HK1150750B - Counter-pressure filtration of proteins - Google Patents

Description

Protein counter-pressure filtration

Cross Reference to Related Applications

This application is entitled to priority from U.S. provisional patent application No.61/017,418, filed on 12/28/2007 as 35u.s.c. § 119(e), which is incorporated herein by reference.

Technical Field

The present disclosure relates generally to filtration methods for protein purification. More particularly, the present application relates to low shear, counter pressure sterile filtration of proteins such as proteins that are susceptible to shear damage when transported in a liquid (e.g., proteins sensitive to shear, proteins of the coagulation cascade).

Background

The purified protein mixture can be administered to a patient for a variety of treatments. The purified protein mixture prepared for infusion into a patient must be sterilized prior to use. Suitable sterilization processes for some proteins include membrane filtration of the purified protein mixture. When the protein is able to pass through the membrane, the filtration membrane may be sized to retain (i.e., remove from the protein mixture) particulates, microorganisms, and some viruses.

However, some proteins cannot be efficiently recovered as purified, sterilized proteins by using conventional methods such as membrane filtration. This effect is most pronounced when attempting to filter proteins that are part of the shear sensitive and/or coagulation cascade. An example of this protein is von willebrand factor (vWF), which circulates in plasma, forms complexes with factor VIII and promotes the regulation of biological coagulation activity. In particular, vWF proteins are sensitive to shear forces generated by the velocity gradient of the transport liquid medium, particularly when vWF proteins cross or near the filter membrane (i.e., resulting in particularly large velocity gradients where the flow gathers and tortuous flow path near the filter pores). Thus, when the filtration device is operated at a pressure sufficiently great to ideally produce the desired process flow rate, the increased flow rate (and concomitant increase in shear forces) tends to reduce the process yield, such as by damaging or destroying the protein, and/or by reducing the filtration rate over time.

It would therefore be desirable to develop a method of filtering a purified mixture of vWF that is substantially non-damaging to vWF and that still allows a suitably high process throughput (i.e., filtration rate) over time. In addition, it would be desirable to develop filtration methods that can be applied generally to any protein so that the general protein can be filtered (e.g., sterile filtered) at a high efficiency rate without substantial damage/loss of the protein.

Disclosure of Invention

The disclosed methods can be used to filter proteins in a liquid mixture in a batch or continuous manner without substantially damaging or otherwise limiting the recovery of the proteins in the filtered filtrate. The present method typically applies opposing pressures to the filtrate in order to accurately reduce and control the pressure differential across the filter. The disclosed method has the advantages that: a relatively high filtration flow rate can be achieved at a relatively low pressure differential, in contrast to a high pressure differential that actually reduces the filtration flow rate of the protein liquid mixture. Still further, the present process allows recovery of substantially all of the protein originally present in the liquid mixture.

More specifically, the present disclosure provides a method of filtering a liquid protein mixture. According to one embodiment, the method comprises: at a first pressure P₁Providing a liquid mixture at a second pressure P₂The lower liquid mixture passes through the filter layer to form a filtrate, and a counter pressure is applied to the filtrate such that P₁-P₂Is not greater than 300 mbar. In another embodiment, the method comprises the steps of: at a first pressure P₁Providing a liquid mixture in a secondPressure P₂Passing the liquid mixture through a filter to form a filtrate and will be sufficient to produce at least about 300 g/min-m²The average filtrate flow rate over the filter surface area is back-pressed against the applied filtrate. The liquid mixture comprises a carrier liquid, at a first concentration C relative to the carrier liquid₁Proteins, and dispersed contaminants. The filtrate comprises a carrier liquid and is in a second concentration C relative to the carrier liquid₂The following proteins. The filter is sized to remove at least a portion of the dispersed contaminants from the liquid mixture.

