Soft Matter

PAPER

Publ

ishe

d on

22

Sept

embe

r 20

14. D

ownl

oade

d by

McM

aste

r U

nive

rsity

on

31/1

0/20

14 1

2:58

:15.

View Article OnlineView Journal | View Issue

Evidence of gues

aShanghai Key Laboratory of Regulatory Bio

Normal University, Shanghai, 200241, P.R.bShanghai Key Laboratory of Magnetic Re

Shanghai, 200241, P.R. China

† Electronic supplementary informa10.1039/c4sm01381f

Cite this: Soft Matter, 2014, 10, 9153

Received 25th June 2014Accepted 22nd September 2014

DOI: 10.1039/c4sm01381f

www.rsc.org/softmatter

This journal is © The Royal Society of C

t encapsulation within G8 and G10dendrimers using NMR techniques†

Naimin Shao,a Tianjiao Dai,a Yan Liu,a Lei Lia and Yiyun Cheng*ab

Encapsulation of guest molecules within the interior cavities of dendrimers is promising, but high

generation dendrimers show limited encapsulation capacity due to their dense surface shell. Here, for

the first time, we prove that high generation polyamidoamine dendrimers, such as generation 8 and

generation 10, are able to encapsulate hydrophobic guests using NMR spectroscopy. Guest molecules

such as phenylbutazone, dexamethasone sodium phosphate and 9-anthracenecarboxylic acid with

molecular weights up to 516 Da are in close proximity to the interior scaffold protons of high generation

dendrimers. This encapsulation behavior depends on guest hydrophobicity. Chemical defects and back-

folding of terminal groups make it possible for these guest molecules to penetrate through the dense

surface shell of high generation dendrimers. These results provide new insights into the host–guest

chemistry of dendrimers.

1. Introduction

Dendrimers are hyperbranched macromolecules with well-dened nanostructures and globular shapes.1–4 These polymersare synthesized in a step-wise manner around a central core.1

During synthesis, each successive reaction step leads to anadditional generation of branching and the number of repeatedcycles is dened as dendrimer generation (denoted as G).Generally, dendrimers with G < 4, 4 # G < 7 and G $ 7 aretermed low, medium and high generation dendrimers, respec-tively. High generation dendrimers are difficult to obtainbecause the synthesis of these large-size polymers requiresnumerous steps.5,6 Considering the serious spatial hindranceon the dendrimer surface, high generation dendrimers usuallyhave defects in their structures.7 In spite of these describeddifficulties, several types of high generation dendrimers (e.g.polyamidoamine (PAMAM) dendrimers up to G10,6 phosphorusdendrimers up to G12,8 polyphenylene dendrimers up to G9(ref. 9) and triazine dendrimers up to G13 (ref. 10)) wereobtained with extremely low polydispersity over the past twodecades. To accelerate the synthesis of high generation den-drimers, researchers have developed several innovativesynthetic methods for simplifying the synthesis of den-drimers.11,12 A G6 dendrimer can now be synthesized within asingle day using click chemistry.13 An alternative approach isusing polymers as repeated units in dendrimer synthesis. This

logy, School of Life Sciences, East China

China. E-mail: yycheng@mail.ustc.edu.cn

sonance, East China Normal University,

tion (ESI) available. See DOI:

hemistry 2014

strategy generates giant dendrimers in a few steps.14 With theaid of these techniques, more and more high generation orlarge size dendrimers will become available in the near future.

