CN110004108B

CN110004108B - Novel animal cell culture amplification method based on three-dimensional shear space

Info

Publication number: CN110004108B
Application number: CN201811589225.XA
Authority: CN
Inventors: 夏建业; 李超; 腾小锘; 易小萍; 庄英萍; 张嗣良
Original assignee: East China University of Science and Technology
Current assignee: East China University of Science and Technology
Priority date: 2018-12-25
Filing date: 2018-12-25
Publication date: 2021-06-25
Anticipated expiration: 2038-12-25
Also published as: CN110004108A

Abstract

The invention relates to a novel animal cell culture amplification method based on a three-dimensional shearing space, which comprises the following steps: (1) firstly, quantitatively analyzing the shearing environments in reactors of different scales by a CFD (computational fluid dynamics) method, and verifying by a laser particle velocimeter (PIV) method; (2) then, determining a shearing parameter which can be used for representing the shearing environment in the reactor according to a laboratory cell culture result, and taking the shearing parameter as an amplification standard; (3) further establishing an optimal three-dimensional shear rate operation space according to the correlation analysis of the live cell quantity and the three characteristic shears of Spodoptera frugiperdaSf9 in the reactors with different scales; (4) and (3) converting the shear rate operation space into a stirring rotating speed operation space according to the established correlation between the shear rate and the blade tip speed, and carrying out experimental verification on reactors of different scales to finally successfully realize the amplification of the cells from the experimental scale to the production scale.

Description

Novel animal cell culture amplification method based on three-dimensional shear space

Technical Field

The invention relates to an animal cell culture amplification method, in particular to a shear sensitive cell amplification method.

Background

Animal cells have strong advantages in protein expression and modification, and suspension culture is a culture mode for efficiently producing antibodies, vaccines and protein drugs, wherein an insect cell-based expression platform such as an insect baculovirus expression system (BEVS) is widely applied to vaccine production. However, in the process of production scale-up, the environments such as mass transfer, mixing and shearing in the reactor can be greatly changed along with the change of the scale and the structure of the reactor. To this end, the relevant scholars have proposed many amplification methods, including theoretical analysis, semi-theoretical analysis, analytic analysis and empirical methods. Tescione et al found different amplification strategies such as power per unit volume (P/V), volumetric oxygen mass transfer coefficient (k) in the course of CHO cell amplification studies_La) And Oxygen Transfer Rate (OTR), the final cell growth conditions vary widely. In the stem cell culture process, Borys and the like research the influence of seven common amplification strategies (average speed, average shear rate, average energy dissipation rate, Reynolds number, leaf tip speed, power input and maximum shear rate) on the process, and as a result, the results show that the stirring rotating speed phase obtained by different methods through calculation is obtainedThe difference is large, and the growth and aggregation state of the cells also vary greatly. Fluid mechanics characteristics (including mass transfer, mixing, shear, etc.) within the reactor during scale-up exhibit strong structural and scale dependencies, and conventional scale-up methods based on single parameter similarity generally fail to achieve consistent critical flow field characteristics within the reactor. A large number of researches show that the flow field characteristic or the fluid mechanics characteristic in the reactor is a main factor influencing the physiological metabolism of cells, and the maintenance of the similarity of key fluid mechanics parameters is expected to become an efficient biological process amplification method.

Insect cells are very sensitive to shear, especially after transfection of viruses, and therefore changes in the shear environment caused by aeration and agitation are often the major cause of scale-up failure. Since shear is difficult to quantify, many scholars use indirect shear parameters such as blade tip speed, energy dissipation rate, power input per unit volume (P/V), energy dissipation rate/cycle time function (EDCF), maximum or average shear rate, etc. as scaling criteria^[14-17]. However, numerous studies have shown that the shear environment within the reactor is highly non-uniform, and for stirred bioreactors the shear rate in the paddle region is typically several orders of magnitude higher than the tank region. Thus, a single shear parameter is difficult to characterize the shear environment within the reactor. How to quantitatively describe the shear environment within a pilot scale reactor and reproduce it in a large scale reactor is key to achieving production scale-up.

