FIELD OF THE INVENTION
The present invention relates to a rapid and efficient method for separation of a non-ionic detergent from a protein formulation containing the said detergent.
BACK GROUND OF THE INVENTION
Importance of non-ionic detergents  in particular polysorbates  is widely recognized in therapeutic protein formulations. Chemically  polysorbates are polyoxyethylene sorbitan fatty acid esters (Kerwin et al.  J Pharm Sci. 2008 Aug; 97(8):2924-35) and are thought to stabilize protein formulation by preventing surface adsorption and aggregation of the protein (Ha et.al  J Pharm Sci. 2002 Oct; 91(10):2252-64). Randolph et al.  (Pharm Biotechnol. 2002; 13:159-75) has extensively reviewed various interactions between proteins and polysorbates.
Polysorabte 20 or 80 are usually used at a concentration range of 0.001 to 3% in therapeutic protein formulations. The appropriate concentration is very dependent on the nature of the protein. Due to the chemically heterogeneous nature of polysorbates  there is a regulatory requirement to quantify the amount of polysorbate in the formulated product.
Different techniques have been developed for detection and quantification of polysorbates. US7531358 proposes the use of a colorimetric method for quantification of polysorbate using ammonium thiocyanate/cobalt nitrate reagent. Adamo et. al.  (J Chromatogr B. 2010 Jul 1; 878 (21):1865-70) proposed a method for quantification of polysorbate 80 by alkali hydrolysis of the ester linkage between oleic acid and the polyoxyethylene sorbitan. The released oleic acid was analyzed by reverse phase high performance liquid chromatography (RP-HPLC). Due to the equimolar ratio of oleic acid and polyoxyeyhylene sorbitane  oleic acid estimation was used as a “surrogate marker” to quantitate the total polysorbate 80 present in the preparation.
Thus  though different techniques for polysorbate quantitation are known in the art  a major challenge lies in the separation of polysorbate from a protein containing formulation. Currently known polysorbate separation methods are primarily based on phase separation between an organic and aqueous layer. US20070082004 proposes a method for separation of polysorbate from a protein preparation by using two sets of reagents. Whereas  the first reagent forms a complex with the detergent  the second reagent precipitated the proteinaceous component of the formulation.
Hewitt et al (Journal of Chromatography A  1215 (2008) 156–160) proposed a method for polysorbate 20 quantitation in which the detergent component was separated from the protein by using an “on-off chromatography approach”. Overall  the process depended on binding polysorbate 20 to a resin  while simultaneously repelling the proteinaceous component of the formulation electrostatically. The bound polysorbate was quantitated by elution of bound polysorbate from the resin employing an organic mobile phase and detection using an evaporative light scattering detector (ELSD).
Though different techniques are available for quantitation of polysorbates in protein preparations  there are considerably fewer techniques available for separation of non-ionic detergents from protein preparations. In fact  efficient quantitation of polysorbate present in a protein preparation depends on the efficiency of polysorbate separation from the protein formulation. Further  currently available separation methods employ organic solvents and are often time consuming. Therefore  there is a need for new techniques which are both simpler and more efficient in separating polysorbates from a protein formulation.
The primary object of the present invention is to provide a method for separation of non-ionic detergents in particular polysorbates from a protein preparation  that is simple and efficient. A further object of the present invention is to develop a method for efficient fractionation of non-ionic detergent in particular polysorbates from a protein preparation without use of any organic solvents. A further object of the present invention is to develop a method for separation of polysorbates from a protein preparation without involving any protein precipitation step. Another object of the present invention is to develop a method for separation of polysorbates from a protein preparation without involving any chromatographic step.
SUMMARY OF THE INVENTION
The present invention describes to a rapid and efficient method for separation of non-ionic detergent in particular polysorbates from a protein preparation containing the said detergent. The disclosed method does not employ use of any organic solvent  protein precipitation or any chromatographic step.
DETAILED DESCRIPTION
Non-ionic surfactants in particular polysorbate 20 and 80 have been extensively used as stabilizers in therapeutic protein preparations. Due to their chemical complexity  heterogeneity and presence in final therapeutic preparation  their quantitation is of utmost importance. Though different techniques are known for polysorbate quantitation  a key challenge lies at the efficient fractionation of polysorbates from a protein preparation.
The present invention provides a simple  rapid and efficient method for separation of non-ionic surfactant in particular polysorbates from a protein formulation. Once fractionated the polysorbates could be quantified by different means such as  colorimetry or chromatography.
