THE BALANCE OF RAS/ERK AND PI3K/AKT/MTOR SIGNALING REGULATES PITUITARY LACTOTROPE HOMEOSTASIS AND IS DYSREGULATED IN
PROLACTINOMA by
ALLYSON KATHLEEN ROOF
B.S., University of Arizona, Tucson, Arizona, 2009 M.S., University of Arizona, Tucson, Arizona, 2012
A thesis submitted to the faculty of the Graduate School of the University of Colorado in partial fulfillment
of the requirements for the degree of Doctor of Philosophy
Integrated Physiology Program 2017
This thesis for the Doctor of Philosophy degree by Allyson Kathleen Roof
has been approved for the Integrated Physiology program
by
Margaret Wierman, Chair Andrew Bradford Kathleen Connell Rebecca Schweppe
Heidi Wilson
Arthur Gutierrez-Hartmann, Advisor
Roof, Allyson Kathleen (PhD, Integrated Physiology)
The Balance of Ras/ERK And PI3K/AKT/mTOR Signaling Regulates Pituitary Lactotrope Homeostasis and is Dysregulated in Prolactinoma
Thesis directed by Professor Arthur Gutierrez-Hartmann
ABSTRACT
Dysregulation of the signaling pathways that govern lactotrope biology contributes to tumorigenesis of prolactin-secreting pituitary adenomas, or prolactinomas, leading to a state of pathological hyperprolactinemia. Elevated serum prolactin suppresses the hypothalamic-pituitary-gonadal (HPG) axis, resulting in osteoporosis and infertility, as well as decreased libido. Dopamine agonists (DAs), the standard therapy for prolactinoma, signal through the dopamine D2 receptor (D2R) on lactotrope cells to inhibit proliferation and prolactin synthesis. However, 10-20% of prolactinoma patients are resistant to DAs and require
surgical tumor resection. Dopamine-mediated D2R activation results in ERK stimulation and PI3K inhibition, and dysfunctional D2R-mediated signaling is thought be the primary
mechanism by which patients become resistant to DA therapies. As such, the work presented here aims to acquire a better understanding of the effects of activated D2R signaling on lactotrope cells. First, I directly examine the role of ERK in GH4T2 cells and show that long-term activation of the Ras/ERK pathway promotes differentiation of the bi-hormonal
somatolactotrope GH4T2 precursor cell into a prolactin-secreting, lactotrope cell phenotype, and not only fails to promote cell proliferation, but also diminishes tumorigenic
characteristics. Next, I aim to identify the mechanisms that allow GH4T2 cells to evade an oncogenic response to Ras, and find that Ral plays a key role in directing Ras signaling
outcome in somatolactotropes. Next, I demonstrate that PI3K signals through AKT/mTOR to drive GH4T2 cell proliferation, but is not sufficient to increase tumorigenicity. I also show that the DA cabergoline signals through D2L, the long isoform of D2R, to inhibit mTOR and reduce lactotrope proliferation. Lastly, I show data to support the hypothesis that there is a balance of ERK and PI3K signaling required to maintain lactotrope homeostasis, and I demonstrate that this balance is dysregulated in human prolactinoma. Taken together, the data presented here reveal an intricate balance of signaling that involves pathway crosstalk and counterregulation, and is required to maintain homeostasis in lactotrope cells, with loss of this equilibrium contributing to lactotrope tumorigenesis. These data suggest that dual inhibition of PI3K and ERK presents a potential alternative approach to therapy for DA resistant prolactinoma.
The form and content of this abstract are approved. I recommend its publication. Approved: Arthur Gutierrez-Hartmann
ACKNOWLEDGEMENTS
I consider myself very fortunate to have had so many exceptional teachers and mentors throughout my education. In particular, I would like to thank Dr. Claudia Stanescu for helping me realize my love for teaching science, and without whom I likely would not have chosen to pursue a doctoral degree. I would also like to thank Dr. Heddwen Brooks, my master’s thesis advisor, for teaching me how to be passionate about science and for being an inspiring example of a hard-working and successful female scientist. And finally, to Arthur, for recognizing that life does not have a pause button, and for reminding me of this fact when I needed to hear it. His willingness to extend his mentorship beyond the realm of science by sharing his wisdom on both science and life has helped me grow into a stronger person. I would also like to thank him for giving me the independence to decide where I wanted this project to take me. This has undoubtedly helped me grow as a scientist, and has been a huge contributing factor in my decision to pursue a career in academia.
I would also like to thank and acknowledge the members of my committee: Dr. Maggie Wierman, Dr. Andy Bradford, Dr. Kathleen Connell, Dr. Rebecca Schweppe, and Dr. Heidi Wilson. All of you have provided invaluable insight and feedback on this project, and you have been admirably committed to helping me succeed. Thank you.
I owe an exceptional amount of gratitude to the Gutierrez-Hartmann lab. I would like to thank and acknowledge Tammy Trudeau for her training and guidance, and I owe her an exceptional amount of thanks for laying the groundwork for this project before I joined the lab. However, I would be remiss if I did not also thank her for being a wonderful friend. Our off-campus lunch adventures make up some of my favorite memories from the last five years. In addition to Tammy, I would like to thank Hany Abdel-Hafiz for his endless wisdom
and sense of humor. Thank you for making me laugh, and for laughing with me, while I struggled to figure out when to stop doing experiments. I would like to thank and acknowledge past members of the Gutierrez-Hartmann lab who have contributed to this work: undergraduate student workers Crystal Gomez, Cynthia Janku, Nicholas Gonyea, and high school student volunteer Joshua Choe. I would also like to acknowledge former lab members Karina Gomez and Adwitiya Kar.
I would like to thank the Integrated Physiology graduate program (formerly
Reproductive Sciences), for charming me into to making the move to Colorado. The students in the program, past and present, have constantly reminded me that being a graduate student is what you make of it, and they made it fun. I’d especially like to thank Linnea Schmidt and Sydney Coates for their friendship and support. I’d also like the thank Jennifer Thurston, Emily Dailey, and Deanne Sylvester, our program administrators, all three of whom have helped restore my sanity in times of utter panic too many times to count. I also want to acknowledge Jim McManaman, our program director, for his support and guidance.
I would like to thank and acknowledge my parents, Jim and Anne Booth, for instilling in me a very strong will which, although at times may manifest as stubbornness, has without a doubt provided me with the courage to always be the best person I can be. Even though they can’t read this, I have to thank my dogs for their ability to always make me smile, even after a long day of failed experiments. Lastly, I would like to give tremendous thanks to my husband, Alex, for his unwavering support and for always motivating me to be my best self. He has patiently and selflessly stood by my side for twelve of my twenty-four years as a student. No more degrees, honey; this really is the end – I promise!
