Dark fermentative bioconversion of glycerol to hydrogen by Bacillus thuringiensis

Accepted Manuscript Short Communication Dark fermentative bioconversion of glycerol to hydrogen by Bacillus thuringiensis Prasun Kumar, Rishi Sharma, Subhasree Ray, Sanjeet Mehariya, Sanjay K.S. Patel, Jung-Kul Lee, Vipin C. Kalia PII: DOI: Reference:

S0960-8524(15)00167-4 http://dx.doi.org/10.1016/j.biortech.2015.01.138 BITE 14572

To appear in:

Bioresource Technology

Received Date: Revised Date: Accepted Date:

21 November 2014 27 January 2015 30 January 2015

Please cite this article as: Kumar, P., Sharma, R., Ray, S., Mehariya, S., Patel, S.K.S., Lee, J-K., Kalia, V.C., Dark fermentative bioconversion of glycerol to hydrogen by Bacillus thuringiensis, Bioresource Technology (2015), doi: http://dx.doi.org/10.1016/j.biortech.2015.01.138

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.

Dark fermentative bioconversion of glycerol to hydrogen by Bacillus thuringiensis

Prasun Kumara, Rishi Sharmaa, Subhasree Raya, Sanjeet Mehariyaa, Sanjay K.S. Patelb, Jung-Kul Leeb, Vipin C. Kaliaa*

a

Microbial Biotechnology and Genomics, CSIR–Institute of Genomics and Integrative Biology (IGIB),

Delhi University Campus, Mall Road, Delhi-110007, India b

Department of Chemical Engineering, Konkuk University, 1 Hwayang-Dong, Gwangjin-Gu, Seoul, Korea, 143-701

*

Corresponding author Mailing address:

a

Microbial Biotechnology and Genomics,

CSIR -Institute of Genomics and Integrative Biology (IGIB), Delhi University Campus, Mall Road, Delhi-11 00 07, India Phone : 091- 11- 27666156; 27666157 Fax: 091- 11 - 27667471; 27416489 E-mail: [email protected], [email protected]

Running title: Glycerol to hydrogen by Bacillus thuringiensis

1

Abstract Biodiesel manufacturing units discharge effluents rich in glycerol. The need is to convert crude glycerol (CG) into useful products such as hydrogen (H2). Under batch culture, B. thuringiensis EGU45 adapted on pure glycerol (PG, 2% v/v) resulted in an H2 yield of 0.646 mol/mol glycerol consumed on minimal media (250 mL) supplemented with 1% ammonium nitrate at 37 °C over 4 days. Here, H2 constituted 67% of the total biogas. Under continuous culture, at 2 days of hydraulic retention time, B. thuringiensis immobilized on ligno-cellulosic materials (banana leaves - BL, 10% v/v) resulted in a H2 yield of 0.386 mol/mol PG consumed. On CG, the maximal H2 yield of 0.393 mol/mol feed consumed was recorded. In brief, B. thuringiensis could transform CG, on limited resources - minimal medium with sodium nitrate, by immobilizing them on cheap and easily available biowaste, which makes it a suitable candidate for H2 production on a large scale.

Keywords: Adaptation, Bacillus, Continuous culture, Crude glycerol, Dark fermentation, Hydrogen

2

1. Introduction Biodiesel has been recognized as an alternative to fossil fuel and is being produced on a commercial scale. In the US and EU, 6.9 million tonnes of biodiesel are being produced each year (Maru et al., 2012). Biodiesel industrial effluent typically consists of 70-75% glycerol along with a few contaminants in small quantities such as methanol, soap, oils, salts and solid organic materials. The generation of crude glycerol (CG) is about 1 kg for every 9 kg of biodiesel produced (Dasari et al., 2005; Donkin, 2008). Since biodiesel production is increasing rapidly, the price of CG is declining at such a rate that its disposal is becoming uneconomical. This scenario has driven researchers to develop procedures to convert glycerol into useful by-products (Wang et al., 2012). Biological hydrogen production (BHP) is an eco-friendly process that has attracted significant attention worldwide (Veziroğlu, 1995). Many bacteria can metabolize pure glycerol (PG) and CG to hydrogen (H2) (Kumar et al., 2015; Lo et al., 2013; Sarma et al., 2012, 2013). Glycerol has been ranked as a better candidate for H2 generation than the most widely utilized glucose because of its being more reduced, and that its metabolism generates comparatively more NAD(P)H (Maru et al., 2012). BHP from glycerol as feed has been reported to the tune of 0.14-1.23 mol/mol (Kumar et al., 2015; Lo et al., 2013; Maru et al., 2012, 2013). Most of the studies have been carried out as batch culture using Citrobacter, Clostridium, Enterobacter, Klebsiella, and Thermotoga (Kumar et al., 2015; Maru et al., 2012). A limited effort has been made to demonstrate H2 production from glycerol in continuous culture (Kumar and Das, 2001; Kumar et al., 2013). Although, Bacillus species have been shown to evolve H2 from pure sugars and biowastes (Kumar et al., 2013; Patel and Kalia, 2013; Patel et al., 2010, 2012), however, the role of glycerol as feed has been confined thus far only to Bacillus coagulans (Kumar et al., 2013). Bacillus as an organism is generally regarded as safe (GRAS), which can be easily grown to high cell densities (Law et al., 2003; Valappil et al., 2007). Bacillus spp. are well recognized as work horses because of their abilities to produce a wide range of enzymes for biotechnological applications on an industrial scale (Schallmey et al., 2004; Westers et al., 2004). Bacillus has been now recognized as a potential candidate for bioenergy and biopolymer production (Kumar et al., 2013). Another unique feature of Bacillus is its ability to compete with other microbes since it can produce bioactive molecules such as acyl homoserine lactonases (Huma et al., 2011), which helps in its survival when mixed cultures are to be used for biotransformation of biowastes (Kumar et al., 2014; Patel et al., 2012, 2015). It is thus desirable to explore the abilities of other Bacillus spp. as H2 producers from glycerol under dark fermentative conditions. Bacillus thuringiensis EGU45 has been recently demonstrated to be effective as a producer of hydrolytic enzymes, H2 and/or polyhydroxyalkanoates (PHAs) (Patel et al., 2010, 2012, 2015; Porwal et al., 2008; Singh et al., 2013). Mixed bacterial cultures involving B. thuringiensis EGU45 have been instrumental in enhancing H2 and PHA co-polymer production, which thus helps in efficient utilization of biowastes as feed (Kumar et al., 2014 a,b; Patel et

