Plant Oils and Products of Their Hydrolysis as Substrates for Polyhydroxyalkanoate Synthesis

Plastics are a ubiquitous part of modern life. They can be modified to meet the demands of almost any conceivable application1. While plastic is extremely useful and has been very important in the development of society since the beginning of the 20th century, it also presents a problem. It is estimated that 25 Mt of plastic waste was generated in the European Union in 2008. As much as 49 % of this waste ended up in landfill sites2. Plastic persists in the environment3 and plastics buried in landfills can leach harmful chemicals into groundwater4. Almost all plastics currently in use are produced using petrochemicals derived from fossil oil and gas. It is Plant Oils and Products of Their Hydrolysis as Substrates for Polyhydroxyalkanoate Synthesis


Introduction
Plastics are a ubiquitous part of modern life.They can be modified to meet the demands of almost any conceivable application 1 .While plastic is extremely useful and has been very important in the development of society since the beginning of the estimated that 8 % of annual petroleum production is used in the production of plastic 5 .There is a need for new methods of producing plastics that are derived from renewable resources and are biodegradable so the environmental impact of industrially produced materials can be reduced.
Polyhydroxyalkanoates (PHA) are polymers naturally synthesized as carbon storage molecules by a range of bacteria including Pseudomonas.Certain bacteria accumulate PHA as a carbon and energy store under conditions where there is an excess of carbon and a limitation in another essential nutrient such as nitrogen or phosphorous 6 .PHA is biodegradable in all biologically active environments and can be produced from renewable resources 7 .Fatty acids have traditionally been the preferred substrates for studies investigating the production of medium chain length (mcl) PHA 8,9 , but they are generally expensive as a purified single fatty acid.
Plant oils as substrates for PHA production may be interesting due to their relative low cost and the minimal processing required compared to fatty acids prior to fermentation.In 2014, the oils tested in this study ranged in price from an average of approximately $700/mt for palm oil to an average of approximately $4000/mt for olive oil 10 .While many studies have investigated the accumulation of scl-PHA from oils, we have focused solely on mcl-PHA production.Furthermore, a comparative study of oils versus fatty acid mixtures is not currently available.Mcl-PHA has been accumulated by Pseudomonas aeruginosa using palm oil 11 , Brassica carinta oil 12 and waste frying oil 13 , however, this species is an opportunistic pathogen and not suitable for PHA production at an industrial level.Pseudomonas chlororaphis HS21 was previously used to produce mcl-PHA from palm oil 14 .Comamonas testosteroni produced mcl-PHA from a range of oils, such as castor seed oil, coconut oil, mustard oil, cottonseed oil, groundnut oil, olive oil, and sesame oil 15 .Pseudomonas resinovorans B2649 was shown to utilize lard, butter oil, olive oil, sunflower oil, coconut oil and soybean oil to produce mcl-PHA 16 , and Pseudomonas saccharophilia B628 was cultured using soybean oil and sunflower oil to produce mcl-PHA 17 .Genetic engineering has also been used to express lipase genes in well-known PHA-producing strains, such as P. putida, P. oleovorans and P. corrugata to allow them to grow directly on oils 18,19 .Surprisingly, none of these studies has examined the ability of the strains to grow to high cell density when cultivated on plant oils, growth has been carried out in shake flasks or to low cell density in batch fermentation.Some of the studies compared the growth on oils to growth on individual pure fatty acids, mainly oleic acid 11,12,13,14,16 , but there has been no comprehensive comparison between growth and PHA production using plant oils and crude fatty acid mixtures (generated from the hydrolysis of oils).
Industrially, fatty acids are predominantly produced through the hydrolysis of plant oils, mainly soybean oil, palm oil, rapeseed oil, sunflower-seed oil, cotton oil, and olive oil), which represent about 85 percent of the fats production worldwide, with the remaining 15 percent coming from animal-origin fats, such as tallow, lard, butter, poultry fat, and fish oil.Industrially, fats and oils are hydrolysed in the splitting process to yield free fatty acids and glycerin.Typical splitting plants operate as continuous or batch processes at temperatures of 210 -250 °C and vapor pressures of 20 -40 bar without catalysts.However, in this study, saponification was used to represent fatty acid production, as this is more appropriate for a laboratory setting due to technical and mechanical constraints 20 .
Oleic acid prices in 2014 were significantly higher than most raw oil prices at approximately $1800/mt, however, it is true that crude fatty acid mixtures would certainly be cheaper to produce than purified single fatty acids, as no distillation is required, and may also produce PHA with different properties to that produced on single fatty acids 10,21 .There has been a limited number of reports of the use of saponified fatty acids for mcl-PHA production.Saponified palm kernel oil has been used to produce mcl-PHA (3 g L -1 CDW 37 % PHA) in Pseudomonas putida PGA1 21,22 .Mcl-PHA has been accumulated by Pseudomonas aeruginosa (3.5-5.5 g L -1 CDW, 66.1 % PHA) from waste free fatty acids from soybean oil 13 and from saponified waste palm oil by Pseudomonas sp G101 (5.2 g L -1 CDW, 42 % PHA) 23 .
In the current study, we have tested both plant oils and fatty acid mixtures generated from the saponification of olive oil, sunflower oil, rapeseed oil and palm oil as substrates for mcl-PHA production by P. putida KT2440, CA3, GO16, and P. chlororaphis 555.We examined the effect of oil type and bacterial strain on PHA accumulation levels and monomer composition, as well as attempted to grow strains to high cell density and high PHA productivity.

