Adrian V.
Hauck
a,
Patric
Komforth
b,
Jessica
Erlenbusch
c,
Judith
Stickdorn
b,
Krzysztof
Radacki
d,
Holger
Braunschweig
d,
Pol
Besenius
c,
Simon
Van Herck
e and
Lutz
Nuhn
*a
aChair of Macromolecular Chemistry, Institute of Functional Materials and Biofabrication, Julius-Maximilians-Universität Würzburg, 97070 Würzburg, Germany. E-mail: lutz.nuhn@uni-wuerzburg.de
bMax Planck Institute for Polymer Research, 55128 Mainz, Germany
cDepartment of Chemistry, Johannes-Gutenberg-Universität Mainz, 55122 Mainz, Germany
dInstitute for Sustainable Chemistry and Catalysis with Boron, Julius-Maximilians-Universität Würzburg, 97074 Würzburg, Germany
eMeinig School of Biomedical Engineering, Cornell University, Ithaca, NY 14850, USA
First published on 8th November 2024
Pharmacokinetics and biodistribution profiles of active substances are crucial aspects for their safe and successful administration. Since many immunogenic compounds do not meet all requirements for safe and effective administration, well-defined drug nanocarrier systems are necessary with a stimuli-responsive drug-release profile. For this purpose, a novel pH-responsive aliphatic cyclic carbonate is introduced with benzyl ketal side chains and polymerized onto a poly(ethylene glycol) macroinitiator. The resulting block copolymers could be formulated via a solvent-evaporation method into well-defined polymeric micelles. The hydrophobic carbonate block was equipped with an acid degradable ketal side group that served as an acid-responsive functional group. Already subtle pH alternations led to micelle disassembly and the release of the active cargo. Furthermore, basic carbonate backbone degradation assured the pH responsiveness of the nanocarriers in both acidic and basic conditions. To investigate the delivery capacity of polymeric micelles, the model small molecule compound CL075, which serves as an immunotherapeutic TLR7/8 agonist, was encapsulated. Incubation studies with human blood plasma revealed the absence of undesirable protein adsorption on the drug-loaded nanoparticles. Furthermore, in vitro applications confirmed cell uptake of the nanodrug formulations by macrophages and the induction of payload-mediated immune stimulation. Altogether, these results underline the huge potential of the developed multi-pH-responsive polymeric nanocarrier for immunodrug delivery.
With tailored sizes between 10 and 100 nm, synthetic polymeric nanoparticles have been reported to enter tumor tissue through leaky blood vessels and remain there owing to ineffective lymphatic draining.6,18,19 This concept, known as the enhanced penetration and retention (EPR) effect, is difficult to translate from in vivo models to applications in the human body because of the high variability between individual patients.20 Nevertheless, a few approved polymeric nanodrug formulations, including DOXIL®, Marqibo® and Genexol®, have demonstrated effectiveness, according to their relevant EPR.21–23 To further enhance this effect, it is necessary to facilitate nanocarriers with stimulus responsiveness that would allow drug release exclusively at the target site. Environmental changes in the tumor microenvironment, such as an increase in reactive oxygen species (ROS) and redox potential as well as a decrease in oxygen and pH levels, seem to be highly attractive triggers.24
Reduced pH gradients are also present upon nanoparticle cell internalization inside endosomes (pH 6.0–5.5) and lysosomes (pH 5.0–4.5).25,26 Acid-responsive functionalities include hydrazones,27 orthoesters, oximes,28 imines,29 acetals,30 and ketals,31 to name a few. Ketals are often used as a linker to modify functional polymer backbones and achieve acid-responsive micelles or nanogels, since they are known to be stable at physiological pH values but degrade within a suitable time window upon acidification.32–35 The degradation kinetics of ketals can be fine-tuned by regulatory neighboring groups, allowing precise adjustment to the given requirements.36 Another advantage of a ketal trigger compared to other pH-sensitive triggers is the release of nontoxic small molecules, like acetone, which is listed on the FDA GRAS (generally recognized as safe) list.37
Herein, we report the development of an innovative biodegradable nanocarrier based on block copolymers with pH-responsive amphiphilicity. Attached to a gradually hydrolysable aliphatic polycarbonate backbone, acid degradable ketal side groups assure pH-responsive drug formulations that ultimately release their active cargo upon acidification (as found already in the tumor microenvironment or later on during endocytosis, Fig. 1). Besides acid degradability of the micelle-stabilizing side groups, such as gradual hydrolytic carbonate backbone degradation, which is generally accelerated under basic conditions, provides a dual-pH responsiveness in both acidic and basic directions.