Methods for quantitatively describing the shear environment within a bioreactor include theoretical analysis, experimentation, and numerical simulation. Through theoretical analysis, a number of studies have correlated the average shear rate within the reactor to the agitation speed, aeration rate, and rheology. However, theoretical analysis is only applicable to certain standard conditions (e.g., specific reactor configuration and agitation configuration) and can only evaluate average or maximum shear rates. And the flow field information in the reactor can be obtained through an experimental method based on Particle Imaging Velocimetry (PIV) or Laser Doppler Velocimetry (LDV), and then local shearing information is obtained through calculation. Wu et al established an empirical correlation of shear rate in the reactor with paddle flow accuracy by theoretical analysis and LDV experiments. Some scholars have also established experimental devices to quantify shear rates and used to study the sensitivity of mammalian and insect cells to shear. In addition, indirect experimental methods or parameters (such as oxygen consumption rate) have been proposed by the scholars to characterize shear damage to cells. However, the experimental study method is only suitable for experimental-scale reactors, and the shear rate has strong dependence on the reactor structure, and the data obtained in the laboratory can not be used for evaluating the shear rate in a large-scale reactor. In recent years, the Computational Fluid Dynamics (CFD) method has been widely used for the study of the flow field characteristics of the reactor, and has a strong expandability and economy as compared with the above two methods. In addition, the CFD method can obtain shearing information in the three-dimensional space of the reactor, and is beneficial to statistical analysis of various shearing parameters. Liu et al analyzed the shearing environment that Carthamus tinctorius l. cells undergo in the reactor using a euler-lagrangian based CFD method and successfully simulated the death and growth of cells in a complex shearing environment. Collignon et al established a correlation of cell death with EDCF (describing the magnitude and frequency of shear forces experienced by the cell) by the method of large vortex CFD simulation. However, these methods mainly focus on studying the relationship between shear and cell damage, and do not propose a perfect shear-sensitive cell amplification method.

Disclosure of Invention

The invention aims to provide an animal cell culture amplification method based on shearing environment similarity so as to overcome the defects of the prior art. The specific technical scheme of the invention is as follows:

the novel animal cell culture amplification method based on the three-dimensional shearing space comprises the following steps:

(1) firstly, quantitatively analyzing the shearing environments in reactors of different scales by a CFD method, and verifying by a PIV method;

(2) then determining three characteristic shearing parameters which can be used for representing the shearing environment in the reactor according to the cell culture result of a laboratory, and taking the three characteristic shearing parameters as an amplification standard;

(3) further establishing an optimal three-dimensional shear rate operation space according to the correlation analysis of the live cell quantity and the three characteristic shears of Spodoptera frugiperdaSf9 in the reactors with different scales;

(4) the shear rate operation space is converted into the stirring rotating speed operation space according to the established correlation of the shear rate and the blade end speed, and experiments are carried out on reactors with different scales, so that the amplification of Spodoptera frugiperda Sf9 cells from the experimental scale to the production scale is finally and successfully realized.

Further, the reactor is a 7.5-1000L reactor.

Further, the scale-up from the experimental scale to the production scale is from 7.5-42L to 30-1000L.

The invention has the beneficial effects that: the insect baculovirus expression system (BEVS) is widely used for the production of various vaccines, however the high sensitivity of insect cells to shear often hinders the scale-up of their production process. Therefore, the research provides an animal cell culture amplification method based on a three-dimensional shear space. First, the shear environment in the experimental (7.5L and 42L) and production (30L, 90L, 350L and 1000L) scale reactors was quantitatively analyzed by Computational Fluid Dynamics (CFD) and the results were verified by laser particle velocimetry (PIV) to establish a quantitative relationship between the shear parameters in the reactors (including paddle zone shear rate, tank zone shear rate, average shear rate and maximum shear rate) and the paddle tip velocity. Through the correlation analysis of the cell culture result of Spodoptera frugiperda Sf9 and the in-tank shearing parameter obtained through experiments, it is found that the production amplification of the culture process cannot be successfully realized by a single shearing parameter, the optimal operation space of the shearing rate can be established in three dimensions by adopting three characteristic shearing parameters, namely the shearing rate of the paddle region, the shearing rate of the tank region and the average shearing rate, and the amplification of the cell culture process can be realized when the three characteristic parameters in the reactor are positioned in the space. Based on the method, the optimal stirring speed operation space of the production scale reactor is established according to the obtained correlation between the shear rate and the blade end speed, verification is carried out on reactors of different levels of production scales, finally, amplification is successfully realized on a 1000L reactor, and the cell growth rate and the living cell amount are consistent with those of a pilot experiment (the final living cell number is about 700 multiplied by 106 cells/mL). The amplification strategy is expected to be applied to the production amplification process of other shear sensitive cells.

