Medium engineering for enhanced production of undecylprodigiosin antibiotic in Streptomyces coelicolor using oil palm biomass hydrolysate as a carbon source

Accepted Manuscript Medium engineering for enhanced production of undecylprodigiosin antibiotic in Streptomyces coelicolor using oil palm biomass hydrolysate as a carbon source Shashi Kant Bhatia, Bo-Rahm Lee, Ganesan Sathiyanarayanan, Hun-Seok Song, Junyoung Kim, Jong-Min Jeon, Jung-Ho Kim, Sung-Hee Park, Ju-Hyun Yu, Kyungmoon Park, Yung-Hun Yang PII: DOI: Reference:

S0960-8524(16)30188-2 http://dx.doi.org/10.1016/j.biortech.2016.02.055 BITE 16105

To appear in:

Bioresource Technology

Received Date: Revised Date: Accepted Date:

15 December 2015 1 February 2016 16 February 2016

Please cite this article as: Bhatia, S.K., Lee, B-R., Sathiyanarayanan, G., Song, H-S., Kim, J., Jeon, J-M., Kim, JH., Park, S-H., Yu, J-H., Park, K., Yang, Y-H., Medium engineering for enhanced production of undecylprodigiosin antibiotic in Streptomyces coelicolor using oil palm biomass hydrolysate as a carbon source, Bioresource Technology (2016), doi: http://dx.doi.org/10.1016/j.biortech.2016.02.055

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 Medium engineering for enhanced production of undecylprodigiosin antibiotic in Streptomyces coelicolor using Oil palm biomass hydrolysate as a carbon source Shashi Kant Bhatiaa, Bo-Rahm Leea, Ganesan Sathiyanarayanana, Hun-Seok Songa, Junyoung Kima, Jong-Min Jeona, Jung-Ho Kima, Sung-Hee Parkb, Ju-Hyun Yuc, Kyungmoon Parkd, YungHun Yanga,d,e,f*

a) Department of Microbial Engineering, College of Engineering, Konkuk University, Seoul, 143701, South Korea b) Food Ingredients Center, Foods R&D, CheilJedang, Guro-dong, Guro-Gu, Seoul 152-051, South Korea c) Center for Industrial Chemical Biotechnology, Ulsan Chemical R&BD Division, Korea Research Institute of Chemical Technology, P.O. Box 107, 141 Gajeong-ro, Yuseong-gu, Daejeon 305-600, South Korea d) Department of Biological and Chemical Engineering, Hongik University, Sejong Ro 2639, Jochiwon, Sejong city, 339-701, South Korea e) Microbial Carbohydrate Resource Bank, Konkuk University, Seoul 143-701, South Korea, f) Institute for Ubiquitous Information Technology and Applications (CBRU), Konkuk University, Seoul 143-701, South Korea.

*Author for correspondence (Fax: +82-2-3437-8360; E-mail: [email protected])

2 Abstract In this study, a biosugar obtained from empty fruit bunch (EFB) of oil palm by hot water treatment and subsequent enzymatic saccharification was used for undecylprodigiosin production, using Streptomyces coelicolor. Furfural is a major inhibitor present in EFB hydrolysate (EFBH), having a minimum inhibitory concentration (MIC) of 1.9 mM, and it reduces utilization of glucose (27%), xylose (59%), inhibits mycelium formation, and affects antibiotic production. Interestingly, furfural was found to be a good activator of undecylprodigiosin production in S. coelicolor, which enhanced undecylprodigiosin production by up to 52%. Optimization by mixture analysis resulted in a synthetic medium containing glucose: furfural: ACN: DMSO (1%, 2 mM, 0.2% and 0.3%, respectively). Finally, S. coelicolor was cultured in a fermenter in minimal medium with EFBH as a carbon source and addition of the components described above. This yielded 4.2 µg/mg dcw undecylprodigiosin, which was 3.2 fold higher compared to that in un-optimized medium.

Keywords: Antibiotic; biosugar; furfural; Streptomyces coelicolor; undecylprodigiosin.

3 1. Introduction Utilization of biomass for economical production of commercialized products is a subject of debate amongst the global scientific community (Pandey et al., 2000). Lignocellulosic raw materials are projected to play a major role in the development of various industrial bioprocesses (Patel et al., 2015). These raw materials are usually composed of cellulose, hemicellulose and lignin, which are associated with each other (Ahn et al., 2012; Jonsson et al., 2013). To use lignocellulose as a raw material, it must be converted to fermentable sugars by physio-chemical pretreatment and enzymatic hydrolysis. Lignin is a complex structure of polyalcohols, and is highly resistant to degradation by microbial enzymes (Arantes & Saddler, 2010). Pretreatment is necessary to improve enzyme accessibility and permeability to break down the complex cell wall, in which utilizable polysaccharides are embedded (Li et al., 2013). Various pretreatment methods using acids (such as sulfuric acid and phosphoric acid) or bases (such as ammonium hydroxide, sodium hydroxide, and calcium hydroxide) have been reported to promote conversion of the complex plant cell wall to fermentable sugars by hydrolytic enzymes (Kang et al., 2013; Sindhu et al., 2015). Lignocellulose pretreatment results in release of inhibitors; i.e., aldehydes (furan aldehydes), ketones, phenolics and organic acids, which influence microbial metabolism to utilize these free sugars as a carbon source for fermentation (Allen et al., 2010). Glucose and xylose degradation resulted in 5-hydroxymethylfurfural (HMF) and 2-furaldehyde (furfural), respectively (Kootstra et al., 2009). Furfural induces production of reactive oxygen species (ROS) in microorganisms, which cause cellular damage and affect growth (Allen et al., 2010). At present, little information about the microbial inhibitors released during pretreatment of lignocellulosic biomass, their mechanism of action, effects on carbon utilization and secondary metabolite

