The 29th Special CU-af Seminar 2021

The 29th Special CU-af Seminar 2021 “Deep Tech for Great Impact on Society” August 25th, 2021 Zoom Conference



The 29th Special CU-af Seminar 2021 August 25, 2021 MESSAGE from PRESIDENT, CHULALONGKORN UNIVERSITY Professor Bundhit Eua-Arporn, Ph.D. Mr. Takuya Shimamura, Chairman of the Asahi Glass Foundation, President of AGC Flat Glass (Thailand) PCL., AGC Chemicals (Thailand) Co.,Ltd., AGC Automotive (Thailand) Co.,Ltd. Distinguished Guests, Ladies and gentlemen Thank you all for joining this virtual seminar and MOU signing ceremony. It is, of course, disappointing that we are not in the situation to be together here at Chulalongkorn University, but I would also like to extend a very warm welcome to all of you to this 29th Special Chulalongkorn University and the Asahi Glass Foundation seminar virtually, as well as my congratulations to all researchers granted this year. As someone once said ‘the show must go on’, the global COVID 19 pandemic might has put the world on pause in many aspect but not in the area of research. As Research and development is one of the major factors in the development of higher education and our country nowadays, we are gearing towards a knowledge-based global community and creative economy. Therefore, the research for new knowledge and technological advancement is indispensable and must be carried out in view of particular needs and practical uses on both national and global levels. The research project revealing in the Keynote lecture session, is one of great example of this ongoing research and development. It should also be emphasized here that research in university can both enhance and enrich our teaching to be more up-to-date and relevant to the contemporary and sustainable society. However, Chulalongkorn University cannot achieve these goals without the long supports from our precious partners like the Asahi Glass Foundation. For nearly forty years, the university has been receiving continuous and generous support from the Asahi Glass Foundation of Japan, enabling us to promote various impactful research works. Amid the outbreak of COVID-19, it is extraordinarily important that this virtual seminar takes place. This CU–af seminar is organized to showcase the research projects supported by the Asahi Glass Foundation. It also allows researchers to publicise their findings and to exchange ideas among distinguished audience. I am confident that this kind of academic exchange will greatly contribute to the existing body of knowledge and enhance the atmosphere of intellectual spirit. I would like to take this opportunity to express my warmest thanks to the Asahi Glass Foundation for the generous support for nearly four decades. As the new MOU between two organizations are signing, my firm believe has ensured that the relationship between us will be even stronger in the future. I would like to express my best wishes for the success of this event and hope that the seminar, as well as the virtual MOU signing ceremony will achieve all of its goals. May I take this opportunity to declare the 29th Special Seminar open. I

The 29th Special CU-af Seminar 2021 August 25, 2021 CONGRATULATORY ADDRESS from CHAIRMAN, THE ASAHI GLASS FOUNDATION Mr.Takuya Shimamura My name is Takuya Shimamura, I am Chairman of the Asahi Glass Foundation and Chairman of AGC Inc. It is both honor and pleasure for me to be this ceremony, as a representative of Asahi Glass Foundation in Japan. Firstly, please allow me to extend my sincere gratitude to Professor Bundhit Eua-arporn President of the University, Professor Chakkaphan Sutthira, Vice President, Professor Kiat Ruxrungtham, a Keynote Speaker, our distinguished guests and the members hosting today’s ceremony. Thank you very much. Secondly, let me introduce Asahi Glass Foundation briefly. It was established in 1933 to celebrate the 25th anniversary of the founding of Asahi Glass Co., Ltd. Asahi Glass Co., Ltd. was renamed the AGC Inc. in 2018. Now we provide 3 major programs such as the Research Grant Program, Scholarship Program and Commendation Program named “Blue Planet Prize”. In addition, we have publicized the annual survey on the global environment named “Environment Doomsday Clock”. In terms of the relationship between AGC Inc. and Thailand, In 1964, AGC began to produce flat glass in Thailand, and expanding its business to automotive glass, chemicals and electric materials. Today total annual sales revenue of group is approximately 284 billion THB (9.4 Billion USD) in Thailand. And many alumni of the Chulalongkorn University have joined AGC Flat Glass, Thailand, AGC Chemicals, Thailand and so on. In 1982,Asahi Glass Foundation initiated our grant program with Chulalongkorn University. This is the 40th grant presentation ceremony. So AGC Inc, and Asahi Glass Foundation have a long history with Thailand and Chulalongkorn University as well. We have supported more than 300 grant themes and contributed approximately 2.3 M USD from the beginning. I believe these researches can contribute enough for the development and prosperity of Thailand. We wish our support can be helpful for your researchers. As one of other major activities of Foundation, We provide “Blue Planet Prize”, from 1992, which has become internationally recognized award successfully. The foundation has awarded the Blue Planet Prize to two people who have made outstanding contributes towards solving global environment issues with a certificate of merit, commemorative trophy and 50 million Japanese yen or 450,000 USD per each in prize money. Prof. David Tilman and Dr. Simon Stuart were winners in 2020. And Prof. Veerabhadran Ramanathan from USA and Prof. Moha Munasinghe from Sri Lanka this year. Please visit our website, which explains the winner’s achievement in detail. And you can also reach the names and research themes of this year’s new grantees and the research reports of the past year’s grantees. As you know, our environmental situation is getting worse every year. Humankind has sacrificed its environment in exchange for its prosperity. That is the reason why we established the “Blue Planet Prize”. I hope the winner of this prize from Thailand will be awarded someday. I sincerely hope all of researchers here can contribute to our precious Blue planet through your researches. II

The 29th Special CU-af Seminar 2021 August 25, 2021 Once again, I’d like to express my deepest appreciation to Professor Bundhit Eua-arporn, Professor Chakkaphan Sutthira, Professor Kiat Ruxrungtham, our distinguished guests, the members hosting today’s ceremony. and all members concerned kindly supported the seminar, and award ceremony as well. As closing my speech, I’d like to express the followings: I firmly believe in for all of you today, Your future is bright. Your potential is unlimited. And your possibilities are endless. Thank you very much. Takuya Shimamura Chairman of Asahi Glass Foundation August 25th, 2021 III

The 29th Special CU-af Seminar 2021 August 25, 2021 REPORT ADDRESS by VICE PRESIDENT, CHULALONGKORN UNIVERSITY Professor Chakkaphan Sutthirat, Ph.D Mr. Takuya Shimamura, Chairman of the Asahi Glass Foundation, Professor Dr. Bundhit Eua-arporn, President of Chulalongkorn University, Distinguished Guests, Ladies and gentlemen Back in 1982, Chulalongkorn University received the first support from the Asahi Glass Foundation for research. Initially, this support was focused on the research in the field of engineering. Subsequently, it has expanded to other vital areas, such as environmental science, medical science, pharmaceutical science, and information technology. While in this year, we can see that the support has also opened a new door to research project in the field of social science. Up to now, the findings of these research projects are compiled and published with the support of the Asahi Glass Foundation for dissemination to various agencies worldwide. Moreover, the seminars, that have been organized regularly over the past 28 years, are based on the research findings that help providing venues for the exchange of knowledge and expertise among academics, external organizations and agencies in both the governmental and private sectors. It is expected that the seminar will promote the application of the research in the industrial sector as well. We are proud that this special seminar in the year 2021 marks the 29th (twenty-ninth) of such gathering. We are particularly honored by the presence of today keynote speaker Professor Kiat Ruxrungtham, founder of Chula Vaccine Research Centre, on the very current topic of “ChulaCov19 mRNA vaccine: from Bench to Clinic and Manufacturing”. After his lecture, the fifteen research papers will be presented. Eight projects are from Faculty of Science, four from Faculty of Engineering, one from Faculty of Medicine, together with projects from Metallurgy and Materials Science Research Institute and Institute of Biotechnology and Genetic Engineering. Please allow me to take this opportunity to express our sincere appreciation to the Asahi Glass Foundation, Japan, for the continuous support for the research projects of Chulalongkorn University. We also would like to thank all our researchers who have worked with strong commitment on their research projects that will help enrich their teaching and their societies. May I now invite Professor Dr. Bundhit Eua-arporn, President of Chulalongkorn University, to give an address on an opening ceremony. Thank you very much. IV

The 29th Special CU-af Seminar 2021 August 25, 2021 The 29th Special CU-af Seminar 2021 “Frontier research for sustainable society” August 25, 2021 Zoom Conference 12.30 – 12.45 Registration 12.45 – 13.30 Opening Ceremony • Welcome Remark and Report Prof. Chakkaphan Sutthirat, Ph.D., DGemG. Vice President, Chulalongkorn University • Opening Address Prof. Bundhit Eua-arporn, Ph.D. President, Chulalongkorn University • Greetings from The Asahi Glass Foundation, Japan Mr. Takuya Shimamura Chairman, The Asahi Glass Foundation, Japan • MOU Signing Ceremony - Chairman of The Asahi Glass Foundation (Mr. Takuya Shimamura) - Senior Executive Director of The Asahi Glass Foundation (Mr. Hiroyuki Watanabe) - President of Chulalongkorn University (Prof. Bundhit Eua-arporn) - Vice President of Chulalongkorn University (Prof. Chakkaphan Sutthirat) Grantees (1) Dr. Pyone Myat Thu Institute of Asian Studies (2) Dr. Pattarin Tangtanatakul Faculty of Allied Health Sciences (3) Dr. Vorrapon Chaikeeratisak Faculty of Science (4) Dr. Nut Songvorawit Faculty of Science (5) Dr. Manunya Okhawilai Metallurgy and Materials Science Research Institute (6) Dr. Sorapat Niyomsin The Petroleum and Petrochemical College (7) Assist. Prof. Dr. Sawekchai Tangaramvong Faculty of Engineering V

The 29th Special CU-af Seminar 2021 August 25, 2021 (8) Assist. Prof. Dr. Dao Suwansang Janjaroen Faculty of Engineering (9) Dr. Jitti Kasemchainan Faculty of Science (10) Dr. Jenyuk Lohwacharin Faculty of Engineering (11) Dr. Manaswee Suttipong Faculty of Science (12) Dr. Yotwadee Hawangchu Aquatic Resources Research Institute 13.30 – 13.35 Virtual Group Photo 13.40 – 14.00 Keynote Speech ChulaCov19 mRNAvaccine: from Bench to Clinic and Manufacturing Prof. Kiat Ruxrungtham, M.D. ChulaVRC (Vaccine Research Center), Faculty of Medicine 14.00 – 16.00 Concurrent Sessions (Session 1, Session 2) Concurrent Sessions Session 1: Health Science Chairperson: Dr. Pattarin Tangtanatakul Department of Transfusion Medicine and Clinical Microbiology, Faculty of Allied Health Sciences 14.00 – 14.15 Investigation of Effects of Physiological and Hemodynamic Changes Observed in Patients with Diabetic Nephropathy on Glomerular Fluid and Macromolecule Filtration through a Mathematical Simulation Employing Hindered Transport Theory Assistant Professor Dr. Panadda Dechadilok Department of Physics, Faculty of Science 14.15 – 14.30 Crosstalk Between Liver Cancer Cells and TumorAssociated Macrophages in a Three-Dimensional Spheroid Culture Professor Dr. Tanapat Palaga Department of Microbiology, Faculty of Science 14.30 – 14.45 Cold plasma therapy attenuates multi-drug resistant bacteria induced infected-wound mouse-models through the neutralization of bacteria and bacterial biofilm with inducing anti-inflammatory immune cells (neutrophil) Associate Professor Dr. Asada Leelahavanichkul Imunology Unit, Department of Microbiology, Faculty of Medicine VI

