summary from one of the following research articles

homework_7_due_3.21.16.pdf

Home >Biology homework help >summary from one of the following research articles

Homework 7 Due: 3/21/2016 (Monday) Question: Each student should write a page of review summary from one of the following research articles. You can combine the knowledge that you learn from the biofuels classes. Please choose your interested article and upload your homework on the Blackboard. The summary should contain following scientific component: 1) Background, 2) experimental design; 3) interpretation of the data; 4) statistical analysis; 5) impact; and 6) conclusion. Research articles: (The full version of the articles are attached below)

1. Sivakumar et al. 2014. Bioprocessing of Stichococcus bacillaris strain siva2011. Biotechnology for Biofuels, 7:62.

2. Sivakumar et al. 2014. Biomass and RRR-a-tocopherol production in Stichococcus bacillaris strain siva2011 in a balloon bioreactor. Microbial Cell Factories, 13: 79

RESEARCH Open Access

Bioprocessing of Stichococcus bacillaris strain siva2011 Ganapathy Sivakumar1*, Kwangkook Jeong2 and Jackson O Lay Jr3

Abstract

Background: Globally, the development of a cost-effective long-term renewable energy infrastructure is one of the most challenging problems faced by society today. Microalgae are rich in potential biofuel substrates such as lipids, including triacylglycerols (TAGs). Some of these algae also biosynthesize small molecule hydrocarbons. These hydrocarbons can often be used as liquid fuels, often with more versatility and by a more direct approach than some TAGs. However, the appropriate TAGs, accumulated from microalgae biomass, can be used as substrates for different kinds of renewable liquid fuels such as biodiesel and jet fuel.

Results: This article describes the isolation and identification of a lipid-rich, hydrocarbon-producing alga, Stichococcus bacillaris strain siva2011, together with its bioprocessing, hydrocarbon and fatty acid methyl ester (FAME) profiles. The S. bacillaris strain siva2011 was scaled-up in an 8 L bioreactor with 0.2% CO2. The C16:0, C16:3, C18:1, C18:2 and C18:3 were 112.2, 9.4, 51.3, 74.1 and 69.2 mg/g dry weight (DW), respectively. This new strain produced a significant amount of biomass of 3.79 g/L DW on day 6 in the 8 L bioreactor and also produced three hydrocarbons.

Conclusions: A new oil-rich microalga S. bacillaris strain siva2011 was discovered and its biomass has been scaled-up in a newly designed balloon-type bioreactor. The TAGs and hydrocarbons produced by this organism could be used as substrates for jet fuel or biodiesel.

Keywords: Algae, Bioreactor, Hydrocarbon, Jet fuel, Triacylglycerol

Background Worldwide consumption of crude oil is predicted to grow continuously. It is clear that in spite of improvements in the recovery of traditional fossil fuels, alternative renew- able energy resources will at some point be needed. Moreover, such renewable fuels offer the prospect of minimizing increases in atmospheric CO2 by recycling carbon from the atmosphere back into a viable liquid fuel (or perhaps eventually sequestering it entirely). Over a large number of cycles, the net effect could be a significant reduction in the addition of CO2 into the at- mosphere compared to continued reliance only on fossil fuels. A wide variety of existing biofuel technologies have been tested, but none have proven to provide a suitable source of replacement liquid fuels. Although current alter- natives such as ethanol and biodiesel can provide carbon neutrality, fuels derived largely from normally edible plant

sources affect the food supply negatively [1-3]. For these reasons algae feedstocks are being explored as an alterna- tive [4]. The development of a suitable algal-based jet fuel from algal biomass may also impact air transportation. The jet fuel approach is to chemically process tria-

cylglycerols (TAGs) to alkanes. This could be done by catalytic hydrotreating, breaking the TAG molecule and removing the oxygen to form alkanes. While this product meets diesel specifications, it can be further upgraded into jet fuel or naphtha by hydrocracking, isomerization and catalytic reforming [5]. The by-product propane can be used for residential central heating. However, not all microalgae are capable of producing sufficient TAGs and hydrocarbons for effective fuel production. While others might produce abundant TAGs, they might not necessar- ily be the optimum TAGs for production of high-value products such as aviation fuel. For production of such spe- cialized fuels, the selection of the algal species is the key to success. Carbon profiles for selecting algal strains [6] and catalytic hydrothermal decarboxylation of fatty acids for aviation fuel [7] have been studied.

* Correspondence: [email protected] 1Arkansas Biosciences Institute and College of Agriculture and Technology, Arkansas State University, PO Box 639, Jonesboro, AR 72401, USA Full list of author information is available at the end of the article

Sivakumar et al. Biotechnology for Biofuels 2014, 7:62 http://www.biotechnologyforbiofuels.com/content/7/1/62

mailto:[email protected]

http://creativecommons.org/licenses/by/2.0

Recently, Stichococcus bacillaris Naegeli was proposed as a potential and dedicated candidate for use in fuel pro- duction [8]. S. bacillaris is a green soil microalga which in- cludes over 14 species [9]. Cells are approximately 2 to 3 μm in diameter. The state of filamentous or unicellular structures depends on salinity [10]. This species can grow both in freshwater and seawater with different growth kin- etics [11], while tolerating high salinities [12]. In addition, this alga has adapted to low temperatures and is found in ice-free areas of Antarctica [13]. Moreover, it also contains high NADPH-GDH activity [14], low CO2 resistance [15] and has unique microtubule organization in prophase [16]. The NADPH-GDH plays an important role in photo- synthetic microalgae, which is associated with photore- gulation and the incorporation of ammonia into amino acids. The changes in NADPH-GDH were shown in different culture conditions such as photoautotrophic, heterotrophic and mixotrophic [17]. Compared to am- monium, nitrate-grown S. bacillaris had higher activity of NADPH-GDH [18]. S. bacillaris is fairly abundant glo- bally, can remove heavy metals from hazardous environ- ments [19] and is also capable of biotransforming phenols [20]. These characteristics suggest that S. bacillaris could minimize water contamination or improve water quality. In addition, this organism has a short life cycle and is

tolerant to different ranges of pH. Most importantly, over 30% of its dry mass can be produced as oil that can be readily converted to biodiesel [21]. Moreover, this alga produced a high percentage of C16 to C18 carbon fatty acids. Therefore, the goal was to isolate Stichococ- cus species for the study of aviation fuel. Other proposed algal strains either produced triterpene hydrocarbons that are difficult to convert cost-effectively to usable fuels or grew too slowly to be useful [22,23]. Some other TAGs are produced from algae but they typically yield a low biomass [24]. Thus, the aim of this research has been: 1) to isolate new Stichococcus algal species produ- cing significant quantities of lipids and hydrocarbons, es- pecially those suitable for production of aviation fuel; and 2) to evaluate the scale-up potential of this alga in a new design balloon-type bioreactor.

Results and discussions Stichococcus bacillaris strain siva2011 identification A new axenic microalga S. bacillaris strain siva2011 was isolated from an in vitro plant. Microscopic examination demonstrated green rod-shaped cells 5 to 10 μm in length and 2 to 3 μm in diameter. The cells are often presented in chains. The 18S sequence data confirmed that this new alga is a strain of genus Stichococcus with the greatest similarity to S. bacillaris. However, there is a large difference between this and existing strains at nu- cleotides 610 to 980 of the 18S rDNA (see Additional file 1). The 23S sequence did not match with existing S.

bacillaris sequences, providing further confirmation that this is a new species. Due to taxonomic problems of the Stichococcus species and intraspecific biodiversity [25], this new alga was named ‘Stichococcus bacillaris strain siva2011’. The two partial sequences were deposited into the National Center for Biotechnology Information (NCBI) [GenBank:JN168788 and JN168789]. A neighbor-joining tree was created using a Clustal X2.0.12 set to exclude posi- tions with gaps and correct for multiple substitutions. Based on the 18S rDNA sequences, 1,000 bootstrap trials were used to show the relationship of the S. bacillaris and the strain siva2011 (Figure 1). Previously, a similar phylo- genetic tree was reported for S. bacillaris strains NJ-10 and NJ-17 [13].

Bioreactor culture of S. bacillaris strain siva2011 To further explore the potential for enhanced fuel applica- tion, production of the S. bacillaris strain siva2011 was cul- tured using a low-cost indoor balloon-type bioreactor (Figure 2). This alga required both light and sugar sub- strates for optimum growth. The higher biomass accu- mulation was noticed in 1% fructose supplemented medium when compared to other sugars. It is also im- portant to know that in a 6-day culture, a comparison of

Trebouxiophyte sp. UR55/3

Trebouxiophyte sp. UR47/41

Stichococcus deasonii UTEX 1706

Stichococcus bacillaris K4-4

Stichococcus mirabilis CCAP 379/3

Stichococcus sp. MBIC10465

Stichococcus bacillaris D10-1

Stichococcus bacillaris strain siva2011

Stichococcus jenerensis D4

716

582

591

695

409

311

0.001

Figure 1 Phylogenetic tree showing the relationship of S. bacillaris strain siva2011 based on 18S rDNA sequences. A distance of 0.001 is indicated by the scale.

