Pulsating Polymer Micelles via ATP-Fueled ... - ACS Publications

Sep 29, 2017 - ABSTRACT: Energy dissipation underlies dynamic behaviors of the life system. This principle of biology is explicit, but its in vitro mi...
0 downloads 4 Views 3MB Size
This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Letter pubs.acs.org/macroletters

Pulsating Polymer Micelles via ATP-Fueled Dissipative Self-Assembly Xiang Hao,† Wei Sang,† Jun Hu,‡ and Qiang Yan*,† †

State Key Laboratory of Molecular Engineering of Polymers, Department of Macromolecular Science, Fudan University, Shanghai 200433, China ‡ State Key Lab of Polymer Physics and Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun 130022, China S Supporting Information *

ABSTRACT: Energy dissipation underlies dynamic behaviors of the life system. This principle of biology is explicit, but its in vitro mimic is very challenging. Here we use an energy-dissipative self-assembly pathway to create a life-like polymer micellar system that can do periodic and self-adaptive pulsating motion fueled by cell energy currency, adenosine triphosphate (ATP). Such a micelle expansion−contraction behavior relies on transient supramolecular interactions between the micelle and ATP fuel. The micelles capturing ATPs will deviate away from the thermodynamic equilibrium state, driving a continuous micellar expansion that temporarily breaks the amphiphilic balance, until a competing ATP hydrolysis consumes energy to result in an opposing micellar contraction. As long as ATP energy is supplied to keep the system in out-of-equilibrium, this reciprocating process can be sustained, and the ATP level can orchestrate the rhythm and amplitude of nanoparticulate pulsation. The man-made assemblies provide a model for imitating biologically time-dependent self-assembly and periodic nanocarriers for programmed drug delivery. Here we describe a “living” polymer micellar system in which each micelle nanoparticle can undergo rhythmically pulsating motion driven by a crucial biochemical fuel, adenosine triphosphate (ATP). In order to achieve this dynamic selfassembly cycle, our design is to couple switchable micelles to an ATP-driven transient supramolecular interaction, using the following design elements: (i) a class of specific polymer micelles bearing ATP-binding receptor units can switch between an expansive and contractive state with altering the polymeric hydrophilic−hydrophobic ratio; (ii) by capturing ATP, the formation of the ATP/polymer complex can increase the polymer hydrophobicity, which drives a micellar expansion due to a shift of amphiphilic balance (as a forward path); (iii) a deactivation effect through the consumption of ATP can restore the systemic hydrophobicity, leading to an opposing micellar contraction (as a backward path); (iv) two distinct routes for “activating” and “deactivating” the noncovalent interactions constitute a feedback loop for lasting this micellar pulsation. The long-life operation of this nonequilibrium system can be maintained via a constant source of external energy, as shown in Figure 1. To meet the criteria (i)∼(iv) above, we developed a class of amphiphilic block copolymer composed of a poly(ethylene oxide) block (PEO) and a functional block (PCD) appended with biguanidine-cyclodextrin side units (Figure 1, bottom

O

ne of the most inherent attributes and intriguing abilities of living systems is that they are dynamic, with the autonomous motion regulation.1 For instance, deformable motion of cells and dynamic lengthening/shortening of microtubules underlie vital functions such as cell motility, cell division, and cell replication.2,3 Biomimicking these natural dynamic systems is a long-term dream of chemists.4−6 Although some studies have demonstrated that artificial molecular assemblies can undergo reversible deformation in response to external stimuli,7−11 they are not yet comparable to those biological counterparts due to their essential distinction in selfassembly processes.12 Synthetic self-assembly systems run on thermal equilibrium, producing structures that are stable in time. In contrast, natural self-assembly systems operate away from equilibrium, requiring a persistent energy input to sustain their continuous structural evolution. This is referred to as dissipative self-assembly,13,14 and it plays a central role in underpinning the biological dynamics. Therefore, cooperating this energy-induced self-assembly pathway with man-made assemblies might create a life-like system, with structural adaption and autonomous motion features. Currently, the synthetic systems fabricated by dissipative selfassembly are very limited. The few examples are focused on active macroscopic materials such as gels15−18 and films.19 Nanoscaled dissipative self-assembly systems have received relatively little attention.20 A main reason is that integrating energy input into an intricate self-assembly system to propel a dynamic and time-dependent self-assembly evolution is quite difficult. © XXXX American Chemical Society

Received: August 24, 2017 Accepted: September 29, 2017

1151

DOI: 10.1021/acsmacrolett.7b00649 ACS Macro Lett. 2017, 6, 1151−1155

Letter

ACS Macro Letters

Figure 1. Schematic illustration of the pulsating polymer micelle via ATP-fueled dissipative self-assembly process (top) and the structure of PEO-bPCD diblock copolymer and ATP molecule.

