IFMBE Proceedings

Volume 53

Series Editor Ratko Magjarevic

Deputy Editors Fatimah Binti Ibrahim Igor Lackovic´ Piotr Ładyzy˙ nski´ Emilio Sacristan Rock The International Federation for Medical and Biological Engineering, IFMBE, is a federation of national and transnational organizations representing internationally the interests of medical and biological engineering and sciences. The IFMBE is a non-profit organization fostering the creation, dissemination and application of medical and biological engineering knowledge and the management of technology for improved health and quality of life. Its activities include participation in the formulation of public policy and the dissemination of information through publications and forums. Within the field of medical, clinical, and biological engineering, IFMBE’s aims are to encourage research and the application of knowledge, and to disseminate information and promote collaboration. The objectives of the IFMBE are scientific, technological, literary, and educational.

The IFMBE is a WHO accredited NGO covering the full range of biomedical and clinical engineering, healthcare, healthcare technology and management. It is representing through its 60 member societies some 120.000 professionals involved in the various issues of improved health and health care delivery.

IFMBE Ofﬁcers President: Ratko Magjarevic, Vice-President: James Goh Past-President: Herbert Voigt Treasurer: Marc Nyssen, Secretary-General: Shankhar M. Krishnan http://www.ifmbe.org

More information about this series at http://www.springer.com/series/7403 Tomaž Jarm · Peter Kramar Editors

1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015)

Portorož, Slovenia, September 6–10, 2015

ABC Editors Tomaž Jarm Peter Kramar University of Ljubljana University of Ljubljana Faculty of Electrical Engineering Faculty of Electrical Engineering Ljubljana Ljubljana Slovenia Slovenia

ISSN 1680-0737 ISSN 1433-9277 (electronic) IFMBE Proceedings ISBN 978-981-287-816-8 ISBN978-981-287-817-5 (eBook) DOI 10.1007/978-981-287-817-5

Library of Congress Control Number: 2015949816

Springer Singapore Heidelberg New York Dordrecht London c Springer Science+Business Media Singapore 2016 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein orforany errors or omissions that may have been made.

The IFMBE Proceedings is an Ofﬁcial Publication of the International Federation for Medical and Biological Engineering (IFMBE)

Printed on acid-free paper Springer Science+Business Media Singapore Pte Ltd. is part of Springer Science+Business Media (www.springer.com) Preface

Welcome to the 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC2015), the first scientific congress with the goal to bring together researchers and users of electroporation-based and related technologies from academic and medical institutions and the industry. This meeting presents an excellent opportunity to meet with people from different disciplines possessing diverse expertise, who are involved in basic or applied research or are developing applications based on electroporation and the use of pulsed electric fields of high intensity. The aim of the WC2015 is to create an environment stimulating interdisciplinary interactions, to present latest achievements and to demonstrate current knowledge. To achieve this, the Congress will consist of plenary sessions given by distinguished invited lecturers, poster-viewing sessions, and parallel scientific sessions, interspersed with generous coffee and lunch breaks to foster lively discussion and exchange of ideas. More than 360 scientific contributions will be presented over four days by authors from nearly 50 countries from all six continents. The following main areas will be covered: • Application of pulsed electric fields technology in food: challenges and opportunities • Electrical impedance measurement for assessment of electroporation yield • Electrochemistry and electroporation • Electroporation meets electrostimulation • Electrotechnologies for food and biomass treatment • Food and biotechnology applications • In vitro electroporation - basic mechanisms • Interfacial behaviour of lipid-assemblies, membranes and cells in electric fields • Irreversible electroporation in clinical use • Medical applications: electrochemotherapy • Medical applications: gene therapy • Non-electric field-based physical methods inducing cell poration and enhanced molecule transfer • Non-thermal plasmas for food safety, environmental applications and medical treatments • PEF for the food industry: fundamentals and applications • PEF process integration - complex process chains and process combinations in the food industry • Predictable animal models • Pulsed electric fields and electroporation technologies in bioeconomy • Veterinary medical applications The WC2015 is also dedicated to encouraging young researchers to develop their skills and expertise in electroporation- related areas through the Young Investigator Competition. YIC is organised in three categories: Biomedical Engineer- ing (sponsored by IFMBE - the International Federation for Medical and Biological Engineering); Food, Biotechnology and Environment (sponsored by DIL - German Institute of Food Technologies); and Medicine and Biology (sponsored by Frank Reidy). Two satellite events will be incorporated in the WC2015, namely BFE2015 - the 3rd International Bio & Food Electrotech- nologies Symposium; and Bioelectrics 2015 - The 12th International Bioelectrics Symposium. WC2015 is organized by COST TD1104 Action (European network for development of electroporation-based technologies and treatments). COST TD1104 Action is a large international consortium connecting 569 researchers from 235 institutions and 42 countries. For more information please visit the COST TD1104 Action’s website at www.electroporation.net. For more information about the Congress visit wc2015.electroporation.net. We hope that you will enjoy the WC2015 and we wish you a pleasant stay in Portorož

Damijan Miklavciˇ cˇ The Chair of the COST TD1104 Action On behalf of the International Organising Committee and the Scientiﬁc Programme Committeee Satellite Events Incorporated in WC2015

BFE2015

The 3rd International Bio & Food Electrotechnologies Symposium

Bioelectrics 2015

The 12th International Bioelectrics Symposium 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food and Environmental Technologies (WC2015)

September 6-10, 2015 Portorož, Slovenia

Organisation Committees

International Organising Committee

Volker Heinz, Germany LluisM.Mir,France Richard Heller, USA Javier Raso, Spain Damijan Miklavciˇ c,ˇ Slovenia

Scientiﬁc Advisory Board

Gustavo V. Barbosa-Canovas, USA Boris Rubinsky, USA Dietrich Knorr, Germany Justin Teissié, France Eberhard Neumann, Germany Karl H. Schoenbach, USA Carlo Ricardo Rossi, Italy James C. Weaver, USA

Scientiﬁc Programme Committee

Hidenori Akiyama, Japan Guillermo Marshall, Argentina Sid Becker, New Zealand Olga Martín-Belloso, Spain Roman Buckow, Australia Mihaela G. Moisescu, Romania Maja Cemažar,ˇ Slovenia Marie-Pierre Rols, France Rafael V. Davalos, USA Gregor Serša, Slovenia David A. Dean, USA Declan Soden, Ireland Giovanna Ferrari, Italy Mounir Tarek, France Wolfgang Frey, Germany Stefan Töpﬂ, Germany Julie Gehl, Denmark P. Thomas Vernier, USA Mark J. Jaroszeski, USA Eugène Vorobiev, France Hao Lin, USA C.Y. Xiao, China James Lyng, Ireland Ken-Ichi Yano, Japan

Organising Committee (University of Ljubljana)

Tomaž Jarm, Slovenia Tadeja Forjanic,ˇ Slovenia Peter Kramar, Slovenia Saša Haberl Meglic,ˇ Slovenia Samo Mahnic-Kalamiza,ˇ Slovenia Alenka Macekˇ Lebar, Slovenia Damijan Miklavciˇ c,ˇ Slovenia Bor Kos, Slovenia Matej Kranjc, Slovenia and Lea Rems, Slovenia Tina Batista Napotnik, Slovenia Barbara Zorec, Slovenia Janja Dermol, Slovenia XII Committees

Young Investigator Competition Committee

Nikolai Lebovka, Ukraine Justin Teissié, France James Lyng, Ireland P. Thomas Vernier, USA Boris Rubinsky, USA

Scientiﬁc Reviewers

Oliver Bals, France Orjan Martinsen, Norway Farid Chemat, France Olga Martín-Belloso, Spain Maja Cemažar,ˇ Slovenia Damijan Miklavciˇ c,ˇ Slovenia Maria Laura Fernandez, Argentina Sea Min, Korea Giovanna Ferrari, Italy LluisM.Mir,France Wolfgang Frey, Germany Thomas Nacke, Germany Alexander Golberg, Israel Andrew Nelson, UK Muriel Golzio, France Gianpiero Pataro, Italy Nabil Grimi, France Uwe Pliquett, Germany Richard Heller, USA Javier Raso, Spain Antoni Ivorra, Spain Marie-Pierre Rols, France Nadica Ivosevic Denardis, Croatia Boris Rubinsky, USA Henry Jäger, Austria Martin Sack, Germany Tomaž Jarm, Slovenia Saulius Satkauskas, Lithuania Tony Jin, USA Gintautas Saulis, Lithuania Juergen Kolb, Germany Karl H. Schoenbach, USA Bor Kos, Slovenia Gregor Serša, Slovenia Igor Lackovic,´ Croatia Emanuela Signori, Italy Nikolai Lebovka, Ukraine Justin Teissié, France Hao Lin, USA Christian Theel, Germany James Lyng, Ireland P. Thomas Vernier, USA Felipe Horacio Maglietti, Argentina Eugène Vorobiev, France Faisal Mahmood, Denmark Ken-Ichi Yano, Japan Marija Marcan,ˇ Slovenia Günther Zeck, Germany Guillermo Marshall, Argentina Sponsors

Platinum Sponsors

• IGEA S.p.A. (Carpi, Italy) - http://www.igeamedical.com/ • OncoSec Medical, Inc. (San Diego, USA) - http://oncosec.com/

Gold Sponsors

• Elea Vertriebs- und Vermarktungsgesellschaft mbH (Quakenbrueck, Germany) - http://www.elea-technology.com • Electroblate, Inc. (Burlingame, USA) - http://www.electroblate.com/

Silver Sponsors

• Inovio Pharmaceuticals, Inc. (Plymouth Meeting, USA) - http://www.inovio.com/ • MaxCyte, Inc., (Gaithersburg, USA) - http://www.maxcyte.com/

Bronze Sponsors

• Arc Aroma Pure AB (Lund, Sweden) - http://www.arcaromapure.se/ • Bioelectrochemical Society - http://www.bioelectrochemical-soc.org/ • CNR - Institute of Translational Pharmacology (Rome ,Italy) - http://www.cnr.it • EnergyPulse Systems, Lda. (Lisboa-Carnide, Portugal) - http://www.energypulsesystems.pt • Pulsemaster BV (Bladel, The Netherlands) - http://www.pulsemaster.us/ • SFNano - French Society for Nanomedicine (Paris, France) - http://www.sfnano.fr/ • Transient Plasma Systems, Inc. (Torrance, USA) - http://www.transientplasmasystems.com/ • Trojan Technologies (London, Canada) - http://www.trojantechnologies.com/ • Virginia Tech Department of Biomedical Engineering and Mechanics (Blacksburg, USA) - http://www.beam.vt.edu/ Table of Contents

Part I: Invited Plenary Lectures

About the First Industrial Scale PEF – Plants and Heinz Doevenspeck’s Role – A Historical Review ...... 3 W. Sitzmann

Harnessing the Structure Modifying Potential of Pulsed Electric Fields (PEF) – Food Processing Examples in Product Stabilization, Process Acceleration and Compound Extraction ...... 7 J.G. Lyng, C. Arroyo, O. Cregenzán Alberti, D. Frontuto, C. Apel, N. Brunton, M. O’Sullivan, P. Whyte

Fundamental and Applied Aspects of Pulsed Electric Fields for Microbial Inactivation ...... 11 J. Raso

How Imaging Molecule Uptake into Cells can Reveal the Mechanisms of Membrane Electropermeabilization ...... 15 M.-P. Rols

Tissue Reactions to Electroporation and Electrochemotherapy: Vascular Effects that have Implications in Tumor Treatment ...... 20 G. Sersa, M. Cemažarˇ

Nanosecond Pulses and Beyond – Towards Antenna Applications ...... 24 Karl H. Schoenbach, Shu Xiao

Optimal Irreversible Electroporation Techniques in the Treatment of Locally Advanced Liver and Pancreatic Cancer ...... 28 Robert C.G. Martin

Electrotransfer of Antiangiogenic shRNA against Endoglin for Effective Cancer Treatment ...... 32 M. Cemažar,ˇ T. Dolinsek, N. Tesic, M. Stimac, G. Sersa

Abiotic Gene Transfer – A Rarity or a Ubiquity? ...... 36 Tadej Kotnik, James C. Weaver

Part II: Application of Pulsed Electric Fields Technology in Food: Challenges and Opportunities

Antimicrobial Effect of Stevia rebaudiana Bertoni against Listeria monocytogenes in a Beverage Processed by Pulsed Electric Fields (PEFs): Combined Effectiveness ...... 43 A. Rivas, S. Sansano, M.C. Pina Pérez, A. Martínez, D. Rodrigo

Pulsed Electric Field Technology Enhances Release of Anthocyanins from Grapes and Bioprotective Potential against Oxidative Stress ...... 47 S.Y.Leong,I.Oey,D.J.Burritt

Effects of Pulsed Electric Fields on Selected Quality Attributes of Beef Outside Flat (Biceps femoris) ...... 51 F. Faridnia, P. Bremer, D.J. Burritt, I. Oey XVI Table of Contents

Study on a Solid-State Pulse Generator Based on Magnetic Switch for Food Treatments by Pulsed Electric Field (PEF) ...... 55 Song Li, Jingming Gao, Martin Sack, Hanwu Yang, Baoliang Qian, Georg Mueller

Part III: Electrical Impedance Measurement for Assessment of Electroporation Yield Modeling Dynamic Electrical Impedance Spectroscopy Measurements on Electroporated Cells ...... 63 Tomás García-Sánchez, Antoine Azan, Isabelle Leray, Javier Rosell-Ferrer, Lluis M. Mir, Ramon Bragós Conductance Dynamics to Monitor and Control Cell Electropermeabilization ...... 67 J. Teissié Electric Field Mapping in ex vivo Anisotropic Muscle Tissue Using DT-MREIT ...... 71 W.C. Jeong, S.Z.K. Sajib, T.I. Oh, H.J. Kim, O.I. Kwon, E.J. Woo Electrorotation as a Versatile Tool to Estimate Dielectric Properties of Multi-scale Biological Samples: From Single Cell to Spheroid Analysis ...... 75 C.I. Trainito, E. Bayart, E. Bisceglia, F. Subra, O. Français, B. Le Piouﬂe Electrical Measurements for Monitoring Electroporation ...... 79 U. Pliquett Comparison of Single-Shot Rapid Acquisition with Relaxation Enhancement and Echo Planar Current Density MRI Sequences for Monitoring of Electric Pulse Delivery in Irreversible Electroporation ...... 83 I. Serša, F. Bajd, M. Kranjc, H. Busse, N. Garnov, R. Trampel, D. Miklavˇciˇc Micro Electro-permeabilization System for Cell Medium Conductivity Change Measurement of Erythrocytes Cells.... 87 L.C. Ramos, G.B. Pintarelli, D. Altenhofen, D.O.H. Suzuki Magnetic Resonance Electrical Impedance Tomography for Monitoring Electrical Conductivity during Delivery of Electric Pulses in Irreversible Electroporation ...... 91 M. Kranjc, I. Serša, D. Miklavˇciˇc

Part IV: Electrochemistry and Electroporation Mathematical Modeling of Electrochemical Phenomena at the Electrode-Solution Interface in a PEF Treatment Chamber ...... 97 G.Pataro,G.M.Barca,G.Ferrari Electrolytic Ablation Dose Planning Methodology ...... 101 E. Luján, H. Schinca, N. Olaiz, S. Urquiza, F.V. Molina, P. Turjanski, G. Marshall

Part V: Electroporation Meets Electrostimulation Electroporation and Electrostimulation of Blind Retina Using Micro-electrode Arrays ...... 107 G. Zeck, T. Herrmann

Part VI: Electrotechnologies for Food and Biomass Treatment High Voltage Electric Discharges Assisted Extraction of Stilbenes from Grape Stems ...... 113 S. Brianceau, X. Vitrac, M. Turk, E. Vorobiev Table of Contents XVII

Part VII: Food and Biotechnology Applications

Microbial Inactivation in a Non-commercial Juice of Mango and Papaya Submitted to Pulsed Electric Fields in Presence of a Stevia rebaudiana Bertoni Extract ...... 119 C.M. Belda-Galbis, A. Martínez, D. Rodrigo

Effects of Pulsed Electric Fields on Four Residual Fungicides in White Wines ...... 124 C. Delsart, C. Franc, N. Grimi, G. de Revel, E. Vorobiev, M. Mietton Peuchot

Control of Polyphenoloxidase and Peroxidase Activities in Mango and Papaya Juice by pulsed Electric Field and Stevia rebaudiana Bertoni Extract Combined Process ...... 128 M.N. Criado, A. Martínez, D. Rodrigo

Antimicrobial Capacity of a Cauliﬂower By-product Infusion Combined with the PEF Treatment against S. Typhimurium ...... 132 M. Sanz-Puig, L. Santos-Carvalho, L.M. Cunha, M.C. Pina-Pérez, D. Rodrigo, A. Martínez-López

Steviol Glycosides Stability after Pulsed Electric Technologies and Ultrasounds Treatments in Fruit Juice Blend Sweetened with Stevia rebaudiana ...... 136 M. Buniowska, J.M. Carbonell-Capella, A. Frígola, M.J. Esteve

Pulsed Electric Field Processing Optimization of Ascorbic Acid in a Mango and Papaya Beverage Sweetened with Stevia rebaudiana ...... 140 J.M. Carbonell-Capella, M. Buniowska, M.J. Esteve, A. Frígola

Part VIII: In vitro Electroporation – Basic Mechanisms

Study of Transmembrane Voltage Kinetics during 100μs Pulse Using Voltage Sensitive Dyes ...... 147 A. Silve, C. Poignard, M. Sack, R. Straessner, W. Frey

Electroporation of a Bladder Cancer Cell Line in Presence of Calcium: Efﬁcacy Dependence on Electric Field Strength and Calcium Concentration...... 151 S. Romeo, E.L. Hansen, S.K. Frandsen, J. Gehl

Extremely Low Frequency Electromagnetic Stimulation Alters Osteoblast Actin Filament Morphology ...... 155 A.-M. Bique, T. Keskinen, M. Paulasto-Kröckel

The Effect of Pulsed Electric Field on Mesenchymal Stem Cell Direct Migration ...... 159 K. Jezierska-Wo´zniak, J. Wojtkiewicz, Ł. Grabarczyk, M. Barczewska, A. Habich, S. Lipinski, W. Maksymowicz

Cell Sensitization is Induced by a Wide Range of Permeabilizing Electric Fields ...... 163 J. Dermol, O.N. Pakhomova, S. Xiao, A.G. Pakhomov, D. Miklavˇciˇc

Effect of Electrode Distance in Electrochemotherapy: From Numerical Model to in vitro Tests ...... 167 A. Ongaro, L.G. Campana, M. De Mattei, F. Dughiero, M. Forzan, A. Pellati, E. Sieni, C.R. Rossi

Numerical Analysis of Split Dose Protocols for nsPEF-Electroporation ...... 171 P. Lamberti, S. Romeo, M.R. Scarfì, V. Tucci, L. Zeni

The Effect of Temperature on Protein Extraction by Electroporation and on Bacterial Viability ...... 175 S. Haberl Megliˇc, E. Leviˇcnik, E. Luengo, J. Raso, D. Miklavˇciˇc XVIII Table of Contents

Effects of the Content of Cholesterol on the Permeability of Vesicles Membranes Induced by Pulsed Electric Fields ...... 179 Zhi-Wei Liu, Xin-An Zeng, Zhong Han

Electrostransfer of Plasmid gWIZ Blank into B16-F10 and TS/A Increase Expression of Cytosolic DNA PRRs ...... 183 K. Znidar, M. Bosnjak, L.C. Heller, M. Cemažarˇ

Glucose Derivatives as Efﬁcient Markers of Cell Reversible Electropermeabilization ...... 187 E. Raeisi, Y. Lemoigne, L.M. Mir

Effects of Nanosecond Pulsed Electric Fields on Cell-Cell Communication in a Monolayer ...... 192 A. Steuer, A. Schmidt, P. Babica, J.F. Kolb

Spheroids, a Three-Dimensional Preclinical Model for Electrochemotherapy and Gene Electrotransfer ...... 196 S. Kranjc, M. Cemažar,ˇ G. Sersa

Mixed Spheroids as a Relevant 3D Biological Tool to Understand Therapeutic Window of Electrochemotherapy ...... 200 L.Gibot,M.Madi,R.Vézinet,M.-P.Rols

Part IX: In vivo Electroporation – Basic Mechanisms

Modeling of in vivo Tissue Electroporation and Cellular Uptake Enhancement ...... 207 Bradley Boyd, Sid Becker

Dynamic Modeling of Electroporation for the Computation of the Electric Field Distribution Inside Biological Tissues during the Application of the Pulse Voltage ...... 211 L.M. Mir, C. Poignard, R. Scorretti, A. Silve, D. Voyer

Effects of Pulse Addition in Electropermeabilization: Theoretical Insights on the Electric Conductivity ...... 215 C. Suárez, A. Soba, F. Maglietti, N. Olaiz, G. Marshall

Antimetastatic Potential in Mice after Gene Therapy with Plasmid AMEP ...... 219 M. Bosnjak, U. Kamensek, A Sedlar, M. Cemažar,ˇ J. Zavrsnik, B. Turk, Celine Bouquet, G. Sersa

Incorporation of the Blood Vessel Wall into Electroporation Simulations ...... 223 L. Silve, R. Qasrawi, A. Ivorra

Design of an Applicator for nsPEF Exposure of Newborn Mice ...... 228 C. Merla, A. Pafﬁ, P. Monaco, T. Calderaro, F. Apollonio, C. Marino, P.T. Vernier, M. Liberti

Ex vivo Evaluation of Transdermal Drug Delivery by Means of Electroporation with in vivo Oriented Experimental Protocols ...... 232 B. Zorec, J. Jelenc, D. Miklavˇciˇc, N. Pavšelj

Part X: Interfacial Behaviour of Lipid-Assemblies, Membranes and Cells in Electric Fields

Sensitivity of Cells to Nanosecond Pulsed Electric Fields is Dependent on Membrane Lipid Microdomains ...... 239 Jody C. Ullery, Hope T. Beier, Bennett L. Ibey

Phase Transition to the Gel State Creates Long-Lived Electropores in Biomembranes ...... 243 S.V. Gomonov Table of Contents XIX

Part XI: Irreversible Electroporation in Clinical Use

Safety of Clinical Irreversible Electroporation ...... 249 K.R. Thomson, H. Kavnoudias, R.N. Neal II

Percutaneous and Intraoperative Irreversible Electroporation of Liver Cancer – A Monocenter Experience ...... 252 M. Moche, J. Fuchs, T.-O. Petersen, R. Jantschke, M. Bartels, T. Kahn, P. Voigt

Web-Based Tool for Patient-Speciﬁc Planning of Electroporation-Based Tumor Treatments in Different Tissue ...... 256 Marija Marˇcan, Denis Pavliha, Bor Kos, Tadeja Forjaniˇc, Gregor Serša, Damijan Miklavˇciˇc

Part XII: Medical Applications: Electrochemotherapy

Electrochemotherapy of Colorectal Liver Metastases – Trial Update ...... 263 I. Edhemovic, E. Brecelj, A. Ivanecz, G. Gasljevic, M. Marolt Music, T. Jarm, B. Kos, M. Bosnjak, M. Cemažar,ˇ D. Miklavˇciˇc, S. Potrc, E Gadzijev, G. Sersa

Treatment of Primary Liver Tumors with Electrochemotherapy – Clinical Trial ...... 267 M. Dokic, B. Trotovsek, V. Sojar, D. Stanisavljevic, R. Jansa, P. Popovic, M. Cemažar,ˇ D. Miklavˇciˇc,N. Pozar, P. Kavcic, A. Tomazic, M. Petric, G. Sersa

Technical Approach for Coupling Treatment Planning with Navigation Systems for Electroporation-Based Treatments ...... 271 B. Kos, A. Grošelj, M. Cemažar,ˇ J. Urbanˇciˇc, G. Kragelj, M. Bošnjak, B. Veberiˇc, P. Strojan, D. Miklavˇciˇc, G. Serša

Proper Patient and Treatment Parameters Selection for Electrochemotherapy of Deep Seated Head and Neck Tumors...... 275 A. Groselj, B. Kos, M. Cemažar,ˇ J. Urbanˇciˇc, G. Kragelj, M. Bosnjak, B. Veberic, T. Jarm, P. Strojan, D. Miklavˇciˇc, G. Sersa

Gemcitabine + Cisplatin Combination Electrochemotherapy for Triple Negative Breast Cancers: An in vitro Model Study ...... 280 R. Sundararajan, V. Raman, V. Masterson, S. Madhivanan, M. Raakesh, I.G. Camarillo

Prototype of a Flexible Grid Electrode to Treat Large Surfaces by Means of Electrochemotherapy...... 285 L.G. Campana, F. Dughiero, M. Forzan, C.R. Rossi, E. Sieni

Histological Characteristics of Soft Tissue Sarcomas Correlated to Electrical Resistance ...... 290 A.L. Tosi, L.G. Campana, F. Dughiero, M. Forzan, M. Rastrelli, C.R. Rossi, E. Sieni

3D Assessment of Irreversible Electroporation Treatments in Vegetal Models ...... 294 Q. Castellví, J. Banús, A. Ivorra

Modeling of Tumor Growth Inhibition in Mice Following Electrochemotherapy ...... 298 T. Forjaniˇc, D. Miklavˇciˇc

A Novel Sample Preparation Concept for Sepsis Diagnostics Using High Frequency Electric Fields ...... 302 K.J. Wassermann, T. Maier, F. Keplinger, J.R. Peham

Cryopreservation of Human Umbilical Stem Cells in Combination with Trehalose and Reversible Electroporation ..... 307 B. Dovgan, J. Dermol, A. Barliˇc, M. Kneževi´c, D. Miklavˇciˇc XX Table of Contents

Part XIII: Medical Applications: Gene Therapy

Constructing Clinically Applicable Plasmids for Cancer Gene Therapy ...... 313 U. Kamensek, N. Tesic, G. Sersa, M. Cemažarˇ

Evaluation of Smooth Muscle γ Actin Promoter Suitability for Tissue-speciﬁc Gene Delivery of Interleukin 12 ...... 317 N. Tesic, U. Kamensek, G. Sersa, M. Cemažarˇ

Utilization of Multi-array Electrodes for Delivery of Drugs and Genes in the Mouse Skin ...... 321 S. Kos, T. Blagus, M. Cemažar,ˇ J. Jelenc, G. Sersa

Melanoma Cell Viability is Reduced after Endoglin Silencing with Gene Electrotransfer ...... 325 T. Dolinsek, G. Sersa, M. Cemažarˇ

Part XIV: Non-electric Field Based Physical Methods Inducing Cell Poration and Enhanced Molecule Transfer

Custom Experimental Apparatus Design for the Investigation of Ultrasound as an Active Enhancement Method in Transdermal Drug Delivery ...... 331 J. Robertson, S. Becker

Magnetofection: An Effective, Selective and Feasible Non-viral Gene Delivery Method ...... 335 L. Prosen, M. Cemažar,ˇ G. Sersa

Emulating Exposures of Biological Samples to Lightning Strokes ...... 339 M. Reberšek, I. Marjanoviˇc, S. Beguš, F. Pillet, M.-P. Rols, D. Miklavˇciˇc, T. Kotnik

Mouse Melanoma (B16-F1) Cell Viability After Sonoporation in Acidic and CO2 Saturated Media ...... 343 J. Jelenc, J. Škafar, J. Jelenc, D. Miklavˇciˇc, A. Maˇcek Lebar

Part XV: Non-thermal Plasmas for Food Safety, Environmental Applications and Medical Treatments

Antibacterial Efﬁcacy of a Novel Plasma Reactor without an Applied Gas Flow ...... 349 C.M. Edelblute, M.A. Malik, L.C. Heller

Part XVI: PEF for the Food Industry: Fundamentals and Applications

Effect of Pulsed Electric Fields on Water Distribution in Apple Tissue as Monitored by NMR Relaxometry ...... 355 N. Dellarosa, L. Ragni, L. Laghi, U. Tylewicz, P. Rocculi, M. Dalla Rosa

Pulsed Electric Field Effects on Wine Yeast in Deﬁned Grape Juice Medium ...... 359 V. Kethireddy, P. Bremer, I. Oey

Improving the Extraction of Juice and Anthocyanin Compounds from Blueberry Fruits and Their By-products by Pulsed Electric Fields ...... 363 R. Bobinait˙e, G. Pataro, R. Raudonis, P. Vškelis, C.ˇ Bobinas, S. Šatkauskas, G. Ferrari

Electroeradication of Escherichia Coli is Under the Control of the Conductance of the Pulsing Buffer ...... 367 V. Blanckaert, A. Salles, M.L. Thomas, J. Teissié Table of Contents XXI

Part XVII: PEF Process Integration – Complex Process Chains and Process Combinations in the Food Industry Comparison of the Efﬁcacy of Pulsed Electric Fields Treatments in the Millisecond and Microsecond Range for the Extraction of Betanine from Red Beetroot ...... 375 E. Luengo, J.M. Martinez, I. Álvarez, J. Raso Extraction of Non-polar Molecules from Green Alga Chlorella vulgaris by Electroporation ...... 379 T. Elersek, A. Kapun, J. Golob, K. Flisar, D. Miklavˇciˇc Extraction of Sugar Solution from Sugar Beet Cossettes by Electroporation and Compressive Load ...... 384 Jan Gjörek, Karel Flisar, Damijan Miklavˇciˇc, Nataša Ulrih Poklar, Janvit Golob

Part XVIII: Predictable Animal Models Animal Models for Translational Cancer Immunotherapy Studies ...... 391 F. Riccardo, V. Rolih, F. Cavallo Gene Electro Transfer to Left Ventricular Myocardium in Rat and Porcine Models ...... 395 A.A. Bulysheva, B. Hargrave, N. Burcus, C.G. Lundberg, L. Murray, R. Heller Methodology for Assessing the Degree of Degeneration of the Porcine Intervertebral Discs Based on Magnetic Resonance Imaging ...... 399 Seweryn Lipi´nski, Katarzyna Jezierska-Wo´zniak, Aleksandra Habich, Monika Barczewska, Joanna Wojtkiewicz, Piotr Walczak, Wojciech Maksymowicz

Part XIX: Pulsed Electric Fields and Electroporation Technologies in Bioeconomy The Use of Pulsed Electric Fields for Protein Extraction from Nanochloropsis and Chlorella ...... 405 Mathilde Coustets, Justin Teissié A Microfluidic Device for the Real-Time Characterization of Lipid Producing Algae Cell Population Submitted to a Pulsed Electric Field ...... 409 P. Bodénès, F. Lopes, D. Pareau, O. Français, B. Le Pioufle Design, Optimization and Scale-Up of Pulsed Electric Field Technologies for Large Flow Applications in Heterogeneous Matrices ...... 414 S.R. Sarathy, C. Sheculski, J. Robinson, J. Walton, A. Soleymani, M.A. Kempkes, M.P.J. Gaudreau, D. Santoro Electro-biorefinery as a Potential Tool for Valorization of Mango and Papaya By-products...... 418 Oleksii Parniakov, Francisco J. Barba, Nabil Grimi, Nikolai Lebovka, Eugène Vorobiev

Part XX: Veterinary Medical Applications Electrochemotherapy in Veterinary Medicine – Our Clinical Experiences ...... 425 N. Tozon, G. Sersa, M. Cemažarˇ Electrochemogene Therapy as an Effective and Safe Treatment of Canine Cutaneous Mast Cell Tumors ...... 429 D. Pavlin, M. Cemažar,ˇ N. Tozon

Numerical Study for Needle Electrode in Treatment of Cutaneous Tumor Model ...... 433 J.A. Berkenbrock, M.M.M. Rangel, D.O.H. Suzuki XXII Table of Contents

Beneﬁts and Side-Effects of Electrochemotherapy in Veterinary Patients ...... 437 R.J. Lowe

In vitro and in vivo Evaluation of a Plasmid Encoding Canine Interleukin 12 ...... 441 U. Lampreht, U. Kamensek, M. Stimac, M. Bosnjak, S. Kranjc, G. Sersa, M. Cemažarˇ

Electrochemotherapy in Non-satisfactory Responding Tumors in Vet Patients: Combined Administration of Bleomycin, Systemic and Local ...... 445 F. Maglietti, M. Tellado, N. Olaiz, S. Michinski, G. Marshall

Author Index ...... 449

Keyword Index ...... 453

Part I Invited Plenary Lectures

About the First Industrial Scale PEF – Plants and Heinz Doevenspeck's Role – A Historical Review

W. Sitzmann Technical University of Hamburg Solids Process Engineering & Particle Technology, Hamburg, Germany

Abstract Inspired by the work of Sven Carlson, Heinz deals with defined homogeneous "pulsed electric fields, Helmut Doevenspeck, a German engineer, was starting in 1958 PEF". Their generation, application and impact on cell to develop his so-called "Elektroimpulsverfahren", which used membranes were reported for the first time by the German for the first time defined discharges of capacitors to generate engineer Heinz Doevenspeck who therefore can be declared homogeneous, pulsed electric fields. He applied these fields on intellectual father of the pulsed electric field technology. dispersed systems of inorganic or organic origin to influence the membranes of plant or animal cells or the surface charges There are several publications about the historical develop- of particles. Cracking of cells, growing and/or killing of micro- ment of PEF technology. Overviews are given by Sitzmann organisms, acceleration of fermentation processes and treat- /1/ and Töpfl /2/. The actual contribution is not a scientific ment of wastewater had been identified, described and pa- work as such. It describes Doevenspeck's own research tented by Doevenspeck as suitable applications already in the projects between 1958 and 1985 as far as they are known to early 1960s. Until mid of the 1980s, working as an independent the author and their common work from then on until Doe- consultant, he did many trials at technical scale plants without venspeck's death in 1993. reaching a breakthrough of his technology. From 1985 until 1993 he was cooperating with Krupp Company, Hamburg. In 1988 the first industrial scale plants were built and during this II. DOEVENSPECK`S PROJECTS period several applications and process mechanisms were FROM 1958 UNTIL 1983 identified, mathematically modeled and presented to the scientific public for the first time. Heinz Helmut Doevenspeck was born in 1917 in Minden, Germany, and died 1993 in Bremen, Germany. In Bremen Keywords Elektroimpulsverfahren, Doevenspeck, Pulsed he served his apprenticeship as a locksmith (AG Weser). Electric Fields, ELCRACK©, electroporation. After that he studied electrical engineering at the University of Applied Sciences in Bremen, working afterwards in his I. INTRODUCTION own office in Minden as a consulting engineer. Knowing about Sven Carlson's publications on the acceleration of More than a hundred years ago when electricity was first growth of microorganisms by applying electric current he used in industrial applications and shortly after the aware- started his own experiments in 1958. In his opinion Carl- ness rose that electric current can be used for warming up son's mistake was to change the physical properties of the fluid media, first methods were developed in order to kill treated materials in a way that particularly a targeted elec- germs in milk. Scientists experimented with continuous and tric field strength could not be adjusted during the whole alternating currents with different frequencies and ampli- duration of treatment. According to Doevenspeck defined tudes. Known as "Ohmic Heating", this process is still being portions of electric charges determined by the size of a used nowadays, for instance in the food industry. At the end capacitor have to be used to overcome this problem. From of the 1940s Flaumenbaum found out in the USSR that fruit this time on he was obsessed by the idea to influence sub- cells in mashed apples can be cracked by electric AC fields. stance systems by pulsed electric fields. Doevenspeck was This process became known as "Electroplasmolysis". In the an autodidact and not a scientifically educated person. As 1950s there were reports on abrupt capacitor discharges Fig. 1 (original document) shows he fiddled as an interested which lead to killing effects of microorganisms in fluids non-professional with physical, chemical and - in this case - and to cracking of composite materials. This process, then biological details to understand how the material systems he titled "Electrohydraulic Treatment", is nowadays used for was working with were constructed and how "his" pulses example by the German Institute of Food Technology in could impact these materials. He converted the basement of Quakenbrück, Germany, to change the structure of his one-family house in Minden to a professional laborato- ham. Contrary to the before mentioned applications of elec- ry, containing pulse generators, experimental equipment tricity the 1. World Congress on Electroporation however made of glass, microscopes, centrifuges, filters etc. etc.

© Springer Science+Business Media Singapore 2016 3 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_1 4 W. Sitzmann

Although Doevenspeck had no idea about the mechanism that damaged cell membranes irreversibly he described the effects rather well. Only during our close cooperation in the 1980s the author could confirm and visualize Doevenspeck's assumptions concerning cell destruction in PEF treated herring by producing pictures with a scanning electron microscope (Fig. 2).

Fig. 2 Herring cells before (left) and after (right) PEF /4/. 4.900 Fig.1 Doevenspeck`s studies on oilseeds fold magnification.

In spite of his limited financial and scientific possibilities In cooperation with the German company Baumgarten and his very empirical approach he early came to astonish- Fischindustrie in Bremerhaven, in the early 1960s Doevens- ing conclusions which he described 1960 in his first patent peck produced fishmeal and fish oil from poor stinking application /3/. In detail he identified that PEF shows the herring offal by cracking the cells with pulsed electric fields following advantages compared to existing technologies: before separating solids and liquids by using a screw press. - extensive suppression of electrolysis The original correspondence between Doevenspeck and his - wide prevention of temperature increase partners shows that the experts on site especially noticed - high profitability by low energy consumption that contrary to expectations the whole process was very - mild treatment of raw materials and preservation of low-odor. biological activity Also together with the German fishmeal producing facto- - killing of pathogenic germs ry Lohmann & Co. in Cuxhaven he experimented on fish processing. Fig. 3 displays an original flow diagram from He postulated in this patent application that surface 1964 drawn by Doevenspeck personally. One can see a charges of inorganic and organic substances can be influ- typical fishmeal production line as it is still being used no- enced by PEF. Thus unwanted ingredients in wastewater wadays. The only difference is the PEF system instead of a can be flocculated and separated and oil/water-emulsions conventional cooking system. Original reports and the can be divided into their components. Doevenspeck was correspondence between Lohmann and a participating feed living near to the coast, he was an enthusiastic sailor and producing company show that Doevenspeck's fishmeal was made contact to fish processing companies in his region. high quality and had a long shelf life. The vitamin A content There in the early 1960s he did several trials and found did not change essentially during six months storage. Addi- out that wastewater ("stickwater") could get rid of its sus- tionally there was almost no oxidation of the residual oil in pended proteins by means of electro flocculation during the final fish meal. Feeding experiments with laying hens PEF treatment. showed that "Doevenspeck" fishmeal produced from poor, More important however were his observations concerning low protein containing raw materials was much better than cell cracking. During his experiments he noted that one conventional fish meal from Peru in respect to feed conver- could get cell contents like oils or fats from animal raw mate- sion rate. Moreover independent experts certified that the rials. He described 1960 that when applying PEF muscle eggs produced by the above mentioned laying hens tasted fascicles contracted and cell containing liquids were set free. distinctively more palatable than the eggs from hens fed The same happened to oil containing deposit cells of fish. with conventional meal.

IFMBE Proceedings Vol. 53

About the First Industrial Scale PEF Plants and Heinz Doevenspeck's Role A Historical Review 5

Krupp Industrietechnik GmbH, a German company for plant construction and engineering, located in Hamburg, was very successful in selling this type of extraction plants for many years. In this critical phase Ernst Wilhelm Münch (Fig. 5) the area manager of the above mentioned extraction plants at Krupp company became acquainted with Doe- venspeck. Münch realized the possibilities of Doevens- peck's new technology which could be an alternative for the old (banned) solvent extraction process. In 1985 he engaged him as a consultant.

III. COOPERATION DOEVENSPECK - KRUPP UNTIL 1993 Fig. 3 Production line for fishmeal and -oil from herring (original diagram In the technical centre of Krupp company in Hamburg a from Doevenspeck, 1964). PEF pilot plant was erected and taken into operation in Doevenspeck realized that the impact of electric pulses 1985. The main component of this plant was Doevenspeck's on biological matter strongly correlates to the electric field 80 kW pulse generator (Fig.4). Frequencies in a range from strengths being applied. During experiments with E. coli 1 to 16.7 Hz and capacities of 5, 10, 15, 20, 25, 30 and bacteria in his own lab in 1962/63 he discovered that cul- 35 µF could be adjusted; the maximal charging voltage U0 tures being treated with field strengths below 3 kV/cm was 8 kV. seemed to increase their growth rates whereas those being treated with field strengths above this value showed reduced growth velocities. And very high field strengths obviously killed microorganisms. Motivated by these observations he extracted microorganism containing corn mash from a technical scale fermenter and applied electric pulses with a field strength of about 2,4 kV/cm. As a result he could measure a distinct acceleration of the fermentation process. In cooperation with the sewage plant operators of the German city of Nienburg, Doevenspeck achieved a 20% increase of methane yield by using PEF. 1967 he applied a patent together with a well-known big German brewery in Dort- mund. Subject of their common invention was a low temperature (25°C) conservation method for beer with the positive side effect of almost no changes of color and taste during the subsequent storage phase. At the end of the Fig. 4 Doevenspeck and his pulse generator 1970s and the beginning of the 1980s Doevenspeck was in (Krupp pilot plant, 1985) contact with the Institute of Biophysics of the University of Hannover (Glubrecht, Niemann) and the Institute of Micro- Krupp was mainly interested in gentle cell cracking of biology of the Hannover Medical School (Hülsheger, Potel) animal and herbal materials. During 1986 and 1990 the where his working hypothesis on germ killing was con- author's (Fig. 5) task as a developing engineer was to inves- firmed and scientifically examined for the first time. tigate different products with respect to their processability, Until 1983 Doevenspeck conducted many trials with in- to research the mechanism of action of PEF and to make the terested companies, communities and research institutes. He results available to a scientific public /5–7/. In order to took part in symposia and was able to inspire people being protect Doevenspeck's ideas several patents were applied in touch with him. But in the end with respect to a remarka- and Krupp company registered the trademarks ELCRACK© ble financial success of his new process he failed after 20 and ELSTERIL©. years of intense work. In Krupp's continuously operating pilot plant the material Doevenspeck's chance came up in the early 1980s due to throughput could be varied between 100 and 250 kg/h. The the ban of perchlorethylene as a solvent for the extraction of distances of the discharge electrodes were 50 and 70 mm oils and fats from animal raw materials in rendering plants. respectively.