In another embodiment, the method is capable of filtering an aqueous protein mixture and comprises the steps of: at a first pressure P₁Providing an aqueous mixture at a second pressure P₂Passing the aqueous mixture through a porous membrane filter to form a filtrate and applying a counter pressure to the filtrate such that a pressure difference P is created₁-P₂Not greater than about 90 mbar. The aqueous mixture comprises water and water in a first concentration C₁Lower vWF. The filtrate comprises water and has a second concentration C relative to the water₂Lower vWF. The porous filter membrane filter includes pores ranging in size from about 0.1 μm to about 0.5 μm.

In any of the above embodiments, the protein is preferably a shear sensitive protein and/or a protein of the coagulation cascade. Further, the protein is recovered in the filtrate, preferably at a recovery ratio C₂/C₁At least about 0.95, and more preferably at least about 0.99. In addition, the pressure difference P₁-P₂Preferably no greater than about 90mbar and the first pressure P1 is preferably at least about 200mbar gauge. Preferred embodiments of the above method include wherein the carrier liquid is water and the dispersed contaminants include microorganisms. The protein may include von willebrand factor, factor VIII, factor XIII, and mixtures thereof. The filter preferably comprises a porous filter membrane filter having a pore size in the range of from about 0.1 μm to about 0.5 μm, more preferably a pore size of about 0.2 μm or about 0.22 μm. Preferably, the filtrate product is substantially free of dispersed contaminants.

Filtration of the protein under the application of back pressure allows the protein to be recovered at a relatively high concentration, relatively high filtrate flow rate, whereas it is not possible to achieve a substantially constant filtrate flow rate without back pressure. This is in contrast to the general application of filtration theory, at least in terms of filtrate flow rate, as theory predicts that filtrate flow rate will increase (i.e., in the absence of back pressure) with increasing pressure differential across the filter.

Further aspects and advantages will be apparent to those skilled in the art from the following detailed description, taken in conjunction with the accompanying drawings. The compositions, films and kits described herein are susceptible of embodiment in various forms, and specific embodiments within the following description are understood to illustrate the invention and are not intended to limit the invention to the specific embodiments described herein.

Drawings

Two figures are attached here to facilitate understanding of the present disclosure.

FIG. 1 is a cross-section of an axisymmetric cartridge filter for backpressure filtration of proteins.

Fig. 2 is a comparison of filtration rate data obtained using a conventional filtration method and a counter-pressure filtration method.

Detailed Description

The methods described herein are generally applicable to the filtration purification of proteins in liquid mixtures in a manner that does not substantially impair or otherwise limit the recovery of the proteins during filtration. In addition to the protein, the liquid mixture also includes a carrier liquid and dispersed contaminants. The carrier liquid is a suspension medium for the protein and is generally not limited. Preferably, the carrier liquid is water. Similarly, dispersed contaminants are also not particularly limited and may include any dispersed solid matter that is an undesirable component of the final purified protein filtrate. In the case of a sterile filtration operation, the dispersed contaminants typically include any of a variety of microorganisms (i.e., bacteria) that may be present in the liquid mixture.

The disclosed methods are particularly preferably applied to those proteins that are shear sensitive, those proteins that are part of the human coagulation cascade, or both (i.e., some suitable proteins such as vWF, which can be classified as both shear sensitive and coagulation cascade proteins).

Shear sensitive proteins suitable for purification using the disclosed counter-pressure filtration methods include those proteins that are susceptible to damage, disruption, loss of activity, and/or reduced filtration rates when transported as a suspension in a carrier liquid that is resistant to large shear forces (i.e., relatively large velocity gradients). Generally, a shear sensitive protein is one that exhibits an inverse relationship between filtration rate and applied shear (or applied pressure) above the critical applied shear. At the point of critical applied shear (or applied pressure), the direct and inverse relationship of filtration rate and applied shear (applied pressure) switches (i.e., switching occurs when the filtration rate is at a maximum), and the critical applied shear may be different for different shear-sensitive proteins. For example, the critical applied shear may be at least about 2000S^-1(or at least about 4000S)^-1) And not greater than about 8000S^-1(or not more than 12000S)^-1). However, the qualitative behavior of different shear sensitive proteins is expected to be similar. Shear sensitive proteins include vWF, although the disclosed methods are not particularly limited thereto. vWF exists in plasma in a series of oligo/multimeric forms, the molecular weight range of these forms being from about 1000kDa (kilodalton) to about 20000kDa based on a dimer of 520kDa, the disclosed method not necessarily being limited to a particular molecular weight range.

