SELF-ASSEMBLY OF BIOHYBRID POLYMERS .1 Polymer-DNA Hybrids

Một phần của tài liệu macromolecular self assembly (Trang 214 - 223)

SELF-ASSEMBLY OF BIOHYBRID POLYMERS

7.2 SELF-ASSEMBLY OF BIOHYBRID POLYMERS .1 Polymer-DNA Hybrids

DNA is a remarkable molecule carrying the genetic information. DNA is involved in most biological processes. However, thein vivouse of bare nucleotide sequences is restricted due to their low cellular penetrability and uptake ratio as well as short plasma half-life [26, 27]. Thus a DNA-polymer conjugation that enables these issues to be overcome is currently widely being investigated. Upon conjugation with a suitable polymer, the resulting biohybrids may form nanostructures that might infer cell penetrability and thus may serve as cargos for the delivery of active compounds. Moreover, the DNA higher order structures such as hybridized complementary strands oraptamers, which are synthetic single-stranded nucleotide sequences that undergo molecular recognition as antibodies [28–31], provide a pool for post-synthesis modifications and can ensure either specific detection or targeting.

There is therefore a growing interest in designing DNA-decorated structures of potential use in drug delivery and biosensing [32, 33].

7.2.1.1 Linear Block Copolymers Coupling of a hydrophobic segment to a single-stranded nucleotide sequence results in macromolecules that are composed of the water-soluble, flexible nucleic acid strand linked to the nonsoluble segment, which could be regarded as high-molecular-weight analogues of surfactants. Along this line, Vebertet al. described the grafting of an aptamer, a single-stranded nucleic acid sequence that specifically recognizes the immunoglobulin E (IgE) protein, an allergy biomarker, to amino acid modified poly(2-alkyl-2-oxazoline) (POX) [27].

The resulting copolymer, of high-hydrophilic-weight fraction, exhibits surface activity that can be used in an emulsification process to stabilize submicrometer size oil-in-water emulsion droplets. Functionality of the aptameric strand engaged in the structure formation was assessed by reflectometry, confocal laser scanning microscopy, and atomic force microscopy imaging.

Depending on the DNA-copolymer composition, both chemical and physical incompatibility drives the structure formation through self-assembly. Coupling and structure formation may, however, affect the properties of the DNA strand involved in the assembly process, which stresses the need to establish a mechanism of functional structure formation through investigations of the modes of interaction of the resulting self-assemblies.

Although composed of nucleic acid strands, the examples reported reveal that linear DNA-based block copolymers do not undergo specific interactions such as

k k intermolecular hydrogen bonding or conformal stacking of the bases or sugar rings

in which the nucleic acid sequence can engage. As exemplified in the following dis- cussion, DNA-copolymer self-assembly is primarily driven by the attraction of the hydrophobic segment coupled to the nucleic acid strand.

For instance, the conjugation of the hydrophilicc-mycantisense strand, which suppresses smooth muscle cell proliferation [34], to the biodegradable hydrophobic poly (D,L-lactic-co-glycolic acid) (PLGA) induced self-assembly into well-defined micellar structures of narrow size distribution [35]. Being micelles composed of a PLGA hydrophobic core surrounded by a hydrophilic shell of nucleic acid strands, gradual degradation of the PLGA in the intracellular medium resulted in the release of the antisense strand into the surrounding and enhanced cellular uptake subsequent to endocytosis. A general strategy to generate polystyrene-based DNA-copolymers through solid phase synthesis was reported by Mirkin et al. [36]. Self-assembly into spherical core-shell micelles was induced through a dialysis-mediated solvent exchange from a DMF to an aqueous solution. Functionality of the nucleic acid strands engaged in structure formation was assessed through the assembly of higher order structures by hybridization with the complementary nucleic acid strands tethered to gold nanoparticles.

However, due to the high glass transition temperature (Tg) of the hydrophobic segments (about 40∘C and 100∘C), PLGA- and PS-based copolymer micelles were likely kinetically “frozen” and might not have reached thermodynamic equilibrium [37]. To overcome this issue, Herrmannet al. conjugated polypropyleneoxide (PPO) to 22-mer long nucleic acid sequences through automated DNA synthesis [37].

