UNIVERSITY OF OULU P .O. Box 8000 F I -90014 UNIVERSITY OF OULU FINLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

University Lecturer Tuomo Glumoff

University Lecturer Santeri Palviainen

Postdoctoral research fellow Sanna Taskila

Professor Olli Vuolteenaho

University Lecturer Veli-Matti Ulvinen

Planning Director Pertti Tikkanen

Professor Jari Juga

University Lecturer Anu Soikkeli

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-952-62-1822-9 (Paperback)ISBN 978-952-62-1823-6 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

U N I V E R S I TAT I S O U L U E N S I S

MEDICA

ACTAD

D 1452

AC

TAJanne C

apra

OULU 2018

D 1452

Janne Capra

DIFFERENTIATION AND MALIGNANT TRANSFORMATION OF EPITHELIAL CELLS

3D CELL CULTURE MODELS

UNIVERSITY OF OULU GRADUATE SCHOOL;UNIVERSITY OF OULU,FACULTY OF MEDICINE;BIOCENTER OULU

ACTA UNIVERS ITAT I S OULUENS I SD M e d i c a 1 4 5 2

JANNE CAPRA

DIFFERENTIATION AND MALIGNANT TRANSFORMATION OF EPITHELIAL CELLS3D cell culture models

Academic dissertation to be presented with the assent ofthe Doctoral Training Committee of Health andBiosciences of the University of Oulu for public defence inAuditorium F202 of the Faculty of Medicine (Aapistie 5 B),on 16 March 2018, at 12 noon

UNIVERSITY OF OULU, OULU 2018

Copyright © 2018Acta Univ. Oul. D 1452, 2018

Supervised byDocent Sinikka EskelinenProfessor Tuomo Karttunen

Reviewed byDocent Varpu MarjomäkiDocent Satu Kuure

ISBN 978-952-62-1822-9 (Paperback)ISBN 978-952-62-1823-6 (PDF)

ISSN 0355-3221 (Printed)ISSN 1796-2234 (Online)

Cover DesignRaimo Ahonen

JUVENES PRINTTAMPERE 2018

OpponentDocent Juha Klefström

Capra, Janne, Differentiation and malignant transformation of epithelial cells.3D cell culture modelsUniversity of Oulu Graduate School; University of Oulu, Faculty of Medicine; Biocenter OuluActa Univ. Oul. D 1452, 2018University of Oulu, P.O. Box 8000, FI-90014 University of Oulu, Finland

Abstract

The epithelial cells form barriers that compartmentalize the organs. Carcinomas are cancersstemming from epithelial cells and are the most common cancer type. The aim of this study wasto understand the differentiation and malignant transformation of epithelial Madin-Darby caninekidney (MDCK) cells and to analyse the electrophysiological parameters which regulate theirtransport capacity. Emphasis was placed on comparing different culture environments, both in 2Dand 3D. First, the effects of drugs or basal extracellular fluid composition on MDCK cell, cyst andlumen volumes were analysed using time-lapse microscopy. The results showed that MDCK cellswere capable of both water secretion and reabsorption. The cells were able to perform thesefunctions in a hyperpolarizing or depolarizing environment; change in osmolality of basal fluidwas not required. Taken together, these results validate MDCK cells as a good basic model forstudying kidney function. Next, the aim was to analyse the effect of 2D and 3D cultureenvironments on the gene expression of untransformed MDCK and temperature sensitive ts-Src -transformed MDCK cells and the changes a single oncogene can induce. Microarray analysisrevealed a decrease in the expression of survivin, an inhibitor of apoptosis protein, when switchingthe untransformed cells from 2D environment to 3D. This downregulation of survivin occurs inadult tissues as well, indicating that the cells grown in 3D are closer to the in vivo state than 2Dcells. Src oncogene induced disintegration of cell junctions, but did not downregulate E-cadherinexpression. The last part was to study further the factors controlling survivin expression and itssignificance to cell survival. MDCK cells grown in 3D did not suffer apoptosis if the cellsremained in contact with the extracellular matrix. If MDCK cells were denied of ECM contactsthey were more susceptible to apoptosis than survivin-expressing ts-Src MDCK cells. Finally, ifcells were denied of cell-cell junctions, cells lacking survivin suffered apoptosis even though theyhad proper cell-matrix contacts. Taken together, these results highlighted the importance ofcellular contacts to the cells: MDCK cells needed ECM contacts to differentiate and cell-cellcontacts to avoid apoptosis.

Keywords: apoptosis, live cell microscopy, MDCK cells, polarity, Src kinase, survivin,transepithelial transport

Capra, Janne, Epiteelisolujen erilaistuminen ja pahanlaatuistuminen. Kolmi-ulotteiset soluviljelymallitOulun yliopiston tutkijakoulu; Oulun yliopisto, Lääketieteellinen tiedekunta; Biocenter OuluActa Univ. Oul. D 1452, 2018Oulun yliopisto, PL 8000, 90014 Oulun yliopisto

Tiivistelmä

Epiteelisolut ovat erikoistuneet toimimaan rajapintana elimen ja ympäristön välillä. Ihmistenyleisin syöpä on epiteelisoluista alkunsa saanut karsinooma. Tämän tutkimuksen tarkoituksenaoli ymmärtää Madin-Darby-koiran munuaisen solujen (MDCK) erilaistumista ja pahanlaatuistu-mista sekä analysoida sähköfysiologisia tekijöitä, jotka säätelevät näiden solujen kuljetustoimin-taa. Erityisenä kiinnostuksen kohteena oli erilaisten kasvuympäristöjen vertailu. Farmakologis-ten aineiden tai basaalisen, solunulkopuolisen nesteen koostumuksen vaikutusta MDCK-solu-jen, -kystan sekä luumenin kokoon tutkittiin valomikroskooppisten aikasarjojen avulla. Tuloksetosoittivat MDCK-solujen olevan kykeneviä sekä veden eritykseen että absorptioon, niin hyper-polarisoivassa kuin depolarisoivassakin ympäristössä. Basaalisen nesteen osmolaliteetin muutos-ta ei tarvittu. Nämä tulokset osoittavat MDCK-solujen olevan hyvä munuaisen tutkimuksenperusmalli. Seuraavaksi analysoitiin kaksi- ja kolmiulotteisten (2D ja 3D) viljely-ympäristöjenvaikutusta ei-transformoitujen MDCK-solujen ja lämpötilaherkkien ts-Src-transformoitujenMDCK-solujen geenien ilmentymiseen sekä yhden onkogeenin aktivoimisen aikaansaamia muu-toksia. Microarray-analyysi osoitti apoptoosin estäjän, surviviinin, ilmentymisen vähenemisen,kun kasvuympäristö vaihdettiin 2D-ympäristöstä 3D-ympäristöön. Koska surviviinin vähenemi-nen on normaali tapahtuma aikuisissa kudoksissa, voitiin todeta, että 3D-ympäristössä kasvatetutsolut ovat lähempänä luonnonmukaista olotilaa kuin 2D-ympäristössä kasvaneet. Src-onkogeenisai aikaan soluliitosten hajoamisen, mutta ei vähentänyt E-kadheriinin ilmentymistä. Tutkimuk-sen viimeinen osa keskittyi surviviinin ilmentymistä säätelevien tekijöiden analysoimiseen jasurviviinin merkitykseen solujen eloonjäämiselle. 3D-ympäristössä kasvaneet MDCK-soluteivät kärsineet apoptoosista edellyttäen, että solut pysyivät kosketuksissa soluväliaineeseen. Jossolut irtautuivat soluväliaineesta, ne päätyivät herkemmin apoptoosiin kuin surviviinia ilmentä-vät ts-Src MDCK-solut. Mikäli solujen väliset liitokset pakotettiin avautumaan, solut joutuivatapoptoosiin, vaikka ne olivat kosketuksissa soluväliaineeseen. Yhteenvetona nämä tuloksetkorostavat solujen kontaktien merkitystä: MDCK-solut tarvitsevat soluväliainekontakteja erilais-tumiseen ja solujen välisiä kontakteja välttyäkseen apoptoosilta.

Asiasanat: apoptoosi, elävien solujen mikroskopointi, MDCK-solut, polariteetti, Src-kinaasi, surviviini, transepiteelinen kuljetus

7

Acknowledgements

The work for this thesis was conducted in Biocenter Oulu and Cancer Research and

Translational Medicine Research Unit under the supervision of Docent Sinikka

Eskelinen.

I am tremendously thankful to Sinikka, who has basically raised me as a

scientist. You were the one who first sparked my interest in microscopy, and

introduced me to the intricate world of epithelial cells. Time and time again you

showed me that something that felt like a dead-end just needed to be looked at from

a different angle and you were never out of ideas for a publication. I truly admire

your creativity and immense flexibility in forming hypotheses that could be

answered with our limited resources and how you are very rarely without an answer.

During these years, you have always been there for me, supporting and guiding,

from the very first day of my one month orientation in the Eskelinen group to the

final touches of this thesis. Truly, could have not done it without you.

I want to express my gratitude to Professors Tuomo Karttunen, Markus

Mäkinen, Johanna Myllyharju and Taina Pihlajaniemi, together with all other

professors and group leaders of Biocenter Oulu for creating and maintaining such

a functional and supportive research environment.

My thanks to the current and past members of the Eskelinen group, Madhura,

Mira and Marja-Liisa, whom I have had the pleasure to work with.

I wish to thank Docents Varpu Marjomäki and Satu Kuure for sharing their

expertise and knowledge as the pre-examiners of this thesis. Your insightful and

valuable comments remarkably improved the thesis manuscript.

For support and guidance, I am very grateful to my thesis follow-up group

members Docent Lauri Eklund, Dr Irina Raykhel and especially the chairperson

Docent Tuomo Glumoff, who has also been mentoring me ever since I started my

Bachelor’s Degree at the University of Oulu.

I want to thank coordinators Ritva Saastamoinen, Pirkko Huhtala and Anthony

Heape as well as Teija Luoto, Anne Vainionpää and Irmeli Nykyri for their help

with practical matters.

I wish to thank all the research staff of the 4th floor of the Kieppi building for

maintaining what I believe to be a truly unique research environment where people

like to help each other. Especially, I want to thank laboratory technicians Riitta

Jokela and Jaana Träskelin. Even though we didn’t work in the same group, you

both were always more than happy to help me if I had any questions or problems.

8

I want to thank Antti Viklund for all his help with computers and for all the

enjoyable chats we have had during the years. Knowing I could always come to

talk to you whether I needed to get a network drive working, your opinion about a

computer game or some fresh information about an upcoming movie, was

invaluable to me.

My sincere thanks to Veli-Pekka Ronkainen. We have shared an office for years

and I have had more enjoyable conversations with you than I can count. In addition

to teaching me and helping with the usage of microscopes, your passion towards

light microscopy has truly inspired me.

I am grateful to my friends Emilia, Tiina, Joonas, Jani and Joona for supporting me

all these years. Being with you guys allows me to take my mind off of science or

other pressing concerns and to truly unwind. Special thanks to Lea for her

friendship and showing how this PhD thing is done.

Haluan kiittää vanhempiani, jotka opettivat minut arvostamaan tietoa ja

koulutusta sekä erityisesti äitiäni, joka oli tukenani, kun päätin ryhtyä

tavoittelemaan jotain niinkin hullua kuin tohtorin titteliä.

And finally, I want to thank my amazing wife Lisette. You know me better than

anyone, and I can talk to you about anything. You are an endless source of support

and encouragement. You know the hardships of being a scientist and don’t mind if

I want to talk about something else than work. I would be nothing without you.

This work was supported by grants from Munuaissäätiö and Scholarship Fund

of University of Oulu.

9

Abbreviations

[Cl-]i or [Cl-]o Concentration of chloride ions in the intracellular fluid

and extracellular fluid, respectively

[K+]i or [K+]o Concentration of potassium ions in the intracellular fluid

and extracellular fluid, respectively

[Na+]i or [Na+]o Concentration of sodium ions in the intracellular fluid

and extracellular fluid, respectively

1D One-dimensional, in this study 1D refers to cells grown

in suspension

2½D Two-and-a-half-dimensional, in this study 2½D refers to

cells grown on top of a layer of Matrigel

2D Two-dimensional, in this study 2D refers to cells grown

on uncoated plastic or glass surface

3D Three-dimensional, in this study 3D refers to cells

grown encased in Matrigel

A1/BFL-1 BCL2-related protein A1

AE Anion exchanger

AIF Apoptosis-inducing factor

AJ Adherens junction

AIIB2 β1 integrin blocking antibody

Akt Serine/threonine protein kinase Akt/protein kinase B

AMIS Apical membrane initiation site

Ano Anoctamin

APAF Apoptotic protease activating factor

API Active pharmacological ingredient

aPKC Atypical protein kinase C

AQP Aquaporin

ARM Armadillo

Arp Actin-related proteins

ARVCF Armadillo repeat protein deleted in velo-cardio-facial

syndrome

ATG Autophagy-related genes

ATP Adenosine triphosphate

BAD Bcl-2-associated death promoter

BAK Bcl-2 homologous antagonist/killer

BAX Bcl-2-associated X protein

10

Bcl B-cell lymphoma

BH BCL-2 homology domain

BH3-only Bcl-2 homology 3-only

BIR Baculovirus IAP repeat

BOK Bcl-2 related ovarian killer

BRCA Breast cancer tumour suppressor gene

CaCC Calcium-activated chloride channels, a group of channel

proteins

Caco Colon carcinoma cell line

CAF Cancer-associated fibroblasts

cAMP Cyclic adenosine monophosphate

CAS Crk- and Src-associated substrate

CBD Catenin-binding domain in cadherin family proteins

CDH Cadherin

CD Cluster of differentiation

CDC Cell division control protein

CF Cystic fibrosis

CFTR Cystic fibrosis transmembrane conductance regulator

CHIP28 Channel-forming integral membrane protein of 28 kDa

cIAP Cellular inhibitor of apoptosis protein

ClC Chloride channel family of proteins

CPC Chromosomal passenger complex

Crb Crumbs

Csk Carboxy-terminal Src kinase

c-Src Cellular-Src

DLBCL Diffuse large B-cell lymphoma

dlg Discs large

ECM Extra-cellular matrix

EGF Epidermal growth factor

EGFR Epidermal growth factor receptor

EGTA Ethylene glycol-bis(β-aminoethyl ether)-N,N,N',N'-

tetraacetic acid

EMEM Eagle’s modified essential medium

EMT Epithelial-mesenchymal transition

ENaC Epithelial sodium channel

ER Endoplasmic reticulum

ERBB2 Receptor tyrosine-protein kinase erbB-2

11

ERK Extracellular signal-regulated kinase

F Faraday’s constant

FADD Fas-associated death domain

FAK Focal adhesion kinase

FBS Foetal bovine serum

FGF Fibroblast growth factor

FITC Fluorescein isothiocyanate

FOXO Forkhead box protein O1 transcription factor

GATA A family of transcription factors with the ability to bind

the DNA sequence GATA

GDI Guanosine nucleotide dissociation inhibitor

GEF Guanine nucleotide exchange factor

GTPase Guanosine triphosphate hydrolysing enzyme

HBSS Hank’s balanced saline solution

HBXIP Hepatitis B X-interacting protein

HGF Hepatocyte growth factor, also known as scatter factor

HIF Hypoxia-inducible factor

IAP Inhibitor of apoptosis

ILP Inhibitor of apoptosis protein -like protein

INCENP Inner centromere protein

JMD Juxtamembrane domain in cadherin family proteins

KCNQ/KCNE Potassium voltage-gated channel subfamily Q/E

complex

kDa Kilodalton

KLF Krüppel-like family of transcription factors

Lats Large tumour suppressor homolog

LEF Lymphoid enhancer-binding factor

lgl Lethal giant larvae

LRRC8 Leucine-rich repeat-containing protein 8

MAPK Mitogen-activated protein kinase

MARCH8 Membrane-associated RING-CH 8, also known as c-

MIR

MCL-1 Induced myeloid leukemia cell differentiation protein

MDCK Madin-Darby canine kidney cell line

MET Mesenchymal-epithelial transition

ML-IAP Melanoma inhibitor or apoptosis protein

MMP Matrix metalloprotease

12

MOM Mitochondrial outer membrane

MOMP Mitochondrial outer membrane permeabilization

mRNA Messenger-RNA

Mst Mammalian STE20-like protein kinase

mTOR Mechanistic target of rapamycin

Myc Myelocytomatosis

NAC N-acetyl cysteine

NADPH Nicotinamide adenine dinucleotide phosphate

NAIP NLR family apoptosis inhibitory protein

NBC1 Na+-HCO3- cotransporter 1

NBL-2 The Naval Biosciences Laboratory 2, original name for

the MDCK cell line

NF-κB Nuclear factor-kappa B

NHE1 Na+-H+ exchanger 1

NHEK Normal human epidermal keratinocytes

NKCC Na+-K+-2Cl- cotransporter

NOX NADPH oxidase

PALS1 Protein associated with Lin Seven 1

PAR Partitioning-defective

PATJ PALS1-associated tight junction protein

PCD Programmed cell death

pCl Membrane permeability for chloride ions

PI3K Phosphatidylinositol-4,5-bisphosphate 3-kinase

PIP2 Phosphatidylinositol 4,5 bisphosphate

PIP3 Phosphatidylinositol 3,4,5 trisphosphate

pK Membrane permeability for potassium ions

PKA Protein kinase A

PKC Protein kinase C

PKD Polycystic kidney disease

PKR Protein kinase R

PL Piperlongumine

pNa Membrane permeability for sodium ions

pp2 Amino-5-(4-chloro-phenyl)-7-(t-butyl)pyrazolo[3,4-

d]pyrimidine

PTEN Phosphatase and tensin homologue deleted on

chromosome 10

PTP Protein-tyrosine phosphatase

13

R Universal gas constant

rab Ras genes from rat brain

Rac Ras-related C3 botulinum toxin substrate

Ras Rat sarcoma, a small GTPase with oncogenic properties

rho Ras homolog

RING-CH Really interesting new gene, which has a cys residue in

the fourth position and a His in the fifth

ROI Region of interest

ROS Reactive oxygen species

rr1 Antibody specific to the extracellular domain of canine

E-cadherin

RSV Rous sarcoma virus

RTK Receptor tyrosine kinases

RT-PCR Reverse transcription polymerase chain reaction

RVD Regulatory volume decrease

SCRIB Scribble

SH Src homology domain

siRNA Small interfering RNA

SIRT NAD-dependent deacetylase sirtuin

SLUG Snail 2

SMAC Second mitochondria-derived activator of caspases

SNAIL Snail family transcriptional repressor 1

Sp Specificity protein

Src A non-receptor tyrosine kinase, name is derived from

sarcoma

STAT Signal transducer and activator of transcription

STE20 Sterile 20

T Temperature in Kelvin

t½ Half-life

TAZ Transcriptional co-activator with PDZ-binding motif

TCF T-cell factor

TEF Transcription enhancer factor

TER Transepithelial electric resistance

TGF-β Transforming growth factor-β

TJ Tight junction

TMACl Tetramethyl ammonium chloride

14

TMA-DPH 1-(4-Trimethylammoniumphenyl)-6-Phenyl-1,3,5-

Hexatriene p-Toluenesulfonate

TMEM16A Transmembrane member 16A

TNF Tumour necrosis factor

TNFR Tumour necrosis factor receptor

TRADD TNFR-associated death domain

ts-Src MDCK Temperature-sensitive Src MDCK cell line

TWIST A basic Helix-Loop-Helix transcription factor

VAC Vacuolar apical compartment

VEGF Vascular endothelial growth factor

Vfinal Volume at the end of the experiment

Vinit Volume at the start of the experiment

Vm Membrane potential

VRAC Volume-regulated anion channel

VSORCC Volume-sensitive outwardly rectifying Cl- conductance

v-Src Viral-Src

Wnt Wingless-related integration site

XIAP X-linked inhibitor of apoptosis protein

YAP Yes-associated protein

ZO Zonula occludens

15

List of original publications

This thesis is based on the following publications, which are referred to throughout

the text by their Roman numerals:

I Capra JP, Eskelinen SM (2017) MDCK cells are capable of water secretion and reabsorption in response to changes in the ionic environment. Can J Physiol Pharmacol 95(1): 72-83.

II Töyli M, Rosberg-Kulha L, Capra J, Vuoristo J, Eskelinen S (2010) Different responses in transformation of MDCK cells in 2D and 3D culture by v-Src as revealed by microarray techniques, RT-PCR and functional assays. Lab Invest 90(6):915-928.

III Capra JP, Eskelinen SM (2017) Correlation between E-cadherin interactions, survivin expression, and apoptosis in MDCK and ts-Src MDCK cell culture models. Lab Invest 97(12):1453-1470.

16

17

Table of contents

Abstract

Tiivistelmä

Acknowledgements 7 

Abbreviations 9 

List of original publications 15 

Table of contents 17 

1  Introduction 21 

2  Review of the literature 23 

2.1  Epithelium ............................................................................................... 23 

2.1.1  Membrane transport properties ..................................................... 23 

2.1.2  Active pharmaceutical ingredients (API) affecting plasma

membrane transport mechanisms ................................................. 25 

2.1.3  Absorptive and secretory epithelia ............................................... 26 

2.2  Kidney structure and function ................................................................. 27 

2.2.1  Water absorption and reabsorption in the nephrons ...................... 29 

2.2.2  Aquaporin protein family and their function in kidney ................ 29 

2.2.3  Chloride channels are important for water transport .................... 30 

2.2.4  Monovalent cations ...................................................................... 32 

2.3  Madin-Darby canine kidney (MDCK) cells ............................................ 35 

2.4  Cellular junctions and their constituents ................................................. 36 

2.4.1  Tight junctions .............................................................................. 37 

2.4.2  Adherens junctions and desmosomes ........................................... 37 

2.4.3  Cadherin superfamily ................................................................... 38 

2.4.4  Catenins ........................................................................................ 39 

2.4.5  Cadherin signalling ....................................................................... 41 

2.4.6  Cadherin recycling ....................................................................... 43 

2.5  Epithelial cell polarity ............................................................................. 43 

2.5.1  Lumen formation and maintenance .............................................. 46 

2.6  Malignant transformation ........................................................................ 47 

2.6.1  Oncogenes and tumour suppressors ............................................. 48 

2.6.2  Epithelial-mesenchymal transition ............................................... 49 

2.7  Programmed cell death has many forms ................................................. 50 

2.7.1  Inhibition of apoptosis .................................................................. 53 

2.8  Src was the first identified proto-oncogene ............................................. 56 

2.8.1  Temperature-sensitive Src MDCK (ts-Src MDCK) cells ............. 58 

18

2.9  Reactive oxygen species ......................................................................... 59 

2.9.1  Piperlongumine is a small molecule that selectively targets

cancer cells ................................................................................... 60 

3  Aims of the study 63 

4  Materials and methods 65 

4.1  Cell lines and experimental procedures ................................................... 65 

4.1.1  Use of MDCK cell lines ............................................................... 66 

4.1.2  Image collection and analysis ....................................................... 67 

5  Results 69 

5.1  The effects of ionic environment on water secretion and re-

absorption (I) ........................................................................................... 69 

5.1.1  Basal fluid lacking in monovalent cations induces water

influx into the lumen (I) ............................................................... 69 

5.1.2  Basal fluid lacking in chloride ions induces water re-

absorption (I) ................................................................................ 70 

5.1.3  Sodium gluconate does not affect the tight junctions (I) .............. 70 

5.1.4  Complete depolarization of the cells causes cell swelling

(I) .................................................................................................. 70 

5.1.5  Inhibiting or activating chloride channels leads to changes

in lumen, cyst and cell size (I) ...................................................... 71 

5.2  Src-induced events in MDCK cells (II and III) ....................................... 73 

5.2.1  Phenotypes of untransformed MDCK and ts-Src

transformed MDCK cells in different culture systems

analysed using confocal fluorescence microscopy (II and

III) ................................................................................................. 74 

5.2.2  Expression of cadherins in MDCK and ts-Src MDCK cells

in different culture environments (II) ........................................... 75 

5.2.3  Microarray analysis revealed the differences in gene

expressions of cells cultivated in different environments

(II) ................................................................................................. 76 

5.2.4  Src-induced functional changes in the mitochondrial

activity and E-cadherin endocytosis in ts-MDCK cells (II) ......... 77 

5.2.5  Relation of survivin expression to the cell phenotype and

occurrence of apoptosis in different conditions (II and III) .......... 79 

5.2.6  Ts-Src MDCK cells are more susceptible to ROS than

untransformed MDCK cells, and can be rescued by

antioxidant supplementation (III) ................................................. 81 

19

5.2.7  Summary ...................................................................................... 82 

6  Discussion 85 

6.1  Plasticity of epithelial cells can be seen in the rapid response of

MDCK cells to changes in ionic environment ........................................ 85 

6.2  Transformation of MDCK cells by v-Src causes changes in gene

expression, cell phenotype and behaviour ............................................... 87 

6.3  The importance of E-cadherin interactions and survivin

expression on cell fate in untransformed and ts-Src MDCK cells .......... 89 

6.4  Ts-Src MDCK cells as a model for malignant transformation of

epithelial cells ......................................................................................... 91 

6.5  Comparison of the effects of 2D and 3D growth environment on

cell behaviour .......................................................................................... 93 

6.6  Use of time-lapse imaging of live cells as a tool for monitoring

cellular processes .................................................................................... 95 

7  Summary and conclusions 97 

References 99 

Original articles 113 

20

21

1 Introduction

A cell and its functions are baffling in their complexity. Even more incredible is the

intricacy of the cooperation that cells in a multicellular organism are capable of.

For the organism to function, the cells need to be at the right place at the right time,

and be able to discern when it is their turn to be active. Epithelial cells are a special

cell type that lines many organs, like intestine, liver and salivary glands, and the

whole organism as skin. They act as a barrier, compartmentalizing organs and

different parts of the body and enabling selective interactions between them,

forming the basis for communication throughout the whole organism. Epithelial

cells also have crucial active roles, especially in secretion and absorption. As a part

of the barrier between compartments, epithelial cells are required to polarize by

having two distinguishable membrane domains, the apical and the basal. The apical

side usually faces the fluid-filled opening, the lumen, in lungs, glands, kidney

tubules etc., whereas basal membrane faces the extracellular matrix in tissues.

The kidneys are an example of an organ, where epithelial cells have an active

role. Kidneys are a pair of vital organs that regulate the electrolyte balance in the

blood, maintain pH homeostasis and remove the waste products from the blood.

Kidneys participate in maintaining the acid-base and fluid balance, reabsorption of

water, glucose and amino acids and in regulating blood pressure and filtering the

blood to produce urine. The tubules of the kidney are comprised of epithelial cells

that specialize in secretion or reabsorption of different solutes depending on the

part of kidney they are located in. The kidney epithelial cells mainly transport ions

and water. The transport direction is dictated by the surrounding environment and

hormonal control.

As a barrier, epithelial cells are defined by their ability to form and maintain

cell-cell and cell-matrix junctions. The cells are so reliant on these junctions that

they easily suffer apoptosis if impeded from forming them. The junctional proteins,

like cadherins in epithelial cells, are also important for cell signalling, and

participate in several pathways that regulate cell survival, proliferation,

differentiation and other functions, and well-formed cell-cell junctions are a

hallmark of mature, fully differentiated epithelial cells.

Carcinomas are cancers caused by epithelial cells, and are the most common

cancer type. Epithelial cells that have transformed into cancer cells have overcome

their susceptibility to apoptosis, as well as the regulation of proliferation, among

other things, and usually acquire the phenotype of mesenchymal cells. Survivin is

22

a protein that is almost universally overexpressed in cancers, but not in healthy

adult tissues. Survivin protects cells from apoptosis and aids in proliferation.