The disclosed methods are also generally applicable to the purification of proteins of the human coagulation cascade (i.e., coagulation factors). For example, factor II (molecular weight approximately 37kDa), VII (approximately 50kDa), VIII: c (about 260kDa), IX (about 55kDa to about 70kDa), X (about 100kDa), XIII (about 350kDa), vWF (as described above), and combinations thereof, some of which are also shear sensitive, can be efficiently filtered and recovered by using back pressure.

Some particularly preferred purified proteins include single proteins such as factor XIII or vWF, and multi-protein compositions such as factor VIII: C/vWF complexes. Protein compositions that are not particularly shear sensitive are preferred, but are still successfully filtered through the use of back pressure, are FEIBA VH mixtures ("(steam heated) factor octa-inhibitor bypass active factor", from baxter, Deerfield, illinois) and are useful in controlling spontaneous bleeding and/or treating hemophilia a/B patients. FEIBA VH mixtures include factors II, IX and X (mainly unactivated), factor VIII (mainly activated) and factor VIII: c (approximately 1 to 6 units/ml). FEIBA VH mixtures benefit from the use of back pressure because when back pressure is applied, the protein mixture can be filtered at a relatively high filtration rate with little or no loss of protein activity.

The liquid mixture containing the protein and dispersed contaminants is then purified through a filter. Filters suitable for use in accordance with the disclosed methods are not particularly limited and may include surface filters such as dead-end filters (i.e., liquid is filtered vertically through the filter surface) and cross-flow filters (i.e., liquid is filtered moving parallel to the filter surface). See, for example, Kirk-Othmer, encyclopedia of chemical technology, volume 10, pages 788 and 853 ("filtration") (4 th edition 1993). The filters are also not particularly limited with respect to their size of classification (i.e., dispersed material above this size is retained on the filter and dispersed material below this size passes into the filtrate). Once the filtration fraction size is selected for a particular application (i.e., the dispersed material to be retained relative to the dispersed material entering the filtrate), the filter should be operated to take into account the shear sensitivity of the amount of shear generated by the carrier liquid flowing through the filter relative to the particular protein being filtered.

Preferred filter media are porous filter membranes, typically of different sizes (i.e., filter surface area; e.g., ranging from 0.001 m)²To about 5m²) And configurations (i.e., filter disc, cartridge). The porous filter membrane may be formed of: such as cellulose nitrate, cellulose acetate, vinyl polymers, polyamides, fluorocarbons, and polyethersulfones. Porous filtration membranes typically comprise pores of highly uniform size, depending on the size of the dispersed contaminants to be removed from the liquid mixture. For example, where it is desired to remove microorganisms during the sterile filtration process (while allowing proteins to pass through the filter into the filtrate), the pore size is preferably in the range of about 0.1 μm to about 0.5 μm, or about 0.15 μm to about 0.25 μm, for example about 0.2 μm or about 0.22 μm. Suitable porous membrane filters may also include 0.2 μm/0.22 μm filters and coarse prefilters (e.g., about 0.45 μm) to increase flux and limit cake build-up on the 0.2 μm/0.22 μm filter surface. Examples of suitable commercial porous filter membrane filters for sterile filtration of protein-containing liquid mixtures include: SARTOBRAN P0.2 μm cellulose acetate membrane (prefilter membrane comprising 0.2 μm pores and having 0.45 μm pores, available from Sartorius AG,germany) and SUPOR EKV 0.2 μm polyethersulfone membranes (available from Pall corporation, East Hills, new york).