To access such carriers, first the synthesis and characterization of a cyclic carbonate monomer bearing an acid degradable ketal side group was demonstrated. Subsequently, this monomer was polymerized under controlled ring-opening polymerization (ROP) conditions into well-defined aliphatic polycarbonates. The acid responsiveness of these polymers was investigated by 1H nuclear magnetic resonance (NMR) diffusion-ordered spectroscopy (DOSY) and size-exclusion chromatography (SEC). Chain extension of the poly(ethylene glycol) yielded amphiphilic block copolymers that could be formulated into narrowly dispersed polymeric micelles. A benzyl moiety stabilizes these micelles via π–π-stacking interactions.38,39 Since this benzyl group is linked to the polymers backbone via the ketal functionality, acidification led to rapid micelle disassembly. This degradation process was intensively investigated in a wide range of pH levels by dynamic light scattering (DLS) and UV/vis spectroscopy. Since the polymeric micelles’ hydrophobic core allows for the encapsulation of water-insoluble pharmaceutically active ingredients, these micelles can be used as acid-responsive nano-sized drug carriers. The encapsulation of the model compound Nile red revealed selective cargo release exclusively upon micelle degradation, which could be triggered under both acidic as well as basic conditions. Furthermore, drug loading was investigated using the hydrophobic Toll-like-receptor 7/8 agonist CL075. Its stimulus-responsive release in RAW-Blue macrophages demonstrated the polymeric micelles remarkable delivery performance as a multi-pH-responsive biodegradable immunodrug nanocarrier.
In parallel, ethylene glycol (4) was asymmetrically protected with trimethyl orthoacetate (5) following a standard procedure via a cyclic five-membered orthoacetate ring.40,41 Hydrolysis led to the ethylene glycol monoacetate (EGMA, 6), again in a good yield (91%). The transacetalization of BisBnKetal (3) with EGMA (6) yielded the side group component as an acetate-protected alcohol (7). Since this type of transacetalization is now a more statistic exchange of benzyl alcohol (2) and EGMA (6), a product mixture consisting of the two symmetric ketals BisBnKetal (3) and Bis(EGMA)Ketal and the desired asymmetric product (7) was inevitable. Nonetheless, careful reaction performance allowed accessing yields up to 46%. The basic deprotection of (7) afforded the alcohol 2-hydroxyethoxy-2-benzyloxypropane (8) almost quantitatively.
In a second synthesis path, 5-methyl-1,3-dioxane-2-oxo-5-carboxylic acid (MTC-OH, 13) was obtained by a three-step synthesis following a procedure established by Al-Azemi and Hedrick et al.42–44 First, bis-MPA (9) was protected with benzyl bromide (10) for its acid functionality. Subsequently, the six-membered cyclic carbonate could be formed with ethylene chloroformate. Finally, reductive benzyl ester cleavage via Pd/C catalyzed hydrogenation led to MTC-OH (13) in an overall yield of 55% (recently, a one-pot synthesis strategy toward MTC-OH (13) has been reported without the use of toxic ethylene chloroformate, opening future opportunities for safer syntheses45). All the intermediates were characterized by 1H and 13C NMR spectroscopy as well as electron-spray ionization mass spectroscopy thoroughly (ESI-MS, Fig. S1–S21†).
The desired monomer (MTC-OEtKBn, 14) was finally synthesized by a Steglich esterification of MTC-OH (13) and 2-hydroxyethoxy-2-benzyloxypropane (8) using N,N′-diisopropylcarbodiimide (DIC) and 4-dimethylaminopyridine (DMAP). It could be isolated in a good yield (91%) as a colorless crystalline solid. The monomer was successfully characterized by 1H, 13C, and 2D NMR spectroscopy as well as ESI-MS (Fig. S22–S27†). It is worth mentioning that other coupling approaches, e.g., via acid chloride chemistry and active pentafluorophenyl ester chemistry, could not yield the desired monomer (14). We propose that even under basic conditions the liberated acidic intermediates may still catalyze unfavorable ketal hydrolysis.
Since the ring-opening polymerization of cyclic carbonates is highly prone to side reactions, the purity of all the reactants is essential. Thus, the monomer was carefully purified by column chromatography and subsequent recrystallization from diethyl ether. The collected material was finally characterized by crystal structure determination, with the results provided in Fig. S28 and Table S1.† The twisted ring confirmation suggested a preferable reactivity toward ring-opening polymerization (ROP).
