JP5710311B2 - Addition medium addition control method and cell culture apparatus using the method - Google Patents

Description

  The present invention relates to culturing of cells that produce substances that are main raw materials such as pharmaceuticals, for example, and relates to an addition medium addition control method applied at that time, and a cell culture apparatus using the method.

  Some pharmaceuticals including antibody pharmaceuticals contain a substance produced by cells as a main component. Since such substances are secreted and produced by animal cells, for example, they can be obtained by culturing animal cells and separating and purifying the target substance secreted in the culture solution. In the process of culturing cells, the cells proliferate by repeating division, but if the culture environment deteriorates, the number of cells that cause cell death increases, and if the culture environment continues to deteriorate, all cells eventually die. End up. Factors that deteriorate the culture environment include mechanical crushing by agitation, nutrient depletion, and accumulation of waste products such as ammonia and lactic acid secreted by cells.

  Cell culture method can be batch culture, continuous culture, fed-batch culture (semi-batch culture, fed batch culture), depending on the difference in the culture medium replacement method. being classified.

  Batch culture is a method in which a new medium is prepared each time and a new medium is not added until planting after planting the strain. This batch culture is characterized in that the quality of individual cultures varies, but the risk of contamination can be dispersed and reduced.

  Continuous culture is a culture method in which a medium is supplied to a culture system at a constant rate, and at the same time, the same amount of culture solution containing a product is extracted. This continuous culture is characterized in that the culture environment is always kept constant and productivity is stable. On the other hand, once the contamination occurs, the contamination is persistent.

  Fed-batch culture is a culture method in which the medium itself or specific components in the medium are added during the culture, and the culture solution containing the product is not extracted until the end of the culture. This fed-batch culture is performed for the purpose of, for example, optimizing growth by adjusting the cell density, and maintaining productivity by diluting harmful substances accumulated during the culture.

  In particular, in fed-batch culture and continuous culture, the feeding method of the supplemented medium, that is, depleting the nutrient source by devising the addition method, and the use of carbon source (glucose), which is one of the energy sources, By shifting from lactic acid metabolism to ATP (Adenosine Triphosphate) using complete oxidation by TCA cycle (tricarboxylic acid cycle), energy sources are used efficiently, accumulation of harmful metabolites is prevented and cell density is increased. A fed-batch method has been developed to achieve high cell concentrations.

JP 2007-244341 A JP-T 62-503146

  By the way, in any culture method as described above, due to the inaccuracy of the component composition ratio in the added medium, the concentration of one or more nutrient components deviates from the target concentration (control value) as the culture progresses, and the nutritional balance is lost. Disintegration, cell death such as apoptosis due to nutrient depletion, and inefficient metabolism at excessive nutrient concentration caused an increase in accumulation of harmful substances such as ammonia and lactic acid, resulting in lower production efficiency of the target product.

  Therefore, in fed-batch culture, various nutrients such as glucose and glutamine are set at low concentrations to effectively suppress the accumulation of harmful substances, and the minimum nutrient concentration that allows the cells to be cultured to grow is maintained. It is considered to culture while. However, in this case, the various nutrient components are only maintained at the minimum nutrient concentration at which the cells can grow, so that the cells cannot be efficiently densified, and the reduction in the production efficiency of the target product can be prevented. No further improvement can be expected.

  Therefore, in view of the above-described circumstances, the present invention increases the accuracy of the component composition ratio of the supplemented medium, and controls the concentration of each nutrient component in the culture solution to maintain the control value, thereby achieving an excellent yield of the target product. It is an object of the present invention to provide an addition medium addition control method capable of producing at a high rate, and a cell culture apparatus using the method.

  In order to solve the above problems, the present invention uses the ratio of each nutrient component consumed by cells as the component composition ratio of the supplemented medium, and the accuracy of the component composition ratio of the supplemented medium calculated using intracellular metabolism analysis And an additional medium that provides a control value for each nutrient component of the culture solution is supplied. When maintaining each nutrient component constant, by adding the above-designed addition medium, when changing the concentration of a specific nutrient component during the culture, supply the above-designed addition medium, and further increase the concentration. With respect to the nutritional component that changes the concentration, the concentration of each component in the culture solution is controlled by adding a difference amount of the control value. Specifically, it is as follows.

(1) The supply of the supplemented medium is controlled with the component composition ratio of the supplemented medium as the ratio of nutrient consumption of the cells.
(2) The supply and supply amount of the supplemented medium are determined according to the growth rate of the cells, and the nutrient components in the culture solution are maintained at a constant concentration.
(3) When changing the concentration of one or more specific nutrients during the cultivation, supply the supplemental medium designed as described above, and further change the amount of the difference from the control value for the nutrients whose concentration is to be changed. Is added to control the concentration of each component in the culture solution.
(4) The cell growth rate is measured by counting the number of cells.
(5) Measurement of cell growth rate is performed using glucose consumption, lactic acid secretion, and glucose and lactic acid addition as indices.
(6) The component composition ratio of the supplemented medium is determined by using GC-MS / MS (gas chromatography tandem mass spectrometer) in the intracellular metabolic flux analysis.
(7) The component composition ratio of the supplemented medium is determined by intracellular metabolic flux analysis using two or more isotope-labeled substrates.

  According to the addition medium addition control method and the cell culture apparatus using the method according to the present invention, inefficient metabolism due to overnutrition and cell death due to nutrient depletion can be prevented by preventing excess or deficiency of medium components. And high-quality cells can be efficiently produced with an excellent yield.

1 is a schematic configuration diagram of a fed-batch culture device as an embodiment of a cell culture device according to the present invention. It is the conceptual diagram which showed the structure of one Embodiment of the estimation system of an intracellular metabolic flux. It is the flowchart which showed the algorithm of the estimation method of the ratio of intracellular metabolic flux performed with the estimation system of intracellular metabolic flux. It is the figure which showed the change of the carbon atom skeleton of each substrate, metabolite, and product which concerns on an intracellular metabolic pathway model. It is the figure which showed the Example of the metabolic-reaction formula set up paying attention to the carbon position and number in the carbon skeleton of a substrate. It is a figure which shows the simultaneous equation based on the metabolic reaction formula derived | led-out based on the reaction path | route. It is explanatory drawing about the change of a carbon atom skeleton at the time of adding the isotope labeled tyrosine. 6 is a flowchart showing an embodiment of a method for determining an upper limit value of a confidence interval in obtaining a confidence interval of an estimated value of metabolic flux. It is the figure which showed the confidence interval of the value of the metabolic flux of each metabolic pathway. It is the figure which showed the growth curve of the number of living cells. It is a figure which shows the relationship between glutamine consumption and the time integration of the number of living cells. It is a figure which shows the relationship between the sum total of glucose consumption and lactic acid consumption, and the time integration of the number of living cells.

  An additive medium addition control method and a cell culture apparatus using the method according to an embodiment of the present invention will be described with reference to the drawings. In the description, the cells to be cultured are not limited to animal cells, plant cells, insect cells, bacteria, yeasts, fungi, algae, etc., but in the following, proteins such as antibodies and enzymes are produced. An example of an animal cell will be described.

<1. Cell culture device>
FIG. 1 is a schematic configuration diagram of a fed-batch culture device as an embodiment of a cell culture device according to the present invention.

  In addition, in the fed-batch culture apparatus 1 shown in FIG. 1, in order to facilitate the understanding of the configuration, the supply of one additional medium 3 as a nutrient component of the cultured cells is controlled from one additional medium supply tank 56 to the culture tank 10. An apparatus system will be described as an example. However, the apparatus system is not limited to this, and a plurality of supplemented medium supplies each corresponding to a plurality of supplemented mediums 3,... Having different nutrient components and / or component ratios. The present invention can also be applied to an apparatus system that controls supply from the tanks 56 to the culture tank 10. In addition, the specific configuration of the device is not limited to the illustrated device configuration, and does not preclude the realization of a similar device system using another device having a similar function.

  In FIG. 1, the fed-batch culture device 1 includes a culture tank 10 and a feed medium system 50. The feed medium system 50 includes a culture medium analyzer 52, a controller 54, an added medium supply tank 56, and a pump 58.

  In the culture tank 10, the culture solution 2 inoculated with the cultured cells, that is, the initial medium is stored at the start of the culture. The initial medium is appropriately selected according to the type of cells to be cultured, and examples include those containing amino acid components, carbon source components, minerals, vitamins, serum replacement components, and the like. Here, examples of the amino acid component include 20 kinds of amino acids constituting natural proteins, cystine, hydroxylysine, hydroxyproline, thyroxine, O-phosphoserine, desmosine, etc. Examples of the carbon source component include And sugars such as glucose, lactose and galactose, and alcohol.

  In the tank, a stirring blade 22 of a stirring mechanism 21 that stirs the culture solution 2 is provided. The shaft portion of the stirring blade 22 is connected to a drive unit 23 disposed outside the tank while maintaining the airtightness of the culture tank 10. The stirring blade 22 is rotatable in the tank in response to the operation of the drive unit 23. The operation of the drive unit 23 is controlled by the controller 54 of the feed medium system 50.

  Further, the culture tank 10 is stored in a sampling nozzle 24 for collecting the culture solution 2 stored in the tank, a detection device such as a temperature measurement electrode, a pH electrode, or a DO electrode (not shown), or in the tank. A temperature control device 26 such as a heater for adjusting the temperature (culture temperature) of the culture solution 2 by heating or cooling the culture solution 2 is attached.

  The temperature of the culture solution 2 is monitored by a temperature measuring electrode, the pH of the culture solution 2 is monitored by a pH electrode, and the dissolved oxygen concentration of the culture solution 2 is monitored by a DO electrode. These monitored values are supplied to the controller 54 and used by the controller 54 for culture temperature control, pH control, dissolved oxygen concentration control, etc. of the culture solution 2. For example, in the culture temperature control of the culture solution, the controller 54 controls the temperature control device 26 based on the temperature of the culture solution 2 supplied from the temperature measurement electrode, so that the temperature of the culture solution 2 is set to the target temperature. To control.

