Preparation methods for a novel generation of biological safe KLH products used for cancer treatment, for the development of conjugated therapeutic vaccines and as challenging agents

Abstract

The present invention relates to the provision of a biologically safe hemolymph sera, preferably hemocyanin, more preferably KLH (keyhole limpet hemocyanin). The hemocyanin is purified using anion exchange chromatography.

Claims

1. A method of isolating and/or purifying hemocyanin or immunocyanin comprising the steps of: a) providing a hemocyanin formulation, wherein the hemocyanin formulation is a hemolymph serum derived from a marine mollusk; b) reducing the conductivity of the formulation provided in step a) to a value of between 30 mS/cm and 10 mS/cm by adding a dilution buffer to the formulation provided in step a); c) adding the diluted hemocyanin formulation obtained in step b) to an anion exchange chromatography column, wherein the diluted hemocyanin formulation is a diluted hemolymph serum, and wherein the diluted hemolymph serum contains serum proteins other than KLH; d) rinsing the column with a second buffer; and e) eluting the hemocyanin or immunocyanin from the column by adding a third buffer.

2. The method of claim 1, wherein the matrix material is an anion exchange matrix including particles of a particle size of more than 30 m, preferably 50 m +/ 5 m.

3. The method of claim 1, wherein the anion exchange matrix includes pores of a pore size between 3,000 and 8,000 Angstroms.

4. The method of claim 1, wherein the rinsing in step (d) removes salts, carbohydrates and proteins other than KLH.

5. The method according to claim 1 wherein the second buffer has a pH between 7 and 8.

6. The method of claim 5, wherein the anion exchange matrix includes pores of a pore size between 3,000 and 8,000 Angstroms.

7. The method of claim 5, wherein hemocyanin is eluted, wherein the third buffer has a pH between 7 and 8 and comprises Ca.sup.++and Mg.sup.++ions.

8. The method of claim 7, wherein the matrix material is an anion exchange matrix including particles of a particle size of more than 30 m, preferably 50 m+/5 m.

9. The method of claim 8, wherein the anion exchange matrix includes pores of a pore size between 3,000 and 8,000 Angstroms.

10. The method of claim 5, wherein immunocyanin is eluted, wherein the third buffer has a pH between 9 and 10 and is free from Ca.sup.++and Mg.sup.++ions.

11. The method of claim 10, wherein the matrix material is an anion exchange matrix including particles of a particle size of more than 30 m, preferably 50 m+/5 m.

12. The method of claim 11, wherein the anion exchange matrix includes pores of a pore size between 3,000 and 8,000 Angstroms.

13. The method of claim 1, wherein hemocyanin is eluted, wherein the third buffer has a pH between 7 and 8 and comprises Ca.sup.++and Mg.sup.++ ions.

14. The method of claim 13, wherein the matrix material is an anion exchange matrix including particles of a particle size of more than 30 m, preferably 50 m+/5 m.

15. The method of claim 13, wherein the anion exchange matrix includes pores of a pore size between 3,000 and 8,000 Angstroms.

16. The method of claim 1, wherein immunocyanin is eluted, wherein the third buffer has a pH between 9 and 10 and is free from Ca.sup.++and Mg.sup.++ions.

17. The method of claim 16, wherein the matrix material is an anion exchange matrix including particles of a particle size of more than 30 m, preferably 50 m+/5 m.

18. The method of claim 16, wherein the anion exchange matrix includes pores of a pore size between 3,000 and 8,000 Angstroms.

Description

EXAMPLES

Example 1: Quarantine of Megathura Crenulata

(1) Since the start of the production activities at our facility in Carlsbad, Calif. in February 2002, we have collected 13 batches of animals at various time periods. The collection site ID is the zone number as defined by the Southern California Fisheries Chart (SCFC).

(2) The weather condition in California during the collection of animals for batches MC-001 to MC-013 were not unusual. However, during the more recent batch of animals collected, MC-014, California had experienced unusual rainy conditions prior to animal collection, however, on the day of animal receipt we experienced no rains.

