, allowing the reaction to proceed for 3 min. Lipase activity was determined by the rate of p-nitrophenol production (pNP), measured at 405 nm in a model DU640 spectrophotometer (Beckman Coulter, Fullerton, CA). Lipolytic activity was determined, using substrate free blanks as control. The reaction rate was calculated from the slope of the absorb- ance curve, using software installed by the manufacturer (Beckman Coulter). The extinction coefficient under the conditions described was 17454 LÆmolÆcm )1 [6]. One unit of enzyme activity is defined as the amount of enzyme forming 1 lmol of pNP min )1 . Lipase specific activity was expressed as unitsÆmg protein )1 . When examining the effect of tem- perature on activity, enzyme preparations were incubated at different temperatures for 5 min and assayed at the incubation temperature. During growth studies, cell bound and cell free lipase activities were determined as follows: cells were pelleted, washed twice, and resuspended in sterile LNPS prior to assay. Supernatants were filtered (0.2 lm Tuffryn mem- brane, Gelman Laboratories, Ann Arbor, MI, USA) and assayed separately. Protein concentration Protein was measured using the method of Bradford [13], with BSA as standard. Detergent-compatible BCA protein assay (Pierce, Rockford, Il) was used to determine protein concentration in samples containing Triton X-100. Total protein of cellular fractions was determined after cell disruption by sonication (3 · 30 s) using a Branson model 450 sonicator fitted with a 1.0-mm microtip. Protein concentration was routinely used as a measure of cell growth in hydrophobic media [14,15]. Lipase purification After incubation for 48 h in LNPS amended with 10 m M hexadecane, cells were removed by centrifugation (10 000 g) at 4 °C and the supernatants pooled. To increase the yield of lipase, 2 m M CaCl 2 and 2 m M MgCl 2 were added during magnetic stirring and the crude supernatant centrifuged a second time. To remove residual hexadecane, the combined sample was allowed to stand for 30 min prior to passage through a coarse glass fiber filter, after which the sample was filtered (0.2 lm). Supernatants were concentrated by ultrafiltration, using an Amicon RA2000 filter unit fitted with an S1Y10 spiral membrane (10 000 MW, 20 psi) and the sample volume reduced approximately 10–15-fold to 200 mL. After con- centration, a serine protease inhibitor, phenylmethanesulfo- nyl fluoride, was added (0.2 m M ) to reduce loss from proteolytic activity. Concentrated supernatants were ultra- centrifuged at 141 000 g (1 h) at 4 °C. Supernatants con- taining lipase were divided into 35 mL aliquots and stored at )80 °C prior to chromatography. Preliminary experiments to investigate binding properties of the lipase, using ion exchange resins, showed little (10%) affinity for the matrices under the conditions employed. However, other proteins were effectively bound, yielding significant purification. Therefore, anion exchange was employed as a preliminary step to hydrophobic interaction chromatography (HIC). Samples were dialyzed overnight against 20 m M Tris/HCl buffer (pH 8.0), followed by passage through an Econo-Pac Mono Q cartridge (5 mL) (Bio-Rad, Hercules, CA) at approximately 1 mL min )1 , prior to loading on the hydrophobic matrix. HIC methods were employed as described elsewhere [16,17]. An equal volume of Buffer 1 (30 m M Tris/HCl; 2 m M CaCl 2 ;2 m M MgCl 2 ;0.5 M NaCl) was added to supernatant that had equilibrated at room temperature for 30 min. Aliquots were added to 25 mL of Butyl Sepharose Fast Flow 4 hydrophobic medium (Amersham Pharmacia Biotech, Piscataway, NJ, USA) equilibrated in the same buffer. Equilibration of HIC gel and samples at room temperature allowed for more than 90% of the lipase to be bound. The protein slurry was degassed and loaded onto a glass Econo- Column (Bio-Rad) fitted with a column adaptor, for a total column volume of 20–25 mL. Lipase was eluted using two linear gradient profiles and flow rate of 1 mLÆmin )1 .Two column volumes of Buffer 1 were passed through the column, followed by two volumes of a decreasing salt gradient (0.5 M )0 M NaCl in Buffer 1) and washed in six volumes of the same buffer without salt. A second gradient of 10 column volumes of Triton X-100 (0–1.0%) in 30 m M Tris/HCl pH 8.0, followed by an additional 60 mL 1% Triton X-100, was used to elute LipA and other highly hydrophobic proteins. Absorbance (280 nm) was monitored and fractions (8.0 mL) collected from the detergent gradient were assayed for lipase activity. Fractions containing pure lipase, based on SDS/PAGE results, were pooled and concentrated, using ultrafiltration (PM10 membrane, Millipore, Bedford, MA) or an Ultrafree-4 centrifugal filter unit (10 000 MW cut-off, Millipore) and stored at )80 °C. Detergent removal Excess Triton X-100 was removed from the protein preparations by incubating samples at 4 °C with polymeric absorbent Amberlite XAD-2 (Sigma, St. Louis, MO). The absorbent was cleaned following the manufacturer’s instructions and fine particles removed by siphoning after 5772 E. A. Snellman et al.