4 production is available. Hence, the generation of inhibitors during the production of lignocellulosic hydrolysate and their effects on microbial fermentation warrant investigation. Streptomyces coelicolor, an actinobacterium, produces various industrially important secondary metabolites (Borodina et al., 2008). S. coelicolor can synthesize two chemically distinct antibiotics as secondary metabolites—actinorhodin (a diffusible blue pigment) and the cell wallassociated, red-pigmented undecylprodigiosin (Kim et al., 2015a). Antibiotic production is controlled by many factors, such as transcriptional regulators, quorum-sensing molecules (γbutyrolactone) concentrations, metabolic and nutritional status (van Wezel et al., 2000; Yang et al., 2005), and the proposed coupling of antibiotic synthesis and resistance genes (Hindra et al., 2010). These regulatory mechanisms can be altered by varying the culture conditions and addition of various factors (Schaberle et al., 2014). Undecylprodigiosin is a focus of interest due to its immunosuppressive and anticancer properties (Williamson et al., 2006). In this study, the effects of biomass-derived inhibitors (furfural and hydroxymethylfurfural) on S. coelicolor growth, carbon source utilization and antibiotic production were investigated. Moreover, a minimal medium comprising EFB hydrolysate (EFBH) as a carbon source was designed to enhance production of undecylprodigiosin. 2. Materials and methods 2.1 Microorganisms, media and culture conditions All reagents were purchased from Difco Laboratories (Becton-Dickinson Franklin Lakes, NJ, USA) and other chemicals; i.e., furfural and hydroxymethylfurfural (HMF) from Sigma-Aldrich (St. Louis, MO, USA). Streptomyces coelicolor A3 (2) M145 used for antibiotic production was purchased from the Korean Culture Type Collection (KCTC, South Korea). For spore production S. coelicolor was cultivated on R5 agar plates for 7 days at 30°C, harvested by scraping and

5 suspended in 20% (v/v) glycerol and stored at –80°C. Seed culture of S. coelicolor was prepared by inoculating spores in 50 ml of Luria-Bertani (LB) liquid medium, with five 3 mm glass beads, and incubated at 30°C with shaking at 200 rpm. The germinated spores were harvested by centrifugation (3200 × g, 4°C, 10 min) and resuspended in 5 ml of ion-free water. A 0.1 ml volume (2 × 106 CFU) of germinated seed culture was used as the inoculum. For production of undecylprodigiosin, S. coelicolor was cultured in minimal medium (Difco Laboratories) containing EFB hydrolysate (EFBH) as a carbon source. EFB hydrolysate was prepared by hot water treatment of empty fruit bunch at 190°C for 15 min (Bench Top Reactor, Model 4526, Parr Instruments), mechanical refining of the pretreated slurry using a valley beater (L&W, Norway), enzymatic hydrolysis of the solid fraction of the pretreated EFB with Cellic CTec2 (Novozymes, Denmark), and finally solid–liquid separation of the hydrolysate by centrifugal filtration (Eom et al., 2015). An HPLC system equipped with a Bio-Rad Aminex HPX-87H column (Bio-Rad Co., Hercules, CA, USA) was used to analyze sugar, furfural and HMF contents. A mobile phase of 5 mM H2SO4 at a flow rate of 0.6 ml/min was used, and the column temperature was maintained at 50°C. The EFBH concentration was adjusted to 1% glucose, and additional furfural (2 mM), DMSO (0.3%), and acetonitrile (0.2%) were added. S. coelicolor was cultured in 10 ml volumes in the capped tubes with a capacity of 25 ml at 30°C, with agitation at 160 rpm for 72 h, and then subjected to analysis of antibiotic production. 2.2 Antibiotic extraction and quantification At intervals of 24 h, 2 ml culture samples were removed, and biomass and antibiotic production were estimated. The samples were divided into two aliquots. For actinorhodin estimation, an equal volume of 1 N NaOH was added to the culture aliquot. The sample was then centrifuged at 4000 g for 5 min and the actinorhodin concentration was determined by measuring the absorbance at 633