Sivakumar et al. Biotechnology for Biofuels 2014, 7:62 Page 2 of 9 http://www.biotechnologyforbiofuels.com/content/7/1/62

the sugar concentrations and the total inorganic carbon in the media showed no significant difference (data not shown). Light limitation is one of the critical factors in algal biomass scale-up in the photobioreactor. This strain grows well under a low light intensity (15 to 30 μE m-2 s-1

for 10 hours) and requires room temperature. The newly designed reactor has a larger headspace, which efficiently

captures light and also has good media circulation to en- hance photosynthesis. In other words, this unique bioreac- tor configuration could minimize the factors limiting the overall rate of photosynthesis in a high density culture. After autoclaving the media, the pH dropped from 6.0

to 4.2 because of fructose degradation. The sugar deg- radation could be minimized by sterilizing filtration; however, this procedure increases the percentage of cul- ture contamination. Figure 3A,B illustrates the kinetics of S. bacillaris strain siva2011 grown for 6 days in 4 L and 8 L airlift balloon bioreactors with 0.05, 0.1, 0.2 or 0.5% CO2, respectively. Of the four concentrations of CO2 tested, 0.2% yielded the highest biomass. The expo- nential growth phase was noticed on day 4 and station- ary phase on day 6. Although the culture medium nutrients did not deplete in 6 days, the pH dropped to highly acidic levels (Figure 4A,B). The dropping pH could be caused by the degradation of fructose or CO2 effects or HNO3 accumulation in the culture medium. For instance, in the aqueous phase, nitrogen oxide from the media could react with oxygen to form nitrogen di- oxide, which could then react with the hydroxyl radical to form nitric acid. In order to bring the pH back to an optimum level of 4.5 to 5.0 in fructose supplemented medium, every 6 days the culture was subcultured and old media was removed by centrifugation. Alternatively, adding sodium hydroxide to the culture could perhaps have maintained the pH; further studies are needed for verification. This alga did not grow well in algal Bold’s Basal Medium (BBM) (data not shown). On day 6 with 0.2% CO2, a maximum biomass of

3.79 g/L dry weight (DW) was achieved in 8 L and 3.45 g/ L DW in 4 L, respectively. Since the new strain requires a very low 0.2% of CO2, the input cost on large-scale could be minimized. When compared to 0.5% CO2, algal cells grown in a 4 L bioreactor were higher in biomass, with 0.55, 0.986 and 1.45 g/L DW in the 0.05, 0.1 and 0.2% CO2, respectively. Similarly, in the 8 L bioreactor, biomass accumulation was 0.793, 1.107 and 1.79 g/L DW after

Figure 2 Balloon-type bioreactor 20 L culture of S. bacillaris strain siva2011 (working volume 8 L).

0.5

1.5

2.5

3.5

0 1 2 3 4 5 6

D ry

W ei

gh t

(g /L

)

Days

0.05% CO2 0.10% CO2 0.20% CO2 0.50% CO2

A 0

0.5

1.5

2.5

3.5

4.5

0 1 2 3 4 5 6

D ry

W ei

gh t

(g /L

)

Days

0.05% CO2 0.10% CO2 0.20% CO2 0.50% CO2

Figure 3 Growth kinetics of S. bacillaris strain siva2011. (A) Growth for 6 days in a 4 L airlift balloon bioreactor with 0.05, 0.1, 0.2 or 0.5% CO2. (B) Growth for 6 days in an 8 L airlift balloon bioreactor with 0.05, 0.1, 0.2 or 0.5% CO2.

Sivakumar et al. Biotechnology for Biofuels 2014, 7:62 Page 3 of 9 http://www.biotechnologyforbiofuels.com/content/7/1/62

6 days of culture. Both growth and pH kinetic trends were similar in 4 L and 8 L bioreactors. The data supports the notion that this strain does not require light intensity over 15 to 30 μE m-2 s-1; therefore, achieving high density bio- mass may not be a problem. However, the optimization of air flow is needed for better media circulation and growth. The media circulation is critical for efficient photosyn- thesis. The oil-rich alga Ettlia oleoabundans accumulates 2.28 g/L dry biomass in BBM in approximately 22 to 27 days [24], while S. bacillaris strain siva2011 accumu- lates approximately 1.5 g/L higher biomass within 6 days in modified Murashige and Skoog (MS) medium. For scale-up studies, modeling is necessary because it

provides detailed estimates regarding prediction and

validation. For instance, DW of S. bacillaris strain siva2011 has been regarded as an indicator for produc- tivity in the bioprocess. Prediction of DW is needed and helpful for scale-up of S. bacillaris strain siva2011 biomass in large-scale reactors. Quinn et al. [26] mod- eled microalgae growth and lipid accumulation in an outdoor photobioreactor by using MATLAB for biofuel applications. This model qualitatively captures the growth trends with variations in time. Figure 5 shows the pre- diction of normalized DW for 0.2% CO2 in 4 L and 8 L bioreactors compared with experimental/measured data obtained from each test. An overall trend of predicted DW for 8 L was similar to the 4 L data. The R2 value has been estimated in the 8 L as 85.8%, while the 4 L exhibits

0.5

1.5

2.5

3.5

4.5

0 1 2 3 4 5 6

p H

Days

0.05% CO2 0.10% CO2 0.20% CO2 0.50% CO2

A 0

0.5

1.5

2.5

3.5

4.5

0 1 2 3 4 5 6

p H

Days

0.05% CO2

0.10% CO2

0.20% CO2

0.50% CO2

Figure 4 pH kinetics of S. bacillaris strain siva2011. (A) Grown medium pH for 6 days in a 4 L airlift balloon bioreactor with 0.05, 0.1, 0.2 or 0.5% CO2. (B) Grown medium pH for 6 days in an 8 L airlift balloon bioreactor with 0.05, 0.1, 0.2 or 0.5% CO2.

0.2

0.4

0.6

0.8

1.2

1.4

1.6

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

N or

m al

iz ed

D ry

W ei

gh t,

D W

Normalized Time, t*

4L Measured with 0.20%

8L Measured with 0.20%

4L Predicted with 0.20%

8L Predicted with 0.20%

CO2

Figure 5 Prediction of normalized S. bacillaris strain siva2011 dry weight (DW) for 0.2% CO2 in 4 L and 8 L balloon-type bioreactors.

Sivakumar et al. Biotechnology for Biofuels 2014, 7:62 Page 4 of 9 http://www.biotechnologyforbiofuels.com/content/7/1/62

87.4%. Both tests give an error under 15%. This suggests that the selected variables are not adequate for accurate validation and mapping within an acceptable error. Appli- cation of polynomial equations with more variables in the modeling or another mathematical approach might be warranted for future optimization if better accuracy is needed.

Hydrocarbons and FAMEs Figure 6 shows the gas chromatography–mass spectrom- etry (GC-MS) total ion chromatogram of the hydro- carbon fraction harvested from S. bacillaris strain siva2011 biomass. This chromatogram shows that this strain biosynthesizes three free hydrocarbons, namely n-nonadecane (C19H40), nonacosane (C29H60) and hep- tadecane (C17H36). These alkanes are also found in trad- itional and non-traditional liquid fuels. The S. bacillaris strain siva2011 contained 1.36% of total hydrocarbons: the C19H40, C29H60 and C17H36 were 6.3, 4.1 and 3.2 mg/g DW, respectively, with 0.2% CO2. In addition, it was un- changed in both the 4 L and 8 L bioreactor studies. Some of these hydrocarbons were previously reported in other algal species [27,28]. In cyanobacteria, a two-step alkane biosynthetic pathway was reported: 1) acyl-acyl carrier protein (ACP) reductase converted ACP into aldehyde by reduction; and 2) an aldehyde-deformylating oxygenase converted aldehyde into alkane or alkene by oxidation [29,30]. The overexpressed ACP reductase and aldehyde- deformylating oxygenase cyanobacteria LX56 strain bio- mass accumulated 1.1% DW of alkane in a column photobioreactor [30]. In general, alkane accumulation was toxic to algal cells, therefore, production was lowered. However, the longer chain (over C12) of alkane accumula- tion was insignificant with respect to toxicity in Saccharo- myces cerevisiae cells, while C8 to C11 alkanes were cytotoxic [31]. Although the S. bacillaris strain siva2011

biosynthesize longer chain hydrocarbons, the alkane level is lower than in the TAGs. Since the S. bacillaris strain siva2011 is a new species, a reference genome is necessary for transcriptome analysis, and subsequent metabolic en- gineering is unavailable. The S. bacillaris strain siva2011 biomass contains two

free fatty acids (FFAs) (palmitic acid (C16:0), linolenic acid (C18:3)) and phytol (Figure 6). Figure 7 illustrates the fatty acid methyl ester (FAME) profile and total lipid (35.82%) content of S. bacillaris strain siva2011 biomass from the 8 L bioreactor culture with 0.2% CO2. This data shows that this strain contains 31.62% of total FAMEs, 2.3% of FFAs and 1.9% of unidentified fatty acids, which is higher compared to the 4 L bioreactor. In addition, 1.2% of phytol was also found. The S. bacillaris strain 158/11 biomass contained 32% of total lipid and 1% of phytol [21]. The results show that this strain con- tains a high degree of unsaturated fatty acids. The main unsaturated FAMEs detected are methyl hexadecatrieno- ate (C16:3), methyl oleate (C18:1), methyl linoleate (C18:2) and methyl linolenate (C18:3). The predominant saturated FAME is methyl palmitate (C16:0). This profile is consistent with the other S. bacillaris strains, NJ-10 and NJ-17 [13]. When S. bacillaris strain siva2011 was scaled-up in the 8 L bioreactor with 0.2% CO2, the C16:0, C16:3, C18:1, C18:2 and C18:3 were 112.2, 9.4, 51.3, 74.1 and 69.2 mg/g DW, respectively; in the 4 L bio- reactor the FAMEs were 102, 8.1, 49.4, 71.7 and 65.3 mg/g, respectively, which is higher compared to the 0.5% CO2. It suggests not only that S. bacillaris strain siva2011 biomass was scaled-up, but also that the TAG production metabol- ism appeared to be scaled-up as well. However, the FAME profiles were unchanged. Previously this reactor was used for plant root culture and was successful at commercial- scale (10,000 L) cultivation of small molecules [32,33]. It was also demonstrated that the resveratrol metabolic

600

500

400

300

200

100

7.5 10.0 12.5 15.0 17.5 20.0 Minutes

k C

ou n

n -N

on ad

ec an

C 16

P h

yt ol

C 18

N on

ac os

an e

H ep

ta d

ec an

Figure 6 GC-MS profile of the hydrocarbons and free fatty acids from S. bacillaris strain siva2011 biomass. GC-MS, gas chromatography–mass spectrometry.

