Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Targeted therapy in low-grade serous ovarian carcinoma : characterization of MEK inhibitor response in… Dawson, Amy 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_november_dawson_amy.pdf [ 19.36MB ]
Metadata
JSON: 24-1.0319145.json
JSON-LD: 24-1.0319145-ld.json
RDF/XML (Pretty): 24-1.0319145-rdf.xml
RDF/JSON: 24-1.0319145-rdf.json
Turtle: 24-1.0319145-turtle.txt
N-Triples: 24-1.0319145-rdf-ntriples.txt
Original Record: 24-1.0319145-source.json
Full Text
24-1.0319145-fulltext.txt
Citation
24-1.0319145.ris

Full Text

TARGETED	THERAPY	IN	LOW-GRADE	SEROUS	OVARIAN	CARCINOMA:	CHARACTERIZATION	OF	MEK	INHIBITOR	RESPONSE	IN	NOVEL	PATIENT-DERIVED	CELL	LINES	by  Amy Dawson B.Sc., The University of Victoria, 2014   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Reproductive and Developmental Sciences) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) October 2016  © Amy Dawson, 2016 	 ii	ABSTRACT		Low-grade	serous	ovarian	carcinoma,	or	LGSOC,	predominantly	occurs	in	pre-menopausal	women	at	lower	frequency	than	the	more	common	high-grade	serous	ovarian	carcinoma.	It	tends	to	harbour	few,	but	distinct,	mutations,	with	a	significant	proportion	of	RAS	oncogene	mutations.	Aberrations	of	the	RAS-MAPK	signaling	pathway	in	LGSOC	have	led	to	the	initiation	of	several	clinical	trials	of	MEK	inhibitors	in	this	disease.	Due	to	the	relative	rarity	of	LGSOC,	few	representative	model	systems	are	available.	Accordingly,	there	has	been	a	lack	of	preclinical	investigational	drug	testing	in	this	disease.	In	this	study,	11	histotype-specific	cell	lines	have	been	derived	and	molecularly	characterized	in	order	to	further	elucidate	the	molecular	biology	of	the	disease,	and	serve	as	an	important	platform	to	test	novel	therapeutics	in	LGSOC.	The	most	common	missense	mutations	by	HotSpot	mutational	profiling	analysis	were	oncogenic	KRAS/NRAS	mutations,	and	varying	levels	of	copy	number	aberration	were	seen.	All	cell	lines	were	uniquely	identifiable	by	short-tandem	repeat	analysis,	with	the	exception	of	two	patient-paired	cell	lines.		The	phenotypic	and	molecular	responses	to	the	4	MEK	inhibitors	(MEKi)	trametinib,	selumetinib,	refametinib,	and	binimetinib	have	been	characterized	in	selected	LGSOC	cell	lines	through	IC50	analysis,	proliferation,	viability,	and	apoptosis	assays,	and	on-target	drug	effects	by	Western	blot.	Of	the	4	MEKi,	trametinib	exhibited	the	best	anti-proliferative	effects	and	inhibition	of	MEK	kinase		 iii	activity.	Biological	and	on-target	data	from	two	of	the	cell	lines	reveals	an	exquisite	sensitivity	to	MEK	inhibition.	Reverse-phase	protein	array	and	quantitative	mass	spectrometry	global	proteomic	analysis	on	control	and	MEKi-treated	LGSOC	cell	lines	was	performed	in	order	to	examine	basal	proteomic	differences	between	MEKi-sensitive	and	resistant	LGSOC	cell	lines	and	gather	data	for	examining	proteomic	changes	upon	MEKi	treatment.	Three	significantly	differentially	expressed	protein	candidates	(PKCα,	EGFR,	and	Smac/DIABLO	were	identified	between	sensitive	and	resistant	cell	lines	under	control	and	MEKi	treatment	conditions	by	both	proteomic	platforms.		Results	from	therapeutic	testing	in	these	pathologically	reviewed,	histotype-specific	LGSOC	models	allow	for	a	comparison	of	the	overall	efficacies	of	the	4	MEKi,	characterization	of	drug	sensitivity	in	novel	cell	lines,	and	identification	of	potential	functional	markers	of	MEKi	response.											 iv	PREFACE			 Data	from	Chapters	2	and	3	has	been	submitted	to	the	American	Journal	of	Cancer	Research	for	publishing:	Llaurado	Fernandez,	M.,	DiMattia,	G.,	Dawson	A.,	Bamford,	S.,	Anderson,	S.,	Hennessey,	B.,	Anglesio,	M.,	Shepard,	T.,	Salamanca,	C.,	Hoenisch,	J.,	Tinker,	A.,	Huntsman,	D.,	Carey,	M.,	Differences	in	MEK	inhibitor	efficacy	in	molecularly	characterized	low-grade	serous	ovarian	cancer	cell	lines.	Accepted	September	2016.	DOCUMENTATION	OF	MY	SPECFIC	CONTRIBUTIONS	TO	THE	CAREY	LAB’S	WORK	ON	LOW	GRADE	SEROUS	OVARIAN	CARCINOMAS		Development	and	molecular	characterization	of	LGSOC	cell	lines:		 Some	of	the	cell	lines	and	primary	cultures	in	this	study	were	previously	established	by	Clara	Salamanca	and	Dr.	Marta	Llaurado:	VOA-1056,	1312,	3448,	3723,	3993,	4627,	4698,	and	4881.	The	LGSOC	cell	line	iOvCa241	was	established	by	Dr.	Gabriel	DiMattia	at	the	University	of	Western	Ontario.	I	am	currently	responsible	for	continual	culture	and	maintenance	of	all	LGSOC	cell	lines	and	frozen	cell	stock	library,	as	well	as	development	and	enrichment	of	new	primary	patient	samples	from	LGOSC	or	serous	borderline	ovarian	(SBOT)	tumors	and	ascites.	Cell	lines	and	primary	cultures	that	I	have	been	instrumental	in	establishing	since	the	commencement	of	this	study	include	VOA-6406,	VOA-6800,	and	several	others	(VOA-6857,	VOA-7263,	VOA-7604,	VOA-7608)	not	discussed	in	this	thesis.	Incucyte®	images	were	obtained	from	experiments		 v	performed	by	myself	and	Dr.	Marta	Llaurado.	I	was	responsible	for	amplifying	cell	cultures	and	collecting	cell	pellets	for	DNA	extraction	from	all	LGSOC	lines,	performing	DNA	extraction	and	quantification	in	collaboration	with	Dr.	Marta	Llaurado.	The	HotSpot	panel	for	iOvCa241	was	run	by	Dr.	Gabriel	DiMattia,	and	data	analyzed	and	Sanger	sequencing	primers	designed	by	Dr.	Marta	Llaurado.		Sanger	sequencing	validation	of	HotSpot	candidates	was	performed	by	Janine	Senz.	STR	profiling	analysis	was	performed	by	GeneWiz	Inc.	Copy	number	variation	experiments	for	iOvCa241	were	performed	by	Dr.	Gabriel	DiMattia,	and	for	all	other	LGSOC	cell	lines	by	Génome	Québec	and	McGill	University	Innovation	Centre.	CNV	data	was	analyzed	by	Dr.	Marta	Llaurado	and	the	Vancouver	Prostate	Centre	bioinformatics	research	team.	Investigation	of	drug	effects	of	MEKi	in	LGSOC		 Study	design	was	conceptualized	by	Dr.	Marta	Llaurado	and	Dr.	Mark	Carey.	I	was	involved	in	design,	cell	culture,	execution,	and	data	analysis	of	IC50	experiments	and	Incucyte™	proliferation	experiments,	and	execution	of	MTS	viability	staining	with	support	from	Dr.	Marta	Llaurado.		I	performed	preparation	of	lysates	and	execution	of	all	Western	blots	with	support	from	Dr.	Marta	Llaurado.	Capillary	isoelectric	point	focusing	(cIEF)	work	and	data	analysis	was	performed	by	Sylvia	Bamford	and	Dr.	Marta	Llaurado,	using	lysates	prepared	by	Dr.	Llaurado	and	myself.		I	was	involved	with	preparation	of	cell	cultures	and	plates	for	caspase	experiments,	which	were	executed	and	the	data	analyzed	by	Dr.	Llaurado.	Josh	Hoenisch	provided	support	to	experiments	run	from	January	2016	onwards.			 vi		Proteomic	experiments	(Reverse	phase	protein	analysis	(RPPA)	and	mass	spectrometry	(MS))		 I	was	responsible	for	experimental	design,	cell	culture,	drug	treatment,	cell	harvest,	adjusting	concentrations,	and	denaturing	of	resultant	lysates	for	all	samples	submitted	for	RPPA	analysis	with	support	from	Sylvia	Bamford	and	Dr.	Marta	Llaurado,	as	well	as	analyzing	RPPA	data	with	Dr.	Mark	Carey	and	Dr.	Llaurado.	The	RPPA	experiments	including	preliminary	data	analysis	and	technical	support	were	run	by	Dr.	Bryan	Hennessey’s	lab	at	the	Royal	College	of	Surgeons	in	Ireland.	SPSS	statistical	analysis	and	heat	maps	of	RPPA	data	was	performed	by	Dr.	Mark	Carey	and	myself.		 I	assumed	a	lead	role	in	the	mass	spectrometry	arm	of	the	research	project.	I	was	responsible	for	performing	preliminary	research	on	the	MS	techniques,	and	with	this	knowledge	helped	design	the	experiments	and	provided	input	on	the	technical	aspects	of	the	MS.	I	was	also	responsible	for	cell	culture,	drug	treatment,	and	cell	harvest	for	all	samples	submitted	for	mass	spectrometry	analysis.	Sample	preparation	(purification,	digestion,	and	tandem	mass	tagging)	of	24	and	48	hour	mass	spectrometry	samples	was	performed	in	collaboration	with	Shane	Colborne	from	Dr.	Gregg	Morin’s	laboratory,	who	was	responsible	for	all	technical	aspects	of	the	mass	spectrometry	experiment	as	well	as	for	the	cursory	and	Protein	expression	control	analysis	(PECA).	As	part	of	the	project	I	observed	the	work-flow	for	the	MS	sample	preparation	and	prepared	a	summary	for	our	research	group.	Investigation	into	the	fidelity	and	overall		 vii	quality	of	the	mass	spectrometry	data	was	performed	by	myself,	Josh	Hoenisch,	and	Shane	Colborne.		Deeper	analysis	of	the	mass	spectrometry	data	was	performed	by	myself,	Josh	Hoenish,	and	Dr.	Marta	Llaurado.																						 viii		TABLE	OF	CONTENTS	ABSTRACT	................................................................................................................................................	ii	PREFACE	..................................................................................................................................................	iv	TABLE	OF	CONTENTS	.......................................................................................................................	viii	LIST	OF	TABLES	......................................................................................................................................	x	LIST	OF	FIGURES	...................................................................................................................................	xi	LIST	OF	ABBREVIATIONS	.................................................................................................................	xiv	ACKNOWLEDGEMENTS	.....................................................................................................................	xvi	COLLABORATIVE	PERSONNEL:	.................................................................................................................	xvii	CHAPTER	1.	INTRODUCTION,	BACKGROUND,	AND	AIMS	.........................................................	1	1.1.	 INTRODUCTION	..............................................................................................................................................	1	1.2.	BACKGROUND	.....................................................................................................................................................	2	1.3.	GENERAL	HYPOTHESES	AND	AIMS	..............................................................................................................	10	CHAPTER	2.	DEVELOPMENT	AND	MOLECULAR	CHARACTERIZATION	OF	NOVEL	LOW-GRADE	SEROUS	OVARIAN	CARCINOMA	CELL	LINES	.................................................................	11	2.1.	AIMS	..................................................................................................................................................................	11	2.2.	METHODS	.........................................................................................................................................................	12	2.2.1.	Establishment	of	LGSOC	cell	cultures	from	patient	samples	...............................................	12	2.2.2.	Molecular	characterization	of	newly-established	LGSOC	cell	cultures	...........................	13	2.3.	RESULTS	AND	DISCUSSION	...........................................................................................................................	15	CHAPTER	3.	EVALUATION	OF	BIOLOGICAL	AND	ON-TARGET	EFFECTS	OF	4	COMMERCIALLY	AVAILABLE	MEK	INHIBITORS	IN	PATIENT-DERIVED	LGSOC	CELL	LINES	........................................................................................................................................................	27	3.1.	HYPOTHESIS	AND	AIMS	.................................................................................................................................	27	3.2.	METHODS	.........................................................................................................................................................	27	3.2.1.	Biological	and	on-target	drug	effects	of	4	MEKi	.......................................................................	27	3.3.	RESULTS	AND	DISCUSSION	...........................................................................................................................	31	3.3.1.	Biological	effects	.....................................................................................................................................	31	CHAPTER	4.	GLOBAL	PROTEOMIC	ANALYSIS	IN	SENSITIVE	AND	RESISTANT	LGSOC	CELL	LINES	.............................................................................................................................................	56	4.1.	HYPOTHESIS	AND	AIMS	.................................................................................................................................	56	4.2.	METHODS	.........................................................................................................................................................	57	4.2.1.	Reverse	Phase	Protein	Array	(RPPA)	signaling	pathway	analysis	of	MEKi	sensitive	and	resistant	LGSOC	cell	lines.	......................................................................................................................	57		 ix	4.2.2.	Global	Quantitative	Mass	Spectrometry	of	MEKi	sensitive	and	resistant	LGSOC	cell	lines.	..........................................................................................................................................................................	58	4.2.3	Western	blot	validation	of	RPPA	and	MS	results	.......................................................................	60	4.3.	RESULTS	AND	DISCUSSION	...........................................................................................................................	61	CHAPTER	5.	CONCLUSIONS	AND	FUTURE	DIRECTIONS	..........................................................	74	5.1.	CONCLUSIONS	..................................................................................................................................................	74	5.2.	FUTURE	DIRECTIONS	......................................................................................................................................	77	REFERENCES	..........................................................................................................................................	79	APPENDIX:	REAGENTS	AND	MATERIALS	.....................................................................................	86																														 x					LIST	OF	TABLES		Table	2.3.1.	Primary	pathology,	pathology	at	time	of	sampling,	and	patient	information	for	each	LGSOC	cell	line	and	primary	culture	currently	established.	..........	17		Table	2.3.2.	Short	tandem	repeat	(STR)	profiling	data	for	8	LGSOC	cell	lines.	..................	20		Table	2.3.4.	Copy	number	variation	data	(%	genome	change,	total	copy	number	[CN]	aberrations,	%	loss	of	heterozygosity	[LOH],	type	of	platform,	CN	gains	and	losses)	in	select	LGSOC	cell	lines.	................................................................................................................................	24		Table	4.3.1.	SPSS	results	from	independent	samples	Mann	Whitney	U	test:	significantly	different	protein	expression	levels	of	lysates	treated	with	control	(DMSO)	and	0.1μM	trametinib/1μM	refametinib	between	sensitive	(iOvCa241	and	VOA-1312)	and	resistant	(VOA-1056,	VOA-3993,	VOA-3723,	VOA-3448,	VOA-4627,	VOA-4698,	and	VOA-4881)	LGSOC	cell	lines.	...............................................................................................................................	65		Table	4.3.2.	Mass	spec	data	(48hr	drug	treatment	experiment)	for	statistically	significant	RPPA	candidates.	Raw	protein	abundance	values	for	VOA-1056	and	VOA-4627	were	normalized	to	iOvCa241	abundance	levels	for	both	the	control	and	the	drug-treated	data	sets	in	order	to	illustrate	fold-change	differences	in	each	data	set.	Values	were	obtained	from	averaging	raw	abundance	data	from	three	biological	replicates.	.........................................................................................................................................................	68		Table	4.3.3.	Mass	spectrometry	PECA-analyzed	data,	showing	slr	(log2	of	the	fold	change)	and	p.fdr	(false-discovery	rate	adjusted	p-values)	for	statistically	significant	RPPA	candidates	by	SPSS.		........................................................................................................................	70		Table	A2.	List	of	proteins	tested	in	RPPA	array.	Known	oncogenic	candidates,	as	well	as	moieties	associated	with	apoptotic	and	MAPK-related	pathways	were	chosen	to	include	in	the	array.	Two	different	antibodies	against	phospho-MAPK/ERK1/2	were		 xi	used,	that	which	was	used	in	the	Western	blots	in	this	study	and	one	provided	by	the	RPPA	service.	...................................................................................................................................................	95		LIST	OF	FIGURES		Figure	1.2.1.	RAS-MAPK	signaling	pathway.	Extracellular	ligand	binding	to	receptor	tyrosine	kinases	(RTK)	induces	activation	of	RAS-GDP	to	RAS-GTP	through	a	series	of	intermediate	mechanisms.	Activated	RAS	initiates	a	downstream	signaling	cascade	through	RAF,	MEK,	and	ERK,	resulting	in	enhanced	proliferation,	survival,	angiogenesis,	migration,	and	cell	cycle	regulation[43].	..................................................................................................................................................................	6		Figure	2.3.1	Morphology	of	LGSOC	cell	lines	developed.	Images	were	taken	in	a	Corning	96-well	plate	on	an	Incucyte Live	Imaging	System	at	10X	magnification	(Essen	BioScience).	..................	19		Figure	3.3.1	IC50	values	for	4	commercially	available	MEK	inhibitors	in	11	newly-established	LGSOC	cell	lines.	A.	IC50	values	organized	by	drug.	B.	IC50	values	organized	by	cell	line.		Cell	lines	were	treated	for	72	hours	with	a	spectrum	of	drug	dosages	from	0-100μM,	stained	with	0.25%	crystal	violet,	and	absorbance	read	at	595nm	on	a	microplate	reader.	Cells	were	seeded	so	that	control	wells	would	be	100%	confluent	at	time	of	staining.	......................................................	