Bacterial growth medium & strain maintenance
Minimal Salt Medium (MSM), prepared as previously described 24 was used as the growth medium for all strains in shaken flask experiments.Pseudomonas putida KT2440, CA-3 and Pseudomonas chlororaphis 555 were grown on Pseudomonas isolation agar (PIA) (Fluka analytical).Pseudomonas putida GO16 was grown on MSM medium supplemented with sodium terephthalate (TA) (1.95 g carbon L -1 ) and solidified with 15 g L -1 agar as it grew poorly on PIA.TA was used as a selective carbon source, to lower the risk or contamination by other bacterial strains, as P. putida GO16 was originally isolated using this carbon source 54 Growth conditions for PHA accumulation in shaken flask cultures Growth experiments were carried out in 250 mL Erlenmeyer flasks containing 50 mL of nitrogen-limited MSM medium supplemented with the appropriate carbon source at a concentration of 3 g L -1 .Ammonium chloride in this medium was limited to 0.25 g L -1 .An overnight culture was prepared by inoculating 2 mL of MSM supplemented with the appropriate carbon source with a single colony from a plate and incubating overnight at 30 °C and shaking at 250 rpm. 100 µL of the 2 mL MSM overnight culture was used as an inoculum for the 50 mL flasks, which were then incubated under the same conditions for 48 h.All of the oils used in this study were of technical grade and obtained from Sigma Aldrich.All four strains were tested in this manner on all of the oils and all of their hydrolysis products.
Fermentation of Pseudomonas chlororaphis 555 on rapeseed oil and a fatty acid mixture derived from rapeseed oil Inoculation culture conditions.Four 50 mL cultures of P. chlororaphis were prepared in MSM as described above, with either rapeseed oil (RO) or hydrolysed rapeseed oil fatty acids (HROFA) as the carbon source.The cultures were grown for between 16 and 18 h before transfer to the bioreactor.
Fermentation conditions.Fermentations were performed in a Biostat B+ bioreactor with a 5 L working volume (Sartorius).MSM was used as the base media for all fermentations.The fermentations had a final volume of 5 L, with initial agitation at 500 rpm.The pH was controlled at 6.9 +/-0.1 by addition of 20 % NH 4 OH solution or 15 % (v/v) H 2 SO 4 .The NH 4 OH also provided a nitrogen source.Foaming was controlled by addition of antifoam solution (polypropylene glycol P2000, Sigma).Dissolved oxygen (DO) was controlled at 20 % of maximum by increasing agitation.O 2 was supplied as air at a constant flow rate of 5 litres per minute.An oil concentration of 60 g L -1 was used for the fermentation on pure rapeseed oil.An initial substrate concentration of 1 g L -1 (HROFA) was used for the fed-batch fermentation on fatty acids derived from rapeseed oil.90 g L -1 of HROFA was fed over 25 hours by peristaltic pump, the feed rate was de-signed to ensure that the concentration of fatty acids in the bioreactor never exceeded 20 mmol L -1 .To achieve this, the culture was grown at an average set µ of 0.2 h -1 by limiting the rate at which the HROFA were supplied to the fermentation.This rate of feeding was increased exponentially every hour in order to maintain the average set µ.

Oil hydrolysis
An amount of 100 g of oil was added to 100 mL of 6 mol L -1 NaOH.The mixture was heated to 60 °C for 1.5 h.The hydrolysed oil was then mixed with 200 mL of 6 mol L -1 HCl.The mixture was decanted into a separating funnel and the fatty acid phase separated from the aqueous phase containing salt and glycerol.The fatty acid phase was washed with approx.10 volumes (1000 mL) of distilled water.