6 nm (Horinouchi & Beppu, 1984). Undecylprodigiosin is a membrane-associated red pigment, and was extracted from the cell pellet. An aliquot of S. coelicolor culture was harvested by centrifugation (4000 g for 5 min), suspended in methanol and incubated at 37°C with shaking at 200 rpm for 1 h. Cells were removed by centrifugation at 4000 g for 5 min and 0.1 N HCl was added to adjust pH. To quantify undecylprodigiosin, absorbance at 533 nm was measured and the concentration calculated (Horinouchi & Beppu, 1984). 2.3 Targeting sub-lethal concentrations of furfural and HMF EFB hydrolysate has furfural as the main inhibitory component, with trace amounts of HMF. The effect of these components on growth and antibiotic production by S. coelicolor was investigated. It is not possible to assess the effect of these inhibitors in complex EFBH, so a minimal medium containing glucose as carbon source was used. S. coelicolor was cultured in minimal medium in the presence of 1% glucose, 0.1% yeast extract and 0-4 mM inhibitors at 30°C for 72 h. The minimum inhibitory concentration (MICs) of the inhibitors against S. coelicolor were calculated. Antibiotic extraction was also performed to estimate actinorhodin and undecylprodigiosin concentrations. To make stock solutions, compounds were dissolved in water and sterile filtered prior to their addition to the fermentation media. 2.4 Effect of furfural on antibiotic production and carbon utilization EFBH contains glucose as the major sugar together with trace amounts of xylose, so the effect of furfural on the utilization of these sugars was investigated. S. coelicolor was cultured in minimal medium, supplemented with 10 g/l glucose or 10 g/l xylose or both, in the presence of furfural (3 mM). Samples were collected at 24 h intervals for 72 h and biomass, antibiotic production, carbon source and furfural utilization evaluated.

7 2.5 Effect of furfural 2.5.1 Field emission scanning electron microscopy (FESEM), total fatty acid profile and transcriptional analysis To assess the effect of furfural, S. coelicolor was cultured with and without furfural and morphological changes monitored using FESEM. Samples were prepared (Ishii et al., 2004) and FESEM was performed using a SUPRA 55VP FESEM (Carl Zeiss, Oberkochen, Germany) using a 15 kV accelerating voltage, and photographic images were captured digitally. Total fatty acid of S. coelicolor, cultured with and without furfural was extracted and analyzed for composition as described already (Bhatia et al., 2015a). Transcriptional analysis of antibiotic regulatory genes (redD and actII-orf4) was performed using mRNA extraction and RT-PCR (Kim et al., 2015b) 2.5.2 Fluorescence spectroscopy To assay reactive oxygen species (ROS), S. coelicolor cells grown in the presence of furfural for 72 h were subjected to 2',7'-dichlorodihydrofluorescein diacetate staining, as described previously with slight modification (Dwivedi et al., 2014). A stock solution of 2', 7'dichlorodihydrofluorescein diacetate (1 mg/ml) was prepared in DMSO and stored in the dark. S. coelicolor staining was performed by incubating 200 µl of culture with 2', 7'dichlorodihydrofluorescein diacetate (50 µg/ml) for 18 h at room temperature. Fluorescence spectroscopy was performed to determine ROS using a Perkin Elmer LS-55 fluorescence spectrometer with maximum excitation and emission spectra at 495 and 529 nm, respectively. 2.5.3 Effect of antioxidants and organic solvents on undecylprodigiosin Antioxidants; i.e., ascorbic acid, glycine betaine, dithiothreitol (DTT) and dimethyl sulphoxide (DMSO) were added (0.1%) to the culture medium to reduce the effects of ROS on S. coelicolor.

8 The effects of acetone, acetonitrile and methanol on undecylprodigiosin production was also investigated by adding (0.1%) in the culture media. 2.6 Mixture design for interaction effect of glucose: furfural: ACN: DMSO To assess the interaction effect and impact of various variables; i.e., glucose: furfural: ACN: DMSO on antibiotic production by S. coelicolor, a mixture analysis using Minitab 16 was performed. Mixture analysis experiments were designed using a simplex lattice method and 19 sets of experiments were performed. The experimental design with response is shown in Table 1. Experiments were performed in 10 ml volumes using glucose (20%), furfural (80 mM), acetonitrile (ACN; 20%) and DMSO (20%), at predetermined ratios in minimal medium. Regression analysis using ANOVA was performed and model fitting methods were applied for the data analysis. Contour surface plots were created to determine the interaction effects of the four components on biomass production, and actinorhodin and undecylprodigiosin synthesis. The combination of predictor settings that optimized the fitted response was used to verify the model. 2.7 Antibiotic production using EFB hydrolysate EFB hydrolysate (EFBH) was used as a carbon source in minimal medium to enhance undecylprodigiosin production by S. coelicolor. Medium composition was adjusted according to optimized synthetic media composition using EFBH as carbon source. S. coelicolor was cultured for 72 h and antibiotic extraction and estimation was performed as described above. Scale up of the undecylprodigiosin production process was also performed in 100 ml and 500 ml scale using a 250 ml conical flask and 1l capacity fermenter (Ferementec Co. Ltd.) respectively, with the optimum parameters of 10 ml scale.