34		Figure	3.3.2.	(previous	page)	Incucyte®	proliferation	curves	for	LGSOC	cell	lines	treated	with	4	different	MEKi	at	a	1X	dose	determined	by	literature	review	and	IC50	data.	Cells	were	treated	at	10-20%	confluence,	and	MTS	viability	staining	applied	when	control	wells	reached	~100%	confluence.	.......................................................................................................................................................................	36		Figure	3.3.3.	MTS	viability	staining	following	5-7	day	Incucyte®	proliferation	assays.	Reagent	was	applied	for	3.5	hours	prior	to	reading	absorbance	at	490nm.	........................................................	36		Figure	3.3.4.	Incucyte®	proliferation	curves	for	two	selected	resistant	(VOA-1056,	VOA-4627),	and	one	sensitive	(iOvCa241)	cell	lines	treated	with	low	(0.1	or	1μM)	and	high	(0.5	or	5μM)	doses	of	MEKi.	.................................................................................................................................................................	38		Figure	3.3.5.	Basal	levels	of	phospho	and	total	ERK1/2	and	MEK1/2	for	LGSOC	cell	lines.	All	cell	lines	were	treated	with	DMSO	with	the	exception	of	VOA-6800	and	VOA-6406.	.....................	40			 xii	Figure	3.3.6.	Western	blot	evaluation	of	on-target	inhibition	of	ERK	phosphorylation	by	MEK	inhibitors	for	8	LGSOC	cell	lines.	Cell	lines	were	treated	with	drugs	for	24	hours	prior	to	EGF	stimulation	and	lysate	preparation.	Note:	All	phospho-ERK1/2	blots	were	overexposed	to	show	as	much	signal	as	possible	after	MEKi	treatment.	For	each	cell	line,	all	antibodies	come	from	the	same	blot	and	are	exposed	at	the	same	time,	however	images	were	cut	and	re-ordered	to	ensure	the	same	sample	order	for	presentation	purposes.	.........................................................................	42		Figure	3.3.7.	[A]	cIEF(NanoPro)	results	for	iOvCa241	and	VOA-1056	[B]	cIEF	results	for	VOA-3723	and	VOA-4627.	All	four	cell	lines	were	treated	with	trametinib,	selumetinib,	binimetinib,	and	refametinib.	Stimulation	with	EGF	leads	to	forced	signaling	and	more	pronounced	peaks.	An	antibody	against	ERK1/2	was	used	to	identify	ERK	isoforms	and	phosphorylation	in	each	sample	condition.	Shift	of	peaks	towards	a	more	acidic	isoelectric	point	(pI)	represent	increasing	levels	of	phosphorylation	of	ERK.	....................................................................................................	45		Figure	3.3.8.		Dose-response	Western	blots.	LGSOC	cell	lines	treated	with	two	doses	of	MEKi.	Cells	were	harvested	at	24	and	72	hours,	with	no	EGF	stimulation.	Note:	All	phospho-ERK1/2	blots	were	overexposed	to	show	as	much	signal	as	possible	after	MEKi	treatment.	For	each	cell	line,	all	antibodies	come	from	the	same	blot	and	are	exposed	at	the	same	time;	however	images	were	cut	and	re-ordered	to	ensure	the	same	sample	order	for	presentation	purposes.	................	48		Figure	3.3.9.	Caspase	3/7	activity	assay	(Caspase-Glo)	on	three	LGSOC	cell	lines.	Increases	in	cleaved	caspase	over	the	baseline	control	(values	subtracted)	are	shown.	........................................	51		Figure	3.3.10.	RPPA	results	for	phosphorylated	and	total	ERK1/2,	phosphorylated	MEK1/2	and	total	MEK1	for	cell	lines	treated	with	the	4	MEKi	and	control	DMSO	with	and	without	EGF	stimulation	after	24	hours.	Red	indicates	low	expression	values.	Biological	triplicate	data	was	averaged.	Antibody	signal	was	normalized	by	serial	dilution	curves	according	to	protocols	established	by	the	Royal	College	of	Surgeons,	Ireland.	................................................................................	53		Figure	3.3.11.	RPPA	results	for	phospho	and	total	ERK1/2,	phospho	MEK1/2	and	total	MEK1	for	one	selected	sensitive	(iOvCa241)	and	two	selected	resistant	(VOA-3723	and	VOA-4627)		cell	lines	treated	with	the	4	MEKi	and	control	DMSO	with	and	without	EGF	stimulation	after	24	and	72	hours.	Red	indicates	low	expression	values.	Biological	triplicate	data	was	averaged................................................................................................................................................................................................	54		Figure	4.2.1.	Triple	quadrupole	mass	spectrometer	schematic.	..............................................................	60	Figure	4.3.1.	RPPA	heatmaps	of	LGSOC	sensitive	and	resistant	cell	lines	treated	with	control	or	0.1μM	trametinib.	Averaged	data	from	three	biological	replicates	was	used.	Red	indicates	low	expression	values.	.........................................................................................................................................................	63		 xiii		Figure	4.3.3.	Western	blot	validation	for	two	selected	candidates,	PKCα	and	EGFR,	identified	as	differentially	expressed	at	basal	level	(non-DMSO	treated)	between	sensitive	and	resistant	LGSOC	cell	lines.	.............................................................................................................................................................	73		Figure	A1.	Western	Blot	analysis	of	VOA-1312	and	VOA-1056	for	cleaved	PARP	induction	after	24	hour	treatment	with	0.1μM	trametinib	or	Control	(DMSO).	Lysates	were	non-EGF	stimulated.	.......................................................................................................................................................................	93		Figure	A2.	Mass	spectrometry	experimental	design	of	24	and	48	hour	sample	sets.	.....................	97		Figure	A3.	Reverse-phase	HPLC	pre-fractionation	of	TMT-labeled	peptides	for	24	and	48	hour	mass	spectrometry	experiments.	Samples	were	run	on	a	Phenomenex	EVO	C18	Core-Shell	Column	(15cm	length	x	2.1mm	diameter;	1.7μM	diameter	beads)	and	48	fractions	were	collected	from	minute	19	to	minute	73	of	80.	..................................................................................................	98																	 xiv		LIST	OF	ABBREVIATIONS		AKT	 Protein	Kinase	B	ARID1A	 AT-Rich	Interaction	Domain	1A					BIN	 Binimetinib			BRAF	 serine/threonine-protein	kinase	B-Raf				BRCA1/2		 Breast	Cancer	1/2	c-PARP	 Cleaved	poly-ADP	ribose	polymerase	cIEF	 Capillary	isoelectric	point	focusing	COD	 Collision-induced	dissociation	DMSO	 Dimethyl	sulfoxide	DUSP	 Dual-specificity	phosphatase	EGF	 Epidermal	growth	factor	EGFR	 Epidermal	growth	factor	receptor	ERBB2	 Erb-B2	Receptor	Tyrosine	Kinase	2		ERK1/2	 Mitogen-activated	protein	kinase	1/2	ESI	 Electrospray	ionization	FBS	 Fetal	bovine	serum	FGFR3	 Fibroblast	growth	factor	receptor	3	HBSS	 Hank’s	balanced	salt	solution	HGSOC	 High-grade	serous	ovarian	carcinoma	HNF1B	 Hepatocyte	nuclear	factor-1-beta		HPLC	 High-performance	liquid	chromatography	HRAS	 Harvey	Rat	Sarcoma	Viral	Oncogene	Homolog		IC50	 Inhibitory	concentration	required	to	kill	50%	of	cells	JAK3	 Janus	kinase	3	KDR	 Kinase	insert	domain	receptor	KIT	 KIT	Proto-Oncogene	Receptor	Tyrosine	Kinase		KRAS	 Kirsten	rat	sarcoma	viral	oncogene	homolog	LGSOC	 Low-grade	serous	ovarian	carcinoma	MAPK	 Mitogen-activated	protein	kinase	MEK1/2	 Mitogen-activated	protein	kinase	kinase	1/2		MEKi	 MEK	inhibitors	MET	 MET	Proto-Oncogene,	Receptor	Tyrosine	Kinase		MILO	 MEK	Inhibitor	in	Low-grade	serous	Ovarian	cancer	MPSC1	 Micropapillary	serous	carcinoma	1	cell	line	p-ERK1/2	 Phosphorylated	mitogen-activated	protein	kinase	1/2	p-MEK1/2	 Phosphorylated	mitogen-activated	protein	kinase	1/2	PI3K		 Phosphoinositide-3	kinase	PARP	 Poly-ADP	ribose	polymerase	PECA	 Protein	expression	control	analysis		 xv	PIK3CA	 Phosphatidylinositol-4,5-Bisphosphate	3-Kinase	Catalytic		subunit	alphaSubunit	Alpha		PKCα	 Protein	kinase	C	alpha	subunit	RAF	 Rapidly	accelerated	fibrosarcoma	kinase	RAS	 Rat	sarcoma	GTPase	REF	 Refametinib	RPPA	 Reverse-phase	protein	array	SEL	 Selumetinib;	AZD6244	STIC	 serous	tubal	intra-epithelial	carcinoma	STR	 Short	tandem	repeat	TMT	 Tandem-mass	tagging	TP53	 Tumor	protein	p53	TRA	 Trametinib	WB	 Western	blot	XIAP	 X-linked	inhibitor	of	apoptosis																		 xvi	ACKNOWLEDGEMENTS			 This	work	is	supported	by	the	infrastructure	of	the	OvCaRe	research	team	(a	multidisciplinary	group	of	clinicians,	scientists,	and	research	staff),	providing	reagents,	work	space,	and	collaboration	opportunities	to	our	team.	OvCaRe	is	funded	by	the	BC	Cancer	Foundation.	 		 I	would	like	to	offer	immense	gratitude	for	the	guidance	and	support	offered	to	me	by	Dr.	Mark	Carey	and	Dr.	Marta	Llaurado	for	the	duration	of	my	graduate	program.	I	am	especially	grateful	for	the	independence,	critical	thinking	skills,	and	research	insight	I	have	developed	under	their	supervision.	I	also	offer	immense	thanks	to	Sylvia	Bamford	and	Clara	Salamanca	for	their	technical	expertise	and	contribution	to	my	organizational	and	tactile	skills	development,	and	to	Josh	Hoenisch	for	his	research	support	during	his	stay	in	our	laboratory.	To	the	investigators	Dr.	Gregg	Morin,	Dr.	Bryan	Hennessey,	and	Dr.	Gabriel	DiMattia,	I	offer	appreciation	for	the	resources,	services,	and	technical	guidance	provided	to	complete	this	project.		 Finally,	I	would	like	to	extend	immeasurable	thanks	to	my	family	for	their	constant	love	and	support.	To	my	sister	Naomi,	for	her	unerring	faith	in	my	ability, my	grandparents	for	their	trust	in	my	skills	and	perseverance,	and	my	parents	Elaine	and	Clive	Dawson	for	the	guidance,	motivation,	and	encouragement	I	continue	to	benefit	from.			 With	love	and	appreciation.			 xvii	COLLABORATIVE	PERSONNEL:		Dr.	Mark	Carey,	Principal	Investigator,	M.Sc.	Supervisor,	University	of	British	Columbia	(UBC),	Reproductive	and	Developmental	Sciences,	Department	of	Obstetrics	and	Gynecology,	UBC	Faculty	of	Medicine	Dr.	Marta	Llaurado,	Research	Associate,	University	of	British	Columbia	(UBC),	Department	of	Obstetrics	and	Gynecology,	UBC	Faculty	of	Medicine	Dr.	Gregg	Morin,	Head	of	Proteomics,	Michael	Smith	Genome	Sciences	Centre,	BC	Cancer	Agency	Dr.	Gabriel	DiMattia,	Department	of	Obstetrics	and	Gynecology,	University	of	Western	Ontario,	London,	Ontario,	Canada	Dr.	Bryan	Hennessey,	Royal	College	of	Surgeons,	Dublin,	Ireland	Clara	Salamanca,	Technician,Dr.	David	Huntsman	Lab,	OvCaRe,	University	of	British	Columbia	(UBC),	Department	of	Pathology	Sylvia	Bamford,	Technician,	University	of	British	Columbia	(UBC),	Department	of	Obstetrics	and	Gynecology,	UBC	Faculty	of	Medicine	Josh	Hoenisch,	Co-operative	education	student,	University	of	Victoria	Faculty	of	Biochemistry	and	Microbiology						 1	CHAPTER	1.	INTRODUCTION,	BACKGROUND,	AND	AIMS	1.1. INTRODUCTION		Ovarian	epithelial	carcinomas	represent	the	most	lethal	gynecologic	malignancy	with	1	in	every	71	Canadian	women	being	diagnosed	with	an	ovarian	cancer	in	her	lifetime.		It	is	a	difficult	disease	to	diagnose	in	early	stages	due	to	clinical	presentation	and	many	forms	quickly	develop	resistance	to	traditional	chemotherapy	methods	[Canadian	Cancer	Statistics,	2015].	Low-grade	serous	ovarian	carcinoma	(LGSOC)	represents	only	6-10%	of	all	ovarian	epithelial	carcinomas,	and	while	they	have	lower	malignant	potential	than	high-grade	serous	ovarian	carcinomas	(HGSOC)	[1]	they	show	a	striking	resistance	to	traditional	chemotherapy	(4%	response	rate	in	the	relapsed	or	neoadjuvant	setting)	or	anti-hormone	therapy	(9%	response	rate)	[2-6].			Additionally,	despite	LGSOC	following	a	more	indolent	disease	course	than	HGSOC,	multiple	recurrences	are	common	and	eventually	tend	to	be	fatal	[7-11].	As	such,	effective	treatment	options	for	this	disease	are	lacking	and	it	is	crucial	that	new	therapeutic	strategies	are	developed	[5].	In	2010,	the	Gynecologic	Cancer	InterGroup	Consensus	Conference	affirmed	this,	publishing	a	statement	regarding	the	urgent	need	to	develop	new	therapeutic	options	for	patients	with	advanced	or	recurrent	LGSOC	[12].	Due	to	the	rarity	of	low-grade	serous	ovarian	carcinomas,	research	into	this	area	has	been	limited.	Due	to	a	lack	of	research	models,	relatively	little	is	known	about	the	molecular	biology	of	this	disease,	and	clinical	trials	for	LGSOC	are	still	in	the	stages	of	infancy.		Therefore	we	developed	a	cohort	of	cell	lines	to	serve	as	important		 2	experimental	models	for	characterizing	important	aspects	of	the	disease,	and	allowing	us	to	investigate	the	molecular	biology	of	this	disease.	In	this	study,	we	have	used	these	histotype-specific	cell	line	models	to	evaluate	the	efficacy,	biological	and	on-target	effects	of	four	MEK	inhibitors.	These	important	experimental	models	will	be	used	to	discover	important	predictive	biomarkers	of	treatment	response	or	molecular	determinants	of	prognosis.	1.2.	BACKGROUND		Ovarian	carcinomas	are	comprised	of	mucinous,	endometrioid,	clear	cell,	and	serous	histotypes.	Mucinous	ovarian	carcinomas	and	borderline	tumors	frequently	contain	mutations	in	KRAS,	amplifications	in	ERBB2,	and	a	50%	mutation	rate	in	TP53.	They	tend	to	have	a	favourable	prognosis	compared	to	other	ovarian	cancer	subtypes.	Endometrioid	and	clear	cell	ovarian	carcinomas,	shown	to	be	associated	with	atypical	endometriosis,	have	a	high	frequency	of	ARID1A	mutations,	and	a	high	(clear	cell)	or	moderate	(endometrioid)	frequency	of	PIK3CA	mutations.	Both	clear	cell	and	endometrioid	ovarian	carcinomas	tend	to	have	a	loss	of	PTEN	expression.	Clear	cell	ovarian	carcinomas	have	high	expression	of	HNF1B[13].		Serous	ovarian	carcinomas	are	the	most	common	histotype	of	ovarian	cancer.	They	are	divided	into	low-grade	and	high-grade	subtypes,	the	former	being	a	much	more	indolent	disease.	However,	the	10-year	survival	rates	for	patients	with	both	HGSOC	and	LGSOC	is	similar	[14].	HGSOC	are	hallmarked	by	ubiquitous	TP53	mutations,	and	a	high	frequency	of	hereditary	BRCA1/2	germline	mutations	or	somatic	BRCA1/2		 3	dysfunction.	While	HGSOC	usually	have	high	levels	of	genomic	instability,	LGSOC	tend	to	have	fewer	genomic	aberrations	and	no	genetic	instability	[15].	Also	in	contrast	to	HGSOC,	LGSOC	lack	frequent	TP53	and	BRCA1/2	abnormalities.	It	has	been	previously	reported	that	up	to	35%	of	LGSOC	tumors	have	mutations	in	KRAS	or	BRAF,	and	frequently	upregulated	PI3K/AKT	and	RAS/MAPK	signaling	pathways[15,	16],	both	of	which	contribute	significantly	to	promotion	of	cancer	growth	and	invasion	[15,	17,	18].	It	has	been	shown	that	BRAF	mutations	are	rare	in	advanced-stage	LGSOC	and	predominate	in	early-stage	tumors	that	are	far	more	likely	to	be	surgically	curable	than	advanced/	recurrent	cases[19,	20].	However,	advanced	and	recurrent	cases	of	LGSOC	tend	to	have	a	greater	prevalence	of	RAS	mutations	[20,	21].	Because	mutations	in	these	pathways	are	present	in	a	large	subset	of	metastatic	and	recurrent	LGSOC,	it	is	of	significant	clinical	importance	to	develop	targeted	therapeutic	approaches	to	treat	these	patients	with	better	efficacy	than	current	techniques.	Thus,	well-characterized	experimental	models	which	accurately	reflect	advanced	or	recurrent	LGSOC	biology	will	serve	as	an	important	tool	for	exploring	novel	therapeutic	options	for	cases	of	LGSOC	which	require	alternative	treatments.	Although	aberrant	RAS/MAPK	signaling	pathways	tend	to	harbor	mutations	in	RAS,	it	has	proven	to	be	a	difficult	therapeutic	target	due	to	complex	control	over	many	other	signaling	pathways,	a	difficult-to-target	molecular	structure,	and	multiple	RAS	isoforms[22].	In	cancers	with	enhanced	RAS	pathway	activity,	the	signaling	protein	MEK	has	been	shown	to	be	a	promising	therapeutic	target	due	to	its	unique	structure	among	kinases	and	specific	downstream	target	[23].	However,	success	has	been	limited	in		 4	clinical	application	of	RAS	pathway	inhibitors,	often	due	to	the	development	of	drug	resistance	during	treatment	[24-26].	Additionally,	response	to	specific	targeted	agents	has	been	shown	to	be	context-specific.	A	recent	example	of	these	is	the	differing	success	of	BRAF	inhibitors	in	melanoma	vs.	colon	cancer.	While	application	of	BRAF	inhibitors	in	BRAF-mutant	melanomas	with	a	V600E	activating	mutation	has	been	fairly	successful,	colon	cancers	harboring	the	same	mutation	have	had	limited	responses	[27].			An	important	Phase	2	study	by	Farley	et	al,	in	2013[28]	demonstrated	a	15%	response	rate	in	LGSOC	patients	after	treatment	with	the	MEK	inhibitor	(MEKi)	selumetinib,	providing	a	new	option	for	therapy	with	a	better	response	rate	for	patients	than	traditional	chemotherapies.	However,	the	response	of	LGSOC	to	selumetinib	is	still	lower	than	chemotherapeutic	success	rates	in	other	cancers	such	as	the	more	aggressive	HGSOC,	and	many	cancer	types	[29]	have	been	shown	to	develop	resistance	to	MEKi	therapy.	Currently,	there	are	a	variety	of	MEKi	in	clinical	trials	in	LGSOC	or	RAS-mutant	cancers.	In	this	project,	we	intend	to	focus	on	four:	trametinib,	selumetinib	binimetinib,	and	refametinib[30].	Trametinib	is	an	allosteric	ATP	non-competitive	inhibitor	with	IC50	values	in	the	0.92-3.4nM	range	against	purified	MEK1/2[31].	It	binds	MEK	1	and	2	isoforms	in	a	specific	hydrophobic	pocket	adjacent	to	the	active	site	and	the	activation	loop,	similarly	to	other	MEK1/2	inhibitors.	It	has	been	shown	to	inhibit	dual	phosphorylation	of	ERK1/2	on	both	the	T202	and	Y204	residues,	preventing	its	activation[32].	Binimetinib	and	refametinib	are	also	ATP-noncompetitive	inhibitors	of	MEK1/2	with	IC50	values	against	purified	MEK	protein	in	the	10-40nM	range[33]	[17];	Interestingly,	selumetinib	primarily	targets	the	MEK1	isoform	with	a	IC50	of	14nM		 5	according	to	pharmacologic	studies	and	the	published	drug	data	sheet.[30,	34,	35].	While	each	of	the	four	MEKi	are	allosteric	ATP-noncompetitive	inhibitors,	binding	in	a	region	adjacent	to	the	active	site[36],	their	IC50	values	for	MEK1	are	greater	than	10X	higher	than	those	of	trametinib.	Trametinib	has	shown	potent	anti-tumor	efficacy	in	a	variety	of	RAS-mutant	and	RAS-wild	type	xenograft	models	and	exceptional	specificity	and	potency	against	MEK1/2[32],	making	it	a	promising	new	therapeutic	agent	already	in	many	clinical	trials	in	different	cancer	types.				