High generation dendrimers have unique properties such asrelatively large size and high density of surface functionality.They can be directly observed by atomic force microscopy(AFM)15 and transmission electron microscopy (TEM)16 andeasily characterized by other techniques such as matrix-assistedlaser desorption ionization time-of-ight (MALDI-TOF) massspectrometry7,9 and small-angle X-ray scattering (SAXS).17 Highgeneration dendrimers also bear fruit in terms of applications.For example, dendrimers with sizes above 8 nm can be used asscaffolds for lymphatic or liver targeting.18,19 When these poly-mers are labeled with gadolinium or technetium ions, they canbe used for lymphatic or liver diagnosis by magnetic resonanceimaging (MRI) or single photon emission computed tomog-raphy (SPECT);18,20 high generation dendrimers are more effec-tive in hole formation on cell membranes compared to low andmedium generation dendrimers and this property is applicablein drug and gene delivery;21,22 an acetylated G9 PAMAM den-drimer can be used as a template to synthesize platinumnanoparticles and the resulting material can mimic the size,shape and biological activity of catalase (11 nm);23 a G13 triazinedendrimer with a molecular size of 30 nm can mimic smallviruses.10

Dendrimer-based host–guest systems are of central impor-tance to the applications of dendrimers.24 Dendrimer genera-tion plays a key role in the host–guest interactions. Lowgeneration dendrimers have an open surface and exiblestructure, while medium and high generation dendrimerspossess interior cavities and a more congested surface.25 Thedrug loading capacity of dendrimers generally increases in

Soft Matter, 2014, 10, 9153–9158 | 9153

Soft Matter Paper

Publ

ishe

d on

22

Sept

embe

r 20

14. D

ownl

oade

d by

McM

aste

r U

nive

rsity

on

31/1

0/20

14 1

2:58

:15.

View Article Online

proportion with dendrimer generation for low and mediumgeneration dendrimers, and there is an inversion in guestcapacity for high generation ones.26 The dense surface shell ofhigh generation dendrimers limits the encapsulation of guestmolecules into their interior cavities. Therefore, mediumgeneration dendrimers are usually considered as the idealcandidates for host–guest applications. Until now, few studiesconcerning the host behaviors of high generation dendrimershave been reported.26 Though Simanek et al. recently found thathigh generation triazine dendrimers (G > 7) maintain drugloading ability,26 they do not give evidence whether these highgeneration dendrimers form inclusion structures with guestmolecules (guests might be bound on the dendrimer surface,encapsulated within the dendrimer interior or within den-drimer aggregates/micelles).

Here, we characterize the inclusion complexes of G8 and G10PAMAM dendrimers with guest molecules such as phenylbuta-zone, dexamethasone sodium phosphate, 9-anthracenecarboxylicacid and 1-pyrenecarboxylic acid using nuclear Overhausereffect (NOE) spectroscopy. We prove that high generationPAMAM dendrimers such as G8 and G10 are capable of hostinga list of hydrophobic compounds.

2. Materials and methods2.1 Materials

Ethylenediamine (EDA)-cored and amine-terminated G8 andG10 PAMAM dendrimers were purchased from Dendritech, Inc.(Midland, MI). Deuterium oxide, phenylbutazone, 9-anthrace-necarboxylic acid, 1-pyrenecarboxylic acid and 1-naphthoic acidwere purchased from Sigma-Aldrich (St. Louis, MO). Dexa-methasone sodium phosphate was purchased from Santa CruzBiotechnology, Inc. (Santa Cruz, CA). G8 and G10 PAMAMdendrimers are stored in methanol solution and the solvent wasdistilled before use. These polymers were characterized by 13CNMR and polyacrylamide gel electrophoresis. Other chemicalswere used as received without further purication.

2.2 1H NMR experiments1H NMR spectra were acquired at 298.2 � 0.1 K on a Varian699.804 MHz NMR spectrometer, equipped with a 5 mm stan-dard probe. The 1H NMR spectra were obtained with 32 scansand a 2 s relaxation delay. The samples were maintained for atleast 10 min to avoid the uctuation of temperature before eachacquisition. A water suppression pulse was added to improvesignal sensitivity.

