Polymersomes as a potential tool in nuclear medicine

Polymersomes as a

potential tool in

nuclear medicine

Guanglin WANG

ISBN 978-94-6203-609-3

P

oly

m

er

so

m

es

as

a p

ote

nti

al

to

ol

in

n

ucl

ea

r m

ed

ici

ne

Gu

an

gli

n

WA

N

G

Polymersomes as a

potential tool in nuclear

medicine

Guanglin WANG

Delft University of Technology

Faculty of Applied Sciences

Department of Radiation, Science and Technology

Radiation and Isotopes for Health

Polymersomes as a potential tool in

nuclear medicine

Proefschrift

ter verkrijging van de graad van doctor

aan de Technische Universiteit Delft,

op gezag van de Rector Magnificus prof. ir. K. A. C. M. Luyben.

voorzitter van het College voor Promoties,

in het openbaar te verdedigen op woensdag 2 juli 2014 om 15:00 uur

door

Guanglin WANG

Master of Engineering Techniques and Applications of Synchrotron Radiation,

University of Science and Technology of China, China.

Prof. dr. ir. H. Th. Wolterbeek Copromotors:

Dr. Ir. A. Denkova Dr. hab. E. Mendes

Samenstelling promotiecommissie: Rector Magnificus, voorzitter

Prof. dr. ir. H. Th. Wolterbeek, Technische Universiteit Delft, promotor Dr. Ir. A. Denkova, Technische Universiteit Delft, copromotor

Dr. hab. E. Mendes, Technische Universiteit Delft, copromotor Prof. dr. ir. M. De Jong, Erasmus Universiteit Rotterdam Prof. dr. Ir. E. J. R. Sudholter, Technische Universiteit Delft Prof. dr. R. J. M. Konings, Technische Universiteit Delft Prof. dr. O. C. Boerman, Radboud Universiteit Nijmegen Prof. dr. P. Dorenbos, Technische Universiteit Delft, reservelid

ISBN 978-94-6203-609-3

Copyright © 2014 by Guanglin Wang

All right reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic, or mechanical, including photocopying, recording or by any information storage and retrieval system without written permission of the author. Printed in the Netherlands.

Dedicated to Hui and Yanxi

The research described in this thesis was part of the 3BINDING project, co-funded by the Dutch Ministry of Economic Affairs and the Province of Zuid-Holland, contract number 1-5354. The Chinese Scientific Council is also acknowledged for financial support.

Table of Contents

1 Introduction………...………..……….……….…..1

1.1 Alpha radionuclide therapy………...…..………….…....1

1.2 Polymer vesicles………..…………...….…………..…...4

1.3 Scope and outline of the thesis………....………..….6

1.4 References………...….…..……...…7

2. Polymersomes as radionuclide carriers loaded via active ion transport through the hydrophobic bilayer…..………..11

Abstract……….……….……….…………...11

2.1 Introduction………..…………..……..…..12

2.2 Materials & Methods……….………..……..…13

2.3 Results and discussion………..………..…..…..16

2.4 Conclusions………..……….………....…..29

2.5 References………..………..…..…..29

2.6 Appendix...32

3. Polymersomes under gamma radiation: stability, permeability and morphology changes...………..….35

Abstract…….…….……….………..………....…....35

3.1 Introduction………..……..36

3.2 Materials & Methods………..………...37

3.3 Results and discussion………..………..……..…..40

3.4 Conclusions………..……….………...…..…….56

3.5 References………..………..…..…………..………..……..57

3.6 Appendix...59

4. Pharmacokinetics of polymersomes composed of poly(butadiene-ethylene oxide); healthy vs. tumor bearing mice ………….…….………....63

Abstract………...………...………...….……....63

4.1 Introduction………..………..…..……..64

4.2 Materials & Methods………...………..………65

4.3 Results and discussion………...………….…..……….….68

4.4 Conclusions……….………….………..………….78

4.5 References……….………..………….79 5. Retention studies of recoiling daughter nuclides of 225_{Ac in polymer}

vesicles……….……….………..….81

Abstract…..………..……….………..….…….…...….81

5.1 Introduction………...………..……....……82

5.2 Materials & Methods………...………..………84

5.3 Results and discussion………..………..…..……..91

5.4 Conclusions………….………..………..…..………102 5.5 References………...………...……103 5.6 Appendix...105 6. Conclusion….……….…...………..…..………107 Summary...………..……..………...……….….111 Samenvatting………..………...……….………...…113 Acknowledgements………..……...……….………...…115 Curriculum Vitae...117 Publications...118

Chapter 1

Introduction

1.1 Alpha radionuclide therapy

In 2012 there were 14.1 million new cancer cases worldwide of which 8.2 million cancer-related deaths and 32.6 million people still alive within 5 years of diagnosis. It is likely that by 2025, these figures could increase to an alarming 19.3 million new

cases due to growth and aging of the global population.1 The fight with cancer,

therefore, becomes more and more urgent with each passing year requiring both improved diagnostics and therapy. Although surgery is considered to be one of the most effective way to treat cancer patients, in cases where the disease metastasizes radionuclide therapy is, next to chemotherapy, one of the only proven alternatives. Radionuclides are, therefore, indispensable in both diagnostics as in therapy (i.e. radionuclide therapy).

Radionuclides, depending on their decay characteristics, can emit different particles, such as gamma rays, Auger electrons, beta or alpha particles. Gamma rays and beta particles have both been used for diagnostics and therapy for a long time unlike alpha particles which importance in cancer treatment has only recently been recognized. Alpha particles are positively charged mono-energetic helium nuclei with energy range from approximately 2 to 12 MeV. They have high linear energy transfer (LET) (around 200 keV/µm) and short penetration range (40 to 100 µm) in biological tissue, which is equal to several cell diameters (5 to 10) depending on the energy. These characteristics of alpha particles make them ideal for the treatment of metastases, provided that they accumulate at the tumor site their short penetration range will spare healthy tissue while destroying the malignant site. In addition, high LET radiation causes more direct damage to DNA and more double-strand breaks in comparison to low LET emitters. In general, 3 alpha particle tracks hitting the DNA molecule are sufficient to cause cell death, while several hundred low-LET particles will be needed for the same effect, implying that lower activities can be used in alpha- vs beta-

There are more than 100 radionuclides that can emit alpha particles. However, most of them are either too short or too long-lived or they are difficult to be produced to be of interest in therapeutic applications. At the moment there are just a few alpha

radionuclides having potential applications as therapeutic agents (Table 1).5

Table 1 Alpha radionuclides with potential application in alpha radionuclide therapy.5

Radionuclide Daughters Half-life Emission Energy Production

225_Ac 221_Fr, 217_At, 213_Bi, 213_Po, 209_Tl, 209_Pb 10 d 5α, 3β - 5.8 MeV 233_{U natural decay} Cyclotron 211_At 211_Po, 207_{Bi 7.2 h 2α, 2EC} 5.9 MeV Cyclotron 212_Bi 212_Po, 208_{Tl 60.6 m 2α, 3β}- 6.05 MeV 228_{Th natural decay} 224_{Ra generator} 213_Bi 212_Bi, 209_Tl, 209_{Pb 45.6 m 2α, 3β}- 5.8 MeV 225_{Ac generator} 212_Pb 212_Bi, 212_Po, 208_{Tl 10.6 h 2α, 3β}- 6.05 MeV 224_{Ra generator} 223_Ra 219_Rn, 215_Po, 211_Pb, 211_Bi, 211_Po, 207_Tl 11.4 d 5α, 3β - 5.7 MeV 227_{Ac generator} 227_Th 223_Ra, 219_Rn, 215_Po, 211_Pb, 211_Bi, 211_Po, 207_Tl 18.7 d 6α, 3β- 5.9 MeV 227_{Ac generator}