Figure 2. (a) DLS results for the size change of PEO-b-PCD micelles in the absence and presence of ATP (100 μM). Inset, the micellar expansive rate as a function of ATP treatment time. (b) Fluorescent intensity (FI) at 428 nm plotted versus the incubation time with different ATP levels, using DPH as a probe molecule.

decay of ATP/polymer complex structure, which will dissipate the energy stored in the micelles and further lead to the micellar contraction. The competition of the forward complexation and the backward enzymatic dissociation between ATP and the polymer sets up a closed supramolecular cycle. Continuous ATP energy supply will circulate this micellar swelling−shrinking process, resembling a particulate periodic pulsation (Figure 1, top panel). The forward micellar motion (expansion) triggered by ATP was first studied. PEO-b-PCD copolymer (polymer concentration is 0.4 mg mL−1, which contains ∼184 μM of ATPreceptor units) can form aggregates in aqueous solution owing

panel). In our previous work, this positively charged side unit has proved to be a specific receptor for capturing ATP.21 In this study, we demonstrate that this PEO-b-PCD copolymer (Mn,GPC = 21.7 kDa, per polymer chain contains 10 ATPreceptor units) can form small micellar nanoparticles. Upon addition of ATP, these artificial receptors inside the micelles can associate with ATP species in solution to result in an increase of micellar hydrophobic domains, which provides an uphill energy to favor micellar expansion. On the other hand, potato apyrase22 is a phosphatase that can decompose ATP to phosphate (Pi) and adenosine monophosphate (AMP). The internalization of this enzyme into the micelles enables the 1152

DOI: 10.1021/acsmacrolett.7b00649 ACS Macro Lett. 2017, 6, 1151−1155

Letter

ACS Macro Letters to its amphiphilic nature. Dynamic light scattering (DLS) showed that their hydrodynamic diameter, Dh, is 18.4 nm with PDI of 0.06, indicating a uniform nanostructure formed (Figure 2a, red curve). The TEM image further revealed that they are sphere-like micelles with the mean size of 16 ± 3 nm (Figure S1), in accordance with DLS results. With the addition of ATP (100 μM), interestingly, the micelles began spontaneous expansion. Their aggregated size produced a 3-fold increase to 57.1 nm (PDI = 0.11) within 30 min, while their globular shape was maintained, as confirmed by TEM (Figure 2a, blue curve; Figure S2). The change in micelle size has a near-linear relationship to ATP treatment time, and the swelling kept at a rate of ∼1.2 nm/min (Figure 1a, inset). Since we have proved that the negatively charged ATP can bind to the positively charged ATP receptors by ligand−receptor interactions (KB = 7.52 × 107 M−1),21 the micellar expansion may arise from an increase of systemic hydrophobicity which was caused by the generation of a neutralized ATP/polymer complex (Figure S3). To validate this viewpoint, we used an apolar-sensitive probe, 1,6-diphenyl1,3,5-hexatriene (DPH, λem = 428 nm), to examine the enlargement of the hydrophobic domain in micelles by fluorescent intensity change.23 As expected, an ATP-dependent fluorescence enhancement could be found (7-, 12-, and 18-fold change upon 50, 100, and 200 μM ATP, Figure 2b), which suggests that the more ATP added, the higher the degree of the swollen micelle. Consistent results were obtained by DLS (Figure S4). Different ATP levels can cause different expansive scopes from 18.4 nm (initial) to 30.9 nm (50 μM ATP), 55.6 nm (100 μM ATP), and 70.9 nm (200 μM ATP), respectively. In addition, we found that if the ATP level exceeds 200 μM the micellar size has no obvious change. This implies that there is an upper limit of ATP associating with the polymeric micelles. We used the UV−vis spectrum to ensure this ATP upper limit is 211 μM (Figure S5), slightly larger than the theoretical value (184 μM). After confirming the forward micellar motion driven by ATPinduced supramolecular complexation, we attempted to switch on a backward motion for micellar contraction. To fulfill this goal, installing an additional process to dissipate the chemical energy stored in ATP and break the ATP/polymer complexes is necessary. Potato apyrase is a suitable candidate for this as it can hydrolyze ATP to AMP and two Pi species. Neither AMP nor Pi can form complexes with the copolymer owing to their weak affinity (KB < 102 M−1, Figure S6), guaranteeing that the two decomposed products have no effect on stabilizing the micellar structures. Thus, when ATP (100 μM) was added to the initial micelle system pretreated with this enzyme (12 U/ L), an unexpected micellar evolution could be tracked over time by cryo-TEM. The polymer micelles were first self-dilated, reaching a maximum within 40 min (size from 16 ± 3 nm up to 55 ± 9 nm by averaging over 200 nano objects in TEM, Figure 3a,b); subsequently, they underwent a spontaneous self-decline stage, reverting to their initial state within 55 min (size from 55 ± 9 nm down to 18 ± 4 nm, Figure 3b,c). If we defined this polymer micellar expansion-to-contraction as one cycle (95 min), a new cycle could be restarted via ATP refueling after the system was reset (Figure 3d−f). The kinetics of micellar evolution can be further monitored by DLS. Strikingly, we found that the variation of micellar size plotted versus ATP incubation time is similar to a sine wave (Figure 3g, black curve). The micelles first grew until climbing up to a size extreme (Dh = 54 nm), followed by an autonomous size decay,

