A Novel Nanoparticle Therapy for Glioblastoma

A novel nanoparticle therapy for glioblastoma

Victor M. Lu

A thesis submitted in fulfilment of the requirements for the degree of Doctor of Philosophy (PhD)

University of New South Wales

Faculty of Medicine Prince of Wales Clinical School Adult Cancer Program

Cure Brain Cancer Foundation Biomarkers and Translational Research

January 2019

i THESIS/DISSERTATION SHEET

Thesis/Dissertation Sheet

Surname/Family Name : LU Given Name/s : VICTOR MING YUAN Abbreviation for degree as give in the University calendar : PHD Faculty : MEDICINE School : PRINCE OF WALES CLINICAL SCHOOL Thesis Title : A NOVEL NANOPARTICLE THERAPY FOR GLIOBLASTOMA

Abstract 350 words maximum: (PLEASE TYPE) Glioblastoma (GBM) is a malignant brain tumour with a median overall survival of 15 months, despite best management of surgical resection, radiation therapy (RT) and chemotherapy with temozolomide (TMZ). This outlook has not changed in the last decade. Nanoparticle (NP) therapy presents an exciting novel consideration due to its ability to be formulated and target the brain. This thesis describes the evaluation of rare earth oxide (REO) NP therapies as potential treatment for GBM given the chemical properties of rare earth elements (REEs) have been shown to exert anti-cancer effects. In characterising and testing a series of four different REO NPs, it was shown that they can penetrate into a GBM cell, and that they have potential to not only be cytotoxic to both immortalised and patient-derived GBM cell lines, but also augment the therapeutic effects of RT and TMZ as well. In particular, REO La2O3 was found to be the most cytotoxic NP tested, and further investigation into determining mechanisms of effect were pursued.

The cause of cell death by La2O3 was found to be multifaceted. Involvement of both intrinsic and extrinsic apoptosis pathways was demonstrated, as well as pathways involving direct DNA damage and autophagy. Interaction with adjuvant RT and TMZ was also shown to occur via reactive oxygen species (ROS) and alteration in apoptosis tendencies respectively.

Finally, the biocompatibility and biodistribution of La2O3 NP therapy was evaluated in a balb/c nude mouse model. We observed that not only does the NP reach the brain, it behaves quite similarly to other NP therapies in terms of biodistribution. There were no short- or long-term adverse events observed with our designed therapeutic regimen. Concurrent adjuvant RT and TMZ did not affect the compatibility or distribution of the NP. In summary, this thesis describes the initial investigation of a REO NP therapy to improve GBM management and prognosis. Its anti-GBM effects have been demonstrated and rationalised in a number of GBM cell lines, and its shown biocompatibility and biodistribution in a biological system justify future GBM xenograft survival studies to elucidate the transfer from bench to bedside of this novel therapy.

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

…5… February….…………… 2019………...…….… Signature Witness Signature Date The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

ii ORIGINALITY STATEMENT

ORIGINALITY STATEMENT

‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’

Signed ……………………………………………......

5 February 2019 Date ……………………………………………......

iii INCLUSION OF PUBLICATION STATEMENT

INCLUSION OF PUBLICATIONS STATEMENT

UNSW is supportive of candidates publishing their research results during their candidature as detailed in the UNSW Thesis Examination Procedure.

Publications can be used in their thesis in lieu of a Chapter if: • The student contributed greater than 50% of the content in the publication and is the “primary author”, ie. the student was responsible primarily for the planning, execution and preparation of the work for publication • The student has approval to include the publication in their thesis in lieu of a Chapter from their supervisor and Postgraduate Coordinator. • The publication is not subject to any obligations or contractual agreements with a third party that would constrain its inclusion in the thesis

Please indicate whether this thesis contains published material or not. This thesis contains no publications, either published or submitted for publication ☐ (if this box is checked, you may delete all the material on page 2)

Some of the work described in this thesis has been published and it has been ☐ documented in the relevant Chapters with acknowledgement (if this box is checked, you may delete all the material on page 2)

This thesis has publications (either published or submitted for publication) ☐ incorporated into it in lieu of a chapter and the details are presented below

CANDIDATE’S DECLARATION I declare that: • I have complied with the Thesis Examination Procedure • where I have used a publication in lieu of a Chapter, the listed publication(s) below meet(s) the requirements to be included in the thesis. Name Signature Date (dd/mm/yy) VICTOR M LU 05/02/2019

iv COPYRIGHT STATEMENT

'I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of sis

::�:: ····························

Date .... �.� ...... /!1/fJ ...... Z-Ql. 1......

AUTHENTICITY STATEMENT

'I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion form

Signed ......

Date .... �!...... M./.l:Y...... 2.0{.9...... Abstract

Glioblastoma (GBM) is a malignant brain tumour with a median overall survival of 15 months, despite best management of surgical resection, radiation therapy (RT) and chemotherapy with temozolomide (TMZ). This outlook has not changed in the last decade. Nanoparticle (NP) therapy presents an exciting novel consideration due to its ability to be formulated and target the brain. This thesis describes the evaluation of rare earth oxide (REO) NP therapies as potential treatment for GBM given the chemical properties of rare earth elements (REEs) have been shown to exert anti-cancer effects. In characterising and testing a series of four different REO NPs, it was shown that this novel NP therapy can penetrate into the GBM cell, and that it has potential to not only be cytotoxic to both immortalised and patient-derived GBM cell lines, but also augment the therapeutic effects of RT and TMZ as well. In particular, REO La2O3 was found to be the most cytotoxic REO NP tested, and further investigation into determining mechanisms of effect was pursued.

Finally, the biocompatibility and biodistribution of La2O3 NP therapy was evaluated in a balb/c nude mouse model. It was observed that not only does the NP reach the brain, it also behaves quite similarly to other reported non-REO NP therapies in terms of biodistribution. There were no short- or long-term adverse events observed with the designed therapeutic regimen. Concurrent adjuvant RT and TMZ did not affect the compatibility or distribution of the NP. In summary, this thesis describes the initial investigation of a novel REO NP therapy that has shown basic promise to improve GBM management and prognosis. Its anti-GBM effects have been demonstrated and rationalised in a number of GBM cell lines, as well as its shown biocompatibility and biodistribution in a biological system. Collectively, these findings justify future GBM xenograft survival studies to elucidate the transfer from bench to bedside of this novel therapy.

v Academic publications during candidacy

1. Lu VM, Alvi MA, McDonald KL, Daniels DJ. The impact on survival by the H3K27M mutation in pediatric high grade glioma: a systematic review and meta-analysis. J Neurosurg Ped. 1 December 2018 [Epub]. https://doi.org/10.3171/2018.9.PEDS18419 2. Lu VM, Goyal A, Rovin RA, Lee A, McDonald KL. Concurrent versus non-concurrent immune checkpoint inhibition with stereotactic radiosurgery for metastatic brain disease: a systematic review and meta-analysis. J Neuro-oncol. 4 November 2018 [Epub]. https://doi.org/10.1007/s11060-018-03020-y 3. Lu VM, Texakalidis P, McDonald KL, Mekary RA, Smith TR. The survival effect of valproic acid in glioblastoma and its current trend: a systematic review and meta- analysis, Clin Neurol Neurosurg. 15 September [Epub]. https://doi.org/10.1016/j.clineuro.2018.09.019 4. Lu VM, Goyal A, Vaugh LS, McDonald KL. The impact of hyperglycemia on survival in glioblastoma: a systematic review and meta-analysis, Clin Neurol Neurosurg. July 2018; 170-165-169. https://doi.org/10.1016/j.clineuro.2018.05.020 5. Lu VM, Phan K, Yin JX, McDonald KL. Older studies can underestimate prognosis of glioblastoma biomarker in meta-analyses: A meta-epidemiological study for study- level effect in the current literature. J Neuro-oncol. 17 May 2018 [Epub]. https://doi.org/10.1007/s11060-018-2897-2 6. Lu VM, Jue TR, McDonald KL, Rovin RA. The survival effect of repeat surgery at glioblastoma recurrence and its trend: A systematic review and meta-analysis, World Neurosurg. 12 April 2018 [Epub]. https://doi.org/10.1016/j.wneu.2018.04.016 7. Lu VM, McDonald KL. Isocitrate dehydrogenase 1 mutation subtypes at site 132 and their translational potential in glioma, CNS Oncol; 2018 Jan [Epub]. https://doi.org/10.2217/cns-2017-0019 8. Lu VM, McDonald KL, Rovin RA. The potential of fluid-attenuated inversion recovery imaging in improving glioblastoma outcome. Int J Surg Proced. 2018 Feb; 1(1):104. https://doi.org/10.31021/ijsp.20181104 9. Lu VM, Jue TR, Phan K, McDonald KL. Quantifying the prognostic significance in glioblastoma of seizure history at initial presentation: a systematic review and meta- analysis. Clin Neurol Neurosurg; 2018 Jan;164:75-80. https://doi.org/10.1016/j.clineuro.2017.11.015 10. Lu VM, McDonald KL, Townley, HE. Realising the therapeutic potential of rare earth elements in designing nanoparticles to target and treat glioblastoma. Nanomedicine (Lond). 2017 Oct;12(19):2389-2401. https://doi.org/10.2217/nnm-2017-0193 11. Lu VM, McDonald KL. The current evidence of statin use affecting glioblastoma prognosis, J Clin Neurosci. 2017 Aug;42:196-197. https://doi.org/10.1016/j.jocn.2017.04.036

vi Academic presentations during candidacy

1. Lu VM, McDonald KL. Elucidating the translational potential of La2O3 nanoparticles as a novel therapy in glioblastoma. Poster presentation. Society of Neuro-Oncology (SNO), New Orleans, US. 15-18 November 2018. 2. Lu VM, Texakalidis P, McDonald KL. The survival benefit of valproic acid in glioblastoma trends away from significance with newer studies: a systematic review and meta-analysis. Poster presentation. Society of Neuro-Oncology (SNO), New Orleans, US. 15-18 November 2018. 3. Lu VM, Townley HE, McDonald KL: Radiosensitization by rare earth oxide nanoparticles in glioblastoma: an in vitro proof of concept. Poster presentation. Society Neuro-Oncology (SNO), San Francisco, US. 17-19 November 2017. 4. Lu VM, McDonald KL, Townley HE: The biochemical implications of in vitro rare earth element nanoparticles in glioblastoma. Poster presentation. British Neuro- Oncology Society (BNOS), Edinburgh, UK. 21-23 June 2017.

vii Acknowledgements

To acknowledge all those that contributed to the formulation of this thesis is a thankless task, as it is almost impossible to account for all direct and indirect contributions to it. Hence, I will keep this broad and general as possible to recognise the great breadth of support I have been fortunate to receive throughout my time completing this thesis. The one exception is to my primary supervisor A/Prof. K. McDonald. I am so deeply grateful to you for your faith in me and this project from the outset. The experiences gained both within the laboratory and outside of it have shaped much of my resolve in how I would like to continue my career in cancer research. These lessons are invaluable, and I hope to share them with as many people as possible moving forward. To all the post-doctoral researchers, research assistants, and fellow graduate students in the laboratory group and the wider Adult Cancer Program, past and present, I appreciate very much all your guidance, friendships, and interactions. These too have allowed me to form a greater understanding of what is required for a successful laboratory group to thrive. I only hope the best for you all in the future. It would be amiss not to thank all those involved with the affiliate bodies of the University of New South Wales that have made this thesis and its results possible: Prince of Wales Clinical School, Adult Cancer Program, Electron Microscope Unit, Biological Resources Imaging Laboratory, and Flow Cytometry. Additionally, special thanks to the Prince of Wales Clinical School and the Translational Cancer Research Network for your kind funding. To my dearest friends and colleagues around the world, if I have ever mentioned this thesis to you, thank you so much for listening. I would rather not list individual names at the risk of forgetting someone, but feel free to ask me any time if you are included in this! Finally, to Angie, Tommy, Yaneth, Cissy, and George, this thesis and any fruits that it bears in the future best case scenario, are dedicated wholly and completely to you.

viii Abbreviations

95% CI 95% confidence interval A low-grade astrocytoma AA anaplastic astrocytoma Ab antibody ACEC Animal Care and Ethics Committee ADP adenosine diphosphate AIF apoptosis-inducing factor ALT alanine aminotransferase AMT adsorptive-mediated transcytosis AnnV Annexin V ANOVA analysis of variance AO anaplastic oligodendroglioma AOA anaplastic oligoastrocytoma ATCC American Type Culture Collection AUC area under curve AUMC area under the first movement curve BBB blood brain barrier BCNU 1,3-bis-(2-Chlor-oethyl)-1-nitrosurea BDNF brain-derived neutrotrophic factor BSA bovine serum albumin BUN blood urea nitrogen CD cytochalasin D CFU colony forming unit CH chlorpromazine CI combination index CMV cytomegalovirus CNS central nervous system cPARP cleaved PARP CST Cell Signaling Technology CT chemotherapy CTX chlorotoxin DIABLO direct inhibitor of apoptosis -binding protein with low pI DMSO dimethyl sulfoxide DNA deoxyribonucleic acid dsDNA double stranded DNA ED50 median effective dose EDTA ethylenediaminetetraacetic acid EE early endosome EGF epidermal growth factor EGFR epidermal growth factor receptor EORTC European Organisation for Research and Treatment of Cancer

ix EPR enhanced permeability and retention FDA Food Drug and Administration FGF fibroblast growth factor FLAIR fluid-attenuated inversion recovery FSC forward scatter GBM glioblastoma GBM-O glioblastoma with an oligodendroglial component GTR gross total resection H&E hematoxylin and eosin H2DCFDA 2′,7′-dichlorodihydrofluorescein diacetate Hb haemoglobin HREC Human Research Ethics Committee HRP horseradish peroxidase HSV-1 herpes simplex virus 1 IAP inhibitor of apoptosis protein IC50 half maximal inhibitory concentration ICL immortalised cell line ICP-MS inductively-coupled plasma mass spectroscopy IDH1 isocitrate dehydrogenase 1 IP intraperitoneal IV intravenous LC3 light chain 3 LE late endosome MD mean difference MGMT O6-alkylguanine DNA alkyltransferase MHC major histocompatibility complex MRI magnetic resonance imaging MTS 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4- sulfophenyl)-2H-tetrazolium NA not available NCIC National Cancer Institute of Canada NP nanoparticle NY nystatin O oligodendroglioma OA oligoastrocytoma OR odds ratio PARP poly (ADP-ribose) polymerase PBS phosphate-buffered saline PDCL patient-derived cell line PED polyethylene glycol PI propidium iodide PMSF phenylmethylsulfonyl fluoride POWH Prince of Wales Hospital PS phosphatidylserine

x QIMR Queensland Institute of Medical Research Rb rabbit RBC red blood cell REE rare earth element REO rare earth oxide RES reticuloendothelial system RMT receptor-mediated transcytosis ROS radical oxygen species RT radiation therapy SD standard deviation SDS sodium docecyl sulfate SEM scanning electron microscope Smac second mitochondrial-derived activator of caspase SSC side scatter TEM transmission electron microscope TMZ temozolomide TNF tumour necrosis factor TRAIL TNF-related apoptosis inducing ligand UCL upconversion luminescence UNSW University of New South Wales WHO World Health Organisation

xi Table of contents

THESIS/DISSERTATION SHEET ...... ii ORIGINALITY STATEMENT ...... iii INCLUSION OF PUBLICATION STATEMENT ...... iv Abstract ...... v Academic publications during candidacy ...... vi Academic presentations during candidacy...... vii Acknowledgements ...... viii Abbreviations ...... ix Table of contents ...... xii List of tables ...... xvii List of figures ...... xviii CHAPTER 1 ...... 1 1 Review of the current literature ...... 2 1.1 Glioblastoma: a deadly brain tumour ...... 2 1.1.1 Epidemiology ...... 2 1.1.2 Classifications ...... 3 1.1.3 Clinical features ...... 7 1.1.4 Prognosis ...... 8 1.1.5 Standard therapies ...... 9 1.1.6 Emerging novel therapies ...... 11 1.2 Nanoparticle therapy: a novel approach ...... 13 1.2.1 Design parameters ...... 13 1.2.2 Targeting glioblastoma ...... 15 1.2.3 Clinical mode of delivery...... 18 1.2.4 Potential enhancement by functionalisation ...... 20 1.3 Rare earth elements: a potential formulation ...... 21 1.3.1 Characteristics ...... 21 1.3.2 Therapeutic potential ...... 21 1.3.3 Rare earth elements and glioblastoma ...... 26 1.3.4 Safety profile of rare earth elements ...... 27 1.3.5 Rare earth oxides...... 29 1.4 Specific aims ...... 31 CHAPTER 2 ...... 33

xii 2 Materials and Methods ...... 34 2.1 Cell culture ...... 34 2.1.1 Cell lines ...... 34 2.1.2 Maintenance ...... 34 2.1.3 Subculture ...... 34 2.1.4 Seeding into wells ...... 35 2.1.5 Cryopreservation and revival ...... 36 2.1.6 Cell counting ...... 38 2.2 Cell viability assessment ...... 38 2.2.1 MTS assay ...... 38 2.2.2 Clonogenic assay ...... 39 2.3 Therapeutic treatments ...... 40 2.3.1 Nanoparticle therapy ...... 40 2.3.2 Radiation therapy ...... 41 2.3.3 Chemotherapy ...... 41 2.4 Electron microscopy ...... 42 2.4.1 Scanning electron microscopy ...... 42 2.4.2 Transmission electron microscopy ...... 43 2.5 Flow cytometry ...... 43 2.5.1 Endocytosis measurement ...... 45 2.6 Protein assessment ...... 46 2.6.1 Protein extraction ...... 46 2.6.2 General quantification by BCA assay ...... 46 2.6.3 Detection by western blot ...... 48 2.7 In vivo techniques ...... 52 2.7.1 Animal handling...... 52 2.7.2 Post-mortem organ extraction ...... 53 2.7.3 Determination of organ nanoparticle concentration ...... 55 2.7.4 Assessment of nanoparticle biocompatibility ...... 55 2.8 Statistical analyses ...... 59 CHAPTER 3 ...... 60 3 The therapeutic potential of rare earth oxide nanoparticle therapy in glioblastoma...... 61 3.1 Introduction ...... 61 3.1.1 Exert cytotoxic effect ...... 61

xiii 3.1.2 Augment radiation therapy ...... 61 3.1.3 Augment chemotherapy ...... 62 3.1.4 Objective and aims ...... 63 3.2 Results ...... 64 3.2.1 Nanoparticle characterisation...... 64 3.2.2 Cell line characterisation...... 64 3.2.3 Cellular uptake of nanoparticle ...... 64 3.2.4 Cytotoxicity of nanoparticles ...... 69 3.2.5 Interaction between nanoparticles and radiation...... 72 3.2.6 Interaction between nanoparticles and chemotherapy ...... 77 3.3 Discussion ...... 82 3.3.1 REO NPs enter the cell and localise to the endosome complex ...... 82 3.3.2 REO NPs are cytotoxic to GBM cells...... 83 3.3.3 REO NPs enhance radiation therapy in GBM cells ...... 85 3.3.4 Lanthanum oxide NPs enhance temozolomide in GBM cells ...... 87 3.3.5 General limitations ...... 88 3.4 Conclusions ...... 90 3.5 Supplementary ...... 91 3.5.1 Supplementary Tables ...... 91 3.5.2 Supplementary Figures ...... 95 CHAPTER 4 ...... 97 4 Mechanisms of cell death by lanthanum oxide nanoparticle in glioblastoma...... 98 4.1 Introduction ...... 98 4.1.1 Inducing effect within the cell ...... 98 4.1.2 Cell death by apoptosis ...... 99 4.1.3 Cell death by other mechanisms ...... 101 4.1.4 Objective and aims ...... 102 4.2 Results ...... 103 4.2.1 Cellular uptake by endocytosis ...... 103 4.2.2 Lanthanum ion dissociation ...... 103 4.2.3 Cell death by apoptosis ...... 103 4.2.4 Cell death by direct DNA damage ...... 107 4.2.5 Radical oxygen species ...... 107 4.2.6 Cell death by autophagy...... 110

xiv 4.2.7 Interaction of NPs with radiation therapy ...... 110 4.2.8 Interaction of NP with chemotherapy ...... 113 4.3 Discussion ...... 116 4.3.1 Endocytosis of nanoparticle is clathrin-mediated ...... 116 4.3.2 Free lanthanum ions are present in the cell ...... 117 4.3.3 Nanoparticle activates both apoptosis pathways...... 118 4.3.4 Nanoparticle causes direct DNA damage ...... 119 4.3.5 Generation of reactive oxygen species ...... 120 4.3.6 Nanoparticle induces autophagy ...... 122 4.3.7 Nanoparticle augments radiation therapy ...... 123 4.3.8 Nanoparticle augments chemotherapy ...... 123 4.3.9 General limitations ...... 124 4.3.10 Overall summary of mechanism ...... 125 4.4 Conclusions ...... 127 4.5 Supplementary ...... 128 4.5.1 Tables ...... 128 CHAPTER 5 ...... 137 5 Biocompatibility of lanthanum oxide nanoparticle in a mouse model...... 138 5.1 Introduction ...... 138 5.1.1 Biological tolerance ...... 138 5.1.2 Biodistribution ...... 139 5.1.3 Long-term persistence ...... 139 5.1.4 Objective and aims ...... 140 5.2 Results ...... 141 5.2.1 Delivery to the brain ...... 141 5.2.2 Short-term maximum tolerable dose ...... 141 5.2.3 Pharmacokinetics of nanoparticle ...... 146 5.2.4 Long-term safety profile ...... 146 5.3 Discussion ...... 157 5.3.1 The nanoparticle regimen is tolerable ...... 157 5.3.2 Lanthanum nanoparticles reach the brain ...... 158 5.3.3 Biodistribution of nanoparticle outside the brain...... 159 5.3.4 Systemic considerations ...... 161 5.3.5 Therapeutic consequences ...... 164

xv 5.3.6 General limitations ...... 165 5.4 Conclusions ...... 167 5.5 Supplementary ...... 168 5.5.1 Supplementary Tables ...... 168 CHAPTER 6 ...... 173 6 Chapter 6: Executive summary and future directions ...... 174 6.1 Purpose ...... 174 6.2 Background ...... 174 6.3 Objectives ...... 175 6.4 Key findings ...... 176 6.4.1 Identifying the therapeutic potential of REO NPs in GBM ...... 176 6.4.2 Mechanisms of cell death by lanthanum oxide NP therapy in GBM ...... 176 6.4.3 Biocompatibility and biodistribution of lanthanum oxide NP therapy in a mouse model...... 177 6.5 Future directions ...... 178 6.5.1 Xenograft survival study ...... 178 6.5.2 Nanoparticle enhancement ...... 178 6.6 Overall conclusions ...... 179 REFERENCES ...... 180 7 Reference list ...... 181

xvi List of tables Table 1.1 REE electronic configurations...... 22 Table 2.1 Optimised seeding densities...... 37 Table 2.2 Flow cytometry dyes used...... 44 Table 2.3 Components of cell lysis buffer...... 47 Table 2.4 Western blotting primary antibody list...... 51 Table 3.1 Characteristics of REOs investigated...... 65 Table 3.2 Characteristics of cells...... 67

Table 3.3 IC50 values at 3 days...... 73

Table 3.4 IC50 values at 8 days...... 78 Table 5.1 Pharmacokinetics of lanthanum nanoparticle in brain...... 148

xvii List of figures Figure 1.1 Brain tumour incidence...... 4 Figure 1.2 Histological classification of brain tumours...... 6 Figure 1.3 Nanoparticle movement in cell...... 16 Figure 1.4 Typical electronic configuration...... 23 Figure 1.5 Recent publication trend...... 32 Figure 2.1 Schematic of tolerable dose determination...... 56 Figure 2.2 Schematic for biocompatibility determination...... 58 Figure 3.1 Representative images of NPs and cells...... 66 Figure 3.2 Transmission electron microscopy images of NP in cell...... 68 Figure 3.3 Optimal time point for cytotoxicity in vitro...... 70 Figure 3.4 Primary cytotoxicity results...... 71 Figure 3.5 Radiosensitisation in ICLs...... 74 Figure 3.6 Radiosensitisation in PDCLs...... 76 Figure 3.7 Interaction between NP and TMZ...... 79 Figure 3.8 Congruency between observed and theoretical modelled therapeutic effect...... 80 Figure 3.9 Synergy between NP and TMZ...... 81 Figure 4.1 Endocytosis analysis by flow cytometry...... 104 Figure 4.2 Free ion analysis by ICP-MS...... 105 Figure 4.3 Apoptosis pathway analysis by flow cytometry...... 106 Figure 4.4 Apoptosis marker analysis by western blotting...... 108 Figure 4.5 Radical oxygen species analysis by flow cytometry...... 109 Figure 4.6 Mitochondrial apoptosis analysis by western blotting...... 111 Figure 4.7 Autophagy analysis by western blotting...... 112 Figure 4.8. RT interaction analysis by flow cytometry...... 114 Figure 4.9. RT and CT interaction analyses by western blotting...... 115 Figure 4.10. Overall mechanism...... 126 Figure 5.1 Body weights for maximal tolerable dose...... 142 Figure 5.2 Organ weights for maximal tolerable dose...... 144 Figure 5.3 Organ lanthanum concentrations for maximal tolerable dose...... 144 Figure 5.4 Brain lanthanum concentration over time...... 147 Figure 5.5 Body weights for biocompatibility...... 149 Figure 5.6 Organ weights for biocompatibility...... 151 Figure 5.7 Brain lanthanum concentrations for biocompatibility...... 152 Figure 5.8 Organ lanthanum concentrations for biocompatibility...... 154 Figure 5.9 Comparison of organ lanthanum concentrations between two timepoints...... 155 Figure 5.10 Histological staining of brain specimens...... 156

xviii

CHAPTER 1

1 REVIEW OF THE CURRENT LITERATURE

In order to best contextualise the aims and findings of this thesis, the current concepts of glioblastoma (Section 1.1), nanoparticle therapy (Section 1.2), and rare earth elements and oxides (Section 1.3) are introduced and described in this chapter.

1.1 Glioblastoma: a deadly brain tumour

The term brain tumour describes a heterogenous group of neoplasms originating from, or involving, the central nervous system (CNS). Brain tumours are uncommon, with incidence rates of approximately 6.9-23 per 100,000 population across the United States1 and Australia2.

In addition, there is great diversity within brain tumours, from a biological aspect with respect to histopathology, and from a clinical aspect with respect to clinical course.

Of great concern are malignant brain tumours, which across both the United States3 and

Australia4 constitute one third of all presentations. Of the malignant brain tumours, glioblastoma (GBM) is the most deadly, with 5- and 10-year overall survival rates estimated to be as low as 5% and 2% respectively across all age demographics.1 Also termed glioblastoma astrocytoma grade IV, this disease continually invokes more questions than answers every time there is an attempt made to bring about a cure. By understanding what is already known about the clinical course of GBM, we will be better equipped to devise new and novel approaches to improving treatment and outlook for this deadly brain cancer.

1.1.1 Epidemiology

The annual GBM incidence rate is consistently low worldwide, with estimates ranging from 3.2-4.0 per 100,000 across the United States5 and Australia4. In terms of all brain tumours,

GBM is the third most frequently reported CNS histology accounting for 15% of all brain

tumour presentations, 54% of all gliomas, and is the most common malignant CNS tumour overall.1,6

GBM itself comprises of primary and secondary types, which are viewed as separate molecular entities that derive from different genetic pathways, affecting patients at different ages but nonetheless conferring similarly poor prognosis.7 The vast majority of GBM is primary, occurring in more than 80% of presentations, and typically present in patients over the age of 60 years. Whereas secondary GBM typically occurs in younger patients at age 40 years, and develop from lower grade gliomas.8,9

In terms of demographic predilections, GBM is predominantly an adult disease with overall mean diagnosis age of 65 years, and paediatric cases being extremely rare (Figure 1.1).

For gender, there is a demonstrated statistical increase of GBM in males compared to females, although the rationale behind this is currently unclear.1,10 Finally, the latest statistics indicate that GBM is more prevalent in urban areas and Caucasian ethnicities, however selection and survey biases limit interpretation of these associations.1

1.1.2 Classifications

As alluded to earlier, brain tumours present across a spectrum of histopathologies. To date there has been two classification approaches to brain tumours based on histology and more recently, molecular profiling. As a malignant and diffuse glioma, GBM occupies a specialised niche in both these systems.

1.1.2.1 Histology classification

The first attempts to classify brain tumours were based on histological appearance of their primary cell type under the light microscope.11 Typically, more malignant high-grade glial tumours were associated with a number of observable features including nuclear atypia,

Figure 1.1 Brain tumour incidence.

Age-adjusted incidence rates of brain and other central nervous system (CNS) tumours by selected histologies in adults, with glioblastoma (GBM) coloured red. (Adapted from CBTRUS Statistical Report: NPCR and SEER, 2011–2015; Louis et al. 20181)

increased mitotic activity, with evidence of microvascular proliferation and necrosis.12 Based on different combinations of these characteristics, the World Health Organisation (WHO) developed a grading system from I-IV: grade I indicated lesions with low proliferative potential and possibility of cure after surgical resection, across to grade IV describing lesions that were associated with rapid disease evolution and a fatal outcome despite intervention.13

Under this system, GBM is considered grade IV, particularly delineated from other brain and glial tumours by microvascular proliferation and necrosis.14 Although shifting towards molecular profiling, GBM prognostication was once largely influenced by this histology classification.13,15 In fact, it was the histological association of microvascular proliferation and GBM that led to the rise of anti-angiogenic therapies in GBM, and the eventual Foods and Drugs Administration (FDA) approval of bevacizumab in treatment of recurrent GBM.16,17

1.1.2.2 Molecular classification

In the latest iteration of the WHO Classification of CNS Tumours published in 2016, molecular profiling was for the first time integrated into the classification paradigm along with histological findings.18 Although histology continues to provide much of the basis behind the diagnosis, molecular profiling has enhanced our classification of predicting prognosis and thus has become a serious adjunct in histopathological diagnoses. The incorporation of isocitrate dehydrogenase (IDH) mutation and 1p/19q co-deletion evaluation in general provides us with a more robust recognition of clinically relevant subgroups in brain tumours (Figure 1.2).12

With reference to GBM, the primary result of this integration is that immunohistological evaluation of the IDH mutation has created an additional dichotomy which has proven influential in conferring more favourable prognosis.3,19,20 The 1p/19 co-deletion appears not to be prognostically relevant in GBM.21 Although not yet officially incorporated into the WHO Classification of CNS Tumours, another molecular trait for GBM prognosis that

Figure 1.2 Histological classification of brain tumours.

Paradigm shift in classification of GBM, and other gliomas from a A. histologic to an B. integrated histologic and molecular diagnosis of diffuse gliomas. Previously, gliomas formed a histological spectrum, with regard to both cellular phenotype and malignancy grade. The introduction of the molecular markers IDH mutation and 1p/ 19q co-deletion in the World Health Organization (WHO) 2016 definition of gliomas allows for a more robust recognition of clinically relevant subgroups of gliomas in adults, as delineated by the triangles in the second figure. The tumours in the blue triangle in are now considered as “canonical oligodendrogliomas,” those in the pink and red triangles as astrocytic neoplasms, with the latter typically behaving as a primary or de novo GBM. This paradigm shift necessitates further study of how histologic malignancy grade and/or additional molecular markers can be optimally used for prognostication within these newly defined categories. Of note, this simplified scheme does not adequately cover the situation for diffuse gliomas in the paediatric age group. (Adapted from Perry et al., 201612)

A, low-grade astrocytoma; O, oligodendroglioma; OA, oligoastrocytoma; AA, anaplastic astrocytoma; AO, anaplastic oligodendroglioma; AOA, anaplastic oligoastrocytoma; GBM, glioblastoma; GBM-O, glioblastoma with an oligodendroglial component.

has been thoroughly investigated is that of methylation status, specifically methylation at the

O6-methylguanine-DNA methyltransferase (MGMT) promoter, which is associated with improved response to chemotherapy in GBM patients.22,23 These molecular factors, along with a host of other factors and biomarkers currently under investigation, are crucial in further enhancing our classification of not only GBM, but also its optimal treatment strategies, as we move forward.24

1.1.3 Clinical features

The presentation of GBM shares many similarities to that of primary brain tumours.

Typically, they present with subacute neurologic signs and symptoms that manifest over weeks to months; the most common at presentation being non-specific headache (50-60%), seizures

(20-50%) and focal neurologic deficits (10-40%) which include memory loss, motor weakness, and deficits in language, cognition and personality.25

Imaging in the workup of GBM is crucial, for it underpins the decisions involved in much of the possible treatments available. Upon T1-weighted magnetic resonance imaging

(MRI), GBM typically appear as hypointense lesions, with heterogeneous enhancement throughout, intense rim enhancement and central clearing, a common feature due to the associated histological necrosis.26 Peritumoural oedema, as well as increased T2 and fluid- attenuated inversion recovery (FLAIR) signal intensity tend to also be seen in GBM.27

However, despite these features, definitive diagnosis of GBM can only be made on tissue sample analysis.

1.1.4 Prognosis

The prognosis of GBM is infamously dismal. Despite best standard of care treatment, the median overall survival ranges from 14-15 months after diagnosis across both the United

States and Australia.28,29 In terms of 5- and 10-year survival outcomes, only 5% and 2% of

GBM patients respectively are expected to survive.1 This compares to the 5- and 10-year survival rates of 35.0% and 29.3% for all malignant CNS tumours, and 90.7% and 87.2% for all non-malignant CNS tumours.1 It was also shown that patients who were older, male, and from rural areas all trended toward poorer survival, and these trends have been shown consistently in the last decade.1,3,30 However, similar to incidence, we should acknowledge selection and survey biases may confound the strength of these associations.

A significant contributor to the poor prognosis of GBM is the tendency of the disease to recur, despite best efforts of primary treatment. The inevitable return of the disease most often occurs 6-9 months after primary treatment.31,32 At this point, estimate suggest median overall survival of 6-8 months.33,34 This is despite best intervention efforts for the recurrence when deemed safe.

1.1.4.1 Prognostic factors

As previously mentioned in Section 1.1.2.2, there are a host of molecular markers that assist in the prognostication of GBM. Patients that are IDH1 mutant positive35, MGMT methylation positive36, and TERT promoter mutation negative37 have all been thoroughly shown to experience superior prognosis. Other factors that possess potential prognostic potential include epidermal growth factor receptor (EGFR) amplification, p53 expression, loss of heterozygosity of 9p and PD-L1 expression.38 Additionally, there are clinical features such as seizure history39, and hyperglycaemia40, which all may further assist in prognosticating

GBM patients and their survival in the future.

1.1.5 Standard therapies

The conventional treatment modalities for GBM are surgical resection, followed by adjuvant radiation therapy and chemotherapy by temozolomide (TMZ).

1.1.5.1 Surgery

Maximal, safe, surgical resection is a mainstay in the management of GBM. It has been shown that maximal safe resection of the GBM bulk at first diagnosis, and it alone can be associated with up to 5 months of survival benefit.41,42 Many studies41-43 have demonstrated this clearly, establishing surgery as an independent prognostic factor for either subtotal and gross total resections (GTR) compared to no surgery. This benefit appears to derive from the immediate alleviation of tumour burden in both physical and biological senses.44 It, however, must be acknowledged that surgery in GBM is rarely performed with curative intent, for this disease inevitably recurs even in the cases of GTR.45,46 Nonetheless, repeat surgery in these cases has been shown to still confer a survival benefit, thus implicating surgery has a continuous role throughout the management course of GBM.47

1.1.5.2 Radiation therapy

Adjuvant radiation therapy (RT) directed at both gross and residual GBM has been shown to improve local control rates and survival after surgical resection compared to surgery alone.48,49 However, similar to surgery, it is rarely administered with curative intent. RT is typically delivered at 60 Gy, and focuses on targeting both the original lesion and up to a 2 cm peritumoural margin in normal brain tissue, as 80-90% recurrence is believed to manifest in this zone.50 To date, RT has proven itself to be a required modality in GBM treatment, however, how feasible RT is in the setting of GBM recurrence is unclear, and clinical trials are ongoing.51

The use of RT to target this disease relies on the production of local reactive oxygen species (ROS).52 The energy from radiation reacts with surrounding water molecules about the

− cancer to generate ROS such as superoxide anion O2 , hydrogen peroxide H2O2, and hydroxyl radical HO·. These ROS are unstable and highly reactive compounds that can damage cellular

DNA, compromise cellular repair mechanisms and trigger apoptotic processes within the GBM, ultimately inducing cellular death and tumour clearance.53,54

1.1.5.3 Chemotherapy

Cytotoxic chemotherapy represents the most widely researched modality in GBM to date.55 The current standard chemotherapy agent is temozolomide (TMZ), with its therapeutic efficacy and safety in concurrent combination with radiation therapy established in clinical trial28 by the European Organisation for Research and Treatment of Cancer/ National Cancer

Institute of Canada (EORTC/NCIC) in 2005. It has since remained at the helm on conventional cytotoxic chemotherapy treatment of GBM.

In brief, TMZ is an oral alkylating agent, with its primary mechanism of cytotoxicity being that of DNA methylation within a cell, which can overwhelm the normal mismatch repair pathway leading to DNA double-strand breaks, halting cellular replication and ultimately causing cell death.56 While TMZ is the standard of cytotoxic agents in GBM, other previous agents it superseded, such as procarbazine57 and nitrosureas58, relied on cytotoxicity by similar mechanisms to exert their therapeutic effect as well.

A major limitation in the use of cytotoxic TMZ is the phenomenon of resistance, which has become well documented over time, in part due to adaptations of the programmed cell death pathways.56,59 The epigenetic landscape of GBM has been found to be under constant molecular evolution, which arguably renders the current mode of cytotoxicity by TMZ suboptimal.60,61 Thus, the discovery of alternative, novel therapies that cause cell death by other mechanisms would greatly assist in improving optimisation of cytotoxic therapies.

1.1.6 Emerging novel therapies

In addition to the standard treatment options, there is growing understanding of how novel treatment therapies may assist in improving current prognosis of GBM.

1.1.6.1 Immunotherapy

The emerging concept of immunotherapy to treat GBM involves harnessing the immune system of a GBM patient to target the tumour, by inducing, enhancing or suppressing immunes responses to reject cancerous cells.62 This can be via active and/or passive mechanisms targeting specific components of the immune system. For example, targeting checkpoint inhibition63 and adoptive T-cell therapy64 have proven effective in the setting of other solid organ tumours. Unfortunately however, early experience with immunotherapy in

GBM has shown only modest benefit at best. The tolerability of checkpoint inhibitors pembrolizumab and nivolumab appear feasible in patients, however, the occurrence of complete response to therapy has yet to be clearly demonstrated in recurrent GBM either with bevacizumab, or compared to bevacizumab.65,66

It is recognised that translating the success of immunotherapy in other tumours to GBM may prove not so facile, for a number of reasons. Firstly, the CNS is an immune-privileged site, with the blood brain barrier (BBB) preventing free diffusion of immune cells to reach the GBM site.67 Secondly, the microenvironment of GBM in the brain is hostile to the required components of an effective immune response, such as decreased major histocompatibility complex (MHC) expression, and altered expression of T cell costimulatory molecules. 68

Finally, GBM itself secretes a variety of immunosuppressive cytokines, including TGFβ and

IL-10.69 Although these challenges exist, it is not to declare immunotherapy irrelevant yet, and more investigation is required to ascertain how applicable immunotherapy may be in GBM.

There are a number of clinical trials currently active, and their results will assist in determining how effective this novel therapy is in GBM.62

1.1.6.2 Viral therapy

Another novel therapy receiving much attention is viral therapy. This approach involves the use of genetically engineered viruses with decreased neurovirulence that are directly applied to the tumour bed of GBM.70 These specialised viruses lack the ability to replicate in normal neurons and healthy brain tissue, but can still replicate in other types of cells, and are oncolytic, inducing a systemic tumour-specific immunity leading to cancer cell death.71,72 This approach has proven fruitful in the case of some cancers, with an adenovirus for head and neck cancers73, as well as a herpes simplex virus 1 (HSV-1) for melanoma74, having already been approved for marketing.

There has been recent evidence that not only can viral therapy be administered safely to patients with recurrent GBM, and it may provide significant improvements in prognosis. A study involving a cytomegalovirus (CMV) has shown incredible results, with median survival of 41 months, compared to the standard median of 15 months.75 In addition to survival, other viruses have shown significant multiyear durable response in recurrent GBM such as the vector vocimagene amiretrorepvec (Toca511)76 and a recombinant polio virus (PVSRIPO)77. Thus, similar to immunotherapy, this novel therapy for GBM requires greater investigation to validate how exactly its promising concepts may translate to clinically meaningful improvements in all GBM presentations when compared to what our current standard treatments afford.

1.2 Nanoparticle therapy: a novel approach

Nanoparticle (NP) therapy in cancer is an emerging and exciting area. NPs can be designed to act as an efficacious physical vector for microscopic anti-cancerous formulations to reach cancer cells, and in some circumstances, deliver cargo to cancer cells that would be otherwise suboptimal or not deliverable.78 The primary attraction for using NPs in the brain is their proven ability to reach and infiltrate brain matter.79-81 Indeed, in vivo animal studies 82-84 studying intravenous delivery of NPs up to 160 nm in size have shown presence in brain matter using transmission electron microscopy. Furthermore, there is evidence that NPs are not intrinsically toxic to normal brain tissue, making them an attractive consideration in targeting diseases of the brain.85 Thus, in the setting of GBM, NP therapy is quite unknown, but nonetheless, presents an exciting avenue as a novel therapy given its ability to potentially overcome the BBB and deliver anti-cancerous formulations to the cancer cells.

1.2.1 Design parameters

There are a number of key parameters which need to be considered when designing

NPs for use in GBM treatment; they must be able to reach the target site, infiltrate, and then undergo uptake by the cancer cells. Uptake across the cell membrane is both size and shape dependent,86 and also related to the material composition. Interactions of the NPs with the membrane can therefore include electrostatic and hydrophobic forces, and uptake will be a combination of bending and stretching forces in the membrane.

1.2.1.1 Size

To be effective, NPs must be small enough to remain in circulation to reach the GBM, however large enough to avoid excessive excretion by renal and hepatic clearance.87 Based on these requirements, the optimal NP size for cellular uptake in cancer cells has been proposed to be <100 nm.88,89 Given that NPs need to be able to accumulate in the cancer interstitium in

vivo after leaving the circulation, size is a compromise between penetration and retention when administered systemically. Perrault et al.87 found that larger NPs >100 nm appeared to stay nearer the vasculature, permeated less and were retained for longer at the target sites when compared to smaller NPs ~20 nm which penetrated deeper, diffused more quickly throughout the interstitium, but were retained for shorter amounts of time. Finally, a study by Chithrani et al. 90 showed that NPs 50 nm in diameter exhibited the greatest overall uptake in brain cancer cells.