  The sampling nozzle 24 has a configuration including a three-way valve in this embodiment. The three-way valve has a first connection opening in the culture tank 10, a second connection opening through the sampling tube 34, and a culture fluid analyzer 52, and a third connection opening in the high-pressure steam generator. (PS generator, Pure Steam Generator) 25. The sampling nozzle 24 collects the culture solution 2 in the tank and sterilizes the sampling tube 34 by controlling the three-way valve. Specifically, the three-way valve allows communication between the inside of the culture tank 10 and the culture medium analyzer 52 via the sampling tube 34 so that the culture medium analyzer 52 can culture the culture tank 10 in the culture tank 10. A necessary amount of liquid 2 can be sucked out. The three-way valve communicates between the high-pressure steam generator 25 and the culture medium analyzer 52 via the sampling pipe 34, so that the steam generated by the high-pressure steam generator 25 is collected in the sampling pipe 34 and the culture medium analyzer. 52 can be distributed. Then, after the steam is distributed under a predetermined sterilization condition, the flow of the sampling nozzle 24, the sampling tube 34, and the culture solution analyzer 52 can be sterilized by stopping the distribution of the steam and drying each part. . As sterilization conditions, for example, steam distribution conditions such as 121 ° C. and 20 min are employed. On the other hand, during cell culture in the culture tank 10 except during sampling by the culture medium analyzer 52, the three-way valve passes between the high-pressure steam generator 25 and the culture medium analyzer 52 via the sampling tube 34. The inside of the culture tank 10 is kept in a shut-off state with respect to the outside of the tank. The supply of steam from the three-way valve of the sampling nozzle 24 or the high-pressure steam generator 25 is controlled by the culture solution analyzer 52 or the controller 54 in accordance with the analysis operation of the culture solution analyzer 52, that is, sampling.

  Further, in addition to the sampling pipe 34 connected via the sampling nozzle 24, the culture tank 10 includes a submerged aeration gas supply pipe 31, a gas phase gas supply pipe 32, and an added medium supply pipe 33. The tank 10 is provided so as to communicate from the outside of the tank to the inside of the tank while maintaining the airtightness of the tank 10. In addition, piping such as exhaust circulation, culture medium discharge, and pH adjustment chemical injection is connected to the culture tank 10 from outside the tank to the inside of the tank while maintaining the airtightness of the culture tank 10. Although not shown in FIG.

  The submerged aeration gas supply pipe 31 is communicated with a submerged aeration diffuser pipe (sintered sparger) 35 made of sintered metal as an aeration apparatus disposed on the bottom side of the culture tank 10. Yes. The submerged aeration diffuser pipe 35 discharges the aeration gas supplied from an aeration gas adjusting device (not shown) into the culture solution in the tank. The valve 36 is controlled by the control controller 54 that controls the dissolved oxygen concentration of the culture solution together with the aeration gas control device, and controls supply or stop of supply of the aeration gas and the supply amount at the time of supply.

  Control of the dissolved oxygen concentration of the culture solution 2 is performed by controlling the operation of the aeration gas regulator and the valve 36 based on the measurement output of the DO electrode, which the controller 54 monitors the dissolved oxygen concentration in the culture solution. This is performed by supplementing an appropriate amount of oxygen consumed during culture from the submerged aeration diffuser 35 through the gas supply pipe 31. When increasing the dissolved oxygen concentration in the culture solution, the controller 54 supplies the oxygen-containing gas to the submerged aeration tube 35 from the aeration gas regulator and also decreases the dissolved oxygen concentration in the culture solution. In this case, the dissolved oxygen concentration in the culture solution is controlled to a target value by supplying a nitrogen-containing gas from the aeration gas regulator to the submerged aeration diffuser 35.

  The gas-phase gas supply pipe 32 is provided in communication with the ceiling portion in the tank of the culture tank 10. The gas phase gas supply pipe 32 discharges the gas phase gas supplied from a gas phase gas adjusting device (not shown) to the gas phase portion in the tank. The valve 37 is controlled by a control controller 54 that controls the pH of the culture solution together with the gas phase gas regulator, and controls supply or stop of the gas phase gas and the supply amount at the time of supply.

  The pH of the culture solution is controlled by the control controller 54 based on the measurement output of the pH electrode that monitors the pH in the culture solution, and the carbon dioxide concentration in the gas phase gas from the gas phase gas regulator and the valve 37. The gas phase gas is supplied from the gas phase gas supply pipe 32 to the gas phase in the tank. In addition, the control controller 54 moves the pH of the culture solution to the acidic side due to the growth of the cells due to cell growth, and adjustment of the pH of the culture solution cannot be handled only by this gas-phase gas supply control. Adjusts the pH of the culture solution 2 by adding an appropriate amount of an alkaline solution such as sodium hydroxide via a pH adjusting agent injection pipe (not shown).

  The supplemental medium supply pipe 33 is connected to the supplemental medium supply tank 56 of the feed medium system 50 through a pump 58 in the middle of the pipe. The pump 58 is controlled to operate by a controller 54 that performs medium supply control, and controls the supply or supply stop of the added medium from the added medium supply tank 56 to the tank of the culture tank 10 and the supply amount at the time of supply.

  On the other hand, in the feed medium system 50, the culture medium analyzer 52 sucks and takes out a necessary amount of the culture medium in the culture tank 10 from the sampling nozzle 24 via the sampling tube 34, and extracts the collected culture liquid. Analyze the condition. The culture medium analyzer 52 measures, for example, the concentration of glucose, glutamine, lactic acid, ammonia, etc., the concentration of the product, the number of living cells, etc. in the cell culture collected from the culture tank 10 by this sampling. Sampling for analyzing the state of the culture solution is performed a plurality of times at predetermined intervals during cell culture. One sampling measurement step by the culture medium analyzer 52 requires an analysis time of, for example, at least 20 minutes or more, preferably 40 minutes or more, more preferably 60 minutes or more. The culture medium analyzer 52 is not limited to one device, and may be configured by combining a plurality of dedicated analyzers according to the measurement target. The culture medium analyzer 52 is controlled and operated by the controller 54, and data such as glucose concentration and the number of living cells measured by sampling are supplied to the controller 54. These data are used by the controller 54 for controlling the concentration of the culture fluid components.

  In the additional medium supply tank 56, an additional culture solution for supplying the culture tank 10, that is, the additional medium 3 is stored. The addition medium supply tank 56 is also provided with the stirring mechanism 21, the heater 26, and the like, similarly to the culture tank 10, and the culture management of the addition medium 3 is performed by the controller 54.

  The added medium 3 stored in the added medium supply tank 56 is supplied to the culture tank 10 through the added medium supply pipe 33 by the operation of the pump 58. The addition (supply) of the added medium 3 from the added medium supply tank 56 to the culture tank 10 is performed under the control of the culture medium component concentration performed by the controller 54, that is, the added medium addition control. At that time, the controller 54 controls the drive of the pump 58 while monitoring the liquid level measurement output of a liquid level measurement sensor (not shown) provided in the supplemental medium supply tank 56, so Control the amount added (amount supplied). Further, when the pump 58 is a metering pump capable of metering liquid, the controller 54 directly manages and drives the pump 58 without depending on the liquid level measurement sensor. It can also be set as the structure which controls the addition amount (supply amount).

  In the culture solution component concentration control, the controller 54 performs the sampling until the next sampling based on the analysis result of the state of the culture solution 2 supplied from the culture solution analyzer 52 every time the culture solution 2 in the culture tank 10 is sampled. The required amount of addition medium 3 is calculated and determined. Then, the controller 54 controls the driving of the pump 58 based on the determined addition amount, and the added medium 3 corresponding to the calculated addition amount is added to the culture solution in the culture tank 10 until the next sampling. 2 is added (liquid feeding).

  In addition, in the fed-batch culture apparatus 1 configured to control the supply of a plurality of supplemental media from the corresponding supplemental media supply tanks 56, for example, the supplemental culture medium supply tank 56, the pump 58, and the supplemental culture medium supply pipe 33 are provided. Can be accommodated by providing for each additional medium. For example, when it is necessary to change the concentration of one or more kinds of specific nutrients in the additive medium 3 during the culture using the additive medium 3, as shown in FIG. The added medium supply tank 56, the pump 58, and the added medium supply pipe 33 according to the added medium 4 in which the concentration of these specific nutrients is increased are connected to the added medium supply tank 56, the pump 58, and the added medium supply according to the added medium 3. The pipe 33 is provided separately. In addition, when it is necessary for the controller 54 to change the concentration of one or more kinds of specific nutrients in the added medium 3 during the culture, the controller 54 uses the added medium 3 as described above. After supplying the culture medium 10 from 56 to the culture tank 10, the additional medium 4 is added from the additional medium supply tank 56 according to the difference from the control value for these specific nutrient components, so that the culture solution 2 in the culture tank 10 is added. The concentration of each component will be controlled.

  The control controller 54 is configured by a computer device having an input / output device (not shown), and the culture medium analyzer 52, the pump 58, etc. in the feed medium system 50, the drive unit 23 provided in the culture tank 10, and the heater 26. , Detection devices and valves 36 and 37 of the gas supply pipes 31 and 32, respectively. The controller 54 controls the operations of these units and executes various controls such as the above-described culture temperature control, pH control, dissolved oxygen concentration control, and culture medium component concentration control.

<2. Control of concentration of culture fluid components>
Next, the culture solution component concentration control performed by the controller 54 will be described in detail. The concentration control of the culture medium component uses the additional medium 3 stored in the additional medium supply tank 56, and controls the addition of the additional medium 3 so that the nutrient concentration of the culture medium 2 in the culture tank 10 is constant. Is done by. The addition medium 3 is added, for example, at a predetermined addition amount and addition timing by dividing the feed medium into a plurality of times during the sampling measurement process. At that time, the additive medium 3 stored in the additive medium supply tank 56 is an additive medium whose component composition ratio is determined based on simulations and culture experiments.
Therefore, first, a method for determining the component composition ratio of the additional medium 3 stored in the additional medium supply tank 56 will be described.