2.3 NOE experiments

The concentration of G8 and G10 PAMAM dendrimers in eachsample is xed at 2 mg mL�1. The molar ratio of guests todendrimers is 256 : 1 and 1024 : 1 for G8 and G10 PAMAMdendrimers, respectively. NOE experiments were conductedusing standard pulse sequences. A water suppression pulse wasadded to improve signal sensitivity. 1 s relaxation delay, 146.63ms acquisition time, a 6.5 ms 90� pulse width, and 32 transientswere averaged for 512 � 1024 complex points. To avoid indirect

9154 | Soft Matter, 2014, 10, 9153–9158

magnetization transfer caused by a spin diffusion effect, weconducted the NOE experiments for dendrimer/guestcomplexes at different mixing times (50 ms, 80 ms, 100 ms, 150ms and 300 ms). All the data were processed with NMRpipesoware on a Linux workstation.

3. Results and discussion

NOE spectroscopy is capable of revealing spatial relationshipsamong atoms in a molecule or in a complex of molecules.27 It iswidely used to prove host–guest interactions in inclusioncomplexes.24 Generally, if two atoms in a complex are in closeproximity to each other (within 5 A), the cross-peak betweenthem appears in the NOE spectroscopy. The intensity of a cross-peak depends on several parameters such as atom distance, thenumber of equivalent atoms andmixing time.24 Mixing time is akey parameter in an NOE experiment. For macromolecules,cross-peaks in the NOE spectrum at a relatively short mixingtime are generated by primary NOEs between spatially closeatoms. But at mixing times beyond a threshold value, spindiffusion in macromolecules causes secondary NOEs and cross-peaks between atoms that are not spatially close can also beobserved in the spectrum.28 We previously proved that 300 ms isbelow this threshold value for a G5 PAMAM dendrimer with amolecular weight of 28 826 Da.29 Considering G8 and G10PAMAM dendrimers in this study, whose ideal molecularweights (233 383 and 934 720 Da, respectively) are much largerthan the G5 dendrimer, it is necessary to determine an appro-priate mixing time for these high generation dendrimers beforeNOE experiments.

Phenylbutazone with a molecular weight of 308 Da is used asa guest. NOE experiments for the G10 dendrimer/phenylbuta-zone complex are conducted at 50 ms, 80 ms, 100 ms, 150 msand 300 ms, respectively and the intensities of cross-peaksbetween the G10 dendrimer and phenylbutazone in the spectraare measured by NMRpipe soware (Lorentz function). Forprimary NOEs, the cross-peak intensity increases in linearproportion with mixing time. However, this intensity maydecrease at long mixing times due to the presence of secondaryNOEs.28,29 As shown in Fig. 1, intensities of cross-peaks betweenthe G10 dendrimer (Ha,c,c0) and phenylbutazone (H2,3) increasein proportion with mixing time within the range of 50–150 msand further decrease at a mixing time of 300 ms, indicating thatspin diffusion also contributes to the cross-peaks in the NOEspectrum at 300 ms. The G8 PAMAM dendrimer has a smallersize than the G10 dendrimer and its threshold value for mixingtime should be higher than 150 ms. To ensure that the cross-peaks in the NOE spectrum are generated by primary NOEs, wechose the mixing times of 100 ms and 150 ms to conduct NOEexperiments for G10 and G8 PAMAM dendrimers, respectively.

Phenylbutazone molecules may interact with the cationicdendrimer surface via ionic interactions and with the den-drimer interior via hydrophobic interactions. The downeldshis of peaks (Hb0 and Hd0, located on the outermost layer ofthe dendrimer) in the presence of phenylbutazone (Fig. S1 inthe ESI†) prove that ionic interactions occur between thecationic dendrimers and negatively charged phenylbutazone

This journal is © The Royal Society of Chemistry 2014

Fig. 1 Chemical structures of G8 and G10 PAMAM dendrimers (a) andphenylbutazone (b) with proton labeling. (c) The intensity of NOEcross-peaks between phenylbutazone (H2,3) and the G10 PAMAMdendrimer (Ha,c,c0) at different mixing times.

Fig. 2 1H–1H NOE spectra of G10 PAMAM/phenylbutazone (a) and G8PAMAM/phenylbutazone (b) complexes at a mixing time of 100 ms.The blue arrows indicate NOE cross-peaks between the dendrimerand phenylbutazone.

