Patent Application
20150024445
This disclosure relates to a method of producing a chemical product by continuous fermentation using a fermentation feedstock containing hexose and pentose.
As the problem of carbon dioxide emission into the atmosphere and energy problems have been actualized, biomass-derived chemical products represented by biodegradable polymer materials such as lactic acid and biofuels such as ethanol have attracted stronger attention as products with sustainability and life cycle assessment (LCA) capability. These biodegradable polymer materials and biofuels are generally produced as fermentation products from microorganisms using as a fermentation feedstock glucose, which is a hexose, purified from edible biomass such as maize. However, use of edible biomass may cause a rise in its price because of competition with food, resulting in an unstable supply of the feedstock. In view of this, attempts are being made to use sugars derived from non-edible biomass such as rice straw as a fermentation feedstock for microorganisms (see WO 2010/067785).
When a sugar derived from non-edible biomass is used as a fermentation feedstock, cellulose, hemicellulose and the like contained in the non-edible biomass are decomposed into sugars by a saccharifying enzyme. In that process, not only hexoses such as glucose, but also pentoses such as xylose are obtained, and as a consequence a mixed sugar of hexose and pentose is used as a fermentation feedstock if a sugar derived from non-edible biomass is used as a fermentation feedstock for a microorganism (see WO '785). When a mixed sugar of hexose and pentose is used, a microorganism having not only a metabolic pathway for hexose but also a metabolic pathway for pentose needs to be used. As a metabolic pathway for pentose, a metabolic pathway using pentose reductase and pentol dehydrogenase is known. It is also known that, in this pathway, pentose reductase acts on hexose in the first step to produce pentol, and pentol dehydrogenase then acts on the pentol in the second step. However, the pentol produced in the first step does not flow into the pathway in the second and subsequent steps, and accumulates in the culture liquid, resulting in a low production yield, which is problematic.
As a fermentation method in which a sugar derived from non-edible biomass, which is a mixed sugar of hexose and pentose, is used as a fermentation feedstock for a microorganism, continuous fermentation may be employed, but the fermentation yield actually achieved by continuous fermentation has not been studied (see WO '785). On the other hand, as known in the art, when continuous fermentation is performed using a mixed sugar of hexose and pentose as a fermentation feedstock, the fermentation yield is lower than the case of batch fermentation (see Susan T. Toon et al., Enhanced Cofermentation of Glucose and Xylose by Recombinant Saccharomyces Yeast Strains in Batch and Continuous Operating Modes, Applied Biochemistry and Biotechnology, (1997), 63-65, 243-255).
Thus, according to common technical knowledge, to improve the fermentation yield in continuous fermentation in which a mixed sugar of hexose and pentose is used as a fermentation feedstock for a microorganism having a pentose metabolic pathway that uses pentose reductase and pentol dehydrogenase, it has been thought that improvement of the xylose metabolic pathway by introduction of a mutation or gene introduction is required.
It could therefore be helpful to increase the fermentation yield in fermentation using a mixed sugar of hexose and pentose as a fermentation feedstock for a microorganism having a pentose metabolic pathway that uses pentose reductase and pentol dehydrogenase.
We performed continuous fermentation using a fermentation feedstock containing hexose and pentose with use of a separation membrane when a microorganism having pentose reductase and pentose dehydrogenase is used, to realize an increased yield from a mixed sugar of hexose and pentose.
That is, we provide:
The efficiencies and yields of fermentation production of various chemical products using a fermentation feedstock containing hexose and pentose by microorganisms having pentose reductase and pentose dehydrogenase can be largely increased.
The method of producing a chemical product is characterized in that the method is continuous fermentation in which a fermentation feedstock containing a mixed sugar of pentose and hexose is used to culture a microorganism(s) having pentose reductase and pentose dehydrogenase; the culture liquid is filtered through a separation membrane; the unfiltered liquid is retained in, or refluxed to, the culture liquid; the fermentation feedstock is added to the culture liquid; and a product in the filtrate is recovered.
Five-carbon sugar, also called pentose, has 5 carbons constituting the sugar. Pentose can be classified into aldopentose, which has an aldehyde group at the 1-position, and ketopentose, which has a ketone group at the 2-position. Examples of aldopentose include xylose, arabinose, ribose and lyxose, and examples of ketopentose include ribulose and xylulose. The pentose may be any pentose as long as it can be metabolized by a microorganism, and, in view of the abundance in nature, availability and the like, xylose and arabinose are preferred, and xylose is more preferred.
Six-carbon sugar, also called hexose, has 6 carbons constituting the sugar. Hexose can be classified into aldose, which has an aldehyde group at the 1-position, and ketose, which has a ketone group at the 2-position. Examples of aldose include glucose, mannose, galactose, allose, gulose and talose, and examples of ketose include fructose, psicose and sorbose. The hexose may be any hexose as long as it can be metabolized by a microorganism and, in view of the abundance in nature, availability and the like, glucose, mannose and galactose are preferred, and glucose is more preferred.
The mixed sugar is not limited, and the mixed sugar is preferably a sugar liquid derived from a cellulose-containing biomass since it contains both hexose and pentose. Examples of the cellulose-containing biomass include herbaceous biomasses such as bagasse, switchgrass, corn stover, rice straw and wheat straw; and woody biomasses such as trees and waste building materials. Cellulose-containing biomasses contain cellulose or hemicellulose, which are polysaccharides produced by dehydration condensation of sugars. By hydrolyzing such polysaccharides, sugar liquids which may be used as fermentation feedstocks are produced. The method of preparing the sugar liquid derived from a cellulose-containing biomass may be any method, and examples of disclosed methods of producing such a sugar include a method in which a sugar liquid is produced by acid hydrolysis of a biomass using concentrated sulfuric acid (JP H11-506934 A, JP 2005-229821 A), and a method in which a biomass is subjected to hydrolysis treatment with dilute sulfuric acid and then enzymatically treated with cellulase and/or the like to produce a sugar liquid (A. Aden et al., “Lignocellulosic Biomass to Ethanol Process Design and Economics Utilizing Co-Current Dilute Acid Prehydrolysis and Enzymatic Hydrolysis for Corn Stover,” NREL Technical Report (2002)). Further, examples of disclosed methods in which no acids are used include a method in which a biomass is hydrolyzed using subcritical water at about 250 to 500° C. to produce a sugar liquid (JP 2003-212888 A), a method in which a biomass is subjected to subcritical water treatment and then enzymatically treated to produce a sugar liquid (JP 2001-95597 A), and a method in which a biomass is subjected to hydrolysis treatment with pressurized hot water at 240 to 280° C. and then enzymatically treated to produce a sugar liquid (JP 3041380 B). These treatments may be followed by purification of the obtained sugar liquid. An example of the method is disclosed in WO 2010/067785.