(Eur. J. Biochem. 269) Ó FEBS 2002 each of several washes in distilled water. The absorbent was equilibrated in 30 m M Tris/HCl at pH 8.0 prior to use. Under these conditions, approximately 50 mg detergent was removed per gram (wet) of absorbent. The amount of detergent (%) remaining was calculated by plotting A 289 of samples against a standard curve. The absorbent was removed from the supernatant by centrifugation (8000 g, 30 min) at 4 °C. Electrophoresis and preparation of zymograms SDS/PAGE gel electrophoresis was performed, according to the methods of Laemmli [18]. Protein samples were prepared in sample buffer (0.5 M Tris/HCl, pH 6.8) con- taining 2% SDS, 2.5% 2-mercaptoethanol, 0.1% bromo- phenol blue, and 8 M urea. Electrophoresis was performed using 12% polyacrylamide gels containing 5 M urea. Broad- range molecular weight standards (Bio-Rad) were used for mass determinations. Proteins were stained with Coomassie Blue R-250. Nondenaturing or native PAGE was performed using the discontinuous gel system of Orstein [19] and Davis [20]. Gels were cast with a 4% stacking gel and 6% resolving gel. Proteinswereallowedtostackat80mVandseparateat 160 mV. Prior to incubation with activity gels, native gels were rinsed three times with distilled water and equilibrated in 30 m M Tris/HCl pH 8.0 containing 1% Triton X-100 for 30 min at 25 °C. IEF determination of LipA pI was performed using precast IEF gels, according to manufacturer’s instructions (Bio-Rad). The anode and cathode buffers were 7 m M phosphoric acid and 20 m M lysine/20 m M arginine, respect- ively. IEF standards were purchased from Sigma and consisted of trypsin inhibitor (pI 4.6); carbonic anhydrase (pI 6.6); and lentil lectin (pI 8.2, 8.6, 8.8). Zymograms were accomplished by two methods. LipA activity against triolein (olive oil) was demonstrated accord- ing to the method of Gilbert et al.[21].Insummary,gel overlays were prepared from a 5% olive oil (Sigma) emulsion in 50 m M Tris/HCl (pH 8.5) containing 0.01% (w/v) Victoria Blue B dye (pH indicator) and 1.3% agarose (Fisher Scientific, Pittsburgh, PA). Victoria Blue B was added as a solution in 70% ethanol. Zymograms were also prepared from an emulsion of 1% tricaprylin (C 8:0 )in 25 m M Tris/HCl and 5 m M CaCl 2 (pH 8.0) [22]. Gels were cast between glass plates at 4 °C. Activity staining was accomplished by overlaying native gels with the zymograms in a closed glass dish and incubating at 37 °Cfrom4to 16 h. The incubation chamber was kept humid by addition of paper toweling saturated with Tris buffer. Lipase activity against olive oil was recorded by appearance of a dark blue band. Positive indicator of tricaprylin hydrolysis was recorded by appearance of a zone of clearing. Effect of pH on lipase activity and stability Concentrated lipase preparations were diluted fourfold in NaH 2 PO 4 -NaOH buffer (50 m M ) at various pH values and incubated at 30 °C for 1 h. Lipase activity was determined in the same buffer plus 0.1% (w/v) gum arabic. To determine the effect of pH on enzyme stability, concentrated lipase preparations were diluted fourfold in various buffers and incubated for 24 h at 20 °C. Buffers (50 m M )usedwere sodium acetate (pH 5.0–5.6), Tris/malate (pH 5.8–7.5), Tris/HCl (pH 7.5–8.5), 2-amino-2-methyl-1, 3-propanediol (pH 8.2–9.5), and glycine-NaOH (pH 8.6–10.6). Substrate specificity LipA activity toward substrates with different acyl chain lengths was determined under standard conditions using various esters of p-nitrophenyl (pNP). Substrates and chain lengths examined were as follows: pNP acetate (C 2 ); pNP butyrate (C 4 ); pNP caproate (C 6 ); pNP caprylate (C 8 ); pNP caprate (C 10 ); pNP laurate (C 12 ); pNP myristate (C 14 ); pNP palmitate (C 16 ); and pNP stearate (C 18 ). Substrate stock solutions (36 lL) were added to the reaction mixture (final concentration, 0.3 m M ) containing lipase and the reactionallowedtoproceedfor3min. Effect of Ca 2+ on lipase stability Results of lipase sequence analysis in our laboratory suggested the presence of a Ca 2+ -binding site in LipA [6]. However, the importance of Ca 2+ -sequestering to enzyme stability in LipA was not investigated. We chose to examine the effect of Ca 2+ loss on stability by incubating LipA in the presence and absence of Ca 2+ and examining effects over time. Concentrated enzyme preparations in 30 m M Tris/ HCl, pH 8.0 (approximately 3.5 UÆmL )1 ) were