HPMA-based block copolymers promote differential drug delivery kinetics for hydrophobic and amphiphilic molecules

Accepted Manuscript HPMA-based block copolymers promote differential drug delivery kinetics for hydrophobic and amphiphilic molecules Stephanie Tomcin, Annette Kelsch, Roland H. Staff, Katharina Landfester, Rudolf Zentel, Volker Mailänder PII: DOI: Reference:

S1742-7061(16)30006-X http://dx.doi.org/10.1016/j.actbio.2016.01.006 ACTBIO 4059

To appear in:

Acta Biomaterialia

Received Date: Revised Date: Accepted Date:

24 June 2015 9 November 2015 5 January 2016

Please cite this article as: Tomcin, S., Kelsch, A., Staff, R.H., Landfester, K., Zentel, R., Mailänder, V., HPMAbased block copolymers promote differential drug delivery kinetics for hydrophobic and amphiphilic molecules, Acta Biomaterialia (2015), doi: http://dx.doi.org/10.1016/j.actbio.2016.01.006

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Submitted to Acta biomaterialia

HPMA-based block copolymers promote differential drug delivery kinetics for hydrophobic and amphiphilic molecules Stephanie Tomcin†,‡, Annette Kelsch§, Roland H. Staff†, Katharina Landfester†,*, Rudolf Zentel§,*, Volker Mailänder†,‡,* †

§

Max Planck Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany

Johannes Gutenberg University Mainz, Institute of Organic Chemistry, Duesbergweg 10-14, 55128 Mainz, Germany

‡

University Medicine of the Johannes Gutenberg University, III. Medical Clinic, Langenbeckstr. 1, 55131 Mainz, Germany

* joined senior authorship KEYWORDS: drug release, HPMA, nanoparticles, miniemulsion, PDLLA

1

ABSTRACT

We

describe

a

method

how

polymeric

nanoparticles

stabilized

with

(2-

hydroxypropyl)methacrylamide (HPMA)-based block copolymers are used as drug delivery systems for a fast release of hydrophobic and a controlled release of an amphiphilic molecule. The versatile method of the miniemulsion solvent-evaporation technique was used to prepare polystyrene (PS) as well as poly-D/L-lactide (PDLLA) nanoparticles. Covalently bound or physically adsorbed fluorescent dyes labeled the particles’ core and their block copolymer corona. Confocal laser scanning microscopy (CLSM) in combination with flow cytometry measurements were applied to demonstrate the burst release of a fluorescent hydrophobic drug model without the necessity of nanoparticle uptake. In addition, CLSM studies and quantitative calculations using the image processing program Volocity® show the intracellular detachment of the amphiphilic block copolymer from the particles’ core after uptake. Our findings offer the possibility to combine the advantages of a fast release for hydrophobic and a controlled release for an amphiphilic molecule therefore pointing to the possibility to a ‘multi-step and multi-site’ targeting by one nanocarrier.

INTRODUCTION Nanocarriers typically contain several compounds. In the simplest case this would be a higher molecular weight component like a polymer and a payload which is typically of a lower molecular weight. Furthermore other components like surfactants are needed for preparation of nanoparticles. Release of payload is of utmost importance for a drug or gene delivery system applied in cancer treatment, for instance. The fates of the other components like carrier polymer or surfactant are also important as these may cause side effects or toxicological effects.

2

Alternatively two components could serve as therapeutic payloads with different delivery and release mechanisms.[1] Previously we have shown that a short contact of a nanocarrier to a lipophilic surface like a phospholipid bilayer as a model for a cell membrane as well as a mammalian cell itself can release the cargo. So the cargo and the carrier take different paths.[2] While this is an untargeted effect it could be avoided by using a shielding of the hydrophobic core carrying the hydrophobic cargo.[3] Alternatively this could be exploited for the first contact a nanocarrier encounters. To date the most well-known mechanism for an enhancement of the delivery of nanocarriers to an area in the body is based on the (EPR effect)[4, 5] which gives an imperfect enhancement of accumulation of polymeric drug delivery systems in many tumor vasculatures. One of the ideas to enhance the enhanced permeability and retention effect (EPR) effect is to use nanocarriers that deliver nitroglycerin as a hydrophobic prodrug[6] to the tumor endothelium as endothelial cells are the first cell type to be contacted when a nanocarrier wants to leave the blood stream. Release of nitroglycerin would increase the influx of a second type of nanocarriers with an anti-tumor agent. While the dilatation of vessels and the increase of the leakiness takes place on a macroscopic and supracellular level we also wanted to investigate the cellular and subcellular fate of a hydrophobic drug load and other components like amphiphilic ones as it may be advantageous to deliver certain drugs directly by uptake[1] or indirectly, e.g. as shown for hydrophobic molecules in nanocarriers by cell surface contact.[2] The aim of this work was to investigate the release and distribution of hydrophobic drugs and the other components of the nanocarrier which are mediated by contact to and uptake into cells. Therefore, polystyrene (PS) and poly-D/L-lactide (PDLLA) nanoparticles containing a hydrophobic fluorescent dye were used as models for drug-mediating systems. The particles were stabilized by amphiphilic HPMA-based block copolymers consisting of poly(N-(2hydroxypropyl)methacrylamide) and poly(laurylmethacrylate) (P(HPMA)-b-P(LMA)) with a