Currently, there are two main strategies to use alpha radionuclides for therapy. In the first strategy, the alpha radionuclide is either bound to a targeting agent or it is enclosed in a nano-carrier, in both cases the alpha radionuclide is targeted to the

diseased site.6 _{There are several kinds of targeting agents either based on antibodies}

(e.g. monoclonal antibodies, also referred to as ‘magic bullets’) or peptides. The application of monoclonal antibodies, chelated through a linker molecule to radionuclides, have originated in the 1970’s leading to the so-called immunotherapy, which is at the moment considered to be one of the most promising strategy for alpha radionuclide therapy. In beta radionuclide therapy two radiopharmaceuticals, i.e.,

Zevalin and Bexxar (using 90_{Y (t}

1/2=2.67 d) and 131I (t1/2=8.07 d)), both based on

monoclonal antibodies have already been approved by the FDA. On the other hand, nano-carriers have been much less investigated. The most promising pre-clinical studies have focused on the use of gold coated lanthanide phosphate nano-particles

in which 225Ac has been incorporated.7

In the second approach, the natural affinity of the radionuclide to accumulate at the

site of interest is used. One illustrative example is 223RaCl3, the first alpha radionuclide

radiopharmaceutical that has been approved by the FDA, which is applied for treating

castration resistant prostate cancer and bone metastases.8 223Ra in an analogue of

Calcium and it accumulates primarily in the bone, making it especially suitable for pain

relief caused by bone cancer.9

Therapeutic efficiency of targeted alpha radionuclide therapy has been proven in several preclinical studies including therapy of leukemia/lymphoma, ovarian carcinoma, melanoma, disseminated peritoneal tumor disease, brain tumor, breast

cancer metastases, pancreatic adenocarcinoma, and HIV infection.9-16 _{At the moment,}

there are also several clinical trial studies such as the treatment of myeloid leukemia

cells by anti-CD33 antibody conjugated with 213_Bi.17 _{Phase I study using} 211_At-(mu)

MX35 F(ab’) to treat recurrent ovarian carcinoma18 _and 213_{Bi-cDTPA-9.2.27 to treat}

metastatic melanoma has also been initiated.19 Besides the typical applications in

oncology, alpha radionuclide therapy seems to be also promising for the treatment of

HIV. In 2014 the first clinical trial using 213Bi-labeled monoclonal antibody 2556 is

expected to start.20

Although targeted alpha therapy has already been translated into the clinic, there are several challenges that still need to be faced. First, due to the short range, the alpha emitter should be situated in the vicinity of the cell in order to cause sufficient damage to the DNA. Second, almost all long lived alpha radionuclides are part of a decay chain which passes through several alpha emitters before reaching a stable isotope. The consequence of the alpha decay is that each daughter radionuclide receives energy in the order of 100 keV, which basically leads to the rupture of chemical bonds and subsequently to the distribution of the free radionuclides in the body. The daughter radionuclides, often being alpha emitters themselves, can therefore cause significant damage to healthy tissue. The traditional immunotherapy is, therefore, not a good approach for radionuclides having one or more daughters that are alpha emitters. A solution of the recoil problem is the encapsulation of the alpha emitter and its daughters into nano-carriers capable of retaining the daughter atoms. Different carriers have been investigated so far, such as liposomes, metal nanoparticles and

zeolites with varying success.7,21 _{Sofou et al. have first reported a theoretical and}

experimental combination study in which 225_{Ac has been encapsulated in liposomes,}

size, and showing an overall recoil retention of less than 10 %.21 Recently, McLaughlin et al. have investigated gold coated lanthanide phosphate nano-particles

containing 225Ac as vehicles in alpha radionuclide therapy. The results show 221Fr

retention of more than 88 % when thicker gold layers are deposited on the particles,

but the analysis method does not allow the determination of 213Bi retention in the

particle.7 _{This thesis will focus on polymer vesicles as carriers to prevent the escape of}

recoiling daughters.

1.2 Polymer vesicles (polymersomes)

Polymer vesicles, which are also called polymersomes, are made from amphiphilic block polymers and consist of a bilayered hydrophobic membrane enclosing an

aqueous cavity (Fig. 1).23,24 _{In solution, amphiphilic block copolymers can}

self-assemble into different morphologies depending on the hydrophilic and the hydrophobic block length. Generally, when the ratio (f) of the hydrophilic block to the total mass is between 25 % and 45 %, vesicles will be formed. Block copolymers with f > 45 % typically self-assemble into micelles, whereas at f < 25 % inverted

microstructures are expected to appear.25 _{The physicochemical properties relevant for}

medical applications can be easily fine-tuned by simple variation of the chemical nature and the molecular weight of the hydrophobic and hydrophilic blocks, facilitated

by the large variety of block copolymers that can be synthesized.26 _{Vesicles that are}

biodegradable, magnetically responsive, pH-, temperature, oxidation, and UV-

sensitive, can all be designed according to the desired application.27-30 _{Fig. 2 shows a}

schematic drawing expressing the different functionalities of polymersomes.31

Fig. 1 Schematic drawing of the self-assembly process of a polymersome from amphiphilic

Fig. 2 Schematic drawing of some of the many possible functionalities of polymersomes.31

There are two most commonly used ways to prepare polymersomes. In the first approach the block copolymer is dissolved in an organic solution that is a good solvent for both the lyophilic and the lyophobic block, followed by the addition of water that is a poor solvent for the hydrophobic block. Alternatively polymersomes can be prepared by direct dissolution of the block copolymers in aqueous solution, either by film rehydration or electro-formation. In general, the first method results in polymersomes having narrow size distribution, while the second method leads to vesicles having dimensions ranging from several tens of nanometers to several micrometers. In both cases, however, the size can be adjusted by sonication, freeze-thawing or extrusion through membrane filters resulting in samples with

reduced polydispersity.24

Polymersomes have been widely recognized as carriers in drug delivery due to their ability to enclose both hydrophobic and hydrophilic substances making them ideal for both diagnostic and therapeutic applications. Specifically for cancer treatment, the targeting of these structures to the tumor site, almost entirely relies on the so-called Enhanced Permeation and Retention (EPR) effect, although in some cases targeting

vectors such as antibodies have been added as well.32 _{The EPR effect occurs at a}

certain tumor size when more nutrients are needed to sustain growth and the body needs to make new blood vessels. The new blood vessels are often defective allowing the escape of molecules from the blood and their entrance in malignant

tissue. The reduced lymphatic drainage in turn leads to an accumulation of larger entities at the tumour. Christian et al. have shown the tumor shrinkage potential of two different polymersomes in mice, indirectly revealing the importance of the EPR

effect.33 _{Surprisingly, several therapeutic studies on polymersomes can be found,}

while investigations focusing on the bio-distribution and the pharmacokinetics of these nano-carriers are still scarce. Photos et al. and Brinkhuis et al. are one of the few

authors that have investigated the fate of polymersomes in vivo in healthy mice.34,35

The first group has used fluorescent dyes to obtain pharmacokinetic data while the second has applied diagnostic radionuclides in combination with pre-clinical SPECT (Single Photon Emission Computed Tomography), providing images with high spatial resolution. Brinkhuis et al. are also one of the only authors combining radionuclides with polymersomes, despite of the clear potential of these carriers in molecular

imaging using radioactive probes and radionuclide therapy.35 _{In contrast to liposomes,}

a commonly applied nano-carrier in health-related research, polymersomes have variable thickness of the membrane, better stability and reduced permeability, all factors that are of importance in molecular imaging and radionuclide therapy.