Figure 3. (a−f) Cryo-TEM images over time, visualizing the expansion and contraction of the polymer micelles in the first cycles (a−c) and the second cycle (d−f) of ATP-fueled micellar pulsation (scale bar is 200 nm). (g) DLS tracking the periodic micellar pulsation under various ATP levels adding to the system in the presence of potato apyrase with constant concentration (12 U/L): 200 μM ATP (blue curve), 100 μM ATP (black curve), and 50 μM ATP (red curve), respectively. ATP fuels are repetitively supplied to the system as the micellar system resets to the initial state. (h) DLS experiments showing the irregular, damped micellar pulsation under various enzyme levels in the presence of ATP with constant concentration (100 μM): 2 U/L apyrase (green curve) and 40 U/L (pink curve).

and finally to the beginning state (Dh = 19 nm). This process can be reinitiated by continuous ATP supply over five cycles when the micelles returned back to the minimum, as if they proceed by the reciprocating expansion−contraction. A static light-scattering (SLS) experiment corroborated that this expansion and contraction occurred in a single micelle (Table S1) but not intermicellar fusion or fission. In view of the rhythmic movement in each nanoparticle under energy influx, this periodic expansion and contraction is in many ways reminiscent of the mechanical “pulsating” feature. The two extreme sizes can be identified as the transient self-assembling structures far from equilibrium, and these polymeric nanoparticles can periodically run along the two nonequilibrium states. In an effort to assess the possibility to tune the periodicity and amplitude of micellar pulsation, we varied the ATP or 1153

DOI: 10.1021/acsmacrolett.7b00649 ACS Macro Lett. 2017, 6, 1151−1155

Letter

ACS Macro Letters enzyme concentrations to promote or inhibit the supramolecular complexation. Clearly, with the increase of ATP from 50 to 100 to 200 μM (enzyme level fixed at 12 U/L), micellar pulsating periodicity accelerated from 148 to 95 to 60 min, and their extreme sizes reached 33, 54, and 68 nm, respectively (blue, black, and red curve in Figure 3g), which means that the forward ATP/polymer association depends upon the level of ATP fuel. However, if we only changed the enzyme level (from 12 U/L to 2 U/L or 40 U/L) while fixing the ATP at 100 μM, the rate of backward dissociation was seriously disturbed, leading to a pulsation disorder (Figure 3h). These indicate that the enzyme level is the key to stabilizing the lifetime of micellar pulsation. In addition, we found that a normal pulsation began to attenuate after at most seven cycles, presumably due to the accumulation of chemical wastes (AMP and Pi) in this closed system (Figure S7). Insights to the dynamic system governed by the transient supramolecular cycle were verified by UV−vis and 31P NMR recording. The association/disassociation between ATP and ATP-receptor can be tracked by the change of characteristic UV absorption of ATP at 261 nm.24 The absorption gradually enhanced as micelles grew and subsequently weakened as micelles shrank inversely, which indicates reversible association of ATP/polymer complexes. 31 P NMR analysis further revealed this reversible process (Figure S9). ATP alone gave three phosphorus signals at δ = −9.5 (γ-P), −10.7 (α-P), and −21.4 (β-P) ppm. During the forward micellar expansion, the three signals shifted to downfield and were accompanied by spin−spin splitting at δ = −7.5 (d, 3Jβ,γ = 18.4 Hz), −8.6 (d, 3Jα,β = 10.5 Hz), and −19.3 (dd) ppm, indicating the ATP bound to the copolymer. Similar results can be found in other ATP-mediated responsive polyionic complex (PIC) micelles.25,26 While these micelles backward shrink, a new signal at 1.2 ppm assigned to Pi species appeared, which is indicative of the breakdown of the complexes. According to the above results, we proposed a possible energy dissipation mechanism based on self-adaptive supramolecular cycle to explain this micellar behavior: (i) Without ATP, the system is in thermal equilibrium, and there are numbers of positively charged ATP receptor units inside the micelles (Figure 4a, left panel). (ii) Upon addition of negatively charged ATP, they can bind with the micelles. The charge neutralization enlarges the micellar hydrophobic domains, which renders the system to increasingly deviate away from equilibrium. To compensate the increased free energy, a continuous micellar expansion will be initiated. The micelles do not stop growing until the ATP fuels are used up (Figure 4a, right panel). (iii) Since the competitive enzymatic reaction can reversibly destroy the supramolecular interactions, the micellar charge can recover for releasing the energy stored in the system, which results in an autonomous micellar contraction to the initial state. This dissipative cycle can be refueled by introduction of ATP energy input. To survey this mechanism, zeta potential offered key evidence on the change of micellar surface charge (Figure 4b).27 Identical to the micellar expansive and contractive states, the zeta potential (ζ) periodically changes from 73 ± 5 mV to 19 ± 3 mV (Figure 4b, blue curve), accompanied by the solution pH fluctuation from alkaline 9.1 ± 0.5 to near neutral 7.6 ± 0.2 (Figure 4b, red curve). The oscillation-like zeta potential and pH variation arise from the switchable electro-neutralization by the formation of a transient ATP/polymer complex, which confirms the cyclical switch of