<2-1. Intracellular metabolic flux estimation system>
The component composition ratio of the additional medium 3 stored in the additional medium supply tank 56 is determined to be the ratio of the amount of each nutrient component consumed by the cultured cells. The component composition ratio of the supplemented medium 3 is a metabolic flux (a flow of a substance in the intracellular metabolic network) related to each nutrient component in the intracellular metabolic flux analysis, and the ratio of the estimated metabolic flux between the nutrient components is calculated. Ask by. The component composition ratio of the additional medium 3 is determined based on the intracellular metabolic flux estimated using the intracellular metabolic flux estimation system.

FIG. 2 is a conceptual diagram showing a configuration of an embodiment of an estimation system for intracellular metabolic flux.
In the estimation system 100 for intracellular metabolic flux according to this embodiment, the estimation processing device 110 is connected to the intracellular metabolic pathway database 120 and the culture experiment analysis device 130. In addition, a confidence interval determination device 160 is connected to the estimation processing device 110, and a component composition ratio setting device 170 is attached to the confidence interval determination device 160. The estimation processing device 110, the confidence interval determination device 160, and the component composition ratio setting device 170 are configured by the same computer device having an input / output device (not shown).

  The estimation processing device 110 includes a random number section 112, a forward calculation processing section 114, an isotope ratio distribution processing section 116, and a QP (Quadratic Programming) subproblem processing section 118.

  The intracellular metabolic pathway database 120 stores intracellular metabolic pathways for each cell and for each predetermined substrate.

  The culture experiment analyzer 130 uses GC-MS (Gas Chromatography Mass Spectrometer), GC-MS / MS for intracellular metabolites obtained by the cell culture experiment 140 on a predetermined substrate using isotope labeling. Analysis using a gas chromatography tandem mass spectrometer is performed to measure intracellular metabolites.

  Intracellular metabolic flux estimation system 100 determines the component composition ratio of additive medium 3 stored in additive medium supply tank 56 prior to the concentration control of the culture fluid components, and performs simulations and estimation experiments by estimation processor 110. The intracellular metabolic flux is estimated from the intracellular metabolite measurement result based on the GC-MS and GC-MS / MS analysis results based on the culture experiment 140 acquired by the analyzer 130.

  The estimation processing device 110 may also have a configuration that serves as the control controller 54 of the feed medium system 50 that is configured by a computer device that also includes an input / output device.

  Next, a method for estimating the intracellular metabolic flux of the cell to be analyzed, performed by the intracellular metabolic flux estimation system 100, will be described with reference to the drawings.

  FIG. 3 is a flowchart showing an algorithm for estimation of intracellular metabolic flux executed by the estimation processing device 110 in the estimation system 100 for intracellular metabolic flux shown in FIG.

  In the simulation by the estimation processing device 110, first, the cultured cells to be analyzed and each nutrient component consumed by the cultured cells, that is, each substrate forming the added medium are estimated by an input / output device (not shown). Set to device 110.

  When the cell and substrate to be analyzed are set and input, the estimation processing device 110 searches the intracellular metabolic pathway database 120 corresponding to the cell and substrate to be analyzed from the intracellular metabolic pathway database 120 and sends it to the input / output device. indicate. The intracellular metabolic pathway model 150 used in this simulation is set by determining a desired intracellular metabolic pathway from the searched intracellular metabolic pathways using an input / output device. In this way, the estimation processing device 110 determines the intracellular metabolic pathway model 150 used in the current simulation and acquires it from the intracellular metabolic pathway database 120.

  For example, if a Chinese hamster ovary cell (Chinese Hamster Ovary (CHO) cell), which is often used in the manufacture of biopharmaceuticals such as antibody drugs, is input as the cell corresponding to the cell to be analyzed, the Provost study ( A. Provost et al., Metabolic Design of Macroscopic Models: Application to Chinese Hamster Ovary cells, Bioprocess Biosyst Eng, 2006, 29: 349-366) and KEGG (Kyoto Encyclopedia of Genes and Genomes) database Information on the metabolic pathway map that has been made public is retrieved, and the estimated processing apparatus 110 can use these retrieved metabolic pathways as the intracellular metabolic pathway model 150 of the CHO cell to be analyzed.

  In addition, the estimation processing apparatus 110 relates to the intracellular metabolic pathway model 150 used in this simulation determined in this way, and the intracellular metabolic flux estimation processing for the analysis target substrate of the analysis target cell. I do.

  Here, for example, it is assumed that an intracellular metabolic pathway model 150 as shown in FIG. 2 is determined as the intracellular metabolic pathway model used in this simulation for the substrate A of the analysis target cell.

  In FIG. 2, in the intracellular metabolic pathway model 150 determined for the substrate A used in the simulation of the cell to be analyzed, when the substrate A is given, the cell passes through the metabolic pathway R1 and passes through this substrate A. Is taken into the cell as metabolite B. Then, the cell generates a metabolite D from the metabolite B via the metabolic pathway R2, and similarly generates a metabolite C and a metabolite E from the metabolite B via the metabolic pathway R4. The metabolite E in this case corresponds to a by-product in the production of the metabolite C. Further, the cell generates a metabolite D through the metabolic pathway R5 for the metabolite C generated therein. In other words, in the intracellular metabolic pathway model 150, the cell generates the metabolite D from the metabolite B through the two metabolic pathways of the metabolic pathway R2 and the metabolic pathways R4 and R5. In addition, the cell generates the product F from the generated metabolite D via the metabolic pathway R6, and discharges the product F out of the cell.

  At this time, the surplus metabolite D is returned to the metabolite B through the metabolic pathway R3. On the other hand, when the metabolite B taken up via the metabolic pathway R1 and / or returned through the metabolic pathway R3 becomes redundant, the metabolite produced as a byproduct of the metabolite C by the metabolic pathway R4 Apart from E, it produces a metabolite E. In the illustrated example, assuming that the amount Bm of the metabolite B is the same amount, the amount Em of the metabolite E generated from the surplus metabolite B is a metabolite generated as a byproduct of the metabolite C by the metabolic pathway R4. This corresponds to twice the amount Em of the object E.

  In this way, the estimation processing device 110 relates to the cell to be analyzed and the substrate A. When the intracellular metabolic pathway model 150 used in this simulation is determined, the estimation processing device 110 is based on the acquired intracellular metabolic pathway model 150. Then, paying attention to each carbon position and number of carbon atom skeletons of substrate A, intracellular metabolites B to D, and product F, a reaction path (that is, metabolic path) is established (FIG. 3, step S010).

  FIG. 4 is a diagram showing changes in the carbon atom skeleton of each of the substrate A, metabolites B to E, and the product F related to the intracellular metabolic pathway model 150 shown in FIG.

  As shown in FIG. 4, according to the intracellular metabolic pathway model 150, each of the substrate A, the metabolites B and D, and the product F has three carbon atoms in its carbon atom skeleton, and the metabolite C has 2 The presumed processing apparatus 110 can understand that the metabolite E has a substance structure having one carbon atom in its carbon atom skeleton and one metabolite E in its carbon atom skeleton.

  FIG. 5 relates to the intracellular metabolism model 150 shown in FIG. 2 and shows an example of a metabolic reaction formula established by the estimation processing device 110 by paying attention to the carbon position and number in the carbon atom skeleton of the substrate A. FIG.

FIG. 5A is an example of a metabolic reaction equation derived from the intracellular metabolism model 150 focusing on one carbon position in the carbon atom skeleton of the substrate A. In the case shown in the figure, an example of a metabolic reaction equation obtained by paying attention to one carbon atom A ₂ or A ₃ at the 2-position or 3-position of the carbon atom skeleton composed of three carbon atoms of the substrate A, respectively.

In the illustrated example, the carbon atom A _{2 at} the 2-position of the carbon skeleton of the substrate A becomes the carbon B _{2 at} the 2-position of the carbon skeleton of the metabolite B via the metabolic pathway R1, and carbon B ₂ is 2-position and the carbon D ₂ of the carbon skeleton of metabolites D via a metabolic pathway R2, the 1-position of the carbon skeleton of metabolites C generated through a metabolic pathway R4 be a carbon C ₁ Understandable. It can also be understood that the carbon D _{2 at} the 2-position of the carbon skeleton of the metabolite D becomes the carbon B ₂ at the 2-position of the carbon skeleton of the metabolite B via the metabolic pathway R3.

On the other hand, the carbon atom A _{3 at} the 3-position of the carbon skeleton of the substrate A becomes the carbon B _{3 at} the 3-position of the carbon skeleton of the metabolite B via the metabolic pathway R1, and the carbon at the 3-position of the carbon skeleton of the metabolite B. It can be understood that B ₃ becomes the carbon D _{3 at} the 3-position of the carbon skeleton of the metabolite D through the metabolic pathway R2 and the carbon D ₂ at the 2-position of the carbon skeleton of the metabolite D through the metabolic pathway R5. It can also be understood that the carbon D _{3 at} the 3-position of the carbon skeleton of the metabolite D becomes the carbon B ₃ at the 3-position of the carbon skeleton of the metabolite B via the metabolic pathway R3.

FIG. 5B is a metabolic reaction equation derived from the intracellular metabolism model 150 focusing on two carbon positions in the carbon atom skeleton of the substrate A. The illustrated case, is a three positions 2 and 3 two examples of metabolic reaction formula obtained by focusing on the carbon atoms A ₂₃ carbon atoms backbone consisting of carbon atoms of the substrate A.

In the illustrated example, two carbon atoms A ₂₃ in 2-position and 3-position carbon atom skeleton of three carbon atoms of the substrate A, the metabolites B via a metabolic pathway R1 in 2-position and 3-position of the carbon backbone next two carbon atoms B _23, metabolite two carbon atoms B ₂₃ of the 2- and 3-position of the carbon skeleton of B, the two carbons of the 2- and 3-position of the carbon skeleton of metabolites D via a metabolic pathway R2 It can be understood that become atomic _{D 23.} Also, metabolite D of the two carbon atoms D ₂₃ of 2-position and 3-position of the carbon backbone may be a two carbon atoms B ₂₃ of the 2- and 3-position of the carbon skeleton of metabolites B via a metabolic pathway R3 Can also understand.