1.2.1.2 Shape

The morphology and aspect ratio of a NP will influence the probability and speed of uptake into the cell. Spherical NPs have been shown to be rapidly internalised due to their symmetrical structure.91,92 Rod shaped NPs, on the other hand, are internalised at varying rates dependent on their presenting axis to the cellular membrane given NP asymmetry enhances charge and ligand presentation depending on shape to influence cellular uptake.93 This is because when uptake is initiated, the local curvature of NPs affects the degree to which they fit the contours of target cell membranes, which are rarely symmetrical.94 Shape and size are often considered together in design, as the efficiency of a NP shape is dependent in part by its size. Chithrani et al.95 demonstrated that NPs <100nm favoured a more spherical than rod shape for uptake in cervical cancer cells. In the same cell line, Gratton et al.96 observed greater uptake for NPs >100 nm favoured rods over spheres.

1.2.1.3 Surface charge

The surface charge on NPs is also an influential parameter, as positive charge is associated with greater cellular uptake than negative or neutral charge. Indeed, Slowing et al.97 demonstrated that in cervical cancer cell lines there was a negative linear relationship between

NP surface charge and median effective dose (ED50). The greater effect exerted by positive charge is attributed to the slight negative charge of the cellular membrane and consequently,

cellular uptake is driven by the attracting electrostatic interactions that bind the NPs to the cellular membrane.98 However, in vivo the inevitable process of protein opsonisation of NPs in serum should be taken into consideration since this will likely interfere with NP-membrane interactions and alter long-term retention and clearance parameters.90

1.2.2 Targeting glioblastoma

The blood vessels that transverse the brain matter, facilitating vital exposure for necessary resources to maintain the tissue, are encased in a layer of cerebral endothelial cells which collectively constitutes the blood brain barrier (BBB). The BBB regulates what can penetrate from circulation to brain matter, and serves as a physical barrier to protect the brain from toxic materials; unfortunately, this also includes most therapeutic drugs.99 The BBB is so effective that estimates suggest more than 98% of small-molecule drugs and effectively 100% of large-molecule drugs are prevented from reaching the brain by the BBB.100 Therefore, it is a significant barrier for novel therapies to overcome in order to target the intracranial GBM.

1.2.2.1 Overcoming the blood brain barrier

The primary promise of NPs to treat GBM is their proven ability to penetrate the cerebral endothelial cells of the BBB from the circulation, and to reach intracranial GBM.79-81

The most accepted pathway of natural NP uptake through the BBB appears to be via adsorptive transcytosis and endocytosis (Figure 1.3).101 This type of transport describes a vector movement of molecules within endocytic vesicles across the cerebral endothelial cells, from the luminal (blood) side to the abluminal (brain) side, followed by exocytosis into the brain interstitium.102 The initial surface adsorption at the luminal surface by the NP depends if the process initiates from a specific or a non-specific ligand-interaction, referred to as receptor- mediated transcytosis (RMT) or adsorptive-mediated transcytosis (AMT), respectively. The three best-known ligands that facilitate the RMT of NPs through the BBB are insulin, iron- transferrin and LDL-cholesterol. 103-105 AMT on the other hand is a consequence of electrostatic

Figure 1.3 Nanoparticle movement in cell. A cross-section of an endothelial cell of the blood brain barrier (BBB) that demonstrates the transport of a nanoparticle (NP) from the luminal to abluminal surface. The NP must attach to the membrane from the luminal surface by either (a) receptor-mediated transcytosis (RMT) or (b) adsorptive-mediated transcytosis (AMT). Then, transport through the BBB occurs via vesicles that derive from either (c) caveolae or (d) clathrin-coated pits, which may involve the endosome, to finally reach the abluminal side for exocytosis onto the abluminal surface. The NPs remain too large to cross the BBB between endothelial cells, as do other large molecular compounds. (Adapted from own publication Lu et al. 2017106).

charge-charge interactions between cationic macromolecules and the negative charges on the luminal surface of the BBB. At systemic physiological pH, the luminal surface of the BBB is anionic largely due to the presence of the sialic acid residues of glycoproteins.107 This proves particularly attract for NP therapies where the NP surface charge is positive.

Once adsorbed, the NP interacts at the endothelial cells via one of two entry site types

(either caveolae or clathrin-coated pits) on the luminal surface. Transcytosis vesicles are formed from these sites, triggering an endocytosis process, which then transport the NP into the cell, and ultimately to the abluminal surface of the BBB to release it into the brain interstitium. The caveolae form free intact vesicles that are directly transported across the cell whereas the clathrin-coated pits form vesicles and fuse with an intracellular endosome before being transported across the endothelial cell.108,109 Additionally, it should be noted that the intercellular junctions between adjacent endothelial cells of the BBB can be as small as 1 nm, which are too small for most NPs to pass through, and all molecule compounds for that matter, thereby relying on this adsorptive transcytosis to overcome the BBB.110

1.2.2.2 Enhanced permeability and retention

It was proposed many decades ago that solid tumours possessed unique pathophysiological traits compared to normal healthy tissue.111 Solid tumours exhibit extensive angiogenesis and hence hypervasculature, defective vascular architecture, impaired lymphatic drainage and recovery system, and greatly increased production of a number of permeability mediators.112 Collectively, these phenomena define an enhanced permeability and retention

(EPR) effect for endogenous and exogenous substances to infiltrate tumour interstitium from the vasculature. Specific to the brain, the EPR effect has been associated with a number of changes observed in the locality of GBM. This primarily includes the breakdown of the BBB113, the downregulation of the tight junction protein claudin creating larger intercellular junction gaps114, as well as a redistribution of glial protein aquaporin AQP4 which disturbs the water

balances within the brain115. Clinically, the deterioration of BBB integrity induces cerebral oedema which can manifest as headaches, loss of consciousness or seizures.116

The EPR effect provides a specific avenue for NPs to better penetrate the BBB and the brain to target GBM. This is of particular significance when considering not all GBM cases are amenable to gross total resection due to possible diffuse infiltration and proximity to eloquent grey matter structures. Furthermore, there will be a greater concentration of NPs in the target

GBM cells themselves, and less incidental exposure to non-cancerous cells. However, this effect has been noted to be significantly less pronounced in the brain microenvironment than in cancers elsewhere in the body.117 Consequently, actual delivery load of NPs may be different in GBM compared to other cancer types.118

1.2.3 Clinical mode of delivery

The two main delivery modalities for NPs to target GBM are discussed below, with neither proven superior to the other as of yet. However, the systemic exposure through intravenous delivery may modestly favour trialling direct intratumoural delivery at this point in time. However, until the biocompatibility of a specific NP is investigated, delivery selection remains speculative.

1.2.3.1 Intravenous delivery

Administration via the circulation is the most widely accessible option in clinic for therapeutic drug delivery to GBM. However, the NPs are subject to a number of physiological obstacles including first-pass metabolism, opsonisation, circulation to all organs of the body and the need to efficiently overcome the BBB.106 One study investigating the intravenous delivery of NPs to stereotactically implanted GBM in one lateral ventricle of rats, made three significant conclusions regarding the intravenous approach 119. Firstly, the greatest delivery efficiency of the NP to the brain tissue was 0.93% one hour after administration, whereas

median delivery of NPs to a solid tumour organ is 0.7%.120 Secondly, the concentration of NP formulation was greater in GBM tissue compared to healthy contralateral brain tissue. Thirdly, it was observed that six days after GBM implantation that the magnitude of BBB disruption was sufficient to permit entry of the NPs, presumably due to the EPR effect. It is a timely reminder that while the intravenous approach is able to take advantage of the natural BBB breakdown in the presence of GBM, the BBB is still present and needs to be to effectively overcome, most likely by NP design optimisation.

1.2.3.2 Intratumoural delivery

Delivery directly into the tumour bypasses the issue of the BBB by surgical intervention in a set-up similar to brain needle biopsy.121 A recent study by Cohen et al., 122 analysed the effectiveness of intratumoural administration of a chemotherapy agent coupled to NPs into

GBM in mice. Significantly increased survival in the experimental arm was observed compared to the control arm of mock-treated mice. These results imply a modest degree of success for this delivery approach. A future consideration in refining this approach is the circulation of the cerebral interstitial fluid which affects therapy movement intracranially.123 This may influence the micro-distribution of NPs at delivery prior to their cellular uptake which poses a clinical dilemma because cellular perturbation by NPs to the surrounding non-cancerous cells in the

GBM microenvironment is not desireable.124 This is exemplified by the success of the FDA- approved 1,3-bis-(2-chlor-oethyl)-1-nitrosurea (BCNU, Gliadel ®) wafer, a chemotherapy that is implanted directly into the tumour cavity after surgery. Initially, this approach was seen as a breakthrough as it provided direct targeting to a GBM site overcoming the physiological barriers facing intravenous delivery with modest clinical gains.125 However, even today, it remains unclear how the pharmacokinetics of the drug is defined in both the tumour and surrounding healthy tissue.126 There is no guarantee as of yet that NPs delivered intratumourally will not also suffer a similar fate.

1.2.4 Potential enhancement by functionalisation

Functionalisation is the process of modifying NPs by specific ligand addition to enhance their efficiency in reaching, targeting and infiltrating the appropriate cells of interest. These all remain possible avenues for future research for design NP therapy to treat GBM.

1.2.4.1 Augmenting BBB penetration

A number of ligands conjugated to NPs have been shown to improve BBB penetration, and can be broadly separated into four different groups based on mechanism of effect:

i. mediate adsorption at the BBB interface to initiate transcytosis, e.g. poly(sorbate 80)127

ii. interact with receptors on the BBB to encourage RMT, e.g. transferrin peptide128 iii. increase lipophilicity and decrease negative charge to encourage AMT, e.g. CLFFA129 iv. prolong circulation time to increase exposure to uptake sites along the BBB, e.g.

polyethylene glycol (PEG)130.

1.2.4.2 Enhancing glioblastoma specificity

NPs have been functionalised with a number of targeting molecules to improve brain tumour uptake in animal models, including chlorotoxin (CTX)131, anti-EGFR antibody132, and anti-transferrin receptor antibody133. Additionally, functionalisations have also been successfully shown to augment targeting of and internalisation of NPs into GBM cells specifically. These include peptides such as the Arg-Gly-Asp (RGD) sequence and interleukin-

13134, iron-mimics targeting transferrin and transferrin receptor135, and neurofilamental analogues of tubulin-binding sites136. For more detail regarding the interaction between NPs and functionalisation in determining cellular fate within the brain interstitium, an excellent review by Song et al 137 describes how surface chemistry governs cellular tropism of NPs in the brain and in GBM.

1.3 Rare earth elements: a potential formulation

The rare earth elements (REEs) are a series of non-radioactive elements that consist of seventeen f-electropositive metals; lanthanum (La), cerium (Ce), praseodymium (Pr), neodymium (Nd), promethium (Pm), samarium (Sm), europium (Eu), gadolinium (Gd), terbium (Tb), dysprosium (Dy), holmium (Ho), erbium (Er), thulium (Tm), ytterbium (Yb), and lutetium (Lu). Scandium (Sc) and yttrium (Y) are also considered REEs because of similar chemical and toxicological characteristics. REEs have demonstrated in vitro and in vivo evidence of toxicity, and given their ability to form oxides and produced in NP formulations, they present as an attractive therapeutic anti-GBM therapy that has not yet to date been investigated.106

1.3.1 Characteristics

Despite their name, REEs are not especially rare, and are readily found in the Earth’s crust as rare earth oxides (REOs), as well as naturally within the human body, with average intravenous levels in the order of 1 ng/kg.138

All REEs exhibit preference for the oxidation state +III with an unprecedented uniformity amongst other chemical series in the periodic table. This is attributed to the core- like nature of their unique isolated 4f orbitals (Table 1.1). All RE3+ ions have the configuration

[Xe]4fn6s2 with the exceptions of Ce, Gd and Lu, where the 5d orbital is singly occupied. The particular interest in these elements comes from the number of unpaired electrons these orbitals can hold and how they may interact biologically with GBM cells.

1.3.2 Therapeutic potential

Due to the electronic configuration of these elements, there is potential for therapeutic effect by means of natural cytotoxicity, augmentation of radiation therapy, as well as augmentation of chemotherapy (Figure 1.4).

Table 1.1REE electronic configurations.

Electronic configuration of all rare earth elements (REE) and number of unpaired electrons. Presence of unpaired electrons in the 4f orbitals are demonstrated.

Ground state electronic Rare Earth Element Unpaired electrons configuration Lanthanum, La [Xe] 5d1 6s2 1 Cerium, Ce [Xe] 4f1 5d1 6s2 2 Praseodymium, Pr [Xe] 4f3 6s2 3 Neodymium, Nd [Xe] 4f4 6s2 4 Promethium, Pm [Xe] 4f5 6s2 5 Smarium, Sm [Xe] 4f6 6s2 6 Europium, Eu [Xe] 4f7 6s2 7 Gadolinium, Gd [Xe] 4f7 5d1 6s2 8 Terbium, Tb [Xe] 4f9 6s2 5 Dysprosium, Dy [Xe] 4f10 6s2 4 Holmium, Ho [Xe] 4f11 6s2 3 Erbium, Er [Xe] 4f12 6s2 2 Thulium, Tm [Xe] 4f13 6s2 1 Ytterbium, Yb [Xe] 4f14 6s2 0 Lutetium, Lu [Xe] 4f14 5d1 6s2 1 Scandium, Sc [Ar] 3d1 4s2 1 Yttrium, Y [Kr] 4d1 5s2 1

[Xe] = [Kr] 4d10 5s2 5p6; [Kr] = [Ar] 3d10 4s2 4p6; [Ar] = 1s22s2 2p63s2 3p6

Figure 1.4 Typical electronic configuration.

The ground-state electronic configuration of typical rare earth element gadolinium, which would be found even in oxide formulation. The 4f and 5d orbitals in the dotted box represent the high-energy orbitals that are unique to rare earth elements (REEs) and are filled according to ).

. The represented unpaired electrons in these orbitals are believed to be involved in the anticancer properties of REEs in glioblastoma (GBM) therapy. In nanoparticle (NP) formulation, these mechanisms include (a) direct cytotoxicity, (b) augmenting radiation therapy and (c) augmenting chemotherapy. (Adapted from own publication Lu et al. 2017106).

1.3.2.1 Natural cytotoxicity

REEs have been shown to be cytotoxic in normal cells naturally, and are believe to be due to their impact on intracellular calcium homeostasis that disrupts oxidative stress management systems.139,140 The mobilisation of internal calcium, as well as ionic Ca2+ entry across the plasma membrane, contribute to the ATP-evoked changes in protein phosphorylation within astrocytes.141 These calcium-dependent phosphorylation systems are involved in a wide range of vital cellular functions, including energy metabolism and oxidative stress management, as well as cell motility, proliferation and differentiation.142 The activity of these systems in astrocytes have been shown to be decreased in the presence of REEs such as La.141 This is likely generalisable to certain degree across REEs, for their ionic radii range from 98 to 116 pm.143 Considering the ionic radius of Ca2+ is 99 pm, ionic REEs would readily compete with

Ca2+ ions that are crucial for the regulation of many proteins within astrocytes, and thus theoretically at least, in GBM cells as well which are believed to derive from an astrocytic lineage.139

Furthermore, in each astrocyte there is an overriding tonic inhibition of excitatory glutamatergic neurotransmission triggered by successive rises in the intracellular Ca2+ from adjacent astrocytes caused by local production of ROS.144,145 Any interferences in the intracellular Ca2+ content of astrocytes by REEs can then alleviate this inhibition leading to an excitatory neurotoxicity and eventual death. 146 Hence there appears possible mechanisms by which the ionic radii of REEs can compete with Ca2+ to exert a cytotoxic effect on astrocytic cells. The reality is most likely a synergistic combination of multiple processes.

1.3.2.2 Improving radiation efficiency by radiosensitisation

Radiation therapy (RT) is frequently used to either reduce the size of GBM prior to surgery or as adjuvant therapy.147 RT works by reacting with the intracellular aqueous environment to generate toxic ROS such as superoxide anion, hydrogen peroxide and hydroxyl

radicals. The ROS are unstable and highly reactive compounds that damage cellular DNA, compromise cellular repair mechanisms and trigger apoptotic processes within cells, collectively leading to cellular death.53,54

Various materials and mechanisms can be employed to enhance the efficacy of RT, such as the placement of radiosensitisers at the tumour site. Due to their electronic structure,

REEs display a phenomenon termed ‘the Auger effect’ in response to irradiation which may provide efficient radiosensitisation in GBM. The Auger effect describes the release of an electron from an outer-shell to replenish an inner-shell ionised by radiation, which then leads to an energy emission.148 In the case of the REEs with a large number of electron orbitals of differing energy, the Auger effect can repeat itself across multiple times, creating an Auger cascade which leads to the amplification of the number of electrons released and subsequent energy emitted in addition to the incoming radiation dose. The main weakness of conventional

RT is the lack of cancer specificity and control for sensitivity. By the Auger effect, REE-based

NPs present a novel option to augment incoming RT dose at a target site, reducing the intensity of the incoming dose and its collateral damage to the healthy brain matter it passes to reach the

GBM target.

Radiosensitisation in the presence of REE Gd has been implicated both in vitro and in

149 vivo for GBM. Using their Gd2O3-based NPs in the GBM U87 cell line, Mowat et al. successfully demonstrated a significant decrease in survival of cells exposed to RT, similar to routine X-ray therapy, in the presence of the NPs. While the supporting evidence in human subjects is yet to be reported, it is promising that significantly longer survival times have been implicated in animal models. Le Duc et al.84 evaluated radiosensitisation potential of their

Gd2O3-based NPs in GBM-inoculated rats exposed to highly selective microbeam radiation therapy (MRT) at either 5 or 20 minutes after intravenous NP injection. The lifespan of rats treated at 20 minutes after injection was increased approximately fourfold when compared to those treated 5 minutes after injection. Presumably the time delay between the two groups

allowed for greater infiltration of the tumour by the NPs in the 20-minute group, and thus a greater radiosensitisation and survival effect.

1.3.2.3 Improving chemotherapy delivery by photosensitisation

The ability for REE NPs to augment the yield of chemotherapy is the least researched area with regards GBM management. The most popular, and direct, mechanism is a combined cytotoxicity of REE NPs and an established cytotoxic therapy.

Another possible consideration with REE NPs that requires more investigation is the use of up-conversion luminescence (UCL) and cleavage of pro-drugs, or activation by photosensitisation. 150 UCL, otherwise known as photon upconversion or anti-Stokes emission, is the sequential absorption of two or more low-energy photons which leads to a higher-energy emission or luminescence.151 The generation of UCL NPs could provide an interesting class of materials which are superior to other conventional fluorophores by giving high signal-to-noise ratio, and better photostability. Excitation in the NIR region could also provide deeper tissue penetration and lower photodamage to biological samples.152 Furthermore, NP formulations with REEs could be used to activate photosensitive chemotherapy agents coupled to their surface, and induce a local photodynamic cytotoxicity. This concept has been demonstrated to inhibit growth of GBM in mice using the cytotoxic Chlorin e6 (Ce6) photosensitiser conjugated to Yb and Er NPs.153 The Ce6 group also aided the passage of the REE NPs through the BBB into the GBM cells due to the lipophilic and hydrophobic nature of the photosensitiser.154 Based on these findings, REE NPs have a vast therapeutic potential to be realised in anti-GBM therapy development.

1.3.3 Rare earth elements and glioblastoma

It is possible that REEs may have a natural tendency to accumulate in astrocytic cancer cells, including GBM. Zhuang et al.155 found statistically significant higher concentrations of the REEs La, Ce, Gd, Lu and Sc in astrocytoma cells from human samples compared to age-

matched autopsy controls. A possible explanation for this specific accumulation is that astrocytes are responsible for metal deposition and transport within the CNS, and their dysfunction in cancer leads to aberrant REE processing.156 This does assume that brain cancer cells such as those in the GBM cells possess some similar physiological properties to that of normal astrocyte cells, an assumption inferred by the astrocyte model of GBM.157 It has even been proposed that astrocytes may even possess the capacity to accumulate REEs to less detriment compared to surrounding neurons.158 This is due to the fact that astrocytes exhibit inherent transcriptional and expressional regulations of brain-derived neutrotrophic factor

(BDNF), which protects from oxidative stress159 and govern Ca2+ and free ROS metabolism160.

1.3.4 Safety profile of rare earth elements

There remain many unknowns about the translational process required to take REE- based NPs from bench to bedside in GBM management. One primary area requiring greater validation is the potential for adverse effects of with REE NPs in the brain and the rest of the body.

1.3.4.1 Adverse effects in the brain

If the EPR effect does not operate effectively, then REE NPs could permeate through brain tissue without specific directive to the GBM site, and deposit in surrounding healthy brain tissue. Zhao et al.161 demonstrated in vivo mice that some REEs, such as Ce and Nd, have the ability to induce neuronal inflammation and oedema in healthy brain tissue at a cellular level.

These effects are associated with the potential anticancer cytotoxic mechanisms of REEs, and in the case of non-GBM neuronal death, includes lipid peroxidation, ROS production and downregulation of antioxidant activity.139 Although such undesirable effects at a cellular level in healthy cells occur only at high concentrations of REEs, it remains uncertain as to what these concentrations are and their predictability in vivo. Furthermore, permeability, migration patterns, and clearance rates through healthy brain and tumour interstitium, as well as affinity

for particular neuronal cell types, and toxic threshold of REEs in the healthy brain and GBM are not clearly defined. While these concerns may be more relevant to intravenous delivery, in which all brain areas serviced by vasculature would be affected, intratumoural delivery could still expose healthy brain tissue to REEs albeit at a more localised level. Regardless, the possible induction of healthy neuronal death is a cause for concern and demands thorough investigation in vitro and in vivo before translational studies are considered.

It is currently unknown what the long-lasting effects of accumulated REE NPs at a

GBM site in the brain may be; they may be detrimental to surrounding tissue, they may be cleared by the vascular and lymphatic systems, or they may simply be benign in-situ. If the

NPs cannot be cleared and pose a risk to non-cancerous tissue once their anticancer properties have been utilised, then minimally invasive extraction from the tumour site could be feasible, but far from ideal. Clearly, many surgical factors such as tumour or cavity size, anatomical location, NP delivery media will also need to be considered.

1.3.4.2 Adverse effects in the rest of the body

As inorganic NPs, REE NPs may be sequestered from circulation and have the potential to accumulate at high levels in organs such as the kidneys, liver and spleen.89 This is because the two primary excretion routes from systemic circulation for REE NPs are (i) the renal pathway as urine and (ii) the hepatic pathway as bile in faeces. Renal clearance occurs via glomerular filtration, can be a rapid process with minimal catabolism making it an ideal pathway for NPs to be cleared. However, it is difficult to predict how well unspecific REE NPs will pass through this pathway. This is due to the fact that although the glomerular fenestrations are physically 43 nm in diameter162, physiologically they are significantly smaller at approximately 5 nm in diameter due to the summative effects of the adjacent glomerular capillary layers163. For NPs that cannot pass through the glomerular fenestrations, they enter the spleno-hepatic pathway for clearance. The liver is the primary site for catabolisation and

elimination of foreign materials into the biliary system. It has been reported that the liver most efficiently eliminates NPs up to the size of 20 nm.164

There is a concern of local toxicity if NPs do not clear from the body and accumulate in excretory organs. If REE NPs were to accumulate in the kidney, REEs such as La, Ce, and

Nd could be extremely nephrotoxic at high doses, capable of inducing nephritis, epithelial cell necrosis and oxidative stress to kidney which ultimately may lead to organ failure.165 Of all organs in the body however, there is a distinct predilection for inorganic NPs up to 250 nm in size to rapidly accumulate in the liver, with only slow clearance the alleviating process.166 This poses a unique situation where the risk of REE NP breakdown in the catabolic liver can affect liver function. At high doses, REEs such as Ce, Pr and Gd can induce many unwanted hepatototoxic effects, such as hepatitis, steatosis, jaundice, haematological derangement and necrosis.167,168 Interestingly, REEs may also be cytoprotective in the liver, as evidence exists of mitogenesis promotion in response to insult and P450-dependent monooxygenase inhibition.169,170 Nevertheless, determining the threshold for nephrotoxic and hepatotoxic effects is a significant goal regarding the use of REE in NP formulation targeting the brain.

Approaches that will complement this include designing ultra-small REE NPs to favour renal excretion and functionalising NP surface chemistry to resist hepatic catabolism and encourage greater clearance.

1.3.5 Rare earth oxides

As a consequence of their unique electronic configurations, REEs react vigorously with

O2 to produce highly-stable rare earth oxides (REO); in most cases these are the sequioxides,

(REE)2O3, but other REE with reasonably low fourth ionisation energies form higher oxides,

171 specifically CeO2, Pr6O11 and Tb4O7. These REOs can be readily formulated into nanoparticles (NPs) due to their chemical stability.

1.3.5.1 Applications to date

The combination of high availability, chemical stability and NP formulation of REOs has spurred the interest of applying this technology in the management of brain cancer. REEs are particularly noteworthy for their use as contrast agents for MRI.172 This technique is a non- radiation, high-resolution imaging modality that measures proton relaxation of water molecules after exposure to radiofrequency excitation. Paramagnetic contrast agents, those with unpaired electrons, can enhance the detection of these energy changes by extending relaxation time or augmenting magnetic moment. REEs are an attractive consideration as given their expanded electronic 4f subshells, all RE3+ ions possess at least one unpaired electron and are paramagnetic, with the exceptions of La3+ and Lu3+.173 Amongst the REEs, Gd3+ remains the most commonly used element owing to its seven unpaired 4f electrons, large magnetic moment, and long electronic relaxation time to enhance imaging brightness.174 For a more in-depth analysis on REEs in NP formulation for bioimaging applications relevant to the brain there are a number of excellent reviews published.175,176

1.4 Specific aims

The progress in improving GBM prognosis has been slow. Prior to the inclusion of

TMZ into the surgery and radiation regimens for GBM, median survival was approximately

11-12 months.28 In its defining trial in 2005, TMZ improved median survival to 15 months in the experimental group. Since then, survival gains have been modest, with recent studies reflecting much of the survival gains in that study, with survivals from 15-21 months for similarly treated patients.177-179 This rate of improvement does not appear to reflect the exponential rate at which the current GBM literature continues to grow (Figure 1.5). Rather, it may be a reflection of a persistence to investigate therapeutic modalities with no further capacity for improved survival. Examples such as viral therapy have demonstrated that clinically significant improvements in overall prognosis may require consideration of more novel therapies in the future.

Thus, the overarching goal of this thesis was to explore the feasibility and plausibility of a novel NP therapy made from REO in GBM. Specifically, each of the following experimental chapters reflect a specific aim in achieving this goal:

Aim 1: To establish if a REO NP can infiltrate GBM cells and cause anti-cancerous effects.

Aim 2: To determine by which mechanism(s) such effects occur.

Aim 3: To evaluate the biocompatibility of a REO NP in vivo.

3000 l

) I

(

N /

s 2000

t a

a d

c o

E l

b 1000

u P

0 1960 1980 2000 2020 Year

Figure 1.5 Recent publication trend.

Trend of academic publications regarding glioblastoma (GBM) currently indexed on PubMed database. Search was performed using the term ‘glioblastoma’ only, from inception to December 2018. The landmark publication of the EORTC/NCIC trial28 in 2005 marked an increase in publication output, however, as of today (December 2018), improvement rates in overall survival have not matched the publication rate on this topic.

EORTC/NCIC, European Organisation for Research and Treatment of Cancer/ National Cancer Institute of Canada.

Database accessible at https://www.ncbi.nlm.nih.gov/pubmed.

CHAPTER 2

2 MATERIALS AND METHODS

2.1 Cell culture

2.1.1 Cell lines

From the American Tissue Culture Collection (ATCC; Manassas, USA), glioblastoma

(GBM) immortalised cell lines (ICLs) U87 MG (ATCC no. HTB-14) and M059 K (ATCC no.

CRL-2365) were obtained.

GBM patient-derived cell lines (PDCLs) RN1, GBML1 and WK1 were obtained from our collaborative research partners at the Queensland Institute of Medical Research (QIMR;

Brisbane, Australia). PDCL G53 was a cell line developed by our group from a GBM tissue sample provided directly from our collaborators at the Prince of Wales Hospital (POWH;

Sydney Australia) with ethics approval by the South Eastern Sydney Illawarra Area Health

Service Human Research Ethics Committee (HREC #10/121).

2.1.2 Maintenance

The media used to maintain cell lines was RHB-A medium (Clontech Laboratories Inc.,

Sweden). The media was supplemented with Fibroblast Growth Factor (FGF; Sigma-Aldrich,

USA; Cat. No. F0291) at 1:1000 dilution, and Epidermal Growth Factor (EGF; Sigma-Aldrich,

USA; Cat. No. E9644) at 1:1000 dilution. Cells were maintained in humidified conditions within an incubator (Thermo Fisher Scientific, USA), at 37ºC in 5% CO2 atmosphere. Media was regularly refreshed every 2 to 4 days.

2.1.3 Subculture

Cells were subcultured at 80% confluence and above. Empty NuncTM cell culture flasks

(T-25, T-75, and T-175; Thermo Fisher Scientific, USA) were prepared with BD MatrigelTM

Basement Membrane Matrix solution (BD Biosciences, USA; Cat. No. 354234) at 1:100 dilution in Dulbecco’s Phosphate-Buffered Saline (PBS; Thermo Fisher Scientific, USA) at least three hours prior to subculture (5 mL for T-75, and 2 mL for T-25). Firstly, medium was removed from the flask, and the exposed cell monolayer was washed with PBS to remove any remaining medium and non-adherent cells. Then, cells were exposed to trypsin-like Accutase® solution (Sigma-Aldrich, USA; Cat. No. A6964) for 5 minutes of incubation at 37ºC in 5%

CO2 atmosphere in order for the cells to detach from the flask (2 mL for T-75, and 1 mL for T-

25). Cells were inspected for sufficient dispersion, and then trypsin inhibitor (Sigma-Aldrich,

USA; Cat. No. T9128) was added to neutralise the Accutase (1 mL for T-75, and 0.5 mL for

T-25). At this point, the cells were transferred to a 15 mL falcon tube and centrifuged at 1200

RPM for 3 minutes. The solution was then removed carefully, to expose the cell pellet. This was resuspended in 3 mL of maintenance medium pre-warmed to 37oC, and then divided at the appropriate subculture ratio into the destination culture flask(s). Further medium was added to satisfy required total volume (10 mL for T-75, and 5 mL for T-25), and flask was returned to the incubator at 37ºC in 5% CO2 atmosphere.

2.1.4 Seeding into wells

Cells to be seeded were obtained from the resuspension solution during subculture as described in Section 2.1.3. In order to prevent formation of neurospheres, destination flasks and wells were prepared with MatrigelTM solution in 1:100 PBS at least three hours prior to seeding (100 µL/well for 96-well plate, and 1 mL/well for 6-well plate). All wells were supplemented by maintenance medium to a prescribed total volume unless stated otherwise

(200 µL/well for 96-well plate, and 2 mL/well for 6-well). In the case of the 96-well plates, wells along all four borders were designated evaporation wells, and filled with 200 µL/well of

PBS to provide protection from incubator-associated evaporation affecting the cells and volumes of the inner wells. Only one cell line was seeded per plate.

2.1.4.1 Optimisation

Depending on size of wells and the length of assay, optimal cell density differed. These are listed in Table 2.1. Optimisation was performed prior to any investigation across a range of seeding densities in the appropriate well and length of time. The major determinant for optimal seeding cell density was when cells were just confluent at time of measurement. In the case of clonogenic assays in Section 2.2.2, optimal density was determined when colonies were as numerous as possible without significant overlap.

2.1.5 Cryopreservation and revival

Cells to be cryopreserved were obtained from the cell pellet subculture as described in

Section 2.1.3, with the centrifugation conducted at 4oC.The pellet was resuspended in 3 mL maintenance medium supplemented by dimethyl sulfoxide (DMSO; Sigma-Aldrich, USA; Cat.

No. D8418) in 1:10 maintenance media. 1 mL aliquots were then dispensed into 2 mL freezing cryovials and transferred into -80 ºC for at least 4 hours within a freezing container. The vials were then stored in a vapour-phase liquid nitrogen tank until required again.

In order to revive cryopreserved cells, vials were retrieved from the storage tank and thawed at 37 ºC in a water bath for 2 minutes. The vial solution was then slowly resuspended into warmed 10 mL of maintenance medium at 37oC and then centrifuged at 1000 rpm for 5 minutes. The solution was then removed carefully, to expose the cell pellet. The pellet was carefully resuspended into 1 mL of medium, transferred to its destination flask, and further medium was added to satisfy required total volume (10 mL for T-75, and 5 mL for T-25).

Destination flasks were prepared with MatrigelTM solution in 1:100 PBS at least three hours prior to subculture (5mL for T-75, and 2 mL for T-25).

Table 2.1 Optimised seeding densities.

List of optimised seeding densities for each cell line reported in this thesis for assays of different time lengths. Time points represent the time of therapy exposure to the cells, which was added 24 hours after initial seeding to allow cells to adhere to the surface. Final well volume for 96-well plate wells, 6-well plate wells and T- 25 flasks were 200 µL, 2 mL and 5 mL respectively.

Optimised seeding density (cell number/volume) 96-well 6-well T-25 Cell line Cell type 72 h* 8 d 24 h 72 h 14 d 72 h U87 MG ICL 10,000 . . . 1000 . M059 K ICL 10,000 . . . 1000 . RN1 PDCL 3000 750 200,000 100,000 2,000 300,000 GBML1 PDCL 2500 500 150,000 83,333 2,000 200,000 G53 PDCL 2000 250 100,000 67,777 1,250 150,000 WK1 PDCL 3000 750 200,000 100,000 6,500 300,000

*This seeding density was used for preliminary cytotoxic evaluation across 24-96 hours, with all viability assay results normalised to the negative control relative to each time point.

ICL, immortalised cell line; PDCL, patient-derived cell lines; h, hour; d, day.

2.1.6 Cell counting

Cells were prepared for counting from the resuspension solution obtained during subculture as described in Section 2.1.3. 10 µL of the resuspension solution was first transferred into a 1.5 mL Eppendorf® tube and then mixed by passage with 10 µL of 0.4% trypan blue. Cell counting was then performed with this dyed solution by two different methods, accounting also for the 1:1 dilution by the trypan blue in any calculations.

2.1.6.1 Automatic cell counter

10 µL of the dyed solution was loaded into the chamber port of a chamber slide. This slide was then inserted into the automated Countess ® cell counter (Invitrogen, USA), and a count was performed. Reported results included overall cell number (cells/mL), viable cell count (cells/mL) and viability (%). Viable cell count was used as surrogate for any further calculations, providing that the reported viability was reasonable.

2.1.6.2 Manual hemocytometer

10 µL of the dyed solution was loaded onto a counting slide under a coverslip. The slide was then examined by light microscopy with a 10X objective using a CKX41 light microscope

(Olympus, Japan). A hand tally counter was used to count live, unstained cells in the 4 corner major grid squares. The average of these counts represented viable cell count (x104 cells/mL).

2.2 Cell viability assessment

The viability of cells was assessed in order to characterise cell line proliferation and evaluate the effectiveness of interventions to cause cell death.

2.2.1 MTS assay

This colorimetric utilised the CellTiter 96® Aqueous One Solution Cell Proliferation

Assay (Promega, USA), and was performed primarily in 96-well plates. It is based on the

metabolic reduction of an MTS tetrazolium compound [3-(4,5-dimethylthiazol-2-yl)-5-(3- carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium] by viable, proliferating cells only.

The process generates a yellow-brown coloured formazan product, which can be analysed by photospectrometry.

Cells were seeded according to optimised densities listed in Table 2.1. Once cells were ready for assessment, total 96-well volumes were reduced from 200 to 100 µL by the removal of media. Then, 20 µL of the MTS reagent was added to each well. The cells were then

o incubated at 37 C in 5% CO2 atmosphere for 1 hour. The absorbance was measured then using the SpectraMax ® M2 Multi-Mode Microplate Reader (Molecular Devices, USA) at 490 and

650 nm wavelengths. Data was exported from the SoftMax ® Pro 6 Data Acquisition and

Analysis Software (Molecular Devices, USA) into Prism (GraphPad, California, USA) for complete analysis. Analysis included subtraction of the 650 nm absorbance from the 490 absorbance to account for background, and values were then normalised to the negative control.

2.2.2 Clonogenic assay

The design of this assay was to assess the viability of cells by means of their ability to proliferate and form colonies (colony-forming units, CFUs). This was conducted in 6-well plates with wells coated with MatrigelTM solution in 1:100 PBS at least three hours before seeding. Cells were seeded into 6-well plates in 2 mL of media at densities listed in Table 2.1.

The cell suspension at desired density was made prior to seeding in a separate 50 mL falcon tube to ensure maximal homogeneity. Plates were then placed in the incubator at 37oC in 5%

CO2 atmosphere. For any therapeutic testing, cells were then exposed to therapy the next day, and returned to the incubator for 14 days.

Then, plates were removed from the incubator, and the supernatant removed. Wells were washed gently with 2 mL PBS twice, before 500 µL of 0.5% crystal violet dye in 1:1 distilled water:methanol was added to each well. The dye was allowed to permeate into cells

for 1 hour, and then removed. The dyed CFUs were then washed gently with 2 mL PBS twice, and allowed to air-dry overnight. Plates were then digitised using a 4490 Photoscanner (Epson,

Australia), before CFU counts were quantified using ImageJ (NIH, Bethesda, USA) after representative colony formation and size were confirmed by light microscopy. All counts were normalised to the negative control for subsequent analysis.

2.3 Therapeutic treatments

In order to evaluate therapeutic effect in the selected ICLs and PDCLs, all therapies required titration to establish effect that was not overwhelmingly cytotoxic, but significant enough to induce a measurable change in cell viability.

2.3.1 Nanoparticle therapy

REO NPs were purchased in nanopowder from Sigma Aldrich (Sydney, Australia): cerium (IV) oxide (CeO2) nanopowder (<100 nm), gadolinium (III) oxide (Gd2O3) nanopowder

(<100 nm), lanthanum (III) oxide (La2O3) nanopowder (<100 nm) and neodymium (III) oxide

(Nd2O3) nanopowder (<100 nm).

For the purposes of this study, NPs were prepared in suspension of PBS solution. To achieve desired concentrations, NPs in powder form were accurately weighed, and dispersed in an appropriate volume of PBS. Then, the solution was placed in a sonication bath (Unisonics,

Australia) at 40 kHz for a minimum of 5 minutes to ensure maximal dispersion. Solutions were then vortexed using a Reax top (Heidolph, Germany) at maximum rotation for 1 minute to maximise homogeneity, before immediate administration.

2.3.1.1 Nanoparticle titration

All NPs were examined for cytotoxic effect over different time periods (24, 48, 72 and

96 hours) and concentrations (0-100 µg/mL). For MTS assays, cells were exposed across the concentration range in triplicate 24 hours after seeding into 96-well plates and measured across

the time range to calculate IC50 values where applicable. For clonogenic assays, cells were exposed to NPs across the concentration range only in triplicate 24 hours after seeding into 6- well plates and measured 14 days later to determine concentrations which induced >50% reduction in CFU count.

2.3.2 Radiation therapy

Radiation was administered using the X-Rad 320 Biological Irradiator (Precision X- ray, USA). This was for both cell and animal experiments, and specimens were placed on a specific mount and within a specific holding device respectively to ensure optimal exposure.

2.3.2.1 Radiation titration

For clonogenic assay optimisation, cells were exposed to radiation across different dosages (1, 2, 4, 6 and 8 Gy) in triplicate 24 hours after seeding into 6-well plates and measured

14 days later to determine dosages which induced >50% reduction in CFU count. In vivo experiments utilised a previously validated dosage of 4 Gy as most simulative of a clinical scenario (HREC #14/104A).

2.3.3 Chemotherapy

The chemotherapy agent used in this study was temozolomide (TMZ), the current standard of care agent in GBM management. A TMZ stock solution was prepared from powder form (Sigma-Aldrich, Castle Hill, Australia). The content of one 25 mg vial was dissolved slowly into 800 µL of DMSO. This produced a stock solution of 16095.8 µM. The solution was stored in 50 µL aliquots at -20oC. When required, aliquots were exposed to room temperature 25oC and allowed to thaw. The solution was gently passaged to restore homogeneity before administration at the appropriate concentration.

2.3.3.1 Temozolomide titration

Chemotherapy with TMZ was primarily measured by MTS assay. Cells were exposed to TMZ across different concentrations (0-1000 µM) in triplicate 24 hours after seeding into

96-well plates and measured 8 days later to calculate IC50 values. Eight days has been previously demonstrated by our group to be the required time to observe consistent cytotoxic effect. If the initial concentration range was too broad and induced cytotoxicity throughout, titration was repeated across a much smaller range, e.g. 0-200 µM. Additionally, a negative control of cells exposed to the same concentration of DMSO was also performed to ensure minimal cytotoxic effect caused by the TMZ solvent.

2.4 Electron microscopy

Electron microscopy was used to evaluate physical and in vitro characteristics of our

NPs using scanning (SEM) and transmission electron microscopy (TEM) respectively.

2.4.1 Scanning electron microscopy

The NPs to be characterised were first suspended in MilliQ water at a concentration of

1000 µg/mL, and sonicated in a sonication bath (Unisonics, Australia) at 40 kHz for 5 minutes.

A drop of this suspension was then placed on a Parafilm M film (Bemis NA, USA) surface.

Silica chips were induced with negative charge using the easiGlowTM Glow Discharge

Cleaning System (PELCO, USA), before being inverted onto the NP suspension drop, and allowed to rest for 10 minutes. This would ensure sufficient time for the electrostatic attraction to deposit the NPs on the surface of the silica. The chips then washed by inversion onto three consecutive milliQ water drops, resting on each drop for 2 minutes. The chips were then mounted onto a 12.5 mm pin stub and allowed to dry in a desiccator cabinet (Labec, Australia) at 35oC overnight. The next day, the chips were carbon coated using a Deck Carbon Coater

(NSC, Iran). Chips were mounted onto the NOVATM NanoSEM 230 (FEI Company, USA) for analysis at 5 kV in Immersion Mode and Through-the-Lens Detector turned on.

2.4.2 Transmission electron microscopy

Cells were seeded in thinly Matrigel-coated 27 mm NuncTM Glass Bottom Dishes

(Thermo Fisher Scientific, USA; Cat. No. 150682) and exposed to the NP therapy at 100 µg/mL concentration 24 hours after. One hour later, the media was removed, and cells were fixed in

2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer solution (pH 7.4) for one hour at room temperature 25oC. Cells were then post-fixed with 1% osmium tetroxide for 1 hour at room temperature, and serially dehydrated in increasing percentages of ethanol. They were then embedded in increasing ratios of LX-112 resin/ethanol to 100% resin in a BioWave microwave

(Pelco, USA), and left overnight at 60oC to polymerise. A ultramicrotome (Leica, USA; model

UC6) was used to cut vertical ultrathin (60 nm) sections. These sections were then imaged using F200 TEM (JEOL, USA; model1011) at 80 kV.

