Somatostatin induced SSTR2A endocytosis is regulated through b arrestin2 arf6 EFA6A PLD1 cascade in panceratic b cells (RINm5F)

By using radioligand binding techniques, we found that β-arrestin2, the ADP/ATP exchange factor for Arf6 A EFA6A, ADP-ribosylation factor 6 Arf6, and PLD1 partcipated in the SS-induced S

Somatostatin-induced SSTR2A endocytosis is regulated through

β-arrestin2-Arf6-EFA6A-PLD1 cascade in pancreatic β-cells (RINm5F)

by

Dai Tan Vo

A dissertation submitted to the graduate faculty

in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

Major: Biomedical Sciences (Pharmacology)

Program of Study Committee:

Walter H Hsu, Major Professor

Steven Carlson Timothy A Day Anumantha G Kanthasamy Michael J Kimber Richard J Martin

Iowa State University Ames, Iowa

2011

ii

TABLE OF CONTENTS

ABSTRACT iii

LIST OF ABBREVIATIONS iv

CHAPTER 1: GENERAL INTRODUCTION 1

CHAPTER 2: MATERIALS AND METHODS 22

CHAPTER 3: RESULTS 30

CHAPTER 4: DISCUSSION 66

REFERENCES 74

ACKNOWLEDGEMENTS 93

ABSTRACT

Somatostatin (SS) receptors (SSTRs) undergo endocytosis following agonist stimulation We investigated the mechanism underlying SS-induced SSTR2A endocytosis The results from our previous work demonstrated that SS increases phospholipase D 1 (PLD1) activity through Gβγ leading to the increase in phosphatidylinositol 4,5-bisphosphate (PIP2) levels in clonal β-cells HIT-T15 We also demonstrated that Arf6 and EFA6A are involved in the SS-induced PLD activity

in HIT-T15 In the present study, we found that in clonal β-cells (RINm5F) expressing SSTR2A, SS-14 evoked endocytosis of these receptors By using radioligand binding techniques, we found that β-arrestin2, the ADP/ATP exchange factor for Arf6 A (EFA6A), ADP-ribosylation factor 6 (Arf6), and PLD1 partcipated in the SS-induced SSTR2A endocytosis In addition, immunoprecipitation results showed the bindings between βarrestin2 and Arf6, Arf6 and EFA6A, and Arf6 and PLD1

over-Taken together, our findings strongly support the hypothesis that SS-induced SSTR2A endocytosis is mediated by the β-arrestin2- Arf6-EFA6A-PLD1 cascade in cell model system Therefore, we proposed a novel mechanism regarding SSTR2A endocytosis pathway in which via Gi/o-coupled SSTR2A, SS activates Gβγ, which triggers the GRK2 to phosphorylate SSTR2A, which is followed by recruitment of β-arrestin2 β-Arrestin2 then forms the β-arrestin2-EFA6A complex in the cytosol and translocates it to the plasma membrane to activate Arf6 The activated Arf6 will stimulate PLD1 to catalyze the formation of phosphatidic acid (PA) from phosphatidyl choline (PC) PA then serves as a cofactor to activate phosphatidylinositol 4-phosphate-5-kinase (PIP5K) to increase PIP2 formation PIP2

ultimately recruits endocytic proteins such as adaptor protein 2 (AP2) and clathrin to form vesicle-coat pits, resulting in SSTR endocytosis in β-cells

LIST OF ABBREVIATIONS

AC: Adenylyl cyclase

AP2: Adapter protein 2

Arf6: ADP-ribosylation factor 6

ARNO: Arf Nucleotide-binding site Opener

Bax: Bcl-2–associated X protein

β- Arr: β–arrestin

cAMP: cyclic adenosine monophosphate

cGMP: cyclic guanosine monophosphate

CHO: Chinese hamster ovary

DAG: Dicylglycerol

DAPI: 4’,6-diamidino-2-phenylindole dilactate

dn: dominant negative

EDTA: Ethylenediaminetetraacetic acid

EFA6: Guanine nucleotide exchange factor for Arf6 EFA6A: Guanine nucleotide exchange factor for Arf6 A EGFR: Epidermal growth factor receptor

