|Home | About | Journals | Submit | Contact Us | Français|
Clostridium tyrobutyricum ATCC25755 has been reported as being able to produce significant quantities of hydrogen. In this study, the exo-inulinase encoding gene cloned from Paenibacillus polymyxa SC-2 was into the expression plasmid pSY6 and expressed in the cells of C. tyrobutyricum. The engineered C. tyrobutyricum strain efficiently fermented the inulin-type carbohydrates from Jerusalem artichoke, without any pretreatment being necessary for the production of hydrogen. A comparatively high hydrogen yield (3.7mol/mol inulin-type sugar) was achieved after 96h in a batch process with simultaneous saccharification and fermentation (SSF), with an overall volumetric productivity rate of 620±60mL/h/L when the initial total sugar concentration of the inulin extract was increased to 100g/L. Synthesis of inulinase in the batch SSF culture was closely associated with strain growth until the end of the exponential phase, reaching a maximum activity of 28.4±0.26U/mL. The overall results show that the highly productive and abundant biomass crop Jerusalem artichoke can be a good substrate for hydrogen production, and that the application of batch SSF for its conversion has the potential to become a cost-effective process in the near future.
Hydrogen production technologies have received growing research attention all over the world in recent years, due to the significant increase in hydrogen demand as an alternative energy source as well as a feedstock for the production of chemicals and food-oil industries. Nowadays, approximately 90% of the current hydrogen production is obtained by the conversion of fossil fuels, which requires a mass of energy and emits huge greenhouse gasses1. In contrast to chemical methods, biological processes using carbohydrate-based substrates for hydrogen production, such as indirect photolysis, photo-fermentation, and dark fermentation, are economical and environmentally friendly2, 3. Extensive researches in the past two decades have been focused on the promising route of biohydrogen production via dark fermentation using pure strains in batch and continuous systems, which can be operated at ambient temperature and pressure with minimal energy requirements4. Clostridia are the most significant microorganisms in anaerobic hydrogen fermentation5. So far, a wide range of organic compounds, such as glucose, xylose, ribose, glycerol, hydrolysates of diverse starch sources, and sugar beet molasses, have been used as substrates for hydrogen fermentation using Clostridium species (Table 1). In general, the carbon source is usually used in relatively high concentrations compared with other media components, and thus contributes most to the total cost of raw materials. Considering the cost-efficiency of fermentative hydrogen production, the exploitation of non-food residues originating from agro-industrial activities as carbon sources has therefore been strongly stimulated.
Jerusalem artichoke is one of the least expensive and most widely available non-grain crops6. It shows high resistance to frost and various plant diseases, which resulted in its wide cultivation in northern China for environmental protection6. Fresh Jerusalem artichoke tubers are rich in carbohydrates, of which 70–90% (w/w) is inulin. Moreover, Jerusalem artichoke has been widely utilized as a biotechnological feedstock in recent years, including in the microbial production of lactic acid7, 8, propionic acid9, butyric acid10, ethanol11, 2,3-butanediol12, 13, lipids14 as well as single-cell protein15. However, because most wild-type microorganisms cannot secrete inulinase, an acidic or enzymatic hydrolysis pretreatment of the inulin was traditionally required prior to fermentation, which significantly increases the cost of the process11. Inulin consists of linear chains of β-2,1-linked D-fructofuranose molecules terminated by a glucose residue through a sucrose-type linkage at the reducing end. Inulinases are fructofuranosyl hydrolases, the general reaction of which mainly involves the action of two enzymes: exo-inulinase (EC 184.108.40.206) and endo-inulinase (EC 220.127.116.11). Exo-inulinases can be used for the production of high-fructose syrup from natural inulin (saccharification), while endo-inulinases can be used for the production of fructooligosaccharides of varying lengths (Fig. 1).