Paper Soft Matter

Publ

ishe

d on

22

Sept

embe

r 20

14. D

ownl

oade

d by

McM

aste

r U

nive

rsity

on

31/1

0/20

14 1

2:58

:15.

View Article Online

molecules.30,31 For the inclusion structure of the dendrimer/phenylbutazone complex, as shown in Fig. 2a, NOE cross-peaksbetween the scaffold protons (Ha–d) of the G10 PAMAM den-drimer and the protons (H2,3,7,8) of phenylbutazone can beobserved. These protons are in close proximity with each otherin the G10 dendrimer/phenylbutazone complex. Since protonsHb0 and Hd0 on the G10 PAMAM dendrimer are closer to thedendrimer surface than protons Ha–d localized in the dendrimerinterior, the absence of cross-peaks between these protons (Hb0

and Hd0) and phenylbutazone in Fig. 2a indicates that the cross-peaks are attributed to phenylbutazone entrapped in the den-drimer interior rather than that bound in the outermost den-drimer shell. According to the cross-peaks in Fig. 2a, both thearomatic rings and the aliphatic chain of phenylbutazonemolecules localize in the G10 dendrimer interior, suggestingthat hydrophobic interactions play an important role in theencapsulation process. The localization of phenylbutazonemolecules within the dendrimer's interior cavity is in accor-dance with our previous results observed for medium genera-tion dendrimers such as G5 PAMAM.31 Similarly, strong NOEcross-peaks are observed between protons (Ha–d) of G8 den-drimers and phenylbutazone (Fig. 2b).

Besides phenylbutazone, high generation dendrimers, suchas G8 PAMAM, are able to encapsulate other guest moleculessuch as 9-anthracenecarboxylic acid and 1-pyrenecarboxylicacid. As shown in Fig. 3, obvious NOE cross-peaks are observedbetween the methylene protons of the G8 PAMAM dendrimer(Ha,c) and the aromatic protons of 9-anthracenecarboxylic acidand 1-pyrenecarboxylic acid (Fig. 3a and b). In contrast, no NOEcross-peaks are observed between the G8 dendrimer and 1-naphthoic acid (Fig. 3c). This result clearly demonstrates thatthe G8 PAMAM dendrimer successfully forms inclusion

This journal is © The Royal Society of Chemistry 2014

complexes with 9-anthracenecarboxylic acid and 1-pyrene-carboxylic acid but fails to encapsulate 1-naphthoic acid. Thearomatic rings of 9-anthracenecarboxylic acid and 1-pyrene-carboxylic acid are in close spatial proximity to the dendrimerinterior cavities. Generally, the host behavior of a dendrimerdepends on guest size and hydrophobicity.32 Hydrophobicguests with an appropriate size can be entrapped within thedendrimer interior. Here, 1-pyrenecarboxylic acid, 9-anthrace-necarboxylic acid and 1-naphthoic acid all belong to aromaticacids containing 2–4 aromatic rings and a carboxyl group.Considering that 1-pyrenecarboxylic acid (246 Da) and 9-anthracenecarboxylic acid (222 Da) are larger and more hydro-phobic than 1-naphthoic acid (172 Da), the distinct encapsula-tion behaviors of the guests are attributed to the morehydrophobic properties of 1-pyrenecarboxylic acid and 9-anthracenecarboxylic acid rather than the size effect. Thehydrophobic interactions between 1-pyrenecarboxylic acid andthe dendrimer interior scaffolds are further evidenced by 1HNMR titration experiments. As shown in Fig. S2 (ESI†), theaddition of 1-pyrenecarboxylic acid into G8 PAMAM dendrimersolution causes a signicant upeld shi of the interior methy-lene protons (Ha–d) of the dendrimer, suggesting that

Soft Matter, 2014, 10, 9153–9158 | 9155

Fig. 3 1H–1H NOE spectra of G8 PAMAM dendrimers with 9-anthra-cenecarboxylic acid (a), 1-pyrenecarboxylic acid (b) and 1-naphthoicacid (c) at a mixing time of 150 ms. The blue arrows indicate NOEcross-peaks between the dendrimer and 9-anthracenecarboxylicacid/1-pyrenecarboxylic acid. No cross-peak between the dendrimerand 1-naphthoic acid is observed.