IFMBE Proceedings Vol. 53

6 W. Sitzmann

recycling were new designed. Consequently there were many difficulties during the start-up phase which could not be solved within appropriate time. Forced by short-term economical success the clients decided to stop the commis- sioning. The plants had to be dismantled and replaced by conventional technology. These – preliminary – failures however did neither dis- courage Doevenspeck who continued his work (extraction of wool fat from sheep wool and cleaning the wastewater with PEF) in New Zealand until he died in 1993 nor the author himself who as a consultannt founded NaFuTec ` GmbH. In cooperation with TU Berlin where the author's Fig 5 Münch (left), Sitzmann (right, standing) and diploma student Stempel generator was placed he investigated until 1996 many possible applications. For example cell cracking of fruits in Leading figures especially in food engineering like Prof. order to increase the yield of juices or - in cooperation Barbosa-Canovas, Washington State University, and Prof. with a big German starch producing company - the cracking Knorr, Technical University of Berlin, were very interested of potato cells aiming at low residual water contents in the author's work and the lessons learned up to then on after the consecutive pressing step and at sterilizing the PEF technology. They visited Krupp's pilot plant in the late wastewater. 1980s. In the following years it was primarily their merit to initiate research projects and to disseminate the knowledge CONFLICT OF INTEREST on this new emerging technology worldwide. Based on the work which has been done at Krupp company PEF technol- The author declares that he has no conflict of interest. ogy was made available not only to a scientific but also to an industrial public. Main results of Krupp's R&D work REFERENCES were mathematical models for describing the discharge curves and the killing effects of germs depending on elec- 1. Sitzmann W (1995) High Voltage Pulse Techniques. New methods of food preservation. GW Gould, London, Graham & Hill trical parameters of the materials to be treated, a better un- 2. Töpfl S (2006) Pulsed Electric Fields (PEF) for Permeabilization of derstanding of the operating principle of pulsed electric Cell Membranes in Food and Bioprocessing Applications, Process fields and a way to quantify cell cracking effects by using and Equipment Design and Cost Analysis, doctoral thesis, TU Berlin the impedance of the materials as a measure. Krupp initiated 3. Doevenspeck H (1960) Verfahren und Vorrichtung zur Gewinnung der einzelnen Phasen aus dispersen Systemen. German patent DE in the years 1988 – 1990 together with the Technical Uni- 1237541 versity of Hamburg–Harburg, Germany, an EU–project on 4. Sitzmann, W (1995) Lecture, Convegno Internationale, Il Pesce: continuous sterilization of juices and milk in a pilot plant. Tecnologia, Sicurezza e Nutrizione dei Prodotti Ittici, Milano, Italy Within Grahl's doctoral thesis /8/ the influence of PEF on 5. Sitzmann W and Münch EW (1987) Verarbeitung tierischer Rohstoffe Fat Science Technology 9:368 food components like vitamins, enzymes and flavor carriers 6. Sitzmann W (1988) ELCRACK© und ELSTERIL©- Zwei neue was investigated for the first time. Two industrial scale Verfahren zur gezielten Beeinflussung von Phasengrenzflächen. GVC plants (a 10 t/h plant for producing rendered fats from Fachausschuss, Brussels, Belgium slaughterhouse byproducts in Duisburg, Germany, and a 22 7. Sitzmann W and Münch EW (1989) Elektrische Verfahren zur Keimabtötung. Die Ernährungsindustrie 6:54 - 58 t/h plant in Egersund, Norway, for low temperature produc- 8. Grahl T (1994) Abtöten von Mikroorganismen mit Hilfe elektrischer tion of high grade fishmeal and fish oil from herring) were Hochspannungsimpulse. Hamburg, TU Hamburg-Harburg, Germany engineered and commissioned in these years. Two more plants for processing slaughterhouse offal and bones were designed but never realized. Not only the ELCRACK©- Author: Prof. Dr. Werner Sitzmann Institute: Solids Process Engineering & Particle Technology system was brand new in these plants. Also the double Street: Denickestrasse 15 screw presses for liquid-solid-separation, the fluidized bed City: 21073 Hamburg drier for reduction of moisture in the press cakes and the Country: Germany ultrafiltration plant for wastewater cleaning and protein Email: [email protected]

IFMBE Proceedings Vol. 53

Harnessing the Structure Modifying Potential of Pulsed Electric Fields (PEF) – Food Processing Examples in Product Stabilization, Process Acceleration and Compound Extraction

J.G. Lyng, C. Arroyo, O. Cregenzán Alberti, D. Frontuto, C. Apel, N. Brunton, M. O’Sullivan, and P. Whyte Institute of Food and Health, University College Dublin, Belfield, Dublin 4, Ireland

Abstract Pulsed electric fields (PEF) can electroporate eu- the form of six case studies, three in the area of product karyotic and prokaryotic cells which can be used in food preservation/stabilization and three in the area of cell disin- processing for product preservation and extraction of intracel- tegration. For product stabilization, examples of the use of lular components. In this paper a number of potential PEF in a combined hurdle approach for vegetative cell inac- processing applications for PEF will be considered. These will tivation will be considered, though its potential for spore include its incorporation in hurdle preservation strategies, its use in the termination of enzyme hydrolysis of proteins, its inactivation in conjunction with high temperatures will also application in the waste stream valorization through enhanc- be presented. Product stabilization will also be explored in ing compound extraction and also its potential for accelerating the area of enzyme hydrolysis termination. In terms of cell processes such as meat tenderization or curing. In terms of disintegration, the paper will focus on examples of plant cell product pasteurization, the paired combinations of novel tech- disruption which can assist and accelerate the release of nologies examined certainly inactivated more microorganisms compounds which in turn can be exploited in the valoriza- than each technology alone but the effect of the combinations tion of plant waste streams. The application of PEF in the was largely additive or overlapping with few examples of syn- disruption of animal cells will also be considered with po- ergy. In addition some combinations introduced high specific tential applications in the acceleration of processes such as energy into products (in some cases higher than conventional UHT). In terms of product sterilization, PEF applied at tem- meat tenderization or curing. peratures of 120°C gave a slight though not significant increase in spore inactivation compared to thermal treatments. The II. CASE STUDY 1: PEF use of PEF for endoprotease inactivation in bioactive compound production also proved to be very promising and could IN HURDLE PRESERVATION have application in hydrolyses termination especially where PEF, high intensity ultrasound (US) and high intensity the resultant bioactives are heat labile. PEF pre-treatments of potato peel and brewers spent grain increased the extraction of light pulses (HILP) are three “non-thermal” technologies glycoalkaloids and β-glucans respectively which is promising with the capability to induce structural modifications in very as conventional methods for the extraction of these compounds different ways. PEF uses high voltages to electroporate cells are expensive and the market value of the target compounds is which punctures cell membranes by exceeding their trans- high. In terms of its application in meat processing, PEF can membrane potential. US (10-1000 W/cm2) is different as it have a positive impact on meat tenderization, though age- induces cavitation (i.e. formation of unstable bubbles which ing/conditioning is still necessary. It has also been shown to accelerate the curing of meat when it is applied as a pre- collapse and inactivate microorgansims through localized treatment prior to curing. Overall, these applications are all heating, mechanical/pressure shock waves and the forma- promising though it remains to be seen which of these niche tion of reactive species). High intensity light is a form of areas are likely to achieve significant commercial uptake. broad spectrum (100-1100 nm) light which is applied to foods in the form of short high energy (20-90,000 × sunlight Keywords Microbial inactivation, enzyme hydrolysis, μ valorization, meat processing. on earth) pulses (200-400 s in duration) which disrupt microbial cell structures at the level of DNA, induce cell membrane damage and localized heating. Combining these I. NTRODUCTION I technologies with their differing modes of action could lead Emerging processing technologies such as PEF, ultra- to synergies. In addition, as these are all classed as non- sound and light (e.g. high intensity light) can induce move- thermal technologies, their combination in a hurdle food ment or structural changes in foodstuffs which can be har- preservation strategy could be used in the production of nessed in a variety of ways. This paper will primarily minimally processed beverages where a similar preservation focus on the potential use of PEF in a number of food effect to conventional heat processing is obtained but with processing applications. Applications will be presented in less thermal impact on the product.

© Springer Science+Business Media Singapore 2016 7 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_2 8 J.G. Lyng et al.

When PEF is combined with other hurdles (HILP or US), at temperatures of >100 °C, (a form of PEF which will an increase in microbial inactivation is observed over and define as high temperature pulsed electric fields (HTPEF)). above that obtained when PEF is applied alone (Fig. 1). The objective was to assess whether products could be steri- lized at lower temperatures than conventional ultra-high temperature (UHT) heat processing (generally 135-140 °C) as the temperatures required for this inactivation are such that there is often a large impact on product quality. The heat resistance of Bacillus subtilis endospores to conventional thermal treatment was performed in two different matrixs, namely tap water and reconstituted skimmed milk at temperatures from 90 to 130 °C. In this characterization study, DT values (minutes) for each temperature were obtained. z-values of 16.1 °C and 18.3 °C were calculated for tap water and reconstituted skimmed milk, respectively. The effect of HTPEF processing when applied at initial Fig. 1 An illustration of the impact on microbial destruction of introducing processing temperatures of 100, 110 and 120 °C on the inac- US or HILP with PEF in a combined hurdle approach tivation of B. subtilis endospores was studied in tap water. As a control, thermal-alone effect on microbial inactivation was However, further analysis of this data shows that the na- also carried out by matching thermal effects in comparison ture of this interaction in a combined hurdle approach with HTPEF processing. The comparison between both (CHA) is largely additive (≈50%) or overlapping (≈25%) treatments was based on their FT values. HTPEF inactivation with synergy only accounting for ≈25%. values of B. subtilis endospores at the highest setting PEF- When the novel hurdle approaches incorporating PEF are inlet temperature (i.e. 120 °C) were of up to 4.30 and 3.79 compared to conventional heat processing, the specific log10 cycles in plate count agar (PCA) and selective for Bacil- energy inputs (J mL-1) into the products are high (in some lus cereus agar, respectively. Similar inactivation values of cases higher than conventional UHT processing). Also 4.08 and 3.70 log10 cycles were obtained in PCA and selec- -1 when normalised energy inputs (J mL log10) are compared, tive media respectively, when a comparable thermal-alone results show that energy input per Log inactivation are vari- treatment was applied (similar FT value). This study showed able with Gram + (Listeria innocua IMD 11288 and Staphy- the potential of HTPEF processing to provide a slight, but not lococcus aureus SST 2.4) requiring higher energy input per significant, increase in the inactivation of extremely resistant log cycle reduction, Gram – (E. coli K12 DSM 1607) re- microbial forms such as B. subtilis endospores when com- quiring intermediate and Yeast (Pichia fermentans) requir- pared to conventional heat treatment. Further studies at higher ing less energy. These differences in specific energy are in electric field strengths and elevated processing temperatures line with the knowledge on the impact of inherent structural could be performed in order to establish if greater microbial characteristics of microorganisms on their inactivation by inactivation levels can be achieved at temperatures which are PEF. The thicker peptidoglycan layer in Gram + is more more representative of those encountered in ultra-high tem- challenging for PEF (hence the higher specific energy of perature processing and also for a better understanding of the Gram + microorganisms) while the larger size of yeasts PEF principle of action at high temperature processing. How- generally renders them more susceptible to PEF (hence the ever, as a potential application, PEF with heat alone is unlike- lower specific energy of the yeast). Overall these results ly to be taken up commercially for inactivation of spores. suggest that PEF in a hurdle approach has potential for beverage pasteurization but its application is likely to be IV. CASE STUDY 3: PEF IN HYDROLYSIS restricted to premium products where higher associated TERMINATION uptake costs can be justified. The abolition of milk quotas in April 2015 means the Eu- III. CASE STUDY 2: PEF IN SPORE ropean dairy industry is adapting to the imminent expansion INACTIVATION in the milk pool. One of the ways in which value may be added to traditional protein dairy ingredients by modifying While the impact of PEF on vegetative microorganisms them via enzyme hydrolysis to produce more valuable com- is well documented, its role and potential in the inactivation pounds with bioactive properties (e.g. anti-inflammatory, of microbial spores is less promising when applied at low- satiety etc.). These hydrolyses often start with relatively moderate temperatures. This case study set out to investi- homogenous starting material and are often performed un- gate the potential for PEF to inactivate spores when applied der controlled pH and temperature conditions. When a

IFMBE Proceedings Vol. 53

Harnessing the Structure Modifying Potential of Pulsed Electric Fields (PEF) 9 critical point is reached in the hydrolysis, the enzymes need Rooster potatoes were placed under a fluorescent light for to be inactivated. Heat is the most commonly used method one week to ensure samples with enriched levels of glycoalka- to inactivate enzymes though its use may be at odds with loids. Potatoes were then peeled with a manual peeler then the retention of bioactivity in situations where the resultant air-dried for 4 days. Samples were exposed to a PEF treat- bioactive compounds are heat labile. PEF also has the po- ment prior to extraction in a water, acetic acid, sodium metabi- tential to inactivate enzymes by inducing structural modifi- sulfide solution in a solid liquid ratio of 1/25 and an extraction cations and possible electrochemical effects. time of 1 h at 80 °C under shaking conditions prior to centrifu- Hydrolysis was performed in a 2 litre temperature/pH gation and solid phase extraction cleanup. Results showed the controlled reaction vessel filled with a 10% (w/v) substrate potential for PEF to assist in the extraction of glycoalkaloids ° in distilled H2O, the solution was equilibrated to 50 C and from potato peel by up to 30%. A similar approach was its pH adjusted to 7.0 with 1 N NaOH before enzyme addi- adopted with brewers spent grains and barley which were pre- tion. The mixture was agitated at 300 rpm throughout treated with PEF prior to extraction using an appropriate me- hydrolysis, temperature and pH were logged while the thod with β-glucan levels quantified using a Megazyme kit. degree of hydrolysis (DH) was determined by the pH Results show an elevation in β-glucan yields for samples stat method. While heat is the most commonly used method treated with the highest electrical field strength. to inactivate enzymes, precise data on inactivation of commonly used enzymes is not readily available. Sodium caseinate hydrolysates were produced with a commercial VI. CASE STUDY 5: PEF protease Protamex. IN MEAT TENDERISATION A conventional thermal method to terminate enzymatic hy- The permeabilization effect on microbial and plant cell drolysis was evaluated and complete inactivation of Protamex membranes has recently increased the interest in the poten- was attained following a treatment of 95 °C for 20 s. PEF were tial of PEF application for meat and meat products as it examined in terms of their ability to terminate enzymatic hy- could modify quality and functional attributes (e.g. tender- drolysis. PEF treatment at electric field strengths of 18.2 ness and water holding capacity). kV/cm gave a 70% decrease in Protamex activity in a hydroly- Several microscopic studies have demonstrated physical sate solution with temperature rises below inactivation temper- changes in meat structure caused by PEF [1-5]. Changes in atures of 55 °C which was very promising as it suggests a tissue integrity were also indirectly supported by Töpfl [5] considerable amount of inactivation which was due to PEF who noted that PEF had an impact on sample conductivity alone. While conventional thermal inactivation treatments were and weight loss. more effective at inactivating Protamex in this study, further Conflicting results are reported regarding the effect of optimization (e.g. with moderate heat) may lead to complete PEF on meat tenderisation. PEF induced reductions in shear inactivation using milder treatments to conventional. Given force values for a range of beef muscles from 19-22% have the high value of these products, it may be an area where PEF been reported [3,6-8]. Suwandy et al. [8] reported a faster may find future application. rate of protein degradation in PEF-treated samples using SDS PAGE. Though it has been suggested there is no evi- V. CASE STUDY 4: PEF IN WASTE dence of PEF enhancing the release of protease enzymes ++ VALORISATION from the lysosomes or a faster release of Ca ions for accelerated proteolysis responsible for textural changes so the The Irish food industry produces large quantities of potato biochemical basis of the tenderisation benefit provided by peelings and brewers spent grains annually (approximately PEF warrants further exploration. 300,000 and 160,000 tonnes/annum respectively). Much of In contrast to the above, other studies have reported no this material is either incorporated into animal feed or dis- enhancement of meat tenderness due to PEF at different posed in landfills with other basic valorization strategies stages during the ageing process. Despite the fact that PEF including composting. However, potatoes and brewers spent caused an increased myofibrillar fragmentation in beef m. grains are valuable sources of glycoalkaloids and β-glucans semitendinosus, no changes in texture were revealed by respectively with the former compounds being useful inter- shear force measurements [4]. Similarly, texture of beef m. mediates in steroid drug synthesis while the latter reduce the longissimus thoracis [2], beef m. longissimus thoracis et risk of heart disease by lowering blood cholesterol. The ob- lumborum [9] and turkey breasts [10] was not affected by jective of this work was to optimize the extraction of high PEF application. The apparent discrepancy may be asso- added value products from food waste and to assess if PEF is ciated with differences in the experimental conditions such a viable method to assist in their extraction. as the use of different processing parameters, muscle types,

IFMBE Proceedings Vol. 53

10 J.G. Lyng et al. sample size and sample preparation prior and after PEF CONFLICT OF INTEREST treatment. Other parameters such as color and cook loss have been studied and it is agreed that PEF has no detrimen- “The authors declare that they have no conflict of interest”. tal effect on the color of meat samples [2,4,9-10] whereas cook loss was either improved [7,9] or remained unaffected EFERENCES [2]. According to Faridnia et al. [3] PEF enhanced lipid R oxidation and altered the volatile profile of frozen-thawed 1. Gudmundsson, M., and Hafsteinsson, H. (2001). Effect of electric beef m. semitendinosus during ageing though no alteration field pulses on microstructure of muscle foods and roes. Trends in of the nutritional value of fatty acids was detected. Howev- Fd Sci & Tech 12:122 128 2. Faridnia, F., Bekhit, A. E. D., Niven, B. and Oey, I. (2014a). Impact er, Arroyo et al. [10] found no differences in TBARS for of pulsed electric fields and post-mortem vacuum ageing on beef raw, cooked and frozen turkey breasts after 5 days of sto- longissimus thoracis muscles. Int J Fd Sci & Tech 49:2339 2347. rage post PEF. 3. Faridnia, F., Ma, Q. L., Bremer, P. J., Burritt, D. J., Hamid, N. and Oey, I. (2014b). Effect of freezing as pre-treatment prior to pulsed electric field processing on quality traits of beef muscles. VII. CASE STUDY 6: PEF IN MEAT CURING Innov Fd Sci & Emer Tech, in press. doi:10.1016/j.ifset.2014.09.007. PEF has been shown to improve the uptake of salt during 4. O’Dowd, L. P., Arimi, J. M., Noci, F., Cronin, D. A. and Lyng, J. the production of cured products by enhancing the diffusive G. (2013). An assessment of the effect of pulsed electrical fields on tenderness and selected quality attributes of post rigor beef mass transport along a concentration gradient by breaking muscle. Mt. Sci. 93:303 309. the cells permeability barrier. In this regard, Töpfl and 5. Töpfl, S. (2006). Pulsed Electric fields (PEF) for permeabilization Heinz [11] demonstrated that a PEF treatment prior to im- of cell membranes in food- and bioprocessing Applications, mersion in cover brine could improve the diffusion of salt process and equipment design and cost analysis. PhD Thesis, Technische Universität Berlin. and nitrate of pork haunches. Similarly, McDonnell et al. 6. Lopp, A. and Weber, H. (2005). Untersuchungen zur Optimierung [12] showed that the NaCl content of pork m. longissimus der Zartheit von Rindfleisch: Research into the optimising the thoracis et lumborum increased by 10% and 13% when tenderness of beef from parts of the forequarter. Fleischwirtschaft applying a 1.2 kV/cm and 2.3 kV/cm PEF pre-treatment, 85:111 116. 7. Bekhit, A. E. D. and Hopkins, D. L. (2014). Enhancement of meat respectively. The time required for meat curing and drying quality by pulsed electric field application. Meat and Livestock is essentially determined by the diffusion rate of curing Australia. Available at: http://www.ampc.com.au/site/assets/ agents and water within the tissue, and these studies high- media/reports/2015/A.MQA.0005-Enhancement-of-meat-quality- light the potential for PEF as a pre-treatment to accelerate by-pulse-electric-field-application.pdf. 8. Suwandy V., Carne, A., van de Ven, R., Bekhit, A. E. D. and this process. Further research has to be carried out on the Hopkins, D. L. (2015). Effect of pulsed electric field on the pro- optimisation of PEF efficiency through parameter combina- teolysis of cold boned beef M. Longissimus lumborum and M. tions which result in the lowest effective total treatment Semimembranosus. Mt Sci 100: 222 226. time. The fact that ingredients such as salt will modify tis- 9. Arroyo, C., Lascorz, D., O’Dowd, L., Noci, F., Arimi, J. and Lyng, J. G. (2015). Effect of Pulsed Electric Field treatments at sue has also to be taken into account when setting up the various stages during conditioning on quality attributes of beef PEF processing parameters. longissimus thoracis et lumborum muscle. Mt Sci 99, 52 59. 10. Arroyo, C., Eslami, S., Brunton, N. P., Arimi, J. M., Noci, F. and Lyng, J. G. (2015). An assessment of the impact of Pulsed Elec- VIII. CONCLUSIONS tric Fields processing factors on oxidation, color, texture and sensory attributes of turkey breast meat. Poult Sci, 94,1088-1095. The food industry is poised to adopt new concepts and 11. Töpfl, S. and Heinz, V. (2008). The stability of uncooked cured prod- technologies that offer advantages over conventional sys- ucts. Use of pulsed electric fields to accelerate mass transport opera- tems or just have particular benefits. Each of the 6 case tions in uncooked cured products. Fleischwirtschaft 88, 127 130. 12. McDonnell, C. K., Allen, P., Chardonnereau, F. S., Arimi. J. M. studies presented has commercial potential but only time and Lyng, J. G. (2014). The use of pulsed electric fields for acce- will tell whether they will achieve any commercial uptake. lerating the salting of pork. LWT - Food Science and Technology 59, 1054 1060.

ACKNOWLEDGMENT Author: James Lyng The work reported in this paper was funded under the Institute: University College Dublin Food Institutional Research Measure (FIRM) (Irish De- Street: Belfield City: Dublin 4 partment of Agriculture, Food and the Marine) and Food for Country: Ireland Health Ireland (FHI) (Enterprise Ireland). Email: [email protected]

IFMBE Proceedings Vol. 53

Fundamental and Applied Aspects of Pulsed Electric Fields for Microbial Inactivation

J. Raso Food technology, Faculty of Veterinary, University of Zaragoza, Zaragoza, Spain

Abstract— Pulsed electric field (PEF) is a technology electric field, treatment time) and the cell characteristics that causes electroporation of the cell membranes by (size, shape, orientation in the electric field) the viability of application of intermittent electric fields of high intensi- the electroporated cell can be preserved by recovering the ty for short periods of time (µs). The local defects or membrane integrity, or the electroporation can be perma- pores created by the application of an external electric nent. Electroporation causes inactivation of vegetative field may lead to the loss of the membrane integrity and forms of microorganisms at temperatures below those used uncontrolled molecular transport across microbial in thermal processing [1]. membranes. These events may abolish the microbial capacity to maintain the microbial homeostasis causing II. BASICS PRINCIPLES OF MICROBIAL microbial inactivation at temperatures below those used INACTIVATION BY PULSED ELECTRIC FIELDS in thermal processing. Based in this phenomenon PEF provides an excellent alternative to conventional thermal Although the mechanism underlying microbial inactiva- pasteurization of heat sensitive foods. tion by PEF has not been fully elucidated it is thought that is based on the electroporation of the microbial cytoplasmat- Keywords microbial inactivation, preservation, Pulsed ic membrane. The local defects or pores created in the electric fields. cytoplasmatic membrane by the application of an external electric field lead to the lost of its semipermeable barrier I. INTRODUCTION function and uncontrolled molecular transport across the membrane can occurs. These events may abolish the ho- Currently, the food industry’s primary intervention to meostatic capacity of the microorganisms and will eventual- supply safe foods with an extended shelf-life is thermal ly lead to microbial death [1]. However, the cytoplasmatic treatment. However, heating may cause unacceptable dete- membrane is not the only barrier that separates the cytop- rioration of foods with a heat labile chemical or physical lasm from the environment in microorganisms including structure. Indeed over the last decades considerable research bacteria and yeast. Cytoplasmatic membrane of Gram- efforts have been focus towards the development of non- positive bacteria is surrounded by a thick cell wall made of thermal processing technologies such as high hydrostatic peptidoglycans and tectonic acids. On the other hand the pressure, ultra-violet light, pulse light, ultrasound, high cell wall in Gram negative bacteria is thinner but it is sur- pressure homogenization or pulsed electric fields. The con- rounded by an outer membrane that differs from typical cept of nonthermal technologies referrers to a group of biological membrane because the main molecular constitu- technologies whose effects in foods are similar to effects ents are lipopolysacharides. Similarly to gram positive bac- cause by heating, but the processing temperatures are lower teria yeast cells are surrounded by a cell wall. Although it to those used in thermal processing. has been observed differences in the effect caused by PEF PEF technology is considered a promising no thermal in gram positive and gram negative bacteria, how influence alternative that has received considerable attention for mi- the envelopes surrounding the cytoplasmatic membrane on crobial inactivation in foods. The treatment consists on the electroporation is an aspect still unknown. application of pulses of high voltage and short duration to a Evidence of cytoplasmatic membrane electroporation liquid placed between two electrodes. This voltage results in caused by PEF has been observed by different authors using an electric field which intensity depends on the voltage different techniques such as leakage of intracellular materi- delivered and the gap between the electrodes. PEF cause al, measurement of osmotic response or fluorescent dye some type of structural rearrangement of the cell mem- exclusion assays. branes that consists on the formation of local defects or Several authors have observed leakage of intracellular pores (electroporation) that lead to an increment of the cell compounds from different microorganisms treated by PEF membrane permeability to ions and macromolecules. De- treatments were non-lethal indicating that the temporary pending on the intensity of the treatment applied (external loss of permeability control is not necessarily lethal [3]. © Springer Science+Business Media Singapore 2016 11 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_3

14 J. Raso conditions for PEF pasteurization could be expected to be short residence time (less than 1 s) at moderate electric field different for foods, depending on their pH. strengths. Under this treatment condition the fruit juice shelf life is extended up to 21 days, while maintaining fresh-like IV. FOOD PRESERVATION BY PULSED taste and product quality. At present such products are commercialized, using PEF equipment with a capacity in ELECTRIC FIELDS the range of 1,500 to 8,000 l/h [11]. PEF is viewed as one of the most promising nonthermal methods for inactivating microorganisms in liquid foods. CONFLICT OF INTEREST The efficacy of this technology for inactivating vegetative cells of bacteria and yeast by PEF has widely been demon- The authors declare that they have no conflict of interest strated. However, bacterial spores are resistant to PEF treatments. As bacterial spores are resistant to PEF treat- REFERENCES ments, the primary applications of this technology in food preservation must focus on pasteurization. Traditionally, 1. Álvarez I, Condón S, Raso J. (2006). Microbial inactivation by pulsed pasteurization has been based on thermal processing but, in electric fields. In: Raso J Heinz V, editors. Pulsed electric fields technology for the food industry: Fundamentals and applications. New recent years, it has been demonstrated that there is potential York:Springer. pp 97-129. for several nonthermal technologies to obtain the same goal 2. Mackey, B. M., (2000) Injured bacteria, in: The Microbiological As a consequence, pasteurization has been redefined by the Safety and Quality of Food. (B. M. Lund, T. C. Baird-Parker, and G. National Advisory Committee on Microbiological Criteria W. Gould, eds.), Aspen Publishers, Inc., Gaithersburg, Maryland, pp. 1-16. for Foods as: “any process, treatment, or combination the- 3. Aronsson, K., Rönner, U., Borch, E., (2005) Inactivation of Es- reof that is applied to food to reduce the most resistant mi- cherichia coli, Listeria innocua and Saccharomyces cerevisiae in re- croorganism(s) of public health significance to a level that lation to membrane permeabilization and subsequent leakage of intra- is not likely to present a public health risk under normal cellular compounds due to pulsed electric field processing, Int. J. Food Microbiol. 99:19-32. conditions of distribution and storage”. Therefore, this defi- 4. García, D., Gómez, N., Mañas, P., et al. (2006) Pulsed electric fields nition considers new techniques for food pasteurization cause bacterial envelopes permeabilization depending on the treat- such as PEF. ment intensity, the treatment medium pH and the microorganisms in- Although the main objective of PEF pasteurization is to vestigated, Int. J. Food Microbiol.113:219-217 5. Simpson, R. K., Whittington, R., Earnshaw, R. G. Et al. (1999) guarantee food safety, a large proportion of the population Pulsed high electric field causes all or nothing membrane damage in of vegetative spoilage microorganisms is also inactivated by Listeria monocytogenes and Salmonella the treatment. Therefore, PEF treatments may extend the 6. García, D., Gómez, N., Mañas, P. et al. (2005) Occurrence of suble- shelf-life of foods. thal injury after pulsed electric fields depending on the microorganism, the treatment medium pH and the intensity of the treatment in- PEF is gaining interest as a gentle method of food pre- vestigated, J. Appl. Microbiol. 99:94-104 servation of heat sensitive foods such as fruit juices and 7. García, D., Gómez, N., Mañas, et al. (2005) Biosynthetic require- smoothies. Commercial exploitation of PEF for food pasteu- ments for the repair of sublethal membrane damage in Escherichia rization requires proof that PEF promotes a level of micro- coli cells after pulsed electric fields, J. Appl. Microbiol. 100:428-435. 8. Ho, S. Y., Mittal, G. S., (1996) Electroporation of cell membranes: A bial safety that is equal to that made possible via traditional review, Crit. Rev. Biotechnol. 16:349-362. processing. A 5 Log10 reduction of the most resistant mi- 9. Álvarez, I., Pagán, R., Condón, S. et al. (2003). The influence of croorganism of public health significance has been estab- process parameters for the inactivation of Listeria monocytogenes by lished by the U.S. Food and Drug Administration (FDA) pulsed electric fields. Int. J. Food Microbiol. 87:87-95. 10. Saldaña, G., Puértolas, E., Álvarez, I., et al. (2010) Evaluation of a concerning fruit juice pasteurization. Studies to evaluate the static treatment chamber to investigate kinetics of microbial inactiva- application of PEF for microbial decontamination at room tion by pulsed electric fields at different temperatures at quasi- temperature have shown that to obtain these levels of reduc- isothermal conditions. J. Food Eng., 100:349-356. tion it is necessary to apply long treatments (i.e., >100 µs) 11. Toepfl, S., 2012 Pulsed electric field food processing industrial equipment design and commercial applications. Stewart Postharvest at high electric field strengths (i.e., ≥30 kV/cm). At com- Review 2:1-7. mercial scale, technical and economical limitations exist in applying these intense treatments in continuous flow. Author: Javier Raso However, several studies confirmed that the application of Institute: University of Zaragoza PEF at moderate temperatures provides the possibility of Street: C/ Miguel Servet, 177 City: Zaragoza obtaining substantial microbial inactivation of pathogenic Country: Spain microorganisms that are particularly PEF resistant with a Email: [email protected]

IFMBE Proceedings Vol. 53

How Imaging Molecule Uptake into Cells can Reveal the Mechanisms of Membrane Electropermeabilization

M.-P. Rols1,2 1 Institut de Pharmacologie et de Biologie Structurale du CNRS, 205 route de Narbonne, 31077 Toulouse, France 2 Université de Toulouse III, Toulouse, France

Abstract Cell membranes can be transiently permeabi- known (and still not known) about the mechanisms of mo- lized under application of electric field pulses. This process, lecule electrotransfer. Open questions exist about the beha- called electropermeabilization or electroporation, allows hy- vior of the membranes both while the field is on (µs to ms drophilic molecules, such as anticancer drugs and DNA, to enter into cells and tissues. Single cell imaging experiments time range) and after its application (from seconds to sever- revealed that the uptake of molecules is a complex multi-step al minutes and hours). Also, there is a lack of understanding that takes place in well-defined membrane regions under on how molecules are transported in complex environments, processes that affect the entire membrane, depend on the such as those found in tissues. As our objectives are to give properties of the molecules (size, charge) and in which the a complete molecular description of the mechanisms, our cytoskeleton can play a major role. strategy is to use different complementary systems with Keywords electroporation, cell membrane, cell imaging, increasing complexities (model membranes, cells in culture, intracellular traffic, molecule uptake. spheroids and tissues in living mice) and different microscopy tools to analyze the processes. I. INTRODUCTION The use of electric pulses to deliver therapeutic agents II. MECHANISMS OF MEMBRANE including drugs, proteins and nucleic acids in a wide range ELECTROPERMEABILIZATION AND cells and tissues has been developed over the last decade DNA TRANSFER INTO CELLS. (1-4). It is used in clinics for cancer treatment. Electroche- motherapy is a standard treatment in many cancer centers A. Basic Aspects: Modulation of Membrane Potential around Europe that potentiates the delivery of cytotoxic Difference drugs by local electric pulses delivery (http://guidance. The key effect of an electric field on cells is a position- nice.org.uk/IPG446) (5). This strategy is also promising for dependent change in the resting transmembrane potential the systemic secretion of therapeutic proteins. Vaccination difference, ΔΨo, of their plasma membrane. The electrically and oncology gene therapy are also major fields of applica- induced potential difference, ΔΨE, which is the difference tion of electrotransfer (6, 7). Translation of preclinical stu- between the potential inside the cell, Ψin, and the potential dies into clinical trials in human and veterinary oncology outside the cell, Ψout, at a point M on the cell surface, is has started (8, 9). The first phase 1 dose escalation trial of given by: plasmid interleukin carried out in patients with metastatic melanoma has shown very encouraging results (10). The ⎛ t ⎞ ⎡ ⎜ − ⎟ ⎤ method has also been successfully used for the treatment of ΔΨ =Ψ −Ψ =− λ θ − ⎝ τ ⎠ E ()t in out g( )rEcos( M )1⎢ e ⎥ (1) dogs and horses (8). But the safe and efficient use of this ⎣⎢ ⎦⎥ physical method for clinical purposes requires the knowledge of the mechanism underlying the electropermeabiliza- where t is the time after the onset of the electric pulse, g tion phenomena. Despite the fact that the pioneering work depends on the conductivities λ of the membrane, of the on plasmid DNA electrotransfer in cells was initiated 30 cytoplasm and of the extracellular medium, r is the radius of years ago (11), many of the mechanisms underlying the the cell, E the field strength, θM is the angle between process remain to be elucidated (12, 13). One has to notice the normal to the membrane at the position M and the direc- that even if in vitro electrotransfer is usually efficient in tion of the field, and τ is the membrane charging time (14). almost all cell lines, in vivo gene delivery and expression The field-induced potential difference adds to the resting faces some problems. For instance in tumors, efficiency potential ΔΨo (15): remains low with only a few percent of transfected cell. The focus of the lecture is to describe the basics of cell ΔΨ = ΔΨ +ΔΨ (2) electropermeabilization and the different aspects of what is 0 E

© Springer Science+Business Media Singapore 2016 15 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_4 16 M. P. Rols

Being dependent on the angular parameter θ, the field ef- the bulk into the cytoplasm. The mechanism involved is fect is position-dependent on the cell surface. Therefore, the clearly specific for the physico-chemical properties of the transmembrane potential difference of a cell exposed to an electrotransferred molecule (17). electric field defines the sites (location, extend) where mo- Progress in the knowledge of the involved mechanisms is lecule uptake can take place. still a biophysical challenge. One of our recent objectives was to detect and visualize at the single-cell level the inci- B. Electropermeabilization: A Fast and Localized Process dence of phospholipid scrambling and changes in mem- The use of video microscopy allows visualization of the brane order (18, 19). The pulses induced the formation of permeabilization phenomenon at the single cell level. Pro- long-lived permeant structures and resulted in a rapid phos- pidium iodide uptake in the cytoplasm is a fast process that pholipid flip/flop within less than 1s and were exclusively can be detected seconds after the application of electric restricted to the regions of the permeabilized membrane. pulses. Exchange across the permeabilized membrane is not Our results could support the existence of direct interactions homogeneous on the whole cell membrane. It occurs at the between the movement of membrane zwitterionic phospho- sides of the cells facing the electrodes in an asymmetrical lipids and the electric field. We also performed experiments way where it is more pronounced at the anode-facing side of on lateral mobilities of proteins and showed that electro- the cells than at the cathode (Figure 1), i.e. in the hyperpola- permeabilization affects the lateral mobility of Rae-1, a GPi rized area than in the depolarized area (16), which is in anchored protein. Our results suggest that 10-20 % of the agreement with the above theoretical considerations. membrane surface is occupied by defects or pores and that these structures propagate rapidly over the cell surface. Electrotransfer of plasmid DNA (pDNA) also affects the lateral mobility of Rae-1. Once inserted into the membrane, pDNA is completely immobile and excludes Rae-1; this indicates that the pDNA molecules are tightly packed together to form aggregates. In addition, we took advantage of atomic force microscopy (AFM) to directly visualize the consequences of electropermeabilization and to locally measure the membrane elasticity. We visualized transient rippling of membrane surface and measured a decrease in membrane elasticity. AFM indeed allowed visualizing membrane permeabilization with a different time scale than propidium iodide and other molecules entry. We showed Fig. 1 Molecule electrotransfer mechanisms. Left: During electric pulses that the membrane elasticity decreases slower than the PI application: Plasma membrane is electropermeabilized facing the 2 elec- enter the cell after electric pulses application. This mean trodes (PI uptake). DNA aggregates are formed. This interaction takes that even if membrane permeabilization is locally induced place only on the membrane facing the cathode. Right: About 2 h after electric pulses application, DNA molecules are present at nucleus in specific regions of the cell and is transient (as seen by level. Finally, eGFP expression is detected for hours. The arrow fluorescence imaging), its consequences on stiffness are indicates the direction of the electric field. global and still present when membrane resealing has occurred. Our results obtained both on fixed and living CHO Electropermeabilization can be described as a 3-step cells therefore give evidence of an inner effect affecting the process in respect with electric field: (i) before electropulsa- entire cell surface that may be related to cytoskeleton desta- tion, the plasma membrane acts as a physical barrier that bilization. Thus, AFM appears as a useful tool to investigate prevents the free exchange of hydrophilic molecules between basic process of electroporation on living cells in absence of the cell cytoplasm and external medium; (ii) during electro- any staining (20, 21). pulsation, the transmembrane potential increases which induces the formation of transient permeable structures facing C. Gene Electrotransfer: A Multistep and Localized Process the electrodes and allow the exchange of molecules; (iii) after electropulsation, membrane resealing occurs. Single-cell microscopy and fluorescent plasmids can be A direct transfer into the cell cytoplasm of the negatively used to monitor the different steps of electrotransfection charged small molecules such as siRNA was observed on (22). DNA molecules, which are negatively charged, mi- the side facing the cathode. When added after electropulsa- grate electrophoretically when submitted to the electric tion, siRNA did not penetrate the cells. Therefore, electric field. Under electric fields which are too small to permeabi- field acts on both the permeabilization of the membrane and lize the membrane, the DNA simply flows around the cell in on the electrophoretic drag of the charged molecules from the direction of the anode. However, beyond a critical field

IFMBE Proceedings Vol. 53

How Imaging Molecule Uptake into Cells can Reveal the Mechanisms of Membrane Electropermeabilization 17 value, above which cell permeabilization occurs, the DNA interacts with the plasma membrane. This interaction only occurs at the pole of the cell opposite the cathode and this demonstrates the importance of electrophoretic forces in the initial phase of the DNA/membrane interaction. When the DNA-membrane interaction occurs, one observes the formation of “microdomains” whose dimensions lie between 0.1 and 0.5 µm (Figure 1). The formation of plasmid complexes at fixed sites suggests that membrane domains may be responsible for DNA uptake and their lack of mobility could be due to their interaction with the actin cytoskeleton (23, 24). Translocation of the plasmid from the plasma membrane to the cytoplasm and its subsequent passage towards the nuclear envelope take place with a kinetics ranging from minutes to hours (25). When plasmid has reached the nuclei, gene expression can take place and this can be detected up to several weeks later. Even if the first stage of gene electrotransfection, i.e. migration of plasmid DNA towards the electropermeabi- lised plasma membrane and its interaction with it, becomes understood we are not totally able to give guidelines to improve gene electrotransfer. Successful expression of the plasmid depends on its subsequent migration into the cell. Therefore, the intracellular diffusional properties of plasmid Fig. 2 DNA electrotransfer as a multistep process. During the application DNA, as well as its metabolic instability and nuclear trans- of the electric field: (1) the plasma membrane is permeabilized (orange), location, represent other cell limiting factors (26). (2) the DNA is electrophoretically pushed on the membrane side facing the cathode therefore (3) DNA/membrane interactions occur. DNA aggregates In the conditions induced during electropermeabilisation, are inserted into the membrane and remain there for tens of minutes. After the time a plasmid DNA takes to reach the nuclei is signifi- the application of the electric field and resealing of the membrane (yellow), cantly longer than the time needed for a small molecule and (4) the DNA is internalized via endocytosis (DNA in vesicles). Free DNA is perhaps also internalized through electropores. For gene expression to the complexity of the process as well. Therefore, plasmid occur (5-6), DNA has to cross the cytoplasm toward the nucleus. Our study DNA present in the cytosol after being electrotransferred suggests (5) actin related motion that can be in the form of (5a) actin can be lost before reaching the nucleus. Finally, after the polymerization that pushes the DNA, free or in vesicles (actin rocketing) cytoskeleton, the nuclear envelope represents the last, but and/or (5b) transport via the myosins (in both directions). We observe (6) microtubules related motion which can mean (6a) transport via kinesin and by no means the least important, obstacle for the expression dynein, (6b) DNA interaction with oppositely directed motors and with of the plasmid DNA. (6c) several motors of the same type (6d). Once being in the perinuclear The relatively large size of plasmid DNA (2-10 MDa) region (in vesicles and perhaps also free, 7), DNA has to cross the nuclear envelope, after endosomal escape in case of DNA in vesicles. Finally, makes it unlikely that the nuclear entry occurs by passive DNA is expressed in proteins found in the cytoplasm (8). From ref 27. diffusion. We recently showed how electrotransferred DNA is transported in the cytoplasm towards the nucleus (Figure III. IPID VESICLES AND 3D CELL CULTURES 2) (27). For this purpose, we used single particle tracking L AS OTHER MODELS TO STUDY (SPT) experiments of individual DNA aggregates in living ELECTROPERMEABILIZATION cells (28-30). We showed that fast active transport of the DNA aggregates over long distances occurs. Tracking expe- The characterization of the membranes domains observed riments in CHO cells treated with different drugs affecting during electrotransfer is still an open question. For that both the actin and the tubulin networks clearly demonstrate purpose, we use giant unilamellar vesicles (GUVs) to study that this transport is related to the cellular microtubule net- the effect of permeabilizing electric fields in simple mem- work. Actin is indeed involved in the internalization of the brane models. GUVs represent a convenient way to study DNA by polymerising around it and is important in addition membrane properties such as lipid bilayer composition and for the DNA distribution in the cells and the regulation of membrane tension. They offer the possibility to study and the transport by the microtubules. visualize membrane processes due to their cell like size in

IFMBE Proceedings Vol. 53

18 M. P. Rols absence of any constraint due to cell cytoskeleton. Experi- reconstructed human connective tissue model. This human ments showed a decrease in vesicle radius which is inter- cell model presents multiple layers of primary dermal fibro- preted as being due to lipid loss during the permeabilization blasts embedded in a native, collagens rich extracellular process. Three mechanisms responsible for lipid loss were matrix. We just showed that the cells within this standard- directly observed and will be presented: pore formation, ised 3D normal tissue could be efficiently electropermeabi- vesicle formation and tubule formation, which may be in- lized (39). We believe that a better comprehension of gene volved in molecules uptake (31). We also gave evidence electrotransfer in such a model tissue would help to improve that GUVs are a good model to study the mechanisms of electrogene therapy approaches such as the systemic deliv- electrofusion, with a direct interest to their use as vehicles ery of therapeutic proteins and DNA vaccination. to deliver molecules (32). However, a direct transfer of DNA into the GUVs took place during application of the electric pulses (33). That gives clear evidence that “lipid IV. CONCLUSIONS bubble” is not a cell and a tissue is not a simple assembly of Classical theories of electropermeabilization present single cells. Therefore, in the last past few years, we de- some limits to give a full description of the transport of cided to use an ex vivo 3D-culture cell model, namely tumor molecules through membranes. Certain effects of the elec- multicellular spheroids, for the understanding of the DNA electrotransfer process in tissues. tric field parameters on membrane permeabilization and the Upon growth, spheroids display a gradient of proliferat- transport of molecules are well established but a great deal ing cells. These proliferating cells are located in the outer of what happens at the molecular level remains speculative. cell-layers and the quiescent cells are located more cen- Molecular dynamics give interesting insight into the process trally. This cell heterogeneity is similar to that found in (40-41). Electroinduced destabilisation of the membrane avascular microregions of tumors. We used confocal mi- includes both lateral and transverse redistribution of lipids, croscopy to visualize the repartition of permeabilized cells leading to mechanical and electrical modifications which in spheroids submitted to electric pulses. Electrotransfer of are not yet fully understood. One may suggest that such bleomycin and cisplatin confirmed the relevance of the modifications can be involved in the subsequent transport of model in the case of electrochemotherapy and doxorubicin molecules interacting with them such as DNA. Experimen- showed its potential to screen new antitumor drug candi- tal verification of the basic mechanisms leading to electro- dates for ECT. Confocal microscopy was used to visualize permeabilization and other changes in the membrane re- the topological distribution of permeabilized cells in 3D mains a priority given the importance of these phenomena spheroids. Our results revealed that all cells were efficiently in cell biology, medicine and industrial applications. permeabilized, whatever their localization in the spheroid, even those in the core. The combination of antitumor drugs and electric pulses (ECT) led to changes in spheroid macro- ACKNOWLEDGMENT scopic morphology and cell cohesion, to tumor spheroid growth arrest and finally to its complete apoptosis-mediated Research conducted in the scope of the EBAM European dislocation, mimicking previously observed in vivo situa- Associated Laboratory and of the COST TD1104 action and tions. Taken together, these results indicate that the spheroid supported by the Centre National de la Recherche Scientifi- model is relevant for the study and optimization of elec- que, the Agence Nationale de la Recherche, Projet tromediated drug delivery protocols (34). Small molecules PIERGEN ANR-12-ASTR-0039, the Direction Générale de can be efficiently transferred into cells, including the ones l’Armement and the Midi-Pyrénées Région. The data are present inside the spheroids, gene expression is limited to from PhD students and post-docs I have/had the pleasure to the external layers of cells (35). Taken together, these results, in agreement with the ones obtained by the group of supervise: Muriel Golzio, Cécile Faurie, Emilie Phez, Jean- R. Heller (36), indicate that the spheroid model is more Michel Escoffre, Thomas Portet, Chloé Mauroy, Louise relevant to an in vivo situation than cells cultured as Chopinet, Rosazza Christelle, Amar Tamra, Moinecha Ma- monolayers (37, 38). di, Luc Wasungu, Flavien Pillet and Laure Gibot. Another recent 3D cell culture models we are now using to assess the effects of extracellular matrix composition and CONFLICT OF INTEREST organization, as well as intercellular junctions and communication, in tissue response to electric pulses, is human 3D The authors declare that they have no conflict of interest.