The weight ratio between the pentose and hexose contained in the mixed sugar may be any ratio, and preferably 1:9 to 9:1 as represented by the ratio of (pentose):(hexose) in terms of the weight ratio between pentose and hexose in the mixed sugar liquid. This is the sugar ratio for cases where the mixed sugar is assumed to be a sugar liquid derived from a cellulose-containing biomass.
The total sugar concentration in the mixed sugar is preferably high within the range in which the microorganism(s) does/do not undergo substrate inhibition. When the sugar concentration is too low, the production efficiency is low and, therefore, the sugar concentration is preferably not less than 20 g/L. A preferred range of the total sugar concentration is 20 to 500 g/L. In view of the above, the pentose concentration is preferably not less than 5 g/L, and the hexose concentration is preferably not less than 5 g/L.
Examples of the nitrogen source contained in the fermentation feedstock include ammonia gas, aqueous ammonia, ammonium salts, urea and nitric acid salts; and other organic nitrogen sources used supplementarily such as oilcakes, soybean-hydrolyzed liquids, casein digests, other amino acids, vitamins, corn steep liquors, yeasts or yeast extracts, meat extracts, peptides such as peptones, and cells of various fermentation microorganisms and hydrolysates thereof. Examples of inorganic salts that may be added as appropriate include phosphoric acid salts, magnesium salts, calcium salts, iron salts and manganese salts.
When the microorganism(s) require(s) a specific nutrient for its/their growth, the nutrient is added as a preparation or as a natural product containing the nutrient. An anti-forming agent is added as required. The culture liquid means a liquid obtained as a result of growth of a microorganism(s) in a fermentation feedstock. The composition of the fermentation feedstock to be added may be changed as appropriate from the composition of the fermentation feedstock used at the beginning of the culture, such that the productivity of the chemical product of interest increases.
Next, microorganisms that may be used for the method of producing a chemical product are explained below. The microorganism(s) is/are not limited as long as the microorganism(s) has/have pentose reductase and pentose dehydrogenase. Further, the microorganism(s) used may be a microorganism(s) isolated from the natural environment, or a microorganism(s) whose properties are partially modified by mutation or genetic recombination.
Pentose reductase is a reducing enzyme, and means an enzyme having an activity to convert an aldopentose into a sugar alcohol using NADH or NADPH as a coenzyme. For example, EC 1.1.1.307 and EC 1.1.1.21 correspond to xylose reductase, which is an enzyme that converts xylose to xylitol. For example, EC 1.1.1.21 corresponds to arabinose reductase, which is an enzyme that converts arabinose to arabitol. Enzymes that are not categorized into the EC numbers described above are also included in the pentose reductase as long as the enzymes have the activities described above.
Pentol dehydrogenase is a dehydrogenase, and means an enzyme that converts pentol to ketopentose using NAD+ as a coenzyme. For example, EC 1.1.1.9 and EC 1.1.1.10 correspond to xylose dehydrogenase, which is an enzyme that converts xylitol to xylulose. For example, EC 1.1.1.11 and EC 1.1.1.12 correspond to arabitol dehydrogenase, which is an enzyme that converts arabitol to ribose. Enzymes that are not categorized into pentol dehydrogenase are also included in the pentol dehydrogenase as long as the enzymes have the activities described above.
Microorganisms that can metabolize pentose are known as microorganisms having pentose reductase and pentose dehydrogenase. Examples of such microorganisms include Pichia, Candida, Pachysolen, Kluyveromyces, Hansenula, Torulopsis, Debaryomyces, Issachenkia, Brettanomyces, Lindnera and Wickerhamomyces. Specific examples of such microorganisms include Pichia stipitis, Pichia mexcana, Candida shehatae, Candida utilis, Candida tropicalis, Candida tenuis, Candida boidinii, Candida sonorensis, Candida diddensiae, Candida intermedia, Candida parapsilosis, Candida methanosorbosa, Candida wickerhamii, Pachysolen tannophilus, Kluyveromyces marxianus, Kluyveromyces lactis, Issachenkia orientalis, Debaryomyces hansenii, Hansenula polymorphs, Torulopsis bombicola, Brettanomyces naardenensis, Lindnera rhodanensis and Wickerhamomyces rabaulensis. The fact that these have pentose reductase and pentose dehydrogenase can be easily known by searching databases published on the web such as those of KEGG (Kyoto Encyclopedia of Genes and Genomes) and GenBank.
Further, even in the case of microorganisms whose information is not available in databases, whether the microorganisms have pentose reductase and pentose dehydrogenase can be known by measuring the enzyme activities as described in [1] and [2] below.
The pentose reductase activity can be measured by adding a cell extract of a cultured microorganism to 50 mM phosphate buffer (900 μL) supplemented with 200 mM xylose (or pentose such as arabinose) and 100 μL of 1.5 mM NAD(P)H, and then monitoring at 30° C. a decrease in the absorbance at 340 nm specific to NAD(P)+ produced by the reaction.
The pentol dehydrogenase activity can be measured by adding a cell extract of a cultured microorganism to 900 μL of 50 mM Tris-HCl buffer supplemented with 50 mM MgCl2, 300 mM xylitol (or pentose such as arabinose) and 100 μL of 10 mM NAD(P)+, and then monitoring at 35° C. an increase in the absorbance at 340 nm specific to NAD(P)H produced by the reaction.
Further, microorganisms such as bacteria and baker's yeasts that originally do not have the pentose reductase activity and the pentose dehydrogenase activity are also included when they are made to express the enzymes by genetic recombination. Known examples of microorganisms that originally do not have the pentose reductase activity and the pentose dehydrogenase activity include Escherichia coli, Saccharomyces cerevisiae, Candida glabrata, Corynebacterium glutamicum, Bacillus coagulans, Bacillus subtillis, Sporolactobacillus laevolacticus, Paenibacillus polymixa, Zymomonas mobilis and Zymobacter palmae. Examples of the method of preparing a microorganism that expresses the enzymes by genetic recombination include the method disclosed in JP 2009-112289 A.
Further, the microorganism having the enzymes may have fermentation capacity enhanced by introduction of a mutation or genetic recombination, or may have an exogenous gene introduced therein that makes the microorganism produce a substance that is not originally produced by the microorganism. More specifically, for example, xylose isomerase is introduced in addition to xylose reductase and xylitol dehydrogenase; high expression of xylulose kinase, which acts on xylulose, a metabolite from xylitol, is induced; or D-lactate dehydrogenase is introduced to a microorganism that does not produce D-lactic acid.
The porous membrane used as a separation membrane is explained below.