(3) It may be pointed out that the incorporation of test methods has evolved since we started the activities at our facility. The fecal coliform testing on animals, toxic substance, DDT and PCB testing was initiated with lot # MC-002. The pH and Conductivity testing has been done for all lots, the nitrate, nitrite and ammonia testing was initiated on the sea water sample with lot # MC-006.

(4) The pH and conductivity of sea water has a range of 8.0 to 8.3 and 45.0 to 52.4 respectively. The nitrate, nitrite and ammonia content ranged from 0 to 80 ppm, 0 to 0.25 ppm and 0.25 to 1.0 ppm respectively, the high end of the range corresponds to the values for lot MC-014, which, as has been noted was collected post heavy rains. The fecal coliform of sea water samples were <2 MPN/100 mL for samples collected from MC-002 to MC-013 and were 29, 11 and 49 MPN for the most recent lot MC-014 corresponding to the three water samples 0, 50 and 100 respectively.

(5) The DDT and PCB test results indicate that they are below the detection limit of the respective assays. It appears that this may not be an issue in the collection site 718 and 719 where the animals were collected from.

(6) The animal fecal coliform data suggests that the fecal coliform were generally <18 and a maximum of 20 MPN/100 grams for the lot # MC-002 to MC-013, however for MC-014 the values were very high, 3,500 MPN/100 gram. These results in conjunction with the fecal coliform in the surrounding sea water suggest that these animals tend to concentrate the fecal coliform.

(7) On receipt of the laboratory results for fecal coliform in animals, we decided to send samples of animals from our quarantine tanks, 3 animals were taken from Tank Q1 and 3 from Tank Q2, on Jan. 19 2005. The animals were received into our tanks on Jan. 6 2005 and therefore were present in our tank water for 13 days prior to testing. We initiated this testing to determine if the quarantine procedures that we have incorporated into our manufacturing schedule would have an effect on reducing the animal fecal coliform content. The copy of the test report is attached to this report as an attachment. The results indicate that the animal fecal coliform is <18 MPN/100 gram. These results are very encouraging and suggest that the procedures in place are effective in reducing the animal coliform content, if any are present as with this Lot MC-014.

(8) The analysis of the data relating to animals and sea water suggest that:

(9) The pH and conductivity of sea water provide information on the natural sea water conditions and would be useful to compare with our artificial sea water prepared in-house. We currently have a specification for artificial sea water as 7.5 to 8.5 and conductivity 46-52 ms/cm. These specifications seem to match well with the ranges for natural seawater.

(10) The toxicological screening for sea water from collection site was initiated based on advice from Dr. Robert Mooney, Merkel & Associates, used as an external animal health inspector for the first three lots of animal received, namely MC-001 to MC-004. The results to date suggest that DDT and PCB's are not an issue for the area where the limpets are collected from and released into.

(11) The in-house animal quarantine procedure seems to aid in reducing the animal coliform content and is a very useful procedure. Currently, animals release for manufacture require a minimum of seven days from start of quarantine, the animal fecal coliform data for MC-014 was obtained on animals after 13 days of animal holding in our tanks. It may be necessary to initiate a more systematic investigation into the length of quarantine and reduction of fecal coliform and determine if the current set specification of seven days is sufficient. Such studies to be initiated after collection of animals from waters with high coliform as was the case with MC-014.

Example 2: Direct Chromatographic Isolation of Native Hemocyanin from Hemolymph Sera

(12) Hemolymph Sera based on its' origin from a marine mollusc contains, apart from hemocyanin and other serum components, high levels of Sodium Chloride and other minerals. The conductivity on average is around 50 ms/cm. Under those present conditions KLH cannot be bound to Ion Exchange resins. To achieve quantitative binding of KLH the conductivity has to be reduced to <20 mS/cm.