The invention is further illustrated with reference to the following figures and examples.

Drawings

FIG. 1 shows the basic structure of six reactors, 7.5L and 42L for experimental research and 30-1000L for production.

FIG. 2 is a reactor configuration for PIV experiments, where (A, three dots at different heights represent sampling points for velocity comparison), velocity cloud plot comparison results (B, left half PIV results, right half CFD simulation results), and local velocity comparison results (C-E) at different heights.

FIG. 3 comparison of CFD simulation with average shear rate calculated from empirical formula.

FIG. 4 is a graph of the relationship between the shear rate (SSR) in different reactors and the tip speed of the stirring blades obtained by simulation, where (A: shear rate in the ear-like (EE) blade zone, compared with 30L of propeller blade (APP), B: shear rate in the triclinic blade (PBT) blade zone, C: shear rate in the propeller blade (APP) blade zone, D: shear rate in the tank body, E: overall average shear rate, and F: maximum shear rate).

FIG. 5 analysis of the distribution of the maximum shear rate, wherein (A: ear-like blade EE; B: propeller blade APP; C: combination blade APP + PBT).

FIG. 6 shows the relationship between the shearing parameter and the final viable cell density under different operating conditions. Wherein (blue dots indicate operating conditions where viable cell density is more desirable, red dots indicate operating conditions where viable cell density is less desirable; horizontal dashed lines indicate desirable viable cell density thresholds; two vertical dashed lines indicate boundaries of the magnification criteria).

FIG. 7. three dimensional operating space (cubic area composed of black line frame) and distribution of each experimental operating point (red dot for poor operating condition, blue dot for good operating condition, green dot for operating point used for production enlargement; two sets of numbers beside the dots respectively indicate reactor scale and stirring speed).

FIG. 8 shows the growth of cells in each scale reactor at different stirring speeds (A: small scale reactor; B: larger scale reactor).

FIG. 9 shows the morphology of Spodoptera frugiperda Sf9 cells in a low shear environment (A, 1000L-30rpm) and in a suitable shear environment (B, 1000L-48rpm) (both 84h sampling microscopy results).

Detailed Description

1. The novel animal cell culture amplification method based on the three-dimensional shearing space comprises the following steps:

2. Materials and methods

2.1 bioreactor architecture

Six-scale (7.5-1000L) stirred bioreactors are used in the present invention, the basic structure of which is shown in FIG. 1. Both pilot scale reactors (7.5L and 42L) were equipped with a three-bladed "elephant-ear" paddle (EE) with the 42L paddle positioned above each other. The 30L reactor was equipped with a three-bladed propeller Agitator (APP). The other reactors adopt the same stirring paddle type and respectively comprise a three-blade propelling stirring paddle (APP) and a three-inclined-blade stirring Paddle (PBT), and the diameters of the stirring paddles are different.

2.2 cell culture method

Recombinant baculovirus expression for use in the inventionThe system was Spodoptera frugiperda Sf9, stored in serum-free medium Sf900 II (Invitrogen, USA). Firstly, pre-culturing in a shake flask at the culture temperature of 28 ℃ and the rotation speed of a shaking table of 60rpm, inoculating into a fermentation tank with the experimental scale of 7.5L after culturing to the exponential growth phase, and further expanding to 42L. In industrial production, shake flask seeds are first inoculated into a 30L fermentation tank and then gradually expanded to 1000L (30L-90L-350L-1000L). The initial cell density after inoculation in each fermenter was 80X 10⁶cells/mL. The stirring speed is constant in the culture process, and the dissolved oxygen level is maintained to be not less than 40% through a four-gas control system. Viable cell counts were performed using trypan blue staining with three assays per sample.

