Saturday, November 27, 2010

FORMULATION - Protein Stability | Recombinant Albumin, a Robust Excipient



By Phuong Tran, PHD; Geoffrey Francis; Larissa Chirkova, PHD; and Sally Grosvenor Reduces excipients required in manufacturing process
The development and production of therapeutic proteins and peptides are rapidly expanding in the pharmaceutical industry, with the manufacture of monoclonal antibodies (mAbs) a primary focus. Currently, there are more than 25 approved mAbs worldwide; 240 therapeutics are in clinical studies, and 26 of these are in Phase III clinical trials.1 Recombinant protein molecules are attractive drug candidates due to their site-specific binding and effectiveness at low concentrations, factors which lead to fewer side effects.
During the manufacturing process, transport, and storage, the protein therapeutic can be exposed to a variety of stresses that promote protein instability and degradation. Protein instability can be the result of physical and chemical degradation. Physical protein degradation, such as aggregation, denaturation, and adsorption, is often the result of changes in temperature, shear stresses, and lyophilization, while oxidation of the protein during processing and storage is common. Instability of the protein therapeutic can occur in either the liquid or solid form and is dependent on the protein sequence, isoelectric point, hydrophobicity, and carbohydrate content.2 With the potential to increase the immunogenicity and decrease the efficacy and shelf life of the protein drug product, protein stability is a key issue in the final product.
Figure 1: Suppression of merozoite surface protein (MSP-2) protein aggregation in the presence of rAlbumin. rAlbumin suppressed aggregation of the MSP-2 protein (3.5 mg/ml) in a concentration-dependent manner following a single freeze–thaw cycle.
Figure 1: Suppression of merozoite surface protein (MSP-2) protein aggregation in the presence of rAlbumin. rAlbumin suppressed aggregation of the MSP-2 protein (3.5 mg/ml) in a concentration-dependent manner following a single freeze–thaw cycle.
To protect against degradation, protein therapeutics are usually formulated with excipients to provide the product with an acceptable shelf life for storage and shipping. As defined by the International Pharmaceutical Excipients Council, excipients are any substance other than the active drug that have been appropriately evaluated for safety and are included in a drug delivery system to: (1) aid processing of the system during manufacture; (2) protect, support, or enhance stability and/or bioavailability; (3) assist in product identification; or (4) enhance any other attribute of the overall safety and effectiveness of the drug product during storage and use.3
Human serum albumin (HSA) has been used as an excipient in a number of therapeutic protein formulations, including erythropoietin, antihemophilic factor, and interferon beta-1a.4 Because it is the most abundant protein in human blood, the potential for HSA to elicit an immunogenic response is minimal. However, due to regulatory concerns over the risk of blood-borne contaminants, such as prions or viruses, in these animal-derived products, formulation scientists have moved away from its use in drug formulation.
This article investigates the use of a recombinant human albumin (rAlbumin), expressed in Saccharomyces cerevisiae and manufactured to current good manufacturing practice, as an excipient and examines its ability to prevent or minimize physical and chemical degradation of drug substances in various test formulations. Drug substances, including transforming growth factor-β3 (TGF-β3), insulin-like growth factor–I (IGF-I), and a malarial antigen vaccine protein, were formulated in the presence of recombinant human albumin, and the effect on protein stability was examined.