The current study introduces a modern way to use high-resolution microscopy

to monitor epithelial cell differentiation and function. Specific attention is given to

the cell cyst behaviour in 3D extracellular matrix, and effects of drugs or basal

extracellular fluid on cell, cyst and lumen volumes. The results give a valuable tool

for testing drugs or transport protein inhibitors. It also shows that MDCK cells are

capable of both secretion and absorption of chloride ions and water depending on

the basal extracellular fluid composition. In addition, this work shows that a

downregulation of survivin occurs when untransformed MDCK cells are

transferred from 2D environment to 3D, and that MDCK cells can survive without

survivin as long as their cell-cell junctions are intact. Furthermore, it highlights the

importance of the growth environment to the cell phenotype and behaviour, and

that cells grown in 3D are closer to their in vivo state than cells grown in 2D. This

difference can be crucial when using cell cultures for identification of new drugs,

for example. 3D cultures have many advantages when evaluating the suitability of

potential pharmaceutically active molecules for in vivo use. The early events of the

transformation process induced by v-Src oncogene were monitored and especially

the fate of E-cadherin and the cell response to ROS were in focus of this study.

23

2 Review of the literature

2.1 Epithelium

In the bodies of complex multicellular organisms, the organs form individual,

functional entities. Epithelium forms the outermost layer of many organs. It forms

a tight, cohesive sheet that acts as an interface between the organ and the

environment, and restricts the movement of solutes and water through it.

Epithelium is much more than just an inactive barrier, however. It regulates

permeability, transport, endocytosis and exocytosis. It participates both in excretion

and absorption. Epithelium transports water, ions, oxygen, essential nutrients and

messenger molecules into the organ space and excretes waste and products of the

organ. This selective permeability allows the maintaining of constant internal

environment of the cell and tissue.

For functional epithelium, epithelial cells must be highly organized, with

junctions linking the cells together to form a barrier in the first place, and

specialized membrane regions complete with different transport proteins. These

structural and functional differences divide the cell membrane into three parts:

apical domain (the cell membrane facing the lumen), lateral domain (the membrane

facing the neighbouring cell) and basal domain (the membrane facing basal lamina);

together, they establish epithelial polarity.

Cells can link to each other using multiple different junction types: tight

junctions, adherens junctions, desmosomes and gap junctions. These cell junctions

are all highly specialized for certain functions, and an epithelial cell can express

multiple junction types simultaneously on the same membrane domain (Lodish et

al. 2000).

2.1.1 Membrane transport properties

The ion composition of cells differs greatly from the extracellular environment.

Separation of the cells from their environment is achieved by the plasma membrane

that selectively allows some molecules to pass through while blocking others from

entering. The plasma membrane is naturally permeable to gases, like O2 and CO2,

but only to a small group of molecules. These molecules must be small, uncharged

and polar, such as ethanol and urea. These chemicals can permeate the plasma

membrane via passive diffusion. Other molecules need the help of membrane

24

proteins. Proteins can also aid in the movement of water and urea, molecules that

are able to cross the plasma membrane passively.

Different cells have different requirements for their inner ionic composition,

but basically all cells have an inside pH of around 7.2, and the K+ ion concentration

is multiple times higher inside the cell than outside it, while Na+ ion concentrations

are the opposite. Cytoplasmic concentration of Cl- ions upholds the

electroneutrality of the cell interior and is lower than that of the extracellular

concentration since many proteins, the levels of which are much higher inside the

cell than outside, have a negative net charge. The imbalance of ion concentration

creates an electric potential between the two sides of the cell membrane, called the

membrane potential, and it has an essential role in many biological processes.

Taken together, these charge differences lead to a negative membrane potential,

which in animal cells is usually about -70 mV (cytoplasm is negative in respect to

extracellular environment) (Wills et al. 1996, Lodish et al. 2000). The membrane

potential of the cell can be calculated using the Goldman equation (1)

1

In the above equation, Vm is the membrane potential, R is the universal gas constant,

T is temperature in Kelvin, F is Faraday’s constant, p is the membrane permeability

for K+, Na+ and Cl- and [K+], [Na+] and [Cl-] are the concentrations of potassium,

sodium and chloride, with subscript “o” denoting extracellular fluid and subscript

“i” intracellular fluid.

The ability to regulate cell volume is crucially important to almost all

vertebrate cells. Cell volume affects the concentrations of cellular constituents and

can affect cell signalling and vesicle traffic, among other things, and extreme

swelling can cause the cell to burst (Jentsch 2016). By adjusting the intracellular

concentrations of osmolytes, which, in turn, induce water flux along the osmotic

gradient, cells can regulate their volume. Cell membrane channel proteins play a

pivotal role in this process by regulating ion transport through the membranes

(Jentsch 2016). Epithelial cells are polarized, i.e. they have specialized apical and

basal domains, which often have different representation of channel proteins. These

channel proteins regulate the transepithelial transport which is crucial, e.g. in

kidney function. The importance of ion transport to tissue homeostasis, water

balance and lumen volume has been known for a rather long time, as shown by the

effect of membrane transport inhibitors, like ouabain and amiloride, on the

25

functionality of organs in vivo and on cyst growth and lumen volume in vitro

(Grantham et al. 1989, Mangoo-Karim et al. 1989).

2.1.2 Active pharmaceutical ingredients (API) affecting plasma

membrane transport mechanisms

Several molecules are able to affect the transport properties of the plasma

membrane by blocking channel proteins, creating new channels or via some other

mechanisms. Nigericin is a naturally occurring antibiotic from Streptomyces

hygroscopicus bacteria. In the past, it was used against gram-positive bacteria.

Nigericin is an ionophore that can attach to cell membranes and act as an antiporter

for H+ and K+. When introduced to the cells, nigericin allows swift exchange of

monovalent cations through the cell membrane according to their concentration

gradient, leading to depolarization of the cell. The end result is swelling of the cells.

This is caused by the Donnan effect, i.e. the large charged particles, proteins, unable

to pass through the cell membrane create an asymmetric distribution of charges

across the membranes. When nigericin is used to abolish the concentration

differences of H+ and K+, the electrical gradient vanishes, but the difference in

osmolality still exists due to the charged particles that are not able to pass freely

through the membrane. The osmotic gradient causes water influx into the cells and

leads to cell swelling (Kay 2017 Yakisich et al. 2017).

There are plenty of APIs which activate or inhibit existing channel proteins.

Lubiprostone, for example, is a bicyclic fatty acid derived from prostaglandin E1.

It functions as a chloride channel type 2 (ClC-2) activator, and is approved for use

in the treatment of chronic constipation and constipation-dominant irritable bowel

syndrome. Presence of lubiprostone opens ClC-2 to chloride transport through the

apical side. This, in turn, causes water influx into the lumen (Chan & Mashimo

2013).

Forskolin is naturally produced by the Indian Coleus plant (Plectranthus

barbatus) and acts as an activator of adenylyl cyclase, which converts ATP into

cyclic AMP (cAMP). Introduction of forskolin to the cells increases the

intracellular levels of cAMP. Cyclic AMP is a secondary messenger whose

concentration is increased by the signals from extracellular molecules, like

hormones, and subsequently activates cAMP-sensitive pathways that can have a

multitude of effects. For plasma membrane transport, the most important is the

activation of protein kinase A (PKA), which in turn phosphorylates the apical

chloride channel called cystic fibrosis transmembrane conductance regulator

26

(CFTR). This activates CFTR allowing the flow of Cl- into the lumen (Torres &

Harris 2014).

Another plant-derived compound that can affect the plasma membrane

transport is ouabain, a toxin produced by plants native to eastern Africa,

Acokanthera schimperi and Strophanthus gratus. Ouabain or ouabain-like

substance might also be produced by mammalian adrenal glands to act as a

hormone. Ouabain is a potent inhibitor of Na+-K+-ATPase, an ATP-dependent

channel protein that exchanges intracellular Na+ into extracellular K+. Through

Na+-K+-ATPase, ouabain has an effect on cells and tissues far beyond simple

manipulation of ion balance (Venugopal & Blanco 2017).

Amiloride is a diuretic drug used to treat high-blood pressure and oedema. It

acts as a direct inhibitor of the epithelial sodium channel (ENaC), which is located

on the apical membrane of the epithelial cells of the distal tubule of the kidney. This

channel can also be found in the lung, colon and several gland epithelial cells.

ENaC is permeable to Na+ and is the major facilitator of Na+ reabsorption.

Amiloride promotes sodium and water depletion from the body while preserving

potassium levels. Amiloride also inhibits Na+/H+ antiporter, reducing the secretion

of H+ by the epithelial cells of the proximal tubule (Loffing & Kaissling 2003).

2.1.3 Absorptive and secretory epithelia

Epithelial cells can be either absorptive or secretory. Secretory functions are

performed by various glands (sweat, saliva, mammary etc.), the kidney tubules and

the mammalian lung alveoli that produce surfactant and epithelial lining fluid.

Absorptive epithelia, on the other hand, include epithelia of the small and large

intestine and the proximal tubule cells of the kidney that reabsorb primary urine.

Kidney has both absorptive and secretory roles in urine volume and consistency

regulation. Proximal tubule cells are mainly absorptive by moving e.g. water and

ions out of the lumen of the tubule into the surrounding capillaries, while distal

tubule cells secrete in the direction of the lumen. However, both proximal and distal

tubule cells are capable of performing both tasks if needed due to reasons such as

renal failure. In case the proximal tubules are only capable of absorption, a renal

failure where the glomerular filtration is completely ceased would lead to a collapse

of the tubular lumen as the hydrostatic pressure caused by the water would dissipate

as the primary urine is absorbed. Isolated proximal tubules have been shown to be

able to keep up tubular lumina and move urine through the nephron at a drastically

reduced rate compared to in vivo situation with intact glomeruli. This is useful for

27

mammals suffering from acute renal failure or hypertonic dehydration, where

severe lack of water puts the animal at risk of suffering from hypernatraemia

(Grantham & Wallace 2002). The basic epithelial transport routes through the cell

are presented in Fig. 1.

Fig. 1. The basic epithelial transport routes.

Transepithelial transport can occur via paracellular (through the junctions between

the cells) and transcellular (through the cell) route. Absorption occurs when

transport is from the luminal space into the cell/basal side, while secretion is

transport into the lumen.

2.2 Kidney structure and function

Kidneys are a pair of vital organs located at the rear of the abdominal cavity in

humans. They regulate the electrolyte balance in the blood, maintain pH

28

homeostasis and remove the excess organic molecules (often referred to as waste

products) from the blood. These regulatory functions affect such central

physiological phenomena as maintaining the acid-base and fluid balance,

reabsorption of water, glucose and amino acids and regulation of blood pressure

and filter the blood to produce urine (Hiltunen et al. 2003, Standring et al. 2008).

Kidneys have a bean-shaped structure and consist of cortex and medulla. The

basic functional units of the kidneys are the nephrons, which are connected to the

collecting duct system. Each kidney has about 1,000,000 nephrons, each of which

starts from the cortex and spans through the medulla. On the cortex side, each

nephron has a complex network of capillaries called a glomerulus, where the initial

stages of urine formation take place. The high blood pressure in the capillaries of

the glomerulus filtrates the blood plasma into the surrounding Bowman’s capsule.

The proximal tubule is attached to the Bowman’s capsule and collects the filtrate.

The filtrate is subsequently absorbed back by the epithelial cells of the tubule, and

it moves through the cells into the lateral space between the tubule and capillaries.

Most of the filtrate is then reabsorbed (either passively or actively) through the

endothelial cells back into the surrounding capillaries. The rest of the filtrate is

passed through the loop of Henle, where the concentrated filtrate, urine, is diluted

into a concentration determined by the blood osmolality. Urine will attain its final

concentration in the distal tubule before being taken to the ureter via the collecting

duct system (Hiltunen et al. 2003, Standring et al. 2008). A schematic drawing of

kidney and nephron structure is presented in Fig. 2.

Fig. 2. A schematic drawing of kidney and nephron structure.

29

Aquatic organisms developed the first kidneys, which, due to the essentially

unlimited access to both water and salt, evolved to mainly discharge chemicals that

could be harmful to the organism and could not be disposed by simple diffusion

through the cell membranes (Grantham & Wallace 2002). Once terrestrial animals

began to gain ground, kidneys adopted new roles in water distribution and

preservation of the “internal environment” that must now be kept completely

separate from the new, comparatively dry external environment. Since then,

kidneys have assumed full responsibility for regulating body fluid and salt balance

(Grantham & Wallace 2002).

2.2.1 Water absorption and reabsorption in the nephrons

Water is prerequisite for the function of all cells, and makes up two-thirds of the

body weight of an adult human (Noda et al. 2010). For land-dwelling multicellular

organisms who do not live constantly surrounded by vast amounts of liquids, water

intake and distribution is one of the most crucial and tightly controlled processes.

The role of the kidneys in the water homeostasis is indispensable. An evolutionary

quirk of the glomerulus is the ability to absorb enormous amounts of water from

the blood plasma. This was beneficial for aquatic organisms, but on dry land it

would quickly lead to dehydration if left unchecked. Instead of reducing the amount

of filtrate produced, kidneys developed the means to reabsorb most of the water

back into the blood stream in the proximal tubule (Grantham & Wallace 2002).

Water is able to permeate passively through cell membranes. However, the

permeability of cell membranes to water can be greatly increased by the

introduction of aquaporins, a family of small integral membrane proteins that form

pores to the cell membrane, allowing large amounts of water to pass through

expeditiously (Noda et al. 2010).

2.2.2 Aquaporin protein family and their function in kidney

For a long time, the existence of channel proteins specialized in water transport had

been suspected, until Preston et al. (1992) discovered, by serendipity, a 28 kDa

protein predicted to bear an integral membrane protein with six membrane-

spanning domains. Originally named channel-forming integral protein of 28kDa

(CHIP28), this novel protein was abundant in red blood cells and kidney tubules.

When expressed in Xenopus laevis oocytes, it increased the osmotic water

permeability to levels where cells ruptured from the uncontrolled water influx into

30

the cells (Preston et al. 1992). This protein was later renamed aquaporin 1 (AQP1).

All aquaporin family members have an hourglass shape with a pore in the middle.

Water and some other, selected, molecules pass through the pore one by one, but

the pore is impermeable to charged molecules, maintaining the electrochemical

potential difference of the cell membrane (Noda et al. 2010).

To date, 13 members of the aquaporin family have been discovered (AQP0-

AQP12), and seven of them (AQP1-4, AQP6, 7 and 11) are expressed in human

kidneys (Noda et al. 2010). Cell membranes of epithelial cells of proximal tubules

and descending thin limbs of Henle express aquaporin 1, and it is critical in the

water reabsorption after its initial filtration through the glomeruli (Nielsen et al.

1993a). Water molecules move through the central pore of aquaporin 1 one at a

time, while the complex structure of the protein prevents the entrance of solutes

and is not regulated by vasopressin (Noda et al. 2010). Aquaporin 2 is expressed in

the principal cells of the collecting duct (Nielsen et al. 1993b). The antidiuretic

hormone, vasopressin, causes AQP2 to translocate from intracellular vesicles to

apical plasma membrane, and to transport water from urine into the bloodstream in

response to dehydration (reviewed in Noda et al. 2010). Concomitantly, water

needs to be able to leave the cells as well. For this, aquaporins 3 and 4 are expressed

on the basolateral membrane of the collecting duct and provide such an exit

pathway (Noda et al. 2010). Aquaporin 6 has been suggested to have a function in

promoting urinary acid secretion, as it has low water permeability and acts

primarily as an anion transporter, and is expressed in acid-secreting type-A

intercalated cells (Yasui et al. 1999, Ohshiro et al. 2001). Aquaporin 7 is expressed

in the brush border of proximal straight tubules and works in unison with AQP1 in

the absorption of water from the filtrate, but is also the main player in glycerol

reabsorption (Nejsum et al. 2000, Sohara et al. 2005). The final member of the

aquaporin family expressed in the kidneys is aquaporin 11. One of the newest

members in the family, APQ11 is structurally distinct from other aquaporins and,

instead of cell membrane, localizes to ER. Its function is poorly understood, but

mice with knockout Aqp11 die before weaning due to polycystic kidney disease

(Morishita et al. 2005).

2.2.3 Chloride channels are important for water transport

There are several types of chloride transporters in multicellular organisms. ClCs

are a family of voltage-gated, pH-sensitive epithelial chloride channels that might

have a role in regulating the cell size and are abundant in the kidneys. ClC-Ka and

31

ClC-Kb are expressed in the thin and thick ascending limp Henle’s loop of the

kidneys, respectively, where they participate in the reabsorption of chloride. Both

ClC-Ka and ClC-Kb are rather voltage-independent when compared to other ClCs

(Verkmann & Galietta 2009). ClC-2 is ubiquitously expressed and is activated by

cell swelling (Gründer et al. 1992), and is pharmacologically interesting as a way

to bypass non-functional CFTR in the respiratory epithelia of cystic fibrosis-

patients. ClC-2 has also been shown to be activated by hyperpolarization of the cell

(Norimatsu et al. 2012).

Calcium-activated chloride channels (CaCCs) are passive transport chloride

channels activated by increase in intracellular Ca2+ concentration (Verkmann &

Galietta 2009). One of the confirmed members of CaCCs is anoctamin 1 (Ano1;

originally known as TMEM16A), cloned in 2008 by three independent groups

(Caputo et al. 2008, Schroeder et al. 2008, Yang et al. 2008a). Ano1 is strongly

expressed by the epithelial cells of the proximal tubule, where it regulates protein

reabsorption and acid secretion, but it also contributes to the growth of cysts in

polycystic kidney disease (PKD; Faria et al. 2014, Buchholz et al. 2014).

As the name implies, the cystic fibrosis transmembrane conductance regulator

(CFTR) channel was first described as an integral player in cystic fibrosis pathology.

CFTR is an apical membrane anion channel, with preference to Cl- and HCO3-, but

it is also capable of secreting smaller, but physiologically significant, amounts of

other anions, like glutathione and thiocyanate (Linsdell et al. 1997, Conner et al.

2007). CFTR is abundantly expressed in the apical membranes of the kidney

epithelial cells, but the exact function of this channel protein in reabsorbing

epithelia is not clear. CFTR knock-out mice have normal renal functions, thus

indicating that CFTR functions can be taken over by other channel proteins. CFTR

can also have roles in regulating the conductance of other channel proteins by

interacting with them, and in the acidification of endosomal vesicles in the

cytoplasm of renal epithelial cells (Souza-Menezes et al. 2014). A mutation in

CFTR that affects its ability to localize to the apical membranes or its function as a

chloride channel manifests as CF, a disease whose most striking feature is the filling

of airways with thick mucosa. Excessive CFTR activity, on the other hand,

contributes to production of watery stool in cholera and traveller’s diarrhoea and

the formation of cysts in PKD (Sullivan et al. 1998). Several pathways regulate the

CFTR activity. CFTR is activated by PKA in a cAMP dependent manner (Huang

et al. 2000). CFTR is not the sole chloride transporter in the apical membranes of

epithelial cells; however, it is the most essential (Frizzel & Hanrahan 2012). Due

32

to the important, but contrasting, roles of CFTR in CF and PKD, several direct

and/or specific inhibitors and activators are commercially available.

Volume-regulated anion channels (VRACs) are key players in cell volume

regulation and are found in every cell type. VRACs transport Cl- and organic

osmolytes like amino acids and taurine, creating osmotic gradients that drive water

movement across the cell membrane, usually out of the cell into the extra-cellular

space in an effort to reduce cell volume (Jentsch 2016). Efflux of Cl- also causes

depolarization of the cell. VRACs are a group of channel proteins that are always

heteromers composed of a leucine-rich repeat containing 8A (LRRC8A) and either

LRRC8B, C, D or E. Knocking down all LRRC8 heteromers abolishes the short-

term cell volume increase (causing regulatory volume decrease or RVD),

demonstrating the importance of VRACs to cell volume regulation (Qiu et al. 2014).

In kidney, VRACs have a dominant role in RVD and might protect the kidney

epithelial cells from osmotic harm. LRRC8A knock-out in mice caused kidney

abnormalities and tubular degeneration (Platt et al. 2017).

2.2.4 Monovalent cations

In the cell, the sum of concentrations of monovalent cations is much higher than

the sum of monovalent anions. This is due to the high level of polyvalent anions,

i.e. proteins, and leads to the cell interior being electronegative in comparison to

the extracellular environment. Cells need this electric potential difference to

survive, as they use it for transportation of ions and molecules across the membrane,

ATP creation and other enzymatic functions, but also for signalling (Wills et al.

1996). The gradient is also important for the main kidney functions. As with Cl-,

several ion transporters, pumps and channel proteins participate in moving the

monovalent cations across the cell membrane, both along and against their

concentration gradient. The most important ones are Na+/K+ -ATPase, Na+-K+-2Cl-

cotransporter (NKCC), passive K+ channels and ENaC (Wills et al. 1996).

ENaC is a constitutively active Na+ channel located on the apical membrane

and is the main player in Na+ reabsorption in kidney nephrons and collecting ducts,

but it is also expressed in colon, lung and sweat glands. It has a crucial role in

maintaining salt and water homeostasis, as well as blood pressure. ENaC is highly

sensitive to amiloride (Hanukoglu & Hanukoglu 2016).

Na+/K+ -ATPase is a ubiquitous transport pump found in all epithelial cells, and

is usually located on the basal membrane. The pump uses energy from the

hydroxylation of ATP to move three Na+ out of the cell, while importing two K+

33

into the cell, thus leading to net export of a positive charge per expenditure of ATP.

Na+/K+ -ATPase maintains the resting potential of the cell by constantly sustaining

the concentration gradients of Na+ and K+, and regulates cell volume. It also

functions as a signal transducer that can activate Src, PI3K, ERK1/2, protein kinase

C (PKC), and inhibition of Na+/K+ -ATPase results in reactive oxygen species (ROS)

production (Xie et al. 1999, Clausen et al. 2017).

NKCC is a carrier protein important for Cl- transport in epithelial cells. NKCC1

localizes to basal membranes, while NKCC2, mostly abundantly expressed by the

kidney epithelial cells, localizes to apical membranes. NKCC2 transports one Na+,

one K+ and two Cl- ions from the renal luminal space into the cytosol, resulting in

electroneutral translocation. NKCC2 is the most important Cl- transporter in the

reabsorption of chloride in the kidney and plays a vital role in regulation of water

balance and in salt preservation (Ares et al. 2011).

Passive K+ channels are highly selective for potassium and allow its free flow

along the concentration gradient. There are four classes of K+ channels, categorized

by the method by which they are activated: calcium-activated, inwardly rectifying,

tandem pore and voltage-gated potassium channels. Their activity maintains resting

potential in the cell by returning K+ brought in by Na+-K+ -ATPase and NKCC, and

they have an important role in cell volume regulation and regulating the secretion

of hormones, like insulin (Lawson & McKay 2006). Some common ion transport

pathways and participating channel proteins are presented in Fig. 3.

34

Fig. 3. Common ion transport pathways in kidney epithelia and participating channel

proteins. Several different transport proteins participate in epithelial ion transport. The

most important ions are chloride and the monovalent cations sodium and potassium.

cAMP activates CFTR and other Cl- channels e.g. via activation of PKA. Some of the

most essential ion transporters are presented. AE; anion exchanger 2, AQP; aquaporin,

cAMP; cyclic adenosine monophosphate, CFTR; cystic fibrosis transmembrane

conductance regulator, ClC-2; chloride channel 2, ENaC; epithelial sodium channel,

KCNQ/KCNE; voltage-gated potassium channels, NBC1; Na+-HCO3- cotransporter 1,

NHE1; Na+-H+-exchanger 1, NKCC1; Na+-K+-2Cl--cotransporter 1, PKA; protein kinase A,

VSORCC; volume-sensitive outwardly rectifying Cl- conductance (Modified from Frizzel

& Hanrahan 2012).

35

2.3 Madin-Darby canine kidney (MDCK) cells

As discussed before, the kidney is a complex organ with anatomically different

parts and a multitude of crucial functions. Its vital importance to the survival of the

organism created a need for experimental models that can be used to study the

properties of the kidney. For a cell culture model, the cell line must exhibit proper

cell-cell junctions and secretion and reabsorption properties similar to the

mammalian kidney. Madin-Darby canine kidney (MDCK) cells are widely used for

studying kidney or epithelial function, apico-basal polarity and cell junctions. They

are valued for ease of culture, and their ability to polarize both in 2D and 3D

environments. In 3D they form spherical cysts with clearly defined apical and basal

membranes and a hollow space, lumen, in the middle (O’Brien et al. 2002, Dukes

et al. 2011, Datta et al. 2011). Another advantage of MDCK cells is that they

originate from a healthy dog, while most human-derived epithelial cell lines are of

tumorigenic origin. Dukes and co-workers (2011) identify the canine origin of the

cells as the only palpable downside, as it limits the selection of antibodies, siRNA

reagents and other origin-specific tools commercially available to researchers.

However, the popularity of MDCK cells has opened the market for canine-specific

tools.

MDCK cells were isolated from a healthy adult male cocker spaniel in 1958

by Drs Madin and Darby. For reasons unknown, they did not publish the isolation

of this line (Dukes et al. 2011). The original MDCK cell line, known as Naval

Biosciences Laboratory cell line 2 (NBL-2), was characterized in 1966 (Gaush et

al. 1966) and was found out to be quickly growing (doubling time estimated at 19.2

h), suitable for the study of replication of some viruses and to display contact

inhibition. However, the NBL-2 cell line was not clonal, but displayed a

considerable amount of heterogeneity. From the original cell line, two sub-types of

MDCK cells were isolated: MDCK type I and MDCK type II. Somewhere along

the way, a spontaneous transformation has occurred, allowing continuous

proliferation of MDCK cell lines. Type I cells are of low passage origin while type

II cells were obtained from a higher passage NBL-2 cells. The two separate cell

types exhibit distinct physiological, biochemical and morphological characteristics

(Richardson et al. 1981, Barker & Simmons 1981). The inherent heterogeneity and

ambiguous naming conventions have probably caused a myriad of confusion as

researchers often fail to report the proper strain and place of purchase of the cell

line used in their publications (Dukes et al. 2011).

36

While both cell strains form epithelial monolayers, MDCK type I cells are

generally smaller and flatter, while type II cells are larger and more columnar. Type

I cells display much larger transepithelial electric resistance (TER) values

compared to the type II cells. These differences can be contributed to the distinct

composition of tight junction complexes. Tight junction components ZO-1 and

claudins 1 and 4 are expressed by both cell types, but only type II MDCK cells

express the pore-forming tight junction protein claudin 2. The expression of this

extra claudin might explain the lower TER values of type II cells (Dukes et al. 2011,

Furuse et al. 2001). Type I MDCK cells express CFTR channel protein, whereas

type II cells do not (Mohamed et al. 1997). There are differences between the other

cell junction types in the two strains as well. Both cell strains display adherens

junctions and desmosomes, but type II has a stronger expression of E-cadherin

(Behrens et al. 1989). In addition, only type I cells form gap junctions (Jordan et

al. 1999).

On the basis of their responsiveness to adrenaline and vasopressin and their

TER values, type I cells are claimed to resemble collecting duct epithelia, while

type II cells are akin to distal (McAteer et al. 1987) or proximal (Richardson et al.

1981) tubule epithelia; however, neither cell type is completely identical to their

respective nephron segments, and care should be taken when comparing them to

their in vivo counterparts (Richardson et al. 1981).