On the other hand, in the carbon atom skeleton composed of three carbon atoms of the metabolite B, one carbon atom B _{3 at} the 3rd position of the carbon skeleton and carbon C _{1 at} the 1st position of the carbon skeleton of the metabolite C are metabolized. it 2- and understood to be a two carbon atoms D ₂₃ of 3-position of the carbon skeleton of metabolites D via a route R5.

FIG. 5C is an example of a metabolic reaction equation derived from the intracellular metabolism model 150 focusing on three carbon positions in the carbon atom skeleton of the substrate A. In the case of illustration, it is an example of the metabolic reaction formula acquired by paying attention to the three carbon atoms A ₁₂₃ in the 1st to 3rd positions of the carbon atom skeleton composed of the three carbon atoms of the substrate A.

In the illustrated example, the three carbon atoms A ₁₂₃ in the 1st to 3rd positions of the carbon atom skeleton consisting of the 3 carbon atoms of the substrate A are converted to the 1st to 3rd positions of the carbon skeleton of the metabolite B via the metabolic pathway R1. It can be seen that there are three carbon atoms _B123 . Then, it can be understood that the three carbon atoms B ₁₂₃ from the 1st position to the 3rd position of the metabolite B become the three carbon atoms D ₁₂₃ from the 1st position to the 3rd position of the metabolite D through the metabolic pathway R2. Moreover, metabolites three carbon atoms D ₁₂₃ from position 1 at the 3-position of D becomes three carbon atoms F ₁₂₃ of position 3 from position 1 of the product F via a metabolic pathway R6, metabolism through the metabolic pathway R3 It can also be understood that the three carbon atoms B ₁₂₃ from the 1st position to the 3rd position of the product B are obtained.

On the other hand, in the carbon atom skeleton composed of three carbon atoms of the metabolite B, the two carbon atoms B _{23 at} the 2nd and 3rd positions and the carbon C _{1 at} the 1st position of the carbon skeleton of the metabolite C are metabolized. It can be understood that the three carbon atoms D ₁₂₃ in the 1st to 3rd positions of the carbon skeleton of the metabolite D are obtained via the route R5.

  Therefore, if the intracellular metabolic pathway model 150 for the substrate A to be analyzed is determined in this way, the estimation processing apparatus 110 performs the substrate A, metabolites B to D, and products for the cells to be analyzed. Focusing on the carbon atom position and the number of carbon atoms in the carbon skeleton of each F, a reaction path as shown in FIG. 5 can be established (FIG. 3, step S010).

Then, when the estimation processing apparatus 110 establishes a reaction path as shown in FIG. 5, the number of carbon atoms at the 2nd and 3rd positions in the carbon atom skeleton of the substrate A is calculated based on this reaction path as M (A ₂ ), M (A ₃ ), the number of carbon atoms at the 2nd and 3rd positions in the carbon atom skeleton of the metabolite B are M (B ₂ ), M (B ₃ ), and the number of carbon atoms in the carbon atom skeleton of the metabolite C. The number of carbon atoms at the 1st position is M (C ₁ ), and the number of carbon atoms at the 2nd and 3rd positions in the carbon atom skeleton of the metabolite D is M (D ₂ ), M (D ₃ ),. A simultaneous equation is established with the flux values r _{1 to} r ₆ of the reaction paths R1 to R6 as variables (FIG. 3, step S020).

  FIG. 6 is a diagram showing simultaneous equations derived based on the metabolic reaction equations related to the analysis target substrate A of the analysis target cells derived based on the reaction path.

  In establishing the simultaneous equations shown in FIG. 6, the estimation processing device 110 is configured so that the surplus of metabolites B and D does not accumulate in the cell, and the number of carbon supplies from other metabolites and other values for each metabolite. It is assumed that the number of carbon supplies to metabolites is the same.

  Therefore, in the intracellular metabolic pathway model 150 determined corresponding to the cellular substrate A, the metabolite D and the metabolite B are separated from the metabolite E generated as a byproduct of the metabolite C by the metabolic pathway R4. It is important to reduce the amount of metabolite E produced when the amount of surplus becomes excessive in order to prevent inefficient metabolism due to overnutrition and to efficiently produce high-quality cells in an excellent yield.

Thus, in the change in the carbon atom skeleton of each of the substrate A, the metabolites B to E, and the product F shown in FIG. 5, the carbon C _{1 at} the 1-position of the carbon skeleton of the metabolite C is the metabolite D. Along to become carbon D ₃ of the 3-position of the carbon skeleton, 2-position carbon of the carbon skeleton of the 3-position of the carbon B ₃ metabolites D carbon skeleton of metabolites metabolite was based the generation of C B Being D ₂ results in excellent yield for product F.

  FIG. 6 (a) is a simultaneous equation based on the metabolic reaction equation shown in FIG. 5 (a) when attention is paid to one carbon position in the carbon atom skeleton of the substrate A. FIG.

Specifically, with respect to metabolite C, it is related to carbon atom C ₁ at the 1-position of its carbon atom skeleton, based on the equilibrium of input and output through metabolic pathways R4 and R5,
r ₄ · M (B ₂ ) = r ₅ · M (C ₁ )
Because
(−r ₅ ) · M (C ₁ ) + r ₄ · M (B ₂ ) + 0 · M (D ₂ ) + 0 · M (B ₃ ) + 0 · M (D ₃ ) = 0 · M (A ₂ ) + 0 · M (A ₃ )
Is obtained.

Similarly, with respect to the metabolite B, it is related to the carbon atom B ₂ at the 2-position of the carbon skeleton, and based on the input / output equilibrium via the metabolic pathways R1, R2, R3, R4,
r ₁ · M (A ₂ ) + r ₃ · M (D ₂ ) = r ₂ · M (B ₂ ) + r ₄ · M (B ₂ )
Because
0 · M (C ₁ ) + (− r ₂ −r ₄ ) · M (B ₂ ) + r ₃ · M (D ₂ ) + 0 · M (B ₃ ) + 0 · M (D ₃ ) = − r ₁ · M (A ₂ ) + 0 · M (A ₃ )
Is obtained.

Similarly, with respect to metabolite B, it is related to carbon atom B ₃ at the 3-position of its carbon skeleton, based on the input / output equilibrium via metabolic pathways R1, R2, R3, R5,
r ₁ · M (A ₃ ) + r ₃ · M (D ₃ ) = r ₂ · M (B ₃ ) + r ₅ · M (B ₃ )
Because
0 · M (C ₁ ) + 0 · M (B ₂ ) + 0 · M (D ₂ ) + (− r ₂ −r ₅ ) · M (B ₃ ) + r ₃ · M (D ₃ ) = 0 · M (A ₂ ) + (− r ₁ ) · M (A ₃ )
Is obtained.

Similarly, for metabolite D, it relates to carbon atoms D ₂ of the 2-position of the carbon atom backbone, based on the equilibrium of the input and unloading through the metabolic pathway R2, R3, R5, respectively,
r ₂ · M (B ₂ ) + r ₅ · M (B ₃ ) = r ₃ · M (D ₂ )
Because
0 · M (C ₁ ) + r ₂ · M (B ₂ ) + (− r ₃ ) · M (D ₂ ) + r ₅ · M (B ₃ ) + 0 · M (D ₃ ) = 0 · M (A ₂ ) + 0 · M (A ₃ )
Is obtained.

Similarly, for metabolite D, it relates to a carbon atom D ₃ at the 3-position of the carbon atom backbone, based on the equilibrium of the input and unloading through the metabolic pathway R2, R3, R5, respectively,
r ₂ · M (B ₃ ) + r ₅ · M (C ₁ ) = r ₃ · M (D ₃ )
Because
r ₅ · M (C ₁ ) + 0 · M (B ₂ ) + 0 · M (D ₂ ) + r ₂ · M (B ₃ ) + (-r ₃ ) · M (D ₃ ) = 0 · M (A ₂ ) + 0 · M (A ₃ )
Is obtained.
When these balance equations are represented as simultaneous equations by determinants, FIG. 6A is obtained.

  FIG. 6B is a simultaneous equation based on the metabolic reaction equation shown in FIG. 5B when focusing on two carbon positions in the carbon atom skeleton of the substrate A.

Specifically, with respect to metabolite D, it relates to carbon atom D ₂₃ at the 2nd and 3rd positions of its carbon atom skeleton, based on the input / output equilibrium through metabolic pathways R2, R3, R5,
r ₂ · M (B ₂₃ ) + r ₅ · M (B ₃ × C ₁ ) = r ₃ · M (D ₂₃ )
Because
(−r ₃ ) · M (D ₂₃ ) + r ₂ · M (B ₂₃ ) = (− r ₅ ) · M (B ₃ × C ₁ ) + 0 · M (A ₂₃ )
become. In the above formula, M (B ₃ × C ₁ ) is ₃ of the carbon atom skeleton of metabolite B, which becomes carbon atom D ₂₃ at the 2nd and 3rd positions of the carbon atom skeleton of metabolite D via metabolic pathway R5. Represents the number of each of the carbon atom B _{3 at} the position and the carbon atom C _{1 at} the 1-position of the carbon atom skeleton of the metabolite C.

Similarly, with respect to the metabolite B, it is related to the carbon atom B ₂₃ at the 2nd and 3rd positions of its carbon skeleton, and based on the inflow / outflow equilibrium through the metabolic pathways R1, R2, R3,
r ₁ · M (A ₂₃ ) + r ₃ · M (D ₂₃ ) = r ₂ · M (B ₂₃ )
Because
r ₃ · M (D ₂₃ ) + (− r ₂ ) · M (B ₂₃ ) = 0 · M (B ₃ × C ₁ ) + (− r ₁ ) · M (A ₂₃ )
become.
When these equations are expressed as simultaneous equations by determinants, FIG. 6B is obtained.

  FIG. 6C is a simultaneous equation based on the metabolic reaction equation shown in FIG. 5C when attention is paid to three carbon positions in the carbon atom skeleton of the substrate A.