Analytical techniques Analysis of fatty acids in hydrolysed oils
The mixtures of fatty acids produced from oil hydrolysis were derivatised with N-Methyl-N -(trimethylsilyl)trifluoroacetamide (TMS). 2 mL of chloroform were placed in a gas chromatography vial. 1 µL of the substrate and 20 µL of TMS were added.The vial was sealed and placed at 70 °C for 30 minutes.The derivatised fatty acids were then analyzed using an Agilent 6890N gas chromatograph (GC) fitted with a 5973 series inert mass spectrophotometer (MS).A HP-5 column (12 m × 0.2 mm × 0.33 µm; Hewlett Packard) was used with an oven method of 50 °C for 3 minutes, increasing by 10 °C min -1 to 250 °C and holding for 1 minute.A 10:1 split was used with helium as the carrier gas and an inlet temperature of 250 °C.
Analysis of free fatty acid concentration in oils 2 g of oil were dissolved in 40 mL of a 50:50 mixture of diethylether and ethanol.Two drops of 1 % Phenolphthalein indicator (in ethanol) were added and the mixture titrated against 0.01 mol L -1 NaOH.The following equation was used to determine the free fatty acid content of the oil.OA = oleic acid 25 .
Nutrient and biomass analysis Samples were taken at 0.5-, 1-, or 2-hour intervals during the fermentations.Two 2-mL samples were taken and centrifuged at 17960 g for 3 minutes.The supernatant was decanted into a separate tube, and the cell pellets and supernatant frozen at -80 °C.The cell pellets were then lyophilysed.The lyophilysed pellets were weighed to give an estimation of the cell biomass during the fermentation.50 mL flask cultures were centrifuged at 3220 g for 10 minutes, 2 mL of supernatant were retained and the remainder discarded.Pellets were resuspended in 1 mL of DI water and transferred into Eppendorf tubes.These were then centrifuged at 17960 g for 3 minutes, the supernatant discarded, and the pellets treated as the fermenter samples.The concentration of nitrogen in the supernatant was determined using the method described by Scheiner 26 .The soluble inorganic phosphate concentration was determined using the USA EPA colorimetric method 27 .

Determination of PHA content and monomer composition
PHA content was determined by subjecting lyophilised cells to acidic methanolysis 28,29 .From 5 to 10 mg of dried cells were resuspended in 2 mL of acidified methanol (15 % H 2 SO 4 , v/v), and 2 mL of chloroform containing 6 mg L -1 benzoate methyl ester as an internal standard.The solution was placed in 15 mL Pyrex test tubes, sealed and incubated at 100 °C for 3 hours.The tubes were then placed on ice for 1 minute.1 mL of water was added to each tube and the solution mixed by vigorous vortexing.The phases were allowed to separate, and the organic phase was removed and passed through a filter before further analysis.
The 3-hydroxyalkanoic acid methyl esters were analyzed by gas chromatography (GC) using an Agilent 6890N chromatograph equipped with a HP Innowax column (30 m x 0.25 mm x 0.5 µm), and a flame ionization detector (FID).An oven ramp cycle was employed as follows, 120 °C for 5 minutes, increasing by 3 °C min -1 to 180 °C, 180 °C for 10 minutes.A 20:1 split was used with helium as the carrier gas and an inlet temperature of 250 °C.
Commercially available 3-hydroxyalkanoic acids (Bioplastech Ltd.Dublin Ireland) were methylated as described above for PHA samples, and used as standards to identify PHA monomers.

Polymer isolation
Cells were harvested from cultivations in a Thermo Sorvall Contifuge Stratos continuous flow centrifuge (Fisher Scientific, Dublin, Ireland) at 25040 × g.Harvested cells were frozen at −80 °C for 24 h and then lyophilised (Labconco, Fisher Scientific).PHA was isolated from freeze-dried cells using room temperature acetone.This involved the stirring of 10 g of cells suspended in 100 mL acetone for 24 hours.The mixture was allowed to decant, and the supernatant was filtered using a 0.2μm PTFE filter.Acetone containing PHA was then subjected to rotary evaporation under vacuum until approximately 90 mL of acetone had been recovered.The polymer was precipitated using 2 vol. of a wash solution consisting of 35 % methanol, 35 % ethanol and 30 % distilled water 30 .The supernatant was then decanted, and the precipitated PHA was allowed to dry before further analysis.