1.3 Scope and outline of the thesis

In this research, polymersomes based on the block copolymer poly(butadiene -ethylene oxide) (PB-PEO) have been chosen as carriers for both gamma and alpha emitting radionuclides due to the following reasons: the thickness of the membrane can be tuned, permeability of radionuclides is low, the stability of the polymersomes is expected to be high and the processing methods allow easy and reproducible preparation of vesicles having the desired dimensions. The ultimate goal of this research is the design of polymersomes for nuclear medical applications and in particular alpha radionuclide therapy. In order to achieve this aim, several research questions have to be answered: how can the vesicles be radiolabeled, what are their pharmacokinetics and the biodistribution in vivo, can the stability/loss of radionuclides be controlled and finally what is the recoil retention of recoiled daughters subsequent to an alpha decay as a function of the polymersomes’ characteristics. All of these questions will be addressed in the following chapters. An outline of the thesis is given below.

In chapter 2, a radiolabeling strategy has been developed based on the so-called active loading method which employs a lipophilic agent to carry the radionuclide into the aqueous core of the vesicles wherein a suitable chelator has been previously

enclosed. 111In, a common SPECT radionuclide, has been used. The influence of the

concentration of the lipophilic ligand and the chelator, the 111_{In activity and the}

thickness of the membrane have been investigated. The retention of 111In in the

polymersomes has also been studied.

In chapter 3, the polymersomes have been irradiated by 60_{Co gamma ray source in}

order to induce cross-links in the poly-butadiene, stabilizing in this way the vesicles. The stability of the polymersomes in water and THF was examined for samples that have received different radiation doses. In addition, the influence of cross-linking on both the loading and the retention of the radionuclides has been studied at different solvent compositions.

In chapter 4, the in vivo behavior of 111_{In labeled polymersomes has been}

investigated in mice. Two kinds of administration methods have been used, intravenous injection and subcutaneous injection. The biodistribution and pharmacokinetics of the polymersomes in healthy and tumor bearing mice have been studied and compared to each other.

In chapter 5, the polymer vesicles were radiolabeled with both 225_{Ac and} 213_{Bi and the}

retention of their daughters (221Fr, 213Bi and 209Pb) has been examined by measuring

the daughter atom remaining still inside the polymersomes. A small part of this chapter is dedicated to the uptake and distribution of polymersomes in HeLa cells. Finally, chapter 6 provides conclusions and highlights of the most significant results, including also recommendations for future research.

As the following chapters are based on published or submitted papers, partial information is sometime repeated throughout this thesis. This enables an easier reading of all chapters separately.

1.4 References

1. http://www.iarc.fr/en/media-centre/pr/2013/pdfs/pr223_E.pdf.

2. D. A. Mulford, D. A. Scheinberg, and J. G. Jurcic, J. Nucl. Med., 46, 1 (Suppl) 199s-204s, 2005.

3. W. Martin. Brechbiel, Dalton Trans., 43, 4918-4928, 2007.

Chatal, F. Davodeau, and M. Cherel, Eur. J. Nucl. Med. Mol. I., 32, 601-614, 2005. 5. Y-S. Kim. and M. W. Brechbiel, Tumor Biol., 33, 573-590, 2012.

6. M. R. Jackson, N. Falzone, and K. A. Vallis, Clin. Oncol., 25, 604-609, 2013. 7. M. F. McLaughlin, J. Woodwaard, A. B. Boll, J. S. Wall, A. J. Rondinone, S. J.

Kennel, S. Mirzadeh, and J. D. Robertson, PLOS ONE, 8, e54531, 2013.

8. http://www.fda.gov/newsevents/newsroom/pressannouncements/ucm352363.ht m?source=govdelivery

9. M. R. Harrison, T. Z. Wong, A. J. Armstrong, and D. J. George, Cancer Manag. Res., 5, 1-14, 2013.

10. T. L. Rosenblat, M. R. McDevitt, D. A. Mulford, N. Pandit-Taskar, C. R. Divgi, K. S. Panageas; M. L. Heaney, S. Chanel, A. Morgenstern, G. Sgouros, S. M. Larson, D. A. Scheinberg, and J. G. Jurcic, Clin. Cancer Res., 16, 5303-11, 2010.

11. S. Kneifel, D. Cordier, S. Good, M. C. S. Ionescu, A. Ghaffari, S. Hofer, M. Kretzchmar, M. Tolnay, C. Apostolidis, B. Waser, M. Arnold, J. Mueller-Brand, H. R. Maecke, J. C. Reubi, and A. Merlo, Clin. Cancer Res., 12, 3843–3850, 2006.

12. B. J. Allen, C. Raja, S. Rizvi, Y. Li, W. Tsui, P. Graham, F. Thompson, R. A. Reisfeld, J. Kearsley, A. Morgenstern, and C. Apostolidis, Cancer Biol. Ther., 4, 1318-1324, 2005.

13. A. Morgenstern, F. Bruchertseifer, and C. Apostolidis, Curr. Radiopharm., 4, 295-305, 2011.

14. B. Pfost, C. Seidl, M. Autenrieth, D. Saur, F. Bruchertseifer, A. Morgenstern, M. Schwaiger, and R. Senekowitsch-Schmidtke, J. Nucl. Med., 50, 1700-1708, 2009. 15. H. Song, R. F. Hobbs, R. Vajravelu, D. L. Huso, C. Esaias, C. Apostolidis, A.

Morgenstern, and G. Sgouros, Cancer Res., 69, 8941-8, 2009.

16. E. Dadachova, S. G. Kitchen, G. Bristol, G. C. Baldwin, E. Revskaya, C. Empig, G. B. Thornton, M. K. Gorny, S. Zolla-Pazner, and A. Casadevall, PLOS ONE, 7, e31866, 2012

17. J. G Jurcic, S. M. Larson, G. Sgouros, M. R. McDevitt, R. D. Finn, C. R. Dvigi, A. M. Ballangrud, K. A. Hamacher, D. Ma, J. L. Humm, M. W. Brechbiel, R. Molinet, and D. A. Scheinberg, Blood, 100, 1233-1239, 2002.

18. P. Albertsson, H. Andersson, T. Bäck, J. Elgqvist, F. Henriksson, L. Jacobsson, H. Jensen, S. Lindegren, S. Manuchopour, S. Palm, and R. Hultborn, 8th International symposium for targeted alpha therapy in Oak Ridge National Laboratory. 2013 19. B. J. Allen, A. A. Singla, S. M. Abbas Rizvi, P. Graham, F. Bruchertseifer, C.

Apostolidis, and A. Morgenstern, 7th _{International symposium for targeted alpha}

therapy in Berlin. 2011.

Ridge National Laboratory. 2013

21. S. Sofou, J. L. Thomas, L. Hung-yin, M. R. McDevitt, D. A. Scheinberg, and G. Sgouros, J. Nucl. Med., 45, 253-60, 2004.

22. A. Piotrowska, E. Leszczuk, F. Bruchertseifer, and A. Morgenstern, J. Nanopart. Res., 15, 2082, 2013.

23. D. E. Discher, and A. Eisenberg, Science, 297, 967-973, 2002.

24. C. LoPresti, H. Lomas, M. Massignani, T. Smart, and G. Battaglia, J. Mater. Chem., 19, 3576-3590, 2009.

25. D. E. Discher, and F. Ahemd, Annu. Rev. Biomed. Eng., 8, 323-341, 2006. 26. N. P. Kamat, J. S. Katz, and D. A. Hammer, J. Phys. Chem. Lett. 2, 1612-1623,

2011.