Protects Against Aggregation

Aggregation, generally described as the association of misfolded proteins, is a major problem encountered during the manufacturing process of therapeutic proteins, resulting in significant product loss and potentially increasing the immunogenicity of the drug product. There are numerous process operations during which protein aggregation can occur, such as refolding, purification, mixing, freeze-thawing, freeze-drying, and reconstitution. Aggregation can also occur during transport and storage. The formation of these aggregates is generally concentration dependent, which is a particular challenge for protein therapeutics formulated at high concentrations.
Figure 2: The effect of rAlbumin in protecting insulin-like growth factor-I (IGF-I) against oxidation in a concentration-dependent manner following exposure to H2O2.
Figure 2: The effect of rAlbumin in protecting insulin-like growth factor-I (IGF-I) against oxidation in a concentration-dependent manner following exposure to H2O2.
In this study, rAlbumin was evaluated for its ability to suppress amyloid fibril formation by the merozoite surface protein (MSP-2) after a single freeze-thaw cycle. Amyloid-like fibrils are measured by the effect on light scattering at λ 320 nm. A range of rAlbumin concentrations were evaluated for their ability to suppress aggregation of MSP-2. MSP-2 was chosen as the model to investigate aggregation due to its tendency to form amyloid-like fibril aggregates.5,6
At various concentrations, rAlbumin was dissolved in a solution of phosphate buffered saline (PBS); the MSP-2 protein (3.5 mg/ml) was then added to all samples, followed by a single freeze-thaw cycle. Freeze-thawing is one of the numerous process conditions during which protein aggregation can occur.2 Samples were then plated in a 96-well plate and stored at 2°C to 8°C. Absorbance readings were taken at λ 320 nm at multiple time intervals over a five-day period.
A variety of excipients in common use within the industry to improve protein stability were also compared with rAlbumin for their ability to inhibit protein aggregation. The excipients rAlbumin (15.0 mg/ml), glycine (20.0 mg/ml), PEG 400 (1.0 mg/ml), polysorbate 80 (0.82 mg/ml), and polysorbate 80 (8.2 mg/ml) were tested in the same model described above. Absorbance readings were taken at multiple time intervals at λ 320 nm. Results indicated that aggregation was suppressed by 50% at a 1:1 molar ratio of the antigen to rAlbumin and reduced by 80% in the presence of the highest concentration of rAlbumin (see Figure 1, p. 23). rAlbumin also suppressed aggregation of the MSP-2 antigen to a greater extent compared with other commercially available excipients when the antigen was formulated in PBS pH 6.4 (see Figure 2, p. 24).
Although surfactants like polysorbate 80 have proven beneficial during the manufacturing process by reducing stress-induced aggregation, they are also known to adversely affect protein stability during storage by acting as a pro-oxidant and increasing the oxidation of the therapeutic drug substance.7
Figure 3: The effect of rAlbumin compared to polysorbate 80 in preventing the nonspecific binding of transformng growth factor-β3 (TGF-β3) to plastic polypropylene container.
Click to Enlarge
Figure 3: The effect of rAlbumin compared to polysorbate 80 in preventing the nonspecific binding of transformng growth factor-β3 (TGF-β3) to plastic polypropylene container.
The mechanism by which albumin inhibits aggregation is not well understood. HSA is known to sequester >90% of the Alzheimer’s disease-related peptides Aβ (1-40) and Aβ (1-42) in blood serum, presumably affecting the ability of the Aβ peptides to aggregate.8 HSA is also known to bind and transport metal ions such as Cu. Complexes between Cu and Aβ peptides are involved in Aβ aggregation; therefore, HSA is capable of reducing Cu-induced aggregation.9

rAlbumin Acts as Antioxidant

Oxidation of a protein therapeutic is a problem for manufacturers, particularly during storage. Factors that can affect the oxidation rate of proteins during storage include oxygen (head space), light, the physical state of the product, and temperature. Modifications to proteins through oxidation can lead to a range of functional consequences, such as altered binding activities, increased susceptibility to aggregation and proteolysis, increased or decreased uptake by cells, and altered immunogenicity. It is for these reasons that Food and drug Administration guidelines suggest that oxidation must be controlled in the product formulation of therapeutic proteins. During storage, the methionine residues are often the most susceptible to oxidation; formulation excipients are often used to protect the protein from this oxidation.
IGF-I, an important anabolic growth factor, is susceptible to oxidation, particularly during storage, and was therefore chosen as an appropriate model to investigate rAlbumin’s ability to inhibit oxidation.10 To test the functionality of rAlbumin as an antioxidant, pharmaceutically relevant conditions in protein oxidation were modeled using trace amounts of the oxidizing agent hydrogen peroxide (H2O2). Both rAlbumin (0, 10.0, 15.0, and 20.0 mg/ml) and L-methionine (0, 0.1, 0.2, and 0.3 mg/ml) were dissolved in solutions of PBS buffer. The IGF-I protein (20 µg/ml) was then added to all samples, followed by H2O2 to a final concentration of 0.0005%, and incubated for eight hours at 37°C. The reaction was terminated with catalase, and degree of oxidation was analyzed by reverse-phase high performance liquid chromatography (HPLC). The percentage of oxidized IGF-I was calculated against the main IGF-I peak for all samples. Using hydrogen peroxide as the agent, oxidation of IGF-I was shown to be significantly reduced by the presence of increasing concentrations of rAlbumin (see Figure 3, p. 24). At the highest concentration, oxidation of IGF-I was reduced by 93%. It is noteworthy that the initial IGF-I sample already contained 11.6% of oxidized form due to storage alone.
The ability of rAlbumin to act as an antioxidant following exposure to hydrogen peroxide was also compared to the commonly used antioxidant L-methionine (see Figure 4, p. 25). The oxidative protection of IGF-I by rAlbumin was achieved at molar concentrations ~ 13-fold less than that of L-methionine.
Formulation of therapeutic proteins and peptides that provide optimal product stability during process, storage, and shipping is essential for the biopharmaceutical manufacturer.
HSA is known to have an antioxidant function, primarily due to a single free thiol at position Cys 34, and this single thiol of human serum albumin acts as a potential scavenger for reactive oxygen and nitrogen species. As an antioxidant, albumin also binds and transports metal ions, such as Cu2+ and Fe3+, thus reducing the availability of these ions to cause oxidation.11

