pH-responsive dithiomaleimide-amphiphilic block copolymer for drug delivery and cellular imaging.

A drug delivery system that is integrated with fluorescent imaging is an emerging platform for tumor diagnostic and therapy. A pH-responsive fluorescent polymer that can respond to the surrounding medium is a desired component with which to construct an advanced drug delivery system with bioimaging characteristics and controllable drug releasing. In this work, we synthesized novel amphiphilic block copolymers of poly(ethylene glycol)-b-poly(2-(diisopropylamino) ethyl methacrylate-co-dithiomaleimide) (PEG-b-poly(DPA-co-DTM)) and poly(ethylene glycol)-b-poly(2-(dibutylamino) ethyl methacrylate-co-dithiomaleimide) (PEG-b-poly(DBA-co-DTM)) with pH-responsiveness and fluorescence. The block copolymers exhibited relatively stable fluorescence properties in different solvent and excitation-independent fluorescence behaviours. By copolymerizing the responsive segments in the molecule chain, the doxorubicin (DOX)-loaded micelles could be triggered to disassemble, thus releasing DOX at the corresponding pH values and yielding a pH-responsive drug release. Targeted deliveries of the drug within the cell were demonstrated by using the carrier responding to different pH values. The best antitumor effect was obtained by PEG-b-poly(DPA-co-DTM), which immediately released DOX as soon as it entered the tumor cells, as a result of responding to the regional pH level (pH = 6.3). The pH-responsive copolymers showed excellent biocompatibilities, as nearly 85% of cells with these fluorescent micelles survive when the testing concentration goes up to 200 μg mL-1. In all, these pH-responsive and dithiomaleimide-based fluorescent block copolymers hold great potential in future cancer diagnostic and therapeutic techniques.


Introduction
Fluorescence imaging, a reliable strategy to detect and visualize bio-subjects both in vitro and in vivo [1][2][3], has led to wide applications in clinical practices [4][5][6][7]. The key features for fluorescence imaging technique, normally driven by high specification fluorescent probe, are the high sensitivity and non-radioactivity [1,8] that can acquire feedbacks at molecular level as well as good biocompatibility [1]. Recent developments on quality fluorescence imaging suggested few technical bottlenecks, i.e., aggregationcaused quenching of traditional organic fluorophores [9][10][11], cytotoxicity of semiconductor Quantum Dots [12][13], low luminescence intensity of lanthanide luminescent materials [3]. Therefore, there is a timing demand to enhance and/or seek novel fluorescent probes to address the above technical gaps. At the same time, most probes mentioned above present poor water solubility and short circulation time, thus many polymer-based micelle systems have been developed for transportation of bioimaging probes [14]. For example, Zhu's group reported a glucose-based NIRfluorescence polymer, PMMA-b-P(GATH-co-BOD) for cancer detection with strong recognition towards GLUT1 [15].
Combining bioimaging with smart drug release to get multifunctional nanomaterials is of great significance in diagnostic and therapeutic applications [16][17]. These imaging-guided drug delivery systems can visualize the accumulation of nanoparticles and simultaneously deliver and trace the drug, thereby evaluating the efficiency of drug delivery [17][18]. For molecular therapies, these nanomaterials may be able to understand the pharmacokinetic processes, which are important for nanomedicine [18]. Fluorescent and pH-responsive polymers are popular in imagingguided drug delivery systems, being able to recognize the small pH changes around tumor microenvironment, where the pH value in early endosomes is about 5.9-6.2, but around 5.0-5.5 in the late endosomes/lysosomes [19][20][21][22][23]. However, advanced drug delivery strategy with high pH sensitivity and fast responsive remain yet to be fully exploited.
In this work, we propose a novel fluorescent and pH-responsive material strategy by co-polymerizing fluorescent DTM groups with pH-sensitive polymers (PDPA and PDBA), aiming to improve efficiency for cellular imaging and drug release. DOXloaded micelles self-assembled from PEG-b-poly(DPA-co-DTM) or PEG-bpoly(DBA-co-DTM) were taken up by tumor cells, the pH-responsive parts became positively charged at low pH leading to dissociation of micelles into unimers and therefore releasing DOX. Simultaneously, the DTM probes exhibited bright green fluorescence to achieve cellular imaging. The drug releasing roadmap is shown in Scheme 1.  [25], as well as PEG-Br [42] were synthesized according to a previously reported method.