The	RAS-MAPK	signaling	pathway	(Figure	1.2.1),	a	receptor	tyrosine	kinase	mediated	cascade	characterized	by	the	moieties	RAS/RAF/MEK/ERK,	is	a	critical	cellular	pathway	contributing	to	proliferation,	survival,	angiogenesis,	migration,	and	cell	cycle	regulation.	Dysregulation	has	been	linked	to	a	variety	of	malignancies	[17,	30,	37-41].	The	primary	signaling	moiety	in	the	cascade	is	the	plasma-membrane	associated	small	GTP-ase	RAS	(KRAS,	NRAS,	or	HRAS;	clinically	significant	RAS	members),	which	acts	as	a		molecular	switch	for	regulating	downstream	signal	transduction	activity	when	in	a	GTP-bound	“on”	state	or	a	GDP-bound	“off”	state.	Activating	mutations	in	RAS	genes	result	in	a	persistently	GTP-bound	molecule	and	subsequent	constitutive	signaling.		KRAS	is	one	of	the	most	commonly	mutated	oncogenes,	with	activating	mutations	seen	in	33%	of	all	cancers[42].	KRAS	and	HRAS	were	first	identified	as	potentially	oncogenic	proteins	from	rat	sarcoma	viruses	in	1979,	and	NRAS	mutations	were	identified	in	neuroblastoma	in	1983,	when	it	was	also	discovered	that	chemical	carcinogenesis	induces	RAS	mutations.	In	1989,	the	structure	of	RAS	proteins	was	further	elucidated,	identifying	associated	farnesyl	lipid	groups	which	allow	for	plasma	membrane	localization	of		 6	RAS[43].	Activated	RAS	proteins	recruit	the	serine-threonine	kinase	RAF	to	the	plasma	membrane,	allowing	it	to	become	activated	by	autophosphorylation	and	subsequently	initiating	the	signal	transduction	cascade[44].	Phosphorylation	by	RAF	activates	MEK1/2,	which	then	phosphorylates	serine/threonine	and	tyrosine	residues	on	ERK.	Activated	phospho-ERK1/2	can	then	effect	a	variety	of	cellular	processes	[30,	39,	40].	Activated	RAS	can	effect	signaling	through	many	different	cellular	pathways	affecting	apoptosis,	proliferation,	angiogenesis,	and	metabolism	among	other	traits[45];	including	the	PI3K-AKT	signaling	pathway	which	is	important	for	survival	and	proliferation[46].			Figure	1.2.1.	RAS-MAPK	signaling	pathway.	Extracellular	ligand	binding	to	receptor	tyrosine	kinases	(RTK)	induces	activation	of	RAS-GDP	to	RAS-GTP	by	through	a	series	of	intermediate	mechanisms	(guanine	nucleotide	exchange	factor).	Activated	RAS	initiates	a	downstream	signaling	cascade	through	RAF,	MEK,	and	ERK,	resulting	in	enhanced	proliferation,	survival,	angiogenesis,	migration,	and	cell	cycle	regulation[45].			 7	Due	to	the	large	proportion	of	LGSOC	tumors	demonstrating	upregulation	of	the	RAS-MAPK	signaling	pathway	through	activating	mutations	in	KRAS,	NRAS,	or	BRAF	(RAF	B	isoform),	inhibition	of	the	downstream	signaling	effectors	MEK1	and	MEK2	have	been	touted	as	a	therapeutic	strategy,	as	they	represent	a	critical	junction	between	upstream	regulators	of	the	RAS-MAPK	pathway	and	a	specific	downstream	target[17,	32,	41].	MEK1/2	is	an	attractive	therapeutic	target	due	to	high	selectivity	for	acting	on	ERK1/2	and	a	unique	hydrophobic	allosteric	inhibition	site	compared	to	other	kinases[17,	29].	While	Farley	et	al,	2013,	could	not	correlate	RAS	mutation	status	with	patient	response	to	selumetinib	therapy	in	LGSOC[28],	RAS-mutant	cell	line	and	tumor	models	have	shown	increased	sensitivity	to	MEK	inhibition	in	other	cancer	types[17,	29,	32].		 Because	of	the	rarity	of	the	disease	and	the	difficulty	of	culturing	LGSOC	cells	in	vitro,	there	is	little	preclinical	information	in	cell	line	or	xenograft	models	evaluating	MEKi	efficacy	in	LGSOC.	One	study	performed	in	2005	screened	a	single	MEKi	(CI-1040)	against	a	library	of	serous	ovarian	primary	cultures,	including	MPSC1	(BRAF	mutant).		Only	MPSC-1	was	classified	as	a	low-grade	serous	ovarian	culture,	and	little	information	about	clinical	information	or	pathology	for	the	other	primary	cultures	were	reported.	Results	from	this	study	suggested	the	potential	value	of	using	MEKi	in	KRAS	or	BRAF	mutant	tumors	such	as	LGSOC	[18].	One	other	putative	LGSOC	cell	line	was	reported	in	1985	by	Buick	et	al.	(HOC-7),	and	molecularly	profiled	by	Beaufort	et	al.,	in	2014	however	there	was	a	lack	of	clinical	or	pathologic	data[47].			 8		 There	are	several	clinical	trials	of	MEKi	currently	taking	place	involving	patients	with	advanced	or	recurrent	LGSOC	[NCT00551070,	NCT01849874,	NCT02101788].	However,	the	clinical	study	of	binimetinib	[NCT01849874;	MEK	Inhibitor	in	Low-grade	serous	Ovarian	cancer	(MILO)]	was	recently	closed	when	a	planned	interim	analysis	did	not	show	expected	improvement	on	progression	free	survival.	There	was	no	preclinical	testing	of	binimetinib	in	LGSOC	cell	lines	or	xenografts.	It	is	critical	that	well-characterized	model	systems,	linked	with	tumor	tissue	molecular	pathology,	patient	information,	and	disease	outcomes	are	developed	to	better	understand	the	molecular	biology	of	the	disease	and	develop	new	therapeutic	regimens	and	treatments	for	LGSOC	patients.		 Recent	advances	in	proteomic	technologies,	such	as	quantitative	mass	spectrometry,	have	made	it	possible	to	quantitatively	compare	the	global	protein	content	of	cell	lines	and	tissue	samples,	allowing	for	sensitive	detection	of	changes	in	individual	peptide	markers	between	samples.	In	order	to	better	identify	LGSOC	patients	who	will	benefit	from	targeted	MEK	inhibition,	we	aim	to	identify	predictive	protein	markers	of	MEKi	sensitivity	or	resistance.	By	performing	a	global	proteomic	analysis	on	LGSOC	cell	lines	of	known	drug	response	phenotypes,	we	aim	to	elucidate	a	functional	mechanism	that	accounts	for	a	de	novo	LGSOC	cell	line	drug	resistance.	Additionally,	identification	of	dynamic	proteins	in	response	to	MEKi	therapy	may	represent	mechanisms	contributing	to	acquired	drug	resistance	phenotypes.	Biomarker	and	signaling	pathway	analyses	such	as	these	are	an	important	first	step	in	developing	screening	protocols	that	can	be	applied	to	patient	cohorts,	in	order	to	target	individuals		 9	who	are	most	likely	to	respond	to	a	particular	drug	therapy.	By	characterizing	proteomic	response	in	MEKi-sensitive	and	resistant	LGSOC	cell	lines	by	reverse-phase	protein	array	(RPPA)	and	global	quantitative	mass	spectrometry	(MS)	we	can	compare	the	biological	effects	of	MEKi,	potentially	identifying	predictive	markers	de	novo	drug	sensitivity	or	resistance,	or	targets	for	additional	therapeutics	upon	acquired	drug	resistance.	Using	proteomic	technologies	confers	an	advantage	in	biomarker	discovery	over	sequencing	technologies,	as	functional	protein	candidates	responsible	for	a	particular	drug	response	phenotype	are	examined.	Understanding	the	mechanisms	of	MEK	inhibitor	resistance	in	the	context	of	LGSOC	at	a	protein	level	may	provide	knowledge	that	can	be	transferred	to	other	cancers	being	investigated	for	MEKi	response.	We	are	interested	in	studying	this	disease	because	of	the	great	need	to	improve	treatment	options	for	patients,	and	with	our	established	cell	line	models	we	are	now	in	a	good	position	to	advance	therapeutic	strategies.									 10	1.3.	GENERAL	HYPOTHESES	AND	AIMS			 We	hypothesize	that	the	development	of	a	library	of	molecularly	characterized	LGSOC	cell	lines	will	serve	as	a	preclinical	model	system	with	which	to	test	novel	therapeutics.	Through	MEK	inhibition	with	four	different	drugs	in	these	cell	lines,	we	hypothesize	that	MEKi-sensitive	and	MEKi-resistant	LGSOC	cell	phenotypes	will	be	seen,	there	will	be	varying	efficacy	among	the	four	MEKi,	and	that	subsequent	proteomic	analysis	will	yield	novel	protein	biomarkers	of	MEKi	sensitivity	or	resistance.	Based	on	the	current	state	of	knowledge	and	available	resources	in	LGSOC,	the	overall	research	aims	of	this	study	are	as	follows:	Aim	1.	 To	establish	and	molecularly	characterize	a	library	of	histotype-specific,	pathology-reviewed	cell	lines	from	advanced	and	recurrent	LGSOC	patients	for	use	as	preclinical	models	in	this	disease	(Chapter	2).	Aim	2.	To	evaluate	the	relative	efficacy	of	four	MEK	inhibitors	in	molecularly	characterized	patient-derived	LGSOC	cell	lines	using	molecular	biology	techniques,	and	to	classify	LGSOC	cell	lines	based	on	their	MEKi	sensitivity	or	resistance	(Chapter	3).	Aim	3.	To	compare	the	proteomes	of	patient-derived	LGSOC	cell	lines	of	differential	MEKi	sensitivities	using	reverse-phase	protein	array	(RPPA)	and	mass	spectrometry	(MS)	analyses	(Chapter	4).			 11	Methodology	for	each	aim	of	this	thesis	is	outlined	at	the	beginning	of	the	corresponding	chapter	as	different	techniques	were	used	to	achieve	each	aim.	A	full	description	of	methods,	materials,	and	reagents	can	be	found	in	the	Appendix.	CHAPTER	2.	DEVELOPMENT	AND	MOLECULAR	CHARACTERIZATION	OF	NOVEL	LOW-GRADE	SEROUS	OVARIAN	CARCINOMA	CELL	LINES		2.1.	AIMS		In	order	to	address	our	general	hypothesis	and	the	three	aims	of	this	study,	we	must	develop	a	robust,	well-characterized	library	of	LGSOC	cell	lines	as	models	for	future	research.	Because	of	the	lack	of	model	biological	systems	to	study	LGSOC	in	a	preclinical	setting,	development	of	sustainable	cell	culture	systems	for	testing	novel	therapeutic	strategies	is	crucial.	While	commercial	ovarian	cancer	cell	lines	do	exist,	they	mainly	represent	other	distinct	histotypes	with	very	different	genomic	and	phenotypic	profiles	to	LGSOC,	or	lack	patient	data	such	as	disease	stage	and	previous	drug	treatment	status.	To	address	this	issue,	our	objective	(Aim	1)	is	to	develop	a	library	of	patient-derived	LGSOC	cell	cultures	in	collaboration	with	the	OvCaRe	program	from	well-documented	cases	that	have	undergone	pathological	review	and	better	represent	this	disease	in	a	preclinical	setting.			 12	2.2.	METHODS				 2.2.1.	ESTABLISHMENT	OF	LGSOC	CELL	CULTURES	FROM	PATIENT	SAMPLES		When	pathologically	confirmed	cases	of	advanced	or	recurrent	LGSOC	(by	biopsy)	are	scheduled	for	surgeries,	tissues	for	cell	culture	are	immediately	harvested	in	the	operating	room	from	patients	who	have	signed	gynecologic	tumor	banking	consent	under	University	of	British	Columbia	and	BC	Cancer	Agency	ethics	certification	H14-02859.	Tumors	are	processed	immediately	to	avoid	degradation,	and	samples	of	fresh	tissue	are	snap-frozen	for	future	experimental	use.	Using	sterile	technique,	tissues	are	finely	homogenized	using	surgical	scissors	and	digested	with	approximately	10mL	of	1X	collagenase	for	one	hour	at	room	temperature.	Samples	are	then	washed	several	times	with	Hank’s	balanced	salt	solution	(HBSS)	to	remove	debris,	fat,	etc.,	then	resuspended	in	a	rich	combination	media	(199:105)	supplemented	with	10%	defined	FBS	and	plated	in	coated	tissue	culture	flasks.	Fractions	from	the	initial	tumor	collection	media	and	all	washes	of	the	samples	are	typically	collected	and	plated	to	ensure	maximum	recovery	of	tumor	epithelial	cells.	Ascites	samples	are	centrifuged	in	a	standard	tabletop	centrifuge	for	5-10min	at	1000rpm	(approximately	180	x	g),	resuspended	in	supplemented	199:105,	and	directly	plated.	If	a	large	quantity	of	blood	is	present	in	the	ascites,	erythrocytes	are	lysed	using	0.8%	ammonium	chloride	erythrocyte	lysis	buffer	prior	to	plating.	All	primary	cultures	are	kept	on	1μg/mL	gentamicin	for	one	to	four	weeks	following	processing.		 13		 Primary	cultures	from	tumor	and	ascites	are	left	at	37oC	and	5%	CO2	for	up	to	5	days	after	plating	to	ensure	ample	opportunity	for	any	suspended	epithelial	cells	to	attach	and	establish	in	the	flasks.	Any	remaining	floating	cell	or	groups	of	cells	that	look	viable	are	recovered	from	the	supernatant	of	the	initial	flasks	and	re-plated	into	new	small	flasks	to	recover	any	additional	epithelial	cells	which	may	be	present.	As	many	primary	cultures	from	tumor	samples	contain	a	significant	fibroblast	or	mesothelial	cell	component,	short	applications	of	diluted	trypsin	are	applied	to	remove	contaminating	cell	types	and	leave	the	strongly	attached	epithelial	cell	morphologies	in	the	flasks.			 Primary	cultures	from	LGSOC	samples	tend	to	be	slow-growing,	just	as	in	patient	cases.	As	such,	the	establishment	of	a	robust	primary	culture	from	these	tumor	samples	can	take	as	long	as	6	months	until	the	first	passage.	Once	a	stable	primary	culture	or	cell	line	is	established,	they	typically	require	passaging	once	every	week	to	two	or	three	weeks,	and	are	only	diluted	by	half	prior	to	re-plating.	Only	the	fastest-growing	cell	lines	require	passaging	twice	a	week.	We	have	noted	that	the	LGSOC	cells	can	exhibit	a	change	in	phenotype	due	to	stress	when	they	are	kept	at	low	confluence,	so	it	has	proven	to	be	important	not	to	dilute	them	too	much	when	passaging.	All	cell	lines	and	primary	cultures	used	in	this	chapter	have	been	passaged	between	5	to	>30	times.	Cell	lines	used	for	MEKi	testing	in	this	project	are	able	to	be	passaged	>20	times,	however	experiments	are	performed	on	cultures	between	5-20	passages.	All	cell	line	morphology	has	been	documented	by	light	microscopy.		 2.2.2.	MOLECULAR	CHARACTERIZATION	OF	NEWLY-ESTABLISHED	LGSOC	CELL	CULTURES			 14	Our	second	objective	for	this	aim	is	to	characterize	each	new	LGSOC	cell	line	in	terms	of	its	phenotype,	genotype,	and	drug	responsiveness	profile.	These	data	are	compared	to	patient	information	in	order	to	provide	support	for	further	experimentation	and	results.	Unfortunately,	due	to	the	relative	rarity	of	the	disease,	patient	cases	are	limited,	and	as	such	is	the	development	of	new	cell	lines.	Additionally,	certain	cellular	characteristics	may	provide	some	samples	with	an	advantage	for	in	vitro	growth,	so	our	sample	set	may	be	a	biased	population,	however	every	care	is	taken	to	ensure	that	minimal	selective	pressure	(ex.	through	incomplete	trypsinization)	is	imposed	on	cells	in	culture.		 Once	LGSOC	primary	cultures	are	well	established	and	have	proven	to	passage	well,	they	are	grown	to	confluence,	pelleted,	and	DNA	extracted	using	a	standard	kit	(see	appendix).	The	concentration	of	resulting	DNA	samples	is	quantified		and	they	are	purity	checked	for	an	A260/280	ratio	of	approximately	1.8-2.0	(indicating	minimal	protein	or	phenol	contamination)	using	a	NanoDrop	spectrophotometer,	then	sent	to	the	appropriate	platform	for	analysis.	Profiling	of	50	different	known	oncogenes	and	tumor	suppressor	mutations	was	performed	using	Ion	Torrent	AmpliSeqTM	Cancer	Hotspot	Panel	Version	2.	High	frequency	missense	mutations	were	confirmed	by	Sanger	sequencing.	Short	tandem	repeats	were	analyzed	by	GeneWiz	Inc.,	and	compared	to	the	DSMZ	STR	profile	website	database	to	confirm	the	unique	identity	of	each	of	the	LGSOC	cell	lines.	Copy	number	variation	was	analyzed	using	Illumina®	HumanOmni	2.5M-8	Array	or	Affymetrix	CytoScan®	HD	array	and	the	data	analyzed	using	Nexus	Copy	NumberTM	software.	Data	from	each	platform	was	compared	among	patient-paired	cell		 15	line	samples	to	determine	if	molecular	phenotypes	were	retained	as	similar	at	different	patient	disease	time	points.				2.3.	RESULTS	AND	DISCUSSION	 			 LGSOC	cell	lines	derived	for	the	purpose	of	study	have	been	well-characterized	clinically	(Table	2.3.1),	with	primary	pathology,	disease	stage,	tissue	source,	pathology	at	time	of	sample	collection,	chemotherapy	treatment	status,	and	mutation	status.	All	samples	used	to	create	cell	lines	come	from	cases	of	advanced	or	recurrent	LGSOC,	which	is	important	as	early	stage	LGSOC	is	often	cured	by	surgery[16].	No	immortalization	techniques	were	used	in	order	to	establish	the	LGSOC	cell	lines.	Because	we	aim	to	test	novel	therapeutics	and	better	understand	the	molecular	biology	of	advanced	cases	of	LGSOC,	deriving	cell	lines	from	more	benign	samples	would	not	be	as	high	value	for	investigational	drug	testing.						 16					 17	Table	2.3.1.	Primary	pathology,	pathology	at	time	of	sampling,	and	patient	information	for	each	LGSOC	cell	line	and	primary	culture	currently	established.											Cell	Line	ID	 Pa,ent	 Age	at	diagnosis	 Primary	Pathology	Disease	Stage	Sample	Number	 Source	Pathology	at	Collec,on	 Treatment	Status	Muta,on	Status	Contribu,ng	Ins,tu,on	iOvCa241	 1	 51	 SBOT	with	LGSC	 IIIA	 1	 ascites	 LGSC	 Post-chemotherapy	 KRAS	University	of	Western	Ontario	VOA-1312	 2	 58	SBOT	with	non-invasive	implants	IIB	 2	 ascites	 LGSC	 Naïve	 KRAS	 Vancouver	General	Hospital	VOA-1056	 3	62	Micropapillary	SBOT	with	invasive	implants	IIIC	3	 tumor	 LGSC	 Naïve	 NRAS,	FGFR3,	JAK3	Vancouver	General	Hospital	VOA-3993	 3	 4	 tumor	 Recurrent	LGSC	 Post-chemotherapy	 	Vancouver	General	Hospital	VOA-3448	 4	42	 Micropapillary	SBOT	 IC	5	 ascites	 Recurrent	LGSC	 Post-chemotherapy	KDR,	MET,	PIK3CA	Vancouver	General	Hospital	VOA-3723	 4	 6	 ascites	 Recurrent	LGSC	 Post-chemotherapy	 	Vancouver	General	Hospital	VOA-4627	 5	42	 LGSC	 IIIC	7	 ascites	 Recurrent	LGSC	 Post-chemotherapy	KIT,	PIK3CA,	TP53	Vancouver	General	Hospital	VOA-4698	 5	 8	 ascites	 Recurrent	LGSC	 Post-chemotherapy	 	Vancouver	General	Hospital	VOA-4881	 5	 9	 ascites	 Recurrent	LGSC	 Post-chemotherapy	 	Vancouver	General	Hospital	VOA-6406	 6	 	 Invasive	LGSC	 	 10	 tumor	 Recurrent	LGSC	 Post-chemotherapy		 Unknown	Vancouver	General	Hospital	VOA-6800	 7	 	 Invasive	LGSC	 	 11	 tumor	 LGSC	 Post-chemotherapy	 Unknown	Vancouver	General	Hospital		 18			 A	widely	varying	phenotype	is	seen	between	the	LGSOC	cell	lines,	morphologically	(Figure	2.3.1).	Many	of	the	cell	lines	are	uniquely	identifiable	by	light	microscopy,	allowing	for	reassurance	that	cultures	have	not	been	cross-contaminated	and	ease	in	identifying	unusual	morphologic	changes	in	each	cell	line.	Additionally,	each	of	the	cell	lines	that	have	been	submitted	for	short	tandem	repeat	(STR)	profiling	analysis	(Table	2.3.2)	are	uniquely	identifiable,	with	the	exception	of	the	patient-paired	cell	lines	VOA-3448	and	VOA-3723,	which	retain	the	same	STR	profiles.	This	is	likely	due	to	the	proximity	of	sample	collections	from	the	patient.		A	total	of	11	stable	cell	lines	from	7	patients	have	been	derived	to	date	(Table	2.3.1),	with	5	more	putative	cell	lines	in	ongoing	culture,	representing	the	start	of	a	library	for	use	in	preclinical	therapeutic	testing	and	better	understanding	of	the	disease	(Objective	1).		 19	Figure	2.3.1	Morphology	of	LGSOC	cell	lines	developed.	