Accepted Manuscript

Production of biodiesel from carbon sources of macroalgae, Laminaria japonica

Xu Xu, Ji Young Kim, Yu Ri Oh, Jong Moon Park

PII: S0960-8524(14)00980-8DOI: http://dx.doi.org/10.1016/j.biortech.2014.07.015Reference: BITE 13660

To appear in: Bioresource Technology

Received Date: 5 May 2014Revised Date: 2 July 2014Accepted Date: 3 July 2014

Please cite this article as: Xu, X., Kim, J.Y., Oh, Y.R., Park, J.M., Production of biodiesel from carbon sources ofmacroalgae, Laminaria japonica, Bioresource Technology (2014), doi: http://dx.doi.org/10.1016/j.biortech.2014.07.015

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customerswe are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, andreview of the resulting proof before it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

1

Production of biodiesel from carbon sources of macroalgae, Laminaria 2

japonica 3

Xu Xu a, Ji Young Kim b, Yu Ri Oh b, Jong Moon Park a,b,c * 4

5

6

aSchool of Environmental Science and Engineering , 7

bDepartment of Chemical Engineering, 8

cDivision of Advanced Nuclear Engineering, Pohang University of Science and Technology, 9

San 31, Hyoja-dong, Pohang 790-784, South Korea 10

11

12

13

14

15

16

17

18

19

* Corresponding author. 20

Tel: +82-54-279-2275; Fax: +82-54-279-8299; E-mail: jmpark@postech.ac.kr (J.M. Park) 21

22

23

2

Abstract 24

As aquatic biomass which is called “the third generation biomass”, Laminaria japonica (also 25

known as Saccharina japonica) consists of mannitol and alginate which are the main 26

polysaccharides of algal carbohydrates. In this study, oleaginous yeast (Cryptococcus 27

curvatus) was used to produce lipid from carbon sources derived from Laminaria japonica. 28

Volatile fatty acids (VFAs) were produced by fermentation of alginate extracted from L. 29

japonica. Thereafter, mannitol was mixed with VFAs to culture the oleaginous yeast. The 30

highest lipid content was 48.30 %. The composition of the fatty acids was similar to 31

vegetable oils. This is the first confirmation of the feasibility of using macroalgae as a carbon 32

source for biodiesel production. 33 34

35

Keywords: Laminaria japonica, Cryptococcus curvatus, Manntiol, Alginate, Volatile fatty 36

acids (VFAs), Fatty acid methyl esters (FAMEs), Biodiesel, Macroalgae 37

38

39

40

41

42

43

44

45

46

47

48

49

3

1. Introduction 50

Energy problem has become one of the most serious issues all over the world since energy 51

consumption is inevitable for human existence. At this point, a large amount of alternative 52

energy appears. Among them, microbial lipids can be used as a renewable alternative 53

(biodiesel) to traditional fossil diesel fuel. Biodiesel, a mixture of fatty acid alkyl esters, can 54

be obtained from various renewable lipid resources such as vegetable oils, fats and wastes of 55

cooking oils (Liu et al., 2008). However, use of edible oils for biodiesel production competes 56

with food production and requires land and irrigation water; these requirements and the high 57

lignin content of terrestrial plants reduce the economic viability of this production mode. 58

Especially in the countries like South Korea which has no vast territory to cultivate a large 59

amount of high oil content plants, it is meaningless to use the edible oils for biodiesel 60

production. 61

On the basis of above mentioned issues, macroalgae which is called as “the 3rd generation 62

biomass” is gaining increasingly more attentions as alternative renewable sources of inputs 63

for biofuel production since it can deal with these drawbacks of terrestrial biomass and 64

produce sustainable bioenergy and materials. Macroalgae (seaweed) have several 65

advantageous characteristics such as high yield, low energy cost of production, low cost, low 66

contaminant levels and low nutrient requirements, which can be confirmed from the results 67

shown in Table 1 (Park, 2012). Macroalgae do not need land and freshwater for cultivation 68

which is advantageous to the countries lack of land. In addition, macroalgae have a lower 69

cost of the production of food and energy than other energy crops like corn and wheat. 70

Because seaweed markets are mainly existed in a few East Asian countries where seaweed is 71

utilized as food, hydrocolloids, fertilizer and animal feed (Jung et al., 2013). Macroalgae can 72

convert solar energy into chemical energy with higher photosynthetic efficiency (6-8%) than 73

terrestrial biomass (1.8-2.2%) (Jung et al., 2013). Moreover, the process of macroalgae 74

4

production introduces no contaminants, such as soil which can subsequently lead to operation 75

problems (McKendry, 2002). Comparing to the terrestrial plants, no utilization of chemical 76

fertilizers and pesticides also reduce the contaminants produced during the biomass 77

cultivation. Therefore, macroalgae have high potential as the feedstock of energy production, 78

and may be the best choice as a feedstock for biodiesel production. 79

Seaweed grows very quickly and can be harvested more than four crops per year compared 80

to the crop and forest-derived biomass which are harvested less than two crops and one crop 81

per year respectively (Table 1). Globally, about 7 million tons (wet weight) of macroalgae is 82

harvested in 2008, and the most harvested species are Laminaria japonica (sea tangle), 83