Paenibacillus polymyxa (formerly Bacillus polymyxa) is one of the few bacteria known to be able to produce inulinases and ferment inulin into valuable chemicals without previous hydrolysis16. In the present study, the gene encoding exo-inulinase from P. polymyxa SC-2 was cloned into Clostridium tyrobutyricum ATCC 25755, which enabled the host bacterium to efficiently convert inulin from Jerusalem artichokes into hydrogen. Further experiments were conducted to develop a simultaneous saccharification and fermentation (SSF) process for hydrogen production from inulin utilizing this heterologous inulinase in C. tyrobutyricum.
The plasmids and strains used in the present study are listed in Table 2. P. polymyxa SC-2 was isolated from the rhizosphere of pepper in Guizhou, China17. It was cultivated in 10mL of selective medium (30.0g/L inulin, 1.0g/L (NH4)2HPO4, 0.50g/L MgSO4·7H2O, 1.0g/L NaCl, 2.0g/L (NH4)2SO4, pH 6.0) and cultured at 30°C and 120rpm for 2 days. The host strain E. coli TOP10 was obtained from Novagen and grown in Luria-Bertani (LB) medium with 100μg/mL ampicillin added if necessary. C. tyrobutyricum ATCC 25755 (purchased from Guangdong culture collection center, collection number: GIM 1.262) was used as the host strain. The cultivation medium was composed of 5g/L yeast extract, 5g/L peptone 3g/L (NH4)2SO4, 1.5g/L K2HPO4, 0.6g/L MgSO4·7H2O, 0.03g/L FeSO4·7H2O, and was sterilized by autoclaving at 121°C for 20min as reported previously18. In order to promote plasmid retention in C. tyrobutyricum, 25μg/mL of erythromycin was added to the cultivation medium. The batch SSF cultivations were performed in a 5-L stirred-tank fermentor with a working volume of 2L, as well as a 500-L stirred-tank fermentor with a working volume of 200L. The fermentors were operated at 37°C, agitated at 150rpm, and the pH was controlled at 6.0 by adding 5mol/L NaOH solution facilitated by an on-line sensing and dosing system. Anaerobiosis was maintained by sparging the medium with N2 (10mL/min) for 60min as previously reported19.
Plasmid pSY6 with lac promoter was used as the vector backbone for cloning20. E. coli DH5α was used for vector construction. The inu gene encoding exo-inulinase was amplified by PCR using the genomic DNA of P. polymyxa SC-2 strain as template with the primers F: 5′-CCGCTCGAGATGAACGTTGTTAGGCAAGAG AAAT-3′ and R: 5′-CGCTGTACATCATTTCAGCGCATAAAGCTCCAGC-3′ (the bases underlined are the recognition sites of restriction enzymes XhoI and BsrGI)17. The PCR product was cloned into the XhoI and BsrGI sites of plasmid pSY6, supplemented with ampicillin (100μg/mL), yielding exo-inulinase expression recombinant plasmid pSY6-inu, then methylated in E. coli TOP10 (pAN2)21. The recombinant plasmids extracted from the positive transformants were digested with BgIII. After verification by DNA sequencing (GENEWIZ, China), the recombinant plasmid was used for the electrotransformation of C. tyrobutyricum (Fig. S1). Plasmid pSY6 without the insert was used as the empty vector control. After cultivation on agar plates with 25μg/mL of erythromycin for 2 days, positive transformants were selected at random, used to inoculate the inulinase production medium in replicates, and cultivated at 37°C for 3 days22. The cultures were centrifuged at 12,000×g and 4°C, and the inulinase activity in the supernatants of different transformants was determined as described below.
Jerusalem artichoke tubers were purchased from a local market (Nanjing, Jiangsu, China) during the harvest season from October to November 2015. About 200g of the Jerusalem artichoke tubers were washed, peeled, smashed, suspended in 500mL of distilled water and cooked at 100°C for 5min. The inulin extract was filtered under vacuum and the total sugar concentration in the filtrate was adjusted to 20g per 100mL of the supernatant, which was finally autoclaved at 121°C for 20min.
For protein expression analysis, cell-free extracts (CFE) were mixed with an equal amount of 3-fold concentrated loading buffer (10mmol/L Tris-HCl pH 6.8, 4% (w/v) sodium dodecyl sulfate, 20% (v/v) glycine, 0.2% bromophenol blue, 2% (v/v) 2-mercaptoethanol). After boiling for 10min, 10µl of each sample was analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Protein concentrations were determined by the method of Bradford using bovine serum albumin as a standard. Cell concentrations were analyzed by measuring the optical density of the cell suspension at OD600 with a spectrophotometer (Ultrospec 3300 pro, Amersham Bioscience). Quantitative analysis of acids (butyric and acetic acid) and reducing sugars (glucose and fructose) was performed by HPLC (Agilent 1100, Agilent Technologies) equipped with an HPX 87H column (Bio-Rad, Hercules, CA, USA) and a refractive index detector (Agilent1100, G1362A)23. The H2 and CO2 concentrations in the exhaust gas were determined using the PGM 54 MultiRAE IR gas monitor (RAE system Inc., San Jose, USA).
The inulinase activity was assayed by the method of Mughal et al.24. A reaction mixture containing 0.1mL of the enzyme extract (the supernatant obtained earlier) and 0.9mL of a sodium acetate buffer (0.1M, pH 5.5) containing 2% (w/v) inulin was incubated at 50°C for 15min. The enzyme was inactivated by keeping the reaction mixture at 90°C for 10min. The amount of reducing sugars in the reaction mixture was assayed by the Nelson–Somogyi method25. One unit of inulinase activity is defined as the amount of reducing sugar produced (μmol) per minute under the assay conditions used in this study. All results were obtained from the means of triplicate assays.
It has already been known from previous whole-genome sequencing studies that wild-type C. tyrobutyricum cannot secrete inulinase26. In order to make the engineered C. tyrobutyricum hydrolyze inulin so that it can produce hydrogen directly without the need for a separate inulin hydrolysis process, the exo-inulinase gene cloned from P. polymyxa SC-2 was ligated into the expression vector pSY6, and then transformed into C. tyrobutyricum after being methylated in E. coli TOP10 (pAN2). After DNA transformation, some 300C. tyrobutyricum mutant colonies were chosen from the agar plates and the inulinase activities in the supernatants were examined to ensure that they were positive transformants. It was revealed that most of the clones had inulinase activities lower than 10.0U/mL. Furthermore, there was a 46.3% false-positive rate (139 out of the 300 clones shown in Fig. 2). A relatively high false-positive rate, which was further illustrated by colony PCR identifications not yielding the target band (data not shown), may be partly due to the low transformation efficiency in the genus clostridia. It is generally acknowledged that genetic engineering of the genus clostridia has been developed slowly compared with that of other bacteria, one of the major reasons might be the restriction and modification system which degrades foreign DNA20. However, the transformant J66 was showed the highest activity of 28.2U/mL among the top 16 values (Table 3). As expected, the wild-type strain C. tyrobutyricum had no inulinase activity as shown in Table 3, which was in accordance with the genome sequencing result26.
In order to confirm that the recombinant plasmids encompassing the exo-inulinase gene from P. polymyxa SC-2 had indeed been electro-transferred into C. tyrobutyricum, sequence analysis of the cloned inu gene in transformant J66 was subsequently carried out. The PCR products were obtained with the expected size, approximately 1,458bp (Fig. 3), which encoded a protein of 485 amino acids with a calculated molecular mass of 55.5kDa. The expressed recombinant exo-inulinase with a fused (His)6-tag was further purified by Ni2+ affinity chromatography. The purified exo-inulinase induced by IPTG exhibited an activity of 126.9U/mL, which was 4.5-fold higher than that of the crude recombinant enzyme. The final preparation migrated as a single protein band on SDS-PAGE (Fig. 4), with a mobility corresponding to a molecular mass of about 56kDa, which was in agreement with the predicted molecular mass of the exo-inulinase protein. A comparison of the deduced amino acid sequence of P. polymyxa SC-2 exo-inulinase with entries in the DDBJ database indicated that this enzyme is highly homologous with many Paenibacillus exo-inulinases as expected, for example, exo-inulinases from P. polymyxa Sb3-1 (sequence identity 99%), P. polymyxa ZJ-9 (93%), P. polymyxa SQR-21 (90%), and P. polymyxa YC0136 (86%) (Fig. 5). The invariability of the conserved domain in the exo-inulinase from P. polymyxa SC-2 indicates that its function in substrate binding and catalysis may not be very different from that of all known exo-acting inulinases. As expected, no such PCR products were amplified from the wild-type C. tyrobutyricum which carried no cloned exo-inulinase gene. Therefore, the transformant J66 was used for further investigations as the engineered inulin-consuming strain of C. tyrobutyricum.
Inulin typically has a degree of polymerization (DP) between 3 and 60, and chemical degradation (e.g., acid or low-pH treatment) or controlled enzymatic hydrolysis with endo-inulinases yields fructooligosaccharide compounds with a DP between 2 and 2027. Since wild-type C. tyrobutyricum cannot directly utilize inulin, Jerusalem artichoke need to be hydrolyzed prior to the fermentation step in order to transform inulin into available glucose and fructose10. In order to study the feasibility of directly using inulin as a low-cost substrate for fermentative hydrogen production, batch fermentation of inulin extract as the sole carbon source by C. tyrobutyricum mutant were studied, and compared with the two fermentable monosaccharides glucose and fructose. The total carbon source concentrations were all adjusted to 60g/L in a 5-L fermentor system.
As shown in Table 4, C. tyrobutyricum mutant was able to utilize glucose and fructose to produce hydrogen at high yields. When the cells were grown in the fructose-containing medium, the hydrogen yield (2.8±0.12mol/mol) and overall volumetric production rate (430±50mL/h/L) were about the same as with glucose as the sole carbon source (2.6±0.12mol/mol, and 410±40mL/h/L), which was almost equivalent with that of the wild-type C. tyrobutyricum 28. It should be noted that the hydrogen yield from glucose or fructose depends on the fermentation end-products in C. tyrobutyricum. It has been widely accepted that the highest theoretical yield of 4mol/mol glucose can be obtained if acetic acid is the end by-product of metabolism, while a maximum 2mol/mol glucose can be obtained with butyric acid as the sole by-product. However, actual yields of hydrogen are always associated with a mixture of acetic and butyric acid, and in most cases, are lower than 3mol/mol glucose29.
Since exo-inulinase splits the terminal fructose units from inulin, it was found that the inulin extract was fermented efficiently by the engineered C. tyrobutyricum, enabling high levels of production with a yield of 3.5±0.13mol/mol inulin-type sugar and an overall volumetric production rate of 540±50mL/h/L. Monosaccharides, disaccharides, and trisaccharides were released after inulin was hydrolyzed by the inulinase expressed by the engineered cells, while only a small amount of monosaccharides and oligosaccharides could be detected after the autoclaving process (Fig. S2). This once again proved that the inulinase engineered into the C. tyrobutyricum cells was an exo-inulinase. The results of this study therefore indicate a great potential for producing hydrogen from inulin-containing materials, such as Jerusalem artichoke. In addition, the growth of mutant cells was not significantly influenced by different sugars.
Bacterial growth and productivity tends to be low if high initial concentrations of reducing sugars are used in the medium due to substrate inhibition, which is a common phenomenon in practical applications28. A relevant characteristic of the media formulated with high initial sugar concentrations is the intrinsic high osmolarity due to the elevated amounts of media components. For instance, the osmolality of media with 120g/L of glucose was as high as 1.8 Osm/kg in the most concentrated medium, which would inhibit the growth of saccharolytic clostridia30. By contrast, we used initial concentrations of 60–120g/L of inulin extract in batch SSF mode with engineered C. tyrobutyricum (Table 5) without a problem. When the initial total sugar concentration was set at a level of 100g/L, the highest hydrogen yield of 3.7±0.22mol/mol inulin-type sugar and overall volumetric production rate of 650±70mL/h/L were achieved. Further addition of inulin did not improve hydrogen production, but decreased both the yield and productivity due to the exhaustion of other nutrients, which was consistent with previous studies31.
To examine the feasibility of using Jerusalem artichoke as a low-cost feedstock for fermentative hydrogen production on a pilot scale, the batch SSF of C. tyrobutyricum with 100g/L of inulin extracted from Jerusalem artichoke as carbon source was studied in a 500-L fermentor, and the results are shown in Fig. 6. In this SSF process, the consumption of inulin started at the beginning of the fermentation, while the accumulation of free fructose was negligible throughout the fermentation. The activity of exo-inulinase was the decisive factor with regard to the conversion efficiency of inulin into reducing sugars. The time course of exo-inulinase production by C. tyrobutyricum showed that the heterologously expressed enzyme had a remarkably high inulin hydrolysis activity, reaching 28.4±0.26U/mL in the supernatant at the end of the exponential phase, which was consistent with the efficient utilization of sugar and synthesis of products. The production of hydrogen and other end-products such as organic acids (e.g., butyric, acetic and lactic acid) and gases (e.g., CO2) was maintained at a low level for the first 24h, but increased afterwards and continued in the stationary phase, while the activity peak of exo-inulinase in the culture supernatant of the strain was reached after around 50h of fermentation. The formation of hydrogen was completed at 96h when inulin was no longer consumed. A maximum hydrogen yield of 3.5±0.23mol of H2 per mol of inulin-type sugar was obtained, with a volumetric productivity rate of 620±60mL/h/L based on the consumed total sugar, which was comparable to our previous results with glucose as the carbon source29. Importantly, the fermentative hydrogen production performance in a process model with the same experimental conditions, but with separate hydrolysis and fermentation steps (unpublished data), was much lower compared with the results of the SSF process presented in the current study. The most probable reason for this discrepancy is that the high concentration of inulin was gradually degraded to fermentable sugars in the SSF process, so that the inhibition by high substrate concentrations was avoided32. Therefore, SSF has an absolute competitive advantage when it comes to high initial substrate concentrations, which enables it to be operated with lower reactor volumes and therefore lower fermentation costs.
Jerusalem artichoke is an abundant resource in China and is suited to serve as a cost-effective feedstock for the bio-based fermentation of value-added products. However, producing hydrogen via direct fermentation of inulin-containing materials was historically out of the question, and the necessary inulin hydrolysis was often the rate-limiting step. This study represents the first report, to our best knowledge, that inulin derived from Jerusalem artichoke was directly converted to hydrogen by an engineered C. tyrobutyricum expressing an exo-inulinase coding gene. A high hydrogen yield of 3.5±0.23mol/mol inulin-type sugar with an overall volumetric productivity rate of 620±60mL/h/L was achieved in the SSF process in a 500-L fermentator. The present study may therefore pave the way for economical bio-based hydrogen production on an industrial scale.
This work was supported by the National Science Foundation for Young Scholars of China (21506101, U1603112), the State Key Laboratory of Bio-organic and Natural Products Chemistry, CAS (SKLBNPC15429), the Six Talent Peaks Project in Jiangsu Province (2015-JY-009), and the Environmental Protection Project in Jiangsu Province (2015053).
L.J. and Q.W. performed the experiments, collected and analyzed the data, and drafted the manuscript; Q.X. and L.Y.Z. assisted in conceiving and designing the experiments, and revised the manuscript; H.H. contributed reagents, materials and analytical tools. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Electronic supplementary material
Supplementary information accompanies this paper at doi:10.1038/s41598-017-07207-7
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Liying Zhu, Email: nc.ude.hcetjn@yppahylz.
He Huang, Email: nc.ude.hcetjn@hcetoib.