Accepted Manuscript

Lipid Extracted Algae as a Source for Protein and Reduced Sugar: A Step Closer
to the Biorefinery

Faiz Ahmad Ansari, Amritanshu Shriwastav, Sanjay Kumar Gupta, Ismail

PII: S0960-8524(14)01792-1
DOI: http://dx.doi.org/10.1016/j.biortech.2014.12.047
Reference: BITE 14369

To appear in: Bioresource Technology

Received Date: 22 October 2014

Please cite this article as: Ansari, F.A., Shriwastav, A., Gupta, S.K., Rawat, I., Guldhe, A., Bux, F., Lipid Extracted
Algae as a Source for Protein and Reduced Sugar: A Step Closer to the Biorefinery, Bioresource Technology (2014),
doi: http://dx.doi.org/10.1016/j.biortech.2014.12.047

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
 1 Lipid Extracted Algae as a Source for Protein and Reduced

2 Sugar: A Step Closer to the Biorefinery

4 Faiz Ahmad Ansari, Amritanshu Shriwastav, Sanjay Kumar Gupta, Ismail Rawat, Abhishek

5 Guldhe, and Faizal Bux*

7 Institute for Water and Wastewater Technology,

8 Durban University of Technology, P O Box1334, Durban,4000, South Africa.

10 *Corresponding author

11 Prof. Faizal Bux

12 Director

13 Institute for Water and Wastewater Technology

14 Durban University of Technology

15 P O Box1334, Durban,4000, South Africa.

16 Tel:+27 31 373 2346

17 Fax:+27 31 373 2777

18 Email: [email protected]

19

20

21

22

23

24

25
26 Abstract

27 The objective of this study was to investigate the feasibility of using lipid extracted algae

28 (LEA) as a source for protein and reduced sugar, and the effects of various procedural

29 treatments on their yields. LEA provided comparable yields of protein and reduced sugars to

30 those from total algae. Oven drying provided highest yields of all products followed by freeze

31 drying, while sun drying significantly lowered their yields. Effective cell disruption by

32 microwave and autoclave increased the lipid yields from algae, but resulted in increased loss

33 of other compounds with lipid extracting solvents lowering their yields during sequential

34 extraction. Relatively inefficient cell disruption by ultrasonication and osmotic shock lowered

35 the amount of cell protein lost to the lipid extracting solvents. These results highlight the

36 complexity of concurrent extraction of all value added products from algae, and the need for

37 proper selection of the processes to achieve the objectives of integrated biorefinery.

38

39 Key Words:

40 Lipid extracted algae; sample drying; cell disruption; value added products; integrated

41 biorefinery.
42 1. Introduction

43 Microalgae have drawn much attention in past few decades due to their nutrient removal

44 potential from wastewater (Di Termini et al., 2011; Sawayama et al., 1998). In addition,

45 microalgae are considered to be important resources for biofuels and a viable alternative to

46 limited fossil fuels due to their lipid accumulation potential (Morweiser et al., 2010;

47 Tsukahara & Sawayama, 2005). Recent efforts have focussed on achieving the dual

48 objectives of wastewater treatment and biofuels production from microalgae (Park et al.,

49 2011; Wu et al., 2012). However, algal biofuels are yet to achieve economical sustainability

50 (Lundquist et al., 2010).

51

52 In addition to lipids, microalgae also produce other compounds of great economic value

53 (Olguín, 2012). Algal proteins are an acceptable alternative of conventional food supplements

54 due to their nutritional value and amino acid profiles (Becker, 2007). Similarly, algal

55 polysaccharides can be hydrolyzed to reduced sugars which have great application in the

56 production of bioethanol (Fu et al., 2010; Sun & Cheng, 2002). The use of residual algal

57 biomass after lipid extraction for other applications can reduce the cost of algal cultivation

58 and biofuel production (Rashid et al., 2013). The concurrent extraction of other valuable

59 products in addition to lipids from algal biomass may result in optimal value extraction and

60 economically beneficial algal technology. This is an important objective of integrated algal

61 biorefinery approach for algal biofuels technology (Subhadra, 2010).

62

63 The application of residual algal biomass after lipid extraction has been investigated by

64 several researchers for different objectives. These objectives can broadly be divided in two

65 categories: first for energy production by utilizing the remaining carbon and hydrogen, and

66 secondly for extracting products for their nutritional and economical values. Zhu et al. (2013)
67 studied the hydrothermal liquefaction potential of lipid extracted algae (LEA) for their

68 conversion to liquid fuels. Similarly, Yang et al. (2011) developed a two-stage process to

69 produce hydrogen and methane gases from LEA. The potential of using LEA as a protein

70 source in animal feeds was also established (Ju et al., 2012; Lodge-Ivey et al., 2014). Geun

71 Goo et al. (2013) investigated the application of LEA for bioethanol production. The

72 extraction of valuable compounds from LEA was found to lower the cost of biofuel

73 production in comparison to the utilization of carbon and hydrogen in residual mass for

74 energy production (Gao et al., 2012).

75

76 A major consideration in the concurrent extraction of all valuable products from algae is the

77 effect of various treatments on the individual yields of such products. The algal biomass

78 undergoes many processes during lipid extraction which invariably affects these coproducts.