3

molecular weight of 10 KDa and 10 mol% of hydrophobic LMA. These block copolymers were synthesized by reversible addition–fragmentation chain transfer (RAFT) polymerization[7] in combination with the reactive ester approach allowing a facile functionalization of the polymers.[8, 9] The advantage of using those block copolymers instead of conventional surfactants like sodium dodecyl sulfate (SDS) lies in their biocompatibility.[10] Thus, tedious and time consuming dialysis steps can be avoided. Furthermore, the synthesized block copolymers could be easily modified by attaching e.g. fluorophores like Texas Red covalently to the hydrophilic block. So, highly functionalized particle systems with specifically labeled and easily traceable components could be prepared and were used for in vitro drug release and localization studies. Also active targeting has been achieved with HPMA based systems.[11] The drug release and its distribution over time were investigated by confocal laser scanning microscopy (CLSM). The distribution of particle components was analyzed in the same manner. Moreover, quantitative analysis of the uptake behavior and the distribution of particle components within CLSM images were realized using Volocity® software.

RESULTS AND DISCUSSION

Drug-mediating

Model System.

Polystyrene (PS)

and

poly-D/L-lactide (PDLLA)

nanoparticles were obtained using the miniemulsion process in combination with the solvent evaporation technique (Figure 1) as already shown by Kelsch et al.[12] An amphiphilic HPMAbased

block

copolymer

consisting

of

poly(N-(2-hydroxypropyl)methacrylamide)

and

poly(laurylmethacrylate) (P(HPMA)-b-P(LMA)) was synthesized by RAFT polymerization and used for particle stabilization. The block copolymer had an over-all number molecular weight Mn

4

of 9.5 kg/mol, resulting from a Mn of 8.5 for the P(HPMA) block and 1.0 for the P(LMA) block with a polydispersity of 1,2. N-(2,6-diisopropylphenyl)perylen-3,4-dicarbonacidimide (PMI, see Table S1 in Supporting Information) was applied as a hydrophobic drug model and incorporated in the polymeric matrix during nanoparticle preparation, as indicated in Figure 1. Figure 2 shows the SEM micrographs of the synthesized poly-D/L-lactide (Figure 2A)) or polystyrene (Figure 2B)) nanoparticles respectively synthesized via miniemulsion-solvent-evaporation technique. The obtained nanoparticles show a narrow size distribution in agreement with ref. [18]. Dynamic light scattering experiments of comparable not-fluorescently labeled colloids show a µ2-value of 0,06 -0,1 [18]. Here in total five fluorescently labelled nanoparticle systems of different composition were prepared. It varied –depending on the system (PS or PDLLA as core material and the dye)- between a diameter of 250 and 300 nm (Table 3). In total five nanoparticle systems of different composition were prepared. The particle core was labeled either covalently with Bodipy (structure see Table S1 in Supporting Information) or with the dissolved hydrophobic dye PMI as mentioned previously. Further, the stabilizing block copolymer was also labeled with fluorescent dyes (either Texas Red or Oregon Green) by covalent linkage (see Table S2). For a better understanding the structure of a representative nanoparticle is depicted in Figure 2C). The important characteristics of the synthesized particle systems are listed in Table 3. For more information concerning labeling of nanoparticles’ core and stabilizing block copolymer see Supporting Information Table S1 and Table S2.

In Vitro Release Studies. Former experiments revealed that PS as well as PDLLA nanoparticles stabilized with HPMA-based copolymers were taken up by HeLa cells and no influence of cell viability was observed up to high concentrations of 1200 µg/mL.[12] Referring to these findings release kinetics of PMI from PDLLA nanoparticles NP4 stabilized by Texas

5

Red-labeled P(HPMA)-b-P(LMA) were performed. Thus, HeLa cells were incubated with NP4 for different incubation times (1/4, 1/2, 1, 2, 4, 8 h). After the respective incubation time CLSM imaging of live cells was conducted subsequently after staining cell membranes with CellMask™ Deep Red (CDR). The merged images for each incubation time shown in Figure 3 demonstrate a very fast release of PMI before uptake of nanoparticles occurred. Even after an incubation time of 15 min fluorescence signals of PMI were already detected indicating a very fast influx of the hydrophobic drug model. Signals of Texas Red which is covalently attached to the HPMA-based block copolymer could be clearly observed after 2 h incubation. The signal intensity as well as the amount of Texas Red spots increased with increasing incubation time. In contrast, the fluorescence signals of PMI were also increasing in their intensity over time but the number of spots did not seem to change. The CLSM imges (Figure 3) indicate that PMI was released from nanoparticles as they came in close proximity and contact to the cells. However, nanoparticles had not necessarily to be taken up to initiate that burst release. Obviously, no intracellular colocalization of PMI and Texas Red signals could be detected although there is colocalization in the surrounding medium (red arrows in Figure 3) indicating that PMI was effectively incorporated into the polymer matrix of the nanoparticles. The results of the conducted experiments resemble a first hint of an independent uptake mechanism of the incorporated drug-model and the nanoparticles themselves which has also been shown for other nanocarrier systems.[2] To fortify the previous statement that there was an independency between uptake of nanoparticle and the burst release of PMI, the cellular uptake was investigated by flow cytometry measurements. For the quantification of nanoparticle and PMI uptake HeLa cells were incubated for 1 and 8 h with NP4. The graph in Figure 4 shows that after 1 h of incubation a vast amount of PMI was detected compared with the weak signal of Texas Red. An influence of differences in