Finally, successful ketal hydrolysis in an acidic environment was evidenced by diffusion-ordered 1H NMR spectroscopy (DOSY) and GPC. The addition of trifluoroacetic acid (TFA) to the NMR sample led to the formation of a new polymer species with a smaller diffusion coefficient, whose 1H NMR signals showed the absence of ketal and benzyl alcohol units (red arrows in Fig. 2E and Fig. S36†). Besides, the formation of the small molecule fragments benzyl alcohol and acetone could be observed, underlining the ability of the homopolymer to meet the desired acid responsiveness. The GPC of this side-group-degraded polymer revealed a shift toward smaller molecular weight species compared to the original homopolymer, as expected (Fig. 2C, blue line, Fig. S37†), indicating acid-triggered side chain fragmentation, too.
Fortunately, redissolving the polymer and precipitating it in ice-cold diethyl ether allowed for the removal of the undesired homopolymer (up to 40%) from the targeted block copolymer. The latter could be isolated as a white amorphous powder (18) at yields of at least 60%. A narrow mass distribution (Đ = 1.05) without any shoulder was now obtained by GPC (green line in Fig. 3C and Fig. S39†). MALDI-TOF mass spectrometry provided a narrow signal of significantly higher molecular weights than the mPEG113-OH initiator, too (Fig. 3D.1 and Fig. S40†). The MALDI-TOF mass spectra allowed for assigning each of the single peaks to a specific block copolymer species with distinctive repeating units (Fig. 3D.2). Characterization of the block copolymer by 1H NMR spectroscopy further confirmed the successful synthesis (Fig. S38†). The NMR-derived number of repeating units was 10, which was on the one hand significantly smaller than the expected value of 26.7 (this value was consistent with the observed monomer consumption due to homopolymer formation), but on the other hand, the straightforward purification process and the acceptable yields of pure the block copolymer can facilitate the application of this highly functional and multi-pH-responsive material in further block copolymer self-assembly studies.
The formulated nanoparticles were analyzed by dynamic light scattering (DLS) and a narrow nanoparticle size distribution with a diameter of 35.7 nm and a PDI of 0.17 were found (Fig. S41†). Further analysis of the polymeric micelles by transmission electron microscopy (TEM) confirmed the formulation of spherical nanoparticles at similar small sizes and narrow distributions, too (Fig. 4B and Fig. S42A.1–4†). The diameters of 100 randomly chosen spherical particles were measured by the software ImageJ and an average particle diameter of 29 ± 4 nm was obtained (Fig. S42B and C†). Compared to the DLS results, the particles appeared slightly smaller, which may be due to the fact that the solvent was removed during the TEM sample preparation. Another reason could be a poor contrast with the micelles’ PEG corona. Nonetheless, both methods led to similar sizes and were therefore in good agreement.
To get a deeper understanding of which concentrations these polymeric micelles are stable at, the critical micelle concentration (CMC) was determined. For this purpose, pyrene-loaded polymeric micelles were formulated. Depending on the polarity of the pyrene environment, a change in the I3 fluorescence band can be detected, which can be used to evaluate whether pyrene is encapsulated in the hydrophobic core of a polymeric micelle or if it is exposed to the hydrophilic solvent. Making use of these properties, a dilution series of pyrene-loaded polymeric micelles was investigated, finally giving a CMC of 131 nmol L−1 (Fig. S43†). This low CMC allows for applications even in very diluted conditions, as particularly relevant in in vitro and in vivo studies.
