METHOD FOR DETERMINING THE LOGARITHMIC REDUCTION VALUE LRV OF A SIZE EXCLUSION FILTER

Abstract

The present invention relates to a method for determining the logarithmic reduction value LRV of a size-exclusion filter for a particle of a process solution, which particle is to be clarified, the size-exclusion filter being protected from a blocking adsorbing species present in the process solution by a process adsorber which is connected upstream in series.

Claims

1. A method for determining the logarithmic reduction value LRV.sub.Size-exclusion filter of a size-exclusion filter G for a particle of a process solution, which particle is to be clarified, the size-exclusion filter G being protected from a blocking adsorbing species present in the process solution by a process adsorber P connected upstream in series and the particle to be clarified being retained by the process adsorber P with an LRV.sub.Process adsorber of 0.5 or more, comprising the steps of: (a) providing a test system comprising a size-exclusion filter G, the LRV.sub.Size-exclusion filter of which is to be determined, and a test adsorber T which is connected upstream of the size-exclusion filter G in series and which consists of a similar material to the process adsorber P and by means of which the particle to be clarified is retained with a known LRV.sub.Test adsorber of 2 or less, where LRV.sub.Test adsorber<LRV.sub.Process adsorber; (b) determining LRV.sub.Test system of the test system for the particle to be clarified; and (c) calculating LRV.sub.Size-exclusion filter by subtracting the LRV.sub.Test adsorber from LRV.sub.Test system.

2. The method as claimed in claim 1, wherein the test adsorber T has an LRV.sub.Test adsorber of 1 or less.

3. The method as claimed in claim 1, wherein the process adsorber P has a specific adsorption capacity k.sub.1, is present on a scale M.sub.1 and has a total adsorption capacity of k.sub.1M.sub.1, wherein the test adsorber T has a specific adsorption capacity k.sub.2, is present on a scale M.sub.2 and has a total adsorption capacity of k.sub.2M.sub.2, where $0.1<\frac{k_1.Math.M_1}{k_2.Math.M_2}<10.$

4. The method as claimed in claim 3, where $0.5<\frac{k_1.Math.M_1}{k_2.Math.M_2}<5.$

5. The method as claimed in claim 3, where $0.9<\frac{k_1.Math.M_1}{k_2.Math.M_2}<1.1.$

6. The method as claimed in claim 3, wherein the total adsorption capacities of the process adsorber P and of the test adsorber T are determined by gas sorption measurement, by determination of breakthrough curves, by binding experiments with model proteins or by static incubation.

7. The method as claimed in claim 1, wherein the process adsorber P and the test adsorber T are present as mechanically integral shaped bodies.

8. The method as claimed in claim 7, wherein the process adsorber (P) and the test adsorber T are present in the form of membranes.

9. The method as claimed in claim 8, wherein the blocking adsorbing species is a blocking adsorbing species selected from biopolymers, biopolymer aggregates, biological particles, viral vectors, viruses and microorganisms.

Description

[0057] FIG. 1 shows a production-scale filtration system comprising a process adsorber P and a size-exclusion filter G.

[0058] FIG. 2 shows a production-scale filtration system (left) and a test system (right).

[0059] FIG. 3 shows the result of Example 1 for gold nanoparticles.

[0060] FIG. 4 shows the result of Example 1 for latex beads.

[0061] FIG. 5 and FIG. 6 show the results of the filtrations from Example 3.

[0062] The present invention will be more particularly elucidated on the basis of the following nonrestrictive examples.

EXAMPLES

Example 1: Determination of the Size-Exclusion Limit of Adsorbers

[0063] In this exemplary embodiment, microporous membranes composed of polyamide will be used as adsorbers. In said embodiment, 25006 and 25058 are two membranes which have been produced from the same basic substances (base material), but have different size-exclusion properties: 25058 is a process adsorber in the context of this invention, 25006 is a test adsorber in the context of this invention. The aforementioned materials have the following properties:

25006: polyamide adsorber, nominal pore size 0.45 m,
25058: polyamide adsorber, nominal pore size 0.10 m.

