Photo-fermentative bacteria aggregation triggered by L-cysteine during hydrogen production
© Xie et al.; licensee BioMed Central Ltd. 2013
Received: 26 November 2012
Accepted: 29 April 2013
Published: 3 May 2013
Skip to main content
© Xie et al.; licensee BioMed Central Ltd. 2013
Received: 26 November 2012
Accepted: 29 April 2013
Published: 3 May 2013
Hydrogen recovered from organic wastes and solar energy by photo-fermentative bacteria (PFB) has been suggested as a promising bioenergy strategy. However, the use of PFB for hydrogen production generally suffers from a serious biomass washout from photobioreactor, due to poor flocculation of PFB. In the continuous operation, PFB cells cannot be efficiently separated from supernatant and rush out with effluent from reactor continuously, which increased the effluent turbidity, meanwhile led to increases in pollutants. Moreover, to replenish the biomass washout, substrate was continuously utilized for cell growth rather than hydrogen production. Consequently, the poor flocculability not only deteriorated the effluent quality, but also decreased the potential yield of hydrogen from substrate. Therefore, enhancing the flocculability of PFB is urgent necessary to further develop photo-fermentative process.
Here, we demonstrated that L-cysteine could improve hydrogen production of Rhodopseudomonas faecalis RLD-53, and more importantly, simultaneously trigger remarkable aggregation of PFB. Experiments showed that L-cysteine greatly promoted the production of extracellular polymeric substances, especially secretion of protein containing more disulfide bonds, and help for enhancement stability of floc of PFB. Through formation of disulfide bonds, L-cysteine not only promoted production of EPS, in particular the secretion of protein, but also stabilized the final confirmation of protein in EPS. In addition, the cell surface elements and functional groups, especially surface charged groups, have also been changed by L-cysteine. Consequently, absolute zeta potential reached a minimum value at 1.0 g/l of L-cysteine, which obviously decreased electrostatic repulsion interaction energy based on DLVO theory. Total interaction energy barrier decreased from 389.77 KT at 0.0 g/l of L-cysteine to 127.21 kT at 1.0 g/l.
Thus, the strain RLD-53 overcame the total energy barrier and flocculated effectively. After a short settlement, the biomass rush out will be significantly reduced and the effluent quality will be greatly improved in the continuous operation. Furthermore, aggregation of PFB could enable high biomass hold-up of photobioreactor, which allows the photobioreactor to operate at low hydraulic retention time and high organic loading rate. Therefore, the described flocculation behaviour during photo-hydrogen production is potentially suitable for practicable application.
Global growing concerns about energy shortages and the environmental pollution have led to worldwide use of renewable energy. Hydrogen is considered as a viable energy carrier for the future which could play an important role in the reduction of emissions of greenhouse gases [1, 2]. Recently, biological hydrogen production processes, especially photo-fermentative hydrogen production by PFB has been attracting more and more attention, as it utilizes various renewable sources like biomass and sunlight to produce an ideal, renewable and carbon-free energy for the future . However, it should be realized that most of photo-fermentative processes are based on suspended culture [4–6] in which it is difficult to achieve high biomass concentration, effective retention and separation of PFB biomass, resulting from poor flocculation of PFB . For steady-state operation of photobioreactor, due to the poor flocculability, PFB cells cannot be efficiently separated from supernatant and rush out with effluent from reactor continuously. This increased the effluent turbidity, meanwhile led to increases in pollutants like chemical oxygen demand, total nitrogen, and total phosphate, causing poor effluent water quality. Furthermore, to replenish the biomass washout, substrate was continuously utilized for cell growth rather than hydrogen production [8, 9]. Thus, the poor flocculability not only deteriorated the effluent quality, but also decreased the potential yield of hydrogen from substrate. Therefore, enhancing the flocculability of PFB is urgent necessary to further develop photo-fermentative process.
Previous studies also tried to isolate self-flocculated PFB or enhance flocculation of PFB, but successful case was rare. Watanabe  first and only reported that photosynthetic bacteria Rhodovulum sp. PS88 has a self-flocculating activity. And high density cell culture was obtained under continuous cultivation in a single-tower fermenter . However, there was no report about photo-hydrogen production and flocculation mechanism of Rhodovulum sp. PS88. According to the DLVO theory, the PFB, Rhodopseudomonas acidophila, could not overcome the total energy barrier to flocculate effectively, because contribution of van der Waals interaction energy to the total interaction energy could be neglected resulting from the small effective Hamaker constant (2.27×10-23 J) . As a result, R. acidophila could not overcome the total energy barrier to flocculate effectively. So far, the information about PFB could flocculate and simultaneously improve hydrogen production have been not yet reported, and effective method and detailed mechanism of flocculation in photo-fermentation hydrogen production is still lacking.