2.5 Flow cytometry

Cells were seeded in 2 mL of media in 6-well plates and exposed to the appropriate therapy or therapies 24 hours after seeding. At the chosen time point, supernatant was discarded, and the cells were washed with PBS, before being trypsinised with Accutase®, with tryspin inhibitor added after. The trypsin mixture was then centrifuged at 1200 RPM for 3 minutes at room temperature. The supernatant was removed, to leave a pellet, which was resuspended in

PBS, and re-centrifuged at 1200 RPM for 3 minutes at room temperature. The supernatant was then removed again, and the pellet was resuspended in 100 µL PBS. To this resuspension, fluorescent dyes were added at the appropriate concentration and allowed to incubate at 37oC in 5% CO2 atmosphere for the appropriate amount of time. The dyes used to investigate cell death are described in Table 2.2.

Table 2.2 Flow cytometry dyes used.

List of fluorescent dyes used in single cell assessment by flow cytometry.

Final Absorbance Fluorescent dye Marker Company concentration (nm) Annexin V Apoptosis 1:50 stock 530 Roche PI (propidium iodide) Cell death 1:50 stock 670 Roche H2ACDFDA ROS, general 10 µM 530 TFS MitoSox® ROS, mitochondrial 5 µM 580 TFS

ROS, reactive oxygen species; TFS, Thermo Fisher Scientific.

The dye solution was then analysed using the BD FACSCanto™ II (BD Biosciences, USA).

Initially, voltage was set at 230 V, flow rate at low, with no gating. As events were collected, voltage was adjusted accordingly to ensure complete capture, flow rate adjusted to achieve

500-1000 events per second, and gates were created to incorporate cells of interest only.

Forward (FSC) and side scatter (SSC) measurements were used to identify and exclude doublets, cell debris, and raw NPs. Once the positive control cells were confirmed, a total of

10,000 events were recorded per experimental sample. Quantification analyses were performed using FlowJo 10.5 (FlowJo LLC, Oregon, USA).

2.5.1 Endocytosis measurement

Cell lines were exposed to a variety of different endocytosis inhibitor for 30 minutes: nystatin (NY) inhibits caveolae-mediated endocytosis, chlorpromazine (CH) inhibits clathrin- mediated endocytosis, and cytochalasin D (CD) inhibits macropinocytosis. The concentrations of these inhibitors were based on reported values in the literature: 10 µg/mL for NY and CH, and 1 µg/mL for CD.180-183 All concentrations were first tested in neutral conditions (no NP) to confirm no significant cell death was induced, otherwise it was titrated appropriately down.

Inhibitor solution was then removed, and a stock solution of 100 µg/ml NP was added for 1 hour. NPs were then washed off and prepared for flow cytometry. The relative amount of NP taken up by the cell was determined by utilising flow cytometry to determine and compare the median side scatter intensity metric of SSC-A (a representative value of intracellular granularity which in this case would be caused by presence of NPs).184

2.6 Protein assessment

2.6.1 Protein extraction

Cells were seeded in 5 mL of media in T-25 flasks and exposed to the appropriate therapy or therapies 24 hours after seeding. At the chosen time point, supernatant was collected, the cells were washed with PBS, the washings collected, and then all remaining cell material trypsinised with Accutase®, with trypsin inhibitor added after. This collected mixture was then centrifuged at 1200 RPM for 3 minutes in 4°C conditions. The supernatant was then removed, to leave a pellet, which was resuspended in cold PBS, and then centrifuged at 1200 RPM for 3 minutes in 4°C conditions. This resuspension process was then repeated one more time. Protein was then extracted from the final pellet using cell lysis buffer to create a mixture which was vigorously passaged to ensure maximal cellular exposure (200 µL for control flasks, 100 µL for experimental flasks). The buffer was prepared from reagents as listed in Table 2.3. The resulting suspension was transferred into 1.5 mL Eppendorf® tubes, and stored in -80°C conditions overnight to maximise lysis. Suspensions were then thawed on ice at 4°C when ready for quantification/analysis, and centrifuged at 14,000 RPM for 10 minutes at 4°C. The supernatant was then removed into separate Eppendorf® tubes as the final protein lysate.

2.6.2 General quantification by BCA assay

Quantification was performed using the PierceTM bicinchoninic acid (BCA) Protein

Assay Kit (Thermo Fisher Scientific, USA; Cat. No. 23227). A series of protein standards using a bovine serum albumin (BSA) stock solution (2 mg/mL) was prepared by continual 1:2 dilution with cell lysis buffer up until 31.25 µg/mL. 10 µL of each standard was aliquoted in duplicate in a 96-well plate, along with duplicates 10 µL aliquots of each protein lysate to be quantified. BCA working reagent was then prepared by combining BCA Reagent A (ionic solution) with BCA Reagent B (4% v/v copper sulphate) in a ratio of 50:1. The BCA working

Table 2.3 Components of cell lysis buffer.

Reagents and respective concentrations used to prepare cell lysis buffer for protein extraction. Volumes and masses are calculated to produce 100 mL of buffer stock to be stored in -80oC freezer, and when needed for lysis, to be thawed and supplemented with protease inhibitor (Roche, Switzerland) and phenylmethylsulfonyl fluoride (PMSF; Roche, Switzerland). All reagents use MilliQ as solvent except for PMSF which uses isopropyl alcohol.

Reagent Final concentration Volume/mass 100 mL storage stock EDTA, 0.5 M stock, pH=8.0 1 mM 200 µL Glycerol 10% v/v 10 mL

Na4P2O7 (sodium pyrophosphate) 20 mM 531.8 mg NaCl (sodium chloride) 100 mM 584.4 mg NaF (sodium fluoride) 1 mM 4.2 mg SDS (sodium docecyl sulfate) 0.1% v/v 100 mg Sodium deoxycholate 0.5% v/v 500 mg Tris-Cl, pH=7.4 10 mM 15.8 mg Triton X-100 1% v/v 1 mL 1 mL buffer for lysis (add to 790 µL storage stock) Protease inhibitor (1 tablet in 1.4 mL) 1:7 of tablet 200 µL PMSF, 10 mM stock 100 µM 10 µL

EDTA, Ethylenediaminetetraacetic acid; PMSF, phenylmethylsulfonyl fluoride.

reagent was then added to each well in 200 µL aliquots, and the plate then incubated at 37°C in 5% CO2 conditions for 30 minutes. The plate was then analysed using the SpectraMax ®

M2 Multi-Mode Microplate Reader (Molecular Devices, USA) at 562 nm absorbance. Blank- corrected values of the BSA solutions were then matched to a linear model, in which then protein lysate concentrations could be extrapolated. Lysates were then stored on ice at 4oC for further analysis or returned to -80°C conditions for storage.

2.6.3 Detection by western blot

2.6.3.1 Sample preparation

Lysates were allowed to thaw on ice at 4°C if not already thawed out. The equivalent volume for 10 µg of protein based on the BCA assay for each protein lysate was then aliquoted into separate Eppendorf® tubes. The same volume of Laemmeli dye (Bio-Rad, USA; Cat. No.

1610737) buffer solution (1X Laemmeli buffer to mercaptoethanol in 20:1 ratio) was then added. Each sample was mixed at 1500 RPM for 1 minute using the Eppendorf MixMate®

(Eppendorf, Germany), and then heated for 5 minutes at 95°C using the Eppendorf

ThermostatTM (Eppendorf, Germany) before being placed on ice.

2.6.3.2 Electrophoresis

The electrophoresis set-up consisted of a 4-15% mini-PROTEAN® TGXTM precast gel(s) with 50 µL wells (Bio-Rad, USA; Cat. Nos. 4561085 and 4561084) and a gel holder.

When only one gel was to be run, a dummy plate was used to form a sealed buffer dam in the holder. Otherwise, two gels used at the same time would also form the required seal. The gel holder was then placed in a Mini-PROTEAN® Tetra Vertical Electrophoresis Cell (Bio-Rad,

USA; Cat. No. 1658004).

1X running buffer was prepared by dilution of 10X tris/glycine buffer/SDS running buffer in MilliQ water. Then, it was added to the buffer dam and allowed to overflow to the

designated level of the Electrophoresis Cell, ensuring each well of the gel was washed in the process and completely submerged. The protein lysate Laemelli buffer mixture was loaded into each well by means of fine pipette at the calculated volume. Additionally, 10 µL of Precision

Plus ProteinTM KaleidoscopeTM Prestained Protein Standards (Bio-Rad, USA; Cat. No.

1610375) was loaded into at least one separate well to allow for weight identification after transfer. Electrophoresis was then run down the gel at 100 V for 1 hour.

2.6.3.3 Wet transfer

1X transfer buffer was prepared by dilution of 10X tris/glycine transfer buffer in a 7:2

MilliQ:methanol mixture. The precast gel was removed from the electrophoresis set-up and the gel itself was then released carefully and placed in 1X transfer buffer. Two sheets of filter paper were cut to a template, as well as one sheet of AmershamTM HybondTM ECL Nitrocellulose

Membrane (GE Healthcare, UK). Along with two foam pads, these were all soaked in 1X transfer buffer for 10 minutes at 4°C. The wet transfer set-up was then prepared in wet- conditions as follows: a gel cassette was opened flat, and in sequential order, one foam pad, one filter paper, the gel, the membrane, the other filter paper, the other foam pad, were placed upon the black side of the cassette. Then, the top of the stack was then gently rolled upon by a roller to remove any bubbles that may have occurred, before the cassette was closed and locked.

It was then transferred into an electrode assembly within the Electrophoresis Cell. An ice-pack was placed into the Cell as well, and then the entire Cell was filled with 1X transfer buffer to its rim. The Cell was placed in an esky surrounded by ice to further prevent overheating. The transfer was then run at 100 V for 2 hours.

2.6.3.4 Immunoblot

1X immunblot buffer was prepared with 10X tris-buffered saline into MilliQ water with

0.2% v/v Tween® 20 (Sigma-Aldrich, USA; Cat. No. P9416). The membrane was removed

from the wet transfer set-up, and protein transfer was confirmed by means of Ponceau S

Staining Solution (Sigma-Aldrich, USA; Cat. No. P7170). Visible pink bands confirmed the presence of protein, and the stain solution removed by repetitive washing with 1X immunblot buffer.

The membrane was then blocked using a skim milk blocking solution (Coles, Australia;

5% w/v in 1X immunoblot buffer) for 2 hours at room temperature on constant tilt rotation by a Stoval Belly Dancer Shaker (Marshall Scientific, USA). The membrane was then washed using 1X immunoblot buffer for 10 minutes on constant rotation three times at room temperature. Primary antibodies (and a housekeeper) were then prepared as per Table 2.4 in skim milk blocking solution and allowed to probe the membrane overnight at 4°C on constant tilt rotation.

Then, the primary antibody solution was removed, and the membrane washed using 1X immunoblot buffer for 10 minutes on constant rotation three times at room temperature. The membrane was then probed by a secondary antibody in skim milk blocking solution for 2 hours at room temperature. All tested primary antibodies were rabbit monoclonal antibodies, and the secondary antibody used was Anti-Rabbit (Polyclonal Goat) horseradish peroxidase (HRP)- conjugated antibody (Cell Signaling Technology, USA; Cat. No. P0449) at 1:1000 dilution.

The membrane then washed again three times using 1X immunoblot buffer for 10 minutes on constant rotation. Membranes were then developed using the Amersham ECL Plus

Western Blotting Detection System (GE Healthcare, UK) as per the instructions of the manufacturer. Images were obtained using the LAS-4000 (Fuji Photo Film Co. Ltd., Japan) at an initial exposure time of 5 seconds at high quality, which was the adjusted accordingly based on quality of the image obtained. Image Studio Lite (LI-COR Biosciences, Nebraska, USA) was used to perform densitometry analysis.

Table 2.4 Western blotting primary antibody list.

List of primary antibodies used in this study. As all were rabbit antibodies (Rb Ab), Anti-Rabbit (Polyclonal Goat) horseradish peroxidase (HRP)-conjugated antibody (Cell Signaling Technology, USA; Cat. No. P0449) at 1:1000 was used as the secondary antibody for all immunoblots.

Type Dilution Weight Company Cat. No. Primary antibody Marker (kDa) α-tubulin housekeeper Rb Ab 1:1000 52 Abcam ab4704 bax apoptosis Rb Ab 1:1000 21 CST 2772 bcl-2 apoptosis Rb Ab 1:1000 26 CST 2872 caspase-8, cleaved apoptosis Rb Ab 1:1000 43 CST 9496 caspase-9 apoptosis Rb Ab 1:1000 47 CST 9502 eIF4E housekeeper Rb Ab 1:1000 25 CST 2067 ɣ-H2AX* DNA damage Rb Ab 1:2000 15 CST 9718 LC3, I and II autophagy Rb Ab 1:1000 14, 16 Abcam ab48394 PARP, cleaved apoptosis Rb Ab 1:1000 88 CST 9541

*phosphorylated form.

Rb Ab, rabbit antibody; CST, Cell Signaling Technology.

2.7 In vivo techniques

All in vivo studies reported from hereon in were conducted under ethics that had been approved by the Animal Care and Ethics Committee (ACEC #18/59B, UNSW).

2.7.1 Animal handling

7-week old Balb/c nude mice (Animal Resource Centre, Perth) were the animal models used in this study. Mice were housed in a controlled environment in groups of five and were given access to ad libitum food and water. They were allowed to acclimatise to their surroundings for at least one week after arrival before any therapy was administered. Mice were weighed twice weekly to ensure health and weight status, and regularly monitored for any neurological or functional deficiencies. Once pre-defined end point was reached, or significant adverse events arose that were incompatible with the wellbeing of the mice, mice were humanely euthanised with carbon dioxide overdose by inhalation.

2.7.1.1 Nanoparticle administration

All NP formulations used in vivo were prepared initially as a stock suspension of 1000

µg/mL in PBS. The suspension was then sonicated in a sonication bath (Unisonics, Australia) at 40 kHz for a minimum of 5 minutes. Then, the solution was then vortexed using a Reax top

(Heidolph, Germany) at maximum rotation for 1 minute. The NP suspension was then diluted to the appropriate concentration to obtain the desired dosage within a total volume of 100 µL.

The final suspension was re-agitated prior to administration.

Administration route was intravenous (IV) injection. Mice were prepared by heating under heat lump for 5 minutes to ensure maximum venous dilation. We utilised 0.5 mL 29G

13 mm insulin syringes (Terumo, USA) for injection, as it provided the most first attempt successes during training. One of the two major tail veins was chosen for injection, with a distal site chosen for the first attempt. Should further attempts be required, the injection site was advanced in a proximal direction. Proximal advancement was only attempted twice before that

vein was abandoned. If this occurred, the mouse was returned to its cage, and one hour later, similar attempts were made in the other tail vein.

2.7.1.2 Radiation administration

Radiation was administered under temporarily-induced anaesthesia in the Biological

Irradiator using the ‘mice’ pre-settings. Firstly, anaesthesia was induced by 3% isoflurane in a sealed knock-down box with oxygen flow. Mice were then transferred into a specialised animal holding device with ‘guillotine-like’ head restraint positioned to ensure only the heads of the mice would be exposed to radiation. The bodies of the mice were then padded with tissue, and then covered by a lead lid. This was performed as quick as possible to avoid the anaesthesia wearing off prematurely. Radiation was then administered. After, mice were observed under heat lamp for 5 minutes to ensure no adverse effects occurred.

2.7.1.3 Chemotherapy administration

All temozolomide (Sigma-Aldrich, Australia; Cat. No. 2577) formulations used in vivo were prepared initially as a stock solution of 10 mg/mL (25 mg in 2.5 mL DMSO). The stock solution was aliquoted into 410 µL volumes into separate Eppendorf® tubes and stored in -

80oC conditions. Aliquots were thawed the day of administration.

Administration route was intraperitoneal (IP) injection. The appropriate volume to achieve a final dosage of 10 mg/kg in each mouse was drawn and injected into the peritoneum of the mice. The 10 mg/kg dose was determined by a previous mouse GBM model in our group

(ACEC #16/86A, UNSW). Mice were then observed for any signs of local bleeding after, before being returned to their cage.

2.7.2 Post-mortem organ extraction

Organ examination at pre-determined end-points was performed post-mortem, after mice were humanely euthanised with carbon dioxide overdose by inhalation.

2.7.2.1 Brain

The cranium was released by incision across the cervical spine. To improve visualisation, the skin on the superior aspect of the cranium was incised in a midline fashion.

The posterior occipital musculature was then gently scraped away, to reveal the C1 vertebra.

This was carefully incised and removed to reveal the inferior aspect of the brain stem. The inferior aspect of the skull was slowly resected in a piecemeal fashion. Once the cerebellum was visualised, an incision was made along the midline of the skull. Then, the nasal bone was incised, allowing for complete resection of the remaining skull fragments by piecemeal fashion to reveal the brain. This was slowly peeled away from the base of the skull, with cranial nerve attachments incised where appropriate. The brain was then removed wholly.

2.7.2.2 Abdominal organs

A midline laparotomy was performed from the lower abdomen to the sternum. The upper left quadrant was then exposed by a lateral incision from the centre of the first incision.

The spleen was then visualised, and the careful dissection away from the pancreas was performed to the posterior end of the spleen. The spleen was then removed. Attention was turned to the left kidney, which was located in the retroperitoneum. The renal vasculature was incised, which released the left kidney. A lateral incision towards the right of the abdomen was then performed. The intestinal organs were carefully retracted to reveal the right kidney. In a similar fashion to its left counterpart, the renal vasculature was incised, which released the right kidney. Attention was then turned to the liver in the upper portion of the peritoneum. The liver was first dissected inferiorly along the lesser curvature of the stomach to release that lobe. The oesophagus and hepatic veins were then visualised, and incised. Then, the superior aspect of the liver was gently dissected away from the upper diaphragm and posterior ribs. The hepatic vein was then identified, clamped, and incised. The liver was then removed.

2.7.3 Determination of organ nanoparticle concentration

After extraction, all organs were weighed, and their masses recorded. Then, all organs were processed in individual 15 mL falcon tubes. They were submerged in 1 mL 70% nitric acid (HNO3; Sigma-Aldrich, Cat no. 438073), except for the liver which was submerged in 2 mL of the same solution and allowed to digest overnight at room temperature. The next day, to complete the digestion, samples were warmed to 80oC in a water bath for 2 hours. They were then allowed to cool, before being diluted with MilliQ water to a total sample volume of 10 mL. These samples were then submitted to the Mark Wainwright Analytical Centre, UNSW,

Australia for processing by means of inductively-coupled plasma (ICP) mass spectrometry

(MS) using an Elan 6100 ICP-mass spectrometer (Perkin Elmer Sciex Instruments, USA) to determine the amount of REE in the sample. This was then normalised to the control and calculated as a percentage of the initial mass recorded prior to digestion. The elemental concentration was then converted to NP concentration based on stoichiometry of the REO NP.

2.7.4 Assessment of nanoparticle biocompatibility

2.7.4.1 Tolerable dose

A short-term tolerable dose of the REO NP therapy tested was first evaluated by a dose- escalation approach. Based on the literature that has used REO NPs before in healthy tumour- free mice, a dose of 5 mg/kg administered weekly is sufficient to induce asymptomatic liver dysfunction on blood tests only – there was no mortality, change in behaviour or weight loss or serious systemic symptoms.185 Thus, this dose of 5 mg/kg was categorised as a high tolerable dose, and we investigated a low dose of 1 mg/kg, and a moderate dose of 3 mg/kg first to ensure there were no overt adverse events with our particular REO NP prior to proceeding with the 5 mg/kg tolerable dose.

Firstly, an intravenous dose of 1 mg/kg was chosen to trial first with to ensure no significant side-effects were at risk with our NP, with planned escalation to doses 3 mg/kg and then 5 mg/kg if feasible (Figure 2.1). Feasibility was defined as asymptomatic progress (no

Figure 2.1 Schematic of tolerable dose determination.

Experimental design to determine tolerable dose of NP in balb/c nude mouse model. Mice (n=3 each cohort) were regularly monitored for adverse events, and concentration of NP determined by inductively-couple plasma mass spectrometry (ICP-MS).

significant physical or behavioural deficits, and ≤ 20% weight loss) after 1 week of the first dose. The design of this experiment utilised a NP regimen of one dose per week for four weeks, with regular evaluation. Once the final dose had been administered, mice were observed for 1 week further before being humanely euthanised. The brain was analysed for evidence of NP accumulation, and the liver, spleen and kidneys were evaluated for evidence of any systemic toxicity and NP accumulation.

2.7.4.2 Nanoparticle pharmacokinetics in the brain

The pharmacokinetics of the NP at highest tested tolerable dose was determined by serial measurements of REE concentration in the brain after one dose. At timepoints of 0, 1,

24, 72 and 120 hours after NP injection, mice were humanely euthanised and brain organs were extracted to be analysed by ICP-MS. Timepoint of maximum brain concentration, as well as other pharmacokinetic parameters, were determined by modelling with PKSolver 2.0 (Nanjing,

China).

2.7.4.3 Long-term safety profile

To validate the appropriateness of our NP therapy for future survival studies, the long- term safety profile was determined (Figure 2.2). However, mice were treated 6 weeks after acclimatisation to simulate the time period required for xenograft establishment as previously determined by our group.186 At this point, possible therapeutic combinations of NP (4 weekly doses at highest tested tolerable dose), radiation (one dose at 4 Gy) and chemotherapy (4 weekly doses of 10 mg/kg TMZ) were administered. Radiation and chemotherapy were administered at the corresponding timepoint of maximum brain concentration as determined in

Section 2.7.4.2. Long-term safety was evaluated by regular weekly monitoring the weight and wellbeing of the mice for a period of expected survival, which in the case of the future xenograft model, would be 50 days from first NP dose. Mice were then humanely euthanised and organs analysed as in Section 2.7.4.1.

Figure 2.2 Schematic for biocompatibility determination.

Experimental design to determine long-term safety profile of NP in balb/c nude mouse model. This was conducted in anticipation of a future GBM xenograft model which would require therapy (or therapies) to commence 6 weeks after acclimatisation, with expected survival to be 50 days after commencement of therapy. Temozolomide (TMZ) was administered at 10 mg/kg, and radiation therapy (RT) utilised a 4 Gy dose. Both TMZ and RT was administered at the timepoint of maximum brain NP concentration as determined by our earlier pharmacokinetics study. PBS was used as the negative control therapy. Mice (n=3 each cohort) were regularly monitored for adverse events, and concentration of NP determined by inductively-couple plasma mass spectrometry (ICP-MS). IV, intravenous; IP, intraperitoneal.

2.8 Statistical analyses

All data were presented as mean ± standard deviation (SD). Statistical comparisons were predominantly performed using Prism 7.0 (GraphPad, San Diego, USA). In the case of combination index (CI) for synergy assessment of NP and TMZ, analysis was performed using

CompuSyn (ComboSyn, Paramus, USA). Comparisons of measurements between multiple variables was conducted by two-way analysis of variance (ANOVA) with a Bonferroni’s post- hoc test or students t-test, where appropriate. Tukey’s post-hoc test was applied in the case of multiple comparisons within variables. Statistical significance was two-sided and set at P- value <0.05.

CHAPTER 3

3 THE THERAPEUTIC POTENTIAL OF RARE EARTH

OXIDE NANOPARTICLE THERAPY IN GLIOBLASTOMA

3.1 Introduction

The development of novel therapies in glioblastoma (GBM) is a rigorous process. Not only must the therapy be taken up by the target cells, but once there, it needs to exert a therapeutic effect within the cells. This can be directly as a cytotoxic agent, or indirectly, as an agent that interacts with other established therapies such as radiation therapy and chemotherapy.

Such agents have been described and used in the management of GBM in the past and present, but are not without their own limitations. Thus, investigating novel therapies, such as rare earth elements (REEs) as rare earth oxide (REO) nanoparticles (NPs), is an important component of

GBM research if such therapeutic potential can be effectively demonstrated either in isolation or in combination with other adjuvant therapies.

3.1.1 Exert cytotoxic effect

Various REO NPs have been documented in vitro to be toxic to cancer cells of the cervix187, lung188, liver189, and bladder190. Although REO NP induced cytotoxicity in GBM cells has not been demonstrated, is not inconceivable given these results in other cancers.106 In the case that natural cytotoxicity in GBM is observed, then REO NP therapy would represent another feasible addition to the cytotoxic armamentarium to consider when treating GBM in the future.

3.1.2 Augment radiation therapy

A component of standard GBM protocol is radiation.28 Avenues to explore augmenting the effect of radiation by non-radiation means provides the advantage of reducing the amount

of radiation required to achieve therapeutic effect. This net reduction in radiation dosage would reduce the exposure to the normal healthy brain areas traversed to reach the tumour site, and reduce also the incidence of unwanted radiation toxicity in these areas.191 Furthermore, given the risks of radiation necrosis are not negligible in the peritumoural area, peripheral radiation reduction by means of local radiation reduction would also reduce the risk of this side-effect.192

Thus, given the previously described Auger effect attributable to REEs due to their expanded

4f orbitals, REO NPs may prove to be of therapeutic benefit by augmenting radiation therapy in GBM as well and should be investigated.

3.1.3 Augment chemotherapy

Although one aim of a novel cytotoxic therapy is to replace current standard cytotoxic therapy, i.e. temozolomide (TMZ), another perspective is that a novel therapy may be able to enhance the therapeutic effect of the standard cytotoxic therapy in combination. The use of combination therapy remains experimental in GBM, with TMZ having been tested with a number of cytotoxic agents in the past, including bevacizumab193, irinotecan194 and everolimus195. The rationale behind potential cytotoxic synergy is that both agents trigger a set of events which leads to greater facilitation of cytotoxicity by the other agent.196 Similar to radiation, chemosensitisation would prove advantageous from a clinical perspective as it would lead to a reduction in systemic exposure of both cytotoxic agents. Given the potential for REO

NPs to penetrate cells and induce a cellular effect, the possibility that they may also enhance the cytotoxicity of TMZ, and vice versa, should be investigated further.

3.1.4 Objective and aims

The objective of this chapter was to evaluate and identify possible avenues for therapeutic application of novel REO NPs in GBM in vitro. Specifically, the aims of this chapter were:

1. To characterise basic properties of novel REO NPs and GBM cell lines.

2. To establish whether or not REO NPs are taken up by GBM cells.

3. To determine if there is a natural cytotoxic effect of REO NPs.

4. To investigate interaction between specific REO NPs and radiation therapy.

5. To investigate interaction between specific REO NPs and TMZ.

3.2 Results

3.2.1 Nanoparticle characterisation

The four REO selected for investigation as NPs were CeO2, Gd2O3, La2O3 and Nd2O3.

Due to their location in the periodic table, these REEs had the most unpaired electrons of the

REEs available, and were chosen as such to maximise the possibility of any therapeutic effect caused by this electronic arrangement. The properties of the REO NPs are summarised in Table

3.1. Size and shape were evaluated using scanning electron microscopy (SEM) and shown in

Figure 3.1A. The average diameters of CeO2, Gd2O3, La2O3 and Nd2O3 NPs were 79.3 ± 28.6,

107.7 ± 28.6, 68.4 ± 23.2 and 94.2 ± 26.1 nm respectively, with CeO2, La2O3 and Nd2O3 NPs being primarily round in shape, and Gd2O3 NP rod-shaped. The zeta potentials of CeO2, Gd2O3,

La2O3 and Nd2O3 NPs were all positive.

3.2.2 Cell line characterisation

Six GBM cell lines were investigated: two immortalised cell lines (ICLs) U87 and

M059, and four patient-derived cell lines (PDCLs) RN1, GBML1, G53 and WK1 (Figure

3.1B). All cell lines were stellate in shape, except for M059 and RN1 which were spindle in shape, and this was reflected in process number (Table 3.2). Overall, U87 (doubling time, 42.0 hrs) was the fastest growing cell line, and RN1 (doubling time, 63.8 hrs) was the slowest growing cell line.

3.2.3 Cellular uptake of nanoparticle

Nanoparticle (NP) uptake by the cells was confirmed utilising transmission electron microscopy (TEM) as shown in Figure 3.2. First, the potential organelle of localisation to the intracellular endosome was located in representative GBM PDCL RN1 without exposure to

NPs. Endosomes were successfully observed, and were vacant. Then, after 1 hour exposure to

Table 3.1 Characteristics of REOs investigated.

Characterisation of rare earth oxide (REO) nanoparticles (NPs) investigated in this study. Mean size determined by averaging 50 representative particles observed on scanning electron microscopy (SEM). Values presented as mean ± SD where applicable.

Zeta potential K REO NP Molecular mass (g/mol) Mean size (nm) Shape sp (mV)

CeO2 172.12 79.3 ± 28.6 Round +14.2 -23 Gd2O3 362.50 107.7 ± 28.6 Rod +17.0 1.8 x10

La2O3 325.81 68.4 ± 23.2 Round +14.8

Nd2O3 336.48 94.2 ± 26.1 Round +23.7

Ksp, solubility product constant.

Figure 3.1 Representative images of NPs and cells.

Representative images A. obtained by scanning electron microscopy (SEM) of all rare earth oxide (REO) nanoparticles (NPs) investigated in this study and B. by light microscopy of all cell lines investigated. These include two immortalised cell lines (ICLs) U87 and M059, and four patient-derived cell lines (PDCLs) RN1, GBML1, G53 and WK1. Images obtained by light microscopy at 20X magnification.

Table 3.2 Characteristics of cells.

Characterisation of GBM cell lines used to investigate effect of rare earth oxide (REO) nanoparticles (NPs). Process number was determined by averaging 50 representative cells observed on light microscopy at 40X. Doubling time was determined by means of MTS assay at 24, 48, 72 and 96 hours after seeding in n=3 replicates and calculated by exponential modelling. Values presented as mean ± SD where applicable.

Cell line Type Shape Process number Doubling time (hrs) U87 ICL stellate 2.4 ± 0.02 42.0 ± 0.9 M059 ICL spindle 2.0 ± 0.05 53.3 ± 1.9 RN1 PDCL spindle 2.1 ± 0.01 63.8 ± 1.5 GBML1 PDCL stellate 2.5 ± 0.03 61.8 ± 3.9 G53 PDCL stellate 2.6 ± 0.04 42.5 ± 1.8 WK1 PDCL stellate 2.6 ± 0.04 55.1 ± 1.7

Figure 3.2 Transmission electron microscopy images of NP in cell.

Representative transmission electron microscopy (TEM) images of GBM patient derived cell line (PDCL) RN1. Images A and B have not been exposed to La2O3 nanoparticle (NP), and vacant endosomes have been identified. Images C to E have been exposed to La2O3 NP for 1 hour. Image C has identified the NPs in both early (lower box) and late (upper box) endosomes in the cell. Image D demonstrates the ability of these NPs to aggregate within endosomes to a high degree. Image E shows NPs making contact with the surface of the cell, prior to endocytosis. Black outlines represent a magnified section of the image in the corner of each panel for clarity.

representative REO NP formulated from La2O3, TEM demonstrated the presence of NPs in endosomes, which were seen at both early and late stage endosomes. The ability for NPs to aggregate within endosomes was also demonstrated.

3.2.4 Cytotoxicity of nanoparticles

The potential for these NPs to cause cytotoxic effect to the GBM cells was the first therapeutic effect investigated. Firstly, the of timepoint 72 hours following NP administration to measure cytotoxicity was justified by observing cytotoxic trends in PDCL RN1 did not appear to change from 72 to 96 hours across all REO NPs (Figure 3.3). Then, all four PDCLs were exposed to all REO NPs at concentrations ranging for 0-100 µg/mL and cell viability measured after 72 hours (Figure 3.4).

In the ICLs, La2O3 was observed to be the most cytotoxic, with mean cell viability of

8.8% and 53.5% in U87 and M059 respectively at the 100 µg/mL concentration. CeO2 was the least cytotoxic, with mean cell viability of 103.7% and 90.4% in U87 and M059 respectively at the 100 µg/mL concentration. In terms of general trends, there were significant cytotoxic trends in U87 observed with Gd2O3, La2O3, and Nd2O3, and similarly in M059. There was no significant trend with CeO2 in either ICL. IC50 values could be calculated for La2O3 and Nd2O3

NPs only, with the IC50(La2O3) < IC50(Nd2O3) in both ICLs (Table 3.3).

In the PDCLs, La2O3 was observed to be the most cytotoxic for all cell lines, with mean cell viability of 38.7%, 36.6%, 49.1% and 61.6% in RN1, GBML1, G53 and WK1 respectively at the 100 µg/mL concentration. CeO2 was the least cytotoxic, with mean cell viability of 92.2%,

94.6%, 82.6% and 96.6% in RN1, GBML1, G53 and WK1 respectively at 100 µg/mL dose.

In terms of general trends, there were significant cytotoxic trends observed with the same three NPs Gd2O3, La2O3, and Nd2O3, that had effect in the ICLs in all four PDCLs;

Interestingly, CeO2 had a significant cytotoxic trend in GBML1 PDCL as well. IC50 values

Figure 3.3 Optimal time point for cytotoxicity in vitro.

Cytotoxicity of all four rare earth oxide (REO) nanoparticles (NPs) in PDCL RN1 after 24, 48, 72 and 96 hours across a concentration range of 0-100 µg/mL. Experiments were performed using MTS assay to assess cell viability in n=3 replicates, in n=3 repeats. Values presented as mean ± SD.

Figure 3.4 Primary cytotoxicity results.

Cytotoxicity of all four rare earth oxide (REO) nanoparticles (NPs) in all cell lines after 72 hours across a concentration range of 0-100 µg/mL. Experiments performed using MTS assay to assess cell viability in n=3 replicates, in n=3 repeats. Linear regression for statistical difference compared to null (dotted line at 100% cell viability) was used to identify cytotoxic trends across concentrations, with P<0.05. Values presented as mean ± SD.

were calculable for Gd2O3, La2O3 and Nd2O3 NPs in the four PDCLs, with the general trend being IC50(La2O3) < IC50(Gd2O3) < IC50(Nd2O3) where applicable (Table 3.3).

3.2.5 Interaction between nanoparticles and radiation

To investigate the potential for radiosensitisation by REO NPs, we analysed the ability for radiation therapy (RT) to affect cell proliferation by means of clonogenic assay. This relies on the inherent ability of cells to form colony-forming units (CFUs), over a period of time, typically 14 days.197 Thus, the two ICLs and four PDCLs were first tested for colony-forming capabilities and optimal seeding densities: U87, Mo59, RN1, G53 and WK1 formed CFUs and were thus used for further investigation; GBML1 did not form CFUs was excluded from further investigation (Supplementary Figure 3.1). Their optimal seeding densities that allowed for greatest number of CFUs growth whilst remaining distinguishable from adjacent colonies are listed in Supplementary Table 3.1.

Once optimal seeding was determined, optimal radiation dose for investigation was ascertained for each cell line by means of exposing cells to a range of radiation from 0-8 Gy and assessing for dose that induced moderate decrease in CFUs after 14 days (Supplementary

Figure 3.2). Optimal radiation doses ranged from 1-4 Gy, and specific cell line doses are listed in Supplementary Table 3.1.

The two ICLs were then used to validate the potential for radiosensitisation by any of the REO NPs at the optimal radiation dosage (Figure 3.5). Using all REO NPs at a representative concentration of 50 µg/mL, it was observed that the combination of Gd2O3

NP+RT significantly reduced the relative number of CFUs after radiation in both U87 and

M059, and La2O3 NP similarly in U87 only, when compared to RT only. All other REO NPs did not significantly affect the relative number of CFUs in either ICL after RT.

Table 3.3 IC50 values at 3 days.

Calculated IC50 for all rare earth oxide (REO) nanoparticles (NPs) tested across all cell lines at 72 hours. Standardised cell viability was determined by MTS assay and were derived from n=3 replicates, and n=3 repeats. IC50 was calculated by exponential modelling. Values presented as mean ± SD where applicable, and when not available (NA), this was due to data not converging, and thus IC50 could not be calculated.

IC50 (µg/mL; 72 hours)

Cell type Cell line CeO2 Gd2O3 La2O3 Nd2O3 U87 NA NA 26.2 ±5.1 53.9 ±4.2 ICL M059 NA NA 113.2 ±32.9 123.2 ±19.6 RN1 NA 77.6 ±6.1 53.6 ±2.9 147.3 ±15.6 GBML1 NA NA 14.0 ±2.5 60.1 ±2.5 PDCL G53 NA 478.9 ± 390.2 86.5 ±43.7 NA WK1 NA 552.4 ±217.4 170.6 ±57.9 NA

Figure 3.5 Radiosensitisation in ICLs.

Validation in immortalised cell lines (ICLs) U87 and M059 that potential change in colony formation unit (CFU) after addition of rare earth oxide (REO) nanoparticles (NPs), measured by clonogenic assay. Cells were seeded at optimised seeding density, then the next day, 50 µg/mL of REO NP was added, and 24 hours later, cells were exposed to optimised radiation therapy (RT) dose (3 Gy for both ICLs). CFUs were stained and counted 14 days after initial seeding. Experiments were performed using in n=3 replicates, in n=2-3 repeats. Student t-test were performed to detect difference between the change in CFU following the addition of RT to control vs NP. Statistical significance was set at P<0.05 (*), Values presented as mean ± SD.

The potential for radiosensitisation was then more extensively investigated in PDCLs.

Using RN1 as a representative PDCL, all 4 REO NPs were tested at concentrations ranging from 0-50 µg/mL at optimal radiation dose. The results showed that significant reduction in relative number of CFUs occurred in NP+RT combinations with Gd2O3, La2O3 and Nd2O3 compared to RT only (Figure 3.6A). Specifically, Gd2O3 at concentration of 40 µg/mL; La2O3 at concentrations of 10, 20 and 40 µg/mL; and Nd2O3 at concentrations of 30 and 40 µg/mL.

Of great interest is that a significant increase in relative number of CFUs was observed in the combination of CeO2 NP+RT compared to RT only, suggesting a radioprotective effect at concentrations 40 and 50 µg/mL.

In order to generalise the findings from RN1, the radiosensitisation ability of the REO

NPs in PDCLs was validated by investigating representative La2O3 NP in the other two colony- forming PDCLs, G53 and WK1. The La2O3 NP was chosen in particular as it was the REO NP with the most cytotoxic nature shown in Section 3.2.4, as well as evidence of radiosensitisation ability in U87 and RN1. In a similar design to the RN1 investigation, significant reduction in relative number of CFUs in the combination of La2O3 NP+RT compared to RT only in both

PDCLs was observed (Figure 3.6B). Specifically, in G53 these reductions were observed at concentrations 30, 40 and 50 µg/mL; and in WK1 at concentrations 20 and 30 µg/ mL

(Supplementary Table 3.3).

Figure 3.6 Radiosensitisation in PDCLs.

Investigation of radiosensitisation by rare earth oxide (REO) nanoparticles (NPs) in patient-derived cell lines (PDCLs) in A. RN1 with all four REO NPs and B. G53 and WK1 by La2O3 NP only by clonogenic assay. Cells were seeded at optimised seeding density. The next day, concentrations ranging from 0-50 µg/mL of REO NP were added, and 24 hours later, cells were exposed to optimised radiation dose (4, 4 and 1 Gy for RN1, G53 and WK1 respectively). Colony forming units (CFUs) were stained and counted 14 days after initial seeding. Experiments were performed using in n=3 replicates, in n=2-3 repeats. Data presented as relative percentage of CFUs with NP addition versus CFUs without NP, with no effect at 100% (dotted line). Student t-test were performed to detect difference between the change in CFU following the addition of RT to control vs NP. Statistical significance was set at P<0.05 (*), Values presented as mean ± SD.

3.2.6 Interaction between nanoparticles and chemotherapy

The potential interaction between these NPs and cytotoxic chemotherapy in PDCLs was investigated to evaluate if the two therapies were synergistic, additive or antagonistic in nature when co-administered. Specifically, TMZ was chosen as the standard chemotherapy agent for GBM treatment currently. Given the most significant cytotoxic effect observed in isolation by the REO NP of La2O3 as seen in Section 3.2.4, it was decided to use this NP as the representative REO NP for investigation of interaction with TMZ on cell viability.

Firstly, the IC50 of La2O3 NP and TMZ were optimised across all four PDCLs over a period of 8 days (Table 3.4). GBML1 was the most sensitive cell line to both NP (IC50, 12.6

µg/mL) and TMZ (IC50, 8.8 µM). WK1 was the least sensitive cell line to NP (IC50, 28.9

µg/mL), and G53 the least sensitive cell line to TMZ (IC50, 435.9 µM). Then, concentrations of NP and TMZ ranging from 0.25-2.00X IC50 were administered separately, and in fixed 1:1 ratio together, to all PDCLs (Figure 3.7). There was a significant decrease in relative cell viability observed in NP+TMZ combination when compared to TMZ only in all PDCLs

(Supplementary Table 3.4).

These results were successfully modelled by combination analysis with Compusyn software (ComboSyn, Paramus, USA), with mean R2 values ranging from 0.959-0.989 indicating that the observed therapeutic effects could be successfully modelled and therefore analysed (Figure 3.8). Combination indices (CIs) were then modelled per the Chou and Talalay method (Figure 3.9A). At 97% fraction affected (Fa), mean CIs for RN1, G53, GBML1, and

WK1 were all <0.9, indicating synergy. Furthermore, combination dose-reduction indices

(DRIs) were obtained (Figure 3.9B). At Fa=97%, mean DRIs for La2O3 NP in RN1, G53,

GBML1 and WK1 were all >1, and mean DRIs for TMZ similarly all >1, indicating dose- reduction for both La2O3 NP and TMZ by Fa =97%.

Table 3.4 IC50 values at 8 days.

Calculated IC50 values for La2O3 nanoparticle (NP) and temozolomide (TMZ) for each patient derived cell line (PDCL) measured at 8 days after administration using MTS assay to assess cell viability in n=3 replicates, and n=3 repeats. Values obtained by means of exponential modelling, and are presented mean ±SD.

IC50 (8 days)

Cell line La2O3 NP (µg/mL) TMZ (µM) RN1 26.1± 1.9 293.6 ± 1.0 GBML1 12.6 ± 0.5 8.8 ± 0.4 G53 26.9 ± 1.5 435.9 ± 1.0 WK1 28.9± 2.1 202.2 ± 1.0

Figure 3.7 Interaction between NP and TMZ.

Investigation of temozolomide (TMZ) sensitisation by La2O3 nanoparticle (NP) in all four PDCLs assessed by cell viability MTS assay. Cells were seeded at optimised seeding density. The next day, concentrations ranging from 0.25-2.00X IC50 (8 days) of both TMZ and NP were added individually and in combination (1:1 ratio), and assessed 8 days later. Experiments were performed in n=3 replicates, in n=2-3 repeats. Student t-test were performed to detect difference between the change in cell viability following the addition of TMZ to control vs NP. Statistical significance was set at P<0.05 (*). Values presented as mean ±SD.

Figure 3.8 Congruency between observed and theoretical modelled therapeutic effect.