[0064] The membranes will be used as filter disks. Fluorescent latex beads will be used as sample particles. The size of the beads will be determined using dynamic light scattering in backscattering mode from the z-average. Determination will be carried out by using zeta sizer nano from Malvern.

[0065] Thereafter, the various adsorbers will be used for the filtration of a) gold nanoparticles and b) for the filtration of fluorescent latex beads in order to determine LRV.sub.Test adsorber and LRV.sub.Process adsorber.

[0066] Furthermore, the retention LRV.sub.Test system of these particles in the test system composed of test adsorber T and size-exclusion filter G will be determined, as described in the method according to the invention in step (b).

[0067] For comparison, the retention of these particles in the process system composed of the process adsorber P and the size-exclusion filter G will also be ascertained. The ascertainment of LRV.sub.Size-exclusion filter from said process system in the following step (c) is not in accordance with the invention, but prior art. In this exemplary embodiment, disadvantages of this hitherto customary type of ascertainment will be shown.

[0068] In step (c) of the method according to the invention, the retention of the size-exclusion filter will be determined from the difference LRV.sub.Test systemLRV.sub.Test adsorber.

[0069] In the case of the noninventive calculation according to the prior art, i.e., according to LRV.sub.Size-exclusion filter=LRV.sub.Process systemLRV.sub.Process adsorber, disadvantages arise, as will be described later on in the description of the results of this exemplary embodiment.

[0070] Description of the determination of particle size and of filtration:

[0071] In addition to viruses, nanoparticles are a particle type as to be retained by the size-exclusion filter. In this exemplary embodiment, gold nanoparticles and fluorescently labeled latex particles are used. Polyamide membrane layers (PA membrane layers) are used as adsorbers P and T; the size-exclusion filter is, in this case, a polyethersulfone membrane having a nominal pore size of 20 nm.

[0072] Use is made of particles having a 50 nm (gold/Nanopartz; CAS: A11-50-CIT-100; lot: E2451C) and 200 nm (fluorescent latex beads/Thermo Fisher; Fluoro-Max; CAS: G100) nominal size as per manufacturer data. The size of the particles serves as a model for viruses using PA adsorbers of differing pore size.

[0073] To verify the monodispersity and size of the particles, it is ascertained using dynamic light scattering.

[0074] Determination of the size of the particles on the basis of DLS (dynamic light scattering):

[0075] Description of the method: [0076] DLS settings: [0077] Material: latex, or gold [0078] Dispersant: water at 25 C., holding temp. 180 s [0079] Cell type: ZEN0040 (small cuvette) [0080] Measurement: 173 backscatter [0081] Data processing: general purpose (normal resolution)

[0082] Triplicate determination with three independent samples of the particle solution yields the following variables:

[0083] Gold particles, 50 nm: [0084] z-average: 51.80.7 nm

[0085] Latex beads, 200 nm: [0086] z-average: 201.60.3 nm

[0087] The error relates to the statistically determined standard deviation of the three measurements.

[0088] Filtration and determination of the particle concentrations in nonfiltrate and filtrate:

[0089] The process adsorber P is represented by a layer structure composed of polyamide membranes having a nominal pore size of 0.1 m.

[0090] The test adsorber T is represented by the combination of two different membranes in a layer structure composed of three layers of the polyamide membrane 25005 (0.65 m nominal pore size) and one layer of the membrane 25006 (0.45 m nominal pore size).

[0091] Gold, 50 nm: detection via UV-VIS

[0092] Solution: gold solution containing 0.26% SDS (sodium dodecyl sulfate) having an optical density at 527 nm of about 1

[0093] Filtration:

[0094] The membrane is preflushed with a 0.26% SDS solution using 50 L/m.sup.2 (volume per incident-flow area of the adsorber), but at least five times the dead volume of the filtration apparatus.

[0095] This is followed by filtering the gold-particle solution. Filtered across the adsorbers are 30 L/m.sup.2 of the gold-particle solution, but at least three times the dead volume of the filtration apparatus. Thereafter, two samples are collected separately in order to determine the concentration of the particles: [0096] 2 fractions per 10 L/m.sup.2 are collected separately from one another and analyzed for the particle content.