3

al., 2014, 2015). Use of a versatile organism like B. thuringiensis EGU45 to tackle the glycerol glut for the production of value added bio-products seem to be a promising strategy.

Since glycerol transformation is limited by the presence of high feed concentration, we have used the following to overcome this problem: i) adaptation of bacterial culture, and ii) immobilization of bacteria on various support materials, to achieve a continuous culture digestion of PG as well as CG under dark fermentative conditions.

2. Material and Methods 2.1. Organisms and Growth Conditions B. thuringiensis strain EGU45, reported to produce H2 from glucose (Porwal et al., 2008), was grown on HiMedia nutrient broth (NB) [13.0 g/L distilled water: composed of (g/L): peptic digest of animal tissue 5.0, NaCl 5.0, beef extract 1.5 and yeast extract 1.5] and incubated at 37 °C with stirring at 200 revolutions per min for 16-20 h (Porwal et al., 2008).

2.2. Adaptation of bacterial cultures B. thuringiensis EGU45 culture was adapted to high glycerol concentration by growing it culture on 10 mL NB supplemented with 1, 2, 5, 7.5 and 10% PG (v/v), in separate tubes. Each tube was inoculated with 100 µl bacterial cultures of OD600: 1.0 and incubated at 37 °C for 24 h. At the end of 24 h, 100 µl bacterial culture of OD600: 1.0 was transferred to a fresh 10 mL NB supplemented with 1, 2, 5, 7.5 and 10% PG (v/v), in separate tubes. This cycle was repeated on daily basis for a total of 15 days. Cultures were retrieved for use as inocula at the end of 5, 10 and 15 cycles of adaptation. B. thuringiensis EGU45 so adapted on PG was used as inoculum (10 µg cell protein/mL of the medium) for H2 production. Bacteria grown on NB without glycerol were used as controls.

2.3. Hydrogen Production 2.3.1. Batch culture 250 mL of minimal medium (M-9) (Porwal et al., 2008) supplemented with (1.0 g/L): ammonium chloride (NH4Cl), sodium nitrate (NaNO3), and ammonium nitrate (NH4NO3) as different nitrogen (N) sources and 1-3% v/v PG was inoculated with an adapted culture of B. thuringiensis EGU45 to a final value of 10 µg cell protein/mL of the medium. The pH of the medium was adjusted to 7.0 prior to incubation and the aspirator bottles were made air tight using glass stoppers. There was provision for

4

gas outlet and liquid sampling in the incubation bottle. Argon was purged in all bottles and incubated at 37 °C without shaking. The pH of the cultures was adjusted to 7.0 using 2N NaOH or 2N HCl and re-flushed with argon, on a daily basis. The evolved gases were collected by the water displacement method. Gas collection and analysis of the samples were carried out, until H2 evolution ceased (Patel et al., 2010).

2.3.2. Continuous Culture Immobilization of bacterial cells on banana leaves (BL), coconut coir (CC), groundnut shell (GS) and pea shells (PS) was performed as described earlier (Patel et al., 2010). Briefly, 3 grams of dried ligno-cellulosic materials were packed in to PVC (Polyvinyl chloride) tube (length: 3 cm and diameter: 2 cm) and tied with a 10 cm2 nylon net. Free floating (FF) bacterial cultures and PVC alone as support material were used as controls. Experiments were conducted in aspirator bottles (20 × 10 cm), 1.2 L capacity with working volume of 1.0 L. Inoculum (10 µg cell protein/mL of medium) was added only once in the beginning of the experiment. Daily fed culture digestions were performed with B. thuringiensis EGU45 on different support materials, occupying 10% of the working volume.