Erk: extracellular signal-regulated kinases

FBS: Fetal bovine serum

FLAG-tag: polypeptide protein tag

GAPs: GTPase-activating proteins

GEFs: Guanine nucleotide Exchange Factors

GH: Growth hormone

GIT1: G protein-coupled receptor interacting protein 1

GPCRs: G protein-coupled receptors

GRKs: G protein-coupled receptor kinases

GRP: Cytohesin/general Receptor for Phosphoinosittides HA: Hemagglutinin

HEK: Human embryonic kidney

HRP: Horseradish peroxidase

IP3: Inositol-1,4,5-triphosphate

JNK: c-Jun-N-terminus kinase

KRB: Krebs Ringer Buffer

MAPK: Mitogen-activated protein kinases

PIP3: Phosphatidylinositol 3,4,5-triphosphate

PIP4Ks: Phosphatidylinositol 5-phosphate 4-kinases

PIP5K: Phosphatidylinositol 4-phosphate-5-kinase

PLA2: Phospholipase A2

PLC-β: Phospholipase C-β

PLD: Phospholipase D

PTP: Protein tyrosine phosphatase

RINm5F: Rat insulinoma m5F

RIPA: radioimmunoprecipitation assay

SDS-PAGE: sodium dodecyl sulfate polyacrylamide gel electrophoresis si-RNA: Small interfering RNA

SS: Somatostatin

SSTRs: Somatostatin receptors

STAT3: Signal transducer and activator of transcription 3

TSH: Thyroid-stimulating hormon

CHAPTER 1: GENERAL INTRODUCTION

DISSERTATION ORGANIZATION

This dissertation contains four chapters Chapter1: The general introduction includes

a research objective, background information and literature review Chapter 2: Materials and Methods Chapter 3: Results: SS-induced SSTR2A endocytosis is mediated by β-arrestin2, Arf6, EFA6A and phospholipase D1 signaling pathway Chapter 4: General discussion and future research In addition, a list of references cited and acknowledgements are included

RESEARCH OBJECTIVES

Endocytosis or internalization of G protein-coupled receptors (GPCRs) is a process that is induced when extracellular agonists bind to surface receptors There are many signaling proteins participating in the endocytosis pathway of GPCRs such as

G protein- coupled receptor kinases (GRKs), β-arrestins, ADP-ribosylation factor 6 (Arf6), phospholipase D (PLD), phosphatidylinositol-4-5-biphosphate (PIP2), and adapter protein 2 (AP2), (Doherty and McMahon, 2009; Houndolo et al., 2005; Sorkin and von Zastrow, 2009) Somatostatin (SS) receptors (SSTRs) undergo endocytosis following the agonist stimulation (Roosterman et al., 2008; Duran-Prado

et al., 2007; Liu et al., 2005) However, the signaling pathway of SS-induced SSTR endocytosis is not well-understood In a previous study, we found that SS stimulated PLD activity and promoted PIP2 synthesis in β-cells (Cheng et al., 2005) We also demonstrated that in β-cells SS- stimulated PLD activity was mediated by a guanine nucleotide exchange factor for Arf6 A (EFA6A) (Grodnitzky et al., 2007) In addition, results from our preliminary study showed that upon SS stimulation SSTR2A underwent endocytosis Therefore, we hypothesized that β-arrestins, Arf6, EFA6A, and PLD participate in the SS-induced SSTR2A endocytosis signaling pathway The proposed pathway is shown in Scheme 1 The objective of this dissertation is to

elucidate the role of four important signal proteins (β-arrestins, EFA6A, Arf6, and PLD) and their interrelationships in the SS-induced SSTR2A endocytosis pathway

