G Model
ARTICLE IN PRESS
BIOMAC-4311; No. of Pages 6
International Journal of Biological Macromolecules xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Conversion of fat-containing waste from the margarine manufacturing process into bacterial polyhydroxyalkanoates Cristiana Morais, Filomena Freitas ∗ , Madalena V. Cruz, Alexandre Paiva, Madalena Dionísio, Maria A.M. Reis REQUIMTE/CQFB, Chemistry Department, FCT/Universidade Nova de Lisboa, 2829-516 Caparica, Portugal
a r t i c l e
i n f o
Article history: Received 10 February 2014 Received in revised form 15 April 2014 Accepted 22 April 2014 Available online xxx Keywords: Margarine waste Poly(3-hydroxybutyrate) Cupriavidus necator
a b s t r a c t A fat-containing waste produced from the margarine manufacturing process was tested as a low cost carbon source for cultivation of different polyhydroxyalkanoates (PHAs) producing bacterial strains, including Cupriavidus necator, Comamonas testosteroni and several Pseudomonas strains. The margarine waste was mainly composed of free fatty acids (76 wt.%), namely mystiric, oleic, linoleic and stearic acids. In preliminary shake flask experiments, several strains were able to grow on the margarine waste, but C. necator reached the highest PHA content in the biomass (69 wt.%). This strain was selected for batch bioreactor experiments, wherein it reached a cell dry weight of 11.2 g/L with a polymer content of 56 wt.%. The culture produced 6.4 g/L of polyhydroxybutyrate, P3(HB), within 20 h of cultivation, which corresponds to a volumetric productivity of 0.33 gPHA /L h. The P3(HB) polymer produced by C. necator from the margarine waste had a melting point of 173.4 ◦ C, a glass transition temperature of 7.9 ◦ C and a crystallinity of 56.6%. Although the bioprocess needs to be optimized, the margarine waste was shown to be a promising substrate for P(3HB) production by C. necator, resulting in a polymer with physical and chemical properties similar to bacterial P(3HB) synthesized from other feedstocks. © 2014 Elsevier B.V. All rights reserved.
1. Introduction In the margarine manufacturing process, vegetable oils or animal fats are modified by hydrogenation, rearrangement and fractionation, and blended with a mixture of water, brine and powdered ingredients [1]. Additives (e.g. vitamins, emulsifiers, salt, flavours) are included in the blend to improve the quality of the product and enhance flavour. The mixture is subjected to temperatures of 50–60 ◦ C, resulting in the formation of an emulsion (margarine) that is pasteurized and packed [1,2]. According to IMACE, International Margarine Association of the Countries of Europe (www.imace.org), in 2012, the European production of margarine was 2440 Mton, while worldwide it reached 9374 Mton. Around 1% of this production results in the generation fatcontaining wastes that require adequate disposal or treatment [3,4]. The fat-containing wastes are usually removed from the aqueous effluents produced by the plant in a gravity separator [3].
∗ Corresponding author at: Biochemical Engineering Lab, Chemistry Department, Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa, 2829-516 Caparica, Portugal. Tel.: +351 212 948 300; fax: +351 212948550. E-mail addresses: [email protected], fi[email protected] (F. Freitas).
Though having a low commercial value, it is commonly sold to oil-recycling companies. Fat-containing materials (e.g. tallow generated from the food industry) have been proposed as feedstocks for the production of value-added products, namely polyhydroxyalkanoates (PHA) [5]. The use of such waste materials is a valuable strategy to improve the producing industries sustainability and economical viability, by converting a waste into high-value products. PHAs are hydroxyalkanoic acids that are synthesized by many microorganisms as intracellular carbon and energy reserve materials or reducingpower storage materials [6]. These polymers possess physical characteristics similar to traditional plastics and have received extensive attention mainly due to their biodegradability and biocompatibility [6,7]. Polyhydroxybutyrate, P(3HB), is the most widely studied and best characterized PHA. It is a homopolymer of 3-hydroxybutyrate, which has mechanical properties similar to polypropylene [6]. Due to their properties, PHAs are used as packaging materials, biomedical devices and in the food industry (e.g. edible packaging, flavour delivery agent) [8,9]. However, PHA commercial applications have been limited by their high production cost that is mainly related to the high price of the carbon source [5]. In this context, the use of inexpensive renewable feedstocks is currently
http://dx.doi.org/10.1016/j.ijbiomac.2014.04.044 0141-8130/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: C. Morais, et al., Int. J. Biol. Macromol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.04.044
G Model
ARTICLE IN PRESS
BIOMAC-4311; No. of Pages 6
C. Morais et al. / International Journal of Biological Macromolecules xxx (2014) xxx–xxx
2