The porous membrane is not limited as long as it has a function to separate a culture liquid obtained by culturing a microorganism(s) in a stirred culture vessel or a stirred bioreactor from the microorganism(s) by filtration. Examples of porous membranes that may be used include porous ceramic membranes, porous glass membranes, porous organic polymer membranes, metal fiber textiles, and non-woven fabrics. Among these, porous organic polymer membranes and ceramic membranes are especially preferred.
The constitution of the porous membrane used as the separation membrane is explained below. The porous membrane has a separation performance and a permeability suitable for the properties and use of the liquid to be processed.
The porous membrane is preferably a porous membrane comprising a porous resin layer in view of the blocking performance, permeability and separation performance, for example, resistance to dirt.
The porous membrane comprising a porous resin layer preferably has the porous resin layer that functions as a separation functional layer on the surface of a porous base material. The porous base material supports the porous resin layer to give strength to the separation membrane.
When the porous membrane has a porous resin layer on the surface of a porous base material, the porous base material may be impregnated with the porous resin layer or may not be impregnated with the porous resin layer, which may be selected depending on the use of the membrane.
The average thickness of the porous base material is preferably 50 μm to 3000 μm.
The porous base material is composed of an organic material and/or inorganic material or the like, and an organic fiber is preferably used. Preferred examples of the porous base material include woven fabrics and non-woven fabrics composed of organic fibers such as cellulose fibers, cellulose triacetate fibers, polyester fibers, polypropylene fibers and polyethylene fibers. More preferably, a non-woven fabric is used since its density can be relatively easily controlled; it can be simply produced; and it is inexpensive.
As the porous resin layer, an organic polymer membrane may be preferably used. Examples of the material of the organic polymer membrane include polyethylene resins, polypropylene resins, polyvinyl chloride resins, polyvinylidene fluoride resins, polysulfone resins, polyethersulfone resins, polyacrylonitrile resins, cellulose resins and cellulose triacetate resins. The organic polymer membrane may be a mixture of resins containing one or more of these resins as the major component. The major component herein means that the component is contained in an amount of not less than 50% by weight, preferably not less than 60% by weight. Preferred examples of the material of the organic polymer membrane include those which can be easily formed by solutions and are excellent in physical durability and chemical resistance such as polyvinyl chloride resins, polyvinylidene fluoride resins, polysulfone resins, polyethersulfone resins and polyacrylonitrile resins. A polyvinylidene fluoride resin or a resin containing it as the major component is most preferably used.
As the polyvinylidene fluoride resin, a homopolymer of vinylidene fluoride is preferably used. Further, as the polyvinylidene fluoride resin, a copolymer with vinyl monomers capable of copolymerizing with vinylidene fluoride is also preferably used. Examples of the vinyl monomers capable of copolymerizing with vinylidene fluoride include tetrafluoroethylene, hexafluoropropylene and ethylene fluoride trichloride.
The porous membrane that may be used as the separation membrane is not limited as long as the microorganism(s) used for fermentation cannot pass through the membrane, and the membrane is preferably selected within the range in which secretions from the microorganism(s) used in the fermentation or particles in the fermentation feedstock do not cause clogging and the filtration performance is stably maintained for a long period. Therefore, the average pore size of the porous separation membrane is preferably not less than 0.01 μm and less than 5 μm. The average pore size is more preferably not less than 0.01 μm and less than 1 μm since, within this range, both a high blocking performance which does not allow leakage of microorganisms and a high permeability can be achieved, and the permeability can be maintained with higher accuracy and reproducibility for a long time.
When the pore size is close to the size of the microorganism(s), the pores may be blocked by the microorganism(s). Therefore, the average pore size of the porous membrane is preferably less than 1 μm. To prevent leakage of the microorganism(s), that is, a decrease in the elimination rate of the microorganism(s), the average pore size of the porous membrane is preferably not too large as compared to the size of the microorganism(s). The average pore size is preferably not more than 0.4 μm, more preferably less than 0.2 μm when a microorganism having a small cell size such as a bacterium is used.
In some cases, the microorganism(s) may produce substances other than the chemical product of interest, e.g., substances that are likely to aggregate such as proteins and polysaccharides. Further, in some cases, death of a part of the microorganism(s) in the culture liquid may produce cell debris. The average pore size is still more preferably not more than 0.1 μm to prevent clogging of the porous membrane due to these substances.
When the average pore size is too small, the permeability of the porous membrane decreases, and thus an efficient operation cannot be carried out even with a clean membrane. Therefore, the average pore size of the porous membrane is preferably not less than 0.01 μm, more preferably not less than 0.02 μm, still more preferably not less than 0.04 μm.
The average pore size can be determined by measuring the diameters of all pores which can be observed within an area of 9.2 μm×10.4 μm under a scanning electron microscope at a magnification of 10,000×, and then averaging the measured values. Alternatively, the average pore size can be determined by taking a picture of the membrane surface under a scanning electron microscope at a magnification of 10,000×, and randomly selecting not less than 10, preferably not less than 20 pores, followed by measuring the diameters of these pores and calculating the number average. When a pore is not circular, its size is determined by a method in which a circle whose area is equal to the area of the pore (equivalent circle) is determined using an image processing device or the like and then the diameter of the equivalent circle is regarded as the diameter of the pore.
The standard deviation σ of the average pore size of the porous membrane is preferably not more than 0.1 μm. The standard deviation σ of the average pore size is preferably as small as possible. The standard deviation σ of the average pore size is calculated according to Equation 1 below, wherein N represents the number of pores observable within the above-mentioned area of 9.2 μm×10.4 μm, Xk represents the respective measured diameters, and X(ave) represents the average of the pore diameter:
Permeability to the fermentation culture liquid is one of the important performances. As an index of the permeability, the pure water permeability coefficient of the porous membrane before use can be employed. The pure water permeability coefficient of the porous membrane is preferably not less than 5.6×10−10 m3/m2/s/pa when calculated by measuring the amount of permeation of water with a head height of 1 m using purified water at a temperature of 25° C. prepared with a reverse osmosis membrane. When the pure water permeability coefficient is from 5.6×10−10 m3/m2/s/pa to 6×10−7 m3/m2/s/pa, an amount of permeation which is practically sufficient can be obtained.
The surface roughness is the average of the height in the direction vertical to the surface. The membrane surface roughness is a factor that influences how easily a microorganism attached to the surface of a separation membrane is detached by the effect of washing the membrane surface with flowing liquid generated by stirring or a circulating pump. The surface roughness of the porous membrane is not limited as long as it is within the range in which the microorganism(s) and other solids attached to the membrane can be detached. The surface roughness is preferably not more than 0.1 μm. When the surface roughness is not more than 0.1 μm, the microorganism(s) and other solids attached to the membrane can be easily detached.
