Fermentation of fastidious bacterial strain in perfusion suspension culture

Abstract

The present invention relates to improved processes for culturing bacteria, in particular to processes for perfusion suspension culturing of bacteria in a fermenter, wherein the culture medium including the bacteria is circulated over a separation system in alternating tangential flow, wherein the separation system removes a filtrate containing inhibitory metabolites from the culture medium.

Claims

1. A process for culturing a fastidious bacterial strain in a fermenter comprising the steps of: a) adding a liquid growth medium comprising a carbon source to a fermenter; b) seeding the growth medium with the fastidious bacteria to form a culture medium; and c) growing the fastidious bacteria in perfusion suspension culture, where an average perfusion rate is above 5% of culture volume per hour; wherein the culture medium including the bacteria is circulated over a separation system in alternating tangential flow, and the separation system removes a filtrate containing spent medium from the culture medium and retains the fastidious bacteria in the culture medium, d) reducing the perfusion rate of step c) when the carbon source in the medium is exhausted.

2. The process according to claim 1 wherein the spent medium contains inhibitory metabolites.

3. The process according to claim 1 wherein the separation system comprises a filter module comprising hollow fibre membranes for the removal of inhibitory metabolites from the culture medium.

4. The process according to claim 1 wherein in step c) the average perfusion rate is 10% or above of the culture volume per hour.

5. The process according to claim 1 wherein the perfusion rate in step c) is reduced 1.5 or 2 fold when the carbon source in the medium is exhausted.

6. The process according to claim 1 wherein in step c) the dissolved oxygen level is kept at between 10% and 30% of the initial level.

7. The process according to claim 1 wherein the suspension culture has a volume selected from at least 10 liters, at least 20 liters, at least 50 liters, at least 100 liters, and at least 250 liters.

8. The process according to claim 1 wherein the density of the bacteria reaches at least 10 OD units measured at 650 nm.

9. The process according to claim 1 wherein the bacterial strain is selected from the group consisting of Bordetella pertussis, Neisseria meningitidis, Cornyebacterium diphtheriae, Clostridium tetani, Clostridium difficile, Helicobacter pylori, Haemophilus influenzae, Staphylococcus aureus, Streptococcus pneumoniae, Salmonella species, Spirochetes species, Legionella species and Mycobacterium tuberculosis.

10. The process according to claim 9 wherein the bacterial strain is selected from the group consisting of Neisseria meningitidis serogroup A, Neisseria meningitidis serogroup B, Neisseria meningitidis serogroup C, Neisseria meningitidis serogroup W135 and Neisseria meningitidis serogroup Y.

11. The process according to claim 10 wherein the bacterial strain is Neisseria meningitidis serogroup B.

12. The process according to claim 9 wherein the bacterial strain is a strain of Streptococcus pneumoniae.

13. The process according to claim 12 wherein the bacterial strain is Streptococcus pneumoniae serotype 1, 4, 5, 6A, 6B, 7F, 9V, 12F, 14, 15C, 33F, 18C, 19A, 19F, 22F or 23.

14. A process for producing a biopolymer or an aggregate of biopolymers including the steps of i) culturing a bacterial strain according to claim 1, and ii) harvesting the biopolymer or aggregate thereof from the culture medium or filtrate.

15. The process for producing a biopolymer or aggregate thereof according to claim 14, which comprises a further step of conjugating the biopolymer or aggregate thereof to a saccharide.

16. The process for producing a biopolymer or aggregate thereof according to claim 14, which comprises a further step of conjugating the biopolymer or aggregate thereof to a carrier protein.

17. A process for producing a vaccine comprising the steps of 1) producing a biopolymer or aggregate thereof using the process of claims 14, and 2) formulating the biopolymer or aggregate thereof as a vaccine by adding a pharmaceutically-acceptable excipient.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1 shows a systematic representation of a fermentation system incorporating ATF showing the flow of the fermentation broth to a vessel through the ATF system upon pressurisation.

(2) FIG. 2 shows a systematic representation of the fermentation system shown in FIG. 1 showing the flow of the fermentation broth from the vessel through the ATF system upon exhaust.

(3) FIGS. 3, 4 and 5 show a typical N. meningitidis fermentation profile with the process parameters monitored during 20 L-scale perfusion fermentation when perfusion rate is mediated by oxygen demand as in Example 1.