Undaria pinnatifiida (sea mustard), and Porphyra tenera (sea weed laver) (FAO, 2011). 84

According to the “Food and Agricultural Organization stats”, the highest production of 85

cultured seaweed throughout the world in 2008 was the brown seaweed, especially L. 86

japonica which was 4.8 million tons per year. Among them, L. japonica and U. pinnatifiida 87

were brown algae and L. japonica was accounted for about 65% of global production. In the 88

case of South Korea, the two most cultured species occupying around 70 % of total 89

aquaculture production are L. japonica and U. pinnatifiida. In our previous research, the 90

composition of L. japonica had been analyzed. As the data shown in Table 2, the amount of 91

carbohydrate of L. japonica is higher than that of U. pinnatifiida, which contains 60 – 67 % 92

(W/W dry). However, U. pinnatifiida has more protein (Cho, 1995). 93

In Table 3, it shows carbohydrate profile in brown algae. The two species of brown algae 94

mainly consists of alginate, laminaran, fucoidan (Kim et al., 1995, Pyeun et al., 1977). 95

Alginate and mannitol are the main structure and storage compound which account for about 96

50% (W/W) of total carbohydrates respectively (Kloareg and Quatrano, 1988). 97

Mannitol, a sugar alcohol equivalent to mannose, is one of the major carbon sources of 98

brown algae. It is different from terrestrial biomass storing carbon source as cellulose, 99

5

hemicellulose and lignin. Since cellulose is very difficult to be hydrolyzed and lignin cannot 100

be used in biorefinery, it is required high costs of pretreatment for these compounds, which is 101

the obstacle for commercialization of energy production from the terrestrial biomass. 102

However, as mannitol is soluble and available carbohydrate, it can be directly utilized by 103

microorganisms as carbon source. Alginate, the most abundant polysaccharide in L. japonica, 104

is one of brown algae’s cell wall compounds with cellulose, fucoidan, and protein (Kloareg et 105

al., 1986). The cells are strengthened to the algal tissue by both mechanical strength and 106

flexibility of alginate (Andresen et al., 1977). In the L. japonica, there are two kinds of uronic 107

acids linking each other which are polymannuronate and polyguluronate. The polymers 108

accumulate in blocks and bind divalent metal ions through forming gels which are structural 109

parts of algae. Therefore, it is less available to contact with alginate lyase. Since it is a 110

polysaccharide, the microbial oil production using alginate would require extra 111

saccharification or anaerobic fermentation process (Wang et al., 2013). Alginate has been 112

used as a food additive for a long time. Its uses are based on the properties of thickening, 113

gelling, film formation, stabilizing and general colloidal properties (Aliste et al., 2000). These 114

properties are useful in pie, cake, ice cream, and canned food productions, which can reduce 115

moisture retention, thicken batter and extend the shelf life. 116

It is well known that the first step of anaerobic digestion is hydrolysis which breaks down 117

biopolymers to monomers. In this step, carbohydrates, proteins and lipids are respectively 118

transformed into sugars, amino acids and fatty acids. These small molecules are fermented to 119

carbon dioxide and hydrogen along with several organic acids in the next step called 120

acidogenesis (Angenent et al., 2008). The organic acids produced during the acidogenesis can 121

be used by some microorganisms as carbon sources. Therefore, based on the processes 122

introduced above, L. japonica was chosen as the most appropriate substrate in this study 123

which has more carbohydrates than U. pinnatifida. 124

6

As one of the most important renewable energy sources, biodiesel is produced through 125

transesterification of vegetable oils or animal fats with short chain alcohols. However, high 126

cost of raw materials for biodiesel production has become one of the major obstacles for wide 127

application. Recently, it is gaining more attention to search for new oil sources for biodiesel 128

production. Among them, microbial oils, also known as single cell oils (SCOs), can 129

overcome these problems. In fact, all microorganisms can synthesize lipids, but only those 130

which can produce more than 20% lipids of their dry biomass are called “oleaginous” 131

including bacteria, fungi, and yeasts (Li et al., 2006, Meng et al., 2009). The oleaginous 132

yeasts possess higher specific growth rate than molds and algae, which can be considered as 133

favorable microorganisms for microbial lipids production (Li et al., 2007). The most 134

productive oleaginous yeasts are Yarrowia lipolytica, Cryptococcus curvatus, and 135

Rhodosporidium toruloides. They could produce from 40 % to 70 % of biomass (Ratledge 136

and Cohen, 2008). The microbial lipids mainly consist of triacylglycerols (TAGs), and other 137

components are free fatty acids, other neutral lipids, sterols and polar fractions (Fakas et al., 138

2006). The TAGs produced by oleaginous yeasts are mostly formed of C14 to C18 fatty acids 139

which can be used as biofuel because they are similar to vegetable oils (Li et al., 2007, Meng 140

et al., 2009). 141

C. curvatus, an oleaginous yeast, is able to accumulate up to 60% oils by dry cell weight 142

(DCW) (Ratledge, 1991) using economical carbon source . In this study, we assess whether C. 143

curvatus can be used to produce significant quantities of lipids from mannitol and alginate 144

extracted from L. japonica. 145

Because mannitol is water-soluble whereas alginate is not, a stepwise process was used 146