Specifically, with respect to metabolite B, it is related to carbon atoms B ₁₂₃ at the 1st, 2nd and 3rd positions of the carbon atom skeleton, based on the input / output equilibrium via metabolic pathways R1, R2 and R3,
r ₂ · M (B ₁₂₃ ) = r ₃ · M (D ₁₂₃ ) + r ₁ · M (A ₁₂₃ )
Because
0 · M (F ₁₂₃ ) + r ₃ · M (D ₁₂₃ ) + (− r ₂ ) · M (B ₁₂₃ ) = 0 · M (B ₂₃ × C ₁ ) + (− r ₁ ) · M (A ₁₂₃ )
become. In the above formula, M (B ₂₃ × C ₁ ) is the carbon atom skeleton of metabolite B, which becomes carbon atom D ₁₂₃ at the 1,2- and 3-positions of the carbon atom skeleton of metabolite D via metabolic pathway R5. 2 and 3 position carbon atoms B ₂₃ , and the number of carbon atoms C ₁ in the 1st position of the carbon atom skeleton of metabolite C, respectively.

Regarding metabolite D, it is related to carbon atom D ₁₂₃ at the 1st, 2nd and 3rd positions of its carbon skeleton, based on the input / output equilibrium via metabolic pathways R2, R3, R5 and R6,
r ₅ · M (B ₂₃ × C ₁ ) + r ₂ · M (B ₁₂₃ ) = r ₃ · M (D ₁₂₃ ) + r ₆ · M (D ₁₂₃ )
Because
0 · M (F ₁₂₃ ) + (− r ₃ −r ₆ ) · M (D ₁₂₃ ) + r ₂ · M (B ₁₂₃ ) = 0 · M (B ₂₃ × C ₁ ) + 0 · M (A ₁₂₃ )
become.

Regarding metabolite F, it is related to carbon atom F ₁₂₃ at the 1st, 2nd and 3rd positions of its carbon skeleton, based on the equilibrium of entry / exit via metabolic pathway R6,
M (F ₁₂₃ ) = r ₆ · M (D ₁₂₃ )
Because
1 · M (F ₁₂₃ ) + r ₆ · M (D ₁₂₃ ) + 0 · M (B ₁₂₃ ) = 0 · M (B ₂₃ × C ₁ ) + 1 · M (A ₁₂₃ )
become.

In this way, the estimation processing apparatus 110 sets the isotope carbon reaction pathways (metabolic pathways) R1 to R6 by paying attention to each carbon position and number in each metabolite A to D in the cell to be analyzed. (FIG. 3, step S010), based on each of the obtained reaction paths R1 to R6, the quantity of each of the substrate A, the metabolites B to E, and the product F is determined as M (A), M (B). -M (E) and M (F), and the simultaneous equations as shown in FIG. 6 are established using the flux values r _{1 to} r ₆ of the reaction paths R1 to R6 as variables (FIG. 3, step S020).

In addition, the estimation processing apparatus 110 sets simultaneous equations based on metabolic reaction equations derived based on the reaction pathways R1 to R6 of the determined intracellular metabolic pathway model 150 in the forward calculation processing unit 114. . In addition, since the number A of the substrate A M (A ₂ ), M (A ₃ ), M (A ₂₃ ), and M (A ₁₂₃ ) can be known, the forward calculation processing unit 114 performs metabolism of the simultaneous equations. first value _r 1 ~r ₆ fluxes assigns random values of metabolic fluxes _r 1 ~r ₆ supplied from the random number portion 112 (a random number), metabolites B to E, and the product F respective numbers amount M (B ₂ ), M (B ₃ ), M (B ₂₃ ), M (B ₁₂₃ ), M (C ₁ ), M (D ₂ ), M (D ₃ ), M (D ₂₃ ), M (D D ₁₂₃ ) and M (F ₁₂₃ ) are obtained by calculation.

The estimation processing device 110 then counts the quantity M (B ₂ ), M (B ₃ ), M (B ₂₃ ), M (of the intracellular metabolites B to E and the product F obtained by this calculation simulation. B ₁₂₃ ), M (C ₁ ), M (D ₂ ), M (D ₃ ), M (D ₂₃ ), M (D ₁₂₃ ), M (F ₁₂₃ ) based on the intracellular metabolite B by simulation to E, isotope carbon number ratio of the product F respectively _k B _{to k} D, calculates the _{k F} (FIG. 3, step S030).

In this case, the isotope carbon number ratios k _{B to} k _D and k _F are expressed as metabolites B in the cell via the metabolic pathway R1 in the intracellular metabolic pathway model 150 as shown in FIG. 2, for example. 2-position of the captured substrate a, 3-position carbon atom of _a 2, _{a 3} is, by the value _r 1 ~r ₆ of this assumed metabolic flux, intracellular metabolite B to E, and how the product F This is a parameter value indicating whether the distribution is in a proper ratio.

In addition, in this estimation of intracellular metabolic flux, can the metabolic flux values r _{1 to} r ₆ obtained by the simulation by the estimation processing device 110 be determined as the metabolic flux values of the intracellular metabolic pathway model 150 for the substrate A? In order to confirm whether or not, the observation parameter (the ratio of the number of isotope carbon of each metabolite in the cell) by the culture experiment is required.

Therefore, the estimation processing device 110 performs the culturing while calculating the isotope carbon number ratios k _{B to} k _D and k _{F in} the metabolite based on the calculation simulation using the random number of the random number unit 112 by the forward calculation processing unit 114. Products in the intracellular metabolites B to D measured from the experimental analyzer 130 by performing a culture experiment 140 with an isotope-labeled substrate A in which a carbon atom at a predetermined skeleton position of the carbon atom skeleton of the substrate A is isotope-labeled. Based on the number M of each in F, the isotope carbon number ratios K _{B to} K _D and K _F in the intracellular metabolites B to D and the product F based on the results of the culture experiment are obtained.

Here, the measurement of the observation parameter by the culture experiment will be described.
In the culture experiment by the culture experiment analyzer 130, an isotope-labeled nutrient substrate (for example, glucose) is taken into a cultured cell, and intracellular metabolites (for example, glucose, glycerol) are obtained using GC-MS and GC-MS / MS. , Acetate, citrate, pyruvic acid, etc.) and the value of the observation parameter is obtained.

  In the simulation described above, the carbon position of the isotope label of each metabolite is important for analysis, but in normal culture experiments, NMR (Nuclear Magnetic Resonance: nuclear magnetic resonance apparatus) is used for measurement of intracellular metabolites. ) And GC-MS (Gas Chromatography Mass Spectrometer), the position of the isotope-labeled carbon in the carbon atom skeleton of each intracellular metabolite becomes unclear, and the estimation accuracy of the metabolic flux value is poor. doing. A specific example is shown in FIG.

FIG. 7 is an explanatory diagram showing changes in the carbon atom skeleton when isotopically labeled tyrosine is added as a nutrient substrate to cultured cells. FIG. 7 (a) shows the change of the carbon atom skeleton in intracellular metabolism, and FIG. 7 (b) shows the state of the carbon atom skeleton at the time of analysis by GC-MS and GC-MS / MS, respectively. . In the figure, filled black circle symbol (● symbol) of substrate A indicates isotope-labeled carbon 13 ( ¹³ C), and unfilled white circle symbol (◯ symbol) indicates unlabeled carbon 12. denote the ⁽¹² C).

In the metabolic pathway shown in FIG. 7 (a), tyrosine having four carbon atoms in its carbon skeleton is divided into two carbon atoms through the respective metabolic pathways. In the carbon atom skeleton and acetyl coenzyme A are produced. In this case, assuming that tyrosine is isotopically labeled with carbon 13 ( ¹³ C) only at the 2-position of the carbon atom skeleton, the labeled carbon ( ¹³ C) is located at the 2-position of the carbon atom skeleton of glycine. Will move.

Here, for example, when tyrosine is measured by GC-MS, it is known how many isotope carbon atoms ( ¹³ C) are in the carbon atom skeleton, but the position in the carbon atom skeleton is unknown, Understanding the flow of isotope labeling in metabolic pathways is unclear. On the other hand, when tyrosine is measured by GC-MS / MS, energy is applied at the second excitation voltage, and the carbon atom skeleton of tyrosine is decomposed into two. In GC-MS / MS, by measuring this decomposition product, it is possible to roughly know at which carbon position in the carbon atom skeleton of tyrosine there was an isotope. That is, as shown in FIG. 7B, in GC-MS, tyrosine in which the 1st or 2nd position of the carbon atom skeleton is isotopically labeled with carbon 13 ( ¹³ C) and the 3rd position or Although it is impossible to distinguish and grasp tyrosine labeled with carbon 13 ( ¹³ C) at the 4-position, it is possible with GC-MS / MS. As a result, glycine, which is a metabolite of tyrosine, can be distinguished from acetyl coenzyme A, and each metabolic pathway can also be clearly grasped. By including GC-MS / MS in addition to GC-MS, culture experiment analyzer 130 increases the positional information of isotope carbon in intracellular metabolism and improves the estimation accuracy of metabolic flux based on experiments. Can do.

Therefore, in the culture experiment by the culture experiment analyzer 130, when the nutrient substrate A is given to the cell to be analyzed shown in FIG. 2, as shown by the filled black circle symbol (● symbol) in FIG. As shown in FIG. 4, if the isotope-labeled substrate A ₂ = [2 ⁻¹³ C] A in which only the 2-position of the carbon skeleton is labeled with labeled carbon 13 ( ¹³ C), FIG. The metabolite B ₂ = [2 − ¹³ C] B generated from the metabolite A ₂ through the metabolic pathway R 1, and the metabolite D generated from the metabolite B ₂ through the metabolic pathway R ₂ as shown in FIG. ₂ = [2- ¹³ C] D and metabolite C ₁ generated via metabolic pathway R4 = [1- ¹³ C] C and metabolite D ₃ generated from this metabolite C ₁ through metabolic pathway R 5 = [3- ¹³ C] D and metabolite B ₃ produced from this D ₃ through metabolic pathway R ₃ = [3- ¹³ C] B can be grasped. Similarly, if the isotope-labeled substrate A ₃ = [3- ¹³ C] A in which only the 3-position of the carbon skeleton is labeled with the labeled carbon 13 ( ¹³ C), the metabolite A ₃ Metabolite B ₃ = [3- ¹³ C] B generated from metabolic pathway R1 and metabolite D ₃ = [3- ¹³ C] D generated from this metabolite B ₃ via metabolic pathway R2 and metabolic pathway It becomes possible to grasp the metabolite D ₂ = [2 − ¹³ C] D generated through R5. Further, assuming that the second and third positions of the carbon skeleton are labeled with different isotopes labeled substrates A ₂₃ , metabolites B ₂₃ , B ₁₂₃ , as shown in FIGS. 5 (b) and 5 (c), D ₂₃ and D ₁₂₃ can be grasped.