TABLE OF CONTENTS CHAPTER
INTRODUCTION AND BACKGROUND ... 1
Pituitary Lactotropes And Prolactin Biology ... 1
Signaling Pathways Regulating Lactotrope Biology ... 3
Regulation of pituitary stem/progenitor cells and lactotrope ontogeny. ... 3
Signaling pathways regulating lactotrope homeostasis and physiological expansion. ... 7
Cyclic 3’-5’-Adenosine Monophosphate (cAMP) and Protein Kinase A (PKA) Signaling... 8
Mitogen-Activated Protein Kinase (MAPK/ERK) Signaling. ... 10
Phosphatidylinositol-3-Kinase (PI3K) Signaling. ... 12
Transforming Growth Factor β (TGFβ) Signaling. ... 14
Mutations in signaling pathways associated with prolactinoma and useful mouse models of pituitary adenoma. ... 15
Ras... 15
Menin. ... 17
Heparin Secretory Transforming (hst) Gene. ... 18
Pituitary Tumor Transforming Gene (PTTG). ... 18
Aryl Hydrocarbon Interacting Protein (AIP). ... 19
Guanine Nucleotide Activating Subunit (GNAS). ... 19
Unknown/Unidentified Mutations. ... 20
D2R and Dopamine-Agonist Intolerance/Resistance in Prolactinoma Patients ... 22
Scope of Thesis ... 24
MATERIALS AND METHODS ... 26
GH4T2 and GH4C1 cells. ... 26
293T and BOSC cells. ... 27
Pharmacological Inhibitors, EGF, and Cell Synchronization. ... 27
Transfection and Transduction Methods/Reagents ... 27
Doxycycline-inducible construct cloning and transfection. ... 27
Doxycycline-inducible GH4T2 clones. ... 28
Transient transfections. ... 29
Viral plasmids and transduction. ... 29
Protein Lysate Preparation, SDS-PAGE, and Western Blotting ... 30
Proliferation Assays ... 30
Clonogenicity Assays ... 31
Soft Agar Assays ... 32
Nude Mouse Xenografts ... 32
Immunohistochemistry ... 33
Cytotoxicity Assays ... 34
Luciferase Reporter Assays ... 35
ERK MEDIATES PITUITARY SOMATOLACTOTROPE DIFFERENTIATION ... 36
Introduction ... 36
Long-Term Activation of ERK In Vitro Promotes Lactotrope Differentiation and Reduces Tumorigenic Phenotype ... 38
Stable expression of constitutively active V12Ras increases PRL and reduces GH expression. ... 41
V12Ras expression does not increase proliferation but decreases tumorigenic characteristics. ... 44
Exogenous EGF Recapitulates Lactotrope Differentiation and Reduced Tumorigenic Phenotype Observed with V12Ras Expression ... 46
PRL:GH expression ratio is increased with EGF treatment in GH4T2 cells. ... 46 EGF does not affect proliferation but reduces 2D colony formation and
anchorage-independent growth of GH4T2 cells. ... 48 Doxycycline-Induced 3HA-V12Ras Expression Diminishes GH4T2 Cell
Xenograft Tumor Growth in Nude Mice and Promotes a Lactotrope Phenotype ... 48 Conclusions ... 52 PITUITARY SOMATOLACTOTROPES EVADE AN ONCOGENIC
RESPONSE TO RAS ... 58 Introduction ... 58 Stable Expression of a Dominant Negative N17Ras Construct In Vitro Reduces
GH4T2 Proliferation and Tumorigenic Phenotype, and Does Not Alter The
PRL:GH Expression Ratio ... 61 Menin Knockdown Increases GH4T2 Cell Proliferation, But Does Not Increase
Colony Formation ... 65 Ral is Responsible for One-Third of Ras-Mediated PRL Promoter Activation ... 67 Conclusions ... 72 PI3K DRIVES PITUITARY SOMATOLACTOTROPE PROLIFERATION
AND INHIBITS ERK-MEDIATED PROLACTIN PROMOTER ACTIVITY
VIA CDK4 ... 76 Introduction ... 76 Pharmacological Inhibition of PI3K/AKT/mTOR, but Not MEK, Diminishes
GH4T2 Proliferation and Tumorigenicity ... 78 Activation of PI3K Via Knockdown of PTEN or with Expression of
Constitutively Active PI3K Increases GH4T2 Proliferation but Does Not
Increase Tumorigenicity ... 84 Cabergoline Reduces Cell Viability Most Dramatically in D2L-Expressing
GH4C1 Cells ... 87 Dopamine Signals Through D2L to Inhibit the mTOR Effector s6K ... 91 PI3K Antagonizes Ras-Mediated PRL Promoter Activity and Is Blocked with
Palbociclib ... 92 Conclusions ... 97
ERK AND AKT/MTOR IN HUMAN PROLACTINOMA ... 103
Introduction ... 103
p-ERK and p-AKT Are Increased in Human Prolactinoma Compared to Normal Pituitary Tissue ... 105
Conclusions ... 110
DISCUSSION ... 114
Summary of Findings and Key Conclusions ... 114
Context-Specific Signaling Outcomes within Endocrine Cell Types ... 115
Ras/ERK and PI3K/AKT/mTOR Signaling in Pituitary Cells. ... 115
Lactotrope cells. ... 115
Somatotrope cells. ... 120
Gonadotrope cells. ... 120
Thyrotropes and corticotropes. ... 122
Unique Responses to Ras/ERK and PI3K/AKT/mTOR in Endocrine Cell Types. ... 122
Non-classical responses to Ras. ... 124
Cell cycle control. ... 126
Pro-tumorigenic signaling in pituitary and neuroendocrine tissues. ... 127
Counterregulation and signaling pathway balance. ... 128
Model and Future Directions ... 131
Significance of Research ... 133
REFERENCES ... 135
APPENDIX A. Knockdown of p21 or p27 Increases GH4T2 Cell Proliferation and Modestly Increases Colony Formation ... 154
LIST OF TABLES TABLE
I.1 Clinically Identified Mutations in Human Prolactinoma ... 16 I.2 Animal Models of Prolactinoma ... 21 II.1 List of Antibodies ... 31 VII.1 Physiological responses to activated Ras/ERK and PI3K/AKT/mTOR
LIST OF FIGURES FIGURE
I.1 Embryonic ontogeny and Pit-1 pituitary cell lineage. ... 4 I.2 Adult pituitary stem cells, facultative cell expansion, and pituitary
tumorigenesis. . ... 6 I.3 Lactotrope signaling pathways central to pituitary cell proliferation,
tumorigenesis, and proximal rat prolactin (PRL) promoter activation. ... 10 I.4 The Ras cycle. ... 17 I.5 ERK 1/2 and AKT phosphorylation profiles in pituitaries of 4-month-old
mice. ... 23 III.1 Dox-dependent 3HA-V12Ras expression is stable and tightly regulated. ... 40 III.2 Stable expression of V12Ras increases PRL and reduces GH expression. ... 42 III.3 MKP-2 expression fluctuates with doxycycline addition, does not increase
with EGF treatment. ... 43 III.4 V12Ras expression does not increase GH4T2 cell proliferation but
decreases tumorigenic characteristics. ... 45 III.5 EGF increases PRL to GH expression ratio. ... 47 III.6 EGF does not influence proliferation but reduces 2D colony formation and
anchorage-independent growth of GH4T2 cells. ... 49 III.7 Dox-induced 3HA-V12Ras expression diminishes GH4T2 cell xenograft
tumor growth and promotes a lactotrope phenotype. ... 51 IV.1 Stable expression of N17Ras reduces GH4T2 cell proliferation and colony
formation, and does not change the PRL:GH expression ratio. ... 63 IV.2 Menin knockdown increases GH4T2 cell proliferation but is not sufficient
to increase colony formation. ... 66 IV.3 Ral is responsible for one-third of ras-mediated PRL promoter activation. ... 68 IV.4 Ras effectors RalGEF and ERK, but not PI3K, induce morphological
changes in GH4T2 cells. ... 71 V.1 Treatment of GH4T2 cells with pharmacological inhibitors. . ... 80
V.2 PI3K/AKT/mTOR, not ERK, regulates GH4T2 cell proliferation, colony
formation, and anchorage-independent growth. ... 83 V.3 Stable knockdown of PTEN increases GH4T2 cell proliferation but does
not increase colony formation. ... 86 V.4 Stable expression of constitutively active PI3K increases GH4T2 cell
proliferation but does not increase colony formation. ... 88 V.5 CAB signals through D2L and D2S to reduce GH4C1 cell proliferation. ... 90 V.6 Cell cycle progression inhibits PRL promoter activity via Cdk4/Cyclin
D1-dependent mechanism. ... 93 V.7 PI3K activation reduces ERK-stimulated rPRL promoter activity via
CDK4. . ... 96 V.8 EGF stimulation of p-ERK is greater in the presence of a PI3K inhibitor. . ... 101 VI.1 p-ERK is increased in human prolactinoma in sample set A. ... 106 VI.2 The balance of p-ERK and p-AKT activation is dysregulated in human
prolactinoma in Sample Set B., ... 108 VII.1 Simplified schematic of classical Ras/ERK and PI3K/AKT/mTOR
signaling in non-endocrine cells. ... 116 VII.2 Model of somatolactotrope/lactotrope signaling pathways contributing to
cell differentiation, cell proliferation, and tumorigenesis. ... 118 VII.3 Schematic of Ras/ERK and PI3K/AKT/mTOR pathway crosstalk. AKT
phosphorylates and inhibits c-Raf and b-Raf, and PDK1 inhibits MEK. ... 129 AP.1 Knockdown of p21 or p27 increases GH4T2 cell proliferation and colony
formation. ... 155 AP.2 High doses of cabergoline reduce cell viability in a D2R-independent
LIST OF ABBREVIATIONS ACTH adrenocorticotropic hormone
AHR aryl hydrocarbon receptor
AIP Aryl Hydrocarbon Interacting Protein AKT protein kinase B; aka PKB
AMC Anschutz Medical Campus ANR anterior neural ridge BTE basal transcription element BMP bone morphogenic protein BSA bovine serum albumin CAB cabergoline
cAMP cyclic 3’-5’-adenosine monophosphate CDK cyclin-dependent kinase
CMVp cytomegalovirus promoter CRH corticotropic releasing hormone D2L human long isoform of D2R D2R dopamine receptor D2 D2S human short isoform of D2R
DA dopamine agonist
DAB diaminobenzidine ddH20 double distilled water DHEA dehydroepiandrosterone
DMEM Dulbecco's Modified Eagle's Medium
Dox doxycycline
DTT dithiothreitol
EDTA ethylenediaminetetraacetic acid EGF epidermal growth factor
ER estrogen receptor
ERK extracellular signal-regulated kinase EtOH ethanol
ETS1 E26 transformation-specific transcripton factor F2F FP2 repressor site
FBS fetal bovine serum FGF fibroblast growth factor
FP footprint
FSH follicle-stimulating hormone GABP GA-binding protein
GAP GTPase-activating protein GDP guanosine diphosphate
GEF guanine nucleotide exchange factors
GH growth hormone
GHRH growth hormone-releasing hormone GNAS guanine nucleotide activating subunit GnRH gonadotropin-releasing hormone GPCR G-protein-coupled receptor
GRK GPCR kinases
GTP guanosine triphosphate GTPase guanosine triphosphatase
