R

Eb

AC

ARRAA

K�MACB

1

tthtgsdvocgggofaps

su

m

1h

Biochemical Engineering Journal 78 (2013) 11– 17

Contents lists available at SciVerse ScienceDirect

Biochemical Engineering Journal

jou rna l h om epage: www.elsev ier .com/ locate /be j

egular article

xploring potential use of Australian thraustochytrids for theioconversion of glycerol to omega-3 and carotenoids production

darsha Gupta, Dilip Singh, Colin J. Barrow ∗, Munish Puri ∗

entre for Chemistry and Biotechnology, Geelong Technology Precinct, Deakin University, Waurn Ponds, Victoria 3217, Australia

a r t i c l e i n f o

rticle history:eceived 31 October 2012eceived in revised form 24 April 2013ccepted 25 April 2013vailable online 15 May 2013

a b s t r a c t

Marine microbes have the potential for accumulating large quantities of lipids and are therefore suitablecandidate as feedstock in unsaturated fatty acid production. The efficient utilisation of glycerol as analternative carbon source to glucose was demonstrated in the fermentation of newly isolated thraus-tochytrid strains from the Queenscliff, Victoria, Australia. The isolates exhibited the presence of omega-3and omega-6 polyunsaturated fatty acids, with the major fatty acids for all isolates being (as percent total

eywords:-Carotenearine microalgae

staxanthinanthaxanthin

fatty acid), palmitic acid (25.1–40.78%), stearic acid (4.24–13.2%), eicosapentaenoic acid EPA (2.31–8.5%)and docosapentaenoic acid (7.24–10.9%). Glycerol as a carbon source gave promising biomass growthwith significant lipid and DHA productivity. An approximate three-fold increase in carotenoid content inall isolates was achieved when glycerol was used as a carbon source in the production medium.

© 2013 Elsevier B.V. All rights reserved.

iofuel

. Introduction

Glucose has been used extensively for microalgae fermenta-ion, having yielded high biomass growth and lipid productivityhan other carbon sources [1–5]. These resulting high carbon lipidsave been proposed as a source of sustainable oil production andhus present a feasible alternative for the production of third-eneration biofuels. Microalgae also produce other metabolites,uch as astaxanthin, lutein, arachidonic, eicosapentaenoic (EPA),ocosahexaenoic acids (DHA), all of which are of high economicalue. Maximising lipid productivity depends on the optimisationf an array of nutrient conditions in the growth medium, a pro-ess which must be repeated for new isolated strains [6]. Recently,lycerol has been used as an alternative carbon source instead oflucose due to its abundance and relatively low cost compared tolucose. Biodiesel derived glycerol can be used as a carbon source tobtain value added secondary metabolites such as polyunsaturatedatty acids (PUFAs) [7]. Glycerol could therefore be employed as anlternative carbon source in the fermentation process to cheaplyroduce DHA as a value added co-product, thereby taking anothertep towards economic viability of these fermentation processes.

Thraustochytrids, large-celled marine heterokonts and clas-ified as oleaginous microorganism, have been reported totilise a range of substrates such as glucose, galactose, fructose,

∗ Corresponding authors. Tel.: +61 3 52272325/52271318; fax: +61 3 52272170.E-mail addresses: colin.barrow@deakin.edu.au (C.J. Barrow),

unish.puri@deakin.edu.au (M. Puri).

369-703X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.bej.2013.04.028

mannose, sucrose [6,8], complex organic matter [9] and morerecently, cellulosic biomass [10], for the production of polyun-saturated fatty acids (PUFAs). They are considered to be highlypromising DHA producers [11]. Thraustochytrids have many ben-efits, including the potential use of their lipid rich biomass inbiodiesel and PUFA production, particularly DHA. DHA is a par-ticularly important omega-3 fatty acid due to its health benefitsin humans and commercial use in infant formula [11,12]. Thraus-tochytrids have the potential to produce other metabolites such ascarotenoids; astaxanthin, canthaxanthin, �-carotene, zeaxanthinand echinenone [13]. Carotenoids have potent antioxidant activi-ties and are therefore potentially beneficial to human health as theymay assist in the treatment of cancer and eye vision [14]. In fact,some carotenoids such as zeaxanthin are endogenous in humansand are an important component of eye retina. Thraustochytrids arealso known to secrete enzymes, polysaccharides and carotenoids,squalene and co-enzymes [15,16]. A commercial thraustochytridstrain, Schizochytrium limacinum SR21, has recently been used byresearchers studying PUFA production utilising biodiesel-derivedglycerol, corn steep liquor, and other organic nutrients such aswaste water from barley distilleries, soybean cake, liquid residuesfrom beer and potato processing, and sweet sorghum juice [17–23].

In this study, the utilisation of glycerol as the sole carbon sourcefor the production of value-added lipids and carotenoids by Thraus-tochytrium sp. is documented for strain AMCQS5-5 (a newly isolated

strain from the Queenscliff region, Victoria, Australia). The effectsand optimum levels of glycerol concentration and C/N ratio weredetermined for optimal production of carotenoids and maximumbiomass and lipid content.

1 ginee

2

2

ge(v(aaCrcp

2

anJAfiaaewut22wpwcm(a0(mduihuc

2

4upwa

2

vta

2 A. Gupta et al. / Biochemical En

. Materials and methods

.1. Chemicals

The chemicals used in this study were of analytical and HPLCrade. The other medium components such as glycerol, yeastxtract and mycological peptone (Sigma–Aldrich, USA) and sea saltInstant Ocean, USA) were used for biomass production while sol-ents such as acetone, dimethyl sulphoxide (DMSO), ether, hexanefrom Merck), methanol and ethyl acetate (HPLC grade from Fischernd Honeywell, Australia) were used for carotenoid extractionnd HPLC analysis for carotenoid identification and quantification.arotenoid standards (astaxanthin, zeaxanthin, canthaxanthin,es/meso-astaxanthin, beta-cryptoxanthin, echinenone) were pro-ured from CaroteNature, Switzerland while �-carotene wasrocured from Sigma–Aldrich, Australia.

