Tuning thermoresponsive behavior of diblock copolymers and their gold core hybrids. Part 2. How properties change depending on block attachment to gold nanoparticles

Abstract

Thermoresponsive diblock copolymers of di(ethylene glycol) methyl ether methacrylate (DEGMA) and oligo(ethylene glycol) methyl ether acrylate (OEGA) were synthesized by reversible addition–fragmentation chain transfer polymerization, allowing us to prepare diblocks with a thiol group at the desired chain end, and bond that block to a ∼20 nm gold nanoparticle core. The cloud point and coil–globule transition window were measured by UV–vis spectroscopy. The gold core lowered the cloud point and narrowed the coil–globule transition window of all the diblock hybrids, but raised the cloud point of statistical copolymer hybrids that had similar cloud points. The extent of the change in the thermo-response properties of the hybrid diblock copolymers was more significant when the gold was bonded to the DEGMA block than the OEGA block. This block is less hydrophilic and sterically hindered than OEGA and may adsorb more effectively to the gold so that the hydration of the outer OEGA block is relatively unaffected by the Au core. This work indicates that diblock copolymers allow factors such as steric bulk and the effects on arrangement around a metal core to be effective tools for manipulating thermo-responsive properties that are not as significant with statistical copolymers.

Introduction

This paper reports the effects of bonding diblock copolymers to gold nanoparticles (AuNPs) of ∼20 nm diameter. The study tested the effects of which block was bonded directly to the AuNP on the cloud point (CP) and the temperature range over which the coil-to-globule transition occurred. We also compared the effects to statistical copolymers with similar CPs and their hybrids. The diblock copolymers were made using DEGMA and OEGA and were prepared by reversible addition–fragmentation chain transfer (RAFT) copolymerization to yield well-defined block structures with a dithioester at one end (subsequently reduced to a thiol) and a carboxylic acid group at the other end. The colloidal gold nanoparticle (AuNP) cores were pre-synthesized by citrate reduction [1], [2], [3], [4], [5]. While there are some studies that have explored gold/polymer hybrid nanoparticles (AuHNPs) of homo- and co-polymers [6], [7], [8], [9], [10], [11], [12], [13] this appears to be the first study of the effects of thermoresponsive AuHNPs prepared using diblock copolymers and evaluating the effects of which block is bonded to the AuNP on thermal response properties.

In prior work [14] we showed that the chain end placement of organic amphiphilic end groups (dithioester and carboxylic acid) on the DEGMA-OEGA diblock copolymers could alter the CP by as much as 28 °C and narrowed the coil-to-globule transition range by as much as 70% (from 13 °C to 4 °C) simply by reversing the end group placement on the diblock of an otherwise equivalent composition. In this work we performed a similar test using the AuNP as a “chain end”. This was done by reduction of the dithioester chain end, and bonding the thiol to the AuNP, while maintaining the other chain end as a carboxylic acid.

The rationale for this design was threefold. Firstly, DEGMA and OEGA were selected for the diblock because they are both biocompatible and possess pendant ethylene glycol units. This will allow them to be used in the bloodstream and resist non-specific protein adsorption without “PEGylating” the surface as is typically done with nanoparticles intended for use in the body and delivery via the bloodstream [15], [16], [17].

Secondly, a diblock polymer matrix makes sense for a nanoparticle intended for use as a drug delivery device because one block can be designed to provide a suitable domain for the intended drug with an outer block designed to stabilize the nanoparticle in the bloodstream. In the prior work [14] we compared the relative effects of the length of the less hydrophilic DEGMA and more hydrophilic OEGA blocks, the placement of the amphiphilic chain ends, the “methyl effect” (acrylate vs. methacrylate backbone), and making one block a statistical OEGA/DEGMA block and the other a pure block, on CPs and the coil–globule transition temperature range. The large number of variables allowed one in principle to tune the “inner domain” polarity to make it suitable for a drug, and still control the final CP by adjusting the other variables while maintaining the outer OEGA block for stability in the bloodstream.

The rationale for coupling the diblock structure with an AuNP core was due to our ultimate objective, which was to design a thermo-responsive nanoparticle containing drug(s) for the treatment of cancer. For this purpose we wanted a nanoparticle that could travel safely through the bloodstream with minimal loss of the drug, target cancer cells by bearing a targeting device on the nanoparticle surface and deliver all or most of the therapeutic drug only once inside the cancer cells. The value of using an AuNP core in this design is that it absorbs 527 nm light (the absorption wavelength is tunable to the near IR region), which is safe for living tissue, but it causes the AuNP core to heat up. Therefore we can design the diblock copolymer in an AuHNP to possess a CP significantly higher than 37 °C (e.g. 44 ± 1 °C) to significantly reduce diffusion release of chemotherapy agents in the bloodstream (which is at ∼37 °C), compared to thermo-responsive nanoparticles that are often designed with CPs closer to 37 °C. Therefore a patient would potentially experience fewer side effects because less chemotherapy drug is released outside the cancer cells, and potentially less drug is required to achieve the same positive effect.