being evaluated since their use can reduce the overall production costs in about 40–50% [10]. Fatty acids and vegetable oils, as well as wastes and by-products rich in oils or fats, have been reported as suitable for PHA production by a wide range of microorganisms, including Cupriavidus necator, Comamonas testosteroni and Pseudomonas sp. [11–16]. C. necator is a well known P(3HB) producer able to accumulate high amounts of polymer (up to 80 wt.%) from plant and waste oils [13,15,17]. C. testosteroni has been reported to synthesize medium chain length PHA (mcl-PHA) during cultivation on vegetable oils, accumulating up to 87.5 wt.% [18]. Several Pseudomonas sp., including P. citronellolis, P. oleovorans, P. resinovorans and P. stutzeri have also been described to synthesize mcl-PHA from tallow, fatty acids and biodiesel co-product stream [11,12,14,19]. Oils and fats can be used by many microorganisms, in the presence of extracellular lipase that induces their enzymatic hydrolysis into free fatty acids that are transferred through the cell membrane and metabolized via -oxidation pathway to produce PHA monomers [15]. In C. necator, the synthesis of P(3HB) involves three enzymes and their encoding genes: (1) condensation of two acetyl-CoA molecules to form acetoacetylCoA, catalyzed by -ketothiolase (encoded by phaA gene); (2) reduction of acetoacetyl-CoA to (R)-3-hydroxybutyryl-CoA by the NADPH-dependent enzyme acetoacetyl-CoA reductase (encoded by phaB gene); (3) polymerization of (R)-3-hydroxybutyryl-CoA monomers catalyzed by PHA synthase (encoded by the phaC gene) [6,20]. Another type of PHA biosynthetic pathway is exhibited by Pseudomonas species that derive 3-hydroxyacyl-CoA from the intermediates of fatty acid -oxidation pathway, enoyl-CoA, 3ketoacyl-CoA, and/or S-3-hydroxyalcyl-CoA [20]. In this work, a fat-containing material waste generated by the margarine manufacturing process was, for the first time, tested as carbon source for the production of bacterial PHA. The margarine waste was characterized in terms of its physical and chemical properties. Several bacterial strains were screened for their ability to grow and produce PHA using the margarine waste as the sole carbon source and the highest PHA yielding strains was selected for bioreactor cultivation. The resulting PHA polymer was characterized in terms of its chemical and thermal properties.
2. Materials and methods 2.1. Margarine waste characterization The fat-containing waste from the margarine manufacturing process was supplied by FIMA, SA – Unilever, Portugal. Two margarine waste samples, supplied by the manufacturer at different times, were analyzed to determine their composition and assess their variability. The samples were characterized in terms of density, pH, water content, inorganic compounds content and the composition in organic compounds (total carbohydrates and lipids). All analyses were performed in duplicate. The margarine waste was a solid material at ambient temperature and it had to be melted by placing at a temperature of 50 ◦ C. The margarine waste analyzed for its carbon, hydrogen, nitrogen and sulphur content, using the elemental Analyzer Thermo Finnigan-CE Instruments (Italy), model Flash EA 1112 CHNS. The water content of the margarine waste was determined as the weight loss by a 2 mL sample upon lyophilization, in a percentage basis. The total carbohydrates content was determined using the phenol-sulphuric acid method [21], using glucose solutions (0–200 mg/L) as standards. The total sugar content was expressed as percentage of sugar in the margarine waste sample. To determine the total content in inorganic compounds and their composition, 1 mL of margarine waste was subjected to pyrolysis at 550 ◦ C, for 24 h. The resulting ashes were dissolved in 20 mL 2.3 M H2 SO4 . The
solution was analyzed by inductively coupled plasma-atomic emission spectrometry (ICP-AES), (Horiba Jobin-Y, France, Ultima), for quantification of aluminium, calcium, iron, magnesium, phosphorus, potassium and sodium. The margarine waste content in glycerine, mono-, di- and triglycerides was determined by on-column gas chromatography (GC) (Trace GC Ultra), according to the European norm EN 14105 (Thermo Fisher Scientific Inc.). Standard solutions (Biodiesel Consumables Kit EN), containing glycerin, monolein, diolein, triolein, butanetriol (IS1) and tricaprin (IS2) were used at the concentration specified in the European norm. IS1 (80 L), IS2 (100 L) and of N-methyl-N-(trimethylsilyl)triluoroacetamide (MSTFA reagent) (100 L) were added to 100 mg of margarine waste and the mixture was vigorously shaken. After 15 min, 8 mL of n-heptane were added, and the mixture was used for GC analysis. The free fatty acid content of the margarine waste was determined by automatic titration (TIM 86J Titration Manager) of a solution of margarine waste sample (0.1–0.5 g) in isopropanol (30 mL) with 0.1 M NaOH. The fatty acid composition of the margarine waste was determined by direct transesterification of the lipids to the corresponding methyl esters, according to a modified Lepage and Roy method [22]. The methyl esters were quantified by GC, with a Thermo Trace GC ULTRA gas chromatograph, equipped with a flame ionization detector and a split/splitless injector, according to European norm EN1403PTV (Thermo Fisher Scientific Inc.). Methyl heptadecanoate was used as internal standard.