IFMBE Proceedings Vol. 53

How Imaging Molecule Uptake into Cells can Reveal the Mechanisms of Membrane Electropermeabilization 19

REFERENCES 23. Rosazza, C. et al. 2011. The actin cytoskeleton has an active role in the electrotransfer of plasmid DNA in mammalian cells. Mol Ther 1. Gothelf, A., L. M. Mir, and J. Gehl. 2003. Electrochemotherapy: 19:913-921. results of cancer treatment using enhanced delivery of bleomycin by 24. Rosazza, C. et al. 2011. Cholesterol implications in plasmid DNA electroporation. Cancer Treat Rev 29:371-387. electrotransfer: Evidence for the involvement of endocytotic path- 2. Rols, M. P. 2010. Gene transfer by electrical fields. Curr Gene Ther ways. Int J Pharm. 10:255. 25. Faurie, C. et al. 2010. Electro-mediated gene transfer and expression 3. Andre, F. M., and L. M. Mir. 2010. Nucleic acids electrotransfer in are controlled by the life-time of DNA/membrane complex forma- vivo: mechanisms and practical aspects. Curr Gene Ther 10:267-280. tion. J Gene Med 12:117-125. 4. Mir, L. M. 2008. Application of electroporation gene therapy: past, 26. Lechardeur, D., and G. L. Lukacs. 2006. Nucleocytoplasmic Trans- current, and future. Methods Mol Biol 423:3-17. port of Plasmid DNA: A Perilous Journey from the Cytoplasm to the 5. Caraco, C. et al. 2013. Long-lasting response to electrochemotherapy Nucleus. Hum Gene Ther 17:882-889. in melanoma patients with cutaneous metastasis. BMC Cancer 27. Rosazza, C. et al. 2013. Intracellular tracking of single plasmid DNA- 13:564. particles after delivery by electroporation. Mol Ther. 6. Vandermeulen, G. et al. 2007. Optimisation of intradermal DNA 28. Ruthardt, N., D. C. Lamb, and C. Brauchle. 2011. Single-particle electrotransfer for immunisation. J Control Release 124:81-87. tracking as a quantitative microscopy-based approach to unravel cell 7. Chiarella, P., V. M. Fazio, and E. Signori. 2013. Electroporation in entry mechanisms of viruses and pharmaceutical nanoparticles. Mol DNA vaccination protocols against cancer. Curr Drug Metab 14:291- Ther 19:1199-1211. 299. 29. Braeckmans, K. et al. 2010. Advanced fluorescence microscopy 8. Cemazar, M., T. Jarm, and G. Sersa. 2010. Cancer electrogene thera- methods illuminate the transfection pathway of nucleic acid nanopar- py with interleukin-12. Curr Gene Ther 10:300-311. ticles. J Control Release 148:69-74. 9. Heller, L. C., and R. Heller. 2010. Electroporation gene therapy 30. Brandenburg, B., and X. Zhuang. 2007. Virus trafficking - learning preclinical and clinical trials for melanoma. Curr Gene Ther 10:312- from single-virus tracking. Nat Rev Microbiol 5:197-208. 317. 31. Portet, T. et al. 2009. Visualization of membrane loss during the 10. Daud, A. I. et al. 2008. Phase I trial of interleukin-12 plasmid elec- shrinkage of giant vesicles under electropulsation. Biophys J troporation in patients with metastatic melanoma. J Clin Oncol 96:4109-4121. 26:5896-5903. 32. Mauroy, C et al. 2012. Interaction between GUVs and catanionic 11. Neumann, E., et al. 1982. Gene transfer into mouse lyoma cells by nanocontainers: new insight into spontaneous membrane fusion. electroporation in high electric fields. Embo J 1:841-845. Chem Commun (Camb) 48:6648-6650. 12. Escoffre, J. M. et al. 2009. What is (Still not) Known of the Mechan- 33. Portet, T. et al. 2011. Insights into the mechanisms of electrome- ism by Which Electroporation Mediates Gene Transfer and Expres- diated gene delivery and application to the loading of giant vesicles sion in Cells and Tissues. Mol Biotechnol 41:286-295. with negatively charged macromolecules. Soft Matter 7:3872-3881. 13. Cemazar, M., et al. 2006. Electrically-assisted nucleic acids delivery 34. Gibot, L. et al. 2013. Antitumor drug delivery in multicellular sphero- to tissues in vivo: where do we stand? Curr Pharm Des 12:3817-3825. ids by electropermeabilization. J Control Release 167:138-147. 14. Bernhardt, J., and H. Pauly. 1973. On the generation of potential 35. Wasungu, L. et al. 2009. A 3D in vitro spheroid model as a way to differences across the membranes of ellipsoidal cells in an alternating study the mechanisms of electroporation. Int J Pharm 379:278-284. electrical field. Biophysik 10:89-98. 36. Marrero, B., and R. Heller. 2012. The use of an in vitro 3D melanoma 15. Kotnik, T. et al. 2000. Evaluation of cell membrane electropermeabi- model to predict in vivo plasmid transfection using electroporation. lization by means of a nonpermeant cytotoxic agent. Biotechniques Biomaterials. 28:921-926. 37. Chopinet, L., L. Wasungu, and M. P. Rols. 2011. First explanations 16. Bureau, M. et al. 2004. Intramuscular plasmid DNA electrotransfer: for differences in electrotransfection efficiency in vitro and in vivo biodistribution and degradation. Biochim Biophys Acta 1676:138- using spheroid model. Int J Pharm. 148. 38. Gibot, L., and M. P. Rols. 2013. Progress and Prospects: The Use of 17. Paganin-Gioanni, A. et al. 2011. Direct visualization at the single-cell 3D Spheroid Model as a Relevant Way to Study and Optimize DNA level of siRNA electrotransfer into cancer cells. Proc Natl Acad Sci U Electrotransfer. Current Gene Therapy 13:175-181. S A 108:10443-10447. 39. Madi, M., Rols, M.P. and Gibot, L. 2015. Efficient In Vitro Electro- 18. Escoffre, J. M., et al. 2014. Evidence for electroinduced membrane permeabilization of Reconstructed Human Dermal Tissue. J Mem- defects assessed by lateral mobility measurement of a GPi anchored brane Biol 15. protein. Eur Biophys J. 40. Tokman M. et al. 2013. Electric field-driven water dipoles: nanoscale 19. Escoffre, J. M. et al. 2014. Membrane disorder and phospholipid architecture of electroporation.PLoS One.11;8(4) scrambling in electropermeabilized and viable cells. Biochim Bio- 41. Polak A. et al. 2014. Electroporation of archaeal lipid membranes phys Acta 1838:1701-1709. using MD simulations. Bioelectrochemistry. 100:18-26. 20. Chopinet, L. et al. 2013. Destabilization induced by electropermeabilization analyzed by atomic force microscopy. Biochim Biophys Acta 1828:2223-2229. 21. Chopinet, L. et al. 2013. Imaging living cells surface and quantifying Author: Rols its properties at high resolution using AFM in QI (TM) mode. Micron Institute: IPBS-CNRS 48:26-33. Street: 205 route de Narbonne 22. Golzio, M., J. Teissie, and M. P. Rols. 2002. Direct visualization at City: Toulouse the single-cell level of electrically mediated gene delivery. Proc Natl Country: France Email: [email protected]

Acad Sci U S A 99:1292-1297.

IFMBE Proceedings Vol. 53

Tissue Reactions to Electroporation and Electrochemotherapy: Vascular Effects that have Implications in Tumor Treatment

G. Sersa1 and M. Čemažar1,2 1 Institute of Oncology Ljubljana, Department of Experimental Oncology, Zaloska 2, SI 1000 Ljubljana, Slovenia 2 Faculty of Health Sciences, University of Primorska, Polje 42, SI 6310 Izola, Slovenia

Abstract Electrochemotherapy is an ablative technique of action, and highlight its implications in the treatment of utilizing electroporation for enhanced drug delivery to cells. cutaneous and deep seated tumors. The predominant mechanism of action is enhanced tumor cell cytotoxicity; however, the vascular disrupting effect is also present and is of significant importance. It is limited to tumor II. VASCULAR EFFECTS microvasculature, while electrochemotherapy does not affect OF ELECTROPORATION bigger and normal blood vessels. This clinical indication of electrochemotherapy application is important in the treatment It all started with the observation that tumors become of bleeding metastases, since other ablative techniques lack whitish, pale, after the application of electric pulses. This those mechanisms of action. This was confirmed by the clinical stimulated the initial preclinical studies that were exploring data, where electrochemotherapy proved effective and safe in vascular effects of electroporation on a physiological level. the treatment of metastases located in the vicinity of the major Obtained results demonstrated that the application of elec- hepatic vessels, not amenable to surgery or radiofrequency tric pulses of different amplitude, duration, frequencies and ablation. their number, induced local and transient reduction of the Keywords Electrochemotherapy, vascular effects, vascular perfusion in the exposed tissue [6], where the effect was disrupting effect, bleeding metastases. more pronounced in tumors, than in the normal tissues [6,7]. The difference in the responsiveness was attributed to the I. INTRODUCTION specificity of the tumor vasculature that is chaotic, and less developed than the normal vasculature. Electrochemotherapy is one of local ablative techniques, These first studies stimulated more mechanistically with high effectiveness and no side effects [1]. Its clinical oriented studies from which we developed a model of how use is in the treatment of cutaneous and subcutaneous tu- electric pulses affect the vessels. The model postulated that mors of different histology with the response rate of approx- vascular permeability is induced due to the rounding of imately 70-80% [2]. The treatment is acknowledged in endothelial cells (Fig.1). Consequently, the interstitial fluid many European countries and is entering routine clinical pressure is increased, resulting in the compression of small practice. Lately, this technological approach, using electro- blood vessels. This effect is reversible due to the recovery poration for chemotherapeutic drug delivery, is being trans- of the endothelial cells lining tumor blood vessels after 12 lated also into treatment of deep seated tumors. The tech- to 24 h [8]. nology is being modified and is in clinical testing for liver, The model was based on the findings in preclinical stu- bone, brain and colorectal tumors [3]. dies, which demonstrated that the application of electric The success in translation of electrochemotherapy is due to pulses to the tissue immediately induced a transient reduc- its well-known underlying mechanisms. As the principal tion of blood flow for up to 80%, which then restored [6]. mechanism of action, this drug delivery to tumor cells can This effect was observed in the area exposed to electric potentiate drug cytotoxicity up to 1000 times. Consequently, pulses, tumors or muscle, whereas it had no effect on distant the destruction of tumor cells activates the immune response, tissues. Furthermore, the effect was reproducible on the which by immunological cell death, contributes to the eradi- same tumors, when electric pulses were applied 24 h after cation of the remaining tumor cells and is currently in the the application of the first set of pulses [6]. The changes in spotlight of the research [2,4]. Furthermore, vascular effects tumor perfusion were demonstrated to correlate with of electrochemotherapy contribute significantly to the overall changes in tumor oxygenation, as measured by electron antitumor effectiveness, and are very important in the treat- paramagnetic resonance (EPR) technique [9]. ment of bleeding tumors, and in the treatment of tumors in Later on, in vitro study evaluated the effect of electric the vicinity of the major blood vessels [5]. Therefore, the aim pulses on cytoskeleton of the endothelial cells and on a of this review is to describe the vascular effects of electropo- model of endothelial cells lining. It was demonstrated that ration and electrochemotherapy, concerning its mechanisms electric pulses profoundly disrupt actin and microtubule © Springer Science+Business Media Singapore 2016 20 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_5 Tissue Reactions to Electroporation and Electrochemotherapy: Vascular Effects that have Implications in Tumor Treatment 21 cytoskeletal network and induce loss of cell-to-cell junctions leading to rounding of endothelial cells and consequently the leakiness of the endothelium [10]. The disruption of the microvascular structures in the tumors and abrogation of the blood flow was confirmed also by histological analysis of tumors, i.e. swollen endothelial cells were observed and stacked erythrocytes within the vessels [11]. However, the question, to what extent can the endothelial cells be affected by the applied electric field, remained. This was addressed by a mathematical model, which demonstrated that endothelial cells, lining of the vessels can be affected by the electric field and that they may be exposed to up to 40% higher electric field than the surrounding tumor cells [11]. The latest in vivo real time experiments on the normal and tumor vasculature in the dorsal window chamber model confirmed all the findings from before and added some new Fig.1 Model of the vascular effects in tumors of electroporation and elec- data. These real time experiments using fluorescently la- trochemotherapy that lead to vascular disrupting effect. With permission from Expert Review of Anticancer Therapy. belled dextrans of different sizes demonstrated the differential response of tumor and normal vessel to the application Vascular effects of electrochemotherapy with bleomycin of electric pulses. Both types of vessels became leaky after and cisplatin were investigated, and demonstrated to lead to application of electric pulses within 2 minutes. Vascular the almost complete abrogation of tumor perfusion lock occurred immediately in both types of vessels. The [9,11,14]. This was demonstrated at the level of the blood restoration of blood flow occurred in both types of vessels flow measurements, and tumor oxygenation. The effect was within 10 minutes after application of electric pulses, how- confirmed also by ultrasonography of tumors and staining ever, i.e. blood flow in normal vessels was completely res- of hypoxic areas in the tumors [11,16]. Therefore, these in tored in 1 hour, while in tumor vessels even after 24 h the vivo studies indirectly demonstrated that electrochemothe- restoration was not completed in all vessels [12,13]. rapy damages the tumor vasculature, which accounts for the vascular disrupting effect of electrochemotherapy, as it was III. VASCULAR EFFECTS proposed. The study in vitro also demonstrated that endo- OF ELECTROCHEMOTHERAPY thelial cells are susceptible to bleomycin and cisplatin, in the same or even higher degree as tumor cells [17] and that Parallel to studies, that investigated the effects of electric electrochemotherapy had also significant effect on cytoske- pulses on tumor vasculature, were the studies investigating letal structures, which faster occurred after electrochemo- the effects of electrochemotherapy. The rationale behind therapy (within 10 min) than after electroporation alone was, that if the endothelial cells in the tumors exposed to [18]. Furthermore, the histological analysis of tumors has electric pulses are permeabilized, than the drug molecules confirmed the findings, demonstrating the apoptotic endo- which are in the blood vessels, would enter the endothelial thelial cell death and disruption of the microcirculation of cells as well and should affect the vasculature in the tumors. the tumors [11]. Furthermore, if these endothelial cells would undergo apop- The latest intravital microscopy experiments have con- totic or necrotic cell death, the permeability of small vessels firmed the findings in vitro, and furthermore demonstrated within the tumors can be affected, and blood flow in such that changes in tumor blood vessels occurred faster after blood vessels should be abrogated permanently, leading to electrochemotherapy than after electroporation alone. In the vascular disrupting effect. This vascular disrupting ef- contrast to our proposed model, indicating that endothelial fect is well known, and is observed with some tubulin bind- cells start to die after 8˙hours after the treatment, these new ing agents, like combretastatins, and was described after data showed that the death of tumor endothelial cells starts irreversible electroporation as well [10,14,15]. within the first hour[13]. The model postulated that in addition to the vascular ef- The striking finding of our study, however, was a diffe- fects of electric pulses, the presence of the drug induces rential susceptibility between the tumor vasculature and endothelial cell death that leads to abrogation of tumor the surrounding normal vasculature [13]. The normal vascu- blood flow and consequently to vascular disrupting effect of lature was not affected by electrochemotherapy, which electrochemotherapy (Fig. 1) [8]. could be attributed to the difference in the mitotic rate of the

IFMBE Proceedings Vol. 53

22 G. Sersa and M. Čemažar normal and tumor endothelial cells: i.e. due to the higher effect is limited to tumor micovaculature and has only mi- rate of division the tumor endothelial cells, should be more nor and reversible effect on normal microvasculature. Fur- susceptible, similarly as the tumor vs. normal cells [19]. thermore, electrochemotherapy does not affect the function of bigger and major, well-structured blood vessel, and there- IV. CLINICAL OBSERVATIONS fore exerts antitumor effectiveness through vascular disrup- AND IMPLICATIONS tion of the tumor vessels. Due to non-thermal action of electrochemotherapy and lack of significant damage to Due to the vascular disrupting effectiveness of electro- bigger or major vessels, it is suitable also for the treatment chemotherapy, this therapy is now well established for the of tumors in the vicinity of those. treatment of bleeding cutaneous metastases [20-22]. Several published clinical cases describe effective and immediate ACKNOWLEDGMENT abrogation of bleeding in such metastases, which gives electrochemotherapy an advantage over other established Research was conducted in the scope of LEA EBAM ablative techniques [14]. Furthermore, the vascular disrupt- (French-Slovenian European Associated Laboratory: Pulsed ing effect, which leads to a cascade of cell death in neigh- Electric Fields Applications in Biology and Medicine) and boring tumor cells, may substantially contribute to the COST Action TD1104. The authors acknowledge the finan- overall effectiveness of electrochemotherapy. How much is cial support from the state budget by the Slovenian Re- the contribution of this indirect effect in addition to the search Agency (program no. P3-0003, project no. J3-5505). direct effect on tumor cells, is still not known and needs to be established. Certainly it varies according to the perfusion level or the vascularization of the tumors, therefore it will CONFLICT OF INTEREST be difficult to determine the contribution of the vascular disrupting effectiveness of electrochemotherapy. The authors declare that they have no conflict of interest. In contrast to irreversible electroporation, the effects of electrochemotherapy on the bigger vessels were not evaluated. Irreversible electroporation showed similar, vascular REFERENCES disrupting effects on miscrovasculature, as electrochemo- 1. Yarmush ML, Golberg A, Sersa G et al. (2014) Electroporation-based therapy, although on bigger, well-structured vessels with technologies for medicine: principles, applications, and challenges. strong muscular (media) layer, the scaffold of the vessels Annu Rev Biomed Eng 16: 295-320. remained intact, whereas the endothelial cell lining was 2. Mali B, Jarm T, Snoj M et al. (2013) Antitumor effectiveness of only transiently damaged [15]. Similar effects are also ex- electrochemotherapy: a systematic review and meta-analysis. Eur J pected for electrochemotherapy. Namely, in the clinical Surg Oncol 39: 4-16. 3. Miklavcic D, Sersa G, Brecelj E et al. (2012) Electrochemotherapy: study where colorectal liver metastases were treated with technological advancements for efficient electroporation-based treat- electrochemotherapy, 13 were afterward surgically removed ment of internal tumors. Med Biol Eng Comput 50: 1213-1225. and evaluated histologically, a significant reduction of via- 4. Calvet CY, Famin D, Andre FM et al. (2014) Electrochemotherapy ble tissue was observed. The typical response two months with bleomycin induces hallmarks of immunogenic cell death in mu- after electrochemotherapy was infarct like necrosis of the rine colon cancer cells. Oncoimmunology 3: e28131. 5. Miklavcic D, Mali B, Kos B et al. (2014) Electrochemotherapy: from tumor tissue and the surrounding tumor parenchyma, with the drawing board into medical practice. Biomed Eng Online 13. the encapsulation of the treated tissue (fibrous pseudocap- 6. Sersa G, Cemazar M, Parkins CS et al. (1999) Tumour blood flow sule on the border between the normal liver tissue and the changes induced by application of electric pulses. Eur J Cancer 35: electrochemotherapy treated area). In some of the tumor 672-677. samples, bigger vessels were not affected [23,24]. 7. Gehl J, Skovsgaard T, Mir LM (2002) Vascular reactions to in vivo electroporation: characterization and consequences for drug and gene However, 14 metastases that were located in the close vi- delivery. Biochim Biophys Acta 1569: 51-58. cinity of the major blood vessels of the liver were treated 8. Jarm T, Cemazar M, Miklavcic D et al. (2010) Antivascular effects of with electrochemotherapy only, since they were not amena- electrochemotherapy: implications in treatment of bleeding metastas- ble for surgical resection or radiofrequency ablation due to es. Expert Rev Anticanc 10: 729-746. their location. Nevertheless, very good antitumor effective- 9. Sersa G, Krzic M, Sentjurc M et al. (2002) Reduced blood flow and ness was observed, with 71% of complete responses, and no oxygenation in SA-1 tumours after electrochemotherapy with cispla- side effects recorded. [23,24]. tin. Brit J Cancer 87: 1047-1054. 10. Kanthou C, Kranjc S, Sersa G et al. (2006) The endothelial cytoskeleton as a target of electroporation-based therapies. Mol Cancer Ther 5: V. CONCLUSIONS 3145-3152. 11. Sersa G, Jarm T, Kotnik T et al. (2008) Vascular disrupting action of Electrochemotherapy has a well-defined vascular disrupt- electroporation and electrochemotherapy with bleomycin in murine ing effect that is clinically relevant. The vascular disrupting sarcoma. Brit J Cancer 98: 388-398.

IFMBE Proceedings Vol. 53

Tissue Reactions to Electroporation and Electrochemotherapy: Vascular Effects that have Implications in Tumor Treatment 23

12. Bellard E, Markelc B, Pelofy S et al. (2012) Intravital microscopy at 19. Folkman J (2003) Angiogenesis and apoptosis. Semin Cancer Biol the single vessel level brings new insights of vascular modification 13: 159-167. mechanisms induced by electropermeabilization. J Control Release 20. Gehl J, Geertsen PF (2000) Efficient palliation of haemorrhaging 163: 396-403. malignant melanoma skin metastases by electrochemotherapy. Mela- 13. Markelc B, Sersa G, Cemazar M (2013) Differential mechanisms noma Res 10: 585-589. associated with vascular disrupting action of electrochemotherapy: 21. Gehl J, Geertsen PF (2006) Palliation of haemorrhaging and ulcerated intravital microscopy on the level of single normal and tumor blood cutaneous tumours using electrochemotherapy. Ejc Suppl 4: 35-37. vessels. PLoS One 8: e59557. 22. Snoj M, Cemazar M, Srnovrsnik T et al. (2009) Limb sparing treat- 14. Sersa G, Miklavcic D, Cemazar M et al. (2008) Electrochemotherapy ment of bleeding melanoma recurrence by electrochemotherapy. in treatment of tumours. Eur J Surg Oncol 34: 232-240. Tumori 95: 398-402. 15. Jiang C, Davalos RV, Bischof JC (2015) A review of basic to clinical 23. Edhemovic I, Brecelj E, Gasljevic G et al. (2014) Intraoperative studies of irreversible electroporation therapy. IEEE T Bio-Med Eng electrochemotherapy of colorectal liver metastases. J Surg Oncol 110: 62: 4-20. 320-327. 16. Cor A, Cemazar M, Plazar N et al. (2009) Comparison between 24. Edhemovic I, Gadzijev EM, Brecelj E et al. (2011) Electrochemothe- hypoxic markers pimonidazole and glucose transporter 1 (Glut-1) in rapy: a new technological approach in treatment of metastases in the murine fibrosarcoma tumours after electrochemotherapy. Radiol On- liver. Technol Cancer Res T 10: 475-485. col 43: 195-202. 17. Cemazar M, Parkins CS, Holder AL et al. (2001) Electroporation of Author: Prof. Gregor Sersa human microvascular endothelial cells: evidence for an anti-vascular Institute: Institute of Oncology Ljubljana mechanism of electrochemotherapy. Brit J Cancer 84: 565-570. Street: Zaloska 2 18. Meulenberg CJ, Todorovic V, Cemazar M (2012) Differential cellular City: SI-1000 Ljubljana effects of electroporation and electrochemotherapy in monolayers of Country: Slovenia human microvascular endothelial cells. PLoS One 7: e52713. Email: [email protected]

IFMBE Proceedings Vol. 53

Nanosecond Pulses and Beyond – Towards Antenna Applications

Karl H. Schoenbach1 and Shu Xiao1,2 1 Frank Reidy Research Center for Bioelectrics, Old Dominion University, Norfolk, Virginia, USA 2 Department of Electrical and Computer Engineering, Old Dominion University, Norfolk, VA, USA

Abstract Modeling results and an increasing number of II. NANOSECOND PULSED ELECTRIC experimental results indicate that pulses with durations on the FIELD EFFECTS order or less than the charging time constant of the plasma membrane (which is on the order of 100 ns for cells in solution) The prediction that nanosecond pulses cause intracellular affect the plasma membrane in a different way than classical effects is based on a simple analytical, passive, and linear electroporation pulses. Such nanosecond Pulsed Electric Fields electrical model of the cell [1]. The results of this model (nsPEF) have been shown to also target subcellular structures indicate that for pulses on the order of the charging time and possibly proteins. When the pulse duration is shortened constant of the plasma cell membrane, τm, the applied into the picosecond range (psPEF) the dielectric, rather than electric field reaches into the interior of the cell and the resistive, properties of the outer medium and cytoplasm therefore affects cell substructures, and possibly cell begin to determine the coupling between electric field and cell functions. Since typical charging time constants of cells membranes. Besides entering a new field of electric field - cell with dimensions of 10 µm are on the order of a hundred interactions, the use of picosecond pulses allows the application nanoseconds, submicrosecond pulses were expected to of wideband antennas, rather than invasive electrodes, for the cause different effects on cells than pulses with durations of delivery of psPEF to tissue. microseconds and longer. Keywords nanosecond pulsed electric fields, picosecond The first experimental study on the effect of nanosecond pulses, wideband antennas, cell death, neural stimulation. pulses on the intracellular structures indeed showed that such pulses affect membranes of subcellular structures [1]. These studies, where 60 ns pulses of 60 kV/cm amplitude I. INTRODUCTION were applied to eosinophils seemed to indicate that only the granules in these cells were affected and the plasma The electric fields that are required to achieve membrane was not permeated. However, subsequent studies electroporation depend on the duration of the applied pulse. showed that nanosecond pulses caused poration of the Typical pulse parameters range from tens of milliseconds plasma membrane, however with pore sizes which, at least with amplitudes of several 100 V/cm to pulses of a few for single shots and not exceedingly high electrical field microseconds and several kV/cm. More recently, the pulse intensity, prevented the uptake of larger ions. duration range has been extended into the nanosecond range. In line with the predicted effect of nanosecond pulses on Pulse durations are as brief as several nanoseconds, with subcellular structures it was shown that an increase in pulse amplitudes as high as 300 kV/cm for these short pulses. calcium was not only obtained from extracellular sources The effects of such nanosecond Pulsed Electric Fields through the nanoporated plasma membranes, but from (nsPEF) have been shown to reach into the cell interior. intracellular sources such as the endoplasmic reticulum as The need to generate ever higher amplitude electrical well [2,3]. NsPEFs also cause platelet calcium release and pulses when moving further into the ultrashort pulse regime aggregation in human platelet rich plasma and washed is technically challenging. However, modern pulsed power platelets [4], an effect which is now successfully used in technology, the technology of generating extremely high wound healing. power electrical pulses, is now so advanced that pulses An important result of bioelectric studies with nanosecond down to 200 ps duration and amplitudes of tens of kV can pulses was the generation of apoptosis when high amplitude be produced. The use of such picosecond pulses allows us pulses were applied [5]. First studies indicated the potential of to enter a range of electrical field - cellular membrane such pulses for the controlled elimination of unwanted tissue. coupling with new bioelectric applications. By applying With both treatment regimens and electrode designs such ultrashort pulses it also becomes possible to use optimized, these studies have now led to the development of wideband antennas, rather than direct contact electrodes, to therapeutic methods for cancer treatment, particularly deliver the pulses to tissue. treatment of skin cancer [6].

© Springer Science+Business Media Singapore 2016 24 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_6 Nanosecond Pulses and Beyond – Towards Antenna Applications 25

III. FROM NANOSECOND TO The technical challenge is to connect the pulse generator PICOSECOND PULSES to the load, for example, cell suspension in a cuvette. The extremely short pulse duration requires designs of the pulse When the pulse duration is reduced from nanoseconds to delivery system such that pulse dispersion and the picoseconds, the dielectric properties, rather than the resistive capacitance of the load is minimized [13], and the load is characteristics of the media determine the electric field matched as well as possible to the source in order to distribution. Generally, in modeling the effect of pulsed electric fields on cells, the capacitive component of minimize reflections of the pulse [12]. An example of such cytoplasm and medium are neglected. These are assumptions a pulse delivery system with a cylindrical cuvette is shown that are only valid for a pulse duration that is long compared in Fig. 1. In order to insure minimum reflection, the sample to the dielectric relaxation time (ε/σ) of cytoplasm and chamber (exposure chamber) was designed to ensure an medium, where ε is the permittivity and σ the conductivity. impedance as close as possible to 50 Ω (the impedance of For pulses short compared to the dielectric relaxation time of the source) while providing a homogeneous electric field the cytoplasm and the medium, the coupling of electric field across the sample. with the cell is defined by the permittivity of the cell components and the cell environment [7]. For a membrane with a relative dielectric constant of 8, the electric field in the membrane is approximately 10 times higher than the electric field in the medium, which has a dielectric constant of about 80. The electric field then acts directly on the lipid bilayer and the embedded membrane proteins, rather than causing charging of the membrane, and, if sufficiently strong, can cause direct and instant conformational changes in membrane proteins. Another bioelectric effect which becomes relevant (when the pulse duration is reduced into the lower nanosecond and subnanosecond range) is heating of the membranes through dielectric relaxation. Dielectric spectroscopy on multilamellar liposomes has shown dispersion frequencies related to the diffuse thermal rotational motion of the phosphatidylcholine headgroup and to bound water [8]. Dielectric relaxation in the frequency range of 10 to 100 MHz leads to an increased energy deposition in the plasma membrane in this frequency range [9]. Fig. 1 (left) Schematics of the connector between source and exposure Considering the frequency distribution of square wave pulses, chamber. The chamber has a gap distance and radius of 1.8 mm and 2.4 such dielectric relaxation effects are expected for nanosecond, mm, respectively; (right) Voltage across the exposure chamber [12]. but even more for picosecond pulses. This energy deposition leads to a spike in membrane temperature and possibly to V. BIOELECTRIC EFFECTS OF PICOSECOND changes in the lipid bilayer [10]. PULSED ELECTRIC FIELDS (PSPEF) The electric field applied to the membrane for IV. PICOSECOND PULSE GENERATORS subnanosecond pulse-induced electroporation is, according Two types of picosecond pulse generators have been to molecular dynamics calculations, greater than 5 MV/cm used. One is based on the use of high pressure spark gaps. [14]. This is independent of the type of coupling, which, for Such picosecond pulse generators have the advantage that pulses of approximately 1 ns, are just at the borderline they can be built in-house, and that the voltage amplitude between membrane charging through conductive media and can reach several hundred kV [11,12]. However, the through displacement current in a principally capacitive cell repetition rate is relatively low, generally less than 10 Hz. A and medium. Although the applied electric field need not be second type of picosecond pulse generator utilized solid that high, due to the amplification of the electric field in the state closing switches, so-called FID switches. The voltage membrane when charged through cytoplasm and medium, amplitude of such commercially available pulse generators or due to the continuity of the displacement current density is listed as up to 10 kV for subnanosecond pulse generators at the low permittivity membrane, the electric fields applied at pulse risetimes down to less than 100 ps [FID GmbH, to the cell suspension still need to exceed 500 kV/cm. These Burbach, Germany]. Besides the extremely fast risetime, are values which can only be achieved with special pulsed their advantage is the high repetition rate, up to 100 kHz. power generators.

IFMBE Proceedings Vol. 53

26 K.H. Schoenbach and S. Xiao

Such an 800 ps pulse generator, based on a waveguide shown in Fig. 2 (left). When a fast rising pulse is applied to Marx Generator, is described in reference [11]. The uptake the antenna, the pulse structure in the target focal of trypan blue when electric fields of up to 1 MV/cm were point consists of a prepulse, which resembles the applied applied to B16 (murine melanoma) cells was recorded. For electric pulse, and the main pulse, which has the shape of pulse amplitudes of 550 kV/cm, approximately 50% of the the differentiated applied pulse. This means the faster cells took up trypan blue right after pulsing, whereas only the risetime of the applied pulse, the shorter the pulse at 20% took it up after 1 h. This indicates that the plasma the target. membrane in a majority of the cells affected by the pulses recovers with a time constant of about 1 h. The cells that show trypan blue uptake after this time suffer cell death through apoptosis. The fact that the results matched a scaling law for nanosecond pulses [15] indicates that the effect of 800 ps pulses is still likely dominated by membrane charging, which means, by the conductivities of medium and cytoplasm, rather than by their permittivities. In subsequent experimental studies the pulse duration was shortened to 200 ps. Also, in order to shift the voltage range to amplitudes which are more reasonable for medical applications, studies on thermally assisted bioelectric effects Fig. 2 (left) An antenna with prolate spheroidal reflector; (right) were performed. By moderately heating the cell suspension The schematics of the analytical pulse shape at the focus when the utilizing the Joule heat generated by applied picosecond antenna is fed with a step-function pulse [18]. pulses at a high repetition rate of 10 kHz, it was possible to reduce the cell viability to 20% of its initial value with The reason to use pulses with a risetime of picoseconds, applied electric fields as low as 20 kV/cm [13]. Similarly, by which provide pulses of the same duration as the risetime, is heating the cell suspension for a few minutes to temperatures the size of the focal volume, or the spatial resolution which of up to 47 0C, cell death increased considerably when 200 ps can be achieved with such an antenna. For narrowband pulses at 90 kV/cm amplitude were applied [12]. radiation characterized by a certain wavelength, λ, the spot Another research direction was the study of bioelectric width W is given as r0λ/D indicating that a large aperture, D, membrane effects which require lower electric fields than for a short focal distance, r0, permits a small focus. For those needed for inducing cell death. Studies on primary rat wideband radiation, as is the case of ultrashort pulses, for an hippocampus neurons exposed to subnanosecond pulses were estimate of the spot width we can use the cut-off wavelength performed using a patch clamp method, where a constant of the Fourier spectrum of a square wave pulse, which is current was injected to the neurons before the pulses were given as 2.27 τ c/ε, τ being the pulse duration, c the speed of applied. It could be shown that pulsed electric fields of just 10 light in vacuum, and ε the relative permittivity of the kV/cm (100 pulses at a repetition rate of 500 Hz) caused medium. The shorter the pulse duration and the higher the membrane depolarization, which culminated in an action relative permittivity of the medium, the smaller is the spot potential at injection currents of – 80 pA [16]. More recent size, and consequently the higher is the spatial resolution. For studies on different cell types indicate that even a single a medium with an ε of 80 (water), and for a prolate spheroidal pulse, at electrical fields far below that required for reflector with a D/r0 of approximately one, the pulse duration electroporation, can cause an increase in cytosolic calcium. needs to be in the hundred picosecond range to obtain a The results indicate that such picosecond pulses cause long- spatial resolution of less than a centimeter. lasting opening of voltage gated calcium channels [17].

VI. FROM INVASIVE PULSE DELIVERY SYSTEMS TO ANTENNAS

A practical reason for entering the sub-nanosecond pulse range is the possible use of antennas as pulse delivery systems instead of electrodes. These antenna systems studied at the Frank Reidy Research Center for Bioelectrics are mainly based on the use of a prolate spheroidal reflector, where the pulse is launched from one focal point and Fig. 3 (left) Schematics of a combined reflector-lens system; (right) reflected into a second focal point [18]. Such an antenna is Photograph of the dielectric lens

IFMBE Proceedings Vol. 53

Naanosecond Pulses and Beyond – Towards Antenna Applications 27

However, reducing the pulse duration even further causes 4. Zhang, J., Blackmore, P.F.,Hargrave, B.Y., Xiao, S., Beebe ,S.J., and a problem: the absorption of tissue increases with reduced Schoenbach, K.H. (2008) The Characteristics of Nanosecond Pulsed Electrical Field Stimulation on Platelet Aggregation in Vitro, Arch. wavelength, preventing the electromagnetic radiation to Biochem. Biophys. 471:240-248. reach deeper lying tissues. Other problems which need to be 5. Beebe, S.J., Fox, P.M., Rec, L.J., Willis, L.K., and Schoenbach, K.H. addressed are the impedance mismatch at the skin (~80% (2003) Nanosecond, High Intensity Pulsed Electric Fields Induce reflection, between air and the target’s dielectric constants). Apoptosis in Human Cells, FASEB J., 17, No. 1493Nuccitelli R, Wood R, Kreis M, Athos B, Huynh J, Lui In order to improve the matching between the antenna and 6. K, Nuccitelli P, Epstein EH Jr. (2014) First-in-human trial of the tissue, special lenses, made of layers with increasing nanoelectroablation therapy for basal cell carcinoma: proof of permittivities towards the target can be used, as shown method. Exp Dermatol 23:135-137 schematically in Fig. 3 (together with the prolate spheroidal 7. Schoenbach, K.H., Xiao, S., Joshi, R.P., Camp, J.T., Heeren, T., Kolb, J.F., Beebe, S.J. (2008) The Effect of Intense Subnanosecond reflector). The use of a lens not only improves matching, Electrical Pulses on Biological Cells, IEEE Trans. Plasma Science 2 but also allows for a better resolution: an εr times reduction 36:414-424 in the spot size. 8. Kloesgen, B., Reichle, C., Kohlsmann, S., Kramer, K.D. (1996) Dielectric Spectroscopy as a Sensor of Membrane Headgroup Mobility and Hydration, Biophys. J. 71:3251-3260. VII. CONCLUSION 9. Kotnik, T. and Miklavcic, D. (2000) Theoretical Evaluation of the Distributed Power Dissipation in Biological Cells Exposed to Electric The study of bioelectric effects is a rather recent Fields, Bioelectromagnetics 21:385-394 undertaking, starting less than a decade ago. The 10. Croce, R.P., De Vita, A., Pierro, V., and Pinto, I.M. (2010) A Thermal Model for Pulsed EM Field Exposure Effects in Cells at experimental studies were initially focused on determining Nonthermal Levels, IEEE Trans. Plasma Science, 38:149-155. the pulse parameters for cell death induction, more recently 11. Heeren, T., Camp, J.T., Kolb, J.F., Schoenbach, K.H., Katsuki, S., on cell activation. In parallel to these bioelectric studies, and Akiyama, H., (2007) 250 kV Subnanosecond Pulse Generator efforts are underway to design and build high power with Adjustable Pulsewidth,” IEEE Trans. Diel. Electr. Insul. 14: 884-888. picosecond pulse generators, picosecond pulse delivery 12. Camp, J.T., Jing, Y., Zhuang, J., Kolb, J, Beebe, S.J., Song, J., Joshi, devices, and focusing wideband antennas. Modeling of R.P., Xiao, S., and Schoenbach, K.H. (2012) Cell Death Induced by psPEF effects on lipid bilayer membranes was performed by Subnanosecond Pulsed Electric Fields at Elevated Temperatures, means of molecular dynamics. The results of these studies IEEE Trans. Plasma Science, 40:2334-2347 13. Xiao, S., Guo, S., Vasyl, N., Heller, R., and Schoenbach, K.H. (2011) indicate possible antenna-based applications. The most Subnanosecond Electrical Pulses Cause Membrane Permeabilization promising seems to be neural stimulation, an effect which and Cell Death, IEEE Trans. Biomedical Engineering, 58:1239-1245. has been explored with 200 ps long psPEF. Other 14. Levine, Z.A., and Vernier, P.T. (2010) Life cycle of an electropore: applications include treatment of tumors. However, with field-dependent and field-independent steps in pore creation and annihilation, J. Membr. Biol. 236:2736. present pulsed power and antenna technology, this 15. Schoenbach, K.H., Baum, C.E., Joshi, R.P., and Beebe, S.J. (2009) A application still requires thermal assistance. Scaling Law for Membrane Permeabilization with Nanopulses, IEEE Trans. Dielectrics and Electrical Insulation, 16:1224-1235. 16. Xiao, S., Pakhomov, A., Guo, F., Polisetty, S. Schoenbach, K.H. CONFLICT OF INTEREST (2013) Neurostimulation Using Subnanosecond Electrical Pulses,SPIE Proc., Vol. 8585, 0M. The authors declare that they have no conflict of interest. 17. Pakhomov, A.G., Semenov, Iu., Gianulis, E., Xiao, S., Schoenbach, K.H., and Pakhomova, O.N. (2015) Stimulation and permeabilization REFERENCES by nano- and subnanosecond electric pulses, this conference 18. Xiao, S., Altunc, S., Kumar, P., Baum, C.E., and Schoenbach, K.H. 1. Schoenbach, K.H., Beebe, S.J., and Buescher, E.S. (2001) (2010) A Reflector Antenna for Focusing in the Near Field, IEEE Intracellular Effect of Ultrashort Electrical Pulses, J. Antennas and Wireless Propagation Letters, 9:12-15. Bioelectromagnetics, 22:440-448 2. Vernier, P.T., Sun, Y., Marcu, L., Salemi, S., Craft, C.M., and Gundersen, M.A. (2003) Calcium Bursts Induced by Nanosecond Author: Karl H. Schoenbach Electrical Pulses, BBRC, 310: 286-295. Institute: Frank Reidy Research Center for Bioelectrics 3. Beebe, S.J., White, J.A., Blackmore, P.F. Deng, Y., Somers, K., and Old Dominion University Schoenbach, K.H, (2003) Diverse Effects of Nanosecond Pulsed Street: 4211 Monarch Way, Suite 300 Electric Fields on Cells and Tissues”, DNA and Cell Biology, Country: USA 22:785-796 Email: [email protected]

IFMBE Proceedings Vol. 53

Optimal Irreversible Electroporation Techniques in the Treatment of Locally Advanced Liver and Pancreatic Cancer

Robert C.G. Martin University of Louisville School of Medicine, Department of Surgery, 315 East Broadway, Louisville, KY, USA