(13) In order to reduce the conductivity as described above the Hemolymph Sera is partially desalinated by suitable methods such as gel filtration, electrodialysis, diafiltration or dilution. The removal of salts leads to precipitation of the other serum components i.e protein and carbohydrates. The precipitation is removed by low speed centrifugation, depth filtration or membrane filtration (0.8 u. 0.45 m).

(14) Subsequently the colloidal dissolved high molecular weight KLH is isolated by chromatography procedures i.e. IEX chromatography followed by dissociation and purification of subunits.

(15) Dissociation of Hemocyanin

(16) Method 1:

(17) The native, oxygen-binding hemocyanin protein is purified from the hemolymph by Ion Exchange Chromatography. The hemocyanin is bound to an anion exchanger and then dissociated on the column into the KLH subunits (immunocyanin) in alkaline (pH 9.6) buffer. The immunocyanin is recovered from the column by means of salt gradient elution. The resulting immunocyanin solution is desalinated and concentrated by diafiltration/ultrafiltration. The concentrated immunocyanin solution may be subsequently purified by a further IEX chromatography step.

(18) Method 2:

(19) The native, oxygen-binding hemocyanin protein is purified from the hemolymph by Ion Exchange Chromatography. The hemocyanin is bound to an anion exchanger and then recovered from the column by means of salt gradient elution. The resulting hemocyanin solution is desalinated, concentrated and dissociated into the KLH subunits (immunocyanin) by means of diafiltration, dialysis or ultrafiltration in alkaline (pH 9.6) buffer. Finally the immunocyanin solution is concentrated followed by purification with a further IEX chromatography step.

(20) The hemocyanin obtained from Method 1/Method 2 is dissolved in dissociation buffer (pH 9.6, devoid of Ca.sup.++ and Mg.sup.++). This creates alkaline conditions, which lead to dissociation of the native hemocyanin molecule into its subunits. The entity of these subunits are called immunocyanin.

(21) Method 3:

(22) The native, oxygen-binding hemocyanin protein is purified from the hemolymph by Ion Exchange Chromatography. The hemocyanin is bound to an anion exchanger and then recovered from the column by means of salt gradient elution. The hemocyanin of the resulting hemocyanin solution is isolated by means of ultracentrifugation. The obtained pellets are dissolved in dissociation buffer. Finally, the immunocyanin solution is concentrated and may be purified with a further IEX chromatography step.

(23) Concentration of Immunocyanin Solution

(24) Before final purification (polishing) by gel filtration, the immunocyanin solution is concentrated to a protein content of 20 mg/mL (2.5 mg/mL) by ultrafiltration. For this purpose, low protein binding polysulfone or polyether sulfone membranes (separation limit: 30,000 Dalton; filter area: 700 cm.sup.2) mounted in a stainless steel ultrafiltration unit are used.

(25) After ultrafiltration, the concentrated immunocyanin solution is filtered through a 0.22 m membrane filter.

(26) Purification (Polishing)

(27) The concentrated immunocyanin solution is finally purified by middle pressure liquid chromatography through a gel filtration column.

Example 3: Nanofiltration of Purified KLH Subunits (Immunocyanin)

(28) Due to the origin of native KLH exists a virological risk by human pathogens. To guarantee biological safety the downstream process of biologicals should contain steps for inactivation or removal of potential virus contamination. Commercially available inactivation methods were tested on KLH and found to be not suitable because of their damaging effect on the KLH preparations (pH reduction, heat treatment).

(29) Nanofiltration was previously shown to effectively remove various viruses from protein compositions. However, nanofiltration turned out to not be useful for hemocyanins or native KLH, due to the molecular weight of native KLH from >8 Mill. Dalton. The filters could not discriminate viruses from protein, i.e. the protein is too big to pass the membrane of typical virus filters (pore size between 15 and 35 nm). A reduction to the uniform molecular weight of approx. 400,000 Dalton of KLH subunits was affected. Several virus filters with different pore sizes have been tested in a down scaled process.