H3 histone 3
hCG human chorionic gonadotropin
HPG hypothalamic-pituitary-gonadal (axis) HRP horseradish peroxidase
HST heparin secretory transforming gene IRES internal ribosome entry site
kDa kilodalton
LH luteinizing hormone
Lhx LIM homeodomain transcription factor MAPK mitogen-activated protein kinase
MAPKK mitogen-activated protein kinase kinase MAPKKK mitogen-activated protein kinase kinase kinase
MEK the mitogen activated protein kinase kinase of the ERK cascade MEN1 multiple endocrine neoplasia type I
MEN4 multiple endocrine neoplasia type V MKP MAPK phosphatase
mTOR mammalian target of rapamycin
mTORC mammalian target of rapamycin complex NEAA non-essential amino acids
PBS phosphate-buffered saline PDE4A5 phosphodiesterase 4A5 PFA paraformaldehyde PH plekstrin homology
PI3K phosphatidylinositol-3-kinase PI3KCA catalytic subunit of PI3K
PIP2 phosphatidylinositol 4,5-bisphosphate; aka PI(4,5)P2
PIP3 phosphatidylinositol 4,5-trisphosphate; aka PI(4,5)P3
Pit-1 POU homeodomain protein; aka growth hormone factor 1 (GHF-1) PKA protein kinase A
PKC protein kinase C PLC phospholipase C
PMSF phenylmethylsulfonyl fluoride
PRL prolactin
PRL-R prolactin receptor
Prop-1 paired-like homeodomain factor 1; aka prophet of Pit-1 pSPC pituitary postnatal stem/progenitor cells
PTEN phosphatase and tensor homologue PTTG pituitary tumor transforming gene Rb retinoblastoma
RRE Ras response element
SDS-PAGE sodium dodecyl sulfate-polyacrylamide gel electrophoresis Shh sonic hedgehog
TET tetracycline; aka doxycycline TGF transforming growth factor
TGFb-R transforming growth factor b-receptor TIDA tuberoinfundibular (neurons)
TR thyroid hormone receptor TRE tetracycline response element TRH thyrotropin-releasing hormone TSH thyroid-stimulating hormone VIP vasoactive intestinal peptide
WT wild type
CHAPTER I
INTRODUCTION AND BACKGROUND1 Pituitary Lactotropes And Prolactin Biology
Prolactin (PRL) is a 23 kDa polypeptide hormone that is a member of the growth hormone (GH) family and is primarily synthesized and secreted from lactotrope cells of the anterior pituitary gland. In mammals, PRL acts at the mammary gland to promote growth and development, milk synthesis, and maintenance of milk secretion (1). Knockout of PRL or PRL receptor genes in mice results in impaired growth and development of the mammary gland and absence of milk production (2, 3). The strongest stimulus for PRL secretion from lactotrope cells is suckling, with the duration and intensity of the stimulus corresponding to the amount of PRL secreted into the blood (1, 4, 5).
In addition to its classical actions on the mammary gland, PRL also influences many other physiological systems. The PRL receptor is expressed in the mammary gland, gonads, uterus, brain, pituitary gland, adrenal gland, lung, heart, liver, skeletal muscle, skin, and lymphocytes. Elevated PRL levels act at the gonads to decrease the sensitivity of follicle stimulating hormone (FSH) and luteinizing hormone (LH) receptors. Furthermore,
circulating PRL attenuates pulsatile secretion of gonadotropin releasing hormone (GnRH) from the hypothalamus, reducing LH and FSH secretion from the anterior pituitary gland (6). As a result, increased levels of PRL cause reduced secretion of and sensitivity to LH and FSH, leading to suppression of ovulation in women and hypogonadism in men. During
1
Portions of this chapter were previously published in the following publication and are reprinted here with permission: Booth, A. et al. “Signaling pathways regulating pituitary lactotrope homeostasis and tumorigenesis.” In: Advances in Experimental Medicine and Biology, (2015).
pregnancy, elevated serum PRL has effects that extend beyond the reproductive system. At the adrenal gland, PRL increases androgen and dehydroepiandrosterone (DHEA)
steroidogenesis, and also reduces cortisol and aldosterone secretion (6). In the liver, PRL increases lipoprotein lipase activity in hepatocytes and increases bile secretion. PRL has osmoregulatory effects in the kidney, reducing renal sodium and potassium excretion. PRL also increases sodium and chloride excretion in sweat, and salt and water absorption in the intestine. Lastly, PRL influences the immune system by inducing proliferation of
lymphocytes (6).
As PRL is involved in many different physiological systems, signaling pathways are critical for regulating lactotrope biology from humans to rodents. Pituitary lactotropes have a high basal PRL secretory activity. To maintain PRL homeostasis, tonic inhibition by
dopamine acting via the D2 receptor (D2R) is required to limit PRL production and secretion, lactotrope proliferation, and growth of PRL-secreting adenomas (7–13). During pregnancy and lactation, dopaminergic inhibition is diminished by estradiol, allowing local growth factors from folliculostellate support cells to stimulate lactotropes, promoting
lactotrope hyperplasia and a doubling in pituitary size (7, 14–16). Circulating PRL levels are elevated during pregnancy and lactation, creating a state of physiological hyperprolactinemia. Dysregulation of the signaling pathways that govern lactotrope biology contributes to
tumorigenesis of PRL-secreting adenomas, or prolactinomas (16–18), leading to a state of pathological hyperprolactinemia. Prolactinomas cause infertility and osteoporosis due to hypogonadism; women often present with amenorrhea and galactorrhea, whereas men develop gynecomastia and erectile dysfunction (19–21).
Signaling Pathways Regulating Lactotrope Biology Regulation of pituitary stem/progenitor cells and lactotrope ontogeny.
During embryogenesis, the pituitary first develops from the anterior neural ridge (ANR) of the neural plate. Pituitary organogenesis begins at embryonic day 8.5 (E8.5) with the formation of Rathke’s pouch. The ventral diencephalon, which will ultimately become the hypothalamus, develops from neural plate cells posterior to the ANR (22). The process of pituitary development is dependent upon the homeobox gene Tift1, as well as fibroblast growth factor 8 (FGF8) and bone morphogenic protein 4 (BMP4) signaling from the ventral diencephalon. Knockout of Titf1 results in pituitary aplasia (23). FGF8 signaling and the resulting expression of the LIM homeodomain transcription factor Lhx3 is required for pituitary development to progress beyond the formation of Rathke’s pouch (22). Without BMP signaling from the ventral diencephalon, pituitary development does not progress beyond E10. Sonic hedgehog (Shh) signaling is required for pituitary patterning and
proliferation after E10. Shh works in unison with FGF8 to maintain Lhx3 expression, and it also induces BMP2 expression in the ventral pouch [(21); Figure I.1].
Transient, intrinsic BMP2 and Wnt4 signaling gradients in the developing pituitary gland promote proliferation and establish a pattern that determines localization of specific pituitary cell types (22). Somatotrope and lactotrope cells arise within the caudomedial region of the developing pituitary gland. Before each cell type can progress beyond initial proliferation and localization, expression of cell-fate-specific transcription factors is required. For lactotropes, somatotropes, and thyrotropes, expression of paired-like homeodomain factor 1 (Prop-1) and Pit-1 POU homeodomain protein is required for terminal differentiation (Figure I.1). Prop-1 is required for Pit-1 activation, and is expressed only in the developing
pituitary gland. Deficiency in Prop-1 leads to near complete loss of somatotrope, lactotrope, and thyrotrope cells (24). After E17.5, cells in the Pit-1 lineage exhibit permanent
2
Figure I.1was previously published in the following publication and is reprinted here with permission: Booth, A. et al. “Signaling pathways regulating pituitary lactotrope homeostasis
Figure I.1: Embryonic ontogeny and Pit-1 pituitary cell lineage.2
All hormone secreting cells in the anterior pituitary gland originate from pituitary stem cells. During embryonic development, FGF8 and BMP4 from the
hypothalamus stimulate LIM homeodomain transcription factors (Lhx) 3 and 4. Intrinsic gradient signaling of Wnt4 and BMP2, and expression of the Prop-1 transcription factor, play key roles in determination of pituitary cell fate and
localization. Thyrotropes, somatotropes, and lactotropes are derived from the Pit-1 lineage. Hormone secretion from thyrotropes, somatotropes, and lactotropes is regulated in part by hypothalamic TRH, GHRH, and dopamine, respectively, and the Pit-1 transcription factor is required for cell-specific determination. In lactotropes, Ets1 and ER are also required for prolactin (PRL) production. In somatotropes, the thyroid hormone receptor (TR) is required for growth hormone (GH) secretion. In rats, a somatolactotrope precursor cell gives rise to PRL secreting lactotropes and GH secreting somatotropes. The contribution of such a precursor cell is well described in rats, but has less of contribution in mice. The existence of a somatolactotrope cell in humans, as well as the possibility that lactotropes and somatotropes may give rise to one another in response to physiological demand, has yet to be confirmed in humans.
autonomous commitment and cannot be converted to alternative fates (22). Hormone secretion from differentiated thyrotropes, somatotropes, and lactotropes is regulated by hypothalamic thyroid-releasing hormone (TRH), GH-releasing hormone (GHRH), and dopamine, respectively (Figure I.1).