Images	were	taken	in	a	Corning	96-well	plate	on	an	Incucyte Live	Imaging	System	at	10X	magnification	(Essen	BioScience).	VOA-1056	VOA-1312	VOA-3448	VOA-3993	VOA-4627	VOA-4698	RAS wild-type	 	RAS mutant	 	VOA-3723	iOvCa241	VOA-6800	 VOA-6406	Unknown	RAS	status		 20		Table	2.3.2.	Short	tandem	repeat	(STR)	profiling	data	for	8	LGSOC	cell	lines.			 Marker	and	alleles	present	Sample	 TH01	 D21S11	 D5S818	 D13S317	 D7S820	 D16S539	 CSF1PO	 AMEL	 vWA	 TPOX	iOvCa241	 6,	9	 32.2	 12	 9,	11	 8,	10	 12,	14	 12	 X	 16,	17	 8,	11	VOA-1312	 7,	9	 29	 12	 9,	12	 8,	11	 9,	12	 12	 X	 17,	19	 10,	11	VOA-1056	 6,	9	 30,	32.2	 11,	13	 8,	11	 8,	11	 11,	12	 10,	12	 X	 18,	19	 11,	12	VOA-3993	 6,	9	 30,	32.2	 11,	13	 8,	11	 8,	11	 11,	12	 10,	12	 X	 18	 11,	12	VOA-3448	 9,	9.3	 28,	29	 10,	11	 9,	10	 12	 10,	13	 9,	14	 X	 16,	17	 8,	11	VOA-3723	 9,	9.3	 28,	29	 10,	11	 9,	10	 12	 10,	13	 9,	14	 X	 16,	17	 8,	11	VOA-4627	 6	 30	 13	 11,	12	 11	 8,	10	 11	 X	 15,	17	 9,	12	VOA-4698	 6	 30	 13	 11,	12	 11	 8,	10	 11,	13	 X	 15,	17	 12											 21	HotSpot	mutational	data	from	9	cell	lines	shows	that	the	most	common	mutations	in	these	LGSOC	cell	lines	are	in	the	RAS	pathway	(KRAS/NRAS).	iOvCa241	was	identified	as	a	KRAS	G12D	mutant,	and	VOA-1312	as	a	KRAS	G12V	mutant.	Several	other	missense	mutations	in	TP53,	KDR,	FGFR3,	JAK3,	MET,	PIK3CA,	and	KIT	were	identified	in	some	of	the	cell	lines.	However,	the	only	mutations	identified	as	pathologic	variants	are	KRAS,	NRAS,	JAK3,	and	TP53	(Table	2.3.3).	Interestingly,	the	TP53	mutant	cell	lines	arising	from	the	same	patient	at	different	disease	time	points	(VOA-4627,	VOA-4698,	and	VOA-4881)	have	been	histologically	verified	as	LGSOC,	despite	TP53	mutations	typically	characterizing	HGSOC	[13].	This	patient’s	disease	potentially	arose	from	a	low-grade	serous	tubal	intra-epithelial	carcinoma	(STIC)	in	the	patient	with	a	BRCA1	mutation	[48].	Hotspot	analysis	in	newly	developed	LGSOC	cell	lines	(VOA-6800	and	VOA-6406)	has	yet	to	be	performed.	 22	Table	2.3.3.	Missense	mutation	analysis	(HotSpot)	results	for	LGSOC	cell	lines.	Analysis	for	VOA-6406	and	VOA-6800	has	yet	to	be	performed.		 	 PATIENT 1 PATIENT 2 PATIENT 3 PATIENT 4 PATIENT 5 GENE MUTATION ID iOvCa241 VOA-1312 VOA-1056 VOA-3993 VOA-3448 VOA-3723 VOA-4627 VOA-4698 VOA-4881 FGFR3 COSM1539830     c.1156T>C (Ht) c.1156T>C (Ht)           JAK3* COSM34213     c.2164G>A (Ht) c.2164G>A (Ht)           KDR COSM149673         c.1416A>T (Ht) c.1416A>T (Ht)       KIT COSM28026             c.1621A>C (Ht) c.1621A>C (Ht) c.1621A>C (Ht) KRAS* COSM520,COSM521 c.35G>A (Ht) c.35G>T (Ht)               MET COSM5020653         c.1124A>G (Ht) c.1124A>G (Ht)       NRAS* COSM584     c.182A>G (Ht) c.182A>G (Ht)           PIK3CA COSM328028         c.1173A>G (Ht) c.1173A>G (Ht) c.1173A>G (Ht) c.1173A>G (Ht) c.1173A>G (Ht) TP53* COSM99729              c.701G>A (Ht) c.701G>A (Hm) c.701G>A (Ht) Total mutations - 1/50 genes 1/50 genes 3/50 genes 3/50 genes 3/50 genes 3/50 genes 3/50 genes 3/50 genes 3/50 genes 	 * Missense mutations reported as oncogenic.	 	 	 	 	 	 	 			 23	Cell	lines	generated	from	the	same	patient	at	different	times	and	stages	in	their	disease	retain	similar	phenotypes,	mutation	profiles	and	copy	number	variation	data,	but	also	show	increasing	genetic	instability	in	the	later	samples	typical	of	a	cancer	progression	model	according	to	the	copy	number	data.	Among	the	7	cell	lines	tested	for	copy	number	variation	and	chromosomal	instability,	there	is	a	common	loss	in	9p21.3,	as	well	as	common	amplifications	in	chromosomes	8,	12,	and	20	(Figure	2.3.2).	Among	samples,	chromosomes	4	and	13	remained	relatively	conserved	and	stable.	VOA-4627	showed	the	highest	degree	of	deletions	and	chromosomal	disorganization	(Table	2.3.4).	Additional	work	will	be	performed	by	members	of	our	research	group	to	determine	if	the	genomic	data	from	each	of	the	cell	lines	reflects	that	seen	in	the	original	patient	tumor	or	ascites	samples.													 24	Table	2.3.4.	Copy	number	variation	data	(%	genome	change,	total	copy	number	[CN]	aberrations,	%	loss	of	heterozygosity	[LOH],	type	of	platform,	CN	gains	and	losses)	in	select	LGSOC	cell	lines.			Cell	line	 %	Genome	Change	CN	aberrations	%	LOH	 Array,	Platform	 CN	gains	 CN	losses	iOvCa250	 8.3	 765	 9.3	 CytoScan	HD,	Affymetrix	8,	20	 17q	iOvCa241	 36	 1219	 21.5	 CytoScan	HD,	Affymetrix	3,	5,	7,	8,	12,	14,	20	1p,	6,	15p,	18,	22	VOA-1312	 27.8	 289	 14.8	 HumanOmni	2.5,	Illumina	3q,	5,	8,	12,	14,20	6q,	9	VOA-1056	 1.8	 283	 9.7	 HumanOmni	2.5,	Illumina		 9p	VOA-3993	 15	 302	 9.8	 HumanOmni	2.5,	Illumina	2,	12,	20	 9p	VOA-3723	 40.3	 2038	 23.4	 HumanOmni	2.5,	Illumina	1q,	2,	4,	5,	7,	8q,	10,	11,	12q,	13,	18	1p,	3p,	6,	8p,	9p,	12p,	14,	22	VOA-4627	 40.3	 2366	 9.9	 HumanOmni	2.5,	Illumina	3q,	7p,	8q,	9q,	11q,	14q,	15,	16,	17q,	19,	20	1p,	2q,	5q,	6q,	8p,	9p,	10p,	11p,	17p,	18,	21,	22		 25			Figure	2.3.2.	Sum	of	chromosomal	aberrations	from	copy	number	variation	analysis	among	7	LGSOC	cell	lines.	Analysis	completed	using	Nexus	software.Chromosome # All CNV 100% 50% 0% 50% 100% 1                                 2                             3                        4                         5                       6                     7                   8                 9                 10               11              12              13           14           15         16        17       18      19    20    21   22           X            Y VOA-3723 VOA-3993 VOA-4627 VOA-1056 VOA-1312 iOvCa241 iOvCa250 CNV  Frequency   Figure 1. Chromosome # All CNV 100% 50% 0% 50% 100% 1                                 2                             3                        4                         5                       6                     7                   8                 9                 10               11              12              13           14           15         16        17       18      19    20    21   22           X            Y VOA-3723 VOA-3993 VOA-4627 VOA-1056 VOA-1312 iOvCa241 iOvCa250 CNV  Frequency   Figure 1. Chromosome # All CNV 100% 50% 0% 50% 100% 1                                 2                             3                        4               5          6          7    8            9                 10               11              12              13           14           15         16        17       18      19    20    21   22           X            Y VOA-3723 VOA-3993 VOA-4627 VOA-1056 VOA-1312 iOvCa241 iOvCa250 CNV  Frequency   Figure 1. 	 26	While	cell	lines	fall	short	in	terms	of	biological	relevance	due	to	the	lack	of	a	tumor	tissue	or	host	microenvironment	that	the	testing	of	patient-derived	xenografts	in	animal	models	may	offer,	they	have	great	utility	in	in	vitro	testing,	particularly	when	screening	drug	effects.	They	are	more	cost	effective	than	mouse	models,	and	easier	to	work	with	on	a	daily	basis.	There	is	ongoing	work	by	our	group	to	establish	cell	line	and	patient-derived	xenografts	from	LGSOC	cases	in	order	to	continue	expanding	our	research.	Because	LGSOC	is	a	unique	disease,	the	use	of	commercial	cell	lines	of	other	or	undefined	histotypes	will	not	be	as	unique	of	a	resource	for	disease-specific	therapeutic	testing.	While	HGSOC	tumors	and	cell	lines	are	typically	characterized	by	TP53	mutations	and	large	amounts	of	genetic	instability,	LGSOC	tend	to	have	few,	well-defined	mutations,	the	most	common	of	which	being	in	RAS-related	genes.	Our	library	of	11	LGSOC	cell	lines	contains	4	RAS-mutants	(with	2	cell	lines	of	unknown	RAS	status),	reflecting	the	reported	frequency	(approximately	35%)	of	RAS	mutations	in	LGSOC[10].	When	analyzing	the	cell	lines	by	patient-pairs,	2	out	of	7	patients	(28%)	harbor	RAS	mutations.									 27		CHAPTER	3.	EVALUATION	OF	BIOLOGICAL	AND	ON-TARGET	EFFECTS	OF	4	COMMERCIALLY	AVAILABLE	MEK	INHIBITORS	IN	PATIENT-DERIVED	LGSOC	CELL	LINES		 	3.1.	HYPOTHESIS	AND	AIMS			 			 We	hypothesize	that	varying	efficacy	among	the	MEK	inhibitors	will	be	seen.	We	also	hypothesize	that	differential	sensitivity	to	MEK	inhibition	will	be	seen	among	the	LGSOC	cell	lines	tested	due	to	their	different	molecular	characteristics.	This	chapter	summarizes	our	work	on	Aim	2	of	this	study,	examining	the	relative	efficacy	of	trametinib,	selumetinib,	binimetinib,	and	refametinib,	and	to	examine	drug-response	phenotypes	among	the	LGSOC	cell	lines	tested.	3.2.	METHODS				 3.2.1.	BIOLOGICAL	AND	ON-TARGET	DRUG	EFFECTS	OF	4	MEKI	 		 		 Drug	treatment	experiments	were	conducted	in	order	to	establish	drug	responsiveness	profiles	to	evaluate	the	efficacy	of	each	MEKi.	First,	the	IC50	dose	of	each	of	the	4	MEKi	in	11	LGSC	cell	lines	was	determined.	Cells	were	plated	such	that	the	control	conditions	would	reach	100%	confluence	after	3	days	treatment,	then	a	range	of	MEKi	doses	from	0-100μM	were	tested	for	each	drug	in	biological	triplicate.	Crystal	violet	staining	was	used	to	determine	the	percentage	of	remaining	cells	in	each	well		 28	relative	to	control	values.	The	relative	efficacy	of	each	of	the	drugs	as	well	as	the	dose	required	at	which	50%	of	the	cells	remained	(IC50)	after	3	days	drug	treatment	in	each	LGSOC	cell	line	was	established.	These	experiments	were	performed	in	biological	duplicate	or	triplicate,	with	technical	quadruplicates	for	each	condition,	and	resulting	data	provided	groundwork	for	selecting	doses	for	further	drug	treatment	experiments.		 Subsequently	to	the	IC50	experiments,	10	LGSOC	lines	were	subjected	to	experiments	on	the	Incucyte®	system,	allowing	for	quantitation	of	their	relative	increases	in	confluence	with	and	without	MEKi	treatment	by	phase-contrast	microscopy.		These	experiments	are	referred	to	hereafter	as	proliferation	experiments,	although	only	confluence,	not	cell	cycle	rates	or	increase	in	cellular	size,	was	measured.	LGSOC	cells	were	plated	so	that	they	would	reach	approximately	20%	confluence	at	time	of	treatment.	After	24	hours,	cells	were	treated	with	MEKi	at	0.1μM	(trametinib)	or	1μM	(selumetinib,	binimetinib,	and	refametinib)	and	DMSO	control.	Cells	were	cultured	up	to	one	week	following	a	single	application	of	drug	and	their	confluence	tracked	over	time.	5	technical	replicates	and	2	biological	replicates	of	each	condition	were	performed.	Following	cellular	growth	in	the	control	conditions	to	100%	confluence,	an	MTS-based	viability	stain	(Promega	Cell	Titer	96R	Aqueous	Non-Radioactive	Cell	Proliferation	Assay)	was	applied	to	4	LGSC	cell	line	experiments	to	determine	if	drug	effects	on	proliferation	were	cytostatic,	inhibiting	cellular	growth,	or	cytotoxic,	killing	the	cells.	Viability	of	drug-treated	cells	was	compared	to	cells	treated	with	DMSO	control.		In	order	to	determine	if	a	dose-response	effect	was	seen	with	the	MEKi,	proliferation	experiments	were	repeated	with	three	cell	lines	of	varying	drug	sensitivity		 29	(iOvCa241,	VOA-1056,	VOA-3993)	and	0.1/1μM		and	0.5/5μM		doses	of	each	of	the	MEKi.	Results	from	these	experiments	provided	a	framework	for	structuring	experiments	to	examine	the	on-target	effects	of	the	separate	MEKi	on	the	RAS-MAPK	pathway.		 Prior	to	analyzing	on-target	drug	effects	by	Western	blot,	it	was	important	to	know	the	levels	of	phospho	and	non-phospho	moieties	of	both	MEK	and	ERK	in	the	individual	cell	lines,	so	as	to	understand	basal	RAS-MAPK	signaling	profiles	in	each.		Twenty-four	hour	1μL/mL	DMSO-treated	cell	lysates	obtained	from	confluent	60mm	plates		of	11	LGSC	cell	lines	were	used	to	establish	these	basal	expression	profiles	by	Western	blotting.	In	subsequent	experiments,	10	LGSOC	cell	lines	were	treated	with	1μL/mL	DMSO,	0.1μM	trametinib,	and	1μM	of	refametinib,	selumetinib,	and	binimetinib.	After	24	hours,	plates	were	treated	with	2μL/mL	of	1mg/mL	epidermal	growth	factor	(EGF)	(or	left	untreated)	and	scraped	using	a	house-made	non-ionic	lysis	buffer	to	yield	native	protein	lysates.	These	lysates	were	run	by	Western	blot	to	determine	on-target	inhibition	of	ERK	phosphorylation	(relative	levels	of	phospho-ERK1/2)	by	each	of	the	MEKi	in	most	of	the	LGSOC	cell	lines.	Western	blots	for	p-ERK1/2	were	exposed	for	longer	than	1	hour	by	autoradiograph	in	order	to	collect	any	present	signal.		Native,	non-denatured,	protein	lysates	treated	with	1μL/mL	DMSO	or	0.1/1	μM	MEKi	for	iOvCa241,	VOA-1056,	VOA-3723,	and	VOA-4627	with	and	without	EGF	stimulation	were	run	by	qualitative	capillary	isoelectric	point	focusing	to	examine	ERK	isoform-specific	phosphorylation	events	occurring	upon	MEKi	treatment	using	the	NanoPro1000™	System.			 30		 To	determine	if	differing	levels	of	p-ERK	seen	after	24	hour	MEKi	treatment	(Figure	3.3.6)	was	due	to	dose	or	time-dependence,	5	LGSOC	cell	lines	of	varying	MEKi	sensitivity	(determined	by	proliferation	curves)	were	treated	with	both	1X	doses	(0.1μM/1μM)	and	5X	higher	doses	of	the	MEKi.	Additionally,	lysates	were	harvested	at	both	24	and	72	hours.	Levels	of	apoptosis	induced	by	MEKi	treatment	in	the	LGSOC	lines	were	evaluated	by	both	levels	of	cleaved	PARP	(c-PARP)	by	Western	blotting,	and	caspase	3/7	levels	in	iOvCa241,	VOA-1056,	and	VOA-4627	were	evaluated	in	triplicate	using	a	Promega	Caspase-Glo®	assay	in	order	analyze	how	inhibition	of	ERK	phosphorylation	correlated	to	induction	of	apoptosis,	and	determining	if	high	levels	of	drug	inhibited	proliferation	of	cell	lines	through	effects	other	than	RAS-MAPK	pathway	inhibition	by	suppression	of	p-ERK.	Luminescent	signal	representing	increased	apoptosis	levels	was	measured	using	a	plate-reading	luminometer	(Tecan	Infinite	M200Pro).		 Lysates	from	9	LGSOC	cell	lines	treated	with	1μL/mL	DMSO,	0.1μM	trametinib,	and	1μM	binimetinib,	selumetinib,	and	refametinib	with	and	without	EGF	stimulation,	and	3	LGSOC	cell	lines	treated	with	5μL/mL	DMSO,	0.5μM	trametinib,	and	5μM	binimetinib,	selumetinib,	and	refametinib	with	and	without	EGF	stimulation	were	harvested,	diluted,	and	denatured	for	a	reverse-phase	protein	array	(according	to	protocols	used	by	Dr.	Bryan	Hennessey’s	lab,	Royal	College	of	Surgeons	in	Ireland)	to	screen	a	variety	of	RAS-MAPK	signaling	proteins.	Lysates	were	printed	onto	slides,	then	probed	with	a	panel	of	antibodies	representing	a	variety	of	signaling	proteins	(Appendix	Table	A2).	Data	from	total	and	phosphorylated	ERK1/2,	total	MEK1,	and	phosphorylated	MEK1/2	was	used	in	this	chapter	to	confirm	results	seen	by	Western	blotting.	Heatmaps		 31	were	generated	from	normalized	RPPA	protein	expression	data	using	conditional	formatting	of	cells	according	to	their	numerical	value	in	Microsoft	Excel.	3.3.	RESULTS	AND	DISCUSSION		 	 3.3.1.	BIOLOGICAL	EFFECTS		Among	the	4	MEKi	tested	by	IC50	analysis,	trametinib	was	shown	to	be	the	most	effective,	with	IC50	values	in	the	nM	range	versus	the	μM	range	for	all	other	MEKi	(Figure	3.3.1.A).	Additionally,	three	paired	cell	lines	were	identified	to	be	very	resistant	to	MEKi	treatment,	VOA-4627,	VOA-4698,	and	VOA-4881.	The	remaining	cell	lines	showed	varying	levels	of	response	to	MEKi	treatment	with	selumetinib,	binimetinib,	and	refametinib	(Figure	3.3.1.B).	While	IC50	values	for	the	other	8	cell	lines	seem	to	be	fairly	similar,	results	from	the	Incucyte®	proliferation	curves	(Figure	3.3.2.),	provide	a	more	insightful	approach	to	studying	MEKi	response	and	separating	proliferative	phenotypes	of	the	cell	lines	by	tracking	real-time	growth	under	drug	treatment	conditions.	Because	the	IC50	value	is	simply	a	measure	of	how	much	drug	is	needed	to	have	a	remaining	confluence	of	50%	after	3	days	treatment,	it	does	not	account	for	viability	after	that	time.	In	correlation	with	the	IC50	results,	Incucyte®	curves	showed	that	trametinib	had	the	greatest	anti-proliferative	effects	in	all	LGSOC	cell	lines.	Two	cell	lines	were	identified	to	be	exquisitely	sensitive	to	MEK	inhibition	by	the	Incucyte®	experiments:	iOvCa241	(KRAS	G12D)	and	VOA-1312	(KRAS	G12V)	showed	almost	no	proliferation	following	a	single	dose	of	any	of	the	MEKi.	The	other	cell	lines,	all	either	RAS	wild-type	or	NRAS	mutants,	showed	varying	degrees	of	resistance	to	all	the	MEKi,	exhibiting		 32	proliferation	after	MEKi	application.	Incucyte®	proliferation	curve	results	were	validated	by	the	application	of	a	tetrazolium	viability	stain	for	several	cell	lines,	and	the	percentage	of	viable	cells	remaining	at	the	end	of	the	proliferation	experiments	correlated	closely	with	percentage	of	confluence	in	each	treatment	condition	in	the	proliferation	curves	(Figure	3.3.3).				 33		0.001.002.003.004.005.006.007.008.009.0010.00Trametinib Refametinib Selumetinib BinimetinibIC50	value	(uM)MEK	InhibitorAiOvCa241VOA-1312 VOA-1056 VOA-3993 VOA-3448 VOA-3723 VOA-4627 VOA-4698 VOA-4881 VOA-6406 VOA-6800 	 34		Figure	3.3.1	IC50	values	for	4	commercially	available	MEK	inhibitors	in	11	newly-established	LGSOC	cell	lines.	A.	IC50	values	organized	by	drug.	B.	IC50	values	organized	by	cell	line.		Cell	lines	were	treated	for	72	hours	with	a	spectrum	of	drug	dosages	from	0-100μM,	stained	with	0.25%	crystal	violet,	and	absorbance	read	at	595nm	on	a	microplate	reader.	Cells	were	seeded	so	that	control	wells	would	be	100%	confluent	at	time	of	staining.	012345678024681012141618202224262830IC50 value (µM)Cell Lines BTraRefSelBin	 35		0"10"20"30"40"50"60"70"80"90"100"0" 20" 40" 60" 80" 100" 120" 140"Prolifera)on+(%)+Time+(hours)+VOA1312+Control"Tra"100nM"Azd6244"1uM"Mek162"1uM"Refa"1uM"0"10"20"30"40"50"60"70"80"90"100"0" 20" 40" 60" 80" 100" 120"Prolifera)on+(%)+Time+(hours)+10698D+Control"Tra"100nM"Azd6244"1uM"Mek162"1uM"Refa"1uM"0"10"20"30"40"50"60"70"80"90"100"0" 20" 40" 60" 80" 100"Prolifera)on+(%)+Time+(hours)+VOA1056+Control"Trame6nib"100nM"Selume6nib"1uM"Mek162"1uM"Refame6nib"1uM"0"10"20"30"40"50"60"70"80"90"100"0" 20" 40" 60" 80" 100" 120" 140"Prolifera)on+(%)+Time+(hours)+VOA3993+Control"Tra"100nM"Azd6244"1uM"Mek162"1uM"Refa"1uM"0"10"20"30"40"50"60"70"80"90"100"0" 50" 100" 150" 200"Prolifera)on+(%)+Time+(hours)+VOA73448+Control"Trame6nib"100nM"Selume6nib"1µM"Binime6nib"1µM"Refame6nib"1µM"0"10"20"30"40"50"60"70"80"90"100"0" 10" 20" 30" 40" 50" 60" 70" 80" 90" 100"Prolifera)on+(%)+Time+(hours)+VOA3723+Control"Tra"100nM"Azd6244"1uM"Mek162"1uM"Refa"1uM"0"10"20"30"40"50"60"70"80"90"100"0" 20" 40" 60" 80" 100" 120" 140"Prolifera)on+(%)+Time+(hours)+VOA4698+Control"Tra"100nM"Azd6244"1uM"Mek162"1uM"Refa"1uM"0"10"20"30"40"50"60"70"80"90"100"0" 20" 40" 60" 80" 100" 120"Prolifera)on+(%)+Time+(hours)+VOA4627+Control"Tra"100nM"Azd6244"1uM"Mek162"1uM"Refa"1uM"iOvCa241	Control	Tra	0.1uM	Sel	1uM	Bin	1uM	Ref	1uM	Control	Tra	0.1uM	Sel	1uM	Bin	1uM	Ref	1uM	Control	Tra	0.1uM	Sel	1uM	Bin	1uM	Ref	1uM	Control	Tra	0.1uM	Sel	1uM	Bin	1uM	Ref	1uM	Control	Tra	0.1uM	Sel	1uM	Bin	1uM	Ref	1uM															 															 36	Figure	3.3.2.	