Glycerol concentrations (2% and 5%, v/v) in the M-9 medium were studied at a hydraulic retention time (HRT) of 2 and 4 days. Depending upon the HRT of 2 days (or 4 days), 500 mL (or 250 mL) of the fermented medium was drained from the reactor and replaced with glycerol containing medium, every day. Reactor with free floating (FF) bacterial culture was run as a control. In order to obtain steady state values, the daily batch fed process was continued for a period of 40 days, both at 2 and 4 days HRT. All the fermentation experiments were performed in triplicate and incubated at 37 °C at an initial pH 7.0 and subsequently maintained at the same value, once every day.

2.4. Analytical method 2.4.1. Gas analysis The biogas produced during batch and continuous culture was measured through water displacement method and its composition was determined using gas chromatograph (Nucon GC5765, India) equipped with Porapak-Q and molecular sieve columns using the thermal conductivity detector, as reported earlier (Patel et al., 2010). For all optimization and daily fed culture experiments, H2 yields were calculated on the basis of glycerol consumed during the process.

5

2.4.2. Glycerol estimation The level of unutilized glycerol was estimated at the end of fermentation process (for batch culture) and on daily basis (for continuous culture experiments). 1 mL of effluent from H2 production processes were collected from each experiment. Residual glycerol was estimated by injecting 1 µl of culture supernatant (filtered through 0.45 µm) into Gas chromatograph (Nucon GC5765, India) equipped with Chromosorb 101 packed column (1.21 m × 0.25 mm × 3.0 mm; 80/100 mesh) and flame ionization detector, according to the manufacturer’s instruction which is an adaptation of the method described earlier (Forsberg, 1987). The temperature (°C) of injector, oven and detector were kept at 230, 245 and 250, respectively. Nitrogen was used as carrier gas with a flow rate of 30 mL/min. The final concentration was determined using 1% PG (v/v) as standard and 0.1 M isovaleric acid was used as internal standard. The CG was composed of 85.2% glycerol along with other compounds as impurities.

3. Results and Discussion 3.1. Batch culture H2 production 3.1.1. Effect of adaptation Bioconversion of glycerol to H2 by B. thuringiensis EGU45 was observed to vary with the concentrations of PG in the medium. Total biogas production varied from 255 mL to 890 mL, where H2 component ranged from 50 to 62.9%. The H2 yield increased from 0.263 to 0.282 mol/mol glycerol consumed with an increase in glycerol concentration from 1 to 2% (v/v) in the M-9 medium. Further increase in glycerol concentration resulted in a concurrent decline in H2 yield, which touched a low level of 0.056 mol/mol glycerol at 10% (v/v) glycerol. Thus, higher concentration of glycerol proved inhibitory to H2 evolution process. Such a phenomenon has been reported previously, where mixed microbial culture was used to produce H2 from CG as a sole carbon source. They resorted to ecobiotechnological approach to enhance bacterial tolerance to high glycerol concentration (Sittijunda and Reungsang, 2012). A pre-culture conditioning was proposed to overcome the lag period observed during H2 evolution from glycerol (Sabourin-Provost, and Hallenbeck, 2009). In this study, B. thuringiensis EGU45 culture were adapted by subjecting it to 2-10% of PG containing medium for 5-15 days. A comparison of H2 producing abilities of un-adapted and adapted B. thuringiensis EGU45 revealed that after 5 days of adaptation, observed H2 production varied from 550 mL at 2-5% to 250 mL at 10% v/v PG level (Table S1). The H2 yield was found to decline steadily from 0.452 mol/mol at 2% v/v glycerol to 0.053 mol/mol at 10% v/v glycerol. The positive impact of adaptation of B. thuringiensis EGU45 was evident, since these H2

6

yields were 1.46- 2.75 times higher compared to those recorded with un-adapted bacterial cultures. Subsequently, adaptation period was enhanced to 10 and 15 cycles. Nonetheless, there was no benefit in terms of H2 yields under these conditions.

3.1.2. Effect of Nitrogen source H2 metabolism in prokaryotes takes place through the activities of hydrogenase and nitrogenase enzymes. Nitrogenase releases H2 while transforming molecular nitrogen to ammonia i.e. during biological nitrogen fixation process (Kalia and Purohit, 2008). Previous studies on glycerol to H2 production have employed NH4Cl (1% w/v) as N source (Kivistö et al., 2010). H2 yield has been reported to be influenced by the presence of N sources in the form of ammonium and nitrate (Kumar et al., 1998), suggesting that the H2 production may vary as a function of the combination of different N sources and glycerol concentration. In the present study, 1% (w/v) NH4Cl as N source in 2% (v/v) glycerol gave an H2 yield of 0.452 mol/mol feed consumed. It was higher than 0.261 and 0.252 mol H2/mol glycerol consumed at 1% and 3% (v/v) of feed, respectively. A shift from NH4Cl as N source to NaNO3 resulted in improving H2 yield to 0.748 and 0.570 mol/mol glycerol consumed at 1% and 2% v/v feed material (Table 1). Here, H2 constituted 69.9 to 70.2% of the total biogas evolved.

