1 HIGH 2,3-BUTANEDIOL PRODUCTION 2 FROM GLYCEROL BY Raoultella terrigena CECT 4519 3 Vanessa Ripoll 2, Alberto Rodríguez 3 , Miguel Ladero 1, Victoria E. Santos 1, * 4 1 Department of Chemical and Materials Engineering, Faculty of Chemistry, Universidad 5 Complutense de Madrid, 28040 Madrid, Spain 6 2 Facultad de Ciencias Experimentales, Universidad Francisco de Vitoria, UFV, Ctra. Pozuelo- 7 Majadahonda km 1.800, 28223, Pozuelo de Alarcón (Madrid, Spain) +34 91 351 03 03 8 3 Biological Research Centre (CIB), Spanish National Research Council (CSIC), 28040 Madrid, 9 Spain. +34 91 837 31 12 – 4384/4385 10 * Corresponding author. 11 E-mail ad dd resses: [email protected] (Vanessa Ripoll), [email protected] (Alberto 12 Rodríguez) [email protected] (Miguel Ladero), [email protected] (Victoria E. Santos). 1 13 14 ABSTRACT 15 Bioconversion of derived-biodiesel glycerol into 2,3-butanediol has received recently much 16 attention due to its increasing surplus . It is widely used as bulk chemical for several industrial 17 applications . and its multiple uses in industry as bulk chemical . The influence of initial glycerol 18 concentration on 2,3-butanediol production in batch runs has been studied. A concentration 19 higher than 140 g/L produces an inhibitory effect on the final 2,3-butanediol concentration as 20 well as on and its production rate. In batch operation mode, the best yield was obtained 21 employing 140 g/L as initial glycerol concentration, being 71% the highest yield respect to the 22 theoretical maximum yield ( 71 %) was reached employing 140 g/L as initial concentration 140 23 g/L. Based on these results, fed-batch operation mode was carried out to improve 2,3-butanediol 24 production. a high 2,3-butanediol production has been achieved through a fed-batch strategy. 25 The titre of reached 2,3-butanediol concentration was 90.5 g/L from employing pure glycerol as 26 carbon source and 80.5 g/L from employing raw glycerol. The 2,3 --butanediol yield respect to 27 the theoretical maximum yield was also improved through the fed-batch operation (90 % respect 28 to the theoretical maximum yiel d). To date, this concentration is the highest produced amount 29 employing as biocatalyst a non-pathogenic bacterium (level 1). 30 KEYWORDS 31 Biorefinery; Raoultella terrigena; 2,3-butanediol; Raw glycerol. 32 HIGHLIGHTS 33 High 2,3-butanediol concentration was achieved employing a class 1 microorganism. 34 Initial substrate inhibition has been described in batch experiments. 35 Yield respect to the theoretical maximum yield reaches 90 %. 36 Fed-batch run employing pure glycerol leads to 90.5 g/L of 2,3-butanediol. 37 2 38 39 1 INTRODUCTION 40 Nowadays, the progressive exhaustion of the non-renewable petroleum sources has triggered an 41 energy worldwide crisis. Not only searching Besides the search for new sustainable fuels is 42 required, but also , the develop menting of alternative synthesis routes of bulk chemical is 43 required to complement or substitute the conventional petrochemical ones. In this sense, the 44 advances in the biotechnological industry involve have meant a great contribution in this field 45 (Wilke et al. 2004). 46 2,3-butanediol (2,3-BD) is an one of these interesting commodity chemicals with a large 47 number of industrial applications, as a feedstock for rubber, plastic, and solvents production 48 processes. 2,3-BD dehydration reaction is used to produce methyl ethyl ketone, which is 49 employed as an effective fuel additive and as a solvent for resins and lacquers. It is also 50 significant the direct use of In addition, 2,3-BD is used as antifreeze agent and as octane booster 51 for fuels (Celinska and Grajek 2009; Villet et al. 1981). 52 During the last decade, many authors have been focus inged on the search ing for renewable 53 carbon sources based on of industrial wastes employed as renewable carbon sources to produce 54 2,3-BD by biotechnological routes. The latest published studies have been summarized in Table 55 1. Most of the studied feedstock proceed from agricultural and food industry wastes, whose 56 composition is variable and sometimes includes inhibitory compounds (Palmqvist et al. 2000). 57 To date, the highest 2,3-BD concentration has been reached employing sugar molasses as 58 feedstock. Glucose is the main sugar in this waste, which is, usually, the preferred carbon source 59 fastly uptaken and metabolized by several bacteria (Martinez-Gomez et al. 2012). Although 60 glycerol is an energy-poor carbon source, a number of microorganisms are able also to grow on 61 glycerol as sole carbon source, and metabolize it by means of both oxidative and reductive 62 pathways (Da Silva et al. 2009). 