79 Theoretically, lipid extraction should not result in any loss of other cellular compounds

80 resulting in their fractions in algal mass to increase in the LEA in comparison to whole cell

81 algae, and thus increasing their yield (% w/w) in LEA. However, under realistic conditions,

82 some loss of these compounds is unavoidable as cell disruption would unbind them from cell

83 mass to some degree and the unbound fraction could be lost to the applied solvent during

84 lipid extraction. For example, Lam et al. (2014) found reduction in the carbohydrate content

85 by 7.2 % in the algal biomass after lipid extraction in comparison to whole cell algae. The

86 additional processes used for extracting individual products also have their process

87 limitations, for example hydrolysis of polysaccharides to reduced sugars requires sufficient

88 reaction time for achieving good yields (Fu et al., 2010). It is therefore important to

89 investigate the effects of various treatments during lipid extraction and additional processes

90 for individual product extraction on the final yields of these products. Comparison of yields

91 from whole cell algae in order to establish the feasibility of using LEA as a source of these
92 compounds is also beneficial. Studies investigating the effects of various treatments on

93 product yields from LEA are not available in the literature to the best of our knowledge. This

94 study, investigates the effects of various procedural treatments, i.e. drying and cell disruption,

95 during extraction of lipids, and subsequently protein and reduced sugars from LEA on the

96 product yields and comparison to yields from whole cell algae to identify the optimal

97 treatments and establish the feasibility of LEA as a source for these products. Cell disruption

98 by four methods: microwave, ultrasonication, autoclaving, and osmotic shock with 10 %

99 NaCl were investigated. In addition, main drying processes applied in literature include sun

100 drying, oven drying, freeze drying, drum drying, spray drying, fluidized bed drying etc.

101 However, not all of these are economically sustainable (Brennan & Owende, 2010; Guldhe et

102 al., 2014). Hence in this study, focus has been on investigating the effects of sun drying, oven

103 drying , and freeze drying processes, which are more economic in nature.

104

105 2. Material and methods

106 2.1 Algae culture

107 Scenedesmus obliquus (genbank accession number: FR751179.1) used in this study was

108 isolated from Durban region, KwaZulu Natal, South Africa (Misra et al., 2014). A raceway

109 pond of 300000 L was operated with BG11 medium for algal cultivation under natural sun

110 light (400 – 1200 µmol m-2 s-1) and temperature (18 – 27 °C). The grown biomass was

111 harvested by gravitational settling and then centrifuged to obtain thick algal slurry.

112

113 2.2 Drying of the harvested algae

114 The thickened biomass slurry was dried by three different methods: a) sun drying, b) oven

115 drying, and c) freeze drying. Sun dried biomass was obtained by placing the thickened slurry

116 on a drying bed lined with 1500 micron white plastic for three days under natural sunlight
117 and ambient conditions (400 – 1200 µmol m-2 s-1, 18 – 27 °C). Similarly algal slurry was

118 placed in a hot air oven at 60 °C for 24 hours for oven drying. A freeze dryer (Mini lyotrap,

119 LTE scientific Ltd., United Kingdom) was used to lyophilize the samples after overnight

120 freezing at -84 °C in a biofreezer (Glaciar NU9668E, Nuaire, Japan). Dried biomass was

121 pulverized with mortar and pestle and stored in a desiccator.

122

123 2.3 Lipid extraction assisted by cell disruption

124 Dried biomass of S. obliquus was further subjected to various cell disruption procedures to

125 increase the efficieny of solvent based lipid extraction. Four cell disruption methods were

126 investigated: a) microwave, b) ultrasonication, c) autoclave, and d) osmotic shock with 10 %

127 NaCl. Total lipids were extracted by the method of Folch et al. (1957) using 2:1 (v/v) mixture

128 of chloroform and methanol.

129

130 For microwave assisted lipid extraction, 1 g of dried biomass was mixed with 20 mL of

131 solvent mixture (chloroform and methanol in 2:1 ratio) and heated at 100 °C for 10 minutes

132 at 1000 W in a microwave digester (Milestone S.R.L., Italy; 1200 W of output power).

133 Solvent containing lipids was separated by centrifugation and then vacuum filtered. After

134 such separation from cell debris, solvent was evaporated in oven at 60 °C. The remaining

135 total lipids were quantified gravimetrically. Similarly during ultrasonication, 20 mL of

136 solvent mixture was added to the 2 g of dried biomass in a 50 mL centrifuge tube, after which

137 the entire mixture was sonicated for two minutes at 15 kHz (MisonixXL-2000-010; 100 W of

138 output power, 22.5 kHz of output frequency). The mixture was centrifuged and supernatant

139 transferred to a separate tube. A further 20 mL of the solvent mixture was added to the

140 centrifuged cell debris, sonicated and supernatant separated by centrifugation and mixed with

141 previously recovered solvent. The pooled solvent was vacuum filtered and evaporated in
142 oven at 60 °C to recover the extracted lipids. Autoclaving was also investigated as a cell

143 disruption procedure in which 500 mg of dried biomass was added with 50 mL of ultrapure

144 water and autoclaved at 121 °C and 15 lbs for five minutes (Prabakaran & Ravindran, 2011).

145 The autoclaved sample was mixed with extracting solvent mixture in 1:1 ratio and transferred

146 to a separatory funnel after mixing for five minutes, where two layers formed. The lipid

147 containing layer was carefully removed and lipids were recovered after drying in oven at 60

148 °C. Osmotic shock with 10 % NaCl also results in cell disruption and accessibility to

149 intracellular content due to changing osmotic pressures on both sides of cell walls. 500 mg of

150 dried biomass was mixed with 50 mL of 10 % NaCl solution and vortexed (VM-300,

151 Gemmy, Taiwan) for one minute, after which the biomass was left suspended in NaCl

152 solution for two days. After two days, this suspension was added to lipid extracting solvent

153 mixture in 1:1 ratio and transferred to a separatory funnel after five minutes of mixing. The

154 distinct layer containing lipids was carefully removed and evaporated at 60 °C in oven to

155 recover lipids. After each extraction, the lipid mass was quantified gravimetrically and lipid

156 yields (% w/w) of the processes were calculated.