Sivakumar et al. Biotechnology for Biofuels 2014, 7:62 Page 5 of 9 http://www.biotechnologyforbiofuels.com/content/7/1/62

pathway gene, chalcone synthase, was highly expressed in this type of reactor. Moreover, this reactor design has not only non-agitation hydrodynamics but also enhanced geometry, flow dynamics and kinematics [33,34]. Thus, this configuration could enhance light capturing by mixing the algal cells evenly, which facilitates photosynthesis and bio- mass accumulation as well as possibly up-regulating fatty acid pathway genes. Olivieri et al. [21] showed that the S. bacillaris 158/11 contains C16:0 6.5%, C18:3 5.2%, C18:2 4.6% and C18:1 13.8%. This suggests that the Stichococcus species has unique fatty acid profiles which could be used for high-quality liquid fuel production. However, the actual large-scale feasibility test for algal biomass scale-up is needed for aviation fuel production.

Conclusions Long-term energy demands will eventually greatly out- weigh the world supply of fossil fuels, and their use in- creases greenhouse gases. Therefore, alternative sources and methods of producing fuels must be found. Although algae can capture greenhouse gas emissions while produ- cing oxygen, the need for high biomass and oil accumula- tion are challenging for algal-based bioenergy production. S. bacillaris strain siva2011 is rich in lipids, presumably TAGs, with a suitable carbon range for aviation or other liquid fuels. Indeed, scaling studies showed that at 0.2% CO2 supplementation S. bacillaris strain siva2011 had bet- ter growth and increased FAMEs in the 8 L bioreactor than 4 L. It is likely that the 20 L bioreactor could have substantially lower hydrodynamic stress. However, fur- ther studies on mass transfer at larger scales seem to be warranted. The culture conditions vary from alga spe- cies to species. S. bacillaris strain siva2011 can grow in conditions mentioned in this study. Irrespective of the need to further characterize the biochemical pathways

for this organism, it is nevertheless important to point out that there is already sufficient empirical evidence that it will likely be a possible candidate for renewable production of light liquid fuels based on the copious production of lipids and hydrocarbons, and especially the relatively high degree of unsaturation found therein.

Materials and methods Isolation and identification of Stichococcus bacillaris strain siva2011 Axenic S. bacillaris strain siva2011 cells were isolated from in vitro Lagerstroemia seedlings. The morphology of algal cells was identified by light microscopy. The genus and species were identified by the 18S rDNA region of the nuclear chromosome and the 23S region of the chloro- plast rDNA. PCR was performed using primers to amplify the 23S and the 18S region of the rDNA. The products were then sequenced. The genus Stichococcus was iden- tified based on the 18S rDNA sequence in the NCBI database. The identification was confirmed and authen- ticated by The Culture Collection of Algae at the Uni- versity of Texas at Austin (UTEX), Austin, TX, USA. The phylogenetic tree was created based on the 18S rDNA sequence using a Clustal X2.0.12 set to exclude positions with gaps, correct for multiple substitutions and run 1,000 bootstrap trials.

Bioreactor culture The new alga S. bacillaris strain siva2011 was cultured in 5 L and 20 L liquid-phase airlift balloon-type bioreactors [32-34] with modified MS [35] liquid medium for 6 days. This alga was also tested in BBM for comparison [36]. To evaluate the scale-up potential of the balloon-type bioreac- tor for larger-scale use, the 5 L bioreactor was used for the 4 L working volume and a 20 L bioreactor was used for the

% o

f L

ip id

FAMEs

Figure 7 Total lipid content from S. bacillaris strain siva2011, biomass from 20 L bioreactor and working volume 8 L with 0.2% CO2 on day 6. FA, fatty acid; FAME, fatty acid methyl ester; FFA, free fatty acid.

Sivakumar et al. Biotechnology for Biofuels 2014, 7:62 Page 6 of 9 http://www.biotechnologyforbiofuels.com/content/7/1/62

8 L working volume in order to gain a linear biomass pat- tern for prediction or modeling. Working volumes of bio- reactors for scale-up studies were previously published [37] for root culture and were not repeated here. Balloon biore- actors have a larger headspace. The 5 L bioreactor has an 8 inch diameter and the 20 L has a 12 inch diameter, which may facilitate efficient light absorption and medium circu- lation for algal culture. The modified MS medium contains reduced NH4NO3 0.6 g, KNO3 1.5 g [32,33] with 1% fruc- tose and pH 6.0. The cool white fluorescent room lights at 15 to 30 μE m-2 s-1 for 10 hours followed by 14 hours of dark and 23 to 25°C culture conditions were used. After autoclaving the medium and the bioreactor, the axenic algal cells were cultured into the bioreactor. The inoculum was active cells that were 3 days old and 0.05 g fresh weight (FW)/L. The bioreactor cultures were supple- mented with different concentrations of sterile filtered CO2 such as 0.05, 0.1, 0.2 or 0.5%. The input air mixture CO2 gas flow was set at 0.1 vvm (volume (of air) per volume (of liquid) per minute). To screen growth kinetics of S. bacil- laris strain siva2011, algal biomass was harvested, and medium pH was measured after 1, 2, 3, 4, 5 and 6 days. Algal cells were harvested by centrifugation at 10,000 rpm for 5 minutes. After harvesting, algal biomass were frozen in liquid nitrogen and freeze-dried. DW was recorded after the samples were freeze-dried to a constant weight.

Linear regression A linear regression model has been developed based on 0.2% CO2 experimental data to predict DW of S. bacillaris strain siva2011 production ranging from 4 L to 8 L. Re- lated variables including DW, yCO2, pH, time (t) and vol- ume (V) are listed from constituents in S. bacillaris strain siva2011 production tests, as follows:

DW ¼ f yCO2; pH; t; Vð Þ ð1Þ

To test its feasibility, it has been simplified into a lin- ear equation with two variables, t and V, among the whole variables as shown in Equation (2):

DW ¼ f t; Vð Þ ¼ at þ bð Þ⋅V c ð2Þ

Experimental data and variables have been normalized using maximum DW and t to establish a mathematical model as shown in Equations (3) and (4).

DW% ¼ DW=DW f ð3Þ

t% ¼ t=tf ð4Þ

Where DWf is maximum DW (g/L) produced from 8 L test under 0.2% CO2 fraction and tf is maximum t to reach the maximum DW. The DWf and tf were 3.79 g/L on day 6. V has been standardized by a baseline 8 L in this model. Constants a, b and c in Equation (2) have been determined

using test data and a linear least squares method. The 4 L and 8 L data were used to determine the constants: a, b and c. The final non-dimensional equation is suggested by Equation (5):

DW% ¼ 1:0676⋅t% þ 0:042ð Þ⋅ V 2 V 1

! "0:19177 ð5Þ

V1 was base volume at 4 L and V2 was extended at 8 L volume.

Analysis of hydrocarbons and FAMEs One gram of 6-day-old freeze-dried algal cells were used for analysis of hydrocarbons and FAMEs. The total lipids were evaluated according to Jones et al. [38]. FAMEs were processed according to the AOAC method 996.06 and AOCS method Ce 1 h-05 [39,40]. Each FAME GC- MS spectra were acquired using a Clarus 500 gas chroma- tography (PerkinElmer, Waltham, MA, USA) coupled to a Clarus A mass spectrometer (PerkinElmer). A FAME- WAX column (Restek, Bellefonte, PA, USA) was used for separation of FAMEs (30 m length, 0.25 mm ID, 0.25 μm film thickness). The column conditions were determined prior to analysis using a FAME and hydrocarbon reference mixture. Initially, the gas chromatography temperature was 30°C and ramped 10°C/min to a final temp of 220°C and held for 15 minutes at 220°C. Helium was used as the carrier gas. The flow rate was set at 1 mL/min and the spilt ratio was 1:20. The sample injection volume was 1 μL. The mass spectrometer was set to record ranges of spectra from 50 to 500 m/z. The inlet line temperature was set at 300°C and the source temperature was 180°C. Hydrocarbons were processed and analyzed in GC-MS according to Wang et al. [30]. Quantitative analysis of hydrocarbons and FAMEs in the algal biomass was calcu- lated from the calibration curve of the respective standard. Data acquisition and processing were performed with Tur- boMass software (PerkinElmer).