(previous	page)	Incucyte®	proliferation	curves	for	LGSOC	cell	lines	treated	with	4	different	MEKi	at	a	1X	dose	determined	by	literature	review	and	IC50	data.	Cells	were	treated	at	10-20%	confluence,	and	MTS	viability	staining	applied	when	control	wells	reached	~100%	confluence.				Figure	3.3.3.	MTS	viability	staining	following	5-7	day	Incucyte®	proliferation	assays.	Reagent	was	applied	for	3.5	hours	prior	to	reading	absorbance	at	490nm.				 37		 Dose-response	effects	of	the	MEKi	on	cellular	proliferation	were	evaluated	by	adding	a	high	dose	(0.5μM	or	5μM)	treatment	in	three	selected	cell	lines:	iOvCa241,	VOA-1056,	and	VOA-4627.	As	expected	based	on	previous	results,	iOvCa241	proliferation	rates	were	suppressed	by	both	high	and	low	doses	of	MEKi.	The	other	two	cell	lines	showed	a	dose-dependent	response	to	the	5X	MEKi,	with	a	clear	decrease	in	proliferation	compared	to	the	low	MEKi	doses	(Figure	3.3.4).	While	VOA-1056	remains	fairly	sensitive	to	the	high	doses	of	MEKi,	VOA-4627	continues	to	exhibit	proliferation	with	a	high	dose	of	the	MEKi.				 38		Figure	3.3.4.	Incucyte®	proliferation	curves	for	two	selected	resistant	(VOA-1056,	VOA-4627),	and	one	sensitive	(iOvCa241)	cell	lines	treated	with	low	(0.1	or	1μM)	and	high	(0.5	or	5μM)	doses	of	MEKi.							 39	After	establishing	drug	responsiveness	profiles	by	IC50	and	proliferation	experiments,	the	on-target	effect	of	each	of	the	4	MEKi	in	the	LGSOC	lines	was	evaluated	by	Western	blot,	ELISA,	and	cIEF.	Western	blotting	showed	that	the	basal	phosphorylated	levels	of	the	downstream	target	of	MEK	(p-ERK)	differed	between	cell	lines	irrespective	of	RAS	mutation	status	(Figure	3.3.5).	In	particular,	basal	expression	of	p-ERK	was	lower	in	the	cell	lines	iOvCa241,	VOA-1056,	VOA-3448,	and	VOA-4627,	VOA-4881,	VOA-6800,	and	VOA-6406,	and	higher	in	the	cell	lines	VOA-1312,	VOA-3993,	VOA-3723,	and	VOA	4698.	Interestingly,	levels	of	p-ERK	expression	are	higher	in	the	second	samples	among	the	paired	cell	lines,	with	the	exception	of	VOA-4881.	It	has	been	previously	noted	that	basal	p-ERK	status	does	not	correlate	with	mutation	status	or	RAS-MAPK	pathway	dependence[49].	Also	of	note,	basal	levels	of	total	ERK	are	constant	among	all	cell	lines,	with	the	exception	of	iOvCa241	which	lacks	the	ERK2	isoform	and	VOA-1056	which	predominantly	expresses	ERK2.	These	isoform-specific	expressions	are	also	seen	by	capillary	isoelectric	point	focusing	(Figure	3.3.7).	Levels	of	phospho-MEK1/2	vary	between	samples,	with	higher	expression	in	the	second	paired	sample	(except	VOA-4881)	as	seen	with	phospho-ERK1/2	results.	Levels	of	total	MEK1/2	are	relatively	constant	among	cell	lines,	with	slightly	less	expression	by	VOA-1312	and	VOA-1056.	No	clear	correlation	between	basal	pathway	activation	(by	phosphorylated	ERK1/2	and	MEK1/2)	and	RAS-mutation	status	is	seen.					 40		 		Figure	3.3.5.	Basal	levels	of	phospho	and	total	ERK1/2	and	MEK1/2	for	LGSOC	cell	lines.	All	cell	lines	were	treated	with	DMSO	with	the	exception	of	VOA-6800	and	VOA-6406.								 41	In	order	to	determine	which	drugs	exhibited	an	on-target	effect	by	blocking	the	phosphorylation	of	ERK1/2,	Western	blots	were	run	for	each	cell	line	with	and	without	EGF	stimulation	to	force	RAS/MAPK	pathway	signaling	[50](Figure	3.3.6).	Trametinib	appeared	to	exhibit	the	best	on-target	effects,	with	total	blockage	of	ERK	phosphorylation	in	all	cell	lines	even	after	EGF	stimulation.	The	next	most	effective	on-target	effects	were	exhibited	by	refametinib.	The	drugs	selumetinib	and	binimetinib	exhibited	better	on-target	inhibition	in	the	sensitive	cell	lines	iOvCa241	and	VOA-1312	than	the	other	cell	lines,	but	in	all	cases	there	was	residual	p-ERK	present	for	these	two	drugs,	especially	following	EGF	stimulation.	This	indicates	that	there	is	a	deficiency	by	selumetinib,	binimetinib,	and	refametinib	to	completely	block	signaling	through	ERK1/2.	Additionally,	while	there	is	some	correlation	between	sensitivity	to	drug	treatment	by	Incucyte®	proliferation	experiments	and	residual	p-ERK1/2	expression	after	24	hour	drug	treatment,	sensitive	cell	lines	still	show	ERK	phosphorylation	after	treatment	with	selumetinib	and	binimetinib,	indicating	an	alternate	cause	of	drug	sensitivity,	such	as	off-target	effects.	In	the	cases	of	the	resistant	cell	lines	which	exhibit	some	proliferation	even	after	treatment	with	trametinib	(which	blocks	ERK	phosphorylation	completely),	an	alternate	compensatory	mechanism	such	as	activation	and	signaling	through	other	pathways	potentially	exists	to	bypass	RAS-MAPK	pathway	inhibition.			 42		Figure	3.3.6.	Western	blot	evaluation	of	on-target	inhibition	of	ERK	phosphorylation	by	MEK	inhibitors	for	8	LGSOC	cell	lines.	Cell	lines	were	treated	with	drugs	for	24	hours	prior	to	EGF	stimulation	and	lysate	preparation.	Note:	All	phospho-ERK1/2	blots	were	overexposed	to	show	as	much	signal	as	possible	after	MEKi	treatment.	For	each	cell	line,	all	antibodies	come	from	the	same	blot	and	are	exposed	at	the	same	time,	however	images	were	cut	and	re-ordered	to	ensure	the	same	sample	order	for	presentation	purposes.				 				 43	Capillary	isoelectric	point	focusing	(cIEF)	is	a	qualitative	technique	complementary	to	Western	blotting	that	allows	for	the	visualization	of	isoform-specific	phosphorylation	events.	From	the	LGSOC	cell	lines	tested	by	Western	blot,	4	cell	lines	from	4	different	patients	were	chosen	to	test	by	cIEF:	iOvCa241,	VOA-1056,	VOA-3723,	and	VOA-4627	(Figure	3.3.7).		The	results	from	cIEF	correlated	closely	with	the	Western	blots	from	Figure	3.3.6.	Trametinib	at	0.1μM	resulted	in	little	to	no	detectable	levels	of	p-ERK	in	all	cell	lines	tested,	while	refametinib	was	the	next	best	drug.	The	sensitive	cell	line	iOvCa241	showed	a	complete	inhibition	of	ERK	phosphorylation	by	all	four	MEKi,	while	the	three	resistant	cell	lines	showed	varying	degrees	of	inhibition	of	ERK	phosphorylation,	as	seen	in	the	Western	blots.	Isoform-specific	ERK	expression	as	seen	by	Western	blot	in	Figure	3.3.5.,	was	confirmed	by	cIEF,	showing	only	expression	of	ERK1	isoform	in	iOvCa241,	and	predominantly	ERK2	expression	in	VOA-1056.		 44			A		 45		Figure	3.3.7.	[A]	cIEF(NanoPro)	results	for	iOvCa241	and	VOA-1056	[B]	cIEF	results	for	VOA-3723	and	VOA-4627.	All	four	cell	lines	were	treated	with	trametinib,	selumetinib,	binimetinib,	and	refametinib.	Stimulation	with	EGF	leads	to	forced	signaling	and	more	pronounced	peaks.	An	antibody	against	ERK1/2	was	used	to	identify	ERK	isoforms	and	phosphorylation	in	each	sample	condition.	Shift	of	peaks	towards	a	more	acidic	isoelectric	point	(pI)	represent	increasing	levels	of	phosphorylation	of	ERK.							B		 46	To	determine	the	inhibitory	effects	on	ERK	phosphorylation	of	the	MEKi	over	time,	iOvCa241,	VOA-1312,	VOA-1056,	VOA-3723,	and	VOA-4627	were	chosen	to	analyze	time	and	dose-dependent	effects.	Cell	lines	were	treated	with	low	(0.1μM	trametinib	and	1μM	others)	and	high	(0.5μM	trametinib	and	5μM	others)	of	the	4	MEKi	and	the	phosphorylation	of	ERK	and	analyzed	by	Western	blot	at	24	and	72	hours	(Figure	3.3.8).	In	the	sensitive	iOvCa241,	all	doses	of	all	MEKi	inhibited	ERK	phosphorylation.	In	VOA-1312,	also	highly	sensitive	by	proliferation	assays,	ERK	phosphorylation	is	completely	inhibited	by	trametinib,	and	only	a	small	amount	of	ERK	phosphorylation	is	seen	in	the	low	dose	treatment	conditions	for	selumetinib,	binimetinib,	and	refametinib.	In	VOA-1056,	VOA-3723,	and	VOA-4627,	trametinib	at	low	and	high	dose	completely	blocked	ERK	phosphorylation,	while	there	were	varying	levels	of	activity	amongst	the	other	MEKi.	Refametinib	at	5μM	showed	much	more	effective	inhibition	of	p-ERK	than	either	selumetinib	or	binimetinib	at	high	doses,	as	they	failed	to	suppress	ERK	phosphorylation	at	72	hours	in	VOA-1056,	VOA-3723,	and	VOA-4627.		In	general,	treatment	with	MEKi	induced	an	increase	of	phosphorylated	MEK1/2	in	all	cell	lines,	with	the	exception	of	iOvCa241	treated	with	trametinib.	Both	increases	and	decreases	in	p-MEK	have	been	demonstrated	following	MEKi	treatment	[32]	due	to	differing	cellular	mechanisms	of	compensation	to	MEK	inhibition.	The	increase	in	p-MEK1/2	levels	following	MEKi	therapy	is	likely	context	specific,	and	depends	on	cessation	of	negative	feedback	by	dual	specificity	phosphatases	(DUSPs)	which	dephosphorylate	MEK	in	response	to	high	levels	of	RAS-MAPK	pathway	transcriptional	output.	However,	it	has	been	shown	in	cell	lines	with	altered	RAS-MAPK	signaling		 47	through	BRAF	activating	mutations	that	MEK	inhibition	does	not	induce	an	increase	in	MEK	phosphorylation	as	they	show	an	insensitivity	to	dual	specificity	phosphatase	(DUSP)-mediated	negative	feedback	regulation[49].	Additionally,	it	has	been	theorized	that	trametinib	may	also	inhibit	MEK	phosphorylation	by	RAF	in	a	cell-line	specific	manner	[32].	While	both	VOA-1312	and	iOvCa241	contain	activating	KRAS	mutations,	they	differ	in	ERK	isoform	expression.	Slight	differences	in	their	RAS-MAPK	pathway	status	may	explain	their	differential	p-MEK	expression	levels	in	response	to	MEKi	treatment.								 									 48		Figure	3.3.8.		Dose-response	Western	blots.	LGSOC	cell	lines	treated	with	two	doses	of	MEKi.	Cells	were	harvested	at	24	and	72	hours,	with	no	EGF	stimulation.	Note:	All	phospho-ERK1/2	blots	were	overexposed	to	show	as	much	signal	as	possible	after	MEKi	treatment.	For	each	cell	line,	all	antibodies	come	from	the	same	blot	and	are	exposed	at	the	same	time;	however	images	were	cut	and	re-ordered	to	ensure	the	same	sample	order	for	presentation	purposes.	VOA-1312		 49	Induction	of	apoptosis	following	MEKi	therapy	was	also	evaluated	in	the	LGSOC	cell	lines.	PARP	activation	(cleaved	PARP	or	c-PARP)	was	analyzed	by	Western	blot	(Figure	3.3.8)	using	the	same	low	and	high	doses	of	each	MEKi.	iOvCa241,	VOA-1056	and	VOA-3723	had	negligible	baseline	c-PARP	activity,	but	there	was	induction	of	c-PARP	following	treatment	with	any	of	the	MEKi.	The	analysis	of	apoptotic	response	to	drug	treatment	is	important,	as	it	is	ideal	for	drugs	to	exert	cytotoxic	rather	than	cytostatic	effects	on	the	target	tissue.	VOA-1312	shows	no	baseline	c-PARP	expression,	and	does	not	show	appreciable	c-PARP	induction	after	MEK	inhibition	(Appendix	Figure	A1).		VOA-4627	showed	considerable	c-PARP	activity	at	baseline,	and	there	were	no	appreciable	increases	following	treatment	with	MEKi,	suggesting	that	an	investigation	of	apoptotic	pathway	moieties	and	apoptotic	response	to	treatment	may	be	an	interesting	insight	into	drug	resistance	phenotypes.	Caspase	3/7	activation	was	measured	in	3	of	the	LGSOC	cell	lines	to	validate	the	apoptosis	induction	results	seen	in	the	Western	blot	for	c-PARP	(Figure	3.3.9).	Treatment	with	any	of	the	MEKi	increased	cleaved	caspase	levels	in	iOvCa241,	VOA-1056,	and	VOA-4627,	and	the	magnitude	of	apoptosis	induced	was	most	significant	in	iOvCa241.	VOA-4627	showed	the	least	induction	of	apoptosis,	similarly	to	the	results	seen	in	the	Western	blot	for	c-PARP.	Cellular	apoptosis	can	occur	through	several	different	effector	pathways:	stimulated	by	intrinsic	events	such	as	DNA	damage	and	initiated	by	mitochondrial	proteins,	or	initiated	by	extrinsic	receptor-ligand	mediated	signalling.		Both	cleaved	PARP	levels	and	caspase	3/7	activity	were	chosen	to	analyse	cellular	apoptosis	as	they	represent	downstream	activities	in	both	extrinsic	and	intrinsic		 50	apoptotic	signalling	pathways.	Activity	of	caspases	3	and	7	results	in	cleavage	of	PARP,	a	nuclear	repair	protein,	resulting	in	DNA	degradation	by	endonucleases	during	apoptosis	[51,	52].									 51			Figure	3.3.9.	Caspase	3/7	activity	assay	(Caspase-Glo)	on	three	LGSOC	cell	lines.	Increases	in	cleaved	caspase	over	the	baseline	control	(values	subtracted)	are	shown.		05000100001500020000TRA	0.1uM SEL	1uMBIN	1uMREF	1uMTRA	0.1uM SEL	1uMBIN	1uMREF	1uMTRA	0.1uM SEL	1uMBIN	1uMREF	1uMiOvCa241 VOA-1056 VOA-4627 Luminiscence	(RLU)	Caspase	activity1x	MEKi	treatments24h72h05000100001500020000TRA	0.5uM SEL	5uMBIN	5uMREF	5uMTRA	0.5uM SEL	5uMBIN	5uMREF	5uMTRA	0.5uM SEL	5uMBIN	5uMREF	5uMiOvCa241 VOA-1056 VOA-4627 Luminiscence	(RLU)	Caspase	activityMEKi	treatments	(5x)	in	LGSC	cellsCaspase	3/7	activity	in	LGSC	cells24h72h	 52		 Using	reverse-phase	protein	array,	a	clear	recapitulation	of	the	results	for	ERK	phosphorylation	from	Figure	3.3.6	was	seen.	Levels	of	p-ERK1/2	were	close	to	zero	in	all	the	cell	lines	tested	that	were	treated	with	0.1μM	trametinib,	and	varying	levels	of	p-ERK1/2	were	seen	among	cell	lines	treated	with	the	other	3	MEKi.	Additionally,	it	was	seen	that	EGF	stimulation	allowed	for	some	restoration	of	p-ERK1/2	levels	in	all	of	the	MEKi	treatment	conditions,	except	trametinib,	in	the	majority	of	cell	lines	with	the	exception	of	iOvCa241.	An	interesting	result	was	seen	in	the	RPPA	data	for	VOA-1312,	which	appears	to	lack	the	MEK1	isoform	(Figure	3.3.10).	This	was	not	previously	detected	by	Western	blotting	as	there	is	no	separation	between	the	MEK1	and	MEK2	isoforms	by	SDS-PAGE.		 Results	from	the	24	and	72	hour	dose	response	Western	blots	shown	in	Figure	3.3.8	were	recapitulated	by	RPPA	in	Figure	3.3.11.	For	the	sensitive	cell	line	iOvCa241,	all	MEKi	inhibited	ERK	phosphorylation	even	after	72	hours	treatment.	Phospho-ERK1/2	signal	increased	from	24	to	72	hours	in	the	resistant	cell	lines	VOA-3723	and	VOA-4627,	except	with	trametinib	treatment.	Stimulation	with	EGF	induced	an	increase	in	p-ERK1/2	for	all	treatment	conditions	except	trametinib	in	VOA-3723	and	VOA-4627.			 53			Figure	3.3.10.	RPPA	results	for	phosphorylated	and	total	ERK1/2,	phosphorylated	MEK1/2	and	total	MEK1	for	8	cell	lines	treated	with	the	4	MEKi	and	control	DMSO	with	and	without	EGF	stimulation	after	24	hours.	Red	indicates	low	expression	values.	Biological	triplicate	data	was	averaged.	Antibody	signal	was	normalized	by	serial	dilution	curves	according	to	protocols	established	by	the	Royal	College	of	Surgeons,	Ireland.			Cell	LineTotal	ERK1/2Phospho	ERK1/2Total	MEK1Phospho	MEK1/2Total	ERK1/2Phospho	ERK1/2Total	MEK1Phospho	MEK1/2Total	ERK1/2Phospho	ERK1/2Total	MEK1Phospho	MEK1/2Total	ERK1/2Phospho	ERK1/2Total	MEK1Phospho	MEK1/2SampleCtrlTra	0.1uMSel	1uMBin	1uMRef	1uMCtrl	+EGFTra	0.1uM	+EGFSel	1uM	+EGFBin	1uM	+EGFRef	1uM	+EGFCtrlTra	0.1uMSel	1uMBin	1uMRef	1uMCtrl	+EGFTra	0.1uM	+EGFSel	1uM	+EGFBin	1uM	+EGFRef	1uM	+EGFVOA-4627VOA-4698VOA-1312iOvCa241VOA-1056VOA-3993VOA-3448VOA-3723	 54		Figure	3.3.11.	RPPA	results	for	phospho	and	total	ERK1/2,	phospho	MEK1/2	and	total	MEK1	for	one	selected	sensitive	(iOvCa241)	and	two	selected	resistant	(VOA-3723	and	VOA-4627)		cell	lines	treated	with	the	4	MEKi	and	control	DMSO	with	and	without	EGF	stimulation	after	24	and	72	hours.	Red	indicates	low	expression	values.	Biological	triplicate	data	was	averaged.			 					Cell	LineTotal	ERK1/2Phospho	ERK1/2Total	MEK1Phospho	MEK1/2Total	ERK1/2Phospho	ERK1/2Total	MEK1Phospho	MEK1/2Total	ERK1/2Phospho	ERK1/2Total	MEK1Phospho	MEK1/2SampleCtrlTra	0.1uMSel	1uMBin	1uMRef	1uMCtrl	+EGFTra	0.1uM	+EGFSel	1uM	+EGFBin	1uM	+EGFRef	1uM	+EGFCtrlTra	0.1uMSel	1uMBin	1uMRef	1uMCtrl	+EGFTra	0.1uM	+EGFSel	1uM	+EGFBin	1uM	+EGFRef	1uM	+EGF2472iOvCa241VOA-3723VOA-4627iOvCa241VOA-3723VOA-4627	 55		 Results	from	this	chapter	suggest	exceptional	efficacy	of	trametinib	in	blocking	downstream	MEK1/2	signaling,	even	at	10	times	lesser	dose	than	any	of	the	other	3	MEKi.	A	dose-dependent	effect	is	seen	for	each	of	the	drugs	in	all	of	the	cell	lines	except	iOvCa241	and	VOA-1312,	which	exhibit	exquisite	sensitivity	through	on-target	inhibition	of	ERK	phosphorylation	and	lack	of	proliferation	after	treatment	with	all	of	the	MEKi	even	at	low	doses.	An	investigation	of	molecular	markers	of	MEKi	sensitivity	and	resistance	is	necessary	in	order	to	explain	differential	sensitivity	to	treatment	by	the	panel	of	LGSOC	cell	lines.	Additionally,	the	increased	efficacy	of	trametinib	over	the	other	three	MEKi	in	preclinical	cell	line	tests	suggests	that	it	may	have	a	greater	utility	in	clinical	applications.			 Limitations	of	this	objective	include	a	small	number	of	available	cell	lines,	which	results	in	weak	statistical	correlations	between	mutation	status	and	overall	drug	response.	Additionally,	the	low	proliferation	rate	of	these	cell	lines	can	impede	expedient	data	collection.	Ideally,	more	LGSOC	cell	lines	as	well	as	cell-line	xenograft	and	patient-derived	xenograft	models	would	be	used	to	gather	data	about	MEKi	response	and	LGSOC	molecular	phenotype.	Drug	treatment	studies	are	performed	based	on	a	single	administration	of	a	dose	determined	by	IC50	results	and	review	of	previous	literature	[53-55].	RPPA	analysis	is	limited	by	the	number	and	quality	of	the	primary	antibodies	validated	in	this	technique.			 56	CHAPTER	4.	GLOBAL	PROTEOMIC	ANALYSIS	IN	SENSITIVE	AND	RESISTANT	LGSOC	CELL	LINES			 	4.1.	HYPOTHESIS	AND	AIMS			 Subsequent	to	analysis	of	on-target	and	biological	drug	effects	in	the	LGSOC	cell	lines,	it	became	apparent	that	there	were	two	phenotypes:	sensitive	cell	lines	showing	flat	proliferation	curves	and	low	IC50	values,	and	all	other	cell	lines	which	displayed	a	range	of	resistance	phenotypes	with	some	proliferation	after	treatment.	Based	on	this	observation,	we	decided	to	set	up	a	global	proteomic	arm	of	the	study	to	achieve	Aim	3,	comparing	the	proteomes	of	patient-derived	LGSOC	cell	lines	of	differential	MEKi	sensitivities	using	reverse-phase	protein	array	(RPPA)	and	mass	spectrometry	(MS)	analyses.	