Gel permeation chromatography (GPC).
The average molecular weight (M w ), the molecular number (Mn), and the polydispersity index (PD) of the polymer were measured by GPC using PL gel 5 mm mixed-C + PL gel column (Perkin-Elmer) with PELV 290 UV-vis detector set at 254 nm.Spectroscopic grade chloroform was used as the eluent at flow rate of 1.0 mL min -1 .Sample concentration of 1 % (w/v) and injection volumes of 500 µL were used.A molecular weight calibration curve was generated with polystyrene standards with low polydispersity 31 .

Differential scanning calorimetry (DSC).
The polymer was analyzed by DSC with a Perkin-Elmer Pyris Diamond calorimeter calibrated to Indium standards to determine the glass transition temperature (T g ), melting temperature (T m ), and degradation temperature (T d ).The samples were encapsulated in hermetically sealed aluminum pans and heated from -70 to 100 °C at a rate of 10 °C min -1 .To determine the glass transition temperature (T g ), the samples were held at 100 °C for 1 minute, and rapidly quenched to -70 °C.The samples were then reheated from -70 to 100 °C at 10 °C min -1 to determine the melting temperature (T m ) and T g .Finally, the samples were heated to 350 °C at a rate of 10 °C min -1 to determine the T d 31  .

Results and discussion
Hydrolysed oil fatty acid (HOFA) composition Each of the HOFA were analyzed for their fatty acid content by GCMS (Table 2).HROFA and HOOFA were predominantly composed of oleic acid, while the HSOFA was predominantly composed of linoleic acid.HPOFA was the only oil tested to have a saturated fatty acid predominating with palmitic acid, the predominant acid detected.
P. putida KT2440, the most commonly studied bacterial strain for mcl-PHA production, cannot grow on pure oils, therefore the oils were hydrolysed to produce fatty acid mixtures and glycerol.Glycerol is not a focus of the current study, but numerous studies have shown its usefulness for the production of polyhydroxyalkanoate 40,41,42 .
The fatty acid mixtures produced by hydrolysing oils were analysed by GC-MS to determine their fatty acid composition (Table 2).The fatty acid compositions of the fatty acid mixtures derived from oils in this study agree well with published fatty acid compositions of these oils 43,44 .The common European plant oils; olive, sunflower and rapeseed, consist mainly of unsaturated C 18 fatty acids.Due to this fatty acid profile, these mixtures are liquid at room temperature (~20 °C).Palm oil consists of predominantly C 16 and C 18:1 fatty acids.Palm oil-derived fatty acids are solids at temperatures up to 35 °C.This may pose problems for supply of these substrates to laboratory-scale bioreactors in a fed-batch scenario, as they would need to be heated.