(Fig. 1): first, alginate was fermented under anaerobic condition to obtain VFAs which can be 147

used by C. curvatus, then the VFAs were mixed with mannitol and then added to the lipid-148

producing fermentor. It was the first time to produce lipid by oleaginous yeast only using the 149

7

L. japonica as the sole carbon source without adding any other nutrients. 150

151

2. Materials and Methods 152

2.1. Feedstock 153

Dried kelp, Laminaria japocia, was purchased from Geumil-do Fisheries Cooperation 154

Association (Wando, South Korea), milled to fine powder, and kept in a sealed container at 155

room temperature until it was used. For the trials, 90 g of the powder was mixed in 3 L tap 156

water to make the concentration 30 g/L and stirred for 6 h at room temperature to maximize 157

mannitol recovery rate in the supernatant (based on results of preliminary work), then left 158

unstirred to allow solids to precipitate. After 24 h, the mixture had separated into two layers. 159

The supernatant (~2 L) was mannitol solution; the bottom layer (~1 L) was mainly alginate 160

suspension. The supernatant was aspirated off and steam autoclaved at 121 °C for 20 min. 161

The alginate suspension was used for VFA production. 162

In the case of feedstock for lipid production, the effluent from the alginate fermentation 163

was recovered, centrifuged for 10 min at 400 g-force, and then passed through a 1.20-µm 164

filter (GF/C, Whatman, UK) to separate the liquid and solid components. The liquid part was 165

then steam autoclaved at 121 °C for 20 min. 166

167

2.2. Inoculum 168

Anaerobic digester sludge from a wastewater treatment plant (Daegu, Korea), screened 169

through a sieve (No. 10, diameter: 2 mm), then pretreated by heat at 90 °C for 20 min to 170

inactive methanogens (Lee et al., 2008). First, populations of VFA-producing bacteria were 171

increased and adapted to conditions in a 7 L fermentor (Biotron LiFlus GR Fernmentor) 172

under anaerobic condition. Heat-treated sludge (1.5 L) and 2.5 L of ground kelp (4 L working 173

volume) were added to the fermentor. It was operated at 35 °C and stirring rate was 200 ppm. 174

8

After enrichment and adaptation, the 4 L content of the fermentor was poured into a 7 L 175

continuous stirred tank reactor (CSTR) (Biotron LiFlus GR Fernmentor). Ground kelp mixed 176

with tap water (30 g/L) was supplied to the reactor under anaerobic condition using peristaltic 177

pumps (model No. 7523-30, 7521-57, Masterflex®, USA). The continuous process was 178

operated at 4 d hydraulic retention time with 2.78 mL/min flow rate. After inoculation, the 179

fermentor was purged using N2 gas for 10 min, and all instruments were maintained in strictly 180

anaerobic condition. The pH was controlled at 7 by an automatic pH controller (Model KB-181

250, K&B) using 5 M NaOH (Samchun) and 1 M HCl (Samchun) solution because this pH 182

yielded the highest VFA production rate in a preliminary study. Temperature and stirring rate 183

were maintained at 35 °C and 200 rpm respectively. Ultimately, the effluent from the CSTR 184

was directly used as the inoculum for alginate fermentation. 185

C. curvatus (ATCC 20509) was obtained from the Korean Collection for Type Culture and 186

pre-cultured in a medium composed of 1 g/L peptone, 1g/L yeast extract, and 10 g/L 187

dextrose. C. curvatus was grown in 250 mL Erlenmeyer flasks containing 50 mL of medium 188

and incubated at 28 °C in a rotary shaker which was set to 150 rpm. Before inoculation, seed 189

cells were sub-cultured for 24 h in a medium that contained 10 g/L mannitol, 1 g/L peptone, 190

and 1g/L yeast extract. The initial pH was 7.2 and the inoculum rate was 10% (v/v). 191

192

2.3. Culture conditions 193

Alginate fermentation and lipid production were conducted sequentially (Fig. 1). All 194

instruments and reactors for lipid production were steam autoclaved before use. All batch 195

tests were conducted in triplicate. 196

For VFA production from alginate, a 1.5-L fermentor (Biotron LiFlus GR) containing 1 L 197

alginate feedstock (Section 2.1.1) was seeded with 10 % (v/v) inoculum derived from the 198

CSTR effluents, then purged using N2 gas for 10 min. pH was controlled at 7 by an automatic 199

9

pH controller (Model KB-250, K&B). Temperature was maintained at 35 °C and the 200

fermentor was stirred continuously at 400 rpm. 201

For lipid production by C. curvatus, a 5-L fermentor (Biotron LiFlus GR) was utilized 202

(Fig. 1). It was loaded with 2 L of mannitol supernatant and 1 L VFAs. The pH was 203

controlled at 5.5 by an automatic pH controller (Model KB-250, K&B). Lipid production was 204

aerobic, so air was supplied at 0.5 VVM using an aerator. Temperature was maintained at 205

30 °C and the fermentor was stirred continuously at 400 rpm. 206

207

2.4. Analytical method 208

2.4.1. Physico-chemical analytical method 209

Total solid (TS), volatile solid (VS), total suspended solid (TSS), volatile suspended solid 210

(VSS) and total nitrogen (TN) were measured according to the standard method (APHA-211