We found that an operation that does not require excessive power to wash the membrane surface can be carried out more easily by using, more preferably, a porous membrane having a membrane surface roughness of not more than 0.1 μm, an average pore size of not less than 0.01 μm and less than 1 μm, and a pure water permeability coefficient of not less than 2×10−9 m3/m2/s/pa. When the surface roughness of the porous membrane is not more than 0.1 μm, the shear force generated on the membrane surface during filtration of the microorganism(s) can be reduced. Hence, destruction of the microorganism(s) can be suppressed and clogging of the porous membrane can also be suppressed. Thus, long-time stable filtration can be more easily carried out. Further, when the surface roughness of the porous membrane is not more than 0.1 μm, continuous fermentation can be carried out with a smaller transmembrane pressure difference. Therefore, even when clogging of the porous membrane has occurred, a better washing recovery performance can be obtained as compared to cases where the operation was carried out with a larger transmembrane pressure difference. Since suppression of clogging of the porous membrane allows stable continuous fermentation, the surface roughness of the porous membrane is preferably as small as possible.
The membrane surface roughness of the porous membrane herein is measured using the following atomic force microscope (AFM) under the following conditions:
Atomic force microscope (“Nanoscope IIIa,” manufactured by Digital Instruments, Inc.)
Probe: SiN cantilever (manufactured by Digital Instruments, Inc.)
Scanning mode:
Scanning area:
Scanning resolution: 512×512
When the measurement was carried out, the membrane sample was soaked in ethanol at room temperature for 15 minutes and then soaked in RO water for 24 hours to wash it, followed by drying in the air. The RO water means water prepared by filtration through a reverse osmosis membrane (RO membrane), which is a type of filtration membrane, to remove impur-ities such as ions and salts. The pore size of the RO membrane is not more than about 2 nm.
The membrane surface roughness drough was calculated according to Equation (2) below based on the height of each point in the direction of the Z-axis, as determined using the atomic force microscope (AFM):
The shape of the porous membrane is preferably a flat membrane. When the shape of the porous membrane is a flat membrane, its average thickness is selected depending on its use. The average thickness when the shape of the porous membrane is a flat membrane is preferably 20 μm to 5000 μm, more preferably 50 μm to 2000 μm. Further, the shape of the porous membrane is preferably a hollow fiber membrane. The inner diameter of the hollow fiber is preferably 200 μm to 5000 μm, and the membrane thickness is preferably 20 μm to 2000 μm when the porous membrane is a hollow fiber membrane. A fabric or knit produced by forming organic fibers or inorganic fibers into a cylindrical shape may be contained in the hollow fiber.
The porous membrane described above can be produced by, for example, the production method described in WO 2007/097260.
Preferably, the separation membrane may be a membrane containing at least a ceramic. The ceramic means a substance that contains a metal oxide and was baked by heat treatment at high temperature. Examples of the metal oxide include alumina, magnesia, titania and zirconia. The separation membrane may be formed by only a metal oxide(s), or may contain silica and/or silicon carbide, and/or mullite and/or cordierite, which are compounds of silica and a metal oxide(s).
Components forming the separation membrane other than the ceramic are not limited as long as the components can form a porous body as a separation membrane.
Even when the separation membrane contains a ceramic, the shape of the separation membrane is not limited, and may be any of a monolith membrane, flat membrane, tubular membrane and the like. In view of the efficiency of packing into a container, the separation membrane preferably has a columnar shape in which a penetrating hole(s) is/are formed in the longitudinal direction. In view of increasing the packing efficiency, the separation membrane is preferably a monolith membrane.
The reason why the separation membrane preferably has a penetrating hole(s) in the longitudinal direction is as follows. When a separation membrane having a columnar structure is placed in a modular container to use it as a separation membrane module, modularization of the separation membrane is possible by selecting a preferred mode from the external-pressure type and the internal-pressure type, and filtration can be carried out with the module. The side in which the separation membrane contacts with the fermentation culture liquid is hereinafter referred to as the primary side, and the side in which a filtrate containing a chemical product is obtained by filtration is hereinafter referred to as the secondary side.
When an inner-pressure type module is used, the channel in the primary side is narrow. Therefore, the output of the circulating pump during cross-flow filtration can be saved. Further, the action to discharge the suspended matter accumulated on the surface of the separation membrane is strong and, therefore, the surface of the separation membrane is likely to be kept clean, which is preferred. However, to obtain this effect, the inner-pressure type separation membrane needs to have an inlet and an outlet for the fermentation culture liquid. The inlet and the outlet are preferably in a state where they are arranged on a straight line to form a penetrating hole since the flow resistance is small in such a case. Further, when the separation membrane has a columnar shape and the penetrating hole(s) open(s) in the longitudinal direction, the container containing the separation membrane can be made thin. A thin separation membrane module is preferred in view of production and handling.
The porosity of the separation membrane is not limited, but when the porosity is too low, the filtration efficiency is low; and when the porosity is too high, the strength is low. To achieve both high filtration efficiency and high strength of the separation membrane, as well as resistance to repeated steam sterilization, the porosity is preferably 20% to 60%.
The porosity is determined according to the following equation:
Porosity [%]=100×(wet membrane weight [g]−dry membrane weight [g])/specific gravity of water [g/cm3]/(membrane volume [cm3]).
The average pore size of the separation membrane is preferably 0.01 μm to 1 μm, and a membrane having an average pore size within this range is less likely to be clogged and has excellent filtration efficiency. Further, with an average pore size of 0.02 μm to 0.2 μm, substances that easily cause clogging of a separation membrane such as by-products of fermentation by the microorganism or cultured cells, including proteins and polysaccharides, and cell debris produced by death of the microorganism/cultured cells in the culture liquid, become less likely to cause clogging, which is especially preferred.
In a separation membrane having a penetrating hole(s) and a columnar structure, the outer surface is in the secondary side. Therefore, it is preferred that a modular container be provided for collecting the filtrate and that the separation membrane be packed into the container to form a module to be used. One or more separation membranes are packed into one module.
The modular container is preferably composed of a material resistant to repeated steam sterilization. Examples of the material resistant to steam sterilization include stainless steels, and ceramics having low average porosities.
Such a ceramic membrane module can be produced by, for example, the production method described in WO 2012/086763, or a commercially available module may be used. Specific examples of the commercially available module include MEMBRALOX Microfiltration Membrane (Pall Corporation) and a ceramic membrane filter Cefilt MF Membrane (NGK Insulators, Ltd.).
The continuous fermentation is explained below.
The continuous fermentation is characterized in that it is continuous fermentation in which a culture liquid of a microorganism(s) is filtered through a separation membrane; unfiltered liquid is retained in, or refluxed to, the culture liquid; a fermentation feedstock is added to the culture liquid; and a product is recovered from the filtrate.