9 3. Results and discussion 3.1 Effects of inhibitors on growth and antibiotic production of S. coelicolor Furfural and HMF were investigated in terms of their effects on biomass and antibiotic production of S. coelicolor. Biomass and actinorhodin production by S. coelicolor decreased continuously with increasing furfural concentration, while undecylprodigiosin increased (1.87 µg/mg dcw) up to 3 mM furfural, and decreased thereafter (Fig. 1a). The addition of 2 mM HMF resulted in an increase in biomass (5.0 mg dcw/ml) and undecylprodigiosin production (1.4 µg/mg dcw), but reduced actinorhodin production (Fig. 1b). Actinorhodin production in the presence of furfural and HMF decreased by 43% and 50%, respectively. The MICs for furfural and HMF were 1.9 and 2.3 mM, respectively. Thus furfural is more toxic to microbial fermentation than HMF (Heer & Sauer, 2008). 3.2 Effect of furfural on glucose and xylose utilization and antibiotic production EFB hydrolysate contains glucose and xylose, so the effect of furfural on their utilization was determined. Biomass production, glucose and xylose utilization and antibiotic production were monitored for 72 h. Furfural reduced biomass production (41%, Fig. 2a), glucose utilization (27% decrease, Fig. 2b) and actinorhodin production (41%, Fig. 2c), but increased undecylprodigiosin production (52%, Fig. 2d). Xylose utilization was also reduced (59%, Fig. S1a), with a 44% reduction in biomass (Fig. S1b), and 49% reduction in actinorhodin production (Fig. S1c). Furfural enhanced undecylprodigiosin production by 9% (Fig. S1d), when xylose was used as the carbon source. Furfural also effect glucose: xylose utilization, when used together as carbon source. S. coelicolor prefer to utilize glucose as carbon source along with lower preference towards xylose (Fig. S2a). Biomass was reduced by 21%, with a 50% reduction in actinorhodin and 22% in undecylprodigiosin production, respectively (Fig. S2b, S2c and S2d). Xylose utilization was

10 affected to a greater extent than glucose utilization. Metabolism was affected more severely during xylose consumption than glucose consumption, which resulted in lower productivity when xylose was used as the carbon source (Ask et al., 2013). The furfural level decreased by 40%, 59% and 63% in the presence of glucose, xylose and glucose: xylose, respectively, as carbon sources after 24 h. After 48 and 72 h no furfural was detected, as it may be consumed or degraded into less-toxic products (data not shown). In the presence of furfural, increase in undecylprodigiosin production was observed as compared to control due to change in morphology of cell and expression of antibiotic regulatory genes which is explained in further section. Undecylprodigiosin production is growth-associated, while actinorhodin production begins upon growth cessation (Hobbs et al., 1990). 3.3 Effects of furfural on morphology and fatty acid synthesis FESEM of S. coelicolor showed that cells grow normally and produce mycelia in the absence of furfural, while addition of furfural inhibited mycelium formation. Regarding fatty acid profiles, a 29% increase in 12-carbon fatty acids was observed together with a decrease in long-chain and total fatty acid (17%) accumulation in S. coelicolor cells cultured with furfural as compared to control (data not shown). Undecylprodigiosin synthesis begins from 12-carbon lipids (Craney et al., 2013), and furfural increased the accumulation of 12-carbon lipids in S. coelicolor. Furfural affects the morphology and total fatty acid profile of cells, which influenced antibiotic production by S. coelicolor. The antibiotic synthesis and fatty acid pathways are interrelated (Revill et al., 1996). Transcriptional analysis indicated that change in antibiotic production was due to increased expression of redD and decreased expression of actII-orf4 (Fig. S3). redD and actII-orf4 are transcriptional regulators of antibiotic production in S. coelicolor, and their expression levels affect undecylprodigiosin and actinorhodin production (Fujii et al., 1996; Wang et al., 2014).

11 3.4 Effects of antioxidants and organic solvents on undecylprodigiosin production The effect of furfural on ROS levels in S. coelicolor was investigated. ROS were observed in the presence of furfural (Fig. S4), which decreased biomass production and altered antibiotic production. Addition of ascorbic acid, glycine betaine and DTT did not significantly influence antibiotic production (Fig. 3a). The addition of DMSO and furfural resulted in a 33% increase in biomass and 27% increase in undecylprodigiosin production. DMSO elicits secondary metabolite production in S. coelicolor, as in Streptomyces venezuelae and S. glaucescens, in which it induces chloramphenicol and tetracenomycin C production, respectively (Chen et al., 2000; Pettit, 2011). The effects on undecylprodigiosin production of various organic solvents were investigated. Acetone and ethanol had no significant effect on secondary metabolite production (Fig. 3b). Acetonitrile (ACN) reduced actinorhodin production (35%) and increased undecylprodigiosin production (26%), but did not significantly affect biomass production (Fig. 3b). The effects of various organic solvents and detergents on production of streptomycin by Streptomyces griseus have been reported (El-Shahed et al., 2008). 3.5 Mixture design for an interaction effect of glucose: furfural: ACN: DMSO Contour surface plots for the interaction effects of glucose: furfural: ACN: DMSO ratios on various responses were generated using the Minitab 16 modelling software. ANOVA of various responses was performed, and p <0.05 was taken as indicative of statistical significance. The correlation coefficient r and adjusted coefficient R2 were both >92%, suggesting significance. The contour plot indicates that biomass production is dependent upon glucose and the roles of furfural, acetonitrile and DMSO are negligible (Fig. 4a). All variables (glucose, furfural, ACN and DMSO) had a cumulative effect on undecylprodigiosin production (Fig. 4b). Glucose at higher concentrations inhibited antibiotic production. Furfural and ACN inhibit production of