AWWA-WEF, 1998). Every soluble sample was passed through a 0.45-µm pore 212

polyethersulfone syringe filter (Millipore, USA) to remove insoluble particles. 213

Mannitol, VFAs and ethanol concentrations were analyzed using a high performance liquid 214

chromatography (HPLC, Agilent Technology 1100 series) equipped with a carbohydrate-215

analysis column (Aminex HPX-87H, BIORAD INC., USA), refractive index detector (RID) 216

and diode array detector (DAD). The eluent was 0.004-M H2SO4 and the flow rate was 217

0.6 mL/min. The column temperature was maintained at 50 °C. All liquid samples were 218

diluted (1:5 or 1:10) with water and passed through a 0.2 µm syringe filter (Millipore, USA). 219

220

2.4.2. Determination of dry cell weight 221

For the dry cell weight (DCW) of C. curvatus was measured by forcing a known volume of 222

cell suspension sample through a pre-dried, pre-weighed 0.45-µm nitrocellulose filter 223

(Millipore, USA) using a vacuum pump. The samples were dried to constant weight at 224

10

110 °C in an oven, then weighed. 225

226

2.4.3. Lipid extraction 227

The extraction of total cellular lipid was performed according to Bligh and Dyer (Bligh and 228

Dyer, 1959) with modifications (Bourque and Titorenko, 2009). Lipid was extracted from 229

lyophilized biomass using a mixture of chloroform and methanol (2:1 v/v). The mixture was 230

centrifuged to partition the lipids into the solvent, then the solvent was evaporated using a 231

Nitrogen Evaporation System (Organomation Associates Inc., USA). Lipid content was 232

represented as a percentage of dry cell weight (%, W/W). Lipid concentration was defined as 233

the amount of extracted cellular lipid in per liter working volume (g/L). 234

235

2.4.4. Fatty acid methyl esters 236

The fatty acid compositions of the lipid produced by C. curvatus were determined by 237

analysis of fatty acid methyl esters (FAMEs). The FAMEs were produced by 238

transesterification: methanol was added to the extracted lipid with sulfuric acid (2.5 % V/V 239

H2SO4/CH3OH) as a catalyst; the reaction was allowed to proceed for 45 min at 90 °C. Then 240

1 ml H2O and 2 ml n-hexane were added. The FAMEs dissolved into the n-hexane. The 241

solution was centrifuged at 2000 rpm for 15 min, to separate the water from the 242

FAME/hexane phase which was then transferred into glass vials using Pasteur pipettes. 243

The FAMEs in n-hexane were analyzed using a gas chromatography (6890N, Agilent, 244

USA) equipped with a flame ionized detector (FID) and an INNOWAX capillary column 245

(Agilent, USA, 30 m × 0.32 mm × 0.5 µm). The column temperature was programmed as 246

follows: (1) initial column temperature 100 °C, hold for 5 min, (2) increase to 250 °C at 247

10 °C/min, hold for 30 min. The split ratio was 1:10 (v/v). The injector and the detector 248

temperatures were both set at 250 °C. FAME components were identified and quantified by 249

11

comparing the retention times and peak areas with those of standard solutions. 250

251

3. Results and discussion 252

3.1. VFA production by alginate fermentation using mixed cultivation system 253

As mentioned in the materials and methods, the inoculum of alginate fermentation was the 254

effluent from a CSTR. In fact, the CSTR was a VFA-producing system which can 255

continuously produce acetate, propionate and butyrate using L. japonica as sole carbon 256

source without adding any other nutrients. This system has been operated for over 1000 d in 257

our previous research and stabilized more than 600 d. Therefore, the microbial community in 258

the effluent from the CSTR was quite stable. 259

To evaluate the profile changes of acids during alginate fermentation, mannitol 260

consumption along with the accumulation and consumption of acidogenic products in batch 261

reactor were demonstrated (Fig. 2). Although the mannitol and alginate were separated by 262

natural sedimentation, the liquid (mannitol solution) could not be completely separated from 263

the solid (alginate suspension). Therefore, liquid samples for HPLC analysis after 264

centrifugation retained some mannitol (5.8 g/L initially). The products of alginate 265

fermentation were acetate, succinate, lactate, formate, acetate, propionate, butyrate and 266

ethanol. Ethanol concentration reached 1.0 g/L within 24 h, but gradually decreased to 0 by 267

144 h, probably because ethanol initially produced was converted into acetate by 268

acidogenesis (Angenent et al., 2008). The depletion of ethanol and production of acetate are 269

thermodynamically favorable (Oh et al., 2003). Likewise, succinate, lactate and formate also 270

increased to a certain amount and disappeared later. During the period of acidogenesis, 271

acetate was produced as a major alginate fermentation product, followed by propionate and 272

butyrate. The final concentration of acetate was more than 9.0 g/L at 6.5 d. Propionate 273

concentration slowly increased to 3.5 g/L, and remained at that level until 156 h. Butyrate 274

12

was produced until 24 h and its concentration began to increase from 36 h, ultimately to a 275

concentration which was more than 0.5 g/L. At the end of digestion, the VFAs consisted of 276