Reference Example: Dead EndFiltration Protocol, Virus Filtration of KLH Subunits

(30) A protein of approx. 400 KD in a concentration of approx. 5 mg/ml obtained from California see snail's blood has been filtered through Planova 20N, 0.001 m.sup.2 in Dead-End modus with constant pressure of 2.0 bar. The flow rate was 0.4 ml/min. The protein formulation was at a pH of 9.6 in a glycine/NaOH buffer. 1 g of starting material was applied. With dead-end filtration, 0.1 g of protein in a concentration with 0.5 mg/ml was obtained, i.e. a protein yield of 10%. Also, a reduction of the starting amounts of the protein or the concentration of the protein by a factor of 10 or more did not lead to different results. Also, the reduction of pressure or the increase of size did not lead to changes of protein yield.

Working Example: Cross-Flow Filtration Protocol; Virus Filtration of KLH Subunits

(31) A protein of approx. 400 KD in a concentration of approx. 0.45 mg/ml obtained from California sea snails' blood has been filtered through Planova 20N, 0.12 m.sup.2 in cross-flow modus with a constant pressure of 0.16 bar. Formulation was in a buffer of glycine and NaOH at a pH of 9.6. The amount of starting material was 5,000 g in a concentration of 0.45 mg/ml. The protein amount obtained after nanofiltration was 4,688 g in a concentration of 0.42 mg/ml. This makes up to a yield of 93%.

(32) This example demonstrates that contrary to dead-end filtration, cross-flow filtration enables the filtration of quantitative amounts of hemocyanine protein upon dissociation into its subunits. Nanofilters of a pore size between 15 and 35 nm can be employed which are sufficient to remove the smallest viruses known.

Example 4: Proof of Concept: Feasibility Study

(33) This example demonstrates the suitability of cross-flow filtration for removing small viruses of a diameter of the smallest viruses known. In this example, PPV was tested. PPV has a diameter of 20 nm and the virus was spiked at a concentration of 0.5% per protein. Immunocyanin, a protein of uniform molecular weight of approximately 400,000 Dalton of KLH subunits was spiked with 0.5% PPV. The total virus load in prefiltered was 10,620. The protein amount after virus spiking was 4,814.9 g. Nanofiltration was performed with Planova 20N nanofilters, 0.12 m.sup.2 in cross-flow mode. The flow-rate chosen was 50 mm/min with a constant pressure of 0.28 bar. 4,662.4 g protein were retained in the filtrate. The LRF for PPV was 3.14+/0.32. Accordingly, the protein amount was more than 93% with a virus removal of more than 99.9%.

(34) This example shows that cross-flow filtration is suitable for the preparation of KLH or KLH subunits in a virus free form on a commercially relevant scale with a yield of more than 90%.

Example 5: Stabilisation of Purified KLH Subunits by Means of Desalination

(35) Das Folgende wrde ich noch krzer zusammenfassen, so z.B.!

(36) Principle

(37) KLH BULK LIQUID salt-free is manufactured from purified immunocyanin by desalination and concentration. Both are achieved by a series of ultrafiltration steps using a polysulfone membrane with a nominal separation limit of 30,000 Da.

(38) Preparation of the Desalination Batch

(39) In order to minimize batch-to-batch variations during the desalination process, purified immunocyanin solution is concentrated to an immunocyanin content of between 10 mg/ml and 40 mg/ml (2 mg/ml) by ultrafiltration. For this purpose, low protein binding polysulfone or polyether sulfone membranes (separation limit: 30,000 Da) mounted in a stainless steel ultrafiltration unit are used. Before use, the membranes are conditioned by recirculation with alkaline dissociation buffer at a temperature of +2-8 C. Flow is achieved by a peristaltic pump. Finally, the conditioning is tested by in-process control pH and bacterial endotoxins. Before starting the concentration process, the ultrafiltration unit is checked for integrity.