We used the rat insulinoma m5F (RINm5F) cell line transiently overexpressed with SSTR2A as a model We used radioligand binding techniques to demonstrate endocytosis To knockdown the signaling genes we either used small interfering RNA (si-RNA) techniques or overexpressed dominant negative (dn) constructs to generate mutant signaling proteins In addition, we used immunoprecipitation technique to investigate the interactions between signaling proteins

BACKGROUND AND LITERATURE REVIEW

1 Somatostatin (SS)

SS, a peptide hormone, was first discovered from the extraction of hypothalamus as

a tetradecapeptide and inhibited secretion of growth hormone (Brazeau et al., 1973) Subsequently, SS was found to be secreted from many other tissues including δ-cells of pancreatic islets of Langerhans, nervous systems, and gastrointestinal mucosa, and in a small amount from the thyroid, adrenals, submandibular glands, kidneys, prostate, and placenta (Patel, 1999) It exists in two biologically active forms, SS-14 (14 amino acids) and SS-28 (28 amono acids) Both isoforms are derived from the proteolytic cleavage of the prosomatostatin (a 92-amino-acid protein), which is derived from a precursor of 116-amino-acid preprosomatostatin (Figure 1) SS has a potent inhibitory effect on the secretion of many hormones, including growth hormone (GH), thyroid-stimulating hormone (TSH), insulin and glucagon from the pancreas, and a number of gastrointestinal hormones such as gastrin, secretin, cholecystokinin, motilin, gastric inhibitory polypeptide and vasoactive intestinal peptide (Bloom and Polak, 1987; Kumar and Grant, 2010)

Figure 1 Somatostatin processing and amino acids order of SS-14 and SS-28

PreproSS: preprosomatostatin; proSS: prosomatostatin

2 SS receptors and their distribution

There are five genes on different chromosomes encoding five different SS receptor subtypes (SSTR1, SSTR2, SSTR3, SSTR4, and SSTR5) The name SSTR1-5 was based on the order of discoveries SSTR1 and SSTR2 were identified by Yamada et

al (1992) Soon after that the rest of SSTRs (SSTR3, SSTR4, and SSTR5) were cloned and characterized (Patel et al., 1995, 1996; Patel, 1997) SSTRs have the sizes from 356 to 391 amino acids and have overall sharing of 39-57% sequence homology The most divergences are at the N- and C-terminal segments SSTR2 exists in two spliced variants named SSTR2A and SSTR2B; they are different at the C-terminus and are encoded on the same chromosome 17 SSTR2A is a physiologically active isoform between two of them (Barnett, 2003)

SSTRs distribute in many cells/tissues, including brain, pituitary, islets, stomach, kidneys, liver, placenta, and lungs (Patel, 1999; Tulipano and Schulz, 2007; Olias et al., 2004) However, the distribution of SSTRs is not equal in all tissues There is usually more than one SSTR subtype in each cell or tissue SSTR1 is predominantly expressed in the brain and is also found in the pituitary, islets, and adrenals SSTR2

is predominantly expressed in many normal tissued including brain, vessels, nerve plexuses, adrenal medulla, spleen, kidneys, prostatic stroma, and pancreatic islets (Reubi et al., 2001; Liu et al., 2007; Reubi, 2003) The expression of SSTRs in neoplastic tissues is much higher in density than normal tissues in which SSTR2 is the most frequently expressed receptors of tumors (Reubi et al., 2001; Reubi, 2003; Liu et al., 2005) SSTR2 also has high affinity for SS analogs, therefore; SSTR2 is the most investigated SS receptor

SSTR3 is highly expressed in the cerebellum, but less in the rest of the brain SSTR3 is also highly expressed in the spleen, kidneys, and the liver (Taniyama et al., 2005) SSTR4 is poorly expressed in the brain, but abundantly expressed in the heart and moderately expressed in the lungs and islets SSTR5 is prominent in the

pituitary, intestine, and islets, but poorly expressed in the brain (Taniyama et al.,

2005)