157

158 The algal biomass remaining after lipid extraction was vacuum filtered and air dried at room

159 temperature to obtain lipid extracted algae (LEA) from which proteins and reduced sugars

160 were extracted. This recovered biomass was quantified and utilized in calculating the

161 subsequent yields from LEA.

162

163 2.4 Protein extraction from LEA

164 Dried whole cell algae or LEA biomass was subjected to protein extraction as per Lowry

165 method (López et al., 2010; Lowry et al., 1951). All reagents were prepared as per the

166 procedure outlined by López et al. (2010). The actual procedure was subjected to minor
167 modifications to optimize the protein extraction in the present study. 200 mg of different

168 dried LEA or whole cell algae samples were added to 25 mL of lysis buffer solution. The

169 mixture was ground using a mortar and pestle for five minutes, and subsequently vortexed for

170 two minutes. The mixture was then centrifuged at 3000 rpm for 10 minutes, and supernatant

171 decanted into a separate tube. 25 mL lysis buffer was added to the pellet and again ground,

172 vortexed, and centrifuged. Both supernatants were pooled and mixed. 0.5 mL of this

173 supernatant was mixed with 0.5 mL of SDS solution in a test tube and vortex mixed.

174 Subsequently 5 mL of reagent-C was added in the mixture and vortex mixed. The solution

175 was left to rest for 10 minutes, before 0.5 mL of Folin reagent was added in the test tube and

176 mixed. Absorbance of this mixture was measured at 750 nm using a spectrophotometer

177 (SpectroquantPharo 300, Merck) after 30 minutes of sample hold. Bovine serum albumin

178 (BSA) was dissolved in lysis buffer and this solution was used to develop calibration curve

179 for protein quantification, and the protein yields (% w/w) were calculated from the standard

180 curve (López et al., 2010).

181

182 2.5 Extraction of reduced sugars from LEA

183 Polysaccharides present in the whole cell algae as well as LEA were hydrolyzed to reduced

184 sugars using H2SO4 and autoclaving (Sun & Cheng, 2002). 500 mg of dried algae or LEA

185 were added to 50 mL of H2SO4 (2 % v/v) in a 100 mL conical flask. Hydrolysis was achieved

186 by autoclaving this mixture at 121 °C for 30 minutes. Flasks were removed and cooled to

187 room temperature under running tap water. The hydrolyzed mixture was neutralized with 0.1

188 M NaOH/H2SO4. Total reducing sugars were quantified using the DNS method (Miller,

189 1959). Minor modifications were incorporated to optimize the yield in the present study. 1

190 mL of hydrolyzed sample mixture was added with 1 mL of DNS reagent in a test tube. These

191 tubes were capped and heated to 100 °C by keeping in boiling water for 10 minutes. Samples
192 were cooled to room temperature and 8 mL of ultrapure water was added to each test tube.

193 Absorbance of the sample was measured at 540 nm using a spectrophotometer

194 (SpectroquantPharo 300, Merck). Calibration was done using glucose solutions of different

195 concentrations (Fu et al., 2010).

196

197 2.6 Chemicals and reagents

198 All chemicals and solvents were purchased from Sigma Aldrich, USA and were of

199 analytical/HPLC grade. All solutions were prepared with ultrapure water (Aqua MAX Ultra

200 370, Younglin Korea).

201

202 2.7 Statistical analysis

203 All analyses were performed in triplicate. Two-way ANOVA was carried out to investigate

204 the effects and interactions of various treatments on product yields. Tukey’s honestly

205 significant differences (HSD) test was used for post-hoc analysis of the results.

206

207 3. Results and discussion

208 3.1 Effects of drying and cell disruption procedures on the lipid yield

209 The extracted lipids from S. obliquus have been demonstrated to be suitable for biodiesel

210 conversion based on their fatty acid profile and saponification value (Mandal & Mallick,

211 2009). S. obliquus strain used in this work has also been previously demonstrated to contain

212 suitable composition and characteristics for biofuel conversion (Guldhe et al., 2014).

213 However, the present study found the lipid yields from S. obliquus to significantly depend on

214 various drying as well as cell disruption methods applied during extraction. The resulting

215 lipid yields after drying and cell disruption are presented in Fig. 1. Lipid yields significantly

216 (P < 0.05) varied with different treatments and ranged from 8.78 ± 2.15 % for sun dried
217 samples assisted with cell disruption by osmotic shock with 10 % NaCl solution, to 21.43 ±

218 1.52 % for freeze dried samples with microwave assisted cell disruption. Mata et al. (2010)

219 have also reported similar lipid yields for Scenedesmus sp. Microwave assisted extraction

220 resulted in significantly (P < 0.05) higher lipid yields for all dried samples in comparison to

221 other selected methods, and osmotic shock with 10 % NaCl solution resulted in lowest lipid

222 yields. In addition, sun dried samples had the significantly lower (P < 0.05) lipid yields than

223 all the selected drying techniques.