12 actinorhodin, while DMSO has the opposite effect (Fig. 4c). The glucose: furfural: ACN: DMSO mixture analysis experiments suggest that furfural is a key player in differential production of undecylprodigiosin, and co-feeding of ACN and DMSO further increased undecylprodigiosin production. DMSO acts as an antioxidant and improves biomass production. Mixture design is an interesting approach to assessing the interaction effect of various independent variables on response and their optimization for maximum production (Bhatia et al., 2015b). 3.6 Validation of model A numerical optimization method was applied to predict the values of variables for the desired response. High and low values were adjusted for all responses using optimal parameter settings as recommended by the statistical software (Minitab 16) to obtain the most suitable responses. D is the composite desirability and d is the individual desirability, and their value under optimal conditions was close to 1.0, which confirmed the suitability of the model. To verify the design, we chose a model to produce undecylprodigiosin (3.1 µg/mg dcw) with optimum biomass of 2.5 mg dcw/ml and minimum actinorhodin of 10 µg/mg dcw. The optimum values for glucose: furfural: ACN: DMSO were predicted to be 0.484:0.252:0.151:0.111 (Fig. S5). Use of these predicted values resulted in undecylprodigiosin production of 3.01 µg/mg dcw and biomass of 2.4 mg dcw/ml. The verification revealed the model to have a high degree of accuracy (> 97%), indicating its validity under the test conditions. 3.7 Undecylprodigiosin production using EFB hydrolysate The composition of EFB hydrolysate (EFBH) was determined. Glucose is the major carbon source (12.8%), followed by xylose/galactose/mannose (1.58%); other sugars are present in trace amounts. Hot water treatment of EFB leads to production of byproducts; i.e., furfural (0.04%) and hydroxymethylfurfural (HMF), in trace amounts.

13 EFBH was used as a carbon source and the concentrations of other components (furfural, ACN, DMSO) were adjusted according to the predicted response of synthetic media. S. coelicolor was cultured in minimal medium for 72 h with EFBH: furfural: ACN: DMSO components, and resulted into biomass (3.1 mg dcw/ml), undecylprodigiosin (3.0 µg/mg dcw) and 5.0 µg/mg dcw actinorhodin production (Fig. 5). An EFBH minimal medium without furfural, ACN and DMSO resulted in production of biomass (3.4 mg dcw/ml), undecylprodigiosin (1.6 µg/mg dcw) and actinorhodin (3.4 µg/mg dcw). Engineered synthetic medium with glucose: furfural: ACN: DMSO and engineered EFBH: furfural: ACN: DMSO medium yielded almost identical undecylprodigiosin production, 3.2 and 3.0 µg/mg dcw, respectively (Fig. 5). Medium engineering to utilize EFBH as a carbon source, with the addition of furfural, ACN and DMSO, resulted in an 88% increase in undecylprodigiosin production. 3.8 Scale up of undecylprodigiosin production in EFBH medium Scale up of S. coelicolor to the 100 ml scale produced results identical to those at the 10 ml scale; i.e., 3.01 µg/mg dcw undecylprodigiosin with 3.32 mg dcw/ml biomass, and cells exhibited identical morphology and growth profile. Upon further scale up to the fermenter level (500 ml), 4.2 µg/mg dcw undecylprodigiosin production with 4.8 mg dcw/ml biomass was recorded. Undecylprodigiosin production increased by 26% with 44.5% increase in biomass in comparison to 10 ml scale. Better biomass production likely due to enhanced oxygen-transfer and mixing power in a fermenter (Zhang et al., 1996). Medium engineering using EFBH as carbon source resulted in a 3.2-fold increase in undecylprodigiosin production

14 4. Conclusion Furfural decrease biomass production, utilization of glucose and xylose and enhances undecylprodigiosin production by affecting cell morphology and the expression of the antibiotic regulatory genes redD and actII-orf4. There is need to engineer S. coelicolor cells to tolerate furfural and improving their xylose utilization capability. Liognocellulosic biomass may be a better carbon source and the effects of various biomass-derived inhibitors on microbial fermentation, metabolism and carbon utilization should be investigated to determine the optimum growth medium formulation for maximum productivity. Acknowledgements This study was supported in part by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF2013R1A1A2A10004690 and NRF-2011-619-E0002), by the R & D Program of MOTIE/KEIT (10049674) and an Energy Efficiency & Resources of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea Government Ministry of Trade, Industry and Energy (20133030000300). Additionally, this article was also supported by the KU Research Professor Program of Konkuk University, Seoul, South Korea.