Therefore, as described above, by performing a culture experiment using the isotope-labeled substrate A for the cell to be analyzed, the culture experiment analyzer 130 also uses the values of observation parameters obtained by GC-MS and GC-MS / MS as a basis. In addition, based on the number M of each of the intracellular metabolites B to D and the product F, the ratio of the number of isotope carbons in the intracellular metabolites B to D and the product F based on the results of the culture experiment K _{A to} K _D and K _F can be obtained by calculation.

Returning to the flowchart shown in FIG. 3, based on the algorithm of the method for estimating the intracellular metabolic flux, the estimation processing device 110 uses the forward calculation processing unit 114 to simulate the intracellular metabolites B to D and the product F by the simulation. When the carbon number ratios k _{B to} k _D and k _F are obtained (step S030), the isotopes in the intracellular metabolites B to D and the product F based on the results of the above-described culture experiment on the cell to be analyzed The carbon number ratios K _{B to} K _D and K _F are obtained, and the isotope carbon number ratios k _{B to} k _D and k _F in the metabolite by simulation, the metabolites _{B to D} by experiment, and the product F, respectively. Comparison with the isotope carbon number ratios K _{B to} K _D and K _F is performed. At that time, in the estimation processing device 110, the isotope ratio distribution processing unit 116 has the isotope carbon number ratio values k _{B to} k _D and k _F in the metabolite by simulation and the isotope carbon number in the metabolite by experiment. It is determined whether or not the mean square error between the ratio values K _{B to} K _D and K _F is larger than a predetermined range, and whether or not there is a statistically significant difference between the two. Is discriminated (step S040). Note that the number of intracellular metabolites to be compared for the isotope carbon number ratio may be equal to or greater than the number of metabolic fluxes in which the metabolic pathways R1 to R6 are independent.

Then, as a result of the discrimination by the isotope ratio distribution processing unit 116, the values of the isotope carbon number ratios k _{B to} k _D , k _F in the metabolite by simulation and the values of the corresponding isotope carbon number ratio in the metabolite When the mean square error between K _{B to} K _D and K _F becomes larger than a predetermined range and a statistically significant difference occurs between them, the estimation processing device 110 uses the QP subproblem processing unit. 118 so that the mean square error between the two (k _{B to} k _D , k _F and K _{B to} K _D , K _F ) falls within a predetermined range and is minimized. Among the metabolic flux values r1 to r6 of the metabolic pathways R1 to R6 obtained by the simulation, the metabolic flux values r1 to r6 of the metabolic pathways R1 to R6 related to the error are corrected (step S050).

In addition, in the estimation processing apparatus 110, the forward calculation processing unit 114 obtains the metabolic flux values r _{1 to} r ₆ of the metabolic routes R1 to R6 corrected by the QP subproblem processing unit 118 in step S020. The number of carbon atoms of the metabolites B to E and the product F M (B ₂ ), M (B ₃ ), M (B ₂₃ ), M (B ₁₂₃ ) , M (C ₁ ), M (D ₂ ), M (D ₃ ), M (D ₂₃ ), M (D ₁₂₃ ), M (F ₁₂₃ ) are obtained by calculation, and intracellular metabolites B to E by simulation are calculated. , The isotope carbon number ratios k _{B to} k _D and k _F of the product _F are calculated (step S060).

Then, in the estimation processing device 110, the isotope carbon number ratios k _{B to} k _D and k _F in the metabolite by this simulation, and the isotopes in the metabolite B to D and the product F by the experiment acquired earlier, respectively. The body carbon number ratios K _{B to} K _D and K _F are compared. At that time, in the estimation processing device 110, the isotope ratio distribution processing unit 116 has the isotope carbon number ratio values k _{B to} k _D and k _F in the metabolite by simulation and the isotope carbon number in the metabolite by experiment. It is determined whether or not the mean square error between the ratio values K _{B to} K _D and K _F is larger than a predetermined range, and whether or not there is a statistically significant difference between the two. Is discriminated (step S070).

On the other hand, in the discrimination processing by the isotope ratio distribution processing unit 116 in steps S040 and S070, the values of the isotope carbon number ratio k _{B to} k _D and k _F in the metabolite by simulation and the isotope in the metabolite by experiment When the mean square error between the carbon number ratio values K _{B to} K _D and K _F is within a predetermined range and no statistically significant difference occurs between them, the estimation processing device 110 Determines the metabolic flux values r _{1 to} r ₆ to be compared and discriminated at that time as estimated metabolic flux values r _{1 to} r ₆ for the metabolic pathways (reaction pathways) R _{1 to} R ₆ (step S 080).

Normally, the estimation processing device 110 is set by repeated calculation for each correction by the forward calculation processing unit 114 by correcting the metabolic flux values r _{1 to} r ₆ in the QP partial problem processing unit 118 a few times. The estimated values r _{1 to} r ₆ of the metabolic flux of the metabolic pathway model 150 can be estimated, but the number of isotope carbon number ratios k _{A to} k _E in the metabolite by simulation, If there is a statistically significant difference between the isotope carbon number ratio values K _{A to} K _E in the metabolite by experiment, whether the preset metabolic pathway model 100 is incorrect or the experimental data It is determined that the values K _{A to} K _E are not properly measured. In such a case, the estimation processing device 110 notifies that effect from an input / output device (not shown), returns to step S010 again, resets the metabolic pathway (reaction pathway), and intracellular metabolic flux. Will be estimated again.

  In the estimation system 100 for intracellular metabolic flux according to the present embodiment, the estimation processing device 110 uses the estimation method for the intracellular metabolic flux shown in the flowchart in FIG. Based on this, it is configured to estimate the intracellular metabolic flux.

  Further, in the estimation system 100 for intracellular metabolic flux according to the present embodiment, as shown in FIG. 2, a confidence interval determination device 160 is attached to the estimation processing device 110.

When the estimation processing device 110 determines the estimated value of the metabolic flux of each metabolic path (each reaction path) of each analysis target substrate of the analysis target cell based on the discrimination result by the isotope ratio distribution processing unit 116, The confidence interval determination device 160 obtains a distribution of estimated values of the determined metabolic pathways (reaction pathways) and confidence intervals. For example, the confidence interval determination device 160 uses the estimated values r _{1 to} r _{6 of} the metabolic flux determined by the estimation processing device 110 for each of the metabolic pathways (reaction pathways) R1 to R6 related to the substrate A, as shown in FIG. When the isotope carbon number ratios K _{B to} K _D and K _F in the intracellular metabolites B to D and the product F by the experiment shown in the experiment are obtained from the estimation processing device 110, the confidence interval is obtained. , Using the Antoniewicz method (Maciek R. Antoniewicz et al., Determination of confidence intervals of metabolic fluxes estimated from stable isotope measurements, Metabolic Engineering, 8 (2006) 324-337) It has become.

FIG. 8 is a flowchart showing an embodiment of a method for determining the upper limit value of the confidence interval when the confidence interval determination device 160 obtains the confidence interval of the estimated values r _{1 to} r ₆ of the metabolic flux.

In FIG. 8, the confidence interval determination device 160 calculates and determines the estimated values r _{1 to} r ₆ of the metabolic fluxes of the metabolic pathways R1 to R6 related to the substrate A of the cell to be analyzed, for example. (Step S110), obtaining the estimated values r _{1 to} r ₆ and the data of the isotope carbon number ratios K _{B to} K _D and K _F in the metabolite by the experiment used for the estimation, Focusing on the estimated value r ₁ determined for the metabolic flux of one of the metabolic pathways, for example, the metabolic pathway R1, a predetermined amount ΔN is added to the estimated value r ₁ to update and change the value r ₁ ′ (step S120). . Accordingly, every time the update change process shown in step S120 is executed, the confidence interval determination device 160 increases the metabolic flux update value r ₁ ′ by a predetermined amount ΔN with respect to the initial estimated value r ₁ . To go.

Furthermore, since the substrate A is known, the confidence interval determination device 160 has only the value r ₁ among the metabolic flux values r _{1 to} r ₆ of the metabolic pathways R _{1 to} R 6 in the equation shown in FIG. Is changed to the updated value r ₁ ′, and the metabolic flux values r _{2 to} r ₆ of the remaining metabolic pathways R _{2 to} R ₆ are left as the estimated values r _{2 to} r ₆ determined by the estimation processing device 110, and metabolism is performed. Numbers M (B ₂ ), M (B ₃ ), M (B ₂₃ ), M (B ₁₂₃ ), M (C ₁ ), M (D ₂ ), M for each of the products B to E and the product F (D ₃ ), M (D ₂₃ ), M (D ₁₂₃ ), M (F ₁₂₃ ) are obtained by calculation, and intracellular metabolites B to E and product F according to the update change of this estimated value r ₁ are obtained. isotope carbon number ratio of _k B _{to k} D, calculates the _{k F} (step S130).

Then, the confidence interval determination device 160 uses the isotope carbon number ratio values k _{B to} k _D , k _F in the metabolite calculated according to the updated value r ₁ ′ and the isotope in the metabolite by the experiment. The body carbon number ratio values K _{B to} K _D and K _F are compared. Then, the confidence interval determination device 160 determines the value k _{A to} k _E of the isotope carbon number ratio in the metabolite according to the updated value r ₁ ′ and the value K _A of the isotope carbon number ratio in the metabolite by the experiment. mean square error between ~K _E, it is determined whether or not larger than a predetermined range set in advance, determines whether statistically significant difference between the two has occurred (step S140).