In this work, we first time found that the L-cysteine induced the obvious bioflocculation of Rhodopseudomonas faecalis RLD-53 and at the same time promoted hydrogen production. Traditionally, flocculability of biological cells highly depended on the extracellular polymeric substances (EPS) , bacterial surface characteristics  and electrolyte concentration . However, L-cysteine is unique natural amino acids containing a thiol group, which could form disulfide bond. Disulfide bonds are crucial to the folding and stability of many proteins [16, 17], usually proteins secreted to the extracellular medium. As a predominant component in EPS, proteins have been demonstrated to play a crucial role in the bacterial aggregation [18, 19]. Therefore, the mechanism of aggregation triggered by L-cysteine was explored through combination biological function of L-cysteine and traditional flocculation theory. EPS, surface properties and zeta potential of PFB were investigated for better understanding flocculation characteristics of strain RLD-53 under different concentration of L-cysteine. Effect of disulfide bonds on components of EPS production and conformational changes of proteins in EPS were also determined. Furthermore, contribution of specific EPS protein conformation and cell surface functional groups to bacterial aggregation were further discussed. Finally, the DLVO theory was used to evaluate the flocculability of R. faecalis RLD-53.
Effect of L-cysteine concentration on hydrogen production kinetics and nitrogenase activity
H2yield (mol H2/mol acetate)
H max (ml/l)
R max (ml/l/h)
Nitrogenase activity (nmol C2H4/ml/h)
In order to further demonstrate above results, acetylene reduction was used to determine the activity of nitrogenase, which catalysed the hydrogen production in photofermentation . As shown in Table 1, nitrogenase activities increased with the L-cysteine concentration, reached maximum (1374 nmol C2H4/ml/h) at 1.0 g/l. However, with further increase of L-cysteine to 1.5 g/l, the nitrogenase activity sharply decreased to765 C2H4/ml/h. Thiol group from L-cysteine is a key part of active sites in nitrogenase, which play an important role in structure and function of nitrogenase [20, 21]. In chemical evolution of a nitrogenase model, the ratio of thiol and molybdenum significantly influenced the catalytic activity, and the maximum catalytic activity was obtained at ratio of 1:1 . In this study, nitrogenase activity was enhanced by increasing of thiol from L-cysteine, but strongly depressed by excessive L-cysteine.
In addition, humic substances also increased significantly (Figure 4). Humic substances are a natural organic matter, resulting from the biodegradation of dead biomass, which are resistant to degradation . In the pure culture of photofermentative bacteria, humic substances mainly came from the dead cell decomposition. In free cell culture, cell debris and humic substances from the dead cell decomposition may disperse into the culture broth. However, after the bioflocculation formation, humic substances from dead cell decomposition may be retained in the EPS matrix. As a result, the humic substances also significantly increase, due to the floc formation caused by L-cysteine.
Conformation changes of EPS proteins from R. faecalis RLD-53 at different concentration of L-cysteine
Aggregated strands (%)
Random coil (%)
3-Turn helix (%)
Antiparallel β-sheet/ aggregated strands (%)
With increasing protein and EPS covering on the cell surface, the surface elemental composition and functional groups could be greatly influenced by L-cysteine. The surface elements and functional groups of R. faecalis RLD-53 were studied by X-ray photoelectron spectroscopy (XPS) (Additional file 1: Figure S1), which detected the outermost molecular layers (mainly EPS) of the cell surface (2–5 nm). The major peaks in the spectra identified by XPS were the C 1s and O 1s, and N 1s peak, with minor peaks of P, Na, Cl and Si. The functional groups on the cell surfaces were illustrated by high-resolution XPS spectra of the C 1s, O 1s and N 1s region in Additional file 2: Figure S2. The C 1s spectra were resolved into four individual component groups: CG1, 284.6 ev, C-(C, H) mainly from hydrocarbons; CG2, 286.2 eV, C-(O, N) from proteins and alcohols; CG3, 287.8 eV, C=O or O-C-O from carboxylate, carbonyl, amide, acetals, or hemiacetals, and CG4, 289.2 eV, O=C-OH and O=C-OR commonly from uronic acids. The O 1s peak was decomposed into two peaks, OG1, 531.3 eV, O=C from carboxylate, carbonyl, ester, or amide, and OG2, 532.7 eV, O-(C, H) from hydroxide, acetal, and hemiacetal. The N 1s peak was also resolved into two component peaks, NG1, 399.9 eV, O=C-NH-R from amines and amides, and NG2, 401.3 eV, C-NH2 mainly from basic amino acids. The percentages of surface functional groups were determined from the XPS peaks area after subtraction of a linear background (Table 2). The results showed that the functional groups on the cell surface of R. faecalis RLD-53 were significantly affected by L-cysteine concentration.