Fig. 1 shows the flow of a liquid mixture through a filter 100, such as a cartridge filter having a porous filter membrane. As shown, the liquid mixture enters the inlet chamber 120 of the sealed filter 100 through the inlet 110. The liquid in the inlet chamber 120 is pressurized to have a first pressure P₁Typically above ambient pressure (i.e., about 1bar absolute atmospheric pressure). First pressure P in the inlet chamber 120₁The carrier liquid and proteins in the liquid mixture are driven through the porous filter membrane 130. The material passing through the filter membrane 130 forms a filtrate, and the filtrate in the filtrate chamber 140 includes a carrier liquid and a protein. Filter elementThe liquid in the liquid chamber 140 is also pressurized, having a pressure greater than the first pressure P₁Small second pressure P₂But is also typically above ambient pressure. The purified filtrate then exits the filter through outlet 150. Flow rate through the outlet 150 (and second pressure P₂) Is regulated by a back pressure regulator 160.

The pore size of the filter membrane 130 is sized to remove at least a portion of the dispersed contaminants contained in the liquid mixture, with the removed dispersed contaminants being retained on the inlet side of the filter membrane 130 (i.e., in the inlet chamber 120). Preferably, filter membrane 130 is sized to remove substantially all of the dispersed contaminants originally present in the inlet liquid mixture; thus, the filtrate is substantially free (or free) of dispersed contaminants. In particular, the filtrate should not contain an amount of dispersed contaminants that would adversely affect the use of the filtered and purified protein mixture as a therapeutic composition. For example, where the dispersed contaminants include microorganisms, filter membrane 130 should be capable of removing at least about 10⁷Individual colony forming unit per cm²The filtration area.

First pressure P of inlet liquid mixture₁May be applied by any suitable source such as gravity, compressed gas (e.g. compressed air) or a pump (e.g. low shear pump). Preferably, the first pressure P₁At least about 200mbar gauge, or in the range of about 200mbar to about 1000mbar gauge (i.e., about 1bar gauge), such as 300mbar gauge. The back pressure (or back pressure) may also be applied by any suitable source, such as a conventional valve (back pressure regulator 160 shown in fig. 1), to obtain a second pressure P of the outlet filtrate₂. Alternatively, back pressure may also be applied by flow accumulation, blockage, etc. along the path of the outlet 150. As the first and second pressures P₁And P₂As a result of which the applied counter-pressure generates a (positive) pressure difference P₁-P₂The carrier liquid and protein are driven through the filter into the filtrate. The pressure differential is low and the appropriate maximum pressure differential depends on various factors, such as the particular protein being filtered and the particular filtration usedA device. In particular, the pressure difference is preferably not greater than about 300mbar, 200mbar, 150mbar or 120mbar, more preferably not greater than about 90mbar or 50mbar, even more preferably not greater than about 20mbar or 10mbar, such as not greater than 5mbar or not greater than 3 mbar.

At such pressure differentials, the flow rate through the filter is sufficiently low to create a low shear environment that is substantially non-damaging, destructive, or otherwise diminished in activity of the protein. The low shear environment further does not substantially limit the yield of the filtration process.

In particular, it can be seen that a relatively high and substantially constant filtration rate can be achieved with a low pressure differential. In particular, low pressure differentials may be used to achieve at least about 300 g/min-m²Average filtration flow rate of, for example, about 300 g/min m²To about 1000 g/min m²Or about 400 g/min m²To about 800 g/min m²Where the units are the mass of filtrate (i.e., grams of carrier liquid and protein bound) obtained per unit time (i.e., minutes) and normalized to the surface area per unit of filter (i.e., m)²). The low pressure differential resulting from the application of back pressure may also produce a substantially constant filtration rate, especially after the transient start-up is complete, thus increasing the net capacity of the filter. Higher pressure differences (e.g. over about 100mbar) tend to reduce the filtration rate compared to the filtration rate seen at low pressure differences (which is contrary to general filtration theory, which means that the filtration rate is proportional to the pressure difference across the filter). Thus, conventional filtration methods, applying a pressure differential of at least about 100mbar to about 300mbar, cannot achieve the average flow rates seen in the methods of the present disclosure.