Modelling of four PDCLs against Compusyn software for combination analysis. Cell viability was determined by means of MTS assay were converted to the affected fraction (Fa) metric for each individual and combined (Combo) outcome of La2O3 NP (La) and temozolomide (TMZ). This was performed for RN1 (mean R2, 0.977), GBML1 (mean R2, 0.959), G53 (mean R2, 0.989) and WK1 (mean R2, 0.978). Experiments were performed in n=2 replicates, in n=3 repeats and data for the average of each repeat is presented individually.

Figure 3.9 Synergy between NP and TMZ.

Combination analysis of representative La2O3 nanoparticle (NP) and temozolomide (TMZ) in all four patient-derived cell lines (PDCLs) based on n=3 replicates, in n=2 repeats of individual and combination cell viability data obtained by MTS assay. A. Combination indices (CIs) were modelled across affected fraction (Fa) 5%-97%. The major dotted line at CI=1 represents the theoretical absence of interaction (addition), and the minor dotted lines from CI=0.9-1.1 delineate the more practical region for additive interaction. CI<0.9 indicates synergism, and CI>1.1 indicates antagonism at the given Fa. B. Dose reduction indices (DRIs) were modelled across affected fraction (Fa) 5%-97%. DRI describes the number-fold reduction of the specific therapy in the presence of the other therapy in order to achieve the respective Fa. The dotted line at DRI=1 represents null reduction in drug dose. DRI<1 indicates that less therapy is required, and DRI>1 indicates that more is required to achieve the respective Fa. Values presented as mean ±SD.

3.3 Discussion

The intention of this chapter was to identify if any therapeutic anti-GBM modalities could be observed with the selected REO NPs in immortalised and patient-derived GBM cell lines once uptake was confirmed. Firstly, successful infiltration of a representative NP into representative cells was demonstrated. Then, by means of various assays, the ability of particular REO NPs to exert 1) a cytotoxic effect, 2) a radiosensitisation effect and 3) a chemosensitisation effect across different GBM cell lines was shown. These results are encouraging, and support the greater investigation of how exactly a novel REO NP therapy could be evaluated further for translational potential.

3.3.1 REO NPs enter the cell and localise to the endosome complex

Although often presumed, observation that a therapy is able to infiltrate the cell is the most elemental feature of any investigation. TEM imaging demonstrated that representative

REO NPs not only can enter the GBM cell, but was able to show that they localised and accumulated in endosome complexes.

The endosome complex of a cell regulates the intracellular trafficking of internalised materials into a cell.198 The first stage of the endocytic process is the formation of the early endosome (EE), which serves as the major sorting station of the cell. If the material is destined for the lysosome or autophagosome, the vesicle progresses into a late endosome (LE) within the cell for final delivery to that target.199 Given the NPs were observed in both EE and LE based on morphological examination, it is likely that this is indeed the intracellular path of the novel REO NP therapy as well. This finding is supported by other NP studies200-203 that report an EE-to-LE transition, with final destination to the lysosome and/or autophagosome in both normal and cancer cells.

3.3.2 REO NPs are cytotoxic to GBM cells

The REO NPs demonstrated a wide range of cytotoxicity in the GBM cell lines. Notably,

IC50 values could be calculated for La2O3 NP in all cell lines, which were the lowest of all four

NPs when calculable, whereas the CeO2 NP was benign in most tested conditions. It is this difference between all REO NPs tested with respect to their cytotoxic effect upon GBM cells that highlights while the composition of all NPs were all REOs, this does not necessarily translate to the same cytotoxic potential, and each NP should be evaluated separately. These findings echo the general sentiment of the REE literature, with lanthanum-based therapies being proven to be widely cytotoxic across a number of cancer cell lines204,205, and cerium- based therapies used for their benign nature in order for them to persist intracellularly in cancer cells206,207.

Whether or not trends can be predicted based on physical properties of the NPs can only be speculated at this point. Despite the commercial definition of NPs ≤100 nm in diameter,

SEM imaging indicated that the actual size ranged widely across all NPs within this definition.

Furthermore, shape and zeta-potential also were detected to differ between the NPs.

Collectively, given these physical properties of size, shape and charge can all affect how effectively the NP can enter the cell in order to exert an effect, it is possible the variation in cytotoxicity between our NPs is influenced to a particular degree by this heterogeneity.86,91,92,98

208 Furthermore, Lim et al. observed greater in vitro cytotoxicity with smaller La2O3 NPs compared to larger ones, indicating that size is an important factor to consider. Of note amongst all tested NPs, La2O3 NPs were both the smallest in size and most cytotoxic.

It is even possible that the choice in REE based on chemical properties may also similarly affect these cytotoxic outcomes, for example, the La3+ ionic radii is the closest in size to that of Ca2+, an integral component of the intracellular machinery. If Ca2+-related processes

are disrupted in any way, then both apoptosis and autophagy are possible outcomes.209

Assumptions that intrinsic REE properties in the REO can predict effect must be tempered as in the REO formulation, the majority of these elements remain stably inert, and ionic dissociation is minimal.210 Nonetheless, given that a threshold for ionic concentration of REE to exert a cytotoxic effect has never been established, this remains a plausible consideration until thorough evaluation can be performed.

In this study, different cell lines responded differently to the same NPs, and any expectation of a homogeneous response in all GBM cell lines would be unreasonable. Variance in vulnerability to cytotoxic stimuli between different GBM cells lines has been documented previously.211,212 A possible explanation for this is that GBM cancers commonly possess mutations that are expressed at varying degrees, such as the TP53 gene responsible for regulating the p53 protein.213 In this example, the p53 protein is important in cellular recovery from toxic stimulus such as ROS, and arguably, calcium pathway disruption. This protein acts as a transcription factor to arrest growth at the G1/S regulation point in the cell cycle when

DNA is damaged by toxic stimulus.214 Then p53 is able to activate DNA repair mechanisms before recommencing the cycle. What remains under investigation is quantifying the genotypic and phenotypic variance of TP53 expression of the cell lines used in this study. If theoretically the expression of TP53, or any other gene for that matter involved in regulatory GBM cell function, differs between GBM cell lines, the difference would then manifest as different repair capacities between cell lines to cytotoxic insult. This would be best implicated currently by the differences in doubling time calculated for all cell lines, which has been used as a surrogate for predicting therapeutic response in cancer cells.215

Finally, a challenge worth highlighting in interpreting the translational potential of these cytotoxic trends is the perception that these NPs could be comparable to other cytotoxic agents with respect to their pharmacodynamics and pharmacokinetics. Binary REOs as used in

this study are extremely stable, and largely insoluble in aqueous solution.210 This distinguishes the experimental design for evaluating NP therapy from the more conventional soluble agent therapies by a number of aspects; the use of different concentration units (w/v rather than n/v), the need to homogenise suspension prior to administration, and the understanding that this agent is not metabolised or actively broken down once entering or leaving a cell. Collectively, these factors weaken the applicability of the more traditional IC50 parameter. Thus, it is of great promise that IC50 values for many of the tested REO NPs could in fact be calculated, which stands as testament to the therapeutic potential of cytotoxicity with this novel therapy, in particular that of La2O3 NPs.

3.3.3 REO NPs enhance radiation therapy in GBM cells

The ability for particular REEs to augment the effect of electromagnetic radiation presents another attractive reason to consider REO NPs as a novel anti-GBM therapy. In the presence of Gd2O3, La2O3, and Nd2O3 NPs, radiation reduced the ability of all GBM cell lines tested to form colonies significantly more than when the NPs were absent. It is encouraging to observe this radiosensitisation trend in this study, as this is what much of the literature investigating clinical translation of REE-based therapies in cancer has been primarily based on, including these three specific elements.216,217

Gadolinium-based NPs are the most widely studied in GBM, and have been shown to act as effective ionising radiosensitizers, as well as under other different types of irradiation such as neutron therapy and hadron therapy.218 The radiosensitisation findings of this study are in agreement with previous studies. Analysing the in vitro effects of Gd2O3-based NP in U87,

Mowat et al.149 successfully demonstrated this radiosensitisation phenomenon by reporting a significant decrease in cell survival at radiation energies similar to those used in routine X-ray therapy in the presence of NPs. In theory, this would translate to clinical studies aimed at

improving GBM survival. While the supporting evidence is not extensive, it is promising to note Le Duc et al.84 did demonstrate significantly longer survival times in GBM 9LGS- inoculated mice after radiation with Gd2O3-based NPs.

Less extensively reported are lanthanum-based NPs, which while no reports in GBM have been published, have been shown to successfully enhance ionizing radiation in prostate cancer.219 This particular NP was investigated due to given its significant cytotoxic properties, and found that this effect is the case across all the tested GBM cell lines.

Interestingly, there appears to be no reports of neodymium-based NPs causing radiosensitisation in any form of cancer. This however is more likely to be due to a lack of interest and investigation rather than an absence of the natural phenomenon given the shared electronic features across the REEs. Thus, not only does the results of this Chapter provide the first evidence that REO NPs can confer a radiosensitisation effect in GBM cells, it also describes this for neodymium-based NPs for the first time in cancer cells.

Finally, it is noted that CeO2 NPs exposed to radiation responded differently to that of the other three REO NPs, as there was either a negligible or radioprotective trend across the tested cell lines. This result may not be surprisingly unique to cerium as it is the only REE that possesses two oxidation states (+III and +IV) which can facilitate an antioxidant protective system within a cell.220 It is this antioxidant property that is posited to quench oxidative stress induced by radiation, the main mechanism by which radiation affects a cell.221 Although this property is highly intriguing, its clinical application probably lies outside the realm of a cytotoxic-based therapy for GBM.222 Nonetheless, these findings are the first to validate such potential in GBM cells, and strengthens the movement towards separating cerium-based NPs from other REE-based NPs when considering anticancer applications.

3.3.4 Lanthanum oxide NPs enhance temozolomide in GBM cells

Given the significant cytotoxicity of La2O3 NPs as previously observed, it was decided to investigate if its cytotoxic nature was interactive with the established cytotoxic chemotherapy agent TMZ. Across all PDCL, the interaction between these agents was synergistic. Even if one was to doubt the general delineation of synergy (CI<1) from antagonism (CI>1) given that the neutral additive CI=1.00 value is largely theoretical, lowering the threshold to <0.90 in all tested cell lines at therapeutic fraction affected values still indicated synergy at all moderate to high doses.196 This further distinguishes the robustness of the findings.

Clinically, the attraction of this synergy finding is illustrated by the DRI metric, where for both agents, NP and TMZ, there was less required to achieve the intended fraction affected.

Translation of this into practice would result in less amounts of each cytotoxic agent required, which would, theoretically at least, limit the side-effect profiles of these individual agents.

There is very little reported in the literature regarding the potential for REO NPs, to augment the cytotoxic effect of TMZ in GBM, or any cancer for that matter. Most of the current reports on TMZ enhancement by NP involve increased or improved delivery by conjugation with other cytotoxic agents rather than using the NPs as a source of cytotoxic synergy

223,224 themselves. However, as shown, if the La2O3 NP exerts its own cytotoxic effect, then the potential for synergy with TMZ is not unreasonable. A number of established cytotoxic agents have been shown to synergize with TMZ in GBM, such as paclitaxel224 and decitabine225. The order of which agent plays the faciliatory role to allow for synergy between our La2O3 NP and

TMZ will prove useful in the future in designing the optimal administrative schedule to investigate further the translational potential of this finding.

3.3.5 General limitations

In the determination of therapeutic potential of REO NPs in GBM, inherent natural cytotoxicity has been demonstrated, as well as enhancing interactions with radiation therapy and chemotherapy are possible. However, it is difficult to declare these findings representative of the entire REE family, given the differences observed alone in the set of 4 REOs tested, with

CeO2 trending away from the other NPs as both non-cytotoxic and potentially radioprotective.

Furthermore, given the commercial nature by which these NPs were obtained, it is possible that lack of control for uniformity in terms of physical characteristics may have also influenced the differences between REO NPs observed in this chapter. Ideally, NPs of the same shape and size, would have been used to alleviate this concern, and attribute with greater validity of any different results to the choice of REEs themselves.

As with most in vitro cancer studies, complete generalisation of the results is not possible. Ultimately, every GBM presents a unique anatomical, biological and molecular entity, and whether or not these findings with the REO NPs can be replicated in all GBM is unclear.226

One particular concern is biological heterogeneity, which can be be demonstrated by the rsults from the radiosensitization studies. Although all ICLs and PDCLs showed some evidence of radiosensitization, its trend was not linear. That is to say, there were concentrations between concentrations of radiosensitization that failed to show evidence of such sensitization. This may suggest that the assay at hand may not have been sensitive enough to detect all possible changes, or that in fact there was heterogeneity within the study design. Indeed, the relative nature of comparisons after cytotoxic NP therapy may have contributed to the disappearance of significance at higher doses which would render radiosensitisation effects difficult to detect.

Ultimately, these findings and other biological concerns can only be better accounted for by greater number of biological replicates in the future, as well as study across a wider range of therapy doses.

Attempts were made to improve the generalisability by incorporating both ICLs and PDCLs to look for effect. It is thus promising that the REO NPs produced largely consistent trends across more than one GBM PDCL, and all significant results in the case of La2O3 NPs. Nonetheless, the fact that these cell lines most likely do not represent the complete gamut of GBM possibilities is acknowledged, and the potential for resistance or refraction to this novel therapy by as-of-yet unidentified GBM subgroups cannot be discounted.

3.4 Conclusions

I conducted in vitro experiments using GBM cell lines, both ICL and PDCL, to determine if therapeutic potential of REO NPs existed. Firstly, I showed that these NPs can penetrate GBM cells. Then, I was able to extensively demonstrate within the chosen NPs that

REOs can exert a significantly cytotoxic effect on cell viability across multiple cell lines.

Additionally, the potential for interaction with both radiation therapy and chemotherapy as sensitisers to increase cell death was validated, with La2O3 in particular. Collectively, these properties demonstrate the promise of REO NPs as a novel therapy for GBM, and justify further molecular and in vivo experiments to elucidate if indeed this therapy can be translated to survival benefit in GBM.

3.5 Supplementary

3.5.1 Supplementary Tables

Supplementary Table 3.1

Optimal seeding density and radiation dose of all colony-forming cell lines in a 6-well plate setting for clonogenic assay. For seeding optimisation, cells were seeded at densities ranging from 100-10000 cells/well. For radiation optimisation, cells were exposed to radiation one day after seeding, with doses ranging from 0-8 Gy. All CFUs were stained and counted 14 days after initial seeding.

Cell line Density (per well) Radiation dose (Gy) U87 500 3 M059 750 3 RN1 2000 4 G53 1000 4 WK1 6000 1

Supplementary Table 3.2

Summary of relative colony forming units (CFU) counts comparing the addition of radiation therapy (RT) in both the control and rare earth oxide (REO) nanoparticles (NP) setting as derived from clonogenic assays in n=3 replicates, in n=2-3 repeats. Odds ratio (OR) represents the overall change in CFU count due to the addition of NP to cells. NP and RT administered at 50 µg/mL and 3 Gy respectively for both cell lines. Statistical difference was detected by student t-test, with significance set at P<0.05 and those P-values in bold. Values are presented as mean ± SD.

Relative CFU count (%) Cell line REO NP RT only vs Control NP+RT vs NP only Odds ratio P-value

CeO2 54.6 ±11.9 1.09 0.69

Gd2O3 24.6 ±5.0 0.49 0.049 U87 50.0 ±14.4 La2O3 2.4 ±4.1 0.05 <0.01

Nd2O3 5.6 ±9.6 0.11 0.01

CeO2 56.6 ±12.2 1.13 0.52

Gd2O3 27.5 ±5.0 0.55 0.048 M059 50.3 ±8.7 La2O3 34.7 ±8.7 0.69 0.15

Nd2O3 58.4 ±5.4 1.16 0.32

Supplementary Table 3.3

Summary of relative colony forming unit (CFU) counts comparing the addition of radiation therapy (RT) in both the control and rare earth oxide (REO) nanoparticles (NPs) setting as derived from clonogenic assays in n=3 replicates, in n=2-3 repeats. Odds ratio (OR) represents the overall change in CFU count due to the addition of optimised radiation therapy (RT) (4, 4 and 1 Gy for RN1, G53 and WK1 respectively). Student t-test were performed to detect difference between the change in CFU following the addition of RT to control vs NP. Statistical significance set at P<0.05, with corresponding P-values in bold. Values are presented as mean ± SD.

Relative CFU count (%) Concentration RT only vs NP+RT vs NP Cell line REO NP Odds ratio P-value (µg/mL) Control only 10 29.7 ±6.5 1.20 0.25 20 30.9 ±7.9 1.25 0.24

CeO2 30 24.2 ±2.7 29.5 ±6.9 1.19 0.29 40 36.4 ±7.4 1.47 0.04 50 36.7 ±6.5 1.49 0.03 10 25.7 ±2.3 0.94 0.43 20 22.4 ±5.9 0.82 0.24

Gd2O3 30 27.4 ±2.0 19.3 ±5.4 0.71 0.07 40 16.9 ±3.7 0.62 0.01 50 21.1 ±4.5 0.77 0.10 RN1 10 15.1 ±1.9 0.76 0.01 20 11.5 ±2.1 0.58 <0.01

La2O3 30 19.9 ±0.6 20.1 ±7.1 1.00 0.97 40 7.5 ±7.1 0.38 0.04 50 19.5 ±20.0 0.98 0.97 10 15.0 ±4.1 0.90 0.57 20 16.0 ±5.7 0.96 0.87

Nd2O3 30 16.6 ±1.7 7.9 ±1.6 0.48 <0.01 40 7.6 ±3.3 0.46 0.01 50 10.8 ±9.5 0.65 0.36 10 50.7 ±3.3 0.87 0.07 20 52.9 ±4.3 0.91 0.20

G53 La2O3 30 58.3 ±4.2 44.7 ±5.3 0.77 0.03 40 49.0±1.5 0.84 0.02 50 44.0 ±5.9 0.75 0.03 10 33.2 ±10.6 0.80 0.33 20 22.0 ±1.2 0.53 0.01

WK1 La2O3 30 41.6 ±7.6 17.3 ±3.9 0.41 0.01 40 54.0 ±39.9 1.30 0.63 50 22.2 ±19.2 0.53 0.18

Supplementary Table 3.4

Summary of relative cell viability of all four patient-derived cell lines (PDCLs) comparing the addition of temozolomide (TMZ) in both the control and La2O3 nanoparticle (NP) setting based on MTS assay in n=3 replicates, in n=3 repeats. Odds ratio (OR) represents change in viability due to the addition of TMZ. Student t-test was performed to detect difference between the change in cell viability following the addition of TMZ to control vs NP. Statistical significance set at P<0.05 with P-values in bold. Values presented as mean ±SD.

Relative cell viability (%) Concentration TMZ only vs NP+TMZ (1:1) vs Cell line Odds ratio P-value (X IC50 units) Control NP only 0.25 92.9 ±2.2 89.0 ±2.9 0.96 0.72 0.50 78.7 ±2.2 79.4 ±2.4 1.00 0.31 0.75 70.3 ±2.3 73.8 ±3.8 1.05 0.23 1.00 61.2 ±2.3 49.7 ±4.0 0.81 0.10 RN1 1.25 50.8 ±1.6 37.7 ± 4.4 0.74 0.01 1.50 42.6 ±2.1 32.3 ±7.0 0.76 0.02 1.75 34.9 ±4.6 25.1 ±3.2 0.72 0.049 2.00 32.0 ±2.3 22.3 ±2.5 0.70 <0.01 0.25 67.4 ±3.5 63.4 ±1.5 0.94 0.15 0.50 46.2 ±1.7 40.3 ±2.5 0.87 0.03 0.75 41.5 ±0.7 38.7 ±1.0 0.93 0.02 1.00 40.3 ±0.5 35.7 ±2.4 0.89 0.03 GBML1 1.25 37.5 ±1.0 34.6 ±0.7 0.92 0.02 1.50 36.3 ±1.0 32.4 ±0.7 0.89 <0.01 1.75 33.1 ±0.8 33.6 ±1.7 1.02 0.64 2.00 30.6 ±1.4 32.3 ±4.4 1.05 0.58 0.25 92.9 ±2.2 92.0 ±3.6 0.99 0.72 0.50 78.7 ±2.2 82.3 ±5.0 1.05 0.31 0.75 70.3 ±2.3 72.6 ±1.7 1.03 0.23 1.00 61.2 ±5.4 49.6 ±7.7 0.81 0.10 G53 1.25 50.8 ±1.6 38.5 ±4.7 0.76 0.01 1.50 42.7 ±2.1 31.8 ±4.3 0.75 0.02 1.75 34.9 ±4.6 25.1 ±4.4 0.72 0.06 2.00 32.0 ±2.3 22.6 ±2.7 0.71 0.01 0.25 94.4 ±0.6 90.8 ±3.2 0.96 0.13 0.50 91.6 ±2.1 81.5 ±8.5 0.89 0.11 0.75 82.4 ±6.7 64.7 ±5.0 0.79 0.02 1.00 70.9 ±2.2 45.1 ±7.5 0.64 <0.01 WK1 1.25 64.0±3.3 35.7 ±2.9 0.56 <0.01 1.50 52.3 ±4.3 36.1 ±1.1 0.69 <0.01 1.75 50.1 ±2.8 37.3 ±2.3 0.74 <0.01 2.00 47.5 ±5.7 41.6 ±3.2 0.88 0.19

3.5.2 Supplementary Figures

Supplementary Figure 3.1

Representative images of clonogenic assays obtained for each cell line. Cells were seeded at a range of concentrations 100-10,000 cells/well in 6-well plates, and allowed to proliferate over 14 days. They were then stained with crystal violet and assessed for colonies. All cell lines, except for GBML1, formed countable colonies and proceeded to be further analyzed for interaction between nanoparticles and radiation.

Supplementary Figure 3.2

Optimisation of radiation dose for effect in colony-forming cell lines. Cells were seeded at optimal seeding density, and exposed to range of radiation from 0-8 Gy. Colony-forming units (CFUs) were then stained and counted 14 days after seeding. Data presented as mean percentage of CFUs counted at each dose compared to the control (0 Gy). Experiments were performed in n=1 replicate, in n=1 repeat.

CHAPTER 4

4 MECHANISMS OF CELL DEATH BY LANTHANUM

OXIDE NANOPARTICLE IN GLIOBLASTOMA.

4.1 Introduction

In the previous chapter I demonstrated the ability for REO NPs to cause cell death in

GBM ICLs and PDCLs, as well as augment adjuvant therapies. Thus, the purpose of this chapter was to determine by what molecular mechanisms of cell death occurs, which is important in the evaluation of a novel therapy. The REO La2O3 was selected as the representative REO NP therapy to investigate due to its consistent trends favouring therapeutic effect shown in the previous chapter.

4.1.1 Inducing effect within the cell

In order for a NP therapy to induce effect in the cell, it must first penetrate the cellular membrane. Endocytosis is the primary transport pathway into the cell, with the main three types being 1) clathrin-mediated, 2) caveolae-mediated, and 3) clathrin- and caveolae- independent, which is typified by macropinocytosis.227 To date, the precise endocytic mechanism by which La2O3 NPs specifically may enter cells has not been reported.

Once inside the cell, there needs to be a reactive component of the NP that can then trigger pathways that lead to cell death. An issue is that REOs, such as La2O3, are stable compounds that are largely considered insoluble, at least at a macroscopic level. This is an attractive feature in the design of a NP to penetrate cells which relies on physical characteristics of being a solid, such as surface charge and shape.228 If the NP was composed of a more soluble material, then these features would be lost. But, once penetration of the cell is achieved, if the material is absolutely inert, then most likely, no biochemical disturbance will be observed.

Thus, to explain the results of the previous chapter, despite the relative stability of the

3+ compound, exploration into whether or not ionic La can be detected in the presence of La2O3

NPs is required.

4.1.2 Cell death by apoptosis

The most studied mechanism of cell death is apoptosis.229 Apoptosis involves the disruption of the cellular membrane, which leads to fatal morphological changes and then cell death. It can be caused by different triggers, and include reactive oxygen species (ROS), as well as dysregulation of receptor pathways.229

There are multiple factors that can be measured to validate this process. During apoptosis, internal lipid phosphatidylserine (PS) is exposed to the extracellular environment.230

Annexin V is a calcium-binding protein which binds to PS, and can be detected by fluorescence with flow cytometry. Another approach is measuring poly(ADP-ribose) polymerase (PARP), a protein that is cleaved by activated caspases-3 and -7, a hallmark event underpinning apoptosis.231

Apoptosis involves a number of proteolytic enzymes, primarily members of the caspase family, which are crucial effector molecules.232 Activation of these caspases leads to the initiation of apoptosis, and is achieved by stimulation of either the mitochondria or membranous receptors which correspond to the intrinsic and extrinsic pathways respectively.233 When activated, these caspases are cleaved next to aspartate residues in a cascading fashion to propagate apoptosis. Although both pathways share particular caspases in common, e.g. caspases -3 and -7, there are other caspases specific to each pathway (caspase-9 in the intrinsic pathway; caspase-8 in the extrinsic pathway) which, can be used as measurable markers specific for each pathway.

4.1.2.1 The intrinsic pathway

The intrinsic pathway, or sometimes referred to as the mitochondrial pathway, is initiated when a number of different pro-apoptotic factors are released upon stimulation of the mitochondria to activate caspases.233 They include cytochrome c, apoptosis-inducing factor

(AIF) and Smac (second mitochondria-derived activator of caspase)/DIABLO (direct inhibitor of apoptosis protein (IAP)-binding protein with low PI).234 Collectively, this triggers the formation of an apoptosome complex involving cytochrome c, Apaf-1 and caspase-9 which goes onto activate the common apoptotic pathway caspase-3 leading to cell death.234

4.1.2.2 The extrinsic pathway

The extrinsic pathway, or sometimes referred to as the receptor pathway, differs from the intrinsic pathway in that its activation is not organelle-based. Rather, this pathway is initiated upon stimulation of receptors of the tumour necrosis factor (TNF) death receptor family such as TNF-α and CD95/FAS/APO-1, as well as TNF-related apoptosis-inducing ligand (TRAIL) receptors such as TRAIL/APO-2. Caspase-8 is then exclusively activated to propagate the common effector caspases downstream to lead to apoptosis.235

4.1.2.3 Interactions with mitochondria

Given the intrinsic pathway is also termed the mitochondrial pathway, it is important to recognise the organelle as one that may directly interact with La2O3 NPs. Mitochondrial specific factors involved in apoptosis include pro-apoptotic bax and anti-apoptotic bcl-2. Bax has the ability to open membranous pores of the mitochondria, which in turn facilitate greater release of cytochrome c into the mitochondrial intermembrane space and trigger the intrinsic pathway.236 Bcl-2 antagonises the actions of bax.237 This is relevant to the specificity of the

REO NP therapy for GBM, as bcl-2 family proteins have been reported to be expressed at high levels in GBM, in particular in comparison to glioma of lower grades.238 Finally, these

100

interactions have translational significance where modulation of these mitochondrial factors can facilitate greater response to radiation therapy (RT) and chemotherapy (CT) with temozolomide (TMZ).239

4.1.3 Cell death by other mechanisms

4.1.3.1 Direct DNA damage

Direct DNA damage can be caused by extrinsic factors, such as ionising radiation and cytotoxic drugs.240 In response to DNA damage that cannot be repaired, measured by DNA damage by-product ɣ-H2AX, cell death can be triggered which does not involve the caspase system or mitochondria, aspects that characterise apoptosis.241,242 However, limited activation of death receptors in response to this type DNA damage has been associated with partial activation of the extrinsic apoptosis pathway which may or may not apply to this REO NP therapy.242

4.1.3.2 Autophagy

Autophagy is another cell death pathway, often described as an intracellular degradation response to cellular stress that delivers cytoplasmic content, such as NPs, to the lysosome, which then undergo catabolism.243 The accumulation of these autophagic vacuoles is then thought to lead to programmed cell death due to the major ultrastructural changes of cytoplasmic organelles.244 One possible marker of autophagy is light chain 3 (LC3), a microtubule-associated protein that is conjugated from cytosolic LC3-I to LC3-II during the formation of the autophagosomal complex that ultimately leads to cell death.245

101

4.1.4 Objective and aims

The objective of this chapter was to investigate possible mechanisms in which La2O3

NP therapy induced cellular death in GBM in vitro. Specifically, the aims of this chapter were:

1. To determine the mechanism of NP entry into the cell.

2. To establish if La2O3 dissociates within the cell.

3. To determine the relevance of apoptosis in causing cell death.

4. To determine the relevance of direct DNA damage in causing cell death.

5. To determine the relevance of autophagy in causing cell death.

6. To elucidate interactive aspects of cell death between the NP and adjuvant therapies

of radiation therapy and chemotherapy by TMZ.

102

4.2 Results

4.2.1 Cellular uptake by endocytosis

The types of endocytosis investigated by flow cytometry were caveolae-mediated endocytosis, clathrin-mediated endocytosis, and macropinocytosis, which were inhibited by 10

µg/mL nystatin (NY), 10 µg/mL chlorpromazine (CH) and 1 µg/mL cytochalasin D (CD), respectively doses in the literature which were confirmed not to induce significant cell death in the GBM PDCLs.180-183 Compared to the positive control consisting of NP only, the combination of NP+CH led to statistically significant reductions in uptake across all four

PDCLs (all P<0.01) as demonstrated in Figure 4.1. With respect to both NP+NY and NP+CD combinations, there were no significant differences between the combinations and the positive control across all PDCLs (Supplementary Table 5.1).

4.2.2 Lanthanum ion dissociation

3+ Ionic dissociation of La2O3 NP to free La ion in solution was measured using inductively-coupled plasma mass spectrometry (ICP-MS). Compared to the negative control of MilliQ water, there was a statistically significant presence of La3+ ions in the superficial supernatant of centrifuged NP suspensions (P<0.01) as demonstrated in Figure 4.2. The Ksp

(solubility constant) was calculated to be 4.9 ±0.9 x10-20.

4.2.3 Cell death by apoptosis

Across all PDCLs, evidence of dose-dependent apoptosis involvement was observed with the La2O3 NP therapy (Figure 4.3). Increased proportions of early and late apoptotic cells were measured, correlating with increased NP concentration (Supplementary Table 4.2). To confirm the Annexin-V positivity, western blots to confirm the apoptotic pathway end-product

103

A )

% 150 (

RN1

l n

o GBML1

N 100 G53

o t

* WK1 A

- * * C

S 50 *

e * *

v *

i *

a l

e 0 R l ly Y D H o n N C C tr o + + + n P P P P o N N N N c e v ti a g e N Inhibitor combination

B Inhibitor combination

Negative NP only NP+NY NP+CD NP+CH control

C P

Figure 4.1 Endocytosis analysis by flow cytometry.

Endocytosis mechanisms of La2O3 nanoparticles (NPs) were investigated by means of flow cytometry for 10,000 events A. across four PDCLs, with B. representative results for each PDCL. Cells were exposed to pre-determined concentrations of different mechanism inhibitors for 30 minutes before being removed, and 100 µg/mL NPs added for 1 hour before measurement. Inhibitors were 10 µg/mL nystatin (NY), 10 µg/mL chlorpromazine (CH) and 1 µg/mL cytochalasin D (CD) to inhibit caveolae-mediated endocytosis, clathrin- mediated endocytosis and macropinocytosis respectively. The median side scatter area (SSC-A) measure was used as a surrogate for marker for NP uptake within a cell after solitary NPs were gated out. Values presented as mean ±SD, with percentages of the SSC-A relative to the NP only results (dotted line represents control). Flow cytometry plots are presented as forward scatter area (FSC-A; x-axis) vs SSC-A (y-axis). Data were obtained in triplicate. Statistical analysis was performed using two-way ANOVA analysis against the NP only (control) result, with significance set at P<0.05 (*).

104

Figure 4.2 Free ion analysis by ICP-MS.

The concentration of La3+ ions 72 hours after addition of nanoparticle (NP) to a 1000 µg/mL suspension as determined by means of inductively-coupled plasma mass spectroscopy (ICP-MS). All suspensions were made with neutral MilliQ water, and centrifuged for 10 minutes at 1500 RPM before superficial supernatant was sampled. Data were obtained in technical triplicate. Mean difference was 6.5 µM, with P<0.01. Statistical analysis was performed using student t-test analysis, with significance set at P<0.05 (*).

105

Viable cells Early apoptotic cells

150 50 )

0 µg/mL ) 0 µg/mL

% (

* * * * (

50 µg/mL 40 50 µg/mL

s l

* * * * l

l e

100 100 µg/mL e 100 µg/mL

* * * 30 f

* f

n * *

o i

i 20 t

t * r

50 r * *

o p

p 10 * *

P P 0 0 1 1 3 1 1 1 3 1 N L 5 K N L 5 K R M G W R M G W B B G G

Late apoptotic cells Necrotic cells

50 50 )

0 µg/mL ) 0 µg/mL

% (

* (

40 * 50 µg/mL 40 50 µg/mL

s l

* l l

* l e

* * 100 µg/mL e 100 µg/mL

30 30 f

* * f

o i

20 i 20 t

* t *

r o

* * o * * * * p

10 p 10 o

o * * *

P P 0 0 1 1 3 1 1 1 3 1 N L 5 K N L 5 K R M G W R M G W B B G G

B La2O3 NP concentration

0 µg/mL 50 µg/mL 100 µg/mL

K W

Figure 4.3 Apoptosis pathway analysis by flow cytometry.

The role of the apoptotic pathway causing cell death was first investigating using Annexin V (AnnV) and PI marker staining by means of flow cytometry A. across four PDCLs, with B. representative results for each PDCL. Cells were exposed 0, 50 and 100 µg/mL of NP for 72 hours, and then stained with the respective markers, where 10,000 cells for each sample were analysed. The proportion of cells positive vs negative for both markers were used as indicators for viable (AnnV-/PI-), early apoptotic (AnnV+/PI-), late apoptotic (AnnV+/PI+), and necrotic (AnnV-/PI+) cells after solitary NPs were gated out. Values presented as mean ±SD. Flow cytometry plots are presented as fluorescence of PI (x-axis) vs AnnV (y-axis). Data obtained in triplicate. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (*).

106

cleaved PARP (cPARP) across all four PDCLs were performed (Figure 4.4). Consistent with

Annexin-V staining, the expression levels of cPARP increased with increasing concentration of NP across all PDCLs (Supplementary Table 4.3).

To determine the nature of apoptosis, intrinsic and extrinsic pathways were explored using western blotting antibodies to detect the expression of the pro-form of caspase-9, and the cleaved form of caspase-8 respectively (Figure 4.4). We observed significant trends in favour of both intrinsic and extrinsic pathway activation in dose-dependent manners with the La2O3

NP therapy (Supplementary Table 4.3).

4.2.4 Cell death by direct DNA damage

The occurrence of direct DNA damage caused by the La2O3 NP therapy was investigated by western blotting (Figure 4.4). Expression of double stranded DNA (dsDNA) marker ɣ-H2AX was significantly increased with increasing concentration of NP across all

PDCLs (Supplementary Table 4.3).

4.2.5 Radical oxygen species

To determine potential triggers for the cell death observed, the intracellular generation of ROS after exposure to the La2O3 NP therapy was investigated by means of flow cytometry.

Expression of intracellular ROS marker H2DCFDA was significantly increased with increasing concentration of NP across all PDCLs (Supplementary Table 4.4).

To further localise ROS generation, ROS generation within the mitochondria was then evaluated (Figure 4.5). Similar to H2DCFDA, expression of mitochondrial ROS marker

Mitosox was significantly increased with increasing concentration of NP across all PDCLs

(Supplementary Table 4.4).

107

A cPARP Caspase-9

5000 150 )

) 0 µg/mL 0 µg/mL %

% * * (

( *

4000 y * t

t 50 µg/mL * * * 50 µg/mL

s n

* 100 µg/mL n 100 100 µg/mL

e t

t 3000 *

d e

2000 e

z i

* i 50

l a

* a m

1000 m

o N 0 N 0 1 1 3 1 1 1 3 1 N L 5 K N L 5 K R M G W R M G B B W G G

cCaspase-8 H2AX

600 4000 * )

0 µg/mL ) 0 µg/mL %

* %

(

* y

* y t

t 50 µg/mL 50 µg/mL i

* i 3000 s

s * n

400 * * 100 µg/mL n 100 µg/mL

e t

t * n

* n i

i *

2000 * d

d * *

e z

z * i

i 200 l

l * a

a 1000

o N 0 N 0 1 1 3 1 1 1 3 1 N L 5 K N L 5 K R M G W R M G W B B G G B La2O3 NP concentration (µg/mL)

0 50 100 0 50 100 0 50 100 0 50 100 cPARP (88 kDa) Caspase-9 (47 kDa) cCaspase-8 (41, 43 kDa)

eIF4e (25 kDa)

H2AX (15 kDa)

α-tubulin (52 kDa)

RN1 GBML1 G53 WK1

PDCL Figure 4.4 Apoptosis marker analysis by western blotting.

The presence and trends in apoptosis-related proteins cleaved PARP (cPARP), caspase-9, cleaved caspase- 8 (cCaspase-8), and ɣ-H2AX, were investigated by means of western blotting A. across four PDCLs with B. representative blot images for all four shown. Cells were exposed 0, 50 and 100 µg/mL of NP for 72 hours, and then protein was harvested. Wells were loaded with 10 µg protein. Concentration for both primary antibodies were 1:1000, and eIF4e was used as loading control. Intensity was measured using ImageStudio Lite software, and normalised to the intensity of the 0 µg/mL measurement. Data were obtained in technical triplicate. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (*).

108

H2DCFDA Fluoresence Mitosox Fluoresence )

200 ) 600 %

% * (

* (

Control * Control e

* * * e c

* * c n

* 50 µg/mL n * * 50 µg/mL e

150 * e c

* c * * s

100 µg/mL s 400 100 µg/mL

r o

100 o * * *

l f

f * *

d 200

z i

50 i

r o

0 o 0

N N 1 1 3 1 1 1 3 1 N L 5 K N L 5 K R M G R M G B W B W G G

B C La2O3 NP concentration La2O3 NP concentration

0 µg/mL 50 µg/mL 100 µg/mL 0 µg/mL 50 µg/mL 100 µg/mL

W W

Figure 4.5 Radical oxygen species analysis by flow cytometry.

The generation of radical oxygen species were investigated using general intracellular fluorescent marker H2DCFDA and mitochondria specific fluorescent marker Mitosox by means of flow cytometry A. across four PDCLs, with representative results for each PDCL for B. H2DCFDA and C. Mitosox. Cells were exposed 0, 50 and 100 µg/mL of NP for 72 hours, and then stained with the respective markers, where 10,000 cells for each sample were analysed. The geometric mean of fluorescence intensity at the corresponding wavelengths (H2DCFDA, 530 nm; Mitosox, 585 nm) were then calculated after solitary NPs were gated out. Values presented as mean ±SD. Flow cytometry plots are presented as histogram of H2DCFDA/Mitosox fluorescence. Data obtained in triplicate. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (*).

109

4.2.5.1 Involvement of mitochondria in apoptosis

The involvement of the mitochondria in the intrinsic pathway of apoptosis was investigated by evaluating mitochondria-specific markers by means of western blotting across all PDCLs (Figure 4.6). The observed trends were significant decrease in anti-apoptotic bcl-2 expression with increasing concentration of NP, and significant increase in pro-apoptotic bax expression at 50 µg/mL NP, but negligible change at 100 µg/mL NP (Supplementary Table

4.5).

To determine if the bax results were time-dependent, bax expression at 24, 48 and 72 hours with 50 and 100 µg/mL La2O3 NP concentrations were investigated in PDCL RN1

(Figure 4.6C). Its expression peaked at 72 h following 50 µg/mL NP, and at 48 h following

100 µg/mL NP, suggesting bax expression to be dose- and time-dependent within the mitochondria, and may precede the changes in Bcl-2 expression at higher concentrations

(Supplementary Table 4.6).

4.2.6 Cell death by autophagy

By western blotting, the possibility the La2O3 NP therapy causing cell death by autophagy was also investigated due to the observation of significant vacuolisation in PDCLs after NP exposure (Figure 4.7). The expressions of autophagy markers LC3-I and LC3-II were significantly decreased and increased respectively with increasing concentration of NP across all PDCLs, corresponding to the autophagy-associated transformation of LC3-I into LC3-II

(Supplementary Table 4.7).

4.2.7 Interaction of NPs with radiation therapy

Given the relationship between ROS generation following radiation therapy (RT), and the presumed Auger effect of REEs, the potential for the La2O3 NP therapy to augment the

110

A Bcl-2 Bax

150 200 )

0 µg/mL ) * 0 µg/mL

% % (

* * * * (

y * t

* * * * 50 µg/mL t 50 µg/mL i

i 150 * s

s * n

100 100 µg/mL n 100 µg/mL

e e

n n

i i

100

d d

e e

z z i

50 i

l l a

a 50

m m

r r

o o N 0 N 0 1 1 3 1 1 1 3 1 N L 5 K N L 5 K R M G W R M G W B B G G B La2O3 NP concentration (µg/mL)

0 50 100 0 50 100 0 50 100 0 50 100 Bcl-2 (26 kDa)

Bax (21 kDa)

α-tubulin (52 kDa)

RN1 GBML1 G53 WK1

PDCL

Bax (RN1) Bax (RN1) 200

) * * 24 hr

% 24h 48h 72h (

* y

t 48 hr

i 150 s

n 72 hr 50 µg/mL

n i

100

d e

z 100 µg/mL

i l

a 50

r o N 0 α-tubulin (52 kDa) 50 µg/mL NP 100 µg/mL NP

Figure 4.6 Mitochondrial apoptosis analysis by western blotting.

The presence and trends in mitochondrial apoptosis-related proteins Bcl-2 and Bax were investigated by means of western blotting A. across four PDCLs with B. representative blot images shown. Additionally, C. the time-dependent nature of Bax was investigated in PDCL RN1. Cells were exposed 0, 50 and 100 µg/mL of NP for 72 hours if not specified, and then protein was harvested. Concentration for both primary antibodies were 1:1000, and α-tubulin was used as loading control. Intensity was measured using ImageStudio Lite software, and normalised to the intensity of the 0 µg/mL measurement. Data were obtained in technical triplicate. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (*).

111

A Control @ 72 h La2O3 NP @ 72 h

B LC3-I LC3-II

150 2000 * )

0 µg/mL ) 0 µg/mL

% (

* * * (

y * t

* * * * 50 µg/mL t 50 µg/mL i

i 1500

s n

100 100 µg/mL n 100 µg/mL

t n

n *

1000 d

d * * e

e *

z i

50 i

l a

a 500

o N 0 N 0 1 1 3 1 1 1 3 1 N L 5 K N L 5 K R M G W R M G W B B G G C La2O3 NP concentration (µg/mL)

0 50 100 0 50 100 0 50 100 0 50 100 LC3-I (16 kDa) LC3-II (14 kDa)

eIF4e (25 kDa)

RN1 GBML1 G53 WK1

PDCL

Figure 4.7 Autophagy analysis by western blotting.