The 29th Special CU-af Seminar 2021 August 25, 2021 14.45 – 15.00 Early Detection of Anthracnose on Mango Fruit Using Hyperspectral Imaging Associate Professor Dr. Ubonrat Siriptrawan Department of Food Technology, Faculty of Science 15.00 – 15.15 Functional Identification of Dof Transcription Factors Controlling Auxin Biosynthesis and Starch Degradation in Durian Fruit Ripening Associate Professor Dr. Supaart Sirikantaramas Department of Biochemistry, Faculty of Science 15.15 – 15.30 Development of a Yeast-Based Assay and Screening for Compounds that canAlleviate the Toxicity of HumanAlpha-Synuclein, a Neurodegenerative Disease Associated Protein Associate Professor Dr. Chulee Yompakdee Department of Microbiology, Faculty of Science 15.30 – 15.45 Non-Invasive Blood Glucose Monitoring through Optical Fibre Technology Dr. Charusluk Viphavakit International school of Engineering, Faculty of Engineering 15.45 – 16.00 Non-enzymatic electrochemical detection of cholesterol using β-cyclodextrin immobilised on 3D paper-based device Dr. Sudkate Chaiyo Institute of Biotechnology and Genetic Engineering Session 2: Science and Technology Chairperson: Associate Professor Dr. Sawekchai Tangaramvong Department of Civil Engineering, Faculty of Engineering 14.00 – 14.15 Coal Combustion Product Utilization for Degraded Soil Improvement in Nan Province Professor Dr. Kreangkrai Maneeintr Department of Mining and Petroleum Engineering, Faculty of Engineering 14.15 – 14.30 Application of EBSD to petroleum related strike-slip zones in Thailand Professor Dr. Pitsanupong Kanjanapayont Department of Geology, Faculty of Science 14.30 – 14.45 Utilization of bacteria for self-healing concrete Assistant Professor Dr. Pitcha Jongvivatsakul Department of Civil Engineering, Faculty of Engineering VII

The 29th Special CU-af Seminar 2021 August 25, 2021 14.45 – 15.00 Appraisal of corrosion degree and structural properties on reinforced concrete beam using mill cut steel fiber concrete Associate Professor Dr. Withit Pansuk Department of Civil Engineering, Faculty of Engineering 15.00 – 15.15 Noncovalent Functionalization of Graphene Oxide for Photocatalytic Applications Assistant Professor Dr. Pannee Leeladee Department of Chemistry, Faculty of Science 15.15 – 15.30 Synthesis and electrocatalytic activities of MgNiCoCuZn high entropy oxides for hydrogen and oxygen evolution reactions Assistant Professor Dr. Numpon Insin Department of Chemistry, Faculty of Science 15.30 – 15.45 Utilization of Gypsum-bonded Investment Mold Waste from Jewelry and Accessory Industry as Raw Material for Construction Materials using Geopolymer Technology Dr. Nithiwach Nawaukkaratharnant Metallurgy and Materials Science Research Institute VIII

The 29th Special CU-af Seminar 2021 August 25, 2021 TABLE OF CONTENTS Page I Message form President, Chulalongkorn University II Congratulatory Address form Chairman, The Asahi Glass Foundation IV Report Address by Vice President, Chulalongkorn University V Schedule Keynote Lecture 1 ChulaCov19 mRNA vaccine: from Bench to Clinic and Manufacturing Prof. Kiat Ruxrungtham, M.D. ChulaVRC (Vaccine Research Center), Faculty of Medicine Invited Papers 2 Session 1: Health Science 1. Investigation of Effects of Physiological and Hemodynamic Changes Observed in Patients with Diabetic Nephropathy on Glomerular Fluid and Macromolecule Filtration through a Mathematical Simulation Employing Hindered Transport Theory Assistant Professor Dr. Panadda Dechadilok 2. Crosstalk Between Liver Cancer Cells and Tumor Associated Macrophages 14 in a Three-Dimensional Spheroid Culture Professor Dr. Tanapat Palaga 3. Cold plasma therapy attenuates multi-drug resistant bacteria induced 26 infected-wound mouse-models through the neutralization of bacteria and bacterial biofilm with inducing anti-inflammatory immune cells (neutrophil) Associate Professor Dr. Asada Leelahavanichkul 4. Early Detection of Anthracnose on Mango Fruit Using Hyperspectral 46 Imaging 58 Associate Professor Dr. Ubonrat Siriptrawan 5. Functional Identification of Dof Transcription Factors Controlling Auxin Biosynthesis and Starch Degradation in Durian Fruit Ripening Associate Professor Dr. Supaart Sirikantaramas IX

The 29th Special CU-af Seminar 2021 August 25, 2021 6. Development of a Yeast-Based Assay and Screening for Compounds that 74 can Alleviate the Toxicity of Human Alpha-Synuclein, a Neurodegenerative Disease Associated Protein Associate Professor Dr. Chulee Yompakdee 7. Non-Invasive Blood Glucose Monitoring through Optical Fibre Technology 88 Dr. Charusluk Viphavakit 8. Non-enzymatic electrochemical detection of cholesterol using 100 β-cyclodextrin immobilised on 3D paper-based device Dr. Sudkate Chaiyo Session 2: Science and Technology 1. Coal Combustion Product Utilization for Degraded Soil Improvement 112 in Nan Province Professor Dr. Kreangkrai Maneeintr 2. Application of EBSD to petroleum related strike-slip zones in Thailand 126 Professor Dr. Pitsanupong Kanjanapayont 3. Utilization of bacteria for self-healing concrete 136 150 Assistant Professor Dr. Pitcha Jongvivatsakul 168 4. Appraisal of corrosion degree and structural properties on reinforced concrete beam using mill cut steel fiber concrete Associate Professor Dr. Withit Pansuk 5. Noncovalent Functionalization of Graphene Oxide for Photocatalytic Applications Assistant Professor Dr.Pannee Leeladee 6. Synthesis and electrocatalytic activities of MgNiCoCuZn high entropy 184 oxides for hydrogen and oxygen evolution reactions Assistant Professor Dr. Numpon Insin 7. Utilization of Gypsum-bonded Investment Mold Waste from Jewelry and 196 Accessory Industry as Raw Material for Construction Materials using Geopolymer Technology Dr. Nithiwach Nawaukkaratharnant Appendix 208 X



Keynote Lecture ChulaCov19 mRNA vaccine: from Bench to Clinic and Manufacturing Prof. Kiat Ruxrungtham, M.D.

The 29th Special CU-af Seminar 2021 August 25, 2021 Keynote Lecture ChulaCov19 mRNA vaccine: from Bench to Clinic and Manufacturing Prof. Kiat Ruxrungtham, M.D.1,2* Abstract ChulaCov19 vaccine is an mRNA vaccine candidate developed at the Chula VRC (Vaccine Research Center), Faculty of Medicine, Chulalongkorn University in collaborating with Professor Drew Weissman, University of Pennsylvania, USA. In pre-clinical studies, ChulaCov19 induced high SARS-Cov2 wild-type virus neutralising antibody and specific IFNγ ELISpots T-cell responses in mice and in cynomolgus macaques. ChulaCov19 at low dose 1 and 10 μg given 2 doses 21 days apart in mice induced high Live-virus MicroVNT50 with GMT=11,763, and 54,047, respectively. In the K18-hACE2 transgenic mice SARS-Cov2 challenge model with SARS-Cov2 wild-type, ChulaCov19 vaccine at 1 or 10 μg dose protected mice from Covid-19 clinical symptoms, reduced more than 7 log of viral load in the noses and lung, prevented viremia and induced high neutralizing antibody and specific IgA antibody. Toxicity evaluation of ChulaCoV19 mRNA vaccine in Wistar rats showed no safety concerns. Currently, ChulaCov19 mRNA vaccine (the first clinical lot was manufactured by an U.S. CMO) has been tested in Phase 1/2 trial. The interim results of ChulaCov19 given 2 doses IM at 10, 25 and 50 μg, 21 days apart showed no safety concern. ChulaCov19 elicited robust B-and T-cell responses in dose-dependent and age-dependent relationships. The GMT of neutralizing antibody titer against wild-type virus was higher than that of Covid19 convalescent sera penal, and showed cross neutralizing against Alpha, Beta, Gamma, and Delta variants. In exploratory comparison with a control arm who received Pfizer/BNT vaccine in Phase 2 results, ChulaCov19 elicited significantly higher neutralizing antibody (MicroVNT50 and pseudovirus VNT-50) and T-cell (IFNγ-ELISPOT) than Pfizer/BNT vaccine. The next clinical lot of ChulaCov19 vaccine has now been manufactured in Thailand and is ready for further registrational clinical trials. According to the development plan, ChulaCov19 may be authorized for EUA by the end of 2022. 1ChulaVRC (Vaccine Research Center), Faculty of Medicine, Chulalongkorn University 2School of Global Health, Faculty of Medicine, Chulalongkorn University *e-mail: [email protected], [email protected] 1

Investigations of Effects of Physiological and Hemodynamic Changes Observed in Patients with Diabetic Nephropathy on Glomerular Fluid and Macromolecule Filtration through a Mathematical Simulation Employing Hindered Transport Theory Numpong PUNYARATABANDHU Panadda DECHADILOK and Pisut KATAVETIN

The 29th Special CU-af Seminar 2021 August 25, 2021 Investigations of Effects of Physiological and Hemodynamic Changes Observed in Patients with Diabetic Nephropathy on Glomerular Fluid and Macromolecule Filtration through a Mathematical Simulation Employing Hindered Transport Theory Numpong PUNYARATABANDHU1* Panadda DECHADILOK1* and Pisut KATAVETIN2* Abstract Effects of the alteration of the glomerular barrier morphology and the hemodynamic factors associated with diabetic nephropathy on the first step of renal urine formation is investigated through a mathematical model where the glomerular capillary is assumed to be a network of cylinders with complex multi-layered walls separating blood plasma and primary urine. The filtrated excess fluid across the capillary wall causes the protein and test solutes concentration to increase as a function of the distance from the capillary afferent end. The average sieving coefficient, the ratio between the test solute concentration in the primary urine and that in the blood plasma in the lumen, is obtained from the steady-state finite element solution of the convection-diffusion equation and is compared to the sieving coefficient of ficolls from an in vivo study performed in healthy and early diabetic Wistar rats. 1Department of Physics, Faculty of Science, Chulalongkorn University, Bangkok, Thailand 2Divison of Nephrology, Department of Medicine Faculty of Medicine, Chulalongkorn University Bangkok, Thailand 3

