A versatile synthetic platform for amphiphilic nanogels with tunable hydrophobicity

Alexandra Gruber a, Doğuş Işık a, Bianca Bueno Fontanezi b, Christoph Böttcher c, Monika Schäfer-Korting b and Daniel Klinger *a
aInstitute of Pharmacy (Pharmaceutical Chemistry), Freie Universität Berlin, Königin-Luise Str. 2-4, Berlin D-14195, Germany. E-mail: daniel.klinger@fu-berlin.de
bInstitute of Pharmacy (Pharmacology and Toxicology), Freie Universität Berlin, Königin-Luise Str. 2-4, Berlin D-14195, Germany
cResearch Center of Electron Microscopy and Core Facility, BioSupraMol, Institute of Chemistry and Biochemistry, Freie Universität Berlin, Fabeckstr. 36a, Berlin D-14195, Germany

Received 1st August 2018 , Accepted 15th October 2018

First published on 25th October 2018

We present a new versatile synthetic strategy for amphiphilic nanogels with tunable amphiphilicity. Our approach is based on crosslinked reactive precursor particles, which serve as a platform for secondary functionalization with hydrophilic and hydrophobic moieties: the functionalization of reactive poly(pentafluorophenyl methacrylate) (PPFPMA) nanogels with different amines allows precise adjustment of the internal network composition. Since nanogels with varying amphiphilicity originate from the same uniform precursor particles, they exhibit similar particle sizes, size distributions and homogeneous morphologies. This comparability between nanogels allows accurate investigation of their structure–property relationships. For example, by tuning the amphiphilic network composition, the loading and release profiles for Nile red as a hydrophobic model compound can be controlled through adjusting the hydrophobic interactions with the network. Consequently, our strategy represents a powerful platform for the generation of a new type of amphiphilic nanocarrier that combines the exceptional biocompatibility of hydrophilic nanogels with the transport of hydrophobic cargoes.


Polymeric nanocarriers have evolved as promising materials to address common challenges of hydrophobic drugs, i.e. poor solubility, high aggregation tendency or unfavored partitioning.1–3 Within such drug delivery vehicles, nanogels as crosslinked polymer nanoparticles exhibit a number of advantages due to the unique combination of their colloidal size and internal network structure.4–8 In addition to tunable mechanical properties and drug release profiles, their high flexibility and high water content ensure biocompatibility and colloidal stability, thus making them favorable candidates for drug,9–11 protein,12,13 and DNA14,15 delivery.

However, conventional nanogels exhibit one major drawback: due to their overall hydrophilic network, their potential to efficiently encapsulate hydrophobic drug molecules is very low. Thus, the therapeutic potential of conventional nanogels is mainly restricted to the delivery of hydrophilic drugs.16,17 Consequently, expanding the benefits of nanogels to the versatile delivery of hydrophobic cargoes would represent a promising new approach to increase the therapeutic potential of this class of nanocarriers.

Our strategy to realize this concept is based on introducing partial hydrophobicity into the colloidal polymeric networks. For this, nanogels are designed to contain copolymer networks with hydrophilic and hydrophobic side groups. The resulting amphiphilic nanogels combine multiple benefits from two different carrier types in a single colloidal system: first, the hydrophilic polymer side groups generate a water-swollen hydrogel matrix, which ensures good colloidal stability and biocompatibility. Second, the internal hydrophobic moieties enable the efficient loading of water-insoluble cargoes. This occurs through hydrophobic interactions between the payload and network. Tuning of these interactions can be achieved by varying the type and content of the hydrophobic polymer side groups and enables accurate control of the loading and release profiles. Thus, the ability to precisely tailor the nanogel hydrophobicity, i.e. the hydrophilic/hydrophobic balance, will open up new therapeutic options for various administration pathways including intravenous or topological routes.

While a promising concept, the defined preparation of such tunable amphiphilic nanogels is still in its infancy. It is highly challenging to homogeneously integrate two different network functionalities with opposing solubilities into a single colloidal system. As a result, conventional emulsion-based approaches often result in core/shell structures or complex morphologies.18

To ensure the formation of amphiphilic particles with a homogeneous distribution of hydrophilic and hydrophobic groups in the network, post-modification or multi-step procedures have emerged. One approach is based on the initial synthesis of completely hydrophobic nanoparticles and subsequent hydrolysis of protected hydrophilic comonomer units to generate the amphiphilic networks.19–21 While this strategy is scalable, it is limited in its design flexibility: only a limited number of hydrophilic moieties is accessible through emulsion-based copolymerization. Another strategy focuses on the self-assembly and subsequent crosslinking of amphiphilic (co-)polymer building blocks.22–24 Principally, this strategy allows the utilization of a variety of different polymer backbone chemistries but the concept is limited in its design flexibility since only specific polymer compositions and architectures show the desired assembly characteristics.

In addition, both approaches can show large variations in their colloidal features. Every change in the network composition requires the synthesis of a new particle batch. As a result, nanogels with different network hydrophobicities can also vary in their morphology, particle size, size distribution, crosslinking density, etc. These natural batch-to-batch variations hinder the accurate determination of the structure–property relationships. Since it is difficult to separate the influence of varying network hydrophobicities from the influence of different colloidal features, the ability to tailor nanogel amphiphilicity to a specific pharmaceutical application is limited. Thus, to ensure comparability between nanogels, a new synthetic strategy is required that allows varying both factors independently from each other.

To address this synthetic challenge, we were drawn to particle functionalization methods.25–27 These strategies are based on the preparation of a master batch of crosslinked reactive precursor nanoparticles, which serves as a platform for secondary functionalization with different functional groups. Hereby, decoupling of the interior network functionality from the initial structural parameters of the precursor particles can be achieved.28–33

Based on these considerations, we developed a new precursor particle strategy for the preparation of amphiphilic nanogels with tunable internal network hydrophobicity while keeping as many colloidal features constant as possible (Scheme 1). Although truly similar colloidal features would also include similar surface and mechanical properties, this synthetic platform will serve as the first step towards overall comparable systems. Key to the versatility of our synthetic approach is the utilization of fast and efficient coupling reactions compatible with the functional groups that are introduced in a post-modification reaction.34 For this, we translated the versatility of reactive poly(pentafluorophenyl methacrylate) (PPFPMA)35–42 from linear polymers to crosslinked colloidal networks. The resulting PPFPMA precursor particles can be functionalized with a wide variety of different hydrophilic and hydrophobic amines through the efficient nucleophilic substitution of the reactive pentafluorophenyl esters.41,43,44 By using mixtures of hydrophilic and hydrophobic moieties, the network composition can be tuned in a facile and scalable way, thus giving the opportunity to synthesize a library of amphiphilic nanogels with varying network compositions but comparable colloidal features. As a platform approach, this enables the systematic investigation of the effects of the nanogel amphiphilicity on the loading and release profiles as well as on interactions with biological systems.

image file: c8py01123k-s1.tif
Scheme 1 A versatile synthetic platform for the generation of amphiphilic nanogels with tunable network hydrophobicity: (a) miniemulsion copolymerization of pentafluorophenyl methacrylate (PFPMA) and ethylene glycol dimethacrylate (EGDMA) gives facile and scalable access to reactive precursor particles consisting of reactive copolymers (in the chemical structure, x denotes the repetition units of PFP esters and y the repetition units of the crosslinker). Post-polymerization functionalization of the reactive networks with mixtures of different hydrophilic and hydrophobic amines enables the preparation of a library of amphiphilic nanogels with excellent control over the hydrophilic/hydrophobic balance. In the post-modification reaction, the PFP ester repetition units are substituted by hydrophilic and hydrophobic pendant groups (indicated in the chemical structure as (x–z) for the hydrophilic and z for the hydrophobic groups). (b) Functionalities for the preparation of the amphiphilic library: 2-hydroxypropyl amine (HPA), benzylamine (BENZA), hexylamine (HEXA), linear dodecylamine (DODA), branched dodecylamine (BDODA) and amine-functionalized cholesteryl (CHOLA).