To add to the confusion, several other MDCK cell lines exist (MDCK.1,

MDCK.2, superdome and supertube), and care should be taken not to mistake these

for type I or type II MDCK cells (Dukes et al. 2011).

2.4 Cellular junctions and their constituents

In the course of evolution, the move from single-cell organisms to multicellular

organisms was one of the biggest changes in the complexity of living beings. It

required a multitude of new functions from cells, like communication, group work

and specialization, and ways to stay attached to the cells that comprised the

organism. Cell junctions allow cells to attach to each other, but they also participate

in other critical functions between the cells of an organism mentioned above (le

Bivic 2013). Epithelial cells are the cells specialized in forming continuous sheets

and are often exposed to fluid or air, acting as a selective barrier. This requires cell

junctions that can withstand considerable amounts of mechanical stress: adherens

junctions and desmosomes. In addition to this protective function, epithelial cells

can also be very active in absorbing and excreting ions, hormones, signalling

37

molecules, enzymes, water and other molecules (Hiltunen et al. 2003, Standring et

al. 2008). The junction types and their constituents are presented in Fig. 4.

2.4.1 Tight junctions

Tight junctions (TJ) are located on the apical-most part of the lateral membranes

where they form a strand or several strands around the cell. Tight junctions pull the

cell membranes of two adjacent cells so close to each other that there is virtually

no extra-cellular space between them (Hiltunen et al. 2003, Standring et al. 2008).

This seal halts both the paracellular molecule movement and the intermixing of

apical and basal membrane components. Tight junctions are the most complex and

least understood junction type due to the large number of diverse constituents

(Capaldo et al. 2014). Like adherens junctions, tight junctions attach to actin

filament network, for example via ZO-1. Other well-known tight junction proteins

are the claudin family and occludin (Anderson & van Itallie 2009).

2.4.2 Adherens junctions and desmosomes

Adherens junctions (AJ) are located on the lateral plasma membrane. They

recognize the neighbouring cell of similar origin and join the cells to each other,

but instead of making the epithelial layer impermeable, adherens junctions protect

the epithelial cells from tearing apart from each other. Adherens junction complex

connects the cells to each other via extracellular linkage, and to the actin network

of the cell intracellularly. While providing stable adhesion, adherens junctions are

known for their remarkable plasticity. The way they are built and regulated allows

continuous abolition of existing connections and creation of new ones. This, in turn,

allows changes in cell size, cell shape and cell movement (Coopman & Dijane

2016). The most important components of the adherens junction are the cadherins,

membrane proteins whose extracellular domain binds to the extracellular domain

of cadherin from another cell. On the intracellular side, cadherins bind to the actin

filament network via a group of unrelated proteins called catenins (Maître &

Heisenberg 2013).

Desmosomes, like adherens junctions, use cadherins as the main binding

protein, but instead of actin filaments, they link to intermediate filaments,

especially keratins. They form spot adhesions with dense plaques that can be easily

seen in electron microscope images. Attachment to intermediate filaments provides

great mechanical resilience. Hemidesmosomes are structurally similar to

38

desmosomes, but they link cells to the extracellular matrix via integrins

(Bornslaeger et al. 1996, Hiltunen et al. 2003).

2.4.3 Cadherin superfamily

Cadherins are a super family of single-pass transmembrane glycoproteins (Gall &

Frampton 2013). Their name comes from “calcium-dependent adhesion” as all of

the members mediate cell-cell adhesion with their extracellular domains in a Ca2+-

dependent manner. Cadherins are involved in both adherens junctions and

desmosomes, but in addition they play pivotal roles in diverse processes, from

tissue patterning in embryogenesis to maintenance of the architecture of adult tissue

and growth control during tumorigenesis (Delva & Kowalczyk 2009). The

multitude of these roles elevates them from just simple adhesion molecules holding

cells together like inactive rivets to active and functional agents in the complex

machinery of the cellular processes. Cadherins are divided into four subfamilies

(classical cadherins, desmosomal cadherins, protocadherins and unconventional

cadherins), the best known of which are the classical cadherins (E-cadherin, N-

cadherin and VE-cadherin). Classical cadherins are named after the tissue they

were first identified from; E for epithelial (cadherin 1, CDH1), N for neural

(cadherin 2, CDH2) and VE for vascular endothelial (cadherin 5, CDH5; Maître &

Heisenberg 2013).

E-cadherin

E-cadherin, earlier known by the names liver cell adhesion molecule (in chicken)

and uvomorulin (in mice), among others, is the hallmark of epithelial cell layers

(Gall & Frampton 2013, Schmalhofer et al. 2009). E-cadherin connects the

extracellular linkages to the cytoskeleton (Pece & Gutkind 2000). E-cadherin can

be crudely divided into two domains: extracellular and cytoplasmic. Like other

cadherins, the extracellular domain of E-cadherin binds to the E-cadherin of

another cell in a Ca2+-dependent manner (Shapiro et al. 1995). The cytoplasmic

domain consists of a juxtamembrane domain (JMD) and a catenin-binding domain

(CBD; Gall & Frampton 2013). As the adhesive strength of a single E-cadherin

homodimer (trans-interaction) is weak, the JMD allows gathering of cadherins into

lateral clusters (cis-interaction). If the adhesive strength of a single E-cadherin

homodimer is sufficiently large, cis-interaction formation is limited due to E-

cadherin being constrained in place. Cis-interactions are mediated by p120-catenin

39

(Yap et al. 1998). CBD, on the other hand, links the E-cadherin complex to the actin

filament network via α-catenin and β-catenin (Gall & Frampton 2013). E-cadherin

also participates in several signalling cascades, as it controls the levels of

cytoplasmic β-catenin, which is needed for Wnt signalling. Decreased expression

of E-cadherin is associated with cancer metastasis and is a hallmark of epithelial-

mesenchymal transition (EMT; Kourtidis et al. 2017). Mesechymal-epithelial

transition (MET), on the other hand, is required for the proper development of

kidney and functioning nephrons. MET is defined by increase in the expression or

activity of epithelial genes, like E-cadherin, cytokeratins, desmosomes and

junctional proteins, and decrease in mesenchymal genes like vimentin and

collagens (Chiabatto et al. 2016).

2.4.4 Catenins

Catenins are a group of proteins that form complexes with cadherins. Their name

stems from the Latin word for “chain”, describing their function as a link between

cadherins and the cytoskeleton (Ozawa et al. 1989). With the exception of α-catenin,

all catenins are part of the armadillo (ARM) protein family and contain one or more

Armadillo repeats. On the other hand, α-catenin is an actin-binding protein similar

to vinculin (Gul et al. 2017).

β-catenin is a multifunctional, highly evolutionarily conserved protein.

Binding of β-catenin to E-cadherin is essential for cadherin function and AJ

formation (Kourtidis et al. 2017). In addition to its role in adherens junctions, β-

catenin is involved in the canonical Wnt signalling. In the absence of Wnt signalling,

cytoplasmic β-catenin that is not bound to the adherens junction complex is quickly

marked for proteosomal degradation. However, in the presence of a Wnt signal, the

phosphorylation complex that would mark β-catenin for degradation is sequestered

away, leaving β-catenin free to translocate into the nucleus. Inside the nucleus, β-

catenin regulates gene transcription by binding to TCF/LEF family of co-

transcriptional activators. This pathway is involved in several important processes,

from cell proliferation to differentiation and cell fate specification, and nuclear β-

catenin is a major factor in cancer progression (Boivin et al. 2015, Kourtidis et al.

2017).

γ-catenin, also known as plakoglobin, binds to the same area, the CBD of E-

cadherin, as β-catenin. The localization of these two catenins in cultured cells is

dependent on the time of confluency of the cells: in endothelial monolayers, β-

catenin localizes to the junction sites at the cell periphery upon initial junction

40

formation, before the cells reach confluency, while γ-catenin appears after 48 hours

of confluency (Lampugnani et al. 1995). γ-catenin enhances the barrier function of

endothelial cells and increases their resistance to junction disruption by shear stress

(Venkiteswaran et al. 2002, Schnittler et al. 1997). Together with desmosomal

cadherins (desmoglein and desmocollin), γ-catenin forms a protein complex that

attaches desmosomal junctions to the intermediate filaments.

p120-catenin, also known as δ-catenin, binds to the JMD of E-cadherin, and

can be involved in the localization of E-cadherin to the cell membrane, prevention

of internalization and degradation of E-cadherin, and increasing the stability of

adherens junction. These varying roles are made possible by the interactions of

p120-catenin with E-cadherin. When E-cadherin trans-interactions are intact, p120

is bound to the JMD and is effectively sequestered away from cytoplasm.

Phosphorylation of E-cadherin and other junctional proteins releases p120-catenin

(Davis et al. 2003, Xiao 2003, Kourtidis et al. 2017). p120-catenin is also a

substrate for the tyrosine kinase Src, and contradictory to β-catenin, p120-catenin

is not degraded when not bound to E-cadherin, but is stranded in cytoplasm instead

(Reynolds et al. 1989, Thoreson et al. 2000). Kaiso is a transcription factor that

represses the transcription of genes involved in cell proliferation and tumour

metastasis, and are partly overlapping with those controlled by the β-catenin/Wnt

pathway. p120-catenin that is localized to the nucleus can bind to Kaiso and prevent

the Kaiso-mediated gene repression. These two qualities make p120-catenin a

pivotal player in EMT (Davis et al. 2003)

41

Fig. 4. Schematic presentation of components forming the different epithelial junction

types.

2.4.5 Cadherin signalling

The localization and function of E-cadherin at the cell-cell junctions make it an

excellent hub for transducing both intra- and exogenous signalling. E-cadherin has

even been called an “adhesion-activated cell-signalling receptor” (Yap & Kovacs

2003). In addition to the function in cell adhesion, E-cadherin has a pivotal role in

processes related to cell proliferation, cell polarity, apoptosis, MET, EMT,

migration and invasion. Some of these functions are directly related to the ability

of E-cadherin to interact with β-catenin, p120-catenin, Src and receptor tyrosine

kinases (RTK), or the interactions of the E-cadherin complex with PI3K and actin

cytoskeleton organizing Rho GTPases (Pang et al. 2005, Kourtidis et al. 2017). E-

cadherin can be downregulated by binding of transcription factors SNAIL, SLUG

and TWIST to the E-cadherin promoter (Serrano-Gomez et al. 2016). The role of

downregulation of E-cadherin in cancer progression has been known for years and

is well documented, but recently, more evidence has been uncovered on how the

expression of E-cadherin advances cancer progression. E-cadherin expression has

42

been found to be strong in several cancer types, even those that are metastatic, and

has been shown to be essential for some aggressive tumour types, like inflammatory

breast cancer and certain subtypes of glioblastoma. Kourtidis and co-workers (2017)

list four possible ways in which cancer cells can benefit from E-cadherin: 1) it

participates in the transmission of signals from oncogenes, such as Src, Rac1,

EGFR and ERBB2, 2) it has a central role in a form of metastasis called collective

cell migration, where cancer cells migrate as a sheet, with each cell connected to

another through E-cadherin, 3) it enables cancer cells to divide even when in a

highly confluent environment and 4) E-cadherin increases anchorage-dependent

growth and chemoresistance in Ewing sarcomas (Kourtidis et al. 2017).

The Hippo pathway is a highly evolutionarily conserved pathway of kinase

cascades that has pronounced effects on cell adhesion and participates in regulating

organ size, tissue regeneration and cell proliferation (Michgehl et al. 2017). The

Hippo pathway is composed of four kinases: mammalian STE20-like protein kinase

1 and 2 (Mst1 and 2) and large tumour suppressor homologs 1 and 2 (Lats1 and 2),

and their effector proteins are the transcription factors Yes-associated protein (YAP)

and transcriptional co-activator with PDZ-binding motif (TAZ). When the Hippo

pathway is switched off, YAP and TAZ are dephosphorylated and localize to the

nucleus, where they bind to transcription enhancer factors 1-4 (TEF1-4) and induce

expression of genes involved in cell survival, proliferation and migration. When

the Hippo pathway is on and the kinases are active, YAP and TAZ are

phosphorylated, exported to the cytosol and subsequently degraded. The Hippo

pathway can be activated through physical cues, such as cell contact and

mechanical signals, but also by stress signals, cell cycle and cell polarity signalling

(Crb proteins; Dupont et al. 2011, Ganem et al. 2014, Michgehl et al. 2017). Loss

of Hippo pathway leads to increased activation of YAP and TAZ activated genes,

which results in developmental abnormalities and may contribute to cancer (Harvey

et al. 2013). Switching off the Hippo pathway also leads to dissociation of adherens

junctions and tight junctions due to incorrect distribution of E-cadherin and ZO-1

(Weide et al. 2017). The Hippo pathway also affects cell polarity, as deletion of

both Lats 1 and 2, which leads to increased YAP/TAZ activity, disturbs apico-basal

polarity and results in the loss of apical marker Crb3 from the cell membranes

(McNeill & Reginensi 2017).

E-cadherin can also ligand-independently activate receptor tyrosine kinase

signalling downstream of epidermal growth factor receptor (EGFR), rapidly

leading to increased and sustained activity of mitogen-activated protein kinases

43

(MAPK). The MAPK pathway has a critical role in cell survival, proliferation,

differentiation and embryogenesis (Pearson et al. 2001).

2.4.6 Cadherin recycling

Even though cells exhibit seemingly stable cell-cell junctions, old E-cadherin is

constantly removed from the plasma membrane and degraded while new molecules

are added in their place. This makes the adherens junctions very dynamic structures

that can respond swiftly to both extra- and intracellular signalling (Le et al. 1999).

Most often E-cadherin is removed from the plasma membrane via endocytosis and

either degraded in the lysosomes or recycled back to the plasma membrane,

possibly together with newly synthesized E-cadherin molecules. E-cadherin

endocytosis can occur either via clathrin-dependent or clathrin-independent

mechanism, like caveolin-dependent and macropinocytosis pathways (Bryant et al.

2007, Cadwell et al. 2016). Clathrin-dependent mechanism is the most studied and

most prevalent for E-cadherin. p120-catenin is the key regulator of E-cadherin

endocytosis, and binding of p120-catenin to the JMD of E-cadherin physically

inhibits the endocytosis of E-cadherin, while p120-catenin knock-down causes

continuous endocytosis and subsequent lysosomal degradation of E-cadherin

(Ireton et al. 2002, Xiao et al. 2003). In addition to p120-catenin, other armadillo

family proteins can stabilize E-cadherin at the plasma membrane. δ-catenin and

ARVCF can substitute for p120-catenin in mammalian cells, even though p120-

catenin knock-out is lethal to the embryo (Davis et al. 2003). Cadwell et al. (2016)

postulate this to be due to the need for additional levels of control in cadherin

endocytosis during embryonic patterning.

E-cadherin can also be targeted for degradation via ubiquitination of the JMD

by the E3-ligase Hakai (Fujita et al. 2002). These ubiquitination sites are also

blocked by p120-catenin, further emphasizing its role as the master regulator of E-

cadherin endocytosis. Other E3-ligases might also contribute to E-cadherin

endocytosis, as overexpression of membrane-associated RING-CH 8 (MARCH8,

also known as c-MIR) in zebrafish leads to E-cadherin ubiquitination and reduction

in the levels of E-cadherin localized to the plasma membrane (Cadwell et al. 2016).

2.5 Epithelial cell polarity

As mentioned earlier, the epithelium is highly selective. This selectivity comes with

polarity, i.e. a different composition of cell membrane and the accompanying

44

proteins on different membrane domains of the cell. Epithelial cell membranes can

be divided into three sections in vivo: the basal membrane facing the extracellular

matrix, the lateral domains that are joined to the corresponding ones of the

neighbouring cells, and the apical membrane which faces the hollow, fluid-filled

lumen or acini. This is called apico-basal polarity. In addition, there are several cell

polarity components which only accumulate to certain sections of the cell and can

be used as markers for distinct domains. These can be lipids and membrane proteins,

or cytoplasmic proteins whose localization is specific to the apical or the basolateral

side of the cell. Tight junctions and adherens junctions maintain the segregation of

the cell membrane-bound polarity components to their corresponding domains as

they physically keep cell membrane constituents from mixing (Martin-Belmonte &

Mostov 2008).

Lipids phosphatidylinositol 4,5 bisphosphate (PIP2) and phosphatidylinositol

3,4,5 trisphosphate (PIP3) are important polarity markers which localize to the

inner leaflet of the apical and basolateral domains of the cell, respectively. Their

correct division is vital for cell polarity as they recruit polarity-specific proteins,

and mislocalization of PIP2 and PIP3 results in mistargeting of these proteins. PIP2

can be converted into PIP3 by PI3K, and, reciprocally, PIP3 can be converted into

PIP2 by phosphatase and tensin homologue deleted on chromosome 10 (PTEN).

Artificial introduction of PIP2 to the basolateral domain is sufficient to transform

it into apical and vice versa (Martin-Belmonte & Mostov 2008, Willenborg &

Prekeris 2011, Rodriguez-Boulan & Macara 2014).

Polarity starts to form when the epithelial cell receives polarity cues, often from

a neighbouring epithelial cell. These cues initiate the localization of E-cadherin to

the cell-adhesion sites, and adherens junction is formed. From this initial contact,

polarity complexes are activated and they recruit proteins needed for tight junctions

and commence the trafficking of apical and basal proteins and lipids to their

corresponding domains. Polarity complexes are highly conserved groups of

proteins that work in cooperation. These complexes are the Partitioning-defective

(Par) complex, the Crumbs (CRB) complex and the Scribble (SCRIB) complex.

The PAR complex consists of PAR3, PAR6 and atypical protein kinase C (aPKC),

and is recruited to the adherens junctions at the border between apical and

basolateral domains (Tabuse et al. 1998). This results in the formation of tight

junctions, separation of domain-initiating factors and recruitment of CRB complex.

The largely epithelial specific CRB complex consists of CRB, Protein associated

with Lin Seven 1 (PALS1) and PALS1-associated tight junction protein (PATJ), and

it marks the areas where tight junctions should be formed (Lemmers et al. 2004).

45

The CRB complex also maintains the polarity in fully polarized epithelial cells. The

function of the SCRIB complex is much less understood in mammalian epithelia

than that of CRB or PAR complexes. It consists of scrib, discs large (dlg), and lethal

giant larvae (lgl) proteins and is a prerequisite for the establishment of basolateral

domain (Willenborg & Prekeris 2011). A schematic drawing of polarized epithelial

cell is presented in Fig. 5.

Fig. 5. A schematic presentation of a polarized epithelial cell along with selected

polarity markers. aPKC; atypical protein kinase C, CDC42; cell division control protein

42 homologue, CRB; Crumbs, DLG; Discs large, PALS; Protein associated with Lin

Seven, PAR; Partitioning-defective, PATJ; PALS1-associated tight junction protein, PI3K;

Phosphatidylinositol-4,5-bisphosphate 3-kinase, PIP2; Phosphatidylinositol 4,5

bisphosphate, PIP3; Phosphatidylinositol 3,4,5 trisphosphate, PTEN; Phosphatase and

tensin homologue deleted on chromosome 10, SCRIB; Scribble.

46

If, for some reason, the epithelial cell fails to properly polarize or the

polarization is incomplete, the result might be polycystic kidney disease, cystic

fibrosis, cancer or a host of other diseases (Willenborg & Prekeris 2011).

2.5.1 Lumen formation and maintenance

Epithelial cells delineate hollow structures called lumina, like renal tubuli or

glandular acini. Polarity of the epithelial cells, together with a proper formation of

cell-cell junctions, is crucial for lumen morphogenesis. The cells need to establish

the proper apical side in conjunction with each other and maintain it at all times,

e.g. during cell division (Schlüter & Margolis 2009).

Lumen formation in vivo is a complex process that typically occurs during

development and heavily involves the creation of epithelial cells via MET. First,

mesenchymal cells form an aggregation and start forming cell adhesions. Second,

cells establish an apico-basal axis, where the apical surface is towards the inside of

the aggregate. Finally, lumen formation occurs in the areas where two apical

surfaces of different cells are touching. From there, the cells divide and form

elongate structures, while the lumen expands along the tubule (de novo lumen

formation; Schlüter & Margolis 2009, Gao et al. 2017). In the kidney, both de novo

lumen formation and the use of pre-existing tubule occurs (Saxen & Sariola 1987;

Gao et al. 2017).

In vitro, lumen formation is thought to happen via two distinct mechanisms:

hollowing, where small vesicles coalesce at the cell-cell border, or cavitation,

where cells that end up inside the cyst undergo apoptosis. If cells are able to achieve

polarization rapidly, hollowing is thought to happen. This is usually the case if the

cells are grown in ECM scaffold that contains laminin, like Matrigel, that gives the

cells a strong apical cue to facilitate polarization. However, if cells are grown in

collagen they lack strong apical cues, and thus polarize at a slower rate while still

proliferating at comparable speeds, resulting in cells getting trapped inside the cyst.

These cells are cleared via apoptosis and the lumen formation process is called

cavitation. Lumen formation can also shift between these two processes, depending

on the degree of polarization (Martin-Belmonte & Mostov 2008, Schlüter &

Margolis 2009).

Cyst formation is the process where cells either aggregate and form a

distinctive structure via cell adhesion, or an aggregate forms through cell

proliferation, starting from a single cell or a few parental cells. Hollowing is the

process of plasma membrane separation in the early stages of cyst formation. Once

47

the proper apico-basal polarity has been established, special endocytotic vacuoles

called VACs (vacuolar apical compartment) are transported to the apical surface

between two cells, creating an area called apical membrane initiation site (AMIS;

Vega-Salas et al. 1987, Bryant et al. 2010). VACs contain apical proteins and ECM

fluids, and once at the apical membrane, they release their contents into the space

between the cells, further opening the luminal space. For the lumen to open, the

tight junctions of the two cells need to be intact and properly functioning. This early

stage of lumen is called the pre-apical patch. Lumen opening is furthered by

localization of large transmembrane glycoproteins, such as podocalyxin (also

known as gp135), to the apical membrane (Takeda et al. 2000, Meder et al. 2005).

Podocalyxin is heavily glycosylated and carries a highly negative surface charge.

Its appearance to the apical membrane causes steric repulsion, which pushes the

apical membranes of the two joined cells further apart. Due to this anti-adhesion

role, podocalyxin is required for normal kidney function. Cells continue to divide

and increase the size of the cyst, while simultaneously secreting water into the

lumen to create and uphold hydrostatic pressure. Water can come through

aquaporins on the apical membrane or through paracellular pathways (Schlüter &

Margolis 2009, Willenborg & Prekeris 2011, Sigurbjörnsdóttir et al. 2014).

Cavitation occurs when a cyst is solid or has an imperfect lumen. In this process,

cells that are considered to be inside the lumen, i.e. are lacking an ECM contact,

are cleared via apoptosis. Lumen clearing can occur either via traditional caspase-

dependent apoptosis or a special kind of apoptosis called anoikis that occurs when

cells lack ECM-contacts. Cavitation can be seen in vivo in mammary gland

formation. Inhibition of apoptosis delays the clearing of lumen, but does not

completely abolish it, which means that there are other methods that ensure the

formation of the apical lumen (Willenborg & Prekeris 2011).

3D culture models have made studying lumen formation much more feasible

in vitro. However, in vivo, lumen formation occurs sparsely, and mainly during

embryogenesis. Notable occurrences of lumen formation after embryogenesis are

mammary gland formation during puberty and wound healing or regeneration after

injury.

2.6 Malignant transformation

The hugely complicated machinery that is the functional body of a multicellular

organism requires immensely precise control, coordination and cooperation

between the different cells and cell types. A multitude of partly redundant,

48

overlapping systems try to ensure that cells live, proliferate and die according to

their roles. These normal tissue functions are regulated by proteins expressed

according to the genetic codes in DNA, which, in turn, can be disturbed by

mutations. Single mutation in the DNA of a single cell is rarely catastrophic, as for

it to be harmful to the cell, it needs to be in the coding region of a gene and remain

undetected by the cell’s own repair machinery. If this machinery notices the

mutation, it tries to repair the damage or forces the cell to die via programmed cell

death process. If this fails, extracellular factors like signalling molecules and even

other cells can force the cell to undergo apoptosis. Even when it goes completely

unnoticed, a single mutation is rarely able to cause malignant transformation, i.e. a

process where the cell embodies aspects of cancer cells: uncontrolled proliferation,

resistance to apoptosis (especially anoikis) and enhanced ability to migrate. The

cell carrying mutated DNA might be able to proliferate faster than regular cells, but

will die due to apoptosis signals, or is resistant to apoptosis, but divides too slowly

to pose a threat. Generally, a cell can undergo malignant transformation only when

multiple mutations affecting cell proliferation and making the cell immune to death

signals accumulate. Mutations can be inherited via gametes, leading to hereditary

diseases, or occur spontaneously in somatic cells due to the environment or other

factors. Any cell type can undergo malignant transformation, but cancers of

epithelial origin, carcinomas, are by far the most common (Alberts et al. 1991).

2.6.1 Oncogenes and tumour suppressors

Oncogenes are genes that have the potential to cause cancer if they are mutated.

Proto-oncogenes, or oncogenes that are functioning normally, are typically

important, multifunctional genes that have a role in several different processes,

usually involving proliferation and differentiation. Oncogenic mutations that lead

to malignant transformation are usually gain-of-function mutations, i.e. the

oncogene protein levels or its activity is increased (MacDonald et al. 2004).

Examples of proto-oncogenes include Src, Ras, Wnt, Erk and Myc. Many of the

oncogenes are tyrosine kinases, a family of proteins that transfer a phosphate to a

protein, switching it between inactive and active stages.

Genes whose products naturally antagonize or counteract oncogenes are called

tumour suppressors. These products help maintain the genome, improve cell

polarization, differentiation or cell adhesion, induce apoptosis or reduce

proliferation. Examples of tumour suppressor gene products are p53, PTEN and

APC. Cancer cells usually have mutations not only in the genes coding proto-

49

oncogenes, but also in those coding tumour suppressors. These mutations must

inactivate tumour suppressors or reduce their activity or half-life, cause

mislocalization or otherwise hamper their functions. Taken together, it has been

estimated that to develop malignant tumours, most human cancers require two to

eight mutations in proto-oncogenes or tumour suppressor genes (Vogelstein et al.

2013).

2.6.2 Epithelial-mesenchymal transition

Being able to adapt to the changes in one’s environment is one of the defining

features of the winners in the process of evolution. Cells must also possess a certain

level of plasticity to cope with the changes in their environment, and cells thus have

processes that help in the adaptation. One such process is called the epithelial-

mesenchymal transition (EMT). As the name implies, EMT changes epithelial cells

into mesenchymal phenotype, characterized by lack of polarization and cell

adhesion and increase in motility and invasion. EMT occurs heavily during

embryogenesis in gastrulation and neural crest formation, among others, but also

in wound healing and fibrosis. EMT is divided into three categories: type I occurs

during embryogenesis, type II occurs in wound healing, and type III is the one

utilized by cancer cells (Serrano-Gomez et al. 2016).

EMT is indeed a critical step in the development of malignant cancers, as

epithelial cells have poor motility and invasion capabilities, and the ability to

metastasize is what separates benign tumours from malignant ones. EMT is often

initiated by paracrine factors secreted by cancer-associated fibroblasts (CAFs; De

Wever et al. 2008). These factors include transforming growth factor-β (TGF-β)

superfamily members, matrix metalloproteases (MMPs), tenascin C and other

extracellular matrix components that induce the integrin-mediated cell response

and growth factors that bind to tyrosine kinase receptors including EGF, HGF and

FGF and prostaglandin E2 (Dalla Pozza et al. 2017). The hallmarks of EMT include

removal of E-cadherin from the plasma membrane (often accompanied by

downregulation of E-cadherin expression by transcription factors such as SNAIL,

SLUG and TWIST) and subsequent β-catenin localization to the nucleus, where it

activates the LEF1 and TCF transcription factors, activation of Src kinase,

disruption of cell polarity and apico-basal axis, direct cell contact with collagen

type I and increased expression of vimentin and N-cadherin (Guarino 2010).