.2. Strain selection and biomass production

Thraustochytrid isolates designated as AMCQS5-5 (Genbankccession number JX993841), AMCQS5-3 (Genbank accessionumber JX993839), and AMCQS1-9 (Genbank accession number

X993843) were isolated from the Queenscliff region, Victoria,ustralia. The isolates maintained on GYP agar plates were selected

or this study based on colour exhibiting ability. These isolates weredentified as thraustochytrids based on the colony morphology,ppearance and reproduction pattern (i.e. sporangium formationnd zoospore production) and molecular identification (Guptat al., unpublished). In addition, Schizochytrium S31 (ATCC 20888)as procured from American Type Culture Collection (ATCC) andsed as standard culture. The isolates used in this study were main-ained on GYP medium consisting (g L−1): glucose 5, yeast extract, mycological peptone 2, agar 10 and artificial seawater 50% at5 ◦C and sub-cultured after 15 days. The thraustochytrid isolatesere cultivated in a medium containing (g L−1): yeast extract 2,eptone 2 and artificial seawater 50% for inoculum preparationith shaking at 150 rpm for 2 days at 25 ◦C. The medium was auto-

laved at 121 ◦C for 20 min. Seed medium (0.2% yeast extract, 0.2%ycological peptone, ASW 50%, pH 6.5) and production medium

1% yeast extract, 0.1% mycological peptone, ASW 50%, pH 6.5) wasutoclaved followed by the addition of syringe filtered (0.2 �m).5% glucose and 1% glucose, respectively. Glucose and glycerol1%) were used as carbon source in the production medium. Seed

edium (50 mL) was inoculated from agar plates and grown for 2ays in a shake flask at 25 ◦C, at 150 rpm. Inoculum (5%, v/v) wassed to inoculate 95 mL production medium and cultured for 7 days

n a shake flask at 25 ◦C, and 150 rpm. The resultant biomass wasarvested by centrifugation (10,000 × g, 15 min) and freeze-driedntil further use. Some of the in-house isolates exhibited orangeolour and were considered for carotenoid extraction.

.3. Growth at different glycerol and glucose concentrations

Different concentrations of glycerol and glucose (0.5%, 1%, 2%,%, 6%, 8% and 10%) were used in the fermentation medium to eval-ate their effect on fatty acid production. The fermentation waserformed at 150 rpm and 25 ◦C for 7 days. The biomass growthas monitored by measuring the optical density (OD) at 600 nm

fter every 24 h.

.4. Fatty acid production and cell dry weight

To determine the fatty acid production, the culture was har-ested at the end of 7 days and centrifuged at 10,000 × g for 10 mino obtain the pellet. The cell pellet was freeze-dried and storedt −20 ◦C before proceeding with fatty acid extraction. The cell

ring Journal 78 (2013) 11– 17

dry weight (CDW) was estimated after freeze drying the thraus-tochytrid cells. Results are presented as mean ± SD of duplicatesrepeated twice.

2.5. Fatty acid extraction, esterification and GC analysis

Fatty acid extraction was performed according to Gupta andco-workers [4,11] with some modifications. 10 mg of freeze-driedcells were taken in centrifuge tubes for lipid extraction. The fattyacids were extracted with solvent mixture containing a 2:1 ratio ofchloroform to methanol. The upper layer was removed and driedover nitrogen gas. Lipid content (% dry wt basis) was determinedgravimetrically. For FAMEs, 1 mL toluene was added to the tubefollowed by the addition of 200 �L of internal standard, methylnonadecanoate (C19:0) and 200 �L of butylated hydroxytoluene(BHT). Acidic methanol (2 mL) was also added to the tube and keptfor overnight incubation at 50 ◦C. Fatty acid methyl esters (FAMEs)were extracted into hexane. The hexane layer was removed anddried over sodium sulphate. FAMEs were concentrated using nitro-gen gas. The samples were analysed by a GC-FID system (AgilentTechnologies, 6890N, US). The GC was equipped with a capil-lary column (Supelcowax 10, 30 m × 0.25 mm, 0.25 �m thickness).Helium was used as the carrier gas at a flow rate of 1.5 mL min−1.The injector was maintained at 250 ◦C and a sample volume of1 �L was injected. Fatty acids peaks were identified on compari-son of retention time data with external standards (Sigma–Aldrich,Australia). Peaks were quantified with Chemstation chromatogra-phy software (Agilent Technologies, US). Results are presented asmean ± SD of duplicates repeated twice.

2.6. Carotenoid extraction

Vortexing was applied for extraction of carotenoids from freeze-dried biomass. But later on, a slight modification of carotenoidextraction method from algae and fungi as reported in the liter-ature was followed [15,16]. To 25 mg of freeze-dried biomass, 1 mLof DMSO (preheated at 55 ◦C) was added and kept at 55 ◦C for 60 minundisturbed followed by centrifugation at 4000 rpm for 15 min at15 ◦C. Supernatant was taken and stored at 15 ◦C in dark. This cyclewas repeated 3–4 times until the biomass becomes colourless. Toextract carotenoids from DMSO solution, ether:water (1:1) solventsystem was used in 1:2 ratio (DMSO:solvent system). This solventmixture was centrifuged at 4000 rpm for 15 min at 15 ◦C and keptat −20 ◦C for 10 min. The upper un-freeze solvent layer was trans-ferred in fresh centrifuge tube while bottom freeze DMSO layerwas discarded. The upper layer was washed twice with water toremove traces of DMSO. Solvent was evaporated under nitrogenstream and equivalent volume of acetone was added and storedat −20 ◦C. Final extract volume should be noted down for totalcarotenoid estimation. Carotenoids were identified and quantifiedby RP-HPLC analysis.