Results and discussion

Preparation of well-defined reactive precursor particles

Reactive precursor particles determine the main colloidal features of the resulting amphiphilic nanogels. Thus, a scalable synthetic strategy is required that enables good control over the particle size and size distribution in this initial step. For this, we exploited the power of miniemulsion polymerization to gain access to well-defined particles in multi-gram scale quantities. In contrast to precipitation- or emulsion-based methods, this strategy has a number of advantages. In particular, the inhibition of net monomer diffusion and the circumvention of nucleation and growth mechanisms allow for greater chemical homogeneity of crosslinkers in the particles.45–47 Following this strategy, a masterbatch of crosslinked precursor particles was prepared by free radical copolymerization of pentafluorophenyl methacrylate (PFPMA) and the crosslinker ethylene glycol dimethacrylate (EGDMA) in a miniemulsion using sodium dodecyl sulfate (SDS) as the surfactant (Scheme 1a). The resulting particle dispersions were purified by repeated centrifugation/redispersion in deionized (DI) water and analyzed by dynamic light scattering (DLS) and transmission electron microscopy (TEM). The results showed well-defined particles with hydrodynamic diameters of 115 nm and narrow size distribution (Fig. 1).
image file: c8py01123k-f1.tif
Fig. 1 Miniemulsion copolymerization gives access to well-defined crosslinked PPFPMA precursor particles as demonstrated by TEM. DLS measurements at an angle 173° show a hydrodynamic diameter of 115 nm and a narrow size distribution for these particles.

Nanogel functionalization via efficient substitution of pentafluorophenyl esters

A library of amphiphilic nanogels with controlled amphiphilicity and comparable colloidal features was prepared. For this, post-polymerization modification of the colloidal PPFPMA36–38 networks was performed with a mixture of two different amines.

To introduce hydrophilicity and ensure an overall water-swollen nanogel network, 2-hydroxypropyl amine (HPA) was used as a hydrophilic moiety. The resulting polymer poly(N-(2-hydroxypropyl)methacrylamide) (PHPMA) is known to be water-soluble, nontoxic and biocompatible, thus having found application in multiple polymer-based therapeutics.48–50 In addition, amphiphilicity was introduced into the network by combining the hydrophilic features of PHPMA with the hydrophobic nature of different aliphatic and aromatic amines, i.e. benzylamine (BENZA), hexylamine (HEXA), linear dodecylamine (DODA), branched dodecylamine (BDODA) and an amine-functionalized cholesteryl group (CHOLA) (see Scheme 1b).

To ensure homogeneous particle functionalization with both hydrophilic and hydrophobic moieties, a good solvent for the polymer and the two different amines is required. This enables efficient diffusion of the reagents into the network and ensures quantitative functionalization of the internal pentafluorophenyl esters. To identify a suitable solvent, swelling studies on the reactive precursor particles were performed by measuring the particle diameter in different solvents via DLS. It was demonstrated that dimethylformamide (DMF) as a good solvent for PPFPMA polymers37 ensures a swollen network of the precursor particles (see ESI Fig. S1) and dissolves a large variety of hydrophilic and hydrophobic amines. Therefore, PPFPMA precursor particles were swollen in DMF and they simultaneously reacted with a mixture of functional amines. The network composition was controlled by adjusting the feed ratio between HPA and the different hydrophobic amines to generate a library of amphiphilic nanogels (Table 1). Purification via dialysis enabled the removal of the generated pentafluorophenol, excess of amines, and remaining surfactant. This procedure led to colloidally stable nanogels in water, which could be freeze-dried for storage. Redispersion by swelling in aqueous media gave suspensions with well-defined solid contents that remain stable for more than two weeks in DI water (selected examples were stable for more than five months, see ESI section 7). To evaluate the colloidal stability of the nanogels under physiological model conditions lyophilized nanogels of series 1 in Table 1 were dispersed in phosphate-buffered saline (PBS) and Dulbecco's modified Eagle's medium. The colloidal stability was demonstrated by DLS measurements over a time period of 14 days (see ESI section 7).

Table 1 Overview of the library of amphiphilic nanogels. Amphiphilicity is controlled by varying the type and amount of incorporated hydrophobic groups R

image file: c8py01123k-u1.tif

  Series and sample name (x–z) feed hydrophilic HPMA [mol%] R type of hydrophobic group z feed hydrophobic group [mol%]
1 PHPMA 100
BENZA-20 80 BENZA 20
HEXA-20 80 HEXA 20
BDODA-20 80 BDODA 20
DODA-20 80 DODA 20
CHOLA-20 80 CHOLA 20
2 PHPMA 100
CHOLA-10 90 CHOLA 10
CHOLA-20 80 CHOLA 20
CHOLA-30 70 CHOLA 30
CHOLA-40 60 CHOLA 40
CHOLA-50 50 CHOLA 50

To demonstrate the successful conversion of the pentafluorophenyl (PFP) esters, freeze-dried nanogel powders were investigated by attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy. It was shown that after the reaction with the amines, the band at 1778 cm−1 corresponding to the PFP ester C[double bond, length as m-dash]O stretching vanishes completely. Simultaneously, a new band corresponding to the amide C[double bond, length as m-dash]O stretching at 1662 cm−1 appears, thus indicating successful network functionalization (Fig. 2, ESI Fig. S2).51,52 In addition, quantitative conversion of the PFP ester groups was demonstrated for all nanogels by the disappearance of the PFP ester peaks in 19F-NMR spectra (see ESI Fig. S3).

image file: c8py01123k-f2.tif
Fig. 2 Successful network functionalization is demonstrated by the disappearance of the reactive PFP ester bands and the simultaneous appearance of the functionalized amide bands in ATR-FTIR spectra. Exemplary spectra show PPFPMA precursor particles, hydrophilic PHPMA and amphiphilic CHOLA-20 nanogels.

Controlling the nanogel amphiphilicity requires the ability to tune the ratio between hydrophilic and hydrophobic moieties in the network. This was synthetically addressed by simply changing the feed ratio of the two different amines in the particle modification reaction. However, it needed to be determined how the feed ratio translated into the actual network composition. To assess this effect, the incorporation of the different amines in the nanogels was determined by nuclear magnetic resonance (NMR) studies on the nanogels in deuterated DMF. For the nanogels reacted with 20 mol% of different hydrophobes and 80 mol% of HPA (series 1 in Table 1), the determined incorporation ratios closely resemble the targeted values, i.e. the feed ratios (see ESI Table S1).