224

225 The interactions between and within these two different treatments of drying and cell

226 disruption were analyzed with two-way ANOVA at 95 % confidence level, and results are

227 presented in Table 1. Variations in drying methods significantly affected the lipid yields (P =

228 0.005). Similarly variability in cell disruption processes also had significant effects on the

229 lipid yield (P < 0.001). No significant interaction between these two factors of drying and cell

230 disruption was observed (P = 0.477). Post-hoc analysis with Tukey’s test was carried out to

231 perform pairwise comparisons within each group to investigate their impacts on lipid yields.

232 Different drying processes were compared to all cell disruption procedures. Least square

233 means were calculated to be 11.92 for sun dried samples, 14.19 for oven dried samples, and

234 15.07 for freeze dried samples with a standard error of 0.63. These comparisons are

235 summarized in Fig. 1. Sun drying resulted in significantly lower lipid yields when compared

236 to oven drying (P = 0.044) and freeze drying (P = 0.005), while no significant difference in

237 yields was observed for oven dried and freeze dried samples (P = 0.588). Similar

238 comparisons for different cell disruption procedures were also performed for all drying

239 processes. These processes had least square means of 18.67 for microwave, 12.77 for

240 ultrasonication, 14.48 for autoclaving, and 9.00 for osmotic shock with 10 % NaCl, with an

241 associated standard error of 0.72. Samples undergoing cell disruption with NaCl showed
242 significantly lower lipid yields than the other three processes. Microwave assisted cell

243 disruption resulted in significantly higher lipid yields than the other methods. No significant

244 difference was observed between ultrasonicated and autoclaved samples for such yield (P =

245 0.361). Guldhe et al. (2014) also found better lipid yields with microwave assisted extraction

246 in comparison to ultrasonication for Scenedesmus sp.

247

248 Fig. 1 also depicts the pair wise comparisons of effects of variability within both treatment

249 groups for each individual interaction on lipid yields. For sun dried samples, microwave cell

250 disruption resulted in significantly higher lipid yields in comparison to ultrasonication and

251 NaCl assisted extraction. However, no significant difference in yields was observed in sun

252 dried samples with microwave and autoclave assisted extractions (P = 0.099). Lipid yields of

253 ultrasonication, autoclaving, and NaCl treatment were statistically similar. In comparison, for

254 oven dried samples, microwave assisted extraction resulted in significantly higher yields (P <

255 0.001) from the NaCl assisted process only, while no significant differences were observed in

256 comparison to ultrasonication (P = 0.123) and autoclaving (P = 0.392). Yields for oven dried

257 samples with NaCl were also significantly lower than ultrasonicated (P = 0.044) and

258 autoclaved (P = 0.009) samples, while autoclaved and ultrasonicated oven dried samples

259 resulted in statistically similar lipid yields (P = 0.896). For freeze dried samples, microwave

260 cell disruption provided significantly higher lipid yields than other processes, while NaCl

261 assisted process had significantly lower yields than other processes with the exception of

262 ultrasonication. Autoclaving and ultrasonication did not show any significant difference in

263 the yields (P = 0.582). A pairwise comparison between different drying methods showed a

264 significant difference in lipid yields of microwave assisted extraction of sun dried and freeze

265 dried samples only (P = 0.023).

266
267 These results highlight the inefficiency of sun drying process in comparison to oven or freeze

268 drying. In addition, the longer drying period with sun drying exerts huge land requirement for

269 complete drying in comparison to other processes, and which might be a constraint during

270 full scale application. Widjaja et al. (2009) also reported oven drying at 60 °C and freeze

271 drying to provide best lipid extraction from C. vulgaris. Careful selection of drying methods

272 is important because oven and freeze drying are both energy intensive processes (Guldhe et

273 al., 2014) and may have higher operating costs during long term full scale application.

274 However, higher lipid yields were achieved by these two methods in this study, and which

275 might also be an important factor to selecting the most appropriate drying method.

276

277 Another factor which governed the final lipid yield from algae was the cell disruption

278 procedure assisting the solvent extraction of lipids. Disruption processes result in greater

279 contact between solvent and the cellular lipid than with solvent only and have been found to

280 be more efficient. Microwave, ultrasonication, autoclaving, and osmotic shock are the most

281 common cell disruption methods used for algae (Prabakaran & Ravindran, 2011).

282 Microawave assisted extraction resulted in significantly higher lipid yields among all dried

283 samples in the present study. Higher lipid yields with microwave in comparison to other cell

284 disruption processes are also reported for Botryococcus sp., C. vulgaris, and Scenedesmus sp.

285 (Lee et al., 2010). In addition, they also concluded that microwave assisted lipid extraction of

286 oven dried samples enabled maximum lipid recovery for Scenedesmus sp. which is similar to

287 the findings of the present study. Guldhe et al. (2014) also observed microwave assisted lipid

288 extraction to provide higher yields than ultrasonication. The efficient heating of the whole

289 sample with microwave resulted in better cell disruption, solvent movement, and hence lipid

290 extraction efficiencies (Iqbal & Theegala, 2013). In comparison, osmotic shock with 10 %

291 NaCl resulted in lower lipid yields and which may be a consequence of insufficient cell
292 disruption achieved using this method. The longer time required with osmotic shock (two

293 days) is also a limitation of this process.