Figure 4. (a) Schematic illustration of the mechanism of periodic micellar pulsation based on transient cycle of the formation of ATP/ polymer complex via specific ligand−receptor interactions. (b) The oscillation-like variation of systemic zeta potential (blue curve) and pH (red curve) as a function of micellar pulsation time. ATP (100 μM) refuels the system once the micellar surface potential climbs up to the wave peak.

the micellar system between high and low electropositivity. These results further support the above hydrophobicity-driven phase transition mechanism. Furthermore, the zeta potential/ pH changing periodicity is evaluated to be ∼92 min, concordant with that of micellar pulsation (95 min). Fluorescent study by using DPH-loaded micelles also displayed an oscillation-like intensity variation (Figure S10), indicating that the micellar core hydrophobicity changes periodically. This result accords with the potential results. Considering that the nanoparticles can do regular pulsation, they are promising to be used as smart drug carriers for periodic controlled release and long-lasting circulation modulated by ATP species. Anticancer medication, doxorubicin (Dox), was chosen as a drug model to perform release experiments under different conditions. The cumulative Dox release amount can be obtained by recording its fluorescence change at 417 nm. Clearly, when this system was only treated with apyrase (12 U/ L) the nanocarriers displayed a low-level free release less than 6.5% (Figure 5, blue dash line), indicating that the initial assemblies are compact to inhibit the leakage of internal payloads. In contrast, if only using ATP (100 μM) to trigger this system, the release quantity had a sudden increase to total release (∼95%) within 6 h and then remained a plateau (Figure 5, red dash line). This abrupt release is ascribed to that the highly expanded surface area of the micelles (from 4.1 × 103 to 3.7 × 104 nm2) enhances the membrane permeability, thus resulting in an accelerated Dox escape. Most interestingly, when the micelles were incubated with the enzyme and we applied a pulsing-type ATP stimulus at 150 min interval, the release profile presented a periodic active-dormant feature (Figure 5, black solid line). We deduced that this unexpected release mode is attributed to the periodic change of the micellar permeability caused by rhythmic micellar pulsation. In conclusion, we have developed a pulsatile micellar system that can dissipate chemical energy to maintain the periodic selfassembly behavior in the out-of-equilibrium state. A transient supramolecular cycle established between ATP and the 1154