15 References 1. Ahn, Y., Hu, D.H., Hong, J.H., Lee, S.H., Kim, H.J., Kim, H. 2012. Effect of cosolvent on the spinnability and properties of electrospun cellulose nanofiber. Carbohydr. Polym. 89(2), 340-5. 2. Allen, S.A., Clark, W., McCaffery, J.M., Cai, Z., Lanctot, A., Slininger, P.J., Liu, Z.L., Gorsich, S.W. 2010. Furfural induces reactive oxygen species accumulation and cellular damage in Saccharomyces cerevisiae. Biotechnol. Biofuels. 3(2), 1754-6834. 3. Arantes, V., Saddler, J.N. 2010. Access to cellulose limits the efficiency of enzymatic hydrolysis: the role of amorphogenesis. Biotechnol. Biofuels. 3(4), 1754-6834. 4. Ask, M., Bettiga, M., Mapelli, V., Olsson, L. 2013. The influence of HMF and furfural on redox-balance and energy-state of xylose-utilizing Saccharomyces cerevisiae. Biotechnol. Biofuels. 6(1), 1754-6834. 5. Bhatia, S.K., Yi, D.-H., Kim, Y.-H., Kim, H.-J., Seo, H.-M., Lee, J.-H., Kim, J.-H., Jeon, J.-M., Jang, K.-S., Kim, Y.-G., Yang, Y.-H. 2015a. Development of semisynthetic microbial consortia of Streptomyces coelicolor for increased production of biodiesel (fatty acid methyl esters). Fuel. 159, 189-196. 6. Bhatia, S.K., Yi, D.H., Kim, H.J., Jeon, J.M., Kim, Y.H., Sathiyanarayanan, G., Seo, H.M., Lee, J.H., Kim, J.H., Park, K., Brigham, C.J., Yang, Y.H. 2015b. Overexpression of succinyl-CoA synthase for poly (3-hydroxybutyrate-co-3hydroxyvalerate) production in engineered Escherichia coli BL21(DE3). J. Appl. Microbiol. 119(3), 724-35.

16 7. Borodina, I., Siebring, J., Zhang, J., Smith, C.P., van Keulen, G., Dijkhuizen, L., Nielsen, J. 2008. Antibiotic overproduction in Streptomyces coelicolor A3 2 mediated by phosphofructokinase deletion. J. Biol. Chem. 283(37), 25186-99. 8. Chen, G., Wang, G.Y., Li, X., Waters, B., Davies, J. 2000. Enhanced production of microbial metabolites in the presence of dimethyl sulfoxide. J. Antibiot. 53(10), 114553. 9. Craney, A., Ahmed, S., Nodwell, J. 2013. Towards a new science of secondary metabolism. J. Antibiot. 66(7), 387-400. 10. Dwivedi, S., Wahab, R., Khan, F., Mishra, Y.K., Musarrat, J., Al-Khedhairy, A.A. 2014. Reactive Oxygen Species Mediated Bacterial Biofilm Inhibition via Zinc Oxide Nanoparticles and Their Statistical Determination. PLoS ONE. 9(11), e111289. 11. El-Shahed, K.Y., El-Diwany, A.I., Awad, H.M. 2008. Enhanced production of streptomycin and hydrolytic enzymes by Streptomyces griseus strains using different types of organic solvents and detergent compounds. Indian J. Biotechnol. 7, 341-348. 12. Eom, I.Y., Oh, Y.H., Park, S.J., Lee, S.H., Yu, J.H. 2015. Fermentative l-lactic acid production from pretreated whole slurry of oil palm trunk treated by hydrothermolysis and subsequent enzymatic hydrolysis. Bioresour. Technol. 185, 143-9. 13. Fujii, T., Gramajo, H.C., Takano, E., Bibb, M.J. 1996. redD and actII-ORF4, pathway-specific regulatory genes for antibiotic production in Streptomyces coelicolor A3(2), are transcribed in vitro by an RNA polymerase holoenzyme containing sigma hrdD. J. Bacteriol. 178(11), 3402-3405.

17 14. Heer, D., Sauer, U. 2008. Identification of furfural as a key toxin in lignocellulosic hydrolysates and evolution of a tolerant yeast strain. Microb. Biotechnol. 1(6), 497506. 15. Hindra, Pak, P., Elliot, M.A. 2010. Regulation of a novel gene cluster involved in secondary metabolite production in Streptomyces coelicolor. J. Bacteriol. 192(19), 4973-82. 16. Hobbs, G., Frazer, C.M., Gardner, D.C.J., Flett, F., Oliver, S.G. 1990. Pigmented antibiotic production by Streptomyces coelicolor A3(2): kinetics and the influence of nutrients. Microbiol. 136(11), 2291-2296. 17. Horinouchi, S., Beppu, T. 1984. Production in Large Quantities of Actinorhodin and Undecylprodigiosin Induced by afsB in Streptomyces lividans. Agric. Biol. Chem. 48(8), 2131-2133. 18. Ishii, S., Koki, J., Unno, H., Hori, K. 2004. Two morphological types of cell appendages on a strongly adhesive bacterium, Acinetobacter sp. strain Tol 5. Appl. Environ. Microbiol. 70(8), 5026-9. 19. Jonsson, L.J., Alriksson, B., Nilvebrant, N.O. 2013. Bioconversion of lignocellulose: inhibitors and detoxification. Biotechnol. Biofuels. 6(1), 1754-6834. 20. Kang, Y., Ahn, Y., Lee, S.H., Hong, J.H., Ku, M.K., Kim, H. 2013. Lignocellulosic nanofiber prepared by alkali treatment and electrospinning using ionic liquid. Fibers Polyme. 14(4), 530-536. 21. Kim, M., Yi, J.S., Lakshmanan, M., Lee, D.Y., Kim, B.G. 2015a. Transcriptomicsbased strain optimization tool for designing secondary metabolite overproducing strains of Streptomyces coelicolor. Biotechnol. Bioeng. 15(10), 25830.