6

fluorescence intensity of both fluorophores could be excluded since normalized values were used for calculation of the diagram. Because of the fact that there were no particles attached to the cell membranes or in close proximity in CLSM images (Figure 3) it could be assumed that every detected signal was a consequence of PMI or nanoparticles which were taken up. The significant difference between the determined values of fluorescence intensity of PMI and Texas Redlabeled nanoparticles after 1 h of incubation clearly demonstrates the burst uptake of PMI without an uptake of nanoparticles. The flow cytometry measurements were congruent with the CLSM investigations and supported the postulate of a burst release of an incorporated drug. After 8 h of incubation using NP4 an increase in fluorescence intensity for both PMI and Texas Red could be detected confirming the optical impression of CLSM images (Figure 3). The uptake of nanoparticles occurred temporally delayed in comparison to the uptake and release of PMI respectively. As mentioned before, signals of PMI and Texas Red were detected at different intracellular regions after incubation with the nanoparticles. Hence, experiments were performed to investigate where PMI accumulates after cell-uptake. Hofmann et al.[2] described a similar observation using hydrophobic molecules which were released into cells without nanoparticle uptake. In [2] we have shown that unilamellar vesicles as a model for a cell membrane bilayer are contacted by nanoparticles and a dye is released during this short contact. It has been shown that those molecules accumulated in so-called lipid droplets which are an intracellular lipid storage center. We have also shown that there is no release of PMI in cell culture medium and even the addition of 1% of DMSO (which is used for some experiments in cell culture to add substances) does not trigger release. Only in a lipophilic surrounding like glyeryl trioleate we find the release of PMI. Controlling this process outside of the cell therefore can be done by hindering the direct contact with the cell surface by creating a shell around the hydrophobic,

7

polymeric core such that there is no PMI in this shell. To verify if PMI in our nanoparticulate system accumulated also in lipid droplets, HeLa cells were incubated with PDLLA (NP1) or PS (NP2) nanoparticles whose only labeling was incorporated PMI. The overlay of transmission and fluorescence images of cells after incubation provided a first impression that PMI was localized in lipid-rich regions, which appear as bright or dark, round-shaped spots in transmission images (Figure 5A), D), red circles) due to differences in refractive index with the surrounding medium. The fluorescence signals of PMI (Figure 5B), E)) and lipid-rich regions colocalize, which is shown in the merged images (Figure 5C), F)). To answer the remaining question, whether the polymeric carrier particles are also located at lipid-rich domains, PMI-free PS nanoparticles stabilized by an Oregon Green-labeled block copolymer (NP3) were applied in cell studies in the same manner as described for NP1 and NP2. Even after 24 h of incubation no fluorescent particle signals could be detected in lipid-rich domains (Figure 6). There was no colocalization of bright or dark spots in the transmission image (Figure 6A)) with the fluorescence signals of Oregon Green (Figure 6B)) which is also proved in the merged image (Figure 6C)). These observations of the conducted experiments using particle systems with an either incorporated dye or labeled surfactant-corona were in agreement with our previous results. The fact that PMI and the nanoparticles themselves remained in different intracellular regions after cellular uptake promote the assumption that a release has to have been occurred. To identify precisely those lipid-rich domains, where PMI accumulates, as lipid droplets, an immunofluorescence staining according to Hofmann et al.[2] was performed. HeLa cells were incubated with NP1 for 24 h. The experiments were investigated by fluorescence and bright field CLSM imaging. A red corona around lipid-rich regions in the overlay of transmission image and fluorescence signal of the secondary antibody labeled with Alexa Fluor® 633 (Figure 7A)) indicates those regions as lipid droplets. Figure 7B)

8

provide evidence that PMI (green color) accumulated in lipid droplets (marked by Alexa Fluor® 633-labeled secondary antibody; red color). Further investigations at low temperature (4 °C) demonstrated that the uptake of PMI is controlled rather by diffusion than endocytosis (data not shown) which also supported the fact that the PMI uptake took place without any particle uptake. If we want to exploit in a more controlled way as a “kiss-and-run mechanism” for drug delivery we want to point out that one needs to add another layer which protects first the release of the non-covalently bound lipophilic cargo. This could be done by grafting an outer shell with a high hydrophilicity as a kind of barrier between the nanoparticle and cell membranes.

Fate of the Polymeric Carrier Nanoparticle. The previously presented analysis was dealing with the investigation of PMI release. Therefore PMI was incorporated physically inside the polymeric matrix and a dye-labeled surfactant was used as a particle marker. The conducted experiments had left an open question whether the detected “particle signals” (see Figure 3) represented the whole particle or whether they were the result of a detached and accumulated dye-labeled block copolymer. In this section we focus on the investigation of the destination of the polymeric carrier matrix. For this purpose nanoparticles whose core and stabilizing block copolymer were labeled with covalently attached fluorophores (NP5 and NP6) were synthesized. The core-building polymer was generated by free radical polymerization of a polymerizable Bodipy-based monomer and polystyrene. For nanoparticle formation the synthesized polymer was applied in the miniemulsion-solvent-evaporation procedure as described before.[12] In this case, a Texas Red-labeled P(HPMA)-b-P(LMA) block copolymer was used as a stabilizing agent (Supporting Information Scheme S1). As mentioned before PS and PDLLA particles showed a similar release and cell-uptake behavior. Therefore, PS nanoparticles were applied in this study instead of PDLLA particles since covalently labeled polystyrene could be produced much easier