Statistical analysis All experiments were repeated at least three times, each with three replications except sequencing. The experimen- tal variations were expressed as a mean standard error.

Additional file

Additional file 1: NCBI basic local alignment search tool comparison of the S. bacillaris strain siva2011 18S rDNA sequences with different Stichococcus species. A large difference was found in the Clustal X2.0.12 multiple sequence alignment between the S. bacillaris strain siva2011 and the existing strains at nucleotides 610 to 980. SIVA, S. bacillaris strain siva2011.

Abbreviations ACP: Acyl-acyl carrier protein; BBM: Bold’s Basal Medium; CO2: Carbon dioxide; DW: Dry weight; FA: Fatty acid; FAME: Fatty acid methyl ester;

Sivakumar et al. Biotechnology for Biofuels 2014, 7:62 Page 7 of 9 http://www.biotechnologyforbiofuels.com/content/7/1/62

http://www.biomedcentral.com/content/supplementary/1754-6834-7-62-S1.pdf

FFA: Free fatty acid; FW: Fresh weight; GC-MS: Gas chromatography–mass spectrometry; GDH: Glutamate dehydrogenase; HNO3: Nitric acid; KNO3: Potassium nitrate; MS: Murashige and Skoog; NADPH: Nicotinamide adenine dinucleotide phosphate; NCBI: National Center for Biotechnology Information; NH4NO3: Ammonium nitrate; PCR: Polymerase chain reaction; TAG: Triacylglycerol; UTEX: The Culture Collection of Algae at the University of Texas at Austin.

Competing interests The authors declare that they have no competing interests.

Authors’ contributions GS, JL and KJ made substantial contributions to experimental design or analysis or interpretation of data. Specifically, GS led and designed the experiments and performed the organism isolation, bioprocessing and GC-MS quantitative studies. JL performed the gas chromatography separation and mass spectrometry experiments for compound identification. KJ performed the mathematical model to validate the variables. GS wrote the manuscript, which was reviewed and approved by all authors. All authors read and approved the final manuscript.

Acknowledgments This research was funded by the Arkansas Biosciences Institute (grants 262178 and 200109). Part of the mass spectrometry work was supported by the National Institutes of Health (NIH) (P30 GM103450) through the National Institute of General Medicine, and organism identification was supported by the Department of Energy (DOE) (DE-FG36-08BO88036) through the Midsouth/Southeast Bioenergy Consortium. The authors thank Dr David Nobles at UTEX for identification and authentication of S. bacillaris strain siva2011. The technical assistance of Dr Jianfeng Xu, Christopher Easley, Kelsea Brewer, Veronica Hawes and Saritha Kontham (Arkansas State University, Jonesboro, AR, USA), and Dr Jennifer Gidden (University of Arkansas, Fayetteville, AR, USA) was appreciated.

Author details 1Arkansas Biosciences Institute and College of Agriculture and Technology, Arkansas State University, PO Box 639, Jonesboro, AR 72401, USA. 2College of Engineering, Arkansas State University, Jonesboro, AR 72401, USA. 3Arkansas Statewide Mass Spectrometry Facility, University of Arkansas, Fayetteville, AR 72701, USA.

Received: 26 July 2013 Accepted: 17 March 2014 Published: 15 April 2014

References 1. Sivakumar G, Vail DR, Xu J, Burner DM, Lay JO, Ge X, Weathers PJ:

Bioethanol and biodiesel: alternative liquid fuels for future generations. Eng Life Sci 2010, 10:8–18.

2. Ge X, Burner DM, Xu J, Phillips GC, Sivakumar G: Bioethanol production from dedicated energy crops in Arkansas, USA. Biotechnol J 2011, 6:66–73.

3. Ge X, Green S, Zhang N, Sivakumar G, Xu J: Eastern gamagrass as an alternative cellulosic feedstock for bioethanol production. Process Biochem 2012, 47:335–339.

4. Sivakumar G, Xu J, Thompson RW, Yang Y, Randol-Smith P, Weathers PJ: Integrated green algal technology for bioremediation and biofuel. Bioresour Technol 2012, 107:1–9.

5. Tran NH, Bartlett JR, Kannangara GSK, Milev AS, Volk H, Wilson MA: Catalytic upgrading of biorefinery oil from micro-algae. Fuel 2010, 89:265–274.

6. Fadock MN: Carbon profile matching, algae fatty acids and Jet A fuel properties. Guelph Eng J 2010, 3:1–8.

7. Yang C, Nie R, Fu J, Hou Z, Lu X: Production of aviation fuel via catalytic hydrothermal decarboxylation of fatty acids in microalgae oil. Bioresour Technol 2013, 146:569–573.

8. Olivieri G, Gargano I, Andreozzi R, Marotta R, Marzocchella A, Pinto G, Pollio A: Effects of photobioreactors design and operating conditions on Stichococcus bacillaris biomass and biodiesel production. Biochem Eng J 2013, 74:8–14.

9. Neustupa J, Elias M, Sejnohova L: A taxonomic study of two Stichococcus species (Trebouxiophyceae, Chlorophyta) with a starch-enveloped pyrenoid. Nova Hedwigia 2007, 84:51–63.

10. Brown LM, Hellebust JA: Some new taxonomic characteristics applied to Stichococcus bacillaris (Chlorophyceae). Can J Bot 1980, 58:1405–1411.

11. George EA: A note on Stichococcus bacillaris Naeg. and some species of Chlorella as marine algae. J Mar Biol Assoc UK 1957, 36:111–114.

12. Hayward J: Studies on the growth of Stichococcus bacillaris Naeg in culture. J Mar Biol Assoc 1974, 54:261–268.

13. Chen Z, He C, Hu H: Temperature responses of growth, photosynthesis, fatty acid and nitrate reductase in Antarctic and temperate Stichococcus. Extremophiles 2012, 16:127–133.

14. Ahmad I, Hellebust JA: Transport and assimilation of nitrogen by Stichococcus bacillaris grown in the presence of methionine sulfoximine. Plant Physiol 1985, 79:1125–1126.

15. Iwasaki I, Hu Q, Kurano N, Miyachi S: Effect of extremely high-CO2 stress on energy distribution between photosystem I and photosystem II in a 'high-CO2' tolerant green alga, Chlorococcum littorale and the intolerant green alga Stichococcus bacillaris. Photochem Photobiol B 1998, 44:184–190.

16. Katsaros CI, Varvarigos V, Gachon GMM, Brand J, Motomura T, Nagasato C, Küpper FC: Comparative immunofluorescence and ultrastructural analysis of microtubule organization in Uronema sp., Klebsormidium flaccidum, K. subtilissimum, Stichococcus bacillaris and S. chloranthus (Chlorophyta). Protist 2011, 162:315–331.

17. Ahmad I, Hellebust JA: Control of nitrogen assimilation in Stichococcus bacillaris by growth conditions. Can J Bot 1987, 65:432–437.

18. Hellebust JA, Ahmad I: Nitrogen metabolism and amino acid nutrition in the soil alga Stichococcus bacillaris (Chlorophyceae). J Phycol 1989, 25:48–54.

19. Zhang W, Majidi V: Monitoring the cellular response of Stichococcus bacillaris to exposure of several different metals using in vivo 31P NMR and other spectroscopic techniques. Environ Sci Technol 1994, 28:1577–1581.

20. DellaGreca M, Pinto G, Pollio A, Previtera L, Temussi F: Biotransformation of sinapic acid by the green algae Stichococcus bacillaris 155LTAP and Ankistrodesmus braunii C202.7a. Tetrahedron Lett 2003, 44:2779–2780.

21. Olivieri G, Marzocchella A, Andreozzi R, Pinto G, Pollio A: Biodiesel production from Stichococcus strains at laboratory scale. J Chem Technol Biotechnol 2011, 86:776–783.

22. Lohr M, Schwender J, Polle JEW: Isoprenoid biosynthesis in eukaryotic phototrophs: a spotlight on algae. Plant Sci 2012, 185–186:9–22.

23. Niehaus TD, Okada S, Devarenne TP, Watt DS, Sviripa V, Chappell J: Identification of unique mechanisms for triterpene biosynthesis in Botryococcus braunii. Proc Natl Acad Sci U S A 2011, 108:12260–12265.

24. Yang Y, Xu J, Vail D, Weathers PJ: Ettlia oleoabundans growth and oil production on agricultural anaerobic waste effluents. Bioresour Technol 2011, 102:5076–5082.

25. Handa S, Nakahara M, Tsubota H, Deguchi H, Nakano T: A new aerial alga, Stichococcus ampulliformis sp. nov. (Trebouxiophyceae, Chlorophyta) from Japan. Psychol Res 2003, 51:203–210.

26. Quinn J, Winter L, Bradly T: Microalgae bulk growth model with application to industrial scale systems. Bioresour Technol 2011, 102:5083–5092.

27. Thangalazhy-Gopakumar S, Adhikari S, Chattanathan SA, Gupta RB: Catalytic pyrolysis of green algae for hydrocarbon production using H+ZSM-5 catalyst. Bioresour Technol 2012, 118:150–157.