dialyzed overnight against 1 L 50 m M Tris/HCl, pH 8.0. After dialysis, 20 lL aliquots of protein solution were diluted fourfold in 50 m M Tris/HCl/2 m M MgCl 2 , with and without addition of 2 m M CaCl 2 and incubated at 30 °C. During a 30-h period, samples in triplicate were randomly selected at times indicated and examined for lipase activity. Sensitivity to inhibitors and organic solvents LipA preparations were incubated with various compounds of potential inhibitory activity. Prior to incubation in the presence of these compounds, protein samples were dialyzed overnight against 30 m M Tris/HCl pH 7.2. Lipase samples were diluted with stock solutions of inhibitors (final concentrations, 0.1 m M ,1.0m M , or 10.0 m M ) and incuba- ted at 30 °C for 1 h. At the end of the incubation period, residual activity was determined using pNPP as the substrate. Stability of LipA in organic solvents was meas- ured in a similar fashion using lyophilized enzyme (0.01 mg) incubated in water miscible solvents (15% and 30%, v/v) in 30 m M Tris/HCl (pH 8.0) for 1 h at 30 °C. Activity remaining (%) was measured under standard conditions. RESULTS AND DISCUSSION Acinetobacter sp. RAG-1 has been shown to produce significant amounts of extracellular lipase when grown in LNPS minimal medium amended with hexadecane [6]. However, it was not within the scope of that study to determine the spatial and temporal distribution of the enzyme. In this study, in order to determine optimum time to harvest the lipase for purification, both cell-bound and cell-free lipase activities were measured (Fig. 1). The data show extracellular lipase production in RAG-1 is growth phase dependent. During exponential growth, little cell-free lipase is detected, but a 10-fold increase in extracellular Ó FEBS 2002 Extracellular lipase, LipA, of Acinetobacter sp. RAG-1 (Eur. J. Biochem. 269) 5773 lipase activity was observed during transition to stationary phase. Similar increases in cell-free lipase activity during transition to stationary phase have been reported in other Acinetobacter calcoaceticus strains grown in more complex media [12,23] and in minimal medium supplemented with hexadecane [24]. Kok et al. [24] reported the growth phase dependent pattern of extracellular lipase activity in cultures of A. calcoaceticus BD413 and AAC321-1 grown on hexadecane as the sole carbon and energy source. They found lipase production is primarily regulated by LipA expression (measured by a-galactosidase expression in the lipA:lacZ strain) induced only after exponential growth had ceased, indicating that hexadecane itself did not induce lipase production. Moreover, they suggested that hexade- cane, or one of its degradation products (hexadecanoic acid) may repress LipA expression. Repression of lipase activity by fatty acids has also been reported in Pseudomonas aeruginosa EF2 grown on Tween 80 [25]. It would be of interest to determine if RAG-1 lipase production is also regulated by fatty acid repression of LipA. Purification RAG-1 lipase was purified from stationary phase cells growninLNPSamendedwith10m M hexadecane as the sole carbon source. Under these conditions, lipase accumu- lates in the medium with no apparent loss of activity, making it suitable for purification. LNPS medium supple- mented with various triglycerides as sole carbon sources were also investigated for suitability in lipase purification. However, in these media a significant and rapid reduction in activity was noted as exponential growth ceased (data not shown). This phenomenon has been previously reported for A. calcoaceticus BD413 grown in nutrient rich media [16]. In that study, it was suggested that loss of activity was due to proteolytic degradation that does not occur in a minimal medium amended with hexadecane [16]. LipA was purified 10-fold and 22% yield. A summary of the purification data is presented in Table 1. Prior to separation by HIC, supernatant samples were passed through a Mono Q column to remove contaminating proteins. LipA was effectively bound to the butyl Sepharose resin at 25 °C (90% of total lipase activity bound). The lipase was eluted from the HIC matrix in two gradients: decreasing NaCl gradient, followed by increasing detergent (Triton X-100) gradient. No appreciable lipase activity was detected in fractions collected under conditions of decreas- ing NaCl concentration. LipA is effectively eluted from the hydrophobic matrix only under conditions of increasing detergent concentration. Lipase begins to elute from the column at 0.4% Triton X-100 followed by an activity peak at 0.7% detergent. Remaining LipA activity decreased rapidly in 1% Triton X-100, indicating effective elution of the protein under these conditions. Fractions containing the highest