27. J. A. Opsteen, J. J. L. M. Cornelissen, and J. C. M. van Hest, Pure Appl. Chem., 76, 1309-1319, 2004.

28. G. Y. Liu, C. J. Chen, and J. Ji, Soft Matter, 8, 8811-8821, 2012.

29. M. Hamidi, M-A. Shahbazi, and K. Rostamizadeh, Macromol. Biosci., 12, 144-164, 2012.

30. P. Tanner, P. Baumann, R. Enea, O. Onaca, C. Palivan, and W. Meier, Accounts Chem. Res. 44, 1039-1049, 2011.

31. M. Massignani, H. Lomas, and G. Battaglia, Adv. Polym. Sci., 229, 115-154, 2010.

32. H. Maeda, J. Wu, T. Sawa, Y. Matsumura, and K. Hori, J. Control. Release, 65, 271-284, 2000.

33. D. A. Christian, O. B. Garbuzenko, T. Minko, and D. E. Discher, Macro. Rapid. Commu., 31, 135-141, 2010.

34. P. J. Photos, L. Bacakova, B. Discher, F. S. Bates, and D. E. Discher, J. Control. Release, 90, 323-334, 2003.

35. R. P. Brinkhuis, K. Stojanov, P. Laverman, J. Eilander, I. S. Zuhorn, F. P. J. t. Rutjes, and J. C. M. van Hest, Bioconju.Chem., 23, 958-965, 2012.

!

!

!

!

!

!

!

!

!

!

*_{A modified version of this chapter was published in Soft Matter, 2013, 9, 727-734. 11}

Chapter 2

Polymersomes as radionuclide carriers loaded

via active ion transport through the hydrophobic

bilayer

*

Abstract

Vesicles composed of amphiphilic block copolymers (i.e. polymersomes) have already been shown to have great potential in drug delivery. Nuclear imaging techniques such as Single Photon Emission Computed Tomography (SPECT) are indispensable in the correct evaluation of biodistribution and pharmacokinetics of newly or not fully investigated polymersome formulations. However, to date polymer vesicles, in contrast to their lipid counterparts, have not been loaded with radionuclides. In this chapter, we have investigated the so-called active loading method to trap radionuclides into preformed polymersomes composed of poly(butadiene-b-ethylene oxide) having variable membrane and brush thickness. We have used tropolone as a lipophilic agent to transport the radioactive isotope of

Indium, 111_{In, through the hydrophobic membrane into the aqueous cavity containing}

the strong hydrophilic chelate diethylenetriaminepentaacetic acid (DTPA). The results

show that high loading efficiency of 111_In3+ _{(> 85 %) can be achieved at short}

incubation times in polymersomes with membrane thicknesses twice the size of typical lipid bilayers. However, increasing the molecular weight of the block copolymers results in lower radiolabelling efficiency and a much slower loading rate. In addition, both the DTPA and tropolone concentration have been found to influence the loading efficiency. Finally, we not only demonstrate that a significant amount of this radioisotope can be successfully encapsulated in the polymersomes, but also that a negligible loss (< 5 % in 48 hours) is observed, allowing their safe application in future in vivo studies.

! !

2.1 Introduction

Polymer vesicles, also called polymersomes, are self-assembled structures composed of amphiphilic block copolymers that form a hydrophobic bilayer enclosing an aqueous cavity. During the past decade polymersomes have gained tremendous importance due to the possibility to control their physical, chemical and biological properties, creating in this way a great number of possible applications in various fields with medicine being one of the most important. Indeed, polymersomes have been proven to be especially attractive in drug delivery and diagnostics due to the possibility to encapsulate water-soluble compounds in the closed nano-compartments

as well as water-insoluble substances in the hydrophobic bilayer. 1-6 _{Due to the large}

variety of block copolymers that can potentially self-assemble into vesicles, physico-chemical properties relevant to medical applications can be easily fine-tuned by simple variations of the chemical nature and molecular weight of the hydrophobic and the hydrophilic segments. In this way, polymersome carriers can be engineered to express, for instance, high in vivo stability and long in vivo circulation time comparable to stealth liposomes. Besides these two fundamental nano-carrier properties, copolymer vesicles also allow for tunable membrane permeability and membrane hydrophobicity, an important advantage over their lipid counterparts,

which naturally expands the range of applications that can be envisaged.7

Furthermore, the hydrophilic block can be easily functionalized by the conjugation of different ligands, which in turn enable the coupling of antibodies, proteins, vitamins

etc.8-10 _{The possibility to attach targeting moieties and the fact that polymersomes can}

rely on the Enhanced Permeation and Retention (EPR) effect to stimulate tumor uptake, makes these nano-carriers especially promising in the fight against cancer. Some straightforward examples of the application of polymersomes in medicine

comprise the encapsulation of anti-cancer drugs such as the hydrophobic taxol6 _and

the hydrophilic doxorubicin, the incorporation of MRI contrast agents11,12 _and

near-infrared (NIR) dyes13 _{for diagnostics. More sophisticated illustrations include the}

formation of polymersomes in polymersomes (i.e. vesosomes) for

multi-compartmental loading and self-porating vesicles for controlled drug release.14,15

To date, however, studies that account for the use of polymersomes in nuclear medicine are strikingly low. Only a few studies have been carried out and in none of

them radionuclides have been encapsulated in the aqueous cavity.16

From a practical point of view, loading of radionuclides in the aqueous compartments of preformed polymersomes is very appealing since this would allow nuclear imaging techniques to be utilized in the evaluation of the nano-carriers’ biodistribution and

pharmacokinetics and based on this outcome possible future application in radionuclide therapy. A large number of studies on liposomes show that lipophilic ligands can be used to transport radionuclides through the lipid bilayer into the lumen

achieving in this way high loading efficiency.17, 18 _{However, such an active loading}

mechanism is expected to be much less effective in polymersomes, considering the much more rigid nature, the lower permeability and the higher viscosity of polymeric membranes when compared to lipid bilayers. At the same time, exactly due to these properties polymersomes could offer much better in vivo stability as well as lower loss of radionuclide when compared to liposomes, which makes them especially attractive for nuclear medical applications.

A schematic drawing of the so-called active loading process is depicted in Fig. 1. This figure reveals the transportation of an ion complexed with a lipophilic ligand through the membrane and the subsequent complexation of the ions with a stronger chelate on the inside of the polymersomes.

In this chapter, we focus on the active transport of ions through the hydrophobic membrane of polymersomes composed of poly(butadiene-b-ethylene oxide) (PB-PEO), having at least twice larger bilayer thickness than liposomes. To increase

the relevance of this work we have chosen the radionuclide 111In as an example, since

it is one of the most often applied radiotracers in nuclear imaging (i.e. in Single Photon

Emission Computed Tomography (SPECT)).19

Fig. 1 Schematic drawing of the active loading process.

2.2 Materials & Methods

Chemicals

Four poly(butadiene-b-ethylene oxide) block copolymers with blocks of different molecular weight purchased from Polymer Source (Quebec, Canada) were used in this study. All block copolymers were nearly monodisperse with ratio of the weight

average molecular weight to the number average molecular weight (Mw/Mn) of less

than 1.10. For simplicity reasons the block copolymers were abbreviated to BE. Table 1 shows the molecular weights of each block copolymer and the fraction of the hydrophobic block.

Table 1Characteristics of polymersome-forming block copolymers used in this study

The AG1X-8 (phosphate form) resin was prepared by washing the commercially received chloride form of the resin with sodium hydroxide, Milli-Q water and 1 N

phosphoric acid until no Cl- was found in the wash solution, as verified by the silver

nitrate test for Cl-_{. The anion-exchange resin was then preconditioned with 0.106 M}

sodium phosphate at pH 7.4.