DOI: 10.1021/acsmacrolett.7b00649 ACS Macro Lett. 2017, 6, 1151−1155

Letter

ACS Macro Letters

(3) Desai, A.; Mitchison, T. Annu. Rev. Cell Dev. Biol. 1997, 13, 83− 117. (4) Whitesides, G. M.; Grzybowski, B. Science 2002, 295, 2418−2421. (5) Mattia, E.; Otto, S. Nat. Nanotechnol. 2015, 10, 111−119. (6) Buddingh’, B. C.; van Hest, J. C. M. Acc. Chem. Res. 2017, 50, 769−777. (7) Yan, Q.; Zhao, Y. J. Am. Chem. Soc. 2013, 135, 16300−16303. (8) Huang, Z. G.; Kang, S.-K.; Banno, M.; Yamaguchi, T.; Lee, D.; Seok, C.; Lee, M. Science 2012, 337, 1521−1526. (9) Abdelmohsen, L. K. E.; Williams, D. S.; Pille, J.; Ozel, S. G.; Rikken, R. S. M.; Wilson, D. A.; van Hest, J. C. M. J. Am. Chem. Soc. 2016, 138, 9353−9356. (10) Jie, K.; Zhou, Y.; Yao, Y.; Shi, B.; Huang, F. J. Am. Chem. Soc. 2015, 137, 10472−10475. (11) Chi, X.; Yu, G.; Shao, L.; Chen, J.; Huang, F. J. Am. Chem. Soc. 2016, 138, 3168−3174. (12) Karsenti, E. Nat. Rev. Mol. Cell Biol. 2008, 9, 255−262. (13) Fialkowski, M.; Bishop, K. J. M.; Klajn, R.; Smoukov, S. K.; Campbell, C. J.; Grzybowski, B. A. J. Phys. Chem. B 2006, 110, 2482− 2496. (14) England, J. L. Nat. Nanotechnol. 2015, 10, 919−923. (15) Boekhoven, J.; Hendriksen, W. E.; Koper, G. J. M.; Eelkema, R.; van Esch, J. H. Science 2015, 349, 1075−1079. (16) Boekhoven, J.; Brizard, A. M.; Kowlgi, K. N. K.; Koper, G. J. M.; Eelkema, R.; van Esch, J. H. Angew. Chem., Int. Ed. 2010, 49, 4825− 4828. (17) Debnath, S.; Roy, S.; Ulijn, R. V. J. Am. Chem. Soc. 2013, 135, 16789−16792. (18) Shiraki, Y.; Yoshida, R. Angew. Chem., Int. Ed. 2012, 51, 6112− 6116. (19) Timonen, J. V. I; Latikka, M.; Leibler, L.; Ras, R. H. A.; Ikkala, O. Science 2013, 341, 253−257. (20) Che, H. L.; Buddingh’, B. C.; van Hest, J. C. M. Angew. Chem., Int. Ed. 2017, 56, 12581. (21) Yan, Q.; Zhao, Y. Chem. Sci. 2015, 6, 4343−4349. (22) Molnar, J.; Lorand, L. Arch. Biochem. Biophys. 1961, 93, 353− 363. (23) Makwana, P. K.; Jethva, P. N.; Roy, I. Analyst 2011, 136, 2161− 2167. (24) Bock, R. M.; Ling, N.-S.; et al. Arch. Biochem. Biophys. 1956, 62, 253−264. (25) Wang, C.; Chen, Q. S.; Wang, Z. Q.; Zhang, X. Angew. Chem., Int. Ed. 2010, 49, 8612−8615. (26) Naito, M.; Ishii, T.; Matsumoto, A.; Miyata, K.; Miyahara, Y.; Kataoka, K. Angew. Chem., Int. Ed. 2012, 51, 10751−10755. (27) Velegol, D.; Feick, J. D.; Collins, L. R. J. Colloid Interface Sci. 2000, 230, 114−121.

Figure 5. Controlled release of Dox from the polymer micelles under different conditions: only adding potato apyrase (12 U/L, blue dash line), only adding ATP (100 μM, red dash line), and adding apyrase (12 U/L) and a pulsing-type ATP stimulus in every 150 min (100 μM, black solid line).

polymer micelle becomes a driving force to achieve the dynamic evolution of the micelles. ATP, as a continuous source of chemical fuel, is able to finely regulate the physical parameters of micellar pulsation including periodicity, amplitude, and lifetime. This ATP-fueled dissipative selfassembly pathway represents a new strategy to fulfill the goal of biological motion mimics. The dynamic assemblies have potential to be applies as an “on-demand” drug delivery system. We anticipate that this pulsating nanoparticle will exemplify new opportunities of building life-like assemblies through a man-made molecular system and offer new visions on timedependent dynamic self-assembly with periodic behaviors.



ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsmacrolett.7b00649. Supplementary data, the design, and synthesis of copolymer (PDF)



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Qiang Yan: 0000-0001-5523-2659 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work is supported by National Natural Science Foundation of China (21674022, 51703034) and Open Research Fund of State Key Laboratory of Polymer Physics and Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences.



REFERENCES

(1) Aon, M. A.; Cortassa, S. Dynamic Biological Organization; Springer-Verlag: Germany, 1997. (2) Fletcher, D. A.; Mullins, R. D. Nature 2010, 463, 485−492. 1155

DOI: 10.1021/acsmacrolett.7b00649 ACS Macro Lett. 2017, 6, 1151−1155