A switchover from NaNO3 to NH4NO3 proved to be still better in supporting H2 production process. A total of 1170 mL of biogas having 785 mL H2 was observed at 2% (v/v) glycerol supplemented with 1% NH4NO3 as N source. An H2 yield of 0.646 mol/mol glycerol consumed was the best recorded so far with this bacterial strain. Use of a combination of these two N sources (NH4+ and NO3-) as NH4NO3 was as effective as NO3 alone in enhancing H2 production. The presence of NH4+ inhibits H2 evolution through nitrogenase enzyme (Kumar et al., 1998). Thus, addition of NO3 is likely to relieve nitrogenase from any potential inhibition by NH4+. It further appears that addition of NO3- suppresses the negative effect of NH4+ when both are present together as NH4NO3. These observations on H2 yields are clearly in agreement with those reported for Rhodopseudomonas palustris during photofermentative conditions (Sabourin-Provost and Hallenbeck, 2009). In this case, addition of NH4+ as (NH4)2SO4 to glycerol resulted in negatively influencing the H2 yields, which went down from 4.9 mol/mol glycerol consumed at 0 mM to 1.7 mol/mol glycerol at 4 mM concentration. In fact, NH4NO3 was found to support H2 production from glycerol, better than (NH4)2SO4 as N source (Jitrwung and Yargeau, 2011). A wide range of NH4NO3 (0-6 g/L) used as N source revealed that 1.5 g/L of NH4NO3 in 2% (v/v) glycerol was the best for H2 production, 0.85 mol H2/mol glycerol (Jitrwung and Yargeau, 2011). In this study, 1% NH4NO3 in 2% (v/v) glycerol resulted in 0.646 mol H2/mol of glycerol consumed. However, in the present study, NO3 alone as NaNO3 proved to be as good as NH4NO3.

7

This is the first report of using B. thuringiensis to metabolize glycerol to H2. During batch culture digestions, an H2 yield of 0.570 mol/mol glycerol consumed could be achieved at an incubation period of 4 days. Earlier, only B. coagulans has been reported to produce 2.2 mol H2/mol on PG. It is remarkable to note that, most of the studies carried out with various mesophilic bacterial cultures such as Clostridium, Enterobacter, Escherichia coli, Klebsiella etc. used minimal medium to achieve a yield 0.34-0.89 mol H2/mol (Kumar et al., 2015; Sarma et al., 2012). However, higher yields of 0.23-2.2 mol H2/mol have been achieved by replacing minimal medium with complex substrates including yeast extract, malt extract, tryptone etc. (Kumar and Das, 2001; Kivisto et al., 2010; Kumar et al., 2015; Sarma et al., 2012).

3.1.3. Continuous culture H2 production 3.1.3.1. From pure glycerol 3.1.3.1.1 Effect of hydraulic retention time The effect of HRT on H2 production at a feed level of 2% and 5% glycerol was monitored for 40 days. At an HRT of 4 days, B. thuringiensis EGU45 yielded 0.288-0.410 mol H2/mol glycerol consumed with 2% glycerol and 0.286-0.366 mol H2/mol glycerol consumed with 5% glycerol, which were better than FF bacterial cultures (0.283-0.146 mol H2/mol glycerol consumed). The efficiency of glycerol utilization was in the range of 75-80%. Although, the adapted cultures have been observed to perform quite well even at a higher glycerol level of 5% (v/v), immobilization improved the efficiency, where, BL performed as the best support material. Further reduction of HRT to 2 days resulted in an overall improved H2 production. With BL and CC as support material B. thuringiensis EGU45 was found to yield (mol H2/mol glycerol consumed) 0.237-0.273 and 0.113-0.116 at 2% and 5% glycerol, respectively (Table 2). At a reduced HRT of 2 days, cultures immobilized on ligno-cellulosic wastes (CC and BL) resulted in 1.26-1.45 fold higher H2 yield than FF cultures. Among the two HRTs tested, HRT of 4 days allowed a higher H2 yield of 0.410 mol/mol glycerol consumed in comparison to 0.273 mol/mol glycerol consumed at an HRT of 2 days, however, we preferred to opt for the later, in order to have a reduce reactor size on a larger scale.

3.1.3.1.2. Effect of inorganic nitrogen source Continuous culture H2 production by B. thuringiensis EGU45 from M-9 medium containing 2% PG was observed to be influenced by N sources. In free-floating bacterial cultures of B. thuringiensis EGU45 on M-9 medium containing 2% PG and 1% NH4Cl, H2 production rate was found to decline from 310 mL/day during the first 20 days to 220 mL/day during the next

8

phase i.e. till the end of the experiment. During this period, which is equal to 20 cycles (reactor volume), H2 yield was found to reduce slowly to a final value equal to 0.156 mol/mol glycerol with an overall average H2 yield of 0.188 mol/mol glycerol (Table 3). Similar trends in H2 production were observed with FF cultures of B. thuringiensis EGU45 fed on PG (2% v/v) supplemented with NaNO3 or NH4NO3 (1% w/v). Here, the average H2 yield (mol/mol glycerol consumed) of 0.218 with NaNO3 was higher in comparison to 0.099 with NH4NO3.

