Excipient effect on phenol-induced precipitation of human growth hormone and bovine serum albumin

Abstract

The aim of this study was to investigate the impact of phenol on the precipitation of bovine serum albumin (BSA) and human growth hormone (hGH), in the presence of other excipients frequently used in biological drugs for parenteral delivery. The focus of the study lieson incompatibilities observed in multidose formulations containing non-ionic surfactants and preservatives. Previous research has shown that above a critical concentration, phenol reduces the cloud point of polysorbate surfactants to room temperature or lower.

Here, it is demonstrated that for BSA-polysorbate solutions, phenol-induced incompatibility is primarily controlled by this depression of the surfactant cloud point, resulting in turbidity and/or precipitation. However, for formulations with human growth hormone (hGH) in isotonic salt solutions, the precipitation mechanism is instead driven by protein-phenol interactions. The precipitation is affected by the concentration of sodium chloride and at low salt concentrations the incompatibility is again controlled by depression of the surfactant cloud point. The concentration of salt needed for protein induced precipitation seems to follow the Hofmeister series, with sodium chloride and sodium sulphate inducing precipitation at a lower salt concentration than sodium nitrate.

Notably, non-ionic tonicity agents, such as glucose and mannitol, which are known to impact the surfactant cloud point depression of phenol, do not induce precipitation of hGH in the presence of phenol. In the system containing polysorbate, phenol and hGH, salt-triggered protein precipitation occurs at slightly higher sodium chloride concentrations than in solutions without polysorbate. This indicates a stabilizing effect of polysorbate on hGH below the cloud point. However, the stabilising effect is surfactant dependent, and in the presence of dodecyl maltoside, hGH precipitation occurs at much lower sodium chloride concentrations than for solutions with polysorbates. This illustrates the complexity of the interplay of excipients with each other and with the active ingredient (the protein) in the development of multidose pharmaceutics.

Highlights

• Precipitation in multidose formulations can be due to clouding of polysorbate.
• Precipitation in multidose formulations can be due to phenol–protein interactions.
• The type of precipitation seen is dependent on the protein.
• The type of precipitation can shift when salt is added.
• Salt effects follow the Hofmeister series.

Introduction

In parenteral pharmaceutical products intended for multiple injections, the addition of preservatives assures the microbiological safety of the product (Hodges and Hanlon, 2000). However, it is well known that for formulations containing polysorbate surfactants, there is an incompatibility between these surfactants and the most commonly used preservatives (Ford et al., 2023, Gilbert et al., 2022, Rios, 2006, Rowe et al., 2006a; Stroppel et al., 2023, Zhi Chen, 2015). We have shown in previous work that this is mainly due to a cloud point depression of the surfactant (Wahlgren et al. 2024). This, in turn, means that there is a temperature and concentration range where the system does not cloud, where it is possible to formulate a stable product with acceptable preservative function. For example, systems that become cloudy at room temperature can be reversed by lowering the temperature through refrigeration. However, other excipients like salts and polyols also affect the clouding phenomenon. This is further complicated by the fact that when the phenolic preservative concentration is high enough, proteins are known to aggregate, independent of the presence of polysorbate (Arora et al., 2017, Bis and Mallela, 2014, Bis et al., 2015, Maa and Hsu, 1996, 2004 #62; Thirumangalathu et al., 2006). This aggregation has been linked to perturbation of the tertiary structure of the protein without secondary structure changes (Bis et al., 2015, Hutchings et al., 2013, Thirumangalathu et al., 2006, Zhang et al., 2004). Further, this perturbation does not lead to aggregation if the protein can be stabilised through electrostatic repulsion (Thirumangalathu et al., 2006). Maa and Hsu have shown that a range of phenolic compounds cause aggregation of human growth hormone (hGH) at a preservative concentration of 10 mg/ml (Maa and Hsu, 1996).

In this work two proteins are studied, bovine serum albumin (BSA) and human growth hormone (hGH). Bovine serum albumin was selected as it is a well-known protein used in many biotechnological applications (Xu et al., 2023). It is also known to interact with preservatives such as sodium benzoate and could thus potentially show incompatibility with phenol (Yu et al., 2019). Human growth hormone was investigated as it is commercially available and commonly used as a multidose product containing phenol or metacresol as preservatives. The preservative used is phenol. It is a common preservative for multidose presentations of proteins. It has a solubility in water around 8 wt% (Hill and Malisoff, 1926) which is well above the concentrations where it is used as a preservative which is 0.15–0.5 wt% (Rowe et al., 2006b).