3 63 In the framework of circular economy, raw glycerol, the waste derived from the biodiesel 64 production, stands out as a promising alternative carbon source for fermentation bioprocesses. 65 The increase of biodiesel demand according to European policies, such as the Directives 66 2003/30/EC and 2009/27/EC, has caused a large surplus of raw glycerol. As a result, the drastic 67 fall of its price and the presence of impurities make the conventional glycerol applications 68 related to pharmaceutical, food and cosmetic industries not enough to cope with glycerol 69 production (Gerpen 2005; Pagliaro et al. 2007; Santibañez et al. 2011). 70 In order to establish implant an industrial-scale bio process, safety regulation is thea key aspect. 71 In the biotechnological area, nN on-pathogenic microorganisms belonging to risk group 1 are the 72 most desirable biocatalysts. To date, the best 2,3-BD producer biocatalysts belong to risk group 73 2 2,3-BD producer microorganisms, ( Klebsiella pneumonia and K. oxytoca ), belong to risk 74 group 2 (Cho et al. 2015; Petrov and Petrova, 2010). Recently, Raoultella terrigena CECT 4519 75 has been reported proposed as a promising biocatalyst of the bioconversion process of raw 76 glycerol into 2,3-BD, which accomplishes the safety requirements (Ripoll et al . 2016). 77 According to the available information, this bacterial strain belongs to biosafety level 1 78 (BacDive, CECT). Therefore, it is considered as non-pathogenic biocatalyst. 79 The aim of the present work is to study reach a high 2,3-BD production by means of through 80 glycerol bioconversion employing R. terrigena CECT 4519 as biocatalyst . To reach this goal, 81 the influence of initial glycerol concentration on 2,3-BD production in batch experiments has 82 been studied. Based on the obtained results, different profiles for glycerol addition in fed-batch 83 experiments were carried out in order to enhance the 2,3-BD titre. In addition, aA study 84 regarding the use of industrial raw glycerol as sole carbon source has been also performed. 85 2 MATERIALS AND METHODS 86 2.1 Microorganism and inoculum procedure 87 In the present work, R. terrigena CECT 4519 has beenwas employed as biocatalyst. The ability 88 of this strain to produce 2,3-butanediol ha ds been previously reported demonstrated (Ripoll et 4 89 al. 2010). Strain maintenance and inoculum growth were carried out based on the established 90 protocols previously published (Ripoll et al. 2010; Rodriguez et al. 2017). Cells were stored at 91 -80 ºC in a glycerol-saline serum solution composed by glycerol-saline serum mixture (50:50 % 92 w/w). Two growth steps wereare required to ensure a reproducible inoculum from the stored 93 cellular stock. Both growth phases were carried out in 250-mL unbaffled shaken flask with 50 94 mL of working volume, operating at 30 ºC and 210 rpm in an orbital shaker. The duration of 95 each step was 14 h and 4 h, respectively ., being The initial dry biomass concentration was 0.1 96 g/L in both cases . 97 2.2 2,3-butanediol production experiments 98 Batch and fed-batch fermentations were carried out in a 3-L Biostat® B-Plus (Sartorius AG 99 Germany) with a working volume of 2 - L. The employed bioreactor configuration consist eds 100 ofin a stirred unbaffled cylindrical tank. Medium composition was formulated employing the 101 following amount (per litre of deionized water): formed 2 g of NH 4Cl, 6 g of KH 2PO 4, 12 g of 102 Na2HPO 4, 1 g of NaCl, 0.246 g of MgSO 4·7H 2O2, 0.011 g of CaCl 2 and 1.5 g of yeast extract 103 per litre of deionized water . Different initial glycerol concentration between 35 and 210 g/L 104 were used, employ inged pure glycerol (Panreac, ref. 151339) as well as industrial raw glycerol . 105 Raw glycerol has been provided by a Spanish biodiesel manufacture plant. Raw glycerol 106 samples contented glycerol (from 55 to 85 % w/w), chloride salts (from 17.3 to 56.6 g/L), 107 phosphate salts (from 2.54 to 6.14 g/L), and a low amount of methanol ( ∼0.06 % w/w). 108 At the beginning of each experiment, 10 % v/v of inoculum were added so to reach an initial 109 biomass concentration of 0.25 g/L. Operational conditions were set according to earlier results 110 (Ripoll et al. 2010; Rodriguez et al. 2017). Operational temperature and stirring speed were set 111 at 30 ºC and 400 r.p.m., respectively. Airflow rate was fixed at 1.5 v.v.m. Initial pH was 7.0 and 112 it was not controlled until it reached a value of 5.5 during the fermentation. From that moment 113 on, pH was controlled at 5.5 employing 2 M NaOH and 2 M HCl acid and basic solutions (HCl 5 114 2M and NaOH 2M) .