Our	hypothesis	is	that	candidate	markers	of	de	novo	MEKi	sensitivity	or	resistance	in	LGSOC,	as	well	as	candidates	which	are	dynamic	after	drug	treatment	in	sensitive	or	resistant	lines	can	be	identified.		By	launching	a	mass-spectrometry	based	discovery	phase	experiment,	we	can	narrow	down	a	list	of	targets	to	focus	our	research	into	novel	therapeutic	strategies	for	LGSOC,	or	identify	markers	of	clinical	predictive	value	for	patients	who	may	respond	to	existing	targeted	therapy	with	MEK	inhibitors.	Quantitative	global	proteomic	data	will	provide	inferences	into	oncogenic	pathway	expression.	Investigating	dynamic	proteins	through	the	use	of	TMT	and	MS3	provides	a	cutting-edge	and	highly	accurate	method	of	multiplexing	and	quantifying	treated	LGSOC	cell	line	samples[56].	The	identification	of	dynamic	proteins	following	analysis	and	comparison	of	the	global	proteomes	of	LGSOC		 57	model	cell	lines	basally	and	in	response	to	MEKi	treatment	may	indicate	de	novo	markers	of	sensitivity	or	resistance	to	treatment,	or	pathway	alterations	representative	of	post-treatment	resistance.		4.2.	METHODS			 4.2.1.	REVERSE	PHASE	PROTEIN	ARRAY	(RPPA)	SIGNALING	PATHWAY	ANALYSIS	OF	MEKI	SENSITIVE	AND	RESISTANT	LGSOC	CELL	LINES.			 The	initial	proteomic	analysis	was	done	by	RPPA,	using	the	same	lysates	used	to	confirm	on-target	drug	effects	in	Chapter	3.	A	total	of	91	different	antibodies	assaying	a	variety	of	signaling	moieties	in	the	RAS-MAPK	pathway	and	apoptotic	pathways	(Appendix,	Table	A2)	were	applied	to	RPPA	slides	containing	1μM	lysates	(in	biological	triplicate)	from	LGSOC	cell	lines	treated	with	control,	0.1μM	trametinib,	and	1μM	selumetinib,	binimetinib,	and	refametinib	with	and	without	EGF	stimulation.	Heatmaps	were	created	using	averaged	data	from	the	three	biological	replicates	for	each	sample,	with	conditional	formatting	in	Microsoft	Excel.	An	SPSS	analysis	(Mann-Whitney	U	Test)	of	differentially	expressed	candidates	between	the	two	sensitive	(iOvCa241	and	VOA-1312)	cell	lines	and	6	resistant	(VOA-1056,	VOA-3993,	VOA-3448,	VOA-4627,	VOA-4698,	and	VOA-4881)	cell	lines,	under	control	and	treatment	conditions	without	EGF	stimulation	was	performed	in	order	to	determine	protein	markers	of	statistical	significance	in	this	phase	of	the	study.	Non-averaged	individual	triplicate	data	was	used	for	all	SPSS	analyses.	The	MEKi	treatment	conditions	chosen	to	include	in	the	SPSS	analysis	were	samples	treated	with	1μM	refametinib	and	0.1μM	trametinib,	as		 58	they	exhibited	better	biological	and	on-target	effects	in	LGSOC,	shown	in	Chapter	3.	Additional	SPSS	analysis	of	trametinib-only	treated	samples	in	sensitive	and	resistant	LGSOC	lines	is	ongoing	in	order	to	validate	if	differential	protein	expression	seen	by	RPPA	is	drug-dependent	or	a	pan-MEK	inhibitor	effect.		 4.2.2.	GLOBAL	QUANTITATIVE	MASS	SPECTROMETRY	OF	MEKI	SENSITIVE	AND	RESISTANT	LGSOC	CELL	LINES.			 Concurrently	to	the	time	the	RPPA	experiment	was	being	run,	a	quantitative	mass	spectrometry	experiment	was	constructed	in	order	to	provide	a	more	in-depth	analysis	into	the	global	proteome	of	selected	cell	lines,	and	as	a	corroborative	tool	to	investigate	statistically	significant	candidates	from	the	RPPA	analysis.	iOvCa241,	VOA-1056,	and	VOA-4627	were	treated	with	DMSO,	0.1μM,	and	1.0μM	trametinib	for	24	hours,	then	trypsinized	into	cell	pellets	which	were	digested	into	peptide	fragments	and	labeled	according	to	protocols	established	by	Dr.	Gregg	Morin’s	laboratory.		A	10-plex	tandem	mass	tag	strategy	(TMT)	(using	isobaric-labeled	tags)	was	chosen	to	label	each	of	the	cell	line	and	treatment	conditions	to	avoid	having	to	grow	the	cells	in	specialized	‘heavy’	media	like	that	which	is	used	in	SILAC	quantitative	mass	spectrometry.	This	approach	allows	for	a	determination	of	the	relative	abundance	of	individual	proteins	between	samples.		An	additional	sample	set	was	prepared	with	the	same	three	cell	lines	treated	with	DMSO	and	0.1μM	trametinib	for	48	hours	when	it	became	apparent	that	the	slow-growing	LGSOC	cells	required	longer	treatment	to	show	detectable	changes	in	translation(Figure	A1).	The	higher	dose	of	trametinib	was	omitted	from	the	48	hour	mass	spectrometry	experiment	due	to	significant	cell	death.	Data		 59	collected	from	the	48	hour	experiment	was	used	to	identify	proteins	of	interest	as	it	yielded	a	greater	number	of	differentially	expressed	candidates	between	cell	lines	and	treatment	conditions.		 Briefly,	our	MS/MS/MS	(MS3)	strategy	using	an	Orbitrap®	triple	quadrupole	mass	spectrometer	(Figure	4.1.1)	involves	trypsin-mediated	digestion	of	our	drug-treated	cell	pellets	into	short	peptides,	purified	from	DNA	using	paramagnetic	bead	technology.	These	peptides	are	subsequently	labeled	with	isobaric	tags	consisting	of	a	chemical	linker	group,	a	mass-normalization	region,	and	a	unique	reporter	region.	After	electrospray	ionization	(ESI)	and	injection	into	the	first	quadrupole,	peptides	(precursor	ions)	are	separated	based	on	their	mass	to	charge	(m/z)	ratio	prior	to	entering	the	second	quadrupole,	a	collision	cell.	Collision-induced	dissociation	(CID)	results	in	further	fragmentation	of	peptides	into	product	ions,	as	well	as	the	isobaric	tags,	resulting	in	a	release	of	the	unique	reporter	ion.	In	the	final	quadrupole,	peptides	are	identified	based	on	amino	acid	composition	and	their	relative	abundance	in	each	drug-treated	sample	quantified	by	reporter	ion	abundance[56-58].	Following	detection	of	the	product	ions,	data	was	median-normalized,	log2	transformed,	and	samples	were	compared	pairwise	using	a	PECA	statistical	analysis	package.	The	t	statistics	were	calculated	and	used	to	calculate	p	values,	assuming	normal	distribution,	for	expression	of	each	peptide	between	a	pair	of	samples.		False-discovery	rate	adjusted	p	values	(corrected	for	multiple	testing	due	to	nature	of	peptide-level	detection	methods)	and	log2	fold	change	(slr)	values	were	used	to	determine	protein	candidates	of	interest[59,	60].	Raw	peptide	expression	values	for	iOvCa241,	VOA-1056,	and	VOA-4627	were	used		 60	to	confirm	SPSS-verified	candidates	basally	and	with	MEKi	treatment	from	the	RPPA	data,	and	relative	peptide	abundances	between	pairwise	comparisons	confirmed	using	the	PECA-analyzed	data.	Subsequent	deep	analysis	of	novel	candidates	from	the	mass	spectrometry	data	is	ongoing.		Figure	4.2.1.	Triple	quadrupole	mass	spectrometer	schematic.			 4.2.3	WESTERN	BLOT	VALIDATION	OF	RPPA	AND	MS	RESULTS			 As	an	additional	proteomic	validation	technique,	differentially	expressed	candidates	between	sensitive	and	resistant	LGSOC	cell	lines	that	were	discovered	by	RPPA	and	also	detected	in	MS	data	with	statistical	significance	were	tested	by	Western	blotting	to	visualize	changes	in	expression	between	samples.	Raw	lysates	(non-DMSO	treated)	for	each	LGSOC	cell	line	(lysates	were	unavailable	for	VOA-3448	and	VOA-4881)	were	used	to	compare	basal	expression	levels	among	cell	lines.					ESI	 q1	 q2	 q3	Detec-on	and	data	analysis	Collision	cell	Precursor	ions	 Product	ions		 61		4.3.	RESULTS	AND	DISCUSSION				 Qualitative	visual	analysis	of	RPPA	heatmaps	(Figure	4.3.1)	shows	several	obvious	differences	between	control	and	trametinib-treated	LGSOC	cell	line	samples,	as	well	as	between	sensitive	and	resistant	LGSOC	cell	line	samples	under	both	treatment	conditions.	As	well	as	obvious	suppression	in	all	cell	lines	of	ERK1/2	phosphorylation	caused	by	trametinib	treatment,	the	most	immediately	obvious	difference	between	sensitive	and	resistant	cell	lines	is	a	lack	of	PKCα	expression	in	the	sensitive	LGSOC	lines.	To	further	elucidate	differences	between	sensitive	and	resistant	lines,	an	SPSS	analysis	by	a	Mann	Whitney	U	test,	assuming	non-parametric	data,	was	subsequently	performed.		 62			 63		Figure	4.3.1.	RPPA	heatmaps	of	LGSOC	sensitive	and	resistant	cell	lines	treated	with	control	or	0.1μM	trametinib.	Averaged	data	from	three	biological	replicates	was	used.	Red	indicates	low	expression	values.		 64			 From	the	SPSS-analyzed	RPPA	data,	12	candidates	showing	significant	differences	in	expression	levels	basally	between	sensitive	and	resistant	cell	lines	were	identified,	and	20	candidates	were	differentially	expressed	upon	treatment	with	a	MEK	inhibitor	(Table	4.3.1).	PKCα,	EGFR,	and	Smac/DIABLO	were	significantly	differentially	expressed	between	sensitive	and	resistant	cell	lines	at	basal	(control)	levels	and	upon	drug	treatment,	indicating	that	these	candidates	may	represent	de	novo	markers	of	drug	response.	Box	plots	were	generated	for	all	differentially	expressed	candidates,	demonstrating	that	PKCα	and	EGFR	are	highly	expressed	in	resistant	cell	lines	compared	to	sensitive,	and	Smac/DIABLO	is	highly	expressed	in	sensitive	cell	lines	compared	to	resistant	(Figure	4.3.2).	18	of	these	candidates	(from	both	lists)	which	were	present	in	the	mass	spectrometry	data	were	then	examined	using	the	PECA-analyzed	and	raw	signal	abundance	data.											 65	Table	4.3.1.	SPSS	results	from	independent	samples	Mann	Whitney	U	test:	significantly	different	protein	expression	levels	of	lysates	treated	with	control	(DMSO)	and	0.1μM	trametinib/1μM	refametinib	between	sensitive	(iOvCa241	and	VOA-1312)	and	resistant	(VOA-1056,	VOA-3993,	VOA-3723,	VOA-3448,	VOA-4627,	VOA-4698,	and	VOA-4881)	LGSOC	cell	lines.		 Control	(DMSO)	 Trametinib	0.1μM	Refametinib	1μM	Antibody	 p-value	 p-value	Candidate	1	 	 0.035	Candidate	2	 	 0.043	Candidate	3	 	 0.002	Candidate	4	 	 0.041	Candidate	5	 	 0.000	Candidate	6	 0.009	 0.001	Candidate	7	 0.011	 0.000	EGFR	 0.004	 0.000	Candidate	8	 	 0.001	Candidate	9	 	 0.019	Candidate	10	 	 0.033	Candidate	11	 0.009	 	Candidate	12	 	 0.017	Candidate	13	 0.019	 	Candidate	14	 0.028	 0.006	Candidate	15	 0.016	 	Candidate	16	 	 0.037	Candidate	17	 0.046	 0.004	Candidate	18	 0.000	 	Candidate	19	 0.013	 0.000	PKC_alpha	 0.000	 0.000	Smac/Diablo	 0.009	 0.000	Candidate	20	 	 0.046	Candidate	21	 	 0.027							 66			Figure	4.3.2.	Box	plots	for	select	statistically	significant	candidates	from	RPPA	analysis	between	sensitive	and	resistant	LGSOC	cell	lines	under	DMSO	control	(left	side)	and	0.1μM	trametinib/1μM	refametinib	(right	side)	treatment	conditions.						PKCα	EGFR	Sensi/ve																							Resistant	 Sensi/ve																							Resistant	Sensi/ve																							Resistant	 Sensi/ve																							Resistant	Smac/	Diablo	10.0	5.0	0.0	-5.0	10.0	5.0	0.0	-5.0	PKCα	10.5	10.0	9.5	9.0	8.5	8.0	11.0	10.5	10.0	9.5	9.0	8.5	EGFR	10.0	9.5	9.0	8.5	8.0	7.5	10.5	10.0	9.5	9.0	8.5	8.0	7.5	SMAC/Diablo	DMSO	 TRA/REF		 67		 Similarly	to	the	RPPA	data,	raw	protein	abundance	levels	gathered	from	the	48	hour	mass	spectrometry	experiment	show	a	significantly	higher	expression	level	of	both	PKCα	and	EGFR	in	the	resistant	cell	lines	VOA-1056	and	VOA-4627	compared	to	the	sensitive	iOvCa241	under	drug-treated	and	control	conditions.	Levels	of	Smac/DIABLO	are	lower	in	the	resistant	cell	lines	compared	to	the	sensitive	in	both	drug-treated	and	control	conditions	(Table	4.3.2).	While	these	three	candidates	show	the	most	striking	differences	between	drug-response	phenotypes,	BCL-XL	also	shows	a	decrease	in	the	resistant	lines	compared	to	the	sensitive	in	this	data	set.	While	this	decrease	is	only	shown	to	be	statistically	significant	in	the	RPPA	data	under	MEKi	treatment,	the	RPPA	analysis	is	limited	by	the	quality	of	the	antibodies	being	used,	indicating	that	this	protein	may	still	represent	a	de	novo	marker	of	drug	response	phenotype.	In	order	to	confirm	trends	seen	in	the	raw	protein	abundance	data,	the	PECA-analyzed	data	between	pairs	of	samples	was	consulted	to	eliminate	possible	false	discovery	results	and	to	ensure	statistical	significance	in	differing	expression	levels	seen	by	the	mass	spectrometry	data.								 68	Table	4.3.2.	Mass	spec	data	(48hr	drug	treatment	experiment)	for	statistically	significant	RPPA	candidates.	Raw	protein	abundance	values	for	VOA-1056	and	VOA-4627	were	normalized	to	iOvCa241	abundance	levels	for	both	the	control	and	the	drug-treated	data	sets	in	order	to	illustrate	fold-change	differences	in	each	data	set.	Values	were	obtained	from	averaging	raw	abundance	data	from	three	biological	replicates.			Control	 0.1μM		Trametinib	Gene	 iOvCa241	VOA-1056	 VOA-4627	 iOvCa241	 VOA-1056	 VOA-4627	Candidate	1	 1.000	 0.966	 1.117	 1.000	 0.987	 1.201	Candidate	2	 1.000	 0.910	 0.843	 1.000	 1.030	 0.960	Candidate	3	 1.000	 1.324	 0.372	 1.000	 1.136	 0.375	Candidate	4	 1.000	 0.653	 0.831	 1.000	 0.821	 0.613	Candidate	6	 1.000	 1.075	 1.192	 1.000	 1.106	 1.054	Candidate	7	 1.000	 0.846	 1.192	 1.000	 1.069	 1.134	EGFR	 1.000	 1.918	 2.902	 1.000	 2.103	 3.136	Candidate	22	 1.000	 0.915	 1.032	 1.000	 0.634	 1.092	Candidate	10	 1.000	 1.664	 0.860	 1.000	 2.278	 0.999	Candidate	11	 1.000	 0.751	 0.957	 1.000	 0.868	 1.186	Candidate	13	 1.000	 1.667	 NA	 1.000	 0.641	 NA	Candidate	14	 1.000	 0.818	 1.157	 1.000	 0.889	 0.993	Candidate	23	 1.000	 1.417	 0.957	 1.000	 1.381	 0.858	Candidate	24	 1.000	 1.373	 0.935	 1.000	 1.996	 0.775	Candidate	19	 1.000	 1.014	 1.095	 1.000	 0.928	 1.200	PKCa	 1.000	 7.175	 6.270	 1.000	 4.769	 3.991	SMAC/DIAB	 1.000	 0.284	 0.345	 1.000	 0.358	 0.510	Candidate	20	 1.000	 1.026	 0.872	 1.000	 1.126	 0.864		 					 		 69		 Pairwise	analysis	of	iOvCa241	vs.	VOA-1056	and	iOvCa241	vs.	VOA-4627	by	PECA	(Table	4.3.3)	indicates	statistically	significant	differences	between	levels	of	PKCα,	EGFR,	and	Smac/DIABLO	with	p.fdr	values	<	0.05,	providing	orthogonal	corroboration	of	the	SPSS	analyzed	RPPA	data	under	both	control	and	drug-treatment	conditions.	Differences	in	BCL-XL	expression	were	found	to	be	non-statistically	significant	in	the	PECA	analyses,	eliminating	it	as	a	potential	marker	of	drug	response	phenotype.						 70		Table	4.3.3.	Mass	spectrometry	PECA-analyzed	data,	showing	slr	(log2	of	the	fold	change)	and	p.fdr	(false-discovery	rate	adjusted	p-values)	for	statistically	significant	RPPA	candidates	by	SPSS.		slr p.fdr slr p.fdr slr p.fdr slr p.fdrCandidate	1 0.077 1 -0.106 0.98564 -0.035 1 -0.290 0.08915Candidate	2 0.019 1 0.155 0.53248 -0.113 1 -0.044 1Candidate	3 -0.293 0.22931 0.705 0.00159 0.009 1 1.007 0.00871Candidate	4 0.390 0.14110 0.139 0.53552 0.231 0.96919 0.267 0.64374Candidate	6 -0.028 1 -0.106 0.76937 -0.105 1 -0.093 1Candidate	7 0.229 0.25223 -0.327 0.00226 0.042 1 -0.517 0.18899EGFR -0.542 4.17E-13 -0.969 5.46E-45 -0.541 7.34E-08 -0.939 6.57E-22Candidate	22 -0.093 0.96490 -0.585 0.00033 0.367 0.74642 -0.012 1Candidate	10 -0.516 0.00011 0.086 1 -0.810 9.11E-06 0.108 1Candidate	11 0.309 0.10127 0.042 1 0.240 0.91181 0.083 1Candidate	13 NA NA NA NA 0.303 0.79527 -0.082 1Candidate	14 0.143 0.73196 -0.082 1 0.063 1 -0.059 1Candidate	23 0.357 0.53437 0.082 0.91770 -0.161 1 0.456 0.37518Candidate	24 -0.047 1 0.128 1 -0.949 0.11067 0.288 0.40752Candidate	19 -0.008 1 -0.045 1 0.102 1 -0.013 1PKCa -1.353 1.20E-18 -1.147 3.90E-23 -1.342 2.51E-06 -1.018 4.73E-07SMAC/DIABLO 1.333 5.53E-06 0.802 3.38E-06 0.727 0.01177 0.531 0.04660Candidate	20 0.060 1 0.291 0.30117 -0.013 1 0.222 0.44619Control 0.1uM	TrametinibiOvCa241	vs.	VOA-1056 iOvCa241	vs.	VOA-4627 iOvCa241	vs.	VOA-1056 iOvCa241	vs.	VOA-4627	 71		 Finding	potential	predictive	markers	of	de	novo	MEK	inhibitor	sensitivity	or	resistance	can	have	great	clinical	value	in	identifying	subsets	of	patients	who	will	benefit	from	targeted	therapeutic	strategies.	Through	global	proteomic	analysis	of	2	sensitive	and	6	resistant	LGSOC	cell	lines	by	RPPA,	1	sensitive	and	2	resistant	lines	by	quantitative	mass	spectrometry,	we	have	identified	a	potential	predictive	marker	of	MEKi	sensitivity.	High	levels	of	Smac/DIABLO	are	present	in	the	sensitive	cell	lines	VOA-1312	and	iOvCa241	compared	to	the	6	resistant	lines	by	RPPA,	results	which	are	echoed	by	the	quantitative	mass	spec	results	in	a	small	subset	of	cell	lines.	High	expression	levels	of	the	intrinsic	apoptosis	pathway	member	Smac/DIABLO	in	certain	cancers	have	been	shown	to	increase	the	likelihood	of	the	cancer	cells	to	undergo	apoptosis	in	several	studies	by	sequestering	the	apoptosis-inhibiting	protein	XIAP,	which	may	explain	why	this	marker	is	more	present	in	the	sensitive	LGSOC	cell	lines	[61-63].			 The	proteins	PKCα	and	EGFR	are	shown	to	be	more	highly	expressed	in	MEKi	resistant	LGSOC	cell	lines	by	both	RPPA	and	mass	spectrometry,	results	which	are	validated	by	immunoblotting	(Figure	4.3.3).	Inhibition	of	these	proteins	by	a	targeted	inhibitor	or	siRNA-mediated	knockdown	may	result	in	an	alteration	of	the	MEKi	resistance	phenotype,	and	is	an	important	next	step	in	this	research.	EGFR	expression	has	been	shown	to	be	linked	to	a	drug	resistance	phenotype	in	MEK	inhibitor	resistant	solid	tumors[26,	64,	65],	and	has	been	suggested	as	an	attractive	target	for	combination	therapy	in	addition	to	MEK	inhibition	in	lung	cancers	with	RAS/MAPK	pathway	alterations[66,	67].	PKCα,	an	anti-apoptotic	protein,	has	been	shown	to	contribute	to	a	pro-survival	phenotype	in	mesothelioma	cells	upon	treatment	with	cisplatin		 72	chemotherapy[68],	and	overexpression	of	PKCα	in	the	commercial	ovarian	cancer	cell	line	A2780	has	been	shown	to	decrease	cisplatin	sensitivity[69].	Based	on	the	literature,	there	is	a	rationale	for	exploring	the	functional	roles	of	both	PKCα	and	EGFR	in	MEKi-resistant	LGSOC	cell	lines.	As	a	clinical	option,	it	is	currently	more	feasible	to	target	highly	expressed	proteins	by	the	application	of	a	targeted	therapy	than	to	attempt	to	overexpress	markers	of	sensitivity	in	human	patients.													 73		Figure	4.3.3.	Western	blot	validation	for	two	selected	candidates,	PKCα	and	EGFR,	identified	as	differentially	expressed	at	basal	level	(non-DMSO	treated)	between	sensitive	and	resistant	LGSOC	cell	lines.	Cell	line	lysates	for	VOA-3448	and	VOA-4881	were	omitted	from	this	blot	due	to	lack	of	volume.														β-ac%n	PKCα	EGFR		 iOvCa241	VOA1312	VOA1056	VOA3993	VOA3723	VOA4627	VOA4698	Se	 Re		 74	CHAPTER	5.	CONCLUSIONS	AND	FUTURE	DIRECTIONS		5.1.	