28. Dembitsky VM, Shkrob I, Dor I: Separation and identification of hydrocarbons and other volatile compounds from cultured blue-green alga Nostoc sp. by gas chromatography-mass spectrometry using serially coupled capillary columns with consecutive nonpolar and semipolar stationary phases. J Chromatogr A 1999, 862:221–229.

29. Li N, Chang WC, Warui DM, Booker SJ, Krebs C, Bollinger JM: Evidence for only oxygenative cleavage of aldehydes to alk(a/e)nes and formate by cyanobacterial aldehyde decarbonylases. Biochemistry 2012, 51:7908–7916.

30. Wang W, Liu J, Lu J: Engineering cyanobacteria to improve photosynthetic production of alka(e)nes. Biotechnol Biofuels 2013, 6:69.

31. Chen B, Ling H, Chang MW: Transporter engineering for improved tolerance against alkane biofuels in Saccharomyces cerevisiae. Biotechnol Biofuels 2013, 6:21.

32. Sivakumar G, Yu KW, Paek KY: Production of biomass and ginsenoides from adventitious roots of Panax ginseng in bioreactor cultures. Eng Life Sci 2005, 5:333–342.

33. Sivakumar G, Medina-Bolivar F, Lay JO, Dolan MC, Condori J, Grubbs SK, Wright SM, Baque MA, Lee EJ, Paek KY: Bioprocess and bioreactor: next generation technology for production of potential plant-based antidiabetic and antioxidant molecules. Curr Med Chem 2011, 18:79–90.

Sivakumar et al. Biotechnology for Biofuels 2014, 7:62 Page 8 of 9 http://www.biotechnologyforbiofuels.com/content/7/1/62

34. Sivakumar G: Bioreactor technology: a novel industrial tool for high-tech production of bioactive molecules and biopharmaceuticals from plant roots. Biotechnol J 2006, 1:1419–1427.

35. Murashige T, Skoog F: A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol Plant 1962, 15:23–28.

36. Nichols HW, Bold HC: Trichosarcina polymorpha gen. et sp. nov. J Phycol 1965, 1:34–38.

37. Sivakumar G, Liu C, Towler MJ, Weathers PJ: Biomass production of hairy roots of Artemisia annua and Arachis hypogea in a scaled-up mist bioreactor. Biotechnol Bioeng 2010, 107:802–813.

38. Jones J, Manning S, Montoya M, Keller K, Poenie M: Extraction of algal lipids and their analysis by HPLC and mass spectrometry. J Am Oil Chem Soc 2012, 89:1371–1381.

39. International AOAC: AOAC Official Method 996.06: Fat (Total, Saturated, and Unsaturated) in Foods. Hydrolytic Extraction Gas Chromatography Method. Gaithersburg, MD: AOAC International; 2001.

40. International AOAC: AOCS Official Method Ce 1 h-05: Determination of cis-, trans-, Saturated, Monounsaturated and Polyunsaturated Fatty Acids in Vegetable or Nonruminant Animal Oils and Fats by Capillary GLC. Gaithersburg, MD: AOAC International; 2005.

doi:10.1186/1754-6834-7-62 Cite this article as: Sivakumar et al.: Bioprocessing of Stichococcus bacillaris strain siva2011. Biotechnology for Biofuels 2014 7:62.

Submit your next manuscript to BioMed Central and take full advantage of:

• Convenient online submission

• Thorough peer review

• No space constraints or color figure charges

• Immediate publication on acceptance

• Inclusion in PubMed, CAS, Scopus and Google Scholar

• Research which is freely available for redistribution

Submit your manuscript at www.biomedcentral.com/submit

Sivakumar et al. Biotechnology for Biofuels 2014, 7:62 Page 9 of 9 http://www.biotechnologyforbiofuels.com/content/7/1/62

RESEARCH Open Access

Biomass and RRR-α-tocopherol production in Stichococcus bacillaris strain siva2011 in a balloon bioreactor Ganapathy Sivakumar1*, Kwangkook Jeong2 and Jackson O Lay Jr3

Abstract

Background: Green microalgae represent a renewable natural source of vitamin E. Its most bioactive form is the naturally occurring RRR-α-tocopherol which is biosynthesized in photosynthetic organisms as a single stereoisomer. It is noteworthy that the natural and synthetic α-tocopherols are different biomolecular entities. This article focuses on RRR-α-tocopherol production in Stichococcus bacillaris strain siva2011 biomass in a bioreactor culture with methyl jasmonate (MeJa) elicitor. Additionally, a nonlinear mathematical model was used to quantitatively scale-up and predict the biomass production in a 20 L balloon bioreactor with dual variables such as time and volume.

Results: Approximately 0.6 mg/g dry weight (DW) of RRR-α-tocopherol was enhanced in S. bacillaris strain siva2011 biomass with the MeJa 50 μL/L for 24 hrs elicitations when compared to the control. The R2 value from the nonlinear model was enhanced up to 95% when compared to the linear model which significantly improved the accuracy for estimating S. bacillaris strain siva2011 biomass production in a balloon bioreactor.

Conclusions: S. bacillaris strain siva2011 is a new green microalga which biosynthesizes significant amounts of RRR-α-tocopherol. Systematically validated dual variable empirical data should provide key insights to multivariable or fourth order modeling for algal biomass scale-up. This bioprocess engineering should provide valuable information for industrial production of RRR-α-tocopherol from green cells.

Keywords: Antioxidant, Biomass, Bioreactor, Microalgae, Vitamin E

Background RRR-α-tocopherol is a lipid soluble small molecule and the biologically active form of natural vitamin E. RRR-α- tocopherol is exclusively biosynthesized by photosyn- thetic organisms or green cells including algae, plants, and cyanobacteria [1-3]. Plant-based products are a pri- mary source of RRR-α-tocopherol in the human diet. For example, hazelnut is one of the richest sources of vitamin E [4]. It is known that vitamin E plays an im- portant role in human nutrition as a natural antioxidant. It was recently proposed that vitamin E is active against oxidative stress-related diseases [5]. Reportedly, it sup- presses telomerase activity in ovarian cancer cells [6]. Vitamin E enhances IL-2 production, gene expression, and is an effective therapeutic adjuvant [7]. Vitamin E

deficiency affects both T and B immune cell functions [8], the α-tocopherol transfer protein (α-TTP) gene, and neurologic dysfunctions. In animal models, vitamin E mixtures inhibit colon, lung, mammary, and prostate carcinogenesis [9], as well as prevent diabetes [10]. Cos- metic industries also extensively use vitamin E in skin care products. In addition, RRR-α-tocopherol prolongs the shelf life of meat [11]. Thus, the use of RRR-α-tocopherol continues to increase in nutraceuticals [12]. Bioreactor technology is the key component for the

industrial scale production of bioactive small molecules for pharmaceutical applications. For instance, bioreactor technology was successfully developed and scaled-up to 10,000 L for commercial-scale production of ginseng roots used for human health-related applications [13,14]. This reactor configuration has also been tested for RRR-α- tocopherol production in lab-scale photosynthetic hazel- nut root culture [15]. However, knowledge regarding the bioprocessing of green algal cells for production of RRR-

© 2014 Sivakumar et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Sivakumar et al. Microbial Cell Factories 2014, 13:79 http://www.microbialcellfactories.com/content/13/1/79

mailto:[email protected]

http://creativecommons.org/licenses/by/4.0

http://creativecommons.org/publicdomain/zero/1.0/

α-tocopherol in balloon bioreactors is limited. Previously, Stichococcus bacillaris strain siva2011 biomass was scaled- up in a lab-scale balloon bioreactor (4 L - 8 L), and a lin- ear fitting model for predicting scale-up was proposed [16]. The S. bacillaris strain siva2011 has unique lipid (Figure 1) [16] and vitamin E metabolisms to lead to bio- active RRR-α-tocopherol production. A nonlinear model enables more accurate estimates and should provide insight for quantitatively predicating algal biomass accu- mulation in a large volume bioreactor. The objectives of this study were to: 1) evaluate RRR-α-tocopherol content from S. bacillaris strain siva2011 biomass with MeJa elicit- ation; 2) quantitatively predict S. bacillaris strain siva2011 DW in a 20 L bioreactor with simplified stepwise dual var- iables (time and volume) from 4 L and 8 L data using non- linear regression. This systematic study could provide insight regarding stepwise nonlinear scale-up of S. bacil- laris strain siva2011 biomass for RRR-α-tocopherol pro- duction in a balloon bioreactor.

Results and discussions Natural vitamin E occurs in two general forms, namely the tocopherols and tocotrienols, which are collectively called tocochromanols. Each has four distinct isoforms having α, β, γ, or δ substitution on the chromanol. The natural α-tocopherol contains three methyl groups on the chromanol moiety at positions 5, 7, and 8. The phy- tyl or saturated side chain is attached to the C-2 position of the chromanol ring which has three chiral centers with a single RRR stereoisomer (Figure 2). The chroma- nol groups have two fused rings, a phenol and a tetrahy- dropyran, sharing a 2 carbon bridge [17]. These rings are moderately polar, giving them an affinity for the cel- lular membrane surface while the phytyl tail is hydro- phobic and normally associated with membrane lipids [18]. These structural features of RRR-α-tocopherol are efficiently acted on by the human hepatic α-TTP which is responsible for maintaining plasma α-tocopherol con- centrations [19].