lipase activity were pooled and examined for purity by SDS/PAGE. Only a single major protein band, whose molecular mass (approximately 33 kDa) is consistent with the molecular mass deduced from the nucleotide sequence of lipA, was observed in these fractions (Fig. 2). The fact that LipA binds butyl Sepharose resins under low salt concentration (0.25 M NaCl) and detergent is required to elute the lipase suggests it is hydrophobic in nature. We often observed smearing in our gels (65 kDa region), presumably due to lipase association with residual emulsan during purification. The lipase may associate with the lipophilic component of emulsan through hydrophic inter- action. LipA was determined to have a pI of 5.9 (not shown), in close agreement with the predicted value of 6.2 based on sequence analysis [6]. LipA demonstrates hydrolytic activity toward emulsions of both medium and long chain triacylglycerols (Fig. 3). Areas of olive oil and tricaprylin hydrolysis are clearly seen and correspond with the single protein band stained in native-PAGE gel. In these experiments, we found LipA shows a tendency toward aggregation, as some of the lipase molecules failed to enter the gel, with a corresponding positive indication of lipase activity in those areas of the Fig. 1. Growth and lipase production by Acinetobacter sp. RAG-1 in LNPS medium supplemented with 10 m M hexadecane. Growth (r)of RAG-1, measured as total protein (lgÆmL )1 ). Cell-free (s)andcell bound (j) lipase activity (unitsÆmL )1 ) was determined under standard conditions. One unit of enzyme activity catalyzes the production of 1 lmol of pNPÆmin )1 . Values are means of three replicates ± SE. Table 1. Purification of LipA from Acinetob acter sp. RAG-1. Purification method Protein (mg) Activity (units) a Specific activity (unitsÆmg protein )1 ) Purification (fold) Yield (%) Concentrated supernatant b 20.6 794 38.5 1 100 Mono Q 6.7 615 91.8 2.4 77.5 Butyl Sepharose 0.4 178 412 10.7 22.4 a One unit of enzyme activity catalyzes the production of 1 lmole of pNP min )1 . b Two litres of filtered supernatant were concentrated 10-fold to 200 mL. 5774 E. A. Snellman et al.(Eur. J. Biochem. 269) Ó FEBS 2002 zymograms. Other investigators have also reported aggre- gation of purified lipases to a varying degree [21,26–28]. Application of purified LipA to wells in Spirit Blue agar indicator plates containing an emulsion of tributyrin also indicates activity toward this lipidic substrate (not shown). As LipA is able to hydrolyze long-chain triacylglycerol esters, it merits classification as a true lipase (E.C. 3.1.1.3). Effect of pH on lipase activity and stability Figure 4A shows activity of LipA at various pH values, incubated at 30 °Cwithp-NPP as substrate. The optimal pH was found to be approximately 9.0. The enzyme showed strong activity in a narrow pH range of 8.0–10.0, but activity decreased rapidly at pH exceeding 10.5. These data are in agreement with those for alkaline lipases reported for strains of Pseudomonas [21,28] and Acinetobacter [17,26] and Bacillus subtilis 168 [27]. In comparison, the pH stability curve of LipA showed that the enzyme is stable under a wider pH range and slightly more acidic conditions (Fig. 4B). LipA retained 100% activity at pH 5.8–9.0, when incubated for 24 h at 20 °C. Below pH 5.6, stability of the molecule decreased sharply. The stability data suggest that the sharp decrease in activity below pH 6 (Fig. 4A) was not because of poor stability (Fig. 4B), but may be a result of titration of the imidazole ring of the active site histidine. Further, as significant activity was retained (80%) after incubation at pH 5.6, we suggest the dramatic reduction in stability under more acidic conditions (pH < 5.6) can be explained by titration of the Ca 2+ -coordinating Asp residues, as has been previously suggested [30]. Temperature The optimal reaction temperature for LipA activity towards p-nitrophenyl palmitate is 55 °C (Fig. 5). At this tempera- ture, LipA showed a threefold increase in activity compared with that at 30 °C. LipA remained active at higher temperatures, with activity exceeding 1.5-times that observed at 30 °C at temperatures up to 70 °C. The optimal reaction temperature reported here is higher than that reported for many bacterial lipases under similar experi- mental conditions, although higher temperature optima have been reported for several lipases of Pseudomonas spp. [31]. Activity at high temperature is a useful characteristic for lipases that are used in detergent formulations and biotransformations. Substrate specificity Substrate specificity of LipA was examined using various fatty acid esters of p-nitrophenyl. The enzyme showed activity toward a broad range of acyl chain lengths but maximum activity