CONCLUSIONS			 The	initial	objective	of	this	work	was	to	establish	and	molecularly	characterize	a	library	LGSOC	cell	lines	that	represent	the	disease	more	accurately	than	existing	commercial	ovarian	cancer	cell	lines	for	use	as	a	preclinical	investigational	model.	We	have	been	successful	in	establishing	at	least	two	stable	cell	lines	from	patient	samples	over	the	previous	two	years,	as	well	as	several	promising	primary	cultures	(data	not	shown)	in	addition	to	9	LGSOC	cell	lines	established	by	Clara	Salamanca.	These	11	novel	cell	lines	have	been	molecularly	characterized	by	HotSpot	mutational	profiling,	copy	number	variation	analysis,	and	STR	profiling.	All	cell	lines	are	unique,	and	several	paired	cell	lines	from	the	same	patient	during	different	disease	intervals	have	been	established.	Only	cell	lines	developed	from	advanced	or	recurrent	cases	of	LGSOC	have	been	chosen	to	molecularly	characterize,	as	these	patients	are	those	who	require	alternate	therapeutic	strategies.	This	library	of	characterized	cell	lines	therefore	represents	a	unique	resource	for	preclinical	therapeutic	testing	in	this	disease.			 Due	to	common	upregulations	of	the	RAS/MAPK	pathway	in	LGSOC,	the	signaling	moiety	MEK	has	been	noted	as	an	important	therapeutic	target.	There	are	several	clinical	trials	currently	ongoing	for	MEK	inhibitors	in	LGSOC	and	other	solid	RAS	mutant	tumors.	In	this	study,	4	MEKi	were	selected	to	test	the	biological	and	on-target	effect	of	MEK	inhibition	in	LGSOC.	There	were	distinctly	different	efficacies	between	the	4	MEKi.		 75	Trametinib	showed	the	most	outstanding	results	among	all	cell	lines,	with	low	IC50	values	(nM	range;	vs.	μM	range	for	other	drugs),	greatest	effect	on	proliferation,	and	best	on-target	inhibition	of	ERK	phosphorylation	by	Western	blot,	capillary	isoelectric	point	focusing,	and	RPPA.	The	remainder	of	the	drugs	showed	varying	efficacies:	binimetinib	and	selumetinib	showed	the	least	inhibition	of	proliferation	in	all	resistant	cell	lines,	and	poor	inhibition	of	p-ERK	in	all	cell	lines,	particularly	after	EGF	stimulation	or	longer	treatment	time.	Of	note,	a	recent	clinical	trial	of	binimetinib	in	LGSOC	has	recently	been	cancelled	due	to	no	impact	seen	on	progression-free	survival	[NCT01849874].	The	results	of	our	studies	in	a	preclinical	setting	show	reduced	efficacy	of	binimetinib	compared	to	the	MEKi	trametinib	or	refametinib,	suggesting	that	it	may	not	be	the	most	promising	targeted	inhibitor	in	this	disease.	Also	of	note,	trametinib	has	been	recently	approved	for	the	treatment	of	advanced	melanoma	patients	with	the	BRAF	V600E	activating	mutation	as	a	single	agent	or	in	combination	with	a	BRAF	inhibitor[70-73]		 During	the	evaluation	of	MEK	inhibitor	effects	in	the	newly	characterized	LGSOC	cell	lines,	it	became	apparent	that	there	were	two	distinct	drug	response	phenotypes	among	the	cell	lines.	VOA-1312	and	iOvCa241	were	seen	to	be	exquisitely	sensitive	to	MEKi	therapy	based	on	proliferation	and	Western	blot	experiments.	Both	cell	lines	harbor	KRAS	mutations,	iOvCa241	(G12D)	lacks	the	ERK2	isoform,	and	VOA-1312	(G12V)	lacks	the	MEK1	isoform.		Of	note,	the	study	by	Farley	et	al.,	of	selumetinib	treatment	on	LGSOC	patients	was	unable	to	correlating	RAS	mutation	status	with	patient	response[28],	however	it	has	been	noted	that	depending	on	the	RAS	mutation,		 76	alterations	in	different	downstream	signaling	pathways	can	occur[74].	From	the	clinical	trial	of	selumetinib	[NCT00551070],	one	patient	had	a	ongoing	complete	response	of	greater	than	5	years	with	continuous	selumetinib	therapy	[28].	While	mutational	profiling	of	this	tumor	showed	it	be	wild-type	for	KRAS	or	BRAF	mutations,	next-generation	sequencing	revealed	a	15	nucleotide	delection	in	a	negative	regulatory	region	of	the	MAP2K1	gene	encoding	the	protein	MEK1.	This	activating	event	results	in	oncogenesis	in	mouse	models	through	upregulation	of	the	RAS/MAPK	pathway[75].	In	another	study,	the	lung	cancer	cell	line	NCI-H1437	was	identified	to	be	exquisitely	sensitive	to	MEK	inhibition,	and	further	profiling	also	identified	an	activating	mutation	in	the	MAP2K1	gene[54].		In	melanomas,	activating	gene	fusions	in	BRAF	have	been	shown	to	predict	sensitivity	to	combination	therapy	with	MEK	inhibitors[76].	Taken	together	with	results	from	our	study,	these	data	suggest	that	dependency	on	the	RAS/MAPK	pathway	through	activating	mutations	in	KRAS,	BRAF,	MAP2K1,	and	potentially	other	signalling	pathway	members,	may	predict	tumor	sensitivity	to	MEK	inhibition.			 Results	from	the	RPPA	experiment	of	2	sensitive	and	6	resistant	LGSOC	cell	lines	treated	with	control	and	the	4	MEKi	showed	12	statistically	significant	differentially	expressed	candidates	between	sensitive	and	resistant	cell	lines	in	the	control	condition,	and	20	after	treatment	with	trametinib/refametinib.	Some	of	the	markers	discovered	will	be	cell	line	specific	candidates	of	de	novo	MEKi	response,	and	others	may	be	markers	of	response	following	drug	treatment.	The	three	candidates	with	the	lowest	p-values	that	were	shared	between	both	treatment	conditions	were	Smac/DIABLO,	PKCα,	and	EGFR.	Quantitative	global	proteomic	analysis	by	mass	spectrometry	confirmed		 77	these	three	candidates	were	differentially	expressed	between	one	sensitive	and	two	resistant	LGSOC	cell	lines	under	control	and	trametinib-treatment	conditions	at	48	hours.	PKCα	and	EGFR	were	validated	by	Western	blotting	as	being	highly	expressed	in	resistant	MEKi	cell	lines.	There	is	currently	ongoing	data	analysis	to	identify	candidates	of	interest	which	have	dynamic	activity	upon	drug	treatment	in	sensitive	and	resistant	LGSOC	cell	lines. 5.2.	FUTURE	DIRECTIONS		Deeper	analysis	of	the	24	and	48	hour	mass	spec	data	sets	is	required	in	order	to	elucidate	novel	candidates	of	interest	in	both	de	novo	MEKi	response	and	post-therapy	changes	in	expression	level.	Validation	by	immunoblotting	and	functional	analyses	(overexpression	or	siRNA-mediated	knockdown/targeted	drug	inhibition)	of	statistically-significant	protein	candidates	obtained	from	RPPA	and	mass	spectrometry	analyses	is	a	crucial	next	step	in	order	to	determine	if	sensitivity	or	resistance	to	MEKi	can	be	altered.	Analyzing	more	LGSOC	cell	lines	of	known	drug-response	phenotype	by	quantitative	mass	spectrometry	will	provide	more	statistical	power	in	identifying	novel	candidates	that	identify	de	novo	or	acquired	drug	response	phenotypes.	Phosphoproteomic	analyses	of	drug-treated	sensitive	and	resistant	cell	lines	should	be	performed	in	order	to	determine	basal	signaling	pathway	status	and	changes	in	signaling	following	drug	treatment	in	LGSOC	cell	lines.	A	valuable	insight	into	the	tumor	biology	of	LGSOC	may	be	to	perform	global	proteomic	analyses	on	serial	tumor	biopsies	from	LGSOC	patients	before	and	after	systemic	MEKi	therapy,	to	investigate		 78	potential	markers	of	drug	response	or	acquired	resistance.	Creation	and	proteomic	analysis	of	a	pair	of	isogeneic	cell	lines,	one	with	acquired	resistance	to	MEK	inhibition,	may	also	provide	a	useful	tool	in	the	study	of	acquired	resistance	to	MEKi.	In	addition	to	proteomic	analyses,	whole	genome	sequencing	of	all	developed	LGSOC	lines,	paired	tumor	samples,	and	buffy	coats	if	available	should	be	performed	as	a	discovery	technique	to	determine	if	genomic	abnormalities	such	as	gene	fusions	or	activating	mutations	in	signaling	pathway	genes	exist,	as	well	as	to	validate	that	the	novel	cell	lines	recapitulate	the	molecular	biology	of	the	patient	tumor	tissue	they	were	derived	from.	If	novel	mutations/fusions	are	discovered,	functional	assays	in	cell	line	and	xenograft	models	should	be	performed	to	determine	the	effect	these	genetic	aberrations	have	on	the	biology	and	drug	sensitivity	of	LGSOC.		In	order	to	continue	the	testing	of	novel	therapeutic	options	in	a	LGSOC	model	system,	it	is	important	to	continue	to	develop	appropriate	cell	line	models	and	to	develop	patient-derived	xenograft	mouse	models	as	an	in	vivo	model	of	LGSOC	biology	with	which	to	perform	crucial	preclinical	testing.	Additionally,	as	there	are	few	effective	treatments	and	lack	of	standardized	treatment[5]	for	advanced/recurrent	LGSOC,	patient	data	and	clinical	outcomes	should	meticulously	tracked	for	each	case,	with	the	intention	of	identifying	the	most	effective	therapeutic	regimens	already	approved	for	clinical	use	and	improving	the	standard	of	care	in	this	disease.				 79	REFERENCES		1.	 Vang,	R.,	M.	Shih	Ie,	and	R.J.	Kurman,	Ovarian	low-grade	and	high-grade	serous	carcinoma:	pathogenesis,	clinicopathologic	and	molecular	biologic	features,	and	diagnostic	problems.	Adv	Anat	Pathol,	2009.	16(5):	p.	267-82.	2.	 Grabowski,	J.P.,	et	al.,	Operability	and	chemotherapy	responsiveness	in	advanced	low-grade	serous	ovarian	cancer.	An	analysis	of	the	AGO	Study	Group	metadatabase.	Gynecol	Oncol,	2016.	140(3):	p.	457-62.	3.	 Schmeler,	K.M.,	et	al.,	Neoadjuvant	chemotherapy	for	low-grade	serous	carcinoma	of	the	ovary	or	peritoneum.	Gynecol	Oncol,	2008.	108(3):	p.	510-4.	4.	 Ansquer,	Y.,	et	al.,	Neoadjuvant	chemotherapy	for	unresectable	ovarian	carcinoma:	a	French	multicenter	study.	Cancer,	2001.	91(12):	p.	2329-34.	5.	 Gourley,	C.,	et	al.,	Gynecologic	Cancer	InterGroup	(GCIG)	consensus	review	for	ovarian	and	primary	peritoneal	low-grade	serous	carcinomas.	Int	J	Gynecol	Cancer,	2014.	24(9	Suppl	3):	p.	S9-13.	6.	 Gershenson,	D.M.,	et	al.,	Hormonal	therapy	for	recurrent	low-grade	serous	carcinoma	of	the	ovary	or	peritoneum.	Gynecol	Oncol,	2012.	125(3):	p.	661-6.	7.	 Hsu,	C.Y.,	et	al.,	Characterization	of	active	mitogen-activated	protein	kinase	in	ovarian	serous	carcinomas.	Clin	Cancer	Res,	2004.	10(19):	p.	6432-6.	8.	 Diaz-Padilla,	I.,	et	al.,	Ovarian	low-grade	serous	carcinoma:	a	comprehensive	update.	Gynecol	Oncol,	2012.	126(2):	p.	279-85.	9.	 Nakayama,	N.,	et	al.,	KRAS	or	BRAF	mutation	status	is	a	useful	predictor	of	sensitivity	to	MEK	inhibition	in	ovarian	cancer.	Br	J	Cancer,	2008.	99(12):	p.	2020-8.	10.	 Gershenson,	D.M.,	The	life	and	times	of	low-grade	serous	carcinoma	of	the	ovary.	Am	Soc	Clin	Oncol	Educ	Book,	2013.	11.	 Gershenson,	D.M.,	et	al.,	Recurrent	low-grade	serous	ovarian	carcinoma	is	relatively	chemoresistant.	Gynecol	Oncol,	2009.	114(1):	p.	48-52.		 80	12.	 Stuart,	G.C.,	et	al.,	2010	Gynecologic	Cancer	InterGroup	(GCIG)	consensus	statement	on	clinical	trials	in	ovarian	cancer:	report	from	the	Fourth	Ovarian	Cancer	Consensus	Conference.	Int	J	Gynecol	Cancer,	2011.	21(4):	p.	750-5.	13.	 Anglesio,	M.S.,	et	al.,	Type-specific	cell	line	models	for	type-specific	ovarian	cancer	research.	PLoS	One,	2013.	8(9):	p.	e72162.	14.	 Okoye,	E.,	E.D.	Euscher,	and	A.	Malpica,	Ovarian	Low-grade	Serous	Carcinoma:	A	Clinicopathologic	Study	of	33	Cases	With	Primary	Surgery	Performed	at	a	Single	Institution.	Am	J	Surg	Pathol,	2016.	15.	 Della	Pepa,	C.,	et	al.,	Low	Grade	Serous	Ovarian	Carcinoma:	From	the	molecular	characterization	to	the	best	therapeutic	strategy.	Cancer	Treat	Rev,	2015.	41(2):	p.	136-143.	16.	 Gershenson,	D.M.,	Low-grade	serous	carcinoma	of	the	ovary	or	peritoneum.	Ann	Oncol,	2016.	27	Suppl	1:	p.	i45-i49.	17.	 Miller,	C.R.,	K.E.	Oliver,	and	J.H.	Farley,	MEK1/2	inhibitors	in	the	treatment	of	gynecologic	malignancies.	Gynecol	Oncol,	2014.	133(1):	p.	128-37.	18.	 Pohl,	G.,	et	al.,	Inactivation	of	the	mitogen-activated	protein	kinase	pathway	as	a	potential	target-based	therapy	in	ovarian	serous	tumors	with	KRAS	or	BRAF	mutations.	Cancer	Res,	2005.	65(5):	p.	1994-2000.	19.	 Grisham,	R.N.,	et	al.,	BRAF	mutation	is	associated	with	early	stage	disease	and	improved	outcome	in	patients	with	low-grade	serous	ovarian	cancer.	Cancer,	2013.	119(3):	p.	548-54.	20.	 Wong,	K.K.,	et	al.,	BRAF	mutation	is	rare	in	advanced-stage	low-grade	ovarian	serous	carcinomas.	Am	J	Pathol,	2010.	177(4):	p.	1611-7.	21.	 Tsang,	Y.T.,	et	al.,	KRAS	(but	not	BRAF)	mutations	in	ovarian	serous	borderline	tumour	are	associated	with	recurrent	low-grade	serous	carcinoma.	J	Pathol,	2013.	231(4):	p.	449-56.	22.	 Lu,	S.,	et	al.,	Drugging	Ras	GTPase:	a	comprehensive	mechanistic	and	signaling	structural	view.	Chem	Soc	Rev,	2016.	23.	 Yeh,	J.J.,	et	al.,	KRAS/BRAF	mutation	status	and	ERK1/2	activation	as	biomarkers	for	MEK1/2	inhibitor	therapy	in	colorectal	cancer.	Mol	Cancer	Ther,	2009.	8(4):	p.	834-43.		 81	24.	 Lin,	L.,	et	al.,	The	Hippo	effector	YAP	promotes	resistance	to	RAF-	and	MEK-targeted	cancer	therapies.	Nat	Genet,	2015.	47(3):	p.	250-6.	25.	 Greten,	F.R.,	YAP1	takes	over	when	oncogenic	K-Ras	slumbers.	Cell,	2014.	158(1):	p.	11-2.	26.	 Sun,	C.,	et	al.,	Intrinsic	resistance	to	MEK	inhibition	in	KRAS	mutant	lung	and	colon	cancer	through	transcriptional	induction	of	ERBB3.	Cell	Rep,	2014.	7(1):	p.	86-93.	27.	 Prahallad,	A.,	et	al.,	Unresponsiveness	of	colon	cancer	to	BRAF(V600E)	inhibition	through	feedback	activation	of	EGFR.	Nature,	2012.	483(7387):	p.	100-3.	28.	 Farley,	J.,	et	al.,	Selumetinib	in	women	with	recurrent	low-grade	serous	carcinoma	of	the	ovary	or	peritoneum:	an	open-label,	single-arm,	phase	2	study.	The	Lancet	Oncology,	2013.	14(2):	p.	134-140.	29.	 McCubrey,	J.A.,	et	al.,	Ras/Raf/MEK/ERK	and	PI3K/PTEN/Akt/mTOR	cascade	inhibitors:	how	mutations	can	result	in	therapy	resistance	and	how	to	overcome	resistance.	Oncotarget,	2012.	3(10):	p.	1068-111.	30.	 Akinleye,	A.,	et	al.,	MEK	and	the	inhibitors:	from	bench	to	bedside.	J	Hematol	Oncol,	2013.	6:	p.	27.	31.	 Yamaguchi,	T.,	et	al.,	Antitumor	activities	of	JTP-74057	(GSK1120212),	a	novel	MEK1/2	inhibitor,	on	colorectal	cancer	cell	lines	in	vitro	and	in	vivo.	Int	J	Oncol,	2011.	39(1):	p.	23-31.	32.	 Gilmartin,	A.G.,	et	al.,	GSK1120212	(JTP-74057)	is	an	inhibitor	of	MEK	activity	and	activation	with	favorable	pharmacokinetic	properties	for	sustained	in	vivo	pathway	inhibition.	Clin	Cancer	Res,	2011.	17(5):	p.	989-1000.	33.	 Iverson,	C.,	et	al.,	RDEA119/BAY	869766:	a	potent,	selective,	allosteric	inhibitor	of	MEK1/2	for	the	treatment	of	cancer.	Cancer	Res,	2009.	69(17):	p.	6839-47.	34.	 Adjei,	A.A.,	et	al.,	Phase	I	pharmacokinetic	and	pharmacodynamic	study	of	the	oral,	small-molecule	mitogen-activated	protein	kinase	kinase	1/2	inhibitor	AZD6244	(ARRY-142886)	in	patients	with	advanced	cancers.	J	Clin	Oncol,	2008.	26(13):	p.	2139-46.		 82	35.	 Yeh,	T.C.,	et	al.,	Biological	characterization	of	ARRY-142886	(AZD6244),	a	potent,	highly	selective	mitogen-activated	protein	kinase	kinase	1/2	inhibitor.	Clin	Cancer	Res,	2007.	13(5):	p.	1576-83.	36.	 Wallace,	E.M.,	et	al.,	Progress	towards	therapeutic	small	molecule	MEK	inhibitors	for	use	in	cancer	therapy.	Curr	Top	Med	Chem,	2005.	5(2):	p.	215-29.	37.	 Chang,	L.	and	M.	Karin,	Mammalian	MAP	kinase	signalling	cascades.	Nature,	2001.	410(6824):	p.	37-40.	38.	 Takekawa,	M.,	K.	Tatebayashi,	and	H.	Saito,	Conserved	docking	site	is	essential	for	activation	of	mammalian	MAP	kinase	kinases	by	specific	MAP	kinase	kinase	kinases.	Mol	Cell,	2005.	18(3):	p.	295-306.	39.	 Cheng,	Y.,	G.	Zhang,	and	G.	Li,	Targeting	MAPK	pathway	in	melanoma	therapy.	Cancer	Metastasis	Rev,	2013.	32(3-4):	p.	567-84.	40.	 Friday,	B.B.	and	A.A.	Adjei,	Advances	in	targeting	the	Ras/Raf/MEK/Erk	mitogen-activated	protein	kinase	cascade	with	MEK	inhibitors	for	cancer	therapy.	Clin	Cancer	Res,	2008.	14(2):	p.	342-6.	41.	 Wan,	P.T.,	et	al.,	Mechanism	of	activation	of	the	RAF-ERK	signaling	pathway	by	oncogenic	mutations	of	B-RAF.	Cell,	2004.	116(6):	p.	855-67.	42.	 Baines,	A.T.,	D.	Xu,	and	C.J.	Der,	Inhibition	of	Ras	for	cancer	treatment:	the	search	continues.	Future	Med	Chem,	2011.	3(14):	p.	1787-808.	43.	 Cox,	A.D.	and	C.J.	Der,	Ras	history:	The	saga	continues.	Small	GTPases,	2010.	1(1):	p.	2-27.	44.	 Kohler,	M.,	et	al.,	Activation	loop	phosphorylation	regulates	B-Raf	in	vivo	and	transformation	by	B-Raf	mutants.	EMBO	J,	2016.	35(2):	p.	143-61.	45.	 Pylayeva-Gupta,	Y.,	E.	Grabocka,	and	D.	Bar-Sagi,	RAS	oncogenes:	weaving	a	tumorigenic	web.	Nat	Rev	Cancer,	2011.	11(11):	p.	761-74.	46.	 Mendoza,	M.C.,	E.E.	Er,	and	J.	Blenis,	The	Ras-ERK	and	PI3K-mTOR	pathways:	cross-talk	and	compensation.	Trends	Biochem	Sci,	2011.	36(6):	p.	320-8.	47.	 Beaufort,	C.M.,	et	al.,	Ovarian	cancer	cell	line	panel	(OCCP):	clinical	importance	of	in	vitro	morphological	subtypes.	PLoS	One,	2014.	9(9):	p.	e103988.		 83	48.	 Chay,	W.Y.,	et	al.,	Low	grade	serious	carcinoma	of	the	peritoneum	in	a	BRCA1	carrier	previously	diagnosed	with	a	"low-grade	serous	tubal	intra-epithelial	carcinoma"	(STIC)	on	risk	reducing	surgery.	Gynecol	Oncol	Rep,	2015.	12:	p.	72-4.	49.	 Pratilas,	C.A.,	et	al.,	(V600E)BRAF	is	associated	with	disabled	feedback	inhibition	of	RAF-MEK	signaling	and	elevated	transcriptional	output	of	the	pathway.	Proc	Natl	Acad	Sci	U	S	A,	2009.	106(11):	p.	4519-24.	50.	 McCubrey,	J.A.,	et	al.,	Roles	of	signaling	pathways	in	drug	resistance,	cancer	initiating	cells	and	cancer	progression	and	metastasis.	Adv	Biol	Regul,	2015.	57:	p.	75-101.	51.	 Galluzzi,	L.,	J.M.	Bravo-San	Pedro,	and	G.	Kroemer,	Organelle-specific	initiation	of	cell	death.	Nat	Cell	Biol,	2014.	16(8):	p.	728-36.	52.	 Prokhorova,	E.A.,	et	al.,	Role	of	the	nucleus	in	apoptosis:	signaling	and	execution.	Cell	Mol	Life	Sci,	2015.	72(23):	p.	4593-612.	53.	 Baranski,	Z.,	et	al.,	MEK	inhibition	induces	apoptosis	in	osteosarcoma	cells	with	constitutive	ERK1/2	phosphorylation.	Genes	Cancer,	2015.	6(11-12):	p.	503-12.	54.	 Gannon,	H.S.,	et	al.,	Identification	of	an	"Exceptional	Responder"	Cell	Line	to	MEK1	Inhibition:	Clinical	Implications	for	MEK-Targeted	Therapy.	Mol	Cancer	Res,	2016.	14(2):	p.	207-15.	55.	 Leung,	E.Y.,	et	al.,	Relationships	between	signaling	pathway	usage	and	sensitivity	to	a	pathway	inhibitor:	examination	of	trametinib	responses	in	cultured	breast	cancer	lines.	PLoS	One,	2014.	9(8):	p.	e105792.	56.	 McAlister,	G.C.,	et	al.,	MultiNotch	MS3	enables	accurate,	sensitive,	and	multiplexed	detection	of	differential	expression	across	cancer	cell	line	proteomes.	Anal	Chem,	2014.	86(14):	p.	7150-8.	57.	 Dayon,	L.	and	J.C.	Sanchez,	Relative	protein	quantification	by	MS/MS	using	the	tandem	mass	tag	technology.	Methods	Mol	Biol,	2012.	893:	p.	115-27.	58.	 