Self-assembly in the aqueous solution took place. The micellar core-shell structure is composed of a PPO core surrounded by a DNA corona. Functionality of the DNA strand was assessed by hybridization with the complementary strand modified by various functional groups to carry out diverse organic reactions at the interface between the micelles and their surroundings, such as Michael addition, amide bond formation, and isoindole formation [37]. The impact of formation of specific hydrogen bonds through hybridization of the complementary nucleic acid sequences with the strand engaged in self-assembly of the DNA-b-PPO structures on their morphology was studied as well [38]. Scanning force microscopy (SFM) investiga- tions revealed that hybridization with the fully complementary sequences had a very minor effect on the overall morphology of the block-copolymer micelles. However, hybridization with a much longer sequence, which encoded the complementary nucleic acid strand of the DNA-b–PPO several times, strongly affected the micellar structure. Rod-like structures were observed with a height significantly smaller than that of the initial micellar aggregates and far longer than the pristine structures, which did disintegrate. Material assembly occurred alongside the long nucleotide sequence template through hybridization, without affecting the hydrophobic core (Figure 7.4) [38].

To improve colloidal stability, inhibit nonspecific interactions and, by decreasing the electrostatic repulsion, improve the DNA micelles binding ability, Talom et al. reported on the synthesis and self-assembly of a DNA-triblock-copolymer hybrid obtained through the coupling of a 22-mer ssDNA with a poly(ethylene

k k

20 nm

10 nm

200 nm

20 nm 30 40 50

d / nm 10 0 1 2

20

h / nm

20 nm(a)

(b)

200 nm 200 nm

b a

Figure 7.4 (a) Hybridization of complementary sequences does not affect the shape of the DNA-copolymer micelle. (b) Hybridization with an elongated complementary sequence results in the formation of rod-like structures [38]. Adapted with permission of Wiley-VCH. (See color insert for color representation of this figure)

oxide)-poly(caprolactone) (PEO-b-PCL) diblock copolymer [39]. Self-assembly into micelles took place in aqueous solution. Poly(caprolactone) composed the hydrophobic core, whereas the hydrophilic corona was made of PEO and DNA.

Investigations using the quartz crystal microbalance (QCM) demonstrated that the micelles could be reversibly immobilized on surfaces modified with the complemen- tary nucleotide sequence through hybridization, which revealed functional structure formation.

Synthesis of amphiphilic block copolymers to induce their self-assembly into vesicular structures was described by Vebert-Nardinet al. [40]. Solid phase synthesis was used to couple nucleic acid strands to highly hydrophobic moieties of low glass transition temperature, namely poly(butadiene) or poly(isobutylene). TEM (Figure 7.5), DLS, as well as encapsulation and release studies showed assembly into vesicular structures. Moreover, as assessed by circular dichroism, coupling and self-assembly did not affect the DNA chain configuration.

More recently, a very appealinggrafting-fromroute of preparation of various DNA block copolymers was reported [41]. The approach involved DNA modification with an initiator and monomer addition supported by activators generated by electron transfer(AGET) ATRP. The authors found that this strategy allows the preparation of hydrophilic as well as amphiphilic DNA-based copolymers, which demonstrates the high versatility of this synthesis route. Due to the hydrophobicity of the poly- mer component, the benzyl methacrylate based DNA1-b-PBnMA-co-RMA copoly- mer aggregated into large DNA-latex (DTEX) particles by dialysis-mediated solvent exchange from acetone to water. Functionality of the DTEX particles was evidenced by hybridization.

k k

200 nm

Figure 7.5 PIB31-b-G7A5vesicles observed by TEM. Reproduced with permission from The Royal Society of Chemistry.