2.7. HPLC analysis of carotenoids

The method for carotenoid analysis was adopted from Armentaand co-workers [16]. The Agilent 1200 Series HPLC system withthe Agilent 1200 Series photodiode array detector was used forcarotenoid analysis. Carotenoids were analysed at 477 nm using5 �m Luna C18 reversed-phase column, 4.6 mm × 250 mm (Phen-omenex, USA), and a Security guard column C18, 3.0 mm × 4.0 mm(Phenomenex, USA). This column was equilibrated with mobilephase A consisting of methanol, ethyl acetate and water (88:10:2,

v/v/v) in a gradient mode at a flow rate 0.75 mL min−1. This flow ratewas maintained 10 min. Mobile phase composition was changedto 2:50:48 (mobile phase B) between 10 and 30 min and the flowrate was adjusted to 1.5 mL min−1. This stage was maintained for a

A. Gupta et al. / Biochemical Engineering Journal 78 (2013) 11– 17 13

Fig. 1. (a) Biomass growth of Thraustochytrium sp. AMCQS5-5 as a function oftA

fAabifua

3

3b

wewgdiais[rii

Table 1aProportion of fatty acids by Thraustochytrium sp. AMCQS5-5 at variable glucoseconcentrations.

Glucose concentration (%) SFAs MUFA PUFAs

0.5 42.8 ± 0.4 2.4 ± 0.1 54.3 ± 1.21 38.1 ± 0.2 1.8 ± 0.2 59.5 ± 0.22 37.8 ± 0.1 2.2 ± 0.1 59.5 ± 0.44 35.2 ± 0.3 2.5 ± 0.1 61.6 ± 0.36 46.8 ± 0.3 2.3 ± 0.1 50.5 ± 0.68 39.1 ± 0.2 2.01 ± 0.1 58.1 ± 0.4

10 40.8 ± 0.1 0.8 ± 0.01 57.7 ± 0.3

SFA, saturated fatty acids; MUFAs, monounsaturated fatty acids; PUFAs, polyunsat-

ime. (b) Effect of initial glucose and glycerol concentration on biomass growth ofMCQS5-5.

urther 5 min and then set for column washing with mobile phase for 10 min. Calibration curves for each standard carotenoid suchs astaxanthin, zeaxanthin, canthaxanthin, res/meso-astaxanthin,eta cryptoxanthin, echinenone and �-carotene were prepared for

dentification and estimation of each carotenoid present in dif-erent extracts. 1 mL acetone extract was taken and evaporatednder nitrogen gas followed by addition of 1 mL mobile phase And syringe filtered (�m) before for HPLC analysis of the extracts.

. Results and discussion

.1. Effect of different glucose and glycerol concentrations oniomass growth

To achieve higher biomass concentration, the marine isolatesere grown in a fermentation medium containing glucose or glyc-

rol for 168 h. Highest optical density 12.3 and 10.1 was observedhen growing Thraustochytrium sp. AMCQS5-5 with glycerol and

lucose, respectively (Fig. 1a). When grown under the optimal con-itions the stationary phase was reached. The growth rate of the

solate was studied with the highest specific growth rate measuredt 0.023 h−1 at the end of incubation period of 5 days. Howevern other studies, thraustochytrid-like strains and Schizochytriump. have shown growth rate of 0.16 h−1 and 0.38 h−1, respectively

24,25]. The use of glucose as the carbon source in the mediumesulted in minor improvements in biomass growth at differentnitial concentrations, ranging from 1.0 to 1.44 g L−1, with the max-mum observed value at 2% of initial glucose concentration (Fig. 1b).

urated fatty acids. Results are presented as mean ± SD of duplicates.

However, the effect of different glycerol concentrations resultedin biomass growth ranging from 0.7 to 8.32 g L−1. The highest celldry weight was found to be 8.32 g L−1 at 4% of glycerol concentra-tion where the C:N ratio was 2 and lowest biomass growth wasobserved at 10% glycerol concentration with 0.7 g L−1, showing agradual decrease with increasing concentration after 4% of glycerolconcentration (Fig. 1b). This may be due to the substrate inhibitionat higher glycerol concentration (Fig. 1b). This is in accordance withother reported studies on a commercial thraustochytrid strain, S.limacinum SR21, in which higher concentrations of glycerol werefound to retard biomass growth [20,21,26]. These reports showedthe use of crude glycerol, which has demonstrated itself to be aviable alternative carbon source to glucose [26]. Moreover, crudeglycerol has been utilised for the production of value added prod-ucts such as 1, 3 propanediol, citric acid, polyhydroxyalkanoatesand lipids including omega-3 fatty acids, and also as animal feed[27]. Thus, glycerol may be recognised as more appropriate carbonsource for the cultivation of thraustochytrids.

3.2. Effect of different C:N ratios on biomass growth with glycerol

It was observed that at a C:N ratio of 2 with equal proportionsof yeast extract and peptone, the thraustochytrid isolate exhibitedits highest biomass growth. When the yeast extract and peptoneproportions were changed, there was a shift in the biomass pro-duction maximum from a C:N ratio of 2 to 4 at 11.84 g L−1. Thismay be attributed to the metabolic characteristics of the thraus-tochytrid isolates, which differ among the strains. The highestbiomass growth was reported with yeast extract as the nitrogensource [8]. The cell dry weight obtained (11.84 g L−1) in this studyat 4% of pure glycerol concentration was comparable to the growthof S. limacinum SR21 in the medium supplemented with 9% of pureglycerol (14.4 g L−1) [20]. The effect of biodiesel-derived glycerolon the biomass growth has been studied using other oleaginousspecies such as Rhodotorula glutinis, Pythium irregulare, Chlorellaprotothecoides [28–30]. Furthermore, the use of crude glycerol inthe fermentation medium with S. limacinum SR21 yielded goodbiomass growth both in batch culture mode (∼18 g L−1) and fedbatch mode (37.9 g L−1) [20,31].