The radioactive isotope of Indium (111_In3+_{) was kindly provided by the section Nuclear}

Medicine of the Erasmus Medical Center in the Netherlands and had a specific activity of 1.72 MBq/pmol.

All other chemicals were purchased from Sigma-Aldrich. Polymersomes preparation

The polymersome mixtures were prepared by dissolving the block copolymers in a solution containing different concentrations of DTPA (ranging from 0.1 to 5 mM) and 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) buffer. The mixture was then stirred for a week. The pH of the solution was 7.4. Subsequently, the size of the vesicles was adjusted by extruding them several times through polycarbonate filters with a cut-off membrane of 800, 400, and 200 nm. The remaining free DTPA was separated from the polymersomes using Sephadex G-25 column (DxL = 1x21 cm). The volume fraction containing the highest polymersome concentration was used for the loading.

Polymersomes loading with 111_In3+

After separation of the un-encapsulated DTPA, the nano-carriers were mixed with

111_In3+ _{by slowly adding 200-250 µL of freshly prepared “loading solution” to 0.8 mL}

Polymer formula Designated name Mn (kg/mol) f_BD M_w/M_n(PDI)

PB33-PEO21 BE1 2.7 0.73 1.05

PB120-PEO89 BE2 10.4 0.69 1.10

PB46-PEO23 BE3 3.5 0.77 1.05

polymersome solution. The “loading solution” was typically composed of 111In3+ and

200 µL of 10 mM HEPES at pH 7.4 already containing the dissolved tropolone.

Depending on the desired activity, 2-50 µL111InCl3 (at pH 2) was used in the loading

solution, corresponding to 111_In3+ _{concentration ranging from 0.09 to 21.5 nM. The}

tropolone concentration in the loading solution varied from 1 µM to 1.25 mM, which corresponds to respectively 2 and 250 µM in the final polymersome solution. The polymersome-loading solution mixture was incubated for different periods of time ranging from 5 minutes to 18 hours. The loading process was terminated by adding

0.5 g AG1X-8 (phosphate form) resin to the mixture to remove the 111_In3+_-tropolone

complex still remaining in the solution.

Subsequently, the polymersomes were separated from the resin by filtering the solution through a filter with pore size of 30 µm. The loading efficiency was

determined by measuring the ratio of 111_{In radioactivity in the solution before the}

addition of the resin and the filtrate. The breakthrough of the resin was determined by

adding free 111_{In-tropolone to it and measuring the activity in the eluate. The}

breakthrough was found to be less than 1% even after the resin was washed with 20 mL of HEPES or PBS at pH 7.4.

Alternatively after loading, the unencapsulated 111In-tropolone complexes were

separated from the polymersomes using size exclusion chromatography (a Sephadex G-25 column DxL = 1x30 cm). The loading efficiency was calculated by dividing the activity of the volume fractions containing polymersomes by the total activity before separation. The elution was portioned per mL and the activity in each fraction was determined. The size exclusion column was also used to obtain the separate elution

profiles of polymersomes, 111_{In-DTPA and} 111_{In-tropolone complexes (Fig. 4A). The}

elution profile of polymersomes in the absence of 111In was determined using Dynamic

Light Scattering (DLS). High-purity Germanium (HPGe) detector (Model: LG 22,

Princeton Gamma Tech) was used to measure the activity of 111In3+ and the gamma

energy peak at 245 keV was used to calculate the loading efficiency.

The loading experiments were carried out at room temperature T = 20 ± 1 °C. Determining loss of radiolabel

To evaluate the loss of radiolabel 100 µL of 10 mM DTPA solution was added to 1 ml

of 111_In3+ _{loaded BE1 polymersomes and left to equilibrate for a period of 48 hours at}

room temperature (T = 20 ± 1 °C). Subsequently, the solution was passed through a

Sephadex G-25 column (DxL = 1x21 cm) to separate the 111_{In-DTPA complexes from}

eluted fraction was determined in the same way as described above. The elution

profile of 111_{In-DTPA was measured separately to determine the elution volume of the}

complex. The elution profile of polymersomes in the absence of 111In was again

determined using Dynamic Light Scattering. Both elution profiles can be found in Fig. 4B.

Furthermore, the loss of radiolabel in serum was examined by incubating the polymersomes in serum at 37 °C for 24 hours. Subsequently chromatography on Sepharose 4B column (DxL=1x37 cm) was performed to separate the polymersomes from the serum. Elution was carried out using HEPES buffer.

In both experiments the normalized fraction (i.e. the fraction corresponding to the activity in a certain volume fraction divided by the total activity) was used to estimate the percentage of radiolabel loss.

DLS

The DLS apparatus used in this study consisted of a JDS Uniphase 633 nm 35 mW laser, an ALV sp 125 s/w 93 goniometer, a fibre detector and a Perkin Elmer photon counter. An ALV-5000/epp correlator and software completed the set-up. The DLS tubes were immersed in a temperature-regulated bath (at 20 °C) containing toluene as the index-matching fluid. The intensity autocorrelation function was determined at

90º. The data was fitted using the Contin method20 and the Stokes-Einstein equation

was used to determine the hydrodynamic radius of the polymersomes. Cryo-transmission electron microscopy (Cryo-EM)

2 µL of the polymersome solution containing 0.5 mg/ml block copolymer was

deposited on a holey carbon film (quantifoil 3.5/1) supported on a TEM grid. A filter paper was then used to blot the drop in order to obtain a thin layer on the grid. This sample was vitrified by rapidly immersing into liquid ethane (Vitrobot, FEI, Eindhoven, The Netherlands). The specimen was inserted into a cryo-transfer holder (Gatan

model 626) and then transferred to a Philips CM12 cryo-EM. Images were obtained at

an acceleration voltage of 120 keV and under low-dose conditions on a slow scan CCD camera.

2.3 Results & discussion

This work has focused on the loading of 111In3+ into polymersomes using a strategy

the vesicles and the metal ions are transported through the hydrophobic bilayer using a liphophilic agent. The hydrophilic chelate that has been used in our work is DTPA, while tropolone functions as the lipophilic transport molecule. We have investigated polymersomes composed of four different poly(butadiene-b-ethylene oxide) block copolymers with blocks of different molecular weights as described in detail the materials and methods section. Fig. 2 displays cryo-EM images of polymersomes composed of BE1 and BE2 block copolymers having membrane thickness of around 7 nm (Fig. 2A) and 13 nm (Fig. 2B) respectively. The brush thickness of the BE2 polymersomes has been estimated from the cryo-EM images to be approximately 6 nm (Fig. 2C). The membrane and the brush thickness are in close agreement to the theoretically expected values based on scaling laws found for this block copolymer (dmembrane~(MPB)0.5 and dcorona~ (MEO)1 ).21,22 The brush thickness of the BE1

polymersomes is then estimated to be around 2 nm. The average size of the polymersomes has been found to be between 100 and 200 nm in diameter for all examined block copolymers. Dynamic light scattering data reveals slightly larger diameter of 236 nm (Fig. 3), which is to be expected since the scattered light intensity of a particle is proportional to the sixth power of its size, which tends to shift the apparent particle dimension.

Fig. 2 Cryo-TEM images of A) BE1 and B) BE2 polymersomes as described in table 1. C)

Close up image of BE2 polymersomes. The arrows in C indicate the border of the PEO brush. The polymersomes have been extruded through 200 nm polycarbonate filters.

Fig. 3 Polymersome size distribution as obtained by DLS for a sample extruded through 200

nm polycarbonate filters.