(40) Concentration

(41) The immunocyanin solution is now transferred to the ultrafiltration unit and recirculated at +2-8 C. The concentration is controlled be weighing the obtained ultrafiltrate. The maximal entrance pressure of the ultrafiltration unit should not exceed 1 bar, preferably not 0.5 bar. The immunocyanin solution is recirculated until the calculated amount of ultrafiltrate has been collected. Finally, the concentrate is tested by in-process control pH value, osmolality, conductivity, immunocyanin content.

(42) Desalination

(43) The concentrated immunocyanin solution (=concentrated desalination batch) is either desalinated by dilution 1+1 with water for injections at each ultrafiltration cycle or alternatively by adding the water for injections employing constant volume wash procedure. The desalination process is controlled by weighing the ultrafiltrate, the concentrated desalination batch and testing of conductivity of the ultrafiltrate and the desalination batch.

(44) If the conductivity of the ultrafiltrate has reached <10 S/cm or if the conductivity of the concentrated desalination batch is <150 S/cm, the desalination process is terminated and the immunocyanin content of the desalination batch is determined in order to prepare the final batch of KLH BULK LIQUID salt-free.

(45) Preparation of the Final Batch of KLH BULK LIQUID Salt-Free

(46) The final batch of KLH BULK LIQUID salt-free is prepared from the immunocyanin concentrate by dilution to an immunocyanin content of 20 mg/ml. For this purpose, the filtered immunocyanin concentrate is weighed. The required amount of water for injections is weighed accurately and slowly added to the filtered immunocyanin concentrate. The solution is gently mixed, and a sample for in-process control is removed pH value, density, osmolality, conductivity, immunocyanin content.

(47) The released solution is finally sterilized by filtration through a 0.22 m membrane filter directly into infusion bags.

(48) They are stored at +2-8 C.

(49) During filtration, samples for quality control are removed.

Example 6: Refolding of Virus Filtered KLH Subunits and Final Purification

(50) Principle

(51) High molecular weight KLH is manufactured from concentrated immunocyanin solution by diafiltration (buffer exchange to reassociation conditions, pH 7-8, Ca.sup.++, Mg.sup.++) and concentration. Both are, e.g. achieved by a series of ultrafiltration steps using a polysulfone membrane with a nominal separation limit of 50,000 Da.

(52) Concentration of the Purified Immunocyanin Solution

(53) In order to optimize reassociation conditions, purified immunocyanin solution is concentrated to an immunocyanin content of 20 mg/mL (2 mg/mL) by ultrafiltration. For this purpose, low protein binding polysulfone or polyether sulfone membranes (separation limit: 30,000 Da) mounted in a stainless steel ultrafiltration unit are used. Before use, the membranes are conditioned with elution buffer as follows. The ultrafiltration system is first rinsed with elution buffer at a temperature of +2-8 C., while the filtrate outlets are closed. Flow is achieved by a peristaltic pump. Finally, the elution buffer is completely removed from the system. A sample is removed from the retentate side for in-process control pH, bacterial endotoxins. Before starting the concentration process, the ultrafiltration unit is checked for integrity.

(54) The immunocyanin solution is now transferred to the retentate bag of the ultrafiltration unit. The recirculation is started, and the ultrafiltrate is collected in a weighed beaker. The temperature is kept at +2-8 C. As during conditioning, the maximal entrance pressure of the ultrafiltration unit should not exceed 1 bar. The immunocyanin solution is recirculated until the calculated amount of ultrafiltrate has been collected. The retentate is mixed, while the filtrate outlets are closed, and a sample is removed for in-process control pH value, osmolality, conductivity, immunocyanin content.

(55) Reassociation

(56) In order to refold the KLH subunits a second ultrafiltration system with low protein binding polysulfone or polyether sulfone membranes (separation limit: 50,000 Da) mounted in a stainless steel ultrafiltration unit are used. Before use, the membranes are conditioned with reassociation buffer as follows. The ultrafiltration system is first rinsed with reassociation buffer at a temperature of +2-8 C., while the filtrate outlets are closed. Flow is achieved by a peristaltic pump. Finally, the reassociation buffer is completely removed from the system. A sample is removed from the retentate side for in-process control pH, bacterial endotoxins. Before starting the reassociation process, the ultrafiltration unit is checked for integrity.