In one embodiment  the present invention provides a method for separation of a non-ionic detergent from a protein formulation  wherein the method comprises 
a. bringing at least one solid support that exhibits affinity to the protein  in contact with a protein free buffer  wherein the said protein free buffer has a composition that is same as that of the protein formulation but comprises an amount of non ionic detergent that is same or greater than the amount of non-ionic detergent present in the protein formulation  
b. incubating the mixture of step a for a first period of time under gentle agitation
c. separating the protein free buffer form the said solid support
d. repeating the steps a to c at least twice to obtain a “treated solid support”
e. bringing said “treated solid support” in contact with the protein formulation as described in step a 
f. incubating the mixture of step d for the second period of time under gentle agitation
g. obtaining the “output buffer” from step f by separating the solid support and the liquid phase  wherein the “output buffer” lacks the said protein component but contains almost identical amounts of the non-ionic surfactant as that of the protein formulation in step e.
In a further embodiment  the solid support is one that is capable of binding the protein present in the protein formulation non-covalently. In case of protein preparations  containing more than type of protein  more than one solid support exhibiting affinity to these proteins may be employed.
In a further embodiment  the protein free buffer is incubated with the solid support for about 3 minutes at room temperature under gentle agitation.
In a further embodiment  the protein formulation is incubated with the treated solid support for about 15 minutes at room temperature under gentle agitation.
In a further embodiment  the solid support is separated from the liquid phase by various methods  non limiting examples include sedimentation of the solid support under gravity followed by aspiration of the liquid supernatant  or centrifugation and separation of the solid and liquid phases.
In a further embodiment  the non-ionic detergent is a polysorbate  such as polyoxyethylene sorbitan fatty acid ester
In a further embodiment  the buffer in step g comprises about 90-110% of the non-ionic detergent as that of the protein formulation  preferably  about 95-107% .
The method described in the invention does not employ any organic solvent or precipitation of proteinaceous component/s present in the protein preparation.
In the context of the present invention  the term solid support refers to any immobilized or immobilizable matrix  including resins  monoliths  membranes or other form of affinity or ion exchange support known in the art.
Certain specific aspects and embodiments of the invention are more fully described by reference to the following examples. These are provided only for purposes of illustration and should not be construed as limiting the scope of the invention in any manner.
Example 1
Evaluation of an affinity based solid support for fractionation of polysorbate from a protein preparation.
To evaluate ability of a protein affinity based solid support for fractionation of polysorbate from a protein preparation  a protein A MabSelect resin was employed. About 500µl of MabSelect protein A slurry was taken in a microfuge tube and the resin was allowed to settle under gravity. Excess liquid was discarded by aspiration. To this settled resin about 500µl of rituximab  formulation (described in table 1) was added  and incubated for 15 minutes at room temperature under gentle agitation. The incubation was followed by separation of the solid phase from the liquid phase by centrifugation at 15000g for 5 minutes. The resulting supernatant was collected and percentage recovery of polysorbate from this output buffer was calculated by comparing the obtained amount with that of the expected amount of polysorbate. A recovery of about 60% was observed indicating non-specific binding of polysorbate to the affinity resin. Based on these findings  pre-treatment of the affinity resin with a buffer identical to the protein formulation but lacking any protein was evaluated.
Example 2
Pre-treatment of affinity resin with a Protein free formulation buffer.
Three different types of protein containing varying amounts of polysorbate were employed in the study (Table 1). The table in addition describes the composition of individual protein free buffers used for pre treatment of the affinity resin. The protein free buffer contained identical concentration of polysorbate as that of the protein formulation.
Protein Formulation Protein Free Formulation Buffer
10mg/ml Rituximab
9 mg/ml Sodium chloride
7.35 mg/ml Sodium Citrate dehydrate
0.7 mg/ml Polysorbate 80
pH 6.5 9 mg/ml Sodium chloride
7.35 mg/ml Sodium Citrate dehydrate
0.7 mg/ml Polysorbate 80
pH 6.5
25mg/ml Bevacizumab
60 mg.ml? a ?a-Trehalose dehydrate
5.8mg/ml Sodium phosphate
1.2 mg/ml Sodium phosphate
0.475mg/ml Polysorbate 20
pH 6.2 60 mg.ml? a ?a-Trehalose dehydrate
5.8mg/ml Sodium phosphate
1.2 mg/ml Sodium phosphate
0.475mg/ml Polysorbate 20
pH 6.2
25µg/ml Darbepoetin
0.05 mg/ml Polysorbate 80
2.12 mg/ml Sodium Phosphate monobasic monohydrate
0.66 mg/ml Sodium Phosphate dibasic anhydrous
8.18 mg/ml Sodium Chloride
pH 6.2 ± 0.2 0.05 mg/ml Polysorbate 80
2.12 mg/ml Sodium Phosphate monobasic monohydrate
0.66 mg/ml Sodium Phosphate dibasic anhydrous
8.18 mg/ml Sodium Chloride
pH 6.2 ± 0.2
300µg/ml G-CSF
0.59mg/ml Acetate
50mg/ml Sorbitol
0.03 mg/ml Polysorbate 80
0.035mg/ml Sodium
pH 4.0 0.59mg/ml Acetate
50mg/ml Sorbitol
0.03 mg/ml Polysorbate 80
0.035mg/ml Sodium
pH 4.0
Table 1: Depicts the various protein formulations employed and corresponding protein free formulation buffers employed for pre-treatment of the affinity resins.