[0097] The measurement of the concentration of the particles is done against a calibration curve which is generated from a dilution series of the above-described particle solution.

[0098] Gold nanoparticles are quantified via their absorption at 527 nm.

[0099] The latex beads are quantified on the basis of their fluorescence. This involves using the following settings:

[0100] Solution, 200 nm: 0.5% commercial latex beads solution in ROH.sub.2O containing 0.26% SDS [0101] Gently swirl the latex, then hold for approx. 15 s in an ultrasonic bath.

[0102] Filtration: [0103] Flush the membrane with 5 ml of 0.26% SDS solution (Virosart Max, 0.5 bar; 25006, 0.1 bar) [0104] Flush 3 g containing latex beads solution, then [0105] Collect 21 g fractions via a balance (4 ml glass tubes with lid)

[0106] Measurement: [0107] Set the spectrometer to fluorescence (clear-bottom black plate nunc 96Well CAS: 265301) [0108] Excitation wavelength: 468 nm; emission wavelength: 508 nm [0109] Standard series/%: 100, 75, 50, 25, 0 [0110] Measurement volume: 200 l [0111] Amplification, 200 nm: 86

Size-Determination Results:

[0112] Gold Filtration (See FIG. 3):

[0113] The retention of the Virosart Max (25058=process adsorber) is, on average, 80%; in the case of the membrane 25005, it is about 25% (test adsorber). The retention for gold nanoparticles where d=50 nm for the membrane 25058 corresponds to an LRV.sub.Process adsorber=log(C.sub.feed/C.sub.filtrate)=log (1/0.2)=0.69. For the test adsorber, the result is a retention of LRV.sub.Test adsorber=log (C.sub.feed/C.sub.filtrate)=log (1/0.75)=0.12.

TABLE-US-00001 Determination mode for the concentration of gold nanoparticles Absorption Cuvette Wavelength 527 nm Bandwidth 9 nm Number of flashes 25 Rest time 0 ms

[0114] Latex Filtration (See FIG. 4):

[0115] 200 nm: The retention of the Virosart Max (25058=process adsorber) is, on average, >99.9%; in the case of the membrane 25005 (test adsorber), it is about 10%. The retention for latex beads of 193 nm in size is, for the process adsorber, a large LRV.sub.Process adsorber=log(C.sub.feed/C.sub.filtrate)=log(1/0.001)=3. For the test adsorber, the result is an LRV.sub.Test adsorber=log(C.sub.feed/C.sub.filtrate)=log(1/0.9)=0.04.

TABLE-US-00002 Fluorescence Determination mode for the measurement from concentration of latex beads above Excitation wavelength 468 nm Emission wavelength 508 nm Excitation bandwidth 9 nm Emission bandwidth 20 nm Amplification 86 manual Number of flashes 25 Integration time 20 s Delay time 0 s Rest time 0 ms Plate region D6-D10; E6-H9

[0116] According to the invention, the prefiltration by the process adsorber or the test adsorber is carried out in-line, meaning that it satisfies the guideline of the mechanism of the virus depletion being unambiguously identified.

[0117] The retention of the test systems LRV.sub.Test system is determined for the combination of the test adsorber and, as comparative example, also with the process adsorber for the gold nanoparticles of 50 nm in size and for the latex beads of 193 nm in size. The concentrations of the particles are determined as for the determination of LRV.sub.Test adsorber.

[0118] In this exemplary embodiment, the size-exclusion filter used is a Virosart CPV (polyethersulfone membrane having a nominal pore size of 20 nm) virus filter, available from Sartorius Stedim Biotech GmbH.