Abstract Initially diagnosed stage 3 (locally advanced) all unaffected and left healthy by this treatment [1]. This Pancreatic adenocarcinoma (LAPC) remains an aggressive expands the scope of treatment of lesions near major vascu- tumor with an overall poor prognosis (current median survival lar and biliary structures when compared to conventional chemotherapy and radiation therapy of 9-12 months). Im- thermal injury ablative techniques. The major disadvantage provement in survival can be achieved in the small percentage is the need for general anesthesia (deep paralysis) for its that can undergo an R0 resection or where all macroscopic tumor can be cleared by local ablation. Long duration (>4-6 energy delivery [3]. IRE can be performed open, laparos- months) chemoradiotherapy is not tolerated by all patients and copically, or percutaneously. Repeated reports have demon- still fails to prolong survival alone. Neoadjuvant treatment also strated theadvantages of IRE against other thermal ablation has limited results on pain control or tumor downstaging since techniques include no heat sink effect, tumor specific im- most current modalities do not shrink or downsize the tumor. munological reactions, little impact on the collagen network In recent years, there has been a growing interest in the use of within treated tissues, and the potential to ablate tumor local ablative therapy for the treatment of nonresectable tu- tissues near large vessels [4]. mours in various organs. Ablation techniques are based on In contrast to other physical ablation modalities IRE has direct application of chemical, thermal, or electrical energy to a distinctive advantage: it does not involve thermal damage a tumor, which leads to cellular necrosis without removal of the tumor. With ablation, local control, and relief from symp- and therefore is not influenced by the so-called heat sink toms can be obtained in the majority of the patients when effect that appears in the vicinity of large vessels. This im- appropriate patient selection and technique is utilized. LAPC plies that IRE has the potential to prevent tumor recurrence has been treated by various ablation techniques in the last few close to the vessels while preserving the surrounding years with promising results. We present the current status of healthy tissue [5]. local ablative therapies in the treatment of LAPC and investigate on the efficiency and future trends. II. LOCAL TISSUE FACTORS THAT AFFECT IRE Keywords IRE, Locally Advanced Liver Tumors, Locally Advanced Pancreatic Cancer. When IRE is delivered appropriately it affects the target tissue that is bracketed by the IRE probes and does not damage the surrounding structures, PROVIDED the IRE I. INTRODUCTION probes where placed atraumatically. Proteins, the extracellular matrix, and critical structures such as blood vessels and Irreversible Electroporation (IRE) is a biophysical phe- nerves are not affected and left intake by this treatment [6]. nomenon, which is becoming increasingly used as a defini- Authors have demonstrated that blood vessels and bile ducts tive tumor ablation technique for locally advanced liver and are not permanently damaged by IRE [6, 7]. The reason for pancreatic cancers. The exact mechanism of IRE is based on this, it that higher collagenous connective tissue and elastic electric field pulses being delivered across the target tissue, fiber contents lack a normal cellular membrane that is ef- which forms permanent pores in the cell lipid bilayer mem- fected by IRE. The electric currents of IRE that affect the brane, thereby increasing its permeability and inducing cell cell membrane may travel through the gap junctions from death [1]. This ablation technique takes advantage of the one cell to the other without changing or disrupting the electrical potential gradient that exists across cell mem- integrity of the smooth muscle cell membrane [2]. branes. This technique was reported as early as the 1750s, Preclinical studies have demonstrated the potential effi- however its use for medical purposes dates back only 30 cacy of IRE in ablation of HCC in an animal model [8]. years [2]. Its advantages compared to conventional thermal Additional pre-clinical work by Bower et al [9] and Char- ablation techniques are its non-thermal injury delivery me- pentier et al [7, 10] have demonstrated that the electrical chanism and was first developed in conjugation with che- pulses do damage the cellular membrane of normal tissue motherapy. When properly applied, theoretically, it only (e.g. blood vessel endothelium, biliary epithelium), however affects the target tissue. Proteins, the extracellular matrix, the collagenous structures remain intact. Intact adventitia and critical structures such as blood vessels and nerves are and laminae is visible at 2 days with no smooth muscles

© Springer Science+Business Media Singapore 2016 28 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_7 Optimal Irreversible Electroporation Techniques in the Treatment of Locally Advanced Liver and Pancreatic Cancer 29 cells present [11]. The endothelium is largely repopulated at vary and their particular skills should be addressed before 2 days with the smooth muscle repopulation at 2 weeks. performing IRE. Some factors to consider before perform- This slow method of repopulation has been demonstrated in ing IRE are: the size of ablation zone, the number of probes pre-clinical studies [9] to allow for vital structures to remain needed for the procedure, the distance between the probes, intact, patent, and viable out to at least one month in both and the length of the active electrode tip. With a set plan of acute and chronic animal studies as well as in human evalu- the procedure there will be less of a chance for complica- ation [12, 13]. tions or mistakes. After gaimning access to the liver the probes are inserted using an continuous ultrasound to ensure III. TECHNIQUE OF PERFORMING IRE IN LIVER accurate placement, but also to avoid mechanical damage to FOR TUMORS WITH VASCULAR PROXIMITY the hepatic inflow, bile ducts or hepatic outflow based on the lesion location. Spacers that connect to the probes help The clinical indication for IRE of liver tumors must be fully encompass each tumor within the dimensions of the made based on 1) Tumor Biology, 2) Tumor <4cm in size, probes. Once the probes are in the correct positions, the 3) Tumor within 5mm or less of vital structure that needs to electrical pulses are delivered from the NanoKnife system be spared, and 4) the ability to undergo general endotracheal [17]. This can last from 10-60 min [18]. Once the electrical anesthesia. IRE is not a replacement for Microwave abla- pulses are finished, the patient is closed and is sent into tion or Radiofrequency ablation. The current commercially recovery. Detailed procedural related steps have been available system consists of a computer controlled pulse presented for hepatic tumors [19]. generator that delivers a maximum 3,000 volts between IRE of the liver has well defined criteria based on Abla- each probe pair based on the number and spacing configuration recurrence being defined as persistent viable tumor as tion of the IRE probes. A minimum of 90 pulses must be defined by dynamic imaging in comparison to pre-IRE scan delivered which last from 20 to 100 μsec each. The most or tissue diagnosis. Ablation success was defined as the common pulse length is 70 to 90 μsec, based on the degree ability to deliver the planned therapy in the operative room of electrical resistance that is encountered. Patient selec- and at 3 months to have no evidence of residual tumor on tion is of paramount importance. A recent (<1 month from cross-sectional imaging of treating-team’s choice such as treatment), multi-dimensional thin cut (0.7mm to 1mm) CT CT, MRI or PET (if they had a preoperative PET avid scan) scan or contrast enhanced dynamic MRI must be performed. [6, 18]. From those images, a three dimensional reconstruction can be performed in order to plan 1) # of IRE needles required, IV. CLINICAL RESULTS OF IRREVERSIBLE 2) Needle trajectory, and 3) Access – Open, Laparoscopic, ELECTROPORATION FOR HEPATIC or Percutaneous. The tumor dimensions are input into the MALIGNANCIES IRE pulse generator, which will recommend the number and possible spacing of probes needed to create the desired All patients were followed on an IRB approved human electroporation zone based on a mathematical algorithm. subjects protocol. Optimal Probe spacing is critical to the safty and efficacy of The initial IRE use for liver was reported by Martin et al the device with the optimal spacing being 1.5cm to 2.3cm. with 45 patients undergoing 51 total IRE procedures [20]. Spacing less than 1.5cm can lead to ineffective electropora- Lesions were in proximity to vital structures in 40 (88%) tion (called Reversible) or thermal damage (defined as >54 patients. Initial success was achieved in 50 (98%) treat- degrees C for >10sec) [14]. Probe spacing too far - >2.3cm ments. Five patients had 9 adverse events, with all compli- will lead to ineffective electroporation. This precision of cations resolving within 30days. LRFS at 3, 6, and 12 spacing places the burden on the physician to ensure high months was 97.4%, 94.6%, and 59.5%. There was a trend quality Ultrasound is used during needle placement toward higher recurrence rates for tumors over 4cm (HR and to document this spacing. The probes themselves are 19 3.236, 95% CI 0.585-17.891; p=0.178). They were able to gauge diameter and radio-opaque to aid in intraprocedural conclude that IRE was a safe treatment for hepatic tumors in identification of the probe tip [15]. Consideration of proximity to vital structures. A Significant inflection point intra-operative navigation systems should also be consi- occurred for all tumors greater than 3cm with higher local dered for physicians who are not expert in Liver Ultrasound recurrence rates see. Thus initial use for new user was [16]. The pulses delivered from the NanoKnife system are recommend to start for hepatic tumors in proximity to vital synchronized with the patients ECG to avoid cardiac structures that were <3cm in size. arrhythmias [6]. The true technical reports of how and tumor locations Every patient is unique in the care and service they need were appropriate for IRE especially within the hepatic hi- to have a successful treatment. The abilities of each surgeon lum were published my Martin et al [19]. The pulse voltag-

IFMBE Proceedings Vol. 53

30 R.C.G. Martin es and duration are based on preclinical studies [7, 9, 10]. Pancreatic ductal adenocarcinoma is still one of the most Treatment planning is based on three dimensional preopera- aggressive cancers and is the fourth most frequent tumor tive imaging with CT scanning in which the tumor dimen- related cause of death in the Western world [21] Locally sions and location of surrounding structures are measured. advanced disease is difficult to control, and limited im- From the preoperative scan, the tumor dimensions are input provement in outcomes have been achieved in the last 30 into the pulse generator, with a set planned margin. Multiple years despite the advances of diagnostic modalities and monopolar probes are used (maximum of 6), with greater therapeutic options. For all stages combined, the 1-year numbers of probes needed for larger ablation zones. The survival rate is 20 %, and the overall 5-year survival rate maximum effective probe spacing can vary from only from has remained dismally poor at 5 % [22]. Complete surgical 1.4 to 2.2 cm apart. If the probe spacing is either <1.4 or resection remains the only curative treatment for pancreatic >2.2 cm, then the effectiveness of the electroporation is cancer. The advanced T-stage of pancreatic adenocarcinoma reduced and will lead to an incomplete ablation. The probes is defined according to the involvement of the superior themselves are 19 gauge in diameter and radio-opaque to mesenteric artery, the celiac axis, long segment portal vein aid in intraprocedural identification of the probe tip. The occlusion or their combination on cross-sectional imaging maximum probe exposure utilized in liver IRE is 2 to 2.5 [23, 24]. cm exposure. Optimal technique requires the user to place In the future, IRE has the potential to be an efficacious the needle along the longest axis of the tumor – most com- cancer treatment. More complex treatments of larger lesions monly the caudal to cranial plane (coronal plan) and then and lesions with greater vascular involvement will become perform sequential pullbacks to achieve both cranial and available with more research and clinical trials. caudal margins. It is recommended not to attempt to per- Martin et al reported on a larger study of 54 patients who form the “over-lapping ablation” technique that was first underwent combination of chemotherapy, chemoradiation popularized with RFA, secondary to the fact that IRE thera- therapy with consolidative IRE in comparison to a control py induces artifact and human error to ensure precise spac- group of chemotherapy/chemo-radiation therapy for LAPC ing would lead to a greater incidence or ineffective therapy [25]. All patients were confirmed Stage 3 LAPC based on (i.e. reversible electroporation). staging CT and/or MRI due to encasement of the superior A final review from Martin et al from a prospective IRB mesenteric artery, celiac axis, or long segment occlusion of approved evaluation of 107 consecutive patients from 7 the SMV/PV. IRE was performed through an open supine institutions with tumors that had vascular invasion were midline incision or in a laparoscopic fashion. After a me- treated with IRE from 5/2010 to 1/2012. Local advanced dian follow-up time of 15 months, 15 of the 54 patients tumors were defined as primary tumor with <5mm from appeared to have local disease recurrence. The adverse major vascular structure based on pre-operative dynamic events that were IRE-related were two cases of bile leakage imaging or intra-operative criteria. IRE as utilized in local- and two cases of duodenal leakage. However, the duodenal ly advanced tumors in the liver (N=42, 40%) and pan- leaks occurred after the removal of a duodenal stent and creas(N=37, 35%), with a median number of lesions being 2 placement of the IRE needle. The 90 day mortality in the with a mean target size of 3 cm. IRE attributable morbidity IRE patients was 1(2%). In a comparison of IRE patients to rate was 13.3% (total 29.3%) with high-grade complications standard therapy we have seen an improvement in Local seen in 4.19%(total 12.6%). No significant vascular compli- progression free survival (14 vs 6 months, P=0.01), Distant cations were seen, and of the high-grade complications, progression free survival (15 vs 9 months, p=0.02), and bleeding(2), biliary complications(3) and DVT/PE(3) were overall survival (20 vs 13 months, p=0.03). The investiga- the most common. Complications were more likely with tors concluded that IRE as a consolidative therapy of locally pancreatic lesions (p=0.0001) and open surgery(p=0.001). advanced pancreatic tumors remains safe. In the appropriate Calculated local recurrence free survival was 12.7 months patient who has undergone standard induction therapy for a with a median follow up of 26 months censured at last fol- minimum of 4 months, IRE can achieve greater local pallia- low up. The tumor target size was inversely associated with tion and potential improved overall survival when compared recurrence free survival (b=0.81, 95% CI: 1.6 to 4.7, p val- to standard chemoradiation-chemotherapy treatments. ue=0.02) but this did not have a significant overall survival The established technique for IRE of LAPC has been impact. They were able to conclude that IRE represents a well published and described. Recent from Martin et al, novel therapeutic option in patients with locally advanced reported on the optimal technique for both the LAPC of the tumors involving vital structures that are not amenable to pancreatic head and LAPC of the pancreatic neck/body [26, surgical resection. Acceptable to high local disease control 27]. Representative case would involve a patient who and the long local relapse free survival can be achieved with presents a LAPC of the pancreatic head who has been this therapy in combination with other multi-disciplinary treated with induction chemotherapy, who now has a mass therapies. of <3.5cm in size with clear vascular involvement. Given

IFMBE Proceedings Vol. 53

Optimal Irreversible Electroporation Techniques in the Treatment of Locally Advanced Liver and Pancreatic Cancer 31 the size of the tumor, at least 4 needles are placed in a 12. Martin RCG, McFarland K, Ellis S, Velanovich V (2012) Irreversible bracketing fashion, covering the entire tumor and the vital electroporation therapy in the management of locally advanced pancreatic adenocarcinoma. J Am Coll Surg 215:361 9. structures, which in this case would include the SMA, 13. Martin RCG, McFarland K, Ellis S, Velanovich V (2013) Irreversible SMV, and the bile duct. electroporation in locally advanced pancreatic cancer: potential im- Thus in conclusion, local consolidative therapy for proved overall survival. Ann Surg Oncol 20 Suppl 3:S443 9. LAPC can be effective in local disease control when per- 14. Dunki-Jacobs EM, Philips P, Martin RCG (2014) Evaluation of thermal injury to liver, pancreas and kidney during irreversible elec- formed in collaboration with a multi-disciplinary team and troporation in an in vivo experimental model. Br J Surg 101:1113 21. appropriate sequencing of all three therapies – chemothera- 15. Cannon R, Ellis S, Hayes D, et al (2013) Safety and early efficacy of py, radiation therapy, and IRE. irreversible electroporation for hepatic tumors in proximity to vital structures. J Surg Oncol 107:544 9 16. Agle S, Philips P, VanMeter T, et al (2014) Intra-operative Naviga- CONFLICT OF INTEREST tion of a 3Dimensional Needle Localization System for Precision of Irreversible Electroporation Needles in Locally Advanced Pancreatic The author is a paid consultant for Angiodynamics. Cancer. J Surg Res 186:643 644. 17. Dunki-Jacobs EM, Philips P, Martin RCG (2014) Evaluation of resistance as a measure of successful tumor ablation during irreversi- REFERENCES ble electroporation of the pancreas. J Am Coll Surg 218:179 87. 18. Philips P, Hays D, Martin RCG (2013) Irreversible electroporation 1. Davalos R V, Mir ILM, Rubinsky B (2005) Tissue ablation with ablation (IRE) of unresectable soft tissue tumors: learning curve irreversible electroporation. Ann Biomed Eng 33:223 31 evaluation in the first 150 patients treated. PLoS One 8:e76260. 2. Lee EW, Chen C, Prieto VE, et al (2010) Advanced hepatic ablation 19. Martin RCG (2013) Irreversible Electroporation: a Novel Option for technique for creating complete cell death: irreversible electropora- Treatment of Hepatic Metastases. Curr Colorectal Cancer Rep 9:191 197. tion. Radiology 255:426 33. 20. Cannon R, Ellis S, Hayes D, et al (2013) Safety and early efficacy of 3. Cannon R, Ellis S, Hayes D, et al (2013) Safety and early efficacy of irreversible electroporation for hepatic tumors in proximity to vital irreversible electroporation for hepatic tumors in proximity to vital structures. J Surg Oncol 107:544 9. doi: 10.1002/jso.23280 structures. J Surg Oncol 107:544 9. 21. Spinelli GP, Zullo A, Romiti A, et al (2006) Long-term survival in 4. Guo Y, Zhang Y, Nijm GM, et al (2011) Irreversible electroporation metastatic pancreatic cancer. A case report and review of the litera- in the liver: contrast-enhanced inversion-recovery MR imaging ap- ture. JOP 7:486 91. proaches to differentiate reversibly electroporated penumbra from ir- 22. Jemal A, Thomas A, Murray T, Thun M Cancer statistics, 2002. CA reversibly electroporated ablation zones. Radiology 258:461 8. Cancer J Clin 52:23 47. 5. Rubinsky B, Onik G, Mikus P (2007) Irreversible electroporation: a 23. Callery MP, Chang KJ, Fishman EK, et al (2009) Pretreatment as- new ablation modality--clinical implications. Technol Cancer Res sessment of resectable and borderline resectable pancreatic cancer: Treat 6:37 48. expert consensus statement. Ann Surg Oncol 16:1727 33. 24. Varadhachary GR, Tamm EP, Abbruzzese JL, et al (2006) Borderline 6. Martin RCG, Philips P, Ellis S, et al (2014) Irreversible electropora- resectable pancreatic cancer: definitions, management, and role of tion of unresectable soft tissue tumors with vascular invasion: effec- preoperative therapy. Ann Surg Oncol 13:1035 46. tive palliation. BMC Cancer 14:540. 25. Martin RCG, McFarland K, Ellis S, Velanovich V (2013) Irreversible 7. Charpentier KP, Wolf F, Noble L, et al (2011) Irreversible electropo- electroporation in locally advanced pancreatic cancer: potential im- ration of the liver and liver hilum in swine. HPB 13:168-73. proved overall survival. Ann Surg Oncol 20 Suppl 3:S443 9. 8. Guo Y, Zhang Y, Klein R, et al (2010) Irreversible electroporation 26. Martin RCG (2013) Irreversible electroporation of locally advanced therapy in the liver: longitudinal efficacy studies in a rat model of he- pancreatic head adenocarcinoma. J Gastrointest Surg 17:1850 6. patocellular carcinoma. Cancer Res 70:1555 63. 27. Martin RCG (2015) Irreversible electroporation of locally advanced 9. Bower M, Sherwood L, Li Y, Martin R (2011) Irreversible electropo- pancreatic neck/body adenocarcinoma. J Gastrointest Oncol 6:329 ration of the pancreas: definitive local therapy without systemic ef- 35. doi: 10.3978/j.issn.2078-6891.2015.012 fects. J Surg Oncol 104:22 8. 10. Charpentier KP, Wolf F, Noble L, et al (2010) Irreversible electropo- Author: Robert C.G. Martin ration of the pancreas in swine: a pilot study. HPB (Oxford) 12: Institute: University of Louisville School of Medicine 348 51. Street: 315 East Broadway 11. Maor E, Ivorra A, Leor J, Rubinsky B (2007) The effect of irreversi- City: Louisville, Ky 40202 ble electroporation on blood vessels. Technol Cancer Res Treat Country: United States of America 6:307 12. Email: [email protected]

IFMBE Proceedings Vol. 53

Electrotransfer of Antiangiogenic shRNA against Endoglin for Effective Cancer Treatment

M. Čemažar1,2, T. Dolinsek1, N. Tesic2, M. Stimac1, and G. Sersa1 1 Institute of Oncology Ljubljana/Department of Experimental Oncology, Zaloska 2, 1000 Ljubljana, Slovenia 2 Faculty of Health Sciences, University of Primorska, Polje 42, 6310 Izola, Slovenia

Abstract Endoglin (CD105) is involved in activation and resistant to therapies, while others develop acquired resistance proliferation of tumor endothelial cells. Its pronounced role to these therapies. Besides the resistance of tumors to therapy, was recently described in tumors that became resistant to other limitations, such as decreased perfusion of tumors, standard anti-VEGF(R) antiangiogenic therapies. In our ongo- reducing the amount of chemotherapeutic reaching the ing studies, the feasibility of targeting endoglin with gene ther- tumors, increased aggressiveness of tumors and selection of apy using RNA interference technology was elaborated at the more resistant (stem cells) clones in tumors can occur during molecular, cellular and organism level. We prepared plasmids encoding shRNA against endoglin under the control of differ- the antianiogenic therapy course. Several approaches, focused ent promoters and evaluated their therapeutic potential in on alternative proangiogenic pathways involved in the electrogene therapy of different cancers. In TS/A adenocarci- development of acquired resistance, could be taken into noma tumors, which do not express endoglin, anti-endoglin consideration to overcome these limitations. Therefore, in our gene therapy resulted in vascular targeted effect that was more group, we have focused on pathway involving endoglin pronounced when plasmid encoding shRNA was used, than (CD105). when siRNA molecules against endoglin were employed. Fur- thermore, the plasmid encoding shRNA against endoglin II. ENDOGLIN under the control of endothelial cell specific promoter was equally effective as the plasmid with the strong constitutive The primary function of endoglin is the regulation and promoter. In addition, in melanoma tumors, which express maintenance of cardiovascular homeostasis. It is a homodi- endoglin, besides vascular targeted effect, a significant antitu- meric transmembrane glycoprotein that acts as a type III mor and anti-metastatic effects were obtained. Collectively, β ( ). our results demonstrate that endoglin is a suitable target for auxiliary receptor of transforming growth factor TGF-β vascular targeted gene therapy approach and promote further Under normal conditions, endoglin promotes TGF-β signal- studies with clinically suitable and safe plasmids devoid of ing through two downstream activating receptor-like kinas- antibiotic resistance gene and under the control of tissue es, ALK1 and ALK5. Signalling through ALK1 results in specific promoters. quiescent endothelium, while signaling through ALK5 promotes proliferation, thus promoting normal homeostasis. Keywords endoglin, vascular targeted therapy, antiangi- Reduced or increased expression of endoglin consequently ogenic, melanoma, electroporation. leads to deregulation of normal homeostasis and thus to the development of different diseases, such as hereditary he- I. VASCULAR TARGETED THERAPIES morrhagic telangiectasia, pre-eclampsia etc. [2]. In cancer, Small dormant tumors can persist in the body for a long the increased expression of endoglin was demonstrated in time, until the angiogenic switch enables them to acquire activated tumor endothelial cells, as well as in certain tumor blood vessels and thus continuous growth. Although many histiotypes, including melanoma. Endoglin targeting with different growth factors, their receptors and cytokines, have monoclonal antibodies alone or bound with radiotoxins been implemented in the tumor angiogenesis, the essential for have already been tested for antivascular and antitumor tumor angiogenesis is vascular endothelial growth factor action and systemic treatment with anti-endoglin antibody (VEGF) and its receptor (VEGFR). Therefore, many different TC105 is currently in clinical trials [1]. Another approach to therapeutic approaches were developed for targeting either of target endoglin is by RNA interference, where short inter- them or their signaling pathways by monoclonal antibodies fering RNAs (siRNA) or miRNA are used to specifically (bevacizumab), receptor traps (aflibercept) or by small bind to mRNA of targeted protein and thus diminishes molecules, tyrosine kinase inhibitors [1]. Although, in theory, translation of the protein or in other word silence gene ex- because tumor endothelial cells are more genetically stable pression. In our ongoing studies, we used this approach for than tumor cells, tumors should not develop resistance to vascular targeted gene therapy in combination with electro- vascular targeted therapies, certain tumors are intrinsically poration, named gene electrotransfer..

© Springer Science+Business Media Singapore 2016 32 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_8 Electrotransfer of Antiangiogenic shRNA against Endoglin for Effective Cancer Treatment 33

III. SiRNA AGAINST ENDOGLIN effects after treatment demonstrated a quick onset (after 1 day) of the vascular disrupting effect, supporting the action In the first part of our studies, several human and murine of endoglin silencing only in activated endothelial cells in siRNA molecules were selected and tested for potential normal capillaries [4]. Namely, we presume that at the site therapeutic effectiveness. In vitro studies on endothelial of tumor cells injection, the release of angiogenic factors cells showed that siRNAs against endoglin effectively re- from tumor cells increased expression of endoglin and duced proliferation and tube formation. The proliferation of therefore activated endothelial cells of normal vessels. Con- TS/A adenocarcinoma tumor cells was not affected, as these sequently, these cells were killed by the gene electrotransfer cells do not express endoglin. In vivo, gene electrotransfer of plasmid encoding shRNA molecule against endoglin. of siRNA molecules into palpable (3 mm) TS/A tumors Furthermore, when we treated tumor with established growing in Balb/c mice resulted in significantly reduced blood vessels, 3 days after the treatment these vessels were mRNA level of endoglin, number of tumor blood vessels damaged, without blood flow and the surrounding tumor cells and tumor growth [3]. Here, for the first time, we had dem- were becoming necrotic. Furthermore, the antiangiogenic onstrated that endoglin silencing has a potential antitumor effect of endoglin silencing was also demonstrated, since the effectiveness that is mediated by vascular targeted effects. tumors were stable in growth and no new vessels were formed within the tumor during the observation period of 6 IV. VASCULAR TARGETED EFFECTS OF shRNA days [4,5]. Interestingly, the effects of the treatment on acti- AGAINST ENDOGLIN vated vessels at the site of tumor cells’ injection, and the normal vessels in the surrounding tissue, in the dorsal win- Due to the short-lived siRNA molecules, in our next dow chamber were distinguished. We observed that the nor- study, we constructed the plasmid encoding short hairpin mal tissue vessels remained intact and functional, although RNA (shRNA) molecules against endoglin to obtain a long they were in the treated area and therefore transfected with lasting antitumor effect. The expression of shRNA mole- the plasmid. To our best knowledge, our study was the first to cules was driven by U6 promoter, which is a polymerase III clearly demonstrate dual, antiangiogenic and the vascular promoter responsible for transcription of shRNA molecules disrupting effect of endoglin targeted therapy in vivo [4]. in eukaryotic cells. Firstly, the therapeutic effectiveness of To bring this therapy closer to the clinical studies, we newly constructed plasmid was evaluated in endothelial changed previously used constitutive U6 promoter with the cells in vitro. After gene electrotransfer, reduced mRNA endothelial cell specific promoter for endothelin-1 in plas- and protein levels of endoglin were observed in murine mid (pET-antiCD105), to achieve controlled temporal and 2H11 endothelial cells that can be used as a model of tumor spatial expression of shRNA for endoglin silencing. Firstly, endothelial cells. These lower levels also affected biological tissue specificity of the endothelin-1 promoter was demon- properties of endothelial cells and resulted in reduced mi- strated in vitro in several cell lines. Secondly, by reduced gration, proliferation and invasion, meanwhile adhesion of tube formation of 2H11 cells, we demonstrated the compa- cells was increased. These biological properties are perti- rable antiangiogenic efficacy of both plasmids. In vivo, the nent to activated endothelial cells, thus silencing of endog- TS/A tumor model was used to compare the effectiveness lin in vitro by gene electrotransfer attenuated activated en- of gene electrotransfer of both plasmids silencing endoglin, on dothelial cell properties. To demonstrate that gene small avascular and bigger vascularized tumors. The good electrotransfer of plasmid encoding shRNA against endog- antitumor effect was obtained in smaller as well as in bigger lin can also substantially affect the tumor growth via vascu- tumors, regardless of the plasmid that was used.. Additionally, lar targeted effect, TS/A tumors, lacking expression of en- the histological analysis revealed increased necrosis and de- doglin, were treated.. Such treatment, using plasmid, creased number of blood vessels in therapeutic groups to fur- resulted in a longer tumor growth delay than siRNA treat- ther support vascular targeted effect of anti-endoglin therapy ment [3-5]. Furthermore, to fully elucidate the vascular [5]. The results of this study indicated on the potential use of targeted effect of the tratement using gene electrotransfer of plasmid encoding shRNA against endoglin under the control plasmid, tumors growing in a dorsal window chamber in of endothelin-1 promoter in cancer gene therapy. mice were used and were treated in two different vascularization states.To demonstrate antiangiogenic action, tumors V. ANTITUMOR AND ANTIMETASTATIC were treated before the formation of vasculature, and after EFFECTS OF shRNA AGAINST ENDOGLIN the formation of tumor vasculature, to demonstrate vascular-disrupting action. Our results clearly showed that anti- As already stated, endoglin is not overexpressed only on endoglin gene therapy has dual, antiangiogenic and vascular endothelial cells, but also on certain tumor cells, including disrupting, action. The time dependence of the vascular melanoma. Additionally, our plasmid encoding shRNA

IFMBE Proceedings Vol. 53

34 M. Čemažar et al. molecule against endoglin under the control of endothelin-1 (Fig 1 and Table 1). Tumor growth delay of melanoma promoter would also be specific for melanoma cells, as it tumors treated with this plasmid was the same as tumor was shown that besides stimulation of angiogenesis, endo- growth delay of melanoma tumors treated with control pro- thelin has also a promigratory and proinvasive effect in moter. Similarly, this clinically relevant plasmid was also melanoma cells. Therefore, this plasmid could potentially equally effective in reducing the development of distant have a strong specificity towards both, endothelial and me- metastases in mice. lanoma cells. To prove this hypothesis, preclinical studies were performed in vitro, where anti-proliferative and anti- Table 1 Reduced metastatic development of melanoma after gene angiogenic effects were obtained, and in vivo, where mod- electrotransfer of pET-antiCD105 and pET-antiCD105-ORT. erate antitumor and pronounced anti-metastatic effect was observed in highly metastatic B16F10 melanoma. These Group N Metastasis free mice (%) results further support our hypothesis that endoglin is a Ctrl 5 20 valid and promising target in cancer therapy [6]. EP 5 40 In order to prepare the plasmid that could be used in clin- pET-antiCD105 5 40 ical trials, we also constructed the plasmid encoding shRNA pET-antiCD105-ORT 5 20 against endoglin under the control of endothelin-1 promoter pET-antiCD105+EP 5 80 without the antibiotic resistance gene (pET-antiCD105- pET-antiCD105-ORT+EP 5 100 ORT). Namely, the regulatory agencies, U.S. Food and N number of mice. The percentage of metastasis free mice was calculated Drug Administration (FDA) and European Medicine Agen- as the number of mice with metastasis divided with the number of all mice in each group. Intratumoral injection of H2O alone (Ctrl) or in combination cy (EMA), recommend avoiding the use of antibiotic resis- with application of electric pulses (EP), intratumoral injection of plasmid tance genes or the use of ones that are not commonly used alone pET-antiCD105 and pET-antiCD105-ORT or in combination with to treat human infections, in order to avoid possible hori- EP (pET-antiCD105+EP and pET-antiCD105-ORT+EP) zontal gene transfer. Also, this plasmid was tested in vitro and in vivo, and compared to previously used, control, VI. COMPLIANCE WITH ETHICAL plasmid pET-antiCD105. The obtained results show, that REQUIREMENTS this clinically relevant plasmid had equal effects on biological properties of B16F10 melanoma as the control plasmid A. Statement of Animal Rights All the studies that included laboratory animals were N.S. conducted in accordance with EU directive (2010/63/EU), 10 * 3R’s rule and Slovenian legislation. All procedures were performed in compliance with the guidelines for animal * 8 experiments of the EU directive and the permission from the Veterinary Administration of the Ministry of Agricul- 6 ture, Forestry and Food of the Republic of Slovenia (permission no. 34401-4/2012/4). 4

Tumor grow delay (days) delay grow Tumor 2 VII. CONCLUSIONS

0 The results of our ongoing studies demonstrate that en- 5 T P EP 10 R E doglin is a valid target for cancer gene therapy. Our plasmid D -O 5+ T+EP 5 0 tiC R 10 -O with endothelium specific promoter devoid of antibiotic D 5 tiCD1 ET-an tiC n p resistance gene is prepared for direct translation into clinical ET-an pET-a p studies. However, further preclinical studies evaluating gene pET-antiCD10 therapy against endoglin with established cancer treatment are needed to further promote alternative vascular targeted Fig. 1 The significant growth delay of melanoma tumors after gene electrotransfer of pET-antiCD105 and pET-antiCD105-ORT. Intratumoral injec- therapies. tion of H2O in combination with application of electric pulses (EP), intratumoral injection of plasmid alone pET-antiCD105 and pET-antiCD105- ACKNOWLEDGMENT ORT or in combination with EP (pET-antiCD105+EP and pET-antiCD105- ORT+EP). * P < 0.05 compared to all other groups. N.S. not statistically The authors acknowledge the financial support from the significant. state budget by the Slovenian Research Agency (program

IFMBE Proceedings Vol. 53

Electrotransfer of Antiangiogenic shRNA against Endoglin for Effective Cancer Treatment 35 no. P3-0003, projects no. J3-4211, J3-6793, J3-4259, J3- 3. Dolinsek T, Markelc B, Sersa G et al. (2013) Multiple delivery of siRNA 6796). The research was conducted in the scope of LEA against endoglin into murine mammary adenocarcinoma prevents angi- EBAM (French-Slovenian European Associated Laborato- ogenesis and delays tumor growth. PLoS One 8: e58723. 4. Dolinsek T, Markelc B, Bosnjak M et al. (2015) Endoglin Silencing has ry: Pulsed Electric Fields Applications in Biology and Med- Significant Antitumor Effect on Murine Mammary Adenocarcinoma icine) and is a result of networking efforts within the COST Mediated by Vascular Targeted Effect. Curr Gene Ther 15: 228-244. TD1104 Action. 5. Stimac M, Dolinsek T, Lampreht U et al. (2015) Gene electrotransfer of plasmid with tissue specific promoter encoding shRNA against endoglin CONFLICT OF INTEREST exerts antitumor efficacy against murine TS/A tumors by vascular targeted effects. PLoS One. In press. DOI:10.1371/journal.pone.0124913 6. Tesic N, Kamensek U, Sersa G et al. (2015) Endoglin (CD105) silencing The authors declare that they have no conflict of interest. mediated by shRNA under the control of endothelin-1 promoter for targeted gene therapy of melanoma. Mol Ther Nucleic Acids. In press. REFERENCES Author: Prof. Maja Cemazar Institute: Institute of Oncology Ljubljana 1. Rosen LS, Gordon MS, Robert F et al. (2014) Endoglin for targeted Street: Zaloska 2 cancer treatment. Curr Oncol Rep 16: 365. City: Ljubljana 2. Kapur NK, Morine KJ, Letarte M (2013) Endoglin: a critical mediator of Country: Slovenia cardiovascular health. Vasc Health Risk Manag 9: 195-206. Email: [email protected]

IFMBE Proceedings Vol. 53

Abiotic Gene Transfer – A Rarity or a Ubiquity?

Tadej Kotnik1 and James C. Weaver2 1 Department of Biomedical Engineering, Faculty of Electrical Engineering, University of Ljubljana, Slovenia 2 Harvard MIT Division of Health Sciences and Technology, Massachusetts Institute of Technology, Cambridge, MA, USA

Abstract Phylogenetic studies of the last two decades have prokaryotes [5], and three main natural mechanisms of such revealed that horizontal gene transfer (HGT) has been playing transfer have been identified: a prominent role throughout evolution and contributed importantly to the genetic variability of species inhabiting our pla- • competence: ability to uptake freely floating DNA from net. Three main biotic mechanisms of HGT competence for the environment; the term transformation is also often used DNA uptake, conjugation, and viral transduction have been interchangeably for this mechanism and the effect of its identified and by now rather thoroughly investigated, but it is action on the organism; questionable whether they can account for all natural occur- • conjugation: DNA transfer between two organisms in rences of HGT. Namely, most eukaryotes lack ability for either conjugation or competence, while transduction mostly direct contact; proceeds among organisms that are phylogenetically very close • transduction: transfer of DNA from one organism to to each other; yet there is mounting evidence of HGT from another via a viral infection. prokaryotes to eukaryotes, and even between eukaryotes. Here, we posit that of the four laboratory techniques most In this paper, we discuss whether all natural HGT can be widely used for artificial genetic transformation chemical, explained by these three mechanisms, and we posit that of freeze-and-thaw, microbeads-agitation, and electroporation- the four laboratory techniques most widely used for artifi- based transformation at least three have analogues in nature cial genetic transformation, at least three have analogues in that can act as abiotic mechanisms of gene transfer. In particu- nature that can also function under circumstances where this lar, we show that these abiotic mechanisms of HGT, while is difficult to envisage for the mechanisms listed above. possibly inferior in importance to biotic HGT in those organisms and environments where the latter proceeds efficiently, can also act under circumstances where this is difficult to II. DO THE THREE BIOTIC HGT MECHANISMS envision for the three biotic mechanisms. SUFFICE?

Keywords Horizontal gene transfer, evolution, chemo- Each of the three widely recognized HGT mechanisms transformation, freeze-and-thaw transformation, microbeads- listed above is biotic, in the sense that it is based on biomo- agitation transformation, electrotransformation. lecules synthesized by the organism(s) involved in HGT: competence involves a DNA translocase and a type-IV I. INTRODUCTION system in the recipient organism [6], conjugation requires a DNA relaxase and a pilus in the donor organism [7], while During the last two decades, progress in DNA sequenc- transduction utilizes the viral base plate, its tail, and/or ing has revolutionized the studies of evolutionary distances lysozymes [8]. between the organisms, but at the same time started to re- As the most important biomolecules involved in each of veal that evolution does not proceed merely by natural se- these three evolution-enhancing mechanisms are proteins, it lection of random mutations and their inheritance (termed is clear that these mechanisms are themselves products of vertical gene transfer – VGT), but also by heritable inter- evolution. Thus, while it is now widely assumed that HGT change of genetic material between phylogenetically distant has been ongoing since the earliest stages of evolution (see organisms (horizontal gene transfer – HGT). This started to e.g. Fig. 3 in [3], Fig. 1 in [9], and Fig. 1 in [10]), it is an emerge in the mid-1990’s from comparing genomes, which outstanding question how it could have proceeded before revealed an increasing number of genes in eukaryotes ab- any of its biotic mechanisms had developed. sent from archaea, yet present in phylogenetically more It also seems questionable whether all occurrences of distant bacteria [1,2], and was corroborated by comparing HGT identified to date can be explained by the three biotic nucleotide sequences of individual highly-conserved genes, mechanisms; while each of them has been shown to occur in from which it became clear that phylogenetic trees inferred both prokaryotic kingdoms, i.e. in archaeal [6,11] as well as from different such genes can differ considerably [3,4]. in bacterial species [6–8], it is far from clear whether Heritable horizontal gene transfer is by now widely they can occur in every such species. Thus, it is estimated recognized as a major contributor to genetic variability of that only about 1% of all bacterial species are naturally

© Springer Science+Business Media Singapore 2016 36 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_9 Abiotic Gene Transfer A Rarity or a Ubiquity? 37 competent [12], and likely an even lower fraction of arc- mechanisms, and farther still from a steadfast conclusion that haeal ones [6,13]; a further hindrance to competence is the known HGT mechanisms are the only such mechanisms. rapid degradation of free DNA in natural habitats [14]. For conjugation, though it was recently assessed as the most III. LABORATORY HGT TECHNIQUES ARE ALL important of the three HGT mechanisms [15], its efficiency ABIOTIC quickly decreases with increasing genetic distance, as the protein machinery utilized is highly adjusted to a particular To emulate natural processes in laboratories, we often rely organism’s envelope [16]. Regarding transduction, most on biomolecules used by living organisms for these viruses infect selectively, only transferring genes among processes. For DNA replication in vitro, the PCR technique genetically very close organisms (bacterial phages often thus uses DNA polymerase, which also catalyzes DNA repli- even only bind to a single strain of a bacterial species). And cation in living cells (strictly speaking not in these same cells, finally, there is increasing evidence of HGT in eukaryotes, as PCR relies on the highly heat-resistant variant of DNA even in multicellular animals [17], in which neither conju- polymerase isolated from the bacterium Thermus aquaticus). gation nor competence exist, nor are they in general infecta- In contrast, none of the laboratory techniques for artificial ble by bacteriophages or archael phages. gene transfer is based on the biological machinery of natural As so often in biology, there are exceptions to these gen- HGT; instead, each relies on a rather elementary mechanism eral limitations. Among competent prokaryotes, it was shown of limited membrane disruption: that the bacterium Acinetobacter baylyi can take up and express even highly fragmented and degraded DNA [18]. It A. Chemotransformation is also known that mechanisms similar to conjugation can It has long been known that supraphysiological (> ~1 mM) lead to gene transfer from various bacteria into some euka- concentrations of Ca2+ cause membrane disruption [29], as ryotes: from Escherichia coli into the yeast Saccharomyces well as facilitate contact between extracellular DNA and the cerevisiae [19], from Agrobacterium tumefaciens into cells membrane [30]. Together, these two effects allow for DNA of some flowering plants where they induce tumorigenesis transfer, and in the first report of artificial transformation, [20], and from E. coli into Chinese hamster ovary cells [21]; published in 1970, viral DNA was introduced into E. coli in still, except for A. tumefaciens-induced tumorigenesis in a medium with 100 mM Ca2+ [31]. In 1972, this method was plants that occurs amply in nature, transfer was performed also shown to work with plasmid DNA [32], and in 1974 its in laboratories, using plasmid DNA engineered for stability efficiency was improved by adding to Ca2+ other divalent and transferrability. Regarding transduction, viruses with a ions such as Mg2+ and Mn2+ [33]. While efficiencies of up broad host range and/or adaptive host specificity are known to 105–106 transformants per µg DNA were reported, typi- to exist both in bacteria [22] and eukaryotes [23], and bacte- cally they are much lower in organisms with a cell wall riophages can be artificially modified for gene delivery into (which is much less sensitive to Ca2+), unless other chemi- eukaryotic cells [24]. Finally, in mid-2000’s a fourth biotic cals such as detergents are added; moreover, it is technically HGT mechanism was identified: gene transfer agents (GTAs), difficult to bring the action of the Ca2+ ions to an abrupt virus-like particles synthesized by some anaerobic bacterial halt, and until they are removed (typically by dilution), their species [25] and allowing for rapid interchange of genetic disruptive action persist, and its effects accumulate, result- material between them, particularly in marine habitats [26]. ing in high loss of viability of the exposed organisms [34]. With so many exceptions, as well as exceptions to the exceptions, it is likely impossible to pinpoint a case of HGT B. Freeze-and-thaw transformation for which it could be inferred, let alone rigorously proved, In the first attempts to replace a chemical exposure with a that none of the known biotic HGT mechanisms, or yet physical mechanism of permeabilization whose action can unknown natural adaptations thereof, can explain its occur- be brought to a halt more abruptly and thus better con- rence. Still, the GTAs offer a lesson: until they were discov- trolled, researchers opted for cooling and heating: first ered, many reviews stated that the three biotic mechanisms freezing the microorganisms and then thawing them. As listed on the preceding page are sufficient to account for with chemotransformation, freeze-and-thaw transformation most if not all HGT, and this statement even recurs in some was initially performed with E. coli and viral DNA, with the recent prominent reviews [27], although other prominent first report in 1972, using freezing at –70°C and thawing at studies suggest that GTAs is a major, and perhaps even the 42°C [35]. This technique was subsequently shown to work prevalent mechanism of HGT in the oceans [26,28]. This with other bacteria, e.g. A. tumefaciens in 1978 [36], but its shows that while our knowledge and understanding of HGT simplicity was overshadowed by its low efficiency, which is rapidly increasing, we are far from a reliable assessment reaches at most ~103 transformants per µg DNA unless of the relative importance of each of the known HGT augmented by addition of permeabilizing chemicals [37].

