Self-assembled polymeric micelles as powders for pulmonary administration of insulin

FERNANDA RAQUEL DA SILVA ANDRADE PHD IN PHARMACEUTICAL SCIENCES PHARMACEUTICAL TECHNOLOGY SELF-ASSEMBLED POLYMERIC MICELLES AS POWDERS FOR PULMONARY ADMINISTRATION OF INSULIN PORTO 2015 RUA DE JORGE VITERBO FERREIRA N.º 228 4050-313 PORTO - PORTUGAL WWW.FF.UP.PT THESIS SUBMITTED TO THE FACULTY OF PHARMACY OF THE UNIVERSITY OF PORTO FOR APPROVAL OF THE PHD DEGREE Fernanda Raquel da Silva Andrade SELF-ASSEMBLED POLYMERIC MICELLES AS POWDERS FOR PULMONARY ADMINISTRATION OF INSULIN This page was intentionally left in blank Fernanda Raquel da Silva Andrade Self-assembled polymeric micelles as powders for pulmonary administration of insulin Thesis Submitted in fulfilment of the requirements to obtain the PhD degree in Pharmaceutical Sciences, Pharmaceutical Technology Specialty Tese do 3.º Ciclo de Estudos Conducente ao Grau de Doutoramento em Ciências Farmacêuticas na Especialidade de Tecnologia Farmacêutica Work developed under supervision of Prof. Dr. Bruno Sarmento and co-supervision of Prof. Dr. Domingos de Carvalho Ferreira and Prof. Dr. Mafalda Videira May, 2015 ii The full reproduction of this thesis is allowed for research purposes only, through a written declaration of the person concerned, to which he commits to. É autorizada a reprodução integral desta Tese apenas para efeitos de investigação, mediante declaração escrita do interessado, que a tal se compromete. Fernanda Raquel da Silva Andrade iii “To try and fail is at least to learn; to fail to try is to suffer the inestimable loss of what might have been.” Chester Barnard iv v Acknowledgments I would like to express my gratitude to all the people and institutions that receive me, help me and support me during the performance of this work. Thus, I would like to acknowledge: My supervisor Professor Bruno Sarmento and my co-supervisors Professor Domingos de Carvalho Ferreira and Professora Mafalda Videira for believe in my capabilities and for accepting the challenge of participate and supervise this work and the present thesis. Professor Bruno Sarmento from Instituto de Engenharia Biomédica (INEB) and Instituto de Investigação e Inovação em Saúde (I3S) da Universidade do Porto, Porto, Portugal, and from Instituto de Investigação e Formação Avançada em Ciências e Tecnologias da Saúde (IINFACTS) do Instituto Superior de Ciências de Saúde do Norte (ISCS-N) da Cooperativa de Ensino Superior Politécnico e Universitário (CESPU), Gandra Portugal, for all the guidance and support in the easiest and most difficult moments. All the friendship and trust placed in me to perform this and other works and for all the opportunities provided to make me grow as a researcher. Also for all the scientific discussions, comments and corrections performed to the common publications and the present thesis. Professor Domingos de Carvalho Ferreira from Laboratório de Tecnologia Farmacêutica da Faculdade de Farmácia, Universidade do Porto (FFUP), Porto, Portugal, for all the friendship, understanding and support in the most fun and boring moments of the work. All the availability provided to make possible the progression of this work. Professor Mireia Oliva from Departament de Farmàcia i Tecnologia Farmacèutica, Facultat de Farmàcia, Universitat de Barcelona (UB), Barcelona, España, from Nanoprobes and Nanoswitches Group, Institute for Bioengineering of Catalonia (IBEC), Barcelona, España, and from Centro Investigación Biomédica en Red – Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Madrid, España for all the availability to receive me in her laboratory at UB and to introduce me in IBEC to perform part of the work. For all the hospitality, friendship, and trust and all the scientific discussions, comments and corrections performed to the common papers and the present thesis. vi Professor Mafalda Videira from Instituto de Investigação do Medicamento da Faculdade de Farmácia, Universidade de Lisboa (iMed.ULisboa), Lisboa Portugal for all the support and for receiving me for a stay to perform part of the work in her research group. For all the comments and corrections performed to the common papers and the present thesis. The members of Laboratório de Tecnologia Farmacêutica da FFUP, especially to its Director, for accepting me as PhD student in the department and for their support to this work. The members of Departament de Farmàcia i Tecnologia Farmacèutica, Facultat de Farmàcia, and Servei de Desenvolupament del Medicament (SDM) at UB for their support to this work. The members of Nanoprobes and Nanoswitches group from IBEC, especially Professor Fausto Sanz, Professor Pau Gorostiza, and Dr. Marina Giannotti, for receiving me in the group for a stay and their support to this work. The members of ISCS-N/CESPU, especially Professor Vítor Seabra for all the support to this work and for allowing me to perform the in vivo experiments at ISCS-N. The members of Departament de Farmàcia i Tecnologia Farmacèutica, Facultat de Farmàcia, UB, especially Professor Ana Calpena and Mireia Mallandrich, and Departament de Fisicoquímica, Facultat de Farmàcia, UB, especially Professor Marisa García for their support to this work. The members of REQUIMTE, Departamento de Ciências Químicas da FFUP, especially Professor Salette Reis, Dr. Marina Pinheiro, Ana Rute Neves, and Ana Catarina Alves for their support to this work. The members of Centre d’Investigacions en Bioquímica I Biologia Molecular en Nanomedicina (CIBBIM-Nanomedicina), Vall d’Hebron Institut de Recerca (VHIR), especially Dr. Simó Schwartz, Dr. Petra Gener, Diana Rafael, Dr. Juan Sayos and Dr. Aroa Ejarque for the assistance in the uptake studies with macrophages. xiii Table of contents Acknowledgments ............................................................................................................. v Abstract .............................................................................................................................. ix Resumo............................................................................................................................... xi List of figures .................................................................................................................. xix List of tables ................................................................................................................... xxv Abbreviations ................................................................................................................ xxix Chapter 1 State-of-art ....................................................................................................... 1 1. Drug delivery systems: innovation and technology ......................................................... 2 1.1. Nanotechnology in the development of drug delivery systems ............................... 4 1.1.1. Lipid-based nanoparticles .................................................................................... 7 1.1.2. Polymer-based nanoparticles .............................................................................. 9 1.1.3. Polymeric micelles ............................................................................................. 10 1.2. The role of amphiphilic polymers in the development of drug delivery systems……………………………………………………..………………………………….13 1.2.1. Synthesis of copolymers .................................................................................... 16 1.2.2. Characteristics of copolymers and copolymer-based structures ........................ 17 1.2.2.1. Stimuli-responsiveness ................................................................................... 17 1.2.2.2. Self-assembly: the crucial phenomenon ......................................................... 17 1.2.2.3. Hydrophilic surface: stability and functional role ............................................. 20 1.3. Safety of nanocarriers .......................................................................................... 21 2. Therapeutic peptides and proteins ............................................................................... 22 2.1. Properties ............................................................................................................. 22 2.2. Stability and formulation challenges ..................................................................... 23 xiv 2.3. Administration routes of therapeutic peptides and proteins .................................. 23 3. Pulmonary administration as non-invasive route for systemic delivery of therapeutic peptides and proteins ...................................................................................................... 24 3.1. Brief history of inhalation ...................................................................................... 24 3.2. Anatomo-physiological characteristics of lungs and airways ................................ 25 3.3. Pulmonary biodistribution of inhaled peptides and proteins .................................. 27 3.4. Formulation requirements for pulmonary delivery of drugs ................................... 29 3.4.1. Aerodynamic properties of particles .................................................................. 30 3.4.2. Excipients used in the development of inhalatory formulations .......................... 32 3.4.3. Inhalation devices ............................................................................................. 35 3.5. Limitations of pulmonary administration ............................................................... 36 4. State-of-art on therapeutic peptides and proteins for inhalation ................................... 37 4.1. The new era of pulmonary administration: nanomedicine-based formulations ...... 41 4.1.1. Lipid-based formulations ................................................................................... 42 4.1.2. Polymeric nanoparticles .................................................................................... 45 5. State-of-art of micelles as drug delivery systems by inhalation .................................... 47 5.1. Lipid-polymer micelles .......................................................................................... 48 5.2. Copolymer-based micelles ................................................................................... 50 Chapter 2 Aims and Goals ............................................................................................ 55 Chapter 3 Design and characterization of self-assembled micelles for insulin delivery .............................................................................................................................. 59 1. Introduction ................................................................................................................. 60 2. Experimental ............................................................................................................... 61 2.1. Materials .............................................................................................................. 61 2.2. Production of micelles .......................................................................................... 61 2.3. Determination of size, zeta potential, association efficiency, and osmolality of formulations ................................................................................................................ 62 2.4. Morphological characterization of micelles ........................................................... 63 2.5. Statistical analysis ................................................................................................ 63 3. Results ........................................................................................................................ 63 xv 3.1. Size, surface charge and association efficiency of micelles .................................. 63 3.2. Morphological characterization ............................................................................. 67 4. Discussion ................................................................................................................... 70 5. Conclusions ................................................................................................................. 74 Chapter 4 Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization............................................................ 75 1. Introduction .................................................................................................................. 76 2. Experimental ................................................................................................................ 76 2.1. Materials .............................................................................................................. 76 2.2. Production of micelles and lyophilization .............................................................. 77 2.3. Determination of size and zeta potential of formulations ....................................... 77 2.4. Thermal analysis .................................................................................................. 77 2.5. X-ray diffraction (XRD) experiments ..................................................................... 78 2.6. Raman spectroscopy ............................................................................................ 78 2.7. Surface analysis ................................................................................................... 78 2.8. Assessment of insulin conformation ..................................................................... 79 2.9. Scanning electron microscopy .............................................................................. 79 2.10. Powder’s particle size distribution and aerodynamic diameter ............................ 80 2.11. In vitro aerosolization and deposition properties ................................................. 80 2.12. Insulin in vitro release study ............................................................................... 81 2.13. Stability studies .................................................................................................. 81 2.14. Statistical analysis .............................................................................................. 82 3. Results ........................................................................................................................ 83 3.1. Determination of size and zeta potential of formulations ....................................... 83 3.2. Thermal analysis .................................................................................................. 83 3.3. XRD analysis ........................................................................................................ 84 3.4. Raman spectroscopy ............................................................................................ 85 3.5. Surface analysis ................................................................................................... 88 3.6. Protein conformation ............................................................................................ 89 3.7. Morphology and particle size distribution of powders ............................................ 91 xvi 3.8. Deposition profile of formulations ......................................................................... 93 3.9. Determination of the insulin release pattern from micelles ................................... 95 3.10. Stability of formulations upon storage ................................................................ 96 4. Discussion ................................................................................................................. 103 5. Conclusions ............................................................................................................... 110 Chapter 5 In vitro biological assessment of powder formulations for inhalation of insulin ......................................................................................................................... 113 1. Introduction ............................................................................................................... 114 2. Experimental ............................................................................................................. 115 2.1. Materials ............................................................................................................ 115 2.2. Production of micelles and lyophilization ............................................................ 115 2.3. Conjugation of polymers with 5-DTAF ................................................................ 116 2.4. Production and characterization of fluorescent micelles ..................................... 116 2.5. Cell lines and culture conditions ......................................................................... 117 2.6. Assessment of cytotoxicity ................................................................................. 117 2.7. Permeability of insulin through pulmonary epithelium ......................................... 118 2.8. Interaction of micelles with macrophages ........................................................... 119 2.9. Statistical analysis .............................................................................................. 120 3. Results ...................................................................................................................... 120 3.1. In vitro assessment of the effect of formulations on cell membrane toxicity and viability ...................................................................................................................... 120 3.2. Determination of the apparent permeability coefficient of insulin through pulmonary epithelium ................................................................................................................. 123 3.3. Characterization of fluorescent micelles ............................................................. 125 3.4. Uptake of micelles by human macrophages ....................................................... 126 4. Discussion ................................................................................................................. 129 5. Conclusions ............................................................................................................... 131 xvii Chapter 6 In vivo pharmacological and toxicological assessment of powder formulations for inhalation of insulin ........................................................................ 133 1. Introduction ................................................................................................................ 134 2. Experimental .............................................................................................................. 134 2.1. Materials ............................................................................................................ 134 2.2. Production of powder formulations ..................................................................... 135 2.3. Animals .............................................................................................................. 135 2.4. In vivo pharmacological activity of insulin ........................................................... 136 2.5. Sub-acute toxicity of insulin-loaded polymeric micelles ...................................... 137 2.6. Histological analysis ........................................................................................... 137 2.7. Statistical analysis .............................................................................................. 138 3. Results ...................................................................................................................... 138 3.1. Pharmacological activity of insulin-loaded polymeric micelles............................. 138 3.2. Sub-acute toxicity ............................................................................................... 141 4. Discussion ................................................................................................................. 147 5. Conclusions ............................................................................................................... 150 Chapter 7 General conclusions and future perspectives ...................................... 151 References.…………………………………………………………………....….……...……….155 xviii xix List of figures Chapter 1 State-of-art Figure 1.1 Schematic representation of a multi-functional DDS….......................................2 Figure 1.2 Schematic representation of a liposome…………………………………………...8 Figure 1.3 Schematic representation of SLN (S) and NLC (B)…………………………........9 Figure 1.4 Schematic representation of a polymeric nanoparticle……………………........10 Figure 1.5 Schematic representation of a micelle………………………………………........11 Figure 1.6 Schematic representation of micellization………………………………………...18 Figure 1.7 Schematic representation of the bronchial epithelium…………………………..26 Figure 1.8 Schematic representation of the alveolar epithelium…………………………….27 Figure 1.9 Schematic representation of absorption routes…………………………………..29 Figure 1.10 Deposition profile of particles on the different areas of the respiratory system according to their aerodynamic diameter………………………………………………………31 Chapter 2 Aims and Goals Figure 2.1 General structure of Soluplus® (A) and Pluronic® (B)………………………….…………………………………………………………………........56 Chapter 3 Design and characterization of self-assembled micelles for insulin delivery Figure 3.1 Mean hydrodynamic diameter, polydispersity index (PdI) and zeta potential of SOL (black bars and squares) (A), F68 (grey bars and triangles) (A), F108 (black bars and squares) (B) and F127 (grey bars and triangles) (B) empty micelles, containing just PBA micelles (empty:PBA), insulin-loaded micelles with different polymer:insulin ratio (10:0.1, 10:0.2, 10:0.3, 10:0.4, 10:0.5, 10:0.75 and 10:1) and insulin-loaded containing PBA micelles with 10:1 polymer:insulin ratio (10:1:PBA) after production (mean ± SD, n≥3)…………………………………………………………………………………………………65 xx Figure 3.2 FE-SEM micrographs of SOL (A), F68 (B), F108 (C) and F127 (D) insulinloaded micelles……………………………………………………………………………….…...67 Figure 3.3 TEM images of SOL (A-C) and F68 (B-D) empty micelles (A-B) and insulinloaded micelles (C-D)………………………………………………………………………........68 Figure 3.4 TEM images of F108 (A-C) and F127 (B-D) empty micelles (A-B) and insulinloaded micelles (C-D)…………………………………………………………………………….68 Figure 3.5 AFM images of SOL (A-B) and F68 (C-D) insulin-loaded micelles (A-C) and insulin-loaded micelles containing PBA (B-D)…………………………………………………69 Figure 3.6 AFM images of F108 (A-B) and F127 (C-D) insulin-loaded micelles (A-C) and insulin-loaded micelles containing PBA (B-D)…………………………………………………70 Chapter 4 Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization Figure 4.1 Mean hydrodynamic diameter, polydispersity index (PdI) and zeta potential of SOL (black bars and squares), F68 (dark grey bars and triangles), F108 (medium grey bars and squares) and F127 (light grey bars and triangles) based empty, containing just PBA (empty:PBA), insulin-loaded (Mic:Ins) and insulin-loaded containing PBA (Mic:Ins:PBA) lyophilized micelles after dispersion in water (mean ± SD, n≥3)…………………………………………………………………………………………………82 Figure 4.2 DSC thermograms of raw materials, polymer insulin physical mixture, insulinloaded (polymer:Ins) and insulin-loaded lyophilized micelles containing PBA (polymer:Ins:PBA) of SOL (A), F68 (B), F108 (C), and F127 (D)……………………………84 Figure 4.3 XRD patterns of insulin-loaded lyophilized micelles (Mic:Ins) and insulin-loaded containing PBA (Mic:Ins:PBA) lyophilized micelles of SOL (A), F68 (B), F108 (C), and F127 (D)……………………………………………………………………………………………85 Figure 4.4 Raman spectra of insulin-loaded (Mic:Ins) and insulin-loaded containing PBA (Mic:Ins:PBA) lyophilized micelles of SOL (A), F68 (B), F108 (C), and F127 (D)…………………………………………………………………………………………………..86 Figure 4.5 Area-normalized second-derivative amide I spectra of insulin solution 30 mg/mL, insulin-loaded (polymer:Ins), and insulin-loaded containing PBA (polymer:Ins:PBA) lyophilized micelles of SOL (A), F68 (B), F108 (C), and F127 (D)…………………………………………………………………………………………………..89 xxi Figure 4.6 far-UV CD spectra of insulin-loaded (polymer:Ins) and insulin-loaded containing PBA (polymer:Ins:PBA) lyophilized micelles of SOL (A), F68 (B), F108 (C), and F127 (D)………………………………………………………………………………………….……….91 Figure 4.7 SEM micrographs of insulin-loaded formulations composed of SOL (A), F68 (B), F108 (C), and F127 (D), without (top panel) or with (bottom panel) PBA. Scale bar: 400 µm in formulations without PBA and 100 µm in formulations with PBA……………….92 Figure 4.8 In vitro release profiles of insulin from different formulations in PBS (pH 7.4) without glucose (A) and with 1.2 mM glucose (B). Results are presented as mean ± SD (n=3)………………………………………………………………………………………………..95 Figure 4.9 Mean hydrodynamic diameter, polydispersity index (PdI) and zeta potential of SOL (A) and F68 (B)-based lyophilized insulin-loaded (Mic:Ins) and insulin-loaded containing PBA (Mic:Ins:PBA) micelles stored for 1 month (black bars and squares), 3 months (medium grey bars and squares), and 6 months (light grey bars and squares) at 4 ºC and 20 ºC after redispersion in water (mean ± SD, n=3)………………………………….97 Figure 4.10 Mean hydrodynamic diameter, polydispersity index (PdI) and zeta potential of F108 (A) and F127 (B)-based lyophilized insulin-loaded (Mic:Ins) and insulin-loaded containing PBA (Mic:Ins:PBA) micelles stored for 1 month (black bars and squares), 3 months (medium grey bars and squares), and 6 months (light grey bars and squares) at 4 ºC and 20 ºC after redispersion in water (mean ± SD, n=3)………………………………….98 Figure 4.11 Area-normalized second-derivative amide I spectra of insulin solution 30 mg/mL and insulin-loaded micelles (polymer:ins) after lyophilization (t0) and upon 1 month (t1), 3 months (t3) and 6 months (t6) of storage at 4 ºC and 20 ºC…………………………99 Figure 4.12 Area-normalized second-derivative amide I spectra of insulin solution 30 mg/mL and insulin-loaded micelles containing PBA (polymer:ins:PBA) after lyophilization (t0) and upon 1 month (t1), 3 months (t3) and 6 months (t6) of storage at 4 ºC and 20 ºC………………………………………………………………………………………………….101 Figure 4.13 far-UV CD spectra of insulin-loaded lyophilized micelles (polymer:Ins) and insulin-loaded lyophilized micelles containing PBA (polymer:Ins:PBA) of SOL and F68 (A and C) and F108 and F127 (B and D) stored for 6 months at 20 ºC (A and B) and 4 ºC (C and D)…………………………………………………………………………………………….103 xxii Chapter 5 In vitro biological assessment of powder formulations for inhalation of insulin Figure 5.1 Reaction schematic for the conjugation of the polymers with 5-DTAF via nucleophilic aromatic substitution by an addition-elimination mechanism. At basic pH, the terminal hydroxyl group of PEG blocks presented in the polymers, attack the reactive moiety (2-amino-4,6-dichloro-s-triazine) on the 5-DTAF molecule, promoted by strong electron-withdrawing groups (N) of the s-triazine ring………………..……………………..116 Figure 5.2 Formulations’ toxicity profile regarding cell viability of Raw 246.7, Calu.3 and A549 cell lines. Results are expressed as mean ± SEM (n=5)……………….……………121 Figure 5.3 Formulations’ toxicity profile regarding membrane integrity of Raw 246.7, Calu.3 and A549 cell lines. Results are expressed as mean ± SEM (n=5)……..………...122 Figure 5.4 Permeability of insulin through A549 (A) and Calu-3 (C) cell monolayers, expressed as the percentage of insulin added to the apical chamber of Transwell® system; and transepithelial electrical resistance (TEER) values as percentage of the of the values prior to experiment during permeability studies across A549 (B) and Calu-3 (D) cell monolayers. Results are presented as mean values ± SD (n=3)…………………………..124 Figure 5.5 Confocal microscopy micrographs of SOL (A), F68 (B), F108 (C) and F127 (D) micelle’s internalization by PMA-stimulated THP-1 and U937 macrophages. Each image provides a xy plane through a cell layer, and the cross-sectional view of the same section of the cell layer in the x–y and y–z orientation. Blue, green, and red fluorescence are from DAPI (nucleus), 5-DTAF-polymer (micelles) and CellMask® Deep Red (membrane), respectively………………………………………………………………………………………126 Figure 5.6 FACS quantification of micelles uptake by PMA-stimulated THP-1 and U937 macrophages. The values are expressed as the percentage of cells emitting green fluorescence after 4h incubation with micelles at a concentration of 1 mg/mL…………...127 Figure 5.7 FACS quantification of micelles uptake by PMA-stimulated THP-1 and U937 macrophages. The values are expressed as the percentage of cells emitting green fluorescence after 4h incubation with micelles at a concentration of 2 mg/mL…………...128 xxix Abbreviations 5-DTAF – 5-([4,6-dichlorotriazin-2-yl]amino)fluorescein hydrochloride AAC – Area above the curve AE – Association efficiency AFM – Atomic force microscopy AmB – Amphotericin B AO – Area of overlap AUC – Area under the curve BALF – Bronchoalveolar lavage fluid BALT – Bronchus-associated lymphoid tissue BCA – Biocinchoninic acid BSA – Bovine serum albumin bw – Body weight CC50 - half maximal cytotoxic concentration CINC-3 – Cytokine-induced neutrophil chemoattractant 3 Cmax – Maximum concentration observed CMC – Critical micelle concentration CMT – Critical micellization temperature COPD – Chronic obstructive pulmonary disease CsA – Cyclosporin A CSO-SA – Chitosan oligosaccharide-stearic acid dae – Aerodynamic diameter DAPI – 4′,6-diamidino-2-phenylindole DDS – Drug delivery system DLS – Dynamic light scattering DMEM – Dulbecco’s modified eagle medium DMSO – Dimethyl sulfoxide DPI – Dry powder inhaler DSC – Differential scanning calorimetry DSPE-PEG – 1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-N-methoxy(poly(ethylene glycol)) DSPE-PEG-PHEA – 1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-Nmethoxy(poly(ethylene glycol))-α,β-poly(N-2-hydroxyethyl)-DL-aspartamide xxx EDTA – Ethylenediaminetetraacetic acid ELISA – Enzyme-linked immunosorbent assay EMA – European Medicines Agency EPR effect – Enhanced permeability and retention effect F108 – Pluronic® F108 (PEG-PPO-PEG) F127 – Pluronic® F127 (PEG-PPO-PEG) F68 – Pluronic® F68 (PEG-PPO-PEG) FACS – Fluorescence-activated cell sorting FAE – Follicle associated epithelium far-UV CD – far-ultraviolet circular dichroism FBS – Fetal bovine serum Fc – Fragment crystallizable FDA – US Food and Drug Administration FELASA – Federation of Laboratory Animal Science Associations FE-SEM – Field emission scanning electron microscopy FPF – Fine particle fraction FTIR – Fourier transform infrared spectroscopy GLP-1 – Glucagon-like peptide 1 GnRH – Gonadotropin-releasing hormone GRAS – Generally recognized as safe GSD – Geometrical standard deviation H&E – Hematoxylin and eosin H40-PCL-PEG – Hyperbranched aliphatic polyester Boltorn H40-poly(ε-caprolactone)- poly(ethylene glycol) HA-C18 – Hyaluronic acid-g-octadecyl HbA1c – Glycated hemoglobin HIV-TAT – Human immunodeficiency virus-transactivator of transcription HLB – Hydrophilic-lipophilic balance HPAE-co-PLA/DPPE – Poly[(amine-ester)-co-(D,L-lactide)]/1,2-dipalmitoyl-sn-glycero-3phosphoethanolamine HPESO – Hydrolyzed polymers of epoxidized soybean oil HPLC – High-performance liquid chromatography HPSO – Hydrolyzed polymers of soybean oil IAA – Insulin autoantibodies xxxi IC50 – Half maximal inhibitory concentration ICH – International Conference on Harmonization IL-1 – Interleukin 1 IL-2 – Interleukin 2 IL-4 – Interleukin 4 IL-6 – Interleukin 6 IL-13 – Interleukin 13 INF-α – Interferon-α INF-γ – Interferon-γ LC – Loading capacity LD50 – Median lethal dose LDH – Lactate dehydrogenase LEBP – Lung epithelial binding peptides LHRH – Luteinizing-hormone-releasing hormone MALT – Mucosa-associated lymphoid tissue MBCP-2 – Pluronic P104-b-di(ethylene glycol) divinyl ether MIC – Minimal inhibitory concentration MMAD – Mass median aerodynamic diameter mPEG-b-PVL – Methoxy poly(ethylene glycol)-b-poly(valerolactone) mPEG–DSPE – Methoxy poly(ethylene oxide)-b-distearoyl phosphatidyl-ethanolamine MRP – Multidrug resistance–associated protein MRW – Mean residual weight MTT – 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide MW – Molecular weight NALT – Nasal-associated lymphoid tissue nanoDDS – Nanothechnology-based drug delivery system NLC – Nanostructured lipid carriers P(MAA-g-EG) – Poly(methacrylic acid-grafted-poly(ethylene glycol)) PA – Pharmacological availability PAGE-b-PLA – Poly(allyl glycidyl ether)-b-polylactide Papp – Apparent permeability coefficient PBA – Phenylboronic acid PBCA – Poly(n-butyl cyanoacrylate) PBS – Phosphate buffer saline pH 7.4 xxxii PC – Phosphatidylcholine PCL – Poly(ε-caprolactone) PCL-b-COS-b-PEG – Poly(epsilon-caprolactone)-b-chitooligosaccharide-b-poly(ethylene glycol) PCL-PEG-PCL – Poly(ε-caprolactone)-b-poly(ethylene glycol)-b-poly(ε-caprolactone) PDE – Permitted daily exposure PDEAEMA-PAEMA – Poly(diethylaminoethyl methacrylate)-poly(aminoethyl methacrylate) PdI – Polydispersity index PDT – Photodynamic therapy PE – Phosphatidylethanolamine PEG – Poly(ethylene glycol) PEG-b-PAA – Poly(ethylene glycol)-b-polyacrylic acid PEG-b-(PLL-IM) – Iminothiolane-modified poly(ethylene glycol)-b-poly(L-lysine) PEG-b-PBC – Poly(ethylene glycol)-b-poly(α-benzyl carboxylateε-caprolactone) PEG-b-PCL – Poly(ethylene glycol)-b-poly(ε-caprolactone) PEG-b-PHOHH – Poly(ethylene glycol)-b-poly(3-hydroxyoctanoate-co-3-hydroxyhexanoate) PEG-chitosan – Poly(ethylene glycol)-chitosan PEG-DACH-platin – Poly(ethylene glycol)-dichloro(1,2-diaminocyclohexane)platinum(II) PEG-g-PAE – Poly(ethylene glycol)-g-poly(b-amino ester) PEG-MOG – Poly(ethylene glycol)-monooleylglyceride PEG-PAsp – Poly(ethylene glycol)-poly(aspartic acid) PEG-PBCA – Poly(ethylene glycol)-poly(n-butylcyano acrylate) PEG-PDLA – Poly(ethylene glycol)–poly(d-lactide) PEG-PEI – Poly(ethylene glycol)-poly(ethylene imine) PEG-PGlu – Poly(ethylene glycol)-poly(L-glutamic acid) PEG-PHis – Poly(ethylene glycol)-poly(L-histidine) PEG-PLA – Poly(ethylene glycol)–polylactic acid PEG-PLLA – Poly(ethylene glycol)–poly(l-lactide) PEG-PPO-PEG – Poly(ethylene glycol)-b-polypropylene oxide-b-poly(ethylene glycol) (also known as Pluronic®) PEG-PPS – Poly(ethylene glycol)-b-poly(propylene sulfide) PEG-PPS-PEG – Poly(ethylene glycol)-b-poly(propylene sulfide)-b-poly(ethylene glycol) PEI – Polyethylenimine PER – Permeability enhancement ratio xxxiii PGA-co-PDL – Poly(glycerol adipate-co-ω-pentadecalactone) PHEA – α,β-poly(N-2-hydroxyethyl)-DL-aspartamide PHEA-g-PDTC – Poly-a,b-[N-(2-hydroxyethyl)-L-aspartamide]-g-poly(2,2dimethyltrimethylene carbonate) PHOHH – Poly(3-hydroxyoctanoate-co-3-hydroxyhexanoate) PLA – Polylactide or polylactic acid PLA-b-PEG-b-PHis – Poly(L-lactic acid)-b-poly(ethylene glycol)-b-poly(L-histidine) PLA-chitosan – Polylactide-chitosan PLGA – Poly(D,L-lactide-co-glycolic acid) PLGA-chitosan – Poly(D,L-lactide-co-glycolide)-chitosan PLGA-PEG – Poly(D,L-lactide-co-glycolide)-b-poly(ethylene glycol) PLGA-PEG-PLGA – Poly(D,L-lactide-co-glycolide)-b-poly(ethylene glycol)-b-poly(D,L-lactideco-glycolide) PLLF-g-(PLF-b-PLG) – Poly(l-lysine-co-l-phenylalanine)-g-poly(l-phenylalanine)-b-poly(lglutamic acid) PMA – Phorbol 12-myristate 13-acetate pMDI – Pressurized metered-dose inhaler PPEGMEA-g-PMOMMA – Poly[poly(ethylene glycol) methyl ether acrylate]-gpoly(methacrylate acid) PPO – Polypropylene oxide PVA – Polyvinyl alcohol PVA-acyl chains – Polyvinyl alcohol-modified with acyl chains PVP – Polyvinylpyrrolidone PVP-b-PDLLA – Poly(N-vinyl-2-pyrrolidone)-b-poly(D,L-lactide) RAFT – Reversible addition-fragmentation chain transfer RES – Reticuloendothelial system RGD – Arginine-Glycine-Aspartic acid rhDNase – Recombinant human desoxyribonuclease I ROP – Ring-opening polymerization SA-BPEI – Stearic acid-branched polyethyleneimine SAGly-DA – Poly[(sodium N-acryloyl-Lglycinate)-co-(N-dodecylacrylamide)] SALeu-DA – Poly[(sodium N-acryloyl-Lleucinate)-co-(N-dodecylacrylamide)] SAPhe-DA – Poly[(sodium N-acryloyl-Lphenylalaninate)-co-(N-dodecylacrylamide)] SAVal-DA – Poly[(sodium N-acryloyl-L-valinate)-co-(N-dodecylacrylamide)] xxxiv SAVal-OA – Poly[(sodium N-acryloyl-L-valinate)-co-(N-octylacrylamide)] SCC – Spectral correlation coefficient SEDDS – Self-emulsifying drug delivery systems SEM – Scanning electron microscopy SIMS – Secondary ion mass spectroscopy SIS – Styrene-isoprene-styrene SLN – Solid lipid nanoparticles SOD – Superoxide dismutase SOL – Soluplus® – Polyvinyl caprolactam-polyvinyl acetate-poly(ethylene glycol) graft copolymer SP-A – Surfactant protein A SP-D – Surfactant protein D t1/2 – Half-life time TEER – Transepithelial electrical resistance TEM – Transmission electron microscopy TFA – Trifluoroacetic acid THALWHT – Threonine-Histidine-Alanine-Leucine-Tryptophan-Histidine-Threonine Tmax – Time of maximum concentration observed TNF-α – Tumor necrosis factor alpha TPGS – D-alpha-tocopheryl-co-PEG 1000 succinate VEGF – Vascular endothelial growth factor XPS – X-ray photoelectron spectroscopy XRD – X-ray diffraction ZO-1 – Zonula occludens-1 Chapter 1 I State-of-art ___________________________________________________________________________________ 1 Chapter 1 State-of-art The information presented in this chapter was partially published in the following publications: Fernanda Andrade, Mafalda Videira, Domingos Ferreira, and Bruno Sarmento, Nanocarriers for pulmonary administration of peptides and therapeutic proteins, Nanomedicine (Lond), 6(1):123-41, 2011. Fernanda Andrade, Mafalda Videira, Domingos Ferreira, and Bruno Sarmento, Micelle-based systems for drug pulmonary delivery and targeting, Drug Delivery Letters, 1 (2):171-185, 2011. Fernanda Andrade, Diana Rafael, Mafalda Videira, Domingos Ferreira, Alejandro Sosnik, and Bruno Sarmento, Nanotechnology and pulmonary delivery to overcome resistance in infectious diseases, Advanced Drug Delivery Reviews, 65 (13–14):1816–1827, 2013. Fernanda Andrade, Catarina Moura, Bruno Sarmento, Pulmonary Delivery of Biopharmaceuticals in Mucosal Delivery of Biopharmaceuticals: Biology, Challenges and Strategies, José das Neves and Bruno Sarmento (Eds), Springer, 2014, ISBN 978-1-46149524-6. Diana Rafael, Mafalda Videira, Mireia Oliva, Domingos Ferreira, Bruno Sarmento, Fernanda Andrade, Amphiphilic Polymers in Drug Delivery in Encyclopedia of Biomedical Polymers and Polymeric Biomaterials, Munmaya Mishra (Ed.), CRC Press, 2015, ISBN 9781439898796. Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 2 1. Drug delivery systems: innovation and technology The advent of pharmaceutical industry brought the nececity to control the biodistribution of drugs, aiming to enhance their therapeutic efficacy. The concept of drug delivery system (DDS) that control the release of the drugs and target them to specific locations in the body represents a major clinical breakthrough. This concept is in close agreement with those predicted by Paul Ehrlich in the early 20th century; however, we still cannot achieve the desired 'magic bullet' (1, 2). The pharmacological properties, clinical use, marketability, and competitiveness of drugs are highly dependent on the nature and properties of the DDS used. Thus, pharmaceutical companies are continuously seeking for new and improved DDS to deliver both new and existing drugs, focusing on its effectiveness, safety and market value. Since the success of a DDS relies on several aspects related to the route of administration, specific drug properties or disease physiopathology, distinct strategies must be applied during its rational development according to the desired application. Ideally, a DDS should possess characteristics such as (i) appropriate circulation time in the body to promote a therapeutic or diagnostic action, (ii) protect the drug from degradation and from premature clearance, (iii) organ/tissue selectivity, (iv) therapeutic concentration of the drug at the target anatomical site, (v) release the compound in response to specific stimuli, and (vi) improve the therapeutic index of the drug (2-4). Although some studies already refer the development of multifunctional systems (Figure 1.1), the development of the ideal DDS is still in its infancy. Figure 1.1 Schematic representation of a multi-functional DDS. Chapter 1 I State-of-art ___________________________________________________________________________________ 3 Passive or active targeting of drugs to specific organs and tissues, enhancing its therapeutic efficacy and decreasing the side effects, can be achieved via different mechanisms. By increasing the systemic circulation time of drugs through a reduction of their uptake by the reticuloendothelial system (RES) (e.g. by conjugating drugs or coating particles with poly(ethylene glycol) (PEG), i.e. PEGylation), drugs are more likely to suffer an enhanced permeability and retention effect (EPR effect). Thus, there will be a passive targeting to tissues with increased vascular permeability such as solid tumors, being this mechanism extensively used by DDS of anticancer drugs. It can also occur at infection or inflammation sites. On the other hand, active targeting can be achieved using carriers with stimulisensitiveness once several pathological processes are characterized by changes in pH, temperature or redox potential. Thus, an active targeting can be achieved by using carriers that release drugs only after exposed to certain conditions (stimulus-sensitive). Other potential approach for active targeting might be achieved through the use of specific antibodies, molecules recognized by certain cell receptors, or receptors for molecules that are overexpressed in certain disease states. These include integrins and vascular endothelial growth factor (VEGF) presented in vascular cells of various solid tumors, as well as transferrin and folate residues, whose receptors are overexpressed on the surface of various tumor cells (4-8). The translation of this concept to pulmonary administration leads, for instance, to the identification and selection of lung epithelial binding peptides (LEBP), namely LEBP-1, LEBP2 and LEBP-3 as peptides that bind selectively to receptors of the alveolar epithelium cells, therefore promoting a specific alveolar targeting (9). LEBP-binding DNA complexes presented higher in vitro transfection efficiency to lung epithelial cells (L2 cell line) when compared to the same formulation without LEBP (10). In another study, Jost and co-workers identified a peptide with the amino acid sequence Threonine-Histidine-Alanine-LeucineTryptophan-Histidine-Threonine (THALWHT) that selectively binds to airway epithelial cell lines and can be used as targeting moiety for gene delivery (11). Several reports on moieties explored to achieve active targeting to lungs like surfactant protein A (SP-A) (12), transferrin (13), lectin (14), folate (15) or human immunodeficiency virus-transactivator of transcription (HIV-TAT) peptide (16) can be found on the literature. Mannose and its derivatives have been also proposed to target the mannose receptor present at the surface of alveolar macrophages and improve the treatment of intracellular pathogens like Mycobacterium tuberculosis (17, 18). Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 4 In the last decades, several studies have been conducted with the aim of developing innovative pharmaceutical forms, arising some of the most promising advances from the application of nanotechnology to the production of DDS (19-21). 1.1. Nanotechnology in the development of drug delivery systems The application of nanotechnology in medicine has been capturing growing interest over recent years, having emerged the concept of nanomedicine. This is explained by the nano and micrometer scale of cellular and subcellular structures (6). The goal of nanomedicine is to allow a more accurate and timely diagnosis and to provide the most effective treatment without side effects (22). Currently, the main application areas of nanomedicine are imaging and cancer therapy. However, studies in various areas such as peptides and proteins delivery, vaccination, gene therapy, tissue engineering or production of devices for the administration of drugs are also being carried out (6). Both pharmacokinetics and pharmacodynamics of a drug are highly dependent on its physical and chemical features, and are influenced by the type of formulation and dosage form used to deliver it. NanoDDS like nanoparticles, liposomes or micelles can modulate and improve the performance of many drugs to an extent not achievable by conventional formulations. For example, nanoDDS can be capitalized to encapsulate drugs and thereby (i) increase their solubility, (ii) protect them from degradation, (iii) enhance their epithelial absorption, (iv) escape from the in vivo defensive systems, thus increasing their blood circulation time, (v) target the drugs to specific cells/tissues/organs, releasing them in a controlled manner as a response to a specific stimulus, or (vi) enhance their uptake by cells (19, 23). They also allow the reduction of the immunogenicity of proteins, thus decreasing the toxicity of the formulation (24). In addition, combined nanoDDS can simultaneously detect and treat a disease by encompassing both imaging and therapeutic compounds, an emerging field known as theranostics (25). In the near future, nanomedicine could play a key role to achieve the so desired personalized medicine. Chapter 1 I State-of-art ___________________________________________________________________________________ 11 Polymeric micelles have been developed to modify several major intrinsic characteristics of incorporated drugs, i.e. drug aqueous solubility, in vivo stability, release pattern, pharmacokinetics and biodistribution (5, 39). They allow the formulation and administration of highly hydrophobic drugs that would be withdrawn from development in the stage of drug formulation using conventional formulations (61). Despite being especially used to formulate hydrophobic drugs, like liposomes, polymeric micelles allow the encapsulation of drugs with different polarities. Hydrophobic drugs are incorporated into the micelle core being the solubilization capacity of drugs proportional to the hydrophobicity of the micelle core, while water-soluble drugs are adsorbed on the micelle shell and/or surface. Drugs with intermediate polarity are distributed along the amphiphilic molecules (62, 63). In addition to the higher stability compared to liposomes, micelles present low size and high encapsulation efficiency, which, together with the possibility of being sterilized by filtration, make these systems an interesting alternative to drug delivery (5, 62, 64). Figure 1.5 Schematic representation of a micelle. A few polymeric micelles-based formulations are currently in clinical trials (Table 1.2) (65, 66), one of them (Genexol-PM®, Samyang Co.) granted a pre-market authorization in Korea for the treatment of breast cancer and non-small cell lung cancer (67-69), while SP1049C (Surpatek Pharma, Inc.) granted orphan drug designation by US Food and Drug Administration (FDA) for the treatment of gastric cancer (70). Besides the formulations that are ongoing on clinical trials, polymeric micelles have successfully enhanced the therapeutic Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 12 index of various drugs from anticancer and anti-inflammatory drugs (71-75), to genetic material (76). They are also been proposed as delivery systems for diagnostic agents (77). Due to its small size, generally lower than 200 nm, and hydrophilic surface, micelles are poorly recognized by RES and present long circulation times in bloodstream suffering the EPR effect at solid tumor sites, reason why have been extensively exploited as drug delivery systems for anticancer agents (78-80). Promoting drug selective targeting to specific organs and tissues can be also achieved either using stimuli-responsive micelles (63, 79, 81, 82), or by modulating their surface with active-targeting ligands (79, 83). By sharing some structural and functional features with natural transport systems, e.g. virus and lipoproteins, polymeric micelles can be a useful strategy to solve the problem of drug resistance (84). Pluronic®- based micelles have shown to interfere with the activity of P-glycoprotein and multidrug resistance–associated protein (MRP), increasing the therapeutic effectiveness of anticancer agents in multidrug resistant cancer cell lines (56, 85-87). Table 1.2 Examples of polymeric micelles-based formulations that enrolled in clinical trials. Formulation Polymer Drug Indication Clinical phase Reference SP1049C Pluronic® L61 and F127 Doxorubicin Advanced adenocarcinoma of the esophagus, gastroesophageal junction and stomach II/III (70) Genexol-PM® PEG-PLA Paclitaxel Breast cancer, non-small cell lung cancer, advanced pancreatic cancer, ovarian cancer, and head and neck cancer I/II/III/IV (68, 88-90) NK012 PEG-PGlu-SN38 conjugated SN-38 Breast cancer, colorectal cancer and small cell lung cancer I/II (65, 91, 92) NK105 PEG-PAsp Paclitaxel Advanced or recurrent gastric cancer and breast cancer I/II/III (66, 93) NC-4016 PEG-DACHplatin Oxaliplatin Advanced solid tumors and lymphoma I (94) Chapter 1 I State-of-art ___________________________________________________________________________________ 13 Formulation Polymer Drug Indication Clinical phase Reference NC-6004 (Nanoplatin®) PEG-PGlucisplatin conjugated Cisplatin Solid tumors, breast cancer, pancreatic cancer I/II/III (95, 96) Paxceed® PEG-PLA Paclitaxel Rheumatoid arthritis, psoriasis II (97, 98) CRLX101 Polymercyclodextrincamptothecin conjugated Camptothe cin Advanced solid tumors, ovarian cancer I/II AquADEK® TPGS Vitamins and antioxidants Multivitamin supplement in cystic fibrosis I/II (99, 100) BIND-014 PEG-PLA and PLGA-PEG Docetaxel Non-small cell lung Cancer and prostate cancer I/II (101-103) PEG-DACH-platin – Poly(ethylene glycol)-dichloro(1,2-diaminocyclohexane)platinum(II); PEG-PAsp – Poly(ethylene glycol)-poly(aspartic acid); PEG-PGlu – Poly(ethylene glycol)-poly(L-glutamic acid); PEG-PLA – Poly(ethylene glycol)-polylactic acid; PLGA-PEG – Poly(D,L-lactide-co-glycolide)-bpoly(ethylene glycol); TPGS – D-alpha-tocopheryl-co-PEG 1000 succinate 1.2. The role of amphiphilic polymers in the development of drug delivery systems Amphiphilic copolymers are heterogeneous compounds composed by both hydrophilic and hydrophobic units disposed in sequential blocks (generally diand triblock-copolymers) or grafts. By varying either the type or the chain length of the units, it is possible to modulate the polymer properties (41, 104). It is their versatility that makes them suitable for industrial and pharmaceutical applications. Regarding the last one, they have been used for a long time in different pharmaceutical dosage forms as excipients like emulsifiers, wetting, thickening or gel forming agents, and stabilizing agents of suspensions and colloidal dispersions. Still, they gained an increased interest in the last decades in the development of new DDS driven by the progresses seen in the pharmaceutical sciences and nanomedicine field (41, 105). Amphiphilic polymers are used in the development of different types of DDS, such as tablets, capsules, gels, microparticles, with emphasis in nanoDDS, namely polymeric micelles. Table 1.3 presents examples of DDS in development using amphiphilic polymers. Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 14 Table 1.3 Examples of DDS in development using amphiphilic polymers. Type of DDS Polymer Drug Reference Micelles CSO-SA Doxorubicin, paclitaxel and AmB (106-108) DSPE-PEG Calcitonin (59) HA-C18 Paclitaxel (109) mPEG-b-PVL Camptothecin (110) PCL-PEG-PCL Rifampicin (111, 112) PEG-b-PAA Mitoxantrone and doxorubicin (113) PEG-b-(PLL-IM) siRNA (114) PEG-b-PBC Paclitaxel (115) PEG-b-PCL Paclitaxel, indomethacin, curcumin, plumbagin and etoposide (61, 115) PEG-Phis Doxorubicin (116, 117) PEG-chitosan Methotrexate (118) PEG-g-PAE Doxorubicin (119) PEG-PAsp Lysozyme, irinotecan (120, 121) PEG-PEI DNA (122) PEG-PLA Paclitaxel and CsA (53, 123, 124) PHEA-g-PDTC Prednisone and tegafur (125) PLA-b-PEG-b-Phis Doxorubicin (126) PLGA-PEG-PLGA Curcumin, DNA (127, 128) Pluronic® F68, F127, L61 and P85 DNA (76, 128130) Pluronic® P105 and P105/L101 mix and Pluronic® P105/PCL mix Paclitaxel (131, 132) (133) Poly(sodium N-acryloyl-Laminoacidate-coalkylacrylamide)s Griseofulvin and non-steroidal anti-inflammatory drugs (134-136) PVA-acyl chains Doxorubicin (137) Cylindrical brushes PLLF-g-(PLF-b-PLG) Doxorubicin (138) Chapter 1 I State-of-art ___________________________________________________________________________________ 15 Type of DDS Polymer Drug Reference Microparticles PEG-PLA BSA (139) Pluronic® F68 BSA (140) Nanoparticles H40-PCL-PEG 5-fluorouracil and Paclitaxel (141) HPAE-co-PLA/DPPE Paclitaxel (142) HPESO and Pluronic® F68 Doxorubicin and mitomycin C (143, 144) PDEAEMA-PAEMA siRNA and proteins (145) PEG-PBCA Docetaxel (146) PEG-PLA Paclitaxel, DNA (147, 148) Hydrogels MBCP-2 DNA (149) P(MAA-g-EG) Insulin, interferon β and calcitonin (150-152) PEG-chitosan BSA (153) PLGA-PEG-PLGA Dexamethasone and calcitonin (154, 155) Pluronic® F127 Deslorelin, GnRH, Vitamin B12 and naproxen (156-158) Patches SIS Methyl salicylate, capsaicin, and diphenhydramine hydrochloride (159) Capsules PEG-MOG Risperidone, ketoconazole, indomethacin, hydrocortisone and CsA (160) Solid solutions (Extrudates) Soluplus® Danazol, fenofibrate and itraconazole (161) Tablets Soluplus®, Pluronic® and Vitamin E-TPGS® Amine drugs like donepezil, olanzapine or tamsulosin (162) AmB - Amphotericin B; BSA – Bovine serum albumin; CsA – Cyclosporin A; CSO-SA – Chitosan oligosaccharide-stearic acid; DSPE-PEG – 1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-Nmethoxy(poly(ethylene glycol); GnRH – Gonadotropin-releasing hormone; H40-PCL-PEG – Hyperbranched aliphatic polyester Boltorn H40-poly(ε-caprolactone)-poly(ethylene glycol); HA-C18 – Hyaluronic acid-g-octadecyl; HPAE-co-PLA/DPPE – Poly[(amine-ester)-co-(D,L-lactide)]/1,2dipalmitoyl-sn-glycero-3-phosphoethanolamine; HPESO – Hydrolyzed polymers of epoxidized soybean oil; MBCP-2 – Pluronic P104-b-di(ethylene glycol) divinyl ether; mPEG-b-PVL – Methoxy poly(ethylene glycol)-b-poly(valerolactone); P(MAA-g-EG) – Poly(methacrylic acid-grafted-poly(ethylene glycol)); PCL – Poly(ε-caprolactone); PCL-PEG-PCL – Poly(ε-caprolactone)-b-poly(ethylene glycol)-b-poly(εcaprolactone); PDEAEMA-PAEMA – Poly(diethylaminoethyl methacrylate)-poly(aminoethyl methacrylate); PEG-b-(PLL-IM) – Iminothiolane-modified poly(ethylene glycol)-b-poly(L-lysine); PEG-b- Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 16 PAA – Poly(ethylene glycol)-b-polyacrylic acid; PEG-b-PBC – Poly(ethylene glycol)-b-poly(α-benzyl carboxylateε-caprolactone); PEG-b-PCL – Poly(ethylene glycol)-b-poly(ε-caprolactone); PEGchitosan - Poly(ethylene glycol)-chitosan; PEG-g-PAE – Poly(ethylene glycol)-g-poly(b-amino ester); PEG-MOG – Poly(ethylene glycol)-monooleylglyceride; PEG-PAsp – Poly(ethylene glycol)- poly(aspartic acid); PEG-PBCA – Poly(ethylene glycol)-poly(n-butylcyano acrylate); PEG-PEI – Poly(ethylene glycol)-poly(ethylene imine); PEG-PHis – Poly(ethylene glycol)-poly(L-histidine); PEGPLA – Poly(ethylene glycol)–polylactic acid; PHEA-g-PDTC – Poly-a,b-[N-(2-hydroxyethyl)-Laspartamide]-g-poly(2,2-dimethyltrimethylene carbonate); PLA-b-PEG-b-PHis – Poly(L-lactic acid)-bpoly(ethylene glycol)-b-poly(L-histidine); PLGA-PEG-PLGA – Poly(D,L-lactide-co-glycolide)-bpoly(ethylene glycol)-b-poly(D,L-lactide-co-glycolide); PLLF-g-(PLF-b-PLG) – Poly(l-lysine-co-lphenylalanine)-g-poly(l-phenylalanine)-b-poly(l-glutamic acid); PVA-acyl chains – Polyvinyl alcoholmodified with acyl chains; SIS – Styrene-isoprene-styrene. Among the different structures available, the most extensively explored for the production of DDS are composed of PEG as hydrophilic block and (i) polypropylene oxide (PPO); (ii) poly(ester)s like PCL or PLGA; (iii) poly(amino acid)s such as poly(L-aspartic acid) and poly(L-glutamic acid); or (iv) lipids like phosphatidylethanolamine (PE) as hydrophobic block (5). Particular interest has been given to poly(ethylene glycol)-b-polypropylene oxide-bpoly(ethylene glycol) (PEG-PPO-PEG) block copolymers (poloxamers and poloxamines) (163, 164), mainly due to their commercial accessibility in a wide range of compositions and molecular weight (MW) (Pluronic®/Lutrol®/Kolliphor P® and Tetronic®). 1.2.1. Synthesis of copolymers Due to environmental concerns, the synthesis of these polymers has evolved in the last decades, in order to achieve a cleaner production, in the scope of the ''green-chemistry''. Among the different methods, reversible addition-fragmentation chain transfer (RAFT) polymerization and ring-opening polymerization (ROP) of lactones, lactides and cyclic anhydrides are extensively used, mainly due to its ability to prepare both homoand copolymers of different MW and architectures with well-defined structures or end-groups (165-167). Enzyme-catalyzed ROP has also been used to eliminate the use of organometallic catalysts, but only low MW polymers can be obtained using this method (168, 169). The ROP can be performed either as a bulk polymerization, in solution, emulsion or dispersion (166), while RAFT is commonly performed in emulsions (167). Copolymers can be produced by the sequential addition of monomers or by the conjugation of the preformed homopolymers, e.g. by reactive extrusion using transesterifcation at high temperature (166). For example, Chapter 1 I State-of-art ___________________________________________________________________________________ 17 poloxamers are synthesized using anionic polymerization in the presence of an alkaline catalyst, generally sodium or potassium hydroxide. The PPO central unit is formed by the polymerization of propylene oxide monomers followed by the addition of ethylene glycol to the PPO end-groups (105). 1.2.2. Characteristics of copolymers and copolymer-based structures 1.2.2.1. Stimuli-responsiveness These polymers present some characteristics that make them suitable for human administration, namely their water-solubility, biodegradability, biocompatibility and low immunogenicity. Among all characteristics presented by some amphiphilic polymers for the development of advanced controlled nanoDDS, one of the most interesting is the stimuliresponsiveness (170, 171). Stimuli-responsive polymers are a class of “smart” polymers that undergo physical or chemical changes as response to a specific stimulus, e.g. temperature, pH, redox potential, magnetic field or light (172, 173). Monomers of N-alkyl substituted acrylamides and acrylate/methacrylate derivatives are commonly used in the development of thermo-responsive and pH-responsive copolymers, respectively (171, 174). pH-responsive hydrogels were able to protect proteins from the harsh gastric environment, promoting their release only at distal parts of intestine. At acidic pH the existence of intermolecular polymer complexes lead to the formation of compact gels that swells just at basic pH (150, 152). Due to their properties, the hydrogels enhanced the in vivo intestinal absorption and hypoglycemic effects of oral insulin (152). 1.2.2.2. Self-assembly: the crucial phenomenon Amphiphilic copolymers are able to form nanoscopic structures with different morphologies, e.g. polymersomes, nanocapsules, nanospheres, nanogels or dendrimers, although polymeric micelles are the most commonly used and studied (105, 175, 176). Polymeric micelles are supramolecular structures formed by self-assembly of amphiphilic copolymers into spherical nanosized particles with a hydrophilic corona and hydrophobic core. The selfassembly or micellization occurs at or above a threshold level of concentration (CMC) and temperature (CMT), which are specific for the polymer (Figure 1.6) (5). Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 18 Figure 1.6 Schematic representation of micellization. In water, this process is driven by an increase in entropy of the solvent molecules in contact to the hydrophobic units and a consequent decrease of free energy in the system as the hydrophobic components are withdrawn from the aqueous media to form the micelle core (39, 177). Two forces are involved in the micelle formation, an attractive force that leads to the association of molecules and a repulsive force that prevents unlimited growth of the micelles (178). The free energy of the micellization process, Δ°Gm, is given by the follow Equation 1.1. ∆°𝐺𝑚=𝑅𝑇ln𝐶𝑀𝐶 Equation 1.1 where R is the gas constant and T is the temperature of the system. MW, molecular architecture, temperature, solvent-polymer interactions or salt concentration are parameters that influence the self-assembly of polymers. Increasing temperature of the system, the solvency of hydrophilic unit as well as the CMC value will decrease, promoting the micellization. Similarly, this phenomena is favored when the attractive hydrophobic interactions increases as a result of a gain in the MW of the hydrophobic domain (179). For example, poly(ethylene glycol)-b-poly(3-hydroxyoctanoate-co-3-hydroxyhexanoate) (PEG-bPHOHH) copolymers present a CMC value of 5.50 and 0.93 mg/L for poly(3hydroxyoctanoate-co-3-hydroxyhexanoate) (PHOHH) segments with 1500 and 7700 g/mol, respectively (180). During the development of amphiphilic copolymers for DDS is imperative to determine the CMC value, which is generally estimated by a steady-state fluorescence method using pyrene as a probe. In Table 1.4 are presented estimated CMC values of some amphiphilic copolymers. Chapter 1 I State-of-art ___________________________________________________________________________________ 19 Table 1.4 Estimated critical micelle concentration (CMC) values for some amphiphilic copolymers. CopolymerMw (g/mol) or mol% CMC (mg/L or mM) Reference CSO-SA 140 (106) HA-C18* 10-37.3 (109) HPESO4866 0.075-0.080 (181) HPSO3800 0.055 (181) mPEG-b-PVL2000-10000 0.01-0.1 (110) PAGE-b-PLA 60-160 (182) PCL1050-7850-PEG6000-20000PCL1050-7850 2.2.10-3-39.10-3 (111) PEG5000-b-PHOHH1500-7700 5.50-0.93 (180) PEG-MOG5-10 300-4000 (160) PEG-PPS 0.0076-0.027 (183) PEG-PPS-PEG 0.0015-0.015 (183) PLGA-PEG-PLGA 5 (128) Pluronic® F108 3.08 (184) Pluronic® F127 0.56 (184) Pluronic® P103 0.14 (184) Pluronic® P105 0.46 (184) Pluronic® P123 0.05 (184) PPEGMEA-g-PMOMMA 1.63 (185) PVP-b-PDLLA27 or 38 4.3 and 2.6 (186) SAGly-DA16 2.9 (134) SALeu-DA16 0.4 (134) SAPhe-DA16 1.5 (134) SAVal-DA9 or 16 0.9 and 4.5 (187) SAVal-OA16 22 (136) CSO-SA – Chitosan oligosaccharide-stearic acid; HA-C18 – Hyaluronic acid-g-octadecyl (* Octadecyl moiety with various substitution degrees); HPESO – Hydrolyzed polymers of epoxidized soybean oil; HPSO – Hydrolyzed polymers of soybean oil; mPEG-b-PVL – Methoxy poly(ethylene glycol)-bpoly(valerolactone); PAGE-b-PLA – Poly(allyl glycidyl ether)-b-polylactide; PCL-PEG-PCL – Poly(εcaprolactone)-b-poly(ethylene glycol)-b-poly(ε-caprolactone); PEG-b-PHOHH – Poly(ethylene glycol)- b-poly(3-hydroxyoctanoate-co-3-hydroxyhexanoate); PEG-MOG – Poly(ethylene glycol)- monooleylglyceride; PEG-PPS – Poly(ethylene glycol)-b-poly(propylene sulfide); PEG-PPS-PEG – Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 20 Poly(ethylene glycol)-b-poly(propylene sulfide)-b-poly(ethylene glycol); PLGA-PEG-PLGA – Poly(D,Llactide-co-glycolide)-b-poly(ethylene glycol)-b-poly(D,L-lactide-co-glycolide); PPEGMEA-g-PMOMMA – Poly[poly(ethylene glycol) methyl ether acrylate]-g-poly(methacrylate acid); PVP-b-PDLLA – Poly(Nvinyl-2-pyrrolidone)-b-poly(D,L-lactide); SAGly-DA – Poly[(sodium N-acryloyl-Lglycinate)-co-(Ndodecylacrylamide)]; SALeu-DA – Poly[(sodium N-acryloyl-Lleucinate)-co-(N-dodecylacrylamide)]; SAPhe-DA – Poly[(sodium N-acryloyl-Lphenylalaninate)-co-(N-dodecylacrylamide)]; SAVal-DA – Poly[(sodium N-acryloyl-L-valinate)-co-(N-dodecylacrylamide)]; SAVal-OA – Poly[(sodium N-acryloyl-Lvalinate)-co-(N-octylacrylamide)]. 1.2.2.3. Hydrophilic surface: stability and functional role While the hydrophobic core of micelles, in addition to the solubilization and protection of drugs, provides appropriate mechanical properties for the desired application, hydrophilic shell masks the particle from the biological environment, reducing the protein absorption and cellular adhesion, thereby enhancing the particle stability (188, 189). Opsonin proteins present in the blood serum promptly bind to particles, allowing their recognition by macrophages that will quickly remove the encapsulated drugs from bloodstream (190). The presence of hydrophilic layer on the surface of particles (stealth particles) will reduce or delay the opsonization via steric repulsion forces, thus increasing the plasma circulation time of particles and the half-life time (t1/2) of drugs (190, 191). PEG and poloxamers are generally used as hydrophilic polymers to cover the surface of many nanoDDS (146). Although PEGylation presents advantages, may not always be necessary. Indeed, unnecessary or excessive PEGylation could lead to a non-desirable increase in the particle size. Additionally, an increase in the cost of the final product could also be noticed. Thus, the quantification of hydrophilic groups at the surface of particles must be performed using techniques like XPS or secondary ion mass spectroscopy (SIMS) (192, 193) in order to assess the convenience of PEGylation (or other method to increase hydrophilicity) and to predict the nanoDDS behavior in vivo. From the technological point of view, the importance of the hydrophilic surface on nanoDDS relies not only in the stealth properties but also in their functionalization potential by chemical modification or bioconjugation. The functionalization is mainly used to bind targeting moieties to the particles surface that allows the biodistribution control of nanoDDS (194) and different ligands can be used accordingly to the required application (195), as referred before. The conjugation of folate at the surface of hyperbranched aliphatic polyester Boltorn H40-poly(εcaprolactone)-poly(ethylene glycol) (H40-PCL-PEG) nanoparticles enhances the in vitro drug Chapter 1 I State-of-art ___________________________________________________________________________________ 27 The epithelium of the respiratory system has also lymphoid tissue called mucosa-associated lymphoid tissue (MALT) that is responsible for its immunological activity. The lymphoid follicles present in the airways (nasal-associated lymphoid tissue (NALT) and bronchusassociated lymphoid tissue (BALT)) have many immune cells such as dendritic cells and T and B lymphocytes. The epithelium covering the lymphoid follicles called follicle associated epithelium (FAE) possesses M cells that are involved in uptake, transport and presentation of antigens in the respiratory lumen (216). Figure 1.8 Schematic representation of the alveolar epithelium. Taking into account the physiological characteristics of the respiratory system, it becomes clear that this route provides a non-invasive alternative presenting a large surface area, a thin epithelial barrier, extensive blood supply (flow 5 L/min), and lower enzymatic activity and efflux systems compared to other organs and tissues (e.g., gastrointestinal tract). Moreover, the reduced volume of fluid allows high concentrations of drug near the bloodstream, and the first-pass metabolism is avoided by pulmonary administration, which is especially useful for drugs that suffer high hepatic metabolism (251). These features are the reason for the higher bioavailability of inhaled peptides and proteins (10-200 times higher) compared with other non-invasive routes (84, 252). 3.3. Pulmonary biodistribution of inhaled peptides and proteins After inhalation, particles will undergo lung deposition and be subjected to the existing clearance mechanisms of the respiratory system. Being the place of gas exchange and constant contact with the exterior ambient, respiratory system developed complex and not Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 28 fully clarified defense mechanisms working as barrier for foreign particles that could impair the efficient delivery of drugs (248). The complex geometry and humidity of the airways hampers the passage of the larger particles to the deep lung, and the movement of the bronchial cilia, known as mucociliary escalator, transports the particles trapped in the mucus layer to the gastrointestinal tract. Particles/compounds capable of evade the mentioned barriers and reach the deep lung have to face other defense mechanisms like phagocytosis by alveolar macrophages, alveolar lining fluid, intraand extra-cellular catabolism, and the epithelium to attain the bloodstream (253, 254). Phagocytized compounds can be transported through the alveolar surface, undergo translocation to the lymphatic system or degradation by the intracellular enzymatic lysosomal system (69). In some studies, the rate of protein clearance did not significantly change in the presence or absence of an endotracheal tube, suggesting a reduced role of mucociliary escalator in the lung’s protein clearance (255). Also, demonstrate that only small amounts of proteins are found in macrophages (255). Despite the higher phagocytic capacity of macrophages compared to the endocytic capacity of pneumocytes, the first ones do not play an important role in clearance of proteins before 48 hours after exposure (254, 255). In addition, several large-sized proteins reach the circulation in intact form after instillation, suggesting a lower impact of catabolism in lung protein clearance (248, 254). However, degradation by proteolysis is relevant for proteins with small MW (< 3 kDa) (256). The use of enzyme inhibitors such as bacitracin, chymostatin, leupeptin or nafmostato mesylate reduces proteolysis and, thereby, increase the bioavailability of proteins prone to suffer high catabolism (8). In addition to the mechanisms/barriers described above and able to influence the permeability of compounds from the respiratory tract to the bloodstream and their bioavailability, the alveolar and airway epithelium arise as a major barrier to absorption of drugs. The absorption of macromolecules through the respiratory tract is a complex and enigmatic process that involves various mechanisms that are not yet well characterized, being apparently dependent on the hydrophilicity and the size of macromolecules (257-259). Different studies suggest that the rate of absorption is inversely proportional to the MW of the macromolecule. This influence not only the percentage of drug absorbed, but also the time necessary to the absorption occurs. For example, the t1/2 of the alveolar absorption of macromolecules increases with their MW (inulin with MW: 5250 Da and t1/2: 225 min; dextran with MW: 20000 Da and t1/2: 688 min; dextran with MW: 75000 Da and t1/2: 1670 min) (260). Chapter 1 I State-of-art ___________________________________________________________________________________ 29 Two major mechanisms were proposed to characterize pulmonary absorption of proteins: paracellular diffusion and transcytosis (Figure 1.9). Transcytosis may further be classified into vesicular endocytosis or pinocytosis, and receptor dependent transcytosis. Proteins with small MW are apparently absorbed by the paracellular route, diffusing through the tight junctions, while molecules with higher MW seem to suffer endocytosis (248). Peptides can be absorbed by receptor mediated transcytosis using the high-affinity peptide transporter 2 (247), while immunoglobulins are absorbed by a conjugation of pinocytosis with receptor mediated transcytosis. After suffer endocytosis, immunoglobulins bind with the fragment crystallizable (Fc) receptors that prevent the fusion with lysosomes and are release in the basolateral side of epithelial cells (261). This immunoglobulin transport pathway is used to deliver proteins by conjugation of the therapeutic macromolecule with Fc regions of immunoglobulins (261, 262). Other strategy proposed to enhance the pulmonary absorption of proteins is their coupling with specific peptidic sequences that not alter the biologic activity but promote their translocation through the epithelium probably by receptor mediated transport (263). Figure 1.9 Schematic representation of absorption routes. 3.4. Formulation requirements for pulmonary delivery of drugs Different aspects related to the formulation, inhalation device, and patient influence the aerosolization and deposition of drugs and, consequently, their therapeutic efficacy. The airways geometry, respiratory capacity (tidal volume, inspiratory flow rate and breathing Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 30 frequency), inhaler handling, smoking, and pathologies affecting the lungs will be responsible for therapeutic inter-individual variations. Formulation plays an important role in the inhaled drugs performance, in terms of stability, deposition and absorption. It should maintain the drug in the active state and deliver it to a specific site of action to be absorbed or released for systemic or local action, respectively. Additionally, the formulation must be stable upon storage. Since proteins are labile drugs, suitable to lose their activity through physical and chemical instability, maintenance of the active conformation is a challenge and a series of considerations should be taken into account during their production and storage. Temperature, pH, agitation, ionic strength or presence of surfactants needs to be controlled in order to avoid aggregation, degradation or conformation lost (264, 265). The stability of proteins and their therapeutic performance can be improved through the incorporation of some excipients to the formulation as detailed in the section 3.4.2. 3.4.1. Aerodynamic properties of particles Among the different formulation characteristics, aerodynamic diameter (dae) plays a key role in the deposition pattern and therapeutic efficiency of the aerosolized particles. Aerodynamic diameter is the diameter of a unit density sphere that has the same terminal settling velocity in still air as the particle in consideration (266) and is defined by the following Equation 1.2. d𝑎𝑒=d𝑒𝑞√ρ𝑝 ρ𝑜𝜒 Equation 1.2 where deq is the geometric diameter of an equivalent volume sphere of unit density, ρp and ρo are particle and unit densities, respectively, and χ is the dynamic shape factor. When determined based on the mass size of particles through methods like aerosolization using impactors, dae receives the denomination of mass median aerodynamic diameter (MMAD). Chapter 1 I State-of-art ___________________________________________________________________________________ 31 Figure 1.10 Deposition profile of particles on the different areas of the respiratory system according to their aerodynamic diameter. After inhalation, depending on dae or MMAD, particles will move through the airways and deposit in different parts of the respiratory track or be exhaled. For deposition at the lower regions of lungs, particles in the range of 1–100 nm and 0.5–5 µm are required. Particles larger than 5 µm will impact in the throat and be swallowed, while the middle sized particles will be essentially exhaled (Figure 1.10) (198, 267). Different forces, namely inertial impaction, sedimentation, diffusion and interception, will govern the particles fate and are related to the aerodynamic and hydrophilic properties of particles and shape of airways (253, 268, 269). By manipulating the particle size, is possible to target specific regions of the respiratory tract (more than 50% of deposition). For systemic delivery, alveolar deposition is needed, while for local action, delivery at bronchial level is preferred. As referred before, although alveolar macrophages are part of the respiratory defense system, they are sometimes the therapeutic target, for example in the treatment of tuberculosis. Targeting of alveolar macrophages could be achieved by surface-decoration with ligands of the lectin-like receptors present at the membrane of macrophages (17), or by deliver particles with a size that promote their phagocytosis (270, 271). Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 32 Formulations based on peptides and proteins loaded into nanoor microparticles have been widely proposed in the last years as strategies to overcome the limitations of conventional formulations. Since a wide range of the developed nanoparticles falls within the particle’s size range liable to suffer exhalation, the agglomeration of nanoparticles into micron-sized particles (nanocomposites) with proper aerodynamic characteristics that disaggregate after deposition have been exploited (34, 272). One possible advantage of use agglomerates of nanoparticles instead of microparticles relies on the capacity of nanoparticles to easily evade mucociliary clearance and phagocytosis by alveolar macrophages. Some studies show that smaller particles are internalized at a lower extent than particles higher in size (273, 274). 3.4.2. Excipients used in the development of inhalatory formulations Besides drugs, pharmaceutical excipients constitute an integral part of pharmaceutical formulations. They provide physical, chemical or microbiological stability, bulk properties that improve handling and metering, while controlling the mechanical and pharmaceutical properties of formulations such as release and permeation (267, 275). At the moment, only a small number of excipients are authorized for pulmonary delivery, but a variety of new excipients is under evaluation. Since lungs have limited buffer capacity, only compounds that are biocompatible or endogenous to the lung and that are easily metabolized or cleared can be used in inhaled formulations (275). Since formulations for nebulization are liquid solutions or suspensions, the common excipients used are salts (e.g., NaCl) to adjust the osmolarity (300 mosmol/L), HCl, NaOH, phosphates to adjust the pH to neutrality, and surfactants such as polysorbates, sorbitan monostearate, oleic acid, and soya lecithin to facilitate the formation of liquid droplets. Ethanol can be used as co-solvent and permeation enhancer only in small concentration due to its irritation potential. Preservatives such as parabens and benzalkonium chloride, antioxidants like ascorbic acid or chelating agents such as ethylenediaminetetraacetic acid (EDTA) can also be used to enhance stability (267). The excipients used in pMDI are similar to those found in preparations for nebulization excepting the gas propellants. The most widely excipient used as propellant in pMDI is hydrofluoroalkane, a non-toxic, non-flammable, and chemically stable gas without carcinogenic or mutagenic effects. Due to the absence of ozone-depleting properties, hydrofluoroalkane have been replacing chlorofluorocarbon-based propellants (276). DPI were Chapter 1 I State-of-art ___________________________________________________________________________________ 33 developed as a response to the limitations regarding stability and environmental aspects of pMDI and nebulizers (275). DPI are considered the most advantageous devices for inhalation regarding long-term stability of formulation, absence of gas propellants and patient convenience since are breathactuated and easy to use without the need of hand-lung inhalation co-ordination (277, 278). However, the development of particles with a narrow particle-size distribution and good flowability, suitable for aerosolization and lung deposition of peptides and proteins in the active state is challenging and depends on the appropriate use of powder technology and particle engineering. Techniques like microcrystallization, micronization by jetor ball-milling, lyophilization, spray-drying, spray-freeze-drying or supercritical fluid technology can be used to produce solid particles (267, 279). All the methods present advantages and disadvantages, and should be chosen according to the effect on the stability of the proteins, the characteristics of particles required to a specific formulation, scale-up, cost-effectiveness and safety issues (279). As stated before, the capacity to produce an aerosol with a narrow particle-size distribution will influence the deposition pattern of the drugs. Taking this in consideration, it is crucial to produce powders with good dispersibility. Solid particles are subject to cohesive and adhesive interactions with the surrounding environment, that need to be break during the aerosolization. Different forces are involved in particle’s interactions and include electrostatic and van der Waals forces, capillary forces from to the presence of residual water at the surface of particles, and mechanical interlocking due to surface roughness (275). Distinct aerosolization properties could be obtained playing with these forces by specific particle engineering. For example, an efficient drying of the particles needs to be provided by the production method to reduce moisture and capillary forces, but extra drying should be avoided due to the formation of charges at the surface of particles that promote electrostatic interactions. One of the main factors affecting the particle’s interactions is their surface area. The larger surface area, the greater will be the interactions between particles, and lower will be the flowability. Surface area is dependent on size, shape and morphology of particles (269, 275). Particles in the size range suitable for inhalation possess high surface areas and are generally mixed with larger coarse carrier particles of excipients to improve their flow properties. In DPI, the coarse carrier particle is the major component of the formulation (>95%, w/w). It provides bulk properties and reduces the cohesion forces between drug particles, facilitating the aerosol dispersion and defining deposition pattern (280). Lactose is the main excipient used as coarse and larger carrier particles and, to a lower extent, as cryoprotectant when particles are prepared using lyophilization (281). There Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 34 are commercially available a variety of inhalation grade lactose with different characteristics and narrow particle size distribution (Flowlac®, Granulac®, Respitose®, Lactohale®, Inhalac®, etc.) that should be carefully selected during the development of the formulation (282, 283). Other sugars such as glucose, trehalose and mannitol are also used as cryoprotectants and coarse carriers (267). Magnesium stearate is approved for inhalation to protect drug from moisture and to reduce the cohesion and adhesion between particles (284). The characteristics of the carriers and the adhesive forces between carrier and drug particles influence the performance of the formulation and need to be assessed and optimized. Blending of drugs with coarse carriers is a critical point during the development of a DPI and also object of optimization (285-289). Regarding inhaled nanoDDS, most of the components are generally not approved for inhalation. In this case, new excipients proposed for inhalatory formulations need to pass through the entire process of safety evaluation, including complete in vitro toxicological evaluation and in vivo assessment of non-clinical and clinical safety prior to licensing. Unfortunately, there is a lack of specific regulatory guidance regarding the toxicological assessment of excipients for inhalation (290-292). Excipients generally recognized as safe (GRAS) or those approved for other routes of administration need a more limited number of experiments for safety evaluation, being its acceptance by regulatory agencies usually easier. This is the case of phosphatidylcholine (PC), a lipid surfactant and one of the constituents of lung surfactant that is commonly used in the production of liposomes. Studies show that PCbased liposomes do not affect or slightly decrease the viability of human A549 alveolar cells after 24h of exposure. The effects on cell viability are dependent on the PC derivate and the concentration used (293). Other compounds used in the development of nanoDDS such as dextran, alginate, carrageenan or gelatin also possess GRAS status. It should be mentioned that those regulations apply not only to the material itself but also to its source. For example, contrary to what happen with shrimp-derived chitosan, chitosan obtained from Aspergillus niger is in the process to obtain GRAS status (294). Chitosan does not present significant toxicity to pulmonary tissue and cell lines after inhalation (295, 296). In fact, some studies showed some protective effect against oxidative stress (295). PLGA is another biodegradable polymer extensively used in the production of nanoDDS and present in various approved medicines including Trelstar® Depot (Pfizer), Risperidal® Consta (Johnson & Johnson), Sandostatin LAR® Depot (Novartis), Suprecur® MP (Aventis) or Lupron Depot® (TAP). The cytotoxicity of PLGA nanoparticles with different coatings and surface charges was assessed. Results showed that the cytotoxicity of the nanoparticles to human bronchial Calu-3 cells was Chapter 1 I State-of-art ___________________________________________________________________________________ 35 very limited, with the absence of inflammatory response (297). Cyclodextrins have been tested as complexing agent and excipient of inhalatory formulations. Various approved formulations containing cyclodextrins such as Voltaren® (Novartis), Clorocil® (Laboratório Edol), Brexin® (Chiesi Farmaceutici) or Vfend® (Pfizer), are daily used in the clinical practice for administration routes other than inhalation (298). Recently, carrier-free formulations that presented good aerosolization properties and deposition patterns have been developed and proposed as promising inhalatory formulations (299, 300). This “carrierless” strategy prevented the need of long and expensive safety studies, facilitating the authorization by regulatory agencies. Some engineered drug particles alone fulfill the requirements for inhalation, which is possible by the development of large porous/hollow particles (301-303). Due to its small density, particles with high geometric size and, consequently, reduced cohesive forces, present appropriate aerodynamic diameters. For example, salbutamol particles prepared by thermal ink-jet spray-freeze-drying with mean geometric diameter of 35 µm and mean aerodynamic diameter lower than 8.7 µm, present a percentage of FPF comparable to a salbutamol commercial formulation (302). At the moment, there are commercially available carrier-free DPI composed by agglomerates of pure terbutalin and budesonide particles, namely Bricanyl Turbohaler® (AstraZeneca) and Pulmicort Turbohaler ® (AstraZeneca), respectively. 3.4.3. Inhalation devices Apart from the characteristics of particles, pulmonary administration demands an inhaler device that produces an appropriate aerosol. There are numerous devices available with distinct properties and specifications that are more appropriate for each type of formulation – nebulizer for liquids, pMDI for liquids and powders, and DPI for powders (280, 304, 305). Since the device greatly influences the particle aerosolization and deposition pattern, the choice of the right inhaler for a specific formulation is one of the most time-consuming and challenging stages during the development of inhalable medicines, and it is imperative to ensure the appropriate drug efficacy (280, 306). For example, Alexander and co-workers tested the aerosolization efficacy of Ambisome® (Gilead Sciences) using different nebulizers and they found differences among all of them (307). The ideal inhaler should generate an aerosol with a FPF and reproducible drug dosage, guaranty protection and stability of the product during storage, be accurate, small, easy to handle, discrete, and user friendly in order to be accepted. One of the claimed reasons for the market failure of Exubera® (Pfizer) was Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 36 the low patient compliance partially due to the complex inhaler device (308). Owing to the great therapeutic potential of this alternative administration route, the last decades witnessed the development of devices with greater FPF and lung deposition patterns of approximately 40–50% of the nominal dose as compared with the low levels between 10 and 15% verified in the past (276). However, it is worth mentioning that this is a growing and dynamic field and innovative technologies are emerging. For example, power-assisted devices activated by vibration (309, 310) or pneumatic technology (311), called active inhalers, have been developed for the delivery of systemically proteins and active drugs that have narrow therapeutic windows as well as for inhalatory nanoDDS (275). Examples of such devices are the AERx® from Aradigm, Respimat® from Boehringer or AeroDose® from Aerogen Inc (8). Also, the development of adjusted inhalers that present specific inhalation patterns that are useful for the treatment of certain pulmonary diseases can be faced (276). It is also important to consider the cost relative to the production of the device and its impact on the final product price. A balanced price/therapeutic efficacy relationship is a crucial factor that can limit the clinical application of a certain product. 3.5. Limitations of pulmonary administration The main limitation faced by pulmonary administration of drugs relates to the reproducibility of the dose. From the delivery device until be absorbed in the lungs, the drug undergoes consecutive losses during aerosolization and deposition. Thus, the absorbed dose is usually lower than the dose present in the delivery device. As stated before, the deposition of particles at the lower respiratory tract is a complex phenomenon and its efficiency depends on several factors like the type of formulation, delivery device used and their capacity to produce the aerosol (250, 277, 312, 313). Unfortunately, there is no inhaler device producing only particles within the size limits appropriate to the lung deposition which results in a very low rate of dose emitted (277). Another issue of relevance arises from the fact that more than 50% of patients improperly uses the delivery device, leading to non-reproducibility (312). Thus, an intensive education of patients by health care professionals is required to an increase the effectiveness of the treatment (277). The respiratory capacity of patients also plays an important role on the delivery efficiency of particles, being reduced in patients with lower respiratory capacity like children, elderly Chapter 1 I State-of-art ___________________________________________________________________________________ 43 evaporation technique followed by lyophilization (374). The diameter of the aggregates of liposomes obtained lies within the limits compatible with alveolar deposition of particles from 3.5 to 4.3 µm possessing high encapsulation efficiency (66-72%). After intratracheal instillation in rats, liposomes promoted higher serum luteinizing hormone levels than the solution of leuprolide and an increased t1/2 when compared to solutions for pulmonary and subcutaneous administration. However, the bioavailability of liposomes was only 50% compared with subcutaneous injection and it is necessary to change the formulation in order to increase the bioavailability of leuprolide (374). Liposomal inhaled cyclosporin A (CsA) to treat rejection of lung transplants was also proposed.. Pulmonary administration to dogs resulted in higher concentration of CsA in the lungs instead of the liver, kidneys, spleen, heart and blood compartment. However, three hours after administration, there was some accumulation of CsA in the kidney and spleen, so the effects on the kidney should be studied (375). In another study, pulmonary delivery of liposome-encapsulated CsA promoted an accumulation of CsA in the lungs. The formulation was stable after nebulization and maintains the immunosuppressive activity of CsA after encapsulation (376). Gilbert and co-workers administered liposomal CsA by inhalation to healthy humans and found a preferential deposition of the particles in the alveolar region (70% of inhaled dose) (377). When the formulation was administered via a mouth-only face mask for 45 minutes, no changes were observed in pulmonary function. In contrast, the administration using a nebulizer mouthpiece has proved troublesome, leading to coughing and throat irritation and a slight decrease in lung function. This may be due to the fact of particles, when administered by a nebulizer mouthpiece suffer greater impaction in the throat. The observed differences do not appear to be due to the formulation, because in both cases were used the same formula, with the same particle size at the same dose (377). In an in vivo study conducted by Khanna and co-workers in healthy dogs, the pulmonary administration of interleukin 2 (IL-2)-loaded liposomes led to an activation of the immune response by a significant increase in leukocyte levels in the lung and serum mononuclear cells (378). This response was more efficient compared to inhaled free IL-2. The observed differences may be due to increased cellular uptake or decreased clearance of liposomal IL2. In vitro studies showed that activation of pulmonary leukocytes leads to inhibition of proliferation of tumor cells (378). Another study of the same researchers conducted in dogs with primary lung carcinoma or lung metastasis of other carcinomas, has shown that there is no significant toxicity associated with pulmonary administration of liposomal IL-2 (379). It was also found that liposomal Il-2 stimulates the immune system after inhalation. Moreover, in Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 44 some animals it was observed the complete regression of metastasis and the nonprogression of primary lung carcinoma in another animal. Most significantly results were observed in treating pulmonary metastasis from osteosarcoma (379). In a phase I clinical trial, the pulmonary delivery of liposomes containing IL-2 to patients with various carcinomas affecting the lungs (pulmonary sarcoma, renal cell carcinoma, melanoma and metastatic osteosarcoma) was well tolerated and there was no significant toxicity observed at doses that may possess therapeutic effect (380). However, in this study it was not determined the clinical efficacy of the formulation. In another phase I clinical trial was studied the utility of inhaled lipossomal IL-2 in patients with common variable immunodeficiency (381). IL-2 retained its biological activity after encapsulation. No changes were observed in pulmonary function or significant side effects during treatment. Although the patients treated with IL-2 liposome declare sense of improving their condition was not detected alterations of the immune response within the blood compartment. Such lack of response evidence may be due to low specificity of the markers used in the study (381). Superoxide dismutase (SOD) is an antioxidant enzyme, kidnapper of free radicals, ubiquitous in mammalian cells. It causes a decrease of reactive oxygen species responsible for oxidative stress, involved in phenomena such as carcinogenesis, inflammation and neurodegeneration (382). Studies demonstrate the success of the use of SOD in the treatment of rheumatoid arthritis and ischemia-reperfusion injury. However, after intravenous and oral administration, the SOD has a reduced circulation t1/2 and a high-level gastrointestinal degradation, respectively. With the aim of developing a non-invasive formulation that promotes an increased t1/2 of SOD, Kaipel and co-workers produced liposomes for pulmonary administration (382). In vivo studies conducted in pigs showed a prolonged release of SOD into the systemic circulation and an increase in its t1/2. There were no side effects such as irritation or inflammation in the lung, as well as changes in pH and plasmatic O2 and CO2 pressure (382). Different liposomal formulations were developed for aerosol delivery of a cationic α-helical peptide called CM3 with antimicrobial and antiendotoxin activity (383). The pulmonary delivery of CM3 allows the treatment of local infections and reduction of systemic effects. Of the several formulations developed, the best results in terms of the encapsulation and nebulization efficiencies and maintenance of liposomal integrity during nebulization were achieved with the combination of dimyristoyl phosphatidylcholine and dimyristoyl phosphatidylglycerol with a 3:1 molar ratio. With this formulation liposomes with diameter of 262 nm were obtained, presenting an encapsulation efficiency of 73%. Using a mathematical Chapter 1 I State-of-art ___________________________________________________________________________________ 45 model it was possible to predict the deposition profile and distribution in lungs of liposomes in adults and children of different ages. Data showed a pulmonary deposition in the all lungs, particularly in the tracheobronchial region. In this region, the minimum inhibitory levels of CM3 can be reached in the adult model, and can be exceeded in pediatric model subjects (383). SLN have been proposed for pulmonary administration of insulin. Examples are the nanoparticles of lecithin obtained by emulsification method with an average diameter of 300 nm, and alveolar deposition of 45% (w/w) of the dose delivered after lyophilization. It was confirmed the retention of the primary, secondary and tertiary structure of insulin after processing (322). In another study, Liu and co-workers produced SLN containing micelles inside by double emulsion with an average diameter 115 nm and an encapsulation efficiency of 98% (329). In vitro studies showed prolonged release of insulin. It was also demonstrated to retain the integrity of insulin after encapsulation and stability of the formulation after 6 months of preparation at 4 °C (329). 4.1.2. Polymeric nanoparticles A variety of polymeric nanocarriers have been explored for the pulmonary administration of proteins. A study compared six different types of nanocarriers composed by gelatin, chitosan, alginate, PLGA, poly(D,L-lactide-co-glycolic acid)-chitosan (PLGA-chitosan), and Poly(D,Llactide-co-glycolide)-b-poly(ethylene glycol) (PLGA-PEG), as DDS for inhalation of proteins using bovine serum albumin (BSA) and erythropoietin as model proteins (384). Excepting those of PLGA-PEG and alginate, particles presented a mean diameter lower than 300 nm. Gelatin and PLGA nanoparticles presented the best in vitro cytocompatibility and uptake by human type I alveolar epithelial cells. Additional in vivo studies in rats, showed that inhalation of these systems led to the release and retention of the proteins in lung tissue up to 10 days (384). Thus, these systems could be explored for the pulmonary administration of proteins with local therapeutic activity. In another study, powders for inhalation based on poly(glycerol adipate-co-ω-pentadecalactone) (PGA-co-PDL) nanoparticles co-spray with L-leucine were assessed (385). Powder presenting high FPF (> 75%) and a MMAD compatible to deep lung deposition (1.21 ± 0.67 μm) had shown to be compatible with pulmonary cell lines (A549 and 16HBE14o-). Also, the primary and secondary structure of the encapsulated BSA was maintained (385). The same research group developed a similar system based on poly(ethylene glycol)-co-poly(glycerol adipate-co-ω-pentadecalactone) (PEG-co-(PGA-co- Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 46 PDL)) for pulmonary administration of α-chymotrypsin and DNase I. The obtained powders maintain the activity of both enzymes and presented characteristics compatible with good lung deposition after inhalation (386). Inhalation of insulin encapsulated in polymeric nanoDDS has also been proposed. Huang and co-workers have developed nanoparticles of low MW chitosan using the emulsification/solvent evaporation technique (330). The resulting particles have a spherical shape with average diameter of approximately 400 nm, zeta potential of about +42 mV and encapsulation efficiency of 96%. The release profile of insulin was characterized by a burst effect followed by prolonged release for 24 hours. When administered to diabetic rats, the formulation has demonstrated a hypoglycemic effect similar to subcutaneous administration of insulin solution but prolonged in time (330). In a series of studies, Grenha and co-workers produced chitosan nanoparticles with and without lipid coating obtained by ionic gelation and then spray-dried to obtain nanocomposites. The nanoparticles obtained had an average diameter between 380 and 450 nm and encapsulation efficiency between 65 and 81%, while the spray-dried powder particles a dae of 2.5 – 2.8 µm (34, 272, 333). In vitro studies demonstrate a rapid release of insulin in the case of lipid nanoparticles without coating and a prolonged release in formulation containing lipid coating (34, 333). Additionally, formulations showed to be compatible with the epithelial respiratory A549 and Calu-3 cell lines (296). In vivo studies in rats show that after intratracheal administration, uncoated lipid nanoparticles reach the alveolar region and promote a greater reduction of systemic levels of glucose, compared to insulin solution (272, 387). Kawashima and co-workers obtained PLGA nanoparticles with average diameter of 400 nm, and alveolar deposition of 75% (w/w) using the modified emulsification/solvent evaporation method (332). In vitro dissolution tests showed an insulin release profile characterized by an initial burst effect followed by extended release. In vivo studies show a significant reduction in systemic levels of glucose which lasts for a period exceeding 48 hours, compared with an aqueous solution of insulin for inhalation (332). This biphasic release of insulin can mimic the marketed injectable insulin mixtures of short and long duration of action. In another study, PLGA nanoparticles were produced using the emulsification/solvent evaporation technique followed by granulation (331). In vitro and in vivo showed an alveolar deposition of approximately 45% (w/w) of emitted dose and pharmacological effect of prolonged over 12 hours when compared with solutions of insulin administered intratracheally and intravenously (331). Poly(n-butyl cyanoacrylate) (PBCA)/dextran nanoparticles obtained by Zhang and co-workers using the in situ polymerization method have a diameter of 255 nm and an encapsulation efficiency of 79% Chapter 1 I State-of-art ___________________________________________________________________________________ 47 (334). In vitro assays demonstrate an insulin release profile characterized by an initial burst effect followed by prolonged release. In vivo studies are characterized by a prolonged therapeutic effect over time when compared to insulin solution administered intratracheally and a bioavailability of 57% compared with subcutaneous administration (334). PLGA nanoparticles coated with chitosan obtained by the emulsion and solvent diffusion technique have been proposed to administer calcitonin by pulmonary route (388). Due to the mucoadhesion promoted by chitosan, coated nanoparticles were eliminated more slowly from the lung compared to those not coated. Moreover, the opening of tight junctions also promoted by chitosan led to an increased absorption of calcitonin and a decreased in in vivo systemic levels of calcium (388). The pharmacological effect was prolonged for 24 hours after inhalation. The particles with a diameter of 650 nm and were nebulized with success and the release profile of calcitonin is characterized by a burst effect followed by prolonged release over time (388). Based in the previous study and to enjoy the advantages of solid calcitonin formulations for inhalation over the liquid ones (389), chitosan-modified PLGA nanocomposites were produced by spray drying fluidized bed granulation (Agglomaster®) and dry powder coating technique (Mechanofusion®) (390). Powders obtained with the Agglomaster® showed improved redispersibility of powders in liquid media and higher in vivo lung retention and hypocalcemic effect, which could be explained by the lower strong aggregation of particles using the spray drying fluidized bed granulation technique (390). Chitosan nanoparticles co-sprayed with mannitol as powders for inhalation of calcitonin were also proposed by other research group, presenting appropriate aerodynamic properties and good absorption to the systemic circulation after pulmonary administration to rats (391). 5. State-of-art of micelles as drug delivery systems by inhalation In the last two decades, pulmonary administration of nanoDDS has been a growing topic of interest among researchers (278, 384, 392). However, micelles as platforms for inhalation of drugs have been poorly explored. Numerous examples of liposomes intended for inhalation have been proposed over the years presenting good results (393, 394), with at least two formulations enrolling clinical trials Arikace® (Insmed at phase I/II) (395) and Pulmaquin® (Aradigm at phase III) (396). Since polymeric micelles presented advantages over liposomes, like higher stability and high capacity of solubilization of hydrophobic drugs, as described before, they possess the potential for pulmonary delivery of drugs. Additionally, the capacity demonstrated by some micelles to overcome multidrug resistance (56, 397) and enhance the Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 48 transfection of genetic material to cells (130), make them promising vehicles for local delivery of anticancer drugs to treat lung cancer, and genetic material to target cells. Also, the small size of micelles confers them the opportunity to escape easily from phagocytosis by alveolar macrophages and allow the delivery and absorption of drugs with systemic action through an epithelium with high surface area and reduced enzymatic activity. Thus, the feasibility of micelles as DDS for pulmonary administration of drugs has been explored in the last years with some studies reviewed in this section. Taking into account the results so far, is expected an increase in the upcoming years of the studies using formulations based on polymeric micelles as inhaled DDS. As referred before, a careful and detailed assessment of the safety of such formulations should be performed, especially in the case of the ones expected to be chronically administered. 5.1. Lipid-polymer micelles As referred previously, lipid-polymer micelles are composed by polymers conjugated with phospholipids or long-chain fatty acids. Several combinations of polymers and lipids/phospholipids have been tested for the preparation and application of micelles. Chitosan oligosaccharide-stearic acid (CSO-SA) micelles were developed by Gilani and coworkers, for the pulmonary administration of amphotericin B (AmB) to treat invasive pulmonary fungal infections in some patients receiving immune suppressive treatments (108). Local administration avoids the systemic side effects of AmB and improves their bioavailability (398). After encapsulation into micelles, AmB presented the same antifungal activity of Fungizone® but lower toxicity (108), a phenomenon intimately related to its aggregation state (399). The micelles possessed positive charges with mean diameters between 100-250 nm, and were efficiently nebulized using an Air-jet nebulizer to particles with FPF up to 52%, making them suitable for pulmonary delivery of AmB (108). The same research group developed chitosan-stearic acid micelles encapsulating itraconazole for pulmonary delivery (400). Positively charged micelles with mean diameters inferior to 200 nm were efficiently and stably nebulized presenting FPF up to 48%. The antifungal activity of itraconazole against C. albicans, A. fumegatus, and A. niger was maintained after its encapsulation into micelles (400). Thus, the solubilization of hydrophobic itraconazole into polymeric micelles could be an effective approach to deliver the drugs by inhalation in order to treat pulmonary fungal infections. Chapter 1 I State-of-art ___________________________________________________________________________________ 49 Different research groups have been studying the administration of drugs to treat tuberculosis using polymeric micelles. Stearic acid-branched polyethyleneimine (SA-BPEI) micelles encapsulating rifampicin were spray-dried and powders with a drug content of 48%, a MMAD lower than 2.5 µm, and FPF of 67% obtained. Moreover, micelles showed in vitro biocompatibility up to 75 μg/mL concentration and taken up by THP-1 cells differentiated to macrophages (401). Thus this system could be explored to target alveolar macrophages, the reservoir of Mycobacterium tuberculosis. Methoxy poly(ethylene oxide)-b-distearoyl phosphatidyl-ethanolamine (mPEG–DSPE) micelles where also proposed as carriers for rifampicin, although its therapeutic efficacy was not assessed (60). Rifampicin was entrapped with high encapsulation efficiency into these micelles that sustained its release over 3 days in vitro. Formulations presented a fraction of fine particles of approximately 40% after nebulization (60). The potential of 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-methoxy(poly(ethylene glycol)) (DSPE-PEG) as pulmonary DDS was evaluated. Paclitaxel was successfully encapsulated in DSPE-PEG micelles with an encapsulation efficiency of 95% and was slowly released in simulated surfactant lung fluid, reaching 90% of drug release after 8 hours. Additionally, micelles showed to be stable in water during 3 months of storage. After intratracheal administration of paclitaxel-loaded micelles in rats, the lung concentration of paclitaxel was significantly higher as compared to intravenous administration of micelles and intratracheal administration of Taxol®. While paclitaxel is mainly accumulated in organs such liver and spleen and rapidly cleared from lungs after intravenous administration of micelles and intratracheal administration of Taxol®, respectively, around 50% of the paclitaxel concentration remains in lungs 12 hours after intratracheal administration of paclitaxel-loaded micelles. These results show that pulmonary administration of paclitaxel leave to low systemic exposure, resulting in localized chemotherapy to the lungs and avoiding the unwanted side effects to other tissues (402). In another study, inhaled DSPE-PEG micelles of 16 nm in size encapsulating doxorubicin showed higher accumulation in the lungs and lower distribution in non-target organs compared to its intravenous administration (403). These results reinforce the usefulness of local administration of anticancer drugs through inhalation to treat lung cancer. In addition, micelles showed a strongest tendency to be accumulated to the lungs for the longer periods of time compared to mesoporous silica nanoparticles, dendrimers, and quantum dots (403), demonstrating the feasibility of micelles as nanoDDS for inhalations of drugs. Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 50 DSPE-PEG micelles have also been proposed for pulmonary delivery of anti-inflammatory drugs like beclomethasone dipropionate (404, 405) or budesonide (406). Gaber and coworkers developed micelles of 22 nm diameter and a high encapsulation efficiency (> 96%) presenting a sustained release profile and a FPF suitable for pulmonary delivery, with lower deposition of particles in the throat (404). In another study DSPE-PEG was conjugated with α,β-poly(N-2-hydroxyethyl)-DL-aspartamide (PHEA) in order to further improve the pulmonary delivery of corticosteroids to treat bronchial inflammatory diseases (407). Beclomethasone dipropionate was efficiently encapsulated in 1,2-distearoyl-sn-glycero-3phosphoethanolamine-N-methoxy(poly(ethylene glycol))-α,β-poly(N-2-hydroxyethyl)-DLaspartamide (DSPE-PEG-PHEA) micelles presenting a sustained release in vitro with less than 30% of the drug being released after 48 hours. The obtained micelles showed to be biocompatible with human bronchial epithelial cells (16HBE14o-) and enhanced the drug uptake by the same cell line after 48 hours of incubation (407). Due to its cationic nature, chitosan can be complexed with negatively charged DNA and be used as non-viral vector for gene therapy. Hu and co-workers synthesized CSO-SA in order to produce polymeric micelles to delivery pEGFP-C1 (408). The CSO-SA/DNA micelles efficiently protected the condensed DNA from enzymatic degradation by DNase I, presented lower cytotoxicity and comparable transfection efficiency in A549 cells compared to Lipofectamine® 2000 (408), making these micelles a promising gene delivery system in the treatment of pulmonary diseases. The same research group developed CSO-SA micelles for the delivery of drugs like paclitaxel (107) or doxorubicin (106, 409). The encapsulation of doxorubicin in CSO-SA micelles resulted in higher uptake and accumulation by A549 cells and a decreasing in the half maximal inhibitory concentration (IC50) value (45). Although the promising results, the feasibility of such formulations as inhaled or intravenous delivery systems needs to be confirmed with in vivo studies. 5.2. Copolymer-based micelles As referred previously, besides the use of lipids as hydrophobic segment of polymeric micelles, is possible to produce this kind of micelles with amphiphilic copolymers composed by segments with different water affinity nature, being ones hydrophilic and others hydrophobic. Chapter 1 I State-of-art ___________________________________________________________________________________ 51 Laouini and co-workers developed different systems, namely polymeric micelles, liposomes, SLN and nano-emulsions, intended for pulmonary administration of vitamin E and assess their aerodynamic properties using different techniques (410). PEG-b-PCL micelles of 154 nm in size, low PdI (0.09) and high AE (87.4%) (411), presented a dae of 5.8 µm and a FPF of 29% as determined by laser diffraction analysis of the nebulized micelles dispersion (410). On the other hand, a MMAD of 3.2 µm and a FPF of 78% were obtained when the formulations were assessed by cascade impaction. Additionally, an in silico prediction of the aerodynamic behavior of nebulized micelles using the multiple-path particle dosimetry, resulted in ~52% of lung deposition (410). These results evidence the importance of multiple technique assessment of the aerodynamic properties of inhaled formulations, in order to better predict its in vivo behavior. The incorporation of a hydrazone bond between the PEG and PCL blocks allowed the development of pH-sensitive micelles (411), that could be explored for the targeting of drugs to cancer or inflammation tissues. As referred previously, polymeric micelles have been proposed as platforms for the inhalation of antitubercular drugs. Wu and co-workers synthesized and characterized polylactidechitosan (PLA-chitosan) copolymers with different molar ratios for delivery of rifampicin (412). As polylactide (PLA) molar ratio increased, the micelle size and drug-loading content increased with a decreasing in rifampicin release rate (412). In another study, enantiomeric poly(ethylene glycol)–polylactic acid (PEG-PLA) stereocomplex micelles composed by a equimolar mixture of enantiomeric poly(ethylene glycol)–poly(l-lactide) (PEG-PLLA) and poly(ethylene glycol)–poly(d-lactide) (PEG-PDLA) were developed to sustained release of rifampicin (413). The stereocomplex micelles presented lower CMC and mean diameter, higher stability in water and encapsulation efficiency than those observed with the single enantiomeric micelles. In vitro drug delivery release was characterized by an initial burst release followed by a sustained release until 48 hours (413). Silva and co-workers evaluated the feasibility of poly(ethylene glycol)-poly(aspartic acid) (PEG-Pasp) micelles as delivery system for tuberculostatic agents. The conjugation of isoniazid and pyrazinamide with PEGPasp improved the activity against Mycobacterium tuberculosis by reducing the minimal inhibitory concentration (MIC) of the drug (58, 414). Poly(ε-caprolactone)-b-poly(ethylene glycol)-b-poly(ε-caprolactone) (PCL-PEG-PCL) micelles surface modified with chitosan and galatomannan were also proposed as nanoDDS for delivery of rifampicin. The presence of galatomannan increased the uptake by macrophages, being these micelles proposed for improved therapy of tuberculosis by targeting alveolar macrophages (18). Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 52 For the treatment of fungal infections, in a 12 days study in mice, the inhalation of itraconazole:polysorbate 80:F127 nanostructured aggregates triggered higher lung concentrations and lung-to-serum ratios of itraconazole, improved survival of infected animals while decreasing toxicity compared with orally administered itraconazole formulations (415418). No signs of lung inflammation or changes in pulmonary histology were detected (419). It is worth stressing that a lower drug dose was required to achieve lung and serum therapeutic levels using inhaled formulation comparing to oral administration (416). Liu and co-workers developed an tri-block copolymer consisting of Poly(epsiloncaprolactone)-b-chitooligosaccharide-b-poly(ethylene glycol) (PCL-b-COS-b-PEG) for delivery of doxorubicin (420). The obtained polymer presents the capacity to form micelles with encapsulation efficiency of doxorubicin close to 50%. Genipin post-crosslinking did not affect the macroscopic characteristics of the micelles but delayed the in vitro release of doxorubicin from the micellar reservoir (420). Similar results were obtained by Chen and coworkers using chitosan-poly(ε-caprolactone)-poly(ethylene glycol) to encapsulate paclitaxel and rutin with glutaraldehyde post-crosslinking (44). In another study, Kontoyianni and coworkers synthesized a new type of amphiphilic polymer based on the PEGylation of a hyperbranched aliphatic polyester (BH40-PEG polymer) (421). The polymer produces micelles presenting 20 nm of mean diameter that could encapsulate paclitaxel. Paclitaxel water solubility significantly increased after encapsulation into micelles, and the polymer showed to be non-toxic to A549 cells up to a concentration of 1.75 mg/mL, while the median lethal dose (LD50) was 3.5 mg/mL (421). The obtained results make these systems promising for administration of anticancer agents, however, further studies are required in order to assess their feasibility as inhaled drug delivery systems. Besides the conventional chemotherapy, it is possible resort to photodynamic therapy (PDT) to treat lung cancer. PDT consists in the administration of photosensitizer agents that generate reactive oxygen species after activation of the system by light exposure at the targeted tissues. Pluronic L122 was used by Yang and co-workers to encapsulate hematoporphyrin for pulmonary delivery. Micelles with 98% of encapsulation efficiency exhibited higher cellular uptake and cytotoxicity against A549 cells as compared to the free drug. In addition, micelles were efficiently incorporated into lactose microparticles with ~2 μm by spray-drying, making the system a suitable DPI for pulmonary administration (422). Gene therapy aiming to treat pulmonary pathologies was also explored. An amphiphilic polyethylenimine (PEI) derivative was synthesized by Roesler and co-workers (423). Although PEI is one of the most effective polycations for gene delivery due to is high-density Chapter 3 I Design and characterization of self-assembled micelles for insulin delivery ___________________________________________________________________________________ 59 Chapter 3 Design and characterization of self-assembled micelles for insulin delivery The information presented in this chapter was partially published in the following publication: Fernanda Andrade, Pedro Fonte, Mireia Oliva, Mafalda Videira, Domingos Ferreira, Bruno Sarmento, Solid state formulations composed by amphiphilic polymers for delivery of proteins: characterization and stability, International Journal of Pharmaceutics, 2015, 486:195-206 Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 60 1. Introduction In the last decades, the use of polymers in the development of drug delivery systems has gained a new breath as consequence of the progresses seen in the fields of polymer engineering and nanotechnology applied to health. Among them, amphiphilic polymers have emerged as platforms for advanced delivery of a variety of drugs (41). A multitude of monomers and polymers can be conjugated in order to obtain amphiphilic polymers with modulated characteristics (136). The most commonly used are poloxamers, triblock copolymers of polyoxyethylene and polyoxypropylene, commercially known as Pluronic® (164). However, PLGA-PEG, poly(ethylene glycol)-b-poly(ε-caprolactone) (PEG-b-PCL) as well as their derivatives (18, 127) are also commonly used. Nanotechnology-based delivery systems have been explored to solve the drawbacks of conventional formulations such as instability and degradation, reduced permeation through biomembranes and bioavailability (25, 428). Polymeric micelles are spherical shape nanosized structures composed by amphiphilic polymers or polymers conjugated with lipids that are suitable as drug delivery systems. The inner core of micelles presents the capacity to encapsulate hydrophobic drugs, while the shell can associate the hydrophilic ones (79). Due to its small size, micelles generally escape from the reticulo-endothelial system, presenting higher bloodstream circulation time (40). Also, some studies suggest the capacity of polymeric micelles to inhibit the drug efflux mechanisms and consequently multidrug resistance (113, 397). In addition, as liposomes, the surface of polymeric micelles can be easily tailored with specific ligands for targeted delivery (141). Nevertheless, micelles present the advantage of being more stable than liposomes (63). The versatility of micelles explains why they have been proposed as vehicles for solubilization and delivery of a variety of drugs like doxorubicin (119), paclitaxel (66, 68), rifampicin (18), calcitonin (59), CsA (124) among others, being some formulations in clinical trials (66, 68). Due to the development noted in the biotechnology field, biopharmaceuticals have emerged as an alternative to conventional drugs in the treatment of many diseases. Since the commercialization of insulin obtained by biotechnology processes in 1982, biopharmaceuticals have gained an increased share in the global pharmaceutical market (207). Despite their well-known therapeutic efficacy, the major drawback of biopharmaceutical drugs is the difficulty of administration via non-invasive routes in their active conformation. Among the different non-invasive routes of administration, inhalation appears as a promise one due to the physiological characteristics of lungs and airways resulting in higher Chapter 3 I Design and characterization of self-assembled micelles for insulin delivery ___________________________________________________________________________________ 61 bioavailability than for other non-invasive routes (213). In addition, inhalation of insulin presented better therapeutic results when compared to the oral administration. As a result, an insulin-based formulation (Exubera®) achieved market authorization from FDA and EMA. However, the product was withdrawn due to commercial and financial reasons, being also reported cases of adverse effects and lung cancer after the use of this product (238). More recently, FDA approved a new and improved powder formulation for insulin inhalation (Afrezza®). In this chapter the production of micelles with characteristics appropriated to pulmonary delivery and the association of insulin to the system was explored. 2. Experimental 2.1. Materials SOL, F68, F108 and F127 were kindly provided by BASF (Ludwigshafen, Germany). Lyophilized human insulin, PBA and phosphate buffer saline pH 7.4 (PBS) were purchased from Sigma-Aldrich (St. Louis, MO, USA). The other reagents used were methanol and ethanol absolute from analytical grade; acetonitrile and trifluoroacetic acid (TFA) from highperformance liquid chromatography (HPLC) grade (Merck, Germany) and Type 1 ultrapure water (18.2 MΩ.cm at 25 ºC, Milli-Q®, Billerica, MA, USA). 2.2. Production of micelles Micelles were prepared using the thin-film hydration technique. Briefly, each polymer was individually weight and dissolved in methanol or a mixture of methanol:ethanol (1:1). Then, the solvent was removed under vacuum and the film was left to dry overnight at roomtemperature to eliminate any remained solvent. The film was then hydrated with Type I ultrapure water or PBS at 37 ºC in order to obtain a 1 % (w/v) solution and vortexed for 5 min. The obtained dispersion was filtered through a 0.22 µm syringe filter to remove possible dust and aggregates. PBA containing micelles were prepared by dissolving PBA with the polymers in the solvents prior to the production of the film at a ratio of 10:1 (polymer:PBA). Insulin formulations were prepared by adding different amounts of insulin in the form of solution in PBS during the film hydration, to obtain polymer:insulin ratios ranging from 10:0.1 to 10:1 Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 62 (w/w). The other steps were the same as for plain formulations. After preparation, the pH of all formulations was measured, ranging between 6.1 and 7.1. 2.3. Determination of size, zeta potential, association efficiency, and osmolality of formulations Particle mean hydrodynamic diameter and polydispersity index (PdI) were measured without previous sample dilution by dynamic light scattering (DLS) at both 25 °C and 37 °C using a detection angle of 173° and zeta potential by laser doppler micro-electrophoresis using a NanoZS (Malvern Instruments, UK). For each type of formulation were produced and analyzed at least three replicates. The osmolality of formulations was determined at room temperature using a Micro-Osmometer M3320 (Advanced Instruments, Inc., MA, USA). Triplicates of each formulation were analyzed. AE, i.e. the amount of insulin associated with the micelles, and the LC, i.e. the mass percentage of insulin of the total mass of the particles was calculated according to the Equation 3.1 and 3.2, respectively. The free insulin in filtrate was recovered after filtration of the formulations by centrifugation for 10 min at 10,000 rpm at 37 ºC, using a 30k pore filter (Nanosep® Centrifugal Devices, Pall Corporation, Spain). A previously validated HPLC method was used to quantify the insulin (432). Briefly, the mobile phase consists of acetonitrile:0.1% (v/v) TFA aqueous solution initially set to a ratio of 30:70 (v/v), which was linearly changed to 40:60 (v/v) over 5 min. From 5 to 10 min the ratio was kept constant at 40:60 (v/v). The mobile phase was pumped at a constant flow rate of 1 ml/min, the injection volume was 20 µl and the detection wavelength used was 214 nm. The HPLC (UltiMate® 3000 UHPLC+ focused, Dionex, USA) system was equipped with a Purospher® STAR RP18e (5µm) LiChroCART® 250-4.6 (Merck, Germany) and a LiChrospher® 100 RP-18 (5 µm) LiChroCART® 4-4 guard column (Merck, Germany). All experiments were performed at 20 ºC and the total area of the peak was used to quantify insulin. For each type of formulation were analyzed at least three replicates. AE= total amount of insulin – free insulin in filtrate total amount of insulin × 100 Equation 3.1 LC= total amount of insulin – free insulin in filtrate total weight of micelles × 100 Equation 3.2 Chapter 3 I Design and characterization of self-assembled micelles for insulin delivery ___________________________________________________________________________________ 63 2.4. Morphological characterization of micelles Different microscopic techniques, namely atomic force microscopy (AFM), field emission scanning electron microscopy (FE-SEM) and transmission electron microscopy (TEM), were used to characterize the morphology of the micelles. AFM imaging was performed using a MultiMode VIII microscope (Bruker AXS Inc., Madison, WI, USA) with a NanoScope V controller (Veeco Instruments Inc., Plainview, NY, USA). One drop of formulation was placed on top of freshly cleaved highly oriented pyrolytic graphite and left for 15 min before being removed and replaced with PBS, with no drying step in between. Analysis was conducted under fluid tapping mode using sharp silicon tips on nitride levers (model SLN-10 A, Bruker AFM Probes) with pyramidal shape and nominal tip radius of 2 nm and nominal spring constant 0.35 N.m-1. For FE-SEM performed in a Hitachi H-4100FE (Hitachi Ltd., Tokyo, Japan), a drop of formulation was placed on top of freshly cleaved highly oriented pyrolytic graphite and dried in a desiccator overnight prior to carbon coating. Regarding TEM, samples were placed on a grid, treated with uranil acetate and then observed in a JEM-1400 Transmission Electron Microscope (JEOL Ltd., Tokyo, Japan). 2.5. Statistical analysis One-way ANOVA was used to investigate the differences between formulations. Post hoc comparisons were performed according to Tukey’s HSD test (p0.05 was accepted as significant different) using Prism 6.02 software (GraphPad Software, Inc., CA, USA). 3. Results 3.1. Size, surface charge and association efficiency of micelles For the production of micelles, a first screening with different combinations of polymers (SOL, F68, F108 and F127) evaporation solvents, namely methanol and a mixture of methanol and ethanol (1:1), and hydration solvent (H2O or PBS) were tested and the characteristics of the micelles are presented in Table 3.1. No significant differences were observed in respect to the evaporation and hydration solvents used. Thus, in order to control the pH of formulations Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 64 and reduce the percentage of methanol used, the combination of methanol:ethanol (1:1) and PBS was chosen for further studies. Also, a filtration step after the hydration of the polymeric film was added to the production protocol to eliminate the aggregates observed for Pluronic®- based micelles. Table 3.1 Mean hydrodynamic diameter, polydispersity index (PdI) and zeta potential of micelles produced with different evaporation and hydration solvents. Samples were analyzed at 25 ºC. The results are expressed as mean values ± SD, n≥3. Polymer Evaporation Solvent Hydration Solvent Hydrodynamic Diameter (nm) PdI Zeta Potential (mV) SOL Methanol H2O 53.6±11.4 0.031±0.007 -4.9±15.1 PBS 43.0±1.9 0.063±0.042 -7.8±8.3 Methanol: Ethanol H2O 50.4±2.5 0.022±0.001 -2.7±1.0 PBS 50.9±6.6 0.038±0.015 -8.9±11.3 F68 Methanol H2O 56.4±20.8 0.566±0.161 -5.7±7.4 PBS 337.9±219.5 0.364±0.060 -9.2±2.1 Methanol: Ethanol H2O 148.3±39.1 0.203±0.036 -20.1±4.7 PBS 87.0±21.6 0.294±0.122 -4.3±4.8 F108 Methanol H2O 417.1±98.3 0.279±0.387 -22.5±22.9 PBS 400.8±312.2 0.680±0.267 -2.3±2.3 Methanol: Ethanol H2O 550.6±268.0 0.406±0.029 -16.9±30.9 PBS 455.6±105.3 0.345±0.148 0.2±3.6 F127 Methanol H2O 283.6±112.7 0.144±0.081 -2.6±2.5 PBS 165.1±118.2 0.360±0.100 -4.2±3.4 Methanol: Ethanol H2O 268.4±159.1 0.254±0.308 -9.1±22.0 PBS 131.6±76.3 0.356±0.070 -1.9±6.2 The results regarding the characterization of the selected micelles in terms of size and surface charge are presented in Figure 3.1. Empty micelles of SOL presented a mean hydrodynamic diameter of 55.10 ± 7.72 nm and 288.80 ± 37.35 nm, and a PdI of 0.026 ± 0.015 and 0.298 ±0.115 at 25 ºC and 37 ºC, respectively, showing a clear temperature dependence behavior (p<0.05). The incorporation of insulin at different ratios up to 10:1 reduced the size of micelles although no statistical differences were observed (p>0.05). In the Chapter 3 I Design and characterization of self-assembled micelles for insulin delivery ___________________________________________________________________________________ 65 same way, the incorporation of PBA to micelles did not produce significant differences compared to the plain micelles (p>0.05). All micelles presented higher mean diameters and PdI at 37 ºC compared to 25 ºC (p<0.05). Regarding the surface charge, particles showed negative zeta potential values near zero without significant differences among formulations (p>0.05). Figure 3.1 Mean hydrodynamic diameter, polydispersity index (PdI) and zeta potential of SOL (black bars and squares) (A), F68 (grey bars and triangles) (A), F108 (black bars and squares) (B) and F127 (grey bars and triangles) (B) empty micelles, containing just PBA micelles (empty:PBA), insulin-loaded micelles with different polymer:insulin ratio (10:0.1, 10:0.2, 10:0.3, 10:0.4, 10:0.5, 10:0.75 and 10:1) and insulin-loaded containing PBA micelles with 10:1 polymer:insulin ratio (10:1:PBA) after production (mean ± SD, n≥3). Pluronic®-based micelles presented different results depending of the polymer used. The mean diameter and PdI of empty micelles at 25 ºC was 92.47 ± 21.46 nm and 0.233 ± 0.087 for F68; 53.19 ± 29.91 nm and 0.575 ± 0.176 for F108; and 29.54 ± 10.09 nm and 0.202 ± 0.060 for F127. At 37 ºC the micelles were somewhat larger in size, presenting higher polydispersity, being the differences not significant (p>0.05) for the majority of formulations. The incorporation of insulin up to 10:1 increased the diameter of micelles prepared by F68 (p<0.05), and did not promote significant changes in micelles prepared by F108 and F127 Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 66 (p>0.05). On the other hand, the incorporation of PBA did not significantly alter the characteristics of F68, F108 and F127 micelles (p>0.05). Pluronic®-based micelles also presented a surface charge slightly negative and close to neutrality without significant differences among the formulations (p>0.05). Micelles composed by 10:1 polymer:insulin ratio were chosen to proceed with the production and characterization, since they showed to possess similar values of size, PdI and surface charge of the ones containing lower insulin payloads. The AE and LC were determined for micelles with a polymer:insulin ratio of 10:1. As seen in Table 3.2, excepting for F68, all the formulations presented an AE higher than 80% and LC of at least 7%. The presence of PBA didn´t affect the values of AE and LC (p>0.05). Table 3.2 Association efficiency (AE), loading capacity (LC) and osmolality of the different insulinloaded formulations. Results are presented as mean values ± SD (n≥3). Sample AE (%) LC (%) Osmolality (mOsm/Kg) SOL:Ins 94.63±3.24 8.60±0.29 397±15 SOL:Ins:PBA 84.03±5.14 7.31±0.45 402±6 F68:Ins 76.22±14.56 6.93±1.32 325±12 F68:Ins:PBA 49.31±36.80 5.60±3.13 314±9 F108:Ins 87.28±11.57 7.93±1.05 316±10 F108:Ins:PBA 87.63±3.63 7.62±0.32 330±14 F127:Ins 80.90±14.92 7.35±1.36 325±18 F127:Ins:PBA 83.15±10.05 7.23±0.87 322±10 The osmolality of formulations was superior to 300 mOsm/Kg, being SOL-based samples similar between them (p>0.05) and different from the Pluronic®-based micelles (p<0.05). No differences were observed between the different Pluronic®-based formulations (p>0.05). Chapter 3 I Design and characterization of self-assembled micelles for insulin delivery ___________________________________________________________________________________ 67 3.2. Morphological characterization The morphology of micelles was analyzed with different microscopy techniques. FE-SEM images are presented in Figure 3.2, while TEM images presented in Figure 3.3 and 3.4, and AFM images depicted in Figure 3.5 and 3.6. All the microscopy techniques used showed that the micelles of the different polymers are mainly spherical in shape. Figure 3.2 FE-SEM micrographs of SOL (A), F68 (B), F108 (C) and F127 (D) insulin-loaded micelles. It also revealed the presence of some aggregates of smaller particles. FE-SEM analysis revealed a higher degree of aggregation compared to TEM and AFM due to the dried state of samples during the analysis. No visible differences between empty, insulin-loaded micelles and micelles containing PBA were observed. Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 68 Figure 3.3 TEM images of SOL (A-C) and F68 (B-D) empty micelles (A-B) and insulin-loaded micelles (C-D). Figure 3.4 TEM images of F108 (A-C) and F127 (B-D) empty micelles (A-B) and insulin-loaded micelles (C-D). Chapter 4 I Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization _______________________________________________________________________________________ 75 Chapter 4 Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization The information presented in this chapter was partially published in the following publications: Fernanda Andrade, Pedro Fonte, Mireia Oliva, Mafalda Videira, Domingos Ferreira, Bruno Sarmento, Solid state formulations composed by amphiphilic polymers for delivery of proteins: characterization and stability, International Journal of Pharmaceutics, 2015, 486:195-206 Fernanda Andrade, José das Neves, Petra Gener, Simó Schwartz Jr, Domingos Ferreira, Mireia Oliva, Bruno Sarmento, Biological assessment of self-assembled polymeric micelles for pulmonary administration of proteins, submitted for publication Fernanda Andrade, Pedro Fonte, Ana Costa, Cassilda Cunha Reis, Rute Nunes, Carla Pereira, Domingos Ferreira, Mireia Oliva, Bruno Sarmento, In vivo pharmacological and toxicological assessment of self-assembled polymeric micelles as powders for inhalation of proteins, submitted for publication Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 76 1. Introduction Inhalation of compounds has been performed since ancient cultures to treat diseases mainly affecting the respiratory system like asthma. However, a change in the paradigm is occurring and pulmonary administration of drugs with systemic action has been proposed (84, 448). Recently, a large extent of research has been performed regarding the development of formulations for inhalation and improved delivery devices, especially dry powder inhalers, due to their advantages over liquid formulations, namely long-term stability and patients convenience (449). The efficacy of an inhaled drug is dependent on its deposition pattern in the respiratory system, which is affected by several factors related to the properties of the formulation and the inhaler device, as well as physiologic characteristics of patients. Particles with a MMAD lower than 5 μm are assumed to deposit in the lungs and reach the alveoli if MMAD is below 3 μm, therefore becoming available to either exert a local effect or to undergo systemic absorption (275, 277, 449, 450). It was already demonstrated that the in vitro particle size and aerosolization profiles correlate in an acceptable way with in vivo lung deposition pattern (451), reason why the regulatory agencies require in vitro data regarding particle size distribution and aerosolization properties of the inhalable formulations before approval. Accordingly, FPF (particles < 5 µm), defined as the percentage of particles from the total emitted dose that are able to reach and deposit in the airways and deep lung, is used as an indicator for formulation efficiency. Compendial devices like the eight-stage Andersen nonviable Cascade Impactor are commonly used to assess the deposition profile of inhaled formulations. In this chapter it is described the development and characterization of powders through the lyophilization of the insulin-loaded micelles described in the Chapter 3. 2. Experimental 2.1. Materials SOL, F68, F108 and F127 were kindly provided by BASF (Ludwigshafen, Germany). Lyophilized human insulin, PBA, PBS, and D-glucose were purchased from Sigma-Aldrich (St. Louis, MO, USA). The other reagents used were acetonitrile and TFA from HPLC grade (Merck, Germany) and Type 1 ultrapure water (18.2 MΩ.cm at 25 ºC, Milli-Q®, Billerica, MA, USA). Chapter 4 I Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization _______________________________________________________________________________________ 77 2.2. Production of micelles and lyophilization Micelles were prepared using the thin-film hydration technique. Briefly, each polymer was individually weight and dissolved in a mixture of methanol:ethanol (1:1). Then, the solvent was removed under vacuum and the film was left to dry overnight at room-temperature to eliminate any remained solvent. The film was then hydrated with PBS at 37 ºC in order to obtain a 1 % (w/v) solution and vortexed for 5 min. The obtained dispersion was filtered through a 0.22 µm syringe filter to remove possible dust and aggregates. PBA containing micelles were prepared by dissolving PBA with the polymers in the solvents prior to the production of the film at a ratio of 10:1 (w/w) (polymer:PBA). Insulin-loaded micelles were prepared by hydrating the polymeric films with an insulin solution in PBS to obtain polymer:insulin ratios of 10:1 (w/w). The other steps were the same as for plain formulations. After production micelles were lyophilized in an AdVantage 2.0 BenchTop Freeze Dryer (SP Scientific, Warminster, PA, USA). The cycle used was the follow: the samples were frozen at -30 ˚C and the temperature maintained for 60 min, the primary drying was set at 20 ˚C for 480 min at 150 mTorr and the secondary drying for another 480 min at 30 ˚C and 100 mTorr. 2.3. Determination of size and zeta potential of formulations Particle mean hydrodynamic diameter and PdI was measured without dilution of the samples by DLS at both 25 °C and 37 °C using a detection angle of 173° and zeta potential by laser doppler micro-electrophoresis using a NanoZS (Malvern Instruments, UK). For each type of formulation were produced and analyzed at least three replicates. 2.4. Thermal analysis The thermal behavior of the pure compounds, physical mixtures (1:1) and lyophilized formulations was assessed by DSC. Thermograms were obtained using a Shimadzu DSC-60 system (Shimadzu, Kyoto, Japan). 5 mg of each powder sample in an aluminum crimp was exposed to a controlled thermal treatment, specifically heated from 30 to 300ºC at a rate of 10ºC/min under constant purging of nitrogen at 40 mL/min, and the heat flow measured. Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 78 2.5. X-ray diffraction (XRD) experiments Crystallization properties of powder samples were analyzed by XRD. Spectra were acquired using X’Pert PRO MPD / powder diffractometer of 240 mm of radius (PANalytical B.V., Almelo, Netherlands) in a configuration of convergent beam with a focalizing mirror and a transmission geometry. Samples were sandwiched between films of polyester of 3.6 µm of thickness and scanned at 45 kV, 40 mA using Cu Kα1 radiation (λ = 1.5418 Å) at the range 2/ scans from 2 to 60 º2 with a step size of 0.026 º2 and a measuring time of 400 s per step. 2.6. Raman spectroscopy The micro-Raman spectra of powder formulations were acquired using dispersive high resolution micro Raman spectrometer (Jobin-Yvon LabRam HR 800) coupled with an optic microscope (Olympus BXFM) with a 50X objective. A laser of 532 nm wavelength and 2.5 mW of potency and a charge coupled device detector cooled at -70 ºC were used. The spectra were acquired and analyzed with the software LabSpec 5 (Horiba, Kyoto, Japan). 2.7. Surface analysis The elemental composition of the surface of particles in powder state was analyzed by XPS. XPS experiments were performed in a PHI 5500 Multitechnique System (Physical Electronics, MN, USA) equipped with a monochromatic X-ray source (Aluminium Kalfa line of 1486.6 eV energy and 350 W), placed perpendicular to the analyzer axis and calibrated using the 3d5/2 line of Ag with a full width at half maximum of 0.8 eV. The analyzed area was a circle of 0.8 mm diameter, and the selected resolution for the survey spectra was 187.85 eV of pass energy and 0.8 eV/step. A low energy electron gun (less than 10 eV) was used in order to discharge the surface when necessary. All measurements were made in an ultrahigh vacuum chamber with a pressure between 5x10-9 and 2x10-8 torr. The spectra were acquired, analyzed, and the atomic concentration of the elements quantified using the MultiPak 6 software (Physical Electronics, MN, USA). Chapter 4 I Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization _______________________________________________________________________________________ 79 2.8. Assessment of insulin conformation FTIR and far-UV CD were used to analyze the conformation of insulin in the formulations in order to assess its stability after lyophilization. Infrared spectroscopy analysis was conducted in a FTIR spectrometer ABB MB3000 (ABB, Switzerland) equipped with a deuterated triglycine sulphate detector and using a MIRacle single reflection horizontal attenuated total reflectance accessory (PIKE Technologies, USA) with a diamond/Se crystal plate. All spectra were acquired with 256 scans and 4 cm−1 resolution in the region of 4000–600 cm−1 using a triplicate set of samples and the related blank sample (raw polymer) after a background, and insulin spectra were obtained by a double subtraction procedure (452). After subtraction, spectra were derived using a 15 points Savitzky–Golay second-derivative and the amide I region (1590–1710 cm−1) was selected. The spectra were baseline corrected using a 3–4 point adjustment and area-normalized. All spectra treatment was executed using the Horizon MB FTIR software (ABB, Switzerland). Quantitative comparison of the overall similarity of the FTIR spectra between native insulin and insulin-loaded micelles was assessed by using spectral correlation coefficient (SCC) and area of overlap (AO) algorithms (Origin software, OriginLab Corporation, MA, USA) (453). The far-UV CD spectra were acquired using a Jasco J-815 spectropolarimeter (JASCO International Co., Ltd., Japan) at 20 ºC. In the far-UV region the spectra were recorded in a 1 cm cell from 250 to 190 nm, using a step size of 0.5 nm, a bandwidth of 1.5 nm and a speed of 50 nm/min with the lamp housing purged with nitrogen flow at 10 mL/min to remove oxygen. For all spectra and average of 5 scans was obtained. Appropriate references were used to subtract the signal of the polymers from the spectra of the protein-loaded. The mean residue ellipticity [θ] for insulin was calculated as the CD signal (θ) × mean residual weight (MRW) (116 Da for each insulin residue)/[10 × cell pathlength (cm) × insulin concentration (g/ml)] (454). 2.9. Scanning electron microscopy The shape and morphology of lyophilized formulations was observed by scanning electron microscopy (SEM) on a FEI ESEM Quanta 200 (FEI, Hillsboro, USA). Samples were mounted onto metal stubs and coated with a carbon layer prior to observation. Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 80 2.10. Powder’s particle size distribution and aerodynamic diameter The geometrical particle size distribution of the 20 mg samples of lyophilized formulations was determined by laser diffraction using a Malvern Mastersizer 2000® laser diffractometer equipped with a dry sampling system (Scirocco® 2000, Malvern Instruments, UK) The powders were passed through a 840 µm sieve prior to analysis to eliminate possible aggregates. The volume particle size distribution was characterized by D0.1 (10% of the particles volume has a diameter below that value), D0.5 also known as mass median diameter (50% of the particles volume has a diameter below that value), and D0.9 (90% of the particles volume has a diameter below that value). Values presented are the average of at least three replicates. The theoretical dae, Carr’s index, and Hausner ratio were estimated from the geometrical particle size and tapped density (ρ, determined by tap density measurements) data according to Equation 4.1, 4.2, and 4.3, respectively. dae=𝐷0.5√𝜌 𝜌𝑜𝜒 Equation 4.1 where D0.5 is assumed as geometrical mean diameter, ρ0 is the reference density of a 1 g/cm3 sphere, and χ is the dynamic shape factor. Carr′s index= tap density−bulk density tap density × 100 Equation 4.2 Hausner ratio= tap density bulk density Equation 4.3 2.11. In vitro aerosolization and deposition properties An eight-stage Andersen non-viable Cascade Impactor (ACI, Copley Scientific, UK) was used to determine the aerosolization and deposition properties of formulations in vitro. Hard gelatin nº 4 capsules were manually filled with the powder formulations sieved through a 300 µm sieve, and individually loaded into a Rotahaler® (GlaxoSmithKline, RTP, NC) inhaler device. Chapter 4 I Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization _______________________________________________________________________________________ 81 The experiments were performed at a flow rate of 28.3 L/min and four liters of air passed through the system, as recommended by European Pharmacopoeia (2.9.18. Preparations for inhalation: aerodynamic assessment of fine particles) (455). The stages were individually weighted and the FDF calculated as the amount of the particles deposited in stage 3 or lower in the cascade impactor (particles < 4.7 μm) as a percentage of the initial amount of particles. MMAD and geometrical standard deviation (GSD) were estimated through the cumulative masses of powder deposited in the impactor using a mathematic software (MMADcalculator) developed by Dr. Jay Holt (456). Each experiment was run in triplicate. 2.12. Insulin in vitro release study Insulin-loaded micelles were dispersed in 10.0 mL of PBS with and without D-glucose (1.2 mmol/L) and incubated at 37°C under magnetic stirring. Samples of 0.5 mL were taken at predetermined time intervals of 15, 30, 45 min, 1, 2, 4, 6, 8 and 24 hours and replaced with fresh medium maintained at the same temperature. The collected samples were centrifuged for 10 min at 10,000 rpm and 37 ºC, using 100k pore filters (Nanosep® Centrifugal Devices, Pall Corporation, Spain) and insulin quantified by the HPLC methodology described in Chapter 3. All samples were run in triplicate. The similarity factor (f2) used for comparison of the different formulations was calculated according to the Equation 4.4. 𝑓2=50 × log{[1 + (1 𝑛) ∑ (𝑅𝑡−𝑇𝑡) 𝑛 𝑡=1 ]−0.5 ×100} Equation 4.4 where n is the number of time-points considered, and Rt and Tt are the percentage of insulin released at each time-point (t) for reference and test formulations, respectively. Insulin release profiles with values of f2 between 50 and 100 were assumed to be similar (457). 2.13. Stability studies In order to assess the stability of formulations, samples were stored in closed vials and in the dark at both 20 ºC and 4 ºC after production and lyophilization. At predetermined times (1, 3 Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 82 and 6 months), formulations were characterized regarding mean hydrodynamic diameter and zeta potential after redispersion in liquid, and the insulin structure assessed by FTIR and farUV CD as described previously. 2.14. Statistical analysis One-way ANOVA was used to investigate the differences between the formulations and controls. Post hoc comparisons were performed according to Tukey’s HSD test (p0.05 was accepted as significant different) using Prism 6.02 software (GraphPad Software, Inc., CA, USA). Figure 4.1 Mean hydrodynamic diameter, polydispersity index (PdI) and zeta potential of SOL (black bars and squares), F68 (dark grey bars and triangles), F108 (medium grey bars and squares) and F127 (light grey bars and triangles) based empty, containing just PBA (empty:PBA), insulin-loaded (Mic:Ins) and insulin-loaded containing PBA (Mic:Ins:PBA) lyophilized micelles after dispersion in water (mean ± SD, n≥3). * p<0.05 compared to the liquid micelles. Chapter 4 I Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization _______________________________________________________________________________________ 83 3. Results 3.1. Determination of size and zeta potential of formulations Nanocomposites were dispersed in water and the size and surface charge of redispersed micelles analyzed to study the effect of lyophilization on their characteristics. Results expressed in Figure 4.1 showed an increase in the size of micelles, being this increase not significant for the majority of formulations. Also, no changes on the surface charge of micelles were observed (p>0.05). 3.2. Thermal analysis DSC thermograms of insulin, polymers, PBA, physical mixtures and micelles with a polymer:insulin ratio of 10:1 are presented in Figure 4.2. Insulin thermogram showed a board endothermic peak at 94.16 ºC corresponding to the glass transition and denaturation of insulin and, at some extent, water lost (218). After 230 ºC a group of peaks can be detected, as a result of the degradation of the protein. PBA presented a sharp endothermic peak at 221.26 ºC, as a result of its melting. SOL presented a board endothermic peak at 69.2 ºC corresponding to the glass transition of the polymer. Thermograms of F68, F108 and F127 presented melting endothermic peaks at 46.89 ºC, 58.33 ºC and 50.84 ºC, respectively. In both physical mixtures and micelles the peaks corresponding to insulin glass transition and denaturation and melting of PBA are not detected, while the peaks of Pluronic® copolymers shifted to 51.43 ºC, 52.38 ºC and 47.62 ºC for F68, 57.3 ºC, 57.12 ºC and 51.77 ºC for F108, and 51.51 ºC, 57.62 ºC and 49.23 ºC for F127, in physical mixtures, insulin-loaded micelles and insulin-loaded micelles containing PBA, respectively. Self-assembled polymeric micelles as powders for pulmonary administration of insulin __________________________________________________________________________ 84 Figure 4.2 DSC thermograms of raw materials, polymer insulin physical mixture, insulin-loaded (polymer:Ins) and insulin-loaded lyophilized micelles containing PBA (polymer:Ins:PBA) of SOL (A), F68 (B), F108 (C), and F127 (D). 3.3. XRD analysis The X-ray diffractograms of the pure compounds and lyophilized micelles with a polymer:insulin ratio of 10:1 are depicted in Figure 4.3. All samples presented two small peaks at around 2θ of 16.30º and 42.90º derivate from the polystyrene films. The crystalline nature of PBA was confirmed by the numerous sharp peaks between 2θ of 10º and 30º. Two main peaks at 2θ of 19.13º and 23.27-23.32º also indicated that Pluronic® copolymers possess a crystalline nature. On the other hand, the absence of distinct peaks in SOL spectra indicates its amorphous nature. Insulin presented few small peaks between 2θ 2-10º indicating a low degree of crystallization. Regarding micelles, distinct peaks at 2θ of 27.4º, 31.7º, 45.54º, 53.9º and 56.5º deriving from the NaCl existent in the PBS used to produce the formulations, can be detected. On the contrary, the peaks of PBA as well as the ones Chapter 4 I Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization ________________________________________________________________________________ 91 peaks and a shift in the positive peak, indicating a slight loss in the secondary structure of the protein. Figure 4.6 far-UV CD spectra of insulin-loaded (polymer:Ins) and insulin-loaded containing PBA (polymer:Ins:PBA) lyophilized micelles of SOL (A), F68 (B), F108 (C), and F127 (D). 3.7. Morphology and particle size distribution of powders The shape and morphology of lyophilized formulations was analyzed by SEM. Micrographs of formulations are presented in Figure 4.7, evidencing the presence of a mixture of needle-shape and plate-shape structures resembling parts of an incomplete polymeric network. The incorporation of PBA to the systems did not affect the morphology and structure of powders. It can be noted the presence of pores and spherical nanocomposites smaller than 5 µm attached to the surface of bigger structures. SOLbased formulations presented smaller and more needle-shape structures, while F68 seemed to produce more compact and bigger particles, explaining the differences observed in the particle size distribution of powders (Table 4.3). Self-assembled polymeric micelles as powders for pulmonary administration of insulin _______________________________________________________________________ 92 Figure 4.7 SEM micrographs of insulin-loaded formulations composed of SOL (A), F68 (B), F108 (C), and F127 (D), without (top panel) or with (bottom panel) PBA. Scale bar: 400 µm in formulations without PBA and 100 µm in formulations with PBA. As seen in Table 4.4, all the formulations presented dae smaller than the geometric diameter due to their low densities (p < 0.22 g/cm3, data not shown). SOL led to the formation of powder particles with lower D0.5 and dae, while F68 and F108-based Chapter 4 I Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization ________________________________________________________________________________ 93 formulations showed higher amount of larger particles/aggregates (D0.9). Both F68 and F108 formulations present higher CMC and size of micelles at 25ºC before lyophilization, which could be influencing the final particle size and percentage of aggregates of powders. On the other hand, SOL and F127 could be originating more compact and stable micelles that suffer lower degree of aggregation during lyophilization. Nevertheless, no significant differences were observed between different samples (p>0.05). With the exception of F68-based particles, all the formulations presented D0.5 values lower than 25µm and dae values lower than 6 µm. All formulations presented high Carr’s index (≥ 26) and Hausner ratio values (≥ 1.35), predicting poor flowability of powders according to European Pharmacopoeia (2.9.36. Powder flow) (455). Table 4.4 Particle size distribution over the volume, aerodynamic diameter, Carr’s index, and Hausner ratio of the different insulin-based formulations. Results are presented as mean values ± SD (n=3). Sample D0.1 (µm) D0.5 (µm) D0.9 (µm) dae (µm) Carr's index Hausner ratio SOL:Ins 4.5±0.6 15.3±0.1 44.3±1.9 3.1±0.2 49.2±11.3 2.03±0.45 SOL:Ins:PBA 4.6±0.9 16.0±2.6 42.8±6.9 2.9±0.4 51.1±1.9 2.05±0.08 F68:Ins 9.5±2.3 39.1±11.3 328.9±127.8 13.8±4.6 44.4±19.3 2.00±0.87 F68:Ins:PBA 8.7±0.3 29.7±0.1 410.9±314.1 8.1±0.3 50.0±7.1 2.03±0.29 F108:Ins 6.5±0.5 23.4±2.8 358.3±304.5 5.8±0.2 55.7±5.2 2.28±0.25 F108:Ins:PBA 6.9±0.3 23.3±1.3 235.9±153.0 5.4±0.8 49.2±8.0 2.00±0.29 F127:Ins 6.7±1.1 21.6±3.6 81.9±31.0 5.6±0.8 51.9±3.2 2.08±0.14 F127:Ins:PBA 7.4±0.3 24.2±1.3 68.3±7.2 5.5±0.1 47.7±9.3 1.95±0.33 3.8. Deposition profile of formulations The aerosolization properties of formulations were assessed in vitro following pharmacopeial instructions using an ACI impactor and a Rotahaler® as inhaler device and are presented in Table 4.5. Powders composed by SOL and F127 presented the higher FPF (around 48% and 44%, respectively) while F68 and F108 showed lower FPF (around 27% and 26%, respectively). Expecting for F108-based powders, the presence of PBA did not affect the FPF of powders (p>0.05). All the formulations showed a MMDA lower than 6.6 µm and GSD under 2.1 µm. 94 Table 4.5 Deposition profile of formulation powders after aerosolization into an Andersen Cascade Impactor via a Rotahaler® and estimation of mass median aerodynamic diameter (MMDA) and geometrical standard deviation (GSD). The results of aerosolization profile and fine particle fraction (FPF) are expressed as the amount of particles deposited in each stage as a percentage of the initial amount of particles, and the results of MMAD expressed as size in micrometers (mean ± SD, n=3). T+C+D is composed by throat, capsule and inhaler device. Stage Size (µm) SOL:Ins SOL:Ins: PBA F68: Ins F68:Ins: PBA F108: Ins F108:Ins: PBA F127: Ins F127:Ins: PBA T+C+D > 10 6.2±10.7 10.8±9.4 12.7±20.6 8.4±8.2 20.6±6.9 0.6±1.0 1.0±1.0 8.0±7.7 0 9.0-10.0 23.5±8.4 19.6±4.8 29.7±7.9 25.3±5.6 21.7±7.6 20.0±8.4 20.2±3.8 30.9±6.9 1 5.8-9.0 8.4±1.4 11.5±2.0 13.6±2.4 18.4±0.6 20.5±2.4 21.0±3.0 22.0±3.9 20.3±4.3 2 4.7-5.8 18.0±2.8 20.8±5.4 16.9±4.7 14.7±1.8 11.3±1.7 19.8±1.7 14.6±2.1 11.0±7.1 3 3.3-4.7 18.7±5.0 19.4±3.1 17.0±3.1 15.3±4.5 10.9±0.5 17.7±3.5 14.1±1.8 13.3±1.8 4 2.1-3.3 3.9±2.1 1.3±2.2 4.1±3.7 5.5±2.5 3.4±1.5 3.8±0.2 4.6±2.3 1.2±1.1 5 1.1-2.1 7.2±6.4 5.3±3.2 3.0±5.2 8.3±8.0 4.6±2.7 7.7±2.2 8.1±5.4 6.8±4.2 6 0.65-1.1 7.8±5.2 6.1±9.0 1.2±2.1 4.7±3.1 2.2±2.0 9.5±1.6 10.4±4.1 2.8±1.1 7 0.43-0.65 10.2±2.5 8.6±8.2 1.7±1.7 1.5±2.6 4.7±5.4 5.1±1.0 6.9±2.9 7.0±1.3 FPF <4.7 47.8±15.8 40.6±16.6 27.1±12.0 35.3±7.8 25.8±1.3 43.9±3.1 44.0±10.3 31.2±6.5 MMAD 4.8±0.7 4.9±0.5 5.8±0.2 5.6±0.2 6.1±0.3 5.1±0.3 5.1±0.6 6.6±1.1 GSD 1.8±0.2 1.7±0.7 1.9±0.5 1.7±0.3 1.9±0.3 1.5±0.0 2.1±0.2 2.1±0.5 Chapter 4 I Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization ________________________________________________________________________________ 95 The presence of PBA did not affect the MMAD and GSD values of formulations (p>0.05). No differences (p>0.05) in the MMAD and GSD values were observed between the different powders. 3.9. Determination of the insulin release pattern from micelles Release studies of insulin from micelles were performed at physiological pH and temperature (pH 7.4 and 37 °C) in the absence or presence of glucose (1.2 mM). Results of in vitro release studies are reported in Figure 4.8 as percentage of protein released over time. Insulin release presented a biphasic pattern, with notorious burst release in the first 15 minutes followed by a sustained release of the protein over the following 24 hours. Both F68:Ins and F68:PBA:Ins formulations released around 85-95 % of insulin in the absence of glucose (Figure 4.8A), presenting similar release profiles (f2>50, Table 4.5). SOL:Ins and SOL:PBA:Ins released 40-55 % and 50-65 % of the total insulin, respectively. The high difference in the percentage of insulin released between SOL and F68-based formulations could be related to the MW of polymers and the structure of micelles. Having higher MW and lower CMC, SOL could present more compacted micelles which difficult the release of insulin. The presence of PBA seemed to affect specially F108 and F127-based formulations since, in both cases, the totality of insulin was released from PBA containing micelles (f2>50), whereas only 80 % and 60 % of the protein was release after 24 hours from F108:Ins and F127:Ins formulations, respectively. Despite the different percentage of total insulin release, F108:Ins and F127:Ins also presented similar release profiles (f2>50). Figure 4.8 In vitro release profiles of insulin from different formulations in PBS (pH 7.4) without glucose (A) and with 1.2 mM glucose (B). Results are presented as mean ± SD (n=3). Self-assembled polymeric micelles as powders for pulmonary administration of insulin _______________________________________________________________________ 96 The presence of glucose did not affect the release of insulin (f2>50) from formulations (Figure 4.8B and Table 4.6), excepting for SOL:Ins formulations (f2<50). Although the presence of PBA did not confer glucose-sensing properties to the formulations, it seemed to increase the release of insulin for the majority of formulations by a mechanism independent of glucose concentration. Table 4.6 Similarity factor (f2) values between insulin release profiles of the different formulations in PBS (pH 7.4) without glucose (white columns) and with 1.2 mM glucose (grey columns). SOL: Ins SOL:Ins: PBA F68: Ins F68:Ins: PBA F108: Ins F108:Ins: PBA F127: Ins F127:Ins: PBA SOL:Ins --- 44.1 31.1 29.5 42.1 24.2 48.4 21.2 SOL:Ins: PBA 48.0 --- 22.0 20.9 62.6 17.1 62.6 14.9 F68:Ins 20.0 25.3 --- 72.9 21.1 48.3 23.8 42.2 F68:Ins :PBA 19.1 24.3 68.7 --- 20.1 54.1 22.7 45.8 F108:Ins 35.6 44.2 24.6 23.9 --- 16.6 58.0 14.4 F108:Ins: PBA 13.3 17.2 41.7 43.9 17.4 --- 18.6 58.4 F127:Ins 45.9 60.8 25.3 24.3 52.6 17.3 --- 16.4 F127:Ins: PBA 12.6 16.2 38.9 40.7 16.4 71.9 16.4 --- 3.10. Stability of formulations upon storage In order to assess the stability of the lyophilized formulations upon storage, samples of each formulation were produced and stored at two different temperatures, namely 20 ºC and 4 ºC, and characterized after 1, 3 and 6 months. Results regarding mean hydrodynamic diameter and surface charge of micelles after the dispersion of powders in water are presented in Figure 4.9 and 4.10. The results of the majority of samples stored for 1 month were similar to the ones obtained for redispersed powders after lyophilization (p>0.05), excepting for F68-based insulin-loaded micelles containing PBA and F108- Chapter 4 I Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization ________________________________________________________________________________ 97 based micelles that presented higher micelles size when stored at 20 ºC and analyzed at 25 ºC (p<0.05). Figure 4.9 Mean hydrodynamic diameter, polydispersity index (PdI) and zeta potential of SOL (A) and F68 (B)-based lyophilized insulin-loaded (Mic:Ins) and insulin-loaded containing PBA (Mic:Ins:PBA) micelles stored for 1 month (black bars and squares), 3 months (medium grey bars and squares), and 6 months (light grey bars and squares) at 4 ºC and 20 ºC after redispersion in water (mean ± SD, n=3). * p<0.05 compared to the formulations after lyophilization, ** p<0.05 between different months of storage, *** p<0.05 between 20 ºC and 4 ºC. Neither time nor temperature of storage affected in a high extension the characteristics of SOL and F127-based micelles (p>0.05, for almost all formulations). On the other hand, F68 and F108-based micelles stored at 20 ºC seems to be more sensitive to the time and temperature of storage, since some formulations stored at 4 ºC presented smaller mean hydrodynamic diameter (p<0.05), and had variable size upon storage. The zeta potential of the formulations remained slightly negative as at liquid state and recently lyophilized samples. Self-assembled polymeric micelles as powders for pulmonary administration of insulin _______________________________________________________________________ 98 Figure 4.10 Mean hydrodynamic diameter, polydispersity index (PdI) and zeta potential of F108 (A) and F127 (B)-based lyophilized insulin-loaded (Mic:Ins) and insulin-loaded containing PBA (Mic:Ins:PBA) micelles stored for 1 month (black bars and squares), 3 months (medium grey bars and squares), and 6 months (light grey bars and squares) at 4 ºC and 20 ºC after redispersion in water (mean ± SD, n=3). * p<0.05 compared to the formulations after lyophilization, ** p<0.05 between different months of storage, *** p<0.05 between 20 ºC and 4 ºC. Regarding the protein conformation, FTIR spectra of the area-normalized secondderivative amide I region of native insulin without storage and lyophilized insulin-loaded micelles and insulin-loaded micelles containing PBA stored up to 6 months at both 4 ºC and 20 ºC are presented in Figure 4.11 and 4.12, respectively. Chapter 4 I Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization ________________________________________________________________________________ 99 Figure 4.11 Area-normalized second-derivative amide I spectra of insulin solution 30 mg/mL and insulin-loaded micelles (polymer:ins) after lyophilization (t0) and upon 1 month (t1), 3 months (t3) and 6 months (t6) of storage at 4 ºC and 20 ºC. Self-assembled polymeric micelles as powders for pulmonary administration of insulin _______________________________________________________________________ 100 AO and SCC values depicted in Table 4.7 show that, similarly to micelles after production and lyophilization, SOL-based micelles presented the lower percentage of native-like insulin conformation. After 6 months of storage, both AO and SCC values were lower (p<0.05) than the ones obtained after production and lyophilization, expecting for F127 insulin-loaded micelles stored at both temperatures as well as F108 insulin-loaded micelles, and SOL and F127 insulin-loaded micelles containing PBA stored at 4 ºC (p>0.05). The storage temperature did not seemed to affect extensively the secondary conformation of insulin, since differences (p<0.05) in AO values were only observed for F68 and F127 insulin-loaded micelles containing PBA, while for SCC values were observed for F127 insulin-loaded micelles containing PBA. In addition, the presence of PBA only affected the AO values of SOL and F68 insulin-loaded micelles stored at 4 and 20 ºC, respectively; the SCC values of SOL and F108 insulin-loaded micelles stored at 4 ºC and the SCC values of F68 insulin-loaded micelles stored at 20 ºC. In Table 4.8 are presented the values related to the percentage of reduction in the AO and SCC values of freeze-dried micelles after 6 months of storage. It is possible to notice that SOL-based micelles suffered the higher reduction in the conformation of insulin upon storage with a decrease of 15.0 and 22.7% in the AO and SCC values, respectively. Regarding the spectra it was noticed that, for SOL insulin-loaded micelles, at 4 ºC insulin structure changed from a random coil organization at t0 to a dominant low-frequency βsheet assignment at 1616 cm-1 after 1 month of storage, while at 20 ºC this modification was just observed after 6 months. For SOL insulin-loaded micelles containing PBA a similar pattern of insulin structural modifications was observed. At 4 ºC the modification of random coil structure into a low-frequency assignment after 1 month was observed, but after 3 and 6 months was observed a tendency to an organization of both random coil and β-sheet. At 20 ºC this latter structural organization of a random coil and β-sheet bands was observed since 1 month until 6 months of storage. Considering the Pluronic® formulations, just a few changes were observed in the α-helix band intensity, which seemed to increase during time up to 6 months of storage, both for 4ºC and 20ºC storage conditions. However, a band peak at 1600 cm-1 appeared after 1 month of storage in all the Pluronic®-based micelles containing PBA. Chapter 4 I Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization ________________________________________________________________________________ 107 to the formation of the observed structures. Similar network-like structure was observed for freeze-dried PCL–PEG–PCL micelles (< 6% w/v) (112). Size, density and shape of particles are the main characteristics affecting the aerosolization and deposition properties of particles (449, 477). The theoretical aerodynamic diameter (dae) of powder particles was calculated using a dynamic shape factor of 1.6 as the mean value described by Hassan and Lau (2009) (477) for needleshape (χ=1.7) and plate-shape (χ=1.5) particles (Table 4.3). For many years it was assumed that aerosolized dry particles should present a size ranging 1-5 µm for efficient lung deposition. Nowadays is accepted that large porous particles presenting small particle mass density (p < 0.4 g/cm3) and geometric size above 5 µm can be properly released from inhalers and reach the deep lung (449). Despite presenting the same aerodynamic diameter, large porous particles have a lower surface-to-volume ratio compared to small nonporous particles, thus aggregating less and behaving more as single entities during aerosolization. Additionally due to its high geometric size they could avoid more easily phagocytosis by macrophages (301). For example, insulinencapsulated PLGA/cyclodextrin microspheres presenting geometrical diameter similar to our particles (D0.5 of 26 µm) were able to reach the deep lung and promote a significant hypoglycemic effect when tested in vivo in rodents (24). Due to its shape and elevated Carr’s index (≥ 26) and Hausner ratio (≥ 1.35) values, formulations could present some limitations regarding flowability and by that, lung deposition. Despite being extensively used to predict the quality of powders regarding flowability and deposition; precaution must be taken during the analysis of Carr’s index and Hausner ratio, since for some powders a direct correlation between lower Carr’s index and Hausner ratio values and higher FPF was not observed (477, 478). Additionally, nonspherical particles, especially fibers, could present some dispersion limitations due to the attractive forces existent between adjacent particles (269). Nevertheless, in this study good aerosolization properties were obtained as observed by the low MMAD and high FPF fractions, in spite of the low theoretical flowability. SOL and F127 presented the higher FPF and best aerosolization properties (Table 4.5) correlating to the particle size distribution determined by laser diffraction, were both formulations presented the smallest mean diameter values and absence of large size aggregates. The lower CMC values of SOL and F127 originated more compact, smaller micelles that could lead to small powder particles. The FPF of insulin dry powders varies according to the formulation, the technique and the parameters used to assess it, which turns the comparison between formulations a rather intricate task. A formulation of freeze-dried insulin with lactose as coarse carrier showed a FPF of ~52% at a flow rate of 60 L/min Self-assembled polymeric micelles as powders for pulmonary administration of insulin _______________________________________________________________________ 108 (479), while spore like particles showed FPF >69% (480), and Exubera® a FPF of 33-45% (481). Unlike many authors that take stage 2 (corresponds to the pharynx with a cutoff diameter of 5.8 µm) and lower into account for the calculation of FPF (482-484), in this work only particles that are able to deposit in the trachea (stage 3 with cutoff diameter of 4.7 µm) or lower airways and alveoli were considered. The pharmacological efficacy and clinical outcome of aerosol formulations is related to its deposition profile in the lungs (449), thus higher amounts of insulin are expected to reach the lungs and become available to be absorbed in SOL and F127-based formulations. Future improvements in the efficiency of the formulations could be explored by screening an inhaler device that best fits the characteristics of the developed powders, since the inhaler device has shown to influence the performance of the formulations (280). Additionally, the incorporation of an appropriate coarse carrier to the formulation could improve the deposition of particles in the deep lung by reducing the particle to particle interaction and aggregation of nanocomposites, as well as their electrostatic interaction with the capsule and the inhaler device (283, 485). Release studies of insulin from micelles were performed in the absence or presence of 1.2 mM glucose. The concentration of glucose used was based on determinations of lung glucose levels performed in diabetic patients without lung disease (486). Insulin release profiles followed a biphasic pattern (Figure 4.8) as reported by others (333). F68 and F108 showed the faster and higher release of insulin, while SOL and F127 presented a more sustained release of insulin over time, which is related to the higher stability upon dilution of micelles composed of polymers with lower CMC. Therefore, the former could be explored as formulations of rapid-acting insulin, while the latter ones as long-acting insulin powders. Furthermore, with a proper mixture of different micelles a formulation with both post-prandial and long-acting effects could be achieved, which holds an advantage over formulations that only present fast-acting or prolonged released of insulin (334, 481, 487). PBA did not confer glucose-sensitive properties to formulations but promoted a faster in vitro release of insulin from the micelles. At pH 7.4, PBA presents mainly neutral hydrophobic moieties (433) (~96 % estimated by in silico simulation using the Marvin Suite software from ChemAxon, Hungary), thus potentially being present in the inner core of the micelles as confirmed by XPS analysis, not being able to react with the glucose present in liquid media. Nevertheless, the small percentage of ionic species might increase the hydrophilicity of micelles, promoting some destabilization and a faster release of insulin when compared to the formulations without PBA, even in the absence of glucose. Grafting PBA to the hydrophilic segments of the polymers could ensure the presence PBA at the surface of micelles and provide them with the desired glucose- Chapter 4 I Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization ________________________________________________________________________________ 109 sensitive properties (436). Also, a covalent modification of insulin as recently performed by Chou and co-workers (488) prior to its encapsulation into the micelles is other possible approach. Regarding SOL-based formulations, further studies are required to analyze deeply the effect of glucose concentration of the release properties of SOL and study the possible stimuli-responsive properties of the polymer in future applications. Lyophilized formulations were stored at two different temperatures and characterized in terms of size, surface charge and insulin conformation after 1, 3 and 6 months of storage (Figure 4.9 and 4.10). Excepting for some F68 and F108-based micelles for which time and storage temperature induced some degree of particle aggregation, formulations presented similar particle sizes after dispersion in water compared to recently lyophilized formulations. In addition, no changes in the surface charge of micelles were observed. Redispersed micelles presented particle size lower than 600 nm and neutral surface charge in all formulations. SOL and F127-based micelles presented the best results during the storage period studied. Regarding protein stability, storage seemed to induce, to a small extension, loss on insulin conformation in all formulations as seen by a decrease of AO and SCC values in stored samples (Table 4.7). The higher percentages of conformational loss were observed for insulin-loaded SOL micelles stored at 4ºC, with a reduction of 15.04±0.20% and 22.65±1.15% in AO and SCC values, respectively. The storage temperature did not show to influence to a high extent the secondary structure of insulin. In SOL-based micelles insulin structure was dominated by a random coil and lowfrequency β-sheet organization. Considering the Pluronic® formulations it was observed that no significant changes in insulin structure band positions was observed. Indeed, the α-helix band of insulin in all those formulations was maintained in its characteristic band range, maintaining also happened to its high and low β-sheets assignments. The FTIR spectra results for these formulations, justify in fact the high AO and SCC values obtained, since minor band position changes were observed. Pluronic® copolymers showed to enhance the conformational stability of different proteins against different processing methods and thermal stress, including salmon calcitonin (489), interleukin-1 (IL-1) receptor antagonist (490), which can be due to the PEG chains as already mentioned. Nevertheless, for PBA containing formulations although the α-helix and β-sheets highfrequency assignments did not drastically change over the 6 months of storage at both 4 ºC and 20 ºC, the appearance of a band peak at 1600 cm-1 suggested that PBA may affect insulin stability during storage. These results regarding PBA containing formulations were not corroborated with the far-UV CD experiments performed to the samples stored during 6 months (Figure 4.13), since no significant differences were observed between Self-assembled polymeric micelles as powders for pulmonary administration of insulin _______________________________________________________________________ 110 micelles with and without PBA. However, a close attention to the effect of PBA on insulin structure should be paid. Taking into account the results, lyophilization of insulin-loaded micelles may improve the storage shelf-life of the final product, since the acquisition of amorphous powders will improve the physical and chemical stability of formulations. As referred previously, in liquid state micelles tend to aggregate creating particles on micron size range. Also, powders for pulmonary administration are considered advantageous over liquid formulations for many reasons including higher stability. Regarding the protein, solutions of insulin shown to be more prone to instability due to hydrolytic activity of water, namely with the earlier loss of conformation upon storage (453). In addition, in lyophilized formulations insulin shown to be at amorphous state, which may improve its stability, since contrary to what happens with many proteins, insulin presents greater stability in amorphous state (491). This assumption is supported by the maintenance of high percentages of native-like conformation of insulin in lyophilized formulations upon storage compared to recently formulated samples, as well as taking into consideration the example of Exubera®, a spray-dried insulin formulation in amorphous state that shown to be stable when stored at room-temperature (481). 5. Conclusions Lyophilization of micelles allowed the production of powder nanocomposites that were easily redispersed when in contact to liquid media, originating micelles smaller than 300 nm and neutral charge at body temperature. This behavior can contribute for an increase in the bioavailability of insulin as micelles should prevent partially the clearance of micelles by alveolar macrophages. Insulin shown to be at amorphous state as evidenced by DSC, XRD and Raman spectroscopy and partially at the surface of micelles as evidenced by XPS. In addition, these formulations can be recovered by dispersion, in a manner of preserving the structure and potentially maintaining the activity of insulin. Lyophilized low-density powders composed by large-porous particles presented, in general, aerodynamic diameters compatible with good lung deposition patterns. Analysis of powder morphology shown that they are formed by a non-homogenous population of needle-shape particles mixed with a kind of plate-shape structures which could result in limited flowability of powders, as predicted by Carr’s index and Hausner ratio. However, the in vitro deposition profiles observed presumes interesting in vivo performance of some formulations. Chapter 4 I Micelle-based nanocomposites as solid formulations for pulmonary insulin delivery: design and characterization ________________________________________________________________________________ 111 PBA did not significantly affect the characteristics of formulations but promote a faster in vitro release of insulin from the micelles. However, it did not confer the formulations with glucose-sensitive properties. Formulations showed to be physically stable upon storage with minimal loss of the insulin’s native-like structure. Pluronic®-based formulations presented the best results regarding the maintenance of protein conformation analyzed by FTIR, however, far-UV CD studies showed that SOL-based micelles can also maintain a good amount the nativelike structure. In conclusion, powders formulations have shown promising results as delivery systems of inhaled proteins justifying further in vitro and in vivo assessment. Self-assembled polymeric micelles as powders for pulmonary administration of insulin _______________________________________________________________________ 112 Chapter 5 I In vitro biological assessment of powder formulations for inhalation of insulin ________________________________________________________________________________ 113 Chapter 5 In vitro biological assessment of powder formulations for inhalation of insulin The information presented in this chapter was partially published in the following publication: Fernanda Andrade, José das Neves, Petra Gener, Simó Schwartz Jr, Domingos Ferreira, Mireia Oliva, Bruno Sarmento, Biological assessment of self-assembled polymeric micelles for pulmonary administration of proteins, submitted for publication Self-assembled polymeric micelles as powders for pulmonary administration of insulin _______________________________________________________________________ 114 1. Introduction Biopharmaceuticals hold the potential for the treatment of numerous diseases, for many of which there is no current available cure provided by small molecule drugs (492). This explains the high demand for research and development of formulations based on biopharmaceuticals and the exponential increase in the approval and market share of these formulations that has been witnessed over the past few years. Since biopharmaceuticals are generally administered by injection, a variety of approaches to develop systems for non-invasive administration of such drugs have been proposed recently. For example, oral and pulmonary administration of insulin is being pursued by many researchers, and positive advances in the field have been achieved (326, 493-495). Inhalation appears as a promising non-invasive route for systemic delivery of biopharmaceuticals, reasoned by characteristics of the respiratory system, such as the high area of absorption and blood supply, and the absence of hepatic first-pass metabolism, thus allowing high bioavailability values compared to other non-invasive routes (213, 492). Additionally, reduced costs associated to the production, transport and administration, as well as higher stability and patient compliance are expected for solid formulations for inhalation as compared to parental administration (492). However, the complexity of inhalation and respiratory system (geometry, humidity, defense mechanisms) allied to the challenges posed by the development of biopharmaceuticalsbased inhaled formulations, are underneath the fact that only a very restricted number of these products has reached the market so far (213). Thus, several research groups have developed new and improved formulations based on advances observed in molecular biology and particle engineering technology. Nanocarriers, including polymeric micelles, have been proposed as advanced inhaled drug delivery systems with optimized pharmacokinetics and pharmacodynamics. Among other things, nanocarriers could increase the permeation of compounds through the epithelium, reduce the clearance by mucociliary escalator through the penetration of particles in the mucus as well as reduce the recognition of particles by alveolar macrophages (496). In this chapter, the suitability of powders developed in Chapter 4 as delivery systems for inhalation of proteins was assessed in vitro using pulmonary epithelial cell lines and macrophages. Chapter 5 I In vitro biological assessment of powder formulations for inhalation of insulin ________________________________________________________________________________ 115 2. Experimental 2.1. Materials SOL, F68, F108, and F127 were kindly provided by BASF (Ludwigshafen, Germany). Lyophilized human insulin (potency ≥ 27.5 units per mg), PBA, PBS, 3-(4,5Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), Triton X-100, βmercaptoethanol, phorbol 12-myristate 13-acetate (PMA), and 5-([4,6-dichlorotriazin-2-yl] amino)fluorescein hydrochloride (5-DTAF) were purchase from Sigma-Aldrich (St. Louis, MO, USA), while sodium bicarbonate and dimethyl sulfoxide (DMSO) from Merck KGaA (Darmstadt, Germany). The other reagents used were methanol, ethanol and acetic acid from analytical grade; acetonitrile and TFA of HPLC grade (Merck, Germany) and Type 1 ultrapure water (18.2 MΩ.cm at 25 ºC, Milli-Q®, Billerica, MA, USA). Dulbecco’s modified eagle medium (DMEM) supplemented with L-glutamine (DMEMGlutaMAX®), fetal bovine serum (FBS), non-essential amino acids, 10000 U/mL penicillin and 10000 µg/mL streptomycin, 0.25% Trypsin-EDTA were purchased from Gibco (Life Technologies Ltd., Paisley, UK). CellMask® DeepRed Plasma membrane Stain, 4′,6diamidino-2-phenylindole (DAPI) and ProLong® Gold Antifade Mountant were purchased from Molecular Probes (Life Technologies Ltd., Paisley, UK). RPMI-1640 was purchased from Lonza (Basel, Switzerland). 2.2. Production of micelles and lyophilization Micelles were prepared using the thin-film hydration technique. Briefly, each polymer was individually weight and dissolved in a mixture of methanol:ethanol (1:1). Then, the solvent was removed under vacuum and the film was left to dry overnight at room-temperature to eliminate any remained solvent. The film was then hydrated with PBS at 37 ºC in order to obtain a 1 % (w/v) solution and vortexed for 5 min. The obtained dispersion was filtered through a 0.22 µm syringe filter to remove possible dust and aggregates. PBA containing micelles were prepared by dissolving PBA with the polymers in the solvents prior to the production of the film at a ratio of 10:1 (w/w) (polymer:PBA). Insulinloaded micelles were prepared by hydrating the polymeric films with an insulin solution in PBS to obtain polymer:insulin ratios of 10:1 (w/w). The other steps were the same as for plain formulations. After production micelles were lyophilized in an AdVantage 2.0 BenchTop Freeze Dryer (SP Scientific, Warminster, PA, USA). The cycle used was the follow: the samples were Self-assembled polymeric micelles as powders for pulmonary administration of insulin _______________________________________________________________________ 116 frozen at -30 ˚C and the temperature maintained for 60 min, the primary drying was set at 20 ˚C for 480 min at 150 mTorr and the secondary drying for another 480 min at 30 ˚C and 100 mTorr. 2.3. Conjugation of polymers with 5-DTAF Polymers were fluorescently conjugated with 5-DTAF in an aqueous medium via nucleophilic aromatic substitution by an addition-elimination pathway as previously described (497). Briefly, a stock solution of 20 g/L 5-DTAF in DMSO was diluted in 0.1M sodium bicarbonate (pH 9.3) and added to a 6 % (w/v) polymer solution in 0.1M sodium bicarbonate (pH 9.3) to a final molar ratio of 1:2 (polymer:5-DTAF). The reaction proceeded overnight in the dark at room-temperature. The labelled polymer was purified from the excess of unreacted 5-DTAF by dialysis (12,000-14,000 MWCO Spectra/Por® membrane from Spectrum Europe BV, The Netherlands) against Type I ultrapure water. The dialyzed polymer solutions were lyophilized as described in the manuscript and stored in closed containers protected from light. A schematic representation of fluorescent-labeled polymers preparation is depicted in Figure 5.1. Figure 5.1 Reaction schematic for the conjugation of the polymers with 5-DTAF via nucleophilic aromatic substitution by an addition-elimination mechanism. At basic pH, the terminal hydroxyl group of PEG blocks presented in the polymers, attack the reactive moiety (2-amino-4,6-dichloro-striazine) on the 5-DTAF molecule, promoted by strong electron-withdrawing groups (N) of the striazine ring. 2.4. Production and characterization of fluorescent micelles Fluorescent-micelles were produced and lyophilized like empty micelles substituting polymers by 5-DTAF-conjugated polymers. Particle mean hydrodynamic diameter and PdI