9

than covalently labeled PDLLA. For cell studies HeLa cells were incubated with double labeled particles NP5 (see Table 3) for 8 and 24 h. After the incubation time and staining of cell membranes with CDR, CLSM imaging was subsequently performed on live cells. Figure 8 show CLSM images which indicate a colocalization of Bodipy (Figure 8A)) and Texas Red signals (Figure 8B)) in the merged image (Figure 8C)) after 8 h incubation in HeLa cells. Thus, uptake of the whole nanoparticle (core and block copolymer) has to have been occurred. However, focusing on incubated cells after 24 h (Figure 8D)-F)) it could be clearly seen that an intracellular detachment of the block copolymer occurred, indicated by the appearance of much more fluorescent spots of Texas Red without a corresponding signal of Bodipy. To confirm the hypothesis of an intracellular detachment a quantitative analysis of CLSM images was conducted using the program Volocity®. For this purpose CLSM images of 20 single HeLa cells were analyzed after 8 and 24 h of incubation with NP5. Here later time points were chosen as up to 8 h nanoparticle core and blockcopolymer colocalized and earlier time points therefore yield the same picture as Figure 8A and the 8 h time points in Figure 9 and 10. The program Volocity® was used to calculate fluorescence signals of Bodipy and Texas Red in relation to their amount of spots within each cell. The following histograms show the amount of fluorescence spots after 8 h (Figure 9A)) and 24 h (Figure 9B)) incubation for every single cell (20 in total). This quantitative evaluation of CLSM images clearly illustrates that the amount of fluorescent spots of both fluorophores was almost equal. Thus, an intracellular detachment of Texas Red-labeled block copolymers from the particles’ core did not occur up to 8 h of incubation. Additionally, the moderate uptake behavior is reflected in Figure 9A) which is congruent with the observations of CLSM images (Figure 8). After 24 h (Figure 9B)) incubation the amount of Bodipy spots did not show a significant change whereas the amount of Texas Red

10

spots increased drastically. This leads to the assumption that the uptake of particles was already saturated after 8 h of incubation and the block copolymers got detached intracellularly from the polymer nanoparticles and potentially aggregated in micellar structures. The appearance of more, but smaller areas of Texas Red spots after 24 h in comparison to 8 h incubation time confirms the previous hypothesis of the formation of micellar structures after detachment of the block copolymers. Calculations of the area dimensions of Bodipy and Texas Red spots were also performed utilizing the software Volocity®. While Bodipy spots exhibited nearly the same area dimensions after 8 and 24 h of incubation (Figure 10A)), Texas Red spots showed a significant difference in area size comparing incubation times of 8 and 24 h (Figure 10B). After 8 h incubation the amount and also the dimensions of Texas Red areas corresponded to the area number and area size of Bodipy (Figure 10, black bars). This supported the assumption that the block copolymer was still adsorbed at the particle surface after 8 h incubation time. After 24 h the number of Texas Red spots with a small area size increased drastically in comparison to 8 h incubation (Figure 10, red bars). Interestingly, the number and area size of Bodipy spots do not show any significant changes and can be considered as constant. These results contributed evidence to the hypothesis of an intracellular detachment of block copolymers and their subsequent aggregation to micelle-like formations. In summary, it can be stated that an intracellular detachment of the block copolymer P(HPMA)-b-P(LMA) from nanoparticles was observed. This was probably initiated by intracellular processes whose investigation was out of the scope of this work. Thus nanoparticles stabilized by HPMA-based block copolymers are considered as a promising carrier system for hydrophobic drugs. Figure 11 gives a schematic as an overview of possible mechanisms for the release of PMI as a hydrophobic drug model and the intracellular fate of nanoparticles after uptake. All previously

11

discussed results and considerations are summarized in Figure 11. It could be shown that the release of a hydrophobic drug did not necessarily occur intracellularly, but nanocarriers like polymeric nanoparticles can release their cargo due to contact with cell membranes. In addition, nanoparticle uptake occurred conspicuously after the initial burst release of the hydrophobic drug model PMI. The fate of nanoparticles’ core and stabilizing block copolymer was investigated by using NP5. It could be clearly demonstrated that an intracellular detachment of block copolymers took place after 8 h incubation and single copolymer chains presumably formed micelles or at least smaller aggregates.

CONCLUSION The versatile applicability of the miniemulsion process was utilized for the preparation of polymeric particles loaded with hydrophobic cargos in their polymeric matrix due to hydrophobic interactions. PS as well as PDLLA nanoparticles were synthesized using the miniemulsion solvent-evaporation technique. The applied HPMA-based block copolymer served as a polymeric surfactant in the synthesis process and replaced the commonly used surfactant SDS. PMI was loaded in the particles’ core during the particle formation and represents a model system for hydrophobic drugs. The synthesized nanoparticles were used without further purification in cell-uptake studies due to the non-cytotoxic behavior of the surface-active block copolymers, which is advantageous in comparison to SDS-stabilized particles. Applying the polymer nanoparticle to cell experiments, PMI was released very fast into cells and accumulated in so-called lipid droplets, even before the cellular uptake of the nanoparticles. Further studies with double labeled nanoparticles showed an intracellular detachment of the HPMA-based block

12

copolymer starting 8 h after particle uptake. An individual detection of the particle core and the block copolymer was ensured by separately labeling each with a different covalently bound fluorophore. These findings were also supported by image analysis with Volocity®. Our findings offer the possibility to combine the advantages of a fast release for hydrophobic and a controlled release for an amphiphilic molecule therefore pointing to the possibility to a ‘multi-step and multi-site’ targeting by one nanocarrier. In general, it would be possible to incorporate two different drugs in the very same nanoparticle: One in the hydrophobic core, which undergoes a fast release and one covalently bound to the block copolymer which shows a (controlled) release after a longer time period due to enzymatic or hydrolytic degradation. Furthermore it will be of utmost importance to investigate and exploit the intracellular uptake route of the nanoparticle and therefore the fate of the covalently bound molecules.