Fig. 4 1H–1H NOE spectrum of the G8 PAMAM/dexamethasonesodium phosphate complex at a mixing time of 150 ms. The bluearrows indicate NOE cross-peaks between the dendrimer and dexa-methasone sodium phosphate. The black arrows indicate NOE inter-actions between dexamethasone sodium phosphate molecules.

Soft Matter Paper

Publ

ishe

d on

22

Sept

embe

r 20

14. D

ownl

oade

d by

McM

aste

r U

nive

rsity

on

31/1

0/20

14 1

2:58

:15.

View Article Online

hydrophobic interactions occur between 1-pyrenecarboxylicacid and the G8 PAMAM dendrimer.30

We also test whether the G8 PAMAM dendrimer can encap-sulate a guest larger than phenylbutazone. As shown in Fig. 4,dexamethasone sodium phosphate with a molecular weight of516 also shows strong NOE interactions with the G8 PAMAMdendrimer. The chemical shi of protons in dexamethasonesodium phosphate was assigned according to a previousstudy.33 The protons (H12,18,19) on dexamethasone sodiumphosphate are in close proximity to the interior protons (Ha–d)

9156 | Soft Matter, 2014, 10, 9153–9158

of the G8 PAMAM dendrimer. The dexamethasone sodiumphosphate molecules in the complex are in slow exchangebetween free- and bound-states (Fig. S3 in the ESI†). Theseresults clearly demonstrate that high generation dendrimers,such as G8 and G10 PAMAM, are able to encapsulate a list ofguest molecules with different sizes in their interior cavities.

Previous molecular simulation results on PAMAM andtriazine dendrimers found that low and medium generationdendrimers have a exible structure in aqueous solution, whilehigh generation ones possess a more rigid scaffold with a denseand crowding shell.25,26 The exible structure of low andmedium generation dendrimers allows water molecules topenetrate close to the dendrimer core, however, the crowdingshell and the congested interior of high generation ones preventwater penetration. As a result, the microenvironment in theinteriors of high generation dendrimers is very hydrophobic. Inthis study, we proved that several guests with much larger sizesthan water molecules penetrate into the interior cavities of G8and G10 PAMAM dendrimers. This encapsulation behavior isattributed to the following two reasons: (1) the defects on highgeneration dendrimers and (2) the back-folding of surfacegroups into dendrimer interiors. The presence of severe defectson high generation dendrimers is widely reported on PAMAMdendrimers.34 For example, the molecular weight of G10PAMAM (Dendritech) detected by MALDI-TOF is 579 kDa, whichis only 62% of its ideal molecular weight.7 The G8 and G10PAMAM dendrimers in this study are also obtained from Den-dritech. Though these dendrimers show high purity in 13C NMR(Fig. S4 in the ESI†) and polyacrylamide gel electrophoresis(Fig. S5 in the ESI†) analyses,35 the defects cannot be avoideddue to numerous steps in the synthesis and serious spatialhindrance on the surfaces of G8 and G10 PAMAM dendrimers.These chemical defects make the surface of dendrimers lesscrowded, providing opportunities for the penetration of guestmolecules into the interior of high generation dendrimers. Thepresence of back-folding of surface groups in the dendrimerstructure is reported for nearly all the dendrimer generations bymolecular simulations over the past decade.25,36–38 Guest

This journal is © The Royal Society of Chemistry 2014

Scheme 1 Proposed inclusion structure of a high generation den-drimer with phenylbutazone, 1-pyrenecarboxylic acid, dexametha-sone sodium phosphate and 9-anthracenecarboxylic acid.

Paper Soft Matter

Publ

ishe

d on

22

Sept

embe

r 20

14. D

ownl

oade

d by

McM

aste

r U

nive

rsity

on

31/1

0/20

14 1

2:58

:15.