The possibility of this NP therapy causing autophagy was first considered when evidence of significant vacuolisation was observed after 72 hours of exposure to the La2O3 NP with A. a representative image of this phenomenon observed in PDCL WK1 with 100 µg/mL of NP at 40X magnification using light microscopy. The presence and trends in autophagy-related proteins LC3-I and LC3-II were investigated by means of western blotting B. across four PDCLs with C. representative blot images for all four shown. Cells were exposed 0, 50 and 100 µg/mL of NP for 72 hours, and then protein was harvested. Concentration for both primary antibodies were 1:1000, and eIF4e was used as loading control. Intensity was measured using ImageStudio Lite software, and normalised to the intensity of the 0 µg/mL measurement. Data were obtained in technical triplicate. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (*).

112

ROS produced by RT by means of flow cytometry (Figure 4.8). There were significant increases in general ROS marker H2DCFDA in the NP+RT combination compared to NP only and RT only across all four PDCLs (Supplementary Table 4.8).

To clarify the possible downstream interaction between the La2O3 NP therapy and RT, the expression of dsDNA damage marker ɣ-H2AX was also investigated by means of western blotting (Figure 4.9A). There was a significant increase in ɣ-H2AX expression in the NP+RT combination compared to NP only and RT only in PDCL RN1 (Supplementary Table 4.9).

4.2.8 Interaction of NP with chemotherapy

To determine downstream interaction between the La2O3 NP therapy and CT agent

TMZ, anti-apoptotic protein bcl-2 expression was investigated by means of western blotting

(Figure 4.9B). There was a significant decrease in bcl-2 expression in the NP+CT combination compared to NP only and CT only in PDCL RN1 (Supplementary Table 4.9).

113

A H2DCFDA Fluoresence

) 250

% (

Control

e * c

n 200 * * NP e * c * s * * RT e 150 r * *

o * * NP+RT

l f

100

i l

a 50

m r

o 0 N 1 1 3 1 N L 5 K R M G W B G

B Combination

Control NP RT NP+RT

K W

Figure 4.8. RT interaction analysis by flow cytometry.

The generation of radical oxygen species was investigated using general intracellular fluorescent marker H2DCFDA after radiation in the presence of NPs by means of flow cytometry A. across four PDCLs, with B. representative results for each PDCL. Cells were exposed to 50 µg/mL of NP for 24 hours, and then exposed to 4 Gy radiation, before being stained with the respective markers 24 hours later, where 10,000 cells for each sample were analysed. The geometric mean of fluorescence intensity at the corresponding wavelength (H2DCFDA, 530 nm) were then calculated after solitary NPs were gated out. Values presented as mean ±SD. Flow cytometry plots are presented as histogram of H2DCFDA fluorescence. Data obtained in triplicate. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (*).

114

A H2AX (RN1) 1250 H2AX (RN1)

) Control %

( *

y 1000

t NP i *

s NP - + - +

n * RT e

t 750 RT - - + +

n i

NP+RT d

e 500

z i

l H2AX (15 kDa) a

m 250

r o

N α-tubulin (52 kDa) 0 1 N R B Bcl-2 (RN1)

200 Bcl-2 (RN1)

) Control

(

t NP i 150 NP - + - + s *

n CT e

t * CT - - + +

n i

100 * NP+CT

z i

l Bcl-2 (26 kDa)

a 50

r o

N α-tubulin (52 kDa) 0 1 N R Figure 4.9. RT and CT interaction analyses by western blotting.

The presence and trends in A. radiation-associated ɣ-H2AX protein in combinations of NP and radiation therapy (RT) and B. anti-apoptotic Bcl-2 protein in combinations of NP and chemotherapy (CT) agent temozolomide (TMZ), were investigated in PDCL RN1 by means of western blotting with accompanying representative blot images. Cells were exposed 50 µg/mL of NP for 24 hours, and then exposed to either 4 Gy RT or 200 µM TMZ. All proteins were then harvested 24 hours later, with 10 µg loaded into each gel well. Concentration for both primary antibodies were 1:1000, and α-tubulin was used as loading control. Intensity was measured using ImageStudio Lite software, and normalised to the intensity of the 0 µg/mL measurement. Data were obtained in technical triplicate. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (*).

115

4.3 Discussion

The behaviour and mechanisms of La2O3 NP therapy in GBM cells have not been previously reported in the literature. Based on the results of this chapter, the NPs likely enter the cell by clathrin-mediated endocytosis and are able to produce ionic La3+. It is likely that these free La3+ ions then induce a myriad of anti-cancer effects, including intrinsic and extrinsic apoptosis, direct dsDNA damage, and autophagy to cause cell death, with ROS generation a plausible trigger. Additionally, the NP therapy interacted with cellular mitochondria to augment the effects of adjuvant RT and CT.

4.3.1 Endocytosis of nanoparticle is clathrin-mediated

Using flow cytometry, cellular uptake of the NP was measured.184 The combination of

CH+NP reduced NP uptake, indicating the uptake of the La2O3 NP therapy involved clathrin- mediated endocytosis.246 This pathway was shown to be quite predominant, as both combinations of NY+NP (inhibits caveolae-mediated endocytosis), and CD+NP (inhibits micropinocytosis) did not result in significant changes to NP uptake into the GBM cells.247

This observation aligns with much of the literature describing NP uptake into all types of cells.

In terms of REO NPs, clathrin-mediated endocytosis was the primary uptake pathway for a gadolinium-based NP therapy into the immortalised GBM cell line U87 MG248, as well as for

249 250 cerium oxide (CeO2) NP therapies in both normal keratinocytes and colorectal cancer cells.

Outside REO NPs, a large body of research into other NPs and nanostructures have reported similar conclusions. Widely studied gold NP therapies have similarly been shown to be transported into normal90,251 and cancerous252 cells by means of this type of endocytosis, as well as nano-therapies involving nanosheets253 and nanocages254.

One assumption that requires further validation is that uptake of the NP therapy is exclusively by clathrin-mediated endocytosis. It may be more adequate to describe this pathway as the primary pathway for uptake, because the cellular membrane is such a dynamic

116

entity it would be unwise to assume that the membrane does not interact by other endocytic pathways as well. For instance, other nano-therapies have been shown to favour multiple pathways in addition to clathrin-mediated endocytosis in normal cells, including the caveolae- dependent and -independent pathways inhibited by NY and CD respectively.255,256 These pathways could still be involved for this NP therapy in GBM cells, but to a much lesser extent than could have been detected by the concentrations of inhibitors used in this experiment.

4.3.2 Free lanthanum ions are present in the cell

The presence of free La3+ ions in aqueous solution was demonstrated using ICP-MS.

Despite the simplicity of this experiment, the observed presence of free La3+ ion is integral in understanding how a very stable REO may induce cytotoxic effect in GBM cells.139,257 Based

-20 on the concentrations used, the Ksp was calculated to be 4.9 ±0.9 x10 , which is within

-22 approximately 2-orders of magnitude to the theoretical Ksp of ~10 for all REO involving a

RE(+III).258

3+ In normal cells, the IC50 for La has been shown to be approximately 52 µM after 24 hours of exposure, which is within 1-order of magnitude to the calculated concentration of 6.5

259 µM based on the calculated Ksp after 72 hours. Therefore this result seems plausible due to the cytotoxic effects observed within 24 and 72 hours shown in the previous chapter.

It is most interesting to note that CeO2 is a unique REO in the sense that is extremely

-54 258 insoluble, with a theoretical Ksp as low as ~10 . In the previous chapter, CeO2 NP therapy was the least cytotoxic of all tested REO NP therapies, and given a significant increase in solubility was demonstrated with the more cytotoxic La2O3 NP therapy, a relationship between solubility and REO NP effect within GBM cells is proposed.

This experiment was carried out in MilliQ water (pH=7.0) at room temperature (25oC) which may have confounded the calculated results. This is because the cellular environment is a complex assortment of organelles, each with a different pH; of particular note are the early

117

(pH=6.5) and late (pH=6.0) endosomes in which the NPs enter by clathrin-mediated endocytosis, the lysosomes (pH<5.5) in which the NPs are delivered to, all while surrounded by the cytoplasm (pH=7.2).260 These cellular organelles in which the NPs would find themselves in are thus more acidic than this experiment, which would push the solubility equilibrium in favour of greater dissociation, more free La3+ ion production, and more cytotoxic effect. Additionally, the cells were maintained at 37oC, which reflects a more realistic in vivo representation, and given this is warmer than the experimental in vitro conditions, it would also theoretically push the equilibrium further in favour of more free ion La3+ production.261

4.3.3 Nanoparticle activates both apoptosis pathways

A dose-dependent response of Annexin V intensity was observed across all PDCLs which indicated that the presence of the La2O3 NP therapy was associated with apoptosis in

GBM cells. This observation has yet to be reported for any La2O3 formulation in the literature.

However, free La3+ ion formulation (chloride) has been shown to induce apoptosis by increasing Annexin V expression in non-cancerous endothelial262 and cancerous colorectal205 cells.

To confirm the occurrence of apoptosis, and alleviate any concern for gating inconsistencies inherent in flow cytometry, the expression of cPARP was evaluated and a similar dose-dependent trend was observed across all PDCLs. This is the first report of such a trend with REO NP formulation in GBM cells. Previously, similar changes in cPARP expression have been reported for free La3+ ion formulations (chloride and citrate) in hippocampal263 and HeLa264 cells respectively.

Involvement of the intrinsic apoptotic pathway was evident based on the decrease in uncleaved (unactivated) caspase-9 expression in response to the NP therapy, implying its activation for formation of the apoptosome complex. This is a novel finding, for there has been no published data regarding apoptotic pathways involved with La2O3 NP. There is anecdotal

118

evidence however, in favour of this finding, that free La3+ ions (citrate) specifically activate the intrinsic pathway in HeLa cells.264,265 Furthermore, the intrinsic pathway appears to be the most relevant pathway to NP-caused cell death in general, supporting the notions of this caspase-9 observation.266

The extrinsic pathway was also investigated. Surprisingly, its involvement was indicated based on the increase in cleaved (activated) caspase-8, which was shown to be consistently elevated after NP exposure. This was a novel finding for there is little evidence in the literature describing any NP therapy actively involving both intrinsic and extrinsic pathways simultaneously.267 Although there are no reports in the literature associating lanthanum-based NPs to the extrinsic pathway, other REE free ions such, as gadolinium in osteosarcoma cells268 and yttrium in leukaemic cells269, have been linked with the extrinsic pathway to cause apoptosis. It should be noted that the contribution of the extrinsic pathway to the overall apoptosis process in GBM may not be equal to that of the intrinsic pathway due to the naturally low abundance of extrinsic pathway mediators such as caspase-8, evidenced also by a greater required exposure time for immunoblotting compared to the more common intrinsic pathway caspase-9.270

4.3.4 Nanoparticle causes direct DNA damage

Given cell death can also be a nuclear event, the strong dose-dependent response in ɣ-

H2AX expression with the La2O3 NP therapy in all PDCLs implied that in addition to apoptosis, which can occur outside the nucleus, this therapy also caused direct nuclear damage in GBM cells. This is another novel finding to report, as to date, there is no evidence in the literature associating lanthanum in any formulation directly with ɣ-H2AX generation in vitro. The closest would be a study by Paiva et al., who utilised a comet tail assay to demonstrate that free

La3+ ion (nitrate) was associated with DNA damage in normal Jurkat cells.257

119

With respect to other REO NP therapies, cerium NPs (CeO2) have also been shown to increase ɣ-H2AX in normal blood cells.271 However, this association may not necessarily be generalised to all REEs, for in a study that compared NPs coated with cerium to another REE europium, there was no significant change in ɣ-H2AX expression with the latter combination.272 There is that emerging evidence suggests that the dsDNA damage represented by ɣ-H2AX can also lead to generation of universally cytotoxic ROS by inhibition of NADP(H) oxidase.273 This highlights the interconnectivity of cell death mechanisms, as it is then possible that ROS generation would then propagate not only ROS-dependent dsDNA damage mechanisms, but also the ROS-associated apoptosis mechanisms in the cell. Greater molecular investigation is required to ascertain sequence and significances of these mechanisms.

4.3.5 Generation of reactive oxygen species

A plausible trigger for the aforementioned cell death events is the generation of ROS, which was shown to occur in a dose-dependent manner to the La2O3 NP therapy based on the increase in general intracellular and mitochondria-specific ROS markers H2DCFDA and

Mitosox respectively across all PDCLs.236 Associations between lanthanum and ROS have been previously reported in the literature, although examples of La2O3 specifically are not common. One study was able to show that La2O3 NPs in the presence of antioxidant ascorbic acid caused less cell death, indirectly implicating ROS as a main effector of cell death.274

Otherwise, free La3+ ion (chloride) studies have been able to correlate the presence of lanthanum with corresponding increases in ROS both in vitro275 and in vivo276.

The significance of this finding is that ROS plays a major role in the induction of apoptosis by both pathways.236 In terms of the intrinsic pathway, much of the effect by ROS appears to be mitochondria-mediated. The release of apoptosome complex component cytochrome c from the mitochondria, and then the subsequent Apaf1 aggregation and caspase-

9 activation, appears to be directly triggered by ROS in the mitochondria.277,278

120

In terms of the extrinsic pathway, the involvement of ROS is less defined. Given the triggering of the receptor pathway can be cell-wide, the localisation of ROS to an organelle has not been proven.236 Rather, its presence within the cell has been linked to a number of receptor interactions. The accumulation of intracellular glutathione has been shown to downregulate the

Fas receptor pathway, and given ROS consumes glutathione, it would logically then reduce the resistance threshold to this type of apoptosis.279 ROS has been shown to promote interactions with TNF and TRAIL receptor families, key features of the extrinsic pathway.280,281 Thus, involvement of the extrinsic pathway by ROS generation appears feasible, however requires more intense investigation to establish to what extent this occurs.

4.3.5.1 Other interactions with mitochondria

Although the La2O3 NP therapy correlated with increase in mitochondria-specific ROS, the significant involvement of other mitochondrial factors of the intrinsic pathway was demonstrated by apoptotic changes in expression of pro-apoptotic protein bax and anti- apoptotic protein bcl-2 across all PDCLs. Similar studies associating any REO NP therapy and these mitochondrial factors are lacking in the literature, as well as in GBM studies. Yet, these specific trends have been previously reported for free La3+ ion (chloride) in both noncancerous in vitro282 and in vivo263 models, which is promising.

Interestingly, the bax protein did not demonstrate a dose-dependent response to the

La2O3 NP therapy as most of the other factors reported in this chapter. Pro-apoptotic bax expression peaked at the lower concentration when compared to both control and higher concentration in all PDCLs. The implication of this is that the effect of the La2O3 NP therapy on bax is plausibly both dose- and time-dependent. Although this has not been previously reported regarding La2O3 or other REO NP therapies, there have been many reports demonstrating this similar dual-dependence of bax along with other apoptosis aspects; examples include silica NPs in GBM cells283, silver NPs in cervical cancer cells284, and zinc

121

NPs in endothelial cells285. Taken collectively, this particular finding suggest bax expression may be one of the earliest changes induced by the La2O3 NP therapy in the apoptosis processes within GBM.

4.3.6 Nanoparticle induces autophagy

While apoptosis was initially thought to be the primary mechanism for cell death in the

GBM cells with these NPs, vacuolisation observed under light microscopy during apoptosis study suggested autophagy, an alternative apoptosis-independent pathway to cell death.286 This suspicion was validated by the dose-dependent changes in autophagy markers LC3-I and II across all PDCLs, which is a novel finding with respect to La2O3 NP therapy in GBM, and suggests multiple mechanisms of cell death may be involved simultaneously with this therapy.

With respect to GBM, there have been in vitro studies previously reporting autophagy after the use of NP therapy, however, these were not by lanthanum or REO NP therapies, but other materials such as silica- and iron-based NPs.283,287 However, autophagy has been

222,288 previously reported after exposure to La2O3 NP in non-cancerous cells. Furthermore, this effect appears to not be limited to La2O3 alone, with other REO NPs being reported to also induce autophagy including cerium (CeO2), gadolinium (Gd2O3), samarium (Sm2O3),

222,288-290 europium (Eu2O3), and ytterbium (Yb2O3). REO neodymium (Nd2O3) NPs in lung cancer has also been shown to cause autophagy.188

Although it is posited that lysosomal damage is an upstream trigger for autophagic processes, it is currently unclear how the La2O3 NP therapy trigger and modulate autophagy exactly. One possible theory in the case of La3+, and other REEs, is that their ionic size is comparable to that of intracellular calcium ion Ca2+, an integral molecular in multiple cellular homeostatic processes, including cell recycling and thus autophagic structures.291 In REO NP form, these elements have demonstrated dysregulation of intracellular Ca2+ levels, which may prove in the future with more investigation to be one of the earliest triggers of autophagy by

122

this type of NP therapy.292 Once cell death is triggered, it can positively feedback into the autophagic pathway leading to an accumulative tumour cell death which would be of great therapeutic significance.

4.3.7 Nanoparticle augments radiation therapy

The ability for the La2O3 NP therapy to augment radiation therapy (RT) was demonstrated mechanistically by statistically significant increases in both ROS and ɣ-H2AX in NP+RT combination when compared to NP alone and RT alone in PDCLs.217 The potential for radiosensitisation by this therapy in GBM has not been previously indicated by ROS or ɣ-

H2AX to date in the literature. However free La3+ ions have been shown to generate more ROS when combined with RT in prostate cancer cells293 and melanoma cells294. The REE literature regarding radiosensitisation is limited despite the well-known Auger effect and require greater investigation. Lutetium-based peptides are one such REE that has been reported to successfully augment ɣ-H2AX levels in blood lymphocytes295 and cervical cancer cells296 after RT.

4.3.8 Nanoparticle augments chemotherapy

To date, a synergistic relationship between the La2O3 NP therapy and CT TMZ has not been reported. Based on the significant decrease in anti-apoptotic mitochondrial factor bcl-2 in the NP+CT combination compared to NP alone and CT alone observed in the PDCLs, such a relationship seems possible.297,298

Given how many cytotoxic therapies in GBM treatment utilise components of the apoptosis pathways, the validity of this finding is plausible.232 Similar synergy studies in GBM with other non-NP therapies have been performed by Hanif and colleagues, who have investigated the drug Verapamil297 as well as a synthetic acetamide299 in separate combinations with TMZ. They observed similar combination index results to indicate synergy, and noted that when in combination, bcl-2 expression was similarly significantly decreased. In another study,

123

Zhang et al.298 noted in their combination studies of formononetin and TMZ in GBM decreased bcl-2 expression when combined, as well as appropriately corresponding pro-apoptotic changes in other components of the intrinsic pathway.

4.3.9 General limitations

The primary limitations of this chapter concern the inherent difficulties in determining molecular mechanisms. Although the evidence was consistent for the most part in favour of dose-dependent trends, it is acknowledged that this could be expanded upon further in the future. For example, the demonstration of time-dependency with respect to bax protein expression indicates similar trends with time could be likely for other markers within shorter time frames and lower NP concentrations. Admittedly this would not as much strengthen the therapeutic potential argument for the La2O3 NP therapy, as it would at least better inform future in vivo trials in optimisation of dosage and administration scheduling.

The complex and multi-faceted nature of cell death caused by La2O3 NP therapy tempers the expectation each molecular event happens in sequence rather than simultaneously.

The vast potential for any interplay between factors between and within apoptosis, DNA damage cell death, and positive-feedback with autophagy, makes establishing the sequence of events difficult. Additionally, there remains a myriad of potential cell death related factors that could also be evaluated, further complicating the already complex mechanistic picture. For example, intracellular calcium has been shown to change in the presence of La3+ ion, and given the integral role of calcium within the cell, it is difficult to see there being no molecular

300 consequences after exposure to the La2O3 NP therapy.

Full study of every individual component related to cell death is practically not feasible, and investigators should be guided based on rationale and translational potential, otherwise they run the risk of complicating the overall picture. Taken collectively, the mechanisms

124

presented in this chapter should not be viewed as perfectly complete, but rather as starting points for greater molecular investigation.

4.3.10 Overall summary of mechanism

The mechanism of the La2O3 NP therapy in GBM cells to cause cell death is intricate and multi-faceted. The series of experiments in this chapter has demonstrated this (Figure

4.10). A dot-point summary is provided below.

• Cellular uptake is by clathrin-mediated endocytosis, before the NPs are localised to the

endosome system within the cell. The La2O3 NP then undergoes ionic dissociation to

generate free La3+ ion. This ion itself is capable of creating ROS.

• In the mitochondria, ROS not only directly activates the necessary machinery to initiate

the intrinsic apoptosis pathway by caspase-9, it also modulates apoptosis-related

proteins such as bax and bcl-2 in such a fashion that promotes further apoptosis.

• Within the cytoplasm, ROS can trigger the aggregation of markers responsible for

initiating the extrinsic apoptosis pathway by means of activating caspase-8, most likely

via increasing TNF receptor expression. All caspases coalesce into the common

apoptosis pathway via caspase-3, and promotes apoptosis to lead to cell death, in the

process cleaving PARP.

• When ROS is able to infiltrate the cell nucleus, it causes direct DNA damage which

can eventuate into dsDNA damage, and lead to cell death and create ɣ-H2AX as a by-

product. Furthermore, DNA damage can potentially further promote the extrinsic

pathway.

• Although the pathway is not clear, the presence of the La3+ ion induces autophagy by

dysregulating the lysosomal system, possibly via ROS- or Ca2+-dependent mechanisms,

ultimately also leading to cell death.

125

Figure 4.10. Overall mechanism.

The proposed mechanisms of cell death in glioblastoma (GBM) caused by La2O3 nanoparticle (NP) therapy. Once it enters the cell by endocytosis, the NP localises to a number of potential organelles, and induces one of three possible pathways to cell death; 1) apoptosis, intrinsic and extrinsic; 2) direct double stranded DNA (dsDNA) damage; and 3) autophagy, although cell death can also promote autophagy itself, leading to a cyclic pathway. Changes to protein expression were categorised into increase (green) or decrease (red) after exposure to the NP therapy. Interactions with radiation therapy and chemotherapy temozolomide were shown to promote the ROS/ɣ-H2AX and bcl-2 trends respectively. Increase in tumour necrosis factor (TNF) was assumed based on the current literature, but not proven in this thesis (dotted square). LC3; light chain 3; PARP, Poly (ADP-ribose) polymerase; ROS, reactive oxygen species.

126

4.4 Conclusions

In this chapter I investigated the mechanisms by which the La2O3 NP therapy causes cell death in GBM cells, molecular events which have not previously been reported in the literature.

The mechanism for cellular uptake and distribution involves clathrin-mediated endocytosis and ionic dissociation, with cell death mechanisms such as apoptosis, DNA damage, and autophagy all appearing relevant to the therapy. ROS generation and mitochondrial involvement was demonstrated to be an important feature of how this therapy causes cell death, as well as interacts with adjuvant therapies such as radiation and TMZ. Given multiple PDCLs were utilised, it lends credibility to the translational potential of the findings. However further study into additional molecular processes and pathways will greatly enhance the understanding of the mechanisms involved, and improve design of future in vivo GBM studies.

127

4.5 Supplementary

4.5.1 Tables

Supplementary Table 4.1

Summary of changes in relative side scatter area (SSC-A) after 30 minute pre-exposure of different endocytosis inhibitor combinations compared to cells exposed to NPs only, as measure by flow cytometry. Inhibitors were nystatin (NY), chlorpromazine (CH) or cytochalasin D (CD) to inhibit caveolae-mediated endocytosis, clathrin-mediated endocytosis and macropinocytosis respectively. Cells were exposed to NPs (100 µg/mL) for 1 hour before measurement. Data are presented as mean difference (MD) between two combinations. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (bold).

Relative SSC-A vs NP only (%) NP+NY NP+CD NP+CH Negative control Cell line MD P-value MD P-value MD P-value MD P-value RN1 -8.4 0.59 2.5 0.99 -50.8 <0.01 -86.2 <0.01 GBML1 3.0 0.98 5.0 0.85 -39.6 <0.01 -79.8 <0.01 G53 -5.1 0.88 -2.0 0.99 -65.3 <0.01 -76.4 <0.01 WK1 0.4 0.99 0.8 0.99 -47.4 <0.01 -82.9 <0.01

128

Supplementary Table 4.2

Summary of cell population proportions as determined by flow cytometry across four PDCLs after exposure to NP for 72 hours at 0, 50 and 100 µg/mL concentrations. Markers Annexin V (AnnV) and PI were used to indicate the different cell types as defined as viable (AnnV-/PI-), early apoptotic (AnnV+/PI-), late apoptotic (AnnV+/PI+), and necrotic (AnnV-/PI+) cells. Data are presented as mean difference (MD) between two concentrations. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (bold).

Proportion of cells (%) 0 vs 50 µg/mL 50 vs 100 µg/mL 0 vs 100 µg/ml Cell line Cell type MD P-value MD P-value MD P-value Viable -15.5 <0.01 -17.2 <0.01 -32.7 <0.01 Early apoptotic 8.7 <0.01 1.8 0.47 -10.5 <0.01 RN1 Late apoptotic 4.3 <0.01 15.8 <0.01 4.3 <0.01 Necrotic 2.6 0.04 0.2 0.97 2.8 0.02 Viable -27.4 <0.01 -15.1 <0.01 -42.4 <0.01 Early apoptotic 2.7 0.28 2.1 0.40 4.7 0.02 GBML1 Late apoptotic 17.5 <0.01 12.3 <0.01 29.7 <0.01 Necrotic 7.2 <0.01 1.8 0.19 9.0 <0.01 Viable -15.9 <0.01 -17.6 <0.01 -33.5 <0.01 Early apoptotic 9.0 <0.01 1.9 0.46 10.9 <0.01 G53 Late apoptotic 4.5 <0.01 15.3 <0.01 19.9 <0.01 Necrotic 2.6 0.04 0.3 0.92 2.9 0.01 Viable -16.2 <0.01 -20.0 <0.01 -36.2 <0.01 Early apoptotic 4.3 0.02 -0.5 0.94 3.8 0.03 WK1 Late apoptotic 8.6 <0.01 19.2 <0.01 27.8 <0.01 Necrotic 3.5 <0.01 1.4 0.22 4.9 <0.01

129

Supplementary Table 4.3

Summary of changes in normalised intensity of apoptosis-related protein markers cleaved PARP (cPARP), caspase-9, cleaved caspase-8 (cCapase-8), and ɣ-H2AX, as determined by western blotting across four PDCLs after exposure to NP for 72 hours at 0, 50 and 100 µg/mL concentrations. Data are presented as mean difference (MD) between two concentrations. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (bold).

Normalised intensity (%) 0 vs 100 0 vs 50 µg/mL 50 vs 100 µg/mL µg/ml Cell line Protein MD P-value MD P-value MD P-value cPARP 410 0.30 542 0.14 952 <0.01 Caspase-9 -4 0.77 -17 0.02 -21 <0.01 RN1 cCaspase-8 60 0.01 52 0.04 112 <0.01 ɣ-H2AX 834 <0.01 435 0.12 1259 <0.01 cPARP 2361 <0.01 630 0.08 42.4 <0.01 Caspase-9 -24 <0.01 -11 0.20 -34 <0.01 GBML1 cCaspase-8 120 <0.01 227 <0.01 348 <0.01 ɣ-H2AX 866 <0.01 428 0.13 1294 <0.01 cPARP 436 0.26 220 0.64 656 0.04 Caspase-9 -8 0.41 -16 0.02 -24 <0.01 G53 cCaspase-8 213 <0.01 11 0.88 224 <0.01 ɣ-H2AX 652 0.03 1653 <0.01 2305 <0.01 cPARP 1511 <0.01 144 0.85 1368 <0.01 Caspase-9 -24 <0.01 -8 0.33 -32 <0.01 WK1 cCaspase-8 239 <0.01 4 0.97 235 <0.01 ɣ-H2AX 1079 <0.01 1763 <0.01 2842 <0.01

130

Supplementary Table 4.4

Summary of changes in geometric mean fluorescence intensity of general intracellular fluorescent marker H2DCFDA and mitochondria specific fluorescent marker Mitosox across all PDCLs by means of flow cytometry. Cells were exposed 0, 50 and 100 µg/mL of NP for 72 hours, and then stained with the respective markers, where 10,000 cells for each sample were analysed. Data are presented as mean difference (MD) between two combinations. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (bold).

Normalised fluorescence (%) 0 vs 50 µg/mL 50 vs 100 µg/mL 0 vs 100 µg/ml Cell line ROS Marker MD P-value MD P-value MD P-value H2DCFDA 28 <0.01 5 0.51 32 <0.01 RN1 Mitosox 113 <0.01 46 0.09 159 <0.01 H2DCFDA 23 <0.01 22 <0.01 45 <0.01 GBML1 Mitosox 109 <0.01 163 <0.01 272 <0.01 H2DCFDA 44 <0.01 5 0.39 49 <0.01 G53 Mitosox 149 <0.01 83 <0.01 231 <0.01 H2DCFDA 62 <0.01 7 0.20 70 <0.01 WK1 Mitosox 152 <0.01 199 <0.01 351 <0.01

131

Supplementary Table 4.5

Summary of changes in normalised intensity of mitochondria-related protein markers Bcl-2 and Bax, as determined by western blotting across four PDCLs after exposure to NP for 72 hours at 0, 50 and 100 µg/mL concentrations. Data are presented as mean difference (MD) between two concentrations. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (bold).

Normalised intensity (%) 0 vs 50 µg/mL 50 vs 100 µg/mL 0 vs 100 µg/ml Cell line Protein MD P-value MD P-value MD P-value Bcl-2 -34 <0.01 -16 0.18 -50 <0.01 RN1 Bax 46 <0.01 -40 <0.01 6 0.80 Bcl-2 -35 <0.01 -16 0.08 -51 <0.01 GBML1 Bax 36 <0.01 -41 <0.01 -6 0.76 Bcl-2 -16 0.09 -27 <0.01 -43 <0.01 G53 Bax 23 0.049 -54 <0.01 -32 <0.01 Bcl-2 -44 <0.01 -18 0.05 -62 <0.01 WK1 Bax 19 0.13 -25 0.03 -6 0.78

132

Supplementary Table 4.6

Summary of changes in normalised intensity of mitochondria-related protein marker Bax, as determined by western blotting across three time points in PDCL RN1 after exposure to NP for 24, 48 and 72 hours at 50 and 100 µg/mL concentrations. Data are presented as mean difference (MD) between two concentrations. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (bold).

Normalised intensity (%) 24 vs 48 hrs 48 vs 72 hrs 24 vs 72 hrs NP concentration Cell line MD P-value MD P-value MD P-value (µg/mL) 50 23 0.09 23 0.08 46 <0.01 RN1 100 33 0.01 -26 0.04 6 0.82

133

Supplementary Table 4.7

Summary of changes in normalised intensity of autophagy-related protein markers LC3-I and LC3-II, as determined by western blotting across four PDCLs after exposure to NP for 72 hours at 0, 50 and 100 µg/mL concentrations. Data are presented as mean difference (MD) between two concentrations. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (bold).

Normalised intensity (%) 0 vs 100 0 vs 50 µg/mL 50 vs 100 µg/mL µg/ml Cell line Protein MD P-value MD P-value MD P-value LC3-I -29 <0.01 -13 <0.01 -42 <0.01 RN1 LC3-II 308 0.05 167 0.39 -475 <0.01 LC3-I -18 <0.01 -30 <0.01 -47 <0.01 GBML1 LC3-II 468 <0.01 185 0.32 653 <0.01 LC3-I -21 <0.01 -19 <0.01 -40 <0.01 G53 LC3-II 147 0.48 249 0.14 397 0.01 LC3-I -2 0.83 -23 <0.01 -25 <0.01 WK1 LC3-II 1024 <0.01 374 0.02 1397 <0.01

134

Supplementary Table 4.8

Summary of changes in geometric mean fluorescence intensity of general intracellular fluorescent marker H2DCFDA all PDCLs after exposure to NPs and radiation therapy (RT) in different combinations as measured by flow cytometry. Cells were exposed 50 µg/mL of NP for 24 hours, and then exposed to 4 Gy radiation, before being stained with the respective markers 24 hours later. Data are presented as mean difference (MD) between two combinations. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (bold).

Normalised fluorescence (%) Combination vs. Combination vs Combination vs control NP alone RT alone Cell line MD P-value MD P-value MD P-value RN1 52 <0.01 12 <0.01 20 <0.01 GBML1 41 <0.01 15 <0.01 31 <0.01 G53 10 0.049 14 <0.01 11 0.02 WK1 16 <0.01 8 0.14 15 <0.01

135

Supplementary Table 4.9

Summary of changes in normalised intensity of ɣ-H2AX and Bax proteins in different combinations with radiation therapy (RT) and chemotherapy (CT) agent temozolomide (TMZ) respectively, as determined by western blotting in PDCL RN1. Cells were exposed 50 µg/mL of NP for 24 hours, and then exposed to either 4 Gy RT or 200 µM TMZ. All proteins were then harvested 24 hours later. Data are presented as mean difference (MD) between two concentrations. Statistical analysis was performed using two-way ANOVA analysis, with significance set at P<0.05 (bold).

Normalised intensity (%) Combination vs. Combination vs Combination vs

control NP alone therapy alone Cell Adjuvant Protein MD P-value MD P-value MD P-value line Therapy Radiation ɣ-H2AX 520 <0.01 176 0.02 470 <0.01 therapy (RT) RN1 Chemotherapy Bax -70 <0.01 -37 0.03 -30 0.04 (CT)

136

CHAPTER 5

137

5 BIOCOMPATIBILITY OF LANTHANUM OXIDE

NANOPARTICLE IN A MOUSE MODEL.

5.1 Introduction

The in vitro results of the previous two chapters demonstrated that La2O3 NP therapy has promising anti-GBM properties. The next questions posed relate to how this therapy behaves in a biological system: 1) can it cross the blood brain barrier and penetrate the brain? 2) is it well tolerated? 3) what is its biodistribution and half-life? In order to begin answer these questions, I aimed in this chapter to establish how compatible the novel La2O3 NP therapy is in the biological system of balb/c mice. These were chosen as the anticipated xenograft studies for this therapy would be performed in this species.

5.1.1 Biological tolerance

The adverse-effects of intravenous REOs have been previously reported. They include organ inflammation, fibrosis and atrophy which ultimately prove a concern for any REO-based

NP therapy.301-303 Particular attention should be paid to the liver and spleen, as the phagocytic cells of the reticuloendothelial system (RES) located in the liver and spleen attract much of the systemic NP.304 Furthermore, the kidneys are the primary organs for filtration of blood plasma, which was an important to consider if the administration route is intravenous (IV) delivery.305

If indeed any of these organs are severely impaired, then the dosage of the La2O3 NP therapy will need to be further titrated before progression to xenograft studies to ensure that it can be administered without significant toxicity concerns in future studies. One surrogate marker for toxicity is organ and body weights, which will be the primary measures used.305

138

5.1.2 Biodistribution

NP formulations are known to favour particular organs in terms of biodistribution, including the liver and spleen.306-309 However, unclear is the extent of accumulation within the brain.310 Although some proponents of the literature argue that accumulation is negligible, if it can be proven that the stable La2O3 NP can infiltrate the blood brain barrier (BBB), then anti-

GBM effects may be achieved due to its potential slow clearance. It is encouraging to note other NP therapies have been reported to infiltrate intracranial GBM models via intravenous

311-314 administration. Thus, until specific investigation is performed for the La2O3 NP therapy, its distribution to the brain remains largely anecdotal due to the paucity in the current literature, and requires clarification.

5.1.2.1 Effect of adjuvant therapies

The intention of this novel NP therapy was not to replace the current standard therapy regimen of surgery, radiation therapy and chemotherapeutic TMZ.315 Rather, it was hoped that this therapy would provide another adjunct to current convention. There is evidence of BBB disruption by non-invasive modalities that increase NP uptake in the brain, although evidence is lacking for lanthanum based NPs.316 X-ray radiation317 and ultrasound318 have been successfully demonstrated to increase silver and gold NP uptake across the BBB in brain tumour models respectively. In order to better model the biological behaviour of this NP therapy, confirmation was needed that neither radiation nor TMZ affected NP uptake which would have consequences on optimizing in vivo dosage.

5.1.3 Long-term persistence

The stability of REO NPs is advantageous should the NPs be shown to successfully penetrate the in vivo BBB and reach the target GBM site without significant metabolism or

139

break down. Yet, another consequence of this is biological persistence. In the liver and spleen,

319,320 321 REO CeO2 NPs and REE gadolinium nanostructures formulations have been reported up to 90 days after final dose in these organs. Such a finding raises concerns for the potential of long-term toxicity from this La2O3 NP therapy, however, is yet to thoroughly established.

Furthermore, there has been little study in the longevity of REO NPs in the brain, assuming sufficient acute accumulation. Understanding of how the La2O3 NP therapy persists in a biological system is required to ensure that adverse effects are avoided when titrating for the in vitro observations into an in vivo system. Particular attention to the long-term profile in the brain is required due to the location of GBM and the associated neurological adverse effects associated with some NP therapies and free-ion lanthuam.322,323

5.1.4 Objective and aims

The objective of this chapter was to establish the short- and long-term biocompatibility and biodistribution of the La2O3 NP therapy to prepare for future xenograft GBM survival study.

Specifically, the aims of this chapter were:

1. To establish whether or not La2O3 NP therapy can infiltrate the brain by IV

administration.

2. To determine the short- and long-term tolerance and biodistribution of the La2O3

NP therapy in a biological system.

3. To evaluate the effect of adjuvant radiation therapy and chemotherapy on the

biological profile of the La2O3 NP therapy.

140

5.2 Results

5.2.1 Delivery to the brain

In order to validate La2O3 NPs as a viable therapy for GBM, evidence that the NP could access the brain via IV administration was required. Using a preliminary 1 mg/kg dose that

310 had been previously shown to be non-toxic in mice for various REO NPs , the La2O3 NP was administered in 4x weekly doses to a balb/c mouse model. By ICP-MS, the final post-mortem brain lanthanum concentration was determined to be 0.143 ±0.013 µg/g of organ one week after the final dose, which was significantly greater than control indicating presence that the

La2O3 NP therapy had successfully infiltrated the brain. There were no changes in mice weight or behaviour over the course of the 4 weekly doses (Figure 5.1).

5.2.2 Short-term maximum tolerable dose

Following on observation that the La2O3 NP could be detected in the brain at 1 mg/kg dose without weight or behaviour changes, the highest tolerable dose for the therapy following the regimen of 4x weekly doses was tested in a dose escalation manner, at first 3 mg/kg, and if no tolerance concerns developed, then 5 mg/kg.

5.2.2.1 Body weight

Body weight of mice were measured one week pre-regimen, weekly during regimen, and then one week post-regimen (Figure 5.1). Mean pre-regimen weights for the control, 1 mg/kg, 3 mg/kg and 5 mg/kg doses were statistically comparable, as too were the post-regimen weights. Compared to each other, pre- and post-regimen weights of each dose regimen were not statistically different to that of the control.

All weekly weights during each NP dosed regimen were also not statistically different to control (Supplementary Table 5.1), except for the 5 mg/kg dose at Dose 1 compared to

141

22 Control 1 mg/kg

) 3 mg/kg

g 20 (

t 5 mg/kg

i e

W 18

16 n 1 2 3 4 n e e e e e e m s s s s m i o o o o i g D D D D g re re - t- re s P o P Time

Figure 5.1 Body weights for maximal tolerable dose.

Mean body weight over the course of NP therapy regimen of 4 weekly doses administered in escalating fashion to determine highest tolerable dose. Pre- and post-regimen measurements taken one week before Dose 1 and one week after Dose 4 respectively There were n=3 mice per dose at each measurement. Weights are presented as mean ±SD. Statistical analysis was performed using student t-test to compare mean weights of each dose to corresponding mean weight of control, with significance set to P<0.05. *Weight for 5 mg/kg regimen at Dose 1 was significantly higher compared to control (MD, 1.8 g; P=0.009) and the 1 mg/kg regimen (MD, 1.5 g, P=0.029).

142

control (MD, 1.77 g heavier). It is noted that the 5 mg/kg dose mice were 2 weeks older than that of the control mice at time of regimen commencement, and all subsequent weekly weights were comparable from Dose 2 onwards to post-regimen.

5.2.2.2 Brain and major organ weights

One week post-regimen, the mean weights of the brain, liver, kidneys and spleen dosed at 5 mg/kg dose vs control were compared, with only the liver (6.47 ±0.26 vs 6.86 ±0.19 % total body weight respectively) being an organ that was significantly different between the two cohorts (Figure 5.2). For the organs measured in the 1 and 3 mg/kg dose cohorts, compared to control, only the 3 mg/kg dose showed a statistically significant difference in liver weight vs control, 6.94 ±0.24 vs 6.47 ±0.26 % total body weight respectively. All other organ weights were statistically comparable (Supplementary Table 5.2). At extraction, there were no signs of atrophy, ischemia, or inflammation involving any of the organs.

5.2.2.3 Nanoparticle concentrations

Lanthanum was detected in all investigated cohort organs upon post-mortem ICP-MS one week post-regimen, with concentration positively correlated to the escalating doses administered in all organs. The mean concentrations of lanthanum in the brain (Figure 5.3A), liver (Figure 5.3B), kidneys (Figure 5.3C) and spleen (Figure 5.3D) were all statistically greater than control. For all organs measured in the 1 and 3 mg/kg dose cohorts, lanthanum concentrations were all significantly elevated in comparison to the control cohort, with the 3 mg/kg dose cohort all showing higher levels compared to control as well (Supplementary

Table 5.3).

Finally, in terms of total administered dose, by 1 week after final dose of a four 1, 3, and 5 mg/kg weekly dose-regimens, 1x10-3 ±5x10-5, 5x10-3 ±5x10-4, 1.8x10-2 ±3x10-3 % of

143

10 )

l Brain a

t *

o Liver

t 8 * /

n Kidney

a g

r 6 Spleen

, %

( 4

h g

i 2

e W 0 l l l l o 1 3 5 o 1 3 5 o 1 3 5 o 1 3 5 tr tr tr tr n n n n o o o o C C C C NP dose (mg/kg, weekly for 4 weeks)

Figure 5.2 Organ weights for maximal tolerable dose.

Mean weight of organs extracted one week post-regimen of escalating NP doses of control, 1, 3 and 5 mg/kg, administered weekly for 4 weeks. Weights are presented as mean ±SD. Statistical analysis was performed using student t-test to compare mean weights of each dose to corresponding control, with significance set to P<0.05 and marked by (*).