The 29th Special CU-af Seminar 2021 August 25, 2021 Introduction and Objectives The first step of renal urine formation is believed to be filtration of fluid and solutes through the multi-layered glomerular capillary wall consisting of the fenestrated endothelial cell layer, the glomerular basement membrane (GBM), and the epithelial cell layer with the slit diaphragm connecting the podocytes (Haraldsson et al., 2008). The changes of the capillary nanostructure are found to be associated with proteinuria and glomerular filtration rate (GFR) change: symptoms often seen in patients with renal diseases (Deen et al., 2001). Diabetic nephropathy which is a leading cause of end-stage renal diseases is found to be associated with the glomerular barrier structural changes that include GBM thickening and damage of the glycocalyx layer lining the lumen (Ponciardi et al., 2013), and the alteration in hemodynamic factors such as glomerular hypertension (Andersen et al., 2000). The objective of the present study is to examine the effects of the altered capillary wall structure and the change in the hemodynamic factors observed in patients with diabetic nephropathy on fluid and macromolecule filtration through the glomerular barrier through a computational simulation employing hindered transport theory. The relationship between the glomerular barrier structure and its selectivity has been a topic of interest among theoretical researchers for decades (Deen et al., 2001).Previous mathematical models include those completed by assuming that the glomerular barrier is a porous medium (Maddox et al., 1992; Rippe and Haraldsson, 1994) or a fibrous medium (Curry and Michel, 1990) or a combination of a porous and fibrous media (Oberg and Rippe, 2013). Up until now, the model with the simplified geometry of the filtration barrier most closely resembling the mammalian glomerular capillary wall is the “ultrastructural model” proposed by Drumond and Deen (1994, 1995) and developed further by Edwards et al. (1999) where the capillary wall consisted of repeating three-layered subunits; GBM and the epithelial slit were assumed to be the layers mainly responsible for transport restriction based on the images of the slit diaphragm obtained from transmission electron microscopy (TEM) by Rodewald and Karnovsky (1974). Later electron microscopic images from scanning electron microscopy (SEM) by Gagliardini et al. (2010) and helium ion microscopy (HIM) by Rice et al. (2013), however, demonstrate that the mean spacing between fibers of the epithelial slit can be up to approximately 5 times larger than previously believed. Punyaratabandhu et al. (2017) employed the ultrastructural model but with the linear dimension of the epithelial slit fiber spacing being those reported by Gagliardini et al. (2010) and Rice et al. (2013) and the endothelial cell layer also responsible for restricting fluid transport and macromolecule sieving; the computed sieving coefficient (the ratio between the macromolecule concentration in the primary urine in Bowman’s Space and that in the capillary lumen) is found to agree well with sieving coefficients of ficolls, highly crosslinked polysaccharide with the structure in aqueous environment resembling rigid spheres, from an in vivo study in healthy humans (Blouch et al., 1997). The aim of the present study is to theoretically explore the effects of the structural and hemodynamic changes associated with diabetic nephropathy on glomerular fluid and macromolecule filtration. The glomerular capillaries are viewed as a network of parallel long cylinders containing blood plasma with the cylinder wall structures following the ultrastructural model; the wall is assumed to be three-layered; the filtrated fluid and macromolecules are transported through the endothelial fenestrae, across GBM and through the epithelial slit spacing into primary urine in Bowman’s Space. Next, the effect of the glomerular mesangium on fluid and solute transport is explored assuming that two-third of the capillary wall surface is the three-layered glomerular filtration and one-third of the surfaces is the four-layered barrier with the glomerular mesangium being between the endothelial cell layer and GBM. Computed sieving coefficients of uncharged 4

The 29th Special CU-af Seminar 2021 August 25, 2021 rigid spheres is, then, compared to sieving coefficients from an in vivo study (Rippe et al., 2007). Methods The computation simulation was completed by assuming that the glomerular capillaries is a network of long parallel cylinders with complex and multi-layered surfaces separating blood plasma (inside the cylinders) and primary urine (outside the cylinders). Under an assumption that convection dominates the process of protein transport inside the cylinders (Maddox et al., 1992), the flux of the plasma protein in the axial direction can be expressed as (1) where q(x) is the axial plasma flow rate in a single capillary and Cprotein is the protein concentration. Cprotein,afferent and qafferent is the protein concentration and the fluid flux at the afferent end of the cylinders. x is the axial distance from the afferent end; at the efferent end, x = L .capillary The axial fluid flux, q, declines as a function of x due to fluid leakage across the cylinder wall; its change can be written as follows. (2) where Jv is the fluid flux across the cylinder wall, and s is the cylinder surface. Substituting the expression in Eq. (2) into Eq. (1) and applying the chain rule for finding derivatives, one obtains the following expression for the change in the protein concentration along the capillary length. (3) The transcapillary fluid flux (Jv) is dependent on the transcapillary hydraulic and osmotic pressure differences (ΔP and ΔΠ) as follows. (4) where k is the glomerular hydraulic permeability. In other words, (1/k) is the glomerular filtration surface hydraulic resistance. During the first phase of the present study, only the contribution of the glomerular filtration surface consisting of the endothelial cell layer, GBM and the epithelial cell layer (as shown in Figure 1) is considered. 5

The 29th Special CU-af Seminar 2021 August 25, 2021 Figure 1: Schematic drawing of an idealized geometry of a subunit of a glomerular filtration surface where the fluid and solutes are transported through the endothelial fenestrae, GBM and the epithelial slit into the primary urine in Bowman’s Space. The total glomerular hydraulic permeability can be calculated as the summation of the resistances of the three filtration surface layers as shown in Figure 1; 1/k≅ 1/kfs = 1/ken + 1/kGBM + 1/kep where ken, kGBM and kep are the hydraulic permeabilities of the endothelial cell layer, GBM and the epithelial cell layer. ken and kep are computed with the method employed by Punyaratabandhu et al. (2017), whereas kGBM obtained based on the expression proposed by Drumond and Deen (1994) based on the solution of the continuity equation. However, because, in the present study, GBM is assumed to contains fibers with two different sizes, its Darcy permeability is calculated using the “volume weighted resistivity” approach proposed by Claque and Phillips and proven to be a good approximation in the calculation of the Darcy permeability of a fibrous medium with two different types of fibers by Mattern and Deen (2008). The osmotic pressure difference (ΔΠ) appeared in Eq. (4) is dependent on Cprotein; in the present study it is calculated using the empirical expression proposed by Maddox et al. (1992) as (5) The variation of the hydraulic pressure difference (ΔP) also appeared in Eq. (4) as a function of x, on the other hand, is computed as that of a flow in a cylindrical tube with the fluid flux leaking across the wall being much smaller than the axial fluid flux, q (Maddox et al., 1992). Substituting the expression indicated in Eqs. (4) and (5) into Eq. (3), one obtains the first order nonlinear differential equation as a governing equation for Cprotein. It was solved using the method of Runge-Kutta order 4 (MATLAB, Netick, MA, USA) with the solution satisfies the condition that Cprotein(x = 0) = Cprotein,afferent where Cprotein,afferent is set to be the systemic oncotic pressure. The obtained Cprotein(x) scaled with Cprotein,afferent is shown in Figure 2; because of the transcapillary fluid flux causing q to declines as a function of the axial distance from the afferent end (x), Cprotein/ Cprotein,afferent grows as a function of x (scaled with the capillary length, Lcapillary). After Cprotein(x) and, subsequently, ΔΠ as well as Jv have been computed, it is possible to calculate the concentration of test solutes in the capillary lumen (Ctracer). Assuming that, inside the capillaries, the axial test solute flux is mainly convective, the axial variation of the test solute flux is due to the solute flux (Jtracer) that leaks through the capillary wall: 6

The 29th Special CU-af Seminar 2021 August 25, 2021 (6) where θfs(x) is the local sieving coefficient (the ratio between the local concentration of the test solute filtrated through the filtration surface into Bowman’s Space and that in the lumen); it is determined using the method described below. Substitutions of the expressions in Eqs. (1) – (5) into Eq. (6) yields (7) where Ctracer,afferent is the test solute concentration at the afferent end of the capillary. The protein concentration scaled with the protein concentration at the afferent end of the capillaries are presented in Figure 2. Because q declines as a function of the axial distance from the afferent end (x) and the convective flux is conserved, C Cprotein/ protein,afferent grows as a function of x/Lcapillary. Figure 2: The protein concentration in the blood plasma in the lumen (Cprotein) scaled with the afferent protein concentration (C )protein,afferent as a function of the distance from the capillary afferent end (x) scaled with the capillary length. The transcapillary fluid leakage caused the protein concentration to increase as a function of x/Lcapillary. Prior to obtaining C /C ,tracer tracer,afferent one must determine the local sieving coefficient (θfs). In the present study, it is computed as the product of the sieving coefficient across the three layers shown in Figure 1 as follows. (8) where θen, θGBM and θep are the local sieving coefficient across the endothelial fenestrae, that across GBM and that across the epithelial cell layer. They are computed as discussed by Punyaratabandhu et al. (2017). θen and θGBM are computed from steady-state solutions of the convection-diffusion equation governing a solute concentration in a fibrous membrane. It is known that, inside the fibrous medium, the solute diffusivity is reduced due to the hydrodynamic and steric interaction between the solute and the fibers. GBM is assumed to be a homogenous and isotropic material containing fibers of two different sizes; 90% of its volume is that of water, whereas 5% of its volume is contributed by the combined volume of the fibers the same size as that of type IV collagen, and the rest of its volume is contributed by the volume of the fibers the same size as that of GAGs. The solute hydrodynamic interaction with both types of fibers is taken into account by using the approximation similar to that of Oseen in the calculation of the drag on a sphere from two parallel plates (Oseen, 1927), whereas only its steric interaction with collagen is included as GAGs are flexible (Punyaratabandhu et al., 2017). In addition to changing the solute diffusivity, the presence of the fibers in the fibrous medium also alters 7

The 29th Special CU-af Seminar 2021 August 25, 2021 the solute convection rate. In the present work, the “apparent” test solute velocity in GBM is computed as the velocity of a freely suspending sphere (of the same size) in the Brinkman medium with the same Darcy permeability as that of GBM (Punyaratabandhu et al., 2017). As for the calculation of θen, the endothelial fenestrae only contains GAGs; only the hydrodynamic interaction between the test solutes and the fibers the same size as that of GAGs is included in the calculation of the test solute diffusivity, whereas the “apparent” test solute velocity in the fenestrae is computed as the velocity of a freely suspending sphere (of the same size) in the Brinkman medium with the same Darcy permeability as that of the fibrous media filling the fenestrae computed using the expression of Amsden (Amsden, 1998). θep is computed as the average sieving coefficient through a row of parallel cylinders using the expression obtained by Drumond and Deen (1995): the spacing between fibers follows the lognormal distribution with the mean and variance based on the observation of Rice et al. (2013). After the local test solute sieving coefficient is known, Eq. (7) is solved using the Runge-Kutter order 4 method (MATLAB, Netick, MA, USA) with the condition that Ctracer/Ctracer,afferent(x=0) = 1. The test solutes concentration scaled with the concentration at the afferent end of the capillaries are presented in Figure 3. Because the volume flow rate declines as a function of the axial distance from the afferent end (x) and the convective flux is conserved, C /Ctracer tracer,afferent grows as a function of x/Lcapillary. Figure 3: The test solute concentration in the blood plasma in the lumen (Ctracer) scaled with the afferent test solute concentration (C )tracer,afferent as a function of the distance from the capillary afferent end (x) scaled with the capillary length. The transcapillary fluid leakage caused the test solute concentration to increase as a function of x/Lcapillary with the increase of C /Ctracer tracer,afferent of the test solutes with larger solute radii (α = 4.07 nm) being more rapid than that of the test solutes with smaller radii (α = 2.75 nm). It is desired to relate our computed results to the quantities that can be measured from experiments such as the sieving coefficient of ficolls from an in vivo study (Rippe et al., 2007). The observable sieving coefficient of the test solutes is that averaged over the entire length of the capillary as (8) where, as Bowman’s Space is a dead-end chamber, the test solute flux across the glomerular barrier can be written as (9) 8