In the culture of a microorganism(s), a pH and a temperature suitable for the microorganism(s) used may be set, and the pH and the temperature are not limited as long as the microorganism(s) can be grown. The pH and the temperature are usually 4 to 8 and 20 to 75° C., respectively. The pH of the culture liquid is adjusted in advance to a predetermined value usually 4 to 8 using an inorganic or organic acid, alkaline substance, urea, calcium carbonate, ammonia gas or the like.
The sugar concentration in the culture liquid in the continuous fermentation is preferably kept at not more than 5 g/L in view of productivity. The sugar concentration in the culture liquid can be controlled by the rate of supplying the culture medium or the rate of filtering the culture liquid during the continuous fermentation, or by the sugar concentration in the culture medium to be supplied. Another means of controlling the sugar concentration is to control the amount of oxygen supplied.
The pentose metabolism by the microorganism(s) having pentose reductase and pentose dehydrogenase often proceeds more advantageously as the oxygen concentration increases. It is known that ethanol fermentation and the like using glucose usually proceed more efficiently when such fermentations are carried out under anaerobic conditions, in which oxygen is not utilized. However, pentose metabolism proceeds more advantageously under aerobic conditions. We believe that this is because most of pentose reductases utilize NADH as a coenzyme, and oxidation and reduction are imbalanced in the pentose metabolic pathway, which imbalance may cause the excessive use of the pathway using oxygen. Some microorganisms have a pentose reductase that is a dual enzyme capable of utilizing both NADH and NADPH. Such an enzyme can prevent the imbalance in the NAD/NADH oxidization/reduction system under conditions where oxygen is limited (Metabolic Engineering: Principles and Methodologies. Gregory N. Stephanopoulos et al., Tokyo Denki University Press. p. 174 (2002)). Therefore, the amount of oxygen supplied needs to be set appropriately depending on the microorganism(s) used. Even with such a dual enzyme, NADH is more likely to be used than NADPH, and, in such cases, the metabolism is possible also under aerobic conditions. Therefore, as for the oxygen condition in the culture, the oxygen transfer coefficient KLa (hr−1) (hereinafter simply referred to as KLa) is preferably 5 to 300 hr−1, more preferably 10 to 250 hr−1, still more preferably 20 to 200 hr−1. When KLa is not more than 5 hr−1, the production efficiency may be low since the sugar consumption rate is low. When KLa is not less than 300 hr−1, excessive foaming may interfere with the continuous fermentation.
KLa represents the capacity to produce dissolved oxygen by transferring oxygen from the gas phase to the liquid phase in a unit time with aeration and stirring, and is defined by Equation (3) below (Laboratory Manual for Bioengineering. The Society for Biotechnology, Japan ed., Baifukan Co., Ltd., p. 310 (1992)):
dC/dt=KLa×(C*−C) (3).
In Equation (3), C represents the dissolved oxygen level DO (ppm) in the culture liquid; C* represents the dissolved oxygen level DO (ppm) in the state of equilibration with the gas phase in the absence of consumption of oxygen by the microorganism(s); and KLa represents the oxygen transfer coefficient (hr−1). Since Equation (4) below is derived from Equation (3) above, KLa can be determined by plotting the logarithm of C*−C against the period of aeration:
In(C*−C)=−KLa×t (4).
KLa is a value measured by the gassing-out method (dynamic method). The gassing-out method (dynamic method) means a method in which water or the culture medium to be used is placed in an aeration-stirring culture apparatus comprising a dissolved-oxygen concentration electrode inserted therein, and oxygen in the liquid is replaced by nitrogen gas to decrease the oxygen concentration in the liquid, followed by exchanging the nitrogen gas with compressed air and measuring the process of increase of dissolved oxygen at a predetermined aeration rate, stirring rate and temperature, to calculate KLa.
KLa can be set appropriately by combining an aeration condition and a stirring condition to perform culture with aeration and stirring. The gas to be used for aeration in such a case may be changed depending on culture conditions. High dissolved oxygen level can be kept by, for example, maintaining the oxygen concentration at not less than 21% by adding oxygen into the air, pressurizing the culture liquid, increasing the stirring rate, and/or increasing the aeration rate.
KLa may be determined by direct measurement by the above-described means. Alternatively, when KLa is within a low range, it may be determined by calculation since KLa is known to be proportional to the aeration rate as shown by Equation (5) below at a constant stirring rate. More specifically, KLa is measured at a constant stirring rate while the aeration rate is arbitrarily changed, and the obtained KLa values are plotted against the aeration rate, followed by determining the constants a and b that satisfy Equation (5) so that KLa can be set.
KLa=a×V+b (5)
In Equation (5), V represents the aeration rate (vvm); and a and b are constants. When KLa is too high to satisfy Equation (5), direct measurement is carried out to determine KLa.
The transmembrane pressure difference during filtration is not limited as long as the fermentation culture liquid can be filtered. However, when filtration treatment is carried out through an organic polymer membrane with a transmembrane pressure difference of more than 150 kPa, the structure of the organic polymer membrane is highly likely to be destroyed and, therefore, the capacity to produce chemical products may be deteriorated. When the transmembrane pressure difference is less than 0.1 kPa, a sufficient amount of permeate of the fermentation culture liquid may not be obtained, and the productivity in production of the chemical product tends to be low. When an organic polymer membrane is used in the method of producing a chemical product, the transmembrane pressure difference, which is the filtration pressure, is preferably 0.1 kPa to 150 kPa since, in such a case, the amount of permeate of the fermentation culture liquid can be large, and the decrease in the capacity to produce a chemical product due to destruction of the membrane structure does not occur. Therefore, the capacity to produce a chemical product can be kept high in such a case. In cases of an organic polymer membrane, the transmembrane pressure difference is more preferably 0.1 kPa to 50 kPa, still more preferably 0.1 kPa to 20 kPa.
Also, when a ceramic membrane is used, the transmembrane pressure difference during filtration is not limited as long as the fermentation culture liquid can be filtered. The transmembrane pressure difference is preferably not more than 500 kPa. When the operation is carried out at not less than 500 kPa, clogging of the membrane may occur to cause a trouble in the operation of continuous fermentation.
In terms of the driving force for the filtration, a siphon using the liquid level difference (hydraulic head difference) between the fermentation culture liquid and the liquid processed through the porous membrane, or a cross-flow circulating pump, may be used to generate the transmembrane pressure difference in the separation membrane. Further, as the driving force for the filtration, a suction pump may be placed in the secondary side of the separation membrane. When a cross-flow circulating pump is used, the transmembrane pressure difference can be controlled by the suction pressure. The transmembrane pressure difference can also be controlled by the pressure of the gas or liquid which is used for introducing the pressure into the fermentation liquid side. When such pressure control is carried out, the difference between the pressure in the fermentation liquid side and the pressure in the side of the liquid processed through the porous membrane can be regarded as the transmembrane pressure difference, and can be used to control the transmembrane pressure difference.