3 SS receptor signaling

The actions of SS are mediated through SSTRs that are coupled to pertussis sensitive Gi/o subfamily They are all GPCRs with seven transmembrane-spanning domains (TM1-7) (Patel, 1999; Lahlou et al., 2004), which are connected by three extracellular and three intracellular loops (Hu et al., 2010)

toxin-G protein structures and actions

GPCRs are activated by extracellular ligands, leading the receptors to interact with and activate G proteins G proteins are composed of three subunits (α, β, and γ) and are referred to heterotrimeric G proteins They are activated by diverse ligands, which vary from single photons through ions, odorants, amino acids, fatty acids, neurotransmitters, peptides/polypeptides, and proteins Upon activation, the GPCRs undergo conformational changes leading to the activation of G proteins by the exchange of GDP/GTP at Gα subunit The Gβγ dimer then dissociates from the Gα

subunit Both moieties are active to stimulate downstream effectors and thereby initiate unique intracellular signaling responses (Tuteja, 2009) In humans, there are

21 Gα subunits encoded by 16 genes, 6 Gβ subunits encoded by 5 genes and 12 Gγ

subunits encoded by 12 genes (Downes & Gautam, 1999) Based on sequence homology and functional similarities of α subunits, heterotrimeric G proteins can be grouped into four subfamilies: Gs stimulates adenylyl cyclase (AC); Gi/o inhibits AC and voltage-dependent Ca2+ channels, and acitvates K+ channels; Gq/11 activates phospholipase C-β (PLC-β); and G12 activates Rho, a small G protein, through linking with a Rho GEF

G protein functions for SSTRs

Following the agonist stimulation, all 5 SSTRs inhibit AC (Patel et al., 1994); leading

to the reduction of intracellular cAMP concentrations Protein kinase A (PKA) is then

inactivated, thereby inhibiting the hormone secretion or gene expression (Patel et al., 1994) Other signaling pathways of the SSTR are dependent on many factors such as SSTR subtypes, signaling elements, desensitization, internalization and/or receptor crosstalks (Kumar and Grant, 2010)

SSTR1, SSTR2 and SSTR5 have been shown to directly inhibit voltage-dependent

Ca2+ channels (L-type, N-type, and T-type) via Go2 leading to the closure of the channels (Ikeda and Schofield 1989; Kleuss et al 1991; Tallent et al., 1996; Roosterman et al., 1998) As a result, the concentration of intracellular Ca2+ is decreased Upon activation of SSTR4, SS activates phospholipase A2 (PLA2) that leads to increase of arachidonic acid synthesis (Schweitzer et al., 1990)

SSTRs are directly coupled to several K+ channels to open them, thereby causing cell hyperpolarization (White et al., 1991; Sims et al 1991; Akopian et al 2000; De Weille et al., 1989) In smooth muscle cells, SS stimulates formation of IP3 through activation of phospholipase C-β3 via the Gβγ of SSTR3 (Murthy et al., 1996)

Interactions between SSTRs and mitogen-activated protein kinases (MAPK)

MAPK pathway plays an important role in SS-induced cell growth regulation (Lahlou

et al., 2004) Activated SSTRs have been demonstrated to either stimulate or inhibit MAPK such as extracellular signal-regulated kinases 1/2 (Erk1/2), c-Jun-N-terminus kinase (JNK), p38 and p44/p42 depending on SSTRs subtypes (Smalley et al 1999; Sellers et al 2000; Lahlou et al 2003) SSTR1 and SSTR4 stimulate Erk1/2 phosphotrylation SSTR4-mediated Erk1/2 activation leading to inhibition of cell proliferation is Ras-independent, whereas SSTR1-mediated Erk1/2 activation inhibiting cell growth is Ras-dependent (Seller, 1999; Florio et al., 1999) SSTR4 can also stimulate p38 SSTR2 can activate both p38 and Erk2 SSTR2-mediated Erk2 activation is Ras-Raf-dependent leading to cell growth inhibition (Lahlou et al 2003), whereas SSTR5 inhibits p42 to mediate cell growth inhibition and the inhibition of p42 is cyclic guanosine monophosphate (cGMP)-dependent (Cordelier et al., 1997;