3.3. Effect of glycerol concentrations on fatty acid composition

It has been reported that nutrient sources in the medium (glu-cose and glycerol) affected the fatty acid composition in otheroleaginous microorganisms such as Isochrysis sp. and Rhodosporid-ium sp. [32,33]. In this study, the effect of different glucoseconcentrations on the fatty acid composition varied. PUFA and SFA

proportions of total fatty acids (TFA) were found to be highest(61.6% and 46.8%) at 4% and 6% of glucose concentration respec-tively (Table 1a). No improvement was observed in the DHA profilewith different concentrations of glucose as the carbon source,

14 A. Gupta et al. / Biochemical Engineering Journal 78 (2013) 11– 17

Table 1bEffect of variable glucose concentrations on fatty acid profile of Thraustochytrium sp. AMCQS5-5.

Glucose concentration (%, w/v) % TFA

C16:0 C18:0 C18:1n9 C20:4n6 C20:5n3 C22:5n6 C22:6n3

0.5 29.8 ± 0.13 11.3 ± 0.17 0.9 ± 0.02 3.3 ± 0.06 7.4 ± 0.15 8.1 ± 0.41 31.6 ± 0.451 27.6 ± 0.05 9.03 ± 0.05 0.87 ± 0.04 3.8 ± 0.01 8.1 ± 0.02 8.8 ± 0.06 35.1 ± 0.042 27.6 ± 0.01 8.9 ± 0.04 0.86 ± 0.03 3.41 ± 0.06 7.9 ± 0.08 8.7 ± 0.15 35.8 ± 0.044 26.4 ± 0.13 7.1 ± 0.06 0.80 ± 0.05 3.52 ± 0.02 8.5 ± 0.03 9.3 ± 0.19 36.3 ± 0.016 32.2 ± 0.18 13.2 ± 0.1 0.63 ± 0.05 2.7 ± 0.1 6.4 ± 0.02 8.7 ± 0.26 29.6 ± 0.068 28.2 ± 0.05 9.4 ± 0.05 0.69 ± 0.01 2.9 ± 0.03 7.7 ± 0.04 8.8 ± 0.25 34.5 ± 0.06

10 29.9 ± 0.07 9.6 ± 0.03 – 3.3 ± 0.01 7.1 ± 0.07 8.7 ± 0.1 35.7 ± 0.07

C16:0, palmitic acid; C18:0, stearic acid; C18:1n9, oleic acid; C20:4n6, arachidonic acid; C2as mean ± SD of duplicates.

Table 2aProportion of fatty acid by Thraustochytrium sp. AMCQS5-5 at different initial glyc-erol concentrations. Fig. 2b Effect of variable glycerol concentrations on fatty acidprofile of Thraustochytrium sp. AMCQS5-5.

Glycerol concentration (%) SFAs MUFA PUFAs

0.5 33.1 ± 0.3 3.8 ± 0.1 62.9 ± 0.71 35.2 ± 0.04 3.2 ± 0.04 60.4 ± 0322 45.4 ± 0.1 6.8 ± 0.1 47.2 ± 0.14 48.8 ± 0.09 3.2 ± 0.04 47.6 ± 0.146 43.5 ± 0.1 7.4 ± 0.1 48.6 ± 0.28 38.1 ± 0.04 8.3 ± 0.04 53.3 ± 0.07

10 39.2 ± 0.2 13.5 ± 0.2 47.2 ± 0.5

Ss

patPaicco3ipanDtittiat[

ment with the findings of Scott and co-workers [37], who reported

TE

Ca

FAs, saturated fatty acids; MUFAs, monounsaturated fatty acids; PUFAs, polyun-aturated fatty acids. Results are presented as mean ± SD of duplicates.

ossibly due to the inability of the isolate to consume glucose as nutrient source (Table 1b). However, as the glycerol concentra-ions increased, the fatty acid profile revealed a decrease in theUFAs proportion from 62.9% to 47.2% of TFA, showing highestccumulation at 0.5% of glycerol concentration and demonstrat-ng the inefficiency of the isolate in translating the higher glyceroloncentrations into more PUFA. Moreover, higher glycerol con-entrations may lead to higher viscosity, which is not a desirableutcome [26]. However, an increment in saturated fatty acids from3.1% to 48.8% of TFA up to 4% of glycerol concentration and declin-

ng at higher concentrations was observed (Table 2a). This mayresent an advantage in biodiesel production when using glycerols feedstock. Palmitic acid (40.8% of TFA) was a dominant compo-ent in the saturated fatty acids at 4% of glycerol concentration andHA (37.02% of TFA) was prominent at 0.5% of glycerol concentra-

ion (Table 2b). The decline in the DHA profile was also observed inncreases of different pure and biodiesel-derived glycerol concen-rations with S. limacinum SR21 [22,26], which may be attributedo insufficient oxygen supply due to viscosity issues with increas-ng glycerol content. Further, palmitic acid was found to increase as

proportion of TFA relative to DHA at higher glycerol concentra-ions, which corresponds to the findings of Pyle and co-workers22] but contrasts recent findings where DHA proportions were

able 2bffect of variable glycerol concentrations on fatty acid profile of Thraustochytrium sp. AM

Glycerol concentration (%, w/v) % TFA

C16:0 C18:0 C18:1n9

0.5 25.21 ± 0.3 6.59 ± 0.03 2.11 ± 0.1 28.88 ± 0.01 4.24 ± 0.01 1.63 ± 0.2 38.21 ± 0.03 4.61 ± 0.01 4.86 ± 0.4 40.78 ± 0.05 4.25 ± 0.01 1.69 ± 0.6 35.76 ± 0.01 5.23 ± 0.02 5.32 ± 0.8 28.6 ± 0.01 6.02 ± 0.01 6.04 ± 0.

10 27.9 ± 0.1 8.72 ± 0.03 11.55 ± 0.

16:0, palmitic acid; C18:0, stearic acid; C18:1n9, oleic acid; C20:4n6, arachidonic acid; C2s mean ± SD of duplicates.