2.3CFD simulation method

During insect cell culture, Weidner et al found that bubble size was not a critical factor in cell damage, which is primarily related to the size of the gas-liquid interface. Because of the low ventilation used in this study (1)<0.1vvm), the gas content and the gas-liquid interface are small, and in addition, the shear rate caused by aeration is difficult to quantitatively analyze, so that the fermentation system is simplified into single-phase flow in the CFD simulation process. The medium used in the simulation was replaced with water because of the relatively dilute Sf900 II medium, which has a rheological profile close to that of water. Flow field characteristics within the reactor were analyzed using commercial CFD software ANSYS CFX 15.0(ANSYS inc., USA). The reactor grids are divided by ANSYS ICEM CFD 15.0.0 (ANSYS Inc., USA), and are all unstructured tetrahedral grids, and the grids of the blade area are partially encrypted, so that the total grid number is about 100 ten thousand. Each reactor simulates five stirring rotating speeds, the corresponding speed range of the blade end of the paddle is 0.3-1.5m/s, the minimum stirring Reynolds number is 21000, and the flow reaches a complete turbulent flow state. The turbulence simulation uses a standard k-epsilon model to seal the turbulence equation, and the blade rotation is described using a Multiple Reference Frame (MRF) method. The momentum and continuity equations are described using the time-averaged NS equation. The invention obtains the velocity component U according to CFD simulation_x、U_yAnd U_zShear Rate (SSR) was calculated as follows:

monitoring and calculating residual error and blade torque in the simulation process, and when the residual error is less than 10^-4And the blade torque is stable, and the iterative process is considered to be converged. The solving operation was performed on a 96-core eosin server (Sugon co., ltd., China) consisting of 5 nodes.

2.4PIV Experimental method

The PIV experimental platform adopted for flow field verification in the invention is the same as the previous report, and mainly comprises a laser emitter (Leamtech, Nd: YAG, 200mJ, 15Hz), a PIV camera (PCO2000, 2048 multiplied by 2048pixels), a synchronizer and Vidpiv 4.7 data processing software. In addition, a shaft encoder is added in the research to lock the angle of the stirring paddle blade, so that PIV experimental data of angle analysis are obtained. The reactor for PIV flow field measurement is a 42L acrylic transparent tank provided with two elephant ear type stirring paddles (EE), and the basic structure is shown in figure 2A. The flow field measurement area is a vertical plane in the middle of the two baffles. Due to the shielding of the stirring shaft and the reflection of the bottom of the tank, the measuring area only comprises a half of the straight cylinder area of the tank body. Detailed experimental scale-up and data processing methods are described in the literature of the prior art.

3. Results and discussion

3.1 validation of the CFD model by PIV experiments

The CFD model is verified through a PIV flow field measurement experiment in a 42L reactor, and the stirring speed adopted in the experiment is 50 rpm. Comparing and analyzing the velocity cloud charts obtained by the PIV experiment and the CFD method, the flow fields obtained by the two methods are very similar, and a larger axial circulation is integrally formed on one side of the reactor (figure 2B). To further compare and analyze the local velocities within the reactor, the local average velocities at three different heights along the reactor axis were selected for comparison and analysis (FIGS. 2C-E). From the results, it can be seen that the peak velocity of the blade discharge area obtained by the simulation is slightly higher than the experimental value, and the average velocities at other positions are very similar to the experimental value. Singh et al also showed that the k-epsilon turbulence model predicts the turbulence energy level better than other models (SST, SSG-RSM and SAS-SST), but the predicted blade region velocities are higher. Liu and the like similarly find that the peak speed of a blade area obtained by predicting by a k-epsilon turbulence model is higher. Wu et al indicate that the peak speed of the paddle region obtained from the simulation is high even with high precision large vortex simulation. The experimental data were found by statistical analysis to be not significantly different from the data obtained by the simulation (p values at Z ═ 0.29m, 0.21m and 0.11m were 0.346, 0.192 and 0.259 respectively, T-test). Therefore, the flow field data obtained by the CFD method simulation provided by the invention has sufficient precision, and the shear rate obtained by the velocity field calculation is also credible.

3.2 quantitative analysis results of shear Environment

The invention adopts the CFD method after experimental verification to analyze the flow field conditions in the reactors with different scales and obtain the shearing related information. By simulating the shear strength at different stirring speeds, a correlation between the manipulated variables (e.g., stirring speed, tip speed, etc.) and the shear parameters can be established. In order to further verify the accuracy of the simulation, the average shear rate under the condition of the single-layer stirring paddle obtained by the simulation is compared and analyzed with the average shear rate obtained by calculation through an empirical formula reported in a literature, and the calculation method comprises the following steps:

wherein

Is the average shear rate(s)^-1)，N_QIs a dimensionless axial flow norm and can be obtained by calculating the axial flow velocity^[34]N is the stirring speed(s)^-1)。