Watt et al., 2009) SSTR5 also activates JNK Activated or inactivated MAPKs then activate or inactivate nuclear transcriptional factors such as signal transducer and activator of transcription 3 (STAT3) and finally stimulate or inhibit gene expressions and cell responses such as cell proliferation and cell differentiation (Lahlou et al., 2004)

SS has been recognized for its antiproliferative functions through SSTRs to trigger cell growth arrest or apoptosis (Sharma et al., 1996; War et al., 2011) SS-induced apoptosis is regulated through SSTR2 and SSTR3 Pro-apoptotic function of SSTR3 requires the activation of tumor suppressor protein p53 and induction of Bcl-2–associated X protein (Bax), Bax then promotes apoptosis (Sharma et al., 1996; Sharma and Srikant, 1998) Pro-apoptotic function of SSTR2 is protein tyrosine phosphatase (PTP)-dependent, but does not require activation of p53 (Teijeiro et al., 2002) Upon stimulation, cytosolic PTP is translocated to the plasma membrane to activate signaling cascade (Sharma et al., 1999) In addition, SS stimulation causes the dissociation of SSTR2-p85 complex (p85 is the regulatory subunit of phosphatidylinositol-3 (PI3) kinase which interacts with SSTR2 under basal condition) leading to p85 tyrosine dephosphorylation and PI3 kinase inactivation (Bousquet et al., 2006) As a result, cell growth is inhibited and apoptosis is induced SSTRs are coupled to Na+/H+ exchanger (NHE) to inhibit activity of the NHE1 receptor (Barber et al 1989; Hou et al 1994) NHE1 participates in the pH homeostasis and cell volume regulation by allowing Na+ to enter the cell to exchange H+ The activation of NHE plays a role in a variety of cellular events such

as cell proliferation, cell differentiation, and cell migration (Fliegel, 2005) SSTR1, SSTR3, and SSTR4, but not SSTR2 or SSTR5, mediate inhibition of NHE1 activity (Hou et al., 1994; Lin et al., 2003) Finally, Upon SSTR activation, the Gβγ of Gi/o

stimulates PLD activity and increases PIP2 synthesis in β-cells (Cheng et al., 2005)

4 SS analogs and therapeutic applications

Since SS has a short half-life of only one to three minutes, it cannot be used for clinical application SS analogs with longer half-life have been designed for clinical uses Octreotide is the first SS analog available for treatment with the half-life being about 90 minutes and duration of action being about 6-8 hours (Bauer et al., 1982)

To date, long-acting forms of octreotide and lanreotide are used for the treatments of acromegaly, Islet cell tumors such as insulinoma, thyrotropin-secreting pituitary adenomas, and carcinoids (Liu et al., 2007; Tulipano and Schulz, 2007) as well as other neuroendocrine tumors expressing high density of SSTRs (Oberg, 2009) The response of the anologs to the treatments is dependent on the expression of the SSTRs subtypes Unfortunately, the currently available SS analogs (octreotide and lanreotide) for clinical uses bind only to SSTR2 with high affinity (Patel, 1999) They bind to SSTR3 and SSTR5 with lower affinity and do not bind to SSTR1 and SSTR4 (Reubi et al., 2001) About one-third of patients with acromegaly and insulinoma have been reported to resist to the SS analogs treatments (Casarini et al., 2009; Feelders et al., 2009) The long-term treatment of insulinoma has been reported resistance to SS analogs (Lamberts et al., 1988) This resistance could be due to the desensitization/down-regulation of SSTRs and SSTRs endocytosis might play a role

in the desensitization/down-regulation process

5 Mechanism of GPCRs endocytosis

Endocytosis or internalization is a process by which cells engulf extracellular molecules to transfer them into the cytoplasm The GPCR endocytosis occurs when agonists bind to receptors Endocytosis of GPCRs is regulated by a series of important signaling proteins (Doherty and McMahon, 2009; Houndolo et al., 2005; Sorkin and von Zastrow, 2009) In general, there are two categories of GPCR endocytosis including clathrin-dependent and clathrin-independent pathways (Donaldson, 2009) The most-studied endocytosis pathway is clathrin-dependent one that is mediated by clathrin-coated pits (Wolfe and Trejo, 2007; Doherty and