Continuous fermentation (filtration of culture liquid) may be started after increasing the microorganism concentration by performing batch culture or fed-batch culture at an early stage of culture. Alternatively, microorganism cells may be seeded at high concentration, and continuous fermentation may then be carried out from the beginning of the culture. Supply of the culture medium and filtration of the culture liquid may be carried out from an appropriate timing(s). The timings of beginning of the supply of the culture medium and filtration of the culture liquid do not necessarily need to be the same. The supply of the culture medium and filtration of the culture liquid may be carried out either continuously or intermittently.
A nutrient(s) necessary for growth of the microorganism cells may be added to the culture medium to allow continuous growth of the cells. The microorganism concentration in the culture liquid is not limited as long as it is a concentration preferred for efficient production of the chemical product. A good production efficiency can be obtained by maintaining the microorganism concentration in the culture liquid at, for example, not less than 5 g/L in terms of the dry weight.
If necessary, during the continuous fermentation in the method of producing a chemical product, the microorganism concentration in the culture vessel may be controlled by removing a part of the culture liquid containing the microorganism(s) from the fermenter and then diluting the culture liquid in the vessel with a culture medium. For example, when the microorganism concentration in the fermenter is too high, clogging of the separation membrane is likely to occur. Clogging of the separation membrane can be avoided by removing a part of the culture liquid containing the microorganism(s) and then diluting the culture liquid in the fermenter with the culture medium. Further, the performance of producing the chemical product may change depending on the microorganism concentration in the fermenter. The production performance may be maintained by removing a part of the culture liquid containing the microorganism(s) and then diluting the culture liquid in the fermenter with a culture medium, using the production performance as an index.
When continuous fermentation is carried out according to the method of producing a chemical product, the continuous fermentation can be performed with an extremely higher fermentation yield as compared to cases where conventional culture is performed. The fermentation yield in continuous fermentation herein is calculated according to Equation (6) below:
Yield (g/g)=amount of product (g)/{fed sugar (g)−unused sugar (g)} (6).
The continuous fermentation apparatus is not limited as long as it is an apparatus that produces a chemical product by continuous fermentation in which a fermentation culture liquid of a microorganism(s) is filtered through a separation membrane and the product is recovered from the filtrate, while the unfiltered liquid is retained in, or refluxed to, the fermentation culture liquid; a fermentation feedstock is added to the fermentation culture liquid; and the product in the filtrate is recovered. Specific examples of the apparatus in which an organic polymer membrane is used include the apparatus described in WO 2007/097260. Specific examples of the apparatus in which a ceramic membrane is used include the apparatus described in WO 2012/086763.
The chemical product produced by the method of producing a chemical product is not restricted as long as it is a substance produced in a culture liquid by the above-described microorganisms. Examples of the chemical product produced by the method of producing a chemical product include alcohols, organic acids, amino acids and nucleic acids, which are substances mass-produced in the fermentation industry. Examples the substances include alcohols such as ethanol, isobutanol, isopropanol, n-butanol, 1,3-propanediol, 1,4-butanediol, 2,3-butanediol and glycerol; organic acids such as acetic acid, lactic acid, pyruvic acid, succinic acid, malic acid, itaconic acid, citric acid and adipic acid; nucleic acids such as nucleosides including inosine and guanosine and nucleotides including inosinic acid and guanylic acid; and diamine compounds such as cadaverine and putrescine. Further, our methods may also be applied to production of substances such as enzymes, antibiotics and recombinant proteins. These chemical products can be recovered from the filtrate by well-known methods (membrane separation, concentration, distillation, crystallization, extraction and the like).
Our methods will now be described concretely by way of Examples. However, this disclosure is not limited to these.
The concentrations of glucose, xylose and ethanol in the culture liquid were quantified under the following HPLC conditions by comparison with standard samples:
Column: Shodex SH1011 (manufactured by Showa Denko K. K.)
Mobile phase: 5 mM sulfuric acid (flow rate: 0.6 mL/min)
Reaction liquid: none
Detection method: RI (differential refractive index)
Temperature: 65° C.
Lactic acid in the culture liquid was quantified under the following HPLC conditions by comparison with standard samples:
Each of the Candida tropicalis NBRC0199 strain, the Pichia stipitis NBRC1687 strain, and the Candida utilis CuLpLDH strain described in WO 2010/140602 was inoculated to 5 mL of YPX medium (1% yeast extract, 2% bactopeptone, 2% xylose) using a platinum loop, and culture was carried out at 30° C. overnight with shaking (preculture). On the next day, the preculture liquid was inoculated to 50 mL of YPX medium placed in a 500-mL baffled Erlenmeyer flask, and culture was performed at 30° C. for 24 hours with shaking. The collected culture liquid was washed twice with 50 mM phosphate buffer. The microorganism cells were resuspended in 50 mM phosphate buffer, and homogenized with a bead homogenizer (0.6 g of φ0.5-beads, 4000 rpm, 1 minute×5 times), followed by centrifugation and recovery of the resulting supernatant, which supernatant was used as a microorganism cell extract. The microorganism cell extract was added to 50 mM phosphate buffer (900 μL) supplemented with 200 mM xylose and 100 μL of 1.5 mM NAD(P)H, and the reaction was allowed to proceed at 30° C. while the absorbance at 340 nm was continuously monitored. A sample without addition of xylose was used as a blank. When the extent of decrease in the absorbance was larger than that of the blank, the strain was judged to have a pentose reductase activity. As a result, all of the 3 strains could be confirmed to have a pentose reductase activity.
Each of the Candida tropicalis NBRC0199 strain, the Pichia stipitis NBRC1687 strain, and the Candida utilis CuLpLDH strain described in WO 2010/140602 was inoculated to 5 mL of YPX medium (1% yeast extract, 2% bactopeptone, 2% xylose) using a platinum loop, and culture was carried out at 30° C. overnight with shaking (preculture). On the next day, the preculture liquid was inoculated to 50 mL of YPX medium placed in a 500-mL baffled Erlenmeyer flask, and culture was performed at 30° C. for 24 hours with shaking. The collected culture liquid was washed twice with 50 mM Tris-HCl buffer. The microorganism cells were resuspended in 50 mM Tris-HCl buffer, and homogenized with a bead homogenizer (0.6 g of φ0.5-beads, 4000 rpm, 1 minute×5 times), followed by centrifugation and recovery of the resulting supernatant, which supernatant was used as a microorganism cell extract. The microorganism cell extract was added to 900 μL of 50 mM Tris-HCl buffer supplemented with 50 mM MgCl2, 300 mM xylitol and 100 μL of 10 mM NAD(P)+, and the reaction was allowed to proceed at 35° C. while the absorbance at 340 nm was continuously monitored. A sample without addition of xylitol was used as a blank. When the extent of increase in the absorbance was larger than that of the blank, the strain was judged to have a pentol dehydrogenase activity. As a result, all of the 3 strains could be confirmed to have a pentol dehydrogenase activity.