2.2. Microbial cultivation experiments The bacterial cultures used in this study were Pseudomonas oleovorans strains NRRL B-14682, NRRL B-14683, NRRL B-778 and NRRL B-3429, P. resinovorans strains NRRL B-2649 and NRRL B4205, P. citronellolis NRRL B-2504, P. stutzeri strains NRRL B-775 and P. stutzeri NRRL B-2461, Comamonas testosteroni NRRL B-2611 and Cupriavidus necator DSM 428. All Pseudomonas strains and C. testosteroni were offered by the National Center for Agricultural Utilization Research, USA, and C. necator was purchased from the Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Germany. Luria Bertani (LB) medium (bactotryptone, 10.0 g/L; yeast extract, 5.0 g/L; NaCl, 10.0 g/L), pH 7.0, was used for reactivation of the cryopreserved bacterial cultures. Solid LB medium was prepared by adding agar (15 g/L). A nitrogen-limited mineral medium, with the composition described by Freitas et al. [23] was used for all shake flasks and bioreactor experiments. The mineral medium was supplemented with the margarine waste (20 g/L) as the sole carbon source. The margarine waste was autoclaved separately and added while hot (∼50 ◦ C) to the mineral medium. Prior to culture inoculation, the shake flasks containing the medium supplemented with the melted margarine waste were placed in an orbital shaker (at 30 ◦ C, 200 rpm, 24 h) to obtain an homogenous mixture. The stability of the mixture thus obtained was confirmed by no phase separation being observed after leaving the flasks at rest for several days. In the bioreactor experiments, the hot melted margarine waste was added to the mineral medium and the mixture was stirred (400 rpm) until homogenous mixtures were obtained prior to culture inoculation. The cultures were reactivated by inoculating LB agar plates with a sample of the cryopreserved microorganisms and incubation at 30 ◦ C for 24 h. Afterwards, isolated colonies were inoculated into 50 mL liquid LB and incubated in an orbital shaker at 200 rpm and 30 ◦ C, for 72 h. 10 mL of the cultures thus obtained were used as inocula for the 100 mL shake flask cultivations. In all shake flask experiments, the cultures were incubated under the same
Please cite this article in press as: C. Morais, et al., Int. J. Biol. Macromol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.04.044
G Model
ARTICLE IN PRESS
BIOMAC-4311; No. of Pages 6
C. Morais et al. / International Journal of Biological Macromolecules xxx (2014) xxx–xxx
conditions for 72 h. At the end of experiments, 20 mL samples were used for biomass and PHA quantification. Batch cultivations (three experiments) were performed in a 5 L bioreactor (BioStat® B-Plus, Sartorius) with a working volume of 4 L. In all experiments, the temperature and the pH were kept at 30 ± 0.1 ◦ C and 7.0 ± 0.1, respectively. pH was controlled by the automatic addition of 5 M NaOH and 2 M HCl solutions. A constant aeration rate (1 vvm, volume of air per volume of reactor per minute) was kept during all experiments. The dissolved oxygen concentration (DO) was controlled at 30% of air saturation by automatically adjusting the stirring speed between 300 and 800 rpm. Foam formation was automatically suppressed by addition of Antifoam A (Sigma). Samples (25 mL) were periodically taken from the bioreactor for quantification of the cell dry weight, PHA and margarine waste. For quantification of the cell dry weight (CDW), cultivation broth samples (10 mL) were lyophilized for 48 h. The dried broth was washed twice with 10 mL hexane and 10 mL deionized water. Finally, the washed cell pellets were resuspended in deionized water, transferred to pre-weighed flasks, lyophilized and reweighed. These fractions were used to determine the cell concentration expressed as the mass of dried biomass (g) per volume of the cultivation broth (L). For quantification of the margarine waste concentration, only the final sample of the bioreactor experiments was analyzed since large broth volumes (3 mL × 20 mL) were required for accurate (in triplicate) measurement, which were not possible to be withdrawn from the bioreactor throughout the cultivation run. The lyophilized broth (20 mL) was mixed with hexane (20 mL), stirred in the vortex (2500 rpm, 30 s) and centrifuged (16,743 × g, 10 min). Subsequently, the upper hexane layer was transferred to pre-weighed flasks and the solvent was evaporated, at room temperature, until a constant weight was obtained, corresponding to the samples’ content in margarine waste. PHA content and composition were determined after methanolysis of dried cell samples (3–5 mg) in 1 mL 20% (v/v) sulphuric acid in methanol and 1 mL benzoic acid in chloroform (1 g/L), at 100 ◦ C, for 3.5 h. The resulting methyl esters were analyzed by GC, as described by Albuquerque et al. [24]. All analyses were performed in duplicate. 2.3. Calculations The maximum specific growth rate (max , h−1 ) was determined from the linear regression slope of the exponential phase of Ln Xt versus time, where Xt (g/L) is the active biomass (i.e. cells without PHA) at time t (h). The residual biomass was determined by Eq. (1): Xt = CDWt − PHAt
