Remote Key Loading

[Channing Rupnick]

Thankyou for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Poly(lactic-co-glycolic acid) (PLGA) long-acting release depots are effective for extending the duration of action of peptide drugs. We describe efficient organic-solvent-free remote encapsulation based on the capacity of common uncapped PLGA to bind and absorb into the polymer phase net positively charged peptides from aqueous solution after short exposure at modest temperature. Leuprolide encapsulated by this approach in low-molecular-weight PLGA 75/25 microspheres slowly and continuously released peptide for over 56 days in vitro and suppressed testosterone production in rats in an equivalent manner as the 1-month Lupron Depot. The technique is generalizable to encapsulate a number of net cationic peptides of various size, including octreotide, with competitive loading and encapsulation efficiencies to traditional methods. In certain cases, in vitro and in vivo performance of remote-loaded PLGA microspheres exceeded that relative to marketed products. Remote absorption encapsulation further removes the need for a critical organic solvent removal step after encapsulation, allowing for simple and cost-effective sterilization of the drug-free microspheres before encapsulation of the peptide.

Peptide drugs with molecular weights from a few hundred to a few thousand Daltons form a unique class of drugs with both unique mechanisms of the pharmacological action and physical-chemical properties1. Once beyond a few amino acids peptides are often difficult to deliver to the body owing to poor bioavailability by noninvasive routes of drug administration and short serum half-lives2. Two common methods to improve systemic exposure and minimize injections of peptides include half-life extension via covalent modification and microencapsulation in biodegradable PLGA systems3,4. Microencapsulation has the advantages of much longer intervals between injections (>weeks to months) and no new active pharmaceutical ingredient (API) needs to be developed, thus reducing regulatory obstacles.

Conventional microencapsulation approaches to manufacture PLGA microspheres include solvent evaporation, coacervation, and spray-drying. In each of these methods, the API is combined with PLGA dissolved in an organic solvent before forming microspheres. This combination creates a number of undesirable issues: (a) the peptide-loaded microspheres most often cannot be terminally sterilized, thus requiring expensive aseptic processing with organic solvents and numerous unit operations; (b) the API is expensive, and therefore it is undesired to discard poorly formed drug-polymer microspheres (tiny fines, large aggregated microspheres, or debris on mixing equipment) and yields can be far less than 100%; (c) the complexity of unit operations and components necessary to form microspheres can be problematic to scale-up to large-scale manufacturing; (d) some products have one or more residual organic solvents, which pose challenges to storage stability of the final products;5 (e) there is little opportunity to manipulate the polymer structure once the peptide-PLGA matrix is formed, limiting the ability to engineer release kinetics6,7; and (f) mixing organic solvent/water mixtures in the presence of peptides, particularly with higher-order structure, can be detrimental to drug stability8.

Mechanistic analysis of peptide/PLGA interactions revealed a plausible solution to many of the above difficulties. Recently we discovered that two of the peptides currently used in commercial PLGA long-acting release depots (LARs), leuprolide and octreotide, long known to bind to the surface of uncapped PLGA (PLGA-COOH), can in fact be absorbed rapidly into the polymer phase of PLGA-COOH at high drug content with peptide desorption slow enough for potential long-term controlled release9. Organic solvent-free remote absorption encapsulation of aqueous peptide with drug-free PLGA-COOH particles could provide a number of clear advantages relative to conventional encapsulation approaches: (a) the process involves simple equilibration of peptides with the polymer particles and gentle mixing at modest temperature, bypassing complex unit operations and kinetic variations of particle formation and drying; (b) no organic solvent is used during encapsulation providing additional control over undesired residual solvents in the final product; (c) as the polymer particles are pre-formed before peptide exposure there is an opportunity to (i) sterilize the particles (e.g., via gamma irradiation) and (ii) pretest encapsulation on a small scale in order to reduce costs of goods associated with aseptic large-scale manufacture with organic solvents, and (iii) adjust the properties of the polymer particles (e.g., size distribution, porosity, residual organic solvent) before encapsulation for improved control of product performance; and (d) similar to use of remote loading liposomes10 and commercial transfecting agents11,12 sterile pre-formed microspheres are capable of encapsulating peptides at essentially 100% yields and with far less peptide needed compared to conventional encapsulation methods where even bench-scale procedures often require tens of milligrams of drug.