IFMBE Proceedings Vol. 53

38 T. Kotnik and J.C. Weaver

Furthermore, while it is easier to abruptly terminate cooling exposure, as well as the electric pulse amplitude can be than chemical exposure, the heating process can only be preset to a high accuracy), and with suitably designed pulse- gradual if irreversible damage is to be avoided. delivering electrodes all the treated organisms are exposed quite uniformly. C. Microbeads-agitation transformation The next technique developed for artificial gene transfer IV. ABIOTIC HGT MECHANISMS IN NATURE? was mechanical disruption; this allowed for quick and thus well-controlled triggering and halting of the permeabiliza- While naturally occurring concentrations of Ca2+ may not tion process, but with rather variable intensity of permeabi- suffice for chemopermeabilization (even in seawater they lization of individual microorganisms. The initial form of rarely exceed 12 mM), natural conditions similar to the three this technique, published in 1984, consisted of scraping the physical techniques of transformation are easily imagined: cells growing on a surface with a rubber stick [38], but this was supplanted in 1988 by agitation with small glass beads • freeze-and-thaw cycles: limiting our considerations to (~500 µm diameter) in a vortex mixer [39]. Unlike chemo- aqueous environments where DNA diffuses easily, tempera- transformation and freeze-and-thaw transformation that ture fluctuations resulting in alternations of freezing and were predominantly tried with bacteria, the microbeads- thawing regularly occur in polar seas; with the sea ice cov- agitation technique was first demonstrated in yeasts (i.e., ering an area of ~2.5×1013 m2, and with seawater typically unicellular fungi) [39], and in 1990 in microalgae (i.e., containing ~1011 microorganisms per m3, it is quite unicellular algae) [40], yet in both cases with efficiencies of straightforward to show that the even if freeze-and-thaw at most ~300 transformants per µg DNA. fluctuations are largely limited to its edge, and even if they only amount to several millimeters a day, every such fluctu- D. Electrotransformation ation freezes and thaws trillions of microorganisms per day; The likely reason that despite its simplicity, microbeads- • sand and gravel agitation: limiting us again to aqueous agitation transformation never gained prominence (even environments, the action of water-driven movement of sand with yeast and fungi) was the contemporaneous emergence and gravel on microorganisms inhabiting the eulittoral (fo- of the most universally applicable and controllable tech- reshore) of seas and lakes, as well as in riverbeds, clearly nique of artificial genetic transformation known to date: one resembles the effects to which they are exposed by glass based on electroporation – membrane and wall disruption microbeads in a vortex mixer; the famous coastline paradox by means of exposure to short (microseconds to millise- can be disregarded here, as it is the volume of agitated water conds) and intense (hundreds to thousands of V/cm) electric that matters; coastline length measured at a resolution of 1m pulses. The first report on electroporation-based gene trans- suffices; at this resolution the global sea coastline length is fer was published in 1982 [41], but it was achieved on ~1.2×109 m, and while the fraction of this length with eulit- mammalian cells (which lack a wall) and moreover not toral covered by sand or gravel does not seem to have been heritable, while in organisms with a wall the initial experi- measured, it is again a safe estimate (even with lakes and ments performed in 1983 suggested this to be achievable rivers disregarded) that the volume in which sand and gra- only after its complete removal [42]. This misapprehension vel agitation occurs contains trillions of microorganisms; was, however, due to insufficient amplitude of the applied • lightning-triggered electroporation: focusing again on electric pulses, and development of stronger pulse genera- aqueous habitats, it follows from electrical properties of tors quickly led to transformation of microorganisms with lightning strokes and seawater that a typical lightning stroke an intact wall: yeasts in 1985 [43], bacteria in 1987 [44], entering seawater induces in it an electric field sufficient for and archaea and microalgae in 1991 [45,46]. Compared to electroporation in a volume of ~0.003 m3; with ~3×104 such the earlier techniques of artificial transformation, electro- strokes per day (~1% of ~3×106 cloud-to-ground strokes hit transformation is thus applicable to a broader range of mi- the seas, and the rest hit land) and ~1011 microorganisms croorganisms, but what was likely decisive in its rapid rise per m3, it seems safe to conclude that also electroporation 10 to prevalence was its higher efficiency: up to ~10 trans- affects trillions of microorganisms per day. formants per µg DNA for Gram-negative bacteria, up to ~107 for Gram-positive bacteria and archaea (as they have a In conclusion, these considerations suggest that abiotic thicker wall), and up to ~106 for microalgae and yeasts (as mechanisms of gene transfer are present in nature, and they have both a wall and a nuclear membrane). This large- while possibly inferior in importance to biotic HGT in those ly reflects the fact that with electric pulses, the permeabiliz- organisms and environments where the latter proceeds effi- ing mechanism only acts during the pulse, conditions are ciently, they can also act under circumstances where this is highly controlled and adjustable (the start and end of the difficult to envision for the three biotic mechanisms.

IFMBE Proceedings Vol. 53

Abiotic Gene Transfer A Rarity or a Ubiquity? 39

We will present further quantitative considerations at the 17. Boto L (2014) Horizontal gene transfer in the acquisition of novel conference, and in more detail in a forthcoming article. traits by metazoans. Proc R Soc B 281(20132450):1 8 18. Overballe-Petersen S, Harms K, Orlando LAA et al. (2013) Bacterial natural transformation by highly fragmented and damaged DNA. Proc ACKNOWLEDGMENT Natl Acad Sci USA 110:19860 19865 19. Heinemann JA, Sprague GFJr (1989) Bacterial conjugative plasmids The work was supported by the Slovenian Research mobilize DNA transfer between bacteria and yeast. Nature 340:205 209 Agency (ARRS grant P2-0249) and the National Institutes 20. Zupan J, Muth TR, Draper O et al. (2000) The transfer of DNA from of Health (NIH grant GM063857), with research conducted Agrobacterium tumefaciens into plants: a feast of fundamental in- in the LEA EBAM European Associated Laboratory and sights. Plant J 23:11 28 within networking efforts of the European Cooperation in 21. Waters VL (2001) Conjugation between bacterial and mammalian cells. Nat Genet 29:375 376 Science and Technology (COST actions TD1104 and 22. Koskella B, Meaden S (2013) Understanding bacteriophage specificity TD1308). in natural microbial communities. Viruses 5:806 823 23. Bandin I, Dopazo CP (2011) Host range, host specificity and hypothe- CONFLICT OF INTEREST sized host shifts among viruses of lower vertebrates. Vet Res 42(67):1 15 24. Poul MA, Marks JS (1999) Targeted gene delivery to mammalian cells The authors declare that they have no conflict of interest. by filamentous bacteriophage. J Mol Biol 288:203 211 25. Stanton TB (2007) Prophage-like gene transfer agents. Anaerobe 13:43 49 REFERENCES 26. McDaniel LD, Young E, Delaney J et al. (2010) High frequency of horizontal gene transfer in the oceans. Science 330:50 1. Hilario E, Gogarten JP (1993) Horizontal transfer of ATPase genes 27. Syvanen M (2012) Evolutionary implications of horizontal gene the tree of life becomes a net of life. Biosystems 31:111 119 transfer. Annu Rev Genet 46:341 358 2. Lawrence JG, Ochman H (1998) Molecular archaeology of the Esche- 28. Kristensen DM, Mushegian AR, Dolja VV et al. (2010). New dimen- richia coli genome. Proc Natl Acad Sci USA 95:9413 9417 sions of the virus world discovered through metagenomics. Trends 3. Doolittle WF (1999) Phylogenetic classification and the universal tree. Microbiol 18:11 19 Science 284:2124 2128 29. Trump BF, Berezesky IK (1995) Calcium-mediated cell injury and cell 4. Zhaxybayeva O, Gogarten JP (2004) Cladogenesis, coalescence and death. FASEB J 9:219 228 the evolution of the three domains of life. Trends Genet 20:182 187 30. Weston A, Brown MGM, Perkins HR et al. (1981) Transformation of 5. Bapteste E, O’Malley MA, Beiko R et al. (2009) Prokaryotic evolution Escherichia coli with plasmid deoxyribonucleic acid: calcium induced and the tree of life are two different things. Biol Direct 4(34):1 20 binding of deoxyribonucleic acid to whole cells and to isolated mem- 6. Johnsborg O, Eldholm V, Håvarstein LS (2007) Natural genetic trans- brane fractions. J Biotechnol 145:780 787 formation: prevalence, mechanisms, and funcion. Res Microbiol 31. Mandel M, Higa A (1970) Calcium-dependent bacteriophage DNA 158:767 778 infection. J Mol Biol 53:159 162 7. Chen I, Christie PJ, Dubnau D (2005) The ins and outs of DNA trans- 32. Cohen SN, Chang AC, Hsu L (1972) Nonchromosomal antibiotic fer in bacteria. Science 310:1456 1460 resistance in bacteria: genetic transformation of Escherichia coli by 8. Cann AJ. (2015) Principles of Molecular Virology, 6th edition. Aca- R-factor DNA. Proc Natl Acad Sci USA 69:2110 2114 demic Press, London, pp. 112 119 33. Lederberg E, Cohen SN (1974) Transformation of Salmonella typhi- 9. Smets BF, Barkay T (2005) Horizontal gene transfer: perspectives at a murium by plasmid deoxyribonucleic acid. J Bacteriol 119:1072 1074 crossroads of scientific disciplines. Nat Rev Microbiol. 3:675 678 34. Aune TEV, Aachmann FL (2010) Methodologies to increase the 10. Koonin EV (2009) Darwinian evolution in the light of genomics. transformation efficiencies and the range of bacteria that can be trans- Nucleic Acids Res 37:1011 1034 formed. Appl Microbiol Biotechnol 85:1301 1313 11. Luo Y, Wasserfallen A (2001) Gene transfer systems and their applica- 35. Dityatkin SY, Lisovskaya KV, Panzhava NN et al. (1972) Frozen- tions in archaea. Syst Appl Microbiol 24:15 25 thawed bacteria as recipients of isolated coliphage DNA. Biochim Bi- 12. Brochier-Armanet C, Moreira D (2015) Horizontal gene transfer in ophys Acta 281:319 323 microbial ecosystems. In Bertrand JC (Ed.) Environmental Microbiolo- 36. Holsters M, Dewaele D, Depicker A et al. (1978) Transfection and gy: Fundamentals and Applications. Springer, Dordrecht, pp. 445 484 transformation of Agrobacterium tumefaciens. Mol Gen Genet 13. Lipscomb GL, Stirrett K, Schut GJ et al. (2011) Natural competence in 163:181 187 the hyperthermophilic archaeon Pyrococcus furiosus facilitates genetic 37. StepanovAS, PuzanovaOB, Dityatkin SY et al. (1990) Glycine induced manipulation. Appl Environ Microbiol 77:2232 2238 cryotransformation of plasmid into Bacillus anthracis. J Gen Micro- 14. Lorenz MG, Wackernagel W (1994) Bacterial gene transfer by natural biol 136:1217 1221 genetic transformation in the environment. Microbiol Rev 58:563 602 38. McNeil PL, Murphy RF, Lanni F et al. (1984) A method for incorpo- 15. Halary S, Leigh JW, Cheaib B et al (2010) Network analyses structure rating macromolecules into adherent cells. J Cell Biol 98:1556 1564 genetic diversity in independent genetic worlds. Proc Natl Acad Sci 39. Costanzo M, Fox TD (1988) Transformation of yeast by agitation with USA 107:127 132 glass beads. Genetics 120:667 670 16. Guglielmini J, de la Cruz F, Rocha EPC (2013) Evolution of conjuga- 40. Kindle KL (1990) High-frequency nuclear transformation of Chlamy- tion and type IV secretion systems. Mol Biol Evol 30:315 331 domonas reinhardtii. Proc Natl Acad Sci USA 87:1228 1232

IFMBE Proceedings Vol. 53

40 T. Kotnik and J.C. Weaver

41. Wong TK, Neumann E (1982) Electric field mediated gene transfer. 44. Chassy BM, Flickinger JL (1987) Transformation of Lactobacillus Biochem Biophys Res Commun 107:584 587 casei by electroporation. FEMS Microbiol Lett 44:173 177 42. Shivarova N, Förster W, Jacob HE et al. (1983) Microbiological 45. Micheletti PA, Sment KA, Konisky J (1991) Isolation of a coenzyme implications of electric field effects: VII. Stimulation of plasmid trans- M-auxotrophic mutant and transformation by electroporation in Me- formation of Bacillus cereus protoplasts by electric field pulses. Z Allg thanococcus voltae. J Bacteriol 173:3414 3418 Mikrobiol 23:595 599 43. Hashimoto H, Morikawa H, Yamada Y et al. (1985) A novel method 46. Brown LE, Sprecher SL, Keller LR (1991) Introduction of exogenous for transformation of intact yeast cells by electroinjection of plasmid DNA into Chlamydomonas reinhardtii by electroporation. J Cell Biol DNA. Appl Microbiol Biotechnol 21:336 339 11:2328 2332

IFMBE Proceedings Vol. 53

Part II Application of Pulsed Electric Fields Technology in Food: Challenges and Opportunities

Antimicrobial Effect of Stevia rebaudiana Bertoni against Listeria monocytogenes in a Beverage Processed by Pulsed Electric Fields (PEFs): Combined Effectiveness

A. Rivas, S. Sansano, M.C. Pina Pérez, A. Martínez, and D. Rodrigo Instituto de Agroquímica y Tecnología de Alimentos (IATA CSIC). Dpto. Conservación y Calidad. Avda. Agustin Escardino, 7, 46980 Paterna, Valencia, Spain

Abstract The aim of this work was to evaluate inactivation food preservation technologies, the use of pulsed electric of L. monocytogenes cells suspended in a beverage (a mixture fields (PEF) seems to be an effective alternative for inacti- of oat milk and orange, papaya, and mango juices) by a hurdle vating a large variety of pathogens in the treatment of liquid treatment of PEF, Stevia rebaudiana Bertoni (leaf infusion, products, especially fruit- and vegetable-based beverages, 8.33% w/v), and refrigerated storage. Beverage samples with and milk products [6, 7]. However, long, intense treatments and without stevia underwent five PEF treatments (10 40 are needed, requiring high energy consumption, in order to kV/cm, 50 800 µs), with two energy levels (100 and 800 J/mL), and were stored at two temperatures (5 and 10 °C), for a 48 achieve inactivation levels suitable for pasteurizing foods. hours period. The results showed the existence of a synergic One strategy for reducing treatment intensity without de- PEF-stevia effect, achieving inactivation levels exceeding 5 creasing microbiological inactivation involves combining decimal reductions of L. monocytogenes during refrigerated those mild preservation technologies, in a hurdle process. storage of samples after treated by PEF. This strategy has been investigated with PEF [8] and with These results show that stevia has antimicrobial capacity other non-thermal preservation technologies [9]. against L. monocytogenes and that the PEF-stevia combination Considering the growing interest in incorporating stevia in is a valid strategy to increase the microbiological safety of this fruit- and vegetable-based beverages, such as fruit juices, and type of beverages. the demand for minimally processed products that retain all Keywords Pulsed electric fields, Listeria monocytogenes, their quality attributes, it is needed to evaluate the combined Stevia rebaudiana Bertoni, Inactivation effect of this natural ingredient in beverages processed by non-thermal technologies, in order to achieve products of greater quality that meet the demands of the modern consum- I. INTRODUCTION er. Consequently, the aim of the present study was to eva- Consumer food preferences have changed, especially luate, the combined effect of PEF and stevia on survival of L. during the last decade. Today, consumers not only demand monocytogenes in a food substrate consisting of oat milk and foods that are microbiologically safe and with high quality, fruit juices (papaya, mango, and orange). but also they want perceive them as “fresh-like” products, if possible without chemical additives, that conserve all II. MATERIALS AND METHODS their nutritional value at the time of consumption. There are many natural substances that have been used in A. Listeria monocytogenes culture preparation gastronomy, not only as food condiments but also for their Listeria monocytogenes CECT 4032 was provided by the beneficial health properties. Stevia rebaudiana Bertoni, Spanish Type Culture Collection, and was isolated from a known as sweet leaf, sweet herb, honey leaf, and candy leaf patient with meningitis associated to contaminated cheese. [1], is a perennial shrub native from South-America [2]. It Briefly, a fresh culture of L. monocytogenes (pre-incubated contains sweet steviol glycosides [3], which can be used as for 6 hours) was inoculated into Tryptone and Soy Broth substitutes for sucrose preventing caries, and by individuals (TSB) (Scharlau S.A., Barcelona, Spain), followed by incu- with certain pathologies, such as diabetes, obesity, and bation to stationary phase, cooled in ice, washed twice with hypertension [4]. As a result of the use of steviol glycosides TSB, placed in 2-mL sterile plastic cryogenic vials contain- extract as a calorie-free food sweetener legalization in EU, ing TSB supplemented with 20% (v/v) glycerol in a 1:1 [5], its use in the European food industry has grown. Al- proportion, and stored at –80 °C. The concentration was though it is fundamentally used as a sweetener, various approximately 5 × 109 CFU/mL. recent studies have attempted to make a deeper investigation of the antimicrobial activity of stevia [1, 2]. B. Food substrate formulation and preparation Another way that is envisaged as being most appropriate Stevia and PEF treatment were applied in combination to for achieving foods with “fresh-like” properties is the use of a substrate inoculated with L. monocytogenes. The substrate non-thermal preservation technologies. Within non-thermal formulation was: oat milk (20% (v/v)); orange juice (7.5% © Springer Science+Business Media Singapore 2016 43 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_10

46 A. Rivas et al. of stevia in the samples treated at greater intensity provided 5. EU 1131/2011. COMMISSION REGULATION (EU) of 11 November an additional stress that caused the death of more PEF-treated 2011 amending Annex II to Regulation (EC) No 1333/2008 of the Eu- ropean Parliament and of the Council with regard to steviol glycosides. L. monocytogenes cells. A study [8] indicated that, after a 6. Rivas A, Sampedro F, Rodrigo D et al. (2006) Nature of the inactiva- storage period (12 h, 8 °C), the death of C. sakazakii cells tion of Escherichia coli suspended in an orange juice and milk beve- previously treated by PEF increased (by up to 3 decimal rage. Eur Food Res Technol 223: 541 545. reductions) (15 kV/cm, 3000 µs) in the presence of cocoa 7. Sampedro F, Rodrigo D, Martínez A (2011) Modelling the effect of pH and pectin concentration on the PEF inactivation of Salmonella en- powder. terica serovar Typhimurium by using the Monte Carlo simulation. Food Control 22(3 4): 420 425. 8. Pina-Pérez M C, Martínez-López A, Rodrigo D (2013) Cocoa powder IV. CONCLUSIONS as a natural ingredient revealing an enhancing effect to inactivate Cro- nobacter sakazakii cells treated by Pulsed Electric Fields in infant milk The addition of Stevia rebaudiana to a beverage as a ca- formula. Food Control 13(1): 87 92. lorie-free sweetener may be a valid strategy to increase the 9. Pina-Pérez M C, Silva-Angulo A B, Martínez-López A et al. (2012) A safety of foods treated by pulsed electric fields. It not only preliminary exposure assessment model for Bacillus cereus cells in a acts as an additional control barrier during refrigerated sto- milk based beverage: evaluating High Pressure Processing and antimicrobial interventions. Food Control 26: 610 613. rage of products treated by PEF but also makes it possible 10. Gómez N, García D, Álvarez I et al. (2005) Modelling inactivation of to achieve 5D in a population of L. monocytogenes in a high Listeria monocytogenes by pulsed electric fields in media of different value added beverage because of the synergic effect attri- pH. Int J Food Microbiol 103: 199 206. butable to the PEF, stevia and storage temperature. 11. Zhao W, Yanga R, Shena X et al. (2013) Lethal and sublethal injury and kinetics of Escherichia coli, Listeria monocytogenes and Staphylococcus aureus in milk by pulsed electric ﬁelds. Food Control 32: 6 12. 12. Sampedro F, Rivas A, Rodrigo D et al. (2007) Pulsed electric ﬁelds ACKNOWLEDGMENT inactivation of Lactobacillus plantarum in an orange juice milk based beverage: Effect of process parameters. J Food Eng 80: 931 938. Authors thank the Spanish Ministry of Economy and 13. Engels C, Weiss A, Carle R et al. (2012) Effects of gallotannin treat- Competitiveness as well as FEDER for their financial sup- ment on attachment, growth, and survival of Escherichia coli O157:H7 and Listeria monocytogenes on spinach and lettuce. Eur. Food Res. port through the project AGL2013-48993-C2-2-R. Technol 234: 1081 1090. 14. Criado M N, Barba F J, Frígola A et al. (2013) Effect of Stevia rebaudiana on oxidative enzyme activity and its correlation with antioxidant ca- CONFLICT OF INTEREST pacity and bioactive compounds. Food Bioprocess. Technol 7: 1518-1525. 15. Amiali M, Ngadi M O, Smith J P et al. (2007) Inactivation of Escheri- The authors declare that they have no conflict of interest. chia coli O157:H7 and Salmonella enteritidis in liquid egg white using pulsed electric field. J Food Sci 71: 88 94. 16. Martínez-Viedma P, Sobrino-López A, Omar N B et al. (2008) En- REFERENCES hanced bactericidal effect of enterocin AS-48 in combination with high-intensity pulsed-electric field treatment against Salmonella ente- 1. Jayaraman S, Manoharan M, Illanchezian S (2008) In-vitro antimi- rica in apple juice. Int J Food Microbiol 128: 244 249. crobial and antitumor activities of Stevia rebaudiana (Asteraceae) leaf 17. Lemus-Mondaca R, Vega-Gálvez A, Zura-Bravo L et al. (2012) Stevia rebaudiana Bertoni, source of a high-potency natural sweetener: A extracts. Tropical J Pharmaceutical Res 7: 1143 1149. comprehensive review on the biochemical, nutritional and functional 2. Sivaram L, Mukundam U (2003) In vitro culture studies on Stevia aspects. Food Chem 132: 1121 1132. rebaudiana. In Vitro Cell Dev Biol Plant 39: 520 523. Author: Antonio Martínez López 3. Brandle J E, Telmer P G (2007) Steviol glycoside biosynthesis. Phyto- Institute: Instituto de Agroquímica y Tecnología de Alimentos. chem 68: 1855 1863. Street: Agustin Escardino, 7 4. Ghanta S, Banerjee A, Poddar A, et al. (2007) Oxidative DNA damage City: Paterna preventive activity and antioxidant potential of Stevia rebaudiana (Ber- Country: Spain toni) Bertoni, a natural sweetener. J Agric Food Chem 55: 10962 10967. Email: [email protected]

IFMBE Proceedings Vol. 53

Pulsed Electric Field Technology Enhances Release of Anthocyanins from Grapes and Bioprotective Potential against Oxidative Stress

S.Y. Leong1,2, I. Oey1, and D.J. Burritt2 1 Department of Food Science, University of Otago, Dunedin, New Zealand 2 Department of Botany, University of Otago, Dunedin, New Zealand

Abstract Pulsed electric field (PEF) processing affects the pre-treatment methods such as pectinase enzymatic treat- cell permeability of plant tissue. This technology is capable of ment [11] and moderate thermal cell disintegration [12]. enhancing mass transfer and releasing valuable compounds Moreover, as compared to PEF, both conventional grape (e.g. juice, health-promoting phytochemicals) from plant cells at low electric fields. This study aimed to assess the influence mash pre-treatment methods would cause greater oxidation of PEF (1033 square wave bipolar 20 μs pulses of 1.4 kV/cm at and degradation of the phenolics compounds and thus a frequency of 50 Hz for a total treatment time of 20.66 ms) on reduce the quality of the final red wine. the release of anthocyanins from grapes (Vitis vinifera L. cv. While many previous studies have alert to the fact that Merlot) after 48 h of PEF treatment and on its potential to PEF improved extraction of anthocyanins from grapes protect Caco-2 cells from H2O2-induced oxidative stress. Mer- [13,14], little is known whether an increase in the release of lot anthocyanins were characterized and quantified using grape anthocyanins by PEF is directly referred to an in- LC-MS. To assess the bioprotective potential, cell viability was crease in bioprotective potential on a physiologically rele- measured using 3-(4, 5-dimethythiazol-2-yl)-2, 5-diphenyl tetrazolium bromide (MTT) assay that indicates the mitochon- vant model of human intestinal epithelial (Caco-2 cells) drial metabolic activity. A significant increase in the release of from oxidative stress. anthocyanins was obtained immediately after PEF. After 48 h, The objective of this study was to compare the amount of PEF had facilitated the release of malvidin, delphinidin and anthocyanins released from Merlot grapes immediately and petunidin glucosidic derivatives of anthocyanins. Because of 48 h after PEF processing (1.4 kV/cm, 50 Hz and 1033 the increase in amount and types of Merlot anthocyanins re- pulses). Subsequently, the ability of Merlot juice in protect- leased due to PEF within 48 h, the obtained grape juice was ing Caco-2 cells from H2O2-induced oxidative stress by more effective in protecting Caco-2 cells from oxidative dam- determining the cell viability using 3-(4, 5-dimethythiazol- age. This finding provides evidence that PEF can increase 2-yl)-2, 5-diphenyl tetrazolium bromide (MTT) assay was the release of anthocyanins as well as delivering bioactivity important in the field of functional foods. investigated.

Keywords Pulsed electric fields, grape, anthocyanins, health II. MATERIALS AND METHODS benefits, oxidative stress A. Raw Material Preparation and PEF Processing I. INTRODUCTION Merlot grapes were destemmed and divided into two batches. Juice from the first batch of crushed Merlot were Regular consumption of anthocyanin-rich fruits and veg- immediately collected and referred as ‘untreated Merlot 0 h’ etables has been reported to deliver a broad spectrum of samples. The second batch of Merlot grapes (200 g per human health benefits, such as delaying the initiation of batch) was processed with PEF (ELCRACK-HVP 5 batch various cancers [1], obesity and diabetes prevention [2], configuration, German Institute of Food Technologies, improving brain functions [3] and protecting cellular sys- Quakenbrück, Germany) at an electric field strength of 1.4 tems from oxidative stress [4-7]. kV/cm, 1033 square wave bipolar pulses, pulse width of 20 In grapes, anthocyanins are the major class of phyto- μs and pulse frequency of 50 Hz for a total treatment time chemicals that present exclusively in the skin cells [8]. of 20.66 ms. The batch treatment chamber of PEF (100 mm Pulsed electric field (PEF) is a non-thermal food processing length × 80 mm width × 50 mm height, 400 mL capacity) technology that affects cell permeability [9]. This technolo- consisted of two stainless steel electrodes of 5 mm thickness gy has been shown useful to enhance mass transfer and separated by a distance of 80 mm. Merlot juice was sepa- release of valuable compounds from plant cells at low elec- rated from the berries immediately after PEF processing and tric fields [10]. PEF processing is also proven as a more referred as ‘PEF-treated Merlot 0 h’ sample. From each preferable method in promoting greater extraction of antho- batch, the untreated and PEF-treated Merlot berries cyanins as compared to other conventional grape mash were also allocated for a 48 h contact time with the juice.

© Springer Science+Business Media Singapore 2016 47 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_11 48 S.Y. Leong, I. Oey, and D.J. Burritt

The grapes were sealed into independent sterile container with culture medium. Cells were cultured for 24 h, the cul- and stored at 10 ± 2 oC under dark condition. Merlot juice ture medium was then removed and the Caco-2 cells were was separated from the berries after 48 h of contact. The washed with fresh culture medium without FBS. The induc- experiment was independently carried out in triplicate for tion of oxidative stress was carried out in which the Caco-2 all of the experimental conditions. The juice was snapped cells were then exposed to 500 μM H2O2 for 1 h. Four inde- frozen in liquid nitrogen and stored at -80 oC until analysis. pendent cell culture experiments were conducted to determine the cell viability. B. Extraction and Determination of Anthocyanins Juice was extracted in acidified methanol in a 1:1 ratio. D. Cell Viability MTT Assay After centrifugation, the supernatant was collected, filtered Caco-2 cells were incubated with 0.5 mg/mL MTT dye in and directly injected (20 μL) into the liquid chromatography culture medium without FBS. Next, the medium was re- (LC) system (Dionex Ultimax 3000, California, USA) at- moved and formazan crystals were dissolved in 10% so- tached to a quadrupole-time of flight mass spectrometry dium dodecyl sulfate in 0.01 M HCl. The absorbance was (MS) detector with electrospray ionization (ESI) interface then measured at 570 nm using PerkinElmer 1420 multila- (Bruker micrOTOF-Q, Bruker Daltonics, Massachusetts, bel counter (PerkinElmer, San Jose, California, USA). Cell USA). The system was equipped with a reversed-phase viability is expressed as percentage compared to the un- Synergi Hydro column (250 × 4.6 mm i.d., 4 µm particle treated cells. size, 30 oC; Phenomenex, Milford, New Zealand). The mo- bile phase composed of methanol (solvent A) and a mixture E. Statistical data Analysis of 1% (v/v) formic acid, 5% (v/v) acetic acid and 5% (v/v) The statistical significance of the difference was deter- methanol (solvent B), with a flow rate of 1 mL/min. The mined using a one-way ANOVA followed by Tukey’s test elution conditions were as followed: 0-5 min, 0% A; 5-20 (SPSS Statistics version 20; IBM Corporation, New York, min, 0-95% A; 20-25 min, 95-40% A; and 25-35 min, 40- USA). The criterion employed for statistical significance of 0% A. The LC eluate was introduced directly into the ESI the difference was p<0.05. interface. The ESI voltage was 4.0 kV in positive ion mode. A nebulizing gas pressure of 0.4 bars and a drying gas of 4 III. RESULTS AND DISCUSSION L/min and temperature (453 K) were applied for ionization using nitrogen in both cases. The identification of individual A. Effect of PEF on the Release of Anthocyanins from anthocyanins was performed by matching the retention Merlot times with absorbance at 520 nm and mass spectra obtained Table 1 compares the content of anthocyanins released with authentic standards, otherwise correlating with those from untreated and PEF-treated Merlot. As compared to previously reported in the literature regarding Merlot grape other wine grape cultivars, Merlot is known for its thicker cultivar [15, 16]. The content of anthocyanins was estimated cell wall with multiple layers of skin cells [18]. Therefore, using external calibration curves that were constructed us- this implies a limitation in the release of anthocyanins from ing the respective authentic standards based on peak area. Merlot grape skin cells. In this study, there was no statistically significant difference in the juice recovery compared C. Cell Culture and Induction of Oxidative Stress between untreated and PEF-treated Merlot mash at different Merlot juice was subjected to gastric and intestinal diges- contact times. The juice yield collected from all tested sam- tive enzymes and pH changes [17] before exposing to the ples was in the range of 33 and 36%. Despite this, it was cells. Human Caco-2 cell lines (HTB-37; American Type clearly evidenced that almost no anthocyanins were ex- Culture Collection, Rockville, Maryland, USA) were cul- tracted from the untreated Merlot. After allowing a total tured in Dulbecco's modified Eagle's medium (Gibco, New contact time of 48 h between the juice and berries, a minim- York, USA), supplemented with 1% non-essential amino al amount and type of anthocyanins was released from un- acids, 1% L-glutamine, 20% heat-inactivated fetal bovine treated Merlot. serum (FBS), 100 units/mL of penicillin and 100 μg/mL of A PEF pretreatment on Merlot immediately resulted in a streptomycin. Cells were grown in a humidified atmosphere release of anthocyanins totaled at 0.87 ± 0.02 mg per 100 ml of 5% CO2 at 37 °C and the culture medium was replaced juice from the skin cells. Within 48 h contact time, all of the every 2 to 3 days. Cells were cultivated at weekly intervals, anthocyanins detected in the skin of Merlot berries were sub-cultured after reaching approximately 80% confluence released into the juice as a result of PEF processing, as in all experiments. The untreated cells were taken as ‘con- compared to only three malvidin glucosidic derived- trol’. The culture medium was removed and replaced with anthocyanins being released from untreated Merlot. This fresh medium containing 25% (v/v) of the digest diluted notable difference in the anthocyanins composition of juice Pulsed Electric Field Technology Enhances Release of Anthocyanins 49

released from untreated and PEF-treated Merlot is owing to Table 2 Potential bioprotective effects of Merlot juice, obtained the ability of PEF in affecting the cell wall structure of immediately (0 h) and after 48 h with and without PEF processing, on grape skin cells and altering the anthocyanins’ localization Caco-2 cell viability against H2O2-induced oxidative stress in the cell [19], thus facilitated the release of more antho- Cell viability (%) cyanins. Samples Cell not exposed to Cell exposed to 500 μM H2O2 (no stress) H2O2 for 1 h (stress) Table 1 Content of anthocyanins (in mg per 100 ml) released from Merlot Control 100 54 mash immediately (0 h) and after 48 h with and without PEF processing Untreated Merlot 0 h 107.75 ± 4.66 63.00 ± 2.80 Untreated Merlot 48 h 107.00 ± 5.28 61.50 ± 4.65 Untreated Merlot PEF-treated Merlot Anthocyanina PEF-treated Merlot 0 h 104.50 ± 5.55 77.75 ± 4.29 * 0 h 48 h 0 h 48 h PEF-treated Merlot 48 h 101.25 ± 3.47 94.50 ± 3.48 ** Mal-glu n.d. b 1.76 ± 0.16 0.72 ± 0.01 7.84 ± 0.82 Del-glu n.d. n.d. n.d. 0.15 ± 0.02 Mean ± SEM, n 4. * p<0.05 and ** p<0.001 when compared with the control (stress). Pet-glu n.d. n.d. n.d. 0.24 ± 0.02 Cyn-glu n.d. n.d. 0.02 ± 0.01 0.17 ± 0.01 Mal-acet-glu n.d. 0.25 ± 0.01 0.10 ± 0.01 0.94 ± 0.06 the juices did not exert toxicity on the cells. Addition of 500 Del-acet-glu n.d. n.d. n.d. 0.07 ± 0.01 μM H2O2, however, resulted in cell damage in which the Pet-acet-glu n.d. n.d. n.d. 0.09 ± 0.01 cell viability dropped to 54% without any juice supplemen- Mal-cou-glu n.d. 0.06 ± 0.01 0.03 ± 0.01 0.32 ± 0.01 Pet-cou-glu n.d. n.d. n.d. 0.06 ± 0.01 tation. Juice for untreated Merlot, regardless obtained at 0 Total anthocyanin n.d. 2.07 ± 0.22 0.87 ± 0.02 9.88 ± 0.83 or 48 h, provide no critical bioprotection against oxidative damage. The number of viable cells maintained between 61 Mean ± SEM, n 3. a Abbreviation: Mal, malvidin; Del, delphinidin; Pet, petunidin; Cyn, cyanidin; glu, glucoside; acet-glu, acetyl-glucoside; cou- and 63% and was not significantly different from the con- glu, p-coumaroyl-glucoside. b n.d.: not detected. trol stressed cells. This low number of viable cells under oxidative stress condition is probably because of the limited In this study, it was found that a total of 7.35% antho- amount and type of anthocyanins extracted into the juice. cyanins was released from the solid remains (grape pomace On the contrary, juice obtained immediately after PEF consisted of skins and pulps) within 48 h after PEF processing on Merlot significantly (p<0.05) attenuated the processing whereas only 1.79% of total anthocyanins was H2O2-induced oxidative stress by recovering a high number released from the solid remains of untreated Merlot during of viable cells. The greatest protection on cell viability the 48 h contact time with juice. Previous study also re- against oxidative stress was achieved for juice from Merlot ported that PEF accelerated the release of grape anthocya- after 48 h of PEF processing. The number of viable cells nins [20] and extraction of other valuable intracellular com- under oxidative stress condition was completely restored at pounds from various biological cells [21, 22]. Moreover, this the similar level as the cells without exposing to H2O2. The study demonstrated the release of anthocyanins by PEF is a observed complete bioprotection evoked by the PEF treated continual process and the amount and type of anthocyanins grape juice is probably because of the presence of all types being released increased with juice-skin contact time. of Merlot anthocyanins at high amount. The finding from this study highlights the important contribution of antho- B. Evaluation on the Bioprotective Effect of Merlot Juice cyanins to prevent the deleterious effect of oxidative This study employed a physiologically relevant model of damage. In cellular system, anthocyanins are reported to human intestinal epithelial (Caco-2 cells) to assess the bio- suppress oxidative damage by activation of glutathione- protective effect of Merlot juice containing varying amount related enzymes such as glutathione reductase, glutathione and types of anthocyanins released with and without facili- peroxidase and glutathione S-transferase, and subsequently tated by PEF. This approach takes into the consideration on elevating the level of reduced glutathione [24]. the way Merlot anthocyanins associate, interact and per- meate the membrane of the cells [23] as well as the degree IV. CONCLUSIONS they are protecting the cell viability under oxidative stress condition. Table 2 summarizes the bioprotective effect of This study has demonstrated that PEF enhanced the Merlot juice on Caco-2 cells, either in the absence or pres- release of anthocyanins from Merlot. The enhancement in ence of H2O2-induced oxidative stress. the anthocyanins content was revealed to promote signifi- When compared to the basal value of Caco-2 cells with- cant bioprotection effect on Caco-2 cells from oxidative out any juice supplementation (i.e. control) and absence of damage. The bioprotective effect was further maximized H2O2-induced oxidative stress, the exposure to various juice when the amount and types of anthocyanins released by samples did not affect the cell viability. This implies that PEF increased with time. Taking the advantage of PEF in 50 S.Y. Leong, I. Oey, and D.J. Burritt releasing anthocyanins into the juice, this will warrant a 12. El Darra N, Grimi N, Maroun R, Louka N, Vorobiev E (2013) Pulsed electric field, ultrasound, and thermal pretreatments for better phenolic health-promoting benefits in human upon consumption. extraction during red fermentation. Eur Food Res Technol 236(1): 47-56 13. Donsì F, Ferrari G, Fruilo M et al. (2010) Pulsed electric field-assisted vinification of Aglianico and Piedirosso grapes. J Agri Food Chem ACKNOWLEDGMENT 58(22): 11606-11615 14. López N, Puértolas E, Condón S et al. (2008) Application of pulsed SY Leong is supported by the University of Otago Doc- electric fields for improving the maceration process during vinification toral Scholarship. The research was funded by University of of red wine: Influence of grape variety. Eur Food Res Technol 227(4): 1099-1107 Otago Priming Partnership and the grapes were provided by 15. Dimitrovska M, Bocevska M, Dimitrovski D et al. (2011) Anthocyanin Villa Maria Estates winery. composition of Vranec, Cabernet Sauvignon, Merlot and Pinot Noir grapes as indicator of their varietal differentiation. Eur Food Res Technol 232(4): 591-600 CONFLICT OF INTEREST 16. Romero-Cascales I, Ortega-Regules A, López-Roca JM et al. (2005) Differences in anthocyanin extractability from grapes to wines accord- The authors declare that they have no conflict of interest. ing to variety. Am J Enol Vitic 56(3): 212-219 17. Glahn RP, Lai C, Hsu J et al. (1998) Decreased citrate improves iron availability from infant formula: Application of an in vitro diges- REFERENCES tion/Caco-2 cell culture model. J Nutr 128(2): 257-264 18. Ortega-Regules A, Ros-García JM, Bautista-Ortín AB et al. (2008) 1. Wang LS, Stoner GD (2008) Anthocyanins and their role in cancer Differences in morphology and composition of skin and pulp cell walls prevention. Cancer Lett 269(2): 281-290 from grapes (Vitis vinifera L.): Technological implications. Eur Food 2. Prior RL, Wilkes SE, Rogers TR et al. (2010) Purified blueberry an- Res Technol 227(1): 223-231 thocyanins and blueberry juice alter development of obesity in mice 19. Cholet C, Delsart C, Petrel M et al. (2014) Structural and biochemical fed an obesogenic high-fat diet. J Agri Food Chem 58(7): 3970-3976 changes induced by pulsed electric field treatments on Cabernet Sau- 3. Krikorian R, Nash TA, Shidler MD et al. (2010) Concord grape juice vignon grape berry skins: Impact on cell wall total tannins and poly- supplementation improves memory function in older adults with mild saccharides. J Agri Food Chem 62(13): 2925-2934 cognitive impairment. Br J Nutr 103(5): 730-734 20. Corrales M, Toepfl S, Butz P et al. (2008) Extraction of anthocyanins 4. Ghosh D, McGhie TK, Zhang J et al. (2006) Effects of anthocyanins from grape by-products assisted by ultrasonics, high hydrostatic pres- and other phenolics of boysenberry and blackcurrant as inhibitors of sure or pulsed electric fields: A comparison. Innov Food Sci Emerg oxidative stress and damage to cellular DNA in SH-SY5Y and HL-60 Technol 9(1): 85-91 cells. J Sc Food Agri 86(5): 678-686 21. Ohshima T, Sato M, Saito M (1995) Selective release of intracellular 5. Heo HJ, Lee CY (2005) Strawberry and its anthocyanins reduce oxida- protein using pulsed electric field. J Electrostat 35(1): 103-112 tive stress-induced apoptosis in PC12 cells. J Agri Food Chem 53(6): 22. Parniakov O, Lebovka NI, Hecke E et al. (2014) Pulsed electric field 1984-1989 assisted pressure extraction and solvent extraction from mushroom 6. Youdim KA, Martin A, Joseph JA (2000) Incorporation of the elder- (Agaricus bisporus). Food Bioprocess Technol 7(1): 174-183 berry anthocyanins by endothelial cells increases protection against 23. Spencer JPE, Abd El Mohsen MM, Rice-Evans, C (2004) Cellular oxidative stress. Free Radical Biol Med 29(1): 51-60 uptake and metabolism of flavonoids and their metabolites: Implica- 7. Zhang B, Kang M, Xie Q et al. (2010) Anthocyanins from Chinese tions for their bioactivity. Arch Biochem and Biophys 423(1): 148-161 bayberry extract protect β cells from oxidative stress-mediated injury 24. Shih PH, Yeh CT, Yen GC (2007) Anthocyanins induce the activation via HO-1 upregulation. J Agri Food Chem 59(2): 537-545 of phase II enzymes through the antioxidant response element pathway 8. Pinelo M, Arnous A, Meyer AS (2006) Upgrading of grape skins: Sig- against oxidative stress-induced apoptosis. J Agri Food Chem 55(23): nificance of plant cell-wall structural components and extraction tech- 9427-9435 niques for phenol release. Trends Food Sci Technol 17(11): 579-590 9. Toepfl S, Heinz V, Knorr D (2005) Overview of pulsed electric field Address of the corresponding author: processing for food. Academic Press, London 10. Angersbach A, Heinz V, Knorr D (2000) Effects of pulsed electric fields on cell membranes in real food systems. Innov Food Sci Emerg Author : Professor Indrawati Oey Technol 1(2): 135-149 Institute : Department of Food Science, University of Otago 11. Puértolas E, Saldaña G, Condón S, Álvarez I, Raso J (2009) A com- Street : PO Box 56 parison of the effect of macerating enzymes and pulsed electric fields City : Dunedin 9054 technology on phenolic content and color of red wine. J Food Sci Country : New Zealand 74(9): C647-C652 Email : [email protected]

Effects of Pulsed Electric Fields on Selected Quality Attributes of Beef Outside Flat (Biceps femoris)

F. Faridnia1, P. Bremer1, D.J. Burritt2, and I. Oey1 1 Department of Food Science, University of Otago, Dunedin, New Zealand 2 Department of Botany, University of Otago, Dunedin, New Zealand

Abstract Pulsed electric field (PEF) processing is a non- between enzymes and their substrates are enhanced. PEF thermal food processing technology that applies brief (µs) has previously been used to improve the microdiffusion of a electrical pulses of high voltage to food products placed be- brine solution into ham during processing, to enhance the tween two electrodes. Depending on the process intensity, the permeability of cellular components and to induce cellular process affects cell membrane permeability due to localised changes in muscle foods [1-4]. However, PEF use in meat structural changes. There are few reports on the use of PEF for the processing of solid foods like meat. Therefore, the cur- processing is relatively limited despite the release of several rent project was designed to assess the impact of PEF process- industry briefs and reports highlighting its potential bene- ing on beef muscles and its potential use to reduce meat ageing fits, particularly as a method to improve muscle texture by time and cost. Post rigor Biceps femoris (a low value beef mus- enhancing cell disintegration. cle) cuts were exposed to PEF (electric field strength of In this study the potential use of PEF for improving the 1.7 - 2.0 kV/cm and pulsed electrical energy of 185 kJ/kg) quality of selected low-value beef muscles at various ageing o processing and sampled after aging at 4 C for 3, 7, 14 or 21 times was investigated. days. Samples were assessed for temperature increases, electrical conductivity, pH, purge loss, cooking loss, tenderness and colour stability. The microstructure of PEF treated and un- II. MATERIALS AND METHODS treated meat samples were also investigated. Our results showed that temperature, electrical conductivity, and pH were A. Raw Materials significantly (P<0.05) affected by PEF treatment conditions. Beef Biceps femoris muscles from 6 animals (mean cold Tenderness, as indicated by a reduced shear force value, sig- carcasses weight of 245–285 kg) were obtained at 24 h post- nificantly (P<0.05) increased following PEF treatment com- mortem from a local commercial slaughter-house (Silver pared to the untreated samples. Cooking loss was not affected by PEF treatment, whilst purge loss significantly (P<0.05) Fern Farms Ltd., Finegand Plant, Balclutha). Upon arrival, increased after PEF treatment and during ageing. The PEF all visible fat and connective tissue was removed by trim- treated meat samples showed a dramatic increase in the num- ming. On the day of treatment, the muscles for both control ber of myofibrils ruptured along the z-lines compared to the and PEF treatment were cut parallel to the fibre direction untreated samples during ageing. This resulted in beef muscle into triangular pieces using a guided chopping board fitted with a more porous structure compared to the untreated sam- with a tailored-made stainless steel triangular blade. The ples and accounts for the observed increase in electrical con- form and dimension of the triangular blade were the same as ductivity and purge loss. These results suggest that PEF those of PEF batch treatment chamber (6 cm height × 4 cm induces changes in the microstructure and texture of meat and width × 6 cm length). The weight of the sample was ~70 g could potentially be used to improve tenderness, decrease ageing time or to alter functional properties. for each piece and the fibre direction was arranged to be perpendicular to the electric current. Keywords Pulsed electric field processing, tenderness, beef, ageing, meat quality. B. PEF Treatments The pulsed electric field treatment was carried out using I. INTRODUCTION Elcrack-HPV5 (DIL, Quakenburck, Germany) in batch mode. The batch treatment chamber consisted of two paral- Pulsed electric field (PEF) processing is a non-thermal lel stainless steel electrodes separated by a distance of 40 food processing technology that applies brief (µs) electrical mm. The operating variables used in this experiment were pulses of high voltage to food products placed between two as follows: constant pulse width of 20μs, electric field electrodes. PEF is of interest due to its ability to cause local- strength of 1.7kV/cm, constant frequency of 50 Hz, and ised structural changes to cell membranes that can result in total specific energy input of 185kJ/kg. Pulse shape (square increased cell membrane permeability, physical changes at a wave bipolar) was monitored on-line with an oscilloscope cellular level and/or faster rates of reactions, as interactions (Model UT2025C, Uni-Trend Group Ltd., Hong Kong,

© Springer Science+Business Media Singapore 2016 51 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_12

54 F. Faridnia et al.

REFERENCES 8. Lebovka N I, Bazhal M I, Vorobiev E (2002) Estimation of characteristic damage time of food materials in pulsed-electric fields. J 1. Toepfl S, Heinz V, Knorr D (2006) Pulsed Electric Field Technology Food Eng 54(4): 337 346. for the Food Industry: Fundamentals and Applications. Springer, 9. Arroyo C, Eslami S, Brunton N P et al. (2015). An assessment of the New York. impact of pulsed electric fields processing factors on oxidation, 2. Gudmundsson M, Hafsteinsson H (2001) Effect of electric field- color, texture, and sensory attributes of turkey breast meat. Poult Sci, pulses on microstructure of muscle foods and roes. Trends Food in press. Sci. Technol. 12:122 128. 10. Arroyo C, Lascorz D, O’Dowd L, Noci F, Arimi J, Lyng J G (2015) 3. Toepfl S (2006) Pulsed electric fields (PEF) for permeabilization of Effect of pulsed electric field treatments at various stages during con- cell membranes in food and bioprocessing Applications, process and ditioning on quality attributes of beef longissimus thoracis et lumbo- equipment design and cost analysis. PhD Thesis. Technische Univer- rum muscle. Meat Sci. 99: 52 59. sitat Berlin, Berlin, Germany. 4. McDonnell C K, Allen P, Chardonnereau F S, Arimi J M, Lyng J G 11. Lopp A, Weber H (2005) Untersuchungen zur optimierung derzar- (2014) The use of pulsed electric fields for accelerating the salting of theit von rindfleisch: Research into the optimizing the tenderness of pork. Food Sci. Technol. 59:1054 1060. beef from parts of the forequarter. Fleischwirtschaft.85:111 116. 5. O’Dowd L P, Arimi J M, Noci F, Cronin D A, Lyng J G (2013) An 12. Bekhit A E D, van de Ven, R, Suwandy V, Fahri F, Hopkins D L assessment of the effect of pulsed electrical fields on tenderness and (2014) Effect of pulsed electric field treatment on cold bonedmuscles of selected quality attributes of post rigor beef muscle. Meat Sci. 93: different potential tenderness. Food BioprocessTechnol 7:3136 3146. 303 309. 6. Faridnia F, Bekhit A E D, Niven B, Oey I (2014) Impact of pulsed Corresponding author: electric fields and post-mortem vacuum ageing on beef longissimus thoracis muscles. Int. J. Food Sci. Technol. 49:2339 2347. Professor Indrawati Oey 7. Faridnia F, Ma Q L, Bremer P J, Burritt D J, Hamid N, Oey I (2015) Department of Food Science, University of Otago Effect of freezing as pre-treatment prior to pulsed electric field proc- 276 Leith Walk, Dunedin 9054, New Zealand essing on quality traits of beef muscles. Innov. Food Sci. Emerg. Email:[email protected] Technol. 29: 31-40.

56 S. Li et al.

Table 1 Typical key parameters of PEF for food treatment

Institute V/I E. Field Pulse Fre. Duration Tem. Polarity Electrode others Target Year (kV/kA) (kV/cm) number (Hz) ( μs ) (°C) shape Shape UW Varying Monopolar Radial apple ider 5.1 / 1 40 3 1-7 < 50 2700 J 2011 with duration square 1.27mm KIT Monopolar suger beet 350/6-8 > 2 NI 20 1.4 NI special 15 t/h 2005 exponent KIT Monopolar 800 kg/h grape 300/4 < 60 NI 20 1.5 room Parallel 2009 exponent Industry Vegetables KU Parallel 30/NI 6 < 3000k 200 NI room Monopolar 5 J/pulse Fruits 2013 5cm pork TUB Monopolar Parallel < 40/NI 0.5-5 1000 < 17 10-400 NI - shoulder 2009 exponent 1300mm2 UST Monopolar Cylinder 1.0-2.7 egg white 25/NI 21-36 200 1 0.7-1.7 25-30 2002 exponent < 5mm J/pulse Parallel fish IFL 1.2,2.0 20-120 Monopolar HP 200 - 20/NI 1-4 400 25 1125cm2, shell 2006 > 2 > 90 exponent 300MPa 8cm UW University of Waterloo, Canada; KIT - Karlsruhe Institute of Technology, Germany; KU Kumamoto University, Japan; TUB Berlin University of Technology, Germany;UST University´ des Sciences et Techniques, France; IFL Iceland Fishes Laboratory, Iceland;

Table 2 Detailed parameters of the generator

Primary stage Pulse transformer Magnetic compression stage Load (treatment chamber)

L L C (mF) L (μH) p s M C (nF) L (mH) L (μH) T(μs) Z (Ω) C (pF) L (nH) 0 0 μ (mH) 1 uns sat z z z ( H) 0.8 1.5 5.4 4.2 0.85 20 20 9 17 25-125 120* 5

Lp, Ls primary and secondary inductance of pulse transformer; M coupling coefficient of pulse transformer; Lsat, Luns inductance of the MS at saturated and unsaturated state; T saturated time of the MS; Zz, Cz, Lz characteristic impedance, capacitance and inductance of the load;

* calculated at εr 100 stage, step-up stage, magnetic compression stage and the change and relatively lower price [19-21]. Double parallel load. Working process of the generator can be described as winding are rolled around the cores for decreasing the stray following. Initially, the power source transfers standard impedance and improving magnetism uniformity [22]. Total power supply (SPS) from the normal electric network weight and volume of the generator is expected to be (220V/50Hz, China) to 2kV/DC for charging the primary approximately 90kg and 0.20m3, respectively. Detailed capacitor (C0). After the semiconductor switch (PCT) is parameters of the generator are shown in Table 2. closed, the energy goes into the pulse transformer (PT) through the resonance inductance (L0). III. CIRCUIT SIMULATION AND ANALYSIS Amplitude of the voltage is increased by the pulse trans- OF THE GENERATOR former and transferred to the discharge capacitor (C1). With a proper designed Volt-Sec product, the magnetic switch Characteristics of the output pulses directly influence the (MS) is closed when the voltage on C1 hits its peak ampli- performance of the electroporation which is generally con- tude. Then, the energy is delivered to the load (Z) establish- sidered as the main effect of the PEF treatments. Based on ing an electric field for food treatment applications. the above mentioned results, full circuit model of the pulse According to the research in our group, an air-core Tesla generator, as shown in Fig. 1, is developed using the P- type pulse transformer is proper for the step-up stage with Spice software with a customized magnetic switch part transfer ratio over 20. A phase control thyristor with rated adopting sub-circuit method [23]. voltage and current of 4.2kV and 4.32kA, respectively, is Theoretically, the impedance of the chamber (dummy selected as the semiconductor switch balancing the perfor- load in the simulation) changes with different kind of foods mance and the cost [18]. As the working frequency of mag- and temperature. It is necessary to analysis the performance netic switch is 25kHz approximately, Fe-based amorphous of the generator with the varying impedance. As the con- magnetic cores are selected for their high flux density ductivity of the most of target foods are ranging from

IFMBE Proceedings Vol. 53

58 S. Li et al. tested with the experimental platform. Simulation and expe- CONFLICT OF INTEREST rimental results are shown in Fig. 7. It can be drawn that pulse with peak current of 1.35kA and duration of 1.1μs is The authors declare that they have no conflict of interest. generated in the chamber when there is a flash over. Con- siderations of the insulation inside the treatment chamber REFERENCES are needed for security reason. 1. L. Barsotti, P. Merle, J. C. Cheftel. (2009). Food processing by 0 10203040 1500 1500 pulsed electric fields. I. Physical aspects. Food Reviews International, 15:2, pp: 163-180.

1000 1000 2. L. Barsotti, J. C. Cheftel. (2009). Food processing by pulsed electric experimental fields. II. Biological aspects. Food Reviews International, 15:2, pp: simulation 500 500 181-213. 3. H. Bluhm. "Pulsed Power System: Principles and applications", Springer Express. Current [ A ] A [ Current 0 0 4. M. Gaudreau, T. Hawkey, J. Petry, M. Kempkes. (2005). Solid-state short circuit termination Power System for Pulsed Electric Field (PEF). Processing. IEEE 19th -500 -500 0 10203040 International Conference on Pulsed Power. Time [ μs ] 5. W. H. Jiang, K. Yatsui, et al. (2004) Compact solid-state switched pulsed power and its applications. IEEE. 6. J. Raso, V. Heinz. Pulsed electric firlds technology for the food Fig. 7 simulation and experimental results with short-circuit load industry foundationals and applications. Springer express. 7. Georg. Mueller, W. An, Th. Berghöfer, et al. (2011). Status and recent Progress in Pulsed Power Applications at Karlsruhe Institute of Technology (KIT). IEEE international Pulsed Power Conference, V. CONCLUSION USA. 8. M. Sack, C. Schultheiss, et al. (2005) Triggered Marx Generators for In conclusion, we investigate a solid-state pulse generator the Industrial-Scale electroporation of Sugar Beets. IEEE Trans. on Ind. Applications, 41(3), pp: 707-714. based on the magnetic switch both numerically and experi- 9. Frey, W., Eing, C., Göttel, M., et al. (2012), Pulsed electric field mentally. The generator have potential advantage of high treatment of microalgae benefits for microalgae biomass repetitive rate achievability and long life time reliability, processing. 4th Euro-Asian Pulsed Power Conf. (EAPPC 2012)/19th Internat. Conf. on High-Power Particle Beams (BEAMS2012), which is proper to be used for food treatments by pulsed Karlsruhe, Germany. electric field (PEF). Specially, the pulse generator is de- 10. H. Akiyama, T. Sakugawa, et al. (2007) Industrial Applications of signed and the total weight and volume is expected to be Pulsed Power Technology. IEEE Transactions on Dielectrics and 90kg and 0.20m3, approximately. Circuit of the generator is Electrical Insulation. 14(5), pp: 1051-1064. 11. M. Kristiansen, Pulsed power applications, (1993) Digest of 9th IEEE simulated using the P-Spice software. Influence of the im- International Pulsed Power Conference, Vol.1, pp: 6-10. pedance of a dummy load is analyzed. The peak voltage 12. K. H. Schoenbach, S. Katsuki, R. H. Stark, et al. (2002) Bioelec- over 20kV and duration of 1μs-2.5μs can be achieved on the trics New applications for pulsed power technology. IEEE Trans. on Plasma Science, 30 (1), pp: 293-300. dummy load. As important parameters, characteristics of the 13. Toepfl. Stefan. Pulsed Electric Fields (PEF) for Permeabilization of cores used for establishing the magnetic switch were meas- Cell Membranes in Food- and Bioprocessing Applications, Process ured at the actual working frequency. Additionally, the and Equipment Design and Cost Analysis, dissertation. magnetic switch with winding number of 30turns and vo- 14. M. Sack, J. Sigler, et al. (2010) Research on Industrial-Scale Electro- 3 poration Devices Fostering the Extraction of Substances from Biolog- lume of 6000cm was tested on an equivalent experimental ical Tissue. Food Eng Rev, 2, pp: 147 156. platform. Typical current with peak amplitude over 520A, 15. M. Akiyama, S. Gnapowski, Y. Shigematsu, (2013). Softening of duration of 1.52μs was obtained with a dummy load of vegetables by pulsed power. IEEE 19th International Conference on 45Ω. Importantly, the short-circuit termination was also Pulsed Power. tested for possible flashover in the treatment chamber. Ex- 16. Laura Barsotti, Eliane Dumay. (2002) Effects of high voltage electric pulses on protein-based food constituents and structures. Trends in perimental results show reasonable agreement with the Food Science & Technology 12, pp: 136 144. numerical analysis. 17. Irek Klonowski, Volker Heinz, Stefan Toepfl. (2006). Applications of pulsed electric field technology for the food industry. Iceland Fishes Laboratory Report 06-06. ACKNOWLEDGMENT 18. Zhuzhou CSR, Power devices product guide (in Chinese). 19. R. Hasegawa. (2006) Advances in amorphous and nanocrystalline The authors would like to thank Prof. John Jelonnek and magnetic materials. Journal of Magnetism and Magnetic Materials. Prof. Jun Zhang for their encouragement. 304, pp: 187 191.

IFMBE Proceedings Vol. 53

Study on a Solid State Pulse Generator Based on Magnetic Switch for Food Treatments by Pulsed Electric Field (PEF) 59

20. R. Burdt, R. D. Curry, K. F. McDonald. (2006) Evaluation of nano- Author: Song Li crystalline material, amorphous metal alloys, and ferrites for magnet- Institute1: College of Opto-electronic Science and Engineering ic pulse compression applications, Journal of Applied Physics, Street1: Deya road 107#, Kaifu district 99(08D911). City1: Changsha 21. Liyuan. (2015). User guide of Amorphous magnetic cores. Country1: China 22. Jingming Gao, Hanwu Yng, Song Li. (2014) Investigation on a high Institute2: Institute for Pulsed Power and Microwave Technology power, low impedance, and long pulse generator based on magnetic switches. IEEE trans. On Plasma Science. 42(4), pp: 988-992. Street2: Hermann-von-Helmholtz-Platz 1 23. D. M. Barrett. (1989) Modeling the characteristics of a magnetically City2: Eggenstein-Leopoldshafen switched pulse-forming line. In Proc. 7th IEEE Int. Pulsed Power Country2: Germany Conf., pp: 167-170. Email: [email protected]

IFMBE Proceedings Vol. 53

Part III Electrical Impedance Measurement for Assessment of Electroporation Yield

Modeling Dynamic Electrical Impedance Spectroscopy Measurements on Electroporated Cells

Tomás García-Sánchez1, Antoine Azan2,3, Isabelle Leray2,3, Javier Rosell-Ferrer1, Lluis M. Mir2,3, and Ramon Bragós1 1 Universitat Politècnica de Catalunya, Department of Electronic Engineering, Electronic and Biomedical Instrumentation Group, Barcelona, Spain 2 CNRS, UMR8203 Vectorologie et thérapeutiques anti cancéreuses, Institut Gustave Roussy, Villejuif, France 3Univ. Paris Sud, UMR8203, Orsay, France

Abstract In the present work, electrical impedance spec- The conventional impedance spectroscopy technique con- troscopy (EIS) measurements were performed during electro- sists in sequentially applying and registering AC signals of poration of C2C12 cells monolayers. The measuring strategy, different frequencies (frequency sweep). The drawback of based on multisine excitations, allows acquiring full impedance this approach is the long time required to measure a full spectra at a rate of 1 spectrum/ms during the interval between spectrum and is not suitable for the study of fast membrane pulses of a traditional electroporation treatment. The multi- dynamics. However, more sophisticated methods using time frequency information provided by the measuring system was domain broadband signals, where the sample is simulta- then studied using models. In this study, the Cole model and an electrical equivalent circuit were used. The results show the neously excited at multiple frequencies, have been proposed ability of the system for monitoring the fast impedance dynam- to overcome this situation [6]. ics of cells during electroporation. The differences in the in- In the present work we make use of a broadband multi- formation shown at different frequencies suggest the ability of sine-based signal to register impedance during electropora- the system to measure different phenomena at the same time, tion with a high time resolution (1 spectrum/ms). Thanks to namely: membrane changes and medium conductivity varia- the speed of the proposed method, the system is able to tions. The use of the Cole model reveals the need of more com- measure multifrequency impedance evolution during the plex models to correctly separate and interpret the results. The inter-pulse time gap of a traditional electroporation treat- use of the proposed circuital model enables studying the dy- ment. The measurements are performed during in situ elec- namics of pore formation independently of the conductivity variations. This work confirms the advantages of performing troporation of adherent cultured cell monolayers thanks to EIS measurements in order to have a more complete snapshot the characteristics of the microelectrode assembly used. of the system behavior. The advantage of obtaining multifrequency information relies on the fact that different models can be used to im- Keywords Electrical impedance spectroscopy, electropo- prove the understanding of the effects caused by electric ration dynamics, adherent cells. field pulses enabling to give a more complete description of the phenomenon. In this work the traditional impedance I. INTRODUCTION Cole model and an equivalent circuit model are proposed to The study of the kinetics of the plasma membrane per- follow the dynamics of electroporated cells in vitro. turbations during electroporation has focused the interest of many researchers as an essential approach to understand the II. MATERIALS AND METHODS deep mechanisms that govern the phenomenon. Due to the transient behavior of some of the membrane permeable A. Measurement Setup structures within the range of milliseconds, the development Summarizing, the reference measurement signal consists of fast dynamical measurement methods is necessary. of a multisine burst with duration of 1 ms comprising 21 Among others, the measurement of the electrical properties frequencies following a Bilateral Quasi-Logarithmic (BQL) of membrane during pulse application has been extensively distribution from 5 kHz to 1.313 MHz. The signal is gener- applied to reveal the temporal evolution of pores [1-5]. Usually, these measurements consist in the observation ated in a NI PXIe-1062Q chassis from National Instruments of the applied current-voltage waveforms in order to extract using the NI PXIe-5122 arbitrary waveform generator; the conductivity variations produced in the system [1-4]. In voltage and current are acquired with a high-speed digitizer these studies, an abrupt increase in the electrical conductivi- PXIe-5122. A custom-built analog front-end is used to in- ty is detected in response to pulse application. The other terface the microelectrodes and the measurement hardware available approach consists in the study of the passive elec- enabling the possibility of recording current and voltage in a trical properties of the sample in the frequency domain [5]. four-wire configuration.

© Springer Science+Business Media Singapore 2016 63 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_14 64 T. García Sánchez et al.

Fig. 1 Schematic representation of the system and microelectrode assembly used. In the zoom view the connections both for electroporation and EIS measurements are shown.

Additionally, the NI PXI-2530 high-density multiconfi- A low-conductivity electroporation (LCE) buffer was guration multiplexer-matrix is used for controlling connec- used during the electroporation pulse delivery and imped- tions between the custom-built pulse generator used, the ance measurements. It consisted of 250 mM sucrose, 10 measurement system and the microelectrodes. mM Tris and 1 mM MgCl2 (pH 7, osmolarity 287 mmol/kg, conductivity 0.1 S/m). B. Microelectrodes Finally, as a permeabilization reporter, fluorescent The microelectrode assembly used in this study was specifi- nucleic acids stain Yo-Pro-1 Iodide (λex=491 nm, λem=509 cally designed to perform electroporation to adherent cell mo- nm, Life technologies, Saint Aubin, France) was added to nolayers cultured in standard 24 multiwell plates. The detailed the permeabilization buffer at a final concentration of 1 μM. explanation of the microelectrode concept and fabrication procedure can be found in [7]. Summarizing, the microelec- D. Experimental Procedure trodes consist of Printed Circuit Board (PCB) discs (15 mm For experiments, cells were removed from incubator and diameter), comprising six equally spaced spiral lines coiled in washed once with LCE buffer. Subsequently, 200 μl LCE + parallel (75 μm width and 150 μm spacing). The microelec- Yo-Pro (1 μM) were added. The complete experiment trode is designed to be positioned momentarily above the cell was performed in a controlled temperature environment monolayer during electroporation/measurements procedure (37 ºC) to avoid the effect of temperature changes in the avoiding contact between cells and electrodes by means of 10 measurements. μm microseparators patterned on the external end of each spiral The sequence of electroporation/measurements begins line. The microelectrode topology and a zoom showing the with 100 continuous initial pre-electroporation measure- connections is shown in Fig.1. ments (100 ms). Subsequently, the pulse generation is in- C. Cells and Chemicals itiated (8 biphasic pulses, duration 100 μs, frequency 1 Hz). C2C12 mouse myoblasts were seeded into 24 muti-well After a fixed delay of 15 ms the connections are switched to plates at an initial density of 15×103 cells/well with 1 ml of the measuring system and 860 multisine bursts are conti- growth medium comprising Dulbecco’s Modified Eagle’s nuously generated and acquired. Immediately, connections Medium (DMEM; High Glucose, GlutaMAXTM Supple- are switched again to the pulse generator and the system ment, pyruvate. Life technologies, Saint Aubin, France) waits for the next pulse (125 ms). The sequence is repeated supplemented with 10% fetal bovine serum (FBS) and 1% until completing the 8 pulses. The complete system is con- penicillin-streptomycin. Cells were kept in an incubator trolled by a custom software in Labview (National Instru- at 37 ºC under a 5% CO2 atmosphere for 24 h until full ments). Three different voltage to distance ratio (E=V/d) of confluence. 600, 1000 and 1400 V/cm were applied to cells.

IFMBE Proceedings Vol. 53

Modeling Dynamic Electrical Impedance Spectroscopy Measurements on Electroporated Cells 65

Once the sequence was finished, the microelectrodes were carefully removed and cells were left during 15 additional minutes before washing twice with PBS and adding fresh culture medium before inspection under inverted microscope Zeiss Axio Observer Z1 (Carl Zeiss, Jena, Germany) to detect Yo-Pro fluorescence.

III. RESULTS AND DISCUSSION Once acquired, measurements were processed in Matlab to calculate the impedance spectral information and subsequently fitted to the proposed models. First, the widely known Cole model was used for its simplicity. In Fig. 2a the evolution of the obtained Cole arcs during a complete electroporation procedure applying 1400 V/cm is shown as an example. Figs. 2b and 2c show the evolution of the corresponding Cole parameters R0 and R∞ for the same experiment. The dynamics of R0, which correspond to the impedance at low frequency, show clearly two different behaviors: i), a rapid impedance recovery immediately after each pulse com- patible with the fast resealing of membrane and ii), a continuous impedance decrease over time that relates with the formation of more stable membrane pores during the treat- Fig. 2 Cole model results. a) Evolution of the obtained arcs during a com- ment. However, if we observe the evolution of parameter R∞, plete experiment applying 1400 V/cm. b) and c) show the evolution of which corresponds to the high frequency impedance beha- model parameters R0 and R∞, respectively for the same experiment. vior, this interpretation is not completely coherent. R∞ exhibits a continuous decay pulse after pulse that, according to the In Fig. 3b the evolution of Rext, Rint and Rpores is shown impedance basic theory, should not directly correspond with for the same experiment (applying 1400 V/cm) previously membrane-related phenomena. On the contrary, the changes used with the Cole model. The interpretation of the model at high frequency are more likely to be the result of other parameters in this case is more intuitive. According to the collateral effects such as temperature or conductivity varia- behavior of R∞, Rext also follows a continuous decrease over tions produced by the change in the buffer composition due to time that can now be easily explained as a consequence of ion leakage through membrane pores, what also explains the the increase in the conductivity of the extracellular medium. behavior observed in R∞. Consequently, it is necessary to take Rint shows a negligible increase remaining almost constant into account in the previous explanation of R0 that the just during the procedure, what is not exactly the expected mentioned collateral effects also influence the impedance behavior but can be considered correct due to the low sensi- response at low frequency. This makes more complex the tivity of the system to changes in this parameter. More in- correct interpretation of the observed evolution of R0 and teresting is the behavior of the parameter Rpores. Similarly to makes not possible to differentiate between processes that can R0, it shows two different dynamics that can now be correct- follow similar dynamics. ly interpreted as membrane-related events. Due to this limitation, caused by the simplicity of the However, in this case, the term corresponding to the fast Cole model, a more complex equivalent circuit was next recovery of R immediately after each pulse does not used to fit the measurements. pores follow exactly the same evolution than R . This is explained In Fig. 3a the proposed circuit is depicted. It consists of a 0 because the effect of the conductivity variation that disturbs modified version of a 2R-C circuit replacing the capacitor R0 in the Cole model is now specifically modeled by Rext by a constant phase element (CPEm) and adding an extra resistor in parallel to model the opening and closing of and has lower impact in Rpores. Thus, the use of this circuital model allows the observation of the undisturbed fast mem- membrane pores (Rpores). Rext and Rint represent the extracellular and intracellular resistances, respectively. During the brane resealing more accurately. fitting procedure it was necessary to impose some precondi- In the present manuscript only the response of a single tioning due to the wide variety of possible solutions. Name- experiment applying 1400 V/cm has been shown for sim- ly, Rint was forced not to decrease over time, what makes plicity. However, the observed response for the different sense with the expected behavior of the intracellular resis- electric fields shows a clear relation between impedance tance during electroporation. response and field intensity.

IFMBE Proceedings Vol. 53

66 T. García Sánchez et al.

ACKNOWLEDGMENT This research was supported by the COST TD1104 Ac- tion in the frame of the Short Term Scientific Missions (STSM). This work was also supported by CNRS, Universi- ty Paris Sud and IGR. The work was also performed within the scope of the EBAM European Associated Laboratory (LEA).

CONFLICT OF INTEREST The authors declare that they have no conflict of interest.

REFERENCES

1. Kazuhiko Kinosita Jr and Tian Yow Tsong. Voltage-induced conductance in human erythrocyte membranes. Biochimica et Biophysica Acta (BBA) - Biomembranes, 554 (2):479–497, 1979. 2. Mojca Pavlin and Damijan Miklavcic. Effective conductivity of a suspension of permeabilized cells: A theoretical analysis. Biophysical Journal, 85(2):719–729, 2003. 3. Cukjati, D., et al., Real time electroporation control for accurate and safe in vivo non-viral gene therapy. Bioelectrochemistry, 2007. 70(2): p. 501-7.

4. Henning Krassen, Uwe Pliquett, and Eberhard Neumann. Nonlinear Fig. 3 a) Equivalent circuit used for modeling. b), c) and d) correspond curren-voltage relationship of the plasma membrane of single cho prespectively to the evolution of circuital parameters Rext, Rint and Rpores cells. Bioelectrochemistry, 70(1):71–77, 2007. for an example applying 1400 V/cm. 5. Stolwijk, J.A., et al., Impedance analysis of adherent cells after in situ electroporation: Non-invasive monitoring during intracellular mani- IV. CONCLUSIONS pulations. Biosensors and Bioelectronics, 2011. 26(12): p. 4720-4727. 6. Sanchez, B., et al., A new measuring and identification approach for In this work two different models are used to study the time-varying bioimpedance using multisine electrical impedance multifrequency electrical impedance measurements per- spectroscopy. Physiological Measurement, 2013. 34(3): p. 339. formed at a rate of 1 spectrum/ms during the interval be- 7. García-Sánchez, T., Guitart, M., Rosell-Ferrer, J., Gomez-Foix, A.M., tween consecutive pulses of an electroporation procedure. Bragós, R. A new spiral microelectrode assembly for electroporation The system is applied in situ to adherent cells thanks to the and impedance measurements of adherent cell monolayers. Biomedi- cal Microdevices 2014. 16, 575-90. specific design of the microelectrodes. The present results show how the use of spectroscopy measurements instead of single frequency measurements allows the use of models to study different effects separately. In this direc- Author: Tomás García-Sánchez tion, although the Cole model is usually used because of its Institute: Universitat Politècnica de Catalunya, Department of Electronic Engineering, Electronic and Biomedical simplicity and easy implementation, more complex circuits are Instrumentation Group. suitable in order to better interpret and distinguish the effects of Street: Campus Nord, Edifici C4, Jordi Girona, 1-3 different processes on the system. The use of such type of City: Barcelona measurements together with its modeling in real-time could Country: Spain represent a promising tool for monitoring the success of an Email: [email protected] electroporation treatment both in vitro and in vivo.

IFMBE Proceedings Vol. 53

Conductance Dynamics to Monitor and Control Cell Electropermeabilization

J. Teissié1,2 1 CNRS UMR5089 IPBS (Institut de Pharmacologie et de Biologie Structurale), 205 Route de Narbonne, BP 64182, F 31077 Toulouse, France 2 Université de Toulouse, UPS, IPBS, F 31077 Toulouse, France

Abstract Membrane electropermeabilization is observed resistor was monitored by a transient digital oscilloscope. when electric pulses over a critical strength are delivered on a As the voltage pulse applied on the parallel flat electrodes is cell suspension. This is associated with a time dependent square-waved, the voltage is kept constant, and the current change in the suspension conductance. The magnitude and the flowing through the resistor is directly related to the resis- kinetics of the conductivity changes can be used to get access to tance of the pulsing chamber. Measuring the voltage across the molecular event supporting the membrane electropermeabilization. In Bioprocesses, where extraction results from elec- the resistor then gives direct access to the conductance tropermeabilization, the control of the conductivity change of change in the cell suspension because the geometrical cha- the cell suspension allows the definition of optimized safe con- racteristics of the chamber are constant during the pulse. ditions. Very fast transient in current can be observed as the limits are brought by the rise time of the pulse generator (200 ns Keywords Electropermeabilization, conductivity changes, with our device) and the sampling time of the recorder. red blood cells, microalgae. Platinum black is used as a thin film to cover the flat platinum metal electrodes. This brings an effective surface area much I. INTRODUCTION higher than the geometrical surface area of the electrode and, therefore, exhibits a shorter electrode loading time. A cell can be considered from an electrical point of view No change is observed if bacteria- free pulsing buffer is as a sphere with a shell of dielectric containing a conducting used, showing that the contribution of Joule heating is neg- solution and suspended in an external conducting solution. ligible in the case of non-permeabilized cells. It should be Under steady state condition, the conductance of a diluted kept in mind that the conductance of a cell suspension is cell suspension can be considered controlled by the conduc- highly temperature dependent. tance of the external solution. An ohmic behavior is ob- The post pulse behavior can be obtained by applying low served when a voltage pulse is delivered on the cell suspen- AC voltage between the electrodes. This could be used to sion (as long as the Joule heating is small). When the follow the behavior along a train of pulses [8, 12, 13] associated field is large enough, electropermeabilization In the flow bipolar electropulsator (Deex Bio, Betatech), appears. 2 events are present: i) the dielectric shell has con- a different approach was used. A Clamp-on current logger ductive defects ii) there is leak of the internal (cytoplasmic) (Chauvin Arnoux) was used to monitor by a contact less solution. A sharp increase in the cell suspension conduc- approach the current that was delivered to the pulsing tance is observed. It reflects the dynamics of the structural chamber. This was accurate as long pulses (2ms) were ap- events affecting the membrane [1-7]. The relative magni- plied on the sample allowing the use of the reduced band- tude of the change due to the ion leakage is larger when the with offered by the logger. In an improved version of the pulsing (external) solution is with a low ionic content [8- Deex Bio, a Hall effect probe was part of the device and 11]. This is indeed the case in most studies where it is at- used as a current follower. tempted to limit the Joule heating. Conductivity changes are an easy to perform assay of electropermeabilization with a III. RESULTS very fast resolution. It can be used to get access to information on the cascade of events affecting the cell membrane A. Analysis of Human Red Blood Cells and Muscle Fibers during or after the electric pulse. Electropermeabilization: Implication of Membrane Proteins II. METHODS In 10% NaCI/90% sucrose isotonic mixture, for pulse duration of 0.1 ms, the conductivity remained unchanged as A. Conductance Changes of the Cell Suspension long as E did not exceed a threshold of about 1.5 kV/cm. A 1-ohm resistor was inserted in series with the pulsing Due to the low initial conductivity and the short pulse dura- chamber of the cell electropulser. The voltage across this tion (0.1 ms), the Joule heating was low and did not induce

© Springer Science+Business Media Singapore 2016 67 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_15 68 J. Teissié a thermal increase in conductivity. With higher field intensi- C. E coli: Wall Effect ties, conductivity of the erythrocyte suspension started to Electropermeabilization of bacteria is routinely used for increase at the onset of field application. All the cell mem- gene transfer (and expression) or for the extraction of cyto- μ brane in a sample became conductive in less than 1 s. [8] solic soluble proteins and/or DNA [18, 19, 20] In a low ionic medium at least 35% of the conducting Conductance changes were recorded with a high time defects was related to the opening of Na+/K+ ATPase resolution on a E. coli suspension in a low conductivity channels [14]. Indeed the membrane conductance increase buffer [21]. Two successive pulses of 20 ms were delivered generated by the externally applied electric field could be on the bacterial suspension. During the first pulse, a fast partially blocked by a specific inhibitor, ouabain, or by a conductance increase occurred. Ion leakage occured within specific cross-linking reagent, Cu++-phenanthroline, of the milliseconds. During the second pulse, applied 1 min after ATPase. The effect of ouabain was saturable and had a mid- the first, the change in conductance was slow and very point of saturation at 0.15 μM, the physiological inhibition small but the conductance at the beginning of the second constant of the drug. K+ ion in the external medium sup- pulse was already higher than at the end of the first. The pressed the effect of ouabain, as observed in physiological leakage of cytoplasmic constituents present during the pulse studies. This provides a direct evidence that the membrane continued when the field was switched off. The amplitude defects induced by the field pulse are more than just lipid of the slow increase following the first pulse was smaller “electropores” but involved other cellular components such than that detected during the pulse itself, showing that the as electrogenic membrane proteins. main permeabilization effect was present during the 20 ms Similar targets were characterized in muscle fibers. One field application. Studies of fast change in transmitted light 4 ms duration shock of -450 mV resulted in electro confor- confirmed this conclusion indirectly. Light transmitted by a mational loss of function of voltage-gated Na channels and bacterial suspension is controlled by the orientation distri- in functional reductions in muscle cells' excitability [15, 16] bution. An increase in transmitted light during the pulse was To monitor the conductivity changes in the red blood cell associated with the parallel alignment of bacteria along the membrane after the pulsation, a second pulse with the same field lines. However, the most direct information provided amplitude and duration was applied after various intervals. by the transmitted-light experiments is the rise time of the When the interval between the two pulses was sufficiently orientation change. Under field conditions currently used (E short (ms), conductivity continued to increase disregarding >2 kV/cm), the orientation process occurred within 1 ms, i. the intermission. A recovery was apparent during a long e. a fraction of the pulse duration. The comparison of the intermission (min), suggesting a resealing of the membrane kinetics in transmitted light and conductance change dem- defects. Nevertheless the diffusional efflux of ions greatly onstrated that the contribution of orientation process (less increased the conductivity of the extracellular medium and than 1 ms) to the conductance change (lasting more than 20 as such of the cell suspension. In the case of muscle fibers ms) was negligible [22]. spontaneous sealing of the membrane defects (denatured The extent of this change in current was directly con- transmembrane proteins) occurred. This could be improved trolled by bacterial concentration. When working with a by the addition of surface active polymers acting as mem- tenfold- diluted sample, the change was too small to be brane sealing agents. detected. B. Chinese Hamster Ovary Cells: Transport is Delayed D. Protein Electroextraction from Microalgae Conductivity of a cell suspension in a low ionic content It is possible to produce recombinant protein either in cy- isotonic solution was observed to sharply increase when a 1 tosol or chloroplast of microalgae under safe conditions [23]. ms 1 kV/cm pulse was delivered on the suspension. This As for many other cell factories, extraction of proteins is was a fast process that could be detected in less than 1 μs diminished by the cell wall barrier. Optimization of algal cell after the pulse onset. Membrane conducting defects were disruption methods is fundamental from an economic point of induced on such a short delay. Transport of Propidium view [24]. Pulsed electric fields are now recognized as an iodide was observed at the single cell level by a fast fluo- efficient tool for Biotechnological processing [25]. Cytop- rescence assay (again a time resolution in the μs range). lasmic proteins can be extracted from walled species (plant, Transport was detected only after 60 μs [17] bacteria, yeasts as well as microalgae) by electropulsation The main conclusion was that theoretical studies on (pulsed electric field technology, PEF) [19, 26, 27]. lipid assemblies predicting that transport started within the E. Microalgae Electropermeabilization first microseconds of the electropermeabilizing pulse could not be supported by our experimental observations on Protein leakage was induced on Chlorella vulgaris by ap- mammalian cell. plying a train of 15 2 ms pulses with alternating polarities on

IFMBE Proceedings Vol. 53

Conductance Dynamics to Monitor and Control Cell Electropermeabilization 69 a microalgae suspension (5.107 cell/mL) in water followed Research was conducted in the scope of the EBAM Eu- by an overnight incubation in a salty solution containing ropean Associated Laboratory (LEA). DTT. The current delivered by the pulse generator at a given CNRS-IPBS is member of COST Action TD 1104. applied voltage was a direct assay of the conductivity of the solution present in the pulsing chamber . It was observed to CONFLICT OF INTEREST be much higher (about 20% increase) than the value measured before the pulse delivery (200 µS/cm). This high value The authors declare that they have no conflict of interest. was further checked by measuring the conductivity of the recovered pulsed samples. This conductivity increase was REFERENCES larger when the microorganism density was brought from 5.107 to 108 cell/mL. This observation was indicative of the 1. Pavlin M, Miklavcic D. (2003) Effective conductivity of a suspension release of ions present in the cytoplasm as a consequence of of permeabilized cells: a theoretical analysis. Biophys J. 2003 the plasma membrane electropermeabilization [9, 21]. Aug;85(2):719-29 One direct consequence of this increase in conductance 2. Pavlin M, Kanduser M, Rebersek M, Pucihar G, Hart FX, Magjarevic R, Miklavcic D. (2005) Effect of cell electroporation on the conduc- of the sample is that the temperature increase due to the tivity of a cell suspension. Biophys J. 2005 Jun;88(6):4378-90 Joule heating is very high. The joule heating is proportional 3. Pavlin M, Leben V, Miklavcic D. (2007)Electroporation in dense cell to the square of the field intensity, to the cumulated pulse suspension--theoretical and experimental analysis of ion diffusion duration (applied on the sample) and to the sample conduc- and cell permeabilization. Biochim Biophys Acta. 2007 tivity. As expected, we observed that the temperature of the Jan;1770(1):12-23. solution was over 40°C. A transient heat shock was there- 4. Krassen H, Pliquett U, Neumann E. (2007)Nonlinear current-voltage relationship of the plasma membrane of single CHO cells. Bioelec- fore applied on the sample. This did not appear to affect the trochemistry. 2007 Jan;70(1):71-7. protein release as we observed that the use of two succes- 5. Kakorin S, Neumann E. (2002) Ionic conductivity of electroporated sive pulsing chambers where only half of the pulses were lipid bilayer membranes. Bioelectrochemistry. 2002 May 15;56(1- delivered in each (to reduce the Joule heating) was not af- 2):163-6 fecting the protein extraction. But two open questions are 6. Abidor IG, Li LH, Hui SW. (1994a) Studies of cell pellets: I. Elec- present: i) does this heat shock play a role in the alteration trical properties and porosity. Biophys J. 1994 Jul;67(1):418-26. of the wall organization supporting the cytosolic protein 7. Abidor IG, Li LH, Hui SW. (1994 b) Studies of cell pellets: II. Os- motic properties, electroporation, and related phenomena: membrane release ?, ii) is a temperature denaturation of the released interactions. Biophys J. 1994 Jul;67(1):427-35 proteins present? In previous investigations on electroex- 8. Kinosita K, Jr. Tsong TY (1979) Voltage-induced Conductance in traction of proteins from yeasts, we observed that this Human Erythrocyte Membranes Biochimica et Biophysica Acta, 554 second drawback was not present as the specific activities (1979) 479--497. of the extracted proteins was higher than the one of mechan- 9. Kinosita K Jr, Tsong TY. (1978)Voltage-induced changes in the ical extracted proteins [26, 28]. One technical limit is the conductivity of erythrocyte membranes. Biophys J. 24(1):373-5 concentration of microorganisms to be treated. An increase 10. Kinosita K Jr, Tsong TT. (1977) Hemolysis of human erythrocytes by transient electric field. Proc Natl Acad Sci U S A. 74(5):1923-7. in the concentration brought a larger release of ions and a 11. Schmeer M, Seipp T, Pliquett U, Kakorin S, Neumann E (2004) higher temperature increase. Mechanism for the conductivity changes caused by membrane electroporation of CHO cell-pellets Physical Chemistry Chemical Phys- IV. ONCLUSIONS ics. 6: 5564 5574. C 12. Ramos A, Schneider AL, Suzuki DO, Marques JL. (2012) Sinusoidal Conductance changes along the electrotreatment of cell signal analysis of electroporation in biological cells IEEE transactions on Biomedical Engineering, 59 (10) suspension are both a powerful tool for the investigations of 13. El Zakhem H, Lanoisellé JL, Lebovka NI, Nonus M, Vorobiev the very fast events associated with the membrane and wall E.(2006) Behavior of yeast cells in aqueous suspension affected by reorganizations and an indicator of the changes in the puls- pulsed electric field Colloids Surf B Biointerfaces. 47(2):189-97 ing conditions (new ionic content of the solution, tempera- 14. Teissie J, Tsong TY. (1980) Evidence of voltage-induced channel ture increase by associated Joule heating). opening in Na/K ATPase of human erythrocyte membrane. J Membr Biol. 1980 55(2):133-40. 15. Chen W, Zhongsheng Z, Lee RC. (2006) Supramembrane potential- ACKNOWLEDGMENT induced electroconformational changes in sodium channel proteins: a potential mechanism involved in electric injury Burns. 32(1):52-9. This project was partly supported by Betatech (France), 16. Chen W, Han Y, Chen Y, Astumian D. (1998) Electric field-induced the French government supported ANR Debaciem and Eu- functional reductions in the K+ channels mainly resulted from su- ropean FP7 Project (Electroextraction, FP7-SMA-2007-1, pramembrane potential-mediated electroconformational changes. Bi- Grant agreement n°222220). ophys J. 75(1):196-206.

IFMBE Proceedings Vol. 53

70 J. Teissié

17. Pucihar G, Kotnik T, Miklavcic D, Teissié J. (2008) Kinetics of 24. Haberl, S. , Miklavcic, D. , Sersa, G. , Frey, W. , Ru-binsky, B. transmembrane transport of small molecules into electropermeabi- (2013) Cell membrane electroporation Part 2: The Applications lized cells. Biophys J. 95(6):2837-48 IEEE Electrical Insulation Magazine 29 (1): 29-37 18. Chassy BM, Flickinger JL (1987) Transformation of Lactobacillus 25. Ganeva V, Galutzov B and Teissié J (2003) High yield electroextrac- casei by electroporation FEMS Microbiology Letters 44 ( 2): tion of proteins from yeast by a flow process. Anal Biochem. 315:77- 173 177 84.Mahnič-Kalamiza S, Vorobiev E, Miklavčič D. (2014) Electropo- 19. Coustets M, Ganeva V, Galutzov B, Teissie J. (2015) Millisecond ration in food processing and biorefinery J Membr Biol. duration pulses for flow-through electro-induced protein extraction 247(12):1279-304 from E. coli and associated eradication. Bioelectrochemistry. 103: 26. Ganeva V, Galutzov B, Teissie J. (2014) Evidence that pulsed electric 82-91 field treatment enhances the cell wall porosity of yeast cells. Appl 20. Haberl S, Jarc M, Strancar A, Peterka M, Hodžić D, Miklavčič D. Biochem Biotechnol. 172(3):1540-52 (2013) Comparison of alkaline lysis with electroextraction and opti- 27. Ganeva V, Galutzov B, Teissié J. (2004) Flow process for electroex- mization of electric pulses to extract plasmid DNA from Escherichia traction of intracellular enzymes from the fission yeast, Schizosac- coli. J Membr Biol. 246(11):861-7. charomyces pombe. Biotechnol Lett. 26(11):933-7 21. Eynard N, Sixou S, Duran N, Teissie J. (1992) Fast kinetics studies of Escherichia coli electrotransformation. Eur J Biochem. 209(1):431-6 Author: Teissie J 22. Eynard N, Rodriguez F, Trotard J, Teissié J. (1998) Electrooptics Institute: Emeritus IPBS CNRS and University of Toulouse studies of Escherichia coli electropulsation: orientation, permeabiliza- Street: 118 route de narbonne tion, and gene transfer. Biophys J. 75(5):2587-96. City: Toulouse 23. Walker TL, Purton S, Becker DK and Collet C (2005) Microalgae as Country: France Email: [email protected] bioreactors. Plant Cell Rep 24:629-641.

IFMBE Proceedings Vol. 53

Electric Field Mapping in ex vivo Anisotropic Muscle Tissue Using DT-MREIT

W.C. Jeong1, S.Z.K. Sajib1, T.I. Oh1, H.J. Kim1, O.I. Kwon2, and E.J. Woo1,* 1 Kyung Hee University/Biomedical Engineering, Impedance Imaging Research Center, Yongin, Korea 2 Konkuk University/Mathematics, Seoul, Korea

Abstract Accurate coverage of tissue with a sufficiently Until recently, most MREIT techniques produced large electric field is one of the key conditions for isotropic or equivalent isotropic conductivity images. Since successful electroporation. Magnetic resonance electrical the muscle, for example, is known to be anisotropic, electric impedance tomography (MREIT) provides a means to field estimation using the reconstructed equivalent isotropic map the electric filed distribution during electroporation. conductivity values, therefore, could be inaccurate. Lately, To estimate the electric field strength, the magnetic flux Kwon et al. developed a new anisotropic conductivity density data induced by the electroporation pulses are image reconstruction method called DT-MREIT by measured from MREIT scans during electroporation. combining diffusion tensor MRI (DT-MRI) and MREIT [5]. Since biological tissues such as skeletal muscle are The key idea is from the observation that the water diffusion anisotropic, we propose a novel MREIT technique to map tensor and the conductivity tensor share their directions, the electric field in anisotropic as well as isotropic regions. which are determined by directional mobility values. In DT- We utilize the anisotropic conductivity estimation method MREIT, diffusion tensor images are utilized to provide the based on the lately developed DT-MREIT technique directional information of the conductivity tensor and its where diffusion tensor imaging is combined with MREIT. magnitude is estimated from the measured MREIT To estimate the current density in an optimal way, we magnetic flux density data. adopted the projected current density estimation In this study, we performed DT-MREIT imaging algorithm. From ex vivo experiments using bovine muscle experiments using ex vivo muscle tissues. We will describe tissues, we found that the new method produces electric the experimental methods to acquire diffusion tensor images field maps with a wider coverage of electroporation than the previous method. The results suggest that it is without applying any pulse and magnetic flux density images important to properly handle the effects of the tissue with applied pulses. After explaining the computational anisotropy for more accurate mapping of electric field methods to estimate both the current density and anisotropic during electroporation. conductivity, we will present the results of the estimated electric field map in the anisotropic muscle tissue. Keywords DTI, MREIT, Anisotropic conductivity, Electric filed, Electroporation. II. METHOD A. Imaging Experiment I. INTRODUCTION Imaging experiments were performed using chunks of Electroporation utilizes high intensity electric pulses to bovine muscle as shown in Fig 1. We inserted two silver- increase the permeability of the cell membrane. Monitoring wire electrodes into the muscle tissue about 50 mm. The of the electric filed during electroporation is, therefore, electrodes were insulated by heat-shrink tube except 2 mm needed in its clinical applications. Magnetic resonance at the tips. The diameter of silver-wire was 1.5 mm and the electrical impedance tomography (MREIT) utilizes induced distance between them was 30 mm. magnetic flux density data subject to applied electric current Using the 3T MRI scanner (Achieva TX, Philips, pulses to quantitatively visualize current density and Netherlands) and single-shot spin-echo EPI (SS-SE-EPI) conductivity distributions [1, 2]. Since we can estimate sequence in Fig. 2(a), we collected diffusion tensor image electric field intensity from the knowledge of current density data with b-values of 500 sec/mm2 and TR/TE= 4000/72 and conductivity, Kranjc et al. reported the feasibility of the ms. One reference MR data was also obtained without MREIT technique to monitor the electric filed distribution diffusion sensitized gradient to measure the diffusion tensor. during electroporation [3, 4]. The method is based on the Using a voltage stimulator (BSLSTMB, BIOPAC direct application of Ampere’s law and Ohm’s law without Systems, Inc., USA), 128 electric pulses were applied considering the effects of tissue anisotropy. through the electrode pair with 100 V amplitude and 100 μs width for MREIT data collection. We used the multi- gradient echo pulse sequence in Fig. 2(b) to acquire the * Corresponding author. © Springer Science+Business Media Singapore 2016 71 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_16 72 W.C. Jeong et al.

induced magnetic flux density data called Bz data. The imaging parameters were TR/TE = 200/1.6 ms, ES=2.3 ms, NEE= 32, FOV = 200×200 mm2, imaging matrix = 128×128, imaging time = 51.2 s.

B. Current Density Estimation

Since we could acquire only one component Bz of the magnetic flux density B = (Bx, By, Bz), we adopted the projected current density method [6-7], which approximately recovers the best three-dimensional current density using the measured data of Bz: Fig 1. Experimental setup. ⎛ ∂ − 0 ∂−0 ⎞ 1 (Bz BBBzz) ( z ) JJP = 0 + ⎜ − , ,0⎟ (1) μ ⎜ ∂y ∂x ⎟ 0 ⎝ ⎠ where J0 is the computed current density from the three- dimensional model with a homogeneous conductivity distribution subject to the same boundary conditions 0 as in real experiments and Bz is the z-component of the magnetic flux density corresponding to the computed current density J0.

C. Anisotropic Conductivity Tensor Estimation We adopted the assumption about the linear relationship between the conductivity tensor and the water diffusion tensor [8]. Since water molecules and ions are in the same structural environment, they share the same directional property in terms of their mobilities. Therefore, we set the conductivity tensor as a scalar multiple of the water diffusion tensor as

C =ηD (2) where C is the conductivity tensor, D is the diffusion tensor and η is the scalar factor. Using the diffusion tensor D from the acquired diffusion tensor images and the recovered projected current density JP, we propose the diffusion weighted J-substitution method η Fig 2. (a) Synchronized pulses with SPGMRE sequence to measure the to recover the position-dependent scale factor as induced magnetic flux density data. (b) Single shot spin echo EPI sequence to measure the water diffusion. P ⋅∇ 0 η =− J u (3) D∇u00⋅∇u D. Electric Field Estimation Once we obtained both the current density as a 3×1 where, u0 is the computed voltage from the model with the vector and the conductivity tensor as a 3×3 matrix, we homogeneous conductivity distribution. The conductivity computed the electric field by multiplying the inverse of the tensor C is then computed as ηD. conductivity tensor to the current density vector.

IFMBE Proceedings Vol. 53

Electric Field Mapping in ex vivo Anisotropic Muscle Tissue Using DT MREIT 73

III. RESULT

Fig. 3(a) and (b) show the acquired MR magnitude and Bz images, respectively, of the bovine muscle. Fig. 3(c) is the color-coded fractional anisotropy map obtained from the diffusion tensor images. Fig. 4 plots the computed projected current density image using (1). The position-dependent scale factor η was then obtained by incorporating the measured Fig 4. Estimated projected current density image. diffusion tensor with the recovered projected current density. After recovering the conductivity tensor as a scalar multiple of the water diffusion tensor, we computed the electric filed map. Fig. 5(a) is the recovered electric filed map using the anisotropic conductivity tensor. We also computed the electric field map using the equivalent isotropic conductivity obtained by using the J-substitution algorithm [9]. Fig. 5(b) shows the electric field map obtained by using the equivalent isotropic conductivity distribution.

IV. DISCUSSION AND CONCLUSIONS The accurate electric field mapping is important to correctly estimate the coverage of the electroporation treatment. We found that the electric field map obtained by using the anisotropic conductivity tensor from the DT-MREIT method is different from the one obtained by using the equivalent isotropic conductivity from the conventional MREIT method.

Fig 5. Electric field maps determined by using (a) the anisotropic conductivity tensor and (b) the equivalent isotropic conductivity.

To apply the proposed method to real electroporation experiments, we need to use a different MR pulse sequence where the number of applied electroporation pulses can be properly varied. The scan time must be also reduced for monitoring applications by using a fast MR imaging pulse sequence such as RARE or EPI. With these improvements, future studies should include more rigorous validations of the new method through in vivo animal experiments.

ACKNOWLEDGMENT This work was supported by the National Research Foundation (NRF) of Korea grant funded by the Korean government (MSIP) (No.2013R1A2A2A04016066, No.2014 R1A2A1A09006320).

CONFLICT OF INTEREST

Fig 3. (a) MR magnitude image, (b) Bz image and (c) color-coded fraction anisotropic map showing different directions of muscle fiber. The authors have no conflict of interest.

IFMBE Proceedings Vol. 53

74 W.C. Jeong et al.

REFERENCES 6. Park C, Lee B II, Kwon OI (2007) Analysis of recoverable current from one component of magnetic flux density in MREIT and MRCDI. Phys. Med. Biol. 52:3001-3013. doi: 10.1088/0031- 1. Seo JK, Woo EJ (2014) Electrical tissue property imaging at low 9155/52/11/005 frequency using MREIT. IEEE. Trans. Biomed. Eng 61:1390-1399. 7. Jeong WC, Sajib SZK et al. (2014) Focused current density imaging doi: 10.1109/TBME.2014.2298859 using internal electrode in magnetic resonance electrical impedance 2. Seo JK, Woo EJ (2011) Magnetic resonance electrical impedance tomography (MREIT). IEEE. Trans. Biomed. Eng. 61:1938-1946. tomography (MREIT). SIAM Review 53:40-68. doi: doi: 10.1109/TBME.2014.2306913 10.1137/080742932 8. Tuch DS, Wedeen VJ et al. (2001) Conductivity tensor mapping of the human brain using diffusion tensor MRI. Proc. Natl. Acad. Sci.

3. Kranjc M, Bajd F, Sersa I, et al. (2012) Ex Vivo and in silico 98:11697-11701. doi:10.1073/pnas.171473898 feasibility study of monitoring electric field distribution in tissue 9. Kwon OI, Woo EJ (2002) Magnetic resonance electrical impedance during electroporation based treatments. PLoS One 7(9):e45737. doi: tomography (MREIT): simulation study of J-substitution algorithm. 10.1371/journal.pone.0045737 IEEE. Trans. Biomed. Eng. 49:160-167. doi: 10.1109/10.979355

4. Kranjc M, Bostjan M et al. (2015) In Situ Monitoring of Electric The corresponding author: Field Distribution in Mouse Tumor during Electroporation. Radiology 274:115-123. doi: 10.1148/radiol.14140311 Author: Eung Je Woo Institute: Kyung Hee University 5. Kwon OI, Jeong WC et al. (2014) Anisotropic conductivity tensor Street: 1732, Deogyeong-daero, Giheung-gu imaging in MREIT using directional diffusion rate of water City: Yongin molecules. Phys. Med. Biol. 59:2955-2974. doi: 10.1088/0031- Country: Korea 9155/59/12/2955 Email: [email protected]

IFMBE Proceedings Vol. 53

Electrorotation as a Versatile Tool to Estimate Dielectric Properties of Multi scale Biological Samples 77 with the template used to describe the cell. Each sample has quency region between the negative velocity peak and the a typical ROT spectrum, which reflects its own dielectric single shell model [22] leads respectively to the permitivitty properties (Figure 2(a-b)). and conductivity of the outer shell εouter shell=3.5, σouter shell=2.3e-4 S/m (radius of the spheroid = 45 μm, estimated thickness of the outer shell = 8 nm). A more realistic model taking into account the complexity and the heterogeneity of the spheroid, and might correspond to our experimental observation is under investigation. Finally, the system proposed in this paper is an efficient tool to distinguish and furthermore characterize samples by their dielectric properties.

IV. CONCLUSIONS Electrorotation reveals to be a versatile tool since it can be used to characterize samples at different scale levels, from bacteria’s size (few μm) till spheroid’s size (hundreds μm). The dielectrophoresis and electrorotation techniques were successfully combined to estimate the dielectric properties of various biological samples. In the presented system the nDEP force is first applied for cell trapping, then a rotating field is superimposed in order to induce a rotational movement on the sample. The evolution of the rotational velocity as a function of the frequency represents the electrorotation spectrum, the analysis of the latter, done through a fitting algorithm, gives as results the dielectric properties of the sample. Fig. 2 (a) Electrorotation spectra of biosamples (stars for E.Coli, Indeed the electro-rotation, a rotating electric field through circles for Jurkat cell line, triangles for Blood cell and pentagons a sample medium, can be used to elucidate the phenotypic for Sphreoid U87MG) (b) Biosamples centered on the interesting area differences of biological samples. These differences can be of the electrodes’ structure detected as electro-physiological fingerprints intrinsic to a The estimation of the dielectric properties of each sample cell which are both dependent upon structures both on/within is achieved by employing the trust-region-reflective algo- the cell. Electrorotation can be also employed to detect min- rithm, which is part of the Gauss-Newton algorithms, it was ute changes even within the same sample – from the single chosen since it is widely considered reliable and robust [20]. cell up to multicellular spheroids. For single cell, the estimated values obtained are shown As a perspective we expect to characterize the heteroge- in table 1: neity and time evolution of a multicellular spheroid with this method and contribute to establish a behavioural model Table 1 Dielectric of biosamples estimated by fitting the electrorotation of such complex biological system. spectrum

σ ε σ ACKNOWLEDGMENT Sample ε cyt,rel cyt [S/m] mem,rel mem [S/m] Bacteria 64 0.4 1 5e-5 The authors acknowledge the “Ecole Doctorale Sciences Blood cell 70 1.2 7.5 0.8e-6 Pratiques” of ENS Cachan, the “LASIPS” laboratory of Jurkat 40 0.19 6.3 4e-5 excellence, the “Institut d’Alembert” and the “CNRS” for project founding. We would like to thank Bianca Sclavi for Dielectric properties of all analysed samples showed val- supplying the bacteria. ues that are coherent with the literature [21]. The analysis of the first results obtained with the spheroid shows the predominant insulating behaviour of the CONFLICT OF INTEREST outer cell layer of the cell assembly (negative peak of the velocity at low frequencies). A fitting within this low fre- The authors declare that they have no conflict of interest.

IFMBE Proceedings Vol. 53

78 C.I. Trainito et al.

REFERENCES 13. Arnold W.M. and Zimmermann U.(1998) Electro-rotation: development of a technique for dielectric measurements on individual cells 1. Cen G., Colin Dal ton, Youlan Li, Sophia Adamia, linda M. Pilarski, and particles Journal of Electrostatics 21(2-3):151 191 DOI Karan V.I.S. kaler. (2004) A combined dielecrophoresis, traveling 10.1016/0304-3886(88)90027-7 wave dielectrophoresis and electrorotation microchip for the manipu- 14. Laforet, J., Frena-Robin, M., Buret, F., Nicolas, L.(2007) An auto- lation and characterization of human malignant cells. Journal of Mi- mated platform for cell dielectric spectroscopy International Journal crobiological Methods, 58: 387-401 DOI of Bioelectromagnetism 9(1):23-24 10.1016/j.mimet.2004.05.002 15. Han S.I., Joo Y.D. and Han K.H. (2013) An electrorotation technique 2. Huang C., Wu Y., Wang L. and Yu J.. (2006) Negative Dielectropho- for measuring the dielectric properties of cells with simultaneous use retic Force Assisted Determination Differences between Autotrophic and Heterotrophic Algal Cells Using Electrorotation, Proceedings of negative quadrupolar dielectrophoresis and electrorotation, Ana- of the 1st IEEE International Conference on Nano/Micro Engineered lyst, 138:1529-1537. DOI 10.1039/C3AN36261B and Molecular Systems. Zhuhai, China, 2006 pp. 310-315 DOI 16. Morgan, H., Green, N. G. (2003) AC Electrokinetics: colloids and 10.1109/NEMS.2006.334730 nanoparticles. Research Studies Press Ltd., England, 2003, pp 65-79. 3. Gascoyne P., Mahidol C., Ruchirawat M., Satayavivad J., Watchera- 17. Trainito C.I., Français O. and Le Pioufle B. (2015) Monitoring the sit P. and Becker.F.F. (2002) A combined Microsample preparation permeabilization of a single cell in a microfluidic device, through the by dielectrophoresis: isolation of malaria. Lab on Chip, 2:70-75. DOI estimation of its dielectric properties based on combined dielectro- 10.1039/B110990C phoresis and electrorotation in-situ experiments. Electrophoresis 4. Alshareef, M., Metrakos, N., Perez, E. J., Azer, F., Yang, F., Yang, Journal DOI 10.1002/elps.201400482 X., Wang, G. (2013) Separation of tumor cells with dielectrophoresis- 18. Bisceglia E., Cubizolles M., Mallard F., Vinet F., Français O. and Le based microfluidic chip Biomicrofluidics 7:011803 (pp 12) DOI Pioufle B. (2013) Micro-organism extraction from biological samples 10.1063/1.4774312 5. Cetin B. and Li D., Dielectrophoresis in microfluidics technology using DEP forces enhanced by osmotic shock. Lab on a Chip. (2011) Electrophoresis 32(18):2410-2427 DOI 10.1002/elps.201100167 13:901-909 DOI 10.1039/C2LC41128H 6. Castellarnau M., Errachid A., Madrid C., Juarez A. and Samitier J. 19. Arnold, W. M., Zimmermann, U. (1998) Electro-rotation: develop- (2006) Dielectrophoresis as a tool to characterize and differentiate ment of a technique for dielectric measurements on individual cells isogenic mutants of Escherichia Coli. Biophysical Journal, 91(19): and particles Journal of Electrostatics, 21:151-191 DOI 3937-3945 DOI 10.1529/biophysj.106.088534 10.1016/0304-3886(88)90027-7 7. Chopinet-Mayeux L. Effets des champs électriques pulsés milli et 20. Byrd, R.H., Schnabel, R.B., Shultz, G.A.(1987) A Trust Region nanosecondes sur cellules et tissus. (2013) Micro and nanotechnolo- Algorithm for Nonlinearly Constrained Optimization SIAM Journal gies/Microelectronics. Université Paul Sabatier - Toulouse III on Numerical Analysis, 1987, 24(5):1152 1170 DOI 8. Khoshmanesh, K., Nahavandi, S., Baratchi, S., Mitchell, A., Kalan- 10.1137/0724076 tar-zadeh, K., (2011) Dielectrophoretic platforms for bio- 21. P. Gascoyne, R. Pethig, J. Satayavivad, F.F. Becker, and M. Ruchi- microfluidics systems Biosensors and Bioelectronics, 26(5):1800 rawat (1997) Dielectrophoretic detection of changes in erythrocyte 1814 DOI 10.1016/j.bios.2010.09.022 9. Pohl, H.A. (1958) The Motion and Precipitation of Suspensoids in membranes following malarial infection. Biochimica et Biophysica Divergent Electric Fields. Journal of Applied Physics 29(8):1182- Acta (BBA) - Biomembranes, 1323:240- 252 DOI 10.1016/S0005- 1188. DOI 10.1063/1.1700065 2736(96)00191-5 10. Cheng, J., Sheldon, E.L., Wu, L.,(1998) Isolation of Cultured Cervic- 22. Frenea-Robin M. Micromanipulation de particules par diélectropho- al Carcinoma Cells Mixed with Peripheral Blood Cells on a Bioelec- rèse : application au rangement matriciel et au tri de cellules sur puce tronic Chip. Analytical Chemistry 70(11):2321 2326. DOI (2003) Ecole Normale superieure de Cachan. 10.1021/ac971274g 11. Elitas, M., Martinez-Duarte, R., Dhar, N., McKinney, J.D., Renaud, P. (2014) Dielectrophoresis-based purification of antibiotic-treated bacterial subpopulations. Lab on a chip 14:1850-1857 DOI Author: Claudia Irene Trainito 10.1039/C4LC00109E Institute: Ecole Normale Supérieure, SATIE laboratory 12. Pethig, R., Jakubek, L.M., Sanger, R.H., Heart, E., Corson, E.D., Street: 61 Avenue du Président Wilson Smith, P.J.S. (2005) Electrokinetic measurements of membrane capa- City: 94235 Cachan Cedex citance and conductance for pancreatic β-cells IEE Proc. Nanobio- Country: France technology 152(6): 89-193 DOI 10.1049/ip-nbt:20050040 Email: [email protected] cachan

IFMBE Proceedings Vol. 53

Electrical Measurements for Monitoring Electroporation

U. Pliquett Institut für Bioprozess und Analysenmesstechnik, Heilbad Heiligenstadt, Germany

Abstract Biological material shows characteristic electric- ii. THEORETICAL BACKGROUND al behavior between 10 kHz and 10 MHz which is mostly go- verned by cells surrounded by insulating membranes. A. Electroporation Electroporation of cell membranes renders the generally highly resistive lipid structure conductive. This results in dra- Electroporation within a relatively small range of field matic decrease of the low frequency impedance because ions strength and time duration is today well understood given can cross the membrane due to the electrically created aqueous that the widely accepted pore model is supported by expe- pores. Moreover, there is a pronounced impact on conductivity riment and MD-simulation [5]. of the extracellular medium due to an efflux of ions from the The membrane charges up in an outer electric field which cell plasma. forces water into the membrane structure. At about 200 mV – Given the significant impedance changes due to electropo- 1 V across the 5 nm thick lipid membrane, an aqueous pore is ration, electrical characterization is a powerful tool for moni- created when water enters the membrane from both sides toring this process. Since the time course of electroporation is forming a membrane spanning structure (electroporation). short compared to conventional impedance measurements and the material is driven into a non-linear range of the cur- Due to energy constrains, polar headgroups turn into the pore rent/voltage characteristics, only dynamic impedance mea- making it hydrophilic. The high electric field within the pore surements in presence of high dc-offset with sub-millisecond (≈ 1 MV/m) polarizes the water which stabilizes and expands time resolution are suitable for this purpose. the pore during the presence of the electric field. As soon as the field ceases, pores shrink and reseal. At least with respect Keywords electroporation, impedance, dynamic. to the membrane integrity, electroporation with moderate field strength and time duration is reversible [2]. I. INTRODUCTION The application of very short (ns-duration) but extremely intense (> 10 kV/cm) pulses (nanosecond pulsed electric Electroporation is an important method in biotechnology, field, nsPEF) increases the conductivity of cells and tissue food industry, pharmacy and medicine [1]. The impact on dramatically and even more than expected from the pore the integrity of the membranes increases with field strength model. Therefore, this model becomes increasingly ques- and pulse duration. While cells survive at low pulse energy tionable. A pronounced difference between classical elec- [2], they are killed at high field strength and may become troporation with field strength between 200 V/cm and 1 completely disintegrated at very high pulse energy [3]. kV/cm and the reaction to nsPEF is the thermodynamic The technical use of electric field application at biological equilibrium between pore creation and resealing. With material requires optimized protocols for highest economic impact of the method. A feedback based on the electropora- slight increase of the voltage, more pores are created while tion yield would give the opportunity for intelligent control. they reseal when the field strength decreases. This yields a Because of the dramatic increase of the membrane con- voltage regulator effect, clamping the voltage at a level ductivity, electrical characterization by means of impedance between 0.4 V and 0.6 V. If the slope of the stimulus is very measurement is one of the most promising methods for this steep, the voltage across the membrane rises faster than purpose. It is possible to access electrical properties online pores are created. This yields transmembrane voltage up to without labeling [4]. 1.4 V driving the lipid bilayer system away thermodynamic The challenge of using impedance measurement for moni- equilibrium, thereby forcing assemblies which are not fa- toring the electroporation process is that the material is driven vorable under physiological condition. For instance, due to into a nonlinear range of the current/voltage characteristic but space constrains, double tailed lipids arrange in bilayer also the extremely fast changes during and immediately after structures rather than micelles. Hydrophobic hydration high voltage application. A suitable approach is electrical interaction forces vanish in electrified membranes, thereby characterization in time domain with broad bandwidth excita- increasing the probability for micellar structures [6]. These tion signals having amplitudes small enough for operation in structures are not stable without electric field and rearrange a quasi linear range but also a measurement time where the into bilayer membranes on a time scale of nanoseconds after material changes its properties only negligibly. the electric pulse.

© Springer Science+Business Media Singapore 2016 79 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_18 80 U. Pliquett

The unique feature of electroporation is the dramatic in- slope of the I/U-curve (current/voltage characteristic). This crease of membrane conductance on a time scale of nano- to yields the small signal conductance Gss = dI/dU (also microsecond. A recovery of the membrane integrity, if hap- termed differential conductance) which is in this example pen, occurs on a much longer time scale between microse- 10 mS. If the nonlinear behavior is time invariant the small conds up to minutes or even hours. signal conductance as function of voltage is sufficient for electrical characterization. However, when biological ma- B. Impedance terial is electrically manipulated, time related structural The electrical impedance, Z, accounts for the hindrance changes and the complex nature of the electrical properties of the current driven by an electric field. While the real part yields a time varying admittance Y(t) (or Z(t)=1/Y(t)). is dissipative, meaning the electrical energy is dissipated Therefore, the impedance becomes a function of time and into other energy forms like heat, light, chemical or me- electric field. It is recalled that linear system theory is only chanical energy, the imaginary part arises from energy sto- applicable to linear and time invariant objects. Therefore, rage in conservative elements like capacitors and inductors. operations like Laplace or Fourier transformation are not Biological material exhibits pronounced capacitive behavior simply applicable making interpretation of measured results [7]. Permittivity, ε, is the physical quantity accounting for rather complicate. the capability of electrical energy storage. Besides energy storage, most material exhibits electrical conductivity which 1.5 corresponds to the imaginary part of the permittivity. The permittivity is related to the impedance by ε = 1/(jωε0Zk) 1

where k is a geometry factor which is for a typical measur- -1 ω ε ing chamber is around 1 cm . is the angular frequency, 0 / A 0.5 the permittivity of the free space and j the imaginary unit. P

A particular polarization mechanism yields a spectral 0 0 50 100 150 200 dispersion with pronounced changes in impedance and U / V permittivity. For instance the polarization of the cell membranes yields the ß-dispersion. At low frequency, current Fig.1 I/U characteristics of potato. A voltage ramp with 1 MV/s was flows around cells while at higher frequency cell mem- applied at an outer pair of electrodes and voltage between an inner pair of branes are capacitively shortened which decreases the im- electrodes (distance 3.5 mm) measured. The straight line marks the apparent conductance Gapp I/U in point P. pedance. Since cell membranes are not charged anymore, no energy is stored and the permittivity decreases as well. Oth- Dynamic impedance or its reciprocal, the dynamic admit- α er characteristic dispersions are the -dispersion resulting tance, is a suitable measure for electrical characterization of from lateral movement of along cell membranes and γ- time-variant and non-linear materials. If the complex beha- dispersion caused by orientation polarization of dipolar vior of the material is not considered, dynamic impedance molecules. A pronounced dispersion at high frequency (> (admittance) is reduced to dynamic resistance (conduc- GHz) arises from the polarization of water dipoles. tance). The electrical impedance in general is the ratio between Basic features of dynamic impedance measurement are the electric field within a material and the resulting current measurement time, short with respect to impedance changes density. If the behavior between electrodes is considered, it and the use of an excitation signal small enough for working is the voltage divided by the total current. in a quasi linear range around a working point which is determined by an outer electric field. C. Dynamic Impedance and Admittance

An important feature of impedance measurement is the III. MATERIAL AND METHOD use of an electric stimulus which should never change electrical behavior of the material under test. This is in marked Two basic principles of measurement were employed for contrast to electrical manipulation where the material is electrical characterization of different materials due to elec- driven into a non-linear region. In this case, the quotient I/U troporation: (1) impedance measurement before and imme- is an apparent conductance (Gapp). diately after electroporation and the (2) dynamic conduc- A typical non-linear characteristic is found when for in- tance during the presence of the high electric field. stance potato tissue is electrically altered by means of an In order to ensure maximum measurement speed, time outer electric field. The apparent conductance in the point P domain based impedance measurement using step function (Fig.1.) yields Gapp = 4.8 mS which is the slope of the for excitation was used for assessing the impedance spectrum straight line. Practically more relevant information is the between 10 kHz and 10 MHz with sub-millisecond time

IFMBE Proceedings Vol. 53

Electrical Measurements for Monitoring Electroporation 81 resolution. Typically a 100 mV voltage step was applied and interesting for presetting and controlling electroporation the step response (current) was recorded and digitized. The equipment. For more sophisticated investigation, the appli- impedance was calculated using the Fourier transformed cation of multiple ramps with different slope, also as in case excitation and response. For tetrapolar electrode configura- of rectangular functions, with a new specimen for each tion, current excitation (e.g. 100 µA) was used. measurement is favored (Fig.2a). For obtaining the dynamic conductance, either ramp Although important features can be already seen in the functions (current or voltage clamp), sufficient for trigger- raw data, processing of the data can clarify the picture. ing electroporation, or rectangular pulses with extremely Calculating the dynamic conductance yields a sigmoid low ripple (< 1%) were used. While the differential quotient curve where the voltage of maximum slope is considered as dI/dU was calculated for ramp functions, yielding the dy- electroporation or breakdown voltage (Ub). As seen in namic conductance at each voltage level, only the quotient Fig.2b, this voltage increases with increasing slope of the I/U was calculated for rectangular pulses, independent of ramp which is due to the time for pore creation in which the voltage or current clamp. voltage can rise further without considerable changes in Electrodes were made from stainless steel or carbon. Be- conductance. cause high current density disturbs the measurement, separate pairs of electrodes for high voltage application and 0 4 3 5 0 442 kV/s impedance measurements were used. When electrical mea- 0 787 kV/s 3 0 3 4 878 kV/s surements are emphasized, six-electrode configuration with 2 5

two electrodes for electroporation and a fully insulated tetra 0 2 / mS 2 I / A Dy G polar electrode system for impedance measurement was 1 5 0 1 employed. 1

Instrumentation for impedance measurement is sensitive 0 0 5 0 50 100 150 200 250 0 50 100 150 200 250 to high voltage pulses. Therefore, drivers and amplifiers are U / V U / V disconnected during the presence of high voltage either by Fig.2a Current response of sheep liver to voltage ramps of different slope hand, mechanical switching, relays or diode or solid state (see legend) at room temperature. (corrected for displacement current) b: switches. Among all the switches, relays may be slowest dynamic conductance. (several millisecond) but guarantee the best signal quality. Fast mechanical, spring loaded switches used for nanose- Moreover, there is a trend towards reaching higher dy- cond pulses reached a short time from impedance measure- namic conductance with slower ramp, owing to longer time ment through pulse application and back to impedance mea- for pore expansion. surement of 2 ms. Another fast approach, working up to 1 For automatic assessment of important material characte- kV is the supplement of the impedance instrumentation ristics, a biophysical model is favored. A simple thermody- (driver and preamplifier) from high voltage generator which namic model where the entire material is treated as insulat- made switching between pulse application and impedance ing structure which increases conductivity with each measurement unnecessary. increment in field strength is suitable for most application A great variety of materials was tested, for instance ar- but shows considerable shortcomings. Since all geometry of tificial bilayers, single cells in whole cell clamp, cell sus- cells and membranes is neglected, any increment in dynam- pensions (i.e. CHO, Jurkat, M-16, HeLa), tissue (liver, ic conductance cannot be calibrated to a number of pores heart, muscle, potato) and multi-lamellar systems (skin). created. A simple model (eq.1) assumes time invariant I/U- characteristics ∆ . IV. DISCUSSION ON SELECTED RESULTS (1) . A critical material property for electroporation is the I/U Without electric field, conductive structures arising from characteristic. Although it can be tested with high voltage extra cellular pathways exits and are lumped to G0. Any pulses of increasing amplitude, this procedure requires mul- insulating structure (here lipid membranes) may turn con- tiple measurement and, because the material shows pro- ductive with outer field application. The equilibrium con- nounced memristor behavior, a new specimen for each stant between conductive and non-conductive structures is Δ measurement. A much faster approach is the use of a vol- K0. GDy is the maximum increase of the dynamic conduc- tage ramp function with measurement of the current or vice tance (Fig.3) and U0.5 is the voltage where half of the max- versa. Despite of the time variant and complex nature of the imum number of pores is created. The factor Δεε material (capacitive elements), this approach is sufficient b=0.5 0/(kk0T) depends on the difference in permittivity for getting a raw idea about the non-linear behavior as it is Δε between the conductive and non-conductive structures,

IFMBE Proceedings Vol. 53

82 U. Pliquett

geometry of the material (k) and the temperature. k0 is the field strength but also on the pulse duration and the number Boltzman-constant and ε0 the permittivity of free space. of pulses. Below a critical field strength, essentially no effect occurs (Fig.4). The recovery exhibits several phases. Two relaxation mechanisms with about 3 s and 20 s are visible.

V. CONCLUSIONS Electrical measurements form a useful tool for monitoring the yield and time course of electroporation and the recovery hereafter. Since these events are very fast, only methods in time domain with sub-millisecond time resolu- Fig.3 Variables extracted from the GDy/U-characteristic of a CHO-pellet subjected to a 1 MV/s voltage ramp. The distance of the monitor tion are suitable for this approach. For precise measure- electrodes was 1 mm. ments mechanical switches between high voltage application and impedance measurement but also separate electrode An important measure for process control is the electro- systems are favored. poration yield which can be derived from conductivity, σ, before and after high voltage treatment using a disintegration index CDI [8] CONFLICT OF INTEREST The authors declare that they have no conflict of interest. (2)

σ100% is the conductivity after extremely high energy REFERENCES pulses, resulting in 100 % disintegration. 1. Weaver, J. C. (1993) Electroporation: a general phenomenon for mani- Additional information about the course of electropora- pulating cells and tissue, J. Cellular Biochem 51, 426-435. tion and the subsequent recovery mechanism are obtained 2. Glaser, R. W., Leikin, S. L., Chernomordik, L. V., Pastuchenko, V. F., from impedance spectroscopy. Within the ß-dispersion, only and Sokirko, A. I. (1988) Reversible electrical breakdown of lipid lay- the low frequency part will be changed due to increased ers : Formation and evolution of pores, Biochim. Biophys. Acta 940, 275-287. conductivity of cell membranes. At high frequency, only 3. Stacey, M., Stickley, J., Fox, P., Statler, V., Schoenbach, K., Beebe, S. slight changes are expected, because the cell membranes are J., and Buescher, S. (2003) Differential effects in cells exposed to ultra- anyway capacitively shortened. Therefore, in most cases, short, high intensity electric fields: cell survivial, DNA damage, und the dc-resistance extrapolated from the impedance spectrum cell cycle analysis, Mutiation Research 542, 65-75. 4. Ivorra, A. and Rubinsky, B. (2007) in vivo electrical impedance mea- is most informative. surements during and after electroporation of rat liver, Bioelectroche- mistry 70, 287-295. 5. Neumann, E., Sowers, A., and Jordan, C. (1989) Electroporation and Electrofusion in Cell Biology Plenum Press, New York. 6. Pliquett, U., Joshi, R. P., Sridhara, V., and Schoenbach, K. H. (2007) high electrical field effects on cell membranes, Bioelectrochemistry 70, 275-282. 7. Grimnes, S. and Martinsen, O. G. (2015) Bioimpedance and Bioelec- tricity Basics, 3rd edition, Academic Press. 8. Lebovka, N. I., Bazhal, M. I., and Vorobiev, E. (2002) Estimation of characteristic damage time of food materials in pulsed-electric fields, Journal of Food Engineering 54, 337-346.

Fig.4 dc-resistance of sheep liver normalized to pretreatment values during and after a course of 8 rectangular 100µs-pulses (gray area), separated by 1s. Author: Uwe Pliquett Institute: Institut für Bioprozess- und Analysenmesstechnik Street: Rosenhof The field strength E=(153, 307, 461, 615) V/cm, from City: Heilbad Heiligenstadt top to bottom) was measured using an extra pair of elec- Country: Germany trodes. The electroporation yield depends on the electric Email: [email protected]

IFMBE Proceedings Vol. 53

Comparison of Single-Shot Rapid Acquisition with Relaxation Enhancement and Echo Planar Current Density MRI Sequences for Monitoring of Electric Pulse Delivery in Irreversible Electroporation

I. Serša1, F. Bajd1, M. Kranjc2, H. Busse3, N. Garnov3, R. Trampel4, and D. Miklavčič2 1 Institut “Jozef Stefan”, Jamova cesta 39, SI 1000, Ljubljana, Slovenia 2 University of Ljubljana, Faculty of Electrical Engineering, Tržaška 25, SI 1000, Ljubljana, Slovenia 3 Department of Diagnostic and Interventional Radiology, University of Leipzig, Liebigstrasse 20, 04103 Leipzig, Germany 4 Max Planck Institute for Human Cognitive and Brain Sciences, Stephanstr. 1a, 04103 Leipzig, Germany

Abstract Success of electroporation treatment critically method for reconstruction of electric field distribution during depends on coverage of the target tissue with electric field. The electroporation pulse delivery based on magnetic resonance electric field during delivery of the electroporation pulses can electrical impedance tomography (MREIT) was suggested be reconstructed by the magnetic resonance electric impedance [10]. So far, MREIT was primarily used in low voltage/ low tomography, a method that uses current density distribution current regimes as a method for electrical conductivity map- data and electric potentials at the electrodes for reconstruction of electric field in the sample. In this study, two complementary ping [11], while its application in electroporation is relatively MRI methods for current density imaging during delivery of new and was used only in a few recent studies. These include irreversible electroporation pulses are presented. One of the agar phantom experiments [10], ex vivo liver tissue experi- methods is based on the single-shot rapid acquisition with re- ments [12] and in vivo muse tumor experiments [13]. laxation enhancement while the other is based on the echo An essential step in MREIT is an acquisition of current planar imaging MRI method. The methods were compared in density (CD) distributions during the delivery of electropora- terms of their sensitivity and susceptibility to image artifacts by tion pulses. These can be obtained by current density imaging experiments on a liver test sample that were performed on a 2.35 T small bore MRI scanner. In the experiments, a standard (CDI) techniques [14], which rely on mapping of the magnet- protocol for irreversible electroporation where 90 electric ic field change due to application of the electroporation cur- pulses of 100 µs, 3000 V are delivered at 1 Hz was performed. rents and conversion of the maps to CD images using Am- Results of the study confirmed that both methods have compa- pere’s law. In principle, for calculation of the one component rable sensitivity. The RARE-based method was found less sus- of the current density, information on the two components of ceptible to artifacts while the EPI-based method has lower SAR the magnetic field change is needed. This would impose a value and may therefore be a better candidate for use in clinics. rotation of the sample to at least two perpendicular orienta- Keywords MRI, current density imaging, electroporation, tions, as the magnetic field change can be recorded only electric field monitoring. along the direction of the static magnetic field. However, with special orientations of the electrodes with respect to the static I. INTRODUCTION magnetic field, the sample rotation may be avoided and CD maps can be calculated just from a magnetic field change Use of electroporation as a clinical method for tumor map acquired in one sample orientation [10, 13]. treatment is expanding. Two varieties of it are used: elec- Standard CDI sequence [14, 15] is designed for long low trochemo therapy (ECT) and irreversible electroporation voltage electric pulses that are used in a combination with a (IRE). In ECT [1-4] the used electric field is sufficiently high standard spin-echo imaging sequence. The sequence is in- that for a short time makes cell membranes permeable and appropriate for its use in electroporation as it requires deli- thus eases transport of anticancer drugs into the cells, while in very of typically more than 100 electric pulses for acquisi- IRE [5, 6] the exposure to electric field is higher (number of tion of one CD image. As the pulses have high voltage, they pulses and amplitude) so that the electric pulses itself without may significantly alter the sample during image data acqui- the aid of the anticancer drugs results in a cell death in the sition. Therefore, electric pulses can be repeated only very electroprated tumor region. For the success of either of the limited times for acquisition of one image, ideally applica- treatment methods, it is very important that the electric field tion of only one electroporation pulse should be sufficient in the treated region is within the specified range or the for acquisition of one CD image. Possible candidates as treatment efficacy is poor [7-9]. Therefore, the use on an imaging methods for electroporation monitoring are there- efficient method for the electric field monitoring during the fore from the family of single-shot MR imaging methods. delivery of electric pulses is highly demanded. Recently, a Specifically, in this study two such methods were employed

© Springer Science+Business Media Singapore 2016 83 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_19

Comparison of Single Shot Rapid Acquisition with Relaxation Enhancement and Echo Planar Current Density MRI Sequences 85

Both CDI sequences for IRE monitoring were performed image sets (Fig. 3, top row vs. Fig. 3 bottom row) clearly on a 2.35 T horizontal bore small animal MRI scanner. The shows tissue alteration as a result of its exposure to high scanner was based on an Oxford superconducting magnet voltage electric pulses, i.e. electroporation. Moreover, con- (Oxford Instruments, Abingdon, UK), an Apollo NMR/MRI ductivity of the tissue increased which resulted in a higher spectrometer (Tecmag Inc., Houston TX, USA) and MRI current density between the electrodes. probes for MR microscopy (Bruker, Ettlingen, Germany). CDI results shown in Fig. 4 confirmed that a CD distri- The sequences were run with identical geometric and reso- bution can be imaged also by the EPI CDI sequence. Figure lution parameters: field of view 30 mm, imaging matrix 64 4 shows the magnitude CD image of an identical tissue by 64, slice thickness 4 mm and repetition time 1 s, while sample to the previous experiment. As expected, current echo-time parameters were different: inter-echo time in the density was the highest next to the electrodes and then de- two-shot RARE CDI sequence was 2.64 ms and spin-echo creased as distance from the electrodes increased. However, time in the EPI CDI sequence was 35 ms. To maximize CDI trajectory of the highest current was not a straight line con- sensitivity, signal acquisition ordering of lines in the k- necting the electrodes, but was shifted sideways. In the space was centric for the RARE-type of sequence and was experiment 90 electroporation pulses were delivered and the sequential for the EPI-type of sequence. same number of CD images was obtained. The one shown corresponds to CD distribution at the end of the experiment.

III. RESULTS Figure 3 depicts results of the CDI experiment on the liver sample from data acquired by the two-shot RARE CDI sequence. By the sequence vector maps of current distribution as well as the corresponding maps of current density magnitude were obtained. In the experiment 90 electroporation pulses were delivered and 45 CD images were calculated as delivery of two electroporation pulses was needed for calculation of one CD image. Fig. 4 Current density distribution image of a liver sample calculated from data obtained by the EPI-based CDI sequence.

IV. DISCUSSION Results of CDI on the liver test sample confirmed that both tested CDI sequences, the two-shot RARE CDI sequence as well as the EPI CDI sequence, enable clear visualization of current distribution in the electroporated tissue region and are therefore appropriate for tissue electroporation monitoring. However, there are distinctions between both CDI sequences used here. As already pointed in the materials section, the two- shot RARE CDI sequence requires two repetitions of the sequence for acquisition of one image, while the EPI CDI sequence does not have such a limitation and enables image acquisition already after only one sequence run, i.e. after a single electroporation pulse. Therefore, it is a true single-shot sequence. Unfortunately, the EPI CDI sequence has also its drawback. As all EPI sequences are prone to susceptibility (magnetic field inhomogeneity) artefacts, the EPI CDI sequence is no exception. Therefore, it is not very clear if the Fig. 3 Calculated current distribution vector field maps (left) and the unusual current distribution in Fig. 4 is true or it is an EPI corresponding current density magnitude images (right) at the beginning of st artifact. If it is true then it could be attributed to the void in the electroporation experiment (1 row) and at the end of it when 90 high- voltage electric pulses were already delivered to the liver sample (2nd row). middle of the sample, perhaps due to liver tissue folding. The CDI data were acquired by the two-shot RARE CDI sequence. two-shot RARE CDI sequence is more robust in this respect and can produce nice artifact free images also with samples of In Fig. 3, the CD images corresponding to the first two poor magnetic field homogeneity. The EPI CDI sequence has electroporation pulses and to the last two (89th and 90th) also an important advantage over to the two-shot RARE CDI electroporation pulses are shown. Comparison of these two sequence and that is low specific absorption rate (SAR). SAR

IFMBE Proceedings Vol. 53

86 I. Serša et al. is a measure of the rate at which energy is absorbed by the 3. Marty M, Sersa G, Garbay JR, Gehl J, Collins CG, Snoj M et al. human body when exposed to a radio frequency (RF) electro- Electrochemotherapy - An easy, highly effective and safe treatment of cutaneous and subcutaneous metastases: Results of ESOPE (Euro- magnetic radiation. Namely, in the two-shot RARE CDI se- pean Standard Operating Procedures of Electrochemotherapy) study. quence there are many RF refocusing pulses that may contri- Ejc Suppl. 2006;4(11):3-13. bute to a significant tissue RF irradiation. As there are strict 4. Yarmush ML, Golberg A, Sersa G, Kotnik T, Miklavcic D. Electro- regulations regarding SAR in clinical MRI, the performance poration-based technologies for medicine: principles, applications, of two-shot RARE CDI sequence may be impeded when used and challenges. Annual review of biomedical engineering. 2014;16:295-320. in human MRI. For comparison, in the presented study, for 5. Lee EW, Chen C, Prieto VE, Dry SM, Loh CT, Kee ST. Advanced acquisition of one CD image 130 RF refocusing pulses were hepatic ablation technique for creating complete cell death: irreversi- delivered to the sample with the two-shot RARE CDI se- ble electroporation. Radiology. 2010;255(2):426-33. quence, while only one RF refocusing pulse was delivered to 6. Neal RE, 2nd, Rossmeisl JH, Jr., Garcia PA, Lanz OI, Henao- Guerrero N, Davalos RV. Successful treatment of a large soft tissue the sample with the EPI CDI sequence. sarcoma with irreversible electroporation. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. V. ONCLUSION 2011;29(13):e372-7. C 7. Miklavcic D, Corovic S, Pucihar G, Pavselj N. Importance of tumour coverage by sufficiently high local electric field for effective electro- In our study a comparison between two two-shot RARE chemotherapy. Ejc Suppl. 2006;4(11):45-51. CDI sequence and the EPI CDI sequence is presented. Re- 8. Miklavcic D, Beravs K, Semrov D, Cemazar M, Demsar F, Sersa G. sults on the liver tissue test sample ex vivo confirmed that The importance of electric field distribution for effective in vivo elec- both sequences could be used for monitoring of irreversible troporation of tissues. Biophysical journal. 1998;74(5):2152-8. 9. Miklavcic D, Snoj M, Zupanic A, Kos B, Cemazar M, Kropivnik M tissue electroporation, but with different limitations. The et al. Towards treatment planning and treatment of deep-seated solid two-shot RARE CDI sequence was found very robust and tumors by electrochemotherapy. Biomedical engineering online. can produce good results also with samples in conditions of 2010;9:10. poor magnetic field homogeneity, while the EPI CDI se- 10. Kranjc M, Bajd F, Sersa I, Miklavcic D. Magnetic resonance electrical impedance tomography for monitoring electric field distribution quence is prone to susceptibility artifacts. However, the EPI during tissue electroporation. IEEE transactions on medical imaging. CDI sequence has considerably lower SAR than the two- 2011;30(10):1771-8. shot RARE CDI sequence and may therefore be more ap- 11. Seo JK, Woo EJ. Electrical tissue property imaging at low frequency propriate for electroporation monitoring in clinical MRI. using MREIT. IEEE Trans Biomed Eng. 2014;61(5):1390-9. 12. Kranjc M, Bajd F, Sersa I, Miklavcic D. Magnetic resonance electrical impedance tomography for measuring electrical conductivity dur- ACKNOWLEDGMENT ing electroporation. Physiological measurement. 2014;35(6):985-96. 13. Kranjc M, Markelc B, Bajd F, Cemazar M, Sersa I, Blagus T et al. In This study was supported by the Slovenian Research Situ Monitoring of Electric Field Distribution in Mouse Tumor dur- Agency (ARRS) and conducted within the scope of the ing Electroporation. Radiology. 2015;274(1):115-23. 14. Joy M, Scott G, Henkelman M. In vivo detection of applied electric Electroporation in Biology and Medicine European Asso- currents by magnetic resonance imaging. Magn Reson Imaging. ciated Laboratory (LEA-EBAM). This manuscript is a result 1989;7(1):89-94. of the networking efforts of the COST Action TD1104 15. Scott GC, Joy MLG, Armstrong RL, Henkelman RM. Sensitivity of (www.electroporation.net). Magnetic-Resonance Current-Density Imaging. J Magn Reson. 1992;97(2):235-54. 16. Hennig J, Nauerth A, Friedburg H. RARE imaging: a fast imaging CONFLICT OF INTEREST method for clinical MR. Magn Reson Med. 1986;3(6):823-33. 17. Mansfield P. Multi-Planar Image-Formation Using Nmr Spin Echoes. The authors declare that they have no conflict of interest. J Phys C Solid State. 1977;10(3):L55-L8. 18. Sersa I. Auxiliary phase encoding in multi spin-echo sequences: application to rapid current density imaging. J Magn Reson. 2008;190(1):86-94. REFERENCES

1. Miklavcic D, Mali B, Kos B, Heller R, Sersa G. Electrochemothera- Author: Igor Serša py: from the drawing board into medical practice. Biomedical engi- Institute: Jozef Stefan Institute neering online. 2014;13(1):29. Street: Jamova 39 2. Edhemovic I, Brecelj E, Gasljevic G, Marolt Music M, Gorjup V, City: Ljubljana Mali B et al. Intraoperative electrochemotherapy of colorectal liver Country: Slovenia metastases. J Surg Oncol. 2014;110(3):320-7. Email: [email protected]

IFMBE Proceedings Vol. 53

Micro Electro-permeabilization System for Cell Medium Conductivity Change Measurement of Erythrocytes Cells

L.C. Ramos1, G.B. Pintarelli1, D. Altenhofen2, and D.O.H. Suzuki1 1 Federal University of Santa Catarina, Department of Electrical Engineering, Florianópolis, Brazil 2 Federal University of Santa Catarina, Department of Biochemistry, Florianópolis, Brazil

Abstract The mechanism of electroporation is used in a The influence of neighboring cells was considered negligible. wide range of medical applications, genetic engineering, and The effective electric field was calculated by [6] in Equation (2). therapies. Measurements of biological cells suspension conductivity have been used to evaluate the effect of electroporation cells. A micro electro-permeabilization system was built to apply 900 µs pulses with electric field variations in a cells suspension. The electrodes have distance around 200 microme- ters. We observed changing in the cells suspension conductivity as the increased electric field with the micro electropermeabilization system. During electroporation we conclude that different volume fractions leads different conductivities of cells suspension in same electric field, in low-conductive medium increasing the electric field induce a higher conductivity gradient of cells suspension in lower concentration of cells, and for same cells volume fractions and same electric field results different conductivity of cells suspensions changing external medium conductivity.

Keywords electroporation, membrane conductivity, cell suspension, erythrocytes of mice. Fig. 1 Spherical cell radius a in a uniform electric field E0. θ is the angle between the direction of the field and the position vector on the membrane. I. INTRODUCTION

The electroporation is the appearance of pores in cell . 1 0.38 2 membranes due to a high transmembrane potential (200 where p is the cells volume fraction and is the applied mV - 1 V) caused by applying an external electric field [1]. electric field. The phenomenon may be reversible or irreversible, accord- The conductivity of the solution used in the suspension ing to the characteristics of the applied field [2]. Reversible facilitates the movement of ions and consequently the pas- open pores allows the transfer of ions across the membrane sage of electric current. Low-conductive medium typically and water-soluble molecules, such as a number of drugs, DNA, antibodies and plasmids [3]. is used in in vitro conditions and high-conductive medium The electroporation of biological cells in suspension that is more representative for in vivo conditions [7][8]. (bulk electroporation) is a technique that enables a large Multiples devices recently developed to electroporation amount electroporation cells with a single pulse. Vm is the cell suspension use MEMS technology - Micro Electro- transmembrane voltage induced in a cell isolated by a uni- Mechanical System. The advantages of minimizing dimen- form and effective electric field E0, Fig. 1, given by Equa- sions is to achieve high intensity electric fields due to the tion 1. Its value is maximum at the pole of the cell, where θ small distance between electrodes and to have a good dis- = 0, called V0 [4]: persion of heat from the Joule-Effect [9]. The objective of this research was to develop a micro 1.5... cos (1) electro-permeabilization system with reduced distance be- where a is the cell radius. tween the electrodes (<1 mm) to observe the effect compar- The proximity of the cells produces a reduction of ative of intense electric fields in different volume fraction of transmembrane potential and suspension conductivity [5]. cell suspensions and the external medium conductivity.

© Springer Science+Business Media Singapore 2016 87 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_20 88 L.C. Ramos et al.

II. MATERIAL AND METHODS average resistance of the electroporation system. The conductivity of the cell suspensions was calculated by the relationship A. Cell Preparation with the reverse of system resistance and with a proportionali- The used erythrocytes cells were obtained from male al- ty constant relative to the dimensions of the electrodes. bino Wistar rat’s blood (160–190 g) after centrifugation by removing the buffy coat. The conductivity of cell suspensions was measured on the same day of blood collection after washing twice with isotonic and atoxic phosphate- buffered saline solution (PBS) composed of 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4 and pH 7.4. The resulting pellet was resuspended in two volume fractions of cells; p = 0.07 (0.51 × 106 cells/ml) and p = 0.73 (5.3 × 106 cells/ml). Two medium was used to resuspend the cells, each with a different conductivity; a low-conductive medium σ1 = 0.3 S/m and a high-conductive medium σ2 = 1.4 S/m. All the animals were carefully monitored and maintained in accordance with ethical recommendations of the Brazilian Veterinary-Medicine Council and the Brazilian College of Animal Experimentation (Protocol CEUA/PP00398). B. Electroporation System and Driving Circuit An electroporation micro system was built with distance between electrodes of 230±25µm, Fig 2.

Fig. 3 Diagram comprising the driving circuit, electroporation system and reading resistance.

Table 1 Electric field applied and effective

Item p 0.07 p 0.73 Equation E 0.97. E E 0.73. E 0 a 0 a Ea min 22 kV/m 22 kV/m Fig. 2 View of electroporation system detail with suspension cells region highlighted. Ea max 348 kV/m 348 kV/m E0 min 20 kV/m 16 kV/m

This electroporation system was coupled to a reading E0 max 338 kV/m 254 kV/m resistance with value known (Rreading, 10.1Ω measured with FLUKE 179 multimeter) and a hardware driving device, Fig. 3. III. RESULTS A set of square pulses of 900 µs was applied to electro- An example of the applied voltage and the current meas- poration system to generate electric fields effective between ured to the system as seen in Fig 4. 22 and 348 kV/m. Considering the Equation 2, it was The measured conductivity of cells suspensions in S/m calculated the values of electric field, Table 1. The trans- unit as function of the medium conductivity and volumetric membrane potential Vm is also calculate for electric field effective and 0. function is showed in Fig 5 (a) for p = 0.07 and and (b) for The reading of the voltages V1 and V2 was performed with p = 0.73. The initial conductivity was obtained in Ea min = an oscilloscope Tektronix TDS2004C. For each electric field 22 kV/m and the final conductivity in Ea max = 348 kV/m. generated value, there was the average of three values of V1 The values of the transmembrane potential V0 obtained and V2 potencial. It was calculated the current flowing from Equations 1 and 2 are between 0.10 V and 1.5 V, through the resistance reading and applied a low-pass filter 5th Table 2. This range is estimated to be close to electropora- order with fc = 60kHz. Therefore it was calculated the tion [10].

IFMBE Proceedings Vol. 53

Micro Electro permeabilization System for Cell Medium Conductivity Change Measurement of Erythrocytes Cells 89

(a) (b)

Fig. 4 A sample of square pulse of 900 µs and 60V of voltage to the electroporation system (a) and the current measured correspondent (b). Theses curves are obtained at p 0,07 and external media conductivity σ1 0.3 S/m.

(a) (b)

Fig. 5 Relationship between the conductivity [S/m] of the cell suspension in y-axis between the low-conductivity medium (σ1 0.3 S/m) and high- conductivity medium (σ2 1.4 S/m) for applied electric field (Ea min 22 kV/m, Ea max 348 kV/m) for cells volume fractions different, (a) p 0.07 and (b) p 0.73. Data analyzed at 600 µs.

The electrical field Ea min induced transmembrana po- Table 2 Transmembrane potential to the system tential V0 less than 0.2 V, as seen in the Table 2. There were no pores on cells membrane. When applying the electrical Item p 0.07 p 0.73 field E max, the induced transmembrana potential was V min 0.1 V 0.08 V a m more than 1 V. In this case, it is ensured that electroporation Vm max 1.5 V 1.2 V threshold has been reached. The transmembrane voltage values depend only on the volume fraction of the cells and IV. DISCUSSION the electric field strength. There were evidences that cells suspension is diluted in solution which presents characteris- The purpose of this article is to demonstrate that the mi- tics of high conductivity (1.4 S/m), the generated electric cro electro-permeabilization system developed was able to fields provide a higher conductivity of the cell membrane produce and detect changes in the conductivity of cell sus- [3][11]. pension due to electroporation. The time of the pulses and The conductivity effects in relation to pores opening and potentials used were defined to generate electric field able diffusion of molecules into the cells remain unclear. Li et al to inducing electroporation [4]. Beyond time and pulse [12] presents evidences that despite the increase of the ex- strain, the electroporation system is sensitive to the change ternal media conductivity be proportional to the conductivi- in concentration of the suspensions and the change in the ty membrane, propidium iodide entry in the cell was lower. type of conductive medium. About Fig. 5, we conclude that the results of conductivity Given the dimensions of 230 µm between the electrodes of cells suspensions were different in the two cells volume of the electroporation system, the suspension volume neces- fractions (p = 0.07, p = 0.73) in different external medium sary to the measurement is less than 100 µl. The amount of (σ1, σ2) for same incident electric field. In low-conductive cells present among the electrodes is enough to measure the medium (σ1), increasing the electric field induce a higher change in current. conductivity gradient of cells suspension in lower concen-

IFMBE Proceedings Vol. 53

90 L.C. Ramos et al. tration of cells (p = 0.07). More concentration of cells CONFLICT OF INTEREST represents more barriers to the flow of current and lower incidence of electric field. The electric field induces a The authors declare that they have no conflict of interest. weaker variation of the conductivity of the suspension, which is not made clear in the electroporation threshold. In the cells suspension with high-conductivity medium (p = 0.07, REFERENCES p = 0.73) the barriers due to the concentration of cells is not 1. Chen C, Smye SW, Robinson MP, Evans JA (2006) Membrane elec- as effective. It was more certain observing the threshold of troporation theories: a review. Med. Biol. Eng. Comput.; v. 44, p. 5-14. the electroporation. This shows that the system presented in 2. Weaver JC, Chizmadzhev YA. (1996) Theory of electroporation: A this work had better results than Suzuki et al [3] by allowing review. Bioelectrochemistry; v. 41, p. 135-160. the use of higher external conductivities medium. 3. Suzuki DOH, Ramos A, Ribeiro MCM, Cazarolli LH, Silva FRMB, Due to be a square wave drive, several frequencies are Leite LD, Marques JLB (2011) Theoretical and experimental ana- lisys of electroporated membrane conductance on cell suspension. present in the system. A study of the impedance of the elec- IEEE Trans. Biomed. Eng.; v. 58, p. 3310-3318. troporation system must be made to compensate for their 4. Krassowska W, Filev PD (2007) Modeling electroporation in a sin- effects. This system aims to study the effects of conductivi- gle cell. Biophys. J; v. 92, p. 404-417. ty and molecules from entering the cell. And through ma- 5. Schmeer M, Seipp T, Kakorin T, Nneumann E (2004) Mechanism for the conductivity changes caused by membrane electroporation of thematical models explain the inverse of the mechanisms of CHO cell-pallets. Phys. Chem. Chem. Phys.; v. 6, p. 5564 5574. conductivity and transport molecules. 6. Ramos A, Suzuki DOH, Marques JLB (2006) Numerical study of the electrical conductivity and polarization in a suspension of spherical cells. Bioelectrochemistry, v. 68, p. 213 217. V. CONCLUSIONS 7. Pavlin M, Slivnik T, Miklavcic D (2002) Effective conductivity of cell suspensions. IEEE Trans. Biomed. Eng. V. 49, p. 77 80 The micro electro-permeabilization system demonstrated 8. Kinosita KJr, Tsong TY (1979). Voltage-induced conductance of hu- a conductivity variation due to the cell suspension electro- man erythrocyte membranes. Biochim. Biophys. Acta. 554:1015-1019. poration. The reduced distance between electrodes (around 9. He H, Chang DC, Lee, YK (2006) Micro pulsed radio-frequency electroporation chips. Bioelectrochemistry. Netherlands. v. 68(1), 200 µm) allowed intense electric fields with small electric p. 89 97. potential. The use of the system allowed conclude that, 10. Teissie J, Golzio M, Rols MP (2005) Mechanisms of cell membrane purposing the electroporation, it is possible to improve the electropermeabilization: a minireview of our present (lack of?) know- response of the cells suspension conductivity by increasing ledge. Biochimica et Biophysica Acta (BBA)-General Subjects. v. the strength of the electric field, increasing the external 1724(3), p. 270-280. 11. Ivorra A, Villemejane J, Mir LM (2010) Electrical modeling of the medium conductivity or reducing the concentration of cells influence of medium conductivity on electroporation. Phys. Chem. volume fractions. Chem. Phys., 12, pp. 10055 10064. In future work, we intend to define better the read limita- 12. Li J, Tan W, Yu M, Lin, H (2013) The effect of extracellular conduc- tions of electroporation system. The aim is to develop a tivity on electroporation-mediated molecular delivery. Biochimica et Biophysica Acta (BBA)-Biomembranes, v. 1828(2), p. 461-470. study of electroporation system in cell suspension for understanding the phenomenon. Author: Luciana Costa Ramos Institute: IEB UFSC CKNOWLEDGMENT Street: Rua das Azaleias, 1468 A City: São José - SC Country: Brazil Thanks CAPES for financial support. Email: [email protected]

IFMBE Proceedings Vol. 53

Magnetic Resonance Electrical Impedance Tomography for Monitoring Electrical Conductivity during Delivery of Electric Pulses in Irreversible Electroporation

M. Kranjc1, I. Serša2, and D. Miklavčič1 1 University of Ljubljana, Faculty of Electrical Engineering, Tržaška 25, SI 1000, Ljubljana, Slovenia 2 Institut “Jozef Stefan”, Jamova cesta 39, SI 1000, Ljubljana, Slovenia

Abstract Monitoring of electroporation process presents the imaged sample using current density imaging (CDI) [14, one of the most important aspects towards safe and reliable 15]. When current density distribution and electrical con- use of electroporation applications such as irreversible electro- ductivity are obtained, electric field distribution can be poration (IRE) ablation and electrochemotherapy (ECT). calculated using Ohm’s law. Since electroporation process is related to an induced trans- We demonstrated a successful reconstruction of the elec- membrane potential, which is directly proportional to the electric field, we proposed a method to determine electric field tric field distribution during electroporation in an agar phan- distribution by means of magnetic resonance electrical imped- tom [11], ex vivo tissue [16, 17], in silico [17] and in mouse ance tomography (MREIT) and current density imaging tumor in vivo [18]. Each of these objects were subject to (CDI). In our previous studies, CDI was adjusted to acquire electroporation by an application of electric pulses of differ- current density distribution established by electric pulses with ent amplitudes and durations, while the repetition rate was repetition rate in the range of kHz. Therefore we developed a the only parameter that remained at a constant value of 5 new CDI sequence for lower repetition rates, i.e. in the range kHz in all experiments. The repetition rate of 5 kHz is of 1 Hz. In this study we evaluated feasibility of new CDI se- commonly used in electroporation applications such as ECT quence to be applied together with MREIT for electroporation [9], whereas IRE ablation is usually applied at the repetition applications such as IRE ablation. We performed experimental evaluation on ex vivo liver tissue during application of electric rate of around 1 Hz. The CDI sequence used in our previous pulses with the repetition rate of 1 Hz. Results of experiment experiments was adjusted to acquire current density distri- demonstrated successful reconstruction of electrical conductiv- bution established by electric pulses with a high repetition ity and detection of conductivity increase in the tissue. rate, i.e. in the range of kHz, and is not applicable for acquiring current density distribution of electric pulses with a Keywords magnetic resonance electrical impedance tomo- lower repetition rate. Recently, we developed a new CDI graphy, electrical conductivity, current density imaging, elec- sequence for lower repetition rates [19]. The aim of our troporation. study was to evaluate feasibility of the new CDI sequence for its application in MREIT in typical electroporation ap- I. INTRODUCTION plications, as for example IRE ablation. Electroporation is a biological phenomenon which allows normally nonpermanent molecules to cross the cell mem- II. MATERIAL AND METHODS brane and enter the cell [1–3]. Cell membrane becomes permeable when it is exposed to an adequate electric field A. Experimental setup [4, 5], i.e. by applying electric pulses with appropriate pa- We evaluated the proposed method of measuring elec- rameters via electrodes. Monitoring of electroporation trical conductivity during delivery of IRE electric pulses on process presents one of the most important aspects towards ex vivo beef liver tissue, which was primarily intended for safe and reliable use of electroporation applications such as human consumption. We obtained liver tissue from a irreversible electroporation (IRE) ablation [6–8] and elec- slaughterhouse, which operates in accordance to Slovenian trochemotherapy (ECT) [9, 10]. law. The process of slaughtering is regulated by Rules on Since an electroporation process is related to an induced animal protection and welfare at slaughter (Ur. l. RS, N. transmembrane potential, which is directly proportional to 5/2006) which ensures ethical standards of slaughtering the electric field, we proposed a method to determine elec- procedure and is in compliance with European Union Coun- tric field distribution by means of magnetic resonance elec- cil directive on the protection of animals at the time of trical impedance tomography (MREIT) [11]. MREIT is an slaughter or killing (93/119/EC). Temperature of the liver imaging modality primarily used for reconstruction of elec- tissue was maintained at 4°C before the beginning of expe- trical conductivity [12, 13]. The method is based on pre- riment when they were allowed to heat up to the room tem- viously acquired current density distribution established in perature. We sectioned tissues in cylindrical and flat shaped

© Springer Science+Business Media Singapore 2016 91 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_21 92 M. Kranjc, I. Serša, and D. Miklavčič samples with a diameter of 21 mm and height of 10 mm and CDI sequence for monitoring IRE was performed on a placed them in an acrylic glass container. In the tissue we 2.35 T horizontal bore small animal MRI scanner. The inserted two self-built cylindrically shaped platinum-iridium scanner was based on an Oxford superconducting magnet electrodes with a diameter of 1 mm. The distance between (Oxford Instruments, Abingdon, UK), an Apollo NMR/MRI the electrodes was 14 mm. Nine sequences of 10 high- spectrometer (Tecmag Inc., Houston TX, USA) and MRI voltage electric pulses, i.e. altogether 90 pulses, with an probes for MR microscopy (Bruker, Ettlingen, Germany). amplitude of 3000 V and a duration of 100 µs at a repetition Parameters of the sequence were the following: field of rate of 1 Hz were used. Pause between sequences was 10 view 30 mm by 30 mm, imaging matrix 64 by 64, slice seconds due to the recharging of capacitors in the pulse thickness 4 mm, inter-echo time 2.64 ms, and repetition generator. Electric pulses were delivered using customized time 1 s. Centric sampling ordering of k-space lines was Cliniporator Vitae (IGEA, Carpi, Italy) pulse generator. used to maximize CDI sensitivity. We applied acquired B. Current Density Imaging current density distributions for reconstruction of electrical Current density imaging (CDI) was used for acquiring conductivity of ex vivo tissue by means of MREIT and for current density distribution inside conductive samples dur- reconstruction of electric current established during applica- ing the application of electric pulses. The method is based tion of electric pulses using surface integration over current on detecting magnetic field changes caused by applied cur- density distribution. rent. When magnetic field changes are obtained by means of C. Magnetic Resonance Electrical Impedance Tomography MRI, current density distribution can be calculated using Ampere’s law [20]. Magnetic field changes are proportional Electrical conductivity of ex vivo liver tissue during ap- to the frequency shift so they can be acquired by chemical plication of electric pulses was obtained by MREIT using J- shift imaging methods [21]. Usually electric pulses are substitution algorithm. The algorithm is based on solving applied synchronously with the imaging sequence and the Laplace’s equation inside a mathematical model of the mea- magnetic field change is proportional to the phase shift [14]. surement object with the corresponding Neumann and Di- Delivery of electroporation pulses for IRE was monitored richlet boundary conditions on the measurement object by the CDI sequence (Figure 1) based on the single-shot outer boundary and on the electrode boundary, respectively. RARE image acquisition scheme [22]. The sequence has Laplace’s equation were solved iteratively using the finite been described in detail in [19]. element method with the numerical computational environment MATLAB 2015a (MathWorks, Natick, MA) on a desktop PC (Windows 8, 3.5 GHz, 32 GB RAM). The ma- thematic process of MREIT has been described in detail previously [11, 16].

III. RESULTS When the tissue was exposed to 9 sequences of 10 electric pulses, the electric current density was established inside the tissue. Current density distribution was acquired using the CDI sequence after each two successive electric pulses. The current density distribution was then applied to the MREIT J-substitution algorithm for reconstruction of electrical conductivity. Five current density distributions and five electrical conductivity images were obtained for each sequence, i.e. each 10 pulses delivered. Both results,

electrical conductivity and reconstructed electric current for Fig. 1 Two-shot RARE pulse sequence that was used for monitoring IRE. each sequence, are shown on Figure 2. The sequence consists of a current encoding part with a short high-voltage In Figure 3 images of electrical conductivities during the electroporation pulse delivered immediately after the nonselective 90° RF excitation pulse. In the second part of the sequence signal acquisition is first and last sequence are presented. Electrical conductivity performed using the single-shot RARE signal acquisition scheme. Due to in the area between the electrodes has increased for up to auxiliary phase encoding induced by the electric pulse, the RARE sequence 50 % during the last pulse sequence compared to the first is repeated twice, each time with a different phase of the refocusing pulses (0°and 90°), and the corresponding signals are co-added. one.

IFMBE Proceedings Vol. 53

Magnetic Resonance Electrical Impedance Tomography for Monitoring Electrical Conductivity 93

increasing electrical conductivity, reconstructed electric current obtained by means of CDI also increased with every sequence. Observed tissue changes are either a consequence of tissue exposure to electric field, which resulted in tissue electroporation, or a consequence of heating due to Joule heating of the tissue. Similar increase of conductivity was also detected in our previous studies employing electric pulses with a higher repetition rate [16, 17]. Areas of increased electrical conductivity in the tissue can be observed on Figure 3. Electrical conductivity increased predominate- ly in the region between the electrodes where electric field was higher compared to regions further away. Still, two inconsistencies regarding increased electrical conductivity can be observed. First, no significant increase of electrical conductivity was detected right in front of the electrodes even though this was the area where the electric field was Fig. 2 Determined electrical conductivity of tissue ex vivo by means of the highest. Second, large conductivity values were present MREIT (marked with blue circles) and reconstructed electric current by means of CDI (marked with red x). Tissue was exposed to 9 sequences of at regions where we did not expect conductivity alterations, 10 high voltage electric pulses with an amplitude of 3000 V and a duration i.e. near the boundary of the tissue. We believe that low of 100 µs at a repetition rate of 1 Hz. Both results, electrical conductivities conductivity in front of the electrodes is a consequence of and electric currents, are presented for each of pulse sequences and were large distortions in the magnetic field due to the conductive obtained by averaging 5 images of electrical conductivities and electric currents of each sequence, respectively. nature of electrodes. These artefacts were consistently present also in our previous studies [11, 16, 17]. We believe local alterations of conductivity near the boundary are measurements error due to the noise and therefore decreased CDI sensitivity in the region. Reconstructed electrical conductivity by means of the MREIT algorithm depends on acquired current density distribution. In principle, reconstruction does not depend on the repetition rate of electric pulses; same algorithms can be applied for high or low repetitions rates. Still, repetition rate of reconstructions are limited to computational time needed for calculations. Numerical models used for reconstruction of electrical conductivity are usually simple and therefore can be quickly solved using finite element method. In this Fig. 3 Electrical conductivities obtained during first (a) and last pulse sequence (b) by means of MREIT. Results were obtained by averaging 5 study, the computer needed 150 ms for each reconstruction, electrical conductivities obtained during first and last pulse sequence. Pulses i.e. well below 1 second needed for sufficient monitoring of were delivered between two needle electrodes (marked with + and −). IRE ablation. With the dedicated numerical platform the computational time could be even further improved. IV. DISCUSSION In this study we evaluated feasibility of a new CDI se- V. CONCLUSION quence in MREIT. The new sequence is designed specially for application in IRE where electric pulses have high vol- We evaluated feasibility of a new CDI sequence for tage but low repetition rate. Experimental evaluation was MREIT reconstruction of electrical conductivity during performed on ex vivo liver tissue during application of elec- electroporation. We show that the MREIT algorithm can tric pulses with the repetition rate of 1 Hz. Results presented successfully reconstruct electrical conductivity and detect in Figure 2 confirm that the MREIT algorithm was success- conductivity increase in ex vivo tissue exposed to sequences fully applied for reconstruction of electrical conductivity of electric pulses. Together with new CDI sequence they using current density distribution acquired by means of the can be used to determine electric field distribution during new CDI sequence. It can be observed that electrical pulse delivery and can be used for monitoring of electropo- conductivity increased during delivery of pulses. Due to the ration applications such as IRE ablation.

IFMBE Proceedings Vol. 53

94 M. Kranjc, I. Serša, and D. Miklavčič

ACKNOWLEDGMENT 10. Sersa G, Miklavcic D (2008) Electrochemotherapy of Tumours. J Vis Exp. doi: 10.3791/1038 11. Kranjc M, Bajd F, Serša I, Miklavčič D (2011) Magnetic resonance This study was supported by the Slovenian Research electrical impedance tomography for monitoring electric field Agency (ARRS) and conducted within the scope of the distribution during tissue electroporation. IEEE Trans Med Imaging Electroporation in Biology and Medicine European Asso- 30:1771 8. doi: 10.1109/TMI.2011.2147328 ciated Laboratory (LEA-EBAM). This manuscript is a result 12. Woo EJ, Lee SY, Mun CW (1994) Impedance tomography using internal current density distribution measured by nuclear magnetic of the networking efforts of the COST Action TD1104 resonance. Proc SPIE 2299, Math Methods Med Imaging III (www.electroporation.net). 2299:377 385. 13. Seo JK, Woo EJ (2014) Electrical Tissue Property Imaging at Low Frequency Using MREIT. IEEE Trans Biomed Eng 61:1390 9. doi: CONFLICT OF INTEREST 10.1109/TBME.2014.2298859 14. Joy M, Scott G, Henkelman M (1989) In vivo Detection of Applied The authors declare that they have no conflict of interest. Electric Currents by Magnetic-Resonance Imaging. Magn Reson Imaging 7:89 94. REFERENCES 15. Sersa I, Jarh O, Demsar F (1994) Magnetic Resonance Microscopy of Electric Currents. J Magn Reson Ser A 111:93 99. doi: 1. Neumann E, Schaeferridder M, Wang Y, Hofschneider PH (1982) 10.1006/jmra.1994.1230 Gene-Transfer into Mouse Lyoma Cells by Electroporation in High 16. Kranjc M, Bajd F, Serša I, Miklavčič D (2014) Magnetic resonance Electric-Fields. {EMBO} J 1:841 845. electrical impedance tomography for measuring electrical 2. Kotnik T, Kramar P, Pucihar G, et al. (2012) Cell membrane conductivity during electroporation. Physiol Meas 35:985 96. doi: electroporation- Part 1: The phenomenon. IEEE Electr Insul Mag 10.1088/0967-3334/35/6/985 28:14 23. doi: 10.1109/MEI.2012.6268438 17. Kranjc M, Bajd F, Sersa I, et al. (2012) Ex vivo and in silico 3. Yarmush ML, Golberg A, Serša G, et al. (2014) Electroporation-based feasibility study of monitoring electric field distribution in tissue technologies for medicine: principles, applications, and challenges. during electroporation based treatments. PLoS One 7:e45737. Annu Rev Biomed Eng 16:295 320. doi: 10.1146/annurev-bioeng- doi: 10.1371/journal.pone.0045737 071813-104622 18. Kranjc M, Markelc B, Bajd F, et al. (2015) In Situ Monitoring of 4. Miklavcic D, Corovic S, Pucihar G, Pavselj N (2006) Importance of Electric Field Distribution in Mouse Tumor during Electroporation. tumour coverage by sufficiently high local electric field for effective Radiology 274:115 123. doi: 10.1148/radiol.14140311 electrochemotherapy. Eur J Cancer Suppl 4:45 51. doi: 19. Sersa I, Kranjc M, Miklavcic D (2015) Current Density Imaging 10.1016/j.ejcsup.2006.08.006 Sequence for Monitoring Current Distribution during Delivery of 5. Miklavcic D, Towhidi L (2010) Numerical study of the electroporation Electric Pulses in Irreversible Electroporation. Biomed. Eng. Online pulse shape effect on molecular uptake of biological cells. Radiol submitted. Oncol 44:34 41. doi: 10.2478/v10019-010-0002-3 20. Maxwell JC (1865) A Dynamical Theory of the Electromagnetic 6. Garcia P a, Pancotto T, Rossmeisl JH, et al. (2011) Non-thermal Field. Philos Trans R Soc London 155:459 512. doi: irreversible electroporation (N-TIRE) and adjuvant fractionated 10.1098/rstl.1865.0008 radiotherapeutic multimodal therapy for intracranial malignant glioma 21. Manassen Y, Shalev E, Navon G (1988) Mapping of electrical in a canine patient. Technol Cancer Res Treat 10:73 83. circuits using chemical-shift imaging. J Magn Reson 76:371 374. 7. Neal RE, Kavnoudias H, Cheung W, et al. (2013) Hepatic epithelioid doi: 10.1016/0022-2364(88)90124-2 hemangioendothelioma treated with irreversible electroporation and 22. Hennig J, Nauerth A, Friedburg H (1986) RARE imaging: a fast antibiotics. J Clin Oncol 27:422 426. doi: 10.1200/JCO.2012.44.9736 imaging method for clinical MR. Magn Reson Med 3:823 33. 8. Moir J, White S a, French JJ, et al. (2014) Systematic review of irreversible electroporation in the treatment of advanced pancreatic cancer. Eur J Surg Oncol 40:1598 1604. doi: 10.1016/j.ejso.2014.08.480 Author: Matej Kranjc 9. Marty M, Sersa G, Garbay JR, et al. (2006) Electrochemotherapy - An Institute: University of Ljubljana, Faculty of Electrical Engineering easy, highly effective and safe treatment of cutaneous and Street: Tržaška 25 subcutaneous metastases: Results of {ESOPE} (European Standard City: Ljubljana, SI-1000 Operating Procedures of Electrochemotherapy) study. Eur J Cancer Country: Slovenia Email: [email protected]

Suppl 4:3 13.

IFMBE Proceedings Vol. 53

Part IV Electrochemistry and Electroporation

Mathematical Modeling of Electrochemical Phenomena at the Electrode-Solution Interface in a PEF Treatment Chamber

G. Pataro1, G.M. Barca1, and G. Ferrari1,2 1 Department of Industrial Engineering, University of Salerno, via Giovanni Paolo II, 132 84084 Fisciano (SA), Italy 2 ProdAl Scarl University of Salerno, via Ponte don Melillo 84084 Fisciano (SA), Italy

Abstract Electrochemical reactions at the electrode solu- Due to this large number of factors, numerical simulation interfaces are unavoidable when typical conditions for tions could significantly help to predict the occurrence of PEF processing are applied. The aim of this paper was to set electrochemical phenomena during PEF treatment as well as up and validate a mathematical model describing the pheno- assist in setting up optimized PEF treatments able to prevent menon of the metal release from stainless-steel (type AISI or minimize the extent of metal release. 316L) electrodes of a continuous flow parallel plate PEF treatment chamber into model liquid foods. The spatial distri- The aim of this paper was to develop a mathematical bution of the main metallic elements (Fe, Cr and Ni) released model describing the phenomena of metal release (Iron, from the electrodes in the treatment zone as well as their con- Chromium and Nickel) from the stainless-steel (type AISI centration in the bulk of the product exiting the treatment 316L) electrodes of a PEF treatment chamber in model chamber were simulated. All simulations were performed with liquid foods. The model developed was solved with FEM- the software package COMSOL Multiphysics ™. Multiphysics under processing conditions commonly used Keywords Pulsed electric field (PEF), Metal release, Finite for PEF microbial inactivation. A validation of the predicted Element Method (FEM). results was also performed.

I. INTRODUCTION II. THEORY Pulsed Electric Fields (PEF) is a promising technology A. Electrode Reactions which has gained increasing interest in the last two decades From an electrochemical point of view, a PEF chamber, as it promotes the cold pasteurization of most liquid foods which consists of two metal electrodes placed in direct with a minimum impact on their nutritional and organolep- contact with a liquid food (or any electrolyte solution), acts tic properties [1]. The polarization effect induced at each as an electrochemical cell [3]. The current flowing in the electrode-solution interface of the PEF chamber by the metal electrodes, consists of free electrons while in the passage of large current during PEF treatments, may trigger liquid food solution the current is carried by ions rather than electrochemical reactions at the electrode-solution interface, by free electrons [3]. At each electrode-solution interface, in especially those involving the metals release from elec- order to maintain the condition of electroneutrality, an ionic trodes to food. These reactions are undesired and must be double layer, which consists of charged particles and/or of minimized since they may seriously affect food safety and orientated dipoles, is developed even if no external voltage quality as well as electric field distribution and electrode is applied to the electrodes. This layer, which behaves as an lifetime [2]. The extent of these electrode reactions depends electrical capacitor, is often called double layer capacitor on many factors, which can essentially be classified into (Fig. 1). When a potential difference is applied across the three groups: processing parameters, design parameters, and electrodes, a process of electric charging of the double layer treatment medium characteristics (Table 1). takes place [4]. Once the potential drop across the double Table 1 Main parameters affecting electrochemical reactions during PEF layer overcomes the threshold voltage, which is typical of treatment the reaction potential of electrode material (~1–2 V), in order to preserve the charge conservation principle, electro- Process parameters Design parameters Food properties chemical (Faradaic) reactions will occur: oxidation reaction Electric field strength Chamber configuration Composition (i.e. loss of electrons) will take place at the electrode sur- Total Specific energy Electrode material Conductivity face, which behaves as an anode and reduction reaction (i.e. Pulsed width Electrode area pH gain of electrons) at the electrode surface which behaves as Polarity Electrode rugosity Halides a cathode [5]. These electrochemical reactions can be Frequency ...... represented by the simplified electrical equivalent circuit Pulse shape ...... reported in Fig. 2. In this circuit, Rs represents the bulk

© Springer Science+Business Media Singapore 2016 97 T. Jarm and P. Kramar (eds.), 1st World Congress on Electroporation and Pulsed Electric Fields in Biology, Medicine and Food & Environmental Technologies (WC 2015), IFMBE Proceedings 53, DOI: 10.1007/978-981-287-817-5_22

Mathematical Modeling of Electrochemical Phenomena at the Electrode Solution Interface in a PEF Treatment Chamber 99

Ni the molar flux of the charged species i, which is the sum The experimental data on metal release used to validate of three contributions (convection, migration and diffusion). the model were those reported in the previous study [2]. To solve the Eq. (1), the following initial (IC) and boundary conditions (BC) were set: IV. RESULTS AND DISCUSSION I.C.: Ci(t = 0) = 0; B.C.1: Ni at the electrode-solution interface is caused by A. Metal Release the Faradaic current; Preliminary simulations have confirmed that, as ex- B.C.2: Ni = 0 at the electrode-insulator and solution- pected, the parallel plate PEF chamber configuration lead to insulator interface; a constant potential drop across the inter-electrode gap B.C.3: Ci(Z = 0) = 0. which, in turn, determines a uniform distribution of the To quantify the metal release from the electrodes, Eq. (1) primary current density and the electric field strength within has been solved coupled with the equations for conservation the treatment zone (data not shown). As a results, since the of mass, momentum, primary current, which defines the chemical composition of the electrodes as well as of the electric potential distribution in the treatment solution regu- treatment media are homogeneous, instant by instant, also lating the migration of charged species, and secondary cur- the secondary current density is uniform and, consequently, rent, which determines the current densities of the single the molar flux of the metal species through the anodic sur- species, ji, at the electrode-solution interface. The set of face in contact with the fluid is uniform (data not shown). these partial differential equations with their relative initial Fig. 4 shows the predicted bulk concentration profiles, and boundary conditions were solved using the software Ci,bulk, evaluated at the outlet cross-section of the PEF package COMSOL Multiphysics™ (COMSOL AB, Stock- chamber, of Fe, Cr, and Ni in the treatment medium as a holm, Sweden). function of the processing time with simulations performed C. Material Properties assuming E = 21 kV/cm and WT = 60 kJ/kg. The material properties of the insulator were taken from the DuPont database and those of the stainless-steel electrodes were taken from the ASTM/ASME material data- 4000 base. The thermo-physical properties of the model buffer solution, except for the electrical conductivity which was 3000 measured as reported in Pataro et al. [2], were assumed Iron similar to those of water and taken from the NIST/ASME Chromium database. The material properties were considered constant, 2000 Nickel and their values were those at 25 °C.

D. Process Parameters 1000

Different settings of applied peak of electric field Metal Concentration (ppb) strength E (12, 21, and 31 kV/cm) and total specific energy 0 input WT (20-40-60-80-100 kJ/kg) were simulated for Triz- 0246810 ma-HCl buffer solution (pH 7, σ = 2 mS/cm at 25 °C). A Time (s) pulse width of 3.1 µs, an inlet temperature of 25 °C and a flow rate of 2 L/h (Reynolds number = 245 at 298 K) were Fig. 4 Predicted Ci,bulk profiles of Fe, Cr, and Ni in Trizma-HCl buffer solution at the outlet section of the treatment chamber as a function of the chosen. After solving the model, the predicted distributions processing time. Simulation conditions: E 21 kV/cm; WT 60 kJ/kg. of the concentration of Fe, Cr, and Ni were integrated at the outlet cross-section of the PEF chamber and the bulk con- Results clearly show that, regardless the metal species, centrations of metals at the treatment chamber exit, Ci,bulk, the amount of metal released increases with time reaching a with i = Fe, Ni, Cr, were evaluated for each processing time stationary value after about 2 s. However, although a similar with the following relationship: behavior can be observed for all the metallic elements considered, the final values of their concentration differ, ac- (2) cording to the composition of the stainless-steel electrodes. When stationary values are attained (processing time > 2s), it can be observed that the concentration of metals being v the liquid velocity vector and dS the differential released in the liquid increases as the fluid moves from surface vector normal to the flow direction Z. the inlet to the outlet of the PEF chamber while the mass

IFMBE Proceedings Vol. 53