Elsevier

Chemical Engineering Science

Volume 262, 23 November 2022, 118057
Chemical Engineering Science

Salting-Out crystallization of glycopeptide Vancomycin: Phase behavior study to control the crystal habit

https://doi.org/10.1016/j.ces.2022.118057Get rights and content

Highlights

  • Octahedral habit is predominant in salting out room temperature Van crystallization.

  • Needle habit formation can be avoided by control of pH and salt concentration.

  • Yield of batch crystallization of octahedral habit is higher than needle habit.

  • Needle and octahedral crystals exhibit distinct dissolution and thermal stability.

  • Needle and octahedral crystals are mostly similar in their characteristics.

Abstract

Needle-shaped crystals pose downstream processing issues in pharmaceutical manufacturing. Strategies to avoid needle-shaped crystal formation have been developed for small-molecule and protein pharmaceuticals. Controlling crystal habit of peptides, whose molecular weight lies between small molecules and proteins, has never been investigated. The present work studied phase behavior of salting-out room-temperature crystallization of glycopeptide vancomycin to determine crystallization conditions to avoid needle-shaped crystal formation. Batch crystallizations of prominent crystal habits identified from phase behavior study (i.e., needle and non-needle) were subsequently performed. Batch crystallization’s products were evaluated in their production yield/capacity, purity, size distribution, thermal stability, interfacial water content, dissolution profile, and antibiotic activity. The results showed octahedral crystals as the predominant habit produced at higher yield across the range of pH, salt/peptide concentrations investigated. Needle crystal formation was avoided by pH and salt concentration controls. Both crystal habits exhibited many similar properties, but with distinct thermal stability and dissolution profiles.

Introduction

Crystallization is a classical unit operation widely used in the purification of both small-molecule pharmaceuticals and large-molecule biologics (e.g., therapeutic proteins, monoclonal antibody, nucleic acids) (Pu and Hadinoto, 2020). In addition to crystal size, crystal habit (or external shape) has profound influences on the filterability (Bourcier et al., 2016), bulk density (Kim and Koo, 2019), adhesion tendency, which in turn affects flowability (Shah et al., 2014), and compactibility (Pudasaini et al., 2017) of the crystals produced. These properties greatly influence the downstream processing efficiencies in steps such as filtration, centrifugation, drying, tableting, as well as storage and handling (Azad et al., 2021). Moreover, crystal habit also influences the solubility, dissolution, and consequently the bioavailability and therapeutic activity of pharmaceuticals and biologics (Modi et al., 2013, Phan et al., 2021, Ren et al., 2019, Yin et al., 2008).

Needle habit as one of the more commonly produced habits in pharmaceutical crystallization (Cote et al., 2020) is notorious for causing several downstream processing issues and difficult handling. For example, needle crystals tend to align with the flow of the mother liquor, thus blocking the filter pores and they are prone to fracture during filtration creating unwanted fines (MacLeod and Muller, 2012, Steenweg et al., 2022). Aqueous dispersions of needle crystals also exhibit a high viscosity requiring a higher energy requirement to transport them to filtration unit (Wood, 2001). Needle crystals are brittle, prone to solvent inclusion (Meekes et al., 2003), and typically associated with low bulk density, poor tablet quality, and difficulty in loading the required dose into capsules (Black, 2019, Ghazi et al., 2019). For this reason, the design of industrial crystallization processes has been geared towards avoiding needle crystal formation (Lovette and Doherty, 2013).

Crystal habit is governed by the relative growth rates of individual faces of the crystals. Understanding crystal growth mechanism of a compound is essential in order to control its crystal habit to avoid the needle habit formation. The crystal growth rates are governed by both internal crystal structure and crystallization conditions (e.g., supersaturation, temperature, pH) (Lovette et al., 2008). Crystal morphological modellings based on crystal growth theories (e.g., Bravais-Friedel-Donnay-Harker (BFDH), periodic bond chain (PBC), and slice attachment energy theories) have been carried out to predict the crystal habit (Civati et al., 2021, Li et al., 2006). The crystal growth theories are typically coupled with kinetic models to incorporate the effect of crystallization conditions (Nayhouse et al., 2013, Winn and Doherty, 2002). From the crystal morphological modelling, organic small molecules can be categorized as either persistent (largely unavoidable), or controllable in their needle habit formation tendency (Civati et al., 2021). Moreover, the crystal morphological modelling has enabled researchers to develop guidelines on solvent selection to avoid the needle habit formation, which have been validated with experimental data (Lovette and Doherty, 2013, Taulelle et al., 2006, Tilbury et al., 2016).