Nonspecific Adsorption Reduced by rAlbumin

Like most proteins, protein therapeutics are susceptible to nonspecific adsorption to various surfaces. This loss of product can significantly decrease the concentration in solution, altering the efficacy of the drug. In addition to product loss, protein adsorption can lead to structural change, denaturation, and inactivation due to aggregation. Nonspecific protein adsorption is a particular problem for liquid product at low concentrations. Many surfaces the drug product can come into contact with during the manufacturing process lead to product loss due to nonspecific adsorption. These include, but are not limited to, delivery pumps, silicone tubing, and glass and plastic containers.
HSA has been used as a blocking agent to prevent therapeutic proteins from binding to various surfaces. The mechanism is not well understood, but albumin is believed to bind to charged surfaces through opposite charged functional groups on the molecule. Hydrophobic interactions that occur are at lower strength and are more easily reversible.12 TGF-β3, an active pharmaceutical ingredient in scarless wound healing, is a hydrophobic protein with a propensity to adsorb to container surfaces. The percentage loss of TGF-β3 due to nonspecific binding to polypropylene or glass vial surfaces in the absence and presence of rAlbumin was examined.
To test the effectiveness of this model, TGF-β3 (0.5–60 µg/ml) was added to a polypropylene or glass container containing citrate buffer pH 3.6. Each sample was mixed and centrifuged for three minutes at 13,500 rpm. Samples were transferred to HPLC vials for analysis via reverse-phase HPLC. Percentage recoveries of TGF-β3 were calculated against the TGF-β3 reference standard. rAlbumin was then assessed for its ability to prevent the loss of TGF-β3 to the container surface. Citrate buffer pH 3.6 was added to a polypropylene test container, followed by rAlbumin (0–0.5 mg/ml), then TGF-β3 (0.2 µg/ml). Each sample was mixed and centrifuged as described above. rAlbumin (0.1 mg/ml) was also compared against polysorbate 80 (0.1 mg/ml), using the method described above, for ability to protect TGF-β3 (0.2–1.0 µg/ml) against nonspecific adsorption to glass surfaces.
Figure 4: The effect of rAlbumin (0.1 mg/ml) compared to polysorbate 80 (0.1 mg/ml) in preventing the nonspecific binding of transformng growth factor-β3 (TGF-β3) (0.2–1.0 µg/ml) to glass vials.
Click to Enlarge
Figure 4: The effect of rAlbumin (0.1 mg/ml) compared to polysorbate 80 (0.1 mg/ml) in preventing the nonspecific binding of transformng growth factor-β3 (TGF-β3) (0.2–1.0 µg/ml) to glass vials.
In this study, rAlbumin significantly reduced protein loss that occurred due to the nonspecific binding of TGF-β3 to glass and plastic. In the absence of rAlbumin, nonspecific binding of TGF-β3 increased progressively at concentrations less than 60 µg/ml, and the recovery of the protein was significantly reduced at lower concentrations. In the presence of rAlbumin, however, the nonspecific binding of TGF-β3 to vessel surfaces was minimal, with >95% recovery achieved using just 0.05 mg/ml of rAlbumin. The benefit of adding rAlbumin was then compared to that of polysorbate 80 in preventing nonspecific binding of proteins to plastic and glass surfaces. Polysorbate 80 is widely used in the formulation of biotherapeutic products to address aggregation and nonspecific binding, but its suitability as a protein stabilizer raises certain concerns, because it is a potential source of peroxides. In this comparison, rAlbumin prevented the nonspecific binding of TGF-β3 at least as well as polysorbate 80 in plastic containers and significantly better in glass vials.

Multipurpose Excipient

Formulation of therapeutic proteins and peptides that provide optimal product stability during process, storage, and shipping is essential for the biopharmaceutical manufacturer. Numerous excipients that reduce protein degradation, which occurs through physical or chemical pathways, are in common use within the industry. Generally, although each excipient has a specific function in stabilizing a protein substance, it may also stimulate and enhance alternate pathways of protein degradation.
In this study, rAlbumin was found to be an effective multipurpose excipient. rAlbumin protects proteins against aggregation, especially amyloid-like fibril products, acts as an antioxidant in preventing protein oxidation, and serves as a blocking agent to prevent nonspecific adsorption to surfaces. rAlbumin reduces the total number of excipients required, simplifying the formulation strategy. Further, it could be argued that the use of a single or reduced number of excipients to formulate drugs may accelerate development time by more quickly allowing more “universal” excipient mixes to be derived—and that rAlbumin could facilitate such an approach.
Dr. Tran is senior scientist in the development and discovery group, Francis is chief scientist, Dr. Chirkova is discovery and development manager, and Grosvenor is scientific communications manager at Novozymes Biopharma, Adelaide, Australia. For more information, go to www.biopharma.novozymes.com or contact Phuong Tran at phut@novozymes.com.

References

  1. Reichert JM. Antibodies to watch in 2010. MAbs. 2010;2(1):84-100.
  2. Bogard WC Jr., Dean RT, Deo Y, et al. Practical considerations in the production, purification, and formulation of monoclonal antibodies for immunoscintigraphy and immunotherapy. Semin Nucl Med. 1989;19(3):202-220.
  3. International Pharmaceutical Excipients Council (IPEC), Pharmaceutical Quality Group (PQG). The Joint-PQG Good Manufacturing Practices Guide for Pharmaceutical Excipients. 2006. Available at: http://ipecamericas.org/node/132. Accessed September 25, 2010.
  4. Berezenko S. Formulation of biotherapeutics: avoiding human and animal excipients. Eur BioPharm Rev. Summer 2005.
  5. Wang W, Singh S, Zeng DL, et al. Antibody structure, instability, and formulation. J Pharm Sci. 2007;96(1):1-26.
  6. Kueltzo LA, Wang W, Randolph TW, et al. Effects of solution conditions, processing parameters, and container materials on aggregation of a monoclonal antibody during freeze-thawing. J Pharm Sci. 2008;97(5):1801-1812.
  7. Wang W, Wang YJ, Wang DQ. Dual effects of Tween 80 on protein stability. Int J Pharm. 2008;347(1-2):31-38.
  8. Milojevic J, Raditsis A, Melacini G. Human serum albumin inhibits Abeta fibrillization through a “monomer-competitor” mechanism. Biophys J. 2009;97(9):2585-2594.
  9. Perrone L, Mothes E, Vignes M, et al. Copper transfer from Cu-Abeta to human serum albumin inhibits aggregation, radical production and reduces Abeta toxicity. Chembiochem. 2010;11(1):110-118.
  10. Fransson JR. Oxidation of human insulin-like growth factor I in formulation studies. 3. Factorial experiments of the effects of ferric ions, EDTA, and visible light on methionine oxidation and covalent aggregation in aqueous solution. J Pharm Sci. 1997;86(9):1046-1050.
  11. Fransson J, Hagman A. Oxidation of human insulin-like growth factor I in formulation studies, II. Effects of oxygen, visible light, and phosphate on methionine oxidation in aqueous solution and evaluation of possible mechanisms. Pharm Res. 1996;13(10):1476-1481.
  12. Jeyachandran YL, Mielczarski E, Rai B, et al. Quantitative and qualitative evaluation of adsorption/desorption of bovine serum albumin on hydrophilic and hydrophobic surfaces. Langmuir. 2009;25(19):11614-11620.

1 comment:

Subhash said...

thank you sir for taking time to share your expertise on this blog