0:5n3, EPA; C22:5n6, DPA, C22:6n3, DHA; TFA, total fatty acid. Results are presented

found to be higher than palmitic acid [26]. However, EPA proportionwas slightly higher in the current study, which may be attributedto the type of thraustochytrid strain. Other fatty acids recordedwere stearic acid (4.3–8.7% of TFA), oleic acid (1.7–11.6% of TFA),arachidonic acid (3.6–7.3% of TFA), EPA (2.3–7.7% of TFA) and DPA(7.2–10.9% of TFA). The decrease of the DHA proportion in the TFAmay be due to the substrate inhibition with the increase in car-bon concentration which resulted in non-conversion of the carbonsource in to lipids [20]. Higher C:N ratio has been demonstrated toenhance the lipid accumulation in thraustochytrids [2]. The totalfatty acids (TFA) production was found to be higher when thestrain was subjected to utilise glycerol versus glucose. The TFA wasobserved ranging from 48.1 to 67.0 mg L−1 with highest productionat 4% glucose whereas, on using 4% glycerol in the medium, it wasfound to be 1112.4 ± 0.02 mg L−1 (1.12 g L−1) which clearly exhib-ited efficient conversion of glycerol into lipids. DHA was found tobe 24.3 ± 0.1 mg L−1 and 305.70 ± 0.01 mg L−1 at 4% of glucose andglycerol concentration, respectively. This was in accordance withother studies where similar DHA production was achieved using 8%glucose instead of glycerol [34]. On using 3% glucose in fermenta-tion medium, 120 mg L−1 DHA was produced by ThraustochytriumATCC 34304 [35]. Zhou et al. [36] showed Thraustochytriidae sp.Z105 was able to produce 2.41 g L−1 of total lipids in 4% of glucosewith DHA production of 0.78 g L−1.

3.4. Effect of carbon source on biomass and carotenoid production

During the initial phase of experiment Schizochytrium S31,Thraustochytrium AMCQS5-3, Thraustochytrium AMCQS5-5 andSchizochytrium AMCQS1-9 were selected for biomass productionand carotenoid extraction. Biomass production (data not shown),carotenoid production (Fig. 2) and lipid content (data not shown)were found to be significantly higher in all cultures, when fed withglycerol instead of glucose carbon source. These results are in agree-

an increase in biomass and lipid content of thraustochytrids whencultures were fed with glycerol and fish oil derived raw glycerolinstead of glucose. Use of glycerol as the carbon source, resulted

CQS5-5.

C20:4n6 C20:5n3 C22:5n6 C22:6n3

04 4.91 ± 0.02 7.72 ± 0.04 9.78 ± 0.1 37.02 ± 0.301 5.65 ± 0.01 6.15 ± 0.03 10.91 ± 0.10 35.17 ± 0.101 4.54 ± 0.01 2.70 ± 0.03 7.25 ± 0.01 28.39 ± 0.0601 3.61 ± 0.01 3.60 ± 0.02 9.95 ± 0.06 27.49 ± 0.0103 5.29 ± 0.01 2.37 ± 0.01 7.38 ± 0.1 28.83 ± 0.0201 6.92 ± 0.01 2.97 ± 0.01 7.94 ± 0.01 29.13 ± 0.0205 7.33 ± 0.01 2.31 ± 0.02 7.24 ± 0.1 21.65 ± 0.1

0:5n3, EPA; C22:5n6, DPA, C22:6n3, DHA; TFA, total fatty acid. Results are presented

A. Gupta et al. / Biochemical Engineering Journal 78 (2013) 11– 17 15

0

10

20

30

40

50

60

70

ATCC 20888-S31

AMCQS5-3 AMCQS5-5 AMCQS1-9

Car

oten

oid

cont

ent (

µg/g

)

Thraustochytrid isolates

Glucose

Glycerol

Fo

it9MTcsrbCtcbcbtmewgiwttaao

3

ubbrwmfowwcvmbaD[

0

10

20

30

40

50

60

70

80

90

ATCC 20888-S31

AMCQS5-3 AM CQS5-5 AM CQS1-9

Car

oten

oid

cont

ent (

µg/g

) Beta-Caro tene

Echinenone

Canthaxanthin

Astaxanthin

ig. 2. Carotenoid production in selected thraustochytrid isolates fed with glucoser glycerol.

nto increase in carotenoid content in Schizochytrium S31 from 16.8o 20.28 �g g−1 whereas in the case of Schizochytrium AMCQS1-

it enhanced carotenoid content from 5.23 to 10.16 �g g−1.aximum improvement in carotenoid content was observed in

hraustochytrium AMCQS5-3, Thraustochytrium AMCQS5-5, wherearotenoid content increased from 16–17 to 50–60 �g g−1, repre-enting a 3 fold increase (Fig. 2). Armenta and co-workers [16]eported carotenoid content up to 50 �g g−1 using glucose as car-on source in the fermentation of thraustochytrid strain ONC-T18.hatdumrong et al. [38] generated a mutant of S. limacinum BR2.1.2hat produced carotenoid about 8–13 �g L−1 while using glucose asarbon source. Glycerol is one of the cost-effective sources of car-on and its usage will be helpful in bringing down the productionost of the nutraceuticals such as DHA and carotenoids. Increase iniomass, lipid and carotenoid production under glycerol fermenta-ion indicates efficient regulation of enzymes involved in glycerol

etabolism prior to entry into the glycolytic pathway [37], enablingnhanced supply of acetyl-CoA or NADPH for lipid biosynthesis,hich are the driving force for oleaginicity in oleaginous microor-

anisms [39]. However, to the best of our knowledge, this is the firstnstance where an increase in carotenoid content was documented

hen glycerol was used instead of glucose as the carbon source forhraustochytrid cultures. TAG accumulation and carotenoid biosyn-hesis are inter-connected, since both share common precursors;cetyl-CoA and NADPH. The role of enhanced synthesis of trisporiccid (a carotenogenesis activator under glycerol) cannot be ruledut for enhanced carotenoid production [40].