McMahon, 2009) Clathrin-coated pits are composed of clathrin heavy- and chain molecules that form a polymeric lattice together with many of regulatory proteins such as AP2, GRKs, and arrestins (Wolfe and Trejo, 2007) β2-Adrenergic receptors and SSTRs are a typical example of clathrin-dependent endocytosis The binding of agonists to the receptors interacts with clathrin, AP2 or other adaptor proteins and recruits them to form clathrin-coated vesicle The complex is then invaginated inwardly to interact with several other proteins such as GRKs and β-arrestins (Sorkin and von Zastrow, 2009) The sequence of the GPCR endocytosis occurs as follows: The receptors are phosphorylated by GRKs (Hipkin et al., 1997; Tulipano et al., 2004) and are followed by the recruitment of β-arrestin to the receptors to form a complex with clathrin The entire complex is then internalized into the cells (Tulipano et al., 2004; Liu et al., 2007) (Figure 2) This process needs scission protein dynamins, the large GTPases; to bind to the vesicle-coated pit and release it from the membrane into the cytoplasm (Praefcke and McMahon, 2004) Regardless of the mode of endocytosis, the endocytosed receptors will go to endosomes and lysosomes for degradation or to the recycling endosomal carriers that can bring the receptors back to the plasma membrane (Grant and Donaldson, 2009) However, molecular mechanism of GPCR endocytosis is still largely unknown and needs to be fully elucidated The identification of specific signal proteins that participate in endocytosis of distinct receptors will provide new strategies to manipulate receptor signaling and develop new targets for drug discoveries (Wolfe and Trejo, 2007)

light-Figure 2 Mechanism of agonist-induced GPCR endocytosis GRK; G protein

coupled receptor kinase, AP2; adaptor protein 2

6 Endocytosis of SSTRs

Upon agonists’ stimulation, SSTRs undergo internalization (Roosterman et al., 2008; Duran-Prado et al., 2007; Boudin et al., 2000; Liu et al., 2005), however; they are not internalized equally SSTR2, SSTR3, and SSTR5 are internalized at a higher level than SSTR1 and SSTR4 (Roth et al., 1997; Olias et al., 2004; Jacobs and Schulz, 2007) The fates of the receptors after endocytosis are also different among subtypes The marjority (75-85%) of SSTR2 and SSTR5 are recycled to the plasma membrane, whereas SSTR3 is predominantly degraded in the lysosomes (Jacobs and Schulz, 2007; Koenig et al., 1998; Roosterman et al., 2008) SSTR2 is the most

investigated SSTR because of its broad expression and high affinity for SS analogs (Jacobs and Schulz, 2007) SSTRs endocytosis is clathrin-dependent (Koenig et al., 1998; Kreuzer, 2001) and β-arrestins are involved in the endocytosis of SSTR2 and SSTR3 (Kreuzer et al., 2001; Liu et al., 2007)

SSTR2A is the most distributed receptor isoform of SSTR2 in human tissues and neuroendocrine tumors In addition, SSTR2A is the physiologically active isoform between SSTR2A and SSTR2B (Barnett, 2003) In RINm5F cells, SSTR2 is the predominant SSTR isoform (Vo unpublished data) Therefore, in this dissertation I will focus my study on SS-induced endocytosis and signaling pathway of SSTR2A