(4) FIG. 3 shows a typical N. meningitidis fermentation profile ATF with pO2/substrate regulation (Agitation speed (rpm) and dissolved oxygen (%) over time.

(5) FIG. 4 shows a typical N. meningitidis fermentation profile ATF with pO2/substrate regulation (Acid consumption (grams) or base consumption (grams) over time.

(6) FIG. 5 shows a typical N. meningitidis fermentation profile ATF with pO2/substrate regulation (Temperature ( C.) and Head Pressure (bar) and pH).

(7) FIGS. 6, 7 and 8 show typical N. Meningitides fermentation profiles with the process parameters monitored during 20 L-scale fermentation where pre-determined perfusion rates are applied, as in Example 2.

(8) FIG. 6 shows a typical N. meningitidis fermentation profile ATF with constant perfusion rate (Agitation speed (rpm) and dissolved oxygen (%) over time.

(9) FIG. 7 shows a typical N. meningitidis fermentation profile ATF with constant perfusion rate (Acid consumption (grams) and base consumption (grams)) over time.

(10) FIG. 8 shows a typical N. meningitidis fermentation profile ATF with constant perfusion rate (Temperature ( C.) and Head Pressure (bar) and pH).

(11) FIG. 9 is a process flowsheet for batch mode fermentation of N. meningitidis.

(12) FIG. 10 is a process flowsheet for perfusion mode fermentation of N. meningitidis.

(13) FIG. 11 shows a fermentation profile over time of Streptococcus pneumoniae serotype 22F with ATF with the process parameters monitored during 20 L-scale perfusion fermentation as in Example 3.

(14) FIG. 12 shows the accumulation of polysaccharide in the different fractions (permeate and culture) over time from Streptococcus pneumoniae serotype 22F culture with ATF as in Example 3.

(15) FIG. 13 shows the evolution of glucose and lactate concentration in the culture over time from Streptococcus pneumoniae serotype 22F culture with ATF as in Example 3.

DETAILED DESCRIPTION

(16) Referring to FIG. 1 there is shown a fluid filtration system comprising a perfusion fermenter 1 which is connected to a housing assembly 2, containing a hollow-fibre filtration module. The housing assembly 2 is positioned between the fermenter 1 and a diaphragm pump 3, controlled by an an alternating tangential flow controller (ATF) 4. A filtrate pump 5, such as a double headed peristaltic pump, (e.g. a Watson Marlow 505U) is used for controlled addition of the feeding medium from inlet 6 and removal of the filtered stream (exhaust medium) by an outlet 7. The rate of the filtrate pump 5 may be automatically controlled by an online measurement performed by a sensor such as a pO2 probe 11 through a pump controller 12. Unfiltered material remains in the system.

(17) The diaphragm pump 3 provides the means for generating alternating tangential flow between fermenter 1 and filtrate pump 5, through the hollow fibres, thus generating rapid, low shear, tangential flow. The diaphragm pump is partitioned into two chambers with a flexible diaphragm. The first pump chamber serves as a liquid reservoir that registers with the fermenter through the fibre module. The second pump chamber is an air chamber that registers with the pump flow control system. In FIG. 1 (Phase 1, Pressurisation), controlled addition of compressed air from the air inlet 8 via the air filter 9 into the system increases the pressure in the second pump chamber (air chamber) relative to the fermenter. This expands the air chamber, inversely reducing the volume in the first pump chamber (liquid reservoir), driving the liquid content away from the diaphragm pump, through the hollow fibre module and to the fermenter, thus generating tangential flow in one direction.

(18) Referring to FIG. 2 there is shown the fluid filtration system of FIG. 1 in exhaust phase. In FIG. 2 (Phase 2, Exhaust), the pressure in the air chamber is reduced by removal of compressed air from the air outlet 10 and so is reduced relative to the pressure in the fermenter, and liquid flows from the fermenter to the liquid reservoir of the pump. The filtered stream (exhaust medium) is removed by outlet 7 generating tangential flow in the other direction. This movement is helped by fermenter overpressure (0.5 to 0.1 bar).

(19) In order that this invention may be better understood, the following examples are set forth. These examples are for purposes of illustration only, and are not to be construed as limiting the scope of the invention in any manner.

EXAMPLES

Example 1

(20) Example 1 describes a process for growing Neisseria meningitidis in perfusion culture using ATF. Example 1(a) describes a process for making a solid pre-culture of N. meningitidis, Example 1(b) describes a process for making a liquid pre-culture of N. meningitidis, and Example 1(c) describes a perfusion culture using ATF initiated with a liquid pre-culture of N. meningitides as starting material.