These results suggest the ability to tune the network composition by changing the feed ratio in the reaction mixture. However, for the crosslinked nanogels, line broadening in the 1H-NMR spectra limits the determination accuracy of the network composition. Thus, control experiments on linear polymer analogues36 were performed to gain further insight into the efficiency of the functionalization reaction. For this, modification reactions of reactive precursor polymers were carried out with CHOLA as an exemplary hydrophobe. In a systematic study, the feed ratio of CHOLA/HPA was increased from 0 to 50 mol% CHOLA. The incorporated ratio of HPA/CHOLA moieties was determined via integration of respective peaks after deconvolution in the 1H-NMR spectra (Fig. 3a). It was demonstrated that changing the feed ratio of both amines (HPA and CHOLA) allows tuning the network composition: the experimental incorporation values for CHOLA show a linear relationship with the initially added feed (Fig. 3b). Thus, these investigations support the previous results from NMR studies on the crosslinked nanogels. For additional discussion see ESI section 3.

image file: c8py01123k-f3.tif
Fig. 3 The network composition can be controlled by the feed ratio of the hydrophilic/hydrophobic amine in the post-polymerization modification reaction. (a) 1H NMR comparison of a CHOLA-20 copolymer and the functional amine reagents CHOLA and HPA reveals the presence of both functionalities in the copolymer, thus enabling quantitative analysis via integration after peak deconvolution, and (b) incorporation efficiencies of CHOLA by modification of linear PPFPMA polymers show good correlation between the feed ratio and incorporation.

Having demonstrated the ability to precisely control the nanogel amphiphilicity, we investigated the influence of the network composition on the colloidal features. As our strategy is designed to enable the preparation of various amphiphilic nanogels from the same reactive precursor particles, the resulting different nanogels should exhibit a majority of similar colloidal characteristics. To demonstrate this central feature of our synthetic strategy, we investigated the sizes of a series of amphiphilic nanogels each containing 20 mol% of a different hydrophobic group and 80 mol% HPA (series 1, Table 1) (for estimated log[thin space (1/6-em)]P values of the different hydrophobic repeat units see ESI Table S2).

Evaluation of the TEM images from purified, lyophilized and redispersed nanogels with different network amphiphilicity showed no deviations of particle sizes in their dry state (see ESI Fig. S9). For amphiphilic nanogels in aqueous medium, angle-dependent DLS measurements were conducted. These measurements showed that the sizes of the functionalized amphiphilic nanogels do not significantly deviate from the sizes of their respective precursor particles, which were measured directly after polymerization (Fig. 4a). This similarity between swollen nanogels and precursor particles is assumed to result from the fact that the precursor particles are also swollen by the unreacted monomer after polymerization.

image file: c8py01123k-f4.tif
Fig. 4 (a) Angle-dependent DLS measurements of purified, lyophilized and redispersed nanogels in water with different network amphiphilicity show similar colloidal features, i.e. particle size and size distribution. Error bars represent the width of the size distribution as the full width at half maximum (FWHM) at an angle of 90°. The dotted line represents the diameter of the reactive precursor particles and the colored box the respective width of the size distribution as the FWHM at an angle of 90°. Exemplary TEM (b) and cryo-TEM (c) images of CHOLA-20 nanogels show homogeneous particle morphologies.

By comparing the sizes of the different nanogels it can be seen that the purely hydrophilic PHPMA nanogels are slightly larger than the hydrophobically modified nanogels. It is suggested that the smaller size of the amphiphilic nanogels stems from lower network hydration and additional physical crosslinks formed by the interaction of hydrophobic groups in the network. As a result, the swelling is reduced (see ESI section 6 for further information on the swelling degrees). However, for hydrophobic side groups with log[thin space (1/6-em)]P values larger than that of BENZA, almost no deviation of size can be observed among the amphiphilic nanogels (see HEXA-20, BDODA-20, DODA-20, and CHOLA-20 in Fig. 4a). It is suggested that in these cases the difference in hydrophobicities is not sufficient to cause any further significant deviations in swelling.

Thus, the observed negligible deviations demonstrate that our approach allows tuning the network composition without dramatically changing the particle size and particle size distribution. These colloidal features are defined during the preparation of the precursor particles and translate into the different amphiphilic nanogels equally. Another important advantage of our post-polymerization modification approach is the homogeneous distribution of the different hydrophilic and hydrophobic moieties throughout the network. It is assumed that the functionalization of the swollen precursor networks ensures isotropic distribution of the different functionalities in the nanogels, thus circumventing the typical disadvantages of common emulsion polymerization approaches. For example, copolymerization of different monomers in dispersed droplets can often lead to phase separation into core/shell structures, Janus particles or even more complex morphologies. In direct contrast, our approach gave amphiphilic nanogels with homogeneous morphologies as demonstrated by TEM and cryo-TEM of exemplary CHOLA-20 nanogels (Fig. 4b and c) (for TEM of other nanogels see ESI Fig. S7).

To investigate the influence of the hydrophobic groups on the network morphology in more detail, cryo-TEM and electron cryo-tomography (cryo-ET) were performed to compare the morphologies of amphiphilic CHOLA-20 nanogels and overall hydrophilic PHPMA nanogels. The hydrophilic PHPMA nanogel show an evenly loose fibrous network due to the overall hydration of the polymer strands as depicted in the cryo-TEM images in Fig. 5a. This finding is supported by the cryo-ET data. Ten central and summarized slices out of a tomography volume reconstruction presented in Fig. 5b provide additional interior structural information of the nanogels.

image file: c8py01123k-f5.tif
Fig. 5 Cryo-TEM comparisons of amphiphilic and hydrophilic nanogels show homogeneous particle morphologies for both structures with a slightly more compact architecture of the amphiphilic networks. Images show an overall hydrophilic PHPMA nanogel in comparison with a hydrophobic CHOLA-20 nanogel containing hydrophobic moieties: (a) cryo-TEM projections and (b) central stacks of a cryo-ET volume reconstruction.

For CHOLA-20 nanogels, these images also show a rather homogeneous internal structural organization, which points to an equal distribution of the hydrophilic and hydrophobic moieties. However, the higher contrast indicates a slightly more compact structure when compared to PHPMA. This effect is assumed to result from additional internal hydrophobic interactions between the cholesteryl groups, where the dense packing of the network and the formation of defined outer boundaries might prevent the contact of the hydrophobic moieties with the outer polar environment.

Hydrophobic interactions: tuning loading and release profiles

The homogeneous incorporation of hydrophobic moieties into the nanogels enables the encapsulation of poorly water-soluble cargoes through hydrophobic cargo–network interactions. As depicted schematically in Fig. 6, the molecular structures of the hydrophobic network functionalities are suggested to control these interactions for a given payload: stronger hydrophobic interactions favor the encapsulation of hydrophobic cargoes and result in a slower release.
image file: c8py01123k-f6.tif
Fig. 6 Simplified schematic representation of hydrophobic interactions between the hydrophobic cargo and different hydrophobic groups in the amphiphilic networks. Changing the network composition allows tuning of the loading capacity and the release rate.