144

A Brain B Liver

1.5 n 50

)

) t

1.2 t 40

r n

0.9 n 30

m g

0.6 g 20

(

( h

0.3 h 10

a L 0.0 L 0 l Control 1 3 5 o 1 3 5 tr n NP dose (mg/kg, weekly for 4 weeks) o C NP dose (mg/kg, weekly for 4 weeks)

C Kidney D Spleen

n 10 100

)

) t

8 t 80

r n

6 n 60

m g

4 g 40

(

( h

2 h 20

a L 0 L 0 l l o 1 3 5 o 1 3 5 tr tr n n o o C C NP dose (mg/kg, weekly for 4 weeks) NP dose (mg/kg, weekly for 4 weeks)

Figure 5.3 Organ lanthanum concentrations for maximal tolerable dose.

Mean lanthanum concentrations (µg/g of organ) of major organs A. brain, B. liver, C. kidney, and D. spleen in mouse model one week post-regimen of four weekly NP doses control, 1, 3 and 5 mg/kg measured by ICP-MS. Each cohort consisted of n=3 mice. Concentrations are presented as mean ±SD. Statistical analysis was performed using student t-test to compare mean weights of each dose to corresponding control, with significance set to P<0.05. All comparisons to control were statistically significant.

145

total administered La2O3 NP dose reached the brain, which were all significantly different to each other.

5.2.3 Pharmacokinetics of nanoparticle

In order to elucidate how likely the La2O3 NP was to persist in the brain during therapy, the pharmacokinetics were determined by modelling changes in brain lanthanum concentrations by ICP-MS at 0, 1, 24, 72 and 120 hours after a single dose of 5 mg/kg. The presence of lanthanum in the brain was detected within the first hour of administration, which peaked by 24 hours (7x10-3 % total administered dose), and was still present in the brain 120 hours after (Figure 5.4). Noncompartmental analysis indicated area under the curve (AUC) to be 104 µg/g∙h and half-life (t1/2) to be 164 hours. All other calculated pharmacokinetic parameters are listed in Table 5.1.

5.2.4 Long-term safety profile

In preparation for a GBM xenograft model, the time course of survival and management in tumour-free mice was simulated, in which La2O3 NP therapy would commence at 40 days after xenograft implantation. This included combinations with radiation therapy (RT) and chemotherapy (CT) with temozolomide (TMZ). The median survival of this model has been previously determined to be 90 days, and thus the mice were followed for 60 days after the final dose to ensure that no adverse long-term effects by the La2O3 NP therapy would bias interpretation of any survival effects observed in future studies.

5.2.4.1 Body weight

Weekly body weight measurements were taken for 60 days after the last dose of a four weekly 5 mg/kg dose regimen commenced at Day 40 (Figure 5.5). Statistical analysis did not demonstrate any significant differences in body weights during NP regimen and 60

146

n 0.5

t a

r 0.4

)

c r

n 0.3

m 0.2

g n

µ 5 mg/kg dose

( a

h 0.1

n a

L 0.0 0 24 48 72 96 120 Time after injection (h)

Figure 5.4 Brain lanthanum concentration over time.

Mean lanthanum concentration of the brain over time after one injection of 5 mg/kg as determined by ICP- MS at 0, 1, 24, 72 and 120 hours after injection. Each cohort consisted of n=3 mice. Concentrations are presented as mean ±SD. Noncompartmental modelling calculated a half-life of 164 hours.

147

Table 5.1 Pharmacokinetics of lanthanum nanoparticle in brain.

Calculated pharmacokinetic parameters by noncompartmental analysis for our NP in the brain based on lanthanum concentrations measured by ICP-MS after one dose of 5 mg/kg injection. Modelling performed by PKSolver 2.0 (Nanjing, China) software.

Parameter Symbol Value Unit -3 -1 Terminal rate constant 흀z 4.23 x10 h

Terminal half-life t1/2 164 h

Maximum concentration Cmax 0.41 µg/g

Time to Cmax Tmax 24.0 h Area under curve AUC 104 µg/g∙h Area under the first movement curve AUMC 2.52 x105 µg/g∙h2 Mean residence time MRT 241 h

Volume of distribution during terminal phase Vz 11.3 (mg/kg)/(µg/g) Clearance Cl 0.048 (mg/kg)/(µg/g)/h

Volume of distribution at steady state Vss 11.6 (mg/kg)/(µg/g)

148

26 Control

24 NP only

) RT only

g ( 22

t CT only

h g i NP+RT e 20

W NP+CT 18

16 0 ) ) ) ) 5 6 7 8 9 0 1 k 1 2 3 4 k k k k k 1 1 e e e e e e e e e e k k e s s s s e e e e e e e o o o o e e W D D D D W W W W W ( ( ( ( W W 1 2 3 4 k k k k e e e e e e e e W W W W Time

Figure 5.5 Body weights for biocompatibility.

Mean body weight progression of balb/c mice model managed by NP therapy (5 mg/kg), RT therapy (4 Gy) and CT (temozolomide, 10 mg/kg), separately and in combination. NP therapy administered for four doses (Weeks 1-4), RT administered as a single dose (Week 1), and CT administered for four doses (Weeks 1-4). Weights were taken weekly, until 60 days after the last dose. Each cohort consisted of n=3 mice. Weights are presented as mean ±SD. Statistical analysis was performed using student t-test to compare mean weights of each dose to corresponding control, with significance set to P<0.05 and marked by (*). However, no statistical difference in body weight at any time point between any of the cohorts.

149

days afterwards between control, NP only, RT only, CT only, NP+RT and NP+CT cohorts (all

P>0.05).

5.2.4.2 Brain and major organ weights

Sixty days after the last dose of a 5 mg/kg NP dose regimen, mean weights of brain, kidney and spleen in the control vs NP only cohorts were statistically comparable, while the liver was significantly heavier (5.37 ±0.33 vs 6.22 ±0.19 % total body weight respectively)

(Figure 5.6). In terms of the other experimental cohorts, the only significant differences compared to the control occurred in the liver, with both combinations NP+RT (6.34 ±0.14 % total body weight) and NP+CT (6.52 ± 0.27 % total body weight). There was no statistical difference within cohorts that involved NPs, and no difference within cohorts that did not

(Supplementary Table 5.4). At extraction, there were no signs of atrophy, ischemia, or inflammation involving any of the organs.

5.2.4.3 Nanoparticle concentrations

Sixty days after the last dose of a 5 mg/kg NP dose regimen, lanthanum concentrations in the brain in the control, NP only, RT only, CT only, NP+RT and NP+CT cohorts were all significantly greater than control (Figure 5.7). There was no statistical difference within cohorts that involved NPs, and no difference within cohorts that did not involve NPs

(Supplementary Table 5.5). Furthermore, the lanthanum concentration in the liver, kidneys and spleen between the control vs NP only cohorts sixty days after the last dose were evaluated, demonstrating similarly significantly greater concentrations than control (Figure 5.8).

In terms of total administered dose, by 60 days after final dose, 1.1x10-2 ±1x10-3 % of administered dose reached the brain, which is significantly lower than 1.8x10-2 ±3x10-3 % at 1 week after final dose.

150

) l

a 8 *

t Brain o

t * /

n Liver

a 6 g

r Kidneys

, Spleen %

( 4

g i

e 2 W

0 l l l l o ly ly ly T T o ly ly ly T T o ly ly ly T T o ly ly ly T T tr n n n R C tr n n n R C tr n n n R C tr n n n R C n o o o + + n o o o + + n o o o + + n o o o + + o P T T P P o P T T P P o P T T P P o P T T P P C N R C N N C N R C N N C N R C N N C N R C N N Cohort

Figure 5.6 Organ weights for biocompatibility.

Mean organ weight of balb/c mice model 60 days after initiation of NP therapy (5 mg/kg), RT therapy (4 Gy) and CT (temozolomide, 10 mg/kg), separately and in combination. NP therapy administered weekly for four doses (Weeks 1-4), RT administered as a single dose (Week 1), and CT administered weekly for four doses (Weeks 1-4). Each cohort consisted of n=3 mice. Weights are presented as mean ±SD. Statistical analysis was performed using student t-test to compare mean weights of each dose to corresponding control, with significance set to P<0.05 and marked by (*). All therapies involving NP showed significantly greater weight than control in the liver (not marked).

151

Brain

n 1.5

)

e g

c 1.0

/ u

g 0.5

(

n a

L 0.0 l o ly ly ly T T tr n n n R C n o o o + + o P T T P P C N R C N N Cohort

Figure 5.7 Brain lanthanum concentrations for biocompatibility.

Mean brain lanthanum concentration (µg/g of organ) of balb/c mice model 60 days after initiation of NP therapy (5 mg/kg), RT therapy (4 Gy) and CT (temozolomide, 10 mg/kg), separately and in combination measured by ICP-MS. NP therapy administered weekly for four doses (Weeks 1-4), RT administered as a single dose (Week 1), and CT administered weekly for four doses (Weeks 1-4). Each cohort consisted of n=3 mice. Concentrations are presented as mean ±SD. Statistical analysis was performed using student t-test to compare mean weights of each dose to corresponding control, with significance set to P<0.05. All therapies involving NPs were significantly higher than all therapies not involving NPs, but statistically comparable amongst themselves.

152

Based on the findings that lanthanum was still present in the in vivo system 60 days after last injection, the lanthanum concentrations of the brain (Figure 5.9A), liver (Figure

5.9B), kidneys (Figure 5.9C) and spleen (Figure 5.9D) were compared between 1 week

(Section 5.2.2.3) and 60 days after last injection in the control and NP only cohorts. There was no significant difference between lanthanum concentration of all control organs between 1 week and 60 days. In terms of NP cohorts, there was a significant reduction in lanthanum concentration in the spleen at 60 days vs 1 week (63.0 ±7.30 vs 22.9 ±5.49 µg/g of organ).

However, with respect to the brain, liver and kidneys, despite decreasing trends over time, there were no significant differences in lanthanum concentration between these two time points.

5.2.4.4 Brain histology

Haematoxylin and eosin (H&E) histological staining of brain samples from the control and NP only (4x weekly 5 mg/kg doses) cohorts were performed to analyse if any histological disturbance could be detected in the brain due to the presence of La2O3 NPs 60 days after final dose. At both 20X and 40X magnifications, there appeared to be no obvious differences in cell size or shape (Figure 5.10).

153

n 40 Liver

i t

a Kidneys

)

t n

n 30 Spleen

o f

c 20

µ a

( 10

n a

L 0 l l l o ly o ly o ly tr n tr n tr n n o n o n o o P o P o P C N C N C N Cohort

Figure 5.8 Organ lanthanum concentrations for biocompatibility.

Mean lanthanum concentration (µg/g of organ) of liver, kidneys and spleen in balb/c mice model 60 days after initiation of NP therapy (5 mg/kg) measured by ICP-MS. NP therapy administered weekly for four doses (Weeks 1-4). Each cohort consisted of n=3 mice. Concentrations are presented as mean ±SD. Statistical analysis was performed using student t-test to compare mean weights of each dose to corresponding control, with significance set to P<0.05. All NP only groups were statistically greater than respective control.

154

A Brain B Liver

n 1.5 50

)

) t

t 40

g c

c 1.0

r n

n 30

m g

g 20

u g

g 0.5

(

( h

h 10

a L 0.0 L 0 ) ) ) ) ) ) ) ) k s k s k s k s e y e y e y e y e a e a e a e a w d w d w d w d 0 0 0 0 (1 6 (1 6 (1 6 (1 6 l ( ( l ( ( o l P l P r o N P ro o N P t tr N t tr N n n n n o o o o C C C C Cohort Cohort

C Kidney D Spleen

n 10 100

t a

a *

)

) t

t 8 80

r n

n 6 60

m g

g 4 40

(

( h

h 2 20

a L L 0 0 ) ) ) ) ) ) ) ) k s k s k s k s e y e y e y e y e a e a e a e a w d w d w d w d 1 0 1 0 1 0 1 0 ( (6 ( (6 ( (6 ( (6 l l P l l P ro o N P ro o N P t tr N t tr N n n n n o o o o C C C C Cohort Cohort

Figure 5.9 Comparison of organ lanthanum concentrations between two timepoints.

Mean lanthanum concentrations (µg/g of organ) of major organs A. brain, B. liver, C. kidney, and D. spleen in balb/c mice model 60 days after initiation of NP therapy (5 mg/kg) measured by ICP-MS. NP therapy administered weekly for four doses (Weeks 1-4). Each cohort consisted of n=3 mice. Concentrations are presented as mean ±SD. Statistical analysis was performed using student t-test to compare mean weights of each dose to corresponding control, with significance set to P<0.05 and marked by (*).

155

A B

C D

Figure 5.10 Histological staining of brain specimens.

Representative histological H&E staining of brain specimens from control cohort A. 20X and B. 40X, and nanoparticle (NP) only cohort C. 20X and D. 40X,

156

5.3 Discussion

A series of in vivo studies were performed to evaluate the biocompatibility of the La2O3

NP therapy in a mouse model in preparation for a xenograft GBM model in the future. The initial results were promising, as they demonstrated that the NP can be detected in the brain after IV administration, and that a weekly dose of 5 mg/kg for four weeks could be tolerated without overt adverse effects. Furthermore, NP concentration in the brain peaked at 24 hours after administration, but did not completely dissipate within one week. In fact, it was demonstrated that at 60 days after 4 weekly doses of 5 mg/kg, the NP was still detectable at comparable levels in the brain, as well as other major organs including the liver and kidneys.

However, it appears histologically this does not result in any adverse effects in the normal brain.

Finally, the addition of adjuvant therapies did not appear to hinder the long-term biodistribution of the NP, indicating the compatibility between them and the La2O3 NP therapy.

5.3.1 The nanoparticle regimen is tolerable

There was no indication of weight loss throughout the course dose escalation to 5 mg/kg at both 1 week and 60 days after final regimen dose, which was one of the primary morbidity

305 concerns with higher doses of the La2O3 NP therapy. Additionally, other signs of acute toxicity including changes in behaviour, tremor, seizures, decreased activity, discoloration and piloerection were not noted in any mouse exposed to La2O3 NP therapy throughout the series of studies in this chapter. These observations are promising as weight loss324 and behavioural impairment323 have been reported in previous animal models exposed to higher doses (>20 mg/kg) of free-ion lanthanum (nitrate and chloride respectively). Collectively, at the very least, the proposed La2O3 NP therapy of four weekly 5 mg/kg doses appears tolerable in vivo, and validates the IV delivery route as a viable pathway to target intracranial GBM.

157

5.3.2 Lanthanum nanoparticles reach the brain

A key finding of this chapter was that the La2O3 NP therapy was able to infiltrate the

BBB, and enter the brain tissue in a dose-dependent manner by IV administration. It confirms the findings of other in vivo animal studies 82-84,325 that showed IV delivery of NPs up to 160 nm in size can infiltrate the brain matter in a similar manner. However, these were not lanthanum-based NPs, and to date, there has been no investigation in the literature evaluating how La2O3 NP specifically targets the brain in vivo, which makes the presented findings

310 extremely novel. It is somewhat supported by the provision of Das et al. that <1% of La2O3

NP was likely to be detectable in the brain, although they did not provide any further quantification. The measurement by ICP-MS to detect lanthanum in the brain after IV administration provides greater certainty that La2O3 NPs indeed can reach the brain, and ultimately, target in-situ GBM moving forward.

5.3.2.1 Stability in the brain

Via serial monitoring of lanthanum concentrations after one 5 mg/kg dose, statistical modelling calculated multiple pharmacokinetic parameters of the La2O3 NP therapy in the brain organ itself: terminal half-life of 164 hours, concentration peaking by 24 hours, an area under curve (AUC) of 104 µg/g∙h, and a mean residence time (MRT) of 241 hours. These values compare differently to that of conventional TMZ in the brain, which has a half-life of 2-3 hours, peaking at 2.0 hours and an AUC of 2.7 µg/mL∙h in GBM patients.326,327 In fact, it would appear the La2O3 NP therapy may be the exception in terms of brain pharmacokinetics of NP therapies.

Gulyaev et al.328 reported their polymer-based NP had a brain half-life of 1.56 hours, peaking within the first 4 hours of administration, with an AUC of 35 µg/g∙h and MRT of 4 hours in their mice model. A lipid NP by Yang et al.313 showed a AUC of 5.14 µg/g∙h with a MRT of

28 hours in the brain after intravenous injection. Different physical and chemical properties of

158

the NP formulation are likely to contribute to these differences, and highlight their importance in characterising the pharmacokinetics of different NP therapies targeting the brain.

There are then a number of implications for the proposed La2O3 NP therapy based on its remarkable persistence in the brain compared to other therapies.329 When evaluating the optimal dosage for effect in the future, the administered dose to accumulate therapeutic mass could be lower than those therapies that are not so stable, as well as its frequency of administration could be reduced. Yet, the prolonged existence of the La2O3 NP therapy in the brain interstitium raises concerns for potential neurological adverse effects, however, the unremarkable clinical course observed of this study is encouraging. Throughout the 60 days of monitoring, no behavioural changes, seizure activity, or macroscopic evidence of toxicity were observed. Additionally, histological staining did not identify any concerning microscopic changes after the NP regimen when compared to control. Although at 60 days the concentration of lanthanum had not decreased significantly, the proportion of total administered dose had, indicating that with more than 60 days follow-up, this NP could be cleared further to become negligible.

5.3.3 Biodistribution of nanoparticle outside the brain

Both 1 week and 60 days after final dose of the NP regimen, lanthanum accumulation was noted in the liver, kidneys and spleen. These organs were chosen based on implications

329 310 for clearance. In their study, Das et al. noted the presence of La2O3 NPs in these exact same organs, with increasing concentrations from kidney, to liver, to spleen after their final dose. This is the same sequence observed in this study after the final regimen dose. Indeed, it would appear that localisation to these organs by REO NPs is generalisable for the most part, with similar trends shown by them with Gd2O3, CeO2, Sm2O3, Er2O3, Eu2O3, Nd2O3 and Dy2O3

NPs. Within REO NPs alone, in vivo accumulation in these organs outside the brain has been

159

330-333 confirmed by a number of other studies , confirming the La2O3 NP therapy conforms to the basic expected biodistribution pattern of REO NP.

Similar to the brain, there was retention of all organs 60 days after final dose of the NP regimen in this study. Such a result has not been reported previously for La2O3 NP, although the literature reporting findings of the other NPs indicates this is not a surprising finding.

Gereats et al.311 observed longevity in the liver and spleen 90 days after last injection of a titanium oxide NP, with no significant reduction in organ concentration from 1 week to 60 days.

319 In the study by Yokel et al. , they did not notice any significant reduction in CeO2 NPs from

1 week to 90 days after final NP dose in any of the three organs. Interestingly, they noted a reduction in brain concentration 1 week to 30 days after administration, which was not observed in our studies after 60 days.

In fact, another difference in the longevity of the La2O3 NP is that by 60 days, there was a statistically significant reduction in lanthanum concentration in the spleen. It would appear that general trends for NP retention very greatly between NPs, because when Fraga et al.307 examined a gold NP, they saw decreases in NP concentration in all organs by 28 days, whereas Smulders et al.308 who used a silicon NP actually reported an increased in NP in the spleen by day 84 compared to day 7 after final injection. Based on these heterogenous trends in the literature, when one considers all the possible variations in NP chemistry and dosing regimens, comparative conclusions may be most valid when considering studies whose conditions mirror those of the ones being tested. Due to the novel nature of La2O3 NP therapy, such values have not been published to date.

160

5.3.4 Systemic considerations

5.3.4.1 Liver

One consistent observation was that after 4 weekly NP doses of 5 mg/kg, compared to the control, the liver was statistically enlarged; both at 1 week and 60 days later. This does bring attention to the state of the liver when managing the La2O3 NP therapy, however, it should also be recognised this statistical significance may not necessarily indicate a pathological hazard due to the proximity of the absolute values and lack of systemic signs.

Das et al.310 evaluated liver weight at the immediate cessation of a 6 weekly regimen of 5 mg/kg La2O3 NP, and observed a statistical decrease in liver weight compared to control.

This contrary finding suggests that organ mass may not be more the valid surrogate marker for

NP toxicity outside identifying potential organs to analyse further. They also employed liver function tests for alanine aminotransferase (ALT) and histology, which in the cohort exposed to NP, demonstrated asymptomatic increased levels and swollen hepatocytes respectively. As inflammation was not investigated in this study to a similar depth, subacute hepatitis at the administered 5 mg/kg dose is unlikely but cannot be ruled out.

Given the predilection of NPs in general to accumulate in the liver, other studies have evaluated liver health after NP administration with inconsistent results. In the study by Brabu

274 et al. evaluating La2O3 NPs, they did not observe any acute significant changes in liver enzymes or histopathology after administration at a higher dose compared to the regimen of this study. Heckman et al.325 found no evidence of liver pathology on histological study 30 days after injection of REO NP CeO2. A telling study in favour of transient liver response to

309 NP therapy was by Tseng et al. , who observed a rise in ALT with NP CeO2 1 day after last injection, however by day 30 this had normalised. Thus, these results advocate that liver response to NP therapy, its severity and duration, is likely dependent on a number of factors,

161

and that it is possible to engineer NP aspects such as size and charge to facilitate greater clearance, or lower regimen dosage, in order to reduce the chances of a liver reaction.334

5.3.4.2 Kidneys

Accumulation of the La2O3 NPs in the kidneys was observed at both 1 week and 60 days after final dose of the regimen based on lanthanum concentrations. This raises concerns of nephrotoxicity, although the lack of any significant changes in organ weight or macroscopic appearance is encouraging. Referencing the study of multiple REO NPs by Das et al.310, they noted that La2O3 did not increase blood urea nitrogen (BUN), a common marker of kidney health. In fact, of the 10 REO NPs tested, only Tb4O7 and Gd2O3 NPs resulted in statistically higher levels of BUN compared to control. They posited that these effects were not so much a result of the NP therapy itself, but rather the formulation of REO NPs used. For Tb4O7, the possible presence of chromium impurities would have increased BUN for chromium is a proven nephrotoxic agent.335 Gadolinium is known to have nephrotoxic properties at high concentration.302 Other NP formulations have been investigated for nephrotoxicity, with findings similarly inconsistent; an Yb-based NP showing increased creatinine 30 days after final dose331, whereas an iron oxide NP showed no signs of kidney damage 7 days after final dose336.

As blood markers for kidney damage were not investigated in this study, the possibility of subacute nephritis cannot be discounted. Nonetheless, the likelihood of severe renal failure being present during this La2O3 NP regimen is very low based on the reported observations.

One final consideration is that lanthanum carbonate is used to manage end-stage renal disease in humans, which would be counterintuitive if it possessed serious nephrotoxic properties. In fact, 10-year data has demonstrated no significant risk of adverse renal toxicity in human patients with suboptimal renal function to begin with.337

162

5.3.4.3 Spleen

Of all the organs examined, the spleen was the only one which demonstrated a significant reduction in lanthanum concentration at 60 days compared to 1 week after final dose, with steady weight across both time points. Das et al.310 remains the only study to have analysed spleen weight after La2O3 NP, and they found immediately after last dose, that there was a significant increase in spleen weight. However, based on their studies of other REO NPs, there were significant decreases, and non-significant changes as well. Thus, any general trend in spleen weight after REO NP therapy appears difficult to predict. If there was indeed serious toxicity to the spleen, clinical signs of asplenia, including infection and lethargy due to anaemia would be expected in the long-term. Neither of those two signs were observed in the mice 1 week or 60 days after last dose, and they remained active throughout the course of monitoring.

No study has linked La2O3 NP therapy to various haematological parameters, however

Su et al.338 and Tiwari et al.339 evaluated white blood cells (WBCs), red blood cells (RBCs), haemoglobin (Hb), and platelets after injection of titanium oxide and low-dose silver NPs respectively, and neither observed any significant differences with control with steady weights.

Although, with a high-dose silver NP regimen, Tiwari et al.339 did report significant reductions in all these parameters, indicating that splenic function can be adversely affected by NP therapy if administered at elevated doses. Such doses would appear greater than the 5 mg/kg used in the presented La2O3 NP regimen. Future studies will benefit from repeat bloods for deranged blood parameter during titration of optimal NP dose for cytotoxic effect in GBM xenograft models to ensure this is the case.

163

5.3.5 Therapeutic consequences

5.3.5.1 Dosing optimisation

It is recognised that as a potential anti-GBM therapy, the dose of 5 mg/kg may not be sufficient to cause maximal cell death in GBM. The case of TMZ illustrates this point, such that the standard dose as approved by the EORTC/NCIC trial in 2005, does not eradicate all

GBM cells, as much a theoretical dose would be toxic to the patients.340 With the current TMZ regimen, nausea and vomiting are common, with severe haematological and liver enzyme derangements are expected to occur in up to 10% of all patients.315,341

In reference to the La2O3 NP therapy, although a regimen of 5 mg/kg may induce asymptomatic liver enlargement and possible enzyme derangement, it is not certain that this dose is sufficient to cause anti-GBM effects as observed in Chapter 3 until titration in a xenograft GBM model is pursued. Analysis indicated 0.02% of the total NP administered over the four-week regimen at 5 mg/kg was in the brain 1 week later. In the case that it is not sufficient, the possibility of titrating to a higher dose at the risk of systematic complications should be considered if anti-GBM effects are still desired.

5.3.5.2 Monitoring

Concern of toxicity to the liver, kidneys and spleen promotes that they be either directly or indirectly monitored in future studies to ensure no serious adverse effects are induced by the administered dose. As the literature suggests, many of these complications are dependent on the regimen of NP therapy selected, as well as the NP design itself.274,312 Blood serum monitoring presents as the most convenient method to do so, with liver, kidney and spleen health inferable by measuring ALT, BUN/creatinine, and complete blood count parameters respectively. Alternatively, non-invasive imaging such as ultrasound could also detect severe

164

toxicities including organ atrophy and fibrosis, however, by this point derangement would be expected upon blood serum analysis, as well as overt behavioural changes.

5.3.5.3 Viability with adjuvant therapies

When RT was administered simultaneously with the first dose of the La2O3 NP regimen, brain lanthanum concentration was comparable to that of the NP only cohort. This indicated that concurrent RT did not affect the uptake of NP into the brain, or any other organ for that matter. Using in vivo models, Tamborini et al.317 only observed increased NP uptake with pre-

NP radiation exposure in their GBM xenograft models, and not healthy brain, a finding they attribute to the enhance permeation and retention (EPR) effect. Given the RT simulated in this study was concurrent RT for the purpose of augmenting the therapeutic production of ROS by

NPs, rather than facilitate greater NP uptake, increasing brain uptake and therapeutic effect by pre-NP radiation remains a distinct possibility with this NP therapy.342

There were no changes in lanthanum concentrations in all organs when either RT or

TMZ was administered concurrently with NP, supporting the notion that NP pharmacokinetics in the brain and other major organs were unlikely to be affected. Consequently, it generates confidence to investigate biological augmentation of RT and TMZ by La2O3 NP therapy in future xenograft models without the concern of potential synergy being interrupted by adjuvant therapy side-effects.

5.3.6 General limitations

The assessment of how likely the demonstrated in vitro anti-GBM effects will translate to the in vivo environment is challenging as there is potential for significant change in NP behaviour an in vivo model versus an in vitro model. As charged entities, NPs will interact with solvent molecules forming protein coronae and/or adsorption of ions on the surface, which

165

distorts overall surface charge and any potential interaction with cells.343 When comparing in vitro cell culture medium versus in vivo blood plasma, the solvent composition is different and

344,345 so too will be the ultimate surface charge on the La2O3 NP. This has significant implications in the in vivo model as surface charge of the NP effect has been shown to affect infiltration and permeation across the BBB.346 The expectations of directly observing in vitro anti-GBM effects of the La2O3 NP therapy in vivo must be tempered accordingly.

Another consideration to address is possible lanthanum present in the blood during organ extraction, which may have distorted the true concentration of lanthanum in the supplied organ tissues.347 Although perfusion of blood prior to organ extraction was not possible in the experimental setting, it is a valid recommendation for future studies. It is worth noting that it would not appear likely that the La2O3 NP calculated by ICP-MS to be present in brain would have been exclusively in brain vasculature. This is because blood capillaries themselves represent only approximately 1% of total brain volume, and the brain blood vessel endothelial cells 0.1%.312 By proportions, this would equate to an intravenous concentration of 97-970 mg/kg based on the results at 1 week after final dose (100-1000x 0.97 µg/g), which is improbably given the sum of the NP regimen over four weeks was 20 mg/kg (4x 5 mg/kg).

Although the assumption that lanthanum measured in extracted organs were present in the tissue as La2O3 NPs, it cannot be fully discounted that there was a baseline presence in the blood too at extraction, even 60 days after last dose due to the remarkable biopersistence of the

NP.348

166

5.4 Conclusions

In this chapter I explored how compatible the La2O3 NP therapy would be in a biological system. There is paucity in the current literature describing the biodistribution and toxicity profile of this specific NP therapy, and it was ultimately shown to follow trends of other similar sized formulations. There are potential concerns for liver, kidney and spleen toxicity, however the findings reported in this chapter and the literature would suggest this is regimen-dependent, and not likely a serious concern for the tested regimen. Nonetheless, thorough monitoring by blood parameters is encouraged moving forward to validate this interpretation. More importantly, evidence that La2O3 NP therapy can infiltrate the BBB, and remain available in- situ to target intracranial GBM, is exciting. Overall, the proposed novel NP therapy appears feasible in a mouse model, and that this biocompatibility and biodistribution permits titration for anti-GBM effect in future GBM xenograft models.

167

5.5 Supplementary

5.5.1 Supplementary Tables

Supplementary Table 5.1

Summary of mean body weight (g) or mean difference (MD) in weights compared to control dose of all mice over the course of NP therapy regimen of 4 weekly doses administered in escalating fashion to determine highest tolerable dose of 1, 3, and 5 mg/kg. Pre- and post-regimen measurements taken one week before Dose 1 and one week after Dose 4 respectively. There were n=3 mice per dose at each measurement. Control mean weight is presented as mean ±SD. Statistical analysis was performed using student t-test to compare mean weights of each dose to corresponding control, with significance (bold) set to P<0.05.

Dose Control 1 mg/kg 3 mg/kg 5 mg/kg Time Weight (g) MD P-value MD P-value MD P-value Pre-regimen 17.8 ±0.46 -0.57 0.71 0.50 0.78 0.50 0.78 Dose 1 17.3 ±0.51 0.23 0.97 1.20 0.12 1.77 0.01 Dose 2 19.0 ±0.23 -0.53 0.75 -0.20 0.98 0.03 0.99 Dose 3 19.0 ±0.46 0.17 0.99 0.80 0.44 0.30 0.94 Dose 4 20.0 ±0.11 -0.63 0.64 -0.37 0.90 -0.33 0.92 Post-regimen 20.3 ±0.38 -0.73 0.52 -0.20 0.98 -0.40 0.88

168

Supplementary Table 5.2

Summary of mean organ weight (% total body weight) or mean difference (MD) in weights compared to control dose over the course of NP therapy regimen of 4 weekly doses administered in escalating fashion from control to 1,3 and 5 mg/kg. Measurements taken one week after Dose 4 respectively. There were n=3 mice per dose at each measurement. Control mean weight is presented as mean ±SD. Statistical analysis was performed using student t-test to compare mean weights of each dose to corresponding control, with significance (bold) set to P<0.05.

Dose Control 1 mg/kg 3 mg/kg 5 mg/kg Weight (% total Dose MD P-value MD P-value MD P-value body weight) Brain 2.10 ±0.15 0.02 0.99 -0.03 0.99 -0.07 0.95 Liver 6.47 ±0.26 0.01 0.99 0.47 <0.01 0.39 0.03 Kidneys 1.56 ±0.04 -0.05 0.98 0.003 0.99 -0.07 0.96 Spleen 0.44 ±0.01 0.002 0.99 0.20 0.47 0.15 0.70

169

Supplementary Table 5.3

Summary of mean lanthanum concentration (µg/g of organ) or mean difference (MD) in concentrations compared to control dose over the course of NP therapy regimen of 4 weekly doses administered in escalating fashion from control to 1,3 and 5 mg/kg. Measurements taken one week after Dose 4 respectively. There were n=3 mice per dose at each measurement. Control mean weight is presented as mean ±SD. Statistical analysis was performed using student t-test to compare mean weights of each dose to corresponding control, with significance (bold) set to P<0.05.

Dose Control 1 mg/kg 3 mg/kg 5 mg/kg Lanthanum Dose concentration MD P-value MD P-value MD P-value (µg/g of organ) Brain 5x10-4 ±2x10-4 0.14 0.01 0.43 <0.01 0.97 <0.01 Liver 7x10-4 ±2x10-4 5.37 0.01 12.5 <0.01 28.8 <0.01 Kidneys 6x10-4 ±3x10-4 1.11 0.03 2.65 <0.01 5.22 <0.01 Spleen 2x10-3 ±8x10-4 10.3 <0.01 23.7 0.04 57.0 <0.01

170

Supplementary Table 5.4

Summary of mean organ weight of balb/c mice model 50 days after initiation of NP therapy (5 mg/kg), RT therapy (4 Gy) and CT (temozolomide, 10 mg/kg), separately and in combination. NP therapy administered weekly for four doses (Weeks 1-4), RT administered as a single dose (Week 1), and CT administered weekly for four doses (Weeks 1-4). Each cohort consisted of n=3 mice. Weights are presented as mean ±SD. Statistical analysis was performed using student t-test to compare mean weights of each dose to corresponding control, with significance set to P<0.05 and marked in bold.

Cohort Control NP only RT only CT only NP+RT NP+CT Weight (% total Dose MD P-value MD P-value MD P-value MD P-value MD P-value body weight) Brain 1.99 ±0.10 0.06 0.98 <0.01 0.99 0.02 0.99 <0.01 0.99 -0.06 0.98 Liver 5.37 ±0.33 0.84 0.01 0.10 0.99 0.48 0.18 0.96 <0.01 1.15 <0.01 Kidneys 1.43 ±0.08 <0.01 0.99 <0.01 0.99 <0.01 0.99 0.02 0.93 0.01 0.97 Spleen 0.67 ±0.14 0.14 0.90 0.04 0.99 0.15 0.87 0.19 0.70 0.22 0.60

171

Supplementary Table 5.5

Summary results of statistical analysis by two-way ANOVA comparing the brain lanthanum concentration (µg/g of organ) of balb/c mice model 60 days after initiation of NP therapy (5 mg/kg), RT therapy (4 Gy) and CT (temozolomide, 10 mg/kg), separately and in combination measured by ICP-MS. NP therapy administered weekly for four doses (Weeks 1-4), RT administered as a single dose (Week 1), and CT administered weekly for four doses (Weeks 1-4). Each cohort consisted of n=3 mice. Mean difference (MD) of the concentration between the two groups is reported, along with its 95% confidence interval (CI). Statistical significance was set to P<0.05 and marked in bold.

Comparison cohorts MD (µg/g of organ) 95% CI P-value Control vs. NP only -0.8813 -1.048 to -0.7143 <0.01 Control vs. RT only 0.000161 -0.1668 to 0.1671 >0.99 Control vs. CT only 0.000176 -0.1668 to 0.1671 >0.99 Control vs. NP+RT -1.025 -1.192 to -0.8582 <0.01 Control vs. NP+CT -0.8968 -1.064 to -0.7298 <0.01 NP only vs. RT only 0.8814 0.7145 to 1.048 <0.01 NP only vs. CT only 0.8815 0.7145 to 1.048 <0.01 NP only vs. NP+RT -0.1438 -0.3108 to 0.02312 0.11 NP only vs. NP+CT -0.01547 -0.1824 to 0.1515 0.99 RT only vs. CT only 1.44x10-5 -0.1669 to 0.167 >0.99 RT only vs. NP+RT -1.025 -1.192 to -0.8583 <0.01 RT only vs. NP+CT -0.8969 -1.064 to -0.73 <0.01 CT only vs. NP+RT -1.025 -1.192 to -0.8583 <0.01 CT only vs. NP+CT -0.8969 -1.064 to -0.73 <0.01

172

CHAPTER 6

173

6 CHAPTER 6: EXECUTIVE SUMMARY AND FUTURE

DIRECTIONS

6.1 Purpose

The median overall survival of less than 2 years for glioblastoma (GBM) patients has not changed in the past decade, despite the exponential growth in research. There is an argument to be made that improvements to current therapeutic options have stagnated, and that there is a need for more novel therapies to be investigated, such as rare earth oxide (REO) nanoparticle

(NP) therapy, to better improve the current dismal prognosis of GBM.

6.2 Background

GBM is an aggressive, malignant brain tumour with dismal prognosis. Current therapeutic management is a combination of safe maximal surgical resection, followed by radiation therapy and chemotherapy by temozolomide (TMZ). Since this regimen was first introduced in 2005, we have seen very little improvement in median overall survival of 15 months.28,177,178

Novel therapies present an exciting area in GBM management that may provide superior survival outcomes if designed and evaluated properly. One such therapy is that of NPs, nanostructure formulations that interact with living systems as a physical entity, rather than a molecular compound.349 The benefits of this include interactions with a cell surface that can facilitate greater uptake. This would prove most useful in targeting intracranial tumours for currently, the endothelial blood brain barrier (BBB) presents as a significant physical barrier for molecular compounds to access GBM.350

174

With respect to NP formulation, rare earth elements (REEs) in REO form are stable compounds that can be produced as NPs. The attraction of REE as a potential therapeutic agent is their unique chemical structure, which can potentially prove cytotoxic to GBM cells, or in combination with other cytotoxic agents such as TMZ.106 Furthermore, due to this as well as their expanded electronic orbital arrangement, they can interact with incoming radiation to augment therapeutic effect.

6.3 Objectives

Given the absolute novel nature of considering REO NP therapy in GBM management, this thesis set out with three overarching objectives, which are reflected in each experimental

Chapter.

Objective 1: In Chapter 3, to assess and identify possible REO NPs that possess therapeutic potential in the setting of GBM, both as a single agent and in combination with other adjuvant therapies.

Objective 2: In Chapter 4, to rationalise the mechanism(s) by which any observations occur that favour an anti-GBM effect by a REO NP therapy.

Objective 3: In Chapter 5, to determine the biocompatibility and biodistribution of a REO NP therapy in a biological system to determine whether or not GBM xenograft studies are warranted.

175

6.4 Key findings

The key findings of each experimental chapter are summarised as follows.

6.4.1 Identifying the therapeutic potential of REO NPs in GBM

In Chapter 3, four potential commercial REO NP candidates of CeO2, Gd2O3, La2O3, and Nd2O3 were characterised by electron microscopy. They were found to all exist reliably in the NP range, and possess positive surface charge. Additionally, cellular infiltration by the NP was also observed in a patient-derived cell line (PDCL) of GBM. Then, using a series of both immortalised cell lines (ICLs) and PDCLs of GBM, it became apparent that within 72 hours,

Gd2O3, La2O3 and Nd2O3 REO NP therapy could be cytotoxic to GBM cells. Additionally, these NPs also demonstrated ability to augment radiation therapy to which the colony-forming abilities of PDCLs were impaired. La2O3 was identified to be the most cytotoxic REO NP tested, and synergy with cytotoxic chemotherapy TMZ was then demonstrated in PDCLs.

Collectively, this Chapter was able to highlight REO NPs as a therapy that has both cytotoxic potential in GBM cells, as well as the ability to augment the adjuvant therapy modalities of radiation therapy and chemotherapy, justifying investigation of molecular mechanisms.

6.4.2 Mechanisms of cell death by lanthanum oxide NP therapy in GBM

In Chapter 4, the mechanisms of how the most cytotoxic REO NP La2O3 affected GBM cells were investigated. The mechanism of entry into the cell was validated to be clathrin- mediated endocytosis, with limited ionic dissociation likely once inside the cell. Generation of reactive oxygen species (ROS) was likely throughout the cell, causing or facilitating a number of possible cell death mechanisms, including both intrinsic and extrinsic apoptosis pathways, direct DNA damage, and autophagy. Augmentation of adjuvant radiation therapy and TMZ was shown to involve increased ROS and pro-apoptotic mitochondrial changes respectively.

176

The findings of this Chapter indicate the mechanisms by which La2O3 NP therapy causes anti-

GBM effects are complex and multifaceted. It presents a number of anti-cancer mechanisms specific for this REO NP that have not been previously reported.

6.4.3 Biocompatibility and biodistribution of lanthanum oxide NP therapy in a mouse

model.

In Chapter 5, the suitability of the La2O3 NP therapy in a mouse model was determined, to justify its study in future xenograft GBM models. Following intravenous injection, lanthanum was found in the brain of balb/c nude mice, indicating that the NP had overcome the BBB and reached the brain tissue. Furthermore, a regimen of 4 weekly doses of 5 mg/kg

NP was tolerable with no serious acute toxicities observed. Pharmacokinetic study showed that the NP therapy reached maximal brain concentrations within the first 24 hours of injection, and remained in the brain 60 days after final dose of the regimen. Histological sampling of the brain did not indicate that the long-term presence of the NP in the brain matter caused any serious toxicity. Although there was also NP distribution to the main clearance organs, the liver, kidneys and spleen, there were no overt signs of long-term toxicity observed throughout all experiments. Finally, it was shown that adjuvant therapy could be administered with La2O3 NP therapy without increasing the risk of long-term systemic toxicities or impairing NP biodistribution. Effectively, this Chapter justified La2O3 NP therapy to be safe and biocompatible in future in vivo xenograft GBM models, which will be required to evaluate its effect on overall survival.

177

6.5 Future directions

There are two key avenues of investigation to follow based on the findings of this thesis.

6.5.1 Xenograft survival study

Collectively, La2O3 NP therapy has been shown to exert favourable anti-GBM effects in vitro, and feasible biocompatibility and biodistribution in vivo. One future direction to pursue is to determine if this novel therapy can result in improved overall survival, and this can first be investigated by use of a GBM xenograft model in balb/c mice.351 At the time of writing, such a survival study is being organised with the same regimen as described in Section 5.2.4 in similar experimental groups. This will enable the detection not only of survival effect by NP therapy, but also whether or not the in vitro synergy with RT and CT can result in further benefit.

6.5.2 Nanoparticle enhancement

Another avenue that should be considered in the future is the possibility of enhancing or refining the NP therapy. One shortcoming of the therapy in its current form is that the NPs were not produced for purposes of therapy. They were obtained commercially, and as observed with scanning electron microscopy, vary greatly in size on a nanoscale. If NPs could be produced with greater uniformity and indeed at smaller sizes, than it is hypothesised that any potential therapeutic effects will be more uniform and amplified.352,353 Additionally, NPs can also act as coated vectors, with coatings such as TMZ354 and surfactant agents355 being reported in GBM to improve cytotoxicity and delivery respectively. These could potentially be applied to this NP therapy as well to further enhance its therapeutic effect.

178

6.6 Overall conclusions

In summary, I have evaluated and identified a REO NP with potential for therapy in

GBM. In vitro, La2O3 NP exerts cytotoxic effects as a single agent, and in combination with other adjuvant therapies, via multiple mechanisms causing cell death. Furthermore, I have been able to demonstrate that it is both biocompatible in a mouse model, and can reach the brain without causing any severe adverse effects in the long-term at a specific dosage. Thus, it is with excitement I await the results of a planned xenograft survival study in the future. Parallel to that, NP enhancement is another prospect that may contribute to the understanding of how prognosis of this dismal disease can be improved by this novel therapy.