Example 1 (a) Neisseria meningitidis B2486 Solid Pre-Culture

(21) A solid pre-culture was prepared using a frozen seed culture of Neisseria meningitidis serotype B strain H44/76 (MenB). The seed culturability was determined by plating serial dilutions of the seed, as approximately 510.sup.8 colony forming units per ml.

(22) The seed culture was thawed to room temperature and 100 l were used to inoculate a solid pre-culture (Agar plate) containing 15 ml of pre-culture agar medium derived from Frantz (J. Bact. 43(6): 757-761 (1942)).

(23) The inoculated Agar plate was then incubated at 36 C. (1 C.) in an environment regulated at 5% of CO.sub.2 and saturated in humidity, for 23 h (2 h). One 40.sup.th of the resuspended cells was used to inoculate the second pre-culture step (liquid pre-culture) (Example 1 (b)).

Example 1 (b) Neisseria meningitidis B2486 Liquid Pre-Culture

(24) The entire cell layer grown on the solid pre-culture (as described in Example 1(a)) was resuspended in 8 ml (1 mL) of liquid pre-culture medium derived from Frantz (J. Bact. 43(6): 757-761 (1942)), containing 20 g/L glucose, 1.5 g/L NaH.sub.2PO.sub.4.2H.sub.2O, 40 g/L soy peptone and 4 g/L sodium glutamate. A volume of 0.5 ml (0.1 ml) of this suspension was diluted in 4.5 ml of NaCl 0.9%. A volume of 2.0 ml of this dilution was used to inoculate a 2 L Erlenmeyer shake flask containing 400 ml of pre-culture medium.

(25) The inoculated flasks were then incubated at 36 C. (1 C.) and 200 rpm. The pre-cultures were stopped when the optical density at 650 nm (OD.sub.65onm) reached at least 3.0, (around 15 h2 h of incubation). The pre-culture was used to inoculate medium in a fermenter as soon as practical after the culture was stopped (Example 1 (c)).

Example 1(c) 20 L Scale ATF Perfused Fermentation of N, Meningitidis Using a Perfusion Rate Regulated by the Oxygen Demand

(26) A 20-liter fermenter (Biolafitte) was used. The vessel was equipped with an Alternating Tangential Flow 4 device (ATF4 from Refine Technology). The ATF was set up with a 0.2 m filter (GE HealthcareCFP-2-E-8SIP) and the whole system was sterilized by autoclaving prior to being assembled to the fermenter. 13 liters of fermentation medium (a similar medium as in Example 1(b), but containing only 5 g/L glucose) were aseptically transferred into the fermenter. (If necessary, the pH of the medium may at this point be readjusted to 7.0 with base addition.) One ml of undiluted irradiated antifoam (SAG 471) was added to the fermenter. The temperature (36 C.), head pressure (0.5 bar), aeration rate (20 liters sparged air per minute) and initial agitation (stirring) speed (100 rpm) were then set prior to inoculation. The level of dissolved oxygen in these conditions was defined as 100%.

(27) The aeration rate was maintained at a constant level during the fermentation. pH was controlled at 7.0 by an acid/base regulation (H.sub.3PO.sub.4 25%/NaOH 5N) throughout the fermentation.

(28) Inoculation was achieved by the addition of 800 ml of liquid pre-culture (prepared from two Erlenmeyer shake flasks, as described in Example 1(b)).

(29) The fermentation process consisted of 3 phases: a first phase (the pre-perfusion phase) where glucose is not limiting the growth rate and there is no perfusion, a second phase (perfusion phase 1) where glucose is not limiting, but perfusion has been initiated and a third phase (perfusion phase 2) wherein glucose is limiting the growth rate and perfusion is taking place.

(30) During the first phase (pre-perfusion), the temperature was maintained at 36 C., pH was maintained at 7.0, head pressure was maintained at 0.5 bar, and aeration was maintained at 20 liters per minute. The level of dissolved oxygen (DO) was set at 20%. The level of DO was regulated by increasing stirring when the DO fell below 20%. After 2.6 hours (when dissolved oxygen is regulated at its setpoint (20%) by stirring), perfusion was started, thus starting the second phase (perfusion phase 1).

(31) In perfusion phase 1, the head pressure was decreased from 0.5 to 0.3 bar.