View Article Online

molecules such as phenylbutazone, dexamethasone sodiumphosphate, 1-pyrenecarboxylic acid and 9-anthracenecarboxylicacid rst bind to the surface amine groups of G8 and G10dendrimers via ionic interactions and then these bound guestsmight be brought into the hydrophobic interiors of dendrimersduring the back-folding process. Hydrophobic interactions playan important role during the second step.

4. Conclusion

In summary, for the rst time we prove that high generationdendrimers, such as G8 and G10 PAMAM, are capable ofencapsulating guest molecules by NOE spectroscopy.9-Anthracenecarboxylic acid with a molecular weight of 222 Da,1-pyrenecarboxylic acid with a molecular weight of 246 Da,phenylbutazone with a molecular weight of 308 Da and dexa-methasone sodium phosphate with a molecular weight of 516Da localize in the dendrimer interior cavities as revealed by theNOE analysis (Scheme 1). Hydrophobic interactions betweenthese guest molecules and dendrimer interiors play an impor-tant role in the formation of inclusion structures. The chemicaldefects on high generation dendrimers and the back-folding ofsurface groups into dendrimer interiors are the reasons that

This journal is © The Royal Society of Chemistry 2014

explain why these polymers with dense surface shells canencapsulate guest molecules. These results provide newinsights into high generation dendrimer-based host–guestsystems.

Acknowledgements

This work was supported by the Specialized Research Fund forthe Doctoral Program of Higher Education (no.20120076110021), the Program for New Century ExcellentTalents in University of Ministry of Education of China (NCET-11-0138) and the National Natural Science Foundation of China(no. 21274044).

References

1 D. A. Tomalia, Prog. Polym. Sci., 2005, 30, 294–324.2 D. A. Tomalia, Nanomedicine, 2012, 7, 953–956.3 S. Svenson and D. A. Tomalia, Adv. Drug Delivery Rev., 2012,64, 102–115.

4 Y. Cheng, L. Zhao and T. Li, SoMatter, 2014, 10, 2714–2727.5 J. M. J. Frechet and D. A. Tomalia, Dendrimers and OtherDendritic Polymers, Wiley, Chichester, U.K., 2001.

6 D. A. Tomalia, A. M. Naylor and W. A. Goddard III, Angew.Chem., Int. Ed., 1990, 29, 138–175.

7 R. Muller, C. Laschober, W. W. Szymanski and G. Allmaier,Macromolecules, 2007, 40, 5599–5605.

8 M. L. Lartigue, B. Donnadieu, C. Galliot, A. M. Caminade,J. P. Majoral and J.-P. Fayet, Macromolecules, 1997, 30,7335–7337.

9 H. J. Rader, T. T. T. Nguyen and K. Mullen, Macromolecules,2014, 47, 1240–1248.

10 J. Lim, M. Kostiainen, J. Maly, V. C. da Costa, O. Annunziata,G. M. Pavan and E. E. Simanek, J. Am. Chem. Soc., 2013, 135,4660–4663.

11 M. V. Walter and M. Malkoch, Chem. Soc. Rev., 2012, 41,4593–4609.

12 X. Ma, J. Tang, Y. Shen, M. Fan, H. Tang and M. Radosz, J.Am. Chem. Soc., 2009, 131, 14795–14803.

13 P. Antoni, M. J. Robb, L. Campos, M. Montanez, A. Hult,E. Malmstrom, M. Malkoch and C. J. Hawker,Macromolecules, 2010, 43, 6625–6631.

14 A. Hirao, K. Sugiyama, Y. Tsunoda, A. Matsuo andT. Watanabe, J. Polym. Sci., Part A: Polym. Chem., 2006, 44,6659–6687.

15 T. Muller, D. G. Yablon, R. Karchner, D. Knapp,M. H. Kleinman, H. Fang, C. J. Durning, D. A. Tomalia,N. J. Turro and G. W. Flynn, Langmuir, 2002, 18, 7452–7455.