(1)
where CDWt (g/L) and PHAt (g/L) are the cell dry weight and the concentration of polymer at time t (h). This concentration is given by the percentage of polymer accumulated in the cells (calculated on a dry basis). The product storage yield (YP/S , gPHA /gS ) was calculated by Eq. (2): YP/S =
P S
3
membrane, PALL) and the PHA was precipitated in ice-cold ethanol (chloroform/ethanol 1:10). Thereof, the mixture was centrifuged at 16,743 × g for 15 min at 4 ◦ C. The white precipitate was then recovered in a pre-weighed flask and left at ambient temperature, in a fume hood, for solvent evaporation. The polymer’s composition was determined by GC as described above. Thermal analysis was performed by differential scanning calorimetry (DSC) with a DSC Q2000 from TA Instruments interfaced with a cooling accessory (RCS). The DSC runs covered a temperature range from −90 to 200 ◦ C with heating and cooling rate of 10 ◦ C/min. The samples (2–3 mg) were placed in aluminium hermetic pans. Measurements were performed under dry high-purity nitrogen gas (at a flow rate of 50 mL/min). The baseline was calibrated scanning the temperature range of the experiments with two empty pans. Calibration was carried out using high purity Indium for temperature transitions and the heats of fusion. The glass transition temperature (Tg , ◦ C) was taken as the midpoint of the heat flux step and the melting temperature (Tm , ◦ C) was determined at the minimum of the endothermic peak. The crystallinity (c , %) of the samples (Eq. (3)) was determined by comparing the area of the melting peak (Hf , J/g) with the melting enthalpy of 100% crystalline P(3HB) (Hf100% ). The heat of fusion of an infinite crystal of P(3HB) was estimated as 146 J/g [25]. c =
Hf Hf100
× 100
(3)
3. Results and discussion 3.1. Margarine waste characterization According to elemental analysis, the margarine had a carbon and hydrogen contents of 75.25 ± 0.47 and 12.43 ± 0.29, respectively. It had no nitrogen content and only traces of sulfur were detected (0.39 ± 0.13). The margarine waste was mainly composed of free fatty acids (FFA) and triglycerides that accounted for 63.21 ± 2.01 wt.% and 16.13 ± 2.27 wt.%, respectively, of their weight. Minor contents of diglycerides (2.73 ± 0.44 wt.%) and traces of monoglycerides (0.22 ± 0.16 wt.%) were detected, while no glycerol was observed (Table 1). The high content of free fatty acids is in accordance with the low pH value of the margarine waste. The fatty acid profile analysis of the margarine waste samples revealed that they were mostly composed of oleic (51.50 ± 3.96 wt.%), linoleic (24.30 ± 3.82 wt.%) and stearic (11.95 ± 0.64 wt.%) acids, with minor contents of mystiric (6.2 ± 0.02 wt.%), palmitic (3.25 ± 0.07 wt.%) and linolenic (2.80 ± 0.71 wt.%) acids. These results are in accordance with the typical fatty acids composition reported for some vegetable oils and animal fats, where oleic and linoleic acids are commonly the main fatty acids present with (up to 84 wt.% and 62 wt.%, respectively) [26]. The saturated fatty acids (mystiric, palmitic and stearic) accounted for 21 wt.% of the margarine waste
(2)
where P (g/L) is the PHA produced and S (g/L) is the total fatty acids consumed during the cultivation (20 h). The volumetric productivity (rp, g/L h) was calculated by dividing the final PHA concentration (P, g/L) for the total time of fermentation (t, h). 2.4. Polymer extraction and characterization Lyophilized cells obtained as described above were subjected to extraction with chloroform (0.1 g/L) at 37 ◦ C for 72 h, at 200 rpm, on an orbital shaker. The cellular debris were removed by filtration with syringe filters with a pore size of 0.45 m (GxF, GHP
Table 1 Composition of the margarine waste. Component
wt.%
Acylglycerols MonoDiTriFree fatty acids Glycerol Carbohydrates Inorganic compounds Water
0.22 (±0.16) 2.73 (±0.44) 16.13 (±2.27) 63.21 (±2.01) n.d. 10.70 (±0.81)
ARTICLE IN PRESS
BIOMAC-4311; No. of Pages 6
International Journal of Biological Macromolecules xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Conversion of fat-containing waste from the margarine manufacturing process into bacterial polyhydroxyalkanoates Cristiana Morais, Filomena Freitas ∗ , Madalena V. Cruz, Alexandre Paiva, Madalena Dionísio, Maria A.M. Reis REQUIMTE/CQFB, Chemistry Department, FCT/Universidade Nova de Lisboa, 2829-516 Caparica, Portugal
a r t i c l e
i n f o
Article history: Received 10 February 2014 Received in revised form 15 April 2014 Accepted 22 April 2014 Available online xxx Keywords: Margarine waste Poly(3-hydroxybutyrate) Cupriavidus necator