(32) Temperature was maintained at 36 C., pH was maintained at 7.0, aeration was maintained at 20 liters per minute. A double head peristaltic pump was used for both feeding fresh growth medium (similar to in Example 1(b), but containing 10 g/L glucose) into the bioreactor and extracting filtrate from ATF at a similar flow rate of 3.5 L/h (perfusion rate), in order to minimize fermentation volume fluctuations due to perfusion. The ATF rate (the volume of media passing from the fermenter to the ATF device through the filter and vice versa per time unit) was set at 3.5 L/min (Pressurisation flow=Exhaust flow), pressure and exhaust delays were automatically adapted by the system to reach that value.

(33) After 7 hours, (as soon as glucose was becoming limiting for growth), the fermentation went into the third phase (perfusion phase 2) wherein the temperature was decreased to 31.5 C. and the perfusion rate was gradually decreased to 1.5 L/h. Then, the perfusion rate was automatically controlled by oxygen demand (pO2/substrate regulation), decreasing perfusion rate when dissolved oxygen dropped below about 20% and increasing it in the opposite case. At this point the agitation speed was maintained at approximately 700 rpm to ensure a constant and non-limiting oxygen supply. The ATF rate was then increased to 5.0 L/min. Head pressure was maintained at 0.3 bar, pH was maintained at 7.0, aeration was maintained at 20 liters per minute.

(34) During this third phase, temperature was regulated at 31.5 C. and the ATF rate was increased from 5.0 to 7.0 L/min in parallel of biomass increase. At the end of fermentation (24 h), cell paste was collected by centrifugation (5000g, 4 C. for 30 min), and stored at 20 C.

(35) At the end of fermentation (24 h), the following biomass levels were determined:

(36) TABLE-US-00001 TABLE 1 Final OD650 nm Dry Cell Weight Wet Cell Weight 75.7 (10 L) or 58.2* 57.7 g/L (10 L) or 270 g/L (10 L) or 207.7 g/l* 44.4 g/l* *Values calculated for starting medium volume of 13 L

Example 2 20 L Scale ATF Perfused Fermentation of N, Meningitides Using Pre-Determined Perfusion Rates

(37) Fermentation conditions were equivalent to those described in Example 1 except that:

(38) 1) Culture medium contained 10 g/L glucose.

(39) 2) After 3.3 h of culture (when dissolved oxygen was regulated at its set point (20%) by stirring), fermentation medium was added in a perfusion mode through the ATF device at a rate of 3.5 L/h; after 6.5 h of culture, the head pressure was decreased from 0.5 to 0.3 bar.

(40) 3) at 9.5 h of culture (when the OD.sub.650 nm reached approximately 20) when glucose became limiting for growth rate, perfusion rate was decreased from 3.5 to 2.0 L/h and maintained at this setpoint throughout the third fermentation phase (perfusion phase 2). Dissolved oxygen was also regulated by the agitation speed during the whole fermentation. At the end of fermentation (24 h), the following biomass levels were determined:

(41) TABLE-US-00002 TABLE 2 Final OD650 nm 54 (10 L)

(42) FIGS. 11 and 12 show respectively the batch and perfusion processes in the form of a flow sheet. The perfusion process is as described in Example 1(c).

(43) Table 3 shows a comparison of the results of different fermentation modes, all using N. meningitides.

(44) TABLE-US-00003 TABLE 3 Perfusion Batch mode Fed-batch mode mode Culture duration 6.5 h 8.5 h 24 h OD650 nm 13.9 16.6 54-58.2 Dry Cell Weiqht (DCW) 8.7 g/L 44.4 g/L Wet Cell Weiqht (WCW) 46.1 g/L 51 g/l 207.7 g/L

(45) As shown above, batch fermentation reached a maximal cell density of about 13.9.

(46) Further growth may in principle be limited by two factors: the consumption of substrates present in the medium and the possible accumulation of one or more inhibitory metabolites leading cells to growth limitation and, finally, lysis.

(47) Substrate limitation in batch mode cannot be avoided by simply increasing substrate concentrations in the medium, because an increased amount of substrate makes dissolved oxygen level impossible to maintain at a constant value (reach of a maximal agitation speed).