The 29th Special CU-af Seminar 2021 August 25, 2021 The average sieving coefficient through the filtration surface, θfs(avg), is then compared to the sieving coefficient of ficolls from an in vivo study performed in healthy Wistar rats (Rippe et al., 2007) and to the sieving coefficient computed by assuming that the capillaries consist of repeating subunits. As will be demonstrated in Figure 6, the results based on an average over the length of the capillary, θfs(avg), is very close to that obtained from one subunit, θfs(one .unit) As a result, during the second phase of the project where the effect of fluid and solute transport through mesangium is investigated, the sieving coefficient calculated as θfs(one unit) is utilized. It is believed that, whereas two-third of the glomerular barrier is the three-layered filtration surface as shown in Figure 1, approximately one-third of the barrier is four-layered with the glomerular mesangium being between the endothelial cell layer and GBM. If θmesangium is the test solute sieving coefficient across the four-layered structure (that includes the mesangium), the total test solute sieving coefficient, θtotal, can be calculated as (Punyaratabandhu et al., 2021) (10) where Sfs is the surface area of the filtration surface and Smesangium is the surface area of the four-layered barrier that includes the mesangium per glomerulus. kmesangium is the hydraulic permeability of the four-layered barrier; the relationship between the fluid velocity and the pressure gradient is based on Darcy’s Law, and kmesangium is computed from a finite element solution (COMSOL Multiphysics, Stockholm, Sweden) of a Laplace equation as the average fluid velocity per unit pressure difference. θmesangium is calculated from the steady-state finite element solution of the convection-diffusion equation ((COMSOL Multiphysics, Stockholm, Sweden). In the glomerular mesangium, the reduction of the solute diffusivity is computed by assuming that the increased drag on the test solute is that on a sphere translating in a Brinkman medium, and the test solute velocity is that of the freely suspending sphere in the Brinkman medium (Punyaratabandhu et al., 2021). After kmesangium and θmesangium are determined, θtotal is calculated and compared to the ficoll sieving coefficient from a study performed in rats with early diabetes (Rippe et al., 2007). Results and Discussion In Figure 4, the sieving coefficient across the glomerular filtration surface averaged over the capillary length, θfs(avg), is presented as a function of solute radii, α. Computed results are plotted for the volume fraction of GAGs in the endothelial fenestrae, ϕGAG(en), being 0.06, 0.07, 0.08 and 0.095. As shown in the figure, θfs(avg) increases as ϕGAG(en) decreases, indicating that the damage of GAGs in the fenestrae is likely to affect glomerular size-selectivity. The best fit to the ficoll sieving coefficient from an in vivo study performed in healthy Wistar rats (Rippe et al., 2007) for α in the range of 2.74 – 5.50 nm is obtained when ϕGAG(en) = 0.095. For the solute radii exceed 5.50 nm, our calculation greatly underestimates the sieving coefficient (in comparison to the ficoll sieving coefficient from the experiment of Rippe et al. (2007). It is possible that there is a different pathway that allows a passage of larger solutes and is one of the directions for future works. 9

The 29th Special CU-af Seminar 2021 August 25, 2021 Figure 4: The test solute sieving coefficient (the ratio between the test solute concentration in the primary urine and that in the blood plasma in the lumen) averaged over the capillary length, θfs(avg), as a function of solute radii, α. Results are presented for the volume fraction of GAGs in the fenestrae, ϕGAG(en), being 0.06, 0.07, 0.08 and 0.095. Also included is the sieving coefficient of ficolls obtained from an in vivo study performed in healthy Wistar rats (Rippe et al., 2007). To examine the effect of hypertension on macromolecule sieving through the glomerular filtration surface, θfs(avg) as a function of α is presented for different values of the afferent glomerular pressure difference (the difference between the afferent hydraulic and osmotic pressure, ΔPafferent – ΔΠafferent) in Figure 5. Results indicate that the increase in ΔPafferent – ΔΠafferent slightly decreases θfs(avg). Next, the sieving coefficient across the glomerular filtration surface averaged over the capillary length, θfs(avg), is compared with the sieving coefficient computed by assuming that the filtration surface consists of repeating subunit and that the sieving coefficient of the entire filtration surface is similar to the sieving coefficient of one of these subunits, θfs(one .unit) As shown in Figure 6, θfs(avg) is very close to θfs(one .unit) In order to make the problem more tractable, θfs(avg) is replaced with θfs(one unit) in the subsequent calculations of the present study. Figure 5: Test solute sieving coefficient Figure 6: Test solute sieving coefficient averaged over the capillary length, θfs(avg), averaged over the capillary length, θfs(avg), as a function of solute radii, α. Results are is compared to the test solute sieving presented for the afferent glomerular total coefficient calculated from one of the subnits pressure difference, the difference between of the capillary wall (shown in Figure 1) as the afferent hydraulic and osmotic pressure a function of solute radii, α. difference (ΔPafferent – ΔΠafferent) being 9.1, 10.1, 11.1, 12.1 and 13.1 mmHg. 10

The 29th Special CU-af Seminar 2021 August 25, 2021 The effect of solute transport across mesangium on glomerular selectivity is, then, investigated. In Figure 7a, θmesangium (the sieving coefficient across the four-layered barrier with the mesangium being between the endothelial cell layer and GBM) is plotted as a function of solute radii; as expected, θmesangium declines as a increases. In Figure 7b, θfs(one unit) and θtotal (the sieving coefficient based on the average of θfs(one unit) and θmesangium found using Eq. (10) and the physiological parameters of rats with early diabetes (Rippe et al., 2007)) is presented as a function of α. Also presented is the ficoll sieving coefficient from a study performed in rats with early diabetes induced by streptozotocin (Rippe et al., 2007). θfs(one unit) is found to be almost graphically indistinguishable from θtotal, implying that the effect of solute transport across the mesangium is likely to be small. At ϕGAG(en) = 0.088 (slightly smaller than ϕGAG(en) that yields the best fit to the ficoll sieving coefficient from a study performed in healthy Wistar rats shown in Figure 4), θfs(one unit) and θtotal agree quite well with the ficoll sieving coefficient from a study performed in rats with early diabetes for α < 5.50 nm. For larger solutes, our computation greatly underestimates the sieving coefficient, once again implying that there could be another pathway that allows a passage of large solutes. Figure 7: Sieving coefficient as a function of solute radii. Presented results include (a) the sieving coefficient across the four-layered barrier with the mesangium being adjacent to the endothelial cell layer and GBM (θmesangium) and (b) the sieving coefficient across the filtration surface (θfs(one )unit) as well as their average (θtotal) and ficoll sieving coefficients from a study performed in rats with early diabetes (Rippe et al., 2007). Because θfs(one unit) and θtotal are almost graphically indistinguishable, and the calculation of θfs(oneunit) is substantially easier than that of θtotal, in the subsequent calculation, θtotal is replaced with θfs(one .unit) In Figure 9, θfs(one unit) computed with the GBM thickness (δGBM) being 200 nm is presented as a function of solute radii and is compared to θfs(one unit) calculated by assuming that δGBM = 400 nm. As demonstrated in the figure, the change in δGBM almost does not alter θfs(one ,unit) indicating that the change in GBM thickness hardly affects the glomerular size-selectivity. 11

The 29th Special CU-af Seminar 2021 August 25, 2021 Figure 8: Sieving coefficient as a function of solute radii. Presented results are computed for the GBM thickness (δGBM) = 200 and 400 nm. Conclusion In conclusion, this study investigates the effect of the changes associated with diabetic nephropathy on glomerular selectivity. The computed sieving coefficient agrees well with the sieving coefficient of the small and moderate-sized ficolls from an in vivo study performed in healthy Wistar rats and Wistar rats with early diabetes induced by streptozotocin (Rippe et al., 2007). The damage of GAGs in the endothelial fenestrae due to hyperglycemia is likely to affect glomerular size-selectivity of small and moderate-sized macromolecules, whereas the effects of glomerular hypertension, GBM thickening and transport across the glomerular mesangium are smaller. The possible pathway for large macromolecules is one of the directions of future work. References 1. Amsden, B., Polym. Gels Netw., 1998 (6), 13-43. 2. Andersen S., Blouch K., Bialek J., Deckert M., Parving H., and Myers, D.M., Kidney Int., 2000 (58), 2129 – 2137. 3. Blouch, K., Deen, W.M., Fauvel, J.-P., Bialek, J., Derby, G., and Myers, B.D., Am. J. Physiol., 1997 273 (3), F430 - F437. 4. Curry, F.E., and Michel, C.C., Microvasc. Res., 1980 (20), 96-99. 5. Deen, W.M., Lazzara, M.J., and Myers, B.D., Am. J. Physiol. Renal Physiol., 2001 (281), 579-596. 6. Drumond, M.C., and Deen, W.M., Am. J. Physiol., 1994 (266), F1 - F12. 7. Drumond, M.C., and Deen, W.M., J. Biomech. Eng., 1995 (117), 414-422. 8. Edwards, A., Daniels, B.S., and Deen, W.M., 1999, Am. J. Physiol. Renal Physiol., 1999 (276), F892-F902. 9. Gagliardini, E., Conti, S., Benigni, A., Remuzzi, G., and Remuzzi, A., J. Am. Soc. Nephrol., 2010 (21), 2081–2089. 10. Haraldsson, B., Nystrom, J., and Deen, W.M., Physiol. Rev., 2008 (88), 451 - 487. 11. Maddox, D.A., Deen, W.M., and Brenner, B.M., Supplement 25: Handbook of Physiology, Renal Physiology. 1992. 12. Mattern K.J., and Deen, W.M., AIChE J., 2008 (54), 32 – 41. 13. Öberg, C.M., and Rippe, B., Am. J. Physiol. Renal Physiol., 2013 (304), F781–F787. 14. Oseen, C.W., Neuere Methoden und Ergebnisse in der Hydrodynamik, 1927. 12

The 29th Special CU-af Seminar 2021 August 25, 2021 15. Ponchiardi C., Mauer M., and Najafian B., Curr. Diab. Rep., 2013 (13), 592-599. 16. Punyaratabandhu N., Kongoup P., Dechadilok P., Katavetin P., and Triampo, W., J. Biomech. Eng., 2017 (139), 121105-1 – 13. 17. Punyaratabandhu N., Roongthamskul Y., Dechadilok P., and Katavetin P., 2021 (In Preparation). 18. Rice, W.L., Hoek, A.N.V., Paunescu, T.G., Huynh, C., Goetze, B., Singh, B., Scipioni, L., Stern, L.A., and Brown, D., PLoS One, 2013 (8), e57051. 19. Rippe, B., and Haraldsson, B., Physiol. Rev., 1994 (74), 163-219. 20. Rippe, C., Rippe A.R., Torffvit O., and Rippe, B. Am. J. Physiol. Renal Physiol., 2007 (293), F1533-F1538. 21. Rodewald, R., and Karnovsky, M.J., J. Cell Biol., 1974 (60), 423-433. 13