Based on characteristics of the protein three different solid supports which exhibited considerable affinity to the protein were employed. For both the monoclonal antibody preparations a Mab Select Protein A affinity resin was used  whereas for G-CSF and darbepoietin  SP sepharose cation exchange and Capto adhere anion exchange resins were respectively used.
500µl of affinity resin slurry was taken in a microfuge tube  and allowed to settle under gravity. Excess liquid was discarded by aspiration. To this settled resin  about 1.5ml respective protein free buffers (described in table 1) was added. The incubation was carried out for about 3 minutes under gentle agitation. The resin was separated from the protein free buffer by centrifugation at 15000g for 5 minutes followed by discarding the supernatant. This process was repeated at least twice. Following incubation with the protein free buffer  the treated resin was now incubated with 500µl of respective protein formulations (Table 1). This was incubated for at least 15 minutes at room temperature followed by centrifugation at 15000g for 5 minutes. About 400µl of supernatant was aspirated  which was now considered the output buffer  which was used for estimation of the fractionated detergent. HPLC and colorimetry based methods were employed for estimation of polysorbate 80 and polysorbate 20 respectively. Only 400µl of the supernatant was recovered  out of which only 100µl of the supernatant was used to polysorbate quantification.
Table 2 enumerates the percentage recovery of polysorbates from different formulations; the percentage recovery was calculated by comparing expected and absolute amounts of polysorbates recovered in the output buffer.
Sample Polysorbate Type Quantified Value (µg) *Expected Value (µg) Percentage Recovery
Rituximab Polysorbate 80 70.5 70.0 100.73 - 101.16
Bevacizumab Polysorbate 20 45.5 47.5 93.68 - 95.79
Darbepoetin Polysorbate 80 5.2 5.0 104.00
G-CSF Polysorbate 80 3.1 3.0 104.67
Table 2: Percentage recovery of polysorbate from different protein formulations  after incubation with the treated affinity resin. *Expected value was calculated theoretically  based on polysorbate concentration present in the formulation buffer.
As described in table 2 about 95% to 104% of the polysorbate present in the protein formulation could be recovered when the resin was pre treated with a protein free formulation buffer  as opposed about 60% without pre treatment (Example 1). Further  the process of fractionation could be accomplished within about 30 minutes and all operations could be performed at room temperature. No protein was detected in the output buffer by either Bradford’s reagent or absorbance at 280nm  indicating efficient fractionation of polysorbate and protein.
Example 3
Resolution of Polysorbate Fractionation Method
To determine the resolution of this separation method  a darbepoetin formulation was employed. The protein free buffer was prepared and contained 0.045 mg/ml polysorbate 80  2.12 mg/ml sodium phosphate monobasic monohydrate  0.66 mg/ml sodium phosphate dibasic anhydrous  8.18 mg/ml sodium chloride and of pH 6.2 ± 0.2.
The protein formulation was prepared by adding 25µg/ml of darbepoetin to the protein free buffer described earlier. This protein formulation was diluted either 2 folds or 2.5 folds in water and the method as described in example 2 was followed. As described in table 3  after 2.5 and 2 fold dilution  1.8µg and 2.25µg of polysorbate was expected respectively in the output buffer. Following the method described in example 2 recovery of about 99 to 100% was achieved  indicating not only robustness of the assay  but also indicating that the method could resolve differences of 0.5µg of detergent present in a preparation. Lack of protein in the output buffer additionally indicated efficient fractionation.
Sample Polysorbate type Observed value (µg) *Expected value (µg) Percentage Recovery
2.5 fold dilution Polysorbate 80 1.79 1.8 99.4
2.0 fold dilution Polysorbate 80 2.27 2.25 100.8
Table 3: Percentage recovery of polysorbate upon dilution of protein samples up to 2 and 2.5 fold. Difference of at least 0.5µg of polysorbate could be fractionated by the described method. *Expected value was calculated theoretically  based on polysorbate concentration present in the formulation buffer.