Besides crystal morphological modelling, various crystal habit modification techniques have been explored experimentally (Hadjittofis et al., 2018). Solvent selection via its influence on the drug solubility has been the most widely investigated, particularly for small-molecule pharmaceuticals (Li et al., 2016). For example, the aspect ratio of anticholesterol drug lovastatin crystals could be lowered (less needle-like) by using less polar solvents (e.g., hexane, methylcyclohexane, ethyl acetate), in place of water–acetone mixture, or methanol used in the industrial crystallization of lovastatin (Hatcher et al., 2020, Turner et al., 2019). An opposite trend was observed in the crystallization of anti-inflammatory drug ibuprofen, where the use of low-polarity solvents resulted in needle crystals formation (Rasenack and Müller, 2002).

The crystal habit can also be modified by adding a growth inhibiting agent that is adsorbed onto the fast-growing crystal face. For example, the addition of hydrophobic polymers resulted in the formation of plate-like lovastatin crystals, in place of needle crystals produced without additives (Hatcher et al., 2020). The needle habit formation was suppressed by adding Polysorbate-80 surfactant and poly(sebacic anhydrite) in the crystallization of antihypertensive drug nifedipine (Kumar et al., 2015) and antifungal drug griseofulvin (Jarmer et al., 2005), respectively.

Another widely used experimental approach to modify the crystal habit is by precise manipulation of the supersaturation level via temperature cycling. For example, the aspect ratio of needle crystals of aspirin was successfully modified by multiple cycles of heating and cooling (Neugebauer et al., 2018). Successful crystal habit modifications by temperature cycling were also demonstrated in the crystallization of paracetamol (Lovette et al., 2012) and a proprietary active pharmaceutical ingredient (Eren et al., 2021). The needle habit formation can also be suppressed by simultaneous controls of multiple crystallization process variables (e.g., temperature, stirring rate, cooling rate, and seeding) as demonstrated in the crystallization of painkiller celecoxib (Banga et al., 2010).

Successful crystal habit modifications have also been demonstrated for bioactive macromolecules, particularly proteins, albeit they have not been as extensively studied as small-molecule pharmaceuticals. For example, needle habit formation of antimicrobial lysozyme could be avoided by (1) performing the crystallization under acidic condition and low temperature (Yu et al., 2015), (2) addition of ionic liquid as the habit modifier (Judge et al., 2009, Yu et al., 2019), and (3) use of crosslinked polymers as seeds (Grzesiak and Matzger, 2008). Needle habit formation of ovalbumin and catalase, on the other hand, was promoted at lower pH (Dumetz et al., 2008). In general, the aspect ratio of protein crystals was found to be most affected by pH and temperature (Dumetz et al., 2008, Liang et al., 2013).

Besides proteins, another clinically important class of bioactive macromolecules is peptides, whose molecular weight (MW) typically lies between small-molecule pharmaceuticals and proteins (i.e., 1 MW 5 kDa) (Lau and Dunn, 2018). Unlike proteins, crystallization of peptides for purification purposes has not been widely employed (Yu et al., 2020). Ultrafiltration membrane and chromatography, which pose problems of low throughput and high operational costs, remain predominantly employed in industrial purification of bioactive peptides (de Castro and Sato, 2015). The few studies on peptide crystallization were mostly aimed at crystal structure determination (Guo et al., 2021, Karle et al., 2003, Schäfer et al., 1996). To the best of our knowledge, modification of crystal habit in peptide crystallization had not been investigated before.

In the present work, we investigated the feasibility of avoiding needle habit formation in peptide crystallization using vancomycin hydrochloride (Van) - a glycopeptide antibiotic - as the model peptide. Van (MW 1.45 kDa) is widely used to treat infections caused by MRSA (methicillin-resistant Staphylococcus aureus), penicillin-resistant pneumococci, and to treat infections in patients who are allergic to penicillin and cephalosporins (Bruniera et al., 2015). Purification of Van, which is industrially produced by bacterial fermentation, requires multicycles of ion-exchange chromatography with pH adjustments, followed by antisolvent/salting-out crystallization as the final purification step (Lee et al., 2006).

While octahedral Van crystals were produced by hanging drop vapor diffusion crystallization (Schäfer et al., 1996), bulk crystallization of Van resulted in the undesirable needle crystals (Lee et al., 2010). Not unlike protein crystallization, the significant influences of pH and temperature were demonstrated in bulk Van crystallization performed in stirred vessels, where a narrow workable range of pH and temperature existed with pH 2.5 and 10 °C determined as the optimal condition (Lee et al., 2010). Bulk Van crystallization was slow needing 24 h to reach 95 % yield, despite combined cooling/salting-out/antisolvent (with acetone) crystallizations were employed. Even though the bulk crystallization rate could be enhanced by the addition of ionic liquid (Ha and Kim, 2015) and polymer seeds (Kim et al., 2011), needle crystals remained the predominant habit.