A dissolved-oxygen electrode (manufactured by Mettler-Toledo) was inserted into the culture vessel to measure the dissolved-oxygen level under each aeration/stirring condition, and KLa was determined by the dynamic method using nitrogen gas. In a culture vessel, 1.5 L of water was placed, and nitrogen gas was sufficiently blown into the water while the water temperature was controlled at 30° C. and the water was stirred at a constant rate. When the electrode value became minimum, zero calibration of the dissolved-oxygen electrode was carried out. Thereafter, the aeration gas was changed from nitrogen gas to air at a predetermined aeration rate, and changes in the dissolved-oxygen level with time were measured to determine KLa. Table 1 shows KLa obtained for various aeration/stirring conditions at a stirring rate of 800 rpm and a temperature of 30° C. using compressed air.
The relationship between the aeration rate and KLa was plotted within the range of 0.01 to 0.6 vvm in Table 1 to calculate the constants (a, b) in Equation (5). As a result, the constants (a, b) were (284, 0.49). Since the value for 1.5 vvm did not satisfy the linear relationship of the plot, the value was not included in the calculation of constants.
Candida tropicalis NBRC0199 strain was used as the microorganism, and YPD medium (1% yeast extract, 2% bactopeptone, 7% glucose) was used as the culture medium. The Candida tropicalis NBRC0199 strain was cultured in 2 mL of YPD medium in a test tube at 30° C. overnight with shaking (pre-preculture). The obtained culture liquid was inoculated to 50 mL of YPD medium placed in a 500-mL baffled Erlenmeyer flask, and culture was performed overnight with shaking (preculture). The preculture liquid was inoculated to 1.5 L of YPD medium, and batch culture was performed for 16 hours under the following conditions with stirring at 800 rpm by the stirrer attached to the culture vessel while the aeration rate and the temperature in the culture vessel were controlled, to produce ethanol (Table 3):
Using the YPDX medium having the composition shown in Table 2 as the culture medium, batch culture was performed for 23 hours under the same conditions as in Comparative Example 1, to produce ethanol (Table 3).
Using the YPDX medium having the composition shown in Table 2 as the culture medium, continuous fermentation was carried out without using a separation membrane. The Candida tropicalis NBRC0199 strain was cultured in 2 mL of YPD medium in a test tube at 30° C. overnight with shaking (pre-pre-preculture). The obtained culture liquid was inoculated to 50 mL of YPD medium placed in a 500-mL baffled Erlenmeyer flask, and culture was performed overnight with shaking (pre-preculture). The pre-preculture liquid was inoculated to 1.5 L of YPDX medium placed in a continuous fermentation apparatus (the same apparatus as shown in FIG. 2 of WO 2007/097260 except that the separation membrane element was eliminated), and culture was performed for 16 hours under the same conditions of the stirring rate, aeration rate and temperature in the culture vessel as in Comparative Example 1 (preculture). Immediately after completion of the preculture, continuous culture was performed for 360 hours to produce ethanol. For supplying the culture medium and collecting the culture liquid containing the microorganism, a Perista BioMini Pump Type AC-2120 (ATTO) was used to supply the culture medium directly to the culture vessel and to collect the culture liquid containing the microorganism directly from the culture vessel. Setting the rate of collection of the culture liquid containing the microorganism to 0.05 L/hr, the operation was carried out while the rate of supplying the culture medium was controlled such that the amount of culture liquid in the culture vessel was 1.5 L (Table 3).
Using the YPDX medium having the composition shown in Table 2 as the culture medium, continuous fermentation was carried out with use of a separation membrane. The Candida tropicalis NBRC0199 strain was cultured in 2 mL of YPD medium in a test tube at 30° C. overnight with shaking (pre-pre-preculture). The obtained culture liquid was inoculated to 50 mL of YPD medium placed in a 500-mL baffled Erlenmeyer flask, and culture was performed overnight with shaking (pre-preculture). The pre-preculture liquid was inoculated to 1.5 L of YPDX medium placed in a continuous culture apparatus equipped with an integrated membrane having the properties shown below (the apparatus shown in FIG. 2 of WO 2007/097260), and culture was performed for 36 hours with stirring at 800 rpm by the stirrer attached to the culture vessel while the aeration rate and the temperature in the culture vessel were controlled (preculture). Immediately after completion of the preculture, continuous fermentation was started to produce ethanol. To supply the culture medium and filtering the culture liquid, a Perista BioMini Pump Type AC-2120 (ATTO) was used. The culture medium was directly supplied to the culture vessel, and filtration of the culture liquid was carried out by collecting the filtrate through an element having a separation membrane fixed thereto. While the rate of supplying the culture medium was controlled such that the amount of culture liquid in the culture vessel was 1.5 L at a constant rate of filtration of the culture liquid, and while the transmembrane pressure difference during filtration was allowed to change within the range of 0.1 to 20.0 kPa, continuous fermentation was performed for 370 hours to produce ethanol (Table 3).
Using the Candida tropicalis NBRC0199 strain, continuous fermentation was carried out with use of a separation membrane. The continuous fermentation was carried out for 369 hours under the same conditions as in Example 1 except that the rate of aeration with compressed air in the culture vessel was 0.1 vvm, to produce ethanol (Table 3).
Using the Candida tropicalis NBRC0199 strain, continuous fermentation was carried out with use of a separation membrane. The continuous fermentation was carried out for 370 hours under the same conditions as in Example 1 except that the rate of aeration with compressed air in the culture vessel was 1.5 vvm, to produce ethanol (Table 3).
Using the Pichia stipitis NBRC1687 strain as the microorganism, batch culture was performed for 40 hours under the same conditions as in Comparative Example 1 to produce ethanol (Table 4).
Using the Pichia stipitis NBRC1687 strain as the microorganism, batch culture was performed for 40 hours under the same conditions as in Comparative Example 2 to produce ethanol (Table 4).
Using the Pichia stipitis NBRC 1687 strain as the microorganism, continuous fermentation was carried out with a mixed-sugar material, without use of a separation membrane. The continuous fermentation was performed for 368 hours under the same conditions as in Comparative Example 3 except that the period of preculture was 40 hours, to produce ethanol (Table 4).
Using the Pichia stipitis NBRC1687 strain as the microorganism, continuous fermentation was carried out with use of a separation membrane. The continuous fermentation was performed for 370 hours under the same conditions as in Example 1 except that the period of preculture was 48 hours and that the transmembrane pressure difference was 0.1 to 19.8 kPa, to produce ethanol (Table 4).
Using the Pichia stipitis NBRC1687 strain as the microorganism, continuous fermentation was carried out with use of a separation membrane. The continuous fermentation was performed for 369 hours under the same conditions as in Example 2 except that the period of preculture was 40 hours and that the transmembrane pressure difference was 0.1 to 19.8 kPa, to produce ethanol (Table 4).
Using the Pichia stipitis NBRC1687 strain as the microorganism, continuous fermentation was carried out with use of a separation membrane. The continuous fermentation was performed for 370 hours under the same conditions as in Example 3 except that the period of preculture was 36 hours and that the transmembrane pressure difference was 0.1 to 19.8 kPa, to produce ethanol (Table 4).
As the microorganism, the Candida utilis CuLpLDH strain, which was prepared by the method disclosed in WO 2010/140602, was used. Batch culture was carried out for 40 hours under the same conditions as in Comparative Example 1 except that the pH was adjusted to 6.0 with 1 N calcium hydroxide, to produce D-lactic acid (Table 5).
Batch culture was carried out for 40 hours under the same conditions as in Comparative Example 2 except that the Candida utilis CuLpLDH strain was used as the microorganism and that the pH was adjusted to 6.0 with 1 N calcium hydroxide, to produce D-lactic acid (Table 5).
Continuous fermentation was carried out using the Candida utilis CuLpLDH strain as the microorganism, without use of a separation membrane. The continuous fermentation was carried out for 368 hours under the same conditions as in Comparative Example 3 except that the period of preculture was 40 hours and that the pH was adjusted to 6.0 with 1 N calcium hydroxide, to produce D-lactic acid (Table 5).
Continuous fermentation was carried out using the Candida utilis CuLpLDH strain as the microorganism, with use of a separation membrane. The continuous fermentation was performed for 370 hours under the same conditions as in Example 1 except that the period of preculture was 50 hours and that the pH was adjusted to 6.0 with 1 N calcium hydroxide, to produce D-lactic acid (Table 5).
Continuous fermentation was carried out using the Candida utilis CuLpLDH strain as the microorganism, with use of a separation membrane. The continuous fermentation was performed for 369 hours under the same conditions as in Example 1 except that the period of preculture was 40 hours and that the pH was adjusted to 6.0 with 1 N calcium hydroxide, to produce D-lactic acid (Table 5).
Continuous fermentation was carried out using the Candida utilis CuLpLDH strain as the microorganism, with use of a separation membrane. The continuous fermentation was performed for 370 hours under the same conditions as in Example 1 except that the period of preculture was 30 hours and that the pH was adjusted to 6.0 with 1 N calcium hydroxide, to produce D-lactic acid (Table 5).
Batch fermentation was carried out using the Candida tropicalis NBRC0199 strain. As the fermentation medium, YPDX medium having the same composition as shown in Table 2 except that the sugar concentrations were 1% glucose and 9% xylose was used. Ethanol was produced by 42 hours of batch fermentation in which other operating conditions and the like were the same as in Comparative Example 2 (Table 6).
Using the Candida tropicalis NBRC0199 strain, continuous fermentation was carried out without use of a separation membrane. As the fermentation medium, YPDX medium having the same composition as shown in Table 2 except that the sugar concentrations were 1% glucose and 9% xylose was used. Ethanol was produced by 400 hours of continuous fermentation in which other operating conditions and the like were the same as in Comparative Example 3 (Table 6).
Using the Candida tropicalis NBRC0199 strain, continuous fermentation was carried out with use of a separation membrane. As the fermentation medium, YPDX medium having the same composition as shown in Table 2 except that the sugar concentrations were 1% glucose and 9% xylose was used. Ethanol was produced by 420 hours of continuous fermentation in which other operating conditions and the like were the same as in Example 1 (Table 6).
Using the Candida tropicalis NBRC0199 strain, continuous fermentation was carried out with use of a ceramic separation membrane. As the fermentation medium, YPDX medium having the same composition as shown in Table 2 except that the sugar concentrations were 1% glucose and 9% xylose was used. The Candida tropicalis NBRC0199 strain was cultured in 2 mL of YPD medium in a test tube at 30° C. overnight with shaking (pre-pre-preculture). The obtained culture liquid was inoculated to 50 mL of YPD medium placed in a 500-mL baffled Erlenmeyer flask, and culture was performed overnight with shaking (pre-preculture). The pre-preculture liquid was inoculated to 1.5 L of YPDX medium placed in a membrane-separation-type continuous fermentation apparatus (the apparatus shown in FIG. 12 of WO 2012/086763), and culture was performed for 36 hours with stirring at 800 rpm by the stirrer attached to the culture vessel while the aeration rate and the temperature in the culture vessel were controlled (preculture). Immediately after completion of the preculture, continuous fermentation was started. While the transmembrane pressure difference during filtration was kept at not more than 500 kPa, ethanol was produced by 400 hours of continuous fermentation (Table 6).
Batch fermentation was carried out using the Candida tropicalis NBRC0199 strain. As the fermentation feedstock, a cellulose saccharification liquid prepared by the preparation method disclosed in Example 2 of WO 2010/067785 using a nanofiltration membrane was used. The composition of the fermentation medium was adjusted as shown in Table 7 using reagents as appropriate. Ethanol was produced by 72 hours of batch fermentation in which other operating conditions and the like were the same as in Comparative Example 3 (Table 8).
Using the Candida tropicalis NBRC0199 strain, continuous fermentation was carried out without use of a separation membrane. Similarly to Comparative Example 12, the culture medium described in Table 7 was used as the fermentation medium. Ethanol was produced by 400 hours of continuous fermentation in which other operating conditions and the like were the same as in Comparative Example 3 (Table 8).
Using the Candida tropicalis NBRC0199 strain, continuous fermentation was carried out with use of a separation membrane. Similarly to Comparative Example 12, the culture medium described in Table 7 was used as the fermentation medium. Ethanol was produced by 456 hours of continuous fermentation in which other operating conditions and the like were the same as in Example 1 (Table 8).
As shown by these results, chemical products could be produced with high yield from a mixed sugar of pentose and hexose.
The efficiencies of fermentation production of various chemical products using a fermentation feedstock containing pentose and hexose can be largely increased with our methods.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2012-005256 | Jan 2012 | JP | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2013/050437 | 1/11/2013 | WO | 00 |