294

295 3.2 Extraction of protein from lipid extracted algae (LEA)

296 The algal biomass after extraction of lipids was further utilized to recover protein and

297 reduced sugars. Protein contents (% w/w) of lipid extracted algal biomass should

298 theoretically increase in comparison to whole cell algae with their increased fraction to

299 account for the removed mass of extracted lipids. However, the cell disruption and solvent

300 application during lipid extraction also results in loss of some protein from the biomass

301 reducing the total amount available for extraction. Hence it is important to investigate and

302 compare the protein yields from such lipid extracted algae (LEA) to establish their

303 applicability as a protein source. Fig. 2 presents the protein yields from such lipid extracted

304 algae with different treatments. Similar yields from whole cell algae without lipid extraction

305 were also determined, and are presented in Fig. 2. Sun dried samples of S. obliquus without

306 any lipid extraction resulted in the protein yield of 41.13 ± 1.64 %, while significantly higher

307 (P < 0.05) yields of 51.42 ± 1.01 % for oven dried and 51.19 ± 0.81 % for freeze dried intact

308 samples were achieved. Becker (2007) reported protein contents for S. obliquus to be around

309 50-56 % which are similar to those obtained in the present study. LEA demonstrated

310 sufficient protein recovery yields when compared to whole cell algae. Protein yields in LEA

311 depended on the drying and cell disruption procedures for lipid extraction. The highest yield

312 of 58.03 ± 1.29 % was observed for oven dried samples assisted with ultrasonication while

313 sun dried samples with autoclaving resulted in the lowest protein yield of 37.51 ± 1.57 %.

314

315 The effects of different treatments on protein yields were analyzed, and such yields were

316 significantly affected by both different drying (P < 0.001) and cell disruption (P < 0.001)
317 procedures (Table 2). In addition, a significant interaction (P < 0.001) was also observed

318 between these two treatments. Further pair wise comparison was also carried out with post-

319 hoc Tukey’s test (Fig. 2). For different drying processes, least square means were calculated

320 as 42.79 for sun drying, 51.02 for oven drying, and 47.09 for freeze drying with a standard

321 error of 0.36. Oven drying resulted in significantly higher protein yields (P < 0.001) than

322 other drying methods among all cell disruption processes. Similarly, protein yields in sun

323 dried samples were significantly (P < 0.001) lower than other methods. These comparisons

324 were also performed for different cell disruption procedures to compare their effects on the

325 protein yields. Least square mean of 47.91 was calculated for intact samples of whole cell

326 algae without any cell disruption. Means were calculated for LEA samples as 44.11 for

327 microwave, 52.46 for ultrasonication, 43.75 for autoclaving, and 46.53 for NaCl assisted cell

328 disruption. Standard error of such means was calculated as 0.46. Ultrasonicated LEA samples

329 had a significantly higher (P < 0.001) protein yield than whole cell algae. Such protein yields

330 were significantly reduced for microwave (P < 0.001) and autoclave (P < 0.001) assisted

331 LEA samples when compared to the yields from intact samples of whole cell algae. Osmotic

332 shock with 10 % NaCl resulted in LEA which had similar protein yields to whole cell algae

333 (P = 0.245). From all the cell disruption procedures, ultrasoniaction favored highest protein

334 yields after lipid extraction, while such yields significantly reduced for both microwave and

335 autoclave assisted cell disruption.

336

337 Effects of variability on protein yields for each individual interaction within both treatment

338 groups are also compared in Fig. 2. For sun dried samples, ultrasonicated LEA had

339 significantly higher (P < 0.001) protein yield while autoclaved LEA showed significantly

340 lower (P = 0.026) protein yield than whole cell algae. No significant differences were

341 observed between yields of microwave and NaCl assisted LEA and yield of whole cell algae.
342 Ultrasonication assisted LEA also showed significantly higher (P < 0.001) protein yields than

343 whole cell algae for oven dried samples. Similarly autoclaving significantly lowered (P =

344 0.002) protein yield than whole cell algae. For freeze dried samples, whole cell algae showed

345 significantly higher protein yields than LEAs assisted with autoclave (P = 0.01), microwave

346 (P < 0.001), or NaCl (P < 0.001); while ultrasonicated LEA had comparable (P = 0.533)

347 yields to whole cell algae. The effects of different drying methods with each cell disruption

348 procedure is also compared in Fig. 2. Sun dried LEA as well as whole cell algae samples had

349 significantly lower (P < 0.001) protein yield than oven dried samples for all cell disruption

350 methods. Similar yields of sun dried samples were lower (P < 0.001) than freeze dried

351 samples for whole cell algae and autoclaved LEA only. Oven drying also produced higher

352 yields than freeze drying for microwave, ultrasonication, and NaCl assisted LEA.

353

354 These results further highlight the limitation of sun drying, since sun dried samples had

355 significantly lower protein yields for both LEA and whole cell algae. In comparison, oven

356 drying resulted in significantly higher protein yields. Since oven drying is an energy intensive

357 process, higher yields of lipids and proteins obtained should be helpful in making such oven

358 drying more cost effective in nature. The protein yields of different LEA were also compared

359 with whole cell algae, and ultrasonication was found to significantly increase yields, while

360 microwave and autoclave reduced the protein yields. When these results are compared with

361 those for lipid extraction, some important insights are gained into the dynamics of these

362 processes. During lipid extraction, microwave and autoclave assisted processes had higher

363 yields than ultrasonication and osmotic shock (Fig. 1). However, the increased cell disruption

364 associated with microwave and autoclave during lipid extraction resulted in increased loss of

365 protein from the disrupted cell to the solvents as higher amounts of protein became unbound

366 and were washed with lipid extracting solvents. Though these processes resulted in the higher
367 lipid yields, the loss of protein with applied solvent for lipid extraction reduced the protein

368 yields from such LEA. This is further supported by the fact that ultrasonication and osmotic

369 shock which resulted in relatively lower cell disruption and hence lower lipid yields provided

370 higher protein yields from their LEA since the loss to the solvents was reduced. Safi et al.