The first objective of the present work was to carry out a phase behavior study to determine crystallization conditions in which the needle habit could be avoided and to identify the predominant non-needle crystal habit produced. Recognizing the significant influences of solvent, pH, and temperature on the resultant crystal habit, we carried out the phase behavior study in conditions distinct from the ones employed in Lee et al. (2010). Specifically, we employed salting-out crystallization at room temperature using acetate buffer solution as the solvent, in contrast to simultaneous cooling/salting-out/antisolvent crystallization pursued in Lee et al. (2010). Phase behaviors of Van crystallization at different (i) pH, (ii) concentrations of Van and salt, and (iii) incubation time were examined in a high-throughput μ l-scale crystallization setup.

The second objective of the present work was to carry out batch crystallization of the predominant non-needle crystal habit identified in the phase behavior study (i.e., octahedral crystals). The octahedral Van crystals from the batch crystallization were characterized in terms of the (1) production yield and capacity, (2) purity, (3) crystal size distribution, (4) thermal stability, (5) interfacial water content, (6) dissolution characteristics, and lastly (7) antibiotic activity. The characteristics of the octahedral crystals were evaluated relative to the needle crystals, which prior to the present study had been the predominant crystal habit produced from batch crystallization of Van.

Section snippets

Materials

Van (United States Pharmacopeia (USP) grade, ≥ 900 μg/mg) was purchased from Duly Biotech Co. ltd. (Nanjing, China). Glycine (≥99 %), glacial acetic acid, potassium dihydrogen phosphate (KH2PO4), sodium hydroxide (NaOH), sodium acetate (≥99 %), ethanol (≥99.5 %), and paraffin oil (puriss., Ph. Eur.) were purchased from Sigma Aldrich (Singapore). Hydrochloric acid (HCl, 37 %) was purchased from VWR (Singapore). Mueller Hinton Broth (MHB) and phosphate buffered saline (PBS, pH 6.8) were purchased

Van solubility in buffer solutions of different pH and ionic strength

Prior to the phase behavior study, the solubility of Van in buffer solutions of different pH and NaCl concentrations ([NaCl]) was characterized to determine the supersaturation level in the phase behavior study. Van is a basic glycopeptide with pI of 8.1 (Horká et al., 2014). For basic proteins, several studies had advocated crystallization at pH below the pI to enhance the crystallization propensity (Charles et al., 2006, Kirkwood et al., 2015, Zhang et al., 2013). Furthermore, Van exhibited

Conclusions

The phase behavior study revealed the O crystals as the predominant crystal habit in salting-out room-temperature crystallization of Van. The O crystals were produced across the range of pH (2.6 to 5.6), Van (15 to 50 mg/mL), and NaCl concentrations (0.4 to 1.0 M) investigated. The needle crystal formation was largely avoided except at pH 5.6 and at NaCl concentrations in which the salting-in event took place (0.1 to 0.5 M). Increasing the incubation time beyond 24 h resulted in a wider range

CRediT authorship contribution statement

Siyu Pu: Writing – original draft, Writing – review & editing, Methodology, Data curation, Methodology, Investigation, Visualization. Kunn Hadinoto: Conceptualization, Funding acquisition, Supervision, Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

The authors would like to acknowledge the research funding from Ministry of Education Singapore (Grant No. AcRF Tier 1 RG82/20).

References (81)

  • D. Kumar et al.

    Effect of surfactant concentration on nifedipine crystal habit and its related pharmaceutical properties

    J. Cryst. Growth

    (2015)
  • J.L. Lau et al.

    Therapeutic peptides: Historical perspectives, current development trends, and future directions

    Bioorg. Med. Chem.

    (2018)
  • J. Li et al.

    A design aid for crystal growth engineering

    Prog. Mater Sci.

    (2016)
  • R.W. Maurer et al.

    Salting-in characteristics of globular proteins

    Biophys. Chem.

    (2011)
  • M. Nayhouse et al.

    Crystal shape modeling and control in protein crystal growth

    Chem. Eng. Sci.

    (2013)
  • R. Oswald et al.

    A view inside the nature of protein crystals

    J. Cryst. Growth

    (2017)
  • S. Pu et al.