69.2 % acetate, 25.7 % propionate and 5.1 % butyrate. 277

278

3.2. Cell growth and lipid accumulation by C.curvatus using carbon source derived from 279

macroalgae by single batch cultivation system. 280

Most of the microorganisms can synthesize lipids, but only the oleaginous strains may 281

accumulate more than 20 % W/W lipids on dry cell basis (Papanikolaou and Aggelis, 2011a). 282

Microorganisms synthesize lipids by either de novo processes that utilizes hydrophilic 283

substances as substrates, or by ex novo processes that utilize hydrophobic substances as 284

substrates. In this research, all lipid synthesis occurred by de novo processes. In de novo 285

processes, lipid accumulation always happens after depletion of nitrogen (or to a lesser extent 286

of other essential nutrient like phosphorus or sulfate) from the medium (Papanikolaou and 287

Aggelis, 2011b). In addition, to ensure that VFAs are mostly converted into lipids rather than 288

to biomass, these carbon sources should be added to a medium containing enough cells 289

(Fontanille et al., 2012). 290

Total nitrogen (TN) was 134 mg/L in the mannitol supernatant and 48 mg/L in the VFA 291

effluent. Initially 1 L of VFA effluent was added to the 2 L of mannitol substrate; this 292

addition diluted manntiol and VFA concentration in the substrate to 2/3 and 1/3 of pre-293

addition levels, respectively. During the first 12 h, VFAs were not dramatically consumed by 294

C. curvatus. Lipid concentration and DCW increased slightly at the same time. However, 295

lipid content increased rapidly from less than10 % to 30 %. During the next 36 h, lipid 296

content, lipid concentration, and DCW increased to a much higher level until 48 h when all 297

VFAs were used up by yeasts. As for lipid content, it achieved its highest point (48.30%). It 298

is obvious that the concentration of mannitol was almost constant until 48 h, which could 299

13

demonstrate that in lipid-producing metabolic pathway of C. curvatus, VFAs were more 300

preferable than mannitol as a carbon sources under pH of 5.5. After the exhaustion of VFAs, 301

mannitol began to gradually decrease and was consumed up within 84 h. In the meantime, 302

DCW increased to 3.60 g/L until 72 h and then remained stable; whereas lipid content 303

suddenly reduced which resulted in the decrease of lipid concentration. This phenomenon 304

which is called as “lipid turnover” is related with storage lipid degradation (Chen et al., 2012, 305

Peng et al., 2013). Microbial lipid turnover has been extensively studied in Y. lipolytic and C. 306

echinulata (Papanikolaou and Aggelis, 2011a). Oleaginous yeasts consume their own 307

intracellular lipids to maintain lipid-free biomass when carbon source is exhausted or carbon 308

uptake rate decreases. Because there was still some mannitol left in the substrate, the lipid 309

turnover could result from the reduced carbon source uptake rate. When VFAs were 310

exhausted, C. curvatus had to consume mannitol as substitution, which could make the 311

carbon source uptake rate decreased during this process. From 72 h, lipid content increased 312

again with more mannitol assimilated by the yeasts. 313

Several research groups have used VFAs to produce lipids by culturing oleaginous yeasts 314

including Yarrowia lipolytica and Cryptococcus curvatus (Christophe et al., 2012, Fontanille 315

et al., 2012). The most significant difference was that in our study the microorganisms 316

utilized mannitol and VFAs which were all derived from seaweed without addition of any 317

other nutrients and reagents, whereas in other research, oleaginous yeasts used a synthetic 318

medium containing glucose and acetate as substrates. Moreover, in other research the ratio of 319

carbon to nitrogen (C/N) was artificially controlled by continuously adding carbon and 320

nitrogen sources to achieve high lipid concentration and content. But in our work, we just 321

utilized the original mannitol supernatant and alginate fermentation effluents to culture the 322

oleaginous yeasts. 323

We compared our results to other researches that also used VFAs or acetate as carbon 324

14

sources (Table 4). In fact, little research has examined the feasibility of using mannitol as a 325

carbon source for oleaginous yeasts. In addition, the strains and concentration of carbon 326

source used were all different with each research. Therefore, the only relevant comparisons 327

was of the value of lipid content (% W/W). Among these results, apparently the lipid content 328

in this work (48.30%) was higher than that of any other researches. It indicated that the 329

mannitol and VFAs derived from alginate anaerobic fermentation were favorable for 330

microbial lipid production even without using synthetic medium or artificially adding any 331

other nutrients. 332

333

3.3. Fatty acid composition analysis 334

The fatty acids produced in our study consisted mainly of palmitic acid (C16:0), stearic 335

acid (C18:0), oleic acid (C18:1), and linoleic acid (C18:2) (Table 5). Others indicates C8:0, 336

C10:0, C12:0, C14:0, C16:1, C18:3, C20:0, C20:1, C20:2, and C24:0. This composition is 337

similar to those of vegetable oils which have been utilized for industrial production of 338

biodiesel (Adamska ET, 2004). Over time, some changes in lipid composition occurred. 339

Obviously, there were three different trends shown in each fatty acid composition. As for 340

C18:0 and C18:1, they both increased during the period of VFAs consumption until 48 h, 341

while the percentages decreased when C. curvatus started to assimilate mannitol as the sole 342

carbon source. In contrast, in the case of palmitic acid, it was gradually reduced during the 343

first 48 h, then accumulated again until the end. However, the profile of linoleic acid 344

followed a totally different pattern. The percentage increased throughout the period from 345