7.2.1.2 Comb/Graft Copolymers Chienet al. reported on DNA programmable micelles that can undergo a reversible morphological transition induced enzy- matically [42]. Spherical micelles of about 25 nm in diameter resulted from the self-assembly of a 19-mer long DNA brush-type amphiphilic copolymer based on (N-Benzyl)-5-norborene-exo-2,3-dicarboximide and (N-acetyloxy-2,5-pyrrolidine- dione)-5-norborene-exo-2,3-dicarboximide. Since it contained a RNA base (rA) cleavage site, the DNAzyme did cut the strand to affect the hydrophilic weight fraction. A sphere-to-cylinder shape transition could thus be induced by an enzymatic reaction (Figure 7.6). A 19-mer nucleotide sequence (In1, Figure 7.6) added to the self-assembled DNA-copolymer formed a 9-base duplex with the DNA-copolymer to reverse the transition. To induce the sphere-to-cylinder transition back, DNA strand invasion was used. The fully complementary 19-mer sequence of In1 (In2, Figure 7.6) interacted competitively with the 9-mer duplex. The more thermodynam- ically stable In1⋅In2 duplex did leave the shell of the micelles, which did assemble back into cylindrical structures (Figure 7.6).

More recently, Kedrackiet al. presented a novel synthetic approach to generate a comb/graft DNA-based copolymer of high (67%) grafting density through a one-step thiol click chemistry of high coupling efficiency (above 60%) [43]. As demonstrated by microscopy, the generated hybrids assembled into submicrometer size spherical structures in aqueous solution, seemingly of vesicular morphology. The structures were further shown to be functional through hybridization experiments, which also suggested the mechanical stability of these structures, and thus a high potential for application in diagnostic or sensing.

7.2.1.3 Star Copolymers The preparation of multifunctional star molecules was introduced by Das et al. [44]. The multi-arm polymer template composed of oligo(ethylene)oxide methacrylate (OEOMA) and ethylene glycol diacrylate (EGDA) on which DNA was grafted was prepared by the “arm-first” method.

The presence of an azide group at each arm of the star enabled conjugation of alkyne-functionalized DNA strands through highly efficient copper-catalyzed

k k Hydrophilic

DNA brush Hydrophobic particle core + DNAzyme

(a) (b)

Cylindrical Spherical

Spherical

100 nm 100 nm 100 nm

(c)

In1• In2 In2 In1

Figure 7.6 Assembly of DNA-brush copolymers: (a) Initial spherical morphology. (b) Tran- sition to a cylindrical morphology upon the addition of a DNAzyme. (c) Hybridization with a partially matching sequence results in the micelles reassembly. Adapted with permission of Wiley-VCH.

azide-alkyne cyclo-addition (CuAAC) “click” reaction. The authors demonstrated the functionality of the obtained structures by hybridization. Two distinct popu- lations of the polymer multi-arm structures were modified with complementary DNA sequences prior to hybridization (star-DNA1 and star-DNA 2, Figure 7.7A).

As evidenced by DLS measurements after changing the molar ratio between the hybrids, it was possible to control the size of the resulting hybridized multi-arm star superstructures. The reversibility of the process was obtained by designing the duplex with an overhang (“toehold”) available for an invasion of the fully complementary sequence, which caused the disassembly of the superstructure (Figure 7.7B).

In 2012 Maeda and colleagues reported on the synthesis of linear and mikto-arm star-shaped DNA diblock copolymers composed of poly(N-isopropylacrylamide) (PNIPAAM) and on the dependence of their assembly on various block lengths and chain architectures [45]. The PNIPAAM of different architectures were generated by ATRP and functionalized with azide groups for coupling with alkyl modified ssDNA through click-chemistry. DSC-monitored phase transition demonstrated that the PNIPAAM LCST decreased with the PNIPAAM fraction in the copolymer. DLS measurements showed lack of micellization for the copolymer with the shortest

k k

V 35 30 25 20 TEN150

95°C (2 min) to rt

15 10 5 0

0 10 20 30

Diameter (nm)

ii ii

i

iii

i

iv

(A)

(B)

Nanostructures observed 80

75 70 65 60

Median size (nm)

10 8 6 4 2 0

Population

40 50

Figure 7.7 (A) Formation of the superstructures through the hybridization of DNA mod- ified multi-arm unimolecular micelles: (v) DLS monitored formation of the superstruc- tures, black-bare star copolymers (i and ii, DNA1 and DNA2, respectively), green- ratio 1:1 (iii), purple-ratio 1:10 DNA1 to DNA2 (iv). (B) Reverse process; i-schematic representation of the disassembly upon addition of the complementary strand, ii-Mean area of DLS scans, black and red small star particles, green-1:1 ratio, yellow-1:100 ratio star polymer/invading DNA, blue-1:100 ratio star polymer/invading DNA. Adapted with permission of American Chemical Society.