toward medium length fatty acid esters (C 6 and C 8 ) (Fig. 6). LipA showed little esterase activity toward the more soluble substrate, p-nitrophenyl acetate (C 2 ). A similar preference for medium chain esters and triacyl glycerols has been reported for lipases purified from Bacillus subtilis 168 and Aeromonas hydrophila [27,32]. Lipases and esterases share common substrate specificities. However, unlike esterases, lipases often dem- onstrate interfacial activation, i.e. a marked increase in activity upon the formation of a lipid–water interface [33]. Fig. 3. Zymograms demonstrating LipA activity toward olive oil (A) and tricaprylin (B) emulsions. Native PAGE gel was incubated between gels A and B for 16 h at 37 °C in a sealed, humid chamber. A positive indicator of lipolysis is demonstrated in (A) by the release of oleic acid and its interaction with the pH indicator (Victoria Blue B) and in (B) by tricaprylin hydrolysis (zone of clearing). Lane 1, mixed marker proteins (negative control); lane 2, ion exchange fraction (0.06 units); lane 3, purified LipA (0.3 units); lane 4, lipase from Chromobacterium viscosum (Sigma) (0.25 units, positive control). Fig. 2. SDS/PAGE of extracellular lipase purified from Acinetobacter sp. RAG-1. Lane 1, molecular mass standards; lane 2, crude super- natant; lane 3, ion exchange fraction; lane 4, purified LipA (10 lg). Ó FEBS 2002 Extracellular lipase, LipA, of Acinetobacter sp. RAG-1 (Eur. J. Biochem. 269) 5775 Therefore, lipase substrates are typically long chain (‡ C 10 ) fatty acid esters available in micellar form [34,35]. The properties of LipA reported here are consis- tent with this description. LipA is capable of hydrolyzing long acyl chain triglycerides, as in the case of olive oil emulsion used in zymogram preparations, and demon- strated uniform activity against various water-insoluble esters of p-nitrophenyl (Figs 3 and 6). In addition, results of LipA sequence analysis demonstrated overall similarity with other bacterial lipases [6]. These data confirm LipA classification as a true lipase. Stability of LipA in the presence of Ca 2+ The deduced sequence of RAG-1 LipA contains two putative Ca 2+ -binding residues (Asp240, Asp282) that participate in protein stabilization [6]. Comparative sequence analysis showed that residues associated with Ca 2+ -binding and protein stabilization are universally conserved in Group I Proteobacterial lipases [6]. However, biochemical evidence supporting the presence of a Ca 2+ - binding site in LipA was not reported. Here, we examine the effect of Ca 2+ loss on enzyme activity and discuss its stabilization effect on RAG-1 lipase. LipA was incubated in the presence and absence of 2 m M CaCl 2 and activity loss was measured over a 30-h period at 30 °C. During this time, enzyme preparations without calcium showed a linear decrease in activity. In the absence of Ca 2+ , up to 70% of the initial activity is lost, whereas enzyme incubated in the presence of calcium retained 100% activity (Fig. 7). The data clearly showed that Ca 2+ enhan- ces stability of RAG-1 lipase at 30 °C. We attribute the gradual decrease in activity to slow diffusion of Ca 2+ from its binding site, resulting in inactivation. The mechanism explaining inactivation by Ca 2+ loss is unclear, but may be attributed to concomitant destabilization in the local struc- ture surrounding the active site histidine [30,36]. Enzyme stabilization by calcium has been demonstrated in other studies [16,29,37]. Lipase purified from A. calcoaceticus AAC323-1 demonstrated a greater loss in activity in the absence of Ca 2+ , with almost no activity toward p-nitrophenyl palmitate within 3 h [16]. Biochemical studies of lipase prepared from Pseudomonas (Burkholderia) glumae also showed reduction in activity associated with calcium loss [36]. Addition of calcium was found to prevent heat inactivation of lipase prepared from P. fluorescens MC50 incubated at 60 °C [37]. Collectively, these data comprise Fig. 6. Substrate specificity of LipA. Acyl-chain length specificity of purified LipA was determined from its activity toward various esters of p-NP (0.3 m M ). Percentages shown are relative to maximum activity (C 6 ). Fig. 4. The effect of pH on LipA activity (A) and stability (B). (A) Concentrated lipase samples were diluted fourfold in Na 2 PO 4 -NaOH buffer (50 m M ), pH adjusted, and incubated for 1 h at 30 °C. Lipase activity was assayed in the same buffer using p-NPP as the substrate. Results expressed as relative percentage of maximal activity (pH 9.0). (B) Lipase preparations were incubated for 24 h at 20 °Cinselected buffers and remaining activity (% of initial activity) determined under standard conditions (pH 8.0). Buffers: (d), sodium acetate (pH 5.0– 5.6); (h), Tris/malate (pH 5.8–7.5); (m), Tris/HCl (pH 7.5–8.5); (·), 2-amino-2-methyl-1, 3-propanediol (pH 8.2–9.5); (e), glycine-NaOH (pH 8.6–10.6). Fig. 5. Effect of temperature on the activity of LipA. Theenzymewas incubatedin50m M phosphate buffer (pH 8.0) for 5 min at various temperatures and activity ± SE was determined using p-NPP as substrate. The value obtained at 30 °C was taken as 100%. 