Recombinantly produced human growth hormone is used in treatment for both adults and children. For children, it is used to increase growth when the child has conditions that affect normal growth and development (Hindmarsh and Dattani, 2006, Mehta and Hindmarsh, 2002). For adults, it has been used to treat HIV patients to increase body weight and physical endurance (Benedini et al., 2008), as well as to treat short bowel syndrome in adults who are receiving additional nutrition or fluids from intravenous (IV) therapy (Barahona-Garrido et al., 2009). The protein is given as a multidose presentation. There are several hGH products on the market. Most of these contain a combination of a preservative and a surface-active agent such as polysorbate 20 and 80 or poloxamer 188. The presentations also contain buffer salts and tonicity agents. The most commonly used preservatives in hGH products are phenol and metacresol. hGH is a small protein with 191 amino acid residues and a molecular weight of 22 kDa. It has an isoelectric point around pH 5 (Aloj and Edelhoch, 1972)). The structure of hGH from the Protein Data Bank (PDB) is given in Fig. 1.

Serum albumin is the most abundant protein in blood, and both human and bovine serum albumin have numerous uses in biomedical sciences (Xu et al., 2023) such as tissue engineering (Yuan et al., 2020), microspheres and nanospheres for drug delivery (Jun et al., 2011, Sokolik et al., 2018, Zhao et al., 2010), cryopreservation (Riel et al., 2011), and as surface active agent for blocking non-specific binding of other proteins to surfaces (Baldo et al., 1986). Serum albumin binds to a large range of molecules including drugs (Banerjee et al., 2017, Behera et al., 2023), fatty acids and surfactants (Banerjee et al., 2017). BSA contains 583 amino acids and have a molecular weight of 66 kDa (Riel et al., 2011). It has an isoelectric point of around 5 (Guo et al., 2016).

This study aims to investigate the (in)-compatibility of preservatives, in formulations of protein and surfactants, containing other common excipients like salts. Ion effects can be due to both screening of electrostatic forces and specific ion effects. To investigate this, the effects of salt concentration and type of ion were studied. The Hofmeister series of salts has, since the late 19th century, been used to study the specific ion effect (Hofmeister, 1888). It ranks ions based on their effect on macromolecules, primarily proteins. With chloride ions in the middle, the ions are divided into kosmotropes and chaotropes. The kosmotropes show a salting-out effect on proteins, increasing their stability, while chaotropes have a salting-in effect, increasing the protein solubility (Zhang and Cremer, 2006). Chaotropes have shown a destabilising effect on proteins which is linked to preferential binding and exclusion of salts from the protein-solution interface (Broering and Bommarius, 2005). The Hofmeister series has shown a systematic effect on several phenomena, such as the distribution of phenol into ionic liquids, (Asrami and Saien, 2018) and the clouding of surfactants (Dave and Joshi, 2018, Schott, 1984). The aim is to better understand the incompatibility seen in multidose formulations of proteins containing surfactants and preservatives (Stroppel et al., 2023).

Download the full article as PDF here: Excipient effect on phenol-induced precipitation of human growth hormone and bovine serum albumin

or read it here

Material

Recombinant human growth hormone (hGH) used in this study was a kind gift from Ferring Pharmaceuticals and polysorbate 80 (super-refined) was a gift from Croda (UK). BSA was supplied by Sigma Aldrich (>98 %) and was used without further purification. All other chemicals were of Pro Analysis grade and the water used was of Millipore grade.

Protein concentration was 10 mg/ml. The buffer used was 10 mM phosphate with a pH of 7.0 prepared with sodium phosphate monobasic and dibasic (Reag. Ph Eur, Sigma-Aldrich, US) and sodium azide (Reag. Ph Eur, Scharlab, Spain). This gave a pH of around 7.3 for the hGH solution and 7.0 for BSA. The surfactant concentration was if nothing else is stated, 0.1 wt%. The standard titrant was 1 M NaCl in 10 mM phosphate buffer.

Johanna Hjalte, Anna-Maria Börjesdotter, Carl Diehl, Stefan Ulvenlund, Marie Wahlgren, Helen Sjögren, Excipient effect on phenol-induced precipitation of human growth hormone and bovine serum albumin, International Journal of Pharmaceutics, Volume 676, 2025, 125624, ISSN 0378-5173, https://doi.org/10.1016/j.ijpharm.2025.125624.

---

Are you looking for excipients in commercial quantities?