18 22. Kim, S.H., Traag, B.A., Hasan, A.H., McDowall, K.J., Kim, B.G., van Wezel, G.P. 2015b. Transcriptional analysis of the cell division-related ssg genes in Streptomyces coelicolor reveals direct control of ssgR by AtrA. Antonie Van Leeuwenhoek. 108(1), 201-13. 23. Kootstra, A.M.J., Mosier, N.S., Scott, E.L., Beeftink, H.H., Sanders, J.P.M. 2009. Differential effects of mineral and organic acids on the kinetics of arabinose degradation under lignocellulose pretreatment conditions. Biochem. Eng. J. 43(1), 9297. 24. Li, Z., Chen, C.H., Hegg, E.L., Hodge, D.B. 2013. Rapid and effective oxidative pretreatment of woody biomass at mild reaction conditions and low oxidant loadings. Biotechnol. Biofuel. 6(1), 1754-6834. 25. Pandey, A., Soccol, C.R., Nigam, P., Soccol, V.T. 2000. Biotechnological potential of agro-industrial residues. I: sugarcane bagasse. Bioresour. Technol. 74(1), 69-80. 26. Patel, S.K., Kumar, P., Singh, M., Lee, J.K., Kalia, V.C. 2015. Integrative approach to produce hydrogen and polyhydroxybutyrate from biowaste using defined bacterial cultures. Bioresour. Technol. 176, 136-41. 27. Pettit, R.K. 2011. Small-molecule elicitation of microbial secondary metabolites. Microb. Biotechnol. 4(4), 471-8. 28. Revill, W.P., Bibb, M.J., Hopwood, D.A. 1996. Relationships between fatty acid and polyketide synthases from Streptomyces coelicolor A3(2): characterization of the fatty acid synthase acyl carrier protein. J. Bacteriol. 178(19), 5660-7. 29. Schaberle, T.F., Orland, A., Konig, G.M. 2014. Enhanced production of undecylprodigiosin in Streptomyces coelicolor by co-cultivation with the

19 corallopyronin A-producing myxobacterium, Corallococcus coralloides. Biotechnol. Lett. 36(3), 641-8. 30. Sindhu, R., Binod, P., Pandey, A. 2015. Biological pretreatment of lignocellulosic biomass - An overview. Bioresour. Technol. 24(15), 01134-7. 31. van Wezel, G.P., White, J., Hoogvliet, G., Bibb, M.J. 2000. Application of redD, the transcriptional activator gene of the undecylprodigiosin biosynthetic pathway, as a reporter for transcriptional activity in Streptomyces coelicolor A3(2) and Streptomyces lividans. J. Mol. Microbiol. Biotechnol. 2(4), 551-6. 32. Wang, W., Ji, J., Li, X., Wang, J., Li, S., Pan, G., Fan, K., Yang, K. 2014. Angucyclines as signals modulate the behaviors of Streptomyces coelicolor. Proc. Natl. Acad. Sci. U S A. 111(15), 5688-93. 33. Williamson, N.R., Fineran, P.C., Leeper, F.J., Salmond, G.P.C. 2006. The biosynthesis and regulation of bacterial prodiginines. Nat. Rev. Micro. 4(12), 887-899. 34. Yang, Y.H., Joo, H.S., Lee, K., Liou, K.K., Lee, H.C., Sohng, J.K., Kim, B.G. 2005. Novel method for detection of butanolides in Streptomyces coelicolor culture broth, using a His-tagged receptor (ScbR) and mass spectrometry. Appl. Environ. Microbiol. 71(9), 5050-5. 35. Zhang, J., Marcin, C., Shifflet, M.A., Salmon, P., Brix, T., Greasham, R., Buckland, B., Chartrain, M. 1996. Development of a defined medium fermentation process for physostigmine production by Streptomyces griseofuscus. Appl. Microbiol. Biotechnol. 44(5), 568-75.

20 Figure captions Fig. 1 Determination of the sub-lethal concentrations of biomass derived inhibitors, furfural and hydroxymethylfurfural (HMF). Results are means ± S.D (݊=3), ܲ <0.05. Fig. 2 Effects of furfural on (a) biomass production (b) glucose utilization (c) actinorhodin (d) and undecylprodigiosin production by S. coelicolor. Results are means ± S.D (݊=3), ܲ <0.05. Fig. 3 Effects of antioxidants and organic solvents on S. coelicolor in minimal medium with glucose (Glu) as the carbon source and furfural (Fur) as an inhibitor. (a) Effects of ascorbic acid (Asc), glycine beatine (Gly), dithiothreitol (DTT) and dimethyl sulphoxide (DMSO) on antibiotic production. (b) Effects of acetone (Ace), acetonitrile (ACN) and ethanol (Eth). Results are means ± S.D (݊=3), ܲ <0.05. Fig. 4 Mixture contour plots for interactive effects of glucose: furfural: DMSO: ACN (a) biomass production (mg dcw/ml) (b) undecylprodigiosin (µg/mg dcw) and (c) actinorhodin (µg/mg dcw). Fig. 5 Antibiotic production using EFB hydrolysate as a carbon source in minimal medium. (a) Comparison of minimal medium (Glu), engineered minimal medium (Glu:Fur:DMSO:ACN), minimal medium with EFBH as a carbon source and engineered minimal medium with EFBH as a carbon source. .