(57) For reassociation the ultrafiltration system is recirculated with reassociation buffer (between 2- to 10-fold volume of concentrated immunocyanin solution) while the filtrate outlets are closed. The concentrated immunocyanin solution is slowly injected in the recirculated reassociation buffer. The temperature during the whole reassociation process is kept at a temperature of +2-8 C. After complete injection of the concentrated immunocyanin solution the reassociation batch is diafiltrated against between 2- and 10-fold of reassociation buffer applying the constant volume wash procedure. Finally the batch is concentrated to a KLH content of 20 mg/ml. After reassociation, the concentrated KLH solution, is filtered through a 0.22 m membrane filter.

(58) Purification of Refolded KLH by Gel Filtration

(59) The concentrated KLH solution is finally purified by middle pressure liquid chromatography through a gel filtration column.

(60) Biological Activity and PotencyComparability with Native KLH

(61) Native KLH and synthetic KLH obtained after reassociating according to the method of the present invention were compared. Synthetic KLH and native KLH were compared via CD-spectroscopy. Bands in CD-spectroscopy were identical.

(62) The protein bands in SDS PAGE were identical when comparing synthetic and native KLH. In synthetic KLH, no protein fragments are found.

(63) 2-dimensional immunoelectrophoresis was also performed to compare synthetic and native KLH. Anti-KLH1 and anti-KLH sera were used. The immunoelectrophoretic patterns were identical for both, native and synthetic, KLH. Two precipitation maxima (one for KLH1 and one for KLH2) occur for both native and synthetic KLH.

(64) Electromicroscopic investigations both, native and synthetic, KLH show the typical decamers, didecamers, and tridecamers.

(65) Native PAGE and densiometric tests show that both synthetic and native KLH include the typical protein bands. A ratio between KLH1 and KLH2 between 0.9 and 1.0 for both, synthetic and native, KLH was obtained.

Example 9: AnionExchange Chromatography

(66) For the isolation, purification and in situ dissociation of hemocyanin anion exchange chromatography was selected because of its isoelectric point (pI) at pH 6 and the negative charge at pH>7.

(67) Hydrophobic interaction separation was involved because of applying different conditions of those used in ion exchange chromatography (IEX). In this separation, a buffer with a high ionic strength, usually ammonium sulfate, is initially applied to the column. The salt in the buffer reduces the solvation of sample solutes thus as solvation decreases, hydrophobic regions that become exposed are adsorbed by the medium.

(68) In a first series of binding studies using hemocyanin solutions with low salt concentration different strong and weak anion exchange media packed in columns (bed volumes approx. 20-50 ml) as well as different HIC columns were proofed on their binding capacity.

(69) Strong and weak anion exchange media were selected due to their specific binding properties and selectivity:

(70) Poros 50 micron media for perfusion chromatography (HQstrong anion exchanger, functional group: quaternized polyethyleneimine; PIweak anion exchanger, functional group: quaternized polyethyleneimine; Dweak anion exchanger, functional group: dimethyl amino alkyl groups DEAE).

(71) The Poros 50 micron media were used because of their robust chemical stability, particle size of 50 m, and the large pore structure (appr. 6000 ) which enables high flow rates without compromising capacity or resolution.

(72) Q-Sepharose FF and HP (strong anion exchanger) with particle sizes of 90 m and 34 m were also tested alternatively. The functional groups exists of quarternary ammonium. Hydrophobic interaction chromatography tests were performed using a HiTrap HIC Selection Kit, which contains prepacked columns ready to use (HiTrap HIC Phenyl FF high sub 1 ml, HiTrap HIC Phenyl FF low sub 1 ml, HiTrap HIC Butyl FF 1 ml and HiTrap HIC Octyl FF 1 ml).

(73) In summary no or only minor binding of hemocyanin could be obtained with the hydrophobic interaction chromatography media. The Q-Sepharose FF and HP and the weak anion exchanger of the Poros media (PI and D) showed only less to medium binding. The highest binding capacity of approx. 20 mg KLH per ml gel has been proofed with Poros 50 micron HQ media.

(74) Based on the obtained results the developmental work was continued in order to purify KLH directly from the starting material Hemolymph Sera.

(75) Hemolymph Sera based on its origin from a marine mollusc contains apart from hemocyanin and other serum components high levels of sodium chloride and other minerals. The conductivity on average is around 50 mS/cm. Under those present conditions KLH could not be bound to the selected Poros 50 HQ media.

(76) In order to achieve quantitative binding of KLH the hemolymph sera was diluted with TRIS/HCl buffer (pH 7.4, including CaCl.sub.2, MgCl.sub.2 and NaCl), so called IEXdilution buffer. The precipitation occurred during dilution was removed by prefiltration steps.

(77) 4 (four) dilution series were tested to determine the binding capacity of Poros 50 HQ media for KLH (hemolymph sera diluted to approx. 25 mS/cm, 20 mS/cm, 15 mS/cm and 10 mS/cm). Tests were performed on Poros 50 HQ columns (approx. 50 ml gel). The elution of KLH started at approx. 27 mS/cm.

(78) Flow rate sample: 5 ml/min

(79) Flow rate elution: 10 ml/min

(80) Elution gradient: 0.15 to 0.65 M sodium chloride, TRIS/HCl buffer (pH 7.4, including CaCl2, MgCl2 and NaCl).

(81) The binding studies showed that the hemolymph sera has to be diluted to <20 mS/cm to achieve quantitative binding of KLH.

(82) Subsequently additional test runs with hemolymph sera diluted to approx. 19 mS/cm were performed to optimize KLH binding addicted to the sample feeding velocity. 4 (four) different sample flows (4.0 ml/min, 4.6 ml/min, 5.2 ml/min and 8.0 ml/min) were tested under the conditions described above. The obtained results demonstrated the correlation between feeding velocity and binding capacity of the anion exchanger media. With the lowest flow rate of 4.0 ml/min (corresponding to a linear flow rate of approx. 50 cm/h) approx. 23 mg KLH per ml gel were bound and even with the highest flow rate of 8.0 ml/min (corresponding to a linear flow rate of approx. 100 cm/h) the achieved binding of approx. 17 mg KLH per ml gel was excellent. Break through of KLH was monitored by UV detection at 280 and 340 nm. Subsequently further trials were initiated in lab scale as well as during scale up to optimize the chromatography parameter due to an effective and economic isolation and purification process.

(83) In addition in situ dissociation was tested on KLH captured on the Poros 50 HQ column applying the chromatography parameters as described above. After a first rinsing step with TRIS/HCl buffer (pH 7.4, including CaCl.sub.2, MgCl.sub.2 and NaCl) buffer exchange with glycine/NaOH buffer (pH 9.6, including NaCl, EDTA) was performed. No breakthrough of KLH was observed during rinsing the column under dissociation conditions.

(84) Elution of the KLH subunit fraction was achieved using a elution gradient of 0.075 to 0.575 M sodium chloride, glycine/NaOH buffer (pH 9.6, including NaCl, EDTA).

(85) The degree of dissociation was determined by means of analytical MPLC/SEC (FPLC; separating column: Superose 6 HR 10/30, molecular weight separation range 5103-5106 Da) and by means of native PAGE. The measurement of the subunit fraction showed that an almost quantitative dissociation into the KLH subunits was achieved on the column. Due to the concentration of KLH subunits (approx. 20 mg/ml) the filtered fraction (0.22 m membrane filter) was used for further purification by gel filtration without any additional processing.

(86) At the end of the research program, anion exchange chromatography was the method of choice.