.5. Carotenoid profiling of thruastochytrid strains

Existing literature suggested that vortexing should be repeatedntil carotenoid content is either completely extracted or pelletecomes colourless [41,42]. In practice, however, the supernatantecomes colourless after 3–4 extraction cycles, while the pelletetain some colour. This suggests inefficient carotenoid extractionith vortexing. DMSO mediated extraction, a well-documentedethod, was therefore applied for carotenoid extraction from

reeze-dried biomass. Significant improvement (10–350%) wasbserved in carotenoid extraction in all thraustochytrid biomasshen DMSO was used as the extraction solvent instead of vortexingith acetone (Figs. 2 and 3). High dipolar moments and dielectric

onstants of DMSO are the driving force behind its powerful sol-ation capacity, enabling it to efficiently extract carotenoids fromicrobial biomass. However, the extraction efficiency of DMSO is

roadly affected by the type of biomass used for extraction as wells total carotenoid content in the cell [43]. Researchers have usedMSO for carotenoid extraction and quantification extensively

44]. Toxicity issues associated with DMSO mediated carotenoid

Thraustochytrid isolates

Fig. 3. Carotenoid profiling of selected thraustochytrid isolates using HPLC.

extraction are however a major concern that must be addressed.A safe and efficient process must be devised and optimised for fullrecovery of carotenoids from these isolates.

All orange coloured thraustochytrid isolates have astaxanthinor canthaxanthin as dominant carotenoids, with some quan-tities of echinenone. No other significant carotenoids such aslutein, zeaxanthin or beta-cryptoxanthin were observed in thesethraustochytrids (Fig. 3). ATCC strain Schizochytrium S31 has com-parable astaxanthin and echinenone quantities of 29.7 �g g−1

and 33.21 �g g−1, respectively, along with some canthaxanthin(14.63 �g g−1). �-Carotene was not observed in the carotenoidprofile, reflecting the efficient conversion of �-carotene into xan-thophylls in the late phase of this culture. However, the absenceof �-carotene and the presence of significant amounts of astaxan-thin and echinenone reflect the rapid oxidation of �-carotene intoechinenone followed by its oxidation/hydroxylation into astax-anthin via the canthaxanthin route [45]. Thraustochytrid strainsAMCQS5-3 and AMCQS5-5 showed similar growth patterns andcarotenoid profiles. No astaxanthin was observed in the carotenoidprofile of AMCQS5-3 and AMCQS5-5, although canthaxanthin waspresent in quantities of 38.13 �g g−1 and 29.31 �g g−1, respectively,along with echinenone at 20.13 �g g−1 and 19.61 �g g−1, respec-tively and �-carotene at 10.24 �g g−1 and 8.06 �g g−1, respectively.These results are in agreement with the findings of Armentaand co-workers [16], who reported similar carotenoid profiles inthraustochytrid ONC-T18, showing canthaxanthin, echinenone and�-carotene with small quantities of astaxanthin. Based on the infor-mation available in literature and findings of this study, it can beassumed that canthaxanthin remains the major carotenoid presentin Thraustochytrium [13,16] whereas astaxanthin in Schizochytrium[17,38]. The absence of astaxanthin from the carotenoid pro-file of these thraustochytrid strains indicates their inability toadd hydroxyl groups to the canthaxanthin backbone. �-Carotenehydroxylase enzyme is responsible for the addition of hydroxylgroups to canthaxanthin backbone and its functional inactivationmay lead to impaired biosynthesis of astaxanthin [46] and to theaccumulation of canthaxanthin. The absence of other hydroxylatedcarotenoid intermediates such as beta-cryptoxanthin, zeaxanthinand 4-ketoexathanin supports this assumption.

4. Conclusion

The experimental data demonstrated the efficient utilisation ofglycerol as alternative carbon source to glucose in the fermentation

medium of newly isolated thraustochytrid strains from Queenscliff,Victoria, Australia. The C:N ratio appears to be a crucial parame-ter in biomass growth along with TFA production. It appears thatglucose was not fully utilised in the fermentation medium by the

1 ginee

ipweoeosdAtopa

A

CvDaA

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

6 A. Gupta et al. / Biochemical En

solate, resulting in poor biomass growth and low lipid and DHAroductivity. However, glycerol showed promising biomass growthith significant lipid and DHA productivity. Moreover, raw glyc-

rol derived from various sources, such as the biodiesel and fishil industries, can be used as a carbon source instead of pure glyc-rol to offset production costs and enhance commercial viabilityf the process. Fermentation strategies for enhancing cell growth,ubsequent the lipid productivity and fatty acid profile must beefined. Based on carotenoid content, thraustochytrid AMCQS5-3,MCQS5-5 along with ATCC strain S31 appeared suitable as poten-

ial isolates for carotenoid production, having significant amountf xanthophylls. The significant amount of DHA and carotenoidsresent in these isolates presents an opportunity to exploit thems potential bio-refinery systems.

cknowledgements

The authors are thankful to the Strategic Research Centre forhemistry and Biotechnology, Deakin University, Australia for pro-iding funds to support bioprocessing research. One of the authors.S. thanks the DIRI program of Deakin University for providing

scholarship to pursue this research work. Thanks to Dr. Jacquidcock for her help in GC and HPLC data anaylsis.

eferences

[1] P.K. Bajpai, P. Bajpai, O.P. Ward, Optimization of production of docosahexaenoicacid (DHA) by Thraustochytrium aureum ATCC 34304, J. Am. Oil Chem. Soc. 68(1991) 509–514.

[2] A.M. Burja, H. Radianingtyas, A. Windust, C.J. Barrow, Isolation and charac-terization of polyunsaturated fatty acid producing Thraustochytrium species:screening of strains and optimization of omega-3 production, Appl. Microbiol.Biotechnol. 72 (2006) 1161–1169.