5776 E. A. Snellman et al.(Eur. J. Biochem. 269) Ó FEBS 2002 compelling biochemical evidence for conservation of the Ca 2+ -binding pocket in bacterial lipases and its importance in enzyme stabilization. Effects of cations and inhibitors LipA was incubated at 30 °C for 1 h with various cations and inhibitors known to affect the Group I, II lipases, reviewed by Gilbert [31]. LipA showed an increase in activity after exposure to low concentrations (1 m M )of Ca 2+ ,Mg 2+ ,Co 2+ ,Fe 3+ ,andRb + (Table 2). Increasing the metal concentration 10-fold had no further enhancing effect. Zn 2+ (10 m M ), Hg 2+ (1 m M )andCu 2+ inhibited the lipase, reducing activity by 15%, 88%, and 85%, respect- ively. Similar inhibition by heavy metals has been noted [29,32,38]. Metal ions tested here may have variable affects on lipase aggregation and on the substrate–water interface through interaction with free fatty acids [1,28]. EDTA strongly inhibited enzyme activity; treatment with 1 m M resulted in 90% activity loss. This effect was irreversible; i.e. no activity was recovered after incubation overnight at 4 °C with 20 m M CaCl 2 . Inhibition by EDTA probably results from its access to the Ca 2+ binding site and ion removal. LipA was not affected by dithiothreitol (10 m M )or 2-mercaptoethanol (10 m M ), suggesting a putative disulfide bridge is not required for activity. Exposure to the serine- protease inhibitor phenylmethanesulfonyl fluoride did not result in significant inhibition. Other serine hydrolases have shown similar resistance to such inhibitors in aqueous solutions [29]. Stability of the lipase in organic solvents Stability and activity in organic solvents are important characteristics of protein catalysts used in organic synthesis reactions. Therefore, LipA stability in selected water miscible solvents was examined to assess this potential. Lyophilized LipA, incubated with a variety of water-miscible organic solvents for 1 h at 30 °C, showed little effect, i.e. ‡ 90% activity was retained (Table 3). Less than 30% of initial activity was lost after incubation with acetonitrile (30%), a solvent that has been shown to deactivate lipases in concentrations as low as 15% (v/v) [29,39,40]. In contrast, lipase prepared from Chromobacterium viscosum showed high activity in acetonitrile where it is employed in trans- esterification reactions [41]. Pyridine caused significant deactivation, i.e. concentrations exceeding 15% (v/v) resul- ted in complete loss of activity within 1 h. Similar sensitivities to pyridine have been reported [27,39]. Lyophilized LipA also did not show deactivation in the presence of nonpolar solvents in concentrations up to 99.8% (v/v) (data not shown). The lipase demonstrated very little loss in activity Table 2. Effect of various inhibitors on LipA. LipA was incubated with various compounds that may inhibit the enzyme, and the remaining activity was measured under standard conditions. Enzyme preparations were dialyzed against 30 m M Tris/HCl pH 8.0 prior to the experiment. Lipase samples were incubated at 30 °C for 1 h. The remaining activity (%) is expressed relative to the appropriate control value (with no addition) post incubation. Standard error for all experiments is less than 10% of the value reported. ND, not determined. Compound Remaining activity (%) at a concentration (mM) of: 0.1 1.0 10.0 CaCl 2 123 105 MgCl 2 116 96 MnCl 2 98 117 ZnCl 2 110 86 CuCl 2 37 16 CoCl 2 141 118 FeCl 3 133 127 ND HgCl 2 85 12 ND RbCl 150 130 EDTA 11 6 EDTA + Ca 2+ a 98 Phenylmethylsulfonyl fluoride 93 89 88 Dithiothreitol 92 93 2-mercaptoethanol 95 a 20 m M CaCl 2 added post incubation with EDTA and incubated at 4 °C for 20 h. Fig. 7. Stabilizing effect of Ca 2+ on LipA. Enzyme preparations were dialyzed against 30 m M Tris/HCl (pH 8.0) prior to incubation in 30 m M Tris/HCl/2 m M MgCl 2 with (d) and without (s)2m M CaCl 2 . Replicates (three) were examined at various times over a 30-h period and activity plotted as percent of initial activity. Table 3. Stability of LipA in selected organic solvents. Lyophilized lipase was incubated with various organic solvents for 1 h at 30 °C. Residual activity was measured using standard conditions described in the text. Results are expressed as the percentage of activity with no addition of solvent. Standard error for all experiments is less than 5% of the value reported. Solvent Remaining activity (%) at a concentration (%, v/v) of: 15 30 isopropyl alcohol 98 92 dimethylformamide 83 96 acetone 96 93 dimethylsulfoxide 100 91 tetrahydrofuran 91 11 acetonitrile 72 78 pyridine 25 0 Ó FEBS 2002 Extracellular lipase, LipA, of Acinetobacter sp. RAG-1 (Eur. J. Biochem. 