In order to enhance the efficiency of the H2 production process, generally adopted mechanism is to immobilize bacteria on different support materials (Patel et al., 2010, 2014, 2015). With M-9 medium supplemented with 1% NH4Cl B. thuringiensis EGU45 immobilized on PS proved to be the quite fruitful, with an H2 yield of 0.342 mol/mol glycerol consumed (Table 3). It was followed with H2 produced by B. thuringiensis EGU45 immobilized on BL. In these cases, H2 constituted 56-58.6% of the total biogas evolved. Overall, H2 yields improved by 1.82-fold compared to FF and 3.48-fold compared to PVC. In the case of M-9 medium having 2% (v/v) PG and 1% (w/v) NaNO3, B. thuringiensis EGU45 evolved a maximum of 640 mL of H2 equivalent to 0.386 mol/mol glycerol consumed with BL as support material. This performance of B. thuringiensis was followed by 0.32 mol H2/mol glycerol observed with GS as support material. A clear cut difference was noted with respect to H2 component of the biogas, which varied from 69.8 to 71.9% with NaNO3 compared to 56.0-67.9% observed on PG supplemented with NH4Cl as N source. Hence, a shift in N source from NH4Cl to NaNO3 proved helpful on two accounts: H2 yield and H2 component of the biogas. The metabolic shift towards higher H2 yield can be attributed to the impact of NO3 on counteracting the inhibitory effect of NH4+ on nitrogenase activity (Jitrwung and Yargeau, 2011; Kumar et al., 1998; Sabourin-Provost and Hallenbeck, 2009). Further change in N source to NH4NO3 did not help in significantly improving the H2 production process efficiency. H2 yields were observed to be 0.385 mol/mol glycerol with BL and 0.3 mol/mol glycerol with GS. Here also, H2 component of the biogas varied from 69.5 to 71.5%. It may be remarked that, during continuous culture studies, an H2 yield of 0.386 mol/mol glycerol consumed was recorded in a much shorter incubation period - an HRT of 2 days. A potential exists for further improving the H2 yields, since 36% of the feed still remains unutilized in the effluent. It may also be noted that compared to most studies carried out as batch cultures, the present study was quite sustainable, being steady, even after 20 cycles (1 reactor volume equal to 1 cycle) of continuous culture for 40 days, with BL as the best support for immobilizing bacteria. This contributed to an enhancement in H2 yield up to 1.43-fold. Such improvements in H2 production process have been accomplished by bacterial immobilization on ligno-cellulosic materials, especially BL (Patel et al., 2010, 2014, 2015).

9

3.1.3.2. From crude glycerol B. thuringiensis EGU45 adapted on PG was found to evolve 520-880 ml H2/ L feed (CG, 1.7% v/v + 1% NaNO3 w/v) under continuous culture conditions. It amounted to 0.273 to 0.393 mol/mol glycerol consumed. Here, H2 constituted 51.7- 73.8 % of the total biogas evolved. It is important to highlight that the H2 production was quite stable during the incubation period between 21 to 40 days (Fig. 1, Table 4). Overall, H2 yields were almost 2.3-fold higher in comparison to FF or PVC as support material. The highest H2 yield was recorded with BL as support material. It may be remarked the performance of B. thuringiensis EGU45 was quite similar on PG and CG, in terms of their H2 yields and the contribution of H2 to the total biogas evolved (Tables 3 and 4). In addition, the ability of B. thuringiensis EGU45 to utilize glycerol was between 45-55% of the feed, which was enhanced by immobilization up from 48% with FF cultures. The residual glycerol still leaves enough margin for further improving H2 production efficiency of the system. Another possibility to improve the process efficiency can be to metabolize the residual glycerol to PHA (Kumar et al., 2015).

Although the H2 yields are not very high, however, the use of B. thuringiensis has some associated benefits. In previous reports on continuous culture experiments with CG, H2 yields (mol/mol glycerol) have been revealed to be: 0.27 with Clostridium sp. (Fountoulakis and Manios, 2009), 0.77 with Clostridium pasteurianum (Lo et al., 2013) and 2.41 with Thermotoga maritima (Maru et al., 2012). It is important to state that the use of Clostridium as inoculum is a costly affair as it demands energy to maintain strictly anaerobic conditions. Similarly, the use of a thermophilic bacterium such as T. maritima is also energy intensive and hence likely to be more costly than mesophilic organisms. Hence, it may be pertinent to state that Bacillus is a more promising candidate, being mesophilic and can operate under fermentative conditions.

4. Conclusions The adaptation of B. thuringiensis EGU45 allowed us to successfully deploy it towards bioconversion of glycerol to H2. The feasibility of operating the process on a large scale was further demonstrated by running it as a continuous culture with bacteria immobilized on support matrix. This process established on PG was effectively demonstrated on CG as well. Use of defined mixed bacterial cultures can prove effective in further increasing the efficiency of the process.