As mentioned above, SSTR2A undergoes agonist-induced endocytosis in several cell lines and tissues (Boudin et al., 2000; Csaba et al., 2001; Liu et al, 2005) Upon stimulation, SSTR2 is rapidly phosphorylated by GRK2 (Hipkin et al., 1997; Tulipano

et al., 2004) Phosphorylated SSTR2 then recruits β-arrestin to the receptor to form vesicle complex with receptor and clathrin The whole complex is endocytosed into the cytoplasm (Liu et al., 2005; Tulipano et al., 2004) However, the specific proteins participated in the signaling pathway of endocytosis of SSTRs, particularly SSTR2,

is not well understood and needs to be elucidated (Jacobs and Schulz, 2007)

7 G protein coupled receptor kinases (GRKs)

GRKs are the primary mediators of agonist-dependent phosphorylation of GPCRs The GRK family is composed of 7 members, namely GRK1-7, which are classified into 3 subfamilies: GRK1 and GRK7 are in the first subfamily (Rhodopsin kinase subfamily), GRK2 and GRK3 form the second [β-adrenergic receptor kinase (βARK) subfamily], and GRKs 4-6 constitute the third subfamily (GRK4 family) (Moore et al., 2007) GRK2 is the first GRK that was shown to be involved in the endocytosis of the M2-muscarinic acetylcholine receptor (Tsuga et al., 1994) Subsequently, numerous additional studies elucidated that GRK-mediated phosphorylation of GPCR in endocytosis events (Ferguson, 2001) The mechanism is that GRK mediates phosphorylation of GPCR leading to activation of GPCR Activated GPCR

induces translocation of β-arrestin to the plasma membrane to bind to phosphorylated receptors through multiple interactions (Moore et al., 2007; Marchese et al., 2008) β-arrestin-GPCR complex then interacts with other proteins such as clathrin and AP2 to undergo endocytosis (Marchese et al., 2008)

8 β-arrestins and their role in endocytosis

The β-arrestins are a class of cytoplasmic soluble proteins (~48 kDa) that function in concert with GRKs to regulate intracellular signaling (Krupnick and Benovic, 1998) It was originally called S-antigen that was a causative antigen of experimental autoimmune uveitis (Krupnick and Benovic, 1998) Retinal S-antigen was subsequently demonstrated to be the protein regulating light-dependent signal transduction in rod photoreceptor cells (called visual arrestin) (Krupnick and Benovic, 1998) Another retinal arrestin was discovered in cone photoreceptor cells (called cone arrestin) (Krupnick and Benovic, 1998) Subsequently, two nonvisual β-arrestins were cloned (β-arrestin1 and β-arrestin2); they are ubiquitously expressed (Krupnick and Benovic, 1998; Lefkowitz and Shenoy, 2005) In the classical functions, β-arrestins play a key role in the interrelated processes of desensitization and endocytosis of GPCRs (Claing et al., 2001; Luttrell and Lefkowitz, 2002) The stimulation of GPCRs leads to rapidly phosphorylation of the receptor at the cytoplasmic tail by GRKs β-Arrestins then bind the phosphorylated receptor to block further G protein–initiated signaling (DeWire et al., 2007; Reiter and Lefkowitz, 2006) β-arrestin1 and β-arrestin2 are ~78% homologous and the coding differences are at the C-terminus The endocytosis of β2-adrenergic receptor is mediated by β-arrestin2, but both β-arrestins1 & 2 are involved in the endocytosis of angiotensin II type 1A receptor (Kohout et al., 2001) β-Arrestins can interact with endocytic elements, e.g AP2, Arf6 and its GEF ARNO (Lefkowitz and Shenoy, 2005) Both β-arrestins1 & 2 subtypes participate in the signaling pathway in SSTR2A endocytosis (Tulipano et al., 2004; Liu et al., 2007) Upon agonist stimulation, both β-arrestin subtypes are recruited to plasma membrane receptors and form arrestin-receptor

complex The complex is maintained during endocytosis process (Liu et al., 2007) However, during last ten years, it became clear that the arrestin-GPCR complex does not internalize without the participation of other signaling proteins Instead, plenty of evidence indicates that β-arrestins function as multifunctional adaptors and scaffold proteins that can interact with several other proteins including ERK, JNK, p38 and Akt and recruit a broad spectrum of signaling molecules (Ma and Pei, 2007; Lefkowitz and Shenoy, 2005) In this dissertation, we used siRNAs as a molecular tool to knockdown β-arrestin 1 and β-arrestin 2 to explore which β-arrestin subtype is involved in SS-induced SSTR2A endocytosis in our cell model system