The Pit-1 transcription factor binds to promoter regions of GH and PRL genes, and is required for their activation. Pit-1 can associate with coactivators and corepressors, and the Pit-1 binding partners required to activate PRL versus GH gene transcription are involved in activation of signaling pathways. Ras-dependent activation of Ets1/Pit-1 synergy results in PRL gene transcription (25–27). Pit-1 is necessary for cell-specific determination, but it is not sufficient; for lactotropes, estrogen receptor (ER), and Ets transcription factors are also required (26).
Until recently, the dogma was that the embryonic ontogeny pathways were also responsible for facultative responses to meet increased pituitary hormonal demand during periods of physiological stress, including lactotrope expansion during pregnancy. However, the identification of pituitary postnatal stem/progenitor cells (pSPCs) within the past decade has challenged this dogma. A niche containing pSPCs exists into adulthood in the pituitary gland and is the likely source of facultative organ expansions driven by upstream tropic hormones and stromal growth factors in response to increased physiological demand (Figure I.2). Cells from the anterior pituitary gland are capable of forming ‘pituispheres,’ and these cells segregate into the ‘side population.’ This side population contains 1-5% of total
Figure I.2: Adult pituitary stem cells, facultative cell expansion, and pituitary tumorigenesis.3
A niche containing pituitary postnatal stem/progenitor cells (pSPCs) exists into adulthood in the pituitary gland and is the likely source of facultative organ expansions that occur in response to increased physiological demand.
Folliculostellate support cells provide growth factors to stimulate pSPCs, and Notch signaling regulates stem cell homeostasis. The pSPCs express SSEA-4, Oct4, Sox2, GFRa2, Sca1, nestin, Prop-1, and Lhx-3, but do not express embryonic pituitary stem cell makers Hesx-1 or Lhx-4. Transit-amplifying (TAC) cells express Sox-9 and Sca1, but not Sox-2, and proliferate more rapidly than pSPCs to allow for prompt cellular expansions. The precise signaling events that regulate these expansions remain unknown. Expression of cell-specific transcription factors is required for hormone secretion from each cell type. Hormone secretion from differentiated gonadotropes, somatotropes, lactotropes, thyrotropes, and corticotropes is regulated by GnRH, GHRH, dopamine, TRH, and CRH, respectively.
3
Figure I.2 was previously published in the following publication and is reprinted here with permission: Booth, A. et al. “Signaling pathways regulating pituitary lactotrope homeostasis and tumorigenesis.” In: Advances in Experimental Medicine and Biology (2015).
analysis of cells of the side population fraction revealed high expression levels of Sca1, as well as expression of other stem cell markers such as Oct4, nanog, nestin, CD133, and Bmi1 (28). A few years later, three separate studies reported the existence of stem cells in the pituitary gland (29–31). Together, these studies reveal that the periluminal pSPCs express SSEA4, Oct4, Sox2, GFRa2, Sca1, nestin, Prop-1, Lhx3, E-cadherin, and cytokeratins 8 and 18. Importantly, pSPC cells do not express embryonic pituitary stem cell makers Hesx1 and Lhx4, distinguishing these cells from embryonic pituitary stem cells. Notch signaling functions in pSPC homeostasis (32). One study also identified putative transit-amplifying (TAC) cells, which express Sox9, low levels of Sca1, and do not express Sox2 (29). The TAC cells are considered to be capable of rapid proliferation, compared to the slow asymmetric doubling of pSPCs, suggesting that these cells act as an important precursor, allowing for cellular expansions into differentiated cell types as needed to meet adaptive responses (Figure I.2). However, the signaling mechanisms governing these neuroendocrine expansions, the precise role of pSPCs in these adaptive responses, and whether a perturbation in the expansion process leads to prolactinoma tumor formation, all remain unknown (33). Signaling pathways regulating lactotrope homeostasis and physiological expansion.
During pregnancy, the mammalian pituitary gland doubles in size, primarily due to expansion of PRL-producing lactotrope cells (34). However, there is a great deal of debate as to whether this doubling in size is a result of lactotrope hypertrophy or hyperplasia. For obvious reasons, the availability of human pituitary tissue from pregnant women is scarce, and as such many questions remain concerning the morphological changes in the human pituitary gland during pregnancy. Studies in rodents are useful, but are also challenging because human and rodent pituitary physiology is not entirely analogous. In rats, bi-hormonal
somatolactotrope precursor cells retain plasticity, allowing for rapid cell differentiation and expansion in response to hormonal need. Somatolactotropes differentiate into lactotropes during pregnancy and into somatotropes in response to exercise (35–38). No such precursor cell has been explicitly identified in humans, and therefore the use of rodent models to study the pituitary during pregnancy becomes convoluted. Additionally, our understanding of the mechanism whereby expanded lactotropes return to the pre-pregnant state remains unclear. The role of apoptosis, senescence, or simply diminished cell synthesis activity in this process is not understood.
The doubling of the lactotrope cell population during pregnancy illustrates that there is an immense capacity for expansion within these cells. As such, signaling pathways within lactotrope cells are primed to induce rapid cellular expansion. With so much capacity for expansion, there is an increased risk that problems may occur and result in uncontrolled growth. It is very likely that the signaling pathways that are in place to allow lactotropes to undergo recurrent expansions also prime the cell for tumorigenic responses if one or more oncogenic mutations are present. Here, I will introduce and discuss the role of these signaling pathways. I will focus on the pathways that are also known to be involved in mechanisms of tumorigenesis.
Cyclic 3’-5’-Adenosine Monophosphate (cAMP) and Protein Kinase A (PKA) Signaling. cAMP is a second messenger that regulates a diverse set of cellular events. Upon stimulation from an extracellular ligand, G-protein-coupled receptors (GPCRs) become activated and stimulate an associated G-protein. The resulting downstream signaling events depend upon the alpha (α) subunit of the G-protein. Gαs proteins activate adenylate cyclase,
intracellular cAMP and activation of cAMP-dependent PKA. Activation of the cAMP/PKA pathway stimulates the rPRL promoter via the Pit-1 binding sites of footprint I (FPI) and FPIII [(37–39); Figure I.3]. Gαi proteins inhibit adenylate cyclase activity, resulting in
diminished intracellular cAMP levels and reduced PKA activity [(40); Figure I.3].
One of the most studied in the classic regulatory pathways of lactotrope homeostasis is dopaminergic inhibition of lactotrope expansion and PRL secretion. In homeostatic conditions, the secretion of PRL from pituitary lactotropes is inhibited by dopamine. Dopamine binds to the D2R receptor, which is coupled to a Gαi protein, and thus inhibits
intracellular cAMP accumulation (7). Without cAMP, the catalytic subunit of PKA remains sequestered by the regulatory subunit, and cytoplasmic and nuclear target proteins are not phosphorylated, preventing activation of PRL gene transcription and PRL release from the lactotrope cell (Figure I.3). Another level of homeostatic regulation exists within a short feedback loop between the pituitary and hypothalamus. PRL can bind at the prolactin receptor (PRL-R) on hypothalamic tuberoinfundibular (TIDA) neurons, increasing dopaminergic release in response to both acute and chronic increases in PRL (43), and further inhibiting cAMP and PKA signaling in lactotrope cells. However, TIDA neurons become refractory when exposed to prolonged hyperprolactinemia during pregnancy or with prolactinoma.
During pregnancy, placental human chorionic gonadotropin (hCG) stimulates
production of ovarian estradiol. In response to estradiol, hypothalamic tyrosine hydroxylase, the enzyme that catalyzes the hydroxylation of tyrosine to produce dopamine, is
lactating mother, dopaminergic inhibition is relieved and PRL is secreted into the blood (47, 48).