Several pathways induce EMT, including Wnt/β-catenin, Notch, Ras-MAPK,

PI3K/Akt, hypoxia signalling and Hedgehog. With these newly acquired invasive

50

properties, epithelial cells can now detach from other cells and the ECM, make

their way to the bloodstream and exit it in a new location. At this new metastasis

site, the cancer cells usually undergo MET to return to their more epithelial

phenotype (Dalla-Pozza et al. 2017).

2.7 Programmed cell death has many forms

Normal, healthy cells have a limited life-span. For organisms, it is beneficial to be

able to control the death process of the cell, known as programmed cell death (PCD),

which is separate from uncontrolled cell death process called necrosis.

Programmed cell death is used by the organism to clear unwanted or harmed cells

in a way that does not induce immunological response. Necrosis alerts the immune

system, which can be beneficial, but widespread necrosis and accompanied

immunological response are very taxing for the organism and can cause more harm

than the loss of the necrotic cells (Kiraz et al. 2016).

In necrosis, the cell swells and eventually ruptures, leaking all of its

constituents into the cell environment. This can wreak havoc in the neighbouring

cells, as many signalling molecules are spread to the environment without any

control (Proskuryakov et al. 2003). PCD, on the other hand, causes cells to shrink

and their nuclei to condensate, but their plasma membrane and cellular organelles

retain their integrity. PCD is usually beneficial to the organism, as it has a role in

embryogenesis, tissue homeostasis, healing and pathogenesis. For a long time,

programmed cell death was synonymous with apoptosis, but recently mechanisms

different enough from apoptosis have been found. PCD can be divided into three

categories: apoptosis, necroptosis and autophagy (Fuchs & Steller 2015).

Apoptosis is the best studied mechanism of PCD. It is highly conserved and

requires a cascade of cysteine proteases called caspases, which can be activated

either intrinsically (by, for example, DNA damage) or extrinsically (by, for example,

binding of extracellular ligand to the plasma membrane death receptors). For

obvious reasons, apoptosis is carefully regulated. The intrinsic pathway starts with

the mitochondrial outer membrane permeabilization (MOMP), caused by Bcl-2

(see more below). This releases mitochondrial proteins, most important of which

are cytochrome c and second mitochondria-derived activator of caspases (SMAC),

into the cytosol. There, SMAC binds to inhibitors of apoptosis proteins (IAP),

which normally form a complex with caspases, keeping them inactive. Caspases

are divided into initiator caspases (caspase 1, 2, 4, 5, 8, 9, 10, 11 and 12 in mammals)

and executioner caspases (3, 6, 7 and 14 in mammals). Initiator caspases can then

51

activate executioner caspases further down-stream. Activated executioner caspases

initiate processes that lead to cell death. The extrinsic pathway is also caspase-

dependent, and the later stages are identical to apoptosis via intrinsic pathway.

Initiation of the extrinsic pathway, however, requires an extracellular ligand

binding to the membrane-bound death receptors instead of MOMP. This death

receptor can be either TNF-α or Fas, both of which belong to the tumour necrosis

factor (TNF) protein superfamily. The death receptors activate caspases by

interacting with them via adaptor proteins (TNFR-associated death domain

(TRADD) for TNF-α and Fas-associated death domain (FADD) for Fas). The

extrinsic pathway may also induce MOMP (Kiraz et al. 2016). For a graphical

presentation of apoptosis, see Fig. 6.

Another mode of PCD is somewhere halfway between necrosis and apoptosis,

and to reflect this, it was termed necroptosis. This cell death programme is caspase-

independent and is activated when one of the various death receptors, including

TNFR1, TNFR2, Fas, Toll-like receptors and protein kinase R (PKR), is ligated. In

other words, necroptosis uses the same pathways as extrinsically initiated apoptosis

and leads to apoptosis as long as caspases are functional. In case where caspases

are non-functional or inhibited, the necroptosis pathway forms membrane-

disturbing pores that allow Na+ and Ca2+ ions to leak in, causing rapid increase in

osmotic pressure that leads to the rupture of the cell. Thus, necroptosis seems to be

a back-up mechanism in case where the caspases are not functioning properly, but

it might also have an important role in alarming the immune system when a cell

succumbs to bacterial or viral infection. Cells that undergo necroptosis also induce

and propagate the inflammatory response. Biomolecules that are not usually found

freely in the extracellular environment leak out of the cell and thus “alarm” the

immune system. In apoptosis, everything in the cell, is contained in vesicles, the

apoptotic bodies, which are then cleared by macrophages via endocytosis (Fuchs

& Steller 2015).

Autophagy, also known as macroautophagy, involves degradation of cellular

components, packed in autophagosomes, in lysosomes. Autophagy-related (ATG)

genes regulate autophagy and are considered pro-survival, as deletion or silencing

of the ATG genes leads to accelerated cell death. However, it is unclear whether

autophagy is a separate mechanism of programmed cell death or an unsuccessful

attempt at saving the cell from apoptosis, as autophagy-related cell death often

requires functional caspases to proceed (Fuchs & Steller 2015).

52

Fig. 6. Common pathways in apoptosis. Apoptosis requires a cascade of caspases to

occur. Caspases are produced in inactive pro-forms and require an enzymatic cleavage

to be active. Initiator caspases activate effector caspases, which then activate several

pathways that lead to apoptosis of the cell. Caspase activation can occur extrinsically

via ligand binding to plasma membrane receptors, which then cleave initiator caspases,

or intrinsically through leakage of mitochondrial proteins, like SMAC and cytochrome

c. Increase in mitochondrial membrane permeability requires oligomerization of BAD

and/or BAX proteins, which are normally bound to the proteins of the Bcl-2 family. BH3-

only proteins can dislodge BAD/BAX from Bcl-2, leading to pore formation and increase

in membrane permeability. Once in the cytoplasm, cytochrome c can form a complex

called apoptosome with APAF-1 and pro-caspase 9. Subsequently, the apoptosome can

cleave effector caspases. IAPs can prevent apoptosis by binding to caspases or by

inhibiting other pro-apoptotic proteins, like SMAC. APAF-1; apoptotic protease

activating factor 1, BAD; Bcl-2-associated death promoter, BAX; Bcl-2-associated X

protein, Bcl-2; B-cell lymphoma 2 family, BH3-only; Bcl-2 homology 3-only, FADD; Fas-

associated death domain, Fas; Fas receptor, IAP; inhibitor of apoptosis family, SMAC;

second mitochondria-derived activator of caspases, TRADD; tumour necrosis factor

receptor type 1-associated death domain, TNF-α; tumour necrosis factor alpha, TNFR;

tumour necrosis factor receptor.

53

2.7.1 Inhibition of apoptosis

As mentioned above, apoptosis is tightly controlled and regulated. Several proteins

inhibit different steps in apoptosis to prevent it from happening accidentally or

prematurely. Amongst the most important players in apoptosis inhibition are the

members of the inhibition of apoptosis (IAP) protein family. IAPs are structurally

identified by the presence of one or more baculovirus IAP repeats (BIR), a zinc-

finger fold consisting of approximately 70 amino acids, and are often upregulated

in cancers. Their upregulation is also associated with poor prognosis in many

cancers. So far, 8 IAP-members have been found in human genome: NAIP, cIAP1,

cIAP2, XIAP, survivin, Apollon, ML-IAP and ILP-2. IAPs regulate apoptosis by

inhibiting caspase activity, hindering their localization or complex formation or

preventing pro-caspases from being cleaved. Some IAPs also inhibit other pro-

apoptotic proteins, like SMAC and death receptors, and can even indirectly activate

nuclear factor-kappa B (NF-κB; Duffy et al. 2007, Fucsh & Steller 2015). They

also have a role in inhibiting necroptosis (Micheau & Tschopp 2003). Some IAPs

have other functions besides apoptosis inhibition, and they have been shown to be

involved in immune regulation, chromosome segregation and cytokinesis.

Survivin

With only 142 amino acids and a single BIR repeat, survivin is the smallest member

of the IAP family. It was first identified as a member of the IAP family by Altieri

and colleagues in 1997. They also discovered that survivin expression was

downregulated in healthy adult tissues, but upregulated in foetal tissues and several

cancers (Ambrosini et al. 1997). This cancer-specificity makes survivin an enticing

target for anti-cancer drugs.

Survivin cannot bind directly to the caspases, but by forming a complex with

hepatitis B X-interacting protein (HBXIP) or XIAP they can bind and inhibit active

forms of caspase 3 and 9. Survivin also binds to SMAC and can thus prevent it

from binding to XIAP, a potent apoptosis inhibitor (Altieri 2010). Survivin can also

bind to XIAP to enhance its stability and caspase-inhibition capabilities (Dohi et al.

2004). The apoptosis inhibition capabilities of survivin are not only limited to the

caspase-dependent pathway. Survivin has been shown to be able to inhibit the

mitochondrial apoptosis-inducing factor (AIF), which can induce caspase-

independent apoptosis by causing DNA fragmentation (Liu et al. 2004,

Athanasoula et al. 2014).

54

Survivin has other functions in addition to its titular role in inhibition of

apoptosis. It also offers cytoprotection, regulates cell-cycle progression and mitosis,

and participates in cell motility and metastasis. The best characterized is the role of

survivin in mitosis, where lack of survivin causes polyploidy of chromosomes and

apoptosis. Chromosomal passenger complex (CPC) is a checkpoint for cell division

and consists of survivin, inner centromere protein (INCENP), Aurora B and

Borealin (Lens et al. 2006). As part of the CPC, survivin acts in all phases of mitosis,

from the aforementioned checkpoint control and kinetochore attachment to

chromatic separation and cleavage furrow progression in anaphase, and finally

localizes to the midbodies of the divided cells, where it ensures that the cytokinesis

is completed in a timely manner (van der Horst & Lens 2014).

Regulation of the survivin gene is highly complex, and its upregulation

involves several pathways that deal with cell proliferation, cell survival and

development, like mTOR/Ras, TCF-4/β-catenin, NF-κB, Wnt-2, PI3K/AKT,

GATA-1, KLF5, specificity proteins 1 and 3 (SP1 and SP3, respectively) and signal

transducer and activator of transcription 3 (STAT3; Kelly et al. 2011). Several

important tumour suppressors, like p53, PTEN, FOXO and BRCA1-SIRT1, are

involved in survivin downregulation (Athanasoula et al. 2014). Survivin knock-

down mice do not survive past the blastocyst stage, and conditional knock-downs

in adult tissues display extensive mitotic defects (Uren et al. 2000).

Phosphorylation plays a crucial role in the functions of survivin protein and can

affect its stability, subcellular localization, protein-protein interactions and activity

in mitosis and apoptosis inhibition (Altieri 2010). Survivin is a short-lived protein

with a t½ of about 30 min before it is degraded via proteasomes (Zhao et al. 2000).

Survivin expression is highly upregulated in almost all types of cancer. High

levels of survivin correspond with chemotherapy resistance, increased chance for

relapse and poor prognosis. Survivin is also suspected to increase tumour

angiogenesis as survivin inhibition of endothelial cells causes their apoptosis and

vascular regression (Kelly et al. 2011, Unruhe et al. 2016). The cellular location of

survivin in cancers seems to have an effect on patient survival, but results vary from

one cell type to another. Usually, cytoplasmic survivin seems to confer

cytoprotective properties and resistance to apoptosis, while nuclear survivin is

responsible for the functions of survivin in mitosis. However, in some cancer types,

like melanoma and glioblastoma, nuclear survivin is a marker of poor prognosis

and heightened chance for relapse, while cytoplasmic survivin has no effect on

patient survival (Kelly et al. 2011, Saito et al. 2017). On the other hand, two studies

of the intracellular localization of survivin in diffuse large B-cell lymphoma

55

(DLBCL) gave opposing results (Watanuki-Miyauchi 2005, Liu et al. 2007). Kelly

et al. (2011) identify the small number of samples used, non-uniform treatments

applied, differing methods of detection of survivin and the alternative splicing of

survivin as reasons for the difference in the results.

More recent findings have revealed the important role of survivin in stem cell

maintenance. Stem cells express high levels of survivin, and although its exact

functions in stem cells are not well understood, it seems that they mainly need

survivin for apoptosis inhibition. Survivin might also have a role in upstream

controlling of stem cell pluripotency, and expression of survivin in stem cells

enhances cell proliferation, motility and invasion. Cancer stem cells also express

high levels of survivin and benefit similarly from its effects. However, extensive

research is needed to elucidate this novel role for survivin (Altieri 2015).

The expression of survivin correlates with cancer metastasis in several cancers.

Whether survivin has a role in the metastatic phenotype is debated, but it has been

shown that survivin expression is mechanistically linked to increased production of

matrix metalloproteinases, upregulation of α5 integrin together with activation of

Akt, and NF-κB-dependent excretion of fibronectin (McKenzie et al. 2010,

Mehrotra et al. 2010, Gao et al. 2014). Survivin might also increase motility in

normal cells, as knock-down of survivin reduced chemotaxis and caused

disorganized actin filaments to form and changed the shape of vascular smooth

muscle cells without affecting cell viability or proliferation (Nabzdyk et al. 2011,

Altieri 2015).

B-cell lymphoma-2 (BCL-2) family

The BCL-2 protein was the first discovered antiapoptotic protein and is highly

evolutionally conserved in different species. BCL-2 family members are identified

by one or more BCL-2 homology (BH) domains, and the family has nearly 20

members in mammals alone. Interestingly, the BCL-2 family contains both pro-

survival (BCL-2, BCL-XL, BCL-W, MCL-1, A1/BFL-1 and BCL-B) and pro-

apoptotic proteins (BAX and BAK, and possibly BOK) that participate in the

controlling of the intrinsic pathway of apoptosis. In healthy cells, BAX and BAK

are kept inactive by the pro-survival members of the BCL-2 family. However, in

stressed cells, members of distant relatives of the BCL-2 members, the BH3-only

proteins, displace the pro-survival BCL-2 proteins and allow the activation of BAX

and BAK. If allowed to activate, BAX and BAK form large homo-oligomers and

permeabilize the mitochondrial outer membrane (MOM), allowing cytochrome c

56

and SMAC, among other mitochondrial proteins, to leak out and induce the

activation of caspases (Cory et al. 2016).

2.8 Src was the first identified proto-oncogene

The history of the Src tyrosine kinase family is firmly tied to the history of cancer

genetics and molecular biology. Cancer as a disease has been known for thousands

of years, but until early 20th century it was thought to be completely endogenous in

origin. In 1911, Peyton Rous discovered that a sarcoma in chickens can be induced

by a transmissible agent, a virus now known after him as Rous sarcoma virus (RSV).

By the 1950s, the scientific community was ready to accept that tumours can have

a viral origin. In the next 30 years, the transforming entity in the RSV was

recognized as a gene in the viral genome and, finally, in 1980 this gene, named

viral-src (v-src, after sarcoma), was sequenced. By then, gene sequences

homologous to v-src were found from normal avian DNA, suggesting that the src

gene had normal, cellular origins. This cellular-src (c-src) was the first identified

proto-oncogene, a discovery awarded with a Nobel Prize in Physiology and

Medicine in 1989. This also led to the discovery that mutated genes have a role in

cancers. Other members of the Src of non-receptor membrane-associated tyrosine

kinases include Fyn, Yes, Blk, Yrk, Fgr, Hck, Lck, and Lyn. Src, Fyn and Yes are

ubiquitously expressed while the others are highly cell-type specific (Martin 2001,

Guarino 2010).

The protein product of c-src was identified to be a tyrosine kinase, also a first

of its kind (Martin 2001). Src kinase consists of the tyrosine kinase domain (also

known as Src homology 1 domain, SH1) and two non-kinase domains, SH2 and

SH3. Two tyrosine phosphorylation sites are present at the carboxy-terminal end of

the protein: Tyr416 and Tyr527. These two are crucial for the regulation of Src

activity. Phosphorylation of Tyr416 results in activation of the kinase, while

phosphorylation of Tyr527 inhibits its activity (Patschinsky et al. 1982, Cooper

1986, Martin 2001). Both SH domains also have intramolecular interactions with

Src, and these interactions result in a closed conformation of Src, keeping it inactive.

Tyr527 is the key in regulating Src activity, and this site is deleted in the constitutive

active v-Src (Murphy 1993, Superti-Furga 1993, Xu et al. 1997). In vivo, Tyr527 is

phosphorylated by carboxy-terminal Src kinase (Csk), while Tyr416 is

autophosphorylated (Okada & Nakagawa 1989, Martin 2001). The regulation of

Src activity is strict, and even when overexpressed, c-Src is unable to induce a

complete transformation process (Johnson et al. 1985). Src can be directly activated

57

by RTKs, integrins, focal adhesion kinase (FAK), Crk- and Src-associated substrate

(CAS) and protein-tyrosine phosphatase 1B (PTP1B). In addition, reactive oxygen

species (ROS) are able to modify the cysteine residues of Src, increasing its kinase

activity (Giannoni et al. 2008). Failure in the regulation of Src activity leads to

increased cell proliferation, anchorage- and growth factor-independence, survival,

loss of contact inhibition and migration, all of which are required for development

of cancer. Not surprisingly, overexpression or hyperactivity of Src is highly

common in a multitude of cancers, including colorectal, breast, prostate and

pancreatic cancer, melanoma and different types of sarcoma. However, Src itself is

very rarely mutated, making Src a more probable candidate in the maintenance of

cancer cells and tumour progression rather than tumour initiation (Martin 2001,

Guarino 2010). A schematic picture of c-Src and v-Src, together with open and

closed formations is presented in Fig. 7.

Src kinase participates in a myriad of cellular functions, from maintaining

cellular homeostasis to proliferation and survival, cytoskeleton regulation, cell

shape, cell motility, migration, cellular contacts and adhesion. Src is localized to

the cytoskeleton or the inner face of the plasma membrane, but also to cell-matrix

or cell-cell adhesion sites. At its core, Src functions as the messenger, transmitting

external signals to the cell interior by phosphorylating of its substrates (Bjorge et

al. 2000). RTKs can activate Src, and Src, in turn, can enhance their activity,

creating a positive feedback-loop. Integrins can interact with SH3 domain of Src to

induce its activation, or recruit FAK, which also results in Src activation. When

activated by RTKs or integrins, Src can activate the Ras/MAPK and PI3K/Akt

pathways, STAT3 transcription factor or Rac GTPases. Src can also phosphorylate

E-cadherin, β-catenin, vinculin and, importantly, p120-catenin, which is an

essential mediator of anchorage-independent cell growth (Bjorge et al. 2000,

Guarino 2010). Phosphorylation of the E-cadherin-β-catenin complex disintegrates

the adherens junction and releases β-catenin together with p120-catenin into the

cytoplasm (Behrens et al. 1993, Guarino 2010). Cytoplasmic p120-catenin

enhances cell migration, invasion and metastasis, while nuclear p120-catenin

enables transcription of genes involved in cell proliferation (Strumane et al. 2006).

Vinculin is activated through phosphorylation by Src, and can participate in focal

adhesion formation. Non-phosphorylatable vinculin mutant leads to spreading

defects and impaired wound healing (Auerheimner et al. 2015).

58

Fig. 7. A schematic presentation of c-Src and v-Src structures (A) and open and closed

conformations (B). The important tyrosine residues regulating Src activity are marked

in both pictures.

2.8.1 Temperature-sensitive Src MDCK (ts-Src MDCK) cells

Many viruses carry mutations that make them temperature-sensitive (Wyke &

Linial 1973). Usually, this means that the temperature-sensitive viruses are more

infectious at the permissive temperature (< 37°C; commonly around 32–35°C) than

at non-permissive temperature (> 37°C; commonly around 39–41°C). In 1970,

Martin isolated a temperature-sensitive mutant of the RSV (Martin 1970). These

mutant viruses were able to replicate normally at the non-permissive temperature

of 41°C, but failed to transform the host cells. When transferred to the permissive

temperature of 36°C, the cells underwent visible changes in their morphology and

adopted a transformed phenotype. Surprisingly, the cells returned to their normal

phenotype within 4 hours after being transferred back to the non-permissive

59

temperature (Martin 1970). Later, the transformation-causing gene was identified

as src. In 1993, Behrens and co-workers infected MDCK type I cells with murine

leukaemia retroviral construct containing the temperature-sensitive v-src gene

(Behrens et al. 1993). This ts-Src MDCK cell line was capable of inducing rapid

and profound changes to the cell morphology when cultivated at the permissive

temperature of 35°C. In addition, these changes were also reversible when the

temperature was changed to non-permissive (40.5°C). De Diesbach et al. (2008)

attributed the temperature sensitivity to heat-shock proteins, like Hsp90, that bind

to v-Src and, while protecting it from proteolysis, also keep it inactive. A change to

permissive temperature releases v-Src from these chaperones, activating it.

MDCK cells cultivated in the permissive temperature had an invasive,

fibroblast-like phenotype together with diffuse distribution of E-cadherin and

differences in tyrosine phosphorylation of β-catenin and, to a lesser extent, of E-

cadherin. Phosphorylation events occurred as early as 10 minutes after the

temperature shift from non-permissive to permissive. This was followed by visible

changes in cell contacts 15 min after the temperature shift (Behrens et al. 1993). In

later studies, the time at permissive temperature it takes v-Src to become active has

been correlated to the polarization state of the cell and the amount and condition of

cell-cell junctions it has, ranging from 20 min in poorly polarized cells to 60 min

in fully polarized cells (de Diesbach et al. 2008). Due to the simple way the Src

tyrosine kinase can be activated, the ts-Src MDCK cell line is a powerful tool for

studying cell junctions, Src-target proteins and phosphorylation events,

transformation of the cells, and epithelial-mesenchymal transition. The cell line is

especially useful in studying phenomena transpiring at the very beginning of the

aforementioned events, as the start point is known.

2.9 Reactive oxygen species

Reactive oxygen species (ROS) are highly reactive molecules that contain oxygen.

They include oxygen radicals (superoxide, hydroxyl, peroxyl and alkoxyl) and non-

radical oxidizing agents, such as hydrogen peroxide. ROS are formed in organisms

as part of their metabolism, especially in mitochondria. Some oncogenes can also

produce ROS, most commonly by targeting NADPH oxidase (NOX), an enzyme

that generates ROS as its main product instead of a by-product (Prasad et al. 2017).

Cells can also acquire ROS exogenously via pollutants, tobacco, smoke, drugs and

other such mediators. Ionizing radiation produces ROS when it interacts with water.

Cells neutralize ROS with specialized enzymes like superoxide dismutase,

60

intracellular antioxidants, like glutathione, and antioxidants acquired through

consumption of food like vitamin E (Prasad et al. 2017).

In low concentrations, ROS are beneficial to the cell as they act as regulators

of intracellular signalling and homeostasis. At high levels, their reactivity can

damage DNA, proteins and lipids and alter the activation of transcription factors,

kinases such as Src and MAPK, growth factors like VEGF and EGF, matrix

metalloproteases and cytokines. Not surprisingly, ROS also has a major role in the

development of several cancers, including breast, lung, colon, pancreatic and

prostate, as well as melanoma, leukaemia and glioblastoma. Cancer cells also

produce increased amounts of ROS compared to healthy cells, as they eschew the

aerobic respiration in favour of glycolysis and other anaerobic metabolic processes.

This increases intracellular ROS production. However, generating large amounts

of ROS is not without risks for the cancer cells, as they also suffer from ROS-

induced damage that can still induce apoptosis or render the cells unviable due to

the excessive damage to DNA and/or proteins and lipids. Thus, cancer cells are

more reliant on antioxidant molecules and scavenging (ROS eliminating)

mechanisms than regular cells. Many chemotherapeutic and other anti-cancer drugs

take advantage of this vulnerability, and while intracellular ROS levels of all cells

are increased, cancer cells are less able to cope with them than healthy cells

(Acharya et al. 2010, Prasad et al. 2017).

2.9.1 Piperlongumine is a small molecule that selectively targets

cancer cells

Many naturally occurring and synthetic molecules like curcumin,

dehydrozingerone and piperlongumine (PL) have been studied for their anti-cancer

and ROS elevating properties. PL (also known as piplartine), a naturally occurring

alkaloid/amide from the long pepper fruit, has been used widely in Indian and

Chinese medicine. It was shown to kill several types of cancer cells, both in vivo

and in vitro, without exhibiting toxic effects in normal cells (Raj et al. 2011, Niu et

al. 2015). In addition, it has been shown to increase the antitumour activity of

chemotherapeutic drugs (Bezerra et al. 2013). PL affects the expression or activity

of several proteins important for cancer development. It has been shown to increase

ROS, cause G2/M cell cycle arrest, increase apoptosis by upregulating the

expression pro-apoptotic proteins like caspase 3 and p53, and downregulating anti-

apoptotic proteins like survivin, BCL-2 and XIAP, inhibit cell migration, metastasis,

EMT and angiogenesis by downregulating the expression of VEGF, HIF-2, Twist,

61

N-cadherin, vimentin and p120-catenin (Bezerra et al. 2013). PL also inhibits NF-

κB activity by preventing its nuclear localization (Niu et al. 2015). Raj and co-

workers (2011) identified 12 targets for piperlongumine, seven of which are

involved in oxidative stress response, implying that the generation of ROS is highly

important for the anti-cancer activity of PL. This was supported by the finding that

treating cells exposed to PL with N-acetyl cysteine (NAC) rescued the cells from

PL-induced cell death. These findings indicate that PL kills tumour cells by both

elevating their ROS levels and targeting their ability to cope with oxidative stress,

while healthy cells survive due to their higher functioning oxidative stress response

elements.

62

63

3 Aims of the study

Cell differentiation and malignant transformation are highly complex processes that

mainly oppose each other. Numerous changes in gene expression and cell

behaviour need to occur for the cell to transform from a highly regulated,

functioning part of the organ to an uncontrollably dividing and migrating cancer

cell. The aim of this study was to develop ways to monitor the function of

differentiated kidney cells, to analyse the effect of different growth environments

on gene expression and resistance to apoptosis, and to elucidate the role of Src

activation in malignant transformation.

The research questions asked in the current study were:

1. How do electrophysiological parameters regulate the transport capacity of the

kidney epithelial cells?

2. How does the gene expression profile of the epithelial cells change when

cultured in different environments? How does the activation of Src affect the

gene expression?

3. How does the growth environment of the cell affect its susceptibility to

apoptosis and what are the crucial players in this process?

4. How does the activation of Src initiate the malignant transformation and how

does this affect the cell phenotype and lumen maintenance?

64

65

4 Materials and methods

The materials and methods used in this thesis are summarized in the table below.

Detailed information with references can be found in the original papers I-III.

Table 1. Materials and methods used in the original articles.

Method Original article

2D cell culture I, II, III

3D cell culture I, II, III

Suspension cell culture III

Manipulation of membrane potential and ion transporters I

Cell biological experiments I, II, III

Microarray analysis II

RT-PCR II

Western blotting I, II, III

Immunostaining of 2D cultured cells I, II, III

Immunostaining of 3D cultured cells I, II, III

Live-cell imaging of lumen and cyst size II, III

Live-cell imaging of mitochondrial activity II

Confocal microscopy I, II, III

Quantification of lumen, cyst and cell size and cell height I

Statistics I, III

4.1 Cell lines and experimental procedures

Madin-Darby canine kidney cells (MDCK II) are a well-established epithelial cell

line, widely used when studying epithelial properties of cells or cyst forming.