16 J. F. Kukowska-Latallo, A. U. Bielinska, J. Johnson,R. Spindler, D. A. Tomalia and J. R. Baker, Proc. Natl. Acad.Sci. U. S. A., 1996, 93, 4897–4902.

17 Y. Liu, C. Y. Chen, H. L. Chen, K. Hong, C. Y. Shew, X. Li,L. Liu, Y. B. Melnichenko, G. S. Smith and K. W. Herwig, J.Phys. Chem. Lett., 2010, 1, 2020–2024.

18 L. M. Kaminskas and C. J. Porter, Adv. Drug Delivery Rev.,2011, 63, 890–900.

Soft Matter, 2014, 10, 9153–9158 | 9157

Soft Matter Paper

Publ

ishe

d on

22

Sept

embe

r 20

14. D

ownl

oade

d by

McM

aste

r U

nive

rsity

on

31/1

0/20

14 1

2:58

:15.

View Article Online

19 H. Kobayashi and M. W. Brechbiel, Adv. Drug Delivery Rev.,2005, 57, 2271–2286.

20 M. C. Parrott, S. R. Benhabbour, C. Saab, J. A. Lemon,S. Parker, J. F. Valliant and A. Adronov, J. Am. Chem. Soc.,2009, 131, 2906–2916.

21 S. Hong, A. U. Bielinska, A. Mecke, B. Keszler, J. L. Beals,X. Shi, L. Balogh, B. G. Orr, J. R. Baker andM. M. Banaszak Holl, Bioconjugate Chem., 2004, 15, 774–782.

22 L. Xie and Y. Ma, So Matter, 2013, 9, 9319–9325.23 X. Wang, Y. Zhang, T. Li, W. Tian, Q. Zhang and Y. Cheng,

Langmuir, 2013, 29, 5262–5270.24 J. Hu, T. Xu and Y. Cheng, Chem. Rev., 2012, 112, 3856–3891.25 P. K. Maiti, T. Cagin, G. Wang and W. A. Goddard III,

Macromolecules, 2004, 37, 6236–6254.26 J. Lim, G. M. Pavan, O. Annunziata and E. E. Simanek, J. Am.

Chem. Soc., 2012, 134, 1942–1945.27 H. Wang, N. Shao, S. Qiao and Y. Cheng, J. Phys. Chem. B,

2012, 116, 11217–11224.28 J. K. Tzeng and S. S. Hou, Macromolecules, 2008, 41, 1281–

1288.

9158 | Soft Matter, 2014, 10, 9153–9158

29 Y. Cheng, Y. Li, Q. Wu and T. Xu, J. Phys. Chem. B, 2008, 112,12674–12680.

30 J. Hu, Y. Cheng, Y. Ma, Q. Wu and T. Xu, J. Phys. Chem. B,2008, 113, 64–74.

31 W. Yang, Y. Li, Y. Cheng, Q. Wu, L. Wen and T. Xu, J. Pharm.Sci., 2009, 98, 1075–1085.

32 J. Hu, Y. Cheng, Q. Wu, L. Zhao and T. Xu, J. Phys. Chem. B,2009, 113, 10650–10659.

33 K. Yang, L. Weng, Y. Cheng, H. Zhang, J. Zhang, Q. Wu andT. Xu, J. Phys. Chem. B, 2011, 115, 2185–2195.

34 A. M. Caminade, R. Laurent and J. P. Majoral, Adv. DrugDelivery Rev., 2005, 57, 2130–2146.

35 T. Xiao, X. Cao, S. Wang and X. Shi, Anal. Methods, 2011, 3,2348–2353.

36 P. K. Maiti andW. A. Goddard III, J. Phys. Chem. B, 2006, 110,25628–25632.

37 K. X. Moreno and E. E. Simanek, Macromolecules, 2008, 41,4108–4114.

38 P. K. Maiti, Y. Li, T. Cagin and W. A. Goddard III, J. Chem.Phys., 2009, 130, 144902.

This journal is © The Royal Society of Chemistry 2014