Growth and PHA production in shaken flasks with plant oils
The oils used in this study were olive oil (OO), sunflower oil (SO), rapeseed oil (RO) and palm oil (PO).These oils are among the nine major vegetable oils produced globally 31 .The EU is the world's leading producer and consumer of olive oil and rapeseed oil, and a major producer of sunflower oil.Palm oil is not produced in Europe, but is a major import, and is a globally important oil 33 .The pure oils were tested as substrates for growth and PHA production in a range of PHA-producing strains.
Pseudomonas aeruginosa has been shown to convert pure plant oil to mcl-PHA 11,13 .However, as P. aeruginosa is an opportunistic pathogen and is not relevant industrially, it was not used in this study.A number of Pseudomonas strains (P.putida KT2440, CA-3 and GO16, P. chlororaphis 555) were tested for their ability to grow and produce PHA on a range of plant oils.The Pseudomonas putida strains have previously been used to produce mcl-PHA from a variety of substrates 34,35,36 .Other P. chlororaphis strains have been used to produce mcl-PHA from palm kernel oil 14 .P. putida GO16 and P. chlororaphis 555 were capable of utilizing all of the plant oils tested as the sole source of carbon for both growth and PHA accumulation (Table 3).The level of free fatty acids present in the oils (0.2-1.1 % (Table 4)) was not high enough to support growth to the biomass levels seen with plant oils, therefore these bacteria must be using the oils as their carbon source.The low levels of growth observed for P. putida KT2440 and CA-3 can be attributed to the free fatty acids present in the oils, and not growth on the oil itself.P. chlororaphis 555 grew to a higher cell density and accumulated more PHA than P. putida GO16 on all of the oils tested.P. putida GO16 achieved the highest growth and PHA from olive oil (0.56 g L -1 CDW, 25 % PHA), and P. chlororaphis 555 achieved the highest levels of growth and PHA from palm oil (0.86 g L -1 CDW, 34 % PHA).To the best of our knowledge, there have been no reports of wild type P. putida growth or PHA production directly on plant oils; however, P. chlororaphis has previously been shown to produce mcl-PHA from palm kernel oil 14 .
Growth and PHA production in shaken flasks with hydrolysed oil fatty acid (HOFA) mixtures All strains grew and produced PHA on all of the HOFA substrates tested (Table 5).The other strains in this study varied in their optimal substrate for growth.The highest biomass observed for P. putida KT2440 was with fatty acids derived from SO (0.82 g L -1 ), while the maximum PHA (28.61 %) accumulated by this strain was when fatty acids derived from PO were supplied.P. putida CA3 achieved its highest PHA on fatty acids derived from OO and RO (1.02 g L -1 and 36-39 % PHA).The highest biomass observed for GO16 was when it was supplied with fatty acids from hydrolysed OO (1.06 g L -1 ), but it accumulated the highest level of PHA on fatty acids derived from PO (44 %).P. chlororaphis 555 grew to the highest biomass and produced the highest amount of PHA on fatty acids from OO (0.94 g L -1 , 39 %).In another study, Pseudomonas putida PGA1, was grown on saponified palm kernel oil achieving similar biomass yields and PHA content to that reported here 49 .However, to the best of our knowledge, there have been no reports of Pseudomonas putida strains grown directly on any of the other saponified oil fatty acid mixtures used in this study.
Batch and fed batch fermentation using plant oils and P. chlororaphis 555 P. putida CA-3 and P. putida KT2440 showed no significant growth on any of the oils, and since P. chlororaphis 555 produced significantly higher biomass and PHA than P. putida GO16 when grown on plant oils, it was chosen as the most appropriate strain for fermentation studies (Table 2).While the other oils tested allowed for higher levels of growth than Rapeseed oil, economic and geographical factors were taken into account.Rapeseed was used as the substrate because the production of rapeseed oil in Europe, particularly for industrial applications, has been increasing rapidly since 2003 50 .
As oil is not toxic to cells at high concentrations, a batch fermentation can be carried out with a high initial oil concentration.This process is easier to scale up than fed-batch fermentation.Therefore, batch fermentations of P. chlororaphis on 60 g L -1  rapeseed oil were carried out.The fermentations produced 51 g/l biomass containing only 26 % PHA in 25 hours (Figure 1).This represents a high biomass yield (85 %), but the PHA content is still as low as in the shaken flask experiments.While pure oils show potential as low-cost renewable substrates for PHA production, further bioprocess development is required.Pure fatty acids have been shown to give very high biomass and PHA yields when used as single substrates for PHA production.141 g L -1 biomass containing 51.4 % PHA was achieved using P. putida KT2440 grown on oleic acid under inorganic phosphorous limitation for 38 hours with oxygen enriched air 47 .Therefore, mixtures of fatty acids produced by hydrolysing oils may give higher yields of biomass and PHA than the pure oils themselves.To investigate this, we carried out a fermen-tation of P. chlororaphis 555 using HROFA as the substrate.As fatty acids are reported to be toxic to cells at high concentrations 51 a fed-batch strategy was employed to grow 555 with HROFA in a 5L fermenter.This fermentation strategy yielded 76 g L -1 biomass containing 17 % PHA (Figure 2) in 25 hours.Inorganic phosphorous was limited in this fermentation after 18 h as was the case for the RO fermentation.The level of PHA was lower than expected as 34 % of the CDW was PHA in shaken flasks (Table 5).In general, PHA accumulation levels should be higher in bioreactor fermentations compared to shaken flasks where process conditions can be controlled to a higher degree.This PHA yield could potentially be improved by imposing inorganic phosphorous limitation earlier or using nitrogen limitation which may act as a better stimulant for PHA production.may play a role in determining the final application of these types of polymers, and so should not be discounted completely.Considering the variations in monomer composition between strains and on the various oils and fatty acid mixtures, it is important to consider both strain and substrate when designing fermentation strategies, not only in terms on growth and PHA productivity, but also in terms of polymer properties.

Conclusion
Plant oils and fatty acid mixtures from hydrolysed plant oils are good substrates for bacterial growth and mcl-PHA accumulation in shake flask experiments.However, PHA productivity from these substrates in bioreactor experiments needs to be improved in order to progress such processes towards scale-up and application.As shown, untreated oils have potential for mcl-PHA production.As these substrates cost less and are more readily scaled, we would suggest that further study and optimization could provide the further productivity gains required for cost effectiveness.

ACKNOWLEDGEMENT Meg Walsh is funded by the Irish Research Council Employment Based Postgraduate Programme (Ref no. 39826).
R e f e r e n c e s