371 (2014) investigated the effects of different cell disruption methods on the extraction of

372 protein from whole cell algae without any lipid extraction and found the protein yields to

373 increase from manual grinding to high pressure homogenization with intermediate yields

374 from ultrasonication and chemical treatment. These yields also follow the trends observed

375 with lipid extraction in the present study and are a consequence of better cell disruption and

376 availability of cellular unbound protein to the extracting solvent. Despite the loss in LEA for

377 different cell disruption procedures, sufficient protein yields were still obtained thus

378 establishing the feasibility of extraction of these coproducts from algae. In addition, the

379 results also demonstrate the complexity of the objective to optimize the extraction of different

380 products using an integrated algal biorefinery approach. The proper selection and

381 optimization of processes involved is important for simultaneously optimizing the yields of

382 these products.

383

384 3.3 Extraction of reduced sugar from lipid extracted algae (LEA)

385 Similar to protein, LEA are also rich in polysaccharides. Such polysaccharides can be

386 hydrolyzed to reduced sugars which can be extracted for their application in bioethanol

387 production (Sun & Cheng, 2002). Miranda et al. (2012) reported sufficient carbohydrate

388 content from S. obliquus. In addition, they also characterized its composition after reduction,

389 and observed glucose as the major fraction (70 % w/w of all sugars), followed by other

390 monosaccharides such as mannose, galactose, xylose and arabinose. The contents of

391 carbohydrate and composition of reduced products are suitable for bioethanol production
392 from S. obliquus. The feasibility of using LEA as a potential source for such sugars was also

393 investigated in the present work to achieve improved utilization of algal biomass in

394 accordance with biorefinery approach. The yields of such reduced sugars (% w/w) in the

395 whole cell algae depended on the drying process and the efficacy of hydrolysis reaction

396 needed for such conversion of total sugars to reduced sugars and ranged from 12.79 ± 3.44 %

397 for freeze dried to 14.63 ± 3.52 % for sun dried samples (Fig. 3). LEA also provided

398 statistically comparable yields of such reduced sugar for different treatments of drying and

399 cell disruption, ranging from 12.37 ± 3.03 % for freeze dried samples assisted with

400 ultrasonication to 19.51 ± 3.81 % for oven dried samples with autoclaving. The yields of

401 reduced sugar obtained in the present study for both whole cell algae and LEA were

402 comparable to those reported in literature (Fu et al., 2010). The effects of variability between

403 different treatments were also analyzed by two-way ANOVA (Table 3). No significant

404 effects on the yield of reduced sugar were observed due to variability in either drying (P =

405 0.1) or cell disruption (P = 0.273) methods for lipid extraction. Similarly, no significant

406 interaction (P = 0.981) was observed between these two treatments on the yield of reduced

407 sugar. Insignificant difference between all treatments for whole cell algae as well as LEA

408 may be a consequence of increased errors introduced into the extraction due to the hydrolysis

409 process for converting total sugar to reduced sugar in the cell as observed by Fu et al. (2010)

410 also in their hydrolysis yields. No significant difference was observed between the yields of

411 reduced sugar between whole cell algae and LEA undergoing different treatments. These

412 comparable yields of reduced sugar for whole cell algae and LEA qualitatively establish the

413 feasibility of LEA as their source.

414

415 4. Conclusion
416 This study demonstrated the feasibility of using LEA as source of protein and reduced sugars.

417 Microwave assisted extraction of oven dried samples resulted in maximum lipid yield, while

418 minimum yields were obtained from sun dried samples with osmotic shock. In comparison,

419 oven dried LEA with ultrasonication provided significantly higher protein yields, while such

420 yields from microwave and autoclaved samples reduced in comparison to whole cell algae. In

421 addition, yields of reduced sugar from LEAs undergoing different treatments were

422 comparable to those obtained from whole cell algae. These results also contribute towards

423 integrated biorefinery approach for maximum value extraction from algae.

424

425 Acknowledgements

426 The authors hereby acknowledges the National Research Foundation and Durban University

427 of Technology for providing financial assistance.

428

429 References

430 1. Becker, E.W., 2007. Micro-algae as a source of protein. Biotechnol. Adv. 25, 207-

431 210.

432 2. Brennan, L., Owende, P. 2010. Biofuels from microalgae—a review of technologies

433 for production, processing, and extractions of biofuels and co-products. Renew. Sust.

434 Energ. Rev. 14, 557-577.

435 3. Di Termini, I., Prassone, A., Cattaneo, C., Rovatti, M., 2011. On the nitrogen and

436 phosphorus removal in algal photobioreactors. Ecol. Eng. 37, 976-980.

437 4. Folch, J., Lees, M., Stanley, G.H.S., 1957. A simple method for the isolation and

438 purification of total lipids from animal tissues. J. Biol. Chem. 226, 497-509.
439 5. Fu, C.-C., Hung, T.-C., Chen, J.-Y., Su, C.-H., Wu, W.-T., 2010. Hydrolysis of

440 microalgae cell walls for production of reducing sugar and lipid extraction. Bioresour.

441 Technol. 101, 8750-8754.

442 6. Gao, M.-T., Shimamura, T., Ishida, N., Takahashi, H., 2012. Investigation of

443 utilization of the algal biomass residue after oil extraction to lower the total

444 production cost of biodiesel. J. Biosci. Bioeng. 114, 330-333.