Flow accuracy (N) was simulated for 7.5L and 42L reactors like-ear paddles (EE)_Q) 0.4 and 0.5 respectively, while the flow criterion for the 30L reactor pusher blade (APP) is 0.3. Since the 7.5L reactor has a relatively small mixing blade area (mainly because the blade and hub connections are relatively long), the flow accuracy is slightly less than that of the 42L reactor. Of the EE blades reported by Zhu et alThe flow accuracy is higher (0.7), mainly because the adopted stirring paddle structure (blade angle, connecting piece length of the hub of the blade) and the tank type (whether a baffle is arranged) have larger difference with the invention. As can be seen from FIG. 3, the average shear rate obtained by simulation is very close to the result obtained by calculation with an empirical formula, and the average shear rate and the stirring speed have a better linear relationship, which is the same as the literature report, and thus, the shear rate obtained by simulation is more reliable. It has also been found in this study that although the 7.5L and 42L reactors employ very similar agitation patterns (differing only significantly in installation location), the resulting average shear rates differ significantly, whereas the average shear rates calculated using conventional empirical equations (which relate the average shear rate only to the agitation speed) for both reactors will be the same. It can be seen that the average shear rate is not only related to the blade type, but is affected by the reactor size or blade mounting location.

The invention further relates and analyzes different shearing parameters and the speed of the leaf end, including the shearing rate (using impelleter SSR or SSR) of the stirring area_impExpressed), tank area shear rate (excluding the portion of the blade area, using tank SSR or SSR)_tankExpression), global average shear rate (average of all regions, using average SSR or SSR)_avgExpressed) and maximum shear rate (using maximum SSR or SSR)_maxRepresentation), the correlation is as follows:

SSR_i＝K_S,i U_T (3)

wherein K_S,iIs a model coefficient (m)^-1) Mainly in relation to the paddle and tank structures, the subscript i denotes imp, tank, avg and max, U_TStirring blade tip speed (m/s). FIG. 4 depicts the relationship between shear parameters and tip velocity in different scale reactors. It is clear that all shear parameters decrease progressively with increasing reactor size at the same tip speed when the same paddles or the same combination of paddles are used. When different stirring paddles are adopted, the change situation of the shear rate is different. For example, the Tank SSR and overall average SSR for the 30L reactor were significantly lower than for the 42L and 90L reactors (FIGS. 4D and 4E). However, it is not limited toThere was no apparent rule for the maximum shear rate in the different scale reactors (fig. 4F), mainly because the maximum shear has a strong dependence on the paddle type. By analyzing the distribution of the maximum shear rate (fig. 5), it can be seen that the maximum shear rate of about 0.01% of the total volume of the reactor is mainly concentrated at the blade edge or at a special narrow structure near the blade (such as a connecting shaft of the blade and the hub, fig. 5A). Numerous studies have shown that the probability of cells experiencing high shear regions is relatively low and the time of exposure is short, and that the maximum shear rate is not the most critical factor causing cell damage.

Table 1 lists the model parameters K obtained from regression analysis of equation 3_S,iValue of (c), correlation coefficient (R) of regression analysis of all models²) Are close to 0.99. The model provided by the invention can be used for quantitative analysis of shear related parameters in other similar reactors, and provides important reference data for amplification research of shear sensitive cells.

TABLE 1 regression coefficients of shear related parameters

3.2 analysis of key factors for amplification

TABLE 2 highest viable cell density under different operating conditions

For Spodoptera frugiperda Sf9 cells, the ideal viable cell density (viable cell density) for commercial scale-up should be greater than 600X 10⁶cells/mL. According to the invention, the growth conditions of cells under different stirring rotation speeds are firstly studied on reactors of various scales, and the final viable cell density result is shown in Table 2. By a pilot scale optimization study, 7 was determined.Optimum stirring rotation speeds for the 5L, 30L and 42L reactors, however, for large-scale reactors of 90L to 1000L, it is extremely uneconomical to determine the optimum stirring rotation speed for each stage reactor step by step using a conventional optimization method. Because the dissolved oxygen is controlled to be more than 40 percent in the whole process of the cell culture process, and the metabolism of insect cells is generally slow (the specific oxygen consumption rate is 1-10 mmol/10)⁹cells/day, specific glucose consumption rate of 1-4mmol/10⁹cells/day)^[37]Thus oxygen transport and mixing are not critical factors affecting the amplification of Spodoptera frugiperda Sf9 cells, and the key to successful amplification is the control of the shear rate during amplification.