[3] A. Beopoulos, J. Cescut, R. Haddouche, J.-L. Uribelarrea, C. Molina-Jouve, J.-M.Nicaud, Yarrowia lipolytica as a model for bio-oil production, Prog. Lipid Res. 48(2009) 375–387.

[4] A. Gupta, J. Vongsvivut, C.J. Barrow, M. Puri, Molecular identification of marineyeast and its spectroscopic analysis establishes unsaturated fatty acid accumu-lation, J. Biosci. Bioeng. 114 (2012) 411–417.

[5] Y.-H. Chen, T.H. Walker, Fed-batch fermentation and supercritical fluid extrac-tion of heterotrophic microalgal Chlorella protothecoides lipids, Bioresour.Technol. 114 (2012) 512–517.

[6] C. Shene, A. Leyton, Y. Esparza, L. Flores, B. Quilodrán, I. Hinzpeter, M. Rubilar,Microbial oils and fatty acids: effect of carbon source on docosahexaenoic acid(c22:6 n-3, DHA) production by thraustochytrid strains, J. Soil Sci. Plant Nutr.10 (2010) 207–216.

[7] S. Abad, X. Turon, Valorization of biodiesel derived glycerol as a carbon sourceto obtain added-value metabolites: focus on polyunsaturated fatty acids, Bio-technol. Adv. 30 (2012) 733–741.

[8] T. Yokochi, D. Honda, T. Higashihara, T. Nakahara, Optimization of docosa-hexaenoic acid production by Schizochytrium limacinum SR21, Appl. Microbiol.Biotechnol. 49 (1998) 72–76.

[9] L. Bongiorni, A. Pusceddu, R. Danovaro, Enzymatic activities of epiphyticand benthic thraustochytrids involved in organic matter degradation, Aquat.Microb. Ecol. 41 (2005) 299–305.

10] W.-K. Hong, C. Kim, D. Rairakhwada, S. Kim, B.-K. Hur, A. Kondo, J.-W. Seo,Growth of the oleaginous microalga Aurantiochytrium sp. KRS101 on cellulosicbiomass and the production of lipids containing high levels of docosahexaenoicacid, Bioprocess Biosyst. Eng. 35 (2012) 129–133.

11] A. Gupta, C.J. Barrow, M. Puri, Omega-3 biotechnology: thraustochytrids as anovel source of omega-3 oils, Biotechnol. Adv. 30 (2012) 1733–1745.

12] M.B. Johnson, Z. Wen, Production of biodiesel fuel from the microalgaSchizochytrium limacinum by direct transesterification of algal biomass, EnergyFuels 23 (2009) 5179–5183.

13] T. Aki, K. Hachida, M. Yoshinaga, Y. Katai, T. Yamasaki, S. Kawamoto, T. Kakizono,T. Maoka, S. Shigeta, O. Suzuki, K. Ono, Thraustochytrid as a potential source ofcarotenoids, J. Am. Oil Chem. Soc. 80 (2003) 789–794.

14] E.J. Johnson, The role of carotenoids in human health, Nutr. Clin. Care 5 (2002)56–65.

15] S. Raghukumar, Thraustochytrid marine protists: production of PUFAs andother emerging technologies, Mar. Biotechnol. 10 (2008) 631–640.

16] R.E. Armenta, A. Burja, H. Radianingtyas, C.J. Barrow, Critical assessment of

various techniques for the extraction of carotenoids and co-enzyme Q10from the thraustochytrid strain ONC-T18, J. Agric. Food Chem. 54 (2006)9752–9758.

17] T. Yamasaki, T. Aki, M. Shinozaki, M. Taguchi, S. Kawamoto, K. Ono, Utiliza-tion of Shochu distillery wastewater for production of polyunsaturated fatty

[

ring Journal 78 (2013) 11– 17

acids and xanthophylls using thraustochytrid, J. Biosci. Bioeng. 102 (2006)323–327.

18] L. Zhu, X. Zhang, X. Ren, Q. Zhu, Effects of culture conditions on growth anddocosahexaenoic acid production from Schizochytrium limacinum, J. OceanUniv. China 7 (2008) 83–88.

19] B. Quilodran, I. Hinzpeter, A. Quiroz, C. Shene, Evaluation of liquid residuesfrom beer and potato processing for the production of docosahexaenoic acid(C22:6n-3, DHA) by native thraustochytrid strains, World J. Microbiol. Bio-technol. 25 (2009) 2121–2128.

20] Z. Chi, D. Pyle, W. Zhiyou, C. Frear, S. Chen, A laboratory study of produc-ing docosahexaenoic acid from biodiesel-waste by microalgae fermentation,Process Biochem. 42 (2007) 1537–1545.

21] S. Ethier, K. Woisard, D. Vaughan, Z. Wen, Continuous culture of the microalgaeSchizochytrium limacinum on biodiesel-derived crude glycerol for producingdocosahexaenoic acid, Bioresour. Technol. 102 (2011) 88–93.

22] D.J. Pyle, R.A. Garcia, Z. Wen, Producing docosahexaenoic acid (DHA)-richalgae from biodiesel-derived crude glycerol: effects of impurities on DHAproduction and algal biomass composition, J. Agric. Food Chem. 56 (2008)3933–3939.

23] Y. Liang, N. Sarkany, Y. Cui, J. Yesuf, J. Trushenki, J.W. Blackburn, Use of sweetsorghum juice for lipid production by Schizochytrium limacinum SR21, Biore-sour. Technol. 101 (2010) 3623–3627.

24] Z. Perveen, H. Ando, A. Ueno, Y. Ito, Y. Yamamoto, Y. Yamada, T. Takagi, T.Kaneko, K. Kogame, H. Okuyama, Isolation and characterization of a novelthraustochytrid-like microorganism that efficiently produces docosahexaenoicacid, Biotechnol. Lett. 28 (2006) 197–202.