PNIPAAM chain length whereas, for the higher PNIPAAM degree of polymerization (overall target DP=50/1), the assembly of colloidal particles in the size range of few and few tens of nanometers was observed. SAXS investigations suggested the formation of hard spheres with the core composed of PNIPAAM surrounded by a DNA corona. Turbidimetric based DNA detection revealed that the increase in parti- cle density and particle size enhanced aggregation due to van der Waals interaction

k k between the particle cores. These studies demonstrate that DNA hybridization can

thus be controlled by tuning the molecular structure.

A significant number of reports on hydrogels or networks composed of DNA-based copolymers have appeared, being structure formation induced through hybridization. A vast number of reports describe hydrogels based on polyacrylamide DNA conjugates [46–48] or PEG [49]. As this field of research is beyond the scope of this contribution, the reader is referred to the excellent review written by Xiong et al. [50] and an interesting article about the structure formation undergone by norbornene block copolymers composed of nucleic acid strands and ferrocenyl side chains [51].

7.2.2 Polypeptide Block Copolymers

Polypeptide-block copolymers have gained interest because the size and morphology of the self-assembled structures are responsive to solution pH, temperature, and salt concentrations [52]. This stimuli responsiveness allows for greater functions in biotechnology, as compared to the structures resulting from the self-assembly of conventional synthetic block copolymers. Moreover, the properties of polypeptide block-copolymer structures offer many advantages over other block-copolymer self-assemblies such as biocompatibility and enhanced biofunctionality. The follow- ing section focuses on how the complex three dimensional secondary structures of the polypeptide affects the self-assembly properties of peptide-based hybrid diblock copolymers.

The self-organization of poly(butadiene)-b-poly(L-glutamic acid) (PB-PGA) in aqueous solution has been studied by Lecommandoux as well as by Kloket al. [53].

The size and shape of the aggregates, as assessed by DLS, can be triggered by exploit- ing the secondary structure of poly(L-glutamic acid) with pH [54, 55]. In the PB-PGA system, the hydrophobic PB provides the driving force for self-assembly and forms the core of the micelle, while the hydrophilic polypeptide (PGA) forms the corona.

The stimuli responsiveness of the structure relies on the swelling induced by changes in pH, temperature, ionic strength, and solvent type. PB40-b-PGA100is readily solu- ble in water and, at basic pH, well-defined vesicular structures were observed [54].

Even at high ionic strength, when the charges of the polypeptide blocks are strongly screened, the size of the aggregate could still be varied by changing the pH. CD spec- troscopy studies showed that there is a transition from a compact𝛼helical secondary structure at low pH (pH= 4.5) to a more extended random coil structure in basic media (pH=11.5). Such pH induced size variations were completely reversible and could be regenerated numerous times [53].

Lecommandoux et al. further studied the impact of the polypeptide secondary structure on polypeptide-b-polymer self-assemblies by considering three dif- ferent systems based on polybutadiene-b-poly(L-glutamic acid) (PB-b-PGA), polyisoprene-b-poly(L-lysine) (PI-b-PLys), and poly(L-glutamic acid)-b- poly(L-lysine) (PGA-b-PLys) [55]. Basically the self-assembly of either spherical

k k core-shell micelles or vesicles were formed depending on the rod-(polypeptide)-

to-coil block molar ratio as assessed by light scattering. For a value lower than 65%, vesicles could be obtained, whereas for greater values, stable spherical core-shell micelles were assembled.

The pH and ionic strength responsiveness of the micelles and vesicles was fol- lowed by DLS. The variation of the conformation (the transition from 𝛼 helix to coil) of the polypeptide block within the self-assembled structure was studied by CD spectroscopy. This modification of the secondary structure resulted in a change of the overall size of the micelles and vesicles. The observed size variation could be attributed to the neutralization of the polypeptide block that thus did undergo a conformation transition from a charged extended coil to a neutral and compact α-helical rod in a reversible manner. This responsiveness was active and efficient even in media of high ionic strength, in which electrostatic interactions are screened [53].