a b s t r a c t A fat-containing waste produced from the margarine manufacturing process was tested as a low cost carbon source for cultivation of different polyhydroxyalkanoates (PHAs) producing bacterial strains, including Cupriavidus necator, Comamonas testosteroni and several Pseudomonas strains. The margarine waste was mainly composed of free fatty acids (76 wt.%), namely mystiric, oleic, linoleic and stearic acids. In preliminary shake flask experiments, several strains were able to grow on the margarine waste, but C. necator reached the highest PHA content in the biomass (69 wt.%). This strain was selected for batch bioreactor experiments, wherein it reached a cell dry weight of 11.2 g/L with a polymer content of 56 wt.%. The culture produced 6.4 g/L of polyhydroxybutyrate, P3(HB), within 20 h of cultivation, which corresponds to a volumetric productivity of 0.33 gPHA /L h. The P3(HB) polymer produced by C. necator from the margarine waste had a melting point of 173.4 ◦ C, a glass transition temperature of 7.9 ◦ C and a crystallinity of 56.6%. Although the bioprocess needs to be optimized, the margarine waste was shown to be a promising substrate for P(3HB) production by C. necator, resulting in a polymer with physical and chemical properties similar to bacterial P(3HB) synthesized from other feedstocks. © 2014 Elsevier B.V. All rights reserved.
1. Introduction In the margarine manufacturing process, vegetable oils or animal fats are modified by hydrogenation, rearrangement and fractionation, and blended with a mixture of water, brine and powdered ingredients [1]. Additives (e.g. vitamins, emulsifiers, salt, flavours) are included in the blend to improve the quality of the product and enhance flavour. The mixture is subjected to temperatures of 50–60 ◦ C, resulting in the formation of an emulsion (margarine) that is pasteurized and packed [1,2]. According to IMACE, International Margarine Association of the Countries of Europe (www.imace.org), in 2012, the European production of margarine was 2440 Mton, while worldwide it reached 9374 Mton. Around 1% of this production results in the generation fatcontaining wastes that require adequate disposal or treatment [3,4]. The fat-containing wastes are usually removed from the aqueous effluents produced by the plant in a gravity separator [3].
∗ Corresponding author at: Biochemical Engineering Lab, Chemistry Department, Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa, 2829-516 Caparica, Portugal. Tel.: +351 212 948 300; fax: +351 212948550. E-mail addresses: [email protected], fi[email protected] (F. Freitas).
Though having a low commercial value, it is commonly sold to oil-recycling companies. Fat-containing materials (e.g. tallow generated from the food industry) have been proposed as feedstocks for the production of value-added products, namely polyhydroxyalkanoates (PHA) [5]. The use of such waste materials is a valuable strategy to improve the producing industries sustainability and economical viability, by converting a waste into high-value products. PHAs are hydroxyalkanoic acids that are synthesized by many microorganisms as intracellular carbon and energy reserve materials or reducingpower storage materials [6]. These polymers possess physical characteristics similar to traditional plastics and have received extensive attention mainly due to their biodegradability and biocompatibility [6,7]. Polyhydroxybutyrate, P(3HB), is the most widely studied and best characterized PHA. It is a homopolymer of 3-hydroxybutyrate, which has mechanical properties similar to polypropylene [6]. Due to their properties, PHAs are used as packaging materials, biomedical devices and in the food industry (e.g. edible packaging, flavour delivery agent) [8,9]. However, PHA commercial applications have been limited by their high production cost that is mainly related to the high price of the carbon source [5]. In this context, the use of inexpensive renewable feedstocks is currently
http://dx.doi.org/10.1016/j.ijbiomac.2014.04.044 0141-8130/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: C. Morais, et al., Int. J. Biol. Macromol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.04.044
G Model
ARTICLE IN PRESS
BIOMAC-4311; No. of Pages 6
C. Morais et al. / International Journal of Biological Macromolecules xxx (2014) xxx–xxx
2