The first set of experiments has been performed to determine the kinetics of radionuclide transportation through the bilayer and encapsulation in the aqueous core. The presented in this section results correspond to AG1X-8 resin separation of the

encapsulated in the polymersomes 111_{In and the free} 111_{In-tropolone. However, these}

findings have been further verified using size exclusion chromatography (Fig. 4A).

10 100 1000 0,0 0,2 0,4 0,6 0,8 1,0 1,2 Radius (nm)

Fig. 4 A) Elution profiles of different species corresponding to separation using Sephadex

G-25 size exclusion column with dimensions of DxL=1x30 cm. The empty BE1 polymersomes (□) elution profile has been determined by DLS and given in kHz and BE1 polymersomes containing 111In (■), pure 111In-DTPA (●) and pure 111In-tropolone (▲) have been measured with a HPGe detector. The normalized fraction corresponds to the ratio of radioactivity in each volume fraction and the total radioactivity before separation. N.B. 95 % of all 111In-tropolone does not come off the column. B) Elution profiles of 111In-DTPA and empty polymersomes corresponding to separation using Sephadex G-25 size exclusion column with dimensions of DxL=1x21 cm. The empty BE1 polymersomes (□) elution profile has been determined by DLS

and is given in kHz and pure 111In-DTPA (●) has been measured with a HPGe detector. The

normalized fraction corresponds to the ratio of radioactivity in each volume fraction and the total radioactivity before separation.

0 5 10 15 20 25 30 0,0 0,1 0,2 0,3 0,4 0,5 0,6 111_{In-polymersomes} 111_In-DTPA 111_In-tropolone polymersomes DLS No rmalize d ac tivity f raction Volume fraction (mL) A 0 100 200 300 400 500 Intens ity (kHz ) 0 5 10 15 20 25 30 0,00 0,05 0,10 0,15 0,20 0,25 0,30 0,35 111_In-DTPA Polymersomes DLS Volume fraction (mL) No rmallized activity fraction 0 100 200 300 400 500 Inten sity (kH z) B

Fig. 5 presents the obtained loading efficiency as a function of time for two types of polymersomes, BE1 and BE2. The results demonstrate that in the case of the BE1 polymersomes high loading efficiency that remains constant over time is achieved

within 10 minutes of incubation. On the other hand, the encapsulation process of 111_In

in BE2 polymersomes has been found to slowly progress with time, reaching a maximum entrapment efficiency of just 38 % at 18 hours. An explanation for these difference in loading rate and efficiency can be due to the much larger bilayer thickness and the bulkier PEO brush in the case of the BE2 vesicles. Both factors can hinder the transport of the Indium-tropolone complex to the aqueous cavity of the vesicles. The PEO chains are expected to create a steric barrier, which is more likely to be of importance in a lengthier polymer. The brush thickness scales approximately linearly with the molecular weight of the PEO block, which means that the corona layer in the case of the BE2 polymersomes is around three times larger than the one of the BE1 polymersomes. In addition, the fluidity of the membrane has been previously reported to decrease with increasing molecular weight of the hydrophobic

block24, which is expected to decrease the mobility of the In-tropolone complex. In fact,

the permeability of the membrane of polymersomes composed of

poly(butadiene-b-ethylene oxide) has been found to scale to the power -1 as function

of the thickness of the bilayer.25 _{This implies that the permeability of the BE2}

polymersomes is nearly 4 times lower than in the case for the BE1 vesicles. Similar slow entrapment rates have been reported for liposomes containing cholesterol. Cholesterol is known to increase the mechanical rigidity and lower the water permeability of lipid bilayers, inducing mechanical properties to liposome membranes that come closer to the ones of polymersomes including an improved in vivo

stability.26 However, radiolabelling of such rigid liposomes often needs to take place at

high temperatures in order to achieve sufficiently high loading efficiency, which is not

Fig. 5 Effect of the loading time on the loading efficiency in BE1 polymersomes (▲) and BE2

polymersomes (■). The block copolymer concentration is 0.5 mg/ml, the 111In3+ activity is 0.15 MBq (0.09 nM), the tropolone and DTPA concentrations are respectively 20 µM and 1mM. Loading has been carried out at 20 °C at pH 7.4. The lines are guides for the eyes.

In order to get more quantitative information on the influence of the membrane thickness and the effect of PEO length we have performed experiments with polymersomes composed of BE3 and BE4 block copolymers. The BE3 polymersomes have slightly thicker membrane (estimated to be around 8 nm from established

scaling laws d~(MPB)0.5) than the BE1 polymersomes and the same PEO molecular

weight. On the other hand, the polymersomes composed of BE4 have the same membrane thickness as the BE1 polymersomes but slightly smaller PEO molecular weight i.e. 600 vs. 900 g/mol, which corresponds to EO 14 vs EO 21 units and therefore a brush thickness of approximately 1.4 nm. The experiments using these block copolymers revealed no significant differences in loading rate and efficiency between BE1, BE3 and BE4 polymersomes (Fig.6) suggesting that a pronounced effect is only to be expected for block copolymers of much higher molecular weight. In

summary, we have attained high loading efficiency of 111_In3+ _{at short incubation times}

in polymersomes with membrane thicknesses twice the size of lipid bilayers (i.e. 3-4 nm) despite the higher viscosity and lower permeability of polymeric membranes in comparison to liposomes. 0 5 10 15 20 0 20 40 60 80 100 Loading effic iency (%) Time (h)

Fig. 6 Effect of the loading time on the loading efficiency of BE1 polymersomes (▲), BE3

polymersomes (●) and BE4 polymersomes (■). The concentration of polymersomes is 0.5 mg/ml, the 111In3+ activity is 0.15 MBq and the tropolone and DTPA concentrations are respectively 20 µM and 1 mM.

Furthermore, we have investigated the influence of trapped chelate, i.e. DTPA, and the lipophilic ligand concentrations on the loading efficiency. Inspection of Fig. 7

shows that only 10 % of the 111_{In can be encapsulated in the polymersomes in the}

absence of DTPA. At chelate concentrations of 0.5 mM the loading efficiency is nearly 90 % and it just slightly changes as the concentration is further increased. These

results strongly suggest that the 111_In3+ _{is entrapped in the aqueous cavity by forming}

a complex with DTPA since at low chelate concentration only small percentage of the

radionuclide is found in the polymersomes. In the absence of DTPA, 111_In3+ _{is most}

probably portioned between the aqueous cavity and the hydrophobic bilayer but still bound to the tropolone molecules.

0 10 20 30 40 50 60 70 80 90 100 Lo ad ing efficien cy (%) Time (h)

Fig. 7 Loading efficiency of BE1 polymersomes as a function of the DTPA concentration at

tropolone concentration of 20 µM, loaded from a 0.15 MBq 111In3+ (0.09 nM) solution at loading time of 1 hour. The line is a guide for the eye.