Remote loading based on absorption in the polymer phase is very different from a prior technique13 where macromolecular drugs, including peptides, were encapsulated by self-healing (or closing) of the aqueous pores in typical aliphatic ester end-capped PLGAs, which do not absorb peptides into the polymer phase9,14. Because of the minimal interaction between the peptide and end-capped PLGA14, encapsulation efficiency in porous PLGA is extremely low14 (e.g.,

Here, we demonstrate a self-encapsulating PLGA microsphere, formulated from the same PLGA 75/25 employed in the well-known 1-month Lupron Depot, which can be mixed with modest concentrations of aqueous peptide solutions to absorb peptides and form highly desirable peptide LARs. These formulations are capable of encapsulating a number of peptides at high efficiency and high drug loading, and then releasing the drug continuously with an acceptable initial burst release in vitro and in vivo.

ITC curve fitting predicted binding as enthalpically favorable for peptides with ε-amino groups (Lys side chains in octreotide and calcitonin) indicative of strongly favorable intermolecular forces (e.g., a combination of ionic and H-bonding interactions)19,20. Entropically driven binding was determined for peptides with cationic Arg residues but no ε-amino groups (leuprolide and vasopressin), indicative of desolvation-driven interactions21. Cationic pramlintide, with significantly higher order structure22, also displayed high loading (9.8%) and binding to PLGA-COOH via ITC (Supplementary Fig. 5), although binding showed some anomalous behavior, prohibiting estimates of binding energies. Exenatide and protirelin displayed very low binding tendencies, as expected by their absence of positive charge (Supplementary Fig. 5). As with leuprolide, the 1-day remote absorption encapsulation procedure yielded PLGA microspheres that slowly released cationic peptides. Octreotide was released for >8 weeks after remote loading by desirable zero-order kinetics with minimal peptide instability, in contrast with the existing 1-month commercial Sandostatin LAR (Fig. 2b), which releases by sigmoidal behavior with substantial peptide acylation and loss of parent drug. Vasopressin and salmon calcitonin were also slowly and continuously released (Fig. 2c, d), with the former showing behavior consistent with peptide damage8 in the PLGA carrier and the latter only releasing

While the mechanism of absorption of peptides in PLGA-COOH is not fully understood, the available data strongly suggests that the cationic peptide forms a salt with surface carboxylate anions before moving into the highly swollen and mobile polymer phase as a peptide-PLGA ion pair. We previously verified the deep entry of peptides in the polymer phase by multiple orthogonal techniques, including stimulated Raman scattering and laser confocal fluorescence imagining, and serial microtome sectioning of thin PLGA films9. The peptide absorption was strongly inhibited or did not occur when the temperature was reduced to inhibit the mobility of the polymer, the pH of the solution was reduced to decrease the ionization of the PLGA carboxylic acid end groups, or by end-capping the PLGA that also removes polymer chain ionization9. The isothermal calorimetry binding of peptide and PLGA in DMSO (Table 1, Fig. 4 and Supplementary Fig. 5) also distinguished between peptides that were successfully remote loaded in the polymer and those that were not (Supplementary Table 2). The binding by this technique was found to be either enthalpically favored for amine-containing peptides and entropically favored for those peptides with positive charges resulting from arginine residues (Table 1).

In closing, the results here are significant in that we have demonstrated a remarkably easy, efficient, and scalable way of creating an effective 1-month LAR formulation for leuprolide and demonstrated the potential for other important peptides for controlled release (e.g., octreotide). Imagine the simple formulation of drug-free microspheres that are terminally sterilized and used to encapsulate peptides by a single aseptic aqueous mixing step. It is reasonable to expect that utilizing this approach could lead to a significantly reduced cost of goods relative to existing commercial formulations (e.g., Lupron Depot and Sandostatin LAR). Large scale encapsulation could be tested for a given microsphere batch on a small scale first to reduce the risk of large-scale batch failure and ensure final product performance. This simple encapsulation paradigm obviates (a) the complex steps necessary to prevent peptide loss during encapsulation by the solvent evaporation method33,43, (b) the use of oil and multiple organic solvents during the coacervation method43,44, and (c) low yields experienced by spray drying44,45. More work is needed to expand the potential of this encapsulation approach, namely, to determine (a) optimal drug-free microsphere formulation, encapsulation and sterilization conditions, (b) how to improve peptide instability in PLGA and related polymers, (c) how to apply the approach to peptides with neutral to negative net charge and other important charged biomacromolecules like nucleic acid drugs, and (d) how to incorporate additional functionality to the surface of the polymer before encapsulation. With these initiatives many more commercial peptide LARs can be envisioned in the future46.

3a8082e126