Further, it can be seen that a low pressure differential can result in a relatively high recovery of protein. In particular, substantially all of the protein originally present in the liquid mixture is preferably recovered in the filtrate. For example, when the first concentration C is₁Refers to the concentration of protein in the initial liquid mixture relative to the carrier liquid, the second concentration C₂Refers to the protein concentration in the final filtrate relative to the carrier liquid, and the recovery rate C of the disclosed filtration process₂/C₁Preferably at least about 0.95 and more preferably at least about 0.99.

The disclosed back pressure filtration method is useful when applied to proteins because it provides precise control of the filtration flow rate. The pressure source used to drive the liquid mixture through the filter typically provides too high a pressure, resulting in high shear rates, which in turn leads to filter plugging and protein damage. It is often difficult, if not impossible, to adjust a single pressure source relative to ambient pressure so that the resulting pressure differential is sufficiently low and constant to result in an average flow through the filter that is sufficiently low to avoid damage to the proteins and is substantially constant. However, by applying a counter-pressure to the filtrate side of the treatment liquid, a relatively high pressure source (such as the first pressure P)₁) Can be precisely counter-balanced (i.e., the second pressure P is applied)₂) To obtain a low and substantially constant pressure differential (i.e., P)₁-P₂)。

Examples

The following examples are intended to illustrate the invention and are not intended to limit the scope of the invention.

Examples 1 to 8

Aqueous mixtures of the shear sensitive, coagulation cascade protein vWF are sterile filtered to determine the effect of pressure (and back pressure) and other process variables (such as filter size and type) on filtration effectiveness, such as based on the activity of vWF recovered in the filtrate, the average flow rate of the filtrate, and the ability to scale up the process.

An aqueous mixture of recombinant vWF ("rVWF: Rco") having an activity of 125IU/ml and a concentration of 1159 μ g/ml (bicinchoninic acid ("BCA") experiment) was prepared for each of examples 1-8 below. Low molecular weight salts are added to adjust the pH of the mixture to 7.3 and the osmolality to a value of about 400mOsmol/l (osmolality/l), respectively. Because examples 1-8 are intended to isolate hydrodynamic effects (such as shear) on proteins in the vicinity of the filter, no other dispersed contaminants or materials (such as bacteria, other microorganisms, or other proteins) are added to the mixture. The ingredients of the aqueous mixture are summarized in table 1.

TABLE 1 liquid mixture composition of examples 1-8

The aqueous liquid mixture was then tested in the following manner. At least about 500ml of the test amount of the mixture was added to a pressure-resistant stainless steel vessel. The outlet of the vessel was connected to the inlet of a sterilizing filter (steam sterilization) via a first section of silicone tubing. A second section of silicone tubing approximately 10cm long was connected to the outlet of the sterile filtration chamber. Filtrate effluent from the sterile filtration was collected in a beaker placed on a balance to monitor filtration rate. The scale was interfaced with a computer to record the total amount of filtrate collected at intervals as a function of time.

In examples 1-7, a SARTOBRAN P0.2 μm cellulose acetate membrane filter (including a 0.45 μm prefilter) was used. In example 8, a SUPOR EKV 0.2 μm polyethersulfone membrane filter was used. Examples 1-3 used disc membrane filters, while examples 4-8 used cartridge membrane filters. The filter surface area of each filter is provided in table 2.