METHODS Material. Oregon green cadaverin and Texas Red cadaverin were purchased from Invitrogen. N-(2,6-diisopropylphenyl)perylene-3,4-dicorboximide (PMI) was received from BASF and used as received. PDLLA (Mw = 28,000 g/mol) was purchased from Sigma-Aldrich.

Synthesis of 4-cyano-4-((thiobenzoyl)sulfanyl)pentanoic acid (CTP), pentafluorophenyl methacrylate (PFPMA) and macro-CTA 4-cyano-4-((thiobenzoyl)sulfanyl)pentanoic acid was used as chain transfer agent (CTA) and synthesized according to literature in a 3-step reaction.[13] PFPMA was prepared according to the literature.[14]

13

RAFT polymerization of PFPMA using CTP was performed in a Schlenk tube, loaded with PFPMA and 2,2’-azobis(isobutyronitrile) (AIBN), (molar ratio monomer/CTA/AIBN: 50/1/0.1) and finally dissolved in absolute dioxane. After three freeze–vacuum–thaw cycles, the mixture was stirred at 70 °C for 16 h. Afterwards the polymeric solution was precipitated 3 times in hexane, isolated by centrifugation and dried for 12 h at 30 °C under vacuum. A pink powder with a yield of 59% was obtained. H-NMR (CDCl3): δ [ppm] 1.9-2.7 (br), 1.2-1.7 (br). 19F-NMR (CDCl3): δ [ppm] -162.6 (br), -157.3 (br), -150 to -152 (br). Table 1: Characterization of PFPMA Polymer

Mtheo (g/mol)

Mn (g/mol) *

Đa

P(PFPM A)57

14100

14700

1,24

* determined by GPC (THF with LiBr (c = 0.1 mmol/L)

Synthesis of block copolymer Block copolymer was prepared in analogy to literature.[12] The macro-CTA, obtained in the above-mentioned polymerization, LMA and AIBN (molar ratios monomer/macroCTA/AIBN: 4/1/0.17) were dissolved in absolute dioxane. After three freeze–vacuum–thaw cycles the tube was immersed in an oil bath at 70 °C. After polymerization time of 2 d, the solution was precipitated twice in ethanol. After removal of the supernatant the precipitate was dried for 12 h at 30 °C under vacuum, obtaining a slightly pink powder with a yield of 69%. H-NMR (CDCl3): δ [ppm] 1.9–2.6 (br), 1.0–1.7 (br), 0.8–0.9 (br t). 19F-NMR (CDCl3): δ [ppm] -162.2 (br), -157.1 (br), -152 to – 150 (br).

14

Table 2: HPMA-based block Copolymer

P(HPMA)‐bP(LMA)

Monomer ratioa 89:11

Đ

Mn (g/mol)b 9.500

1,21

Removal of dithiobenzoate end group The dithiobenzoate end group was removed according to the procedure reported by Perrier et al.[15] Therefore a 25-fold molar excess of AIBN was added to the polymer dissolved in dioxane. After four hours of heating the solution at 80 °C, the polymer was precipitated in hexane twice and collected by centrifugation. The polymer was dried under vacuum for 18 h; a colorless powder could be obtained. Yield: 91%. The absence of the dithiobenzoate endgroup was confirmed by UV–vis spectroscopy.

Postpolymerization modification of block copolymers Precursor polymer (100 mg) without dithioester endgroup was dissolved in abs. dioxane (4 mL). Texas Red Cadaverin (2.7 mg) dissolved in DMSO (1 mL) as well as triethylamine (0.81 mg) were added and the solution stirred for 8 h at 40 °C. Thereafter, 2-hydroxypropylamine (59.5 mg) and triethylamine (160.4 mg) were added and the solution further stirred for 48 hours. The solution was dissolved in a DMSO/water solution for dialysis. After lyophilization a white powder with a yield of 61% could be obtained. H-NMR (400 MHz, DMSO-d6) δ [ppm]: 0.601.56 (br), 2.62-3.10 (br), 3.19-3.43 (s), 3.41-3.80 (br), 4.50-4.83 (br) and 6.90-7.70 (br).

15

Synthesis of Nanoparticles by Miniemulsion Technique in Combination with Solvent Evaporation. Polystyrene nanoparticles were prepared according to the literature.[16] For the solvent evaporation process 13 mg of polymer (PS or PDLLA) were dissolved in 1.85 g of chloroform. The macroemulsion was prepared by adding the aqueous phase consisting of 10 mg of dissolved block copolymer in 4.5 g water to the organic phase, and subsequent magnetic stirring of the mixture at high speed for 60 min. Afterwards, the macroemulsion was subjected to ultrasonication under ice cooling for 180 s at 70% amplitude in a pulse regime (10 s sonication, 10 s pause) using a Branson sonifier W450 digital, 1/4” tip. The obtained miniemulsion was transferred to a reaction flask with a large size neck and stirred overnight at room temperature for complete evaporation of the organic solvent. In order to purify the nanoparticles they were run over a Sephadex G75 column.