C 16

C 18

Figure 1 Gas chromatography mass spectrometry profile of fatty acid methyl esters from S. bacillaris strain siva2011, biomass from 20 L bioreactor, working volume 8 L with 0.2% CO2 on day 6. Methyl palmitate (C16:0), methyl hexadecatrienoate (C16:3), methyl oleate (C18:1), methyl linoleate (C18:2), and methyl linolenate (C18:3).

Sivakumar et al. Microbial Cell Factories 2014, 13:79 Page 2 of 8 http://www.microbialcellfactories.com/content/13/1/79

The synthetic α-tocopherol called all-racemic-α-tocopherol is not identical to RRR-α-tocopherol. It is an equimolar mixture of eight stereoisomers which possess three chiral centers at positions 2’, 4’, and 8’, giving rise to four diastereoisomeric pairs of enantiomers such as RRR, RSR, RRS, RSS, SRR, SSR, SRS, and SSS [20]. Moreover, α-TTP has a high affinity to RRR-α-tocopherol and has a 3-fold greater binding half-life when compared to syn- thetic α-tocopherol [21]. Thus, the bioactivity and the relative safety are different. Human proteins such as en- zymes and receptors usually exhibit high stereospecificity [20]. Therefore, the natural RRR-α-tocopherol is more bio- active than the synthetic form. RRR-α-tocopherol plays a major role as an antioxidant

which prevents lipid peroxidation. In the photosynthetic cells, it may protect photosystem II during photoinhibi- tion and repair chloroplast mechanisms [18]. This is due to the hydroxyl group on the C-6 position which is the active site that donates a hydrogen atom. The phenolic hydrogen atom is capable of scavenging lipid peroxy radi- cals and quenching singlet oxygen [22]. RRR-α-tocopherol is recycled in the photosynthetic cell by cytosolic ascor- bate which oxidizes one-electron from the tocopheroxyl radical thus regenerating vitamin E [23]. This mechanism might protect the cell membranes. RRR-α-tocopherol is biosynthesized in photosynthetic

cells via two different pathways [24]. The phytyl domain precursor comes from an isoprenoid pathway, and the chromanol domain precursor comes from an alternative shikimate pathway homogentisic acid via complex en- zymatic reactions [25]. RRR-α-tocopherol is found in the chloroplast envelope, thylakoids, and plastoglobuli of the plastid. The vitamin E biosynthetic pathway has been elucidated in Arabidopsis [26]. The genes associated with vitamin E biosynthesis in photosynthetic organisms have been well described in literature [3,27]. A signifi- cant metabolic engineering effort has been made to im- prove vitamin E content both in plants [28] and in

cyanobacteria [29,30]. Moreover, tocopherol production in plant green callus [31], cell [32], and root cultures [15] have also been reported.

RRR-α-tocopherol production in S. bacillaris strain siva2011 Photosynthetic algae are a potential alternative for pro- duction because they biosynthesize an abundance of RRR-α-tocopherol. For instance, freshwater Euglena gra- cilis [33], marine Dunaliella tertiolecta and Tetraselmis suecica [34], model system Chlamydomonas reinhardtii [35], and commercial algae Spirulina platensis [36] have been used to biosynthesize RRR-α-tocopherol. The new green alga, S. bacillaris strain siva2011 [16], produces significant amounts of RRR-α-tocopherol. This microalga has efficient photosynthetic mechanisms which facilitate the quick biosynthesis of vitamin E. Tocopherol production can be enhanced by molecular

elicitation which is non-transgenic. MeJa is a plant stress volatile signaling molecule which up-regulates several defense-related genes [37]. In plants, jasmonates are bio- synthesized via the octadecanoid pathway; exogenous MeJa treatment is up-regulating secondary metabolic pathway genes especially those encoding for stress pro- tection [38]. Therefore, it is used as one of the potential molecular elicitors in plant root culture to enhance sev- eral pharmaceutical molecule productions [15]. For in- stance, MeJa elicitation increases the activity of tyrosine aminotransferase in plant green cell culture which is one of the initial step enzymes involved in tocopherol bio- synthesis [31,39]. To increase RRR-α-tocopherol content in S. bacillaris strain siva2011, MeJa elicitation was also used. The S. bacillaris strain siva2011 characteristics, cul- ture conditions, and bioreactor experimental designs were previously reported for lipid production [16]. Figure 3 illustrates the RRR-α-tocopherol content of S.

bacillaris strain siva2011 biomass under elicitation with various concentrations of MeJa. The unelicited culture

C H 3

CH 3

C H 3

C H 3CH

3 CH

C H 3

(R) (R) (R) 8’4’

2 1

3 45

6 7

Chromanol nucleus

Phytyl chain

Figure 2 Chemical structure of RRR-α-tocopherol.

Sivakumar et al. Microbial Cell Factories 2014, 13:79 Page 3 of 8 http://www.microbialcellfactories.com/content/13/1/79

accumulated 0.7 mg/g DW of RRR-α-tocopherol which was detected starting during the early exponential growth phase. The lower concentration of MeJa, 50 μL/L, after 24 hrs elicitation, enhanced the production of RRR-α- tocopherol to the highest concentration, 1.3 mg/g DW. The higher MeJa concentrations or longer elicitation pe- riods were inhibiting both to the biomass growth and the resultant RRR-α-tocopherol production. MeJa can diffuse to cells either by intercellular migration or while in the vapor phase [40]. In plants, MeJa can transport from leaves to roots [41]. The vapor signaling can be trans- ported to distal plants via air, and the intercellular sig- naling can be transported via vascular process [42]. For instance, in an in vitro root culture the MeJa elicitation could trigger the defensive molecules accumulation via intercellular transport [13-15], whereas in in vivo plants during herbivore attack it can act as a volatile signal [42]. In the algal culture, MeJa elicitation could trigger the RRR-α-tocopherol production by either intercellular signaling or both. The optimum concentration can up- regulate the tocopherol biosynthetic pathway enzymes in S. bacillaris strain siva2011 which could increase the anti- oxidant. MeJa elicitation rapidly activates the defensive genes which also down regulates the photosynthetic sys- tem genes [38]. In addition, MeJa induces reactive oxygen species (ROS) which alter the mitochondrial and chloro- plast dynamics [43]. Thus, the higher MeJa concentrations or longer elicitation periods can produce uncontrolled ROS which can precede chloroplast or photosynthetic dysfunctions which could be inhibiting the tocopherol biosynthesis and biomass accumulation in S. bacillaris strain siva2011. In green cells, chloroplasts are an essential organelle for energy capture and transduction; a decline in photosynthetic activity is closely related to the decrease in the biomass. The typical MeJa elicited cells’ symptoms

were loss of chlorophyll, which causes the decline in the net photosynthetic rate, and degradation of ribulose bisphosphate carboxylase, etc. [44]. For in- stance, 100 μL MeJa at 9 hrs elicitation altered chloroplast morphology and function which is associated with cell death [43]. Even though RRR-α-tocopherol indirectly reg- ulates the amounts of jasmonic acid [18], the decline in chloroplast efficiency could down regulate the RRR-α-toc- opherol metabolism. This suggests that higher MeJa con- centrations or longer elicitation could be cytotoxic beyond what was studied.

Nonlinear regression Compared to green adventitious roots, photosynthetic algae have a higher ability to biosynthesize RRR-α-tocopherol in bioreactor cultures. The balloon bioreactor is a liquid-phase reactor with enhanced geometry and efficient fluid flow dynamics which could help provide higher mass transfer efficiency [14]. When compared to other photobioreac- tors, the balloon bioreactor had a larger headspace which efficiently captured light and enhanced photosynthesis [16]. Thus, S. bacillaris strain siva2011 was scaled-up in a balloon bioreactor (Figure 4) to investigate enhanced bio- mass accumulation. Of the four concentrations of CO2 tested, 0.2% yielded the highest biomass of 3.45 g/L in 4 L and 3.79 g/L (DW) in 8 L on day 6 [16]. The RRR-α-toc- opherol production was unchanged by the 0.2% CO2 (data not shown). To quantitatively predict larger-scale algal biomass

production based on lab-scale test data, the following stepwise structured approach was proposed [45,46]: 1) to set up the simplest model to linearize with dual vari- ables; 2) to model a nonlinear regression with dual vari- ables; and 3) to demonstrate multiple variables based on a nonlinear regression. Although scale-up predication

Figure 3 RRR-α-tocopherol content in S. bacillaris strain siva2011, unelicited and methyl jasmonate elicited culture on day 4, 5, and 6.

Sivakumar et al. Microbial Cell Factories 2014, 13:79 Page 4 of 8 http://www.microbialcellfactories.com/content/13/1/79

requires multiple variables, the 6 days algal culture does not significantly utilize all the media components. There- fore, selection of important dual variables can give insight on the efficiency of parameter selection for lab-scale valid- ation and initial scale-up prediction. In addition, the mul- tiple linear regression models might not incorporate the underlying nonlinear relationships [47]. So, in this study the second systematic approach was conducted to evalu- ate nonlinear modeling for scale-up prediction in a 20 L

balloon reactor using dual variables. Nonlinear regres- sion models are more important tools than linear models because they provide better parsimony, inter- pretability, and prediction [48]. Figure 5 illustrates non- linear modeling of predicted S. bacillaris strain siva2011 biomass accumulation in a 20 L balloon bioreactor with 0.2% CO2. This model was generated using 4 L to 8 L data of S. bacillaris strain siva2011 and shows only small discrepancies between measured and predicted data. The

Figure 4 Biomass production of S. bacillaris strain siva2011 in 5 L balloon type bioreactors (working volume 4 L).

Figure 5 Nonlinear modeling for nondimensionalized dry weight from 4 L to 20 L.

Sivakumar et al. Microbial Cell Factories 2014, 13:79 Page 5 of 8 http://www.microbialcellfactories.com/content/13/1/79

nondimensional DW* can be converted into dimensional DW in g/L by multiplying by the maximum DW experi- mentally obtained from the baseline test which in this case is 4 L. The nonlinear modeling agrees with measured data both qualitatively and quantitatively, where the modeling has enhanced the R2 value up to 95% compared to the lin- ear model value 87.4% [16]. This suggests that the nonlin- ear regression approach enhances accuracy of modeling which provides key scale-up informations of S. bacillaris strain siva2011 biomass for RRR-α-tocopherol production. In this study, the significance of this empirical approach provides insights into the application of a nonlinear re- gression model that increases the R2 value and enhances the quantitative predication of S. bacillaris strain siva2011 scaled-up in a 20 L bioreactor. This allows for the sys- tematic understanding and design of a multi-variable nonlinear regression experiment for significant biomass production.

Conclusions Photosynthetic microalgae are rich in RRR-α-tocopherol and a potential source for this natural antioxidant which is an essential human micronutrient. A significant ad- vantage of this natural source is the maintenance of the specific bioactive form needed for nutrition and the elimination of possible issues with potential unknown or unexpected toxicities from the synthetic conforma- tions. S. bacillaris strain siva2011 has a unique vitamin E biosynthetic mechanism capable of sustaining high levels of production, including inducible enhanced pro- duction which could provide a possible production plat- form of RRR-α-tocopherol for pharmaceutical industries. A nonlinear mathematical model was developed to model scale-up to production in 20 L reactors using an approach with a higher accuracy based on the dual variables tested in 4 L and 8 L reactors. The R2 value from this study dem- onstrates that this nonlinear approach significantly im- proves estimation of S. bacillaris strain siva2011 biomass production in the bioreactor than does the linear model. Additional studies with progressively larger reactors (and models) will be needed to bridge the gap between labora- tory and industrial scale. Nevertheless, this data provides enhanced bioprocess engineering information in the pro- gression towards large-scale pharmaceutical RRR-α-tocoph- erol production from S. bacillaris strain siva2011 biomass.

Methods Bioreactor culture, elicitation and analytics S. bacillaris strain siva2011 cells were cultured in a balloon type bioreactor (4 L and 8 L). The bioreactor experimen- tal design and the biomass harvesting were performed as described by Sivakumar et al. [16]. For elicitation, three concentrations of filter sterilized MeJa 50, 100, and 200 μL/L were added to the S. bacillaris strain

siva2011 culture on the 3rd day. MeJa dissolved in etha- nol and MeJa not dissolved in ethanol were tested, and both had a similar effect. Elicitations were carried out for 24, 48, or 72 hrs. Algal cells were harvested and freeze-dried according to the Sivakumar et al. [16] method. One gram of MeJa elicited and unelicited freeze-dried algal cells were used for analysis of RRR-α- tocopherol. RRR-α-tocopherol was processed according to the Sivakumar et al. [4] method. The reversed phase- high performance liquid chromatography chromato- gram was acquired using a Dionex 3000 (LPG3400A) system (Thermo Scientific, Sunnyvale, CA) equipped with a thermostatted UltiMate 3000 autosampler and a Dionex RF 2000 fluorescence detector. The system was monitored by the software’s chromeleon (6.80) for in- strument control and data acquisition, data reproces- sing, and solute quantification, respectively. An Agilent Zorbax Eclipse XDB-C8 (250 mm × 4.6 mm, mean par- ticle size 5 μm) column or C18 (250 mm × 4.6 mm, mean particle size 5 μm) was used to separate RRR-α- tocopherol. The mobile phase consists of a linear gradi- ent of 90% methanol in water. The flow rate was 1 ml/ min. The total acquisition time was 35 min. The wave- length was set at 290 nm for excitation and 330 nm for emission. The authenticated RRR-α-tocopherol fluores- cence spectra and retention time was used for HPLC confirmation of RRR-α-tocopherol in the samples. The RRR-α-tocopherol liquid chromatography mass spec- trometry spectrum confirmation was acquired using a Shimadzu 8050 mass spectrometer.

Nonlinear regression For a stepwise approach, the first phase of the nonlinear regression method begins with the simplification from five variables to two variables, time (t) and reactor vol- ume (V), as shown in equation 1. The regression will be extended to multivariables when the two variables method is validated.

DW ¼ f yCO2; EC; OrP; pH; t; Vð Þ≈f t; Vð Þ ð1Þ

The associated variables DW, t, and V in equation 1 are nondimensionalized into DW*, t*, and V* as shown in equation (2) to (4).

DW$ ¼ DW

DW max ð2Þ

t$ ¼ t

tmax ð3Þ

V $ ¼ V L½ & 4L

ð4Þ

Where DWmax is the maximum DW [g/L] produced from 4 L test under 0.02% CO2 fraction, and tmax is the

Sivakumar et al. Microbial Cell Factories 2014, 13:79 Page 6 of 8 http://www.microbialcellfactories.com/content/13/1/79

maximum time to reach the maximum DW. In this study, DWmax and tmax were 3.45 g/L and 6 days, respect- ively. The volume was standardized by the baseline: 4 L in this modeling. A nonlinear model was assumed by combining 4th

order polynomials and the power form of the equation as shown in equation (5) in consideration of measured data of 4 L and 8 L.

DW$ ¼ at$4 þ bt$3 þ ct$2 þ dt$2 þ et þ f ! "

V $g ð5Þ

All associated constants a, b, c, and d in equation (5) were determined from a nonlinear regression method as shown in equation (6).

DW$ ¼ # 2:686t$4−6:942t$3 þ 5:109t$2

þ 0:127t þ 0:0184 $ V $0:151 ð6Þ

Equation (6) allows the predicting of DW in reactor volume 20 L at 0.02% CO2 fraction from the correlation obtained from 4 L and 8 L measured data.

Statistical analysis Bioreactor culture, elicitations, and analytical experiments were repeated at least three times, each with three replica- tions. Results were expressed as the mean with standard errors. Stepwise dual nonlinear regressions were used to investigate the relationship between 4 L and 8 L in order to predict 20 L bioreactor scale-up.

Abbreviations C: Carbon; CO2: Carbon dioxide; DW: Dry weight; G: Gram; hrs: Hours; α: Trimethyl; β, γ: Dimethyl; δ: Monomethyl; α-TTP: α-Tocopherol transfer protein; MeJa: Methyl jasmonate; Mg: Milligram; μL: Microliter; L: Liter; %: Percentage; ROS: Reactive oxygen species.

Competing interests The authors declare that they have no competing interests.

Authors’ contributions GS, JL, and KJ made substantial contributions to the experimental design, analysis, and/or interpretation of data. Specifically, GS led and designed the experiments and performed the bioprocessing and HPLC studies. JL performed the mass spectrometry experiments for compound confirmation. KJ performed the mathematical modeling to validate the variables. GS wrote the manuscript which was reviewed and approved by all authors. All authors read and approved the final manuscript.

Acknowledgments This research was funded by the Arkansas Biosciences Institute (grants 262178 and 200109). The technical assistance of Kelsea Brewer and Dez Jones (Arkansas State University, Jonesboro, AR, USA), and Dr. Jennifer Gidden (University of Arkansas, Fayetteville, AR, USA) was appreciated.

Received: 11 April 2014 Accepted: 30 May 2014 Published: 3 June 2014

References 1. Dasilva EJ, Jensen A: Content of α-tocopherol in some blue-green algae.

Biochim Biophys Acta 1971, 239:345–347. 2. Sivakumar G, Bacchetta L: Determination of natural vitamin E from Italian

hazelnut leaves. Chem Nat Compd 2005, 6:654–656. 3. Maeda H, DellaPenna D: Tocopherol functions in photosynthetic

organisms. Curr Opin Plant Biol 2007, 10:260–265. 4. Sivakumar G, Bacchetta L, Gatti R, Zappa G: HPLC screening of natural

vitamin E from Mediterranean plant biofactories – a basic tool for pilot-scale bioreactors production of α-tocopherol. J Plant Physiol 2005, 162:1280–1283.

5. Niki E: Role of vitamin E as lipid-soluble peroxyl radical scavenger: in vitro and in vivo evidence. Free Radic Biol Med 2013, 66:3–12.

6. Bermudez Y, Ahmadi S, Lowell NE, Kruk PA: Vitamin E suppresses telomerase activity in ovarian cancer cells. Cancer Detect Prev 2007, 31:119–128.