8.95 % to 17.15 %, irrespective of which carbon source yeasts used. This composition was 346

comparable to vegetable fatty acids indicating that the fatty acids accumulated by C. curvatus 347

are appropriate for use in biodiesel production (Li et al., 2007, Meng et al., 2009). 348

349

15

4. Conclusions 350

We designed an efficient batch system for lipid production by the oleaginous yeast 351

Cryptococcus curvatus using two different carbon sources derived from a brown macroalga, 352

Laminaria japonica: mannitol and VFAs obtained by anaerobic fermentation of alginate with 353

a mixed culture. This is the first demonstration of microbial oil production by bioconversion 354

of macroalgae without addition of any other carbon sources or nutrients. This study has 355

shown that mannitol and VFAs derived from alginate fermentation are favorable feedstock 356

for lipid production. Future work can focus on optimizing and scaling up this process. 357

358

Acknowledgements 359

This research was supported by Basic Science Research Program through the National 360

Research Foundation of South Korea (NRF) funded by the Ministry of Education, Science 361

and Technology (Grant number 2011-0001108), the Advanced Biomass R&D Center (ABC) 362

of South Korea Grant funded by the Ministry of Education, Science and Technology (ABC-363

2011-0028387), Marine Biotechnology Program Funded by Ministry of Land, Transport and 364

Maritime Affairs of South Korean Government, South Korea and the Manpower 365

Development Program for Marine Energy funded by Ministry of Land, Transportation and 366

Maritime Affairs (MLTM) of South Korean government and by the World Class University 367

(WCU) program through the National Research Foundation of South Korea funded by the 368

Ministry of Education, Science and Technology (R31-30005). 369

370

371

16

Reference 372

[1]Adamska ET, C.T., Kaczmarek Z, Szala L. 2004. Multivariate approach to evaluating the 373 fatty acid composition of seed oil in a doubled haploid population of winter oilseed 374 rape (Brassica napus L.,). J. Appl. Genet 45, 6. 375

[2]Aliste, A.J., Vieira, F.F., Del Mastro, N.L., 2000. Radiation effects on agar, alginates and 376 carrageenan to be used as food additives. Radiat. Phys. Chem. 57, 305-308. 377

[3]Andresen, I.L., Skipnes, O., Smidsrød, O., Østgaard, K., Hemmer, P. 1977. Some 378 biological functions of matrix components in benthic algae in relation to their 379 chemistry and the composition of seawater. ACS Symp. Ser. ACS Publications. pp. 380 361-381. 381

[4]Angenent, L.T., Wrenn, B.A., Wall, J., Harwood, C., Demain, A. 2008. Optimizing mixed-382 culture bioprocessing to convert wastes into bionergy. Bioenergy. ASM Press. pp. 383 179-194. 384

[5]APHA-AWWA-WEF. 1998. Standard Methods for the Examination of Water and 385 Wastewater, 20th ed., American Public Health Association. Washington, D.C. 386

[6]Bligh, E., Dyer, W.J., 1959. A rapid method of total lipid extraction and purification. Can. 387 J. Biochem. Physiol. 37, 911-917. 388

[7]Bourque, S.D., Titorenko, V.I., 2009. A quantitative assessment of the yeast lipidome using 389 electrospray ionization mass spectrometry. Journal of visualized experiments: JoVE. 390

[8]Chen, X.F., Huang, C., Xiong, L., Chen, X.D., Chen, Y., Ma, L.L., 2012. Oil production on 391 wastewaters after butanol fermentation by oleaginous yeast Trichosporon 392 coremiiforme. Bioresour. Technol. 118, 594-7. 393

[9]Cho, D.M., Kim, D.S., Lee, D.S., Kim, H.R., Pyeun, J.H., 1995. Trace Components and 394 Functional Saccharides in Seaweed. 1: Changes in Proximate Composition and Trace 395 Elements According to the Harvest Season and Places. Korean J. Fish. Aquat. Sci. 28, 396 49-59. 397

[10]Christophe, G., Deo, J.L., Kumar, V., Nouaille, R., Fontanille, P., Larroche, C., 2012. 398 Production of oils from acetic acid by the oleaginous yeast Cryptococcus curvatus. 399 Appl. Biochem. Biotechnol. 167, 1270-9. 400

[11]Fakas, S., Papanikolaou, S., Galiotou-Panayotou, M., Komaitis, M., Aggelis, G., 2006. 401 Lipids of Cunninghamella echinulata with emphasis to gamma-linolenic acid 402 distribution among lipid classes. Appl. Microbiol. Biotechnol. 73, 676-83. 403

[12]FAO. 2011. 2008 Fishery and Aquaculture Statistics. Available from: 404 <ftp://ftp.fao.org/FI/CDrom/CD_yearbook_2008/index.htm>. 405

[13]Fei, Q., Chang, H.N., Shang, L., Choi, J.D.R., Kim, N., Kang, J., 2011a. The effect of 406 volatile fatty acids as a sole carbon source on lipid accumulation by Cryptococcus 407 albidus for biodiesel production. Bioresour. Technol. 102, 2695-2701. 408