Characterization of Nanoparticles. Scanning electron microscopy (SEM) was used to study the morphology of the polymer particles, dried on a silica wafer. The images were recorded by using a field emission microscope (LEO 1530 Gemini) working at an acceleration voltage of 0.74 V.

Cell Culture for CLSM Studies. Human cervix adenocarcinoma cells (HeLa cells) established from the epitheloid cervix carcinoma from Henrietta Lacks were purchased from German Collection of Microorganisms and Cell Cultures (DSMZ). Cells were kept in Dulbecco´s Modified Eagle Medium (DMEM) without phenol Red supplemented with 10 vol% fetal calf serum (FCS), 100 units/mL penicillin together with 100 µg/mL streptomycin (Invitrogen, Germany), and 1 vol% GlutaMAXTM (all from Invitrogen, Germany). HeLa cells were grown in a humidified incubator at 37 °C and 5% CO2. One day prior to the experiments,

16

adherent cells were detached using 0.5% trypsin (Invitrogen, Germany) and seeded in ibidi µdish35mm,low (IBIDI, Germany) at a density of 8,500 cells per cm2. After re-adhesion overnight cells were washed once with Dulbecco´s phosphate buffered saline (PBS, Invitrogen, Germany) before treatment. Cells were incubated with nanoparticles at concentrations of 150 µg/mL for incubation times in a range of 15 min till 24 h according to experiment. After incubation time cells were washed three times with PBS, covered with 1 mL medium, and stained with 0.2 µm CellMask™ Deep Red (Invitrogen, Germany), right before imaging on a CLSM (confocal laser scanning microscope) was started.

CLSM Imaging. To demonstrate nanoparticle uptake and intracellular release of a hydrophobic drug model confocal laser scanning microscopy (CLSM) was applied directly after staining cells with Cellmask™ Deep Red. Images were taken with Leica LAS AF Software on a Leica TCS SP5 II microscope equipped with five lasers (multiline argon laser with 458, 476, 488, 496, 514 nm, a 561 nm DPSS laser, a HeNe laser with 594 and 633 nm lasers, and a 592 nm CW STED laser) with a HCX PL APO CS 63×/1.4-0.6 oil-immersion objective. Laser lines for excitation and emission bands of detectors (photo multiplier tube, PMT) are summarized in Table S3 (see Supporting Info) for every fluorophore/ dye used in this study. Images were taken with a pinhole size of 1 AE, a line average of 2, a resolution of 1024 × 1024 8-bit-pixels, and by using PMTs for detection. To avoid crosstalk a serial mode was applied for imaging. Image processing was done with Leica LAS AF software and Image J.

Cell Culture for Flow Cytometry. HeLa cells were seeded in a 6 well plate at a density of 200,000 cells per well. After re-adheasion over night, cells were incubated with nanoparticles NP4 for 1 and 8 h at a concentration of 150 µg/mL. HeLa cells were washed once with PBS after

17

incubation time and detached using 400 µL 0.5% trypsin per well. To stop digestion 400 µL medium per well was added and after centrifugation (1500 rpm for 3 min, centrifuge 5810R, Eppendorf, Germany) each cell pellet was suspended in 1 mL PBS, filtered with a bottle top filter (0.2 µm, ZapCap®, Sigma Aldrich, Germany), and analyzed with FloMax 3.0 software on the flow cytometer CyFlow ML (Partec, Germany). Therefore a 488 nm laser was used for excitation of PMI which was detected on FL1 at 527 nm. Texas Red was excited at 560 nm and detection occurred on FL5 at 610 nm. Data analysis was performed using FCSExpress V4 software. Each value was the average of 6 measurements and was normalized to the fluorescence intensity of Texas Red obtained by measurements on a plate reader (Infinite M1000, Tecan, Germany).

Immunofluorescence Staining. After seeding HeLa cells in ibidi µ-dish35mm,low (IBIDI, Germany) at a density of 8,500 cells/cm² and re-adhesion over night cells were incubated 2 h with NP1. Afterwards immunofluorescence staining was performed according to Hofmann et al.[17] In brief, cells were fixed with a mixture of 4% formaldehyde and 0.025% glutaraldehyde (both Sigma-Aldrich, Germany) for 20 min at room temperature, followed by permeabilization with 0.1% saponin (Sigma-Aldrich, Germany) for 10 min. After washing steps cells were incubated with primary antibody TIP47 guinea pig (Progen, Germany) for 30 min at 37 °C. After several washing steps subsequent incubation with the secondary antibody anti-guinea pig Alexa Fluor® 633 (Invitrogen, Germany) followed for 30 min also at 37 °C. CLSM imaging was performed for the detection of lipid droplets containing PMI released from NP1. Therefore a 633 nm HeNe laser line was used for excitation of Alexa Fluor® 633-labeled secondary antibody which was detected at 650-695 nm. PMI was excited with a 488 nm argon laser line and detected at 505-545 nm.

18

Data Quantification with Volocity®. Calculations with the image processing software Volocity® 6.1.2 (PerkinElmer) were performed to support optical impressions of CLSM images that indicate an intracellular detachment of the block copolymer from the nanoparticle NP5. This software can be used for 3D as well as for 2D imaging as we did in our case. At first 20 single cells of both incubations times (8 and 24 h) were separated from CLSM images whereupon only regions of the cytoplasm were cut out, so that nanoparticles which were only attached to the cell membrane, but not internalized had not to be taken into considerations. Afterwards the amount of fluorescence spots from Bodipy and Texas Red as well as the spots’ areas were calculated for each cell. Therefore a threshold of 5-255 was set for the fluorescence channels of the fluorophores Bodipy and Texas Red. The command “find objects” was applied for both fluorophores separately whereby the amount as well as the area of fluorescence spots was obtained. These values were further edited with the software Excel 2007 and OriginPro 8.5.1G.