By this determination, the confidence interval determination device 160 determines that the estimated flux value r ₁ increased in step S120 when the mean square error is smaller than the predetermined range and there is no statistically significant difference between the two. 'Is again increased by a predetermined amount ΔN to update the update value r ₁ ' (step S120), and the processes of steps S130 and S140 described above are repeated. Meanwhile, confidence interval determination unit 160 is smaller than the mean square error is a predetermined range, if the statistically significant difference between the two occurs, the estimated value r ₁ of the metabolic flux of metabolic pathways R1 reaches the upper limit value The previous update value r ₁ ′ obtained by subtracting the predetermined amount ΔN from the current update value r ₁ ′ is determined as the upper limit value of the estimated value r ₁ of the metabolic flux of the metabolic pathway R1 ( Step S150).

The confidence interval determination device 160 also determines the lower limit value of the estimated value r ₁ of the metabolic flux of the metabolic pathway R1 in the same manner as the determination of the upper limit value described above. In this case, in step S120, and updating changed by subtracting a predetermined amount ΔN values r ₁ and _'the estimated value r ₁ of the metabolic flux in step S150, the metabolic flux of metabolic pathways R1 estimate r ₁ When determining the lower limit value, the previous update value r ₁ ′ obtained by adding the predetermined amount ΔN to the current update value r ₁ ′ is determined as the lower limit value of the estimated value r ₁ of the metabolic flux of the metabolic pathway R1. Is different from the case of determining the upper limit value.

Thus, the confidence interval determination device 160 sets the upper limit value and the lower limit value as shown in FIG. 8 for the estimated values r _{1 to} r ₆ of the metabolic fluxes of all metabolic pathways R1 to R6, Distributions and confidence intervals of the estimated values r _{1 to} r ₆ of the metabolic fluxes of the metabolic pathways R1 to R6 are obtained and displayed on the input / output device.

FIG. 9 is a diagram showing confidence intervals for the metabolic flux values of the metabolic pathways thus determined.
In the figure, the bar graph portion 9A is a confidence interval of the estimated values of metabolic fluxes of the metabolic pathways R1 to R6 obtained from the results of intracellular metabolic flux analysis using only GC-MS without using GC-MS / MS. . On the other hand, the line segment portion 9B is a confidence interval of the estimated value of each metabolic flux R1 to R6 obtained from the intracellular metabolic flux analysis result using GC-MS / MS.

That is, an estimation processing device 110 that executes an estimation algorithm of intracellular metabolic flux as shown in FIG. 3 and a confidence interval determination method as shown in FIG. 8 (the upper limit value and lower limit value of the estimated value of metabolic flux According to the estimation system 100 of the intracellular metabolic flux of this embodiment including the confidence interval determination device 160 provided with the determination method), the confidence intervals of the estimated values r _{1 to} r ₆ of the metabolic fluxes R1 to R6 are calculated. It becomes possible to make a decision by remarkably shrinking.

  Further, depending on the actual individual substrate forming the supplemented medium, there is no problem with the metabolite B to be taken into the cell, but the metabolite D for metabolizing the target product F is the metabolite D. Since it follows various metabolic pathways in the cell before reaching the level and is metabolized as other various metabolites, the labeled isotope in the metabolite D is reduced, and its detection may be difficult. Therefore, for such a substrate, the accuracy of the estimated value of the metabolic flux of the metabolic pathway around the metabolite D for metabolizing the product F can be improved by adding a labeled substrate close to the metabolic pathway. .

  Specifically, glucose can be mentioned as an example of such a substrate. Therefore, in the case of using glucose as a substrate, in the measurement of the observation parameter in the culture experiment by the culture experiment analyzer 130, for example, glutamine is added as a labeling substrate having a close metabolic pathway instead of glucose. The accuracy of the estimated value of the metabolic flux of the metabolic pathway around the metabolite D for metabolizing F can be improved.

In addition, focusing on the flow of carbon positions in the carbon skeleton, substrates with different isotope labeling positions (for example, [U- ¹³ C] glucose labeled with all carbon atoms and carbon atoms at the 1st and 2nd positions) By mixing and analyzing [1,2- ¹³ C] glucose labeled with, the estimation accuracy of the estimation system 100 for intracellular metabolic flux can be increased.

  Further, in the estimation system 100 for intracellular metabolic flux of the present embodiment, as shown in FIG. 2, a component composition ratio setting device 170 is attached to the confidence interval determination device 160. The component composition ratio setting device 170 is based on the estimated value of the flux for each metabolic pathway of each nutrient component consumed by the cultured cells, that is, the substrate forming the added medium, and its confidence interval, obtained as described above. Thus, the component composition ratio of the supplemented medium is obtained by taking the ratio of the metabolic flux for each nutrient component in this intracellular metabolic flux analysis.

  As described above, according to the above-described estimation system 100 for intracellular metabolic flux, it is possible to accurately determine the flux estimated value of the intracellular metabolic pathway, and as a result, the accuracy of the component composition ratio of the supplemented medium is also determined accurately. can do.

<2-2. Control method of addition of added medium>
In the fed-batch culture apparatus 1 shown in FIG. 1, an additive medium 3 having a component composition ratio set, that is, determined by the component composition ratio setting apparatus 170 is used, and the nutrient concentration of the culture solution 2 is a living cell in the culture tank 10. Regardless of the increase in the number, the controller 54 controls the addition of the addition medium 3 so as to be constant for individual cells.

The addition amount and / or addition timing of the additional medium 3 from the additional medium supply tank 56 is based on the specific growth rate μ of the cultured cells calculated by the controller 54 every time the culture medium 2 is sampled by the culture medium analyzer 52. It is determined by the controller 54 as follows.
First, cell proliferation follows the following formula (1).

When the culture solution analyzer 52 can measure the number of living cells in the culture solution 2, the controller 54 sends the number of living cells in the culture tank 10 to the culture solution analyzer 52 at a preset sampling interval. _Xv is measured, and based on the measurement result, the specific growth rate μ _n of the living cells at the sampling time of this time (n-th time) is calculated using the formula (2) obtained by integrating the formula (1). calculate.

The controller 54 uses the value of the specific growth rate μ _n of the living cells at the current (n-th) sampling time to calculate the next ((n + 1) th) sampling time from the current (n-th) sampling time. Predict changes in cell growth over time.

That is, the controller 54 counts the number of living cells X _v (at time t after Δt time from the current t _n within the sampling interval from the current t _n to the next ((n + 1) th) sampling time t _{n + 1.} t) is obtained by equation (3) based on equation (2).

In the above formula (3), since the amount of liquid sampled with respect to the amount of the culture solution 2 stored in the culture tank 10 is sufficiently small, the number of living cells X _v (t) in the culture solution 2 is The change in influence before and after sampling can be ignored. When the amount of the sampling solution cannot be ignored relative to the amount of the culture solution 2, taking into account the influence, the correction term for the number of viable cells that is included in the amount of the sampling solution is reduced in the above equation (3). May be added.

FIG. 10 is a graph showing a growth curve of the number of living cells X _v (t) expressed by the equation (3).
As shown in Equation (3), when the number of living cells at the time (n) sampling t _n is X _vn , the number of living cells X _v (t) thereafter is based on the natural logarithm e, and the exponent is It increases in proportion to an exponential function as a product of the specific growth rate μ and the elapsed time (t−t _n ) from the sampling time t _n .

Therefore, the total number X _{Total of} living cells to be cultured from the current (n-th) sampling time t _n to Δt time shorter than the sampling interval is _calculated as shown in Equation (4). This corresponds to the time integral value of the number _Xv .

Thus, controller 54, if obtaining the number of viable cells X _v at sampling time (n-th), the culture object between sampling time t _n of the current (n-th) until after △ t time The total number of living cells X _Total can be obtained. The component composition ratio of the additive medium 3 added from the additive medium supply tank 56 to the culture tank 10 becomes the ratio of the metabolic flux regarding each nutrient component based on the intracellular metabolic flux analysis of the intracellular metabolic flux estimation system 100. since there, the live cells Total Total X _Total corresponds to the addition amount of the additive medium 3 between the sampling time t _n of the current (n-th) until after △ t time.

Thus, controller 54, and time division sampling time t _n of the current (n-th) next ((n + 1) th) sampling interval until the sampling time t _{n + 1} Furthermore △ plurality every t time, the The addition amount of the supplemented medium 3 at each time-divided timing is determined on the basis of the total number of live cells X _Total for each individual timing.

Accordingly, the controller 54, the amount of added medium 3 until the next timing for each timing of time division, can be determined in accordance with the variation of the viable cells Total Total X _totall, raw this variation The operation of the pump 58 is controlled so that the added medium 3 corresponding to the number of cells is added from the added medium supply tank 56 to the culture tank 10.

Next, determination of the addition amount of the supplement medium 3 at each time-divided timing will be described by taking glutamine as an example.
FIG. 11 is a diagram showing the relationship between glutamine consumption and the time integral of the number of living cells.

In the figure, the time scale plot interval is 12 hours. Proportionality constant (slope) k between the time integral of the glutamine consumption and viable cell count X _v is changed by increasing the number of viable cells X _v with increasing culture time t, the sampling interval △ t is 1 hour In the case of the degree, the time change of the proportionality constant k within the sampling interval Δt is small, and the change of the value can be ignored.

Therefore, when the sampling interval Δt is about 1 hour, the glutamine consumption amount V is an increase in the glutamine consumption amount ΔVt _n during the sampling interval Δt, and the relationship with the time integral value ∫ (live cell number) dt of the number of living cells. Therefore, the following equation (5) is followed.