445 7. Geun Goo, B., Baek, G., Jin Choi, D., Il Park, Y., Synytsya, A., Bleha, R., Ho Seong,

446 D., Lee, C.-G., Kweon Park, J., 2013. Characterization of a renewable extracellular

447 polysaccharide from defatted microalgae Dunaliella tertiolecta. Bioresour. Technol.

448 129, 343-350.

449 8. Guldhe, A., Singh, B., Rawat, I., Ramluckan, K., Bux, F., 2014. Efficacy of drying

450 and cell disruption techniques on lipid recovery from microalgae for biodiesel

451 production. Fuel. 128, 46-52.

452 9. Iqbal, J., Theegala, C., 2013. Microwave assisted lipid extraction from microalgae

453 using biodiesel as co-solvent. Algal Res. 2, 34-42.

454 10. Ju, Z.Y., Deng, D.-F., Dominy, W., 2012. A defatted microalgae (Haematococcus

455 pluvialis) meal as a protein ingredient to partially replace fishmeal in diets of pacific

456 white shrimp (Litopenaeus vannamei, Boone, 1931). Aquaculture. 354-355, 50-55.

457 11. Lam, M.K., Tan, I.S., Lee, K.T., 2014. Utilizing lipid-extracted microalgae biomass

458 residues for maltodextrin production. Chem. Eng. J. 235, 224-230.

459 12. Lee, J.-Y., Yoo, C., Jun, S.-Y., Ahn, C.-Y., Oh, H.-M., 2010. Comparison of several

460 methods for effective lipid extraction from microalgae. Bioresour. Technol. 101, S75-

461 S77.
462 13. Lodge-Ivey, S.L., Tracey, L.N., Salazar, A., 2014. Ruminant nutrition symposium: the

463 utility of lipid extracted algae as a protein source in forage or starch-based ruminant

464 diets. J. Anim. Sci. 92, 1331-1342.

465 14. López, C.V.G., García, M.d.C.C., Fernández, F.G.A., Bustos, C.S., Chisti, Y., Sevilla,

466 J.M.F., 2010. Protein measurements of microalgal and cyanobacterial biomass.

467 Bioresour. Technol. 101, 7587-7591.

468 15. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement

469 with the folin phenol reagent. J. Biol. Chem. 193, 265-275.

470 16. Lundquist, T.J., Woertz, I.C., Quinn, N.W.T., Benemann, J.R., 2010. A realistic

471 technology and engineering assessment of algae biofuel production. Energy

472 Biosciences Institute, University of California, Berkeley, USA.

473 17. Mandal, S., Mallick, N., 2009. Microalga Scenedesmus obliquus as a potential source

474 for biodiesel production. Appl. Microbiol. Biotechnol. 84, 281-291.

475 18. Mata, T.M., Martins, A.A., Caetano, N.S., 2010. Microalgae for biodiesel production

476 and other applications: a review. Renew. Sust. Energ. Rev. 14, 217-232.

477 19. Miller, G.L., 1959. Use of dinitrosalicylic acid reagent for determination of reducing

478 sugar. Anal. Chem. 31, 426-428.

479 20. Miranda, J.R., Passarinho, P.C., Gouveia, L. 2012. Bioethanol production from

480 Scenedesmus obliquus sugars: the influence of photobioreactors and culture

481 conditions on biomass production. Appl. Microbiol. Biotechnol. 96, 555-564.

482 21. Misra, R., Guldhe, A., Singh, P., Rawat, I., Bux, F. 2014. Electrochemical harvesting

483 process for microalgae by using nonsacrificial carbon electrode: a sustainable

484 approach for biodiesel production. Chem. Eng. J. 255, 327-333.

486 perspectives of photobioreactors for biofuel production. Appl. Microbiol. Biotechnol.

487 87, 1291-1301.

488 23. Olguín, E.J., 2012. Dual purpose microalgae-bacteria-based systems that treat

489 wastewater and produce biodiesel and chemical products within a biorefinery.

490 Biotechnol. Adv. 30, 1031-1046.

491 24. Park, J.B.K., Craggs, R.J., Shilton, A.N., 2011. Wastewater treatment high rate algal

492 ponds for biofuel production. Bioresour. Technol. 102, 35-42.

493 25. Prabakaran, P., Ravindran, A.D., 2011. A comparative study on effective cell

494 disruption methods for lipid extraction from microalgae. Lett. Appl. Microbiol. 53,

495 150-154.

496 26. Rashid, N., Rehman, M.S.U., Han, J.-I., 2013. Recycling and reuse of spent

497 microalgal biomass for sustainable biofuels. Biochem. Eng. J. 75, 101-107.

498 27. Safi, C., Ursu, A.V., Laroche, C., Zebib, B., Merah, O., Pontalier, P.-Y., Vaca-Garcia,

499 C., 2014. Aqueous extraction of proteins from microalgae: effect of different cell

500 disruption methods. Algal Res. 3, 61-65.

501 28. Sawayama, S., Rao, K.K., Hall, D.O., 1998. Nitrate and phosphate ion removal from

502 water by Phormidium laminosum immobilized on hollow fibres in a photobioreactor.

503 Appl. Microbiol. Biotechnol. 49, 463-468.

504 29. Subhadra, B.G., 2010. Sustainability of algal biofuel production using integrated

505 renewable energy park (IREP) and algal biorefinery approach. Energy Policy, 38,

506 5892-5901.

507 30. Sun, Y., Cheng, J., 2002. Hydrolysis of lignocellulosic materials for ethanol

508 production: a review. Bioresour. Technol. 83, 1-11.