269) 5777 after exposure to solvents, indicating that it should be a useful catalyst for organic solvent systems. Acinetobacter sp. RAG-1 lipase purified and character- ized in this study demonstrated many properties appropri- ate to a variety of industrial applications. The enzyme is stable for extended periods of time in the presence of calcium. LipA showed wide substrate specificity toward medium and long chain esters of pNP. Maximal reaction temperature is 55 °C and LipA shows strong activity in the presence of metals, inhibitors, and organic solvents. Based on these properties, we are investigating further the potential of LipA to serve as a biocatalyst in selective transesterification reactions. REFERENCES 1. Brockerhoff, H. & Jensen, R.G. (1974) Lipolytic Enzymes. Academic Press, New York. 2. Brockman, H.L. (1984) Lipases. Elsevier. Scince Publishing Co., Inc, New York, NY. 3. Jaeger, K.E., Ransac, S., Dijkstra, B.W., Colson, C., van Heuvel, M. & Misset, O. (1994) Bacterial lipases. FEMS Microbiol. Rev. 15, 29–63. 4. Harwood, J. (1989) The versatility of lipases for industrial uses. Trends Biochem. Sci. 14, 125–126. 5. Mogi, K., Nakajima, M. & Mukataka, S. (2000) Transesterifica- tion reaction between medium- and long-chain fatty acid trigly- cerides using surfactant-modified lipase. Biotechnol. Bioeng. 67, 513–519. 6. Sullivan, E.R., Leahy, J.G. & Colwell, R.R. (1999) Cloning and sequence analysis of the lipase and lipase chaperone- encoding genes from Acinetobacter calcoaceticus RAG-1, and redefinition of a proteobacterial lipase family and an analogous lipase chaperone family. Gene 230, 277–286. 7. Gutnick, D.L., Allon, R., Levy, C., Petter, R. & Minas, W. (1991) Applications of Acinetobacter as an industrial microorganism. In: The Biology of Acinetobacter (Towner, K.J.Bergogne-Berezin, E. & Fewson, C.A., eds), pp. 411–441. Plenum Press, New York, NY. 8. Rosenberg, E., Zuckerberg, A., Rubinovitz, C. & Gutnick, D.L. (1979) Emulsifier of Arthrobacter RAG-1: isolation and emulsi- fying properties. Appl. Environ. Microbiol. 37, 402–408. 9. Belsky, I., Gutnick, D.L. & Rosenberg, E. (1979) Emulsifier of Arthrobacter RAG-1: determination of emulsifier-bound fatty acids. FEBS Lett. 101, 175–178. 10. Goldman, S., Shabtai, Y., Rubinovitz, C., Rosenberg, E. & Gutnick, D.L. (1982) Emulsan production in Acinetobacter calcoaceticus: Distribution of cell-free and cell associated cross- reacting material. Appl. Environ. Microbiol. 44, 165–170. 11. Zuckerberg,A.,Diver,A.,Peeri,Z.,Gutnick,D.L.&Rosenberg, E. (1979) Emulsifier of Arthrobacter RAG-1: chemical and phy- sical properties. Appl. Environ. Microbiol. 37, 414–420. 12. Fischer, B., Koslowski, R., Eichler, W. & Kleber, H. (1987) Bio- chemical and immunological characterization of lipase during its secretion through cytoplasmic and outer membrane of Acineto- bacter calcoaceticaus 69 V. J. Biotechn 6, 271–280. 13. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. 14. Marin, M.M., Pedregosa, A.M., Ortiz, M.L. & Laborda, F. (1995) Study of a hydrocarbon-utilizing and emulsifier-producing Acinetobacter calcoaceticus strain isolated from heating oil. Microbiologia 11, 447–454. 15. Zhang, Y. & Miller, R.M. (1994) Effect of a Pseudomonas rhamnolipid biosurfactant on cell hydrophobicity and biode- gradation of octadecane. Appl. Environ. Microbiol. 60, 2101–2106. 16. Bompensieri, S., Gonzalez, R., Kok, R., Miranda, M.V., Nutgeren-Roodzant, I., Hellingwerf, K.J., Cascone, O. & Nudel, B.C. (1996) Purification of a lipase from Acinetobacter calcoace- ticus AAC323-1 by hydrophobic-interaction methods. Biotechnol. Appl. Biochem. 23, 77–81. 17. Kok, R., van Thor, J.J., Nugteren-Roodzant, I.M., Brouwer, M.B.,Egmond,M.R.,Nudel,C.D.,Vosman,B.&Hellingwerf, K.J. (1995) Characterization of the extracellular lipase, LipA, of Acinetobacter calcoaceticus BD413 and sequence analysis of the cloned structural gene. Mol. Microbiol. 15, 803–818. 18. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. 19. Ornstein, L. (1964) Disc electrophoresis I. Background and Theory. Ann. N.Y. Acad. Sci. 121, 321–349. 20. Davis, B.J. (1964) Disc Electrophoresis II. Method and applica- tion to human serum proteins. Ann. N.Y. Acad. Sci. 121, 404–427. 21. Gilbert, E.J., Cornish, A. & Jones, C.W. (1991) Purification and properties of extracellular lipase from Pseudomonas aeruginosa EF2. J. General Microbiol. 137 (9), 2223–2229. 