To demonstrate this effect, we first focused on the dependence of the loading capacity on the type of hydrophobic group in the network. For this, nanogels with different network compositions (series 1, Table 1) were loaded with Nile red as a hydrophobic model compound53via the co-solvent method:54 aqueous dispersions of the respective nanogels were mixed with a solution of Nile red in acetone. The organic solvent was then evaporated, causing the incorporation of the hydrophobic dye into the nanogels. After the loading process, the particle dispersions were filtered to remove any precipitated Nile red. TEM images of the loaded nanogels did not show any non-encapsulated Nile red aggregates (see ESI Fig. S33). Encapsulated Nile red was determined by UV/Vis measurements on freeze-dried nanogels redispersed in dimethylsulfoxide (DMSO). Quantification was achieved relative to a Nile red calibration curve (see ESI Fig. S34). The loading capacity (LC) was calculated as:

image file: c8py01123k-t1.tif(1)

To demonstrate the influence of the hydrophobic network moieties on the loading profile, the LCs of the nanogels of series 1 (Table 1) consisting of 80 mol% hydrophilic PHPMA and 20 mol% of different hydrophobic groups were compared. It is assumed that the cargo–network interactions, and thus the LC, are strongly dependent on the hydrophobicity of the respective network functionality, i.e. BENZA, HEXA, DODA, BDODA, or CHOLA. To demonstrate this correlation, the loading capacities for the different nanogels were plotted against the calculated logarithmic partition coefficients (log[thin space (1/6-em)]P) of the incorporated hydrophobic structural units (calculated from http://www.molinspiration.com; see the ESI for structures and calculated log[thin space (1/6-em)]P values).55Fig. 7a shows that, in comparison with the overall hydrophilic PHPMA nanogels, the loading capacity is enhanced in the amphiphilic nanogels. Furthermore, it can be seen that the LC increases with increasing hydrophobicity of the network functionality due to the stronger hydrophobic interactions. The incorporation of 20 mol% of hydrophobic groups enabled an increase in loading capacities in the order: BENZA < HEXA < DODA ≈ BDODA < CHOLA. It becomes obvious that the introduction of only 20 mol% of cholesteryl groups into PHPMA nanogels (CHOLA-20 nanogels) can increase the loading capacity of highly water-insoluble Nile red by a factor of 3 to 1.4 wt%. Thus, CHOLA-based nanogels show high potential as new amphiphilic nanocarriers for highly hydrophobic drugs.

image file: c8py01123k-f7.tif
Fig. 7 Network composition of the amphiphilic nanogels determines the loading capacity of Nile red via the types and amounts of hydrophobic network functionalities. (a) Loading capacity increases with the hydrophobicity (log[thin space (1/6-em)]P) of the hydrophobic groups in the network. (b) For CHOLA-containing nanogels, the loading capacity increases linearly with the CHOLA content.

Having demonstrated the influence of the molecular structure of the network hydrophobes on the loading capacities, we examined the influence of the amount of hydrophobic groups. As CHOLA-20 nanogels showed a LC of 1.4 wt%, it was assumed that an increasing CHOLA content can further increase the loading capacity of the amphiphilic nanogels. To examine this effect, a series of nanogels with varying CHOLA contents (10–50 mol%, series 2, Table 1) were prepared and loaded with Nile red. Plotting the loading capacities against the CHOLA contents shows a linear increase of the LC with increasing network hydrophobicity (Fig. 7b). For CHOLA-50 nanogels, a very high LC of almost 3 wt% was determined. Comparing this loading capacity with the literature values for Nile red loaded in different carriers shows very good encapsulation properties of our new carriers. For example, micelles based on amphiphilic block copolymers show a LC of 0.45 wt%,56 amphiphilic dendritic core–multishell (CMS) carriers show a LC of 0.3 wt%57 and completely hydrophobic polystyrene nanoparticles (100 nm) show a LC of 1.4 wt%.58

Since the loading experiments revealed a clear influence of the network hydrophobicity on the cargo–network interactions, it is assumed that this also translates into different release profiles with stronger interactions showing a slower release. Thus, in the next step, the time-dependent cumulative release was examined for nanogels with different network hydrophobicities. In these experiments, hydrophilic PHPMA nanogels were compared to two types of amphiphilic nanogels containing similar amounts of different hydrophobic groups, i.e. HEXA-20 and CHOLA-20 nanogels. All networks were loaded with approximately the same relative amount of Nile red (ca. 0.5 wt% – representing the maximum loading capacity of overall hydrophilic functionalized PHPMA nanogels) to ensure comparability and avoid cross influences of different dye concentrations in each nanogel. After the loading process, the particle dispersions were filtered to remove any precipitated Nile red. The release kinetics were determined using a dialysis set-up in DI water under sink conditions: by offering the system excess amounts of water in the surrounding phase, it was ensured that all incorporated Nile red was soluble in the water phase, thus avoiding limited dye solubility as a restricting factor. As can be seen from the results in Fig. 8a, the hydrophobicity of the nanogels clearly influences the kinetic release profile of the nanogels. As suggested, the hydrophilic PHPMA nanogel showed the fastest release, while the amphiphilic HEXA-20 and CHOLA-20 nanogels showed a much slower release profile. The release of Nile red from CHOLA-20 nanogels was found to be the slowest. In line with the results from the loading experiments, this supports our assumption that CHOLA shows the strongest interaction with Nile red. Therefore, an extended release study was carried out on CHOLA-20 nanogels over a period of 14 days. Fig. 8b shows the respective release profile of Nile red which reaches 80% after approximately ten days. While this strong interaction can be used to realize a sustained release of Nile red as a model compound, the implications of these results are much wider. In future investigations, the ability to accurately tune the network hydrophobicity by changing the molecular structure and amount of incorporated hydrophobic moieties will allow the adjustment of the nanogel hydrophobicity to a specific payload, thus allowing precise tuning of the loading and release profiles. This ability offers great potential to our approach for the utilization of these nanocarriers in delivery applications.

image file: c8py01123k-f8.tif
Fig. 8 Cumulative release kinetics of Nile red (NR) can be tuned by controlling the cargo–network interactions through the nanogel composition. (a) Time-dependent release profiles of PHPMA, HEXA-20 and CHOLA-20 nanogels show a decrease of the release rate with increasing network hydrophobicity (in DI water under sink conditions). (b) Time-dependent release profile over 14 days of CHOLA-20 nanogels shows an extended release of Nile red which reaches 80% after approximately ten days.


The amphiphilic nanogel networks are based on PHPMA as a highly hydrophilic polymer which is known to be non-toxic and biocompatible,50 even if synthesized via the post-polymerization modification of PPFPMA.37 Thus, the process of introducing the HPA groups by the functionalization of the reactive precursor particles is assumed to result in networks that are characterized by the good biocompatibility of PHPMA as the main network component.