Mitogen-Activated Protein Kinase (MAPK/ERK) Signaling.
The MAPK signaling pathways connect a wide variety of extracellular signals to intracellular outcomes, including proliferation, differentiation, and apoptosis. The MAPK pathways consist of a three-level kinase cascade, where a MAPK is phosphorylated by a
4
Figure I.3 was previously published in the following publication and is reprinted here with permission: Booth, A. et al. “Signaling pathways regulating pituitary lactotrope homeostasis
Figure I.3: Lactotrope signaling pathways central to pituitary cell proliferation, tumorigenesis, and proximal rat prolactin (PRL) promoter activation.4
Growth factor receptor tyrosine kinase (RTK) and GPCR signaling pathways regulating lactotrope homeostasis and rat PRL (rPRL) promoter activation are depicted here. The proximal rPRL promoter, with Pit-1 binding sites (FPI, III, IV), Ets1 and GABP binding sites, and the F2F ubiquitous factor binding site are also shown. The MAPK depicted in this figure is ERK. For further details, see the review by Booth, et al. (46).
mitogen-activated protein kinase kinase (MAPKK), which must first be phosphorylated by a mitogen-activated protein kinase kinase kinase (MAPKKK). The ERK pathway is the best studied of the MAPK signaling pathways, as dysregulation of ERK signaling is associated with many human cancers. In the ERK signaling pathway, extracellular growth factors and mitogens bind to receptor tyrosine kinases, activating the GTPase Ras, which leads to
recruitment and activation of the MAPKKK Raf, phosphorylation of the MAPKK MEK, and stimulation of ERK, which ultimately results in phosphorylation of a wide variety of effector proteins including other kinases, phosphatases, and transcription factors [(48); Figure I.3].
The duration of ERK signaling is critical in dictating cellular response (50). Estrogen-induced PRL expression is ERK-regulated (51), and importantly, the estrogenic effect on lactotropes during pregnancy is persistent, lasting for many months. Estrogen stimulates folliculostellate support cells to produce growth factors such as fibroblast growth factor (FGF) that act via the ERK pathway (14). The Ras/ERK pathway regulates the PRL promoter via a composite Ets1/Pit-1 site (25, 26, 41, 52), and via a basal transcription element (BTE) [(55, 56); Figure I.3]. The precise role of ERK signaling in lactotrope proliferation versus differentiation has been somewhat controversial. In vitro studies using rat pituitary
somatolactotrope or lactotrope cell lines have shown that short-term (24–96 h) ERK pathway activation mediates cellular proliferation (14, 55, 56). By contrast, long-term treatment of GH3 or GH4 rat pituitary tumor cells over 4–7 days with epidermal growth factor (EGF), fibroblast growth factor4 (FGF4), or thyrotropin-releasing hormone (TRH) results in decreased GH4 cell proliferation (57–61). Furthermore, a persistent pattern of p-ERK activation has been shown to play a pivotal role in cellular differentiation in other endocrine tumors including thyroid carcinoma and pheochromocytoma (62, 63). The inconsistency in
the reported effects of ERK on lactotrope proliferation or differentiation suggests that the duration of ERK activation determines the response of lactotrope cells.
The specific role of ERK signaling in durable lactotrope proliferation and
differentiation, and whether activated p-ERK is sufficient for lactotrope proliferation and tumor formation remains unknown. Ras mutations and persistently activated p-ERK are found in human tumors (64, 65), including prolactinomas and other pituitary tumors (18, 66, 67). Uncontrolled activation of growth factor signaling pathways, such as the Ras/ERK pathway, results in lactotrope hyperplasia with very delayed adenoma formation in transgenic mice (17, 68). Transforming growth factor alpha (TGFα) activates the epidermal growth factor receptor (EGFR) to stimulate the Ras/Raf/MEK/ERK pathway. TGFα is expressed in lactotropes, and upon overexpression promotes proliferation, suggesting a role for TGFα and ERK signaling in prolactinoma formation (69).
Phosphatidylinositol-3-Kinase (PI3K) Signaling.
The PI3K family of lipid kinases functions to activate signaling cascades that regulate diverse intracellular processes such as cell survival, cell cycle progression, and cell growth. Extracellular growth factors bind to receptor tyrosine kinases, which are associated with an intracellular PI3K. When growth factor binds, the receptor is auto-phosphorylated and PI3K binds to the receptor. The catalytic subunit of PI3K is allosterically activated, resulting in the conversion of phosphatidylinositol 4,5-bisphosphate [PI(4,5)P2 or PIP2] to the second
messenger phosphatidylinositol 4,5-trisphosphate [PI(4,5)P3 or PIP3]. PIP3 anchors AKT
near the membrane via its plekstrin homology (PH) domain, where AKT is phosphorylated by 3′phosphoinositide-dependent kinase 1 (PDK1), which also has a PH domain. AKT is also phosphorylated by the mammalian target of rapamycin (mTOR) 2 complex, mTORC2. Once
phosphorylated, AKT activates and inhibits several targets to ultimately influence cell survival, growth, and proliferation. Phosphatase and tensor homologue (PTEN), a PIP3 phosphatase, can dephosphorylate PIP3 to negatively regulate PI3K/AKT signaling (70).
Constitutive activation of AKT promotes GH3 somatolactotrope cell proliferation, and the proliferative effects of AKT are diminished by the mTOR inhibitor rapamycin as a result of G1 growth arrest (71). Inhibition of AKT also results in decreased GH3
somatolactotrope cell viability, potentially due to decreased activity of NF-κB, a transcription factor that controls cell survival (72). Pharmacological inhibition of PI3K or AKT in GH4C1 somatolactotrope cells results in increased phosphorylation of ERK1/2, as well as Raf1 kinase activity (73). However, these effects of PI3K/AKT inhibition were diminished upon co-treatment with IGF1 (73), suggesting that the ERK and PI3K pathways regulate lactotrope physiology through a delicate balance of intracellular signaling. Preclinical data suggest that increased Ras/ERK and/or increased PI3K/AKT pathway activity may contribute to pituitary tumorigenesis (74).
As discussed previously, activating mutations in the Ras/ERK signaling pathway are not sufficient to promote tumorigenesis of lactotrope cells. Transgenic mice studies targeting growth factors (nerve growth factor, TGFα, and FGF-R4) to pituitary lactotropes resulted in early hyperplasia, occurring within approximately 4 months, followed by delayed adenoma formation at approximately 10 months, but these pituitary cells were resistant to true
carcinogenesis (69, 75–77). Activating mutations in an additional pathway, often PI3K, must also occur to promote tumorigenesis (68, 78–80). Transgenic mice studies targeting
oncogenic Ras to thyroid and ovarian endocrine cells show that activated ERK is necessary, but not sufficient, to mediate proliferative and tumorigenic responses, and that the PI3K
pathway is essential (81–84). These findings support the notion that the ERK and PI3K signaling pathways work in unison to drive lactotrope differentiation and hyperplasia during pregnancy or prolactinoma formation.
Transforming Growth Factor β (TGFβ) Signaling.
TGFβ signaling is important in a wide variety of cellular events, including
proliferation, differentiation, and apoptosis. The TGFβ ligand binds to the heterodimerized TGFβ receptor (TGFβ-R), consisting of type I and type II receptor serine/threonine kinases. Upon dimerization, the type II receptor phosphorylates the kinase domain of the type I receptor, ultimately resulting in the phosphorylation of SMAD effector proteins. Activated SMAD protein complexes are translocated to the nucleus and regulate transcription of target genes (85).
Under basal conditions, TGFβ1 acts on lactotropes to inhibit the effects of estradiol on cell proliferation (12, 86). Dopamine stimulates TGFβ1 secretion and mRNA expression, resulting in inhibited cell proliferation, suggesting that TGFβ1 mediates the inhibitory action of dopamine on lactotropes (13). TGFβ1 also inhibits activity of the rat PRL promoter in GH4T2 cells (87). Lactotropes do not express the TGFβ2 isoform, and the effect of TGFβ3 on lactotrope proliferation is negligible in the absence of high levels of estrogen (15). Activin, a member of the TGFβ family, negatively regulates PRL production in lactotropes by repressing transcription of Pit-1. Activin also stimulates phosphorylation of SMAD3, which interacts with the tumor suppressor menin to inhibit PRL transcription [(90); Figure I.3].
However, upon exposure to increased estrogen concentration, TGFβ3 indirectly increases lactotrope proliferation by simulating production of growth factors from
folliculostellate cells, suggesting that TGFβ3 mediates the mitogenic effects of estrogen (15). Furthermore, this reveals that TGFβ1 and TGFβ3 have opposing actions on lactotrope cell proliferation (86). Together these data suggest that a balance of TGFβ signaling is required for lactotrope homeostasis, and a substantial shift in this balance in favor of TGFβ3 is required for physiological lactotrope proliferation in pregnancy and lactation.