7. Molano A, Meydani SK: Vitamin E, signalosomes and gene expression in T cells. Mol Aspects Med 2012, 33:55–62.

8. Pekmezci D: Vitamin E and immunity. Vitam Horm 2011, 86:179–215. 9. Yang CS, Li G, Yang Z, Guan F, Chen A, Ju J: Cancer prevention by

tocopherols and tea polyphenols. Cancer Lett 2013, 334:79–85. 10. Kamimura W, Doi W, Takemoto K, Ishihara K, Wang DH, Sugiyama H, Oda S,

Masuoka N: Effect of vitamin E on alloxan-induced mouse diabetes. Clin Biochem 2013, 46:795–798.

11. Röhrle FT, Moloney AP, Black A, Osorio MT, Sweeney T, Schmidt O, Monahan FJ: α-Tocopherol stereoisomers in beef as an indicator of vitamin E supplementation in cattle diets. Food Chem 2011, 124:935–940.

12. Ajjawi I, Shintani D: Engineered plants with elevated vitamin E: a nutraceutical success story. Trends Biotechnol 2004, 22:104–107.

13. Sivakumar G, Yu KW, Paek KY: Production of biomass and ginsenosides from adventitious roots of Panax ginseng in bioreactor cultures. Eng Life Sci 2005, 5:333–342.

14. Sivakumar G, Medina-Bolivar F, Lay JO, Dolan MC, Condori J, Grubbs SK, Wright SM, Baque MA, Lee EJ, Paek KY: Bioprocess and bioreactor: next generation technology for production of potential plant-based antidiabetic and antioxidant molecules. Curr Med Chem 2011, 18:79–90.

15. Sivakumar G: Bioreactor technology: a novel industrial tool for high-tech production of bioactive molecules and biopharmaceuticals from plant roots. Biotechnol J 2006, 1:1419–1427.

16. Sivakumar G, Jeong K, Lay JO: Bioprocessing of Stichococcus bacillaris strain siva2011. Biotechnol Biofuels 2014, 7:62.

17. Zingg ZM: Vitamin E: an overview of major research directions. Mol Aspects Med 2007, 28:400–422.

18. Munne-Bosch S: α-Tocopherol: a multifaceted molecule in plants. Vitam Horm 2007, 76:375–392.

19. Liu K, Luo HL, Yue DB, Ge SY, Yuan F, Yan LY, Jia HN: Molecular cloning and characterization of the sheep α-TTP gene and its expression in response to different vitamin E status. Gene 2011, 494:225–230.

20. Jensen SK, Lauridsen C: Alpha-tocopherol stereoisomers. Vitam Horm 2007, 76:281–308.

21. Traber MG, Ramakrishnan R, Kayden HJ: Human plasma vitamin E kinetics demonstrate rapid recycling of plasma RRR-α-tocopherol. Proc Natl Acad Sci U S A 1994, 91:10005–10008.

22. Traber MG, Atkinson J: Vitamin E, antioxidant and nothing more. Free Radic Biol Med 2007, 43:4–15.

23. Nowicka B, Kruk J: Plastoquinol is more active than α-tocopherol in singlet oxygen scavenging during high light stress of Chlamydomonas reinhardtii. Biochim Biophys Acta 1817, 2012:389–394.

24. Sarah C, Hunter SC, Cahoon EB: Enhancing vitamin E in oilseeds: unraveling tocopherol and tocotrienol biosynthesis. Lipids 2007, 42:97–108.

25. Porfirova S, Bergmuller E, Tropf S, Lemke R, Dormann P: Isolation of an Arabidopsis mutant lacking vitamin E and identification of a cyclase essential for all tocopherol biosynthesis. Proc Natl Acad Sci U S A 2002, 99:12495–12500.

26. Tsegaye Y, Shintani DK, DellaPenna D: Overexpression of the enzyme p-hydroxyphenylpyruvate dioxygenase in Arabidopsis and its relation to tocopherol biosynthesis. Plant Physiol Biochem 2002, 40:913–920.

Sivakumar et al. Microbial Cell Factories 2014, 13:79 Page 7 of 8 http://www.microbialcellfactories.com/content/13/1/79

27. DellaPenna D: Progress in the dissection and manipulation of vitamin E synthesis. Trends Plant Sci 2005, 10:574–579.

28. Cahoon EB, Hall SE, Ripp KG, Ganzke TS, Hitz WD, Coughlan SJ: Metabolic redesign of vitamin E biosynthesis in plants for tocotrienol production and increased antioxidant content. Nat Biotechnol 2003, 21:1082–1087.

29. Sattler SE, Cahoon EB, Coughlan SJ, DellaPenna D: Characterization of tocopherol cyclases from higher plants and cyanobacteria. Evolutionary implications for tocopherol synthesis and function. Plant Physiol 2003, 132:2184–2195.

30. Backasch N, Schulz-Friedrich R, Appel J: Influences on tocopherol biosynthesis in the cyanobacterium Synechocystis sp. PCC 6803. J Plant Physiol 2005, 162:758–766.

31. Antognoni F, Faudale M, Poli F, Biondi S: Methyl jasmonate differentially affects tocopherol content and tyrosine amino transferase activity in cultured cells of Amaranthus caudatus and Chenopodium quinoa. Plant Biol (Stuttg) 2009, 11:161–169.

32. Caretto S, Nisi R, Paradiso A, De Gara L: Tocopherol production in plant cell cultures. Mol Nutr Food Res 2010, 54:726–730.

33. Kusmic C, Barsacchi R, Barsanti L, Gualtieri P, Passarelli V: Euglena gracilis as source of the antioxidant vitamin E. Effects of culture conditions in the wild strain and in the natural mutant WZSL. J Appl Psychol 1999, 10:555–559.

34. Carballo-Cardenas EC, Tuan PM, Janssen M, Wijff RH: Vitamin E (α-tocopherol) production by the marine microalgae Dunaliella tertiolecta and Tetraselmis suecica in batch cultivation. Biomol Eng 2003, 20:139–147.

35. Kruk J, Szymanska R, Krupinska K: Tocopherol quinone content of green algae and higher plants revised by a new high-sensitive fluorescence detection method using HPLC–Effects of high light stress and senescence. J Plant Physiol 2008, 165:1238–1247.

36. Mendiola JA, Garcıa-Martınez D, Ruperez FJ, Martın-Alvarez PJ, Reglero G, Cifuentes A, Barbas C, Ibanez E, Senorans FJ: Enrichment of vitamin E from Spirulina platensis microalga by SFE. J Supercrit Fluids 2008, 43:484–489.

37. Turner JG, Ellis C, Devoto A: The jasmonate signal pathway. Plant Cell 2002, 14:153–164.

38. Cheong JJ, Choi YD: Methyl jasmonate as a vital substance in plants. Trends Genet 2003, 19:409–413.

39. Sandorf I, Holländer-Czytko H: Jasmonate is involved in the induction of tyrosine aminotransferase and tocopherol biosynthesis in Arabidopsis thaliana. Planta 2002, 216:173–179.

40. Seo HS, Song JT, Cheong JJ, Lee YH, Lee YW, Hwang I, Lee JS, Choi YD: Jasmonic acid carboxyl methyltransferase: a key enzyme for jasmonate-regulated plant responses. Proc Natl Acad Sci U S A 2001, 98:4788–4793.

41. Zhang ZP, Baldwin IT: Transport of [2-14C]jasmonic acid from leaves to roots mimics wound-induced changes in endogenous jasmonic acid pools in Nicotiana sylvestris. Planta 1997, 203:436–441.

42. Tamogami S, Noge K, Abe M, Agrawal GK, Rakwal R: Methyl jasmonate is transported to distal leaves via vascular process metabolizing itself into JA-Ile and triggering VOCs emission as defensive metabolites. Plant Signal Behav 2012, 7:1–4.

43. Zhang L, Xing D: Methyl jasmonate induces production of reactive oxygen species and alterations in mitochondrial dynamics that precede photosynthetic dysfunction and subsequent cell death. Plant Cell Physiol 2008, 49:1092–1111.

44. Wierstra I, Kloppstech K: Differential effects of methyl jasmonate on the expression of the early light-inducible proteins and other light-regulated genes in barley. Plant Physiol 2000, 124:833–844.

45. Bates D, Watts D: Nonlinear regression analysis and its applications. NY, USA: Wiley-Interscience; 2007.

46. Liu S: Bioprocess engineering: kinetics, biosystems, sustainability, and reactor design. Oxford, UK: Elsevier; 2012.

47. Hassan SS, Farhan M, Mangayil R, Huttunen H, Aho T: Bioprocess data mining using regularized regression and random forests. BMC Syst Biol 2013, 7:S5.

48. Archontoulis SV, Miguez FE: Non-linear regression models and applications in agricultural research. Agronomy J 2013, 105:1–13.

doi:10.1186/1475-2859-13-79 Cite this article as: Sivakumar et al.: Biomass and RRR-α-tocopherol production in Stichococcus bacillaris strain siva2011 in a balloon bioreactor. Microbial Cell Factories 2014 13:79.

Submit your next manuscript to BioMed Central and take full advantage of:

• Convenient online submission

• Thorough peer review

• No space constraints or color figure charges

• Immediate publication on acceptance

• Inclusion in PubMed, CAS, Scopus and Google Scholar

• Research which is freely available for redistribution

Submit your manuscript at www.biomedcentral.com/submit

Sivakumar et al. Microbial Cell Factories 2014, 13:79 Page 8 of 8 http://www.microbialcellfactories.com/content/13/1/79

Homework 7 Due 3.21.16