Therefore, the controller 54 uses the number of viable cells and the amount of glutamine in the culture solution 2 measured by the culture solution analyzer 52 for each sampling, and the live controller between the sampling time t _n and Δt time later. A proportional constant k that defines the relationship between the time integral value of the cell number X _v and the increase ΔVt _n of the glutamine consumption V is calculated, and the calculated proportional constant k and the number of living cells X _v at the sampling time t _n are calculated. Using the viable cell count X _v time integrated value until Δt time calculated based on the above, in the culture solution 2 from this time (nth) sampling time t _n to Δt time later The amount of glutamine consumption ΔVt _n consumed by the living cells is calculated. Then, the controller 54 predicts the glutamine concentration in the culture solution 2 that decreases in accordance with the calculated glutamine consumption ΔVt _n , so that the glutamine concentration in the culture solution 2 does not decrease this time (n determining a medium amount of added medium 3 to be added during the sampling time t _n until after △ t time times eyes).

  Then, the controller 54 controls the operation of the pump 58 with the amount of the added medium determined based on the glutamine concentration in the culture solution 2 to control the supply of the added medium 3 to the culture solution 2 in the culture tank 10. Thereby, since the component composition ratio of the addition medium 3 is the ratio of the metabolic flux with respect to each nutrient component based on the intracellular metabolic flux analysis of the intracellular metabolic flux estimation system 100, each nutrient component other than glutamine In addition, the addition medium 3 is controlled while maintaining a constant concentration.

  In addition, in order to control the concentration of one or more types of specific nutrients to be higher than the current concentration during the culture, the component composition ratio is the ratio of metabolic flux for each nutrient based on intracellular metabolic flux analysis. In addition to the addition control of the additional medium 3, as shown in FIG. 1, the concentration change during the culture can be changed by separately adding the additional medium 4 having only the component whose concentration is changed at this time. It is also possible to cope with the addition control of the required additional media 3 and 4.

  For example, the fed-batch culture apparatus 1 is configured to have a plurality of additional medium supply tanks 56, 56 as indicated by broken lines in FIG. 1, and one additional medium supply tank 56 has an additive medium having a predetermined component composition ratio. 3, the other supplemented medium supply tank 56 stores the supplemented medium 4 in which only a specific nutrient component to be changed among the nutrient components of the additive medium 3 is stored at a high concentration. In this case, the controller 54 cultivates the added medium 4 from the added medium supply tank 56 by a difference between the control value after the specific component concentration change and the control value by the added medium 3 before the change at a predetermined timing during the culture. The tank 10 is supplied. Thus, according to the fed-batch culture apparatus 1, it becomes possible to change the component density | concentration of the specific nutrient in the addition culture medium 3 of a predetermined component composition ratio to the high density | concentration different from the present condition.

  On the other hand, in the case of confirming whether each nutrient component satisfies the control value in the culture solution 2 of the culture tank 10 to which the additive medium 3 has been added, the controller 54 analyzes the nutrient solution of all the nutrient components in the additive medium. Although it is only necessary to be able to monitor by the total 52, the controller 54 performs the addition control of the addition medium 3 based on the nutrient component with the narrowest confidence interval (high accuracy) even if not all the nutrient components can be monitored, It is also possible to evaluate the culture control of cultured cells in the fed-batch culture apparatus 1 by preferentially monitoring nutrient components with a wide confidence interval (poor accuracy) and examining deviations from the control values. .

On the other hand, when the culture solution analyzer 52 cannot directly measure the number of living cells in the culture solution, the controller 54 uses the dissolved oxygen concentration or the medium component concentration as an index as an index in the culture tank 10 as follows. Viable cell count _Xv is calculated.

For example, when dissolved oxygen is used as an index, the controller 54 measures the dissolved oxygen concentration in the culture solution based on the measurement output of the DO electrode, and derives the amount of time variation of the oxygen amount. Since oxygen consumption rate and viable cell numbers, which have a very high correlation, the controller 54, the time variation amount of the measured amount of oxygen, from the correlation equation to derive the number of viable cells X _v in the medium 2 that Can do. If viable cell count X _v is determined, then, viable cells total total becomes culture object between sampling time t _n until after △ t time Similarly this culture solution 2 (n-th) X _Total , i.e. by calculating the time integration value ∫ (viable cell count) dt in the number of viable cells X _v, it is possible to predict the amount of feed medium 3 to be added.

  When the medium component concentration is used as an index instead of the dissolved oxygen concentration, the number of living cells in the culture solution 2 can be predicted using a carbon source such as glucose as an index.

  As a specific example, a case where the number of living cells in the culture solution 2 is derived by measuring the glucose concentration and the lactic acid concentration in the culture solution 2 with the culture solution analyzer 52 will be described first.

In this case, the time _t n of glucose consumption QGlc _n and lactate consumption QLac _n in (n-th sampling), the following equation (6) can be calculated by the equation (7).

In the above formulas (6) and (7), the amount of liquid sampled with respect to the volume of the culture solution 2 stored in the culture tank 10 is sufficiently small, so the number of living cells X _v (t ), The change in influence before and after sampling can be ignored. When the amount of the sampling solution cannot be ignored with respect to the amount of the culture solution 2, taking into account the influence, the above-mentioned equations (6) and (7) reduce the living cells included in the sampling solution amount. A few minutes of correction terms may be added.

FIG. 12 is a diagram showing the relationship between the total of glucose consumption and lactic acid consumption and the time integration of the number of living cells.
A proportional relationship as shown in FIG. 12 is established between the time integration of the number of living cells and the total of the consumption of glucose and the consumption of lactic acid. Or previously obtained in advance by experiments the inclination k of the proportional relationship, or the increase in time integral value and glutamine consumption V in the number of viable cells X _v between the sampling instants t _n described above until after △ t time △ Similarly to the case of obtaining the proportionality constant k that defines the relationship with Vt _n , the amount of glucose and the amount of lactic acid in the culture solution 2 measured by the culture solution analyzer 52 for each sampling are determined by the culture solution analyzer. While monitoring according to 52, the slope k of the proportional relationship is obtained. Controller 54, once the value of the proportionality constant (slope) k, can be calculated the number of viable cells X _v at time t _n. Controller 54, knowing the number of viable cells, it is possible to predict the growth in the number of viable cells X _v of until the next sampling, in a manner similar to the case of the above-mentioned glutamine method, consumption and lactic acid in glucose The supply amount of the supplemental medium 3 can be predicted and controlled on the basis of the consumption amount.

  According to the fed-batch culture apparatus 1 using the addition control method as described above, the addition medium 3 is metabolized with respect to each nutritional component based on the analysis of intracellular metabolic flux by the estimation system 100 of intracellular metabolic flux. Due to the flux ratio, it is possible to prevent excess and deficiency of medium components, prevent inefficient metabolism due to overnutrition and cell death due to nutrient depletion, and produce high quality cells with excellent yield. It can be produced efficiently.

1 fed-batch culture device, 2 culture medium, 3,4 supplemented medium, 10 culture tank,
21 stirring mechanism, 22 stirring blades, 23 drive unit, 24 sampling nozzle,
25 High pressure steam generator, 26 Temperature control equipment,
31 Gas supply pipe for aeration in liquid, 32 Gas supply pipe for gas phase,
33 supplemented medium supply pipe, 34 sampling pipe, 35 submerged aeration diffuser,
36 valves, 37 valves, 50 feed media system,
52 culture medium analyzer, 54 control controller, 56 supplemented medium supply tank,
58 pump, 100 estimation system of intracellular metabolic flux,
110 estimation processing device, 112 random number part, 114 sequential calculation processing part,
116 isotope ratio distribution processing unit, 118 QP partial problem processing unit,
120 intracellular metabolic pathway database, 130 culture experiment analyzer,
140 culture experiment, 150 metabolic pathway model, 160 confidence interval determination device,
170 Component composition ratio setting device, 200 Intracellular metabolic pathway model.

Claims (8)

The component composition ratio of the supplemented medium is defined as the nutrient consumption ratio of the cultured cells, and the nutrient consumption ratio is estimated by the intracellular metabolic flux analysis using two or more isotope-labeled substrates. addition control method of adding a medium, characterized in that it is computed by taking the ratio of the metabolic flux related components. The two or more kinds of isotope-labeled substrates are set for each nutrient component , paying attention to the carbon atom position and the number of carbon atoms in the carbon atom skeleton of the intracellular metabolite and product of the nutrient component.
The addition control method of the addition culture medium of Claim 1 characterized by the above-mentioned. And in the intracellular metabolic flux analysis using two or more kinds of isotope-labeled substrate, a metabolic flux for each nutrient component, I Ri resulting intracellular metabolite each isotope carbon atoms in the simulation of the isotope-labeled each substrate The ratio is estimated based on a comparison result between the isotope carbon number ratio of each intracellular metabolite obtained by an experiment using GC-MS / MS for each isotope-labeled substrate. The method for controlling addition of the supplemented medium described in 1. The addition control method for an addition medium according to claim 3 , further comprising acquiring the accuracy of the estimated metabolic flux for each nutrient component as a confidence interval of the estimated metabolic flux value. Determine the supply amount of supplemented medium according to the growth rate of the cultured cells,
The method for controlling addition of an additive medium according to any one of claims 1 to 4, wherein the nutrient component of the culture solution inoculated with the cultured cells is maintained at a constant concentration. 6. The method for controlling addition of an added medium according to claim 5, wherein the growth rate of the cultured cells is measured using glucose consumption, lactic acid secretion, or glucose or lactic acid addition as an index. The supply of the supplemented medium is controlled based on the nutrient component with the narrowest confidence interval (high accuracy) in the supplemented medium, and the component with the wide confidence interval (poor accuracy) in the supplemented medium is monitored with priority. 5. The method for controlling addition of an added medium according to claim 4 , wherein deviation from the supply control value is examined. The nutrient consumption of the cultured cells determined by taking the ratio of the metabolic flux for each nutrient component of the cultured cells estimated by the intracellular metabolic flux analysis using two or more isotope-labeled substrates . A cell culture characterized by comprising a feed medium system for controlling the supply medium to be fed to the culture tank using the growth rate of living cells in the culture medium of the culture tank inoculated with the cultured cells as an index apparatus.

Images