For examples 1-8, the test amount of aqueous mixture was filtered at constant pressure. In example 1, the applied pressure of the aqueous solution entering the sterilizing filter was about 100mbar gauge (i.e., about 1.1bar absolute atmospheric pressure), depending on the height of the column entering the sterilizing filter from the stainless steel container. In examples 2-8, clean compressed air was used to provide a pressure of the aqueous solution into the sterilizing filter ranging from 150mbar to 300mbar as shown in table 2. In examples 1-4, no counter pressure filtration was performed (i.e., the outlet of the sterile filter was vented when the filtrate was poured into the collection flask), so the pressure differential driving the filtration was essentially the pressure of the aqueous solution entering the sterile filter. For examples 5-8, back pressure was applied to the filtrate by attaching and clamping a clamp to the outlet tube of the sterilizing filter. The pressure differences for examples 5-8 (shown in table 2) were extrapolated from the average filtrate flow rate measured during filtration and the pressure differential-flow rate characteristics of commercial filters known to be used.

Generally, in each example, the mixture was filtered soon after the test amount of aqueous mixture was prepared. However, in example 5, the mixture was filtered after a delay of 8 hours in order to check for any potential effect of storage on the disclosed method.

Table 2 summarizes the experimental parameters for each of examples 1-8.

TABLE 2 filtration test parameters for examples 1-8

The results of the filtration test are summarized in table 3. In table 3, "filtrate amount" refers to the filtrate quality (i.e., including water and any vWF recovered) obtained during a single experiment. For examples 1-4, filtration experiments were performed until the filter was clogged and no substantial amount of filtrate was obtained. For examples 5-8, filtration experiments were performed until hundreds of grams of filtrate were obtained and the filter was not clogged at the end of the experiment and still further filtration was possible. The "filtration capacity" input in table 3 refers to the amount of filtrate per unit area of filter surface that is normalized. The symbol ">" in examples 5-8 means that the filtration capacity is a lower estimate of the actual capacity, since the filter was not clogged during the experiment. The "average flow rate" entry in table 3 refers to the amount of filtrate averaged and normalized per unit area of filter surface during the test.

TABLE 3 filtration results of examples 1-8

As is evident from table 3, the backpressure filtration of the protein increased the filtration capacity and flow rate. Even at suitably low applied pressures (ranging from 100mbar to 300mbar) for the vWF and the filter used (the filter cartridge used here can withstand the highest pressure ranging from about 2bar to about 5.5bar), examples 1-4 show that the filtration capacity is low and decreases dramatically with increasing pressure. In contrast, examples 5-8 exhibited much greater filtration capacity and flow rate when back pressure was applied to the filtrate to reduce the pressure differential. The observed phenomenon is unexpected, since filters are generally characterized by a direct relationship between the filtration flow rate and the pressure difference:

formula (1)

In equation (1), Q is the filtration flow rate (volume or mass per unit time), a is the surface area of the filter, Δ p is the pressure differential across the filter, μ is the viscosity of the fluid being filtered, and R is the experimental resistance of the filter media. The similarity between the results of examples 5-8 further demonstrates that the advantages of counter-pressure filtration can be obtained by using various filter media and/or filter sizes.

The time-dependent filtration data (i.e., the filtrate collected as a function of time) for examples 4 and 6 is shown in fig. 2. As is evident from fig. 2, the high pressure differential of example 4 results in a relatively high filtration rate of the first 10-15 minutes of the filtration test. However, the high pressure variance in example 4 also resulted in more acute clogging of the filter. In contrast, the use of back pressure in example 6 reduced the pressure differential and the initial filtration rate. Specifically, during the initial brief period of about 10 minutes to about 15 minutes, the low pressure differential creates a low shear filtration environment that prevents the filter from becoming clogged with vWF, thus increasing the filtration capacity and filtration flow rate throughout the filter's life, and further resulting in a substantially constant filtration flow rate.