179

REFERENCES

180

7 REFERENCE LIST

1 Ostrom, Q. T. et al. CBTRUS Statistical Report: Primary Brain and Other Central

Nervous System Tumors Diagnosed in the United States in 2011-2015. Neuro-oncology

20, iv1-iv86, doi:10.1093/neuonc/noy131 (2018).

2 Australian Institute of Health and Welfare. Australia’s health 2018. (AIHW, 2018).

3 Ostrom, Q. T. et al. CBTRUS Statistical Report: Primary Brain and Central Nervous

System Tumors Diagnosed in the United States in 2008-2012. Neuro-oncology 17

Suppl 4, iv1-iv62, doi:10.1093/neuonc/nov189 (2015).

4 Dobes, M. et al. Increasing incidence of glioblastoma multiforme and meningioma, and

decreasing incidence of Schwannoma (2000-2008): Findings of a multicenter

Australian study. Surgical neurology international 2, 176-176, doi:10.4103/2152-

7806.90696 (2011).

5 Thakkar, J. P. et al. Epidemiologic and molecular prognostic review of glioblastoma.

Cancer epidemiology, biomarkers & prevention : a publication of the American

Association for Cancer Research, cosponsored by the American Society of Preventive

Oncology 23, 1985-1996, doi:10.1158/1055-9965.Epi-14-0275 (2014).

6 Koshy, M. et al. Improved survival time trends for glioblastoma using the SEER 17

population-based registries. Journal of neuro-oncology 107, 207-212,

doi:10.1007/s11060-011-0738-7 (2012).

7 Kleihues, P. & Ohgaki, H. Primary and secondary glioblastomas: from concept to

clinical diagnosis. Neuro-oncology 1, 44-51, doi:10.1093/neuonc/1.1.44 (1999).

8 Ohgaki, H. et al. Genetic pathways to glioblastoma: a population-based study. Cancer

research 64, 6892-6899, doi:10.1158/0008-5472.Can-04-1337 (2004).

9 Wilson, T. A., Karajannis, M. A. & Harter, D. H. Glioblastoma multiforme: State of

the art and future therapeutics. Surg Neurol Int 5, 64, doi:10.4103/2152-7806.132138

181

(2014).

10 Chakrabarti, I., Cockburn, M., Cozen, W., Wang, Y. P. & Preston-Martin, S. A

population-based description of glioblastoma multiforme in Los Angeles County,

1974-1999. Cancer 104, 2798-2806, doi:10.1002/cncr.21539 (2005).

11 Bailey, P. & Cushing, H. A classification of the tumors of the glioma group on a

histogenetic basis with a correlated study of prognosis., (JB Lippincott, 1926).

12 Perry, A. & Wesseling, P. Histologic classification of gliomas. Handbook of clinical

neurology 134, 71-95, doi:10.1016/b978-0-12-802997-8.00005-0 (2016).

13 Louis, D. N. et al. The 2007 WHO classification of tumours of the central nervous

system. Acta neuropathologica 114, 97-109, doi:10.1007/s00401-007-0243-4 (2007).

14 Hardee, M. E. & Zagzag, D. Mechanisms of Glioma-Associated Neovascularization.

The American journal of pathology 181, 1126-1141,

doi:https://doi.org/10.1016/j.ajpath.2012.06.030 (2012).

15 Chen, L. et al. Classification of microvascular patterns via cluster analysis reveals their

prognostic significance in glioblastoma. Human pathology 46, 120-128,

doi:10.1016/j.humpath.2014.10.002 (2015).

16 Zhu, P., Du, X. L., Lu, G. & Zhu, J. J. Survival benefit of glioblastoma patients after

FDA approval of temozolomide concomitant with radiation and bevacizumab: A

population-based study. Oncotarget 8, 44015-44031, doi:10.18632/oncotarget.17054

(2017).

17 Cohen, M. H., Shen, Y. L., Keegan, P. & Pazdur, R. FDA drug approval summary:

bevacizumab (Avastin) as treatment of recurrent glioblastoma multiforme. The

oncologist 14, 1131-1138, doi:10.1634/theoncologist.2009-0121 (2009).

18 Louis, D. N. et al. The 2016 World Health Organization Classification of Tumors of

the Central Nervous System: a summary. Acta Neuropathol 131, 803-820,

doi:10.1007/s00401-016-1545-1 (2016).

182

19 SongTao, Q. et al. IDH mutations predict longer survival and response to temozolomide

in secondary glioblastoma. Cancer science 103, 269-273, doi:10.1111/j.1349-

7006.2011.02134.x (2012).

20 Hodges, T. R., Choi, B. D., Bigner, D. D., Yan, H. & Sampson, J. H. Isocitrate

dehydrogenase 1: what it means to the neurosurgeon: a review. Journal of neurosurgery

118, 1176-1180, doi:10.3171/2013.3.Jns122282 (2013).

21 Clark, K. H., Villano, J. L., Nikiforova, M. N., Hamilton, R. L. & Horbinski, C. 1p/19q

testing has no significance in the workup of glioblastomas. Neuropathology and

applied neurobiology 39, 706-717, doi:10.1111/nan.12031 (2013).

22 Brandes, A. A. et al. MGMT promoter methylation status can predict the incidence and

outcome of pseudoprogression after concomitant radiochemotherapy in newly

diagnosed glioblastoma patients. Journal of clinical oncology : official journal of the

American Society of Clinical Oncology 26, 2192-2197, doi:10.1200/jco.2007.14.8163

(2008).

23 Zhao, H., Wang, S., Song, C., Zha, Y. & Li, L. The prognostic value of MGMT

promoter status by pyrosequencing assay for glioblastoma patients' survival: a meta-

analysis. World journal of surgical oncology 14, 261, doi:10.1186/s12957-016-1012-4

(2016).

24 Hung, A. L., Garzon-Muvdi, T. & Lim, M. Biomarkers and Immunotherapeutic Targets

in Glioblastoma. World neurosurgery 102, 494-506, doi:10.1016/j.wneu.2017.03.011

(2017).

25 Chang, S. M. et al. Patterns of care for adults with newly diagnosed malignant glioma.

Jama 293, 557-564, doi:10.1001/jama.293.5.557 (2005).

26 Scott, J. N., Brasher, P. M., Sevick, R. J., Rewcastle, N. B. & Forsyth, P. A. How often

are nonenhancing supratentorial gliomas malignant? A population study. Neurology 59,

947-949 (2002).

183

27 Shukla, G. et al. Advanced magnetic resonance imaging in glioblastoma: a review.

Chinese clinical oncology 6, 40, doi:10.21037/cco.2017.06.28 (2017).

28 Stupp, R. et al. Radiotherapy plus concomitant and adjuvant temozolomide for

glioblastoma. The New England journal of medicine 352, 987-996,

doi:10.1056/NEJMoa043330 (2005).

29 Gan, H. K. et al. Management of glioblastoma in Victoria, Australia (2006-2008).

Journal of clinical neuroscience : official journal of the Neurosurgical Society of

Australasia 22, 1462-1466, doi:10.1016/j.jocn.2015.03.029 (2015).

30 Ostrom, Q. T. et al. CBTRUS Statistical Report: Primary brain and other central

nervous system tumors diagnosed in the United States in 2010-2014. Neuro-oncology

19, v1-v88, doi:10.1093/neuonc/nox158 (2017).

31 Mallick, S., Benson, R., Hakim, A. & Rath, G. K. Management of glioblastoma after

recurrence: A changing paradigm. Journal of the Egyptian National Cancer Institute

28, 199-210, doi:https://doi.org/10.1016/j.jnci.2016.07.001 (2016).

32 Choucair, A. K. et al. Development of multiple lesions during radiation therapy and

chemotherapy in patients with gliomas. Journal of neurosurgery 65, 654-658,

doi:10.3171/jns.1986.65.5.0654 (1986).

33 Gallego, O. Nonsurgical treatment of recurrent glioblastoma. Current oncology

(Toronto, Ont.) 22, e273-e281, doi:10.3747/co.22.2436 (2015).

34 Wong, E. T. et al. Outcomes and prognostic factors in recurrent glioma patients

enrolled onto phase II clinical trials. Journal of clinical oncology : official journal of

the American Society of Clinical Oncology 17, 2572-2578,

doi:10.1200/jco.1999.17.8.2572 (1999).

35 Chen, J. R., Yao, Y., Xu, H. Z. & Qin, Z. Y. Isocitrate Dehydrogenase (IDH)1/2

Mutations as Prognostic Markers in Patients With Glioblastomas. Medicine 95, e2583,

doi:10.1097/md.0000000000002583 (2016).

184

36 Binabaj, M. M. et al. The prognostic value of MGMT promoter methylation in

glioblastoma: A meta-analysis of clinical trials. Journal of cellular physiology 233,

378-386, doi:10.1002/jcp.25896 (2018).

37 Arita, H. et al. A combination of TERT promoter mutation and MGMT methylation

status predicts clinically relevant subgroups of newly diagnosed glioblastomas. Acta

neuropathologica communications 4, 79, doi:10.1186/s40478-016-0351-2 (2016).

38 Lu, V. M., Phan, K., Yin, J. X. M. & McDonald, K. L. Older studies can underestimate

prognosis of glioblastoma biomarker in meta-analyses: a meta-epidemiological study

for study-level effect in the current literature. Journal of neuro-oncology,

doi:10.1007/s11060-018-2897-2 (2018).

39 Lu, V. M., Jue, T. R., Phan, K. & McDonald, K. L. Quantifying the prognostic

significance in glioblastoma of seizure history at initial presentation: A systematic

review and meta-analysis. Clinical neurology and neurosurgery 164, 75-80,

doi:10.1016/j.clineuro.2017.11.015 (2018).

40 Lu, V. M., Goyal, A., Vaughan, L. S. & McDonald, K. L. The impact of hyperglycemia

on survival in glioblastoma: A systematic review and meta-analysis. Clinical neurology

and neurosurgery 170, 165-169, doi:10.1016/j.clineuro.2018.05.020 (2018).

41 Lacroix, M. et al. A multivariate analysis of 416 patients with glioblastoma multiforme:

prognosis, extent of resection, and survival. J Neurosurg 95, 190-198,

doi:10.3171/jns.2001.95.2.0190 (2001).

42 Sanai, N., Polley, M. Y., McDermott, M. W., Parsa, A. T. & Berger, M. S. An extent

of resection threshold for newly diagnosed glioblastomas. J Neurosurg 115, 3-8,

doi:10.3171/2011.2.jns10998

10.3171/2011.7.jns10238 (2011).

43 McGirt, M. J. et al. Independent association of extent of resection with survival in

185

patients with malignant brain astrocytoma. J Neurosurg 110, 156-162,

doi:10.3171/2008.4.17536 (2009).

44 Woernle, C. M. et al. Efficacy of Surgery and Further Treatment of Progressive

Glioblastoma. World neurosurgery 84, 301-307, doi:10.1016/j.wneu.2015.03.018

(2015).

45 Brandes, A. A., Bartolotti, M. & Franceschi, E. Second surgery for recurrent

glioblastoma: advantages and pitfalls. Expert Rev Anticancer Ther 13, 583-587,

doi:https://dx.doi.org/10.1586/era.13.32 (2013).

46 Chaichana, K. L. et al. Multiple resections for patients with glioblastoma: prolonging

survival. J Neurosurg 118, 812-820, doi:https://dx.doi.org/10.3171/2012.9.JNS1277

(2013).

47 Lu, V. M., Jue, T. R., McDonald, K. L. & Rovin, R. A. The Survival Effect of Repeat

Surgery at Glioblastoma Recurrence and its Trend: A Systematic Review and Meta-

Analysis. World neurosurgery 115, 453-459.e453, doi:10.1016/j.wneu.2018.04.016

(2018).

48 Chang, C. H. et al. Comparison of postoperative radiotherapy and combined

postoperative radiotherapy and chemotherapy in the multidisciplinary management of

malignant gliomas. A joint Radiation Therapy Oncology Group and Eastern

Cooperative Oncology Group study. Cancer 52, 997-1007 (1983).

49 Corso, C. D., Bindra, R. S. & Mehta, M. P. The role of radiation in treating glioblastoma:

here to stay. Journal of neuro-oncology 134, 479-485, doi:10.1007/s11060-016-2348-

x (2017).

50 Wallner, K. E., Galicich, J. H., Krol, G., Arbit, E. & Malkin, M. G. Patterns of failure

following treatment for glioblastoma multiforme and anaplastic astrocytoma.

International journal of radiation oncology, biology, physics 16, 1405-1409 (1989).

51 Howard, S. P., Krauze, A., Chan, M. D., Tsien, C. & Tome, W. A. The evolving role

186

for re-irradiation in the management of recurrent grade 4 glioma. Journal of neuro-

oncology 134, 523-530, doi:10.1007/s11060-017-2392-1 (2017).

52 Spitz, D. R. & Hauer-Jensen, M. Ionizing radiation-induced responses: where free

radical chemistry meets redox biology and medicine. Antioxidants & redox signaling

20, 1407-1409, doi:10.1089/ars.2013.5769 (2014).

53 Hieda, K., Suzuki, K., Hirono, T., Suzuki, M. & Furusawa, Y. Single- and double-

strand breaks in pBR322 DNA by vacuum-UV from 8.3 to 20.7 eV. Journal of

radiation research 35, 104-111 (1994).

54 Misawa, M. & Takahashi, J. Generation of reactive oxygen species induced by gold

nanoparticles under x-ray and UV Irradiations. Nanomedicine : nanotechnology,

biology, and medicine 7, 604-614, doi:10.1016/j.nano.2011.01.014 (2011).

55 Olson, J. J., Nayak, L., Ormond, D. R., Wen, P. Y. & Kalkanis, S. N. The role of

cytotoxic chemotherapy in the management of progressive glioblastoma : a systematic

review and evidence-based clinical practice guideline. Journal of neuro-oncology 118,

501-555, doi:10.1007/s11060-013-1338-5 (2014).

56 Newlands, E. S., Stevens, M. F. G., Wedge, S. R., Wheelhouse, R. T. & Brock, C.

Temozolomide: a review of its discovery, chemical properties, pre-clinical

development and clinical trials. Cancer Treatment Reviews 23, 35-61,

doi:https://doi.org/10.1016/S0305-7372(97)90019-0 (1997).

57 Yung, W. K. A. et al. A phase II study of temozolomide vs. procarbazine in patients

with glioblastoma multiforme at first relapse. British Journal Of Cancer 83, 588,

doi:10.1054/bjoc.2000.1316 (2000).

58 Wang, Y. et al. Comparison of the clinical efficacy of temozolomide (TMZ) versus

nimustine (ACNU)-based chemotherapy in newly diagnosed glioblastoma.

Neurosurgical review 37, 73-78, doi:10.1007/s10143-013-0490-x (2014).

59 Messaoudi, K., Clavreul, A. & Lagarce, F. Toward an effective strategy in glioblastoma

187

treatment. Part I: resistance mechanisms and strategies to overcome resistance of

glioblastoma to temozolomide. Drug discovery today 20, 899-905,

doi:10.1016/j.drudis.2015.02.011 (2015).

60 Clarke, J. et al. Epigenetic pathways and glioblastoma treatment. Epigenetics 8, 785-

795, doi:10.4161/epi.25440 (2013).

61 Crespo, I. et al. Molecular and Genomic Alterations in Glioblastoma Multiforme. The

American journal of pathology 185, 1820-1833, doi:10.1016/j.ajpath.2015.02.023

(2015).

62 Tivnan, A., Heilinger, T., Lavelle, E. C. & Prehn, J. H. M. Advances in immunotherapy

for the treatment of glioblastoma. Journal of neuro-oncology 131, 1-9,

doi:10.1007/s11060-016-2299-2 (2017).

63 Dyck, L. & Mills, K. H. G. Immune checkpoints and their inhibition in cancer and

infectious diseases. European journal of immunology 47, 765-779,

doi:10.1002/eji.201646875 (2017).

64 Yang, J. C. & Rosenberg, S. A. Adoptive T-Cell Therapy for Cancer. Advances in

immunology 130, 279-294, doi:10.1016/bs.ai.2015.12.006 (2016).

65 Kurz, S. C. et al. PD-1 inhibition has only limited clinical benefit in patients with

recurrent high-grade glioma. Neurology 91, e1355-e1359,

doi:10.1212/wnl.0000000000006283 (2018).

66 Omuro, A. et al. Nivolumab with or without ipilimumab in patients with recurrent

glioblastoma: results from exploratory phase I cohorts of CheckMate 143. Neuro-

oncology 20, 674-686, doi:10.1093/neuonc/nox208 (2018).

67 Jackson, C. et al. Challenges in immunotherapy presented by the glioblastoma

multiforme microenvironment. 2011 (2011).

68 Lampson, L. A. & Hickey, W. F. Monoclonal antibody analysis of MHC expression in

human brain biopsies: tissue ranging from "histologically normal" to that showing

188

different levels of glial tumor involvement. Journal of immunology (Baltimore, Md. :

1950) 136, 4054-4062 (1986).

69 Zhou, Q., Wang, Y. & Ma, W. The progress of immunotherapy for glioblastoma.

Human vaccines & immunotherapeutics 11, 2654-2658,

doi:10.1080/21645515.2015.1081727 (2015).

70 Whitley, R. J. & Markert, J. M. Viral therapy of glioblastoma multiforme. Proceedings

of the National Academy of Sciences of the United States of America 110, 11672-11673,

doi:10.1073/pnas.1310253110 (2013).

71 Todo, T. et al. Systemic antitumor immunity in experimental brain tumor therapy using

a multimutated, replication-competent herpes simplex virus. Human gene therapy 10,

2741-2755, doi:10.1089/10430349950016483 (1999).

72 Fukuhara, H., Ino, Y. & Todo, T. Oncolytic virus therapy: A new era of cancer

treatment at dawn. Cancer science 107, 1373-1379, doi:10.1111/cas.13027 (2016).

73 Garber, K. China approves world's first oncolytic virus therapy for cancer treatment.

Journal of the National Cancer Institute 98, 298-300, doi:10.1093/jnci/djj111 (2006).

74 Andtbacka, R. H. et al. Talimogene Laherparepvec Improves Durable Response Rate

in Patients With Advanced Melanoma. Journal of clinical oncology : official journal

of the American Society of Clinical Oncology 33, 2780-2788,

doi:10.1200/jco.2014.58.3377 (2015).

75 Batich, K. A. et al. Long-term Survival in Glioblastoma with Cytomegalovirus pp65-

Targeted Vaccination. Clinical cancer research : an official journal of the American

Association for Cancer Research 23, 1898-1909, doi:10.1158/1078-0432.Ccr-16-2057

(2017).

76 Cloughesy, T. F. et al. Durable complete responses in some recurrent high-grade glioma

patients treated with Toca 511 + Toca FC. Neuro-oncology 20, 1383-1392,

doi:10.1093/neuonc/noy075 (2018).

189

77 Desjardins, A. et al. Recurrent Glioblastoma Treated with Recombinant Poliovirus. The

New England journal of medicine 379, 150-161, doi:10.1056/NEJMoa1716435 (2018).

78 Ashfaq, U. A., Riaz, M., Yasmeen, E. & Yousaf, M. Z. Recent Advances in

Nanoparticle-Based Targeted Drug-Delivery Systems Against Cancer and Role of

Tumor Microenvironment. Critical reviews in therapeutic drug carrier systems 34,

317-353, doi:10.1615/CritRevTherDrugCarrierSyst.2017017845 (2017).

79 Shilo, M. et al. The effect of nanoparticle size on the probability to cross the blood-

brain barrier: an in-vitro endothelial cell model. 13, 19 (2015).

80 van Rooy, I. et al. In vivo methods to study uptake of nanoparticles into the brain.

Pharmaceutical research 28, 456-471, doi:10.1007/s11095-010-0291-7 (2011).

81 Denora, N., Trapani, A., Laquintana, V., Lopedota, A. & Trapani, G. Recent advances

in medicinal chemistry and pharmaceutical technology--strategies for drug delivery to

the brain. Current topics in medicinal chemistry 9, 182-196 (2009).

82 Rao, K. S., Reddy, M. K., Horning, J. L. & Labhasetwar, V. TAT-conjugated

nanoparticles for the CNS delivery of anti-HIV drugs. Biomaterials 29, 4429-4438,

doi:10.1016/j.biomaterials.2008.08.004 (2008).

83 Shilo, M., Motiei, M., Hana, P. & Popovtzer, R. Transport of nanoparticles through the

blood-brain barrier for imaging and therapeutic applications. Nanoscale 6, 2146-2152,

doi:10.1039/C3NR04878K (2014).

84 Le Duc, G. et al. Toward an image-guided microbeam radiation therapy using

gadolinium-based nanoparticles. ACS nano 5, 9566-9574, doi:10.1021/nn202797h

(2011).

85 Ducray, A. D. et al. Uptake of silica nanoparticles in the brain and effects on neuronal

differentiation using different in vitro models. Nanomedicine : nanotechnology, biology,

and medicine 13, 1195-1204, doi:10.1016/j.nano.2016.11.001 (2017).

86 Xu, A. et al. A physical model for the size-dependent cellular uptake of nanoparticles

190

modified with cationic surfactants. International journal of nanomedicine 7, 3547-3554,

doi:10.2147/ijn.s32188 (2012).

87 Perrault, S. D., Walkey, C., Jennings, T., Fischer, H. C. & Chan, W. C. Mediating tumor

targeting efficiency of nanoparticles through design. Nano Lett 9, 1909-1915,

doi:10.1021/nl900031y (2009).

88 Davis, M. E., Chen, Z. & Shin, D. M. Nanoparticle therapeutics: an emerging treatment

modality for cancer. 7, 771-782 (2008).

89 Longmire, M., Choyke, P. L. & Kobayashi, H. Clearance properties of nano-sized

particles and molecules as imaging agents: considerations and caveats. Nanomedicine

(London, England) 3, 703-717, doi:10.2217/17435889.3.5.703 (2008).

90 Chithrani, B. D. & Chan, W. C. Elucidating the mechanism of cellular uptake and

removal of protein-coated gold nanoparticles of different sizes and shapes. Nano letters

7, 1542-1550, doi:10.1021/nl070363y (2007).

91 Champion, J. A. & Mitragotri, S. Role of target geometry in phagocytosis. Proceedings

of the National Academy of Sciences of the United States of America 103, 4930-4934

(2006).

92 Qiu, Y. et al. Surface chemistry and aspect ratio mediated cellular uptake of Au

nanorods. Biomaterials 31, 7606-7619, doi:10.1016/j.biomaterials.2010.06.051 (2010).

93 Hutter, E. et al. Microglial response to gold nanoparticles. ACS nano 4, 2595-2606,

doi:10.1021/nn901869f (2010).

94 Champion, J. A., Katare, Y. K. & Mitragotri, S. Particle shape: a new design parameter

for micro- and nanoscale drug delivery carriers. Journal of controlled release : official

journal of the Controlled Release Society 121, 3-9, doi:10.1016/j.jconrel.2007.03.022

(2007).

95 Chithrani, B. D., Ghazani, A. A. & Chan, W. C. Determining the size and shape

dependence of gold nanoparticle uptake into mammalian cells. Nano Lett 6, 662-668,

191

doi:10.1021/nl052396o (2006).

96 Gratton, S. E. et al. The effect of particle design on cellular internalization pathways.

Proceedings of the National Academy of Sciences of the United States of America 105,

11613-11618, doi:10.1073/pnas.0801763105 (2008).

97 Slowing, I., Trewyn, B. G. & Lin, V. S. Effect of surface functionalization of MCM-

41-type mesoporous silica nanoparticles on the endocytosis by human cancer cells.

Journal of the American Chemical Society 128, 14792-14793, doi:10.1021/ja0645943

(2006).

98 Arvizo, R. R. et al. Effect of nanoparticle surface charge at the plasma membrane and

beyond. Nano Lett 10, 2543-2548, doi:10.1021/nl101140t (2010).

99 Abbott, N. J., Ronnback, L. & Hansson, E. Astrocyte-endothelial interactions at the

blood-brain barrier. Nature reviews. Neuroscience 7, 41-53, doi:10.1038/nrn1824

(2006).

100 Pardridge, W. M. The Blood-Brain Barrier: Bottleneck in Brain Drug Development.

NeuroRx 2, 3-14 (2005).

101 Hwang, S. R. & Kim, K. Nano-enabled delivery systems across the blood-brain barrier.

Archives of pharmacal research 37, 24-30, doi:10.1007/s12272-013-0272-6 (2014).

102 Hervé, F., Ghinea, N. & Scherrmann, J. M. CNS Delivery Via Adsorptive Transcytosis.

The AAPS Journal 10, 455-472, doi:10.1208/s12248-008-9055-2 (2008).

103 Ulbrich, K., Knobloch, T. & Kreuter, J. Targeting the insulin receptor: nanoparticles

for drug delivery across the blood-brain barrier (BBB). Journal of drug targeting 19,

125-132, doi:10.3109/10611861003734001 (2011).

104 Ulbrich, K., Hekmatara, T., Herbert, E. & Kreuter, J. Transferrin- and transferrin-

receptor-antibody-modified nanoparticles enable drug delivery across the blood-brain

barrier (BBB). European journal of pharmaceutics and biopharmaceutics : official

journal of Arbeitsgemeinschaft fur Pharmazeutische Verfahrenstechnik e.V 71, 251-

192

256, doi:10.1016/j.ejpb.2008.08.021 (2009).

105 Michaelis, K. et al. Covalent linkage of apolipoprotein e to albumin nanoparticles

strongly enhances drug transport into the brain. The Journal of pharmacology and

experimental therapeutics 317, 1246-1253, doi:10.1124/jpet.105.097139 (2006).

106 Lu, V. M., McDonald, K. L. & Townley, H. E. Realizing the therapeutic potential of

rare earth elements in designing nanoparticles to target and treat glioblastoma.

Nanomedicine (London, England) 12, 2389-2401, doi:10.2217/nnm-2017-0193 (2017).

107 Vorbrodt, A. W. Ultracytochemical characterization of anionic sites in the wall of brain

capillaries. Journal of neurocytology 18, 359-368 (1989).

108 Schnitzer, J. E. Caveolae: from basic trafficking mechanisms to targeting transcytosis

for tissue-specific drug and gene delivery in vivo. Adv Drug Deliv Rev 49, 265-280

(2001).

109 Vu Hai, M. T., Lescop, P., Loosfelt, H. & Ghinea, N. Receptor-mediated transcytosis

of follicle-stimulating hormone through the rat testicular microvasculature. Biology of

the cell 96, 133-144, doi:10.1016/j.biolcel.2003.11.008 (2004).

110 Sarin, H. Physiologic upper limits of pore size of different blood capillary types and

another perspective on the dual pore theory of microvascular permeability. Journal of

Angiogenesis Research 2, 14, doi:10.1186/2040-2384-2-14 (2010).

111 Matsumura, Y. & Maeda, H. A new concept for macromolecular therapeutics in cancer

chemotherapy: mechanism of tumoritropic accumulation of proteins and the antitumor

agent smancs. Cancer research 46, 6387-6392 (1986).

112 Maeda, H., Wu, J., Sawa, T., Matsumura, Y. & Hori, K. Tumor vascular permeability

and the EPR effect in macromolecular therapeutics: a review. Journal of controlled

release : official journal of the Controlled Release Society 65, 271-284 (2000).

113 Davies, D. C. Blood-brain barrier breakdown in septic encephalopathy and brain

tumours. Journal of anatomy 200, 639-646 (2002).

193

114 Liebner, S. et al. Claudin-1 and claudin-5 expression and tight junction morphology are

altered in blood vessels of human glioblastoma multiforme. Acta neuropathologica 100,

323-331 (2000).

115 Warth, A., Kroger, S. & Wolburg, H. Redistribution of aquaporin-4 in human

glioblastoma correlates with loss of agrin immunoreactivity from brain capillary basal

laminae. Acta neuropathologica 107, 311-318, doi:10.1007/s00401-003-0812-0 (2004).

116 Papadopoulos, M. C., Saadoun, S., Davies, D. C. & Bell, B. A. Emerging molecular

mechanisms of brain tumour oedema. British journal of neurosurgery 15, 101-108

(2001).

117 Liu, Y. & Lu, W. Recent advances in brain tumor-targeted nano-drug delivery systems.

Expert opinion on drug delivery 9, 671-686, doi:10.1517/17425247.2012.682726

(2012).

118 Hobbs, S. K. et al. Regulation of transport pathways in tumor vessels: role of tumor

type and microenvironment. Proceedings of the National Academy of Sciences of the

United States of America 95, 4607-4612 (1998).

119 Ambruosi, A. et al. Biodistribution of polysorbate 80-coated doxorubicin-loaded

[14C]-poly(butyl cyanoacrylate) nanoparticles after intravenous administration to

glioblastoma-bearing rats. Journal of drug targeting 14, 97-105,

doi:10.1080/10611860600636135 (2006).

120 Wilhelm, S. et al. Analysis of nanoparticle delivery to tumours. Nature Reviews

Materials 1, 16014, doi:10.1038/natrevmats.2016.14 http://www.nature.com/articles/natrevmats201614#supplementary-information (2016).

121 Mehta, A. I., Linninger, A., Lesniak, M. S. & Engelhard, H. H. Current status of

intratumoral therapy for glioblastoma. 125, 1-7 (2015).

122 Cohen, Z. R. et al. Localized RNAi Therapeutics of Chemoresistant Grade IV Glioma

194

Using Hyaluronan-Grafted Lipid-Based Nanoparticles. ACS nano 9, 1581-1591 (2015).

123 Iliff, J. J. et al. A Paravascular Pathway Facilitates CSF Flow Through the Brain

Parenchyma and the Clearance of Interstitial Solutes, Including Amyloid β. Science

translational medicine 4, 147ra111, doi:10.1126/scitranslmed.3003748 (2012).

124 Charles, N. A., Holland, E. C., Gilbertson, R., Glass, R. & Kettenmann, H. The brain

tumor microenvironment. Glia 59, 1169-1180, doi:10.1002/glia.21136 (2011).

125 Chowdhary, S. A., Ryken, T. & Newton, H. B. Survival outcomes and safety of

carmustine wafers in the treatment of high-grade gliomas: a meta-analysis. Journal of

Neuro-Oncology 122, 367-382, doi:10.1007/s11060-015-1724-2 (2015).

126 Fleming, A. B. & Saltzman, W. M. Pharmacokinetics of the carmustine implant.

Clinical pharmacokinetics 41, 403-419, doi:10.2165/00003088-200241060-00002

(2002).

127 Petri, B. et al. Chemotherapy of brain tumour using doxorubicin bound to surfactant-

coated poly(butyl cyanoacrylate) nanoparticles: revisiting the role of surfactants.

Journal of controlled release : official journal of the Controlled Release Society 117,

51-58, doi:10.1016/j.jconrel.2006.10.015 (2007).

128 Wiley, D. T., Webster, P., Gale, A. & Davis, M. E. Transcytosis and brain uptake of

transferrin-containing nanoparticles by tuning avidity to transferrin receptor.

Proceedings of the National Academy of Sciences of the United States of America 110,

8662-8667, doi:10.1073/pnas.1307152110 (2013).

129 Guerrero, S. et al. Improving the brain delivery of gold nanoparticles by conjugation

with an amphipathic peptide. Nanomedicine (London, England) 5, 897-913,

doi:10.2217/nnm.10.74 (2010).

130 Choi, C. H., Alabi, C. A., Webster, P. & Davis, M. E. Mechanism of active targeting in

solid tumors with transferrin-containing gold nanoparticles. Proceedings of the

National Academy of Sciences of the United States of America 107, 1235-1240,

195

doi:10.1073/pnas.0914140107 (2010).

131 Fang, C. et al. Temozolomide Nanoparticles for Targeted Glioblastoma Therapy. ACS

applied materials & interfaces 7, 6674-6682, doi:10.1021/am5092165 (2015).

132 Mortensen, J. H. et al. Targeted antiepidermal growth factor receptor (cetuximab)

immunoliposomes enhance cellular uptake in vitro and exhibit increased accumulation

in an intracranial model of glioblastoma multiforme. Journal of drug delivery 2013,

209205, doi:10.1155/2013/209205 (2013).

133 Patil, R. et al. Temozolomide delivery to tumor cells by a multifunctional nano vehicle

based on poly(beta-L-malic acid). Pharmaceutical research 27, 2317-2329,

doi:10.1007/s11095-010-0091-0 (2010).

134 Gao, H. et al. RGD and interleukin-13 peptide functionalized nanoparticles for

enhanced glioblastoma cells and neovasculature dual targeting delivery and elevated

tumor penetration. Molecular pharmaceutics 11, 1042-1052, doi:10.1021/mp400751g

(2014).

135 Kang, T. et al. Enhancing Glioblastoma-Specific Penetration by Functionalization of

Nanoparticles with an Iron-Mimic Peptide Targeting Transferrin/Transferrin Receptor

Complex. Molecular pharmaceutics 12, 2947-2961,

doi:10.1021/acs.molpharmaceut.5b00222 (2015).

136 Lepinoux-Chambaud, C. & Eyer, J. The NFL-TBS.40-63 anti-glioblastoma peptide

enters selectively in glioma cells by endocytosis. International journal of

pharmaceutics 454, 738-747, doi:10.1016/j.ijpharm.2013.04.004 (2013).

137 Song, E. et al. Surface chemistry governs cellular tropism of nanoparticles in the brain.

Nature Communications 8, 15322, doi:10.1038/ncomms15322 (2017).

138 Inagaki, K. & Haraguchi, H. Determination of rare earth elements in human blood

serum by inductively coupled plasma mass spectrometry after chelating resin

preconcentration. Analyst 125, 191-196 (2000).

196

139 Palasz, A. & Czekaj, P. Toxicological and cytophysiological aspects of lanthanides

action. Acta biochimica Polonica 47, 1107-1114 (2000).

140 Neary, J. T., van Breemen, C., Forster, E., Norenberg, L. O. & Norenberg, M. D. ATP

stimulates calcium influx in primary astrocyte cultures. Biochem Biophys Res Commun

157, 1410-1416 (1988).

141 Neary, J. T. et al. ATP-evoked calcium signal stimulates protein

phosphorylation/dephosphorylation in astrocytes. Brain research 566, 89-94,

doi:https://doi.org/10.1016/0006-8993(91)91684-S (1991).

142 Nestler, E. & Greengard, P. Protein Phosphorylation in the Nervous System. (Wiley,

1984).

143 Shannon, R. Revised effective ionic radii and systematic studies of interatomic

distances in halides and chalcogenides. Acta Cryst. A32, 751-767 (1976).

144 Zhang, J.-m. et al. ATP Released by Astrocytes Mediates Glutamatergic Activity-

Dependent Heterosynaptic Suppression. Neuron 40, 971-982,

doi:http://doi.org/10.1016/S0896-6273(03)00717-7 (2003).

145 Zimmermann, H. & Braun, N. Extracellular metabolism of nucleotides in the nervous

system. Journal of Autonomic Pharmacology 16, 397-400, doi:10.1111/j.1474-

8673.1996.tb00062.x (1996).

146 Tjalkens, R. B., Zoran, M. J., Mohl, B. & Barhoumi, R. Manganese suppresses ATP-

dependent intercellular calcium waves in astrocyte networks through alteration of

mitochondrial and endoplasmic reticulum calcium dynamics. Brain Research 1113,

210-219, doi:http://doi.org/10.1016/j.brainres.2006.07.053 (2006).

147 Delaney, G., Jacob, S., Featherstone, C. & Barton, M. The role of radiotherapy in cancer

treatment: estimating optimal utilization from a review of evidence-based clinical

guidelines. Cancer 104, 1129-1137, doi:10.1002/cncr.21324 (2005).

148 Kobayashi, K., Usami, N., Porcel, E., Lacombe, S. & Le Sech, C. Enhancement of

197

radiation effect by heavy elements. Mutation research 704, 123-131,

doi:10.1016/j.mrrev.2010.01.002 (2010).

149 Mowat, P. et al. In vitro radiosensitizing effects of ultrasmall gadolinium based

particles on tumour cells. Journal of nanoscience and nanotechnology 11, 7833-7839

(2011).

150 Gai, S. et al. Synthesis of Magnetic, Up-Conversion Luminescent, and Mesoporous

Core-Shell-Structured Nanocomposites as Drug Carriers. Advanced Functional

Materials 20, 1166-1172, doi:10.1002/adfm.200902274 (2010).

151 Auzel, F. Upconversion and anti-Stokes processes with f and d ions in solids. Chemical

reviews 104, 139-173, doi:10.1021/cr020357g (2004).

152 Ang, L. Y., Lim, M. E., Ong, L. C. & Zhang, Y. Applications of upconversion

nanoparticles in imaging, detection and therapy. Nanomedicine (London, England) 6,

1273-1288, doi:10.2217/nnm.11.108 (2011).

153 Park, Y. I. et al. Theranostic probe based on lanthanide-doped nanoparticles for

simultaneous in vivo dual-modal imaging and photodynamic therapy. Advanced

materials (Deerfield Beach, Fla.) 24, 5755-5761, doi:10.1002/adma.201202433 (2012).

154 Park, S. Y. et al. A smart polysaccharide/drug conjugate for photodynamic therapy.

Angewandte Chemie (International ed. in English) 50, 1644-1647,

doi:10.1002/anie.201006038 (2011).

155 Zhuang, G. et al. Concentration of rare earth elements, As, and Th in human brain and

brain tumors, determined by neutron activation analysis. 53, 45-49 (1996).

156 Lindahl, L. S. et al. Differential ability of astroglia and neuronal cells to accumulate

lead: dependence on cell type and on degree of differentiation. Toxicological sciences :

an official journal of the Society of Toxicology 50, 236-243 (1999).

157 Rousseau, A. et al. Expression of oligodendroglial and astrocytic lineage markers in

diffuse gliomas: use of YKL-40, ApoE, ASCL1, and NKX2-2. Journal of

198

neuropathology and experimental neurology 65, 1149-1156,

doi:10.1097/01.jnen.0000248543.90304.2b (2006).

158 Xia, Q., Feng, X., Yuan, L. & Yang, X. Brain-derived neurotrophic factor protects

neurons from GdCl3-induced impairment in neuron-astrocyte co-cultures. SCIENCE

CHINA Chemistry 53, 2193-2199 (2010).

159 Kirschner, P. B. et al. NGF, BDNF and NT-5, but not NT-3 protect against MPP+

toxicity and oxidative stress in neonatal animals. Brain research 713, 178-185 (1996).

160 Jackson, G. R., Apffel, L., Werrbach-Perez, K. & Perez-Polo, J. R. Role of nerve

growth factor in oxidant-antioxidant balance and neuronal injury. I. Stimulation of

hydrogen peroxide resistance. Journal of neuroscience research 25, 360-368,

doi:10.1002/jnr.490250313 (1990).

161 Zhao, H. et al. Oxidative injury in the brain of mice caused by lanthanid. Biol Trace

Elem Res 142, 174-189, doi:10.1007/s12011-010-8759-1 (2011).

162 Deen, W. M., Lazzara, M. J. & Myers, B. D. Structural determinants of glomerular

permeability. American journal of physiology. Renal physiology 281, F579-596 (2001).

163 Ohlson, M., Sorensson, J. & Haraldsson, B. A gel-membrane model of glomerular

charge and size selectivity in series. American journal of physiology. Renal physiology

280, F396-405 (2001).

164 Choi, H. S. et al. Renal clearance of quantum dots. Nature biotechnology 25, 1165-

1170, doi:10.1038/nbt1340 (2007).

165 Zhao, H. et al. Oxidative stress in the kidney injury of mice following exposure to

lanthanides trichloride. Chemosphere 93, 875-884,

doi:10.1016/j.chemosphere.2013.05.034 (2013).

166 De Jong, W. H. et al. Particle size-dependent organ distribution of gold nanoparticles

after intravenous administration. Biomaterials 29, 1912-1919,

doi:10.1016/j.biomaterials.2007.12.037 (2008).

199

167 Nakamura, Y., Tsumura, Y., Tonogai, Y., Shibata, T. & Ito, Y. Differences in behavior

among the chlorides of seven rare earth elements administered intravenously to rats.

Fundamental and applied toxicology : official journal of the Society of Toxicology 37,

106-116 (1997).

168 Spencer, A. J. et al. Gadolinium chloride toxicity in the rat. Toxicologic pathology 25,

245-255, doi:10.1177/019262339702500301 (1997).

169 Rai, R. M. et al. Kupffer cell depletion by gadolinium chloride enhances liver

regeneration after partial hepatectomy in rats. Am J Physiol 270, G909-918 (1996).

170 Michael, S. L., Pumford, N. R., Mayeux, P. R., Niesman, M. R. & Hinson, J. A.

Pretreatment of mice with macrophage inactivators decreases acetaminophen

hepatotoxicity and the formation of reactive oxygen and nitrogen species. Hepatology

30, 186-195, doi:10.1002/hep.510300104 (1999).

171 Adachi Gy, G. & Imanaka, N. The Binary Rare Earth Oxides. Chemical reviews 98,

1479-1514 (1998).

172 Chourmouzi, D., Papadopoulou, E., Marias, K. & Drevelegas, A. Imaging of brain

tumors. Surgical oncology clinics of North America 23, 629-684,

doi:10.1016/j.soc.2014.07.004 (2014).

173 Townley, H. E. Applications of the Rare Earth Elements in Cancer Imaging and

Therapy. Current Nanoscience 9, 686-691 (2013).

174 Teo, R. D., Termini, J. & Gray, H. B. Lanthanides: Applications in Cancer Diagnosis

and Therapy. Journal of medicinal chemistry 59, 6012-6024,

doi:10.1021/acs.jmedchem.5b01975 (2016).

175 Dong, H. et al. Lanthanide Nanoparticles: From Design toward Bioimaging and

Therapy. Chemical reviews 115, 10725-10815, doi:10.1021/acs.chemrev.5b00091

(2015).

176 Comby, S. et al. Lanthanide-functionalized nanoparticles as MRI and luminescent

200

probes for sensing and/or imaging applications. Inorganic chemistry 53, 1867-1879,

doi:10.1021/ic4023568 (2014).

177 Tresman, R., Wright, J., Dubois, C., Lewis, J. & Surash, S. RT-33: OUTCOME OF

GLIOBLASTOMA IN THE POST-STUPP ERA: A SINGLE CENTRE

RETROSPECTIVE AUDIT. Neuro-oncology 16, v194-v194,

doi:10.1093/neuonc/nou270.29 (2014).

178 Teo, M. et al. A survival analysis of GBM patients in the West of Scotland pre- and

post-introduction of the Stupp regime. British journal of neurosurgery 28, 351-355,

doi:10.3109/02688697.2013.847170 (2014).

179 Ortega, A. et al. Multiple resections and survival of recurrent glioblastoma patients in

the temozolomide era. Journal of clinical neuroscience : official journal of the

Neurosurgical Society of Australasia 24, 105-111, doi:10.1016/j.jocn.2015.05.047

(2016).