According to the invention, various parameters related to shearing in the reactor under various operating conditions, including paddle shear rate (impelleter SSR), tank shear rate (tank SSR), average shear rate (average SSR), maximum shear rate (maximum SSR), tip velocity (tip velocity) and unit volume power (P/V), are obtained by a CFD method, and the final viable cell density is mapped (see figure 6). It can be seen by analysis that the viable cell density is greater than 600X 10, regardless of the shear parameter used as a scale-up criterion⁶The shear operable regions (two vertical line regions in fig. 6) established under the operating conditions of cells/mL (blue dots in region i in fig. 6) both contain operating points (red dots in region ii in fig. 6) with lower viable cell density, and in particular, with the maximum shear rate, tip velocity and power per unit volume as the scale-up criteria, the worse results contained in region ii are more significant, i.e., a single shear parameter cannot be used as the scale-up criteria. The main reason for this is that the shear environment inside the reactor is very heterogeneous and a single shear parameter does not accurately describe the actual shear environment inside the reactor.

3.3 three-dimensional shearing operation space enlargement Standard

Because the shearing environment in the reactor has high non-uniformity, the traditional one-dimensional shearing parameters are difficult to accurately describe the shearing environment in the reactor, and the amplification of the cell culture process cannot be successfully realized by using the shearing environment as an amplification standard. Thus, the present invention proposes to use three characteristic shear parameters (i.e. the paddle shear rate SSR) within the reactor_impSSR shear rate in tank field_tankAnd overall average shear rate SSR_avg) As a scale-up criterion for shear-sensitive cells, this criterion facilitates a more accurate replication of the cleavage loop of the small test into a large-scale reactor. In the case of the present study, the shear rate SSR of the tank farm_tankSSR to global average shear rate_avgHowever, when a blade with a relatively large size is used, the difference between the two is large, so that the method is not applicable to a two-dimensional shearing operation space, and a three-dimensional shearing operation space is selected as a magnification standard. The boundary of the three-dimensional operation space is formed by living cell density more than 600 multiplied by 10⁶SSR obtained under cell/mL operating conditions_imp、SSR_tankAnd SSR_avgTo increase the redundancy of the method, the invention expands the boundaries of each dimension by 20% on the basis of the existing data. FIG. 7 is a shear rate operating space (black square area) created by this method, and it can be seen that the space contains all living cells greater than 600X 10⁶cells/mL, while the lower viable cell count was excluded. It can also be seen in conjunction with the data in table 2 that the density of viable cells is lower as the operating point deviates further from the operating space.

The three shear parameters in the amplification process are limited by the established shear rate three-dimensional operation space, and the shear rate operation space can be converted into a stirring rotation speed operation space, namely a stirring rotation speed optimal range, through the correlation (equation 3) between the shear parameters and the stirring blade tip speed provided in the foregoing. The calculated optimal stirring rotation speed range of each reactor is shown in table 3, and the optimal stirring rotation speed is gradually reduced along with the increase of the scale of the reactor in general. The optimum stirring speed does not have a clear linear relationship with the scale of the reactor, and thus it is difficult to determine an appropriate stirring speed by the conventional method. The three-dimensional parameter amplification method provided by the invention can accurately predict the better stirring rotating speed of the large-scale reactor, and reduce the operable range of the stirring rotating speed to be within 10rpm, thereby having good guiding significance for the amplification process.

TABLE 3 operable range of stirring speed calculated from shearing operation space and experimental verification points

3.4 Experimental validation of novel amplification method

Culture experiments of Spodoptera frugiperda Sf9 cells were performed on reactors of different sizes according to the optimal stirring rotation speed calculated by the three-dimensional scaling-up strategy (Table 3), and the viable cell density during the culture is shown in FIG. 8. From the results, it can be seen that the growth rates of cells under different shearing environments are greatly different, and the cell density is up to over 700 × 10⁶cells/mL, and at least, can be less than 200X 10⁶cells/mL. The present invention finally selects 60rpm, 55rpm and 48rpm for validation of the amplification method on 90L, 350L and 1000L reactors, respectively, and selects the results at 50rpm, 40rpm and 30rpm, respectively, as controls. Compared with the control, the cell growth rate and the final cell concentration in the industrial scale reactor (90L, 350L and 1000L) obtained by adopting the amplification strategy are both obviously improved, and the viable cell density reaches 700 multiplied by 10 within 110h⁶cells/mL (FIG. 8B), very close to the pilot study results. Comparative analysis of cell morphology patterns obtained in a 1000L reactor at 30rpm and 48rpm (FIG. 9) shows that some cells were aggregated at lower rotation speeds (i.e., lower shear rates) and the cells were more uniformly dispersed at 48 rpm. The results show that under the condition of low shearing, cell agglomeration can block the mass transfer of nutrient substances, thereby influencing the growth of cells and finally leading to lower living cell density. It is expected that if the stirring speed exceeds the upper limit of the three-dimensional shear space, the cells will be damaged, and cell debris will be found by microscopic examination, however, the destructive experiment is not performed on a 1000L reactor in view of the cost of the experiment. Based on the optimized three-dimensional shearing operation space, the shearing environment information obtained by a small test is reproduced in the production scale, and the production amplification of Spodoptera frugiperda Sf9 cells is successfully realized.