In Fig. 8 the radionuclide loading efficiency of BE1 and BE2 polymersomes is displayed as a function of the tropolone concentration. An important remark that needs to be mentioned here is the necessity to have an excess of tropolone to carry the Indium ions into the lumen. The results presented in Fig. 8 correspond to Indium concentrations of just 0.09 nM. Nevertheless, in the case of BE1 polymersome, high efficiency could only be attained when 20 µM tropolone is added to the vesicle solution. However, this does not mean that adding large excess of tropolone will necessary lead to high degree of entrapment. The loading efficiency of BE1 polymersomes that have been diluted 10 times after self-assembly exhibit a well-pronounced maximum at a certain tropolone concentration. The appearance of such an optimal concentration suggests that tropolone might be competing with DTPA for the complexation with Indium ions. In order to get a better insight on the loading process we have estimated the amount of DTPA enclosed in the polymersomes assuming that: the DTPA is evenly distributed in the solution, one DTPA binds to one

111_{In ion and the density of the bilayer membrane is the same as the one of}

poly(butadiene). The detailed calculations can be found in the supporting information. According to these calculations 0.6 nmol of DTPA is present in the polymersomes when 0.5 mg/ml of the BE1 block copolymer is used. At 200 µM (200 nmol) tropolone concentration there is approximately 110 times more free tropolone molecules than DTPA molecules, when taking into account that 3 tropolone molecules bind to one

0 1 2 3 4 5 0 20 40 60 80 100 Ef fici en cy (%) DTPA concentration (mM)

Indium ion. Diluting the polymersome solution 10 times, leads to a ten-fold reduction of the number of polymersomes, assuming that the aggregation number of the polymersomes is unchanged, and therefore to the same reduction of the mol DTPA that remains encapsulated. This results in ~ 1000 times more tropolone in the solution. In the presence of such large excess of tropolone, the competition with DTPA will

become more evident and result in a decreased amount of 111_In3+ _{loaded in the lumen.}

The same line of reasoning can explain the observed decrease in loading efficiency also detected for BE2 polymersomes. The thickness of the membrane in this case reduces the volume of the aqueous cavity that could be occupied by DTPA and, as consequence, less amount of the hydrophilic chelate can be encapsulated in the polymersomes. However, besides competition issues, the kinetics of several processes such as association/dissociation kinetics of In-DTPA and In-tropolone complexes, and the actual transfer kinetics that is mediated by the membrane, are expected to play a very important role.

Fig. 8 Loading efficiency of polymersomes as a function of tropolone concentration at DTPA

concentration of 1 mM, 0.15 MBq 111In3+ (0.09 nM) and loading time of 1 hour. Loading has

been carried out at 20 °C and pH 7.4. The squares represent BE1 polymersomes prepared at block copolymer concentration of 0.5 mg/ml, circles denote the same polymersomes loaded after diluting 10 times and the triangles represent BE2 polymersomes prepared at block copolymer concentration of 0.5 mg/ml. The lines are guides for the eyes.

The loading efficiency dependence on the lipophilic concentration implies the need of a transport mechanism through the membrane. In fact, two possible pathways can be

proposed. In the first case, the octahedral In(tropolone)3 complex diffuses as a whole

0 50 100 150 200 250 0 20 40 60 80 100 120 Lo ad ing efficien cy (%) Tropolone concentration (µM)

through the bilayer until a DTPA molecule having higher affinity for 111In3+ replaces the lipophilic ligands (Fig. 9A) , capturing the ion into the lumen. A second possibility involves the hopping of Indium from tropolone molecule to tropolone molecule present in the membrane until it finds the hydrophilic chelate in the aqueous cavity (Fig. 9B). The tropolone distribution can have an essential role here, for instance, it can form a kind of ion channel through which the Indium ions can be transported, analogous to

transfer mechanisms found in cells.28 _{Although no direct proof is established here, the}

fact that the experiments suggest a minimum tropolone concentration that is at least 100 times higher than the Indium ion concentration may be an indication that the second mechanism plays a more important role. Experiments carried out by Rao and Dewanjee suggesting fast dissociation kinetics of Indium-tropolone complexes further

corroborate this theory.29 _{Such transfer efficiency dependence on the concentration of}

lipophilic ligand has also been reported for liposomes, revealing once again the similarities between the two systems.

Fig. 9 Schematic representation of possible transport mechanisms of 111In through a

polymersome bilayer A) diffusion of In-tropolone complexes as a whole and B) 111In hopping

from a tropolone molecule to a tropolone molecule that may form a pathway through the membrane.

In typical in vivo studies involving liposomes and 111_{In as SPECT radiotracer, high}

radioactivities are being used reaching up to 15 MBq per 200 µL of injected sample.30

The only publication reporting SPECT imaging of polymersomes which outer surface

has been labeled with 111_{In suggests that similar amounts of radioactivity are required}

B A

to obtain high spatial resolution images.16 _{Fig. 10 shows the amount of} 111_In3+ _{given in}

MBq and the loading efficiency as a function of the initial amount of radioactivity added to the solution for a fixed incubation time, tropolone and DTPA concentration. It

has to be noted that even at 30 MBq (i.e. 17.4 nM 111In3+, 17.4 pmol), at least 100

times less Indium is present than tropolone or DTPA. At these low Indium concentrations the loading efficiency appears to be independent of the initial amount

of 111_In3+ _{added to the polymersome solution, leading to a linear dependence between}

the encapsulated and originally provided 111In3+ activity. No saturation in the amount

of radioactivity that could be transferred into the polymersomes has been observed in

these experiments, indicating that even larger amounts of radioactive 111In3+ can be

entrapped in the lumen. These results are in agreement with earlier reported radiolabelling studies of liposomes with similar dimensions, which demonstrate that

the loading efficiency remains unchanged at concentrations up to approximately 1 µM

of 111_In3+_.18

Fig. 10 Total encapsulated 111In activity in BE1 polymersomes (■) and loading efficiency of

BE1 polymersomes (●) as a function of the activity of 111In3+ initially added to the polymersome solution. Loading time is fixed to one hour for all samples at temperature of 20 °C and at pH 7.4. Tropolone and DTPA concentrations are respectively 20 µM and 1mM. The lines are guides for the eyes.

!

Besides a sufficient radionuclide loading, the polymersomes need to be stable under physiological conditions and exhibit a negligible loss of the encapsulated radiotracer to be safely applied in in vivo studies. We have focused on the loss of radiolabel since polymersomes composed of PB-PEO are known to have high stability due to the

0,1 1 10

0,1 1 10

Activity in incubation solution (MBq)

En capsu la ted Act ivi ty (MBq ) 0 20 40 60 80 100 Ef fici ency (%)

inherently slow kinetics of this block copolymer in aqueous solutions, and that have

already been demonstrated to remain intact in serum for several days.7 _{For this}

purpose DTPA (1 mM) has been added to 111In loaded polymersomes and the

solution has been left to equilibrate for 48 hours. Indium released from the

polymersomes binds to the DTPA allowing subsequent separation of the 111In-DTPA

complex and the polymersomes using size exclusion chromatography.31 _{Fig. 11A}

shows the elution profile of a solution containing the 111_{In loaded polymersome in the}

presence of a fixed amount of DTPA. Polymersomes have been found to emerge in the volume fractions 5 to 10 as determined by DLS, while the In-DTPA appears much later at volume fractions 12 to 18. Fig. 11A shows the elution profiles of the

polymersomes and 111In-DTPA obtained separately. Clearly, these results show that

only a small percentage of 111_{In-DTPA can be detected (< 5 %) after 48 hours of}

incubation. This indicates that the loss of 111_{In from within the aqueous cavity of the}

polymersomes is very low, despite of the large amount of DTPA added to the

solutions containing the 111_{In-loaded polymersomes.}

Furthermore, we have examined the loss of 111In in serum at 37 °C at incubation time

of 24 hours. Fig.11B shows that only a very small amount of 111_{In (< 4 %) can be}

found in the serum fraction after chromatography on Sepharose 4B column. In both experiments the presence of tropolone trapped in the bilayer, capable of extracting

some of the encapsulated 111In from the lumen is probably the main reason for the

Fig. 11 A) Elution profile of BE1 polymersomes loaded with 111In (0.15 MBq) and 111In-DTPA after incubation with 1 mM DTPA for a period of 48 hours and separation using Sephadex G-25 size exclusion column (DxL=1x21 cm). The polymersomes elute in the fractions 5 to 9 and the 111In-DTPA in the fractions 11 to 17. B) Elution profile of BE polymersomes loaded with

111_{In (0.15 MBq) after incubation in serum at 37 °C for 24 hours as assessed by Sepharose 4B}

chromatography (DxL=1x37 cm). The polymersomes elute in the fractions 9 to 13 and the serum from 25 to 29.