In addition, the nanogel networks also contain hydrophobic moieties, which influence the surface hydrophobicity of the amphiphilic nanogels. Since such amphiphilic surface properties can influence the interaction of nanoparticles with lipid membranes in biological systems, they can represent a risk factor for nanoparticle cytotoxicity.59,60 To ensure that the amphiphilicity of the amphiphilic nanogels does not result in cytotoxicity, a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) reduction assay was conducted. Primary normal human keratinocytes (NHKs) were employed since they are much more sensitive than conventionally used cell lines such as human umbilical vein endothelial cells (HUVEC).61 In the assay, the overall hydrophilic nanogel PHPMA was compared to amphiphilic nanogels containing 20 mol% of different hydrophobic groups. These samples correspond to series 1 (Table 1) and contain 80 mol% of PHPMA and 20 mol% of BENZA, HEXA, DODA and CHOLA. In addition, CHOLA-50 nanogels were tested to demonstrate the non-toxic effects of high incorporation amounts of the most hydrophobic CHOLA moiety. The results of the MTT assay indicate that all nanogels are well tolerated by the NHKs, confirming good biocompatibility after 24 hours (Fig. 9a). Also, after an extended incubation period of 48 hours nanogels showed acceptable biocompatibility (Fig. 9b).

image file: c8py01123k-f9.tif
Fig. 9 Viability of normal human keratinocytes shows good biocompatibility of amphiphilic nanogels with various network hydrophobicities. Viability (mean ± SD; n = 3–6) was assessed via the MTT assay after exposure to different nanogels for (a) 24 h and (b) 48 h. SDS served as the positive control.


We have developed a new synthetic platform for the preparation of biocompatible, amphiphilic PHPMA-based nanogels with precisely tunable hydrophobicity. Key to the versatility of this approach is the utilization of well-defined reactive precursor particles that serve as the master batch for the subsequent introduction of network amphiphilicity in a post-polymerization modification reaction. This particle modification approach allows overcoming conventional synthetic challenges that are associated with the incorporation of two groups with distinctly different solubilities into one colloidal system. The facile reaction of crosslinked poly(pentafluorophenyl methacrylate) (PPFPMA) precursor particles with a mixture of hydrophilic and hydrophobic amines allows tuning the network composition very efficiently. Thus, this scalable method enables the homogeneous integration of a wide variety of different network functionalities while retaining the main colloidal features from the precursor particles.

Of special importance is the ability to adjust the network amphiphilicity to specific hydrophobic cargoes. By using hydrophobic groups of varying hydrophobicities or by increasing the hydrophobic content during the network functionalization the cargo–network interactions can be controlled. As a result, the loading capacity and the release profile can be programmed by the selection of a specific network composition. Following this approach, we have demonstrated the realization of high loading contents and slow release rates for Nile red as the model compound.

The high level of control over the network functionality and interaction with hydrophobic cargoes demonstrates the great potential of this approach for the generation of new delivery vehicles with good biocompatibility. These findings represent the foundation to further refine the nanogels for specific pharmaceutical applications. In this context, it is of high interest to combine slow release rates with stimuli-responsive cleavable crosslinkers to ensure low drug leakage until triggered release. Additionally, we assume that the network amphiphilicity also has an influence on the surface properties of the nanogels. It is assumed that different surface hydrophobicities will result in different interactions with proteins and biological systems. Investigations on these aspects are currently ongoing and will be published separately.

Materials and methods


All starting materials and reagents were purchased from commercial sources and used without further purification, unless otherwise stated. Anhydrous solvents were obtained from an MBraun MB SPS-800 solvent purification system and ultrapure water from a LaboStar UV 2 water system. The synthesis of pentfluorophenyl methacyrylate (PFPMA),36,62 poly(pentafluorophenyl methacrylate) (PPFPMA),36 and amine-functionalized cholesterol (CHOLA)63,64 was performed according to literature procedures. Modification of PPFPMA was performed following the procedure reported by Gibson et al.37 For detailed information on synthetic procedures see ESI section 9. Moisture or air sensitive reactions were carried out in dry glassware under a nitrogen atmosphere. Dialysis was performed in benzoylated cellulose dialysis tubes from Sigma-Aldrich (width: 32 mm, MWCO: 2000 g mol−1).
Biological materials. The use of human cells was approved by the ethics committee in charge (application number: EA4/091/10). Normal human keratinocytes (NHKs) were isolated from juvenile foreskin from therapeutically indicated circumcision; parents had signed written informed consent for scientific use. NHKs were expanded in low calcium keratinocyte growth medium (KGM BulletKit, Lonza, Cologne, Germany).


Nuclear magnetic resonance. NMR spectra (1H NMR, 13C NMR, and 19F NMR) were recorded at 300 K on a Jeol ECX400 spectrometer operating at 400 MHz or a Jeol ECP 500 spectrometer operating at 500 MHz. Spectra were referenced to residual solvent peaks.65 The chemical shifts are expressed in parts per million (δ). Coupling constants (J) are expressed in Hz. Coupling patterns are designated as s (singlet), d (doublet), t (triplet), q (quartet) and their combinations and as m (multiplet).
ATR-FTIR. IR spectra by attenuated total reflection (ATR) were recorded on a PerkinElmer Spectrum Two spectrometer measuring between 4000 and 650 cm−1 (PerkinElmer, Waltham, United States of America). All data were processed using Spectrum (10.4.4) from PerkinElmer.

UV/Vis was performed on a Tecan Infinite M1000 PRO plate reader (Tecan Trading AG, Männedorf, Switzerland). All data were processed using the software i-control (1.10.).

Dynamic light scattering. Particle size distributions were determined by dynamic light scattering, performed on a Nicomp Nano Z3000 (Particle Sizing Systems, Port Richey, United States of America). The measurements were carried out at room temperature on diluted dispersions in the respective solvents.

Angle dependent measurements were carried out at scattering angles of 70°, 80°, 90°, 100° and 110°. The apparent diffusion coefficient Dapp was provided by a Nicomp Nano Z3000 by cumulant analysis of the autocorrelation function and plotted against the quadratic scattering vector q2. By extrapolation of the plotted data to the y-intercept, the z-average diffusion coefficient Ds was obtained. Ds was used to calculate the hydrodynamic diameter of the nanogels (see ESI Fig. S8).

For the evaluation of the nanogel stability and swelling behavior in DMF measurements were carried out at a fixed scattering angle of 173°. The measurements were carried out at room temperature on diluted dispersions in the respective solvents.