After successful polymeric micelle formulation, the nanoparticles’ degradation upon acidification was investigated. While intact side groups stabilized the polymeric micelles through hydrophobic and π–π-stacking interactions, ketal hydrolysis led to the loss of these adhesive forces accompanied by a transition of the hydrophobic polycarbonate block into a hydrophilic one (Fig. 4C). The loss of amphiphilicity further supported the polymeric micelles’ disassembly due to swelling of the micelles core, pushing the single polymer chains further apart. This degradation process was intensively investigated by DLS. For this purpose, polymeric micelles were incubated in different buffered solutions (PBS or acetate buffer) within a pH range of 3.6–7.4. For that purpose, a concentrated micelle stock solution (β = 10.0 mg mL−1) was first prepared in DI water via the solvent-evaporation method. Then, aliquots of this stock solution were diluted 10:
1 (v/v) into the different buffers. Immediately after mixing, DLS measurements were repeatedly conducted over a period of 56 d. The derived count rates were used as an indication of micellar integrity (Fig. 4D.1). As expected, the polymeric micelles degraded rapidly at low pH values and became gradually more stable under milder conditions. The decline in the normalized derived count rate could be fitted by a one-phase exponential decay function, indicating first-order ketal hydrolysis. From this decay function, the micelles half-life times at different pH values were calculated. The endosomal pH value of 5.5 could be realized by PBS as well as acetate buffer. Both conditions showed the same decay underlining the comparability of the two buffered systems (of note, the PBS sample at pH 5.5 provided some aggregates after 7 d, probably because of salt precipitation formation, and therefore it was not considered anymore after this time point). The micelles’ half-life at this pH value was 2.80 d (PBS) and 2.58 d (acetate buffer), respectively (Table S2†), which were within the desirable range for cargo release.47 For higher pH values, longer delayed degradations were recorded, even at pH 7.0 a decrease in the derived count rate could be observed, however only starting after ∼300 h. This was again still consistent with all the other measurements (of note, at pH 7.4 a minimal increase in count rate was observed that was presumably caused by insufficient sealing of the DLS cuvette leading to water evaporation – this was also the case at pH 6.5; however, a decrease appeared after ∼300 h because of the expected particle degradation). Altogether, the trends for the micelle degradation at pH 3.6–6.0 (as well as the anticipated behaviors at pH 6.5 and 7.0, sketched with dashed lines) are summarized in Fig. 4D.1. In conclusion, under physiologic conditions (pH 7.4) the polymeric micelles remained stable over the whole observation period; while upon gradual acidification, accelerated micelle disassembly could be observed. These findings can be further underlined by the changes in the size distributions comparing at t = 0 d and t = 15 d (Fig. 4D.3). While no change was observed at pH 7.4, a slight shift toward smaller diameters could be seen at pH 6.0. Under endosomal (pH 5.5), lysosomal (pH 4.5), and strongly acidic conditions (pH 3.6), the polymeric micelles were entirely disassembled.
The logarithm to the base 10 (log) of the observed degradation rate constant kobs. was plotted against the pH value (Fig. 4D.2). The observed rate constant kobs. was actually composed of a linear combination of the first-order rate constant for non-catalyzed degradation in water only k0 and the second-order rate constants for its degradation catalyzed by protons kH, as described in eqn (1).48,49
kobs. = k0 + kH[H+] | (1) |
Nile red (2.5 wt%) was added to a concentrated polymer solution (β = 10.0 mg mL−1) in acetone. This solution was mixed with DI water and stored in the fume hood overnight. After complete acetone evaporation, the finished formulation was diluted in different buffers. Alongside physiological (pH 7.4) and endosomal (pH 5.5) model-environments, also a basic (pH 9.5) buffer was chosen. The latter served to investigate the aliphatic carbonate block backbone hydrolysis under base-catalyzed conditions (which was also likely to be enzymatically mediated, as reported for bulk materials at neutral pH52,53). Compared to the acidic degradation, where the ketal side group was hydrolyzed and a hydrophilic block copolymer remained, the base-catalyzed carbonate hydrolysis led to a complete backbone degradation and the formation of mPEG113-OH and small molecule fragments (Fig. 5A). Under physiological conditions (PBS), the micelles could be expected to be stable.
The three samples were simultaneously analyzed by DLS and UV/vis spectroscopy. The DLS-detected derived count rate provided an insight into the polymeric micelles’ integrity, while the UV/vis spectra allowed for quantification of Nile red release from the micellar core. According to the derived count rate, the polymeric micelles remained stable at physiologic pH 7.4 but degraded in acidic (pH 5.5) and basic (pH 9.5) environments (Fig. 5C). This decay was accompanied by a simultaneous decline in the Nile red absorption detected by UV/vis spectroscopy (Fig. 5B.1–4). Interestingly, both observations were very much synchronized and suggested cargo release as an exclusive result of polymeric micelle degradation. Consequently, theses polymeric micelles are considered to enable long-time stable cargo encapsulation as well as targeted release only at anticipated variations of the pH conditions.
Encouraged by these results, in vitro studies were performed to demonstrate the internalization of the polymeric micelles by RAW-Blue™ macrophages. For this purpose, Rhodamine B octadecyl ester-loaded polymeric micelles (β = 1.0 mg mL−1) were formulated under sterile conditions (0.33 wt% dye loading, 13.2% encapsulation efficiency, Fig. S44†). Confocal fluorescence microscopy imaging revealed a distinct polymeric micelle uptake by the macrophages (Fig. 5D), which was in strong contrast to the PBS control group. Furthermore, flow cytometry (FC) experiments showed a pronounced shift of the mean fluorescence intensity, demonstrating effective micelle uptake by all the considered macrophages (Fig. 5E and Fig. S45†).