Synthesis of methacrylate-functional DTM (DTM-MA)
DTM-MA was synthesized via the "click" reaction. Alkyne-DTM (934.14 mg, 3 mmol), AzPMA (558.13 mg, 3.3 mmol) and PMDETA (138.4 mg, 0.8 mmol) were dissolved in 5 mL anhydrous dimethylformamide (DMF). The mixture was degassed via three freeze-evacuate-thaw cycles followed by the addition of CuBr (114.8 mg, 0.8 mmol). The solution was allowed to stir for 24 h under N2 atmosphere at 50 °C. Then the solution was exposed to air, and diluted with dichlormethane (DCM). The mixture was then passed through a neutral alumina and the residue was purified by silica gel column chromatography (hexane : ethyl acetate = 2:1, v/v) to yield yellow liquid (1.2 g, 83%). 1   anhydrous DMF. The mixture was degassed via three freeze-evacuate-thaw cycles followed by the addition of CuBr (2.87 mg, 0.02 mmol). After 12 h of stirring under an N2 atmosphere at 60 °C, the reaction bottle was quickly quenched into an ice bath and exposed to air, then diluted with DCM. The mixture was then passed through a neutral alumina. The filtrate was concentrated and precipitated into excess cold diethyl ether twice and dried in a vacuum to yield yellowish-brown solid.

Preparation of micelle nanoparticles
Micelles were prepared by a dialysis method. PEG-b-poly(DPA-co-DTM) (40 mg) was dissolved in 1.5 mL DMF for 20 min, then 4 mL deionized water was added slowly to the above solution. After another stirring for 2 h, the mixture was dialyzed against

Synthesis and characterization of block copolymers
The fluorescent and pH-responsive amphiphilic block copolymers of PEG-bpoly(DPA-co-DTM) and PEG-b-poly(DBA-co-DTM) were synthesized according to  (Figure S1-S2). The triazole group signal at 7.6 ppm appeared in the 1 H NMR proved the successful synthesis of DTM-MA (Figure 1A and S3). The infrared spectra ( Figure 1C) evidenced the synthesis of AzPMA from the disappearance of hydroxyl group peak at 3334 nm -1 and the onsite of ketone group at 1720 nm -1 , and the formation of DTM-MA from the disappearances of the alkynyl group peak at 3189 nm -1 and N3 group at 2100 nm -1 .
The chemical structure for synthesized block copolymers were also characterized.
For P1 (Figure 1B), the peaks at 3.0 ppm and 2.6 ppm associated with the methylene and methine next to the nitrogen atom in the PDPA and the peak at 7.6 ppm referred to the triazole group in the PDTM proved the block copolymers formation. Whereas for P2 (Figure S4), the peaks referred to the methylene next to the nitrogen atom in the PDBA were found at 2.5-2.7 ppm, and at 7.6 ppm for the triazole group in the PDTM.
The molecular weight results measured by GPC and NMR are shown in Table S1 and Particle size is an important property of polymeric micelles for drug delivery and release. For the designed amphiphilic block copolymers P1 and P2, the hydrophilic segment PEG is the shell and the hydrophobic segment P(DPA-co-DTM) or P(DBAco-DTM) is the core of the micelle. DLS results of intensity indicated that the size distributions were 28±8 nm and 37±6 nm for P1 and P2 micelles, whereas the cryo-TEM results showed smaller size distributions at 16±2 nm and 27±4 nm (Figure 2a and 2b). After loading DOX into the micelles, cryo-TEM observations for DOX-loaded P1 micelles and DOX-loaded P2 micelles revealed that the size distributions were 27±5 nm and 36±8 nm, which were also smaller than the DLS results (39±10 nm for DOX-loaded P1 micelles and 47±15 nm for DOX-loaded P2 micelles) (Figure 2c and 2d). After encapsulating the drug into the micelles, the size distributions of DLS showed approximately a 30% increment for DOX-loaded P1 micelles and 21% increment for DOX-loaded P2 micelles. However, both DOX-loaded micelles were within 200 nm, therefore could penetrate and accumulate in tumor tissues through the enhanced permeability and retention (EPR) effect [43][44].
As it is well known, the PDPA and PDBA polymers hold pH-responsive properties, with a sharp pH responsive transition from hydrophobic to hydrophilic responding to pH changes from 7.4 to 6.3 (for PDPA) and from 7.4 to 5.4 (for PDBA), respectively [25]. To assess the micelle stability at different pH levels of P1 and P2 micelles, we measured the size changes of the micelles (Figure 3a, 3c [25,45]. This characteristic was also confirmed by cryo-TEM, when the pH was altered to 6.0 or 5.0, there were almost no spherical nanoparticles with normal size after 30 min of incubation (Figure 3b and 3d).