PPO44-PLys10(0.33 hydrophilic weight fraction) readily formed vesicular structures at low pH [56]. However, this species has limited solubility at pH>5, and hence the pH responsiveness could not be determined [56]. PPO44-PLys62 and PPO44-PLys44 [57–64] are of 0.76 and 0.69 hydrophilic weight fractions, respectively. At low pH, the PLys blocks are in a fully stretched conformation. The RG/RH ratio estimated from light-scattering experiments was found to be 1.1 and 1.01 for PPO44-PLys62and PPO44-PLys44, respectively. This indicates that these block copolymers did assemble into vesicular structures as evidenced further by TEM measurements. The vesicle formation at these relatively high hydrophilic weight fractions could be explained by the partial hydration of the PPO core. The PPO-PLys block-copolymer assemblies did undergo an assembly-to-unimer transition with decreasing temperature. As the temperature was below the LCST of PPO (8∘C), the block copolymer became double hydrophilic and the copolymer existed as unimers. The same process did arise at low pH (∼3). But, as the pH is increased above the pKa of PLys, the polypeptide became more hydrophobic. In case of PPO44-PLys62,the PLys block became insoluble at pH

>7. Contrary to the assembly occurring at lower pH, the structures formed did consist of a PLys core surrounded by a PPO corona. From DLS experiments, it was observed that the hydrodynamic radii of the vesicles increased from 10 to 75 nm, upon increas- ing the pH to 7. The CD data revealed anα-helical configuration at this pH. At pH 8, the solution became turbid due to the inability of the PPO blocks to stabilize the aggregates.

7.2.3 Block Copolypeptides

In the case of copolypeptide-based systems such as PGA15-b-PLys15, a morpho- logical transition of the self-assembly, was observed in addition to a change in size [65–67]. This class of zwitterionic copolypeptides can reversibly self-assemble into vesicles in moderate acidic or basic aqueous solutions. They are called schizophrenic vesicles. Figure 7.8 schematically represents the responsiveness of the diblock copolypeptide with pH. At neutral pH, the zwitterionic diblock copolymer

k k acid pH (<4)

Neutral pH (5<pH<9)

basic pH (>10)

ΔpH

PLys

PGA PLys

PGA

Figure 7.8 Schematic representation of the schizophrenic self-assembly into vesicular struc- tures of PGA15-b-PLys15. Adapted with permission from American Chemical Society. (See color insert for color representation of this figure).

bears the positive charges of the protonated lysine and negative charges of the deprotonated glutamic acid. The dissolution of the diblock copolymer in aqueous basic and acidic solutions results in spontaneous self-assembly. At acidic pH (pH ∼ 3), the poly(L-glutamic acid) block is neutralized, which changes the sec- ondary conformation from a charged coil to a more compact neutral𝛼helical (“rod”) structure. This structural variation is accompanied with a decrease in solubility, and hydrophobicity becomes the driving force for the self-assembly. The low interfacial curvature of the “rod”-like hydrophobic part induced the formation of hollow struc- tures. The insoluble PGA did form the core of the aggregates, while the PLys block composed the shell. In basic conditions (pH∼12), a reversible process took place.

The protonated poly(L-lysine) block (NH3+) is transformed into a neutral (NH2) group. These NH2 groups are highly insoluble and participate in the formation of the hydrophobic core. The PLys block formed the shell in this case. At intermediate pH, the blocks are charged, the assembled structures disassembled, and precipitation eventually occurs. This precipitate can be re-dissolved and the process completely reversed by adding a suitable acid or base. Even at high salt concentrations, when the electrostatic interaction is expected to be screened, this reversible structural formation can be induced. This spontaneous self-assembly was proven by1H NMR and fluorescence spectroscopy. The sizes of the vesicular structures were determined by DLS. Values of 110 and 175 nm were calculated for self-assembled structures of the dipeptide in water at acidic and basic pH, respectively.

Một phần của tài liệu macromolecular self assembly (Trang 214 - 223)

Tải bản đầy đủ (PDF)

(275 trang)