19

FIGURES

Figure 1. Scheme of particle formation by the miniemulsion process in combination with the solvent evaporation technique to obtain homogeneous PS or PDLLA nanoparticles containing PMI as a hydrophobic drug model.

20

A)

B)

C)

P(HPMA)-b-P(LMA)

dye (e.g. Texas Red)

PMI

Figure 2. SEM images of PDLLA, NP1 (A)) and PS^, NP2 (B)) nanoparticles obtained via the miniemulsion process in combination with the solvent evaporation technique. Scale bar 2 µm. Please see also the Supporting Information for a higher magnification (Figure S2) C) shows schematically the compositions of a nanoparticle. The polymer core is shown in grey, the stabilizing dye-labeled block copolymer P(HPMA)-b-P(LMA) is shown chain of units of LMA (black dots) and HPMA (light blue dots). The covalently coupled dye on the P(HPMA)-bP(LMA) is shown as red star. The physically absorbed PMI as hydrophobic drug model is shown as green dot. Chemical structures are given on the left side of the schematic representation.

21

Figure 3. CLSM images of HeLa cells showing a fast release of the hydrophobic drug model PMI from NP4. The color code is depicted in the legend. Red arrows mark colocalization of PMI and Texas Red outside cells. Scale bar 25 µm.

22

median of fluorescence intensity (a. u.)

1000

1h 8h

800 600 400

40

20

0

PMI

NC (PMI)

TR

NC (TR)

Figure 4. Quantification of nanoparticle uptake in HeLa cells after 1 and 8 h incubation with NP4 determined by flow cytometry. Nanoparticles contained PMI and were stabilized by Texas Red (TR)-labeled block copolymer. NC: negative control (cells without particle treatment measured with the appropriate detector for PMI or Texas Red), n = 6 per group.

23

Figure 5. CLSM images of HeLa cells after incubation with NP1 (PDLLA core) and NP2 (PS core) containing PMI. Bright or dark spots in the transmission (TM) images (A), D)) mark lipidrich regions (examples are marked with a red circle). These regions show colocalization with PMI signals (B), E)) in the merged images (C), F)). Scale bar 25 µm.

24

Figure 6. CLSM images of HeLa cells after 24 h incubation with NP3 stabilized by Oregon Green labeld block copolymer. A) TM image, B) fluorescence signal of Oregon Green labeled NP3, and C) is the merged picture. Enrichment of dye in lipid-rich regions was not observed. Scale bar 10 µm.

Figure 7. CLSM images of immunofluorescence staining of lipid droplets. The color code is depicted in the legend. Lipid droplets featured a red colored corona derived from staining with Alexa Fluor® 633. PMI accumulated in the lipid droplets’ core. Scale bar 10 µm.

25

Figure 8. CLSM images of HeLa cells after 8 (A)-C)) and 24 h (D)-F)) incubation with NP5. A) and D) are images showing fluorescence signals of Bodipy, B) and E) signals of Texas Red. C) and F) are the merged pictures of both fluorescence signals. The color code is depicted in the legend. Scale bar 20 µm.

26

A)

8 h incubation number of fluorescence spots

35

Bodipy Texas Red

30 25 20 15 10 5 0 1

2 3

4

5

6

7 8

9 10 11 12 13 14 15 16 17 18 19 20

cell

B)

24 h incubation number of fluorescence spots

35

Bodipy Texas Red

30 25 20 15 10 5 0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20

cell

Figure 9. Quantitative evaluation of 20 cells with Volocity® demonstrates an intracellular detachment of the block copolymer from the particles’ core starting 8 h after particle uptake. The number of fluorescence spots of Bodipy (core labeling) and Texas Red (block copolymer labeling) is given in A) after 8 h incubation and in B) after 24 h incubation with NP5.

27

A) 250

8h 24 h

number of areas

200

150

100

50

0 0

1

2

3

4

5

6

area Bodipy spots [µm²]

B) 250

8h 24 h

number of areas

200

150

100

50

0 0

1

2

3

4

5

6

7

area of Texas Red spots [µm²]

Figure 10. Area distribution of the fluorescence spots of Figure 8. Bodipy signals after 8 and 24 h incubation with NP5 are shown in A), B) represents the same for Texas Red signals.

28

29

Figure 11. Possible mechanism for the release of hydrophobic drugs from polymeric nanoparticles and the intracellular fate of the carrier system after uptake.

TABLES Table 3. List of nanoparticles composed of different core materials and applied labels which were used in this study. P(HPMA)-b-P(LMA) was used as bare or labeled molecule for each particle system as stabilizing agent (surfactant).

code

core material

core label / attachment type

NP1

PDLLA

PMI /

P(HPMA)-b-P(LMA) Diameter label / attachment type

unlabeled

300 nm

unlabeled

270 nm

Oregon Green /

300 nm

incorporated

NP2

PS

PMI / incorporated

NP3

PS

Unlabeled

covalent

NP4

NP5

PDLLA

PS

PMI /

Texas Red /

incorporated

covalent

Bodipy /

Texas Red /

Covalent

covalent

250 nm

270 nm

ASSOCIATED CONTENT Supporting Information. Figure S2 shows close-up image of NP1, Scheme S1 showing the post-polymerization modification, Table S1 and S2 contain a list of the used dyes and the labeled

30

nanoparticles, Table S3 lists the excitation and emission frequencies of fluorchromes used. This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Author * University Medicine of the Johannes Gutenberg University, III. Medical Clinic, Langenbeckstr. 1, 55131 Mainz, Germany, [email protected] Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

ACKNOWLEDGMENT The authors thank the DFG for financial support (SFB 1066 “Nanodimensionale polymere Therapeutika für die Tumortherapie”). We would like to thank Kaloian Koynov for FCS measurements.