being evaluated since their use can reduce the overall production costs in about 40–50% [10]. Fatty acids and vegetable oils, as well as wastes and by-products rich in oils or fats, have been reported as suitable for PHA production by a wide range of microorganisms, including Cupriavidus necator, Comamonas testosteroni and Pseudomonas sp. [11–16]. C. necator is a well known P(3HB) producer able to accumulate high amounts of polymer (up to 80 wt.%) from plant and waste oils [13,15,17]. C. testosteroni has been reported to synthesize medium chain length PHA (mcl-PHA) during cultivation on vegetable oils, accumulating up to 87.5 wt.% [18]. Several Pseudomonas sp., including P. citronellolis, P. oleovorans, P. resinovorans and P. stutzeri have also been described to synthesize mcl-PHA from tallow, fatty acids and biodiesel co-product stream [11,12,14,19]. Oils and fats can be used by many microorganisms, in the presence of extracellular lipase that induces their enzymatic hydrolysis into free fatty acids that are transferred through the cell membrane and metabolized via -oxidation pathway to produce PHA monomers [15]. In C. necator, the synthesis of P(3HB) involves three enzymes and their encoding genes: (1) condensation of two acetyl-CoA molecules to form acetoacetylCoA, catalyzed by -ketothiolase (encoded by phaA gene); (2) reduction of acetoacetyl-CoA to (R)-3-hydroxybutyryl-CoA by the NADPH-dependent enzyme acetoacetyl-CoA reductase (encoded by phaB gene); (3) polymerization of (R)-3-hydroxybutyryl-CoA monomers catalyzed by PHA synthase (encoded by the phaC gene) [6,20]. Another type of PHA biosynthetic pathway is exhibited by Pseudomonas species that derive 3-hydroxyacyl-CoA from the intermediates of fatty acid -oxidation pathway, enoyl-CoA, 3ketoacyl-CoA, and/or S-3-hydroxyalcyl-CoA [20]. In this work, a fat-containing material waste generated by the margarine manufacturing process was, for the first time, tested as carbon source for the production of bacterial PHA. The margarine waste was characterized in terms of its physical and chemical properties. Several bacterial strains were screened for their ability to grow and produce PHA using the margarine waste as the sole carbon source and the highest PHA yielding strains was selected for bioreactor cultivation. The resulting PHA polymer was characterized in terms of its chemical and thermal properties.
2. Materials and methods 2.1. Margarine waste characterization The fat-containing waste from the margarine manufacturing process was supplied by FIMA, SA – Unilever, Portugal. Two margarine waste samples, supplied by the manufacturer at different times, were analyzed to determine their composition and assess their variability. The samples were characterized in terms of density, pH, water content, inorganic compounds content and the composition in organic compounds (total carbohydrates and lipids). All analyses were performed in duplicate. The margarine waste was a solid material at ambient temperature and it had to be melted by placing at a temperature of 50 ◦ C. The margarine waste analyzed for its carbon, hydrogen, nitrogen and sulphur content, using the elemental Analyzer Thermo Finnigan-CE Instruments (Italy), model Flash EA 1112 CHNS. The water content of the margarine waste was determined as the weight loss by a 2 mL sample upon lyophilization, in a percentage basis. The total carbohydrates content was determined using the phenol-sulphuric acid method [21], using glucose solutions (0–200 mg/L) as standards. The total sugar content was expressed as percentage of sugar in the margarine waste sample. To determine the total content in inorganic compounds and their composition, 1 mL of margarine waste was subjected to pyrolysis at 550 ◦ C, for 24 h. The resulting ashes were dissolved in 20 mL 2.3 M H2 SO4 . The
solution was analyzed by inductively coupled plasma-atomic emission spectrometry (ICP-AES), (Horiba Jobin-Y, France, Ultima), for quantification of aluminium, calcium, iron, magnesium, phosphorus, potassium and sodium. The margarine waste content in glycerine, mono-, di- and triglycerides was determined by on-column gas chromatography (GC) (Trace GC Ultra), according to the European norm EN 14105 (Thermo Fisher Scientific Inc.). Standard solutions (Biodiesel Consumables Kit EN), containing glycerin, monolein, diolein, triolein, butanetriol (IS1) and tricaprin (IS2) were used at the concentration specified in the European norm. IS1 (80 L), IS2 (100 L) and of N-methyl-N-(trimethylsilyl)triluoroacetamide (MSTFA reagent) (100 L) were added to 100 mg of margarine waste and the mixture was vigorously shaken. After 15 min, 8 mL of n-heptane were added, and the mixture was used for GC analysis. The free fatty acid content of the margarine waste was determined by automatic titration (TIM 86J Titration Manager) of a solution of margarine waste sample (0.1–0.5 g) in isopropanol (30 mL) with 0.1 M NaOH. The fatty acid composition of the margarine waste was determined by direct transesterification of the lipids to the corresponding methyl esters, according to a modified Lepage and Roy method [22]. The methyl esters were quantified by GC, with a Thermo Trace GC ULTRA gas chromatograph, equipped with a flame ionization detector and a split/splitless injector, according to European norm EN1403PTV (Thermo Fisher Scientific Inc.). Methyl heptadecanoate was used as internal standard.