Mutations in signaling pathways associated with prolactinoma and useful mouse models of pituitary adenoma.
Neuroendocrine tumors are characterized by excessive secretion of tumor-derived hormone(s), which then inhibit upstream tropic hormones. Despite reductions in tropic hormone levels, the tumor continues to secrete hormone, creating a severely blunted
endocrine feedback mechanism. Prolactinomas are the most common type of pituitary tumor. These tumors secrete excessive amounts of PRL, leading to hypogonadism, infertility, as well as tumor mass effects (19, 20). In this section, I will review mutations in signaling pathways that have been clinically identified in prolactinoma (Table I.1). While each of these genetic mutations accounts for only small proportion of clinical prolactinomas, they provide valuable insight into which signaling pathways are most important in regulation of lactotrope homeostasis.
Ras.
Ras is a small GTPase protein that activates signaling pathways that regulate cellular processes such as proliferation, differentiation, and survival, including the ERK signaling pathway. Extracellular growth factors and mitogens bind to receptor tyrosine kinases (RTKs) and G-protein coupled receptors (GPCRs), resulting in the activation of the GTPase Ras. Ras
is a timed-effector that cycles between a GTP-loaded “on”-state, and a GDP-bound “off”-state. Guanine nucleotide exchange factors (GEFs) promote the exchange of GDP for GTP to facilitate Ras activation, and GTPase-activating proteins (GAPs) accelerate the hydrolysis of GTP resulting in Ras inhibition. See Figure I.4 for further detail.
In humans, there are three Ras genes: HRAS, NRAS, and KRAS. Oncogenic mutations at G12, G13, and Q61 allow Ras to remain in its GTP-bound state, resulting in constitutive activation of Ras signaling. These oncogenic Ras mutations are found in human cancers, with mutations most commonly occurring in KRAS. One unusually invasive human prolactinoma was identified to have an H-Ras G12V point mutation, and was lethal (89). Despite the frequency with which Ras is mutated in human cancer, Ras mutations are rare in pituitary adenomas (90–93).
5
Table I.1 was previously published in following publication and is reprinted here with permission: Booth, A. et al. “Signaling pathways regulating pituitary lactotrope homeostasis and tumorigenesis.” In: Advances in Experimental Medicine and Biology (2015).
Figure I.4: The Ras cycle.6
Following growth factor-mediated auto-phosphorylation of RTKs, the SH2 domain of the GRB2 adaptor protein binds to phosphotyrosine residues on the receptor. Son of Sevenless (SOS), a RasGEF, is then recruited by the SH3 domain of GRB2. SOS facilitates GTP loading of Ras, resulting in Ras activation. Active, GTP-bound Ras stimulates the RAF/MEK/ERK signaling pathway (94).
Menin.
Menin is a tumor suppressor protein that regulates transcription of cyclin-dependent kinase inhibitors such as p18 and p27 by promoting histone methylation (95). Menin serves to regulate pregnancy-associated islet β-cell expansion (96), suggesting a potential role of menin in regulating pSPC-mediated expansions during pregnancy, or prolactinoma
tumorigenesis. Menin is mutated in multiple endocrine neoplasia type I (MEN1) ), a disease characterized by tumors of endocrine tissues including the anterior pituitary, pancreatic islet
6
Figure I.4 was modified from the following publication and is printed here with permission: Smith, G et al. "Activating K-Ras Mutations Outwith ‘Hotspot’ Codons in Sporadic
Colorectal Tumours – Implications For Personalised Cancer Medicine.” Br. J. Cancer sos
cells, and parathyroid glands (97). Menin-null mice develop late-onset pituitary and β-cell tumorigenesis (98, 99). An inactivating mutation on chromosome 11q13, the site of the menin gene, has been reported in sporadic human prolactinoma (90), and 60 % of MEN1-associated pituitary tumors secrete PRL (100). However, a separate study reported that somatic menin mutations do not significantly contribute to prolactinoma tumorigenesis (101), suggesting that mutations in other genes may be necessary for prolactinoma formation. Heparin Secretory Transforming (hst) Gene.
The hst gene was originally identified to function as a transforming gene in malignant stomach cancers (102), and encodes for fibroblast growth factor 4 (FGF4). Expression of hst mRNA was later identified in human prolactinomas (103), and has been shown to be a marker of invasive prolactinoma (104). Overexpression of hst in rat lactotropes results in increased FGF4 production, as well as increased cell proliferation (104).
Pituitary Tumor Transforming Gene (PTTG).
PTTG is found in all classes of human pituitary adenomas, including prolactinoma. PTTG is expressed at low levels in normal human tissues, but shows increased expression in many human tumors and malignant cell lines (105). PTTG functions to regulate the
separation of sister chromatids during mitosis (92), and has been shown to regulate cell division and survival in endocrine tumors (18). PTTG was first isolated from rat
GH-secreting adenoma cells, and has been shown to be induced by estrogen and stimulate FGF2 signaling, resulting in prolactinoma tumor formation and progression in rats (18). Expression of PTTG is associated with lactotrope hyperplasia, angiogenesis, and prolactinoma
However, as of yet, a clear correlation between PTTG and tumorigenesis in human adenomas remains unclear (92).
Aryl Hydrocarbon Interacting Protein (AIP).
AIP associates with the cytoplasmic aryl hydrocarbon receptor (AHR), which is a transcription factor that interacts with cell cycle regulators such as retinoblastoma protein (Rb). AIP directly interacts with AHR to regulate its subcellular localization and nuclear cytoplasmic shuttling. AIP also regulates the localization and activity of phosphodiesterase 4A5 (PDE4A5), an enzyme responsible for the hydrolysis of intracellular cAMP. Mutations in AIP can alter the interactions with AHR and PDE4A5, providing a potential role for AIP to regulate signaling pathways that control tumorigenesis (106). However, the precise mechanisms by which AIP acts as a tumor suppressor in pituitary tumorigenesis have not been specifically identified. Germline mutations in AIP have been reported in some familial types of pituitary adenoma, including prolactinomas (106–108). AIP is considered a pituitary adenoma predisposition (PAP) gene (106). Many patients with mutations in AIP have
pituitary adenomas that secrete both GH and PRL (106), underscoring the shared ontogeny of pituitary lactotropes and somatotropes.
Guanine Nucleotide Activating Subunit (GNAS).
Gain-of-function somatic mutations typically occur in GPCR genes expressed in a tissue-restricted manner, and can lead to neuroendocrine adenoma formation and glandular hyperfunction. The stimulatory G protein, Gαs, is a product of the GNAS gene and regulates
activation of adenylate cyclase to produce intracellular cAMP. Activating mutations in GNAS, resulting in the expression of the gsp oncogene, are associated with somatotrope growth as well as the development of PRL and GH co-secreting adenomas in McCune–
Albright syndrome (109). An invasive prolactinoma that was resistant to DA was observed to transition into a GH-secreting adenoma while simultaneously acquiring a de novo mutation in GNAS (110).
Unknown/Unidentified Mutations.
The aforementioned mutations have been identified clinically in humans. There are many more candidate genes that have been shown to have the potential to promote
prolactinoma tumorigenesis, but that have yet to be identified clinically. The majority of patients that present with prolactinoma can be successfully treated with medical therapy, thus surgical resection of tumor tissue is not necessary. As such, prolactinoma tissue is not
abundantly available for genetic and molecular analyses. Unfortunately, from the tissue that is available, state-of-the-art immunohistochemical, microarray, and proteomic expression analysis, oncogenic mutation studies, and DNA epigenetic approaches have been mostly unproductive. Novel candidate oncogenes are frequently proposed for tumorigenesis of prolactinomas and other neuroendocrine tumors, but minimal progress has been made to implicate a specific oncogene or tumor suppressor, or markers of proliferation, senescence, dormancy, or anti-apoptosis, in pituitary tumorigenesis (111). The difficulty in identifying candidate oncogenes may be a result of a transient phosphorylation event that cannot be detected with traditional proteomics. Correlative studies have provided only modest information and have failed to give insights as to cause. To date, the best clues about the mechanism of pituitary tumorigenesis come from familial pituitary tumor disorders and mouse models, where mutations in conserved signaling pathways and factors that govern the cell cycle are critical in pituitary tumor formation (107, 112).
In Table I.2, I have assembled a list of animal models that have proven useful for studying pituitary adenoma; for more details, see the following references: (17, 68, 94, 98, 112–116). While rodent models have understandable limitations, a great deal has been learned from these models and they provide significant insights into the intracellular pathways that may be altered in abnormal human pituitary and lactotrope physiology. It is important to emphasize that although certain genetic alterations can yield PRL-secreting
Table I.2: Animal Models of Prolactinoma7
7
Table I.2 was previously published in following publication and is reprinted here with permission: Booth, A. et al. “Signaling pathways regulating pituitary lactotrope homeostasis
pituitary tumors in mice, adenoma formation is very delayed and thus is not fully accurate in representing the human disease state. This demonstrates that a single gene mutation or deletion is not sufficient, and that an additional mutation is likely required for true prolactinoma tumorigenesis.
D2R and Dopamine-Agonist Intolerance/Resistance in Prolactinoma Patients As mentioned previously, the dopamine D2 receptor, D2R, is a key regulator of lactotrope biology. Clinically, dopamine agonists (DAs), such as cabergoline and
bromocriptine, are used to treat patients with prolactinoma (19). In many patients, DAs are successful in halting lactotrope cell proliferation, shrinking prolactinoma size, and reducing serum PRL. However, approximately 10-20% of patients are DA resistant, and others do not tolerate prolonged therapy (19, 118, 119). Dysfunctional D2R-mediated signaling and/or downregulated D2R expression is thought to be the primary mechanism by which patients become resistant to DA therapies (119). Indeed, if dopamine signaling is abolished by dysfunction or knockout of D2R in mice, lactotrope homeostasis is lost, resulting in
prolactinoma formation (120). As such, we require a better understanding of the downstream effects of activated D2R signaling.
Two isoforms of D2R exist in rodents and mammals: a short isoform (D2S), and a long isoform (D2L). PKA inhibition has been well described to be the primary mechanism by which dopamine reduces prolactin synthesis in lactotrope cells (7, 43), but the precise
mechanism by which dopamine signals through D2R to reduce lactotrope proliferation has not been defined. More recently, other pathways have been shown to be regulated by D2R, including the ERK and PI3K pathways. In D2L-/- mice, phosphorylation of AKT is increased compared to wild type (WT) mice [Figure I.5; (120)], demonstrating that D2L inhibits
PI3K/AKT signaling when activated by dopamine. Phosphorylation of ERK was also increased in D2L-/- mice compared to WT, indicating that D2L also inhibits ERK signaling [Figure I.5; (120)]. In D2S-/- mice, phosphorylation of ERK and AKT is decreased compared to WT, demonstrating that D2S stimulates each of these pathways when activated by
Figure I.5: ERK 1/2 and AKT phosphorylation profiles in pituitaries of 4-month-old mice.8
Western blot analyses of proteins from pituitary extracts of 4-month-old mice. A: Quantifications of Western blots shown as phospho-ERK 1/2:total ERK 1/2 ratios normalized to the WT values, set as 1 (upper panel). Representative Western blots using phospho-ERK and total ERK antibodies (lower panel). B: Quantifications of Western blots shown as phospho-AKT:total AKT ratios normalized to WT values, set as 1 (upper panel). Representative Western blots using phospho-AKT and total AKT antibodies (lower panel). *p < .05 vs WT extracts; **p < .01 vs WT extracts, ^^p < .01 vs D2S-/-, ^^^p < .001 vs D2S-/-, ###p < .001 vs D2R-/-.
8
Figure I.5 was modified from the following publication and is printed here with permission: Radl, D. et al. “Each Individual Isoform of the Dopamine D2 Receptor Protects from
dopamine [Figure I.5; (120)]. D2R-/- mice, lacking both isoforms of the receptor, display reduced ERK phosphorylation and increased AKT phosphorylation compared to WT [Figure I.5; (120)]. Therefore, although the long and short isoforms of D2R each act independently to oppositely regulate ERK and PI3K signaling, when these effects are summated, dopamine-mediated D2R activation results in ERK stimulation and PI3K inhibition. Furthermore, D2R-/- mice display elevated prolactin and lactotrope hyperplasia, but D2L-/- and D2S-/- mice do not. However, after chronic exposure to estrogen for 10 weeks, D2L-/- and D2S-/- mice both display lactotrope hyperplasia and elevated prolactin to a similar degree as D2R-/- mice (120).
In summary, the study by Radl et al. demonstrates that expression of one isoform of D2R is sufficient to maintain lactotrope homeostasis under physiological conditions, but expression of both isoforms is required to prevent pathological lactotrope hyperplasia and hyperprolactinemia in the presence of an estrogen challenge. Moreover, D2R-mediated upregulation of ERK signaling in lactotropes suggests that activation of the ERK pathway does not promote proliferation, but instead promotes maintenance of lactotrope homeostasis. D2R-mediated inhibition of PI3K/AKT signaling in lactotropes suggests that inhibition of the PI3K pathway is necessary to inhibit lactotrope proliferation.
Scope of Thesis
Dysregulation of the signaling pathways that govern lactotrope biology contributes to tumorigenesis of PRL-secreting adenomas, or prolactinomas, leading to a state of
pathological hyperprolactinemia. Elevated serum prolactin suppresses the HPG axis, resulting in osteoporosis and infertility, as well as decreased libido. Standard DA therapies signal through the dopamine D2 receptor on lactotrope cells to inhibit lactotrope proliferation
and reduce prolactin synthesis, and approximately 10-20% of prolactinoma patients are resistant to DAs. Dopamine-mediated D2R activation results in ERK stimulation and PI3K inhibition, and dysfunctional D2R-mediated signaling is thought be the primary mechanism by which patients become resistant to DA therapies. As such, the work presented here aims to acquire a better understanding of the effects of activated D2R signaling on lactotrope cells. First, in Chapter III, I will explore the role of activated ERK signaling in promoting
lactotrope differentiation. In Chapter IV, I will expand on findings from Chapter III to investigate the mechanism by which lactotrope cells evade a tumorigenic response to oncogenic Ras. In Chapter V, I will shift the focus to PI3K and demonstrate that this pathway regulates lactotrope proliferation. I will also show data to support the hypothesis that there is a balance of ERK and PI3K signaling required to maintain lactotrope
homeostasis, and in Chapter VI, I will demonstrate that this balance is dysregulated in human prolactinoma. Finally, in Chapter VII, I will integrate my data with other reports from the literature to emphasize that conserved signaling pathways have context-dependent roles and distinct phenotypic outcomes, particularly in endocrine cell types. I postulate that dual inhibition of PI3K and ERK presents a potential alternative approach to therapy for DA-resistant prolactinoma. Taken together, the data presented here reveal an intricate balance of ERK and PI3K signaling that is required to maintain homeostasis in lactotrope cells, with loss of this equilibrium contributing to lactotrope tumorigenesis.
CHAPTER II
MATERIALS AND METHODS Cell Lines and Culture Methods GH4T2 and GH4C1 cells.
Cells were grown in a humidified tissue culture incubator at 37°C in 5% CO2. GH4T2 cells were originally established in the Gutierrez-Hartmann lab from GH4C1 rat somatolactotrope cells. Following loss of a TSH response, GH4C1 cells were passaged in rats until this response was regained; GH4 cells from tumor 2 were labeled GH4T2. Cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM, Gibco Life
Technologies, Inc.) supplemented with 15% horse serum and 2.5% fetal bovine serum (FBS; Gibco Life Technologies, Inc.), in addition to 1X HEPES and NEAA. Alternatively, GH4T2 cells were cultured in DMEM supplemented with HEPES, and NEAA, and 0.05% FBS for serum-starved conditions, or 0.5% FBS for low-serum conditions. For the doxycycline (dox)-inducible GH4T2 clones, 2.5% fetal bovine serum (FBS) was replaced with 2.5%
tetracycline system-approved FBS (CLONTECH Laboratories Inc.). GH4C1 cells and D2R-expressing GH4C1 clones (a kind gift from Dr. Paul Albert, Ottawa Hospital Research Institute, Ottawa, ON, Canada) were maintained in F10 nutrient mixture (F10, Gibco Life Technologies, Inc.) supplemented with 10% FBS (Gibco Life Technologies, Inc.).
Fingerprinting methods for authentication of rat cell lines are not commercially available. However, somatolactotrope markers Pit-1, PRL, and GH are detectable by Western blot in GH4T2 and GH4C1 cell lines.
293T and BOSC cells.
293T cells and BOSC cell are modified human embryonic kidney (HEK) 293 cells that were used to package lentivirus and retrovirus, respectively. Cells were grown in a humidified tissue culture incubator at 37°C in 5% CO2, and were maintained in DMEM supplemented with 10% FBS in addition to 1X HEPES and NEAA.
Pharmacological Inhibitors, EGF, and Cell Synchronization.
All pharmacological inhibitors, and CAB were reconstituted and diluted in DMSO. EGF was reconstituted and diluted in BSA. Media with or without treatments was changed every two days. For cell cycle (G0/G1) synchronization, cells were cultured in 0.1% charcoal dextran-stripped serum for 96 hours. Cell cycle re-entry was stimulated by the addition of culture medium with 10% FBS. Distributions of cells at each phase of the cell cycle were determined with flow cytometry analysis.
Transfection and Transduction Methods/Reagents Doxycycline-inducible construct cloning and transfection.
pcDNA3.1+ human H-Ras G12V 3xHA (N-terminus) and pCDNA3.1+ human H-Ras S17 3xHA (N-terminus) plasmids were purchased from UMR cDNA Resource Center
(University of Missouri-Rolla, Rolla, Missouri). The plasmids were cut with HindIII and XbaI to remove the Ras inserts. pTRE-Tight vector was purchased from CLONTECH Laboratories Inc and cut with HindIII, XbaI and then calf intestinal alkaline phosphatase (Fisher Scientific). The inserts and vectors were ligated (3:1 ratio) and products were sequenced to confirm insertion.
Doxycycline-inducible GH4T2 clones.
BS/IRES-M2 clone 13 was established by co-transfecting GH4T2 cells with BS/IRES-M2 plasmid (a kind gift from Dr. Stefania Lamartina, Istituto Di Richerche Di Biologia Molecolare, Rome, Italy) and empty pEGFP-C3 vector (Invitrogen; 10:1 ratio). Aliquots of 3x106 cells in 200 µL of media were added to plasmid DNA and transfected by electroporation at 220 V and 500 µF using a GeneZapper 450/2200 (IBI/Kodak) with 4-mm gap cuvettes. After transfection, cells were plated on 60-mm tissue culture plates in culture media and incubated for 24 hours. Media was changed to DMEM with 300 µg/mL G418 (Corning Cellgro) for selection. Clones were picked and maintained in DMEM with 100 µg/mL G418.
To create dox-inducible V12Ras clones, BS/IRES-M2 clone 13 was co-transfected with pTRE-H-Ras G12V 3xHA plasmid and empty pQCXIP-puro plasmid (CLONTECH; 5:1 ratio). Media was changed to DMEM with 100 µg/mL G418 and 4µg/mL puromycin (Sigma) for selection. Clones were picked and maintained in DMEM with 100 µg/mL G418 and 2 µg/mL puromycin. All experiments were performed in V12Ras clones number 10 and number 20 because these clones required dox for 3HA-V12Ras expression. The figures in Chapter III show experimental results from clone number 10. Comparable results were obtained from clone number 20 for all experiments.
To create dox-inducible N17Ras clones, the same transfection and selection protocol was used as described above for V12Ras clones, using a pTRE-H-Ras S17 3xHA plasmid. All experiments were performed in N17Ras clones number 11 and 37 because these clones required dox for 3HA-N17Ras expression. The figures in Chapter V show experimental results from clone 11. Comparable results were obtained from clone 37 for all experiments.
Transient transfections.
Transient transfections for luciferase assays were completed using electroporation parameters described above, or with Effectene transfection reagent, according to the
manufacturers protocol. Plasmid pRSVRaf-BXB, constitutively active Raf kinase construct, was provided by U. Rapp (National Cancer Institute, Frederick, Md), and pBabe hygro myc p110-CAAX was a gift from Channing Der [Addgene plasmid # 12591; (121)].
Viral plasmids and transduction.
pLKO.1 shPTEN constructs were purchased from the Functional Genomics Facility (Denver, Colorado), which is supported by the Cancer Center Support Grant
(P30CA046934). Here, I labeled the constructs by the final two numbers of each TRC. pLKO shMenin, pLKO.1 shp21, pLKO.1 shp27, and pLKO.1 shScramble constructs were
purchased from Open Bio Systems. To package lentivirus, 293T cells were transfected with 4 µg lentiviral shRNA plasmid and 2 µg of packaging plasmids using Effectene (Qiagen), according to the manufacturer’s protocol.
pBabe puro HA PIK3CA E545K plasmid was a gift from Jean Zhao [Addgene plasmid #12525; (122)] and pBABE-puro was a gift from Hartmut Land, Jay Morgenstern, and Bob Weinberg [Addgene plasmid #1764; (123)]. pBabe puro H-Ras 12V (35S), pBabe puro H-Ras 12V (37G), and pBabe puro H-Ras 12V (40C) were a gifts from Channing Der [Addgene plasmids #12588, #12589, #12590; (121)]. To package retrovirus, BOSC cells were transfected with 1.5 µg of the retroviral vector along with 1.8 µg each of pVsV-g and delta8.9 packaging plasmids using Turbofectin, according to the manufacturer’s protocol.
For both lentiviral and retroviral packaging, media with virus was collected after 48 hours, filtered, and snap frozen in single use aliquots for infection. For infection, GH4T2
cells were grown to ~50% confluency and a mixture of 2 mL freshly thawed virus, 4 mL fresh media, and 6 µL polybrene (Sigma) was added directly to cells. Approximately 16 hours later, a second round of infection was completed. Cells were selected using 1 µg/mL puromycin for up to 7 days. Alternatively, cells were selected overnight with 4 µg/mL puromycin and then maintained in media containing 2 µg/mL puromycin. Selection was removed prior to plating of functional assays.
Protein Lysate Preparation, SDS-PAGE, and Western Blotting
GH4T2 and GH41 cells were washed and harvested with extraction buffer (EB): 10 mM Tris, pH 7.4; 5 mM EDTA; 50 mM NaCl; 50 mM NaF; 0.1% BSA; 1% Triton X-100; 2 mM Na3VO4; 1 mM PMSF; and 1 mM DTT. Protein was quantitated using a Bio-Rad DC
protein assay (Bio-Rad Laboratories). Twenty-five to 50 µg of protein sample were loaded into SDS-PAGE gels. Primary antibodies and secondary antibodies are listed in Table II.1. Blots were stripped using Chemicon mild reblot reagent (Chemicon, Inc.) prior to re-probing with additional primary antibodies. Film was imaged using a scanner. Densitometric analysis was completed with ImageJ software (NIH).
Proliferation Assays
In 96-well tissue culture plates, 9,000 cells were plated in 100 µL media per well. All proliferation assays were incubated in a humidified tissue culture incubator at 37°C in 5% CO2. Fresh culture medium was added every other day. Cells were fixed with 4% PFA for each desired time point, stained with 0.1% crystal violet (Sigma, St. Louis, MO) in 25% methanol, lysed with 10% acetic acid, and absorbance was read using a Synergy HT
Microplate Reader (BioTek Instruments, Inc). For a subset of experiments, cells were stained with trypan blue and manually counted, revealing comparable results (data not shown).
Clonogenicity Assays
In 6-well tissue culture plates, 3,000 cells were plated in 2 mL media per well. All clonogenicity assays were incubated in a humidified tissue culture incubator at 37°C in 5%
Table II.1: List of Antibodies
PRIMARY ANTIBODIES
Protein Manufacturer/Catalog Number Application
c-Myc Sigma C3956 1:1000, 1 hr
GAPDH Applied Biosystems AM4300 1:40,000, 1hr
GH NIDDK AFP-411S 1:1000, 1 hr
HA Covance MMS-101P-200 1:5000, O/N
ERK1/2 (p-Thr202/Tyr204) Cell Signaling 4370 1:1000, O/N
MAPK/ERK (Total) Upstate 06-182 1:7500, 1 hr
Cell Signaling 9107 1:1000, O/N
Menin Bethyl A300-105A 1:1000, 1 hr
MKP-1 Santa Cruz sc-1199 1:200, 1 hr
MKP-2 Santa Cruz sc-1200 1:200, 1 hr
mTOR (p-Ser2448) Cell Signaling 2971 1:1000, O/N
mTOR (total) Cell Signaling 2972 1:1000, O/N
p21 Sigma 029K1786 1:7500, 1hr
p27 Novus Biologicals NB100-1949 1:200, 1hr
Pit-1 BAbCo PRB-230C-500 1:5000, 1hr
PRL NIDDK AFP-131581570 1:5000, 1 hr
PTEN Cell Signaling 9552 1:1000, 1 hr
S6K (p-Thr371) Cell Signaling 9208 1:1000, O/N
S6K (Total) Cell Signaling2708 1:1000, O/N
α-Tubulin Calbiochem CP06 1:10,000, 1 hr
SECONDARY ANTIBODIES
Protein Manufacturer/Catalog Number Application
Goat Anti-Rabbit IgG-HRP BioRad 170-6515 1:5000, 1hr
Goat Anti-Mouse IgG-HRP BioRad 170-6516 1:5000, 1hr
Goat Anti-Monkey IGG-HRP Cappel 55432 1:5000, 1hr