Hughes,	C.S.,	et	al.,	Ultrasensitive	proteome	analysis	using	paramagnetic	bead	technology.	Mol	Syst	Biol,	2014.	10:	p.	757.		 84	59.	 Suomi,	T.,	et	al.,	Using	Peptide-Level	Proteomics	Data	for	Detecting	Differentially	Expressed	Proteins.	J	Proteome	Res,	2015.	14(11):	p.	4564-70.	60.	 Kall,	L.,	et	al.,	Posterior	error	probabilities	and	false	discovery	rates:	two	sides	of	the	same	coin.	J	Proteome	Res,	2008.	7(1):	p.	40-4.	61.	 Martinez-Ruiz,	G.,	et	al.,	Role	of	Smac/DIABLO	in	cancer	progression.	J	Exp	Clin	Cancer	Res,	2008.	27:	p.	48.	62.	 Kashkar,	H.,	et	al.,	XIAP	targeting	sensitizes	Hodgkin	lymphoma	cells	for	cytolytic	T-cell	attack.	Blood,	2006.	108(10):	p.	3434-40.	63.	 Kashkar,	H.,	et	al.,	XIAP-mediated	caspase	inhibition	in	Hodgkin's	lymphoma-derived	B	cells.	J	Exp	Med,	2003.	198(2):	p.	341-7.	64.	 Mizukami,	T.,	et	al.,	EGFR	and	HER2	signals	play	a	salvage	role	in	MEK1-mutated	gastric	cancer	after	MEK	inhibition.	Int	J	Oncol,	2015.	47(2):	p.	499-505.	65.	 Liu,	D.,	X.	Liu,	and	M.	Xing,	Activities	of	multiple	cancer-related	pathways	are	associated	with	BRAF	mutation	and	predict	the	resistance	to	BRAF/MEK	inhibitors	in	melanoma	cells.	Cell	Cycle,	2014.	13(2):	p.	208-19.	66.	 Tricker,	E.M.,	et	al.,	Combined	EGFR/MEK	Inhibition	Prevents	the	Emergence	of	Resistance	in	EGFR-Mutant	Lung	Cancer.	Cancer	Discov,	2015.	5(9):	p.	960-71.	67.	 Li,	S.,	et	al.,	Synergistic	interaction	between	MEK	inhibitor	and	gefitinib	in	EGFR-TKI-resistant	human	lung	cancer	cells.	Oncol	Lett,	2015.	10(4):	p.	2652-2656.	68.	 Muscella,	A.,	et	al.,	PKC-delta/PKC-alpha	activity	balance	regulates	the	lethal	effects	of	cisplatin.	Biochem	Pharmacol,	2015.	98(1):	p.	29-40.	69.	 Righetti,	S.C.,	et	al.,	Molecular	alterations	of	cells	resistant	to	platinum	drugs:	role	of	PKCalpha.	Biochim	Biophys	Acta,	2006.	1763(1):	p.	93-100.	70.	 Awad,	M.M.	and	R.J.	Sullivan,	Dabrafenib	in	combination	with	trametinib	for	the	treatment	of	metastatic	melanoma.	Expert	Rev	Clin	Pharmacol,	2015.	8(1):	p.	25-33.	71.	 King,	J.W.	and	P.D.	Nathan,	Role	of	the	MEK	inhibitor	trametinib	in	the	treatment	of	metastatic	melanoma.	Future	Oncol,	2014.	10(9):	p.	1559-70.		 85	72.	 Thota,	R.,	D.B.	Johnson,	and	J.A.	Sosman,	Trametinib	in	the	treatment	of	melanoma.	Expert	Opin	Biol	Ther,	2015.	15(5):	p.	735-47.	73.	 Chung,	C.	and	S.	Reilly,	Trametinib:	a	novel	signal	transduction	inhibitor	for	the	treatment	of	metastatic	cutaneous	melanoma.	Am	J	Health	Syst	Pharm,	2015.	72(2):	p.	101-10.	74.	 Ihle,	N.T.,	et	al.,	Effect	of	KRAS	oncogene	substitutions	on	protein	behavior:	implications	for	signaling	and	clinical	outcome.	J	Natl	Cancer	Inst,	2012.	104(3):	p.	228-39.	75.	 Grisham,	R.N.,	et	al.,	Extreme	Outlier	Analysis	Identifies	Occult	Mitogen-Activated	Protein	Kinase	Pathway	Mutations	in	Patients	With	Low-Grade	Serous	Ovarian	Cancer.	J	Clin	Oncol,	2015.	33(34):	p.	4099-105.	76.	 Hutchinson,	K.E.,	et	al.,	BRAF	fusions	define	a	distinct	molecular	subset	of	melanomas	with	potential	sensitivity	to	MEK	inhibition.	Clin	Cancer	Res,	2013.	19(24):	p.	6696-702.	77.	 Tone,	A.A.,	et	al.,	Intratumoral	heterogeneity	in	a	minority	of	ovarian	low-grade	serous	carcinomas.	BMC	Cancer,	2014.	14:	p.	982.	78.	 McConechy,	M.K.,	et	al.,	Subtype-specific	mutation	of	PPP2R1A	in	endometrial	and	ovarian	carcinomas.	J	Pathol,	2011.	223(5):	p.	567-73.												 86				APPENDIX:	REAGENTS	AND	MATERIALS			LGSOC	cell	line	development:		Advanced	or	recurrent	LGSOC	patient	samples	(tumor	and	ascites)	were	obtained	from	tumor	banks	(Vancouver	General	Hospital	and	the	British	Columbia	Cancer	Agency	(BCCA);	London	Regional	Cancer	Program,	London,	Ontario,	Canada).	Tumor	bank	protocols	and	the	research	relating	to	this	study	were	approved	through	the	human	ethics	review	board	at	BCCA	and	the	University	of	British	Columbia.		Clinical	information	was	extracted	retrospectively	from	the	patient	medical	records.		Tumor	bank	pathology	must	be	reported	by	certified	gynecological	pathologists	to	ensure	diagnostic	accuracy.	LGSOC	patient-derived	uniquely	identified	cell	lines	were	derived	in-house	through	continuous	in	vitro	culture	of	primary	patient	material	(tumor	tissue	or	ascites)	obtained	through	the	OvCaRe	Tumor	bank.	The	Human	ethics	review	board	at	the	University	of	British	Columbia	and	the	BC	Cancer	Agency	approved	all	research	relating	to	this	study	(H14-02859)	and	patients	provided	consent	for	collection	of	tumor	tissues	in	order	to	establish	the	cell	lines.		Cell	cultures	were	established	and	maintained	in	199:105	(1:1)	media	(#M5017,	#M6395,	Sigma-Aldrich,	Oakville,	Ontario,	Canada)	supplemented	with	10%	defined	fetal	bovine	serum	(dFBS;	#SH30070.03,	Hyclone,	GE		 87	Life	Sciences,	Logan,	UT,	USA)	at	37oC	and	5%	CO2.	A	total	of	11	unique	cell	lines	from	5	LGSOC	patients	have	been	developed	to	date.		Authentication	and	unique	identification	of	cell	lines:		Microsatellite	Analysis	of	Short	Tandem	Repeats	(STRs)	was	performed	for	genetic	mapping,	linkage	analysis,	and	tracing	inheritance	patterns	in	the	DNA	of	the	recently	established	LGSOC	cell	lines/cultures.	STR	analyses	of	10	markers/loci	(TH01,	D21S11,	D5S818,	D13S317,	D7S820,	D16S539,	CSF1PO,	AMEL,	vWA	&	TPOX)	were	performed	by	Genewiz	Inc.	(South	Plainfield,	NJ). Results	were	compared	to	the	DSMZ	STR	profile	website	database	(https://www.dsmz.de/services/services-human-and-animal-cell-lines/online-str-analysis.html),	confirming	that	the	newly-established	cell	lines	were	unique.	HotSpot	Mutational	Profiling	and	Copy	Number	Variation	Analysis:			 DNA	was	extracted	from	all	cell	lines	using	All	Prep	DNA/RNA	Mini	kit	(Cat.	No.	80204,	Qiagen)	according	to	protocol	instructions,	and	quantified	using	a	NanoDrop	2000TM	UV-Vis	(Thermo-Scientific,	Burlington,	ON,	Canada).	Purity	(A260/280	≅	1.8)	was	checked	to	ensure	no	contaminating	phenolic	or	protein	material.	Sequencing	libraries	were	created	from	all	tumor	DNA	for	molecular	characterization	using	Ion	Torrent	AmpliSeqTM	Cancer	Hotspot	Panel	Version	2	(Life	Technologies,	Grand	Island,	NY,	USA)	as	per	manufacturer’s	protocols.	A	total	of	50	oncogenes	and	tumor		 88	suppressor	genes	were	screened	(Table	2.3.2).	Full	list	of	mutations	tested	available	at	https://www.thermofisher.com/order/catalog/product/4475346.	Sanger	sequencing	was	performed	to	confirm	select	high	frequency	missense	mutations,	using	methods	previously	described	[77].	Primer	sequences	are	listed	in	Table	A1.	Priming	sites	for	−12	M13	forward	and	−27	M13	reverse	were	added	to	the	5’	ends	to	allow	direct	Sanger	sequencing	of	amplicons.	Methods	are	as	previously	described	[78].	CNV	analysis	was	done	using	Illumina®	HumanOmni	2.5M-8	Array	or	CytoScan®	HD	array	(Affymetrix,Inc)	according	to	manufacture’s	protocols.	Nexus	Copy	NumberTM	software	was	used	to	analyze	the	copy	number	data	from	this	two	platforms.						 89	Table	A1.	Sanger	sequencing	primers	used	for	HotSpot	result	validation.		 90	IC50	analysis:		Selumetinib	(AZD6244;	Cat.	No.	S1008),	Trametinib	(GSK1120212;	Cat.	No.	S2673),	Binimetinib	(MEK162;	Cat.	No.	S7007),	and	Refametinib	(Bay	86-9766;	Cat.	No.	S1089)	were	purchased	from	Selleck	Chemicals	(Houston,	TX,	USA).	Dimethylsulfoxide	(DMSO)	(D2650)	was	purchased	from	Sigma-Aldrich	(Oakville,	Ontario,	Canada).	In	order	to	determine	the	minimum	dose	required	to	kill	50%	of	the	cells,	96-well	plates	were	prepared	for	each	cell	line	so	that	they	would	reach	confluence	in	96	hours	from	plating.	After	24	hours,	cells	were	then	treated	with	a	spectrum	of	doses	[0.05-100μM]	for	each	of	the	4	MEKi.	After	72	hours	drug	treatment,	cells	were	fixed	using	10%	methanol/10%	acetic	acid	in	H2O	for	10	minutes,	stained	with	0.25%	crystal	violet	in	methanol	for	10	min,	dissolved	in	methanol	with	0.1%	SDS	for	1-2	hours	at	room	temperature	and	absorbance	was	read	at	595nm	on	a	BioTek	Epoch	plate	reader	SN:257811.,	and	the	percentage	of	surviving	cells	was	graphed	according	to	the	control.	2-3	biological	replicates	and	4	technical	replicates	of	each	condition	were	performed	in	order	to	determine	the	IC50	value.		Proliferation	curves:			 LGSOC	cells	were	plated	in	96-well	plates	in	order	to	yield	between	10-20%	confluence	after	24	hours,	at	which	point	the	cells	were	drug-treated.	Cell	growth	curves	over	5	-7	days	treatment	with	the	4	MEK	inhibitors	at	0.1	and	0.5μM	(trametinib)	or	1	and	5	μM	(selumetinib,	binimetinib,	and	refametinib)	and	DMSO	controls	were	tracked	using	an	Incucyte	Live	Imaging	System	(Essen	Instruments	Inc.,	Ann	Arbor,	MI,		 91	USA).	Drug	doses	were	selected	based	upon	literature	review	and	IC50	results.		5	technical	replicates	and	2	biological	replicates	of	each	condition	were	performed.		Phase-contrast	images	of	cell	monolayers	at	six	hour	intervals	for	5-7	days	were	used	to	obtain	proliferation	curves.	Data	analysis	was	performed	using	the	IncuCyte™	cell	proliferation	assay	software.	MTS	Cell	Viability	Assay:		Cell	viability	was	measured	using	MTS-Cell	Titer	96R	Aqueous	Non-Radioactive	Cell	Proliferation	Assay	following	the	manufacturer’s	recommendations	(Cat.	No.		Cat.	No.	G5430,	Promega,	Madison,	WI,	USA).	On	day	5-7	of	the	cell	proliferation	assay,	the	media	in	each	of	the	96	wells	was	replaced	with	100μL	of	fresh	media	and	20μL	of	MTS	reagent.	Plates	were	incubated	for	3.5	hours	at	37°	C	in	humidified	5%	CO2.	Absorbance	at	490nm	was	measured	using	a	microplate	reader	(BioTek	Epoch	SN:257811).	The	cell	viability	after	MEKi	treatment	was	compared	to	DMSO	control	treated	cells.	LGSOC	cell	apoptosis	assay:		Cell	apoptosis	was	measured	using	a	Caspase-Glo®	3/7	Assay	Kit	(Cat.	No.		Cat.	No.	G8090,	Promega,	Madison,	WI,	USA).	Three	cell	lines,		iOvCa241,	VOA-1056	and	VOA-4627	(8x103,	5x103,	4x103	cells/96-well	for	24h	treatment	and,	6x103,	4x103,	3x103	cells/96-well	for	72h	treatment)	were	incubated	with	DMSO	vehicle	alone,	or	the	corresponding	MEKi	(selumetinib,	binimetinib	or	refametinib	at	1μM	concentration,	or	trametinib	at	0.1μM).	At	the	end	of	each	treatment,	luminescence	of	each	well	was	measured	by	a	luminometer	(Tecan	Infinite	M200Pro).	The	experiments	were		 92	performed	in	triplicate	for	all	conditions	and	cell	lines.	Western	blot	analysis:		Cells	plated	in	60mm	dishes	were	treated	for	24	and	72	hours	with	each	of	the	MEK	inhibitors	at	0.1	and	0.5μM	(trametinib)	or	1	and	5	μM	(selumetinib,	binimetinib,	and	refametinib).	Cells	were	harvested	using	in-house	non-ionic	lysis	buffer	(20mM	Bicine	(pH	7.5)	0.6%	CHAPS,	Aqueous	Inhibitor	mix	(40mM	sodium	fluoride,	17mM	beta	glycerophosphate,	1mM	sodium	orthovanadate,	2mM	EDTA,	10mM	EGTA),	DMSO	inhibitor	mix	(1mM	AEBSF,	5ug/mL	aprotinin,	20μM	bestatin,	5ug/mL	E-64	protease	inhibitor,	10ug/mL	leupeptin,	7ug/mL	pepstatin	A,	phosphatase	inhibitor	cocktail	[cantharidin,	bromotetramisole,	microcystin	LR])	and	the	protein	extracts	collected	and	spun	at	10000	rpm	for	15	minutes.	Protein	extracts	were	separated	by	SDS-PAGE	using	8%	polyacrylamide	gels,	then	transferred	to	nitrocellulose	membranes	at	100V	for	2	hours.	Primary	antibodies	included	ERK1/2	(Millipore,	Cat.	No.	06-182),	p-MAPK	(p-ERK1/2)	(Cell	signaling,	Cat.No.	4376S),	MEK1/2	(Santa	Cruz	Biotechnologies,	Cat.	No.	436),	p-MEK1/2	(Cell	Signaling,	Cat.	No.	9154S),	c-PARP	(Cell	Signaling,	Cat.	No.	9541S),	PKCa	(Cell	Signaling,	Cat.	No.	2056),	and	EGFR	(Santa	Cruz	Biotechnologies,	Cat.	No.	71032).	Vinculin	(V9131,	Sigma)	or	B-actin	(A5441,	Sigma)	were	used	as	a	protein	loading	control.	Secondary	antibodies	(goat-anti-mouse	or	goat-anti-rabbit,	Sigma	Cat.	No.	A9917	and	A0545)	were	used	accordingly.	Western	blots	were	imaged	using	Immobilon	HRP	reagent		(Cat.	No.	WBKLS0500,	Millipore,	Etobicoke,	ON,	Canada)	and	developed	by	autoradiograph.			 93		Figure	A1.	Western	Blot	analysis	of	VOA-1312	and	VOA-1056	for	cleaved	PARP	induction	after	24	hour	treatment	with	0.1μM	trametinib	or	Control	(DMSO).	Lysates	were	non-EGF	stimulated.		Capillary	isoelectric	point	focusing:		Native	capillary	isoelectric	point	focusing	(cIEF)	was	performed	to	assess	total	ERK	isoform	expression	using	NanoPro1000™	System	(ProteinSimple™,	Santa	Clara,	CA)	according	to	manufacturer	protocols	[23].	G2	premix	gradient	(pH	5–8)	(Cat.	No.		040–972,	ProteinSimple™,	ERK	1/2	(Cat.	No.	040-474,	ProteinSimple™)	primary	antibody	and	Goat	anti-Rabbit	(Santa	Cruz	sc-2054)	Human	Absorbed	IgG	secondary	antibody	was	used	to	identify	the	protein	isoforms.	EGF	cell	stimulation	(20	ng/ml	for	10	minutes)	was	use	for	both	the	cIEF	and	Western	Blot	experiments.	Apoptosis	assay:		Apoptosis	was	measured	using	a	Caspase-Glo®	3/7	Assay	Kit	(Cat.	No.		Cat.	No.	G8090,	Promega,	Madison,	WI,	USA).	iOvCa241,	VOA-1056	and	VOA-4627	LGSOC	cells	(8x103,	5x103,	4x103	cells/96-well	for	24h	treatment	and,	6x103,	4x103,	3x103	cells/96-well	for	72h	treatment)	were	incubated	for	24	and	72	hours	with	vehicle	alone	(DMSO)	C	 TRA	C	 TRA VOA-1312		VOA-1056		 	c-PARP		 94	or	the	corresponding	MEKi	(trametinib	0.1μM;	or	selumetinib,	binimetinib	or	refametinib	1μM).	At	the	end	of	each	treatment,	luminescence	of	each	well	was	measured	in	a	plate-reading	luminometer	(Tecan	Infinite	M200Pro).	Experiments	were	performed	in	triplicate	for	all	conditions	and	cell	lines.	Reverse-phase	protein	array	(RPPA):		Cell	lysates	from	8	LGSOC	lines,	2	sensitive	and	6	resistant,	were	treated	separately	with	DMSO	control	(1μL/mL)	and	with	the	4	MEKi	at	0.1μM	(trametinib)	or	1μM	(refametinib,	selumetinib,	binimetinib)	with	and	without	EGF	stimulation	in	three	biological	replicates.	Three	of	the	cell	lines,	demonstrating	sensitive	and	resistant	phenotypes,	were	treated	with	control	and	0.1μM(trametinib)	and	1μM	(refametinib,	selumetinib,	binimetinib)	of	MEKi	and	harvested	at	24	and	72	hours.		The	drug	treatment	lysates	were	appropriately	diluted	to	1μg/mL	and	denatured	using	tris(2-carboxyethyl)phosphine	(TCEP),	and	subsequently	submitted	for	proteomic	analysis	to	Dr.	Bryan	Hennessey	at	the	Royal	College	of	Surgeons,	Dublin,	Ireland.	91	antibodies	evaluating	a	variety	of	signaling	moieties	are	included	in	the	array	(Table	A2).							 95	Table	A2.	List	of	proteins	tested	in	RPPA	array.	Known	oncogenic	candidates,	as	well	as	moieties	associated	with	apoptotic	and	MAPK-related	pathways	were	chosen	to	include	in	the	array.	Two	different	antibodies	against	phospho-MAPK/ERK1/2	were	used,	that	which	was	used	in	the	Western	blots	in	this	study	and	one	provided	by	the	RPPA	service.	Antibodies	included	in	RPPA	array	1	 Akt	 31	EGFR	(Y1068)	 61	p27	2	 Akt	(S473)	 32	EGFR	(Y1173)	 62	p38	MAPK	3	 Akt	(T308)	 33	Fak	 63	p38	MAPK	(T180/Y182)	4	 Akt	2	 34	Fak	(Y925)	 64	p53	5	 AMPK	alpha	1	 35	Gab	1	 65	p70	S6K1	6	 AMPK	(T172)	 36	Gab	1	(Y627)	 66	p70	S6K1	(T389)	7	 Apaf-1	 37	Glutamate	Dehydrogenase	1/2	67	Parp	8	 ATP5A	 38	Glutaminase	 68	Parp	cleaved	9	 Bak	 39	GSK	3B	 69	PDK1	10	Bax	 40	GSK	3B	(S9)	 70	PDK1	(S241)	11	Bcl2	 41	G6PD	 71	PI3Kp110alpha	12	Bcl2	(S70)	 42	HER	2	 72	PKC	alpha	13	Bcl2	(T56)	 43	HER	2	(Y1248)	 73	PKC	alpha	(S657)	14	Bcl-xL	 44	HER	3	 74	PTEN	15	Bid	 45	HER	3	(Y1289)	 75	PYGM	16	Bim	 46	Hexokinase	II	 76	S6	Ribosomal	Protein	(S235/236)	17	Caspase	3	 47	hIAP2	(cIAP1)	 77	S6	Ribosomal	Protein	(S240/244)	18	Caspase	7	cleaved	(D198)	48	Hif-1	alpha	 78	Shc	(Y317)	19	Caspase	8	 49	IGF1RB	 79	Smac/DIABLO	20	Caspase	9	 50	kRas	 80	Src	21	Caspase	9	cleaved	(D315)	51	LDHA	 81	Src	(Y527)	22	Caspase	9	cleaved	(D330)	52	MAPK	ERK	1/2	 82	Src	family	(Y416)	23	Chk1	 53	MAPK	ERK	1/2	(T202/Y204)	83	Stat	3	24	Chk1	(S345)	 54	MAPK	ERK	1/2	(T202/Y204)	MC	84	Stat	3	(Y705)	25	c-MET	 55	MEK	1	 85	TIGAR	26	c-MET	(Y1234/1235)	56	MEK	1/2	(S217/221)	 86	VEGFR2	27	c-Raf	 57	Monocarboxylic	Acid	Transporter	4	87	XIAP	28	c-Raf	(S338)	 58	mTOR	 88	Estrogen	Receptor	29	EGFR	 59	mTOR	(S2448)	 89	Estrogen	Receptor	(S118)	30	EGFR	(Y992)	 60	NFkB	p65	(S536)	 90	HER4			 91	Progesterone	Receptor			 96	Mass	Spectrometry	global	proteome	profiling:		VOA-4627,	VOA-1056,	and	iOvCa241	were	chosen	to	represent	two	resistant	and	one	sensitive	cell	line,	respectively.	Each	cell	line	was	treated	with	10μL/mL	DMSO	control,	0.1μM	and	1μM	trametinib	for	24	hours	in	biological	triplicate,	and	DMSO	control	and	0.1μM	trametinib	for	48	hours	in	biological	duplicate.	Cell	pellets	were	harvested	by	trypsinization	and	washed	with	1X	dPBS.	At	least	1.0-5.0x106	cells	per	pellet	were	used	to	prepare	samples	for	analysis.	Peptide	samples	were	trypsin-digested,	purified	with	SP3	paramagnetic	bead	technology,	and	labeled	using	a	10-plex	tandem	mass	tag	(TMT)	approach	using	protocols	published	by	Hughes	et	al.	in	2014[58].	Samples	were	pre-fractionated	using	reverse-phase	high-performance	liquid	chromatograpy	in	a	gradient	of	acetonitrile	and	20mM	ammonium	hydroxide	in	water	(Figure	A2).	A	MS/MS/MS	(MS3)	approach	was	used	to	quantify	the	global	proteomes	between	samples	by	running	fractionated,	purified	peptides	on	an	Orbitrap	Fusion	mass	spectrometer	and	comparing	data	using	PECA	statistical	analysis[56,	59].						 97				Figure	A2.	Mass	spectrometry	experimental	design	of	24	and	48	hour	sample	sets.	iOvCa241	(sensi.ve)	VOA	4627	(resistant)	VOA	1056	(resistant)	DMSO	 TRA	0.1	uM	TRA	1.0	uM	 DMSO	TRA	0.1	uM	TRA	1.0	uM	DMSO	TRA	0.1	uM	TRA	1.0	uM	Pool	 DMSO	TRA	0.1	uM	TRA	1.0	uM	 DMSO	TRA	0.1	uM	TRA	1.0	uM	DMSO	TRA	0.1	uM	TRA	1.0	uM	24h	drug	treatment	 24h	drug	treatment	24h	drug	treatment	Create	pooled	control,	label	each	sample	with	individual	isobaric	tag	iOvCa241	(sensi.ve)	VOA	4627	(resistant)	VOA	1056	(resistant)	DMSO	 TRA	0.1	uM	 DMSO	TRA	0.1	uM	DMSO	TRA	0.1	uM	Pool	 DMSO	 TRA	0.1	uM	 DMSO	TRA	0.1	uM	DMSO	TRA	0.1	uM	48h	drug	treatment	 48h	drug	treatment	48h	drug	treatment	Create	pooled	control,	label	each	sample	with	individual	isobaric	tag		 98		Figure	A3.	Reverse-phase	HPLC	pre-fractionation	of	TMT-labeled	peptides	for	24	and	48	hour	mass	spectrometry	experiments.	Samples	were	run	on	a	Phenomenex	EVO	C18	Core-Shell	Column	(15cm	length	x	2.1mm	diameter;	1.7μM	diameter	beads)	and	48	fractions	were	collected	from	minute	19	to	minute	73	of	80.	

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0319145/manifest

Comment

Related Items