Results of the high-resolution XPS analysis of the C 1s, O 1s and N 1s peak region from cell surface
C 1s (%)
O 1s (%)
N 1s (%)
DLVO theory has been widely applied as both qualitative and quantitative models to explain microbial adhesion and aggregation [39, 40]. Here, DLVO theory was applied to predict the potential energy barrier that hindered aggregation of R. faecalis RLD-53 at different concentration of L-cysteine.
The poor flocculability of photosynthetic H2-producing bacterium, R. acidophila, was attributed to its inherent surface characteristics . The effective Hamaker constant between R. acidophila and water was only 2.27×10-23 J, resulting in the negligible contribution of van der Waals interaction energy (W LW ) to the total interaction energy. As a result, the interaction energy barrier between cells was up to 1665 KT in 0.01 mol/l NaCl solutions. Consequently, the bacterial cells could not overcome the total energy barrier to flocculate effectively. In this study, the effective Hamaker constant between R. faecalis RLD-53 and water was 5.54×10-21 J at 1.0 g/l of L-cysteine, suggesting that van der Waals interaction energy was important for total interaction energy. In addition, the electrostatic repulsive energy (W EL ) decreased with decreasing of absolute zeta potential, resulting from changes of surface charged groups caused by L-cysteine. The interaction energy barrier between cells of R. faecalis RLD-53 was 127.21 KT. Therefore, R. faecalis RLD-53 flocculate effectively at 1.0 g/l of L-cysteine.
In this work, L-cysteine was found to promote the effective flocculation and photo-hydrogen production of R. faecalis RLD-53. This finding suggested L-cysteine can be applied as flocculant for continuous photo-hydrogen production. Results showed that proper L-cysteine concentration (1g/l) improved flocculability and hydrogen productivity of R. faecalis RLD-53. The reasons of flocculation also were analysed. Through formation of disulfide bonds, L-cysteine not only promoted production of EPS, in particular the secretion of protein, but also stabilized the final confirmation of protein in EPS. Research also noted that the cell surface covered by EPS have been changed by L-cysteine, thus absolute zeta potential decreased with L-cysteine and reached minimum at 1.0 g/l, which greatly decreased electrostatic repulsion interaction energy based on DLVO theory. Further analysis indicated that total interaction energy barrier decreased from 389.77 KT at 0.0 g/l of L-cysteine to 127.21 kT at 1.0 g/l. This led to the R. faecalis RLD-53 overcome the total interaction energy barrier and flocculate effectively. Therefore, forming stable floc caused by L-cysteine offers great advantages for the realization of enhancing production yield and scale-up application in bio-hydrogen production.
A better understanding the flocculation behaviour of PFB triggered by L-cysteine not only could help the design of subsequent hydrogen production process by flocculation of PFB, but also might favour the further understanding of the bioflocculation mechanism.
The photo-hydrogen producer used in this study was Rhodopseudomonas faecalis RLD-53 . Acetate was used as the sole carbon source, and glutamate was used as nitrogen source in the medium for hydrogen production. The culture medium of strain RLD-53 was prepared as described in previous report .
The batch culture experiments were carried out in triplicate with 80 ml of the medium in 100 ml sealed reactors and filled with argon to maintain anaerobic conditions .The reactors were autoclaved at 121°C for 15 min. R. faecalis RLD-53 in the mid-exponential growth phase was inoculated into reactors. The light intensity on the outside surface of the reactors was maintained at 150 W/m2 by incandescent lamps (60 W). The reactors were stirred at 120 rpm at constant temperature of 35°C.
Where t is culture time (h); H is cumulative H2 production (ml/l medium); H max is maximum cumulative H2 production (ml/l medium); e=2.71828; R max is maximum H2 production rate (ml/l/h); and λ is the lag-phase time (h).
Surface morphology of the bioflocculation samples was evaluated by a scanning electron microscope . The bioflocculation samples were fixed with 2.5% glutaraldehyde and left for 1.5 h in a 4°C refrigerator. The samples were gently washed with phosphate buffer solution and then dehydrated by successive passages through 50%, 70%, 80%, 90%, and 100% ethanol. Each rinsing and dehydrating step took 10 min. The samples were refreeze dried (Hitachi E-2030, Japan) for 4 h, subsequently coated with gold powder by Sputter Coater (Hitachi E-1010, Japan) and finally attached on to the microscope supports with silver glue. Scanning electron microscope images were taken at 5 kV using an SEM (Hitachi S-3400N, Japan).