Dukes et al. (2011) listed clear apico-basal polarity, well-defined cellular junctions,

rapid growth rate and the ability to polarize both in 2D and 3D environments as

some of the reasons for the popularity of the MDCK II cell line.

CD59, also known as protectin, is a lipid anchored membrane protein which

protects the cells from complement activation of the innate immune system. In this

study, MDCK II cells that expressed CD59 joined to Venus-fluorophore (CD59-

Venus MDCK) were used. In this cell line, CD59-Venus localizes to the apical

membranes and is also secreted into the lumen, making it a fitting model for

studying lumen formation or expansion in 3D environment (Rivier et al. 2010;

Castillon et al. 2013).

Behrens and co-workers (1993) created an MDCK cell line which is

temperature-sensitive (ts-Src MDCK), i.e. the viral-Src is located in the perinuclear

66

region of the cell when cultivated at 40.5°C (non-permissive) but is translocated to

the cell periphery when the temperature is lowered to 35°C, and can initiate the cell

transformation (Frame et al. 2002). The transformation process is rapid; changes in

cell phenotype can be detected within one hour after the temperature shift, and

phosphorylation of Src target proteins can already be detected 10 minutes after the

shift (Behrens et al. 1993, de Diesbach et al. 2008). This cell model has given

tremendous insights into the transformation process of epithelial cells, as it allows

studying it with an easily controlled, single oncogene-induced start.

4.1.1 Use of MDCK cell lines

MDCK and Venus-CD59 MDCK cells were grown in cell incubators at 37°C

together with Eagle’s minimal essential medium with Earle’s salts (EMEM) and

supplemented with 10% foetal bovine serum (FBS), 100U penicillin and 100 µg/ml

streptomycin. Ts-Src MDCK cells were cultivated similarly, but at 35°C or 45°C

instead of 37°C and in Dulbecco’s modified medium (DMEM), supplemented as

the EMEM. For experiments, 2.5% trypsin in PBS was used to detach the cells from

the cell culture flask. Cells were counted and the desired amount was transferred

into various growth environments.

The cellular environment has a significant effect on cell behaviour. Extra-

cellular matrix cues are essential for cell polarity and proper formation of apico-

basal axis (O’Brien et al. 2002). MDCK cells covered with ECM (3D environment)

form cysts with the basal membrane facing the ECM and the apical membrane

facing the lumen. However, MDCK cells grown on Matrigel, without being fully

engulfed by the ECM, do not form cysts. They are different from cells grown on

plastic, and this environment can be called 2½D to distinguish it from 2D and 3D.

2D½, or thin coating, has been used in the culturing and maintaining of embryonic

stem cells and induced pluripotent stem cells to support cell attachment and

expansion. To study the effects of cells grown without any kind of ECM contact,

the cells were forced to grow in suspension by plating them on bacterial culture

plates. In addition, β1 integrin blocking antibody (AIIB2) was used to prevent the

cells from attaching to each other.

In the present study, the cells were cultivated in suspension (1D), on plastic

(2D), on top of a layer of Matrigel (2½D), and encased in Matrigel or a mixture of

collagen and Matrigel (3D). By changing the culture environment and comparing

the cell phenotypes and protein expression levels, the suitability of these cells as 1)

a kidney model and 2) cellular transformation model was analysed. Also, factors

67

affecting differentiation, transportation properties and apoptosis were of special

interest.

4.1.2 Image collection and analysis

When studying dynamic processes, live cell imaging is pivotal. In live cell imaging,

cells are provided with an environment that supports their growth and well-being.

Usually this means an incubator with the correct temperature and gas exchange,

mounted on top of the microscope. Live cell imaging is always balancing between

acquiring the most and the best possible quality data, and cell survival and well-

being. In vivo, most cells live in complete darkness, and are thus susceptible to

light-induced damage, called phototoxicity. Spinning-disc confocal microscope

allows acquiring images swiftly, but still with good resolution, and thus reduces the

irritation caused to the cells.

Live cell imaging has been extensively used in this study by acquiring Z-stacks

of the cysts and combining these stacks to form 3D models, at several time points

(4D). This technique, together with live cell fluorescent markers for plasma

membrane (TMA-DPH and CellMask Orange), apoptosis (FITC-Annexin V,

propidium iodide, Hoechst) and mitochondria (Mitotracker Green and Orange), and

with careful regulation of the extracellular environment and quantitative analysis,

is a powerful tool for studying relatively fast (occurring within a day) events with

high resolution and low phototoxicity. In this study, 4D live cell imaging was used

to acquire new information about how the MDCK cell ion transport functions, how

the phenotype changes when they undergo transformation process and how they

survive when deprived of cell-cell and cell-matrix contacts. For live cell imaging,

cells were transferred into glass bottom dishes. The live cell markers were added

15-30 min before visualizing the cell; during the experiments, the cells were kept

in the incubator chamber of the microscope at either 35 or 37°C. In the experiments

where lumen volume was measured, the MDCK cells were grown on regular plastic

dishes with cover slips under the Matrigel. For the experiment, the cover slips were

carefully picked up and placed upside down against the glass of a glass-bottom dish.

This ensured that the working distance of the object was sufficient for the

visualization of the whole cyst.

Even in the age of high-speed, high-resolution live cell imaging, more

traditional fixing of cells and dyeing them with antibody-conjugated fluorophores

still has a place. Compared to markers usable in live cell imaging, a considerably

wider variety of antibodies that can be utilized when working with fixed cells is

68

available. Also, since the cells are already dead, much higher exposure times and

laser intensities can be used to produce sharper images with better signal-to-noise

ratio. In this study, fixed cells were utilized to characterize the differences in

phenotypes between untransformed MDCK and ts-Src transformed MDCK cells,

especially pertaining to the localization of E-cadherin. The cells for the experiments

performed with fixed cells were grown on microscopy slides fitted with a silicon

ring adapter. After the cells had reached the desired size and the experiments were

performed, the cells were fixed and stained. The silicon ring was only removed

when the cells were ready to be mounted with Immu-Mount, and a cover slip was

placed on them.

69

5 Results

5.1 The effects of ionic environment on water secretion and re-

absorption (I)

How the ionic environment affects the water secretion and re-absorption by the

MDCK cell cysts was studied by switching the normal media of cells cultivated in

Matrigel to buffers lacking monovalent cations or chloride and/or supplementing

the buffer with APIs activating or inhibiting the functions of plasma membrane

channel proteins.

Due to the apical localization of CD59, Venus-CD59 MDCK cysts, coupled

with a plasma membrane marker, TMA-DPH, enable the quantitative analysis of

the lumen, cyst and cell sizes (Fig. 1 in I). Venus-CD59 MDCK cells were utilized

to study the transepithelial transport in an environment where variables were the

ionic composition of the basal side that affects transepithelial potential and the

activation and inhibition of Cl- channels. In addition, nigericin was used to

influence the permeability of the cell membranes to cations.

Venus-CD59 MDCK cell cysts, grown in Matrigel, tolerated well laser

illumination by spinning disc confocal microscope, and the cyst and lumen sizes

remained constant for 10 hours when kept in EMEM medium supplemented with

10% foetal bovine serum (FBS), which contained all the essential nutrients and

growth factors the cells need. When the medium was replaced with HBSS lacking

these nutrients and growth factors, the cysts collapsed after four hours (Fig. 2 and

3, and Table 2 in I). Therefore, the cysts were continuously monitored for up to 2.5

hours in the following experiments.

5.1.1 Basal fluid lacking in monovalent cations induces water influx

into the lumen (I)

When the growth medium of Venus-CD59 MDCK cells grown in Matrigel for four

days was replaced with tetramethyl ammonium chloride (TMACl), a buffer lacking

monovalent cations, the cells were hyperpolarized due to the efflux of monovalent

cations from inside the cells to the basal fluid. Hyperpolarization is known to

induce Cl- efflux into the lumen, and it is likely that this efflux happened in the

presence of TMACl, as well. This, in turn, caused a rapid influx of water to the

cells and its transepithelial exit into the lumen. Subsequently, lumen volume was

70

increased 2.2-fold, and this was observed together with 70% increase in apical

membrane surface area. Part of the water remained in the cells, causing 24%

increase in cell volume and 22% increase in basal membrane surface area within

2.5 hours. Cell height, however, remained constant, indicating that the increase in

cell volume was compensated with the opening of some of the many folds in the

apical membrane instead of reduction of the length of the lateral cell walls or

changes in the cell shape (Fig. 3, 4 and 5B,F and Table 2 in I).

5.1.2 Basal fluid lacking in chloride ions induces water re-absorption

(I)

If the cysts were transferred into chloride-free buffer by incubating them in sodium

gluconate solution, chloride ions were transported out of the cells according to their

concentration gradient, resulting in depolarization of the cell interior. This, in turn,

is followed by water efflux from lumen, through the cells or the paracellular

pathway, and into the basal fluid. The phenomenon was seen as a reduction in

lumen size by 42% while the cell volume increased by only 8% (Fig. 5C,G and 6,

and Table 2 in I).

5.1.3 Sodium gluconate does not affect the tight junctions (I)

Gluconate is capable of buffering Ca2+, and sodium gluconate solution without

added CaCl2 would cause the opening of tight junctions. In this case, the opening

of tight junctions would be the obvious reason for the lumen shrinkage. This was

compensated by adding extra Ca2+ into the gluconate solution, which preserved the

intact tight junctions, as shown by confocal microscopy of the tight junction

component occludin and adherens junction component E-cadherin.

5.1.4 Complete depolarization of the cells causes cell swelling (I)

Nigericin is an ionophore that can be used to depolarize the cells by allowing K+ to

pass freely through the cell membrane according to its concentration gradient. A

smaller reduction in the transmembrane potential of the cells can be achieved by

replacing the cell culture medium with KCl buffer. Hence, in the presence of KCl,

the cell size in the cysts remained constant while lumen size decreased. When cells

were exposed to both KCl and nigericin, it caused a similar decrease in lumen size

as KCl alone, but also led to an increase in overall cyst size due to cell swelling

71

caused by the Donnan effect. These KCl experiments showed that lack of sodium

alone was not sufficient to create a driving force for chloride secretion and water

transport in the way the lack of both sodium and potassium (TMACl) did (Fig.

5D,H and 6, and Table 2 in I).

5.1.5 Inhibiting or activating chloride channels leads to changes in

lumen, cyst and cell size (I)

Water transport into and out of the cells is highly dependent on the activity of the

chloride channels. The goal was to analyse more carefully which chloride channels

were involved in the lumen expansion. Forskolin, a cAMP agonist, induced a 1.3-

fold increase in lumen volume and 1.2-fold increase in luminal surface area.

Lubiprostone, an activator of CFTR and ClC-2 chloride channels, was able to

induce a 1.5-fold increase in lumen volume and 1.3-fold increase in luminal surface

area (Fig. 3 and 5A,E, and Table 2 in I). The results were similar, if not as

pronounced, as with TMACl, but without the cell swelling that accompanied the

TMACl treatment. Hence, TMACl-induced hyperpolarization and chloride channel

activation by forskolin or lubiprostone both lead to increased chloride secretion and

subsequent transepithelial water flow, which can be seen as lumen expansion due

to the increased intraluminal hydrostatic pressure. The dependence of chloride

secretion on transmembrane potential was elucidated by treating the cysts with

lubiprostone while in KCl buffer with added nigericin. Even under these

depolarized conditions, chloride channel activation via lubiprostone increased the

lumen volume by 1.5-fold and luminal surface area by 1.3-fold (Fig. 5D,H and 6,

and Table 2 in I).

Inhibition of chloride channels, on the other hand, had a negative effect on

TMACl-induced lumen expansion. CFTRinh-172, a chloride channel inhibitor that

affects CFTR and VSORCC, completely annulled the lumen expansion. CaCCinh-

A01, an inhibitor of anoctamin 1, was able to only slow down the lumen expansion

(Fig 3 and Table 2 in I).

The manipulation of the cell environment showed that the presence of forskolin,

hyperpolarization of the cells or activation of chloride channels, particularly ClC-

2 and VSORCC, induced lumen expansion while depolarization of the cells caused

lumen shrinkage. The results are summarized in Table 2 and in Fig. 8.

72

Table 2. Manipulation of the cell environment and the effects of active pharmacological

ingredients on water fluxes and cell and lumen volume.

Manipulation of basal

fluid

Basal water flux Apical water flux

Forskolin in HBSS ↑ In Out

TMACl ↑ In Out ↑

TMACl + CFTRinh-172 ↓ - In

TMACl + CaCCinh-A01 ↑ In Out

Lubiprostone in HBSS ↑ In Out

Lubiprostone in KCl +

nigericin

↑ In Out ↑

Na-gluconate ↓ - In

KCl ↓ - In

KCl + nigericin ↓ In In ↑

To calculate changes in lumen and cell sizes, a layer from the Z-stack where the

surface area of the cyst and lumen were the largest was selected for each time point.

This maximal cross-sectional area of the lumen and the cyst in the selected layer

was measured and mean values and standard errors were calculated. Based on the

information that both the cyst and lumen are spherical, the radii of the cyst and

lumen could be calculated. From this information, the volume and surface area of

these two could be calculated, and the volume of the cells was obtained by

subtracting lumen volume from cyst volume.

73

Fig. 8. Schematic presentation of the effects of cell environment constituents and active

pharmacological ingredients on lumen, cyst and cell size.

5.2 Src-induced events in MDCK cells (II and III)

With its abundant expression in all cells and numerous substrates, whose function

affects the fate of the cell from proliferation, attachment and migration to apoptosis,

Src tyrosine kinase is an essential player in malignant transformation and cancer

progression.

The expression levels of Src tyrosine kinase were studied using antibodies for

Western blotting that recognize canine c-Src, active Src (both c-Src and v-Src) and

the avian v-Src present in ts-Src MDCK cells.

The levels of c-Src and v-Src in ts-Src MDCK cells grown in 2D were similar

between the permissive and non-permissive temperatures. However, only ts-Src

MDCK cells cultivated at the permissive temperature expressed the active form of

Src, phosphorylated at the tyrosine 416. Hence, v-Src is expressed at both

permissive and non-permissive temperatures, but is mostly activated at 35°C. Some

active Src was also detected in ts-Src MDCK cells grown at non-permissive

temperature, indicating that the system is not able to completely prevent Src from

activating. Untransformed MDCK cells did not express avian v-Src (Fig. 2B in II).

74

5.2.1 Phenotypes of untransformed MDCK and ts-Src transformed

MDCK cells in different culture systems analysed using

confocal fluorescence microscopy (II and III)

The phenotype of ts-Src MDCK cells cultivated at non-permissive temperature was

similar to untransformed MDCK cells. The activation of temperature-sensitive v-

Src drastically affected the E-cadherin localization. As seen in immunofluorescence

pictures, in confluent ts-Src MDCK cells grown in 2D at non-permissive

temperature, E-cadherin was correctly co-localized with α-catenin, β-catenin and

p120 catenin at lateral walls. Switching to permissive temperature caused a

complete reorganization of junctional complexes. Also, the cell shape was

drastically changed towards a more fibroblast-like phenotype. E-cadherin was seen

in vesicles of variable sizes in the cytoplasm and in the perinuclear areas, co-

localizing with β-catenin and occasionally with p120 catenin (Fig. 4A,B in II).

When untransformed MDCK cells were cultivated in 3D environment

consisting of collagen I and/or Matrigel, they formed spherical cysts with a hollow

space, lumen, in the middle. Actin was localized to the apical membranes facing

the lumen. E-cadherin and β-catenin delineated the lateral walls. ZO-1, a tight

junction marker, was localized to the cell-cell junctions on the apical side of the

untransformed MDCK cell cyst. Some apoptotic cell remnants were also seen

inside the lumen, as indicated by the antibody specific to the cleaved form of

caspase 3. These were fragments of inner cells that were forced to go into apoptosis

during cyst and lumen formation (Fig. 4C in II). Ts-Src MDCK cells grown at

permissive temperature, however, formed loosely spherical clusters without a

lumen. Cells facing the ECM had a different phenotype than the inner cells. The

former were cubic in shape and had E-cadherin delineating the lateral membranes

and basal surfaces, while the latter had a rounded shape and showed no apico-basal

axis as evident by actin localizing to all membranes. ZO-1 was only seen in short

fragments and no apoptosis was detected (Fig. 4D in II). However, the Src-induced

changes were reversible, and after switching the cysts back to the non-permissive

temperature, the cells slowly regained their epithelial characteristics by recovering

the apico-basal polarity and starting to clear the lumen after one day at non-

permissive temperature, relocating E-cadherin to the cell membranes by day two

and having a fully cleared lumen with actin at the apical membranes and E-cadherin

on the lateral walls by day six (Fig. 1B in III).

Cells growing on top of a layer of Matrigel (2½D) formed monolayers similar

to 2D, but with areas of early luminal sites where the apical surfaces formed around

75

areas devoid of cell-cell contact. E-cadherin was on the cell membranes (Fig. 1A in

III).

The early events of the Src-induced transformation process were monitored

using live cell imaging. Ts-Src MDCK cells were cultivated inside Matrigel at the

non-permissive temperature to form cysts with lumen. The temperature was then

shifted to permissive and the events in live cells were monitored using a spinning

disc confocal microscope. Using fluorescent markers for plasma membranes, it was

evident that the change in phenotype started with the opening of cell-cell junctions.

This occurred within one hour after the temperature shift. Following opening of the

junctions, lumen shrinkage occurred, presumably due to water leakage through the

opened tight junctions and the following loss of intraluminal hydrostatic pressure

that kept the lumen open. Thus, during the early stages of transformation of ts-Src

MDCK cells, the lumen was not lost due to the migration of cells into it but due to

collapsing of the lumen (Fig 2A in III). E-cadherin remained on the cell membranes

after 4h at permissive temperature, but in addition to lateral cell membranes, E-

cadherin staining was detectable on other membranes as well (Fig. 2B in III).

5.2.2 Expression of cadherins in MDCK and ts-Src MDCK cells in

different culture environments (II)

In addition to E-cadherin, MDCK cells have also been reported to express low

levels of N-cadherin and cadherin-6 (also called K-cadherin) (Youn et al. 2005,

Stewart et al. 2000). Western blot analysis was used to analyse the expression of E-

cadherin at protein level. Antibodies to both cytoplasmic domain and extracellular

domain (rr1) of E-cadherin and to all cadherin family members (pan-cadherin) were

used. In 2D, both untransformed MDCK and ts-Src transformed MDCK cells had

similar expression levels of E-cadherin, and Src activation did not induce the

expression of other cadherins. In a mixture of collagen I and Matrigel, the

expression levels of E-cadherin in both cell lines were remarkably similar. The only

difference was a band in untransformed MDCK cells in 2D, made visible with pan-

cadherin antibody, which had a slightly larger molecular size than E-cadherin. This

band could indicate N-cadherin, cadherin-6 or a propeptide form of E-cadherin (Fig.

4E in II).

76

5.2.3 Microarray analysis revealed the differences in gene

expressions of cells cultivated in different environments (II)

Microarray analysis is a powerful tool to study expression levels of a very large

number of genes simultaneously. It can be used to identify those genes that are

changed when a certain pathogen infects the cells when testing new drugs or

studying the effects that different growth environments have. Microarray analysis

of untransformed MDCK and ts-Src MDCK cells was used to measure gene

expression levels in three different circumstances:

1. ts-Src MDCK cells grown at 40.5°C vs. ts-Src MDCK cells grown at 35°C

2. untransformed MDCK cells grown at 35°C vs. ts-Src MDCK cells grown at

35°C

3. ts-Src MDCK cells grown at 35°C in the presence of Src inhibitor pp2 vs. ts-

Src MDCK cells grown at 35°C.

All three measurements were done both in 2D and 3D. Microarray analysis revealed

hundreds to thousands of changes in gene expression in each case. Most changes

were observed in cases where pp2-inhibitor was used, due to it inhibiting several

Src-family kinases in addition to c-Src and v-Src (Bain et al. 2003), and due to the

multiple functions Src has in healthy cells (Frame et al. 2002). Least changes were

detected in the comparison between ts-Src MDCK cells at permissive temperature

and ts-Src MDCK cells at non-permissive temperature. Only changes in gene

expression that occurred in the same direction (increased or decreased) in each

three cases and that had at least once a 5-fold or bigger increase were chosen for

further analysis. These changes were considered to be results of Src activation (Fig.

1A in II).

In 2D environment, the Src activation caused a sufficient change in only six

genes. Surprisingly, none of these were catenins or cadherins. However, the cell

signalling pathways are often controlled by modifications affecting the ability of a

protein to interact with others due to tyrosine phosphorylation, whereas protein

expression levels remain stable. In this case, as well, it is likely that the activation

of Src affected the ability of catenins and cadherins to interact at the cell junction

sites, thus leading to the observed changes in phenotype (Fig. 1A in II).

Cells grown in the much more complex 3D environment are closer to their

natural state, and the results obtained from microarray analysis reflect this

complexity. All in all, 114 significant changes were observed between ts-Src

MDCK cells grown in Matrigel at permissive temperature and at non-permissive

77

temperature. The majority of these changes were related to energy metabolism, cell

signalling and cell division. The expression of cytoskeletal components actin and

Arp2/3 was also decreased. Two p120 catenin-binding proteins, Kaiso and Nanos,

both members of the zinc-finger protein family, had increased expression; a change

that might be relevant in malignant transformation (Fig. 1A in II).

The effect of the 3D Matrigel environment on cells was enormous when

microarray results from untransformed MDCK cells grown in Matrigel were

compared to untransformed MDCK cells grown in 2D. The complex 3D matrix

induced a change in the gene expression of 6,474 genes. To obtain more specific

results caused by the 3D environment, all the genes also altered by v-Src were

removed from the results. The resulting 118 genes included diminutive changes in

the expression levels of cadherins or other junctional proteins and more impressive

changes in the expression levels of rab proteins. The expression levels of

rho/racGEF and GDI were decreased. Interestingly, the expression level of survivin,

an apoptosis inhibitor, decreased (Fig. 2 in II).

To confirm the microarray results, the expression levels of rab5, rab7, rab8 and

survivin of untransformed MDCK and ts-Src MDCK cells, grown both in 2D and

3D, were measured using RT-PCR. It showed only very minor differences between

MDCK and ts-Src MDCK cells grown in 2D. However, when comparing 2D

MDCK cells to 3D MDCK cells, there was a strong change in gene expression: 3D

environment induced a 3.7-fold increase in rab5, 3-fold decrease in rab7, 5-fold

decrease in rab8 and 5.5-fold decrease in survivin. In ts-Src MDCK cells, the

expression levels of all of these proteins increased compared to 3D MDCK cells

(Fig. 3 in II).

5.2.4 Src-induced functional changes in the mitochondrial activity

and E-cadherin endocytosis in ts-MDCK cells (II)

Cancer cells have vastly altered metabolism compared to healthy cells. One of

characterizing features of the metabolic anomalies is favouring glycolysis over

aerobic respiration and resistance to apoptosis caused by permeabilized

mitochondrial membranes (Kroemer & Pouyssegur 2008). Mitochondrial activity

can be monitored using mitochondria-specific fluorescent markers suitable for live-

cell imaging. Two such dyes are MitoTracker Green and MitoTracker Orange CM-

H2TMRos. Both accumulate to mitochondria; the former is fluorescent in all

mitochondria, the latter is only fluorescent when the molecule is oxidized by active,

respiring mitochondrion. When used together, a ratio of MitoTracker Orange CM-

78

H2TMRos to MitoTracker Green can be calculated. This gives a mitochondrial

metabolic index that represents the proportion of actively respiring mitochondria

in all mitochondrial mass.

In untransformed MDCK cells grown in 3D, the starting ratio was high, but

then rapidly declined as a result of the loss of membrane potential and/or increase

in mitochondrial membrane permeability (Fig. 5A in II). The decline in the index

coincided with cyst formation and expiration; the ratio was high when the cell cyst

was growing swiftly, but started to decline as the cyst matured, finally reaching its

lowest point when the cells started going into apoptosis. Ts-Src MDCK cells had a

higher mitochondrial metabolic index than untransformed MDCK cells for up to

seven days (Fig. 5B in II). This indicated that active v-Src protected MDCK cells

from the mitochondrial membrane permeability crucial for apoptosis.

Even if Src activation has no effect on E-cadherin protein expression levels, it

can still affect the rates of cadherin recycling to the cell membranes that is not

evident by expression levels. To study this, immunofluorescence of E-cadherin and

β-catenin in ts-Src MDCK cells was used. Cells were grown in 2D at non-

permissive temperature until confluent. Subsequently, the cells were labelled with

rr1 antibody while on ice, and the cells were transferred to 40.5°C or 35°C for 30

or 60 min. After the temperature shift, the cells were fixed and antibodies were used

to determine the localization of β-catenin as well as rr1. These immunofluorescence

experiments showed that E-cadherin internalization occurred both at permissive

and non-permissive temperatures. Due to the fact that the co-localization

coefficient diminished after 60 min at 40.5°C and cadherin returned to the cell

membrane it can be concluded that more E-cadherin had returned to the cell

membranes at the non-permissive temperature of 40.5°C than at the permissive

temperature (Fig. 6 and Table 1 in II).

Quantitative co-localization analysis, using co-localization coefficient, reveals

the portion of pixels in the region of interest (ROI) where signals from two different

channels both contribute. Co-localization analysis of E-cadherin and β-catenin at

permissive and non-permissive temperatures revealed that the co-localization

coefficients were higher in 35°C (Table 1 in II). This suggests that, at permissive

temperature, adherens junctions are disintegrating and both E-cadherin and β-

catenin are internalized together. At non-permissive temperature, the co-

localization coefficient for apical E-cadherin was very low, and this might be

interpreted as these vesicles belonging to the clathrin-mediated recycling route, in

which catenins are released from E-cadherin upon internalization (Delva &

Kovalczyk 2009).

79

5.2.5 Relation of survivin expression to the cell phenotype and

occurrence of apoptosis in different conditions (II and III)

Adult, healthy cells rarely express survivin; in fact, its expression in cancer cells is

a sign of bad prognosis (Altieri 2015). Microarray analysis displayed reduction in

survivin expression levels when untransformed MDCK cells were grown in 3D,

compared to 2D. RT-PCR experiments showed that this decrease in gene expression

is translatable into decreased mRNA production. The expression of survivin protein

levels in various culture conditions was analysed using Western blotting.

In untransformed MDCK cells, survivin was expressed when the cells were

cultivated on plastic plates. When grown in 2½D or 3D, survivin was

downregulated, and was lowest in MDCK cells grown in a mixture of collagen and

Matrigel (Fig. 4F in II). In 2½D, survivin was seen in the nucleus or, depending on

the cell-cycle, between the sister chromatids as part of the CPC. During cytokinesis,

survivin remained in the midbodies left behind the separating sister cells (Fig. 1D

in III). Interestingly, even though the survivin levels were very low in MDCK cells

grown in 3D, the cells avoided apoptosis.

Ts-Src-MDCK cells were remarkably akin to untransformed MDCK cells in

phenotype when grown at non-permissive temperature (Fig. 1B in III). The

expression of survivin in ts-Src MDCK cells was low when cultivated inside

Matrigel at non-permissive temperature. A shift to permissive temperature shows

high expression of survivin in 2D and moderate expression in 3D Matrigel (Fig. 4F

in II). In ts-Src MDCK cells grown in 3D, once Src was activated, survivin

expression was increased within two hours and correlated with the disintegration

of cell junctions (Fig. 2B and C in III). This phenomenon has been shown to be

linked to the phosphorylation of catenins and junctional proteins by Src (Palovuori

et al. 2003). On the other hand, survivin expression of ts-Src MDCK cyst

transferred from permissive to non-permissive temperature slowly waned as the

cells took on a more differentiated phenotype and arranged into a single layer of

cells, with the lumen in the middle (Fig. 1B and C in III).