510 Pet. Inst. 48, 251-259.

511 32. Widjaja, A., Chien, C.-C., Ju, Y.-H., 2009. Study of increasing lipid production from

512 fresh water microalgae Chlorella vulgaris. J. Taiwan Inst. Chem. Eng. 40, 13-20.

513 33. Wu, L.F., Chen, P.C., Huang, A.P., Lee, C.M., 2012. The feasibility of biodiesel

514 production by microalgae using industrial wastewater. Bioresour. Technol. 113, 14-

515 18.

516 34. Yang, Z., Guo, R., Xu, X., Fan, X., Luo, S., 2011. Hydrogen and methane production

517 from lipid-extracted microalgal biomass residues. Int. J. Hydrogen Energy. 36, 3465-

518 3470.

519 35. Zhu, Y., Albrecht, K.O., Elliott, D.C., Hallen, R.T., Jones, S.B., 2013. Development

520 of hydrothermal liquefaction and upgrading technologies for lipid-extracted algae

521 conversion to liquid fuels. Algal Res. 2, 455-464.

522

523

524
525 Table 1
526 Results of two-way ANOVA for different drying and cell disruption procedures on lipid yield
527 from S. obliquus
528
Source of Variation DF SS MS F P
Drying Process 2 63.426 31.713 6.722 0.005
Cell Disruption Process 3 434.251 144.75 30.684 < 0.001
Drying × Cell 6 26.984 4.497 0.953 0.477
Disruption
Residual 24 113.22 4.717
Total 35 637.881 18.225
529
530 DF: Degree of Freedom; SS: Sum of Square; MS: Mean of Square; F: Likelihood Ratio; P:
531 Probability
532
533
534
535
Table 2
Results of two-way ANOVA for different drying and cell disruption procedures on
protein yield from lipid extracted S. obliquus

Source of Variation DF SS MS F P
Drying Process 2 513.299 256.65 132.608 <0.001
Cell Disruption 4 447.578 111.894 57.815 <0.001
Process
Drying × Cell 8 174.192 21.774 11.25 <0.001
Disruption
Residual 30 58.062 1.935
Total 44 1193.131 27.117

DF: Degree of Freedom; SS: Sum of Square; MS: Mean of Square; F: Likelihood Ratio;
P: Probability
 1 Table 3
2 Results of two-way ANOVA for different drying and cell disruption procedures on reduced
3 sugaryield from lipid extracted S. obliquus
4
Source of Variation DF SS MS F P
Drying Process 2 55.763 27.882 2.484 0.1
Cell Disruption Process 4 60.822 15.206 1.355 0.273
Drying × Cell Disruption 8 21.001 2.625 0.234 0.981
Residual 30 336.674 11.222
Total 44 474.26 10.779
5
6 DF: Degree of Freedom; SS: Sum of Square; MS: Mean of Square; F: Likelihood Ratio; P:
7 Probability
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42

25
43 FIGURE CAPTIONS
44
45
46 Fig. 1
47 Effects of different drying and cell disruption procedures on the lipid yields of S. obliquus.
48 Small normal face letters signify level of significance between different cell disruption
49 treatments for an individual drying process. Small bold face letters signify significance level
50 between different cell disruption treatments for overall drying effect. Capital normal face
51 letters signify level of significance between different drying methods for an individual cell
52 disruption process. Capital bold face letters signify significance level between different
53 drying processes for overall cell disruption effect. Processes followed by same letters have no
54 significant difference (2-way ANOVA, Tukey’s test, P < 0.05).
55
56
57 Fig. 2
58 Effects of different drying and cell disruption procedures on the protein yields of lipid
59 extracted S. obliquus. Small normal face letters signify level of significance between different
60 cell disruption treatments for an individual drying process. Small bold face letters signify
61 significance level between different cell disruption treatments for overall drying effect.
62 Capital normal face letters signify level of significance between different drying methods for
63 an individual cell disruption process. Capital bold face letters signify significance level
64 between different drying processes for overall cell disruption effect. Processes followed by
65 same letters have no significant difference (2-way ANOVA, Tukey’s test, P < 0.05).
66
67
68 Fig. 3
69 Effects of different drying and cell disruption procedures on the reduced sugar yields of lipid
70 extracted S. obliquus.
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91 Fig. 1
92

26
 Sun Dried Oven Dried Freeze Dried
A B B

a b b c
25%

aB
aAB
20% aA

aAB
Lipid Yield (% w/w)

bB
aA
bcA

abA
15%

bA

bA

cA
bA
10%

5%

0%
Microwave Ultrasonication Autoclave NaCl
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122 Fig. 2
123

27
124
Sun Dried Oven Dried Freeze Dried
A B C
a b c b a
70%

bB
60%

acA
Protein Yield (% w/w)

acB
aB
aB

bA

cdB

aB
bdA
cB

aA
50%

bA
aA

aA

cA
40%
30%

20%

10%

0%
Total Algae Microwave Ultrasonication Autoclave NaCl

125 Lipid Extracted Algae

28
157
Sun Dried Oven Dried Freeze Dried
Reduced Sugar Yield (% w/w) 25%

20%

15%

10%

5%

0%
Total Algae Microwave Ultrasonication Autoclave NaCl

158 Lipid Extracted Algae

189

29
190 Highlights

191 • Lipid extracted algae (LEA) was used as a source for protein and reduced sugars.

192 • Comparable yields of these products were obtained from total algae and LEA.

193 • Microwave assisted extraction from oven dried samples provided highest lipid yield.

194 • Effective cell disruption for lipid extraction increased loss of other products.

195 • Maximizing the yields of all products requires proper process selection.

196

197

30