[0119] The retention of the gold nanoparticles of the two systems is:

LRV.sub.Test system(25005/25006+Virosart CPV)>3.2

LRV.sub.Test system(25058+Virosart CPV)>3.2

[0120] The retention of the latex beads of the two systems is:

LRV.sub.Test system(25005/25006+Virosart CPV)3.2

LRV.sub.Test system(25058+Virosart CPV)3.2

[0121] Analogous to the ascertainment of the retention of the test adsorbers used and of the process adsorber, the LRV.sub.Size-exclusion filter is now calculated in step (c) of the method according to the invention by subtracting the LRV.sub.Test adsorber from LRV.sub.Test system, i.e., LRV.sub.Size-exclusion filter=LRV.sub.Test systemLRV.sub.Test adsorber.

[0122] For the retention of the size-exclusion filter Virosart CPV, what thereby arises for the retention of the gold nanoparticles using the adsorber 25058 is:

LRV.sub.Size-exclusion filter=LRV.sub.Test systemLRV.sub.Test adsorber(25058)

LRV.sub.Size-exclusion filter3.20.69

LRV.sub.Size-exclusion filter2.51

[0123] Using the method according to the invention with use of the adsorber 25006/25005, what arises for the retention of the gold nanoparticles is:

LRV.sub.Size-exclusion filter=LRV.sub.Test systemLRV.sub.Test adsorber(25006/25005)

LRV.sub.Size-exclusion filter3.20.12

LRV.sub.Size-exclusion filter3.08

[0124] Analogously, the retention for the latex beads is now determined. For the use of the adsorber 25058, what arises is:

LRV.sub.Size-exclusion filter3.23.0

LRV.sub.Size-exclusion filter0.2

[0125] Using the method according to the invention with the adsorber 25006/25005, a higher retention is detected:

LRV.sub.Size-exclusion filter3.20.04

LRV.sub.Size-exclusion filter3.16

Example 2: Determination of the Adsorption Capacity of the Adsorbers from Example 1

[0126] Capacity is ascertained on the basis of static protein adsorption using BCA reagent. The redox reaction of the BCA reagent is used for the detection of protein. The color of solutions of the BCA reagent is quantified as absorbance at 562 nm. A calibration curve of known protein concentrations is created and the measurement values are evaluated using this calibration.

DETAILED DESCRIPTION

[0127] 1. Add 3 ml of -globulin solution (3 mg/ml) per Petri dish. [0128] 2. Using tweezers, place the individual test samples (13 mm diameter) and reference samples in, in each case, an appropriately labeled Petri dish (3.5 cm), avoiding any direct hand contact. (Direct marking of the samples canif necessarybe done with an electrophoresis pen). [0129] 3. Incubate the samples in a shaker (80/min) for 3 h at room temperature (approx. 20 C.). [0130] 4. Using tweezers, remove the samples from the Petri dishes and transfer them to appropriately labeled Petri dishes (10 cm). [0131] 5. Add 15 ml of 0.05 M KPi buffer, pH 7.0 per Petri dish and shake for 15 min, then carefully aspirate the buffer solution using a water aspirator and additionally repeat this procedure 3. [0132] 6. Individually transfer the filter samples to appropriately marked Petri dishes (3.5 cm). [0133] 7. To create the calibration curve, individually pipette 325 l, 350 l, 375 l, 3100 l (calibration samples) and 30 l (=blank sample) samples of -globulin solution (1 g/l) into appropriately labeled Petri dishes (3.5 cm), add 2000 l of BCA reagent per dish and immediately place on the shaking platform, then similarly provide the individual membrane samples with 2000 l of BCA reagent each and shake at room temperature (approx. 20 C.). [0134] 8. The photometric measurement of the samples is done in 1 cm semi-microcuvettes. After one hour, the absorbance value of the blank sample at 562 nm is first determined, noted for control purposes and set to zero for the subsequent measurements. Thereafter, the calibration solutions and membrane samples are measured successively without interruption and in the order of their preparation and the absorbance values are noted in the test record.