21 Table. 1 Experimental design points selected by the simplex lattice methodology for the mixture analysis model using as variables glucose, furfural, acetonitrile (ACN) and dimethylsulphoxide (DMSO), and their subsequent effects on biomass, undecylprodigiosin (red) and actinorhodin (blue).

ID#

Glucose

Furfural

ACN

DMSO

Biomass

Red

Blue

(mg dcw/ml)

(µg/mg dcw)

(µg/mg dcw)

1

0.125

0.625

0.125

0.125

1.65

2.33

2.71

2

0.125

0.125

0.125

0.625

2

1.84

1.98

3

0.000

1.000

0.000

0.000

0.35

0.00

0.00

4

0.000

0.500

0.500

0.000

0.65

0.00

0.00

5

1.000

0.000

0.000

0.000

3.3

0.66

21.9

6

0.125

0.125

0.625

0.125

1.6

2.25

1.76

7

0.333

0.333

0.333

0.000

2.2

2.86

1.05

8

0.625

0.125

0.125

0.125

3.05

2.94

8.99

9

0.000

0.000

0.000

1.000

0.4

0.00

0.00

10

0.000

0.500

0.000

0.500

0.65

0.00

0.00

11

0.500

0.500

0.000

0.000

1.7

2.61

7.05

12

0.333

0.000

0.333

0.333

2.8

2.60

9.71

13

0.000

0.333

0.333

0.333

0.25

0.00

0.00

14

0.500

0.000

0.500

0.000

1.7

2.53

5.30

15

0.250

0.250

0.250

0.250

2

3.42

9.93

16

0.500

0.000

0.000

0.500

2.25

1.99

25.8

17

0.333

0.333

0.000

0.333

2.7

2.76

18.2

18

0.000

0.000

0.500

0.500

0.1

0.00

0.00

19

0.000

0.000

1.000

0.000

0.15

0.00

0.00

22 Fig. 1a

4.0

2.0

7.0

3.0 1.6 2.5 1.4 2.0 Biomass Actinorhodin Undecylprodigiosin

1.5

1.2

1.0 1

2

5.5 5.0 4.5 4.0 3.5

1.0 0

6.0

Actinorhodin (µg/mg dcw)

Biomass (mg dcw/ml)

1.8

Undecylprodigiosin (µg/mg dcw)

6.5 3.5

3

4

3.0

5

Furfural (mM)

Fig. 1b

Biomass Actinorhodin Undecylprodigiosin

Biomass (mg dcw/ ml)

5.0

1.4

4.5

1.2

4.0

1.0

3.5

0.8

3.0

0.6

2.5

0.4 0

1

2

HMF (mM)

3

4

5

7

6

5

4

3

2

Actinorhodin (µg/mg dcw)

1.6

Undecylprodigiosin (µg/mg dcw)

5.5

23 Fig. 2

5

120

100 Glu Glu:Fur

80

3

Glucose (%)

Biomass (mg dcw/ml)

4

2

60 Glu Glu:Fur

40

1 20 0

0

0

20

40

60

80

0

20

Time (h)

a

60

80

b

7

2.5

Undecylprodigiosin (µg/ mg dcw)

6

Actinorhodin (µg/mg dcw)

40

Time (h)

Glu Glu:Fur

5 4 3 2 1

Glu Glu:Fur

2.0

1.5

1.0

0.5

0.0

0

0

20

40

60

80

0

20

Time (h)

c

40

Time (h)

d

60

80

G lu :F ur :D

M SO

G lu :F ur :D TT

3

G lu :F ur :G ly

G lu :F ur :A sc

G lu :F ur

G lu

Biomass (mg dcw/ml)

3.0

Biomass Undecylprodigiosin Actinorhodin 2.5

2.0

2 1.5

1.0

1 0.5

0 0.0

5

4

3

2

1

0

Actinorhodin (µg/mg dcw)

4

Undecylprodigiosin (µg/mg dcw)

24

Fig. 3a

7

6

Fig. 3b

-

Biomass (mg dcw mL ) 4

3

2

1

0 Glu

Glu : Glu :

Glu :

Fu r:D MS O

Fu r:D MS O:

Glu :

Ac

e

Biomass Undecylprodigiosin Actinorhodin

Fu r:D MS O:

25

Glu :

Fu r

AC N

Fu r:D MS O:

E th

5

4

3

2

1

0

-

Undecylprodigiosin (µg mg dcw)

8

6

4

2

0

-

Actinorhodin (µg mg dcw)

26 Fig. 4a

27 Fig. 4b

28 Fig. 4c

lu :F

CN

ur :D M SO :A CN

EF BH

ur :D M SO :A

3 Biomass Undecylprodigiosin Actinorhodin

3

2 2

1 1

0 0

Undecylprodigiosin (µg mg dcw)

-

8

6

4

2

0

-

10

Actinorhodin (µg mg dcw)

4

EF BH :F

G

G lu

-

Biomass (mg dcw mL )

29

Fig. 5

4 14

12

30

31 Highlights


Oil palm biomass hydrolysate has furfural and hydroxymethylfurfural as inhibitors
Furfural effects xylose utilization in S. coelicolor adversely
Mycelia formation is decreased with reduction in actinorhodin production
Furfural elicits undecylprodigiosin production in S. coelicolor upto 52%
Engineered EFB hydrolysate media resulted a 88% higher undecylprodigiosin produ ction