Growing in suspension causes anoikis in MDCK cells but not in ts-Src

MDCK cells (III)

Untransformed MDCK cells forced to grow on bacterial plates clustered together

and were more resistant to anoikis, with only 14% of the cells suffering from it (Fig.

3A and 4A in III). This was probably due to their ability to secrete ECM

80

components (Erickson & Couchman 2001; Mathias et al. 2010). However, if a β1

integrin blocking antibody (AIIB2) was added to the growth medium, MDCK cells

were not able to attach to the ECM they produced, and 25% underwent anoikis (Fig.

3B and 4A in III). The occurrence of early signs of apoptosis increased from 41%

in cells without AIIB2 to 56% in cells with the blocking antibody. Whether with or

without AIIB2, the untransformed MDCK cells did not express survivin (Fig. 3E

in III). Ts-Src MDCK cells, on the other hand, were resistant to anoikis even when

AIIB2 was added, with the occurrence of early signs of apoptosis of 20% of cells

with and 30% without the blocking antibody (Fig. 3C,D and 4A in III). Ts-Src

MDCK cells proliferated more actively while in suspension than untransformed

MDCK cells, and formed large clusters. The differences between occurrences of

both stages of apoptosis were significant when comparing untransformed MDCK

cells to ts-Src MDCK cells. Ts-Src MDCK cells also expressed survivin both with

and without the blocking antibody (Fig. 3E in III). Thus, in suspension, survivin

expression correlated with cell proliferation and the lack of expression with anoikis.

Lack of survivin expression and cell-cell junctions causes anoikis in MDCK

cells in 3D and 2½D, but not in 2D (III)

As shown earlier, untransformed MDCK cells were able to proliferate and

differentiate in 3D without going into apoptosis, even though they did not express

survivin. The correlation between intact cell-cell junctions, survivin expression and

apoptosis was tested with the introduction of EGTA. As a calcium chelator, EGTA

disrupts the formation of cell junction complexes by sequestering Ca2+ away from

them. Untransformed MDCK cells were grown either on plastic cell culture plate

(2D), on top of a layer of Matrigel (2½D) or inside Matrigel (3D), and their fate

after introducing EGTA to the culture was monitored with spinning disc confocal

microscope (Fig. 5A,B and C in III). Cells grown in 2D were largely resistant to

apoptosis, and apoptosis was not detected within the 5h observation period.

However, 2½D and 3D were very susceptible to EGTA-induced apoptosis. After 2h

in EGTA, the 2½D cells started shedding, and after 4h massive amounts of cell

remnants were seen. In 3D, apoptotic bodies appeared even earlier, within three

hours after EGTA supplementation, and after five hours, the majority of the cells

had entered apoptosis. In 2D, survivin expression increased the longer the cells

were exposed to EGTA, reaching the peak value in 4h. In 2½D, survivin levels were

high, but dropped quickly after EGTA supplementation and then remained low. In

3D environment, on the other hand, survivin expression remained constantly low

81

during the EGTA treatment (Fig. 5D in III). Taken together, cells grown in 2D were

more resistant to anoikis and upregulate their survivin expression, whilst the

presence of EMC alone was not sufficient to prevent apoptosis in the absence of

cell junctions, E-cadherin trans-interactions and survivin expression. In conclusion,

only in 3D environment did untransformed MDCK cells not undergo apoptosis

when survivin was downregulated, but apoptosis occurred in these cells when their

E-cadherin trans-interactions where broken. The effects of culture conditions on

the phenotype of the untransformed and ts-Src MDCK cells are presented in Table

3.

5.2.6 Ts-Src MDCK cells are more susceptible to ROS than

untransformed MDCK cells, and can be rescued by antioxidant

supplementation (III)

Reactive oxygen species (ROS) are generated in cells as by-products of cellular

metabolism, but they also have essential roles in signal transduction pathways.

Harmful levels of ROS can cause alterations in cellular functions or even cell death

due to oxidative damage (Trachootham et al. 2009). Cancer cells have greatly

increased ROS production due to their altered metabolism, and are thus more

susceptible than normal cells to drugs that increase the levels of ROS than normal

cells. Piperlongumine (PL) is a naturally occurring small molecule that was found

to increase ROS levels in cancer cells, but not in healthy cells, by targeting the

proteins participating in the cellular response to oxidative stress (Raj et al. 2011).

When untransformed MDCK cell were cultivated together with 15 µM PL, the

cells exhibited a distorted phenotype, but only low level (15%) of apoptosis was

detected (Fig. 4B and 6B in III). In ts-Src MDCK cells the cell clusters started to

disintegrate and apoptotic fragments appeared, with 25% of the cells being cleared

via apoptosis (Fig. 4B and 6B in III). 50 µM PL caused both cell types to swell and

undergo necrosis (Fig. 6C in III). N-acetyl cysteine (NAC), an antioxidant and a

reducing agent, alone had no effect on the phenotype of either of the cell types, but

rescued ts-Src MDCK cell from PL-induced apoptosis (Fig. 6D and E in III).

Piperlongumine treatment of ts-Src MDCK cells caused a change in E-

cadherin localization and downregulation of survivin (III)

In control ts-Src MDCK cells grown at the permissive temperature inside Matrigel,

E-cadherin localized to the cytoplasm and along the cell membranes. No clear

82

apico-basal axis was exhibited by these cells, nor did they form lumen. PL-

treatment changed the cell morphology. Cells appeared swollen and non-polarized,

and E-cadherin clustered to the membranes in thick bands. In the presence of NAC,

however, the cells reverted back to the epithelial phenotype and E-cadherin

localized to the lateral cell membranes in narrow bands. The phenotype of the cells

was the same whether in the presence of NAC or in the presence of both NAC and

PL (Fig. 7A,B and C in III).

PL downregulated survivin expression and induced apoptosis. NAC rescued

the cells from PL-induced apoptosis, but was not able to revert the downregulation

of survivin caused by PL (Fig. 7D in III).

Cell behaviour in different environments can be classified into three categories:

differentiated, proliferative and apoptotic phenotype. The expression of survivin,

susceptibility to apoptosis, and appearance of focal adhesions and adherens

junctions are the hallmarks of each category. The growth conditions in which these

phenotypes appear in MDCK and ts-Src MDCK cells are presented in Table 3. The

results show that the prerequisite for differentiated phenotype is the existence of

trans-interactions of E-cadherins. The proliferative phenotype dominates in 2D or

in any environment when Src is activated. The apoptotic phenotype required

external factors that were capable of disintegrating cadherin trans-interactions or

increase ROS production.

Table 3. The effect of culture conditions on the phenotype of the untransformed and ts-

Src MDCK cells classified into three different phenotype categories.

Phenotype MDCK cell culture condition ts-Src MDCK cell culture condition

Differentiated phenotype

3D, 2½D 3D+PL+NAC

Proliferative phenotype

2D 1D, 2D, 3D

Apoptotic phenotype 1D, 3D+EGTA, 2½D+EGTA 3D+PL

5.2.7 Summary

In this study, sensitive 4D imaging of living cells and quantitative analysis was used

to validate the use of MDCK cysts grown in Matrigel as a fitting model for

functional analysis of cell behaviour, especially when conducting research on

processes occurring within minutes or hours, rather than days. Modern microscopy

enables reliable and sensitive monitoring of such processes, producing supremely

83

valuable data for testing the response of the cell cysts to drugs, toxins or

physiological changes, as well as in processes such as malignant transformation.

84

85

6 Discussion

Epithelial cells have a multitude of roles in basically all of the functions of the body

of a multicellular organism. Their ability to attach to each other and form barriers

that compartmentalize organs and separate them from each other, together with the

plasticity to enable alterations in their functions, make them indispensable for a

functioning organism. Due to their importance, the understanding of their structure

and functions is pivotal, yet much is still unclear.

The aforementioned plasticity is one of the most fascinating aspects of

epithelial cells. As barriers, they must work together and form strong enough bonds

to withstand the physical forces affecting them when the organism moves, and keep

up homeostasis by disallowing free entry of extracellular constituents. Yet, they

cannot be completely impermeable and can rarely maintain constant permeability

to the same molecules, thus requiring a carefully regulated signalling network that

allows changes in plasma membrane permeability according to the needs of the

organism.

The carefully regulated functions of epithelial cells, together with their

abundance, make them also susceptible to mutations leading to malignant

transformation. Against this background, it is not surprising that 90% of all cancers

are epithelia-derived carcinomas. Epithelial cells, by nature, are not exceedingly

migratory or invasive, and thus, for malign carcinomas to develop, they need to

adopt aspects of cells that are more capable in these functions, such as

mesenchymal cells.

Complex signalling networks are needed for regulation of cellular

differentiation and formation of epithelial tissue junctions. Failure in the regulatory

processes can lead to severe metabolic diseases and/or carcinomas. In the present

work, the aim was to develop a simple in vitro model system which enabled

sensitive and detailed analysis of kidney epithelial cell phenotype and behaviour

and the consequences of the activation of an oncogene.

6.1 Plasticity of epithelial cells can be seen in the rapid response

of MDCK cells to changes in ionic environment

Secretory and absorptive organs, kidney included, rely on the asymmetrical

distribution of transporters and channel proteins on their apical and basal

membranes. The expression of these types of proteins varies from tissue to tissue,

and even in different parts of the same organ, such as kidney proximal and distal

86

tubule. In addition, the localization and the activity of these transport proteins vary

depending on the circumstances. Indeed, it seems likely that different parts of the

kidney are able to function in different roles when the preferred method of kidney

function, glomerular filtration, is compromised, i.e. proximal tubule cells are able

to secrete as well as absorb. This redundancy becomes advantageous, when facing

acute or chronic renal failure or hypertonic dehydration.

Cell cultures have been a tremendously successful tool when studying

epithelial secretory and absorptive properties. In these functions, ion transport has

a pivotal role. Traditionally, cellular ion transport properties have been studied

using Ussing chamber technique, where a monolayer of cells growing on a

permeable support covers it fully, forming a barrier between the two halves of the

chamber, thus blocking free transport of molecules through the filter membrane

underneath (Ussing 1947, Hoffmann 2001). When studying ion transport, the

potential difference or electrolyte current across the two halves of the chamber is

measured. This method is excellent when conducting studies regarding ion currents

or cell permeability, and factors affecting these functions, such as drugs, nutrients

or physical elements. Also, ex vivo epithelial samples from the tissue of interest can

be used, if available (Li et al. 2004, Li et al. 2012). A limitation of the system,

however, is that it cannot be utilized to study movement of water or generation of

hydrostatic pressure.

Culturing cells in ECM allows building highly organized, differentiated cysts

with a well-developed apico-basal axis. These cysts resemble epithelial structures

in vivo, and, due to the sealed luminal space, it can be used to study water

movement through epithelial cells. Using activators and inhibitors, different

channel proteins contributing to water movement can be analysed, and the

composition of the buffer on basal side of the cells can easily be changed (Li et al.

2004, Li et al. 2012). As a downside, changing the composition of the apical buffer

is not possible. In most previous works, quantitative analysis was carried out using

low-resolution light microscopy and only the enlargement of the cyst was measured.

Hence, the information concerning lumen and cell volumes is missing (Tanner et

al. 1992, Yang et al. 2008b, Yuajit et al. 2013, Buchholz et al. 2014).

In more detail, the present study shows that, depending on the ionic

environment of the basal fluid and potential difference across the cell membrane,

chloride ions and water are able to flow through MDCK cell cysts into the apical

lumen or out of it. Accumulation of Cl- into the luminal space creates luminal

negative potential which induces water flow into the lumen, as was seen with

TMACl and (direct or indirect) chloride channel activators. On the other hand,

87

decreasing the electrical potential difference between the basal fluid and the cells,

i.e. depolarizing it, using KCl or, for a more profound effect, making the plasma

membrane freely permeable to K+ with nigericin, results in water reabsorption from

the lumen and luminal shrinkage.

The method depicted here works also when examining the various plasma

membrane transporter proteins. Here, the roles of chloride channel activators

forskolin and lubiprostone as well as CFTR- and ANO1-inhibitors in MDCK cyst

lumen expansion were investigated. With highly specific inhibitors, other channel

proteins and their effects on lumen, cyst and cell size could also be studied,

potentially giving valuable information about the influence of drugs on cellular ion

and water transport. The cell swelling could be separated from cyst enlargement,

enabling the identification of the Donnan effect on the epithelium.

Modern live cell imaging techniques enable the collection of high-resolution

3D stacks and calculating the volumes of the cyst, lumen and cells, separately. This

gives valuable information about directions into which water and ions flow through

the epithelial layers. In the present study, a combination of techniques was used to

study living MDCK cysts over short periods of time after a change in their ionic

environment and the results were used to draw conclusions about how the changes

in transepithelial potential directly affects the volume of the cyst, lumen and cells,

without changes in cell numbers due to proliferation or apoptosis hindering the

interpretation. The results from this study show that the method used can be utilized

when studying the effects of chloride channel manipulation, hyperpolarization or

depolarization on cell cyst and lumen volume. These experiments gave support to

the earlier findings that ClC-2 and VSORCC are actively involved in chloride

secretion in MDCK cells (Cuppoletti et al. 2004, Melis et al. 2014).

6.2 Transformation of MDCK cells by v-Src causes changes in

gene expression, cell phenotype and behaviour

Tumour cells differ from healthy cells in several ways: 1) independence from

growth factors via autocrine excretion or altered activity of growth hormone

receptors; 2) anchorage-independent growth; 3) lack of contact inhibition; 4)

reduced adhesion, and 5) continued proliferation without regard to cell density

(Macdonald et al. 2004).

Kinases are prominently involved in the birth and progression of cancer. Src is

a multifunctional tyrosine kinase that has a role in all of the important aspects of

cancer, but interestingly, is rarely mutated in cancer cells. It contributes to the

88

disease by being abnormally activated by other oncogenes (Guarino 2010). In

studies of tumour progression in vivo or in vitro, it is difficult to distinguish the

critical factors affecting cell behaviour with respect to proliferation, migration or

resistance to apoptosis. Thus, in the present study, ts-Src MDCK cell line was used

due to the advantages conferred by the cells’ ability to induce transformation by a

simple temperature change, after which the activation of v-Src occurs within an

hour. This model enables the separate analysis of the migratory and proliferative

capacities of the cells, as well as the occurrence of apoptosis, within a short time

period.

The first aspect of Src function in this study was to analyse the changes

induced by prolonged v-Src activation. In MDCK cells, the temperature-induced

activation of v-Src caused a change in expression of 350 genes when cultivated in

2D, and 1570 genes when cultivated in 3D. The relative small number of changes

in 2D is a sign of malignant behaviour exhibited already by inactive v-Src in ts-Src

MDCK, probably due to the temperature-inactivation of v-Src not being complete,

and thus exhibiting a “leaky” phenotype. Inhibition of Src with pp2 caused

significantly more changes, 3,912 and 6,483, respectively. The sizable difference

between the amount of genes changed in temperature-induced inactivation when

compared to inhibition of Src might be due to pp2 inhibiting other members of the

Src-family as well. When comparing MDCK cells with active v-Src to

untransformed MDCK cells, 2,810 genes changed their expression in 2D, whilst

the expression of 4,315 genes was changed in 3D. All in all, the activation of v-Src

has very different outcomes depending on what kind of environment the cells are

growing in.

When the changes induced by v-Src in 3D were analysed, the expression of

two interesting transcription factors was noticed to be elevated: Kaiso and Nanos.

Both of these transcription factors are linked to p120 catenin, a substrate of v-Src,

and these changes in the expression levels of Kaiso and Nanos might thus play a

role in the transformation process of v-Src transformed epithelial cells. Kaiso is a

transcriptional repressor that has a potential role in tumorigenesis, as it represses

the expression of genes regulated by the TCF/LEF pathway. If p120 catenin is

relocated to the nucleus, it binds to Kaiso and releases the expression of these genes

(van Roy & McCrea 2005). Nanos has an evolutionary conserved function in

embryonic patterning and germ line development and is down-regulated by E-

cadherin. Nanos also induces the cytoplasmic translocation of p120 catenin from

the adherens junctions, impairs cell-cell adhesion and promotes cell motility and

invasion (Strumane et al. 2006).

89

E-cadherin is a pivotal player in cell attachment and in multiple different

signalling pathways, and the localization of E-cadherin to the plasma membrane is

a marker of cell polarization. The downregulation of E-cadherin, on the other hand,

is a hallmark of multiple advantage-stage tumour cells, although the loss of E-

cadherin alone is not sufficient to induce tumorigenesis (Jeanes et al. 2008, Lim &

Thiery 2012). Microarray analysis did not reveal any changes in the expression of

cadherins or catenins when v-Src was activated, but significant changes were seen

in the cell morphology: activation of v-Src caused a complete reorganization of

junctional complexes, E-cadherin was localized to the cytoplasm in endocytic

vesicles together with β-catenin and, in some cases, p120-catenin. In 3D, the cyst

had no lumen or apoptosis and the cells were undifferentiated as the cell cluster

was lacking a lumen, and actin and E-cadherin were not localized as they should in

cells that have a proper apico-basal axis. E-cadherin recycling was also disturbed

when v-Src was activated. Microarray analysis showed a decrease in the expression

levels of rab proteins that are important in vesicle trafficking, and this decrease

could explain the observed lack of recycling at permissive temperatures. However,

the activation of Src caused impairments in transportation machinery also by

disintegrating the cytoskeletal structures. Thus, Src activation can affect E-cadherin

recycling in multiple ways.

One of the very striking changes revealed by the microarray analysis was the

expression of survivin in both untransformed and ts-Src MDCK cells in 2D and ts-

Src MDCK cells in 3D, whereas MDCK cells grown in 3D lack it. The expression

of survivin by untransformed MDCK cells in 2D might explain why they were more

resistant to apoptosis than MDCK cells grown in 3D, and thus less differentiated.

In 3D, low survivin expression levels were accompanied by low mitochondrial

membrane permeability, together with lumen formation and apoptosis, in

untransformed MDCK cells. In ts-Src MDCK cells grown in 3D, upregulated

survivin expression correlated with formation of solid cysts, continued proliferation

and resistance to apoptosis.

6.3 The importance of E-cadherin interactions and survivin

expression on cell fate in untransformed and ts-Src MDCK cells

Similarly to ion transport, also epithelial differentiation and cell polarity has been

studied on cells grown on permeable filters where the cell monolayer forms a

barrier, separating the apical and basal sides. Cells grown on plastic polarize, at

best, partially, since they are unable to take in nutrients from the basal side, whereas

90

filter-grown cells that have access to the cell media from the basal side are able to

polarize fully (Balcarova-Ständer et al. 1984). This complete polarization was

utilized to study cell trafficking, endo- and exocytosis or selective manipulation of

apical or basal surfaces, among other things. Growing cells on filters enabled the

study of vesicle transport routes with a technique where only the apical membrane

of highly polarized cells was permeabilized and cells were supplemented with

radiolabelled isotopes or fluorescent probes (Simons & Virta 1987, Bomsel et al.

1989).

Cells grown on filters are still imperfectly differentiated as they continue to

constantly proliferate (or, if their contact-inhibition mechanism is still functioning,

will only cease once the plate is full) and are more resistant to apoptosis. On the

other hand, when grown inside ECM, cells polarize better and differentiate further

than in 2D (Martin-Belmonte & Mostov 2008). The MDCK cell cysts will stop

growing once a certain size is reached and can maintain that size for prolonged

periods of time if no apoptosis signal is received. They are also responsive to

apoptosis signalling (Edmondson et al. 2014). As a downside, studying secretion

and collecting exocytosis vesicles from the apical side is not possible in 3D

environment.

Once a multicellular organism has matured and reached its adult size, it aims

to maintain that size. The cellular homeostasis in an adult organism seeks to balance

the proliferation and cell death and maintain the cell size. Excessive cell division,

hyperproliferation, is needed in wound healing, but can also result in cancer, while

increase in cell size and mass, hypertrophy, is involved in muscular growth due to

exercise, but can lead to cardiac disease and contribute to aging. Hyperactivity, or

abnormally active secretion, occurs normally in antibody production, as part of the

immune response or wound healing, but has also been linked to neurodegenerative

diseases (Alzheimer’s, Parkinson’s and Huntington’s diseases) and aging (Lloyd

2013). For modelling these processes, a 3D system is needed, as they are better

differentiated and can maintain homeostasis, i.e. are closer to how cells behave in

vivo than cells grown on 2D.

The cell and cyst size is regulated by the Hippo pathway, E-cadherin junctions,

inhibition of proliferation signalling and susceptibility to apoptosis. Studying these

regulators separately from each other is extremely difficult. E-cadherin interactions

between two cells are needed for contact inhibition of cells growing in dense

cultures. Moreover, dense cell cultures and mechanical strain, transduced via E-

cadherin and the actin network attached to it, were shown to activate the Hippo

pathway by inactivating Yap or transporting it out of the nucleus (Kim et al. 2011,

91

Benham-Pyle et al. 2015). Thus, both plasma membrane localization of E-cadherin

and the Hippo pathway are important for the maintenance of the non-proliferating,

differentiated phenotype of epithelial cells.

In the present study, the role of E-cadherin localization and the expression of

survivin was investigated in the occurrence and regulation of proliferation and

apoptosis. It was shown that untransformed MDCK cells were able to achieve a

high level of differentiation when grown inside ECM, and did not suffer apoptosis

even though they were not expressing survivin. A good way to study the importance

of E-cadherin trans-interactions to apoptosis susceptibility is to expose the cells

grown in different environments to EGTA. The untransformed MDCK cysts had

intact E-cadherin trans-interactions, which are needed for both functioning contact

inhibition of proliferation of cells and maintenance of quiescence without

undergoing apoptosis (Benham-Pyle et al. 2015). In the current work, the E-

cadherin junctions were disturbed by chelation of the calcium ions using EGTA in

2D, 2½D and 3D environments. There was a clear correlation between the

expression of survivin and cell response to EGTA, since the cells with

downregulated survivin quickly underwent apoptosis whereas the cells in 2D were

able to avoid apoptosis. Guo and co-workers (2014) showed that dissociation of E-

cadherin trans-interactions by EGTA resulted in cis-interactions between two E-

cadherin molecules located on the plasma membrane of the same cell. This,

together with the results presented here, emphasized the importance of E-cadherin

trans-interactions for cell survival, as cells were prone to apoptosis even though

they were properly polarized and had contact with the ECM. On the other hand,

while incapable of saving the cells from apoptosis, E-cadherin cis-interactions were

sufficient to downregulate survivin in 3D in the presence of EGTA. The significant

difference in apoptosis susceptibility of cells grown in different environments (2D,

2½D and 3D) was shown in the current work, which clearly demonstrated the

plasticity of the cell behaviour in different environments.

6.4 Ts-Src MDCK cells as a model for malignant transformation of

epithelial cells

Cancer cells have traditionally been studied in 2D environment, and analysis has

focused on signal transduction from cell surface receptors into nucleus,

characterization of mutations or measurements of invasion potential of cells grown

on collagen or agarose (Edmondson et al. 2014). 2D environment makes mass-

culture of cells possible for high-throughput analyses and genetic screening, but

92

does not allow studying 3D cysts and the changes that occur in them. In addition,

cancer cells grown in 2D are even more dedifferentiated than cancer cells in 3D,

which can make them more resistant to apoptosis and thus reduce the usefulness of

data gathered from these experiments when considering cancer pathology in vivo

(Edmondson et al. 2014, Ravi et al. 2015, Ravi et al. 2017). Temperature-sensitive

Src MDCK cells allow the study of very early steps in malignant transformation

via activation of Src-oncogene, both in 2D and 3D, with a very simple way to

initiate the transformation process. Obviously, the benefits of this cell line for

studying other oncogenes are more limited.

In the present study, the early phases of cellular transformation were monitored.

In ts-Src MDCK cells grown at permissive temperature, E-cadherin localization

varied from lateral membranes to the cytoplasm, but since survivin expression was

upregulated as soon as v-Src was activated, the cells avoided apoptosis. When

grown on rigid cell culture plates, the survivin expression was upregulated

regardless of the Src activation status. In this environment the untransformed

MDCK cells seemed to have a “semi-differentiated” phenotype. The expression of

survivin might even function as an indicator of cellular differentiation.

Lumen filling is usually depicted to happen as transformed cells proliferate into

the lumen. In this study, ts-Src MDCK cell cysts in Matrigel were cultured first at

non-permissive temperature to form cysts with only a single layer of cells and a

lumen in the middle. Then, the cells were shifted to the permissive temperature and

the consequences of v-Src activation were monitored using a spinning disc confocal

microscope. The time-lapse experiment revealed that the lumen collapsed due to

the loss of tight junctions and subsequent loss of intraluminal hydrostatic pressure.

This occurred before any cell proliferation into the lumen could take place. The

current study shows that v-Src is unable to induce cell migration or downregulation

of E-cadherin and thus is not a very strong oncogene when compared to Ras or Myc.

This is supported by the earlier discovery that Src is unable to induce

transformation without STAT3, which is a signal transducer and a Src-substrate that

is also capable of activating genes regulating cell division and survival, including

survivin. Furthermore, cells where Src is activated without STAT3 suffer apoptosis

(Bromberg et al. 1998, Turkson et al. 1998, Geletu et al. 2013).

Behrens and co-workers (1993) noted that activation of v-Src induces

migratory behaviour in ts-Src MDCK cells when cultivated on top of a layer of

collagen at the permissive temperature. Throughout the current study, no invasive

behaviour or tubulogenesis was detected in ts-Src MDCK cells grown inside

Matrigel or the mixture of collagen I and Matrigel, possibly due to the stronger

93

polarization cues and growth factors provided by Matrigel. Tubulogenesis has been

induced in untransformed MDCK cells grown inside collagen I by adding

hepatocyte growth factor (HGF; also known as scatter factor) to the cells. HGF

launches the EMT process of the cells, and cells take on a mesenchymal phenotype

and start scattering from the cyst and initiate tubulogenesis by sending long

extensions into the surrounding matrix (Montesano et al. 1991; O’Brien et al. 2004).

Since Src alone is not able to induce tubulogenesis, and thus EMT, in MDCK cells,

the ts-Src MDCK is not perfect for modelling the processes involved in EMT, but

does provide a good model for monitoring early events in malignant transformation.

Cancer cells typically produce more ROS than healthy cells, and are thus more

reliant on their antioxidant and free-radical scavenging capabilities (Prasad et al.

2017). This makes them more vulnerable to drugs that elevate the levels of

intracellular ROS, such as PL. MDCK cells transformed by v-Src were also more

susceptible to PL-induced apoptosis than untransformed cells. Pro-oxidants, like

PL, have been shown to inhibit several transcription factors, like STAT3, NF-κB,

Sp1, Sp3 and Sp4, and downregulate expression of genes involved in cell growth,

survival, inflammation and angiogenesis (Pathi et al. 2011, Han et al. 2014, Park

et al. 2014, Bharadwaj et al. 2015). PL-treatment was able to downregulate survivin

expression of the ts-Src MDCK cells as well. NAC, an antioxidant, rescued the

cells from PL-induced apoptosis, but was not able to upregulate survivin expression,

indicating some other mechanism than increased apoptosis resistance via survivin

to be behind the rescue. NAC-treatment caused reappearance of E-cadherin to the

lateral membranes of ts-Src MDCK cells, indicating an improvement in

differentiation state of the cells. Earlier, NAC-treatment has been shown to prevent

gamma-irradiation-induced disruption of TJs and AJs in mouse colon cells and to

improve the differentiation state of normal human epidermal keratinocytes (NHEK)

and Caco-2 colon cancer cells by elevating their E-cadherin expression (Shukla et

al. 2016). Taken together with the results from this study, it seems NAC protects

cancer cells from PL-induced apoptosis by improving their differentiation by

returning E-cadherin to the lateral membranes.