4. Conclusion

To solve the problem of shearingThe invention relates to a production amplification problem of sensitive animal cells, and provides an amplification method based on a three-dimensional shearing operation space. Firstly, quantitative research is carried out on the shearing environment in the reactor with experimental scale and production scale by a computational fluid mechanics method, and the result is verified by a PIV experiment, so that a quantitative relation between each shearing parameter in the reactor and the speed of the end of the stirring paddle is established. The average shear rate calculated by the literature is very close to the result obtained by simulation, and the accuracy of the simulation is further verified. Through the correlation analysis of cell culture results and various shearing parameters in the reactor, the single shearing parameter can not be successfully amplified, but three characteristic shearing parameters SSR are adopted_imp、SSR_tankAnd SSR_avgThe shearing environment in the reactor can be described more accurately, a three-dimensional shearing rate operation space can be established according to the shearing environment, all poor experiment results can be excluded from the space, all good experiment results are included, and the three-dimensional shearing rate operation space is used as an amplification standard, so that three characteristic shearing parameters in the amplification process can be ensured to be consistent with a small experiment. And calculating the optimal stirring rotating speed range of the production scale reactor according to the amplification standard and the correlation of the shear rate and the stirring rotating speed. Finally, the Spodoptera frugiperda Sf9 cell amplification experiment is carried out on a production scale reactor, the cell growth rate and the final cell density are both obviously improved, and the experimental level of a small experiment (the cell density is about 700 multiplied by 10)⁶cells/mL), the feasibility of this amplification method was successfully verified.

In summary, the present invention proposes an animal cell amplification method based on three-dimensional shearing operating space (similar principle of shearing in three dimensions), and is successfully applied to production amplification of Spodoptera frugiperda Sf9 cells (from 7.5L experimental scale to 1000L production scale). The amplification method has important significance for guiding the production amplification of the shear sensitive cells, and can be further popularized to the production amplification process of similar shear sensitive cells.

Claims

1. The novel animal cell culture amplification method based on the three-dimensional shearing space is characterized by comprising the following steps of:

(1) firstly, quantitatively analyzing the shearing environments in reactors of different scales by a CFD (computational fluid dynamics) method, and verifying by a laser particle velocimeter (PIV) method;

(2) then, determining a shearing parameter which can be used for representing the shearing environment in the reactor according to a laboratory cell culture result, and taking the shearing parameter as an amplification standard;

shear rate SSR of stirred zone_impSSR (simple sequence repeat) of representation and tank body region shear rate_tankRepresentative, Overall average shear Rate SSR_avgRepresentative and maximum shear Rate SSR_maxThe correlation is expressed as follows:

SSR_i＝K_S，iU_T (3)

wherein K_S,iIs a model coefficient (m)^-1) Mainly in relation to the paddle and tank structures, the subscript i denotes imp, tank, avg and max, U_TStirring blade tip speed (m/s);

2. The novel animal cell culture amplification method based on three-dimensional shear space of claim 1, wherein the amplification criterion is a quantitative relationship between the shear parameter and the tip speed of the stirring blade.

3. The novel animal cell culture amplification method based on three-dimensional shear space of claim 1, wherein the shear parameters of paddle shear rate, tank shear rate and average shear rate are used to establish an optimal operating space for shear rate in three dimensions, and when the three characteristic parameters in the reactor are located in the space, amplification of the cell culture process is achieved.

4. The novel animal cell culture amplification method based on the three-dimensional shear space as claimed in claim 1, wherein the optimal stirring rotation speed operation space of the production scale reactor is established according to the correlation between the obtained shear rate and the leaf tip speed, so as to realize production amplification.