Crosstalk between liver cancer cells and tumor associated macrophages in a three-dimensional spheroid culture Tanapat PALAGA and Pornlapat KEAWVILAI

The 29th Special CU-af Seminar 2021 August 25, 2021 Crosstalk between liver cancer cells and tumor associated macrophages in a three-dimensional spheroid culture Tanapat PALAGA1* and Pornlapat KEAWVILAI2* Abstract Monocytes and macrophages are innate immune cells that play essential roles in host defense against infections. Their roles in cancer, are increasingly appreciated. Tumor-associated macrophages (TAMs) often correlate with poor disease outcomes by promoting tumor growth and metastasis. Because tumor cells and TAMs interact in a complex three dimension (3D) in vivo, it is more relevant to study the 3D interaction in vitro. In this study, we set up a 3D tumor spheroid culture and investigated the impact on monocytes. 3D spheroid culture of two liver cancer cell lines, PLC/PRF/5 and HepG2, restored expression of liver cancer markers and cancer stem cell markers. Co-culturing cancer spheroids with primary CD14+ monocytes increased expression of CD206 and CD163 while decreased M1 markers. This effect was seen only in infiltrated monocytes. Taken together, these results strongly indicated that direct contact of cancer cells and monocytes conditions TAMs to take the tumor promoting phenotypes. 1Department of Microbiology Faculty of Science, Chulalongkorn University Bangkok, Thailand 2Graduate Program in Biotechnology Faculty of Science, Chulalongkorn University Bangkok, Thailand 15

The 29th Special CU-af Seminar 2021 August 25, 2021 Introduction and Objectives Monocytes and macrophages are innate immune cells that play essential roles in host defense against infections. Monocytes can be found in circulation and macrophages are mainly found as tissue resident cells. The origins and detailed differentiation pathway of monocytes and macrophages in different tissues have been extensively revised in recent years[1]. Some tissue resident macrophages are continuously repopulated from the circulating monocytes and some are exclusively derived from embryonic origin[2]. In addition to their protective roles against infections, the physiological roles of macrophages have expanded in recent years, and their involvement in organ development, tissue homeostasis, and metabolic dysfunctions, such as diabetes and obesity, are increasingly appreciated. Cancer is another area in which macrophages have emerged as a crucial player in the creation of a tumor microenvironment (TME) that supports tumor growth and metastasis, in opposition to their traditional role as an innate immune cell, whose function is to eliminate cancer cells[3,4]. Macrophages are often observed in large numbers in tumor tissues and high numbers of tumor-associated macrophages (TAMs) often correlates with poor disease outcomes, partly by promoting tumor growth, dampening immune responses, and inducing angiogenesis and metastasis[5]. Considering the key roles that macrophages play in these pathological states, various studies attempt to manipulate macrophages for controlling diseases such as in cancer. However, the results of these studies show mixed results, possibly due to the lack of the fundamental knowledges on molecular mechanisms in tumor cell-TAM crosstalk. In addition, whether circulating monocytes or tissue resident macrophages or both represent TAMs. In one study using breast cancer animal model, TAMs are derived from newly recruited monocytes but not tissue resident macrophages [6]. TME and tumor cells exhibit strong impact on infiltrating immune cells. Most studies thus far utilized the 2D co-culture system to elucidate the interaction between tumor cells on TAMs. However, it is now well recognized that there are more complex interactions in a three-dimensional (3D) manner in TME in vivo. Therefore, elucidating the impact of tumor cells-TAMs cross talk that regulate macrophage functions in 3D spheroid culture system is expected to yield relevance and informative insights into how tumor cells and TME condition monocytes/macrophages to become TAMs that function to promote tumor growth and invasion[7]. Liver cancer is one of the leading causes of death in cancer patients in Thailand and around the world. Compared to other cancer types, liver cancer is often diagnosed at late stage and the overall survival of patients are not significantly improved in the past 10 years despite intense search for more effective treatment[8]. Because in the liver, it contains resident macrophages called Kupffer cells, liver presents a unique environment for interactions among newly recruited TAMs and tumor cells. TAMs found in hepatocellular carcinomas (HCC) from patients exhibited correlation between vascular invasion and metastasis and depletion of TAMs in animal models of liver cancers resulted in significant decreased tumor growth and metastasis[9]. In contrast, there are some conflicting reports that TAMs in HCC predicted good disease outcome in HCC but this result did not characterize the phenotypes of TAMs in HCC[10]. Careful analysis revealed that TAMs in HCC exhibited a phenotype of wound healing macrophages or the so called M2 macrophages which function opposing that of pro-inflammatory macrophages, M1[11]. Recent evidence suggested that an evolutionarily conserved Notch signaling plays an important role in regulating macrophage functions during inflammatory responses and in cancer .[12] For TAMs, Notch signaling is required for differentiation from monocyte to TAMs in breast cancer model[6]. Inhibiting Notch signaling during this stage prevent cancer cell growth 16

The 29th Special CU-af Seminar 2021 August 25, 2021 and metastasis in animal models. Therefore, Notch signaling is one of the candidates of the signaling pathway for manipulating TAMs for anti-cancer response. How Notch signaling in TAMs functions in HCC has not been documented. In this study, we aim to address the fundamental questions regarding the molecular mechanisms that underlie macrophage functions in liver cancer. Specific research questions focus on the cross talk between macrophages and tumor cells using 3D spheroid culture and the impact of manipulating Notch signaling in monocytes on tumor cell growth using liver cancer as a model. The objectives of this study are: 1) To establish the three-dimensional (3D) spheroid culture of liver cancer cells 2) To study the phenotypes of macrophages in 3D macrophage-tumor co-culture system 3) To study the impact of manipulating Notch signaling in TAMs on tumor growth and phenotypes in 3D co-culture Methods Establishing HCC 3D cancer cell culture To optimize 3D spheroid culture system, Ultralow attachment plates were used to determinethedurationofculture.Twohepatocellularcarcinomacelllines,PLC/PRF/5(JCRB0406; Hepatitis B virus surface antigen or HBsAg positive) and HepG2 (JCRB1054; HBsAg negative) were seeded in Dulbecco’s Modified Eagle Medium (DMEM), supplemented with 10% (v/v) heat-inactivated fetal bovine serum, 1% (w/v) sodium pyruvate, 1% (w/v) HEPES and 1% (w/v) gentamicin in Ultralow attachment plates. The obtained spheroids were harvested at different times for detecting expression of liver cancer cell markers by quantitative polymerase chain reaction (qPCR). Gene expression profiles in 3D culture model of liver cancer spheroid Total RNA from the spheroid from 3D cell culture system were extracted by direct-zol RNA kits (Zymo Research) according to the manufacturer’s instruction. Total RNA was used as template to convert to cDNA and mRNA expression of genes of liver cancer cell markers and liver cell functions were detected by using quantitative RT-PCR (qRT-PCR). The relative level of gene expression normalized with housekeeping gene, GAPDH and compared with liver cancer cell line cultured in the 2D cell culture system. In this study, the marker genes are listed in table. 17

The 29th Special CU-af Seminar 2021 August 25, 2021 Primary human monocyte isolation The use of healthy donor buffy coat to isolate mononuclear cells has been approved by the Institutional Review Board, Faculty of Medicine, Chulalongkorn University (IRB No.683/62). Peripheral blood mononuclear cells (PBMCs) were separated from buffy coat by using Ficoll-Paque PREMIUM 1.084 (GE Healthcare, UK) density gradient by centrifugation at 1,500 rpm for 45 min. CD14+ monocyte was isolated from PBMCs by using MojoSort™ Human CD14+ Monocyte negative selection Kit (BioLegend) according to the manufacturer’s instruction. The purity of CD14+ monocyte was measured by flow cytometry using CD14 and CD16 as markers to be more than 90% before co-cultivation with liver cancer spheroid. Co-culture of liver cancer cell lines and primary CD14+ monocytes or THP-1 cell line After establishing a 3D liver spheroid cell culture system, the spheroids were co-cultured with primary CD14+ monocytes or modified human monocytic leukemia cell line, THP-1 (JCRB0112). THP-1 cell line used in this study was manipulated to express GFP (THP-1 GFP) as control or Notch1 intracellular domain (THP-1 NICD1). PLC/PRF/5 or HepG2 cell line suspension were added into Ultra low attachment plate to allow for spheroid formation at 37 °C in 5% CO2 incubator for 7 days. Spheroids were disrupted and the primary CD14+ monocytes or THP-1 cell line were added. The co-culture spheroids were co-cultured for 5 and 7 days and the infiltrating monocytes were harvested and characterized by flow cytometry. Spheroid growth To study the impact of monocyte on liver cancer spheroid growth, the co-culture spheroids were observed and visualized by phase contrast microscopy. The areas (size) of spheroid were measured using ImageJ (NIH) software. Phenotypes of spheroid infiltrating monocytes To characterize and confirm the impact of cancer spheroids on infiltrating monocytes, the co-culture spheroids were harvested by using 100 µm Reversible strainer (STEMCELL) to obtain only spheroids. The co-culture spheroids were disassociated by incubation with Accutase at 37 °C, while carefully re-suspended by pipetting up and down every 10 min. The cell suspension contained the liver cancer cell line and infiltrated monocytes. The primary CD14+ monocyte or THP-1 were detected by CD45 expression and GFP, respectively, to identify monocyte population by flow cytometry. To determine the phenotypes of infiltrating monocytes, the surface markers of M1 macrophages (i.e., CD80 and CD86) and M2 macrophages (i.e. CD206 and CD163) were detected by Beckman Coulter Cytomics FC500 Flow Cytometer and the acquired data were analyzed by FlowJo V10 software. Data analysis and statistics All experiments were done in biological triplicates with at least two independent experiments. All results were presented as the mean ± SEM. Unpaired two-tailed Student’s t-test was used for comparison between the two groups. All statistical analyses were performed with GraphPad PRISM 5.03 which p-value ≤ 0.05 was considered statistical significance. 18