As summarized in table 4, the filtrate of each of examples 1-3 and 5 was also analyzed to determine the level of recombinant vWF in the filtrate, and thereby determine whether the filtration process adversely affected the concentration of recombinant vWF present in the initial aqueous mixture. In the absence of counter pressure, examples 1-3 show that even moderately low pressure differences during filtration can damage or destroy proteins, with losses during filtration of up to about half of the recombinant vWF at pressure differences of 200mbar and 300 mbar. This filtration loss can be avoided by using back pressure, as shown in example 5, in example 5 there is no measurable decrease in rVWF: Rco concentration and only a 4% decrease in the measured protein (BCA) content. Thus, substantially all of the protein originally present in the aqueous mixture may be recovered in the filtrate.

TABLE 4 recovery of vWF from the filtrate

Examples 9 to 20

With an aqueous solution containing factor VIII in a pre-filter with a SARTOCLEANIs tested on the filterAn experiment to determine the effect of back pressure on filtration effectiveness, e.g., based on the desired filter surface area. In examples 9-13, no counter-pressure filtration was performed; the initial applied pressure is from 100mbar to 500 mbar. In examples 14-16, the pressure difference was 200mbar, and in examples 17-20, the pressure difference was reduced to less than 150 mbar. The surface area of each filter was 1.2m². Table 5 contains the factor VIII activity before and after filtration in each experiment.

TABLE 5 recovery of FVIII from the filtrate

Previous data demonstrate that much less filtration area is necessary when back pressure filtration is used for the same amount of active substance. The back pressure filtration stabilizes the filtration and improves the average activity yield when the pressure difference is optimized.

Examples 21 to 25

Solutions containing factor XIII inN66 polyamide filters were tested in order to examine the effect of back pressure on the filtration effectiveness, e.g. based on the activity and protein concentration of factor XIII recovered in the filtrate. The area of each filter was 0.82m². In examples 21-23, no counter-pressure filtration was performed; the applied pressure for these examples was 600 mbar. In examples 24 to 26, the pressure difference was approximately 100 mbar. Table 6 contains the protein concentration and activity before and after filtration and the yield for each experiment.

TABLE 6 recovery of FXIII from the filtrate

As shown in table 6, by using counter pressure, the activity yield and protein yield during filtration were considerably improved.

The foregoing description is given for clearness of understanding and no unnecessary limitations should be understood therefrom, as modifications within the scope of the invention may be apparent to those skilled in the art.

Throughout the specification, where a composition is described as including ingredients or materials, it is contemplated that the composition can also consist essentially of, or consist of, the recited ingredients or materials, unless otherwise stated.

The operations of the method disclosed herein, as well as the individual steps thereof, may be performed manually and/or with the aid of electronic equipment. Although the process has been described in terms of specific embodiments, those skilled in the art will readily appreciate that other methods associated with the present method may be used. For example, the order of various steps may be changed without departing from the scope and spirit of the method, unless otherwise indicated. Furthermore, some of the individual steps may be combined, omitted, or further subdivided into further steps.

Claims (30)

1. A method of filtering a liquid protein mixture, the method comprising:

at a first pressure (P)₁) Providing a liquid mixture comprising a carrier liquid, a first concentration (C) relative to said carrier liquid₁) Proteins, and dispersed contaminants;

at a second pressure (P)₂) Passing the liquid mixture through a filter to form a filtrate comprising a carrier liquid and a second concentration (C) relative to the carrier liquid₂) The protein of (b), wherein the peptide hasThe filter is sized to remove at least a portion of the dispersed contaminants from the liquid mixture; and is

Applying a counter-pressure to the filtrate in order to ensure a pressure difference (P) between the first pressure and the second pressure₁-P₂) Greater than 0mbar and not greater than 300 mbar.

2. The method of claim 1, wherein the protein comprises a shear sensitive protein.

3. The method of claim 1, wherein the protein comprises a protein of the coagulation cascade.

4. The method of claim 1, wherein the pressure differential is no greater than 90 mbar.

5. The method of claim 1, wherein at least 95% of the protein present in the liquid mixture is recovered in the filtrate.

6. The method of claim 5, wherein at least 99% of the protein present in the liquid mixture is recovered in the filtrate.