    Continuous crystallization as a downstream processing step of pharmaceutical proteins: a review

    Chem. Eng. Res. Des.

    (2020)
  • S. Pu et al.

    Improving the reproducibility of size distribution of protein crystals produced in continuous slug flow crystallizer operated at short residence time

    Chem. Eng. Sci.

    (2021)
  • N. Pudasaini et al.

    Downstream processability of crystal habit-modified active pharmaceutical ingredient

    Org. Process Res. Dev.

    (2017)
  • M. Schäfer et al.

    Crystal structure of vancomycin

    Structure

    (1996)
  • U.V. Shah et al.

    Effect of crystal habits on the surface energy and cohesion of crystalline powders

    Int. J. Pharm.

    (2014)
  • T.D. Turner et al.

    Habit modification of the active pharmaceutical ingredient lovastatin through a predictive solvent selection approach

    J. Pharm. Sci.

    (2019)
  • P.G. Vekilov

    Nucleation and growth mechanisms of protein crystals

    Handbook Crystal Growth. Elsevier

    (2015)
  • D. Winn et al.

    Predicting the shape of organic crystals grown from polar solvents

    Chem. Eng. Sci.

    (2002)
  • A.K. Wöll et al.

    Analysis of phase behavior and morphology during freeze-thaw applications of lysozyme

    Int. J. Pharm.

    (2019)
  • W. Wood

    A bad (crystal) habit—and how it was overcome

    Powder Technol.

    (2001)
  • X. Yu et al.

    Development of magnetic solid phase extraction platform for the purification of bioactive γ-glutamyl peptides from garlic (Allium sativum)

    LWT

    (2020)
  • H. Yu et al.

    Dry powder inhaler formulation of high-payload antibiotic nanoparticle complex intended for bronchiectasis therapy: spray drying versus spray freeze drying preparation

    Int. J. Pharm.

    (2016)
  • X. Yu et al.

    Evaluating the role of ionic liquids (ILs) in the crystallization of lysozyme

    J. Mol. Liq.

    (2019)
  • P. Aleixa do Nascimento et al.

    A new ecological HPLC method for determination of vancomycin dosage form

    Curr. Chromatogr.

    (2020)
  • M.A. Azad et al.

    Impact of Critical Material Attributes (CMAs)-particle shape on miniature pharmaceutical unit operations

    Aaps Pharmscitech

    (2021)
  • S. Banga et al.

    Modification of the crystal habit of celecoxib for improved processability

    J. Pharm. Pharmacol.

    (2010)
  • S.N. Black

    Crystallization in the pharmaceutical industry

    Handbook Ind.strial Crystallization

    (2019)
  • F.R. Bruniera et al.

    The use of vancomycin with its therapeutic and adverse effects: a review

    Eur. Rev. Medical Pharmacol. Sci.

    (2015)
  • M. Charles et al.

    MPCD: a new interactive on-line crystallization data bank for screening strategies

    Acta Crystallogr. D Biol. Crystallogr.

    (2006)
  • Civati, F., O’Malley, C., Erxleben, A., McArdle, P., 2021. Factors Controlling Persistent Needle Crystal Growth: The...
  • A. Cote et al.

    Perspectives on the current state, challenges, and opportunities in pharmaceutical crystallization process development

    Cryst. Growth Des.

    (2020)
  • A. Eren et al.

    Experimental investigation of an integrated crystallization and wet-milling system with temperature cycling to control the size and aspect ratio of needle-shaped pharmaceutical crystals

    Cryst. Growth Des.

    (2021)
  • N. Ghazi et al.

    Investigating the effect of APAP crystals on tablet behavior manufactured by direct compression

    Aaps Pharmscitech

    (2019)
  • M. Giffard et al.

    Salting-in effects on urate oxidase crystal design

    Crystal Growth Des.

    (2008)
  • Cited by (6)

    • Habit modification in pharmaceutical crystallization: A review

      2024, Chemical Engineering Research and Design
    • A comparative study of antisolvent versus salting-out precipitations of glycopeptide vancomycin: Precipitation efficiency and product qualities

      2023, Powder Technology
      Citation Excerpt :

      In precipitation, the solubility of the target proteins (or peptides) is reduced by manipulating their environmental conditions (e.g., pH, ionic strength, dielectric constant), resulting in the formation of a new solid phase. The solid phase could be either amorphous, or crystalline with varying degree of crystallinity [11] depending on the precipitation kinetics, where slower kinetics lead to crystalline precipitates. Common precipitating agents include salts, organic solvents, and polymers.

    View full text