10

Acknowledgements The authors wish to thank the Director of CSIR-Institute of Genomics and Integrative Biology (IGIB), Delhi, CSIR-WUM (ESC0108), for providing necessary funds and facilities. PK is thankful to CSIR for granting research fellowships. Part of this research work was also supported by 2014 KU Brain Pool of Konkuk University.

References [1] Dasari, M.A., Kiatsimkul, P.P., Sutterlin, W.R., Suppes, G.J., 2005. Low-pressure hydrogenolysis of glycerol to propylene glycol. Appl. Catal. A. Gen. 281, 225-231. [2] Donkin, S.S., 2008. Glycerine from biodiesel production: the new corn for dairy cattle. Braz. J. Anim. Sci. 37, 280-286. [3] Forsberg, C.W., 1987. Production of 1, 3-propanediol from glycerol by Clostridium acetobutylicum and other Clostridium species. Appl. Environ. Microbiol. 53, 639-643. [4] Huma, N., Shankar, P., Kushwah, J., Bhushan, A., Joshi, J., Mukherjee, T., Raju, S.C., Purohit, H.J., Kalia, V.C., 2011. Diversity and polymorphism in AHL-lactonase gene (aiiA) of Bacillus. J. Microbiol. Biotechnol. 21, 1001– 1011. [5] Jitrwung, R., Yargeau, V., 2011. Optimization of media composition for the production of biohydrogen from waste glycerol. Int. J. Hydrogen Energy 36, 9602-9611. [6] Kalia, V.C., Purohit, H.J., 2008. Microbial diversity and genomics in aid of bioenergy. J. Ind. Microbiol. Biotechnol. 35, 403-419. [7] Kivistö, A., Santala, V., Karp, M., 2010. Hydrogen production from glycerol using halophilic fermentative bacteria. Bioresour. Technol. 101, 8671-8677. [8] Kumar, A., Jain, S.R., Kalia, V.C., Joshi, A.P., 1998. Effect of some physiological factors on nitrogenase activity and nitrogenase mediated hydrogen evolution by mixed microbial culture. Biochem. Mol. Biol. Int. 45, 245-253. [9] Kumar, N., Das, D., 2001. Continuous hydrogen production by immobilized Enterobacter cloacae IIT-BT08 using lignocellulosic materials as solid matrices. Enzyme Microb. Technol. 29, 280-287. [10] Kumar, P., Patel, S.K.S., Lee, J.K., Kalia, V.C., 2013. Extending the limits of Bacillus for novel biotechnological applications. Biotechnol. Adv. 31, 1543-1561. [11] Kumar, P., Pant, D.C., Mehariya, S., Sharma, R., Kansal, A., Kalia, V.C., 2014a. Ecobiotechnological strategy to enhance efficiency of bioconversion of wastes into hydrogen and methane. Indian J. Microbiol. 54, 262-267.

11

[12] Kumar, P., Singh, M., Mehariya, S., Patel, S.K.S., Lee, J.K., Kalia, V.C., 2014b. Ecobiotechnological approach for exploiting the abilities of Bacillus to produce co-polymer of polyhydroxyalkanoate. Indian J. Microbiol. 54, 151-157. [13] Kumar, P., Mehariya, S., Ray, S., Mishra, A., Kalia, V.C., 2015. Biodiesel industry waste: a potential source of bioenergy and biopolymers. Indian J. Microbiol. 55, 1-7. [14] Law, K.H., Cheng, Y.C., Leung, Y.C., Lo, W.H., Chua, H., Yu, H.F., 2003. Construction of recombinant Bacillus subtilis strains for polyhydroxyalkanoates synthesis. Biochem. Eng. J. 16, 203-208. [15] Lo, Y.C., Chen, X.J., Huang, C.Y., Yuan, Y.J., Chang, J.S., 2013. Dark fermentative hydrogen production with crude glycerol from biodiesel industry using indigenous hydrogen-producing bacteria. Int. J. Hydrogen Energy 38, 15815– 15822. [16] Maru, B.T., Bielen, A.A.M., Kengen, S.W.M., Constantí, M., Medina, F., 2012. Biohydrogen production from glycerol using Thermotoga spp. Energy Procedia 29, 300-307. [17] Patel, S.K.S., Purohit, H.J., Kalia, V.C., 2010. Dark fermentative hydrogen production by defined mixed microbial cultures immobilized on lignocellulosic waste materials. Int. J. Hydrogen Energy 35, 10674-10681. [18] Patel, S.K.S., Singh, M., Kumar, P., Purohit, H.J., Kalia, V.C., 2012. Exploitation of defined bacterial cultures for production of hydrogen and polyhydroxybutyrate from pea-shells. Biomass Bioenergy 36, 218-225. [19] Patel, S.K.S., Kalia, V.C., 2013. Integrative biological hydrogen production: an overview. Indian J. Microbiol. 53, 3-10. [20] Patel, S.K.S., Kumar, P., Mehariya, S., Purohit, H.J., Lee, J.K., Kalia, V.C., 2014. Enhancement in hydrogen production by co-cultures of Bacillus and Enterobacter. Int. J. Hydrogen Energy 39, 14663-14668. [21] Patel, S.K.S., 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, 136-141. [22] Porwal, S., Kumar, T., Lal, S., Rani, A., Kumar, S., Cheema, S., Purohit, H.J., Sharma, R., Patel, S.K.S., Kalia, V.C., 2008. Hydrogen and polyhydroxybutyrate producing abilities of microbes from diverse habitats by dark fermentative process. Bioresour. Technol. 99, 5444-5451. [23] Sabourin-Provost, G., Hallenbeck, P.C., 2009. High yield conversion of a crude glycerol fraction from biodiesel production to hydrogen by photofermentation. Bioresour. Technol. 100, 3513-3517. [24] Sarma, S.J., Brar, S.K., Sydney, E.B., Le Bihan, Y., Buelna, G., Soccol, C.R., 2012. Microbial hydrogen production by bioconversion of crude glycerol: A review. Int. J. Hydrogen Energy 37, 6473-6490.