7. The method of claim 1, wherein the back pressure is sufficient to generate at least 300 g/min. m to the surface area of the filter²Average filtrate flow rate.

8. Method according to claim 1, characterized in that said first pressure (P)₁) At least 200mbar gauge.

9. The method of claim 1, wherein the carrier liquid comprises water.

10. The method of claim 1, wherein the protein is selected from the group consisting of von willebrand factor (vWF), factor VIII, factor XIII, and mixtures thereof.

11. The method of claim 1, wherein the dispersed contaminants comprise microorganisms.

12. The method of claim 1, wherein the filtrate is substantially free of dispersed contaminants.

13. The method of claim 1, wherein the filter comprises a porous filtration membrane comprising pore sizes from 0.1 μ ι η to 0.5 μ ι η.

14. The method of claim 13, wherein the pore size is 0.2 μ ι η or 0.22 μ ι η.

15. A method of filtering a liquid protein mixture, the method comprising:

at a first pressure (P)₁) Providing a liquid mixture comprising a carrier liquid, a first concentration (C) relative to said carrier liquid₁) Proteins, and dispersed contaminants;

at a second pressure (P)₂) Passing the liquid mixture through a filter to form a filtrate comprising a carrier liquid and a second concentration (C) relative to the carrier liquid₂) Wherein the filter is sized to remove at least a portion of the dispersed contaminants from the liquid mixture; and the number of the first and second electrodes,

applying a counter-pressure to the filtrate in order to ensure a pressure difference (P) between the first pressure and the second pressure₁-P₂) Greater than 0mbar and giving a filter surface area of at least 300 g/min m²Average filtrate flow rate.

16. The method of claim 15, wherein the protein comprises a shear sensitive protein.

17. The method of claim 15, wherein the protein comprises a protein of the coagulation cascade.

18. The method of claim 15, wherein at least 95% of the protein present in the liquid mixture is recovered in the filtrate.

19. The method of claim 18, wherein at least 99% of the protein present in the liquid mixture is recovered in the filtrate.

20. The method of claim 15, wherein the carrier liquid comprises water.

21. The method of claim 15, wherein the protein is selected from the group consisting of von willebrand factor (vWF), factor VIII, factor XIII, and mixtures thereof.

22. The method of claim 15, wherein the dispersed contaminants comprise microorganisms.

23. The method of claim 15, wherein the filtrate is substantially free of dispersed contaminants.

24. The method of claim 15, wherein the filter comprises a porous filtration membrane comprising pore sizes from 0.1 μ ι η to 0.5 μ ι η.

25. The method of claim 24, wherein the pore size is 0.2 μ ι η or 0.22 μ ι η.

26. A method of filtering an aqueous protein solution mixture, the method comprising:

at a first pressure (P)₁) Providing an aqueous mixture comprising water and a first concentration (C) relative to water₁) Von willebrand factor (vWF) of von willebrand factor (vWF);

at a second pressure (P)₂) Passing the aqueous mixture through a porous membrane filter to form a filtrate comprising water and a second concentration (C) relative to water₂) (ii) vWF, wherein the porous filter membrane filter comprises pores having a size of from 0.1 μm to 0.5 μm; and the number of the first and second electrodes,

applying a counter-pressure to the filtrate in order to ensure a pressure difference (P) between the first pressure and the second pressure₁-P₂) Greater than 0mbar and not greater than 90 mbar.

27. The method of claim 26, wherein at least 95% of the vWF present in the aqueous mixture is recovered in the filtrate.

28. The method of claim 27, wherein at least 99% of the vWF present in the aqueous mixture is recovered in the filtrate.

29. The method of claim 26, wherein the pore size is 0.2 μ ι η or 0.22 μ ι η.

30. The method of claim 26, wherein the aqueous mixture further comprises a microbial population, wherein at least a portion of the population is removed from the aqueous mixture by a porous membrane filter.