Surface elements concentrations and functional groups on cell surface were determined by the XPS method, which detected the outermost molecular layers (mainly EPS) of the cell surface (2–5 nm) . After 72 h cultivation, R. faecalis RLD-53 cultured at different L-cysteine concentrations were harvested by centrifugation at 12000 rpm for 10 min and washed twice with double distilled water. Collected cell samples were placed in a freeze-dryer for about 48 h (until freeze-dried). XPS was carried out on a PHI-5600 equipped with a monochromatic Al Kα source and data acquisition and processing were conducted using the PC Access ESCA version 7.2A program. The anode voltage and power were 12.5 kV and 250 W, respectively. The pressure in analysis chamber was maintained at 10-9 Torre during each measurement. All binding energies were referenced to the C 1s neutral carbon peak at 284.6 eV. Spectra were analysed using XPSPeak software (Version 4.1).
EPS was extracted using cation exchange resin  (Dowex Marathon C, 20–50 mesh, sodium form, Fluka 91973). The bacterial cells were collected by centrifugation at 12000 rpm for 10 min. And then the cells were washed twice with 0.9% NaCl solution. Subsequently, the cells were re-suspended in ddH2O and transferred to an extraction beaker. And then the beaker was added resin (70 g/g-VSS) and stirred at 600 rpm for 12 h at 4°C. The samples was centrifuged at 12000 g for 30 min followed by filtration using a 0.45 μm cellulose acetate membrane to remove resin, microorganisms, and residual debris to obtain an EPS sample for further analysis.?>
The FTIR spectra of EPS samples were determined using a Fourier transform infrared spectrophotometer (Spectrum One-B, Perkin Elmer, U.S.). The freeze-dried EPS samples were ground with infrared grade KBr and press into pellets and used for FTIR measurement. For each sample, 350–400 scans were collected over the spectral range of 400–4000 cm-1 at a resolution of 4 cm-1. The protein conformation was analysed from the amide I region . Component peaks were fitted with Gaussian band profiles using the frequencies of the components deduced from the second derivatives.
Application of the DLVO approach required the surface thermodynamic parameters, which was determined by measuring the contact angles and using the Lifshitz van der Waals acid–base approach .
Where A BLB is the effective Hamaker constant. H is the separation distance between the cells. R is the cell radius of R. faecalis RLD-53, determined by the Malvern Mastersizer 2000 (Malvern Instruments Ltd., UK). ψ s and κ represent the stern potential and inverse of the Debye length respectively, which are related to the electric double layer interaction W EL . ψ s could be replaced by zeta potential measurement and κ can be calculated from different electrolyte concentrations.
Light intensity was measured at the surface of reactor with solar power meter TENMARS TM-207 (Tenmars Electronics CO., LTD., Taiwan, China). Biogas was sampled from the head space of the photobioreactor by using gas-tight glass syringes and hydrogen content was determined by using a gas chromatograph (Agilent 4890D, Agilent Technologies, USA). The gas chromatograph column was Alltech Molesieve 5A 80/100. Argon was used as the carrier gas with a flow rate of 30 ml/min. Temperatures of the oven, injection, detector, and filament were 35°C, 120°C, 120°C, 140°C, respectively. Residual acetate in culture broth was determined using a second gas chromatograph (Agilent 7890 A, Agilent Technologies, USA) equipped with a flame ionization detector. The liquor samples were firstly centrifuged at 12,000 rpm for 5 min, and filtered through a 0.2 μm membrane before free acids were analyzed. The operational temperatures of the injection port, the column and the detector were 220, 190 and 220°C, respectively. Nitrogen was used as carrier gas at flow rate of 50 ml/min.
Whole-cell nitrogenase activity was assayed by acetylene reduction following the procedure in our previous report . The polysaccharide content in EPS was determined by the anthrone method  using glucose as a standard. The protein and humic substance in EPS were measured followed the modified Lowry method  using bovine serum albumin and humic acid (Fluka Chemical Corp., USA) as the respective standards. The nucleic acid content was measured by the diphenylamine colorimetric method  using fish DNA as the standard. Thiol (SH) and disulfide bond (SS) in EPS were determined using 5-5′-dithio-bis (2-nitrobenzoic acid) (DTNB) according to the method of Ellman  and the procedure reported by Kalapathy et al. .
Extracellular polymeric substances
Fourier Transform Infrared Spectroscopy
Scanning electron microscope
X-ray photoelectron spectroscopy
5-5′-dithio-bis (2-nitrobenzoic acid).
This research was supported by the National Natural Science Foundation of China (No. 51106040 and 51178140), Shanghai Tongji Gao Tingyao Environmental Science & Development Foundation, China Postdoctoral Science Foundation (No. 2012T50366), Heilongjiang Postdoctoral Financial Assistance (No. LBH-Z11120), Project 51121062 (National Creative Research Groups), 863 Program (No. 2011AA060905), the open Project of State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology (No. HC201212).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.