22. Sommer, P., Bormann, C. & Gotz, F. (1997) Genetic and bio- chemical characterization of a new extracellular lipase from Streptomyces cinnamomeus. Appl. Environ. Microbiol. 63, 3553– 3560. 23. Kok, R.G., Christoffels, V.M., Vosman, B. & Hellingwerf, K.J. (1993) Growth-phase-dependent expression of the lipolytic system of Acinetobacter calcoaceticus BD413: cloning of a gene encoding one of the esterases. J. General Microbiol. 139 (10), 2329– 2342. 24. Kok, R.G., Nudel, C.B., Gonzalez, R.H., Nugteren-Roodzant, I.M. & Hellingwerf, K.J. (1996) Physiological factors affecting production of extracellular lipase (LipA) in Acinetobacter cal- coaceticus BD413: fatty acid repression of lipA expression and degradation of LipA. J. Bacteriol. 178, 6025–6035. 25. Gilbert, E., Jane, Drozd, J.W. & Jones, C.W. (1991) Physiological regulation and optimization of lipase activity in Pseudomonas aeruginosa EF2. J. General Microbiol. 137, 2215–2221. 26. Hong, M. & Chang, M. (1988) Purification and characterization of an alkaline lipase from a newly isolated Acinetobacter radio- resistens CMC-1. Biotechnol. Lett. 20, 1027–1029. 27. Lesuisse, E., Schanck, K. & Colson, C. (1993) Purification and preliminary characterization of the extracellular lipase of Bacillus subtilis 168, an extremely basic pH-tolerant enzyme. Eur. J. Biochem. 216, 155–160. 28. Lin, S.F., Chiou, C.M., Yeh, C.M. & Tsai, Y.C. (1996) Purifica- tion and partial characterization of an alkaline lipase from Pseu- domonas pseudoalcaligenes F-111. Appl. Environ. Microbiol. 62, 1093–1095. 29. Choo,D.W.,Kurihara,T.,Suzuki,T.,Soda,K.&Esaki,N. (1998) A cold-adapted lipase of an Alaskan psychrotroph, Pseu- domonas sp. strain B11–1: gene cloning and enzyme purification and characterization. Appl. Environ. Microbiol. 64, 486–491. 30. Lang, D., Hofmann, B., Haalck, L., Hecht, H.J., Spener, F., Schmid, R.D. & Schomburg, D. (1996) Crystal structure of a bacterial lipase from Chromobacterium viscosum ATCC 6918 refined at 1.6 angstroms resolution. J. Mol. Biol. 259, 704–717. 31. Gilbert, E.J. (1993) Pseudomonas lipases: biochemical properties and molecular cloning. Enzyme Microb. Technol. 15, 634–645. 32. Anguita, J., Rodriguez Aparicio, L.B. & Naharro, G. (1993) Purification, gene cloning, amino acid sequence analysis, and expression of an extracellular lipase from an Aeromonas hydrophila human isolate. Appl. Environ. Microbiol. 59, 2411–2417. 33. Brzozowski, A.M., Derewenda, U., Derewenda, Z.S., Dodson, G.G., Lawson, D.M., Turkenburg, J.P., Bjorkling, F., Huge- Jensen, B., Patkar, S.A. & Thim, L. (1991) A model for interfacial activation in lipases from the structure of a fungal lipase-inhibitor complex. Nature 351, 491–494. 5778 E. A. Snellman et al.(Eur. J. Biochem. 269) Ó FEBS 2002 34. Derewenda, Z.S. & Sharp, A.M. (1993) News from the interface: the molecular structures of triacylglyceride lipases. Trends Biochem. Sci. 18, 20–25. 35. van Tilbeurgh, H., Egloff, M.P., Martinez, C., Rugani, N., Verger, R. & Cambillau, C. (1993) Interfacial activation of the lipase- procolipase complex by mixed micelles revealed by X-ray crys- tallography. Nature 362, 814–820. 36. Noble, M.E., Cleasby, A., Johnson, L.N., Egmond, M.R. & Frenken, L.G. (1993) The crystal structure of triacylglycerol lipase from Pseudomonas glumae reveals a partially redundant catalytic aspartate. FEBS Lett. 331, 123–128. 37. Swaisgood, H.E. & Bozoglu, F. (1984) Heat inactivation of the extracellular lipase from Pseudomonas fluorescens. J. Agric. Food Chem. 32, 7–10. 38. I., Izumi, T., Nakamura, K. & Fukase, T. (1990) Purification and characterization of a thermostable lipase from newly isolated Pseudomonas sp. KWI-56. Agric. Biol. Chem. 54, 1253–1258. 39. Shabtai, Y. & Daya-Mishne, N. (1992) Production, purification, and properties of a lipase from a bacterium (Pseudomonas aeru- ginosa YS-7) capable of growing in water-restricted environments. Appl. Environ. Microbiol. 58, 174–180. 40. Shimada, Y., Koga, C., Sugihara, A., Nagai, T., Takada, S.N., Tsunasawa, S. & Tominaga, Y. (1993) Purification and char- acterization of a novel solvent-tolerant lipase from Fusarium heterosporum. J. Ferment. Bioeng. 75, 349–352. 41. Barros, R.J., Moura-Pinto, P.G., Franssen, M.C., Janssen, A.E. & de Groot, A. (1994) Regioselectivity and fatty acid specificity of Chromobacterium viscosum lipase. Bioorg.Med.Chem.2, 707–713. Ó FEBS 2002 Extracellular lipase, LipA, of Acinetobacter sp. RAG-1 (Eur. J. Biochem. 269) 5779 . sp. RAG-1; protein purification; zymogram. Lipases are glycerol ester hydrolases (EC 3.1.1.3) that catalyze the hydrolysis of triacylglycerols to free fatty acids and glycerol. They resemble. University of New Hampshire, USA; 4 Department of Cell and Molecular Biology, University of Maryland, USA An extracellular lipase, LipA, extracted from Acinetobacter sp.