For this purpose, light-scattering studies can be conducted in human blood plasma and evaluated according to Rausch et al.54 in order to estimate probable interactions between nanoparticles and the blood plasma components. Using the solvent-evaporation method, the TLR7/8 agonist CL075 was again encapsulated in polymeric micelles, aiming for a drug loading of 5.0 wt%. Their nanoparticles sizes were then first determined by conventional DLS (Fig. S46A†) as well as multi-angle DLS measurements (Fig. S46B†), affording similar results. The latter technique was then applied to determine the scattering autocorrelation curves of full human blood plasma alone and incubated with the particles (Fig. S46C†). According to Rausch et al.,54 the autocorrelation curves of nanoparticles in human blood plasma can be evaluated by the linear combination of the determined individual autocorrelation curves of the particles and blood plasma as long as no interaction/aggregation occurs. Fortunately, the CL075-loaded polymeric micelles could nicely be fitted by the two predetermined autocorrelation curves at several scattering angles, which was therefore indicative of there being no protein aggregation (Fig. 6B and Fig. S46C†). Consequently, for controlled systemic applications, the pH-sensitive drug-loaded polymeric micelles could remain stable even under complex biological conditions and, thus, provide ideal features for further applications.
In this respect, in vitro studies were finally performed to evaluate the payload's biological activity. For the Toll-like receptor (TLR) 7/8 agonist CL075, we applied the particles on a RAW-Blue™ macrophage reporter assay. Upon addressing the cells’ TLR and initiating their intracellular NFκB pathway, the cells secrete an alkaline phosphatase into the cell culture medium, whose activity can be determined by the Quanti-Blue™ assay. Afterwards, the cells can further be evaluated by MTT assay for their viability after exposure to micellar particles, too. For this purpose, drug-loaded polymeric micelles were again formulated under sterile conditions, with a targeted CL075 loading of 2.0 wt%. As references, empty micelles and a formulation of free drug were prepared in analogy (note that the drug alone was not soluble in PBS and precipitated out during the formulation process – compare Fig. S47A†). The nanoparticles were characterized by DLS, and the drug loading was determined by UV/vis spectroscopy (1.94 wt% drug loading, 97.2% encapsulation efficiency – Fig. S47†). A dilution series of all the samples was prepared in a CL075 concentration range of 0.02–20 μmol L−1. For a better insight, the assays were also performed with a positive control of CL075 prepared from a 25 mg mL−1 DMSO stock solution diluted into cell culture medium. As shown in Fig. 6C, the micelle-guided CL075 transport stimulated the TLR 7/8 at a similar dose range as the positive CL075 reference. The CL075 formulation in the absence of a carrier on the other hand revealed a decreased immune stimulation by one order of magnitude, while the empty nanocarrier did not activate the NFκB pathway within the investigated concentration window but remained immunosilent (Fig. 6C). These findings underline the necessity of a nanocarrier for the solubilization of hydrophobic drugs and point out the suitability of the elaborated polymeric micelles to serve such a purpose. Thereby, the drugs’ receptor activity was also fully retained.
Besides, the MTT cell viability assay successfully demonstrated the absence of any cellular toxicity for all the investigated formulations in the applied concentration regime. In addition, the cytotoxicity of the side-group-degraded polymers was investigated to exclude any toxicity, even after the delivery process (Fig. S48†). For this purpose, polymeric micelles were formulated in PBS and half of the sample was treated with 10 vol% of 1 M hydrochloric acid (Fig. S48A†). After confirming full degradation by DLS (Fig. S48B†), the former pH value was restored by adding the same amount of 1 M sodium hydroxide. Both samples did not show any cytotoxicity within the tested concentration range (which was equal to the carrier concentration range used in the assays before, Fig. S48C.1†). Furthermore, the cytotoxicity of the degradation products acetone and benzyl alcohol as well as a combination of both was investigated, too. Again, no reduced cell viability could be found (Fig. S48C.2†). Altogether, these findings suggest the pH-responsive polycarbonate block copolymer micelles with hydrophobic benzyl ketal moieties could serve as effective immunodrug nanocarriers with intriguing degradation profiles.
Footnote |
† Electronic supplementary information (ESI) available. CCDC 2371836. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4bm00949e |
This journal is © The Royal Society of Chemistry 2025 |