12

[25] Sarma, S.J., Brar, S.K., Le Bihan, Y., Buelna, G., Rabeb, L., Soccol, C.R., Naceur, M., Rachid, B., 2013. Evaluation of different supplementary nutrients for enhanced biohydrogen production by Enterobacter aerogenes NRRL B 407 using waste derived crude glycerol. Int. J. Hydrogen Energy 38, 2191-2198. [26] Schallmey, M., Singh, A., Ward, O.P., 2004. Developments in the use of Bacillus species for industrial production. Can. J. Microbiol. 50:1–17. [27] Sittijunda, S., Reungsang, A., 2012. Biohydrogen production from waste glycerol and sludge by anaerobic mixed cultures. Int. J. Hydrogen Energy 37, 13789-13796. [28] Valappil, S.P., Boccaccini, A.R., Bucke, C., Roy, I., 2007. Polyhydroxyalkanoates in Gram-positive bacteria: insight from the genera Bacillus and Streptomyces. Antonie van Leeuwenhoek 91, 1-17. [29] Veziroğlu, T.N., 1995. Twenty years of hydrogen movement 1974-1994. Int. J. Hydrogen Energy 20, 1-7. [30] Wang, K., Wang, X., Xizhen, G., Tian, P., 2012. Heterologous expression of aldehyde dehydrogenase from Saccharomyces cerevisiae in Klebsiella pneumoniae for 3-hydroxypropionic acid production from glycerol. Indian J. Microbiol. 52, 478-483. [31] Westers, L., Westers, H., Quax, W.J., 2004. Bacillus subtilis as cell factory for pharmaceutical proteins: a biotechnological approach to optimize the host organism. Biochim. Biophys. Acta 1694, 299-310.

13

Figure caption Fig. 1. Continuous culture biological hydrogen production on crude glycerol by Bacillus thuringiensis EGU45 immobilized on different ligno-cellulosic support materials. Feed: 500 mL M-9 supplemented with 2% crude glycerol (HRT 2 days) Inoculum: Bacterial culture adapted on 2% pure glycerol (10 µg cell protein/mL). FF: Free floating; PVC: Polyvinylchloride tubes; BL: Banana leaves; CC: Coconut coir; GS: Groundnut shells; PS: Pea shells Values are based on three experimental replicates.

14

800 700 Hydrogen (mL)

600

FF PVC BL CC GS PS

500 400 300 200 100 0 0

5

10

15

20 25 Days after incubation

Fig. 1.

15

30

35

40

Table 1 Effect of nitrogen sources on hydrogen production from pure glycerol by Bacillus thuringiensis under batch culture.

Glycerola Biogasb (%)

(mL)

Hydrogen Volc

%

Yieldd

NH4Cl 1

255

160

63.0

0.261

2

710

550

77.5

0.452

3

680

461

67.6

0.252

NaNO3 1

650

455

70.2

0.748

2

990

695

69.9

0.570

3

870

575

65.7

0.314

NH4NO3 1

510

335

65.3

0.547

2

1170

785

67.0

0.646

3

840

545

65.1

0.299

a

M-9 medium supplemented with different nitrogen sources (1% w/v)

b

Biogas: mixture of H2 + CO2;

c

Volume of H2 in mL;

d

mol/mol of glycerol consumed.

Values are based on two replicates and standard error was less than 5%.

16

Table 2 Effect of incubation period and immobilization on continuous culture hydrogen production by Bacillus thuringiensis EGU45.

M-9 medium + Support

2% Glycerol

materiala

5% Glycerol

Hydrogen (H2) Volb

%

Yieldc

Vol

%

Yield

HRTd: 2 Days FF

265

67.9 0.188

215

54.3 0.046

CC

420

58.3 0.237

595

57.4 0.116

BL

510

58.6 0.273

605

61.1 0.113

HRT: 4 Days FF

200

65.1 0.283

240

52.2 0.146

CC

270

62.4 0.288

550

63.6 0.286

BL

365

63.8 0.410

680

61.4 0.366

a

Ligno-cellulosic support material (10% v/v) - FF: Free floating; BL: Banana leaves; CC: Coconut coir;

b

Volume of H2 in mL;

c

mol/mol of glycerol consumed;

d

Hydraulic retention time

Medium supplemented with pure glycerol and 1% (w/v) NH4Cl. Bacterial culture adapted on 2% and 5% pure glycerol was used as inoculum (10 µg cell protein/mL).