TABLE OF CONTENTS

incubation for 30 min

drug model dye

CH3 H2 C

CH3

C

CH2

HN

O

8h

C

m

n O

O

burst release

OH 11 CH3

31

REFERENCES 1. 2. 3.

4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

Tomcin, S., G. Baier, K. LandfesterV. Mailaender, Pharmacokinetics on a microscale: visualizing Cy5-labeled oligonucleotide release from poly(n- butylcyanoacrylate) nanocapsules in cells, Int. J. Nanomed., 9 (2014) 5471-5489. Hofmann, D., C. Messerschmidt, M.B. Bannwarth, K. LandfesterV. Mailander, Drug delivery without nanoparticle uptake: delivery by a kiss-and-run mechanism on the cell membrane, ChemComm., 50 (2014) 1369-71. Kang, B., P. Okwieka, S. Schottler, S. Winzen, J. Langhanki, K. Mohr, T. Opatz, V. Mailander, K. LandfesterF.R. Wurm, Carbohydrate-Based Nanocarriers Exhibiting Specific Cell Targeting with Minimum Influence from the Protein Corona, Angew Chem Int Ed Engl, 54 (2015) 7436-40. Maeda, H., Toward a full understanding of the EPR effect in primary and metastatic tumors as well as issues related to its heterogeneity, Adv. Drug Deliv. Rev., 91 (2015) 36. Matsumura, Y.H. Maeda, A new concept for macromolecular therapeutics in cancer chemotherapy: mechanism of tumoritropic accumulation of proteins and the antitumor agent smancs, Cancer Res., 46 (1986) 6387-92. Fang, J., H. NakamuraH. Maeda, The EPR effect: Unique features of tumor blood vessels for drug delivery, factors involved, and limitations and augmentation of the effect, Adv. Drug Deliv. Rev., 63 (2011) 136-51. Moad, G., J. Chiefari, Y.K. Chong, J. Krstina, R.T.A. Mayadunne, A. Postma, E. RizzardoS.H. Thang, Living free radical polymerization with reversible additionfragmentation chain transfer (the life of RAFT), Polymer Int., 49 (2000) 993-1001. Barz, M., M. Tarantola, K. Fischer, M. Schmidt, R. Luxenhofer, A. Janshoff, P. TheatoR. Zentel, From Defined Reactive Diblock Copolymers to Functional HPMA-Based SelfAssembled Nanoaggregates, Biomacromolecules, 9 (2008) 3114-3118. Nuhn, L., M. BarzR. Zentel, New Perspectives of HPMA-based Copolymers Derived by Post-Polymerization Modification, Macromol. Biosci., 14 (2014) 607-618. Rihova, B., Biocompatibility and immunocompatibility of water-soluble polymers based on HPMA, Composites Part B-Engineering, 38 (2007) 386-397. Tappertzhofen, K., M. Bednarczyk, K. Koynov, M. Bros, S. GrabbeR. Zentel, Toward anticancer immunotherapeutics: well-defined polymer-antibody conjugates for selective dendritic cell targeting, Macromol. Biosci., 14 (2014) 1444-57. Kelsch, A., S. Tomcin, K. Rausch, M. Barz, V. Mailander, M. Schmidt, K. LandfesterR. Zentel, HPMA Copolymers as Surfactants in the Preparation of Biocompatible Nanoparticles for Biomedical Application, Biomacromolecules, 13 (2012) 4179-4187. Thang, S.H., Y.K. Chong, R.T.A. Mayadunne, G. MoadE. Rizzardo, A novel synthesis of functional dithioesters, dithiocarbamates, xanthates and trithiocarbonates, Tetrahedron Letters, 40 (1999) 2435-2438. Eberhardt, M.P. Théato, RAFT Polymerization of Pentafluorophenyl Methacrylate: Preparation of Reactive Linear Diblock Copolymers, Macromol. Rapid Commun., 26 (2005) 1488-1493. Perrier, S., P. TakolpuckdeeC.A. Mars, Reversible addition-fragmentation chain transfer polymerization: End group modification for functionalized polymers and chain transfer agent recovery, Macromolecules, 38 (2005) 2033-2036.

32

16. 17.

Schaeffel, D., R.H. Staff, H.J. Butt, K. Landfester, D. CrespyK. Koynov, Fluorescence correlation spectroscopy directly monitors coalescence during nanoparticle preparation, Nano Lett., 12 (2012) 6012-7. Hofmann, D., S. Tenzer, M.B. Bannwarth, C. Messerschmidt, S.F. Glaser, H. Schild, K. LandfesterV. Mailander, Mass Spectrometry and Imaging Analysis of NanoparticleContaining Vesicles Provide a Mechanistic Insight into Cellular Trafficking, ACS Nano, 8 (2014) 10077-10088.

33

incubation for 30 min

drug model dye

8h

burst release