The 29th Special CU-af Seminar 2021 August 25, 2021 Results and Discussion Characteristics of liver cancer in 3D cancer cell culture To examine the effect of liver cancer model in 3D spheroid culture and 2D culture, we collected mRNA from HepG2 and PLC/PRF/5 spheroids after spheroid formation for 7 and 14 days or cells cultured in 2D system (Figure 1A, B). The levels of liver cancer-associated genes were evaluated. In both liver cancer cell lines, all examined genes (cancer stemness markers, liver cancer marker and epithelial-mesenchymal transition (EMT) markers) were upregulated when cultured in 3D cell culture system (Figure 1C, D). The increased cancer stemness markers are Nanog, Oct4 and CD44 which indicated the self-renewal capacity and spherogeneicity of liver cancer cell lines mimicking the tumor characters[13]. The expression level of liver marker ALB was higher in the 3D spheroids, which suggested that these spheroids may help liver cancer cell line regains some hepatocyte function[14]. The liver cancer marker AFP was higher in 3D PLC/PRF/5 but not in 3D HepG2. Moreover, the expression of mesenchymal related gens VIM was upregulated in both liver cancer spheroids which indicated the metastatic ability of 3D liver cancer spheroids. These data strongly indicated that 3D liver cancer cell culture showed the gene expression characteristics mimicking an in vivo tumor growth than the 2D culture. Figure 1: Characteristics of gene expression in the 3D spheroids of liver cancer cell lines. (A-B) Liver cancer cell lines were cultured in the ultra-low attachment plates and observed under phase contrast microscope. Both liver cancer cell lines formed unique shape of spheroids. (C-D) Differential gene expression between liver cancer cell lines in 2D and 3D culture are shown for HepG2 (C) and PLC/PRF/5 (D). Relative mRNA level was analyzed using quantitative PCR. The data were analyzed by unpaired two-tailed Student’s t-test. * p < 0.05, ** p < 0.01, *** p < 0.001. 19

The 29th Special CU-af Seminar 2021 August 25, 2021 Phenotypes of monocytes/macrophages in 3D co-culture system Co-culturing between primary CD14+ monocytes and liver cancer cell lines in a 3D spheroid cultures were performed to study the crosstalk between liver cancer cells and tumor associated macrophages (TAMs). The co-culture spheroids were dissociated with Accutase to obtain a single cell suspension. Infiltrating monocytes were identified by staining with CD45 and characterized the expression of M1/M2 surface markers by flow cytometry. M1 markers (CD80 and CD86) and M2 markers (CD206 and CD163) were measured in infiltrating monocytes at day 5 and day 7 after co-culturing (Figure 2A, B). Isolated monocytes and peripheral monocytes that did not infiltrate into the spheroids were used as control. The results showed that the infiltrating monocytes expressed low levels of CD80 and CD86 in all conditions of HepG2 spheroids but these levels slightly increased in PLC/PRF/5 spheroids. More importantly, these infiltrating monocytes expressed higher levels of M2 markers, CD206 and CD163, in spheroids of both liver cancer cell lines. Co-culturing with PLC/PRF/5 spheroids showed higher percentages of TAMs with M2 markers than those of HepG2. Moreover, M2 marker expression of infiltrating monocytes at day 7 was significantly higher when compared with monocytes in the peri-spheroids, indicating the critical roles of direct contact of cancer cells-monocytes. Interestingly, the CD206+ TAMs were observed in HCC tissues and correlated with the metastasis stage[11]. In the addition, the impact of infiltrating TAMs on liver cancer spheroid growth revealed that coculturing of monocytes significantly increased spheroid areas (Figure 2C, D). TAMs-derived cytokines and chemokines play an important role in the liver cancer growth and progression[15]. 20

The 29th Special CU-af Seminar 2021 August 25, 2021 Figure 2: The interaction between liver cancer cell lines and monocytes/macrophages affected the phenotype of infiltrating monocytes and spheroid growth. (A-B) Expression of representative M1/M2 markers in freshly isolated monocytes, peri-spheroid monocytes and infiltrating monocytes were analyzed by flow cytometry after co-culturing with HepG2 (A) or PLC/PRF/5 (B) for 5 and 7 days in a 3D cell culture system. (C-D) The areas of spheroid were visualized by phase contrast microscope and the areas were calculated by Image-J analysis. All data were presented as mean ± SEM, the experiments were conducted at least three replicates. * p < 0.05, ** p < 0.01, *** p < 0.001 (unpaired two-tailed Student’s t-test). Impact of manipulating Notch signaling in TAMs on tumor growth and the phenotypes in 3D co-culture To study the impact of manipulating Notch signaling in TAMs on liver cancer in 3D co-culture, the THP-1 NICD1 and THP-1 GFP were co-cultured with either liver cancer cell lines. These THP-1 were modified by retrovirus transduction (Naunpun Sangphech)[16]. NICD1 overexpression causes activating Notch signaling bypassing the ligand-receptor interaction while control THP-1 cell line expressed GFP as control. The co-culture spheroids were collected and dissociated to the single cells. The infiltrating THP-1 was identified by GFP+ cells and the expression of M1/M2 markers of these infiltrating THP-1 were evaluated by flow cytometry (Figure 3 A, B). We observed that THP-1 NICD1 expressed CD86 higher than THP-1 GFP, suggesting that Notch signaling promotes M1 phenotype. From previous reports, Notch1 signaling was activated in the murine macrophage cell line by pro-inflammatory stimuli[17]. Upon co-culturing in cancer spheroids, THP-1 NICD1 further enhanced the percentages of cells positive for CD86. This enhanced CD86 expression is more pronounced in PLC/PRF/5 spheroids with infiltrating cells more than cells in the peri-spheroids. Furthermore, CD206 expression in the infiltrating THP-1 GFP was significantly higher than that in THP-1 NICD1, suggesting that Notch signaling suppressed M2 polarization induced by cancer spheroids. Next, to study the impact of manipulating Notch signaling in TAMs on tumor growth, the area of liver cancer spheroid was measured (Figure 3 C, D). It was found that the areas of co-culture spheroids did not change between THP-1 GFP and THP-1 NICD1 spheroids. 21

The 29th Special CU-af Seminar 2021 August 25, 2021 Figure 3: The impact of manipulating Notch signaling in TAMs on phenotypes and tumor growth in 3D co-culture The phenotypes of infiltrating THP-1 were analyzed by flow cytometry at day5 and day7 that THP-1 NICD1 and THP-1 GFP were co-cultured with HepG2 (A) and PLC/PRF/5 (B). The tumor growth was estimated via areas of co-culture spheroid (C, D), that they were compared with liver cancer monoculture. All data were presented as mean ± SEM. n = 3. * p < 0.05, ** p < 0.01, *** p < 0.001 (unpaired two-tailed Student’s t-test). 22

The 29th Special CU-af Seminar 2021 August 25, 2021 Conclusion TAMs are the major immune population infiltrating in the tumor mass that contribute to protumoral tumor microenvironment. They play a critical role in supporting tumor progression and metastasis. In this study, 3D spheroid culture of liver cancer cell lines was used to study the impact of cell-cell contact between monocytes and liver cancer cells on the phenotypic changes of TAMs. The study showed that 3D spheroid culture of liver cancer cell lines increased mRNA expression of genes related to liver cancer cells in vivo such as hepatocellular carcinoma tumor marker AFP. Addition of primary CD14+ monocytes into these 3D spheroid cultures induced expression of pro-tumoral macrophage (M2) markers such as CD206 and CD163 especially in TAMs that infiltrated the spheroids, suggesting the direct cell-cell contact mechanism in conditioning monocytes to become M2-like TAMs. When monocytic cell line THP-1 was engineered to expressed constitutively active Notch1 intracellular domain (NICD1) that has been previously shown to bias toward M1 polarization, decreased CD206 and increased CD86 were observed. The size of the spheroid did not significantly differ between control and THP-1 NICD1 co-culture experiment. These results strongly indicated that 3D spheroid liver cell culture provides a better model for investigating cancer cell-immune cell interaction. More importantly, liver cancers manipulate infiltrated monocytes toward pro-tumoral M2 phenotypes and targeting TAMs using M1 polarizing signaling pathway may provide effective therapeutic approach for treatment of cancers. References 1. Ginhoux, F., and Jung, S. (2014) Monocytes and macrophages: developmental pathways and tissue homeostasis. Nat Rev Immunol 14, 392-404 2. Hoeffel, G., and Ginhoux, F. (2015) Ontogeny of Tissue-Resident Macrophages. Front Immunol 6, 486 3. Noy, R., and Pollard, J. W. (2014) Tumor-associated macrophages: from mechanisms to therapy. Immunity 41, 49-61 4. Chanmee, T., Ontong, P., Konno, K., and Itano, N. (2014) Tumor-associated macrophages as major players in the tumor microenvironment. Cancers (Basel) 6, 1670-1690 5. 5) Biswas, S. K., Allavena, P., and Mantovani, A. (2013) Tumor-associated macrophages: functional diversity, clinical significance, and open questions. Semin Immunopathol 35, 585-600 6. Franklin, R. A., Liao, W., Sarkar, A., Kim, M. V., Bivona, M. R., Liu, K., Pamer, E. G., and Li, M. O. J. S. (2014) The cellular and molecular origin of tumor-associated macrophages. Science 344, 921-925 7. Broutier, L., Mastrogiovanni, G., Verstegen, M. M., Francies, H. E., Gavarró, L. M., Bradshaw, C. R., Allen, G. E., Arnes-Benito, R., Sidorova, O., and Gaspersz, M. P. J. N. m. (2017) Human primary liver cancer–derived organoid cultures for disease modeling and drug screening. Nat Med 23, 1424 8. Wong, M. C., Jiang, J. Y., Goggins, W. B., Liang, M., Fang, Y., Fung, F. D., Leung, C., Wang, H. H., Wong, G. L., and Wong, V. W. J. S. r. (2017) International incidence and mortality trends of liver cancer: a global profile. Sci Rep 7, 1-9 9. Kong, L., Zhou, Y., Bu, H., Lv, T., Shi, Y., Yang, J. J. J. o. E., and Research, C. C. (2016) Deletion of interleukin-6 in monocytes/macrophages suppresses the initiation of hepatocellular carcinoma in mice. J Exp Clin Cancer Res 35, 1-11 23

The 29th Special CU-af Seminar 2021 August 25, 2021 10. Dong, P., Ma, L., Liu, L., Zhao, G., Zhang, S., Dong, L., Xue, R., and Chen, S. J. I. j. o. m. s. (2016) CD86+/CD206+, diametrically polarized tumor-associated macrophages, predict hepatocellular carcinoma patient prognosis. Int J Mol Sci 17, 320 11. Shu, Q. H., Ge, Y. S., Ma, H. X., Gao, X. Q., Pan, J. J., Liu, D., Xu, G. L., Ma, J. L., Jia, W. D. J. J. o. c., and medicine, m. (2016) Prognostic value of polarized macrophages in patients with hepatocellular carcinoma after curative resection. J Cell Mol Med 20, 1024-1035 12. Palaga, T., Wongchana, W., and Kueanjinda, P. J. F. i. i. (2018) Notch signaling in macrophages in the context of cancer immunity. Front Immunol 9, 652 13. Gheytanchi, E., Naseri, M., Karimi-Busheri, F., Atyabi, F., Mirsharif, E. S., Bozorgmehr, M., Ghods, R., and Madjd, Z. J. C. C. I. (2021) Morphological and molecular characteristics of spheroid formation in HT-29 and Caco-2 colorectal cancer cell lines. Cancer Cell Int 21, 1-16 14. Sun, L., Yang, H., Wang, Y., Zhang, X., Jin, B., Xie, F., Jin, Y., Pang, Y., Zhao, H., and Lu, X. J. F. i. O. (2020) Application of a 3D bioprinted hepatocellular carcinoma cell model in antitumor drug research. Front Oncol 10, 878 15. Capece, D., Fischietti, M., Verzella, D., Gaggiano, A., Cicciarelli, G., Tessitore, A., Zazzeroni, F., and Alesse, E. J. B. r. i. (2013) The inflammatory microenvironment in hepatocellular carcinoma: a pivotal role for tumor-associated macrophages. BioMed Res. Int 2013, 1-15 16. Sangphech, N., Keawvilai, P., and Palaga, T. J. F. O. b. (2020) Notch signaling increases PPARγ protein stability and enhances lipid uptake through AKT in IL‐4‐stimulated THP‐1 and primary human macrophages. FEBS Open bio 10, 1082-1095 17. Palaga, T., Buranaruk, C., Rengpipat, S., Fauq, A. H., Golde, T. E., Kaufmann, S. H., and Osborne, B. A. J. E. j. o. i. (2008) Notch signaling is activated by TLR stimulation and regulates macrophage functions. Eur. J. Immunol 38, 174-183 24