2.2. Microbial cultivation experiments The bacterial cultures used in this study were Pseudomonas oleovorans strains NRRL B-14682, NRRL B-14683, NRRL B-778 and NRRL B-3429, P. resinovorans strains NRRL B-2649 and NRRL B4205, P. citronellolis NRRL B-2504, P. stutzeri strains NRRL B-775 and P. stutzeri NRRL B-2461, Comamonas testosteroni NRRL B-2611 and Cupriavidus necator DSM 428. All Pseudomonas strains and C. testosteroni were offered by the National Center for Agricultural Utilization Research, USA, and C. necator was purchased from the Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Germany. Luria Bertani (LB) medium (bactotryptone, 10.0 g/L; yeast extract, 5.0 g/L; NaCl, 10.0 g/L), pH 7.0, was used for reactivation of the cryopreserved bacterial cultures. Solid LB medium was prepared by adding agar (15 g/L). A nitrogen-limited mineral medium, with the composition described by Freitas et al. [23] was used for all shake flasks and bioreactor experiments. The mineral medium was supplemented with the margarine waste (20 g/L) as the sole carbon source. The margarine waste was autoclaved separately and added while hot (∼50 ◦ C) to the mineral medium. Prior to culture inoculation, the shake flasks containing the medium supplemented with the melted margarine waste were placed in an orbital shaker (at 30 ◦ C, 200 rpm, 24 h) to obtain an homogenous mixture. The stability of the mixture thus obtained was confirmed by no phase separation being observed after leaving the flasks at rest for several days. In the bioreactor experiments, the hot melted margarine waste was added to the mineral medium and the mixture was stirred (400 rpm) until homogenous mixtures were obtained prior to culture inoculation. The cultures were reactivated by inoculating LB agar plates with a sample of the cryopreserved microorganisms and incubation at 30 ◦ C for 24 h. Afterwards, isolated colonies were inoculated into 50 mL liquid LB and incubated in an orbital shaker at 200 rpm and 30 ◦ C, for 72 h. 10 mL of the cultures thus obtained were used as inocula for the 100 mL shake flask cultivations. In all shake flask experiments, the cultures were incubated under the same
Please cite this article in press as: C. Morais, et al., Int. J. Biol. Macromol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.04.044
G Model
ARTICLE IN PRESS
BIOMAC-4311; No. of Pages 6
C. Morais et al. / International Journal of Biological Macromolecules xxx (2014) xxx–xxx
conditions for 72 h. At the end of experiments, 20 mL samples were used for biomass and PHA quantification. Batch cultivations (three experiments) were performed in a 5 L bioreactor (BioStat® B-Plus, Sartorius) with a working volume of 4 L. In all experiments, the temperature and the pH were kept at 30 ± 0.1 ◦ C and 7.0 ± 0.1, respectively. pH was controlled by the automatic addition of 5 M NaOH and 2 M HCl solutions. A constant aeration rate (1 vvm, volume of air per volume of reactor per minute) was kept during all experiments. The dissolved oxygen concentration (DO) was controlled at 30% of air saturation by automatically adjusting the stirring speed between 300 and 800 rpm. Foam formation was automatically suppressed by addition of Antifoam A (Sigma). Samples (25 mL) were periodically taken from the bioreactor for quantification of the cell dry weight, PHA and margarine waste. For quantification of the cell dry weight (CDW), cultivation broth samples (10 mL) were lyophilized for 48 h. The dried broth was washed twice with 10 mL hexane and 10 mL deionized water. Finally, the washed cell pellets were resuspended in deionized water, transferred to pre-weighed flasks, lyophilized and reweighed. These fractions were used to determine the cell concentration expressed as the mass of dried biomass (g) per volume of the cultivation broth (L). For quantification of the margarine waste concentration, only the final sample of the bioreactor experiments was analyzed since large broth volumes (3 mL × 20 mL) were required for accurate (in triplicate) measurement, which were not possible to be withdrawn from the bioreactor throughout the cultivation run. The lyophilized broth (20 mL) was mixed with hexane (20 mL), stirred in the vortex (2500 rpm, 30 s) and centrifuged (16,743 × g, 10 min). Subsequently, the upper hexane layer was transferred to pre-weighed flasks and the solvent was evaporated, at room temperature, until a constant weight was obtained, corresponding to the samples’ content in margarine waste. PHA content and composition were determined after methanolysis of dried cell samples (3–5 mg) in 1 mL 20% (v/v) sulphuric acid in methanol and 1 mL benzoic acid in chloroform (1 g/L), at 100 ◦ C, for 3.5 h. The resulting methyl esters were analyzed by GC, as described by Albuquerque et al. [24]. All analyses were performed in duplicate. 2.3. Calculations The maximum specific growth rate (max , h−1 ) was determined from the linear regression slope of the exponential phase of Ln Xt versus time, where Xt (g/L) is the active biomass (i.e. cells without PHA) at time t (h). The residual biomass was determined by Eq. (1): Xt = CDWt − PHAt