2.4 Conclusions

In conclusion, this study shows that a sufficient amount of radionuclides can be loaded in preformed polymersomes with negligible loss of the encapsulated activity

0 5 10 15 20 25 30 0,0 0,1 0,2 0,3 0,4 No rmallized activity faction Volume fraction (mL) 111_{In-polymersomes} 111_In-DTPA A 0 5 10 15 20 25 30 35 0,0 0,1 0,2 0,3 0,4 No rmallized activity fraction Volume fraction (mL) 111_{In-polymersomes} Serum B

allowing the application of nuclear imaging techniques such as SPECT to assess the best polymersome formulation. Radiolabelling of polymersomes composed of low-molecular weight block copolymers can be carried out with the same ease as in liposome systems when active ion transport is used. We show that this method allows for high radiolabelling efficiency to be achieved within several minutes of loading. However, factors such as steric hindrance and diffusion limitations can play a role in polymersomes having large hydrophilic corona and/or thick hydrophobic bilayer. In addition, excess of transport molecules, i.e. tropolone, has been found to be essential to achieve high entrapment efficiency. Clearly, two parallels can be drawn between liposomes and polymer vesicles that suggest comparable transport mechanism in both systems: 1) the loading time increases as the fluidity of the membrane is reduced and 2) a surplus of a lipophilic agent is necessary to transport sufficient amount of the metal ion to the aqueous cavity.

2.5 References

1. B. M. Discher, Y. Y. Won, D. S. Ege, J. M. C. Lee, F. S. Bates, D. E. Discher

and D. A. Hammer, Science, 284, 1143-1146,1999.

2. J. Z. Du and R. K. O’Reilly, Soft Matter, 5, 3544-3561, 2009.

3. M. H. Li and P. Keller, Soft Matter, 5, 927-937, 2009.

4. P. Tanner, P. Baumann, R. Enea, O. Onaca, C. Palivan and W. Meier, Account.

Chem. Res., 44, 1039-1049, 2011.

5. D. H. Levine, P. P. Ghoroghchian, J. Freudenberg, G. Zheng, M. J. Therien, M. I.

Greene, D. A. Hammer, and R. Murali, Methods, 46, 25-32, 2008.

6. F. Ahmed, R. I. Pakunlu, A. Brannan, F. Bates, T. Minko and D. E. Discher, J.

Control. Release, 116, 150-158, 2006.

7. P. J. Photos, L. Bacakova, B. Discher, F. S. Bates and D. E. Discher, J. Control.

Release, 90, 323-334,

2003.

8. G. P. Robbins, R. L. Saunders, J. B. Haun, J. Rawason, M. J. Therien and D. A.

Hammer, Langmuir, 26, 14089-14096,2010.

9. J.J. Lin, P.P. Ghoroghchian, Y. Zhang and D. A. Hammer, Langmuir, 22,

3975-3979, 2006.

10. S. Egli, M. G. Nussbaumer, V. Balasubramanian, M. Chami, N. Bruns, C. Palivan and W. Meier, J. Am. Chem. Soc., 133, 4476-4483, 2011.

11. H. Gruell, S. Langereis, L. Messager, D. D. Castelli, A. Sanino, E. Torres, E. Terreno and S. Aime, Soft Matter, 6, 4847-4850, 2010.

2009.

13. P. P. Ghoroghchian, P. R. Frail, K. Susumu, D. Blessington, A. K. Brannan, F. S. Bates, B. Chance, D. A. Hammer and M. J. Therien, Proc. Natl. Acad. Sci. U. S. A., 102, 2922–2927, 2005.

14. M. Marguet, L. Edembe and S. Lecommandoux, Angew. Chem. Int. Ed., 51, 1173 –1176, 2012.

15. M. Marguet, O. Sandre and S. Lecommandoux, Langmuir, 28, 2035-2043, 2012.

16. R. P. Brinkhuis, K. Stojanov, P. Laverman, J. Eilander, I. S. Zuhorn, F. P. J. T. Rutjes and J. C. M. van Hest, Bioconjugate Chem., 23, 958−965, 2012.

17. K. J. Hwang, J. E. Merriam, P. L. Beaumier and K. S. Luk, Biochimica et Biophysica Acta, 716, 101-109, 1982.

18. M. R. Mauk and R. C. Gamble, Anal. Biochem., 94, 302-307, 1979.

19. T. J. Wadas, E. H. Wong, G. R. Weisman and C. J. Anderson, Chem. Rev., 110, 2858-2902, 2010.

20. S.W. Provencher, Macromol.Chem., 180, 201-209, 1979.

21. H. Bermudez, A. K. Brannan, D. A. Hammer, F. S. Bates and D. E. Discher, Macromolecules, 35, 8203-8208, 2002.

22. T. P. Smart, O. O. Mykhaylyk, A. J. Ryan and G. Battaglia, Soft Matter, 5, 3607– 3610, 2009.

23. R. Finsy, Adv. Colloid Inter. Sci., 52, 79, 1994.

24. J. C. M. Lee, M. Santore, F. S. Bates and D. E. Discher, Macromolecules, 35, 323-326, 2002.

25. G. Battaglia, A. J. Ryan and S. Tomas, Langmuir, 22, 4910-4913, 2006. 26. S. C. Semple, A. Chonn, and P. R. Cullis, Biochemistry, 35, 2521–2525,1996. 27. J. A. Veiro and P. R. Cullis, Biochim. et Biophys. Acta, 1025, 109-115, 1990. 28. B. Alberts, D. Bray, J. Lewis, et al. Molecular Biology of the Cell. 3rd edition.

New York: Garland Science; 1994. Principles of Membrane Transport 1. Available from: http://www.ncbi.nlm.nih.gov/books/NBK28357/

29. S. A. Rao and M. K. Dewanjee, Eur. J. Nucl. Med., 7, 282-285, 1982.

30. I. O. Umeda, K. Tani, K. Tsuda, M. Kobayashi, M Ogata, S. Kimura, M. Yoshimoto, S. Kojima, K. Moribe, K. Yamamoto, N. Moriyama and H. Fujii, Ann. Nucl. Med., 26, 67–76, 2012.

2.6 Appendix

Calculations of loading capacity:

BE1 polymersomes, D=100 nm

a) Determining the average internal volume per vesicle

The average internal volume per vesicles, Vint, is calculated using the diameter of

the polymersomes and the thickness of membrane. The membrane thickness W=7 nm has been determined by Cryo-EM. The diameter of polymersomes has been set at 100 nm according to the size measured by Cryo-EM.

int 3 3 5 3

4

4

[(

/ 2)

]

[(100 / 2) 7]

3.3 10

3

3

V

=

π

D

−

W

=

π

−

=

×

nm

b) Determining the number of vesicles per gram of polymer

For the block copolymer poly(butadiene-b-ethylene oxide) (PB-PEO) used in this study , the PB content by weight is 66.7 %. The weight of PB per gram of polymer is, therefore, equal to ca. 0.667 g/g. The volume of PB per vesicle is calculated using the equation below:

PB weight of one polymersome is:

Assuming that the density of PB equals 0.93 g/mL, the concentration of polymersomes is 0.5 mg/mL. The number of vesicles per milliliter of solution can be expressed as: 3 3 3 3 5 3 4 4 [( / 2) ( / 2) ) ] [(100 / 2) ((100 / 2) 7) ] 1.9 10 nm 3 3 PB i i i V =

π

π

Average weight of PB per polymersomes − g

×

= = = ×