Cold plasma therapy attenuates multi-drug resistant bacteria induced infected-wound mouse-models through the neutralization of bacteria and bacterial biofilm with inducing anti-inflammatory immune cells (neutrophil) Cong Phi Dang Sirapong Waewseetong Awirut Chareonsappakit Kritsanawan Sae-Khow Decho Thong-Aram and Asada Leelahavanichkul

The 29th Special CU-af Seminar 2021 August 25, 2021 Cold plasma therapy attenuates multi-drug resistant bacteria in- duced infected-wound mouse-models through the neutralization of bacteria and bacterial biofilm with inducing anti-inflammatory immune cells (neutrophil) Cong Phi Dang1,2* Sirapong Waewseetong1,2* Awirut Chareonsappakit1,2* Kritsanawan Sae-Khow1,2* Decho Thong-Aram1,2* and Asada Leelahavanichkul1,2* Abstract Plasma medicine is a utilization of gas ionization energy that might be beneficial for treatment of burn wound. The non-thermal Argon-based plasma flux was tested on macrophages (RAW246.7) and in mouse models of burn wound with or without Staphylococcus aureus infection. Accordingly, plasma flux enhanced reactive oxygen species (ROS), using Dihydroethidium assay, and decreased abundance of NF-κB-p65 (Western blot analysis) in macrophages. In parallel, plasma flux upregulated IL-10 in lipopolysac-charide (LPS)-induced inflammatory macrophages, while down-regulated pro-inflammatory cytokines (IL-1β and IL-6). Additionally, plasma flux improved migratory function of fibroblasts (L929) (fibroblast scratch assay) but not fibroblast proliferation. Moreover, once daily plasma flux for 7 days promoted healing-process in wound with or without infection (wound area and wound rank score). In conclusion, plasma flux induced anti-inflammatory macrophages and promoted burn wound healing process partly through the decrease in macrophage NF-κB. Hence, plasma flux treatment should be tested in patients with burn wound. 1Department of Microbiology, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand 2Translational Research in Infammation and Immunology Research Unit (TRIRU), Department of Microbi-ology, Chulalongkorn University, Bangkok, Thailand 27

The 29th Special CU-af Seminar 2021 August 25, 2021 Introduction and Objectives Plasma is a completely or partly ionized gas that has been categorized as the fourth state of matter in additional to solid, liquid and gas[1]. A plasma, as a state of matter, is similar to a gas, in term of an ability to change volume and shape, but unlike a gas, plasma can also alter the electrical charges[1]. Currently, plasma flux has been used in medicine (plasma medicine)[1] and could be operated at normal atmospheric pressure generating the temperatures, range between 30 and 40 °C, that are suitable for utilization on the living organisms without toxicity, referred to as “non-thermal plasma therapy”[2]. Energy from plasma flux initiates ionizing radiation that induces various active molecules, including reactive oxygen species (ROS) and reactive nitrogen species (RNS), without the heat production[3]. Concentrations of the active molecules could be adjustable by the levels of plasma flux energy using different intensities and treatment durations that are suitable for several specific purposes, ranging from enhanced cell proliferation to induction of the program cell deaths[4-7]. Non-thermal plasma does not destroy normal cells, but selectively eliminates only severely injured cells or cancer cells, at least in part, due to the prominent stress-induced ROS in these cells[8-10]. Additionally, energy from ionizing radiation of plasma flux also demonstrates bactericidal activity, regardless of the anti-biotic re-sistant properties[11]. Due to the selective elimination and promoted proliferation of abnormal cells and healthy cells, respectively, with bactericidal activity, plasma flux has been used in several medical conditions (malignancy, wound care and organismal control)[2]. As such, plasma flux enhanced re-epithelialization, collagen synthesis, angiogenesis and anti-inflammation status in a rodent wound model[12], partly through the induction of several growth factors in keratinocytes .[13] Hence, non-thermal plasma therapy is an emerging treatment strategy ,[1,2,12] especially for diabetic wound [14] and burn injury[15]. Indeed, burn wound, a dermal injury from extreme insults (heat, erosive agents or electricity), is classified upon the depth of wound damage (superficial, partial thickness and full thickness) that lead to several severe complications, including hypovolemia, hypermetabolism, gut permeability defect and opportunistic infections .[16] Unfortunately, the incidence of burn wound is still high, especially in low- and middle-income families, with an extremely high economic im-pact of chronic wound care and treatment for the complications .[17] Then, effective wound care is necessary to reduce mortality rate of burn injury and the country bur-dens. Interestingly, secondary bacterial infection in burn wound is one of the major complications, especially with a currently high prevalence of anti-microbial resistance[18]. However, antibiotic resistant organisms are still vulnerable to the organismal eradication with several physical strategies, including ultraviolet light, radiation and plasma flux energy[19,20]. Hence, evaluation of plasma flux on infected burn wound is interesting. Despite extensive evaluations of plasma flux in several models of traumatic wound[21], data of non-thermal plasma on infected burn wound is still very less, In parallel, macrophages are the immune cells responsible for either organismal control or wound healing process which could be manipulated by non-thermal plasma therapy as demonstrated in a solid cancer model[22]. As such, macrophages are immune cells with pleomorphic functions, referred to as “macrophage polarization”, that consist of pro-inflammatory M1 and anti-inflammatory M2 polarization[23]. The in-duction of macrophage polarization into a proper direction for each situation might be beneficial. For example, an acceleration of M1 and M2 for the conditions with immune exhaustion (tumor micro-environments) and hyper-immune responses (infection), respectively, might improve the clinical outcomes .[24] In wound with infection, pro-inflammatory M1 polarization is necessary for organismal 28

The 29th Special CU-af Seminar 2021 August 25, 2021 control but the excessive pro-inflammation worsen wound healing process[25]. With a proper organismal control, M2 macrophages or anti-inflammatory macrophages enhance several process-es (anti-inflammation, debris removal and matrix remodeling) and promote wound healing[25]. Because i) plasma flux (in a proper energy level) could induce an-ti-inflammatory wound status with bactericidal activity that might be beneficial for wound healing and ii) a lack of data of plasma effect on macrophage function, plasma flux treatment in burn wound with infection is interesting. Hence, plasma flux was tested in vitro experiments, using macrophages (RAW264.7 cell line) and fibroblasts (L929 cell line), and also evaluated in vivo using burn wound mouse models with or without infection. Methods Animal model and plasma flux generator C57BL/6 mice, purchased from Nomura Siam International (Pathumwan, Bangkok, Thailand), were used following an approved animal protocol from the Institution-al Animal Care and Use Committee of the Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand. The burn wound mouse models with and without infection were performed in accordance to the previous publications[26,27]. Briefly, in burn wound model without infection, an aluminum rod with 1 cm in diameter was heated up to 100 °C using a dry block heater (Thermo fisher Scientific, Waltham, MA, USA) for 5 mins before placing on shaved area at dorsal part of the mice under isoflurane anesthesia. After that, mice with heated rod attached to the back were turned ventral upward at dorsum for allowing their own body weight as a controlled pressure on the heated rod for 20 seconds. In infected wound model, the heated rod was placed with the same procedures of the non-infected wound model. Then, 1x107 CFU of Staphylococcus aureus (ATCC 25923, the American Type Culture Collection, Manassas, VA, USA) in 0.1 mL normal saline (NSS) was directly spread onto the wound at 30 min after the injury following a previous publication[26]. After 24 hours, necrotic tissue at the burn wound was removed in both wound models. Subsequently, the treatment by non-thermal plasma flux for 30 seconds along with a visual evaluation of the wound was performed daily until the 7th day of experiments when the mice were sacrificed via cardiac puncture under isoflurane anesthesia with sample collection (blood and tissue at the wound). For the in-house plasma generator, the machine was designed following the principle of atmospheric pressure plasma with the alternating current (AC) electricity at 13 kHz in frequency usingArgon gas flow to release the ionized enriched gas[28] as demonstrated in the diagram (Figure 1A-B). 29

The 29th Special CU-af Seminar 2021 August 25, 2021 Figure 1: Schematic diagram of non-thermal atmospheric pressure plasma. Component of plasma flux generator in schematic diagram (A) and the representative pictures of machine (B, left side) with an electrode-containing syringe (B, middle) and the electrode during plasma flux releasing (B, right side) are demonstrated. Cell viability test (MTT assay) in RAW264.7 cells after plasma flux treatment at 30 seconds (30s) in different voltage levels (C) and at 10 voltage with different exposure durations (in seconds) (D) compared with cells without stimulation (Non) or stimulation with Argon gas alone without plasma energy (Argon) are demonstrated. Independent triplicated experiments were performed. Wound injury score and gut permeability determination The wound was macroscopically evaluated by direct visualization based on per-centage of the injury area compared with the initial injury area at 24 h post-injury[29]. Due to a circular-shape of the wound, wound area was calculated using wound diameters which were measured by a Vernier caliper (Thermo fisher Scientific). Addition-ally, the wound rank score, a score for determination of inflammatory signs, was evaluated as a previous publication[29]. Briefly, burn wound was assessed through 4 scoring criteria, including wound length and depth (wound closure assessment), wellness and redness (inflammatory degree evaluation), using a semi-quantitative scale at 0-3 for each characteristic. Because gut permeability defect (gut leakage) in severe burn wounds is possible[30,31], leaky gut is also tested in the models. As such, fluorescent isothiocyanate-dextran or FITC-dextran at 4.4 kDa (FD4, Sigma, St. Louis, MO, USA), a non-gastrointestinal absorbable substance, at 12.5 mg in 500 µL sterile phosphate buff-er solution (PBS) was orally administered at 3 h prior to the sacrifice. Then serum FITC-dextran was analyzed by fluorescent-spectrometry with Varioskan Flash micro-plate reader (Thermo Fisher Scientific). Presentation of an intestinal non-absorbable substance in serum after an oral administration demonstrates gut permeability defect[32]. Mouse sample analysis The inflammatory cytokines in serum and in the wound tissue were evaluated using Enzyme-linked Immunosorbent (ELISA) assay (Biolegend, San Diego, CA, USA). For cytokine in wound tissue, fresh skin samples were washed in PBS, weighted, ho-mogenized and thoroughly sonicated. After that, supernatant from the samples was used for cytokine evaluation. For bacterial burdens in the wound, tissue from the wound were weighted and minced in small pieces before dissolved in PBS (1 g tissue per 1 mL PBS). After that, the samples were directly streaked onto tryptic soy agar (TSA) plate (Oxoid, Thermo Fisher Scientific) in serial dilutions and incubated for 24 h at 37 oC before colony enumeration. 30