9 PLD and endocytosis

PLD is a widely distributed enzyme that catalyzes the hydrolysis of phosphatidylcholine (PC) to form lipid second messenger phosphatidic acid (PA) (Jenkins and Frohman, 2005; Donaldson, 2009) There are two phospholipase D genes (PLD1 and PLD2) in mammalian and their structures consist of four conserved sequences (I–IV) in which both contain two HKD motifs (HxxxxKxD in which H is histidine, x is any amino acid, K is lysine and D is aspartic acid), pleckstrin homology (PH) and phox homology (PX) domains at their N-terminus (Exton, 1997; 2002; Jenkins and Frohman, 2005; Donaldson, 2009) (Figure 3) These domains are implicated in phospholipid and protein binding in which PX binds

to phosphatidylinositol 3,4,5-triphosphate (PIP3) and PH bind to PIP2 (Exton, 2002; Donaldson, 2009) PLD1 localizes at late endosomes and the Golgi apparatus and can be recruited to the plasma membrane during stimulation, whereas PLD2 consistently localizes at the plasma membrane and endosomes (Donaldson, 2009) PLD catalyzes the generation of PA and diacylglycerol (DAG) that affect membrane trafficking directly or indirectly by recruiting and/or activating signaling proteins (Donaldson, 2009) PA can activate phosphatidylinositol 4-phosphate 5-kinase (PIP5K), the enzyme that mediates the production of PIP2 and can stimulate several GTPase-activating proteins (GAPs) (Donaldson, 2009) PLD activity is stimulated by

Rho, Arf, Ral, PKCα/β, and phosphoinositides such as PIP2 and PIP3 (Exton, 2002; Jenkins and Frohman, 2005; Donaldson, 2009) Arf6 can activate both PLD (Brown

et al., 1993; Grodnitzky et al., 2007) and PIP5K (Honda et al., 1999) PLDs are involved in endocytosis of epidermal growth factor receptor (EGFR) (Shen et al., 2001) Overexpression of either PLD1 or PLD2 enhances endocytosis and degradation of EGFR (Shen et al., 2001) In addition, overexpression of dominant-negative mutant PLDs and treatment of the cells with 1-butanol, a PLD inhibitor, inhibited endocytosis (Shen et al., 2001) These results confirmed the role of PLD in endocytosis PLD2, but not PLD1, has been shown to be required for endocytosis of µ-opioid receptor (Koch et al., 2004) and angiotensin II receptor (Du et al., 2004)

In clonal β-cells HIT-T15, PLD1 is the only PLD isoform identified, which mediates SS-induced PIP2 production SS-induced PLD1 activation is mediated through EFA6A-Arf6 cascade (Grodnitzky et al., 2007) In RINm5F cells, we found both PLD1 and PLD2 (unpublished data) Therefore, we designed experiments to determine which PLD isoform was required for SS-induced SSTR2 endocytosis in RINm5F cells

10 PIP 2 and endocytosis

The lipid PIP2 accounts for only about 1% of total phospholipid in the plasma membrane, but plays a very important role in multiple cellular processes including vesicular transport, cytoskeletal organization, and cell signaling (Di Paolo and De Camilli, 2006) Classically, PIP2 is a precursor of two important second messengers

IP3 and DAG, This process is activated by PLC-β PIP2 is also a substrate for the synthesis of phosphatidylinositol 3,4,5-triphosphate (PIP3), which is activated by phosphoinositide 3-kinