Appl Biochem Biotechnol (2014) 174:1822–1833 DOI 10.1007/s12010-014-1169-7
Mixed Food Waste as Renewable Feedstock in Succinic Acid Fermentation Zheng Sun & Mingji Li & Qingsheng Qi & Cuijuan Gao & Carol Sze Ki Lin
Received: 4 May 2014 / Accepted: 15 August 2014 / Published online: 23 August 2014 # Springer Science+Business Media New York 2014
Abstract Mixed food waste, which was directly collected from restaurants without pretreatments, was used as a valuable feedstock in succinic acid (SA) fermentation in the present study. Commercial enzymes and crude enzymes produced from Aspergillus awamori and Aspergillus oryzae were separately used in hydrolysis of food waste, and their resultant hydrolysates were evaluated. For hydrolysis using the fungal mixture comprising A. awamori and A. oryzae, a nutrient-complete food waste hydrolysate was generated, which contained 31.9 g L−1 glucose and 280 mg L−1 free amino nitrogen. Approximately 80–90 % of the solid food waste was also diminished. In a 2.5 L fermentor, 29.9 g L−1 SA was produced with an overall yield of 0.224 g g−1 substrate using food waste hydrolysate and recombinant Escherichia coli. This is comparable to many similar studies using various wastes or byproducts as substrates. Results of this study demonstrated the enormous potential of food waste as renewable resource in the production of bio-based chemicals and materials via microbial bioconversion. Keywords Biorefinery . Fungal hydrolysis . Actinobacillus succinogenes . Recombinant Escherichia coli . Platform chemical . Solid-state fermentation
Introduction Food waste is a serious global issue. Whilst over one billion people in the world are suffering from starvation and malnutrition, one third of the food produced, approximately 1.3 billion Zheng Sun and Mingji Li equally contributed to this work. Z. Sun College of Fisheries and Life Science, Shanghai Ocean University, Shanghai 201306, People’s Republic of China Z. Sun : M. Li : C. Gao : C. S. K. Lin (*) School of Energy and Environment, City University of Hong Kong, Kowloon, Hong Kong e-mail: [email protected] M. Li : Q. Qi : C. Gao State Key Laboratory of Microbial Technology, Shandong University, Shanda Nanlu, Jinan 250100, People’s Republic of China
Appl Biochem Biotechnol (2014) 174:1822–1833
1823
tons, is wasted every year [1]. It is estimated that in 2012, the food waste issue contributed a loss of $197.8 billion [2] in the USA. Besides, the waste of food leads to significant environmental problems. For example, 10 % of developed countries’ greenhouse gas emissions come from growing the food that is never eaten [3]. Considering these social, economic, and environmental impacts, it is necessary to take actions to alleviate the food waste crisis. According to the 2014 Policy Address by the Hong Kong Government of Special Administrative Region (http://www.policyaddress.gov.hk/2014/eng/p163.html), 40 % of Hong Kong’s municipal solid waste is organic waste. While the bulk of which is food waste, it also includes yard waste. The Hong Kong Government will draw up comprehensive strategies and plans to reduce, recover, and treat organic waste, including the provision in stages of modern facilities to convert organic waste into energy and other useful resources. Food waste-based biorefinery is a novel concept which receives a significant attention in recent years [4, 5]. The possibility of converting no-value food waste into valuable products such as chemicals, materials, or fuels has aroused worldwide interest [6]. Recent research conducted by our group demonstrated that the production of a nutrient-rich stream namely ‘food waste hydrolysate’ via enzymatic hydrolysis of unconsumed bakery waste (including bread, cakes, and pastries) into glucose and free amino nitrogen [7–9]. This suggested that food waste is indeed a valuable source of nutrients. The hydrolysate was subsequently used as a generic feedstock in fermentative succinic acid (SA) production [7, 8]. According to the U.S. Department of Energy, SA was ranked as one of the top platform molecules that can be used as the precursor of various commodity chemicals such as 1, 4-butanediol, tetrahydrofuran, and γ-butyrolactone [10]. These compounds have a wide range of applications, e.g. laundry detergents, plastics, and medicines [11, 12]. The current global market of SA is $225 million with expectations of double-digit growth in next 5 years [13, 14]. Key examples from recent research conducted by our group successfully paved the concept of food waste-based biorefinery. However, they were based on the unconsumed bakeries rather than general food waste, i.e. the mixture of various kinds of food produced in domestic or restaurant kitchen. According to [9], the general food waste is rich in carbohydrates (33 %), proteins (10 %), and lipids (15 %), which is very similar to the bakery waste composition according to the earlier study conducted by our group [8]. Besides, we have previously demonstrated that food waste can be used as a nutrient-complete feedstock in microalgal fermentation [9]. Therefore, the focus of this study is on the valorization of mixed food waste in bio-based SA production. In addition, the performance of fungal and enzymatic hydrolyses of food waste were compared to determine the most suitable method for the hydrolysate production. Results of this study could be highly significant, as it is one of the food waste management strategies to be considered for future upscale and practical applications.
Materials and Methods Chemicals and Microorganisms All chemicals used in this study were purchased from Acros Organics (Morris Plains, NJ, USA) and Sigma-Aldrich (St. Louis, MO, USA) except otherwise specified. Commercial enzymes including fungal α-amylase, glucoamylase, and acidity protease were purchased from Longda Bio-Products (Shandong, China). Microorganisms Aspergillus awamori (ATCC
1824
Appl Biochem Biotechnol (2014) 174:1822–1833
14331) and Actinobacillus succinogenes (ATCC 55618) were obtained from the American Type Culture Collection (Rockville, MD, USA). A recombinant Escherichia coli (MG1655) was constructed by our coworkers at the Shandong University in China [15]. This is a genetically modified strain with all pathways accumulating by-products inactivated. An industrial strain of Aspergillus oryzae isolated from a soy sauce starter was provided by the Amoy Food (Hong Kong). A. awamori and A. oryzae were utilized for the production of amylolytic and proteolytic enzymes, respectively. A. succinogenes and recombinant E. coli were used separately in SA fermentations. Their storage and sporulation for inoculum preparation were conducted according to the procedure previously reported by our group [16, 17]. Handling of Food Waste Mixed food waste containing rice, noodles, meat, and vegetables was collected from canteens located in the Hong Kong Science Park at Shatin, Hong Kong. It was immediately brought to the laboratory for processing. The waste was blended using a domestic kitchen blender for 5 min and kept at −20 °C. For 320–480 g of fresh food waste, the dry weight was equivalent to 100–150 g [9]. Solid State Fermentation Bakery waste was used in fungal inoculum preparation via solid state fermentation (SSF). Bakery waste in this paper refers to the pastry waste collected from the Starbucks Hong Kong. An amount of 10 g of bakery waste (8.5 g dry weight) was transferred into a petri dish. Cryopreserved spores (1 mL) of A. awamori (4.6×105 spores mL−1) or A. oryzae (6.3×105 spores mL−1) were inoculated to the blended bakery waste. These inoculated samples were incubated at 30 °C for 7 days. Enzymatic Hydrolysis Previous study conducted by our groups showed that the fermentation solids at the end of A. awamori and A. oryzae solid-state bioconversions constituted glucoamylase-rich and protease-rich media, respectively [18]. These solids could be considered as crude enzyme sources and added to suspension of food waste to hydrolyse the key carbon and nitrogen components. In this way, a generic fermentation substrate, namely ‘hydrolysate’ that is rich in glucose and nitrogen, as well as other essential nutrients required for the subsequent SA fermentation, could be produced. In this study, SSF fungal mashes from A. awamori and A. oryzae and commercial enzymes (α-amylase, glucoamylase, and protease) were evaluated separately in enzymatic hydrolysis. The hydrolysis was carried out in a 2.5 L bioreactor (BioFlo/CelliGen 115, New Brunswick Scientific, Edison, NJ, USA) which was equipped with automatic temperature control water jacket and stirrers. For fungal hydrolysis, the solid mashes of A. awamori and A. oryzae from each petri dish were added. For commercial enzymes, the amounts added into the food waste were 10 U α-amylase g−1, 120 U glucoamylase g−1, and 150 U protease g−1. The total volume of the blend was adjusted to 1 L by adding tap water at 55 °C. The reaction mixture was stirred at 300 rpm without pH control, as pH remained between 4.0 and 4.5 during hydrolysis. Hydrolysis samples were taken every 1 h for 24 h. After the hydrolysis step, the broth was centrifuged at 11,500g for 30 min and the supernatant was subsequently filtered by vacuum filtration using Whatman No. 1 filter paper. The
Appl Biochem Biotechnol (2014) 174:1822–1833
1825
resultant solution was kept at −20 °C and was subsequently used as the sole fermentation feedstock in the subsequent SA fermentation. All experiments were carried out in duplicate. Concentration of Remaining Solids After hydrolysis, the remaining solids and supernatant were separated by centrifugation at 7,000g for 20 min. The pellet was lyophilized and the dry weight was measured. The diminished or reduced solids refer to the difference of initial dry weight of food waste and dry weight of remaining solids after hydrolysis [19]. All quantifications were carried out in duplicate. Bacterial Fermentation A. succinogenes and recombinant E. coli (MG1655) were used separately in SA fermentations, with either pastry or food waste hydrolysates as feedstock. Initial experiments for comparing the fermentation performance of A. succinogenes and recombinant E. coli were carried out in small anaerobic reactors (SARs) and Duran bottles, respectively. Both were in the scale of 100 mL, each containing 60 mL hydrolysate. The hydrolysate was filtered by 0.2 μ m PTFE membrane filter (Sartorius, Germany) with an inoculum size of 5 % (v/v). For A. succinogenes fermentation, external CO2 was sprayed into the broth at a rate of 0.5 vvm and the sterilized magnesium carbonate (MgCO3) (30 g L−1) was also added as a neutral pH buffer. For E. coli, compressed air was supplied at 0.5 vvm. Fermentations were conducted at 37 °C and 150 rpm. Samples were taken every 8 h to measure the optical density and metabolites concentration. Fermentations were considered finished either when glucose was completely depleted or when no change in total sugar concentration was detected for a period of 5 h. In the bench-top scale study, the aerobic E. coli fermentation was carried out due to rapid biomass and SA production [17]. In this study, 1.5 L food waste hydrolysate was added into a 2.5 L fermentor (Biostat, Sartorius stedim, Germany). The initial glucose concentration was 58 g L−1. The inoculum size was 10 % (v/v). The fermentation was conducted at 37 °C with filtered compressed air at 1 vvm and reactor agitating at 350 rpm. The pH of the fermentation was controlled with the addition of NaOH (10 M) and H2SO4 (0.5 M). Samples were taken every 6 h to measure the optical density and metabolite concentration. Determination of Cell Density, Sugar Concentration, Free Amino Nitrogen (FAN), and Fermentation Metabolites Bacterial growth was determined by optical density (OD) measurements at a wavelength of 660 or 600 nm using the spectrophotometer (UV-1800, Shimadzu, Japan) for fermentation samples from A. succinogenes or E. coli fermentations, respectively [18]. Glucose and metabolite concentrations were quantified using the high-performance liquid chromatography (HPLC) system (Waters, UK), which was equipped with a BIO-RAD column (HPX-87H), a refractive index (RI) detector (Waters, UK), and a photodiode array (PDA) analyser (Waters, UK). The mobile phase was H2SO4 (5 mM) with a flow rate of 0.6 mL min−1. The column and RI detector were equilibrated to 65 and 35 °C, respectively. Injection volume of all samples and standards was set to 20 μL. The FAN concentration of the hydrolysis samples was analyzed using the ninhydrin colorimetric method promulgated in the European Brewery Convention [20].
1826
Appl Biochem Biotechnol (2014) 174:1822–1833
Results and Discussion Food Waste Composition and Hydrolysis The composition of mixed food waste is shown in Table 1 [19]. It was rich in organic carbon and nitrogen, which could be a potential source of nutrients. The starch content (361.5 mg g−1) of mixed waste was higher than that in bakery waste (316.7 mg g−1), and therefore may be hydrolysed into generic medium with higher concentrations of reducing sugars. In enzymatic hydrolysis, key enzymes such as amylolytic and proteolytic enzymes are needed to break down the macromolecules including starch and proteins present in the food waste into simple sugars and amino acids. In this study, we firstly assessed three major commercial enzymes, including protease, α-amylase, and glucoamylase that were purchased from the company. For protease, the optimum condition was evaluated in Duran bottles (100 mL) and was determined as follows: pH 3.0–5.0; temperature 40–50 °C; protease concentration 6,000 U that was equivalent to 150 U g−1 food waste (Fig. 1). Another two commercial enzymes α-amylase and glucoamylase were supplied by the same agent, and their optimum conditions were determined according to the previous studies: pH 4.5, temperature 55 °C, 10 U g−1 food waste of α-amylase and 120 U g−1 food waste of glucoamylase [21]. Under such conditions, hydrolysis was carried out in 1-L scale bioreactor. As shown in Fig. 2, the yields of nutrients increased immediately when hydrolysis started. At 24 h, almost 120 g L−1glucose and 3,000 mg L−1 FAN were obtained, which were much higher than the essential nutritional requirement of A. succinogenes. In general, medium containing 30– 70 g L−1 glucose and 300 mg L−1 FAN would be sufficient for a typical SA fermentation [22]. In the contrary, over supply of nutrients could be unfavorable for the cell growth of A. succinogenes. Our earlier study showed that the cell growth started to be inhibited when the glucose concentration is higher than 70 g L−1 [23]. Meanwhile, food waste hydrolysis using two fungi, namely A. awamori and A. oryzae, were conducted. The fungal mashes of A. awamori and A. oryzae obtained from SSF were used for the production of enzyme complex. SSF is a technique to gain the high enzyme concentration due to the absence of free water and the low protein breakdown [24]. Similar to commercial enzymes, glucose and FAN productions by fungal hydrolysis began immediately when hydrolysis started. According to Table 2, the resultant glucose and FAN concentration were 31.9 g L−1 and 280 mg L−1, respectively, which met the nutritional requirement of A. succinogenes [16]. Compared with commercial enzymes that produced up to 143 g L−1 glucose, hydrolysate produced from fungal hydrolysis contained appropriate carbon and nitrogen concentrations, which is suitable for microbial growth and SA production. Apart from the nutrient-producing capability, another important factor to be considered is the reduction of solid food waste. Outstanding food waste-diminishing capability could be an additional benefit of hydrolysis. The reduced or diminished solids refer to the difference of initial dry weight of food waste and dry weight of remaining solids after hydrolysis. According
Table 1 Composition of bakery waste and mixed food waste (mg g−1) [19]
Bakery waste
Mixed food waste
Carbohydrates
645.5±47.7
470.3±30.5
Starch Proteins
316.7±67.7 98.2±0.9
361.5±13.8 99.3±13.9
Lipids
265.8±34.3
373.7±25.8
Appl Biochem Biotechnol (2014) 174:1822–1833 Fig. 1 Determination of the optimum conditions of protease. Effects of a pH, b temperature, and c the amount of enzyme on the activity were assessed. The hydrolysis was conducted in 100 mL Duran bottles containing 40 g food waste and 40 mL H2O
1827
1828
Appl Biochem Biotechnol (2014) 174:1822–1833
Fig. 2 Enzymatic catalysis of mixed food waste by commercial enzymes. The conditions were as follows: pH 4.5; temperature 55 °C; solid-to-liquid ratio 1:1; 10 U g−1 food waste of α-amylase; 120 U g−1 food waste of glucoamylase; and 150 U g−1 food waste of protease
to Table 2, 80–90 % of the initial solid weight was reduced by fungal hydrolysis, much higher than the hydrolysis using commercial enzymes (66 %). These data suggested the potential of fungal hydrolysis as an alternative to conventional waste treatments such as composting and anaerobic digestion. Results from this part of the study demonstrated that fungal hydrolysis using SSF solid mashes of A. awamori and A. oryzae is the preferred method for producing hydrolysate with the desired nutrient concentrations and superior food waste diminishing capability. Therefore, the resultant hydrolysate produced via fungal hydrolysis was used for the subsequent SA fermentation. Succinic Acid Fermentation In this section, the performance of the two bacteria, namely A. succinogenes and recombinant E. coli in SA fermentations, was separately examined. A. succinogenes is a well-known natural succinate producer and has been successfully used in our previous studies [7, 8]. The recombinant E. coli is a genetically modified strain with all pathways accumulating by-products inactivated,
Table 2 Comparison of hydrolysis using enzymes from fungal solid mashes of A. awamori and A. oryzae and commercial source at 24 h. Both hydrolyses were conducted in 1-L bioreactors Glucose (g L−1) FAN (mg L−1) Weight reduction of food waste after hydrolysis (%)a Fungal solid mashes
31.9 c
Commercial enzymes
119.22
280
80–90b
290.7
66b
a
Percentage of food waste diminished = dry weight of total suspended solids / dry weight of initial food waste × 100 %
b
Pleissner et al. [19]
Combination of protease (150 U g−1 food waste), α-amylase (10 U g−1 food waste), and glucoamylase (120 U g−1 food waste) c
Appl Biochem Biotechnol (2014) 174:1822–1833
1829
Table 3 SA fermentation of A. succinogenes and recombinant E. coli on the hydrolysates produced from mixed food waste A.succinogenes
E. coli
E. coli
Scale
100 mL SARs
100 mL Duran bottle
2.5 L Bench-top fermentator
Total volume of hydrolysate (mL)
60
60
1,500
Initial glucose concentration (g L−1)
32.9
35.1
57.8
OD660/OD600
2.89
13.4
17.5
SA concentration (g L−1)
24.1
26.4
29.9
SA yield (g g−1 glucose) SA productivity (g L−1 h−1)
0.87 0.29
0.98 0.20
0.56 0.48
By-product concentration (g L−1)
13.7
0
0
which highly enhances the SA production [15]. Both bacterial fermentations were initially conducted in 100 mL Duran bottles, in which the total volume of hydrolysate was 60 mL. For A. succinogenes, the initial glucose concentration was 32.9 g L−1 (Table 3). The OD660 significantly increased within the first 16 h, where the maximum value was 4.16. These data indicated that the nutrients in the hyrolysate were sufficient to support the cell growth. As shown in Fig. 3, the SA production started from 8 h and the fermentation lasted for 92 h. After 92 h, 24.1 g L−1 SA was produced, which corresponded to a yield of 0.87 g SA g−1 glucose and a productivity of 0.29 g L−1 h−1. The amount of acid by-products was 13.7 g L−1, which consists of acetic, formic, and pyruvic acids. For small-scale fermentation using E. coli, the initial glucose concentration was 35.1 g L−1. As shown in Fig. 4, the SA concentration continuously increased until the depletion of glucose after 148 h. At the end of fermentation, the SA concentration reached 26.4 g L−1, which
Fig. 3 Fermentation profile of A. succinogenes using the mixed food waste. The fermentation was carried out in 100 mL small anaerobic reactors. The amount of hydrolysate was 60 mL. The average concentrations and error bars of the duplicated experiments are shown
1830
Appl Biochem Biotechnol (2014) 174:1822–1833
Fig. 4 Fermentation profile of recombinant E. coli using the mixed food waste. The fermentation was carried out in 100 mL Duran bottles. The amount of hydrolysate was 60 mL. The average concentrations and error bars of the duplicated experiments are shown
corresponded to a yield of 0.98 g SA g−1 glucose and a productivity of 0.20 g L−1 h−1. It is worth to note that when using mixed food waste hydrolysate as feedstock, the amount of SA (26.4 g L−1) was comparable with that using pastry waste hydrolysate (26.5 g L−1) (Table 4). Again, these results suggested the significant potential of mixed food waste. Compared with A. succinogenes, recombinant E. coli offered several advantages. Firstly, it is a fast-growing strain. The average OD600 value was 13.4, much higher than that of A. succinogenes (OD660 =2.89), indicating significant amount of biomass was produced. Secondly, due to the genetic manipulation, there was no by-product generated during fermentation, so the substantial recovery process can be simplified. Therefore, this recombinant E. coli was selected for bench-top scale fermentation study. It was carried out in a 2.5 L fermentor using food waste hydrolysate with an initial glucose concentration of 57.8 g L−1. As shown in Fig. 5, cell growth immediately began when the fermentation process started. The OD600 increased sharply from 0.176 to 14.4 during the exponential phase (from 0 to 36 h), whilst SA also began to accumulate at the same time. At 60 h, cell growth attained a maximum OD600 of 17.5. The resultant SA concentration was 29.9 g L−1 with a yield of 0.56 g g−1 Table 4 SA fermentations with pastry or food waste hydrolysates as feedstock. Fermentations were conducted in 100 mL Duran bottles, in which the total volume of hydrolysate was 60 mL
SA concentration (g L−1) SA yield (g g−1 glucose) SA productivity (g L−1 h−1)
A. succinogenes
E. coli
Pastry
Food waste
Pastry
Food waste
17.1 0.85
24.1 0.87
26.5 0.79
26.4 0.98
0.24
0.29
0.20
0.20
Appl Biochem Biotechnol (2014) 174:1822–1833
1831
Fig. 5 Fermentation profile of recombinant E. coli using the mixed food waste. The fermentation was carried out in a 2.5 L fermentor. The amount of hydrolysate was 1.5 L. The average concentrations and error bars of the duplicated experiments are shown
glucose, lower than the small-scale fermentation in Duran bottles (yield=0.98 g g−1 glucose). This could be due to glucose utilisation for other maintenance, resulting in higher biomass concentration but a lower SA yield. Interestingly, the SA productivity in fermentor was Table 5 Comparison of SA yields achieved using different food waste substrates Substrate
SA yield (g g−1 glucose)
Overall SA yield (g g−1 substrate)
References
Wheat
0.40
0.40
Du et al. [16, 25]
Wheat flour milling by-product
1.02
0.087
Dorado et al. [18]
Potatoes
N/A
N/A
Delgado et al. [26]
Corncob
0.58
N/A
Yu et al. [27]
Rapeseed meala
0.115
N/A
Chen et al. [28]
Rapeseed mealb
N/A
N/A
Wang et al. [29]
Orange peel
0.58
Negligible
Li et al. [30]
Bread
1.16
0.55
Leung et al. [7]
Cake
0.80
0.28
Zhang et al. [8]
Pastry
0.67
0.35
Zhang et al. [8]
Mixed food waste
0.56
0.224
This study
N/A not available a
Rapeseed meal is treated by diluted sulphuric acid hydrolysis and subsequent enzymatic hydrolysis of pectinase, celluclast, and viscozyme b Rapeseed meal is treated by enzymatic hydrolysis using A. oryzae
1832
Appl Biochem Biotechnol (2014) 174:1822–1833
relatively high (0.48 g L−1 h−1) as compared to fermentations in 100 mL Duran bottles. (0.20 g L−1 h−1). The enhanced oxygen and mass transfers in fermentation broth were due to the use of agitator and the continuous supply of air. In addition, the pH of fermentation broth was maintained at 7 using automated control. This facilitates the SA production by E. coli at its optimal pH range. Therefore, the duration of fermentation in bench-top fermentation (72 h) was only half of that in the Duran bottles (148 h). The final overall SA yield obtained from the mixed food waste was 0.224 g g−1, which is comparable to the results from the earlier studies conducted by our groups using waste bread, cake, and pastry (Table 5).
Conclusions This study investigated the possibility of converting no-value food waste into succinic acid via microbial bioconversion. In the first part of the study, a nutrient-complete medium, namely food waste hydrolysate, was produced using commercial enzymes or fungal mashes with A. awamori and A. oryzae. Results showed that the hydrolysate produced via fungal hydrolysis contains appropriate levels of carbon and nitrogen sources which would be utilized in the subsequent A. succinogenes fermentation. The second part of the study reported the use of food waste hydrolysate as the sole medium in recombinent E. coli fermentation, which led to the production of 29.9 g L−1 SA. Results indicated that E. coli cells were able to utilize the nutrients in the food waste hydrolysate efficiently for cell growth and SA production. The overall yield of SA was 0.224 g g−1 food waste, which is comparable to similar studies using cereal-based wastes or agricultural by-products as substrates. This is the first study reports the bio-based production of SA using mixed food waste from restaurants without any pretreatments. Results of this study demonstrated the enormous potential of mixed food waste as a renewable resource for producing chemicals and materials, suggesting a novel and sustainable approach to alleviate the food waste crisis. Acknowledgments Carol Sze Ki LIN acknowledges the Research Grant Councils in Hong Kong for the provision of General Research Fund (GRF)/Early Career Scheme (ECS) 2013–2014 (Project No. 189713). Zheng SUN acknowledges financial support from the Shanghai Pujiang Program (Grant No. 13PJ1403500) and the Doctoral Fund of Ministry of Education of China (Grant No. 20133104120004).
References 1. Gustavsson, J., Cederberg, C., Sonesson, U., van Otterdijk, R., & Meybeck, A. (2011). Global food losses and food waste. Rome: Food and Agriculture Organization of the United Nations. 2. Buchner, B., Fischler, C., Reilly, J., Riccardi, G., Ricordi, C., & Veronesi, U. (2012). Food waste: Causes, impacts and proposals. Barilla Center for Food & Nutrition. 3. Stuart, T. (2009). Waste: Uncovering the global food scandal. New York: WW Norton & Company. 4. Arancon, R. A. D., Lin, C. S. K., Chan, K. M., Kwan, T. H., & Luque, R. (2013). Energy Science and Engineering, 1, 53–71. 5. Koutinas, A. A., Pleissner, D., Vlysidis, A., Kopsahelis, N., Lopez Garciac, I., Kookosd, I., Papanikolaou, S., & Lin, C. S. K. (2014). Chemical Society Reviews, 43, 2587–2627. 6. Lin, C. S. K., Pfaltzgraff, L. A., Herrero-Davila, L., Mubofu, E. B., Abderrahim, S., Clark, J. H., Koutinas, A. A., Kopsahelis, N., Stamatelatou, K., Dickson, F., Thankappan, S., Mohamed, Z., Brocklesby, R., & Luque, R. (2013). Energy and Environmental Science, 6, 426–464. 7. Leung, C. C. J., Cheung, A. S. Y., Zhang, A. Y., Lam, K. F., & Lin, C. S. K. (2012). Biochemical Engineering Journal, 65, 10–15.
Appl Biochem Biotechnol (2014) 174:1822–1833
1833
8. Zhang, A. Y., Sun, Z., Leung, C. C. J., Han, W., Lau, K. Y., Li, M., & Lin, C. S. K. (2013). Green Chemistry, 15, 690–695. 9. Pleissner, D., Lam, W. C., Sun, Z., & Lin, C. S. K. (2013). Bioresource Technology, 137, 139–146. 10. McKinlay, J. B., Vielle, C., & Zeikus, J. G. (2007). Applied Microbiology and Biotechnology, 76, 727–740. 11. Werpy, T., & Petersen, G. (2004). Top Value Added Chemicals from Biomass, Vol. 1: Results of Screening for Potential Candidates from Sugars and Synthesis Gas (pp. 23–24). Oakridge: U.S. Department of Energy. 12. Bozell, J. J., & Petersen, G. R. (2010). Green Chemistry, 12, 539–554. 13. Taylor, P. (2010). Biosuccinic acid ready for take off? Royal Society of Chemistry, http://www.rsc.org/ chemistryworld/News/2010/January/21011003.asp. 14. de Guzman, D. (2012). Chemical industry awaits for bio-succinic acid potential, ICIS, http://www.icis.com/ Articles/2012/01/30/9527521/chemical-industry-awaits-for-bio-succinic-acid-potential.html. 15. Li, M., Wang, J., Geng, Y., Li, Y., Wang, Q., Liang, Q., & Qi, Q. (2012). Microbial Cell Factories, 11, 19. 16. Du, C., Lin, S. K. C., Koutinas, A., Wang, R., Dorado, P., & Webb, C. (2008). Bioresource Technology, 99, 8310–8315. 17. Li, Y., Li, M., Zhang, X., Yang, P., Liang, Q., & Qi, Q. (2013). Bioresource Technology, 149, 333–340. 18. Dorado, P. M., Lin, S. K. C., Koutinas, A., Du, C., Wang, R., & Webb, C. (2009). Journal of Biotechnology, 143, 51–59. 19. Pleissner, D., Kwan, T. H., & Lin, C. S. K. (2014). Bioresource Technology, 158, 48–54. 20. Koutinas, A. A., Arifeen, N., Wang, R., & Webb, C. (2007). Biotechnology and Bioengineering, 97, 61–72. 21. Yan, S., Yao, J., Yao, L., Zhi, Z., Chen, X., & Wu, J. (2012). Brazilian Archives of Biology and Technology, 55, 183–192. 22. Lin, C. S. K., Luque, R., Clark, J. H., Webb, C., & Du, C. (2012). Biofuels, Bioproducts and Biorefining, 6, 88–104. 23. Lin, C. S. K., Du, C., Koutinas, A., Wang, R., & Webb, C. (2008). Biochemical Engineering Journal, 41, 128–135. 24. Wang, R. (1999). PhD thesis, The University of Manchester, UK. 25. Du, C., Lin, C. S. K., Koutinas, A., Wang, R., & Webb, C. (2007). Applied Microbiology and Biotechnology, 76, 1263–1270. 26. Delgado, R., Castro, A. J., & Vázquez, M. (2009). LWT-Food Science and Technology, 42, 797–804. 27. Yu, J., Li, Z., Ye, Q., Yang, Y., & Chen, S. (2010). Journal of Industrial Microbiology and Biotechnology, 37, 1033–1040. 28. Chen, K. Q., Li, J., Ma, J. F., Jiang, M., Wei, P., & Liu, Z. M. (2011). Bioresource Technology, 102, 1704– 1708. 29. Wang, R., Shaarani, S. M., Godoy, L. C., Melikoglu, M., Vergara, C. S., & Koutinas, A. (2010). Enzyme and Microbial Technology, 47, 77–83. 30. Li, Q., Siles, J. A., & Thompson, I. P. (2010). Applied Microbiology and Biotechnology, 88, 671–678.
Mixed Food Waste as Renewable Feedstock in Succinic Acid Fermentation Zheng Sun & Mingji Li & Qingsheng Qi & Cuijuan Gao & Carol Sze Ki Lin
Received: 4 May 2014 / Accepted: 15 August 2014 / Published online: 23 August 2014 # Springer Science+Business Media New York 2014
Abstract Mixed food waste, which was directly collected from restaurants without pretreatments, was used as a valuable feedstock in succinic acid (SA) fermentation in the present study. Commercial enzymes and crude enzymes produced from Aspergillus awamori and Aspergillus oryzae were separately used in hydrolysis of food waste, and their resultant hydrolysates were evaluated. For hydrolysis using the fungal mixture comprising A. awamori and A. oryzae, a nutrient-complete food waste hydrolysate was generated, which contained 31.9 g L−1 glucose and 280 mg L−1 free amino nitrogen. Approximately 80–90 % of the solid food waste was also diminished. In a 2.5 L fermentor, 29.9 g L−1 SA was produced with an overall yield of 0.224 g g−1 substrate using food waste hydrolysate and recombinant Escherichia coli. This is comparable to many similar studies using various wastes or byproducts as substrates. Results of this study demonstrated the enormous potential of food waste as renewable resource in the production of bio-based chemicals and materials via microbial bioconversion. Keywords Biorefinery . Fungal hydrolysis . Actinobacillus succinogenes . Recombinant Escherichia coli . Platform chemical . Solid-state fermentation
Introduction Food waste is a serious global issue. Whilst over one billion people in the world are suffering from starvation and malnutrition, one third of the food produced, approximately 1.3 billion Zheng Sun and Mingji Li equally contributed to this work. Z. Sun College of Fisheries and Life Science, Shanghai Ocean University, Shanghai 201306, People’s Republic of China Z. Sun : M. Li : C. Gao : C. S. K. Lin (*) School of Energy and Environment, City University of Hong Kong, Kowloon, Hong Kong e-mail: [email protected] M. Li : Q. Qi : C. Gao State Key Laboratory of Microbial Technology, Shandong University, Shanda Nanlu, Jinan 250100, People’s Republic of China
Appl Biochem Biotechnol (2014) 174:1822–1833
1823
tons, is wasted every year [1]. It is estimated that in 2012, the food waste issue contributed a loss of $197.8 billion [2] in the USA. Besides, the waste of food leads to significant environmental problems. For example, 10 % of developed countries’ greenhouse gas emissions come from growing the food that is never eaten [3]. Considering these social, economic, and environmental impacts, it is necessary to take actions to alleviate the food waste crisis. According to the 2014 Policy Address by the Hong Kong Government of Special Administrative Region (http://www.policyaddress.gov.hk/2014/eng/p163.html), 40 % of Hong Kong’s municipal solid waste is organic waste. While the bulk of which is food waste, it also includes yard waste. The Hong Kong Government will draw up comprehensive strategies and plans to reduce, recover, and treat organic waste, including the provision in stages of modern facilities to convert organic waste into energy and other useful resources. Food waste-based biorefinery is a novel concept which receives a significant attention in recent years [4, 5]. The possibility of converting no-value food waste into valuable products such as chemicals, materials, or fuels has aroused worldwide interest [6]. Recent research conducted by our group demonstrated that the production of a nutrient-rich stream namely ‘food waste hydrolysate’ via enzymatic hydrolysis of unconsumed bakery waste (including bread, cakes, and pastries) into glucose and free amino nitrogen [7–9]. This suggested that food waste is indeed a valuable source of nutrients. The hydrolysate was subsequently used as a generic feedstock in fermentative succinic acid (SA) production [7, 8]. According to the U.S. Department of Energy, SA was ranked as one of the top platform molecules that can be used as the precursor of various commodity chemicals such as 1, 4-butanediol, tetrahydrofuran, and γ-butyrolactone [10]. These compounds have a wide range of applications, e.g. laundry detergents, plastics, and medicines [11, 12]. The current global market of SA is $225 million with expectations of double-digit growth in next 5 years [13, 14]. Key examples from recent research conducted by our group successfully paved the concept of food waste-based biorefinery. However, they were based on the unconsumed bakeries rather than general food waste, i.e. the mixture of various kinds of food produced in domestic or restaurant kitchen. According to [9], the general food waste is rich in carbohydrates (33 %), proteins (10 %), and lipids (15 %), which is very similar to the bakery waste composition according to the earlier study conducted by our group [8]. Besides, we have previously demonstrated that food waste can be used as a nutrient-complete feedstock in microalgal fermentation [9]. Therefore, the focus of this study is on the valorization of mixed food waste in bio-based SA production. In addition, the performance of fungal and enzymatic hydrolyses of food waste were compared to determine the most suitable method for the hydrolysate production. Results of this study could be highly significant, as it is one of the food waste management strategies to be considered for future upscale and practical applications.
Materials and Methods Chemicals and Microorganisms All chemicals used in this study were purchased from Acros Organics (Morris Plains, NJ, USA) and Sigma-Aldrich (St. Louis, MO, USA) except otherwise specified. Commercial enzymes including fungal α-amylase, glucoamylase, and acidity protease were purchased from Longda Bio-Products (Shandong, China). Microorganisms Aspergillus awamori (ATCC
1824
Appl Biochem Biotechnol (2014) 174:1822–1833
14331) and Actinobacillus succinogenes (ATCC 55618) were obtained from the American Type Culture Collection (Rockville, MD, USA). A recombinant Escherichia coli (MG1655) was constructed by our coworkers at the Shandong University in China [15]. This is a genetically modified strain with all pathways accumulating by-products inactivated. An industrial strain of Aspergillus oryzae isolated from a soy sauce starter was provided by the Amoy Food (Hong Kong). A. awamori and A. oryzae were utilized for the production of amylolytic and proteolytic enzymes, respectively. A. succinogenes and recombinant E. coli were used separately in SA fermentations. Their storage and sporulation for inoculum preparation were conducted according to the procedure previously reported by our group [16, 17]. Handling of Food Waste Mixed food waste containing rice, noodles, meat, and vegetables was collected from canteens located in the Hong Kong Science Park at Shatin, Hong Kong. It was immediately brought to the laboratory for processing. The waste was blended using a domestic kitchen blender for 5 min and kept at −20 °C. For 320–480 g of fresh food waste, the dry weight was equivalent to 100–150 g [9]. Solid State Fermentation Bakery waste was used in fungal inoculum preparation via solid state fermentation (SSF). Bakery waste in this paper refers to the pastry waste collected from the Starbucks Hong Kong. An amount of 10 g of bakery waste (8.5 g dry weight) was transferred into a petri dish. Cryopreserved spores (1 mL) of A. awamori (4.6×105 spores mL−1) or A. oryzae (6.3×105 spores mL−1) were inoculated to the blended bakery waste. These inoculated samples were incubated at 30 °C for 7 days. Enzymatic Hydrolysis Previous study conducted by our groups showed that the fermentation solids at the end of A. awamori and A. oryzae solid-state bioconversions constituted glucoamylase-rich and protease-rich media, respectively [18]. These solids could be considered as crude enzyme sources and added to suspension of food waste to hydrolyse the key carbon and nitrogen components. In this way, a generic fermentation substrate, namely ‘hydrolysate’ that is rich in glucose and nitrogen, as well as other essential nutrients required for the subsequent SA fermentation, could be produced. In this study, SSF fungal mashes from A. awamori and A. oryzae and commercial enzymes (α-amylase, glucoamylase, and protease) were evaluated separately in enzymatic hydrolysis. The hydrolysis was carried out in a 2.5 L bioreactor (BioFlo/CelliGen 115, New Brunswick Scientific, Edison, NJ, USA) which was equipped with automatic temperature control water jacket and stirrers. For fungal hydrolysis, the solid mashes of A. awamori and A. oryzae from each petri dish were added. For commercial enzymes, the amounts added into the food waste were 10 U α-amylase g−1, 120 U glucoamylase g−1, and 150 U protease g−1. The total volume of the blend was adjusted to 1 L by adding tap water at 55 °C. The reaction mixture was stirred at 300 rpm without pH control, as pH remained between 4.0 and 4.5 during hydrolysis. Hydrolysis samples were taken every 1 h for 24 h. After the hydrolysis step, the broth was centrifuged at 11,500g for 30 min and the supernatant was subsequently filtered by vacuum filtration using Whatman No. 1 filter paper. The
Appl Biochem Biotechnol (2014) 174:1822–1833
1825
resultant solution was kept at −20 °C and was subsequently used as the sole fermentation feedstock in the subsequent SA fermentation. All experiments were carried out in duplicate. Concentration of Remaining Solids After hydrolysis, the remaining solids and supernatant were separated by centrifugation at 7,000g for 20 min. The pellet was lyophilized and the dry weight was measured. The diminished or reduced solids refer to the difference of initial dry weight of food waste and dry weight of remaining solids after hydrolysis [19]. All quantifications were carried out in duplicate. Bacterial Fermentation A. succinogenes and recombinant E. coli (MG1655) were used separately in SA fermentations, with either pastry or food waste hydrolysates as feedstock. Initial experiments for comparing the fermentation performance of A. succinogenes and recombinant E. coli were carried out in small anaerobic reactors (SARs) and Duran bottles, respectively. Both were in the scale of 100 mL, each containing 60 mL hydrolysate. The hydrolysate was filtered by 0.2 μ m PTFE membrane filter (Sartorius, Germany) with an inoculum size of 5 % (v/v). For A. succinogenes fermentation, external CO2 was sprayed into the broth at a rate of 0.5 vvm and the sterilized magnesium carbonate (MgCO3) (30 g L−1) was also added as a neutral pH buffer. For E. coli, compressed air was supplied at 0.5 vvm. Fermentations were conducted at 37 °C and 150 rpm. Samples were taken every 8 h to measure the optical density and metabolites concentration. Fermentations were considered finished either when glucose was completely depleted or when no change in total sugar concentration was detected for a period of 5 h. In the bench-top scale study, the aerobic E. coli fermentation was carried out due to rapid biomass and SA production [17]. In this study, 1.5 L food waste hydrolysate was added into a 2.5 L fermentor (Biostat, Sartorius stedim, Germany). The initial glucose concentration was 58 g L−1. The inoculum size was 10 % (v/v). The fermentation was conducted at 37 °C with filtered compressed air at 1 vvm and reactor agitating at 350 rpm. The pH of the fermentation was controlled with the addition of NaOH (10 M) and H2SO4 (0.5 M). Samples were taken every 6 h to measure the optical density and metabolite concentration. Determination of Cell Density, Sugar Concentration, Free Amino Nitrogen (FAN), and Fermentation Metabolites Bacterial growth was determined by optical density (OD) measurements at a wavelength of 660 or 600 nm using the spectrophotometer (UV-1800, Shimadzu, Japan) for fermentation samples from A. succinogenes or E. coli fermentations, respectively [18]. Glucose and metabolite concentrations were quantified using the high-performance liquid chromatography (HPLC) system (Waters, UK), which was equipped with a BIO-RAD column (HPX-87H), a refractive index (RI) detector (Waters, UK), and a photodiode array (PDA) analyser (Waters, UK). The mobile phase was H2SO4 (5 mM) with a flow rate of 0.6 mL min−1. The column and RI detector were equilibrated to 65 and 35 °C, respectively. Injection volume of all samples and standards was set to 20 μL. The FAN concentration of the hydrolysis samples was analyzed using the ninhydrin colorimetric method promulgated in the European Brewery Convention [20].
1826
Appl Biochem Biotechnol (2014) 174:1822–1833
Results and Discussion Food Waste Composition and Hydrolysis The composition of mixed food waste is shown in Table 1 [19]. It was rich in organic carbon and nitrogen, which could be a potential source of nutrients. The starch content (361.5 mg g−1) of mixed waste was higher than that in bakery waste (316.7 mg g−1), and therefore may be hydrolysed into generic medium with higher concentrations of reducing sugars. In enzymatic hydrolysis, key enzymes such as amylolytic and proteolytic enzymes are needed to break down the macromolecules including starch and proteins present in the food waste into simple sugars and amino acids. In this study, we firstly assessed three major commercial enzymes, including protease, α-amylase, and glucoamylase that were purchased from the company. For protease, the optimum condition was evaluated in Duran bottles (100 mL) and was determined as follows: pH 3.0–5.0; temperature 40–50 °C; protease concentration 6,000 U that was equivalent to 150 U g−1 food waste (Fig. 1). Another two commercial enzymes α-amylase and glucoamylase were supplied by the same agent, and their optimum conditions were determined according to the previous studies: pH 4.5, temperature 55 °C, 10 U g−1 food waste of α-amylase and 120 U g−1 food waste of glucoamylase [21]. Under such conditions, hydrolysis was carried out in 1-L scale bioreactor. As shown in Fig. 2, the yields of nutrients increased immediately when hydrolysis started. At 24 h, almost 120 g L−1glucose and 3,000 mg L−1 FAN were obtained, which were much higher than the essential nutritional requirement of A. succinogenes. In general, medium containing 30– 70 g L−1 glucose and 300 mg L−1 FAN would be sufficient for a typical SA fermentation [22]. In the contrary, over supply of nutrients could be unfavorable for the cell growth of A. succinogenes. Our earlier study showed that the cell growth started to be inhibited when the glucose concentration is higher than 70 g L−1 [23]. Meanwhile, food waste hydrolysis using two fungi, namely A. awamori and A. oryzae, were conducted. The fungal mashes of A. awamori and A. oryzae obtained from SSF were used for the production of enzyme complex. SSF is a technique to gain the high enzyme concentration due to the absence of free water and the low protein breakdown [24]. Similar to commercial enzymes, glucose and FAN productions by fungal hydrolysis began immediately when hydrolysis started. According to Table 2, the resultant glucose and FAN concentration were 31.9 g L−1 and 280 mg L−1, respectively, which met the nutritional requirement of A. succinogenes [16]. Compared with commercial enzymes that produced up to 143 g L−1 glucose, hydrolysate produced from fungal hydrolysis contained appropriate carbon and nitrogen concentrations, which is suitable for microbial growth and SA production. Apart from the nutrient-producing capability, another important factor to be considered is the reduction of solid food waste. Outstanding food waste-diminishing capability could be an additional benefit of hydrolysis. The reduced or diminished solids refer to the difference of initial dry weight of food waste and dry weight of remaining solids after hydrolysis. According
Table 1 Composition of bakery waste and mixed food waste (mg g−1) [19]
Bakery waste
Mixed food waste
Carbohydrates
645.5±47.7
470.3±30.5
Starch Proteins
316.7±67.7 98.2±0.9
361.5±13.8 99.3±13.9
Lipids
265.8±34.3
373.7±25.8
Appl Biochem Biotechnol (2014) 174:1822–1833 Fig. 1 Determination of the optimum conditions of protease. Effects of a pH, b temperature, and c the amount of enzyme on the activity were assessed. The hydrolysis was conducted in 100 mL Duran bottles containing 40 g food waste and 40 mL H2O