The 29th Special CU-af Seminar 2021 August 25, 2021 The in vitro experiments on macrophages Murine macrophages (RAW264.7) (TIB-71TM) (ATCC) were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Thermo fisher Scientific) and Penicillin-Streptomycin in 5% car-bon dioxide (CO2) at 37 ⁰C overnight. After that, the non-stimulated cells were used as a control group. Meanwhile, in the stimulated groups, macrophages at 2x105 cell/ well were administered by plasma flux or Argon gas without energy (another control group) with different intensities and durations before the cell collection. To determine the proper dose of plasma flux, the cell viability test after plasma flux administration was performed following a published protocol using tetrazolium dye 3-(4, 5-dimethylthiazol-2-yl)-2, 5- diphenyltetrazolium (MTT) (Thermo Fisher Scientific)[33]. In brief, the cells in 96-well plate at the indicated time-points were incubated with MTT solution in dark at 37 oC. After 2 hours, the MTT solution was replaced by dimethyl sulfoxide (DMSO; Thermo Fisher Scientific) and the dissolving purple color was measured by Varioskan Flash microplate reader (Thermo Fisher Scientific) at OD 570 nm. With the selected proper dose and duration of plasma flux treatment, reactive oxygen species (ROS) in macrophages was determined by Dihydroethidium (DHE) fluorescent dye (Sigma-Aldrich, St. Louis, MS, USA) following a previous protocol[34]. In brief, DHE diluted in FBS free media at concentration of 20 µM was incubated with the cells at 37 oC for 20 min before analysis by Varioskan Flash microplate reader (Thermo Fisher Scientific) at OD 520 nm. In parallel, protein abundance of the possible down-stream molecules of ROS stimulation, including AMP-activated protein kinase (AMPK) and nuclear factor kappa B (NF-κB) was determined by Western blot analysis following a previous protocol[35]. Briefly, protein was extracted from the samples using a lysis buffer (radioimmunoprecipitation assay buffer; RIPA) in supplement with inhibitors of protease and phosphatase (Thermo Fisher Scientific) and measured protein concentration via Bradford assay. Then, the samples were segregated in 10% SDS (sodium dodecyl sulfate) polyacrylamide gel and transferred into the nitrocellulose membrane. Thereafter, several primary antibodies against AMPK, phosphorylated AMPK, NF-κB p65, phosphorylated NF-κB p65 or the internal control β-actin (Cell signaling, Beverly, MA, USA) were incubated with the membrane prior to horseradish peroxidase (HRP)-conjugated second antibodies and visualized by chemiluminescence (Thermo Fisher Scientific). The qualification of band intensity was calculated by Image Studio Lite Ver 5.2 software. To test effect of plasma flux on inflammatory responses, lipopolysaccharide (LPS), a potent inflammatory stimulator from Gram-negative bacteria, using LPS of Escherichia coli 026: B6 (Sigma-Aldrich) at 100 ng/ mL or media control was added after plasma flux administration. The cell supernatant was used for cytokines measurement using ELISA assay (Biolegend) and the cells were collected to determine expression of sever-al genes by quantitative polymerase chain reaction (qPCR) following a previous proto-col[36] using Trizol and SYBR Green reagents. The list of primers is demonstrated in Table 1. 31

The 29th Special CU-af Seminar 2021 August 25, 2021 Table 1: List of the primers for macrophage polarization and glycolysis pathway were demonstrated. The in vitro experiments on fibroblasts Murine fibroblasts (L929) (CCL1­ ™) (ATCC) were cultured in modified DMEM overnight in similar to RAW246.7 cells. For fibroblast scratch wound assay, the cells were seeded in 24-well plate with 1x105 cell/ well and incubated overnight to gain monolayer with the cell confluence higher than 80% following a previous publication[37]. The scratch in culture plates was performed with a gently scraping using a 200 µL pipette tip on the cell layer. Then cell debris was washed with warm PBS and incubated with modified DMEM following by plasma flux or Argon gas without plasma. The gap between two edges of scratch was photographed over different time-points to represent wound closure of fibroblast. In parallel, impact of plasma on fibroblast proliferation was evaluated through reduction of carboxy-fluorescein diacetate succinimidyl ester (CFDA-SE), a long-term fluorescent dye tracer of cells, in the daughter cells as previously published[38]. In brief, the fibroblasts were stained with 2 µM CFDA-SE (Sigma-Aldrich) diluted in warm PBS for 15 minutes at 37 oC in dark. Then, the cells were washed twice by PBS and further incubated with warm media for 6 hours prior to plasma flux treatment. The daughter cells was collected at 48 h after plasma treatment for detecting fluorescent intensity by flow cytometry BD LSRII cytometer (BD Biosciences) using FlowJo software. Statistical analysis GraphPad Prism 5.0 (GraphPad Software, Inc., San Diego, CA, USA) was used to generate graphs and statistical analysis of the experiments. The in vitro experiments were performed in three independent times and demonstrated by mean ± standard error (SEM). Data analysis between two or more groups were assessed by Student’s t-test or one-way analysis of variance (ANOVA) with Tukey’s analysis, respectively. The significance was determined by the p value less than 0.05 Results and Discussion Administration of non-thermal plasma flux promoted healing process of burn wound through several wound healing promotion factors (anti-inflammatory macro-phages and improved fibroblast migration). 1. Non-thermal plasma flux induced anti-inflammatory macrophages and fibroblast migration, but not fibroblast proliferation 32

The 29th Special CU-af Seminar 2021 August 25, 2021 The in-house non-thermal plasma flux generator using Argon gas in atmospheric pressure with an adjustable plasma energy (electrical frequency and voltage) was developed (Figure 1A-B). Due to the importance of macrophages on wound healing process, different plasma flux energy levels were applied on macrophages (RAW246.7 cell line) before the cell viability test (MTT assay, see method) to determine a proper dose of plasma flux. With duration of plasma treatment at 30 second (s), all of the selected energy levels in a range between 9.5 and 12.5 volts (V) demonstrated a non-statistical difference on cell viability when compared with control non-treatment (Non) or Argon gas alone without plasma flux energy (Argon) (Figure 1C). However, the plasma energy below 10 V showed a tendency of higher survive macrophages (Figure 1C). Then, plasma flux energy with 10 V was applied in the different durations, from 30 second (30s) to 90s, and the duration less than 30s demonstrated the less numbers of macrophage cell death (Figure 1D). Notably, Argon gas alone did not affect macrophages and only Argon with 90s was showed in the figure 1D. Subsequently, plasma flux energy using 10 V in 30s was selected for the further experiments. Because of the well-known enhanced ROS production by plasma flux[3], ROS in macrophages, using DHE fluorescent dye (see method), was measured. As expected, DHE in plasma-treated macrophages was higher than control conditions (Non and Argon) (Figure 2A). Figure 2: Biological impact of plasma flux on cell signaling of RAW264.7 cells. Characteristics of macrophages without stimulation (Non) or after stimulation with Argon gas alone (Argon) or plasma flux for 30 seconds (Plasma 30s) as evaluated by intracellular reactive oxygen species (ROS) using Dihydroethidium (DHE) fluorescent (neutralized by Hoechest nucleus staining dye) (A) and abundance of phosphorylated NF-κB-p65 (p-NF-κB) and phosphorylated- AMPK (p-AMPK) with the representative pictures of Western blot analysis (B-D) are demonstrated. Independently triplicated experiments were performed. In parallel, the possible downstream signals of ROS, including AMPK (a protein sensing cell energy shortage and anti-inflammatory mediator)[39] and NF-κB (a com-mon proinflammatory transcriptional factor)[40] was determined. Accordingly, abun-dance of activated NF-κB (ratio of phosphorylated NF-κB-p65/ NF-κB-p65), but not AMPK (ratio of phosphorylated AMPK/ total AMPK), decreased after plasma flux administration (Figure 2B-D), implying a modulation of macrophage responses without cell energy alteration. Because of i) an important transcriptional factor for cytokine production of NF-κB [40] and ii) a potent pro-inflammatory activation property of lipopolysaccharide (LPS), an organismal molecule[41], effect of plasma flux on macro-phages with or without LPS stimulation was tested. Accordingly, LPS induced pro-inflammatory cytokines (TNF-α and IL-6) with the peak levels at 24 h post-stimulation (Figure 3A, B). Meanwhile, LPS induced pro-inflammatory macrophage polarization as evaluated by expression of IL-1β (peak level at 3-6 h post-LPS) and iN-OS (peak level at 24 h Post-LPS) (Figure 3C, D). In parallel, LPS enhanced an-ti-inflammatory markers including 33

The 29th Special CU-af Seminar 2021 August 25, 2021 IL-10 (similarly between 3-24 h post-LPS) and Arg-1 (peak level at 24 h post-LPS), but not Fizz and TGF-β (Figure 3E-H). Upregulation of anti-inflammatory IL-10 was highest in LPS-stimulated macrophages with plasma flux (Figure 3E). Without LPS (PBS control), plasma flux mildly induced anti-inflammatory macrophages as determined by the upregulated Fizz, but not other genes (IL-10, Arg-1 and TGF-β) (Figure 3E-H). Due to an importance of fibroblasts in wound healing process[9], an impact of plasma on fibroblasts (L929 cell line) was further tested. As such, wound closure as determined by fibroblast migration (the scratch wound assay) using the distance be-tween each edge of fibroblast monolayer in cell culture plates was lesser in the plasma treated group at 24 h post-treatment (Figure 4A, B). However, plasma flux did not enhance fibroblast cell proliferation as evaluated by CFDA-SE fluorescent staining (Figure 4 C, D). 34

The 29th Special CU-af Seminar 2021 August 25, 2021 Figure 3: Plasma flux partially attenuated inflammation in LPS-activated macrophages. Responses of plas-ma-pretreated macrophages (RAW264.7 cells) followed by the stimulation with LPS or phosphate buffer saline (PBS) (negative control) at different time-points as evaluated by supernatant cytokines (TNF-alpha, IL-6) (A, B), gene expression of M1 macrophage polarization markers (IL-1β and iNOS) (C, D) and M2 macrophage polarization markers (IL-10, Arg1, Fizz and TGF-β) (E-H) are demonstrated. Independently triplicated experiments were performed. Figure 4: Plasma flux improved fibroblast migration but not cell proliferation. Response of fibroblasts (L929 cell line) to plasma treatment (30 seconds) (30s), when compared with cells without stimulation (Non) or stimulated with Argon gas alone without plasma energy (Argon), as evaluated by fibroblast migration (fibroblast scratch assay) with the representative microscopic photographs in different time-points and scoring comparison (A, B) with fibroblast proliferation using Carboxyfluorescein diacetate succinimidyl ester (CFDA-SE) fluorescent staining and the representative flow cytometry analysis for CFDA-SE (B, C) are demonstrated. Independently triplicated experiments were performed. 35