25] K. Min, H. Lee, P. Anbu, B. Chaulagain, B. Hur, The effects of culture con-dition on the growth property and docosahexaenoic acid production fromThraustochytrium aureum ATCC 34304, Korean J. Chem. Eng. 29 (2012)1211–1215.

26] T.Y. Huang, W.C. Lu, I.M. Chu, A fermentation strategy for producing docosa-hexaenoic acid in Aurantiochytrium limacinum SR21 and increasing C22:6proportions in total fatty acid, Bioresour. Technol. 123 (2012) 8–14.

27] F. Yang, M. Hanna, R. Sun, Value-added uses for crude glycerol—a byproduct ofbiodiesel production, Biotechnol. Biofuels 5 (2012) 13.

28] Y.-H. Chen, T. Walker, Biomass and lipid production of heterotrophicmicroalgae Chlorella protothecoides by using biodiesel-derived crude glycerol,Biotechnol. Lett. 33 (2011) 1973–1983.

29] S.K. Athalye, R.A. Garcia, Z. Wen, Use of biodiesel-derived crude glycerol for pro-ducing eicosapentaenoic acid (EPA) by the fungus Pythium irregulare, J. Agric.Food Chem. 57 (2009) 2739–2744.

30] C. Saenge, B. Cheirsilp, T.T. Suksaroge, T. Bourtoom, Potential use of oleagi-nous red yeast Rhodotorula glutinis for the bioconversion of crude glycerolfrom biodiesel plant to lipids and carotenoids, Process Biochem. 46 (2011)210–218.

31] Z. Chi, Y. Liu, C. Frear, S. Chen, Study of a two-stage growth of DHA-producingmarine algae Schizochytrium limacinum SR21 with shifting dissolved oxygenlevel, Appl. Microbiol. Biotechnol. 81 (2009) 1141–1148.

32] Y.-H. Lin, F.-L. Chang, C.-Y. Tsao, J.-Y. Leu, Influence of growth phase and nutrientsource on fatty acid composition of Isochrysis galbana CCMP 1324 in a batchphotoreactor, Biochem. Eng. J. 37 (2007) 166–176.

33] J. Xu, X. Zhao, W. Wang, W. Du, D. Liu, Microbial conversion of biodiesel byprod-uct glycerol to triacylglycerols by oleaginous yeast Rhodosporidium toruloidesand the individual effect of some impurities on lipid production, Biochem. Eng.J. 65 (2012) 30–36.

34] A. Taha, T. Kimoto, T. Kanada, H. Okuyama, Growth optimization of thraus-tochytrid strain 12B for the commercial production of docosahexaenoic acid,Food Sci. Biotechnol. 22 (2013) 53–58.

35] Y. Taoka, N. Nagano, Y. Okita, H. Izumida, S. Sugimoto, M. Hayashi, Effectof Tween 80 on the growth, lipid accumulation and fatty acid composi-tion of Thraustochytrium aureum ATCC 34304, J. Biosci. Bioeng. 111 (2011)420–424.

36] P.P. Zhou, M.B. Lu, W. Li, L.J. Yu, Microbial production of docosahexaenoic acidby a low temperature-adaptive strain Thraustochytriidae sp. Z105: screeningand optimization, J. Basic Microbiol. 50 (2010) 380–387.

37] S.D. Scott, R.E. Armenta, K.T. Berryman, A.W. Norman, Use of raw glycerol toproduce oil rich in polyunsaturated fatty acids by a thraustochytrid, EnzymeMicrob. Technol. 48 (2011) 267–272.

38] W. Chatdumrong, W. Yongmanitchai, S. Limtong, W. Worawattanamateekul,Optimization of docosahexaenoic acid (DHA) production and improve-ment of astaxanthin content in a mutant Schizochytrium limacinum isolatedfrom mangrove forest in Thailand, Kasetsart J. Nat. Sci. 41 (2007)324–334.

39] C. Ratledge, Fatty acid biosynthesis in microorganisms being used for singlecell oil production, Biochimie 86 (2004) 807–815.

40] F. Mantzouridou, E. Naziri, M.Z. Tsimidou, Industrial glycerol as a supplemen-tary carbon source in the production of �-carotene by Blakeslea trispora, J. Agric.Food Chem. 56 (2008) 2668–2675.

41] B. Quilodrán, I. Hinzpeter, E. Hormazabal, A. Quiroz, C. Shene, Docosahexaenoicacid (C22:6n-3, DHA) and astaxanthin production by Thraustochytriidae sp.AS4-A1 a native strain with high similitude to Ulkenia sp.: evaluation of liq-

uid residues from food industry as nutrient sources, Enzyme Microb. Technol.47 (2010) 24–30.

42] M.L. Carmona, T. Naganuma, Y. Yamaoka, Identification by HPLC–MS ofcarotenoids of the Thraustochytrium CHN-1 strain isolated from the Seto Inlandsea, Biosci. Biotechnol. Biochem. 67 (2003) 884–888.

ginee

[

[

A. Gupta et al. / Biochemical En

43] R. Sarada, R. Vidhyavathi, D. Usha, G.A. Ravishankar, An efficient method forextraction of astaxanthin from green alga Haematococcus pluvialis, J. Agric. FoodChem. 54 (2006) 7585–7588.

44] J.J. Sedmak, D. Weerasinghe, S. Jolly, Extraction and quantitation of astaxanthinfrom Phaffia rhodozyma, Biotechnol. Tech. 4 (1990) 107–112.

[

[

ring Journal 78 (2013) 11– 17 17

45] L. Gouveia, V. Veloso, A. Reis, H. Fernandes, J. Novais, J. Empis, Evolution of pig-ment composition in Chlorella vulgaris, Bioresour. Technol. 57 (1996) 157–163.

46] M.A. Scaife, A.M. Burja, P.C. Wright, Characterization of cyanobacterial�-carotene ketolase and hydroxylase genes in Escherichia coli, and their appli-cation for astaxanthin biosynthesis, Biotechnol. Bioeng. 103 (2009) 944–955.