1827
1828
Appl Biochem Biotechnol (2014) 174:1822–1833
Fig. 2 Enzymatic catalysis of mixed food waste by commercial enzymes. The conditions were as follows: pH 4.5; temperature 55 °C; solid-to-liquid ratio 1:1; 10 U g−1 food waste of α-amylase; 120 U g−1 food waste of glucoamylase; and 150 U g−1 food waste of protease
to Table 2, 80–90 % of the initial solid weight was reduced by fungal hydrolysis, much higher than the hydrolysis using commercial enzymes (66 %). These data suggested the potential of fungal hydrolysis as an alternative to conventional waste treatments such as composting and anaerobic digestion. Results from this part of the study demonstrated that fungal hydrolysis using SSF solid mashes of A. awamori and A. oryzae is the preferred method for producing hydrolysate with the desired nutrient concentrations and superior food waste diminishing capability. Therefore, the resultant hydrolysate produced via fungal hydrolysis was used for the subsequent SA fermentation. Succinic Acid Fermentation In this section, the performance of the two bacteria, namely A. succinogenes and recombinant E. coli in SA fermentations, was separately examined. A. succinogenes is a well-known natural succinate producer and has been successfully used in our previous studies [7, 8]. The recombinant E. coli is a genetically modified strain with all pathways accumulating by-products inactivated,
Table 2 Comparison of hydrolysis using enzymes from fungal solid mashes of A. awamori and A. oryzae and commercial source at 24 h. Both hydrolyses were conducted in 1-L bioreactors Glucose (g L−1) FAN (mg L−1) Weight reduction of food waste after hydrolysis (%)a Fungal solid mashes
31.9 c
Commercial enzymes
119.22
280
80–90b
290.7
66b
a
Percentage of food waste diminished = dry weight of total suspended solids / dry weight of initial food waste × 100 %
b
Pleissner et al. [19]
Combination of protease (150 U g−1 food waste), α-amylase (10 U g−1 food waste), and glucoamylase (120 U g−1 food waste) c
Appl Biochem Biotechnol (2014) 174:1822–1833
1829
Table 3 SA fermentation of A. succinogenes and recombinant E. coli on the hydrolysates produced from mixed food waste A.succinogenes
E. coli
E. coli
Scale
100 mL SARs
100 mL Duran bottle
2.5 L Bench-top fermentator
Total volume of hydrolysate (mL)
60
60
1,500
Initial glucose concentration (g L−1)
32.9
35.1
57.8
OD660/OD600
2.89
13.4
17.5
SA concentration (g L−1)
24.1
26.4
29.9
SA yield (g g−1 glucose) SA productivity (g L−1 h−1)
0.87 0.29
0.98 0.20
0.56 0.48
By-product concentration (g L−1)
13.7
0
0
which highly enhances the SA production [15]. Both bacterial fermentations were initially conducted in 100 mL Duran bottles, in which the total volume of hydrolysate was 60 mL. For A. succinogenes, the initial glucose concentration was 32.9 g L−1 (Table 3). The OD660 significantly increased within the first 16 h, where the maximum value was 4.16. These data indicated that the nutrients in the hyrolysate were sufficient to support the cell growth. As shown in Fig. 3, the SA production started from 8 h and the fermentation lasted for 92 h. After 92 h, 24.1 g L−1 SA was produced, which corresponded to a yield of 0.87 g SA g−1 glucose and a productivity of 0.29 g L−1 h−1. The amount of acid by-products was 13.7 g L−1, which consists of acetic, formic, and pyruvic acids. For small-scale fermentation using E. coli, the initial glucose concentration was 35.1 g L−1. As shown in Fig. 4, the SA concentration continuously increased until the depletion of glucose after 148 h. At the end of fermentation, the SA concentration reached 26.4 g L−1, which
Fig. 3 Fermentation profile of A. succinogenes using the mixed food waste. The fermentation was carried out in 100 mL small anaerobic reactors. The amount of hydrolysate was 60 mL. The average concentrations and error bars of the duplicated experiments are shown
1830
Appl Biochem Biotechnol (2014) 174:1822–1833
Fig. 4 Fermentation profile of recombinant E. coli using the mixed food waste. The fermentation was carried out in 100 mL Duran bottles. The amount of hydrolysate was 60 mL. The average concentrations and error bars of the duplicated experiments are shown
corresponded to a yield of 0.98 g SA g−1 glucose and a productivity of 0.20 g L−1 h−1. It is worth to note that when using mixed food waste hydrolysate as feedstock, the amount of SA (26.4 g L−1) was comparable with that using pastry waste hydrolysate (26.5 g L−1) (Table 4). Again, these results suggested the significant potential of mixed food waste. Compared with A. succinogenes, recombinant E. coli offered several advantages. Firstly, it is a fast-growing strain. The average OD600 value was 13.4, much higher than that of A. succinogenes (OD660 =2.89), indicating significant amount of biomass was produced. Secondly, due to the genetic manipulation, there was no by-product generated during fermentation, so the substantial recovery process can be simplified. Therefore, this recombinant E. coli was selected for bench-top scale fermentation study. It was carried out in a 2.5 L fermentor using food waste hydrolysate with an initial glucose concentration of 57.8 g L−1. As shown in Fig. 5, cell growth immediately began when the fermentation process started. The OD600 increased sharply from 0.176 to 14.4 during the exponential phase (from 0 to 36 h), whilst SA also began to accumulate at the same time. At 60 h, cell growth attained a maximum OD600 of 17.5. The resultant SA concentration was 29.9 g L−1 with a yield of 0.56 g g−1 Table 4 SA fermentations with pastry or food waste hydrolysates as feedstock. Fermentations were conducted in 100 mL Duran bottles, in which the total volume of hydrolysate was 60 mL
SA concentration (g L−1) SA yield (g g−1 glucose) SA productivity (g L−1 h−1)
A. succinogenes
E. coli
Pastry
Food waste
Pastry
Food waste
17.1 0.85
24.1 0.87
26.5 0.79
26.4 0.98
0.24
0.29
0.20
0.20
Appl Biochem Biotechnol (2014) 174:1822–1833
1831
Fig. 5 Fermentation profile of recombinant E. coli using the mixed food waste. The fermentation was carried out in a 2.5 L fermentor. The amount of hydrolysate was 1.5 L. The average concentrations and error bars of the duplicated experiments are shown
glucose, lower than the small-scale fermentation in Duran bottles (yield=0.98 g g−1 glucose). This could be due to glucose utilisation for other maintenance, resulting in higher biomass concentration but a lower SA yield. Interestingly, the SA productivity in fermentor was Table 5 Comparison of SA yields achieved using different food waste substrates Substrate
SA yield (g g−1 glucose)
Overall SA yield (g g−1 substrate)
References
Wheat
0.40
0.40
Du et al. [16, 25]
Wheat flour milling by-product
1.02
0.087
Dorado et al. [18]
Potatoes
N/A
N/A
Delgado et al. [26]
Corncob
0.58
N/A
Yu et al. [27]
Rapeseed meala
0.115
N/A
Chen et al. [28]
Rapeseed mealb
N/A
N/A
Wang et al. [29]
Orange peel
0.58
Negligible
Li et al. [30]
Bread
1.16
0.55
Leung et al. [7]
Cake
0.80
0.28
Zhang et al. [8]
Pastry
0.67
0.35
Zhang et al. [8]
Mixed food waste
0.56
0.224
This study
N/A not available a
Rapeseed meal is treated by diluted sulphuric acid hydrolysis and subsequent enzymatic hydrolysis of pectinase, celluclast, and viscozyme b Rapeseed meal is treated by enzymatic hydrolysis using A. oryzae
1832
Appl Biochem Biotechnol (2014) 174:1822–1833
relatively high (0.48 g L−1 h−1) as compared to fermentations in 100 mL Duran bottles. (0.20 g L−1 h−1). The enhanced oxygen and mass transfers in fermentation broth were due to the use of agitator and the continuous supply of air. In addition, the pH of fermentation broth was maintained at 7 using automated control. This facilitates the SA production by E. coli at its optimal pH range. Therefore, the duration of fermentation in bench-top fermentation (72 h) was only half of that in the Duran bottles (148 h). The final overall SA yield obtained from the mixed food waste was 0.224 g g−1, which is comparable to the results from the earlier studies conducted by our groups using waste bread, cake, and pastry (Table 5).