In vitro drug release
Recent studies reveals that pH value in tumor issues is a little lower than in normal tissues, which is beneficial for targeting applications. The intracellular pH value is lower than extracellular, typically 5.9-6.2 at early endosomes and 5.0-5.5 at late endosomes/lysosomes [19][20][21] At a pH value of 7.4 (Figure 4), the DOX releasing efficiencies remained relative stable with less than 30% from both the DOX-loaded P1 and P2 micelles in 36 h. At pH 6.0, we observed an acceleration of drug releasing for DOX-loaded P1 micelles, where approximate 53% of the loaded DOX were released in the first 12 h, then it climbed up to 72% at 36 h. Similarly, approximate 42% of DOX were released from DOX-loaded P2 micelles in the first 12 h, then the releasing rate reached 69% at 36 h.
These evidenced that the micelles produced in this work presented accelerated drug release behavior at corresponding lower pH values.

Fluorescence spectroscopy
When two alkyl thiol groups are added into a non-fluorescence dibromomaleimide molecule, the correspondent dithiomaleimide molecule will be formed with strong fluorescence signal [30]. The mechanism was previously reported by other researchers that thiol groups could saturate the C=C double bond of the maleimide, subsequently eliminating the quenching of the fluorophore and eventually making the fluorophore excitative [46]. The above mechanism also applies in our P1 and P2 copolymers.
We assessed the fluorescence property of our samples. Under methanol environment (Figure 5a), the maximum excitation wavelength of both P1 and P2 copolymers were observed at 405 nm, and the maximum emission wavelength were at 550 nm, which was in a good agreement with that of DTM-MA monomer. In DI-water (Figure 5b), the maximum excitation wavelength showed no shift, however, there was a little shift from 550 nm to 520 nm for maximum emission wavelength, known as a solvatochromic emission with a blue-shift when changing solvent polarity [46]. The solvents affected not only the emission wavelength, but also the intensity of emission wavelength. When copolymer P1 and P2 were dissolved in H2O/DMF mixture with different water fractions (fw), the samples exhibited proportional fluorescence quenching, and the emission intensity just decreased to a constant value instead of fully quenching ( Figure   S7-S8). These obvious fluorescence quenching might be greatly caused by the hydrogen bonding between water and the carbonyl moiety of DTM group [46][47]. And after exposing at a UV light of 365 nm, our samples exhibited visible green fluorescence (Figure 5c).

Cellular uptake and cytotoxicity
The cellular uptake experiments were performed for both P1 and P2 micelles against A549 cells to understand the dynamic fluorescence behavior at practical cell environment [48]. For pure micelles, a weak green fluorescence signal appeared in Similarly, the results from free DOX also indicated a low antitumor efficiency, resulting from the diffusion-controlled transportation from extracellular matrix to intracellular system for drug [49].
To further investigate the biocompatibility of copolymers and cytotoxicity of DOXloaded micelles, Annexin V-PE/7-AAD apoptosis detection assay was used. On the other hand, high rates of apoptosis were observed for DOX-loaded micelles, where an apoptosis ratio of 26.19% was found for DOX-loaded P1 micelles and an apoptosis ratio of 19.31% for DOX-loaded P2 micelles. The results were in a good agreement with cytotoxicity profiles where DOX-loaded P1 micelles had a higher antitumor efficiency to A549 cells than DOX-loaded P2 micelles.

Conclusions
In summary, we successfully developed two fluorescent pH-responsive block copolymers with a robust fluorescence performance that could resist the solvent intervention. The DOX-loaded P1 micelles show the best antitumor effect by immediately releasing DOX once entering the tumor cells, as a result of the response to the regional pH level (pH=6.3). This early release of drug not only extends the working time for the drug but also kinetically enhances the toxicity to kill cancer cells. Moreover, both of the copolymers PEG-b-Poly(DPA-co-DTM) and PEG-b-Poly(DBA-co-DTM) show excellent biocompatibility with nearly 85 % of cells with these fluorescent micelles surviving when the tested concentration is up to 200 μg mL -1 . Comparing to previous studies [39,44], this work presents higher drug loading, smarter responsiveness to microenvironment and great fluorescence performance. With superior fluorescence property and sensitive pH-responsive drug release property, we expect these polymeric micelles technology can find potential applications in biomedical and clinical therapies.

Conflicts of interest
There are no conflicts to declare.