Cryo-transmission electron microscopy

Sample preparation. Samples for cryogenic transmission electron microscopy (cryo-TEM) were prepared in a Vitrobot preparation chamber (Model Mark IV, Thermo Fisher Scientific Inc., Waltham (MA), USA) at ambient temperature and 100% humidity. A volume of 4 μL of sample solution was deposited on a hydrophilized holey carbon filmed grid (R1/4 batch of Quantifoil, MicroTools GmbH, Jena, Germany) after 60 s plasma treatment at 8 W using a BALTEC MED 020 device (Leica Microsystems, Wetzlar, Germany). The excess fluid was blotted off (blotting time: 4.5 s, blotting force: −20, drain time: 1 s) to create an ultrathin layer (typical thickness of 200–300 nm) of the solution spanning the holes of the support film. The grids were vitrified in liquid ethane at its freezing point (89 K). Ultra-fast cooling is necessary for an artifact-free thermal fixation (vitrification) of the aqueous solution avoiding crystallization of the solvent or rearrangement of the assemblies.
Conventional cryo-electron microscopy (cryo-TEM). The frozen grids were mechanically fixed with auto grids and transferred into the autoloader of a Talos Arctica transmission electron cryo-microscope (Thermo Fisher Scientific Inc., Waltham (MA), USA) operated at a 200 kV accelerating voltage (X-FEG illumination). Images were recorded with a 4k Falcon III direct electron detector at full resolution at a primary magnification of 28k (pixel size: 0.3733 nm). The exposure time was chosen to be 1.23 s using the direct alignment option of 48 image frames accumulating a total dose of ∼40 e per Å2 per image. The sum image of the aligned frames was used for further data evaluation. The defocus was chosen to be 4.9 μ to obtain sufficient phase contrast in the final image.
Cryo-electron tomography (cryo-ET). Tomography tilt series were recorded using the FEI Tomography software (Version 4.5.0, ThermoFisher Scientific Inc., Waltham (MA), USA) in the tilt range of −65°/65° at 2° increments. A total dose of 180 e Å−2 was accumulated on the specimen at a primary magnification of 28k (pixel size: 3.733 nm). Images were recorded with the Falcon III direct electron detector at full image size (4096 × 4096 pixel) and an exposure time of 0.28 s per image.
Image data processing. The image stack alignment and volume reconstruction were performed in the context of the FEI Inspect 3D software (Version 4.1.2., Thermo Fisher Scientific Inc., Waltham (MA), USA). For the selection of intra-particle stacks, band pass filtering and normalization of image histograms allowing the direct comparison of electron density variations, the IMAGIC V software was employed (Image Science GmbH, Berlin, Germany).
Transmission electron microscopy. Samples were prepared by applying a 10 μL droplet of the nanogel solution (1 mg mL−1 in ultrapure water) on a carbon-coated copper grid (400 meshes, Quantifoil Micro Tools GmbH, Großlöbichau, Germany) for 45 s. The supernatant liquid was then removed by blotting with filter paper. This process was repeated 10 times and the grids were allowed to dry in air overnight. The TEM samples were then incubated in an iodine chamber for 15 minutes and measured afterwards using the TEM mode of a Hitachi Scanning Electron Microscope (SU8030, Hitachi High-Technologies Corporation, Tokyo, Japan) with a working voltage of 30.0 kV at different magnifications.
Gel permeation chromatography. (GPC) measurements were performed on a customized chromatography system (PSS Polymer Standards Service GmbH, Mainz, Germany). A 5 cm precolumn (PSS-SDV in THF, 5 μm particle size) coupled with a 30 cm column (PSS SDV linear M in THF, 5 μm particle size) and a differential refractometer detector was used to separate and analyze the polymer samples. As mobile phase THF was used at a flow rate of 1.0 mL min−1. The columns and the differential refractometer detector were heated to 30 °C. For each measurement, 50 μL of a sample of 1.5 mg mL−1 solution was injected. Data were processed using WinGPC UniChrom from PSS. Molecular weights and molecular-weight distributions were reported relative to polystyrene standards (PSS, Mainz, Germany).


Synthesis of reactive poly(pentafluorophenyl methacrylate) particles. For the aqueous phase sodium dodecyl sulfate was dissolved in DI water (1.25 mg ml−1, in 200 mL DI water, 2.5 wt% w.r.t PFPMA). The dispersed phase consisting of PFPMA (10.0 g), EGDMA (0.4 g, 5 mol% w.r.t PFPMA) (filtered over basic aluminum oxide to remove the stabilizer), hexadecane (0.44 g) and 2,2′-azobis(2-methylpropionitrile) (AIBN, 195 mg) was pre-mixed. Both phases were combined and pre-sonicated in a sonication bath for 10 minutes. Full dispersion was achieved by ultrasonication with a Branson Sonifier SFX 550. In order to assure maximum energy input into the monomer emulsion, and thus small droplets of narrow size distribution, the dispersion was divided into two batches of smaller volumes. Each batch was sonicated separately (pulse duration: 15 seconds, pause duration: 15 seconds, and total pulse duration: 4 minutes) before the two batches were recombined and sonicated for another 4 minutes to assure uniform particle size and size distribution. The emulsions were purged with nitrogen for 10 minutes before the reaction was allowed to proceed for 24 hours at 70 °C. The size of the particles (as synthesized) was determined by dynamic light scattering of diluted dispersions of the non-purified particles (due to particle aggregation during the centrifugation and washing procedure).

Afterwards, the dispersion was centrifuged (30 minutes, 10[thin space (1/6-em)]000 rpm), and the supernatant was removed and replaced by DI water. The particles were redispersed by vortex and short sonication in an ultrasonic bath. The centrifugation–washing–redispersion steps were repeated 5 times before the particles were freeze-dried and obtained as a white powder in an average yield of 66%.

The particles were investigated regarding potential hydrolysis via ATR-FTIR spectroscopy after freeze-drying in their solid state.

Post-polymerization modification of reactive poly(pentafluorophenyl methacrylate) particles. Freeze-dried PPFPMA particles (400 mg, 1.59 mmol w.r.t monomer units of PPFPMA particles, 1.0 eq.) were dispersed in 80 mL DMF by short treatment in a sonication bath. The particles were swollen overnight in DMF before different molar ratios of amine functionalized moieties (3.0 eq. w.r.t. monomer units, see ESI Table S3) and TEA (660 μL, 4.77 mmol, 3.0 eq. w.r.t. monomer units) were added and heated to 50 °C for 24 hours. Afterwards the nanogels were purified by extensive dialysis first against DMF and subsequently against DI water and ultrapure water. The nanogels were then freeze-dried and obtained as a white powder.

The nanogels can be stored and redispersed in DI water or PBS by vortex and short sonication treatment.

The success of the post-polymerization modification reaction was monitored by ATR-FTIR spectroscopy on freeze-dried nanogels.

The size of the amphiphilic nanogels after modification was determined by dynamic light scattering after redispersion of the freeze-dried samples in DI water.

Encapsulation of Nile red into nanogels

Encapsulation of Nile red was performed using the co-solvent method in which 3 mL of nanogel dispersion (1 mg mL−1 in DI water) was stirred with 1.5 mL of Nile red solution (0.2 mg mL−1 in acetone) until the acetone was evaporated. After evaporation of the acetone the nanogel dispersion was filtered over cotton to remove any precipitated Nile red. This was followed by a second filtering step over a 0.8 μm cellulose mixed ester syringe filter. TEM images of loaded nanogels did not show any non-encapsulated Nile red aggregates (see ESI Fig. S33). The samples were freeze-dried and redispersed in 2 mL DMSO. The amount of Nile red in the particle dispersion was determined by UV/Vis measurements performed on a Tecan plate reader, relative to a Nile red calibration curve measured at the absorption maximum at 552 nm (see ESI Fig. S35). As Nile red is marginally solubility in water, a blank sample analog to the normal samples was prepared to determine the natural solubility of Nile red in water. The solubility obtained from this blank control was subtracted from the nanocarrier results to obtain the effective loading. The experiments were performed in triplicate and each sample was measured three times.