Yet another advantage of using an AuHNP with an AuNP core is the potential to couple chemotherapy with hyperthermia. In this design, after allowing appropriate time for the chemotherapy drug(s) to work, an additional treatment with 527 nm light can be applied to heat the local area to even higher temperature to use hyperthermia [18] as a secondary treatment.

In this design then, an AuHNP containing chemotherapy drugs would be designed with an and AuNP inner core to convert 527 nm light to heat, an diblock copolymer shell with an inner block designed to be compatible with the desired drug(s), an outer OEG(M)A or DEG(M)A block for blood stability, and a final targeting device chemically bonded to the outer block for targeting cancer cells (Scheme 1). The CP itself would be controlled to a desired temperature (e.g. 44 ± 1 °C) by control of block lengths and block composition using either acrylate or methacrylates and balancing pure or statistical blocks as desired to get the best balance of CP and drug compatibility. Although our final AuHNP includes a targeting device that we did not include in this paper, the scope of this paper is focused on the study of the effect of the AuNP core on the CP of the block copolymers depending on block length, design, and AuNP attachment site.

The preparation of AuNPs and the attachment of an organic polymer to AuNPs are well known in the literature and not discussed here [19], [20], [21], [22], [23], [24], [25], [26]. We attached the copolymers to the AuNP by a grafting-to approach [27], [28], [29] rather than one-pot direct-synthesis [30], [31], [32], grafting-from [33], [34], [35] or physical adsorption [36], [37] because grafting-to allowed us to better control the AuNP size.

Most of the research reported in the literature on thermoresponsive AuHNPs is concerned with homopolymers, and most often uses thermoresponsive poly(N-isopropylacrylamide) (PNIPAM) with AuNPs. Among these studies are, for example, Zhu and co-workers [6] who observed that the CP of PNIPAM-functionalized AuHNPs decreased to 28.4 °C compared to a CP of 33.5 °C for the original homopolymer. Transition width was also decreased from the original 4.0 °C to 2.6 °C (AuHNP) but no deeper discussion followed. Shan [32] studied how the molar mass of PNIPAM affected the CP of AuHNPs and reported that increasing molar mass increased the CP. The Au core size also affected the CP but this was reported to be less significant than the molar mass of PNIPAM. An effect on the coil–globule transition window (ΔT) was noted [7], and the AuHNPs had a narrower ΔT compared to PNIPAM alone, but again no explanation was provided.

Studies of AuHNPs with copolymer shells include poly(stryrene-b-NIPAM) (PSt-b-PNIPAM) [8], [9] where the effects of the Au core size and MW of copolymers on the CP and ΔT of AuHNPs were reported, but no comparison was made to the homo PNIPAM, nor was the effect of changing the attachment block studied. Kim [10] recently reported preparation of water-soluble AuHNPs stabilized with PEO-b-PNIPAM copolymers but their paper focused on the RAFT synthesis of PEO-b-PNIPAm copolymers without further study on the thermoresponsive behaviors of AuHNPs.

Copolymer shells of oligo(ethylene glycol) (meth)acrylates (OEG(M)A) have attracted interest and a few studies of POEG(M)A-coated AuHNPs were recently reported for application in protein-antifouling [11], [12]. Gibson [13] used homo-POEGMA as the polymer shell in their study and found that the CP of their AuHNPs was lower than that of the free POEGMA. In their study, a more significant decrease of CP was given by an increase of Au core size and CP value was independent of polymer concentration.

Other AuHNP studies include surface plasmon resonance [38], where PNIPAM brushes (with and without a cross-linker) were attached to Au surface in a “grafting-from” approach, using atom transfer radical polymerization (ATRP) [39]. AuHNPs have been prepared with a biotin end [40], in a reversed core/shell structure (polymer as the core and Au as the shell) [41], and a recent report used AuHNPs that offer a thermo-/photo-triggered release of an anticancer drug 5-fluorouracil [42].

To date, no research appears to have discussed how the block sequences of copolymers alter the CP of AuHNPs. This paper is the first to report the effects of block copolymer attachment sequence on the thermoresponsive properties of AuHNPs, and the first to compare those with the equivalent statistical copolymers, and is focused only on copolymers of OEG(M)A and DEG(M)A because they are biocompatible, and either block could be used as the outer block to provide stability in the bloodstream and significant versatility in controlling the CP.