180 Plummer, E. M. & Manchester, M. Endocytic uptake pathways utilized by CPMV

nanoparticles. Molecular pharmaceutics 10, 26-32, doi:10.1021/mp300238w (2013).

181 Tang, H. et al. Effect of inhibitors of endocytosis and NF-kB signal pathway on folate-

conjugated nanoparticle endocytosis by rat Kupffer cells. International journal of

nanomedicine 12, 6937-6947, doi:10.2147/ijn.S141407 (2017).

182 Huerta-Garcia, E. et al. Internalization of titanium dioxide nanoparticles by glial cells

is given at short times and is mainly mediated by actin reorganization-dependent

endocytosis. Neurotoxicology 51, 27-37, doi:10.1016/j.neuro.2015.08.013 (2015).

183 Saha, K. et al. Surface functionality of nanoparticles determines cellular uptake

mechanisms in mammalian cells. Small (Weinheim an der Bergstrasse, Germany) 9,

300-305, doi:10.1002/smll.201201129 (2013).

184 Ibuki, Y. & Toyooka, T. Nanoparticle uptake measured by flow cytometry. Methods in

molecular biology (Clifton, N.J.) 926, 157-166, doi:10.1007/978-1-62703-002-1_11

201

(2012).

185 Das, S. et al. Tissue deposition and toxicological effects of commercially significant

rare earth oxide nanomaterials: Material and physical properties. Environmental

toxicology 32, 904-917, doi:doi:10.1002/tox.22290 (2017).

186 Whittaker, S. et al. Combination of palbociclib and radiotherapy for glioblastoma. Cell

Death Discovery 3, 17033, doi:10.1038/cddiscovery.2017.33 (2017).

187 Durgo, K., Halec, I., Sola, I. & Franekic, J. Cytotoxic and genotoxic effects of the

quercetin/lanthanum complex on human cervical carcinoma cells in vitro. Arhiv za

higijenu rada i toksikologiju 62, 221-227, doi:10.2478/10004-1254-62-2011-2122

(2011).

188 Chen, Y., Yang, L., Feng, C. & Wen, L. P. Nano neodymium oxide induces massive

vacuolization and autophagic cell death in non-small cell lung cancer NCI-H460 cells.

Biochemical and biophysical research communications 337, 52-60,

doi:10.1016/j.bbrc.2005.09.018 (2005).

189 Hussein, B. H., Azab, H. A., el-Azab, M. F. & el-Falouji, A. I. A novel anti-tumor agent,

Ln(III) 2-thioacetate benzothiazole induces anti-angiogenic effect and cell death in

cancer cell lines. European journal of medicinal chemistry 51, 99-109,

doi:10.1016/j.ejmech.2012.02.025 (2012).

190 Momekov, G. et al. Evaluation of the cytotoxic and pro-apoptotic activities of Eu(III)

complexes with appended DNA intercalators in a panel of human malignant cell lines.

Medicinal chemistry (Shariqah (United Arab Emirates)) 2, 439-445 (2006).

191 Sminia, P. & Mayer, R. External beam radiotherapy of recurrent glioma: radiation

tolerance of the human brain. Cancers 4, 379-399, doi:10.3390/cancers4020379 (2012).

192 Siu, A. et al. Radiation necrosis following treatment of high grade glioma--a review of

the literature and current understanding. Acta neurochirurgica 154, 191-201;

discussion 201, doi:10.1007/s00701-011-1228-6 (2012).

202

193 van Linde, M. E. et al. Bevacizumab in combination with radiotherapy and

temozolomide for patients with newly diagnosed glioblastoma multiforme. The

oncologist 20, 107-108, doi:10.1634/theoncologist.2014-0418 (2015).

194 Reynes, G. et al. A phase I study of irinotecan in combination with metronomic

temozolomide in patients with recurrent glioblastoma. Anti-cancer drugs 25, 717-722,

doi:10.1097/cad.0000000000000059 (2014).

195 Chinnaiyan, P. et al. RTOG 0913: a phase 1 study of daily everolimus (RAD001) in

combination with radiation therapy and temozolomide in patients with newly diagnosed

glioblastoma. International journal of radiation oncology, biology, physics 86, 880-

884, doi:10.1016/j.ijrobp.2013.04.036 (2013).

196 Chou, T. C. Theoretical basis, experimental design, and computerized simulation of

synergism and antagonism in drug combination studies. Pharmacological reviews 58,

621-681, doi:10.1124/pr.58.3.10 (2006).

197 Franken, N. A., Rodermond, H. M., Stap, J., Haveman, J. & van Bree, C. Clonogenic

assay of cells in vitro. Nature protocols 1, 2315-2319, doi:10.1038/nprot.2006.339

(2006).

198 Scott, C. C., Vacca, F. & Gruenberg, J. Endosome maturation, transport and functions.

Seminars in cell & developmental biology 31, 2-10, doi:10.1016/j.semcdb.2014.03.034

(2014).

199 Huotari, J. & Helenius, A. Endosome maturation. The EMBO Journal 30, 3481 (2011).

200 Ferrati, S. et al. Cellular communication via nanoparticle-transporting biovesicles.

Nanomedicine (London, England) 9, 581-592, doi:10.2217/nnm.13.57 (2014).

201 Gal, N., Massalha, S., Samuelly-Nafta, O. & Weihs, D. Effects of particle uptake,

encapsulation, and localization in cancer cells on intracellular applications. Medical

engineering & physics 37, 478-483, doi:10.1016/j.medengphy.2015.03.003 (2015).

202 Schutz, I. et al. Lysosomal Dysfunction Caused by Cellular Accumulation of Silica

203

Nanoparticles. The Journal of biological chemistry 291, 14170-14184,

doi:10.1074/jbc.M115.710947 (2016).

203 Zhang, J. et al. Intracellular Trafficking Network of Protein Nanocapsules: Endocytosis,

Exocytosis and Autophagy. Theranostics 6, 2099-2113, doi:10.7150/thno.16587

(2016).

204 Wang, F., Zhu, Y., Fang, S., Li, S. & Liu, S. Effect of lanthanum chloride on tumor

growth and apoptosis in human ovarian cancer cells and xenograft animal models.

Experimental and therapeutic medicine 16, 1143-1148, doi:10.3892/etm.2018.6299

(2018).

205 Benedetto, A. et al. Effects of the rare elements lanthanum and cerium on the growth

of colorectal and hepatic cancer cell lines. Toxicology in vitro : an international journal

published in association with BIBRA 46, 9-18, doi:10.1016/j.tiv.2017.09.024 (2018).

206 Wason, M. S. et al. Sensitization of pancreatic cancer cells to radiation by cerium oxide

nanoparticle-induced ROS production. Nanomedicine : nanotechnology, biology, and

medicine 9, 558-569, doi:10.1016/j.nano.2012.10.010 (2013).

207 Montazeri, A. et al. Radiosensitizing effect of cerium oxide nanoparticles on human

leukemia cells. Pharmaceutical nanotechnology,

doi:10.2174/2211738506666180306161253 (2018).

208 Lim, C. H. Toxicity of Two Different Sized Lanthanum Oxides in Cultured Cells and

Sprague-Dawley Rats. Toxicological research 31, 181-189,

doi:10.5487/tr.2015.31.2.181 (2015).

209 Smaili, S. S. et al. The role of calcium stores in apoptosis and autophagy. Current

molecular medicine 13, 252-265 (2013).

210 Adachi, G., Imanaka, N. & Kang, Z. C. Binary Rare Earth Oxides. (Springer

Netherlands, 2006).

211 Short, S. C. et al. DNA repair after irradiation in glioma cells and normal human

204

astrocytes. Neuro-Oncology 9, 404-411, doi:10.1215/15228517-2007-030 (2007).

212 Blackburn, R. V. et al. Differential induction of cell death in human glioma cell lines

by sodium nitroprusside. Cancer 82, 1137-1145 (1998).

213 Ohgaki, H. & Kleihues, P. Genetic pathways to primary and secondary glioblastoma.

The American journal of pathology 170, 1445-1453, doi:10.2353/ajpath.2007.070011

(2007).

214 Badie, B., Goh, C. S., Klaver, J., Herweijer, H. & Boothman, D. A. Combined radiation

and p53 gene therapy of malignant glioma cells. Cancer gene therapy 6, 155-162,

doi:10.1038/sj.cgt.7700009 (1999).

215 Smit, J. K. et al. Prediction of response to radiotherapy in the treatment of esophageal

cancer using stem cell markers. Radiotherapy and oncology : journal of the European

Society for Therapeutic Radiology and Oncology 107, 434-441,

doi:10.1016/j.radonc.2013.03.027 (2013).

216 Townley, H. E., Rapa, E., Wakefield, G. & Dobson, P. J. Nanoparticle augmented

radiation treatment decreases cancer cell proliferation. Nanomedicine : nanotechnology,

biology, and medicine 8, 526-536, doi:10.1016/j.nano.2011.08.003 (2012).

217 Floersheim, G. L. Modification of radiation sensitivity by salts of the metals beryllium

and indium and the rare earths cerium, lanthanum and scandium. Radiation research

141, 318-323 (1995).

218 Lux, F. et al. Gadolinium-based nanoparticles for theranostic MRI-radiosensitization.

Nanomedicine (London, England) 10, 1801-1815, doi:10.2217/nnm.15.30 (2015).

219 Zou, X. et al. X-ray-induced nanoparticle-based photodynamic therapy of cancer.

Nanomedicine : nanotechnology, biology, and medicine 9, 2339-2351,

doi:10.2217/nnm.13.198 (2014).

220 Hirst, S. M. et al. Bio-distribution and in vivo antioxidant effects of cerium oxide

nanoparticles in mice. Environmental toxicology 28, 107-118, doi:10.1002/tox.20704

205

(2013).

221 Zal, Z. et al. Radioprotective Effect of Cerium Oxide Nanoparticles Against

Genotoxicity Induced by Ionizing Radiation on Human Lymphocytes. Current

radiopharmaceuticals 11, 109-115, doi:10.2174/1874471011666180528095203

(2018).

222 Sack, M. et al. Combination of conventional chemotherapeutics with redox-active

cerium oxide nanoparticles--a novel aspect in cancer therapy. Molecular cancer

therapeutics 13, 1740-1749, doi:10.1158/1535-7163.Mct-13-0950 (2014).

223 Xu, H. et al. Synergistic anti-glioma effect of a coloaded nano-drug delivery system.

International journal of nanomedicine 12, 29-40, doi:10.2147/ijn.S116367 (2017).

224 Xu, Y. et al. The synergic antitumor effects of paclitaxel and temozolomide co-loaded

in mPEG-PLGA nanoparticles on glioblastoma cells. Oncotarget 7, 20890-20901,

doi:10.18632/oncotarget.7896 (2016).

225 Cui, Y., Naz, A., Thompson, D. H. & Irudayaraj, J. Decitabine nanoconjugate sensitizes

human glioblastoma cells to temozolomide. Molecular pharmaceutics 12, 1279-1288,

doi:10.1021/mp500815b (2015).

226 Cloughesy, T. F., Cavenee, W. K. & Mischel, P. S. Glioblastoma: from molecular

pathology to targeted treatment. Annual review of pathology 9, 1-25,

doi:10.1146/annurev-pathol-011110-130324 (2014).

227 Doherty, G. J. & McMahon, H. T. Mechanisms of endocytosis. Annual review of

biochemistry 78, 857-902, doi:10.1146/annurev.biochem.78.081307.110540 (2009).

228 Albanese, A., Tang, P. S. & Chan, W. C. The effect of nanoparticle size, shape, and

surface chemistry on biological systems. Annual review of biomedical engineering 14,

1-16, doi:10.1146/annurev-bioeng-071811-150124 (2012).

229 Elmore, S. Apoptosis: a review of programmed cell death. Toxicologic pathology 35,

495-516, doi:10.1080/01926230701320337 (2007).

206

230 Crowley, L. C., Marfell, B. J., Scott, A. P. & Waterhouse, N. J. Quantitation of

Apoptosis and Necrosis by Annexin V Binding, Propidium Iodide Uptake, and Flow

Cytometry. Cold Spring Harbor protocols 2016, doi:10.1101/pdb.prot087288 (2016).

231 Germain, M. et al. Cleavage of automodified poly(ADP-ribose) polymerase during

apoptosis. Evidence for involvement of caspase-7. The Journal of biological chemistry

274, 28379-28384 (1999).

232 Degterev, A., Boyce, M. & Yuan, J. A decade of caspases. Oncogene 22, 8543-8567,

doi:10.1038/sj.onc.1207107 (2003).

233 Fulda, S. & Debatin, K. M. Targeting apoptosis pathways in cancer therapy. Current

cancer drug targets 4, 569-576 (2004).

234 Saelens, X. et al. Toxic proteins released from mitochondria in cell death. Oncogene

23, 2861-2874, doi:10.1038/sj.onc.1207523 (2004).

235 Walczak, H. & Krammer, P. H. The CD95 (APO-1/Fas) and the TRAIL (APO-2L)

apoptosis systems. Experimental cell research 256, 58-66, doi:10.1006/excr.2000.4840

(2000).

236 Simon, H. U., Haj-Yehia, A. & Levi-Schaffer, F. Role of reactive oxygen species (ROS)

in apoptosis induction. Apoptosis : an international journal on programmed cell death

5, 415-418, doi:10.1023/A:1009616228304 (2000).

237 Delbridge, A. R., Grabow, S., Strasser, A. & Vaux, D. L. Thirty years of BCL-2:

translating cell death discoveries into novel cancer therapies. Nature reviews. Cancer

16, 99-109, doi:10.1038/nrc.2015.17 (2016).

238 Jiang, Z., Zheng, X. & Rich, K. M. Down-regulation of Bcl-2 and Bcl-xL expression

with bispecific antisense treatment in glioblastoma cell lines induce cell death. Journal

of neurochemistry 84, 273-281 (2003).

239 Eisele, G. & Weller, M. Targeting apoptosis pathways in glioblastoma. Cancer letters

332, 335-345, doi:10.1016/j.canlet.2010.12.012 (2013).

207

240 Hoeijmakers, J. H. DNA damage, aging, and cancer. The New England journal of

medicine 361, 1475-1485, doi:10.1056/NEJMra0804615 (2009).

241 Holler, N. et al. Fas triggers an alternative, caspase-8-independent cell death pathway

using the kinase RIP as effector molecule. Nature immunology 1, 489-495,

doi:10.1038/82732 (2000).

242 Podhorecka, M., Skladanowski, A. & Bozko, P. H2AX Phosphorylation: Its Role in

DNA Damage Response and Cancer Therapy. Journal of nucleic acids 2010, 920161,

doi:10.4061/2010/920161 (2010).

243 Mizushima, N. Autophagy: process and function. Genes & development 21, 2861-2873,

doi:10.1101/gad.1599207 (2007).

244 Lockshin, R. A. & Zakeri, Z. Programmed cell death and apoptosis: origins of the

theory. Nature reviews. Molecular cell biology 2, 545-550, doi:10.1038/35080097

(2001).

245 Tanida, I., Ueno, T. & Kominami, E. LC3 and Autophagy. Methods in molecular

biology (Clifton, N.J.) 445, 77-88, doi:10.1007/978-1-59745-157-4_4 (2008).

246 Wang, L. H., Rothberg, K. G. & Anderson, R. G. Mis-assembly of clathrin lattices on

endosomes reveals a regulatory switch for coated pit formation. The Journal of cell

biology 123, 1107-1117 (1993).

247 Dutta, D. & Donaldson, J. G. Search for inhibitors of endocytosis: Intended specificity

and unintended consequences. Cellular logistics 2, 203-208, doi:10.4161/cl.23967

(2012).

248 Moncelet, D. et al. Cellular density effect on RGD ligand internalization in

glioblastoma for MRI application. PloS one 8, e82777,

doi:10.1371/journal.pone.0082777 (2013).

249 Singh, S., Kumar, A., Karakoti, A., Seal, S. & Self, W. T. Unveiling the mechanism of

uptake and sub-cellular distribution of cerium oxide nanoparticles. Molecular

208

bioSystems 6, 1813-1820, doi:10.1039/c0mb00014k (2010).

250 Vassie, J. A., Whitelock, J. M. & Lord, M. S. Endocytosis of cerium oxide nanoparticles

and modulation of reactive oxygen species in human ovarian and colon cancer cells.

Acta biomaterialia 50, 127-141, doi:10.1016/j.actbio.2016.12.010 (2017).

251 Ng, C. T. et al. Clathrin-mediated endocytosis of gold nanoparticles in vitro.

Anatomical record (Hoboken, N.J. : 2007) 298, 418-427, doi:10.1002/ar.23051 (2015).

252 Dreifuss, T. et al. Uptake mechanism of metabolic-targeted gold nanoparticles.

Nanomedicine (London, England) 13, 1535-1549, doi:10.2217/nnm-2018-0022 (2018).

253 Zhu, X. et al. Intracellular Mechanistic Understanding of 2D MoS2 Nanosheets for

Anti-Exocytosis-Enhanced Synergistic Cancer Therapy. ACS nano 12, 2922-2938,

doi:10.1021/acsnano.8b00516 (2018).

254 Suresh, D. et al. Bombesin peptide conjugated gold nanocages internalize via clathrin

mediated endocytosis. Bioconjugate chemistry 25, 1565-1579, doi:10.1021/bc500295s

(2014).

255 Goldshtein, M., Forti, E., Ruvinov, E. & Cohen, S. Mechanisms of cellular uptake and

endosomal escape of calcium-siRNA nanocomplexes. International journal of

pharmaceutics 515, 46-56, doi:10.1016/j.ijpharm.2016.10.009 (2016).

256 Bandmann, V., Muller, J. D., Kohler, T. & Homann, U. Uptake of fluorescent nano

beads into BY2-cells involves clathrin-dependent and clathrin-independent endocytosis.

FEBS letters 586, 3626-3632, doi:10.1016/j.febslet.2012.08.008 (2012).

257 Paiva, A. V. et al. Effects of lanthanum on human lymphocytes viability and DNA

strand break. Bulletin of environmental contamination and toxicology 82, 423-427,

doi:10.1007/s00128-008-9596-1 (2009).

258 Spahiu, K. & Bruno, J. A selected thermodynamic database for REE to be used in

HLNW performance assessment exercises. Report No. 0284-3757, 90 (Sweden, 1995).

259 Palmer, R. J., Butenhoff, J. L. & Stevens, J. B. Cytotoxicity of the rare earth metals

209

cerium, lanthanum, and neodymium in vitro: comparisons with cadmium in a

pulmonary macrophage primary culture system. Environmental research 43, 142-156

(1987).

260 Demaurex, N. pH Homeostasis of Cellular Organelles. Physiology 17, 1-5,

doi:10.1152/physiologyonline.2002.17.1.1 (2002).

261 Abreu, R. D. & Morais, C. A. Purification of rare earth elements from monazite

sulphuric acid leach liquor and the production of high-purity ceric oxide. Minerals

Engineering 23, 536-540, doi:https://doi.org/10.1016/j.mineng.2010.03.010 (2010).

262 Liu, H., Zhang, C. & Huang, K. Lanthanum chloride suppresses oxysterol-induced

ECV-304 cell apoptosis via inhibition of intracellular Ca(2+) concentration elevation,

oxidative stress, and activation of ERK and NF-kappaB signaling pathways. Journal of

biological inorganic chemistry : JBIC : a publication of the Society of Biological

Inorganic Chemistry 16, 671-681, doi:10.1007/s00775-011-0766-y (2011).

263 Yang, J., Liu, Q., Wu, S., Xi, Q. & Cai, Y. Effects of lanthanum chloride on glutamate

level, intracellular calcium concentration and caspases expression in the rat

hippocampus. Biometals : an international journal on the role of metal ions in biology,

biochemistry, and medicine 26, 43-59, doi:10.1007/s10534-012-9593-z (2013).

264 Su, X., Zheng, X. & Ni, J. Lanthanum citrate induces anoikis of Hela cells. Cancer

letters 285, 200-209, doi:10.1016/j.canlet.2009.05.018 (2009).

265 Shen, L. et al. Proteomic analysis of lanthanum citrate-induced apoptosis in human

cervical carcinoma SiHa cells. Biometals : an international journal on the role of metal

ions in biology, biochemistry, and medicine 23, 1179-1189, doi:10.1007/s10534-010-

9368-3 (2010).

266 Mkandawire, M. M. et al. Induction of apoptosis in human cancer cells by targeting

mitochondria with gold nanoparticles. Nanoscale 7, 10634-10640,

doi:10.1039/c5nr01483b (2015).

210

267 Cho, M. H. et al. Magnetic Tandem Apoptosis for Overcoming Multidrug-Resistant

Cancer. Nano letters 16, 7455-7460, doi:10.1021/acs.nanolett.6b03122 (2016).

268 Tsai, Y. F. et al. Gadolinium chloride elicits apoptosis in human osteosarcoma U-2 OS

cells through extrinsic signaling, intrinsic pathway and endoplasmic reticulum stress.

Oncology reports 36, 3421-3426, doi:10.3892/or.2016.5174 (2016).

269 Lee, S. Y. et al. The rare-earth yttrium complex [YR(mtbmp)(thf)] triggers apoptosis

via the extrinsic pathway and overcomes multiple drug resistance in leukemic cells.

Medical oncology (Northwood, London, England) 29, 235-242, doi:10.1007/s12032-

010-9787-6 (2012).

270 Ashley, D. M. et al. Caspase 8 is absent or low in many ex vivo gliomas. Cancer 104,

1487-1496, doi:10.1002/cncr.21323 (2005).

271 Konen-Adiguzel, S. & Ergene, S. In vitro evaluation of the genotoxicity of CeO2

nanoparticles in human peripheral blood lymphocytes using cytokinesis-block

micronucleus test, comet assay, and gamma H2AX. Toxicology and industrial health

34, 293-300, doi:10.1177/0748233717753780 (2018).

272 Lewis, D. J. et al. Intracellular synchrotron nanoimaging and DNA

damage/genotoxicity screening of novel lanthanide-coated nanovectors. Nanomedicine

5, 1547-1557, doi:10.2217/nnm.10.96 (2010).

273 Kang, M. A., So, E. Y., Simons, A. L., Spitz, D. R. & Ouchi, T. DNA damage induces

reactive oxygen species generation through the H2AX-Nox1/Rac1 pathway. Cell Death

Dis 3, e249, doi:10.1038/cddis.2011.134 (2012).

274 Brabu, B. et al. Biocompatibility studies on lanthanum oxide nanoparticles. Toxicology

Research 4, 1037-1044, doi:10.1039/C4TX00198B (2015).

275 Zhang, L. et al. Activation of Nrf2/ARE signaling pathway attenuates lanthanum

chloride induced injuries in primary rat astrocytes. Metallomics : integrated biometal

science 9, 1120-1131, doi:10.1039/c7mt00182g (2017).

211

276 Hong, J. et al. Molecular mechanism of oxidative damage of lung in mice following

exposure to lanthanum chloride. Environmental toxicology 30, 357-365,

doi:10.1002/tox.21913 (2015).

277 Green, D. R. & Reed, J. C. Mitochondria and apoptosis. Science (New York, N.Y.) 281,

1309-1312 (1998).

278 Zamzami, N. et al. Sequential reduction of mitochondrial transmembrane potential and

generation of reactive oxygen species in early programmed cell death. The Journal of

experimental medicine 182, 367-377 (1995).

279 Watson, R. W., Rotstein, O. D., Jimenez, M., Parodo, J. & Marshall, J. C. Augmented

intracellular glutathione inhibits Fas-triggered apoptosis of activated human

neutrophils. Blood 89, 4175-4181 (1997).

280 Liu, B. et al. Glutathione regulation of neutral sphingomyelinase in tumor necrosis

factor-alpha-induced cell death. The Journal of biological chemistry 273, 11313-11320

(1998).

281 Galski, H. et al. P-glycoprotein-dependent resistance of cancer cells toward the

extrinsic TRAIL apoptosis signaling pathway. Biochemical pharmacology 86, 584-596,

doi:10.1016/j.bcp.2013.06.004 (2013).

282 Wu, J. et al. Lanthanum induced primary neuronal apoptosis through mitochondrial

dysfunction modulated by Ca(2)(+) and Bcl-2 family. Biological trace element

research 152, 125-134, doi:10.1007/s12011-013-9601-3 (2013).

283 Kretowski, R. et al. The Effects of Silica Nanoparticles on Apoptosis and Autophagy

of Glioblastoma Cell Lines. Nanomaterials (Basel, Switzerland) 7,

doi:10.3390/nano7080230 (2017).

284 Pandurangan, M. et al. Time and Concentration-Dependent Therapeutic Potential of

Silver Nanoparticles in Cervical Carcinoma Cells. Biological trace element research

170, 309-319, doi:10.1007/s12011-015-0467-4 (2016).

212

285 Liang, S. et al. Role of Cyt-C/caspases-9,3, Bax/Bcl-2 and the FAS death receptor

pathway in apoptosis induced by zinc oxide nanoparticles in human aortic endothelial

cells and the protective effect by alpha-lipoic acid. Chemico-biological interactions 258,

40-51, doi:10.1016/j.cbi.2016.08.013 (2016).

286 González-Polo, R.-A. et al. The apoptosis/autophagy paradox: autophagic

vacuolization before apoptotic death. Journal of Cell Science 118, 3091 (2005).

287 Zhang, X. et al. Radiosensitivity enhancement of Fe3O4@Ag nanoparticles on human

glioblastoma cells. Artif Cells Nanomed Biotechnol, 1-10,

doi:10.1080/21691401.2018.1439843 (2018).

288 Zhang, Y., Yu, C., Huang, G., Wang, C. & Wen, L. Nano rare-earth oxides induced

size-dependent vacuolization: an independent pathway from autophagy. International

journal of nanomedicine 5, 601-609, doi:10.2147/ijn.S11513 (2010).

289 Hussain, S. et al. Cerium dioxide nanoparticles induce apoptosis and autophagy in

human peripheral blood monocytes. ACS nano 6, 5820-5829, doi:10.1021/nn302235u

(2012).

290 Yu, L., Lu, Y., Man, N., Yu, S.-H. & Wen, L.-P. Rare Earth Oxide Nanocrystals Induce

Autophagy in HeLa Cells. Small 5, 2784-2787, doi:10.1002/smll.200901714 (2009).

291 Capiod, T. Cell proliferation, calcium influx and calcium channels. Biochimie 93, 2075-

2079, doi:10.1016/j.biochi.2011.07.015 (2011).

292 Gao, J. et al. Determining the Cytotoxicity of Rare Earth Element Nanoparticles in

Macrophages and the Involvement of Membrane Damage. Environmental science &

technology 51, 13938-13948, doi:10.1021/acs.est.7b04231 (2017).

293 Zou, X. et al. X-ray-induced nanoparticle-based photodynamic therapy of cancer.

Nanomedicine (London, England) 9, 2339-2351, doi:10.2217/nnm.13.198 (2014).

294 Kudinov, K. et al. in Colloidal Nanoparticles for Biomedical Applications XI. 97220V

(International Society for Optics and Photonics).

213

295 Eberlein, U. et al. DNA damage in blood lymphocytes in patients after (177)Lu peptide

receptor radionuclide therapy. European journal of nuclear medicine and molecular

imaging 42, 1739-1749, doi:10.1007/s00259-015-3083-9 (2015).

296 Phaeton, R. et al. Beta emitters rhenium-188 and lutetium-177 are equally effective in

radioimmunotherapy of HPV-positive experimental cervical cancer. Cancer medicine

5, 9-16, doi:10.1002/cam4.562 (2016).

297 Hanif, F., Perveen, K., Malhi, S. M., Jawed, H. & Simjee, S. U. Verapamil potentiates

anti-glioblastoma efficacy of temozolomide by modulating apoptotic signaling.

Toxicology in vitro : an international journal published in association with BIBRA 52,

306-313, doi:10.1016/j.tiv.2018.07.001 (2018).

298 Zhang, X., Ni, Q., Wang, Y., Fan, H. & Li, Y. Synergistic Anticancer Effects of

Formononetin and Temozolomide on Glioma C6 Cells. Biological & pharmaceutical

bulletin 41, 1194-1202, doi:10.1248/bpb.b18-00002 (2018).

299 Hanif, F. et al. N-(2-hydroxyphenyl)acetamide (NA-2) and Temozolomide

synergistically induce apoptosis in human glioblastoma cell line U87. Cancer cell

international 14, 133, doi:10.1186/s12935-014-0133-5 (2014).

300 Rim, K. T., Koo, K. H. & Park, J. S. Toxicological evaluations of rare earths and their

health impacts to workers: a literature review. Safety and health at work 4, 12-26,

doi:10.5491/SHAW.2013.4.1.12 (2013).

301 Hirano, S. & Suzuki, K. T. Exposure, metabolism, and toxicity of rare earths and related

compounds. Environmental health perspectives 104 Suppl 1, 85-95,

doi:10.1289/ehp.96104s185 (1996).

302 Wermuth, P. J. & Jimenez, S. A. Gadolinium compounds signaling through TLR4 and

TLR7 in normal human macrophages: establishment of a proinflammatory phenotype

and implications for the pathogenesis of nephrogenic systemic fibrosis. Journal of

immunology (Baltimore, Md. : 1950) 189, 318-327, doi:10.4049/jimmunol.1103099

214

(2012).

303 Li, R. et al. Surface interactions with compartmentalized cellular phosphates explain

rare earth oxide nanoparticle hazard and provide opportunities for safer design. ACS

nano 8, 1771-1783, doi:10.1021/nn406166n (2014).

304 Kreuter, J. Evaluation of nanoparticles as drug-delivery systems. II: Comparison of the

body distribution of nanoparticles with the body distribution of microspheres (diameter

greater than 1 micron), liposomes, and emulsions. Pharmaceutica acta Helvetiae 58,

217-226 (1983).

305 Rang, H. P. & Hill, R. G. in Drug Discovery and Development (Second Edition) (eds

R. G. Hill & H. P. Rang) 211-225 (Churchill Livingstone, 2013).

306 Jain, T. K., Reddy, M. K., Morales, M. A., Leslie-Pelecky, D. L. & Labhasetwar, V.

Biodistribution, clearance, and biocompatibility of iron oxide magnetic nanoparticles

in rats. Molecular pharmaceutics 5, 316-327, doi:10.1021/mp7001285 (2008).

307 Fraga, S. et al. Short- and long-term distribution and toxicity of gold nanoparticles in

the rat after a single-dose intravenous administration. Nanomedicine : nanotechnology,

biology, and medicine 10, 1757-1766, doi:10.1016/j.nano.2014.06.005 (2014).

308 Smulders, S. et al. Body distribution of SiO(2)-Fe(3)O(4) core-shell nanoparticles after

intravenous injection and intratracheal instillation. Nanotoxicology 10, 567-574,

doi:10.3109/17435390.2015.1100761 (2016).

309 Tseng, M. T. et al. Alteration of hepatic structure and oxidative stress induced by

intravenous nanoceria. Toxicology and applied pharmacology 260, 173-182,

doi:10.1016/j.taap.2012.02.008 (2012).

310 Das, S. et al. Tissue deposition and toxicological effects of commercially significant

rare earth oxide nanomaterials: Material and physical properties. Environmental

Toxicology 32, 904-917, doi:10.1002/tox.22290 (2016).

311 Geraets, L. et al. Tissue distribution and elimination after oral and intravenous

215

administration of different titanium dioxide nanoparticles in rats. Particle and fibre

toxicology 11, 30-30, doi:10.1186/1743-8977-11-30 (2014).

312 Kreuter, J. Nanoparticulate systems for brain delivery of drugs. Advanced Drug

Delivery Reviews 47, 65-81, doi:https://doi.org/10.1016/S0169-409X(00)00122-8

(2001).

313 Yang, S. C. et al. Body distribution in mice of intravenously injected camptothecin

solid lipid nanoparticles and targeting effect on brain. 59, 299-307 (1999).

314 Wadajkar, A. S. et al. Tumor-targeted nanotherapeutics: overcoming treatment barriers

for glioblastoma. Wiley interdisciplinary reviews. Nanomedicine and

nanobiotechnology 9, doi:10.1002/wnan.1439 (2017).

315 Stupp, R. et al. Radiotherapy plus Concomitant and Adjuvant Temozolomide for

Glioblastoma. New England Journal of Medicine 352, 987-996,

doi:10.1056/NEJMoa043330 (2005).

316 Appelboom, G. et al. Stereotactic modulation of blood-brain barrier permeability to

enhance drug delivery. Neuro-oncology 18, 1601-1609, doi:10.1093/neuonc/now137

(2016).

317 Tamborini, M. et al. A Combined Approach Employing Chlorotoxin-Nanovectors and

Low Dose Radiation To Reach Infiltrating Tumor Niches in Glioblastoma. ACS nano

10, 2509-2520, doi:10.1021/acsnano.5b07375 (2016).

318 Etame, A. B. et al. Enhanced delivery of gold nanoparticles with therapeutic potential

into the brain using MRI-guided focused ultrasound. Nanomedicine : nanotechnology,

biology, and medicine 8, 1133-1142, doi:10.1016/j.nano.2012.02.003 (2012).

319 Yokel, R. A. et al. Distribution, elimination, and biopersistence to 90 days of a

systemically introduced 30 nm ceria-engineered nanomaterial in rats. Toxicological

sciences : an official journal of the Society of Toxicology 127, 256-268,

doi:10.1093/toxsci/kfs067 (2012).

216

320 Yokel, R. A. et al. The Yin: An adverse health perspective of nanoceria: uptake,

distribution, accumulation, and mechanisms of its toxicity. Environmental science.

Nano 1, 406-428, doi:10.1039/C4EN00039K (2014).

321 Opina, A. C. et al. Preparation and long-term biodistribution studies of a PAMAM

dendrimer G5-Gd-BnDOTA conjugate for lymphatic imaging. Nanomedicine (London,

England) 10, 1423-1437, doi:10.2217/nnm.14.113 (2015).

322 Jung, S. et al. Intracellular gold nanoparticles increase neuronal excitability and

aggravate seizure activity in the mouse brain. PloS one 9, e91360,

doi:10.1371/journal.pone.0091360 (2014).

323 He, X., Zhang, Z., Zhang, H., Zhao, Y. & Chai, Z. Neurotoxicological evaluation of

long-term lanthanum chloride exposure in rats. Toxicological sciences : an official

journal of the Society of Toxicology 103, 354-361, doi:10.1093/toxsci/kfn046 (2008).

324 Chen, D., Liu, Y., Chen, A. J. & Nie, Y. X. Experimental study of subchronic toxicity

of lanthanum nitrate on liver in rats. Nonlinearity in biology, toxicology, medicine 1,

469-480, doi:10.1080/15401420390271074 (2003).

325 Heckman, K. L. et al. Custom cerium oxide nanoparticles protect against a free radical

mediated autoimmune degenerative disease in the brain. ACS nano 7, 10582-10596,

doi:10.1021/nn403743b (2013).

326 Portnow, J. et al. The neuropharmacokinetics of temozolomide in patients with

resectable brain tumors: potential implications for the current approach to

chemoradiation. 15, 7092-7098 (2009).

327 Marzolini, C. et al. Pharmacokinetics of temozolomide in association with fotemustine

in malignant melanoma and malignant glioma patients: comparison of oral, intravenous,

and hepatic intra-arterial administration. Cancer chemotherapy and pharmacology 42,

433-440, doi:10.1007/s002800050842 (1998).

328 Gulyaev, A. E. et al. Significant transport of doxorubicin into the brain with polysorbate

217

80-coated nanoparticles. 16, 1564-1569 (1999).

329 Cho, K., Wang, X., Nie, S., Chen, Z. G. & Shin, D. M. Therapeutic nanoparticles for

drug delivery in cancer. Clinical cancer research : an official journal of the American

Association for Cancer Research 14, 1310-1316, doi:10.1158/1078-0432.Ccr-07-1441

(2008).

330 Geraets, L., Oomen, A. G., Schroeter, J. D., Coleman, V. A. & Cassee, F. R. Tissue

distribution of inhaled micro- and nano-sized cerium oxide particles in rats: results from

a 28-day exposure study. Toxicological sciences : an official journal of the Society of

Toxicology 127, 463-473, doi:10.1093/toxsci/kfs113 (2012).

331 Chen, J. P. et al. Potential Clinical Risk of Inflammation and Toxicity from Rare-Earth

Nanoparticles in Mice. Chinese medical journal 131, 1591-1597, doi:10.4103/0366-

6999.235105 (2018).

332 Li, Q., Wang, Z., Chen, Y. & Zhang, G. Elemental bio-imaging of PEGylated

NaYF4:Yb/Tm/Gd upconversion nanoparticles in mice by laser ablation inductively

coupled plasma mass spectrometry to study toxic side effects on the spleen, liver and

kidneys. Metallomics : integrated biometal science 9, 1150-1156,

doi:10.1039/c7mt00132k (2017).

333 Nemmar, A., Yuvaraju, P., Beegam, S., Fahim, M. A. & Ali, B. H. Cerium Oxide

Nanoparticles in Lung Acutely Induce Oxidative Stress, Inflammation, and DNA

Damage in Various Organs of Mice. Oxidative medicine and cellular longevity 2017,

9639035, doi:10.1155/2017/9639035 (2017).

334 Wang, B., He, X., Zhang, Z., Zhao, Y. & Feng, W. Metabolism of nanomaterials in

vivo: blood circulation and organ clearance. Accounts of chemical research 46, 761-

769, doi:10.1021/ar2003336 (2013).

335 Szabo, G., Benyo, I., Sandor, J. & Benyo, Z. Estimation of the hepatic blood flow in

the dog with the Xe133 and hydrogen wash-out Au190 -colloid uptake techniques and

218

with the electromagnetic flowmeter. Research in experimental medicine. Zeitschrift fur

die gesamte experimentelle Medizin einschliesslich experimenteller Chirurgie 169, 69-

76 (1976).

336 Yu, Q. et al. Biodistribution and Toxicity Assessment of Superparamagnetic Iron Oxide

Nanoparticles In Vitro and In Vivo. Current medical science 38, 1096-1102,

doi:10.1007/s11596-018-1989-8 (2018).

337 Hutchison, A. J., Wilson, R. J., Garafola, S. & Copley, J. B. Lanthanum carbonate:

safety data after 10 years. Nephrology (Carlton, Vic.) 21, 987-994,

doi:10.1111/nep.12864 (2016).

338 Su, H. et al. Biosafety evaluation of Janus Fe(3)O(4)-TiO(2) nanoparticles in Sprague

Dawley rats after intravenous injection. International journal of nanomedicine 13,

6987-7001, doi:10.2147/IJN.S167851 (2018).

339 Tiwari, D. K., Jin, T. & Behari, J. Dose-dependent in-vivo toxicity assessment of silver

nanoparticle in Wistar rats. Toxicology mechanisms and methods 21, 13-24,

doi:10.3109/15376516.2010.529184 (2011).

340 Bae, S. H. et al. Toxicity profile of temozolomide in the treatment of 300 malignant

glioma patients in Korea. Journal of Korean medical science 29, 980-984,

doi:10.3346/jkms.2014.29.7.980 (2014).

341 Agarwala, S. S. & Kirkwood, J. M. Temozolomide, a Novel Alkylating Agent with

Activity in the Central Nervous System, May Improve the Treatment of Advanced

Metastatic Melanoma. 5, 144-151, doi:10.1634/theoncologist.5-2-144 (2000).

342 Hoopes, P. J. et al. The effect of hypofractionated radiation and magnetic nanoparticle

hyperthermia on tumor immunogenicity and overall treatment response. Proceedings

of SPIE--the International Society for Optical Engineering 10066, 100660D,

doi:10.1117/12.2255981 (2017).

343 Walkey, C. et al. Catalytic Properties and Biomedical Applications of Cerium Oxide

219

Nanoparticles. Environmental science. Nano 2, 33-53, doi:10.1039/c4en00138a (2015).

344 Allouni, Z. E., Cimpan, M. R., Høl, P. J., Skodvin, T. & Gjerdet, N. R. Agglomeration

and sedimentation of TiO2 nanoparticles in cell culture medium. Colloids and Surfaces

B: Biointerfaces 68, 83-87, doi:https://doi.org/10.1016/j.colsurfb.2008.09.014 (2009).

345 Lazarovits, J., Chen, Y. Y., Sykes, E. A. & Chan, W. C. W. Nanoparticle–blood

interactions: the implications on solid tumour targeting. Chemical Communications 51,

2756-2767, doi:10.1039/C4CC07644C (2015).

346 Lockman, P. R., Koziara, J. M., Mumper, R. J. & Allen, D. D. Nanoparticle Surface

Charges Alter Blood–Brain Barrier Integrity and Permeability. Journal of Drug

Targeting 12, 635-641, doi:10.1080/10611860400015936 (2004).

347 Dan, M. et al. Brain microvascular endothelial cell association and distribution of a 5

nm ceria engineered nanomaterial. International journal of nanomedicine 7, 4023-4036,

doi:10.2147/ijn.S32526 (2012).

348 Choi, H. S. et al. Renal clearance of quantum dots. Nature biotechnology 25, 1165-

1170, doi:10.1038/nbt1340 (2007).

349 Zhang, L. et al. Nanoparticles in Medicine: Therapeutic Applications and

Developments. Clinical Pharmacology & Therapeutics 83, 761-769,

doi:10.1038/sj.clpt.6100400 (2008).

350 Abbott, N. J., Rönnbäck, L. & Hansson, E. Astrocyte–endothelial interactions at the

blood–brain barrier. Nature Reviews Neuroscience 7, 41, doi:10.1038/nrn1824 (2006).

351 Malhotra, M. et al. Targeted nanoparticle delivery of therapeutic antisense microRNAs

presensitizes glioblastoma cells to lower effective doses of temozolomide in vitro and

in a mouse model. Oncotarget 9, 21478-21494, doi:10.18632/oncotarget.25135 (2018).

352 O'Brien, M. N., Jones, M. R. & Mirkin, C. A. The nature and implications of uniformity

in the hierarchical organization of nanomaterials. Proc Natl Acad Sci U S A 113, 11717-

11725, doi:10.1073/pnas.1605289113 (2016).

220

353 Sakthivel, T. et al. Morphological Phase Diagram of Biocatalytically Active Ceria

Nanostructures as a Function of Processing Variables and Their Properties.

ChemPlusChem 78, 1446-1455, doi:10.1002/cplu.201300302 (2013).

354 Zheng, M., Wang, S., Liu, Z., Xie, L. & Deng, Y. Development of temozolomide coated

nano zinc oxide for reversing the resistance of malignant glioma stem cells. Materials

science & engineering. C, Materials for biological applications 83, 44-50,

doi:10.1016/j.msec.2017.07.015 (2018).

355 Lei, C., Davoodi, P., Zhan, W., Kah-Hoe Chow, P. & Wang, C. H. Development of

Nanoparticles for Drug Delivery to Brain Tumor: The Effect of Surface Materials on

Penetration into Brain Tissue. Journal of pharmaceutical sciences,

doi:10.1016/j.xphs.2018.12.002 (2018).

221