In vitro release kinetics of Nile red loaded nanogels

A solution of Nile red loaded nanogels (LC: approximately 0.5 wt%) was distributed in several dialysis tubes (1 mg mL−1, 2 mL) and immersed in 5 L DI water at 37 °C, which induces even higher osmotic pressure than buffer solutions. At pre-determined time points, respectively, the dispersion from one dialysis tube was withdrawn and freeze-dried and the obtained powder redispersed in 2 mL DMSO. The absorbance of the samples was measured at 552 nm using a Tecan plate reader. The amount of released Nile red was determined by comparing the encapsulated Nile red of the samples at the different time points with the initial amount encapsulated.

Cytotoxicity study

The cytotoxicity of the amphiphilic, unloaded nanogels was assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) reduction assay by redispersion of the purified and lyophilized nanogels.61 Briefly, 104 NHKs seeded into 96-well plates were incubated with nanogels at concentrations of 0.5 mg mL−1 (0.05%) and 0.05 mg mL−1 0.005% [w/v] in KGM for 24 h and 48 h. Subsequently, 100 μL of MTT solution (0.5 mg ml−1 in KGM) was applied for 4 h. After the removal of the supernatant, 50 μL DMSO was added to dissolve the formazan salt. Optical densities of the respective solutions were measured at 540 nm (FLUOstar Optima, BMG Labtech, Offenburg, Germany). Keratinocytes incubated with 0.05% SDS served as the positive control; the viability of untreated cells serving as reference was set to 100%. A cell viability below 75% predicts cytotoxic effects. Statistical analysis was conducted using GraphPad Prism, version 5.03 (GraphPad Software, San Diego, California). The data are presented as the mean value ± standard deviation (SD) calculated from three to six independent experiments.

Conflicts of interest

There are no conflicts to declare.


This work was supported by the collaborative research center (CRC) 1112 with funding for instrumentation. We acknowledge additional financial support for consumables from Verband der Chemischen Industrie eV (VCI) and Focus Area Nanoscale of the Freie Universität Berlin. We are grateful to Dr Vivian Kral and Nan Zhang for advice on MTT experiments and Anke Schindler for supervision of TEM measurements. A. Gruber especially thanks the CRC 1112 and Focus area Nanoscale for scholarships to finance her PhD.