[14]Fei, Q., Chang, H.N., Shang, L.A., Choi, J.D.R., 2011b. Exploring low-cost carbon 409 sources for microbial lipids production by fed-batch cultivation of Cryptococcus 410 albidus. Biotechnol. Bioproc. E 16, 482-487. 411

[15]Fontanille, P., Kumar, V., Christophe, G., Nouaille, R., Larroche, C., 2012. Bioconversion 412 of volatile fatty acids into lipids by the oleaginous yeast Yarrowia lipolytica. 413 Bioresour. Technol. 114, 443-9. 414

[16]Jung, K.A., Lim, S.R., Kim, Y., Park, J.M., 2013. Potentials of macroalgae as feedstocks 415 for biorefinery. Bioresour. Technol. 135, 182-90. 416

[17]Kim, D.S., Cho, D.M., Lee, D.S., Kim, H.R., Pyeun, J.H., 1995. Trace components and 417 functional saccharides in marine algae. J. Korean Fish. Soc. 28, 270-278. 418

[18]Kloareg, B., Demarty, M., Mabeau, S., 1986. Polyanionic characteristics of purified 419

17

sulphated homofucans from brown algae. Int. J. Biol. Macromol. 8, 380-386. 420 [19]Kloareg, B., Quatrano, R., 1988. Structure of the cell walls of marine algae and 421

ecophysiological functions of the matrix polysaccharides. Oceanogr. Mar. Biol. Annu. 422 Rev. 26, 259-315. 423

[20]Lee, M., Hidaka, T., Tsuno, H., 2008. Effect of temperature on performance and 424 microbial diversity in hyperthermophilic digester system fed with kitchen garbage. 425 Bioresour. Technol. 99, 6852-60. 426

[21]Li, Y., Zhao, Z., Bai, F., 2007. High-density cultivation of oleaginous yeast 427 Rhodosporidium toruloides Y4 in fed-batch culture. Enzyme Microb. Technol. 41, 428 312-317. 429

[22]Li, Y.H., Liu, B., Zhao, Z.B., Bai, F.W., 2006. Optimization of Culture Conditions for 430 Lipid Production by Rhodosporidium toruloides. Chin. J. Biotechnol. 22, 650-656. 431

[23]Liu, Z.Y., Wang, G.C., Zhou, B.C., 2008. Effect of iron on growth and lipid accumulation 432 in Chlorella vulgaris. Bioresour. Technol. 99, 4717-22. 433

[24]McKendry, P., 2002. Energy production from biomass (part 1): overview of biomass. 434 Bioresour. Technol. 83, 37-46. 435

[25]Meng, X., Yang, J., Xu, X., Zhang, L., Nie, Q., Xian, M., 2009. Biodiesel production 436 from oleaginous microorganisms. Renew. Energ 34, 1-5. 437

[26]Oh, S.E., Van Ginkel, S., Logan, B.E., 2003. The relative effectiveness of pH control and 438 heat treatment for enhancing biohydrogen gas production. Environ. Sci. Technol. 37, 439 5186-5190. 440

[27]Papanikolaou, S., Aggelis, G., 2011a. Lipids of oleaginous yeasts. Part I: Biochemistry of 441 single cell oil production. Eur. J. Lipid Sci. Technol. 113, 1031-1051. 442

[28]Papanikolaou, S., Aggelis, G., 2011b. Lipids of oleaginous yeasts. Part II: Technology 443 and potential applications. Eur. J. Lipid Sci. Technol. 113, 1052-1073. 444

[29]Park, J.M. 2012. PETROFED (Petroleum Federation of India). Trends and Prospects of 445 Biorefinery Research in Korea. Available from: <http://www.petrofed.org/19-446 20_April_12_prog.asp>. 447

[30]Peng, W.F., Huang, C., Chen, X.F., Xiong, L., Chen, X.D., Chen, Y., Ma, L.L., 2013. 448 Microbial conversion of wastewater from butanol fermentation to microbial oil by 449 oleaginous yeast Trichosporon dermatis. Renew. Energ 55, 31-34. 450

[31]Pyeun, J.H., Park, Y.H., Lee, K.H., 1977. Factors involved in the quality retention of 451 cultured Undaria pinnatifida. Bull. Korean Fish. Soc 10, 125-135. 452

[32]Ratledge, C., 1991. Microorganisms for lipids. Acta Biotechnol. 11, 429-438. 453 [33]Ratledge, C., Cohen, Z., 2008. Microbial and algal oils: Do they have a future for 454

biodiesel or as commodity oils. Lipid Technol. 20, 155-160. 455 [34]Wang, J., Kim, Y.M., Rhee, H.S., Lee, M.W., Park, J.M., 2013. Bioethanol production 456

from mannitol by a newly isolated bacterium, Enterobacter sp. JMP3. Bioresour. 457 Technol. 135, 199-206. 458

459

460 461

18

Figure legends: 462 463

Fig. 1. Schematic diagram of alginate fermentation and lipid production processes. 464