(48) An alternative would be fed-batch fermentation, which can feed cells slowly in order to control oxygen supply. In this system, the feeding rate is controlled by the oxygen demand and agitation speed is maintained at a constant level. However, as can be seen in Table 3, fed-batch results revealed the same growth limitation as observed for batch mode with a maximal optical density obtained around 16. Cell lysis was observed at this point forward. This observation indicated that substrate is not the main limiting factor, but that possible accumulation of one or more inhibitory metabolites during the culture might inhibit further growth.

(49) As shown above, the perfusion mode using the ATF device wherein cells were slowly supplied with substrate and, at the same time, spent medium was removed, gave a significant increase in yield.

Example 3 20 L Scale ATF Perfused Fermentation of S. Pneumoniae Using Pre-Determined Perfusion Rates

(50) A 20-liter fermenter (Biolafitte) was used. The vessel was equipped with an Alternating Tangential Flow 4 device (ATF4 from Refine Technology). The ATF was set up with a 0.2 m filter (GE HealthcareCFP-2-E-8SIP) and the whole system was sterilized by autoclaving prior to being assembled to the fermenter. A fermentation medium similar to that described in Hoeprich (1955) J Bacteriol 69(6): 682-688 was used with the modification that it contained 45 g/L rather than 12.5 g/L glucose and 200-400 mg/L Choline HCl). 10 liters of fermentation medium were aseptically transferred into the fermenter. (If necessary, the pH of the medium may at this point be readjusted to 7.2 with base addition.) The temperature (36 C.), head pressure (0.1 bar), aeration rate (2 liters air per minute in the headspace of the fermenter) and agitation (stirring) speed (100 rpm) were then set prior to inoculation. The aeration rate was maintained at a constant level during the fermentation as well as the stirring. pH was controlled at 7.2 by base addition (NaOH 5N) throughout the fermentation.

(51) Inoculation was achieved by the injection of 40 L of Streptococcus pneumoniae serotype 22F working seed suspension (the viability of the seed was estimated at 2.7 10.sup.9 colony forming units per ml) directly in the fermenter through a septum in the headplate.

(52) The fermentation process consisted of in 3 phases: a first phase (the pre-perfusion phase) where glucose is not limiting the growth rate and there is no perfusion, a second phase (perfusion phase) where glucose is not limiting, but perfusion has been initiated, and a third phase (post-perfusion phase) wherein the feeding is stopped but the ATF is still in action to extract permeate, and the culture is pursued in batch mode.

(53) During the first phase (pre-perfusion), the growth is initiated. When the optical density (650 nm) reached 2.85 (9 h40 of culture), the second phase (perfusion phase) was started.

(54) In the perfusion phase, all culture parameters were maintained constant. A double headed peristaltic pump was used for both feeding fresh growth medium (of the same composition as the initial medium) into the bioreactor, and extracting filtrate from ATF at a similar flow rate of 3 L/h (perfusion rate), in order to minimize fermentation volume fluctuations due to perfusion. Perfusion rate was maintained at 3 L/h for the first 45 min and then increased to 6 L/h. Alternating Tangential Flow (ATF) was set at a rate of 2.5 L/min for the first 95 min then at 10 L/h. During the perfusion phase, the rate of filtrate extraction progressively decreased while medium level in the fermenter increased (due to the constant feed rate). After 12 hours of culture, the feeding was stopped, while the permeate extraction was maintained (post-perfusion phase). The culture was continued, maintaining the process parameters constant until 13 h20 of culture.

(55) Culture and permeate were sampled periodically in order to evaluate the polysaccharide content and accumulation.

(56) FIG. 11 shows a fermentation profile with the process parameters monitored during 20 L-scale perfusion fermentation, FIG. 12 shows the accumulation of polysaccharide 22F in the different fractions (permeate and culture).

(57) At the end of fermentation (13 h20), the following biomass and Polysaccharide 22F (ELISA assay) levels were determined

(58) TABLE-US-00004 TABLE 4 Ratio with ATF/ Culture duration (hh:mm) With ATF W/O ATF without ATF Experiment SPC1521 SPC1447 Final volume in fermenter (L) 14 17.5 Fermentation duration (h) 13:20 13:35 Max OD (650) 12.1 7.8 1.55 Permeate weight (g) 11848 NA PS cone in fermenter at end 7057 2217 3.18 (mg/L) Total PS in fermenter (mg) 98798 38797 2.55 Total PS in permeate (mg) 9893 NA Overal Total PS (mg) 108691 38797 2.80 Overal volume (L) 25.85 17.5 1.48 Overal PS productivity (mg/L) 4205 2217 1.90