6.5 Comparison of the effects of 2D and 3D growth environment on

cell behaviour

The importance of cell-based assays to the advancement of medical sciences cannot

be understated. Cultured cells provide a simple, fast and cost-effective alternative

to animal testing, and, more importantly, allow the use of human cells without a

94

constant need for patient donors. Established cell lines provide homogenous sample

material and thus improve predictability and reproducibility. The majority of the

experiments made with cultured cells use monolayers of cells grown on hard plastic

in 2D, which differs remarkably as a growth environment from the in vivo

environment where cells are surrounded by other cells and ECM. This difference

between artificial and natural environment can result in data that does not

correspond to the situation in vivo and may thus be misleading or outright false or

artificial. This delays the advancement of science, and, especially in drug

development, causes significant extra costs in terms of both money and time. Some

of these costs could be avoided by using cell-based assays where the cells are closer

to their state in vivo than in 2D cultures. This way, ineffective or toxic compounds

could be eliminated already before clinical trials. 3D culture of cells presents a

method which occupies a space between 2D and living tissue, being still more cost-

effective and less labour-intensive than animal trials, but simultaneously providing

cells an environment more akin to their natural state (Edmondson et al. 2014).

Cells grown in 3D differ from 2D cells in several important ways, both

morphologically and physiologically. 3D environment provides spatial cues for

differentiation and also physically limit the movements of the cells, while cells in

monolayer have increased degrees of freedom for spreading and proliferation.

Physical constraints require cells with invasive tendencies to also be able to break

down the surrounding ECM. Thus, a hallmark of malignant transformation is the

ability to express enzymes that can break down ECM (Partanen et al. 2012).

Monolayer cells also receive homogenous amounts of nutrients and growth factors

from the provided medium, whilst cells in 3D cysts receive nutrients more unevenly

due to the possible penetration problems through the scaffold material. Cells

growing inside the cyst might also receive less nutrients and even less oxygen

compared to the outermost cells of the cyst. This creates a cyst with cells at various

stages, from proliferating to quiescent, apoptotic, hypoxic and even necrotic,

creating heterogeneity perceived in tissues. In addition, cells grown in 3D are able

to execute their internal programming, like form a lumen (Partanen et al. 2007,

Edmondson et al. 2014, Ravi et al. 2015). Importantly, cells grown in 3D are often

less sensitive to drugs than cells in 2D due to this heterogeneity in cell stages and

levels of differentiation (Edmondson et al. 2014, Ravi et al. 2017).

In the present study, the levels of gene expression of untransformed MDCK

cells grown in 2D were compared to cells grown in 3D. The microarray analysis

revealed 6474 gene expression changes when the environment was changed from

2D to 3D. One of the genes that underwent significant changes with the change in

95

environment was survivin, which was expressed in cells grown in 2D, but

downregulated in 2½D and to an even greater extent in 3D. Cells grown in 3D were

also more susceptible to apoptosis than cells grown in 2D when denied of E-

cadherin trans-interactions. These results again show that the 3D ECM environment

correlates better with normal tissue environment than the 2D plastic environment.

The present study also reveals drastic differences in gene expression and phenotype

that activation of a single oncogene, in this case v-Src, can have in 2D when

compared to 3D, effectively highlighting that results obtained from 2D cell-culture

experiments should be analysed with great care and conclusions pertaining to in

vivo situations drawn with utmost vigilance.

6.6 Use of time-lapse imaging of live cells as a tool for monitoring

cellular processes

In highly dynamic and complex structures like cells, a single time-point snapshot,

such as images of fixed cells or Western blots, can give an incomplete picture of

the whole process. Often, continuous observation of the process is needed. For this,

time-lapse imaging of live cells is an invaluable tool. Time-lapse can utilize

transmitted light, or dyes and/or fluorescent probes which stain a specific part of

the cell, or fusion proteins with an attached fluorescent tag produced by the cell

itself. Time-lapse imaging of live cells is always a balancing act between picture

quality and cell well-being, and many aspects have to be optimized for good quality

data. These include temperature, CO2 and O2 levels, pH, availability of nutrients

and minimizing evaporation and phototoxicity (Coutu & Schroeder 2013).

In the present study, time-lapse imaging of live cells has been used extensively.

A method was created for monitoring changes in lumen, cyst and cell volumes,

which can give useful information on the roles of transporter proteins and can be a

future platform for testing drugs affecting those transporters. Time-lapse imaging

was also used to monitor the sequence of events occurring when the v-Src oncogene

was activated in ts-Src MDCK cysts. Previous studies have shown that lumen

filling occurs once MDCK cells in a cyst undergo cellular transformation, and this

has been attributed to proliferation of cells into the luminal space. In the current

study, time-lapse imaging of ts-Src MDKC cysts revealed that, once v-Src is

activated, the lumen quickly collapses as the transformation weakens tight

junctions, and hydrostatic pressure inside the lumen is released. This occurs before

any proliferation due to oncogene activation could happen. These kinds of

sequential events cannot be monitored via “snapshots” taken with fixed cells.

96

97

7 Summary and conclusions

The aim of the current study was to understand the mechanisms behind the

maintenance of a differentiated phenotype and analyse the electrophysiological

parameters which regulate the transport capacity of kidney epithelial cells. Special

emphasis was placed on comparing the effects of different culture environments on

these processes. The first part of the study concentrated on analysing the effects of

drugs or the composition of the basal extracellular fluid on cell, cyst and lumen

volumes of untransformed MDCK cells using time lapse microscopy of living cells.

Lack of monovalent cations caused lumen expansion by inducing transepithelial

water flow towards the lumen, presumably via activation of cAMP signalling

pathway or voltage- and volume-sensitive chloride channels. Evacuation of water

from the lumen could be induced by exposing the cells to extracellular fluid lacking

chloride ions, or by depolarization of the cyst. These experiments showed that

MDCK cells were capable of both water secretion and reabsorption. More

importantly, the cells were able to perform these functions when exposed to

hyperpolarizing or depolarizing environment; a change in osmolality of basal fluid

was not required. Taken together, these results validate the status of MDCK cells

as a good basic model for studying the cell biological characters of secretory

epithelium or any tissue with secretory or absorptive properties. The method

presented here allows the analysis of the effects of external stimuli, be it changes

in ion balance or drugs affecting ion transport, on cell, cyst and lumen volumes

over short periods of time.

The second part of the study aimed to analyse the effects 2D and 3D culture

environments have on gene expression of untransformed MDCK and ts-Src MDCK

cells, and the extent of alterations in genetic profile that activation of a single

oncogene can induce. Microarray analysis revealed thousands of changes between

different conditions, but the most intriguing was the decreased expression of

survivin when switching from 2D environment to 3D. This downregulation of

survivin occurs in adult tissues as well, indicating that the cells grown in 3D are

better differentiated than 2D cells, and thus closer to the in vivo state. This is in line

with mitochondrial activity measurements, as the mitochondrial membrane

permeability of 3D-grown MDCK cells decreased swiftly after the first two days

in Matrigel, whilst the membrane permeability of ts-Src MDCK cells remained

higher for longer. This observation could explain the increased apoptosis in MDCK

cells compared to ts-Src MDCK cells, and indicate pro-survival and anti-apoptotic

function of v-Src.

98

Differences in survivin expression generated interest for further study of

factors controlling survivin expression and its significance to cell survival. MDCK

cells grown in 3D were devoid of survivin, but also did not suffer apoptosis, at least

as long as the cells remained in contact with the ECM, whereas MDCK cells grown

on 2D and ts-Src MDCK cells in all environments expressed survivin. If MDCK

cells were denied ECM contacts by growing them in suspension, they were more

susceptible to apoptosis than survivin-expressing ts-Src MDCK cells. Finally, if

cells were denied cell-cell junctions, cells lacking survivin suffered apoptosis even

though they had a proper ECM environment supporting cell-matrix contacts. Taken

together, these results highlighted the importance of cellular contacts to the cells:

MDCK cells needed ECM contacts to differentiate and cell-cell contacts to avoid

apoptosis. Since v-Src did not downregulate E-cadherin expression, it is unable to

bring about complete EMT. Thus, rather than being the initiator of cancer, v-Src

can be considered to be a mild oncogene that supports cancer cell survival.

99

References

Acharya A, Das I, Chandhok D & Saha T (2010) Redox regulation in cancer: a double-edged sword with therapeutic potential. Oxid Med Cell Longev 3(1): 23-34.

Alberts B, Bray D, Lewis J, Raff M, Roberts K, Watson J (1989) Molecular Biology of the Cell. New York NY, Garland Publishing Inc.

Altieri DC (2010) Survivin and IAP proteins in cell-death mechanisms. Biochem J 430(2): 199-205.

Altieri DC (2015) Survivin - The inconvenient IAP. Semin Cell Dev Biol 39: 91-96. Ambrosini G, Adida C & Altieri DC (1997) A novel anti-apoptosis gene, survivin, expressed

in cancer and lymphoma. Nat Med 3(8): 917-921. Anderson JM & van Itallie CM (2009) Physiology and function of the tight junction. Cold

Spring Harb Perspect Biol 1(2): a002584. Ares GR, Caceres PS & Ortiz PA (2011) Molecular regulation of NKCC2 in the thick

ascending limb. Am J Physiol Renal Physiol 301(6): F1143-59. Athanasoula KC, Gogas H, Polonifi K, Vaiopoulos AG, Polyzos A & Mantzourani M (2014)

Survivin beyond physiology: orchestration of multistep carcinogenesis and therapeutic potentials. Cancer Lett 347(2): 175-182.

Auernheimer V, Lautscham L, Leidenberger M, Friedrich O, Kappes B, Fabry B & Goldmann W (2015) Vinculin phosphorylation at residues Y100 and Y1065 is required for cellular force transmission. J Cell Sci 128(18): 3435-43. Balcarova-Stander J, Pfeiffer SE, Fuller SD & Simons K (1984) Development of cell surface polarity in the epithelial Madin-Darby canine kidney (MDCK) cell line. EMBO J 3(11): 2687-2694.

Barker G & Simmons NL (1981) Identification of two strains of cultured canine renal epithelial cells (MDCK cells) which display entirely different physiological properties. Q J Exp Physiol 66(1): 61-72.

Behrens J, Mareel MM, van Roy FM & Birchmeier W (1989) Dissecting tumor cell invasion: epithelial cells acquire invasive properties after the loss of uvomorulin-mediated cell-cell adhesion. J Cell Biol 108(6): 2435-2447.

Behrens J, Vakaet L, Friis R, Winterhager E, van Roy F, Mareel MM & Birchmeier W (1993) Loss of epithelial differentiation and gain of invasiveness correlates with tyrosine phosphorylation of the E-cadherin/beta-catenin complex in cells transformed with a temperature-sensitive v-SRC gene. J Cell Biol 120(3): 757-766.

Benham-Pyle BW, Pruitt BL & Nelson WJ (2015) Cell adhesion. Mechanical strain induces E-cadherin-dependent Yap1 and beta-catenin activation to drive cell cycle entry. Science 348(6238): 1024-1027.

Bezerra DP, Pessoa C, de Moraes MO, Saker-Neto N, Silveira ER & Costa-Lotufo LV (2013) Overview of the therapeutic potential of piplartine (piperlongumine). Eur J Pharm Sci 48(3): 453-463.

100

Bharadwaj U, Eckols TK, Kolosov M, Kasembeli MM, Adam A, Torres D, Zhang X, Dobrolecki LE, Wei W, Lewis MT, Dave B, Chang JC, Landis MD, Creighton CJ, Mancini MA & Tweardy DJ (2015) Drug-repositioning screening identified piperlongumine as a direct STAT3 inhibitor with potent activity against breast cancer. Oncogene 34(11): 1341-1353.

Bjorge JD, Jakymiw A & Fujita DJ (2000) Selected glimpses into the activation and function of Src kinase. Oncogene 19(49): 5620-5635.

Boivin FJ, Sarin S, Evans JC & Bridgewater D (2015) The Good and Bad of beta-Catenin in Kidney Development and Renal Dysplasia. Front Cell Dev Biol 3: 81.

Bomsel M, Prydz K, Parton RG, Gruenberg J & Simons K (1989) Endocytosis in filter-grown Madin-Darby canine kidney cells. J Cell Biol 109(6 Pt 2): 3243-3258.

Bornslaeger EA, Corcoran CM, Stappenbeck TS & Green KJ (1996) Breaking the connection: displacement of the desmosomal plaque protein desmoplakin from cell-cell interfaces disrupts anchorage of intermediate filament bundles and alters intercellular junction assembly. J Cell Biol 134(4): 985-1001.

Bromberg JF, Horvath CM, Besser D, Lathem WW & Darnell JE,Jr (1998) Stat3 activation is required for cellular transformation by v-src. Mol Cell Biol 18(5): 2553-2558.

Bryant DM, Datta A, Rodriguez-Fraticelli AE, Peranen J, Martin-Belmonte F & Mostov KE (2010) A molecular network for de novo generation of the apical surface and lumen. Nat Cell Biol 12(11): 1035-1045.

Bryant DM, Kerr MC, Hammond LA, Joseph SR, Mostov KE, Teasdale RD & Stow JL (2007) EGF induces macropinocytosis and SNX1-modulated recycling of E-cadherin. J Cell Sci 120(Pt 10): 1818-1828.

Buchholz B, Faria D, Schley G, Schreiber R, Eckardt KU & Kunzelmann K (2014) Anoctamin 1 induces calcium-activated chloride secretion and proliferation of renal cyst-forming epithelial cells. Kidney Int 85(5): 1058-1067.

Cadwell CM, Su W & Kowalczyk AP (2016) Cadherin tales: Regulation of cadherin function by endocytic membrane trafficking. Traffic 17(12): 1262-1271.

Capaldo CT, Farkas AE & Nusrat A (2014) Epithelial adhesive junctions. F1000Prime Rep 6: 1-1. eCollection 2014.

Caputo A, Caci E, Ferrera L, Pedemonte N, Barsanti C, Sondo E, Pfeffer U, Ravazzolo R, Zegarra-Moran O & Galietta LJ (2008) TMEM16A, a membrane protein associated with calcium-dependent chloride channel activity. Science 322(5901): 590-594.

Castillon GA, Michon L & Watanabe R (2013) Apical sorting of lysoGPI-anchored proteins occurs independent of association with detergent-resistant membranes but dependent on their N-glycosylation. Mol Biol Cell 24(12): 2021-2033.

Chan WW & Mashimo H (2013) Lubiprostone Increases Small Intestinal Smooth Muscle Contractions Through a Prostaglandin E Receptor 1 (EP1)-mediated Pathway. J Neurogastroenterol Motil 19(3): 312-318.

Chiabotto G, Bruno S, Collino F & Camussi G (2016) Mesenchymal Stromal Cells Epithelial Transition Induced by Renal Tubular Cells-Derived Extracellular Vesicles. PLOS One 11(7): e0159163.

101

Clausen MV, Hilbers F & Poulsen H (2017) The Structure and Function of the Na,K-ATPase Isoforms in Health and Disease. Front Physiol 8: 371.

Conner GE, Wijkstrom-Frei C, Randell SH, Fernandez VE & Salathe M (2007) The lactoperoxidase system links anion transport to host defense in cystic fibrosis. FEBS Lett 581(2): 271-278.

Cooper JA, Gould KL, Cartwright CA & Hunter T (1986) Tyr527 is phosphorylated in pp60c-src: implications for regulation. Science 231(4744): 1431-1434.

Coopman P & Djiane A (2016) Adherens Junction and E-Cadherin complex regulation by epithelial polarity. Cell Mol Life Sci 73(18): 3535-3553.

Cory S, Roberts AW, Colman PM & Adams JM (2016) Targeting BCL-2-like Proteins to Kill Cancer Cells. Trends Cancer 2(8): 443-460.

Coutu DL & Schroeder T (2013) Probing cellular processes by long-term live imaging--historic problems and current solutions. J Cell Sci 126(Pt 17): 3805-3815.

Cuppoletti J, Malinowska DH, Tewari KP, Li QJ, Sherry AM, Patchen ML & Ueno R (2004) SPI-0211 activates T84 cell chloride transport and recombinant human ClC-2 chloride currents. Am J Physiol Cell Physiol 287(5): C1173-83.

Dalla Pozza E, Forciniti S, Palmieri M & Dando I (2017) Secreted molecules inducing epithelial-to-mesenchymal transition in cancer development. Semin Cell Dev Biol .

Datta A, Bryant DM & Mostov KE (2011) Molecular regulation of lumen morphogenesis. Curr Biol 21(3): R126-36.

Davis MA, Ireton RC & Reynolds AB (2003) A core function for p120-catenin in cadherin turnover. J Cell Biol 163(3): 525-534.

de Diesbach P, Medts T, Carpentier S, D'Auria L, van Der Smissen P, Platek A, Mettlen M, Caplanusi A, van den Hove MF, Tyteca D & Courtoy PJ (2008) Differential subcellular membrane recruitment of Src may specify its downstream signalling. Exp Cell Res 314(7): 1465-1479.

De Wever O, Demetter P, Mareel M & Bracke M (2008) Stromal myofibroblasts are drivers of invasive cancer growth. Int J Cancer 123(10): 2229-2238.

Delva E & Kowalczyk AP (2009) Regulation of cadherin trafficking. Traffic 10(3): 259-267. Dohi T, Okada K, Xia F, Wilford CE, Samuel T, Welsh K, Marusawa H, Zou H, Armstrong

R, Matsuzawa S, Salvesen GS, Reed JC & Altieri DC (2004) An IAP-IAP complex inhibits apoptosis. J Biol Chem 279(33): 34087-34090.

Duffy MJ, O'Donovan N, Brennan DJ, Gallagher WM & Ryan BM (2007) Survivin: a promising tumor biomarker. Cancer Lett 249(1): 49-60.

Dukes JD, Whitley P & Chalmers AD (2011) The MDCK variety pack: choosing the right strain. BMC Cell Biol 12: 43-2121-12-43.

Dupont S, Morsut L, Aragona M, Enzo E, Giulitti S, Cordenonsi M, Zanconato F, Le Digabel J, Forcato M, Bicciato S, Elvassore N & Piccolo S (2011) Role of YAP/TAZ in mechanotransduction. Nature 474(7350): 179-183.

Edmondson R, Broglie JJ, Adcock AF & Yang L (2014) Three-dimensional cell culture systems and their applications in drug discovery and cell-based biosensors. Assay Drug Dev Technol 12(4): 207-218.

102

Erickson AC & Couchman JR (2001) Basement membrane and interstitial proteoglycans produced by MDCK cells correspond to those expressed in the kidney cortex. Matrix Biol 19(8): 769-778.

Faria D, Rock JR, Romao AM, Schweda F, Bandulik S, Witzgall R, Schlatter E, Heitzmann D, Pavenstadt H, Herrmann E, Kunzelmann K & Schreiber R (2014) The calcium-activated chloride channel Anoctamin 1 contributes to the regulation of renal function. Kidney Int 85(6): 1369-1381.

Frizzell RA & Hanrahan JW (2012) Physiology of epithelial chloride and fluid secretion. Cold Spring Harb Perspect Med 2(6): a009563.

Fuchs Y & Steller H (2015) Live to die another way: modes of programmed cell death and the signals emanating from dying cells. Nat Rev Mol Cell Biol 16(6): 329-344.

Fujita Y, Krause G, Scheffner M, Zechner D, Leddy HE, Behrens J, Sommer T & Birchmeier W (2002) Hakai, a c-Cbl-like protein, ubiquitinates and induces endocytosis of the E-cadherin complex. Nat Cell Biol 4(3): 222-231.

Furuse M, Furuse K, Sasaki H & Tsukita S (2001) Conversion of zonulae occludentes from tight to leaky strand type by introducing claudin-2 into Madin-Darby canine kidney I cells. J Cell Biol 153(2): 263-272.

Gall TM & Frampton AE (2013) Gene of the month: E-cadherin (CDH1). J Clin Pathol 66(11): 928-932.

Ganem NJ, Cornils H, Chiu SY, O'Rourke KP, Arnaud J, Yimlamai D, Thery M, Camargo FD & Pellman D (2014) Cytokinesis failure triggers hippo tumor suppressor pathway activation. Cell 158(4): 833-848.

Gao F, Zhang Y, Yang F, Wang P, Wang W, Su Y & Luo W (2014) Survivin promotes the invasion of human colon carcinoma cells by regulating the expression of MMP7. Mol Med Rep 9(3): 825-830.

Gao L, Yang Z, Hiremath C, Zimmerman S, Long B, Brakeman P, Mostov K, Bryant D, Luby-Phelps K, Marciano D (2017). Afadin orients cell division to position the tubule lumen in developing renal tubules. Development 144(19): 3511-3520.

Gaush CR, Hard WL & Smith TF (1966) Characterization of an established line of canine kidney cells (MDCK). Proc Soc Exp Biol Med 122(3): 931-935.

Geletu M, Guy S, Arulanandam R, Feracci H & Raptis L (2013) Engaged for survival: From cadherin ligation to STAT3 activation. JAKSTAT 2(4): e27363.

Giannoni E, Buricchi F, Grimaldi G, Parri M, Cialdai F, Taddei ML, Raugei G, Ramponi G & Chiarugi P (2008) Redox regulation of anoikis: reactive oxygen species as essential mediators of cell survival. Cell Death Differ 15(5): 867-878.

Grantham JJ, Uchic M, Cragoe EJ,Jr, Kornhaus J, Grantham JA, Donoso V, Mangoo-Karim R, Evan A & McAteer J (1989) Chemical modification of cell proliferation and fluid secretion in renal cysts. Kidney Int 35(6): 1379-1389.

Grantham JJ & Wallace DP (2002) Return of the secretory kidney. Am J Physiol Renal Physiol 282(1): F1-9.

Grunder S, Thiemann A, Pusch M & Jentsch TJ (1992) Regions involved in the opening of CIC-2 chloride channel by voltage and cell volume. Nature 360(6406): 759-762.

Guarino M (2010) Src signaling in cancer invasion. J Cell Physiol 223(1): 14-26.

103

Gul IS, Hulpiau P, Saeys Y & van Roy F (2017) Evolution and diversity of cadherins and catenins. Exp Cell Res 358(1): 3-9.

Guo Z, Neilson LJ, Zhong H, Murray PS, Zanivan S & Zaidel-Bar R (2014) E-cadherin interactome complexity and robustness resolved by quantitative proteomics. Sci Signal 7(354): rs7.

Han JG, Gupta SC, Prasad S & Aggarwal BB (2014) Piperlongumine chemosensitizes tumor cells through interaction with cysteine 179 of IkappaBalpha kinase, leading to suppression of NF-kappaB-regulated gene products. Mol Cancer Ther 13(10): 2422-2435.

Hanukoglu I & Hanukoglu A (2016) Epithelial sodium channel (ENaC) family: Phylogeny, structure-function, tissue distribution, and associated inherited diseases. Gene 579(2): 95-132.

Harvey KF, Zhang X & Thomas DM (2013) The Hippo pathway and human cancer. Nat Rev Cancer 13(4): 246-257.

Hiltunen E, Holmberg P, Kaikkonen M, Lindblom-Ylänne S, Nienstedt W (2003) Galenos IV Ihmiselimistö kohtaa ympäristön. Finland, WSOY.

Hoffmann EK (2001) The pump and leak steady-state concept with a variety of regulated leak pathways. J Membr Biol 184(3): 321-330.

Huang P, Trotter K, Boucher RC, Milgram SL & Stutts MJ (2000) PKA holoenzyme is functionally coupled to CFTR by AKAPs. Am J Physiol Cell Physiol 278(2): C417-22.

Ireton RC, Davis MA, van Hengel J, Mariner DJ, Barnes K, Thoreson MA, Anastasiadis PZ, Matrisian L, Bundy LM, Sealy L, Gilbert B, van Roy F & Reynolds AB (2002) A novel role for p120 catenin in E-cadherin function. J Cell Biol 159(3): 465-476.

Jeanes A, Gottardi CJ & Yap AS (2008) Cadherins and cancer: how does cadherin dysfunction promote tumor progression? Oncogene 27(55): 6920-6929.

Jentsch TJ (2016) VRACs and other ion channels and transporters in the regulation of cell volume and beyond. Nat Rev Mol Cell Biol 17(5): 293-307.

Jentsch TJ, Steinmeyer K & Schwarz G (1990) Primary structure of Torpedo marmorata chloride channel isolated by expression cloning in Xenopus oocytes. Nature 348(6301): 510-514.

Johnson PJ, Coussens PM, Danko AV & Shalloway D (1985) Overexpressed pp60c-src can induce focus formation without complete transformation of NIH 3T3 cells. Mol Cell Biol 5(5): 1073-1083.

Jordan K, Solan JL, Dominguez M, Sia M, Hand A, Lampe PD & Laird DW (1999) Trafficking, assembly, and function of a connexin43-green fluorescent protein chimera in live mammalian cells. Mol Biol Cell 10(6): 2033-2050.

Kelly RJ, Lopez-Chavez A, Citrin D, Janik JE & Morris JC (2011) Impacting tumor cell-fate by targeting the inhibitor of apoptosis protein survivin. Mol Cancer 10: 35-4598-10-35.

Kim NG, Koh E, Chen X & Gumbiner BM (2011) E-cadherin mediates contact inhibition of proliferation through Hippo signaling-pathway components. Proc Natl Acad Sci U S A 108(29): 11930-11935.

104

Kiraz Y, Adan A, Kartal Yandim M & Baran Y (2016) Major apoptotic mechanisms and genes involved in apoptosis. Tumour Biol 37(7): 8471-8486.

Kourtidis A, Lu R, Pence LJ & Anastasiadis PZ (2017) A central role for cadherin signaling in cancer. Exp Cell Res 358(1): 78-85.

Lampugnani MG, Corada M, Caveda L, Breviario F, Ayalon O, Geiger B & Dejana E (1995) The molecular organization of endothelial cell to cell junctions: differential association of plakoglobin, beta-catenin, and alpha-catenin with vascular endothelial cadherin (VE-cadherin). J Cell Biol 129(1): 203-217.

Lawson K & McKay NG (2006) Modulation of potassium channels as a therapeutic approach. Curr Pharm Des 12(4): 459-470.

Le Bivic A (2013) Evolution and cell physiology. 4. Why invent yet another protein complex to build junctions in epithelial cells? Am J Physiol Cell Physiol 305(12): C1193-201.

Le TL, Yap AS & Stow JL (1999) Recycling of E-cadherin: a potential mechanism for regulating cadherin dynamics. J Cell Biol 146(1): 219-232.

Lemmers C, Michel D, Lane-Guermonprez L, Delgrossi MH, Medina E, Arsanto JP & Le Bivic A (2004) CRB3 binds directly to Par6 and regulates the morphogenesis of the tight junctions in mammalian epithelial cells. Mol Biol Cell 15(3): 1324-1333.

Lens SM, Vader G & Medema RH (2006) The case for Survivin as mitotic regulator. Curr Opin Cell Biol 18(6): 616-622.

Li H, Findlay IA & Sheppard DN (2004) The relationship between cell proliferation, Cl- secretion, and renal cyst growth: a study using CFTR inhibitors. Kidney Int 66(5): 1926-1938.