465

Fig. 2. Profile of various carbon sources and volatile suspended solid (VSS) variations 466

during the alginate fermentation. (▲) VSS; (◇) mannitol; (○) acetic acid; (▼) 467

propionic acid; (□) butyric acid; (◆) ethanol; (ⅹ) formic acid; (▽) lactic acid; (△) 468

succinic acid. 469

470

Fig. 3. Profile of mannitol (◇), acetic acid (○), propionic acid (▼), butyric acid (□), 471

dry cell weight (DCW)(●), lipid concentration (▲), lipid content (■) during a two-step 472

batch cultivation of Cryptococcus curvatus in a 5 L fermentor. 473

19

Table 1. Comparisons between 1st, 2

nd, and 3

rd generation biomass

Resourcea Crop Forest-derived Seaweed

Harvesting 1 – 2 /yr 1 / 8yr 4 – 6 /yr

Production (ton/ha) 180 9 565

CO2 uptake (ton/ha) 5 – 10 4.6 36.7

Energy yield (%) 30 - 35 20 – 25 > 45

Cost ($/L) 0.2 – 0.3 0.4 0.2

a (Park, 2012)

20

Table 2. Composition of Laminaria japonica and Undaria pinnadifida

(% w/w, dry base) Protein Lipid Ash Carbohydrate

Laminaria japonica 6.8 – 10.3 7.2 – 11.5 13.8 – 21.1 60.9 – 67.0

Undaria pinnadifidaa 13.7 – 28.7 3.6 – 6.2 29.4 – 46.5 26.5 – 42.8

a (Cho, 1995)

21

Table 3. Characteristic algal polysaccharides in Laminaria japonica and Undaria pinnadifida

Algal carbohydrate (% w/w, dry base)

Alginate Fucoidan Laminaran Mannitol Total

Laminaria japonica 14.6 – 29.5 0 – 0.1 3.7 – 4.2 22.5 – 33.2 40.8 – 67.0

Undaria pinnadifidaa 22.3 – 32.9 1.4 3.6 4.1 – 7.3 31.4 – 45.2

a (Kim et al., 1995, Pyeun et al., 1977)

22

Table 4. Lipid production by oleaginous yeasts using different substrate in two-step system

Strains Carbon source (g/L) DCW (g/L) Lipid conc. (g/L) Lipid content (% W/W) Reference

C. albidus VFAs (2.00) 1.16 0.31 27.00 (Fei et al., 2011a)

C. albidus Acetate (6.00) 2.85 0.74 25.80 (Fei et al., 2011b)

Y. lipolytica Acetate (12.00) 5.98 1.84 30.76 (Fontanille et al., 2012)

C. curvatus Mannitol + VFAs (9.30) 3.60 1.30 48.30 This study

a Calculated based on references.

23

Table 5. Fatty acid methyl ester (FAME) profile of C. curvatus

Fermentation time (h)

Fatty acid methyl ester (FAME) profile of C. curvatus (%)

Palmitic AME (C16:0)

Stearic AME (C18:0)

Oleic AME (C18:1)

Linoleic AME (C18:2)

Othersa

0 25.88 12.83 48.16 8.95 4.18

12 25.62 11.64 48.36 10.65 3.73

24 21.16 15.45 49.38 11.41 2.60

36 14.72 22.41 50.34 11.49 1.05

48 14.70 19.46 52.26 12.78 0.80

60 15.33 18.20 52.13 13.29 1.06

72 16.75 15.76 49.31 16.32 1.86

84 18.18 14.03 48.71 17.15 1.93

a Others: C8:0, C10:0, C12:0, C14:0, C16:1, C18:3, C20:0, C20:1, C20:2, and C24:0.

24

Fig. 1. Schematic diagram of alginate fermentation and lipid production processes.

Mannitol Supernatant

Alginate Slurry

Lipid Fermentor

AeratorFlow Meter pH Controller

VFAs

Alginate Fermentor

pH Controller

Substrate

(30 g/L Laminaria japonica)

25

Fig. 2. Profile of various carbon sources and volatile suspended solid (VSS) variations during alginate fermentation. (▲) VSS; (◇)

mannitol; (○) acetic acid; (▼) propionic acid; (□) butyric acid; (◆) ethanol; (ⅹ) formic acid; (▽) lactic acid; (△) succinic acid.

Time (h)

0 24 48 72 96 120 144

Co

nce

ntr

ati

on

(g

/L)

0

2

4

6

8

10

12

VS

S (

g/L

)

0

10

20

30

40

26

Fig. 3. Profile of mannitol (◇), acetic acid (○), propionic acid (▼), butyric acid (□), dry cell weight (DCW)(●), lipid concentration

(▲), lipid content (■) during a batch cultivation of Cryptococcus curvatus in a 5 L fermentor.

Time (h)

0 12 24 36 48 60 72 84

DC

W, M

an

nit

ol a

nd

VF

As (

g/L

)

0

2

4

6

8

10

Lip

id c

on

cn

. (g

/L)

0.0

0.5

1.0

1.5

2.0

Lip

id c

on

ten

t (%

)

0

10

20

30

40

50

27

H I G H L I G H T S

▶This is the first demonstration of microbial oil production by using macroalgae.

▶Oleaginous yeast Cryptococcus curvatus was used for microbial oil production.

▶Carbon sources were all derived from Laminaria japonica.

▶No addition of any other nutrients or synthetic medium was used.

▶The composition of the fatty acids was found similar to vegetable oils.