Conclusions This study investigated the possibility of converting no-value food waste into succinic acid via microbial bioconversion. In the first part of the study, a nutrient-complete medium, namely food waste hydrolysate, was produced using commercial enzymes or fungal mashes with A. awamori and A. oryzae. Results showed that the hydrolysate produced via fungal hydrolysis contains appropriate levels of carbon and nitrogen sources which would be utilized in the subsequent A. succinogenes fermentation. The second part of the study reported the use of food waste hydrolysate as the sole medium in recombinent E. coli fermentation, which led to the production of 29.9 g L−1 SA. Results indicated that E. coli cells were able to utilize the nutrients in the food waste hydrolysate efficiently for cell growth and SA production. The overall yield of SA was 0.224 g g−1 food waste, which is comparable to similar studies using cereal-based wastes or agricultural by-products as substrates. This is the first study reports the bio-based production of SA using mixed food waste from restaurants without any pretreatments. Results of this study demonstrated the enormous potential of mixed food waste as a renewable resource for producing chemicals and materials, suggesting a novel and sustainable approach to alleviate the food waste crisis. Acknowledgments Carol Sze Ki LIN acknowledges the Research Grant Councils in Hong Kong for the provision of General Research Fund (GRF)/Early Career Scheme (ECS) 2013–2014 (Project No. 189713). Zheng SUN acknowledges financial support from the Shanghai Pujiang Program (Grant No. 13PJ1403500) and the Doctoral Fund of Ministry of Education of China (Grant No. 20133104120004).
References 1. Gustavsson, J., Cederberg, C., Sonesson, U., van Otterdijk, R., & Meybeck, A. (2011). Global food losses and food waste. Rome: Food and Agriculture Organization of the United Nations. 2. Buchner, B., Fischler, C., Reilly, J., Riccardi, G., Ricordi, C., & Veronesi, U. (2012). Food waste: Causes, impacts and proposals. Barilla Center for Food & Nutrition. 3. Stuart, T. (2009). Waste: Uncovering the global food scandal. New York: WW Norton & Company. 4. Arancon, R. A. D., Lin, C. S. K., Chan, K. M., Kwan, T. H., & Luque, R. (2013). Energy Science and Engineering, 1, 53–71. 5. Koutinas, A. A., Pleissner, D., Vlysidis, A., Kopsahelis, N., Lopez Garciac, I., Kookosd, I., Papanikolaou, S., & Lin, C. S. K. (2014). Chemical Society Reviews, 43, 2587–2627. 6. Lin, C. S. K., Pfaltzgraff, L. A., Herrero-Davila, L., Mubofu, E. B., Abderrahim, S., Clark, J. H., Koutinas, A. A., Kopsahelis, N., Stamatelatou, K., Dickson, F., Thankappan, S., Mohamed, Z., Brocklesby, R., & Luque, R. (2013). Energy and Environmental Science, 6, 426–464. 7. Leung, C. C. J., Cheung, A. S. Y., Zhang, A. Y., Lam, K. F., & Lin, C. S. K. (2012). Biochemical Engineering Journal, 65, 10–15.
Appl Biochem Biotechnol (2014) 174:1822–1833
1833
8. Zhang, A. Y., Sun, Z., Leung, C. C. J., Han, W., Lau, K. Y., Li, M., & Lin, C. S. K. (2013). Green Chemistry, 15, 690–695. 9. Pleissner, D., Lam, W. C., Sun, Z., & Lin, C. S. K. (2013). Bioresource Technology, 137, 139–146. 10. McKinlay, J. B., Vielle, C., & Zeikus, J. G. (2007). Applied Microbiology and Biotechnology, 76, 727–740. 11. Werpy, T., & Petersen, G. (2004). Top Value Added Chemicals from Biomass, Vol. 1: Results of Screening for Potential Candidates from Sugars and Synthesis Gas (pp. 23–24). Oakridge: U.S. Department of Energy. 12. Bozell, J. J., & Petersen, G. R. (2010). Green Chemistry, 12, 539–554. 13. Taylor, P. (2010). Biosuccinic acid ready for take off? Royal Society of Chemistry, http://www.rsc.org/ chemistryworld/News/2010/January/21011003.asp. 14. de Guzman, D. (2012). Chemical industry awaits for bio-succinic acid potential, ICIS, http://www.icis.com/ Articles/2012/01/30/9527521/chemical-industry-awaits-for-bio-succinic-acid-potential.html. 15. Li, M., Wang, J., Geng, Y., Li, Y., Wang, Q., Liang, Q., & Qi, Q. (2012). Microbial Cell Factories, 11, 19. 16. Du, C., Lin, S. K. C., Koutinas, A., Wang, R., Dorado, P., & Webb, C. (2008). Bioresource Technology, 99, 8310–8315. 17. Li, Y., Li, M., Zhang, X., Yang, P., Liang, Q., & Qi, Q. (2013). Bioresource Technology, 149, 333–340. 18. Dorado, P. M., Lin, S. K. C., Koutinas, A., Du, C., Wang, R., & Webb, C. (2009). Journal of Biotechnology, 143, 51–59. 19. Pleissner, D., Kwan, T. H., & Lin, C. S. K. (2014). Bioresource Technology, 158, 48–54. 20. Koutinas, A. A., Arifeen, N., Wang, R., & Webb, C. (2007). Biotechnology and Bioengineering, 97, 61–72. 21. Yan, S., Yao, J., Yao, L., Zhi, Z., Chen, X., & Wu, J. (2012). Brazilian Archives of Biology and Technology, 55, 183–192. 22. Lin, C. S. K., Luque, R., Clark, J. H., Webb, C., & Du, C. (2012). Biofuels, Bioproducts and Biorefining, 6, 88–104. 23. Lin, C. S. K., Du, C., Koutinas, A., Wang, R., & Webb, C. (2008). Biochemical Engineering Journal, 41, 128–135. 24. Wang, R. (1999). PhD thesis, The University of Manchester, UK. 25. Du, C., Lin, C. S. K., Koutinas, A., Wang, R., & Webb, C. (2007). Applied Microbiology and Biotechnology, 76, 1263–1270. 26. Delgado, R., Castro, A. J., & Vázquez, M. (2009). LWT-Food Science and Technology, 42, 797–804. 27. Yu, J., Li, Z., Ye, Q., Yang, Y., & Chen, S. (2010). Journal of Industrial Microbiology and Biotechnology, 37, 1033–1040. 28. Chen, K. Q., Li, J., Ma, J. F., Jiang, M., Wei, P., & Liu, Z. M. (2011). Bioresource Technology, 102, 1704– 1708. 29. Wang, R., Shaarani, S. M., Godoy, L. C., Melikoglu, M., Vergara, C. S., & Koutinas, A. (2010). Enzyme and Microbial Technology, 47, 77–83. 30. Li, Q., Siles, J. A., & Thompson, I. P. (2010). Applied Microbiology and Biotechnology, 88, 671–678.