  1. C. Charrueau and C. Zandanel, in Polymer Nanoparticles for Nanomedicines: A Guide for their Design, Preparation and Development, ed. C. Vauthier and G. Ponchel, Springer International Publishing, Cham, 2016, pp. 439–503,  DOI:10.1007/978-3-319-41421-8_14 .
  2. S. Mura, J. Nicolas and P. Couvreur, Nat. Mater., 2013, 12, 991 CrossRef CAS PubMed .
  3. D. M. Eckmann, R. J. Composto, A. Tsourkas and V. R. Muzykantov, J. Mater. Chem. B, 2014, 2, 8085–8097 RSC .
  4. M. Das, H. Zhang and E. Kumacheva, Annu. Rev. Mater. Res., 2006, 36, 117–142 CrossRef CAS .
  5. D. Klinger and K. Landfester, Polymer, 2012, 53, 5209–5231 CrossRef CAS .
  6. M. Guvendiren, H. D. Lu and J. A. Burdick, Soft Matter, 2012, 8, 260–272 RSC .
  7. L. Zha, B. Banik and F. Alexis, Soft Matter, 2011, 7, 5908–5916 RSC .
  8. J. K. Oh, R. Drumright, D. J. Siegwart and K. Matyjaszewski, Prog. Polym. Sci., 2008, 33, 448–477 CrossRef CAS .
  9. F. Zhan, W. Chen, Z. Wang, W. Lu, R. Cheng, C. Deng, F. Meng, H. Liu and Z. Zhong, Biomacromolecules, 2011, 12, 3612–3620 CrossRef CAS PubMed .
  10. Z.-Y. Qiao, R. Zhang, F.-S. Du, D.-H. Liang and Z.-C. Li, J. Controlled Release, 2011, 152, 57–66 CrossRef CAS PubMed .
  11. R. T. Chacko, J. Ventura, J. Zhuang and S. Thayumanavan, Adv. Drug Delivery Rev., 2012, 64, 836–851 CrossRef CAS PubMed .
  12. N. Murthy, M. Xu, S. Schuck, J. Kunisawa, N. Shastri and J. M. Frechet, Proc. Natl. Acad. Sci. U. S. A., 2003, 100, 4995–5000 CrossRef CAS PubMed .
  13. T. Nochi, Y. Yuki, H. Takahashi, S.-i. Sawada, M. Mejima, T. Kohda, N. Harada, I. G. Kong, A. Sato, N. Kataoka, D. Tokuhara, S. Kurokawa, Y. Takahashi, H. Tsukada, S. Kozaki, K. Akiyoshi and H. Kiyono, Nat. Mater., 2010, 9, 572 CrossRef CAS PubMed .
  14. M. Ahmed and R. Narain, Mol. Pharm., 2012, 9, 3160–3170 CrossRef CAS PubMed .
  15. S. Toita, S. Sawada and K. Akiyoshi, J. Controlled Release, 2011, 155, 54–59 CrossRef CAS PubMed .
  16. S. Nayak and L. A. Lyon, Angew. Chem., Int. Ed., 2005, 44, 7686–7708 CrossRef CAS PubMed .
  17. A. Pich and W. Richtering, Chemical design of responsive microgels, Springer, 2010 Search PubMed .
  18. F. Tiarks, K. Landfester and M. Antonietti, Langmuir, 2001, 17, 908–918 CrossRef CAS .
  19. B. Lu, M. D. Tarn, N. Pamme and T. K. Georgiou, J. Mater. Chem. B, 2016, 4, 3086–3093 RSC .
  20. R. Tiwari, T. Heuser, E. Weyandt, B. Wang and A. Walther, Soft Matter, 2015, 11, 8342–8353 RSC .
  21. B. Lu, M. D. Tarn, N. Pamme and T. K. Georgiou, J. Polym. Sci., Part A: Polym. Chem., 2018, 56, 59–66 CrossRef CAS .
  22. B. Schulte, A. Walther, H. Keul and M. Möller, Macromolecules, 2014, 47, 1633–1645 CrossRef CAS .
  23. M. J. Robb, L. A. Connal, B. F. Lee, N. A. Lynd and C. J. Hawker, Polym. Chem., 2012, 3, 1618–1628 RSC .
  24. S. Takeda, H. Takahashi, S.-i. Sawada, Y. Sasaki and K. Akiyoshi, RSC Adv., 2013, 3, 25716–25718 RSC .
  25. C. Cheng, X. Zhu, A. Pich and M. Moller, Langmuir, 2010, 26, 4709–4716 CrossRef CAS PubMed .
  26. L. Li, C. Cheng, M. P. Schürings, X. Zhu and A. Pich, Polymer, 2012, 53, 3117–3123 CrossRef CAS .
  27. D. Gan and L. A. Lyon, in Smart Colloidal Materials, ed. W. Richtering, Springer Berlin Heidelberg, Berlin, Heidelberg, 2006, pp. 1–8,  DOI:10.1007/3-540-32702-9_1 .
  28. C. Fleischmann, J. Gopez, P. Lundberg, H. Ritter, K. L. Killops, C. J. Hawker and D. Klinger, Polym. Chem., 2015, 6, 2029–2037 RSC .
  29. K. Han, R. Tiwari, T. Heuser and A. Walther, Macromol. Rapid Commun., 2016, 37, 1323–1330 CrossRef CAS PubMed .
  30. Y. Lee, J. Pyun, J. Lim and K. Char, J. Polym. Sci., Part A: Polym. Chem., 2016, 54, 1895–1901 CrossRef CAS .
  31. J. Zhuang, S. Jiwpanich, V. D. Deepak and S. Thayumanavan, ACS Macro Lett., 2012, 1, 175–179 CrossRef CAS .
  32. R. Tiwari, D. Hönders, S. Schipmann, B. Schulte, P. Das, C. W. Pester, U. Klemradt and A. Walther, Macromolecules, 2014, 47, 2257–2267 CrossRef CAS .
  33. W. Graisuwan, H. Zhao, S. Kiatkamjornwong, P. Theato and V. P. Hoven, J. Polym. Sci., Part A: Polym. Chem., 2015, 53, 1103–1113 CrossRef CAS .
  34. A. Das and P. Theato, Chem. Rev., 2016, 116, 1434–1495 CrossRef CAS PubMed .
  35. H. Zhao and P. Theato, Polym. Chem., 2013, 4, 891–894 RSC .
  36. M. Eberhardt, R. Mruk, R. Zentel and P. Théato, Eur. Polym. J., 2005, 41, 1569–1575 CrossRef CAS .
  37. M. I. Gibson, E. Fröhlich and H.-A. Klok, J. Polym. Sci., Part A: Polym. Chem., 2009, 47, 4332–4345 CrossRef CAS .
  38. M. Barz, M. Tarantola, K. Fischer, M. Schmidt, R. Luxenhofer, A. Janshoff, P. Theato and R. Zentel, Biomacromolecules, 2008, 9, 3114–3118 CrossRef CAS PubMed .
  39. N. K. Singha, M. I. Gibson, B. P. Koiry, M. Danial and H.-A. Klok, Biomacromolecules, 2011, 12, 2908–2913 CrossRef CAS PubMed .
  40. K. A. Gunay and H.-A. Klok, Polym. Chem., 2016, 7, 970–978 RSC .
  41. K. A. Günay, N. Schüwer and H.-A. Klok, Polym. Chem., 2012, 3, 2186–2192 RSC .
  42. P. Theato and H.-A. Klok, Functional polymers by post-polymerization modification: concepts, guidelines and applications, John Wiley & Sons, 2013 Search PubMed .
  43. K. A. Günay, P. Theato and H. A. Klok, J. Polym. Sci., Part A: Polym. Chem., 2013, 51, 1–28 CrossRef .
  44. K. A. Günay, P. Theato and H. A. Klok, in Functional Polymers by Post-Polymerization Modification, ed. P. Theato and H. A. Klok, 2013,  DOI:10.1002/9783527655427.ch1 .
  45. M. Antonietti and K. Landfester, Prog. Polym. Sci., 2002, 27, 689–757 CrossRef CAS .
  46. D. Crespy and K. Landfester, Beilstein J. Org. Chem., 2010, 6, 1132–1148 CrossRef CAS PubMed .
  47. K. Landfester, Angew. Chem., Int. Ed., 2009, 48, 4488–4507 CrossRef CAS PubMed .
  48. R. Duncan, Nat. Rev. Drug Discovery, 2003, 2, 347–360 CrossRef CAS PubMed .
  49. J. Kopecek and P. Kopeckova, Adv. Drug Delivery Rev., 2010, 62, 122–149 CrossRef CAS PubMed .
  50. T. Lammers and K. Ulbrich, Adv. Drug Delivery Rev., 2010, 62, 119–121 CrossRef CAS PubMed .
  51. P. A. Woodfield, Y. Zhu, Y. Pei and P. J. Roth, Macromolecules, 2014, 47, 750–762 CrossRef CAS .
  52. G. B. H. Chua, P. J. Roth, H. T. T. Duong, T. P. Davis and A. B. Lowe, Macromolecules, 2012, 45, 1362–1374 CrossRef CAS .
  53. D. Klinger and K. Landfester, Macromol. Rapid Commun., 2011, 32, 1979–1985 CrossRef CAS PubMed .
  54. H. M. Aliabadi, S. Elhasi, A. Mahmud, R. Gulamhusein, P. Mahdipoor and A. Lavasanifar, Int. J. Pharm., 2007, 329, 158–165 CrossRef CAS PubMed .
  55. J. Simon, T. Wolf, K. Klein, K. Landfester, V. Mailänder and F. R. Wurm, Angew. Chem., Int. Ed., 2018, 57, 5548–5553 CrossRef CAS PubMed .
  56. R. Tang, W. Ji and C. Wang, Macromol. Chem. Phys., 2011, 212, 1185–1192 CrossRef CAS .
  57. E. Fleige, K. Achazi, K. Schaletzki, T. Triemer and R. Haag, J. Controlled Release, 2014, 185, 99–108 CrossRef CAS PubMed .
  58. T. Behnke, C. Würth, K. Hoffmann, M. Hübner, U. Panne and U. Resch-Genger, J. Fluoresc., 2011, 21, 937–944 CrossRef CAS PubMed .
  59. M.-C. Jones, S. A. Jones, Y. Riffo-Vasquez, D. Spina, E. Hoffman, A. Morgan, A. Patel, C. Page, B. Forbes and L. A. Dailey, J. Controlled Release, 2014, 183, 94–104 CrossRef CAS PubMed .
  60. Y. Li, X. Chen and N. Gu, J. Phys. Chem. B, 2008, 112, 16647–16653 CrossRef CAS PubMed .
  61. S. Kumar, N. Alnasif, E. Fleige, I. Kurniasih, V. Kral, A. Haase, A. Luch, G. Weindl, R. Haag, M. Schäfer-Korting and S. Hedtrich, Eur. J. Pharm. Biopharm., 2014, 88, 625–634 CrossRef CAS PubMed .
  62. J.-C. Blazejewski, J. W. Hofstraat, C. Lequesne, C. Wakselman and U. E. Wiersum, J. Fluorine Chem., 1998, 91, 175–177 CrossRef CAS .
  63. Y. Nie, M. Günther, Z. Gu and E. Wagner, Biomaterials, 2011, 32, 858–869 CrossRef CAS PubMed .
  64. M. D. Kearns, A.-M. Donkor and M. Savva, Mol. Pharm., 2008, 5, 128–139 CrossRef CAS PubMed .
  65. G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb, A. Nudelman, B. M. Stoltz, J. E. Bercaw and K. I. Goldberg, Organometallics, 2010, 29, 2176–2179 CrossRef CAS .


Electronic supplementary information (ESI) available: IR spectra, 19F-NMR spectra, TEM images, swelling ratios, stability measurement, Nile red calibration curve, structural units for log[thin space (1/6-em)]P calculation, and detailed synthetic procedures. See DOI: 10.1039/c8py01123k

This journal is © The Royal Society of Chemistry 2018