Protein delivery: Thinking Outside the Nanoparticle

Protein delivery: Thinking Outside the Nanoparticle

Tags: , ,

By: Malgosia Pakulska, PhD, and Molly Schoichet, PhD, FRSC, O. Ont.

Malgosia Pakulska, PhD
Research Associate, Shoichet lab, Department of Chemical Engineering & Applied Chemistry, University of Toronto
Science Communication Specialist, Research2Reality

Molly Shoichet, PhD, FRSC, O. Ont
Professor, Department of Chemical Engineering & Applied Chemistry, University of Toronto
Professor, Biomaterials & Biochemical Engineering, University of Toronto
Tier 1 Canada Research Chair

It started with an odd result, as interesting discoveries often do. A “huh” moment that got the wheels turning.

My colleague Irja Elliott Donaghue, a fellow PhD student in the Shoichet lab at the University of Toronto, had been encapsulating a therapeutic protein, neurotrophic factor 3 (NT-3), in polymeric nanoparticles composed of poly(lactic-co-glycolic acid) (PLGA). By embedding those NT-3-loaded nanoparticles in a water swollen material called a hydrogel (think soft jello), she could control NT-3 release. But the release profile she obtained was not ideal–not enough NT-3 was coming out in the first few days.  

After some discussion, the proposed solution was to mix some free, unencapsulated NT-3 directly into the hydrogel. This seemed like a logical solution; proteins diffuse out of hydrogels very quickly due to their high water content. But it didn’t work. Instead, the protein release profile remained unchanged whether the NT-3 was completely encapsulated within the PLGA nanoparticles or whether some NT-3 was encapsulated and some was freely dispersed in the hydrogel.


This was especially surprising because researchers have been encapsulating drugs in polymeric nanoparticles for several decades and, if anything, they usually have the opposite problem.

Burst release

One of the first studies to use PLGA nanoparticles for drug delivery was back in 1981.1 Robert Gurny and colleagues showed that encapsulating testosterone in PLGA nanoparticles decreased its release rate by more than ten-fold compared to the conventional oil-based formula. Despite this, encapsulated testosterone still had a rapid, “burst” release over the first 12-24 hours, before it slowed down. The authors attributed this “burst” release to drug that wasn’t well encapsulated, but rather stuck to the nanoparticle surface. Since then, many papers about PLGA nanoparticles for drug delivery have been published, several devoted to overcoming this burst release and achieving a constant release profile.

In the Shoichet lab, we have been using PLGA nanoparticles in controlled drug release applications for two decades. One of the drug delivery systems we have designed is composed of PLGA nanoparticles dispersed in a hyaluronan (HA) methylcellulose (MC) hydrogel (HAMC).2 HA is shear thinning, allowing the gel to flow through fine gauge needles, while MC is inverse thermogelling, providing in situ gelation at body temperature. The hydrogel vehicle creates a depot of particles at the injection site, allowing them to degrade and release the encapsulated proteins locally over time. Embedding PLGA nanoparticles in HAMC has always attenuated the initial burst release,3 but not to the extent we were seeing now.  

Encapsulation free

While we were frustrated that our seemingly simple solution did not work, we were also curious. How could this be happening? We decided to go to the extreme: put all the NT-3 directly into the hydrogel with blank PLGA nanoparticles–that is don’t encapsulate any of it. Still, the release profile of NT-3 from the hydrogel remained unchanged whether or not it was encapsulated.

“It was definitely surprising! I needed to repeat the experiment to feel confident it was a real result,” laughs Elliott Donaghue, now a Commercialization Analyst at the Centre for Commercialization of Regenerative Medicine. But even after many replicates, the results stayed the same.

This was a big deal. “Not needing encapsulation reduces the loss of protein activity and more importantly simplifies the ability to control release rate,” says Professor Michael Sefton, a leading biomedical engineer at the University of Toronto who was not involved in the project.

For protein therapeutics, the encapsulation process itself can be very damaging. Sonication, organic solvents, freeze-drying–none of these things are good for protein structure which is so important for function. By skipping the encapsulation altogether, we can protect the fragile protein.

At the time, I was working on the controlled release of stromal cell derived factor 1α (SDF) to promote stem cell migration after spinal cord injury. Fascinatingly, I saw the same thing with SDF: the SDF release profile was slow and linear whether it was encapsulated in PLGA nanoparticles or simply mixed into the hydrogel with blank PLGA nanoparticles. Another PhD student in the lab, Jaclyn Obermeyer, was working on the controlled release of brain derived neurotrophic factor (BDNF) to promote tissue regeneration after stroke. She also observed that the release rate of BDNF was independent of encapsulation, but the presence of the PLGA nanoparticles in the hydrogel was still necessary.

We spent the next several months trying to figure out why this was happening and how we could control it.

“It wasn’t just one protein that could use this encapsulation-free system, but many proteins that had some similar properties. Those similarities helped us uncover the mechanism behind the release,” explains Elliott Donaghue.

It turns out NT-3, SDF, and BDNF are all positively charged at physiological pH while PLGA nanoparticles have a negative surface charge. When Anup Tuladhar, another PhD student in the Shoichet lab, tried to control the release of erythropoietin (EPO) using this same strategy, it didn’t work – EPO is negatively charged at physiological pH. By changing the pH of the hydrogel and the salt concentration, we strengthened the theory that release was governed by electrostatic adsorption of the positively charged proteins to the negatively charged surface of the particles.4 We could control the release rate by modifying the available surface area for adsorption through the particle size or concentration, or by modifying the strength of the interaction through the hydrogel pH.

In hindsight, it seems so simple. The saying “opposites attract” is something you learn with magnets in elementary school; yet, in over 35 years of PLGA research, no one had taken full advantage of this phenomenon. But the clues were out there.

Past studies investigated the effect of charge interactions on drug release from PLGA nanoparticles,5 but no one ever thought to skip encapsulation altogether. Similarly, PLGA nanoparticles have previously been exploited for surface loading of positively charged drugs, but in these cases, there was a very high concentration of protein which saturated the particle surface and masked the magnitude of the effect.6

Encapsulation in PLGA nanoparticles was a breakthrough for controlled drug release, but when it comes to proteins, it may be time to think outside the nanoparticle.


  1. Gurny R, Peppas NA, Harrington DD, et al. Development of biodegradable and injectable latices for controlled release of potent drugs. Drug development and Industrial Pharmacy 1981; 7(1):1-25.
  2. Baumann DB, Kang CE, Stanwick JC, et al. An injectable drug delivery platform for sustained combination therapy. Journal of Controlled Release 2009;138: 205-13.
  3. Stanwick JC, Baumann MD, Shoichet MS. Enhanced neurotrophin-3 bioactivity and release from a nanoparticle-loaded composite hydrogel. Journal of Controlled Release 2012;160: 666-75.
  4. Pakulska MM, Elliott Donaghue I, Obermeyer JM, et al. Encapsulation-free controlled release: electrostatic adsorption eliminates the need for protein encapsulation in PLGA nanoparticles. Science Advances 2016; 2: e1600519.
  5. Balmert SC, Zmolek AC, Glowacki AJ, et al. Positive Charge of “Sticky” Peptides and Proteins Impedes Release From Negatively Charged PLGA Matrices. J Mater Chem B Mater Biol Med. 2015; 3(23):4723-4734.
  6. Cai C, Bakowsky U, Rytting E, et al. Charged nanoparticles as protein delivery systems: A feasibility study using lysozyme as model protein. Eur. J. Pharm. Biopharm. 2008; 61: 31–42.