(1)
where CDWt (g/L) and PHAt (g/L) are the cell dry weight and the concentration of polymer at time t (h). This concentration is given by the percentage of polymer accumulated in the cells (calculated on a dry basis). The product storage yield (YP/S , gPHA /gS ) was calculated by Eq. (2): YP/S =
P S
3
membrane, PALL) and the PHA was precipitated in ice-cold ethanol (chloroform/ethanol 1:10). Thereof, the mixture was centrifuged at 16,743 × g for 15 min at 4 ◦ C. The white precipitate was then recovered in a pre-weighed flask and left at ambient temperature, in a fume hood, for solvent evaporation. The polymer’s composition was determined by GC as described above. Thermal analysis was performed by differential scanning calorimetry (DSC) with a DSC Q2000 from TA Instruments interfaced with a cooling accessory (RCS). The DSC runs covered a temperature range from −90 to 200 ◦ C with heating and cooling rate of 10 ◦ C/min. The samples (2–3 mg) were placed in aluminium hermetic pans. Measurements were performed under dry high-purity nitrogen gas (at a flow rate of 50 mL/min). The baseline was calibrated scanning the temperature range of the experiments with two empty pans. Calibration was carried out using high purity Indium for temperature transitions and the heats of fusion. The glass transition temperature (Tg , ◦ C) was taken as the midpoint of the heat flux step and the melting temperature (Tm , ◦ C) was determined at the minimum of the endothermic peak. The crystallinity (c , %) of the samples (Eq. (3)) was determined by comparing the area of the melting peak (Hf , J/g) with the melting enthalpy of 100% crystalline P(3HB) (Hf100% ). The heat of fusion of an infinite crystal of P(3HB) was estimated as 146 J/g [25]. c =
Hf Hf100
× 100
(3)
3. Results and discussion 3.1. Margarine waste characterization According to elemental analysis, the margarine had a carbon and hydrogen contents of 75.25 ± 0.47 and 12.43 ± 0.29, respectively. It had no nitrogen content and only traces of sulfur were detected (0.39 ± 0.13). The margarine waste was mainly composed of free fatty acids (FFA) and triglycerides that accounted for 63.21 ± 2.01 wt.% and 16.13 ± 2.27 wt.%, respectively, of their weight. Minor contents of diglycerides (2.73 ± 0.44 wt.%) and traces of monoglycerides (0.22 ± 0.16 wt.%) were detected, while no glycerol was observed (Table 1). The high content of free fatty acids is in accordance with the low pH value of the margarine waste. The fatty acid profile analysis of the margarine waste samples revealed that they were mostly composed of oleic (51.50 ± 3.96 wt.%), linoleic (24.30 ± 3.82 wt.%) and stearic (11.95 ± 0.64 wt.%) acids, with minor contents of mystiric (6.2 ± 0.02 wt.%), palmitic (3.25 ± 0.07 wt.%) and linolenic (2.80 ± 0.71 wt.%) acids. These results are in accordance with the typical fatty acids composition reported for some vegetable oils and animal fats, where oleic and linoleic acids are commonly the main fatty acids present with (up to 84 wt.% and 62 wt.%, respectively) [26]. The saturated fatty acids (mystiric, palmitic and stearic) accounted for 21 wt.% of the margarine waste
(2)
where P (g/L) is the PHA produced and S (g/L) is the total fatty acids consumed during the cultivation (20 h). The volumetric productivity (rp, g/L h) was calculated by dividing the final PHA concentration (P, g/L) for the total time of fermentation (t, h). 2.4. Polymer extraction and characterization Lyophilized cells obtained as described above were subjected to extraction with chloroform (0.1 g/L) at 37 ◦ C for 72 h, at 200 rpm, on an orbital shaker. The cellular debris were removed by filtration with syringe filters with a pore size of 0.45 m (GxF, GHP
Table 1 Composition of the margarine waste. Component
wt.%
Acylglycerols MonoDiTriFree fatty acids Glycerol Carbohydrates Inorganic compounds Water
0.22 (±0.16) 2.73 (±0.44) 16.13 (±2.27) 63.21 (±2.01) n.d. 10.70 (±0.81)