Li H, Yang W, Mendes F, Amaral MD & Sheppard DN (2012) Impact of the cystic fibrosis mutation F508del-CFTR on renal cyst formation and growth. Am J Physiol Renal Physiol 303(8): F1176-86.

Lim J & Thiery JP (2012) Epithelial-mesenchymal transitions: insights from development. Development 139(19): 3471-3486.

Linsdell P, Tabcharani JA, Rommens JM, Hou YX, Chang XB, Tsui LC, Riordan JR & Hanrahan JW (1997) Permeability of wild-type and mutant cystic fibrosis transmembrane conductance regulator chloride channels to polyatomic anions. J Gen Physiol 110(4): 355-364.

Liu L, Zhang M & Zou P (2007) Expression of PLK1 and survivin in diffuse large B-cell lymphoma. Leuk Lymphoma 48(11): 2179-2183.

Liu T, Brouha B & Grossman D (2004) Rapid induction of mitochondrial events and caspase-independent apoptosis in Survivin-targeted melanoma cells. Oncogene 23(1): 39-48.

Lloyd AC (2013) The regulation of cell size. Cell 154(6): 1194-1205. Lodish H, Berk A, Zipursky S, Matsudaira P, Baltimore D, Darnell J (1999) Molecular cell

biology. New York NY, W. H. Freeman and Company. Loffing J & Kaissling B (2003) Sodium and calcium transport pathways along the

mammalian distal nephron: from rabbit to human. Am J Physiol Renal Physiol 284(4): F628-43.

105

Macdonald F, Ford C & Casson A (2004) Molecular Biology of Cancer. United Kingdoms, Garland Science/BIOS Scientific Publishers.

Maitre JL & Heisenberg CP (2013) Three functions of cadherins in cell adhesion. Curr Biol 23(14): R626-33.

Mangoo-Karim R, Uchic M, Lechene C & Grantham JJ (1989) Renal epithelial cyst formation and enlargement in vitro: dependence on cAMP. Proc Natl Acad Sci U S A 86(15): 6007-6011.

Martin GS (1970) Rous sarcoma virus: a function required for the maintenance of the transformed state. Nature 227(5262): 1021-1023.

Martin GS (2001) The hunting of the Src. Nat Rev Mol Cell Biol 2(6): 467-475. Martin-Belmonte F & Mostov K (2008) Regulation of cell polarity during epithelial

morphogenesis. Curr Opin Cell Biol 20(2): 227-234. Mathias RA, Chen YS, Wang B, Ji H, Kapp EA, Moritz RL, Zhu HJ & Simpson RJ (2010)

Extracellular remodelling during oncogenic Ras-induced epithelial-mesenchymal transition facilitates MDCK cell migration. J Proteome Res 9(2): 1007-1019.

McAteer JA, Evan AP & Gardner KD (1987) Morphogenetic clonal growth of kidney epithelial cell line MDCK. Anat Rec 217(3): 229-239.

McKenzie JA, Liu T, Goodson AG & Grossman D (2010) Survivin enhances motility of melanoma cells by supporting Akt activation and {alpha}5 integrin upregulation. Cancer Res 70(20): 7927-7937.

McNeill H & Reginensi A (2017) Lats1/2 Regulate Yap/Taz to Control Nephron Progenitor Epithelialization and Inhibit Myofibroblast Formation. J Am Soc Nephrol 28(3): 852-861.

Meder D, Shevchenko A, Simons K & Fullekrug J (2005) Gp135/podocalyxin and NHERF-2 participate in the formation of a preapical domain during polarization of MDCK cells. J Cell Biol 168(2): 303-313.

Mehrotra S, Languino LR, Raskett CM, Mercurio AM, Dohi T & Altieri DC (2010) IAP regulation of metastasis. Cancer Cell 17(1): 53-64.

Melis N, Tauc M, Cougnon M, Bendahhou S, Giuliano S, Rubera I & Duranton C (2014) Revisiting CFTR inhibition: a comparative study of CFTRinh -172 and GlyH-101 inhibitors. Br J Pharmacol 171(15): 3716-3727.

Micheau O & Tschopp J (2003) Induction of TNF receptor I-mediated apoptosis via two sequential signaling complexes. Cell 114(2): 181-190.

Michgehl U, Pavenstadt H & Vollenbroker B (2017) Cross talk between the Crumbs complex and Hippo signaling in renal epithelial cells. Pflugers Arch 469(7-8): 917-926.

Mohamed A, Ferguson D, Seibert FS, Cai HM, Kartner N, Grinstein S, Riordan JR & Lukacs GL (1997) Functional expression and apical localization of the cystic fibrosis transmembrane conductance regulator in MDCK I cells. Biochem J 322 ( Pt 1)(Pt 1): 259-265.

Montesano R, Matsumoto K, Nakamura T & Orci L (1991) Identification of a fibroblast-derived epithelial morphogen as hepatocyte growth factor. Cell 67(5): 901-908.

106

Morishita Y, Matsuzaki T, Hara-chikuma M, Andoo A, Shimono M, Matsuki A, Kobayashi K, Ikeda M, Yamamoto T, Verkman A, Kusano E, Ookawara S, Takata K, Sasaki S & Ishibashi K (2005) Disruption of aquaporin-11 produces polycystic kidneys following vacuolization of the proximal tubule. Mol Cell Biol 25(17): 7770-7779.

Murphy SM, Bergman M & Morgan DO (1993) Suppression of c-Src activity by C-terminal Src kinase involves the c-Src SH2 and SH3 domains: analysis with Saccharomyces cerevisiae. Mol Cell Biol 13(9): 5290-5300.

Nabzdyk CS, Lancero H, Nguyen KP, Salek S & Conte MS (2011) RNA interference-mediated survivin gene knockdown induces growth arrest and reduced migration of vascular smooth muscle cells. Am J Physiol Heart Circ Physiol 301(5): H1841-9.

Nejsum LN, Elkjaer M, Hager H, Frokiaer J, Kwon TH & Nielsen S (2000) Localization of aquaporin-7 in rat and mouse kidney using RT-PCR, immunoblotting, and immunocytochemistry. Biochem Biophys Res Commun 277(1): 164-170.

Nielsen S, DiGiovanni SR, Christensen EI, Knepper MA & Harris HW (1993a) Cellular and subcellular immunolocalization of vasopressin-regulated water channel in rat kidney. Proc Natl Acad Sci U S A 90(24): 11663-11667.

Nielsen S, Smith BL, Christensen EI, Knepper MA & Agre P (1993b) CHIP28 water channels are localized in constitutively water-permeable segments of the nephron. J Cell Biol 120(2): 371-383.

Niu M, Shen Y, Xu X, Yao Y, Fu C, Yan Z, Wu Q, Cao J, Sang W, Zeng L, Li Z, Liu X & Xu K (2015) Piperlongumine selectively suppresses ABC-DLBCL through inhibition of NF-kappaB p65 subunit nuclear import. Biochem Biophys Res Commun 462(4): 326-331.

Noda Y, Sohara E, Ohta E & Sasaki S (2010) Aquaporins in kidney pathophysiology. Nat Rev Nephrol 6(3): 168-178.

Norimatsu Y, Moran AR & MacDonald KD (2012) Lubiprostone activates CFTR, but not ClC-2, via the prostaglandin receptor (EP(4)). Biochem Biophys Res Commun 426(3): 374-379.

O'Brien LE, Tang K, Kats ES, Schutz-Geschwender A, Lipschutz JH & Mostov KE (2004) ERK and MMPs sequentially regulate distinct stages of epithelial tubule development. Dev Cell 7(1): 21-32.

O'Brien LE, Zegers MM & Mostov KE (2002) Opinion: Building epithelial architecture: insights from three-dimensional culture models. Nat Rev Mol Cell Biol 3(7): 531-537.

Ohshiro K, Yaoita E, Yoshida Y, Fujinaka H, Matsuki A, Kamiie J, Kovalenko P & Yamamoto T (2001) Expression and immunolocalization of AQP6 in intercalated cells of the rat kidney collecting duct. Arch Histol Cytol 64(3): 329-338.

Okada M & Nakagawa H (1989) A protein tyrosine kinase involved in regulation of pp60c-src function. J Biol Chem 264(35): 20886-20893.

Ozawa M, Baribault H & Kemler R (1989) The cytoplasmic domain of the cell adhesion molecule uvomorulin associates with three independent proteins structurally related in different species. EMBO J 8(6): 1711-1717.

107

Pang JH, Kraemer A, Stehbens SJ, Frame MC & Yap AS (2005) Recruitment of phosphoinositide 3-kinase defines a positive contribution of tyrosine kinase signaling to E-cadherin function. J Biol Chem 280(4): 3043-3050.

Park KW, Kundu J, Chae IG, Kim DH, Yu MH, Kundu JK & Chun KS (2014) Carnosol induces apoptosis through generation of ROS and inactivation of STAT3 signaling in human colon cancer HCT116 cells. Int J Oncol 44(4): 1309-1315.

Partanen JI, Nieminen AI, Makela TP & Klefstrom J (2007) Suppression of oncogenic properties of c-Myc by LKB1-controlled epithelial organization. Proc Natl Acad Sci U S A 104(37): 14694-14699.

Partanen JI, Tervonen TA, Myllynen M, Lind E, Imai M, Katajisto P, Dijkgraaf GJ, Kovanen PE, Makela TP, Werb Z & Klefstrom J (2012) Tumor suppressor function of Liver kinase B1 (Lkb1) is linked to regulation of epithelial integrity. Proc Natl Acad Sci U S A 109(7): E388-97.

Pathi SS, Lei P, Sreevalsan S, Chadalapaka G, Jutooru I & Safe S (2011) Pharmacologic doses of ascorbic acid repress specificity protein (Sp) transcription factors and Sp-regulated genes in colon cancer cells. Nutr Cancer 63(7): 1133-1142.

Patschinsky T, Hunter T, Esch FS, Cooper JA & Sefton BM (1982) Analysis of the sequence of amino acids surrounding sites of tyrosine phosphorylation. Proc Natl Acad Sci U S A 79(4): 973-977.

Pearson G, Robinson F, Beers Gibson T, Xu BE, Karandikar M, Berman K & Cobb MH (2001) Mitogen-activated protein (MAP) kinase pathways: regulation and physiological functions. Endocr Rev 22(2): 153-183.

Pece S & Gutkind JS (2000) Signaling from E-cadherins to the MAPK pathway by the recruitment and activation of epidermal growth factor receptors upon cell-cell contact formation. J Biol Chem 275(52): 41227-41233.

Platt G, Chou J, Houlihan P, Badran Y, Kumar L, Bainter W, Poliani L, Perez C, Dent S, Clapham D, Benavides F & Geha R (2017) J Allergy Clin Immunol 140(6): 1651-1659.

Prasad S, Gupta SC & Tyagi AK (2017) Reactive oxygen species (ROS) and cancer: Role of antioxidative nutraceuticals. Cancer Lett 387: 95-105.

Preston GM, Carroll TP, Guggino WB & Agre P (1992) Appearance of water channels in Xenopus oocytes expressing red cell CHIP28 protein. Science 256(5055): 385-387.

Proskuryakov SY, Konoplyannikov AG & Gabai VL (2003) Necrosis: a specific form of programmed cell death? Exp Cell Res 283(1): 1-16.

Qiu Z, Dubin AE, Mathur J, Tu B, Reddy K, Miraglia LJ, Reinhardt J, Orth AP & Patapoutian A (2014) SWELL1, a plasma membrane protein, is an essential component of volume-regulated anion channel. Cell 157(2): 447-458.

Raj L, Ide T, Gurkar AU, Foley M, Schenone M, Li X, Tolliday NJ, Golub TR, Carr SA, Shamji AF, Stern AM, Mandinova A, Schreiber SL & Lee SW (2011) Selective killing of cancer cells by a small molecule targeting the stress response to ROS. Nature 475(7355): 231-234.

Ravi M, Paramesh V, Kaviya SR, Anuradha E & Solomon FD (2015) 3D cell culture systems: advantages and applications. J Cell Physiol 230(1): 16-26.

108

Ravi M, Ramesh A & Pattabhi A (2017) Contributions of 3D Cell Cultures for Cancer Research. J Cell Physiol 232(10): 2679-2697.

Reynolds AB, Roesel DJ, Kanner SB & Parsons JT (1989) Transformation-specific tyrosine phosphorylation of a novel cellular protein in chicken cells expressing oncogenic variants of the avian cellular src gene. Mol Cell Biol 9(2): 629-638.

Richardson JC, Scalera V & Simmons NL (1981) Identification of two strains of MDCK cells which resemble separate nephron tubule segments. Biochim Biophys Acta 673(1): 26-36.

Rivier AS, Castillon GA, Michon L, Fukasawa M, Romanova-Michaelides M, Jaensch N, Hanada K & Watanabe R (2010) Exit of GPI-anchored proteins from the ER differs in yeast and mammalian cells. Traffic 11(8): 1017-1033.

Rodriguez-Boulan E & Macara IG (2014) Organization and execution of the epithelial polarity programme. Nat Rev Mol Cell Biol 15(4): 225-242.

Saito T, Sugiyama K, Takeshima Y, Amatya VJ, Yamasaki F, Takayasu T, Nosaka R, Muragaki Y, Kawamata T & Kurisu K (2017) Prognostic implications of the subcellular localization of survivin in glioblastomas treated with radiotherapy plus concomitant and adjuvant temozolomide. J Neurosurg : 1-6.

Saxén L and Sariola H (1987) Early organogenesis of the kidney. Pediatr Nephrol 1(3):385-92.

Schluter MA & Margolis B (2009) Apical lumen formation in renal epithelia. J Am Soc Nephrol 20(7): 1444-1452.

Schmalhofer O, Brabletz S & Brabletz T (2009) E-cadherin, beta-catenin, and ZEB1 in malignant progression of cancer. Cancer Metastasis Rev 28(1-2): 151-166.

Schnittler HJ, Puschel B & Drenckhahn D (1997) Role of cadherins and plakoglobin in interendothelial adhesion under resting conditions and shear stress. Am J Physiol 273(5 Pt 2): H2396-405.

Schroeder BC, Cheng T, Jan YN & Jan LY (2008) Expression cloning of TMEM16A as a calcium-activated chloride channel subunit. Cell 134(6): 1019-1029.

Serrano-Gomez SJ, Maziveyi M & Alahari SK (2016) Regulation of epithelial-mesenchymal transition through epigenetic and post-translational modifications. Mol Cancer 15: 18-016-0502-x.

Shapiro L, Fannon AM, Kwong PD, Thompson A, Lehmann MS, Grubel G, Legrand JF, Als-Nielsen J, Colman DR & Hendrickson WA (1995) Structural basis of cell-cell adhesion by cadherins. Nature 374(6520): 327-337.

Shukla PK, Gangwar R, Manda B, Meena AS, Yadav N, Szabo E, Balogh A, Lee SC, Tigyi G & Rao R (2016) Rapid disruption of intestinal epithelial tight junction and barrier dysfunction by ionizing radiation in mouse colon in vivo: protection by N-acetyl-l-cysteine. Am J Physiol Gastrointest Liver Physiol 310(9): G705-15.

Sigurbjornsdottir S, Mathew R & Leptin M (2014) Molecular mechanisms of de novo lumen formation. Nat Rev Mol Cell Biol 15(10): 665-676.

Simons K & Virta H (1987) Perforated MDCK cells support intracellular transport. EMBO J 6(8): 2241-2247.

109

Sohara E, Rai T, Miyazaki J, Verkman AS, Sasaki S & Uchida S (2005) Defective water and glycerol transport in the proximal tubules of AQP7 knockout mice. Am J Physiol Renal Physiol 289(6): F1195-200.

Souza-Morales J, da Silva Feltran G, Morales M (2014) CFTR and TNR-CFTR expression and function in the kidney. Biophys Rev 6: 227-236.

Standring S, Borley N, Collings P, Crossman A, Gatzoulis M, Healy J, Johnson D, Mahadevan V, Newell R, Wigley C (2008) Gray’s Anatomy The Anatomical Basis of Clinical Practice. Netherlands, Elsevier Limited.

Strumane K, Bonnomet A, Stove C, Vandenbroucke R, Nawrocki-Raby B, Bruyneel E, Mareel M, Birembaut P, Berx G & van Roy F (2006) E-cadherin regulates human Nanos1, which interacts with p120ctn and induces tumor cell migration and invasion. Cancer Res 66(20): 10007-10015.

Sullivan LP, Wallace DP & Grantham JJ (1998) Chloride and fluid secretion in polycystic kidney disease. J Am Soc Nephrol 9(5): 903-916.

Superti-Furga G, Fumagalli S, Koegl M, Courtneidge SA & Draetta G (1993) Csk inhibition of c-Src activity requires both the SH2 and SH3 domains of Src. EMBO J 12(7): 2625-2634.

Tabuse Y, Izumi Y, Piano F, Kemphues KJ, Miwa J & Ohno S (1998) Atypical protein kinase C cooperates with PAR-3 to establish embryonic polarity in Caenorhabditis elegans. Development 125(18): 3607-3614.

Takeda T, Go WY, Orlando RA & Farquhar MG (2000) Expression of podocalyxin inhibits cell-cell adhesion and modifies junctional properties in Madin-Darby canine kidney cells. Mol Biol Cell 11(9): 3219-3232.

Tanner GA, Maxwell MR & McAteer JA (1992) Fluid transport in a cultured cell model of kidney epithelial cyst enlargement. J Am Soc Nephrol 2(7): 1208-1218.

Thoreson MA, Anastasiadis PZ, Daniel JM, Ireton RC, Wheelock MJ, Johnson KR, Hummingbird DK & Reynolds AB (2000) Selective uncoupling of p120(ctn) from E-cadherin disrupts strong adhesion. J Cell Biol 148(1): 189-202.

Torres VE & Harris PC (2014) Strategies targeting cAMP signaling in the treatment of polycystic kidney disease. J Am Soc Nephrol 25(1): 18-32.

Turkson J, Bowman T, Garcia R, Caldenhoven E, De Groot RP & Jove R (1998) Stat3 activation by Src induces specific gene regulation and is required for cell transformation. Mol Cell Biol 18(5): 2545-2552.

Unruhe B, Schroder E, Wunsch D & Knauer SK (2016) An Old Flame Never Dies: Survivin in Cancer and Cellular Senescence. Gerontology 62(2): 173-181.

Uren AG, Wong L, Pakusch M, Fowler KJ, Burrows FJ, Vaux DL & Choo KH (2000) Survivin and the inner centromere protein INCENP show similar cell-cycle localization and gene knockout phenotype. Curr Biol 10(21): 1319-1328.

Ussing HH (1947) Interpretation of the exchange of radio-sodium in isolated muscle. Nature 159(4060): 262.

van der Horst A & Lens SM (2014) Cell division: control of the chromosomal passenger complex in time and space. Chromosoma 123(1-2): 25-42.

110

van Roy FM & McCrea PD (2005) A role for Kaiso-p120ctn complexes in cancer? Nat Rev Cancer 5(12): 956-964.

Vega-Salas DE, Salas PJ & Rodriguez-Boulan E (1987) Modulation of the expression of an apical plasma membrane protein of Madin-Darby canine kidney epithelial cells: cell-cell interactions control the appearance of a novel intracellular storage compartment. J Cell Biol 104(5): 1249-1259.

Venkiteswaran K, Xiao K, Summers S, Calkins CC, Vincent PA, Pumiglia K & Kowalczyk AP (2002) Regulation of endothelial barrier function and growth by VE-cadherin, plakoglobin, and beta-catenin. Am J Physiol Cell Physiol 283(3): C811-21.

Venugopal J & Blanco G (2017) On the Many Actions of Ouabain: Pro-Cystogenic Effects in Autosomal Dominant Polycystic Kidney Disease. Molecules 22(5): 10.3390/molecules22050729.

Verkman AS & Galietta LJ (2009) Chloride channels as drug targets. Nat Rev Drug Discov 8(2): 153-171.

Vogelstein B, Papadopoulos N, Velculescu VE, Zhou S, Diaz LA,Jr & Kinzler KW (2013) Cancer genome landscapes. Science 339(6127): 1546-1558.

Watanuki-Miyauchi R, Kojima Y, Tsurumi H, Hara T, Goto N, Kasahara S, Saio M, Moriwaki H & Takami T (2005) Expression of survivin and of antigen detected by a novel monoclonal antibody, T332, is associated with outcome of diffuse large B-cell lymphoma and its subtypes. Pathol Int 55(6): 324-330.

Weide T, Vollenbroker B, Schulze U, Djuric I, Edeling M, Bonse J, Hochapfel F, Panichkina O, Wennmann DO, George B, Kim S, Daniel C, Seggewiss J, Amann K, Kriz W, Krahn MP & Pavenstadt H (2017) Pals1 Haploinsufficiency Results in Proteinuria and Cyst Formation. J Am Soc Nephrol 28(7): 2093-2107.

Willenborg C & Prekeris R (2011) Apical protein transport and lumen morphogenesis in polarized epithelial cells. Biosci Rep 31(4): 245-256.

Wills N, Reuss L, Lewis S (1996) Epithelial Transport A guide to methods and experimental analysis. United Kingdoms, Chapman & Hall.

Wyke JA & Linial M (1973) Temperature-sensitive avian sarcoma viruses: a physiological comparison of twenty mutants. Virology 53(1): 152-161.

Xiao K, Allison DF, Buckley KM, Kottke MD, Vincent PA, Faundez V & Kowalczyk AP (2003) Cellular levels of p120 catenin function as a set point for cadherin expression levels in microvascular endothelial cells. J Cell Biol 163(3): 535-545.

Xie Z, Kometiani P, Liu J, Li J, Shapiro JI & Askari A (1999) Intracellular reactive oxygen species mediate the linkage of Na+/K+-ATPase to hypertrophy and its marker genes in cardiac myocytes. J Biol Chem 274(27): 19323-19328.

Xu W, Harrison SC & Eck MJ (1997) Three-dimensional structure of the tyrosine kinase c-Src. Nature 385(6617): 595-602.

Yakisich JS, Azad N, Kaushik V, O'Doherty GA & Iyer AK (2017) Nigericin decreases the viability of multidrug-resistant cancer cells and lung tumorspheres and potentiates the effects of cardiac glycosides. Tumour Biol 39(3): 1010428317694310.

111

Yang B, Sonawane ND, Zhao D, Somlo S & Verkman AS (2008) Small-molecule CFTR inhibitors slow cyst growth in polycystic kidney disease. J Am Soc Nephrol 19(7): 1300-1310.

Yap AS & Kovacs EM (2003) Direct cadherin-activated cell signaling: a view from the plasma membrane. J Cell Biol 160(1): 11-16.

Yap AS, Niessen CM & Gumbiner BM (1998) The juxtamembrane region of the cadherin cytoplasmic tail supports lateral clustering, adhesive strengthening, and interaction with p120ctn. J Cell Biol 141(3): 779-789.

Yasui M, Hazama A, Kwon TH, Nielsen S, Guggino WB & Agre P (1999) Rapid gating and anion permeability of an intracellular aquaporin. Nature 402(6758): 184-187.

Yuajit C, Homvisasevongsa S, Chatsudthipong L, Soodvilai S, Muanprasat C & Chatsudthipong V (2013) Steviol reduces MDCK Cyst formation and growth by inhibiting CFTR channel activity and promoting proteasome-mediated CFTR degradation. PLoS One 8(3): e58871.

Zhao J, Tenev T, Martins LM, Downward J & Lemoine NR (2000) The ubiquitin-proteasome pathway regulates survivin degradation in a cell cycle-dependent manner. J Cell Sci 113 Pt 23: 4363-4371.

112

113

Original articles

I Capra JP, Eskelinen SM (2017) MDCK cells are capable of water secretion and reabsorption in response to changes in the ionic environment. Can J Physiol Pharmacol 95(1): 72-83.

II Töyli M, Rosberg-Kulha L, Capra J, Vuoristo J, Eskelinen S (2010) Different responses in transformation of MDCK cells in 2D and 3D culture by v-Src as revealed by microarray techniques, RT-PCR and functional assays. Lab Invest 90(6):915-928.

III Capra JP, Eskelinen SM (2017) Correlation between E-cadherin interactions, survivin expression, and apoptosis in MDCK and ts-Src MDCK cell culture models. Lab Invest 97(12):1453-1470.

Reprinted with permission from Canadian Science Publishing (I) and Springer

Nature (II and III)

Original publications are not included in the electronic version of the dissertation.

114

A C T A U N I V E R S I T A T I S O U L U E N S I S

Book orders:Granum: Virtual book storehttp://granum.uta.fi/granum/

S E R I E S D M E D I C A

1437. Filatova, Svetlana (2017) Incidence of schizophrenia and associations ofschizophrenia and schizotypy with early motor developmental milestones

1438. Käräjämäki, Aki (2017) Non-alcoholic fatty liver disease (NAFLD) : perspectivesto etiology, complications and lipid metabolism

1439. Mikkola, Reija (2017) Determinants and clinical implications of bleeding related tocoronary artery bypass surgery

1440. Hagnäs, Magnus (2018) The association of cardiorespiratory fitness, physicalactivity and ischemic ECG findings with coronary heart disease-related deathsamong men

1441. Huhtaniska, Sanna (2018) The association between antipsychotic andbenzodiazepine use with brain morphology and its changes in schizophrenia

1442. Sundquist, Elias (2018) The role of tumor microenvironment on oral tonguecancer invasion and prognosis

1443. Keto, Jaana (2018) The middle-aged smoker in health care : primary health careuse, cardiovascular risk factors, and physician’s help in quitting

1444. Tiisanoja, Antti (2018) Sedative load and oral health among community-dwellingolder people

1445. Alaraudanjoki, Viivi (2018) Erosive tooth wear and associated factors in NorthernFinland Birth Cohort 1966

1446. Sinikumpu, Suvi-Päivikki (2018) Skin diseases and their association with systemicdiseases in the Northern Finland Birth Cohort 1966

1447. Nortunen, Simo (2018) Stability assessment of isolated lateral malleolarsupination-external rotation-type ankle fractures

1448. Härönen, Heli (2018) Collagen XIII as a neuromuscular synapse organizer : rolesof collagen XIII and its transmembrane form, and effects of shedding andoverexpression in the neuromuscular system in mouse models

1449. Kajula, Outi (2018) Periytyvän rintasyöpäalttiusmutaation (BRCA1/2)kantajamiesten hypoteettinen perinnöllisyysneuvontamalli

1450. Jurvelin, Heidi (2018) Transcranial bright light – the effect on humanpsychophysiology

1451. Järvimäki, Voitto (2018) Lumbar spine surgery, results and factors predictingoutcome in working-aged patients

UNIVERSITY OF OULU P .O. Box 8000 F I -90014 UNIVERSITY OF OULU FINLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

University Lecturer Tuomo Glumoff

University Lecturer Santeri Palviainen

Postdoctoral research fellow Sanna Taskila

Professor Olli Vuolteenaho

University Lecturer Veli-Matti Ulvinen

Planning Director Pertti Tikkanen

Professor Jari Juga

University Lecturer Anu Soikkeli

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-952-62-1822-9 (Paperback)ISBN 978-952-62-1823-6 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

U N I V E R S I TAT I S O U L U E N S I S

MEDICA

ACTAD

D 1452

AC

TAJanne C

apra

OULU 2018

D 1452

Janne Capra

DIFFERENTIATION AND MALIGNANT TRANSFORMATION OF EPITHELIAL CELLS

3D CELL CULTURE MODELS

UNIVERSITY OF OULU GRADUATE SCHOOL;UNIVERSITY OF OULU,FACULTY OF MEDICINE;BIOCENTER OULU