17

Table 3 Effect of nitrogen sources on continuous culture hydrogen production by immobilized Bacillus thuringiensis EGU45 from pure glycerol.

Support Material

a

Biogasb

Hydrogen

(mL)

Vol

c

Residual d

Yield Glycerol

%

(% v/v)

NH4Cl (1% w/v) FF

390 ± 16.2

265 ± 12.9

67.9

0.188

0.84

PVC

225 ± 10.2

135 ± 5.5

60.0

0.098

0.86

PVC+BL

870 ± 37.6

510 ± 23.8

58.6

0.273

0.46

PVC+CC

720 ± 33.1

420 ± 19.7

58.3

0.237

0.54

PVC+GS

570 ± 26.3

345 ± 14.7

60.5

0.204

0.61

PVC+PS

920 ± 43.1

515 ± 21.4

56.0

0.342

0.76

NaNO3 (1% w/v) FF

445 ± 21.3

320 ± 11.9

71.9

0.218

0.79

PVC

265 ± 11.5

185 ± 7.1

69.8

0.131

0.84

PVC+BL

910 ± 39.9

640 ± 22.4

70.3

0.386

0.64

PVC+CC

675 ± 26.7

475 ± 18.7

70.3

0.303

0.71

PVC+GS

745 ± 36.1

530 ± 19.6

71.1

0.320

0.64

PVC+PS

555 ± 24.7

390 ± 11.3

70.3

0.268

0.80

NH4NO3 (1% w/v)

a

FF

190 ± 6.3

135 ± 4.3

71.5

0.099

0.87

PVC

260 ± 12.5

190 ± 8.1

72.5

0.144

0.92

PVC+BL

850 ± 35.8

595 ± 23.3

70.2

0.385

0.73

PVC+CC

560 ± 21.1

400 ± 16.7

71.3

0.268

0.77

PVC+GS

660 ± 23.9

480 ± 13.9

72.7

0.300

0.68

PVC+PS

475 ± 22.9

330 ± 15.8

69.5

0.228

0.81

M-9 medium containing 2% (v/v) pure glycerol and ligno-cellulosic support material (10% v/v) - FF: Free floating;

PVC: Polyvinylchloride tubes; BL: Banana leaves; CC: Coconut coir; GS: Groundnut shells; PS: Pea shells b

Biogas: mixture of H2 + CO2

c

Volume of H2 in mL

d

mol/mol of glycerol consumed

Values are based on 70 observations made over a period of 40 days of fermentation in three replicates at a hydraulic retention time of 2 days.

18

Table 4 Effect of crude glycerol on continuous culture hydrogen production from Bacillus thuringiensis EGU45 immobilized on ligno-cellulosic materials.

Support

Biogasb

Materiala

(mL)

Hydrogen Volc

Residual

% Yieldd Glycerol (% v/v)

NaNO3 (1% w/v)

a

FF

240 ± 11.2

175 ± 7.9

72.7

0.173

0.87

PVC

210 ± 9.7

145 ± 8.6

68.2

0.170

1.00

PVC+BL

595 ± 27.3

440 ± 20.8 73.8

0.393

0.78

PVC+CC

535 ± 25.8

350 ± 15.8 65.1

0.306

0.76

PVC+GS

455 ± 20.5

320 ± 12.7 70.2

0.312

0.86

PVC+PS

500 ± 21.3

260 ± 11.1 51.7

0.273

0.92

FF: Free floating; PVC: Polyvinylchloride tubes; BL: Banana leaves;

CC: Coconut coir; GS: Groundnut shells; PS: Pea shells b

Biogas: mixture of H2 + CO2;

c

Volume of H2 in mL;

d

mol/mol of glycerol consumed

Values are based on observations made over a period between 21 to 40 days of fermentation in three replicates at a hydraulic retention time of 2 days.

19

Dark fermentative bioconversion of glycerol to hydrogen by Bacillus thuringiensis Prasun Kumara, Rishi Sharmaa, Subhasree Raya, Sanjeet Mehariyaa, Sanjay K.S. Patelb, Jung-Kul Leeb, Vipin C. Kaliaa* a

Microbial Biotechnology and Genomics, CSIR–Institute of Genomics and Integrative Biology (IGIB), Delhi University Campus, Mall Road, Delhi-110007, India. b Department of Chemical Engineering, Konkuk University, 1 Hwayang-Dong, Gwangjin-Gu, Seoul, Korea, 143-701

Highlights:
Bioconversion of crude glycerol into H2 by Bacillus thuringiensis
Adaptation of Bacillus thuringiensis on glycerol improved H2 yield by 1.46-fold
H2 evolution of 0.393 mol/mol of crude glycerol consumed by immobilized bacteria

20