[0135] The adsorption capacities thus ascertained are used to determine the quantity of the adsorbers, in this case as area ratios relative to the process adsorber. The results of the capacity determination and the resulting quantities of the corresponding adsorbers are summarized in Table 1:

TABLE-US-00003 TABLE 1 Determination of the binding of IgG at pH = 6 with various adsorbers by means of BCA assay (column: BCA_IVIG) and presentation of the scaling factors between process adsorbers and test adsorbers (column: Area ratio according to BCA). Service life Pore size Area ratio of IVIG HC (nominal) BCA_IVIG according as end filter Category Type [m] [g/cm.sup.2] to BCA [l/m.sup.2 in 4 h] Polyamide adsorber Test 25004 0.8 52 1.62 89 adsorber Test 25005 0.65 59 1.42 91 adsorber Test 25006 0.45 71 1.18 83 adsorber Process 25058-SF 0.1 84 1 84 adsorber pH = 6 pH = 6 Negative charge: strong cation exchangers AMPS-MBAm) Test modif. 15445 52 1.62 200 adsorber Process modif. 87 0.97 165 adsorber 15458_3/0.6 Process modif. 60 1.40 134 adsorber 15458_2/0.4 Reference: 15458 0.1 3.3 25.45 without charge modification 25058 147 AMPS: 2-Acrylamido-2-methylpropanesulfonic acid MBAm: N,N-ethylenebisacrylamide

Example 3: Scaling of a Test Adsorber which has a Differing Size-Exclusion Limit in Relation to the Process Adsorber

[0136] In Table 1 above, the scaling factors arising from the binding capacities of the BCA assay are already presented. The practical application of said factors will be described below.

Filtration of a Human IgG Protein Solution Across Various Polyamide Prefilters:

[0137] For the filtration, use was made of a solution of a human IgG (5% solution, SeraCare, catalog No. HS-475-1 L) which was diluted with a 50 mM KPI buffer solution, pH 7.2, to a concentration of 0.5%.

Membranes Used:

[0138] Prefilter (VF).sub.1: Nylon 6 [0139] Nominal pore size: 0.35 m [0140] Flow time for water [ml/(min cm.sup.2 bar)]: 23 [0141] Bubble point with water, visual [bar]: 2.9

[0142] Prefilter (VF).sub.2: Nylon 6 [0143] Nominal pore size: 0.1 m [0144] Flow time for water [ml/(min cm.sup.2 bar)]: 6 [0145] Bubble point with water, visual [bar]: 7

[0146] End filter (EF): surface-modified, virus-retentive polyethersulfone Virosart HC membrane (surface-modified polyethersulfone membrane having a nominal pore size of 20 nm), double layer [0147] Nominal pore size: 20 nm [0148] Flow time for water [ml/(min cm.sup.2 2 bar)]: 0.19 [0149] Effective incident-flow area: 4.7 cm.sup.2

[0150] Both filtrations were operated in-line, i.e., VF and EF were connected to one another via the Luer connector before the start of filtration. The VF.sub.1 module has an effective incident-flow area of 4.7 cm.sup.2, whereas the VF.sub.2 module has an effective incident-flow area of 3.11 cm.sup.2.

[0151] The filtrations were carried out in parallel with the same stock solution and under a pressure of 2 bar. The different slopes in FIGS. 5 and 6 are caused by the differing flow resistance of the two adsorbers.

[0152] FIGS. 5 and 6 show that the filtered volume is virtually identical for the two filtrations, because the adsorber quantity used, referred to here as VF1 and VF2, was scaled such that the same binding capacity for impurities was achieved.

[0153] Owing to the appropriate scaling of the membrane areas of the prefilters, an approximately identical service life was achieved.

Selection Criteria for the Area Scaling of the Prefilters:

[0154] One way of measuring the nonspecific protein adsorption of a membrane is found in the BCA method. In said method, the protein adsorbs to a membrane and is quantitatively determined photometrically using BCA reagent (2,2-biquinoline-4,4-dicarboxylic acid disodium salt dihydrate).

[0155] The following results were ascertained in this connection:

VF.sub.1: 52.7 g/cm.sup.2
VF.sub.2: 84.2 g/cm.sup.2

[0156] To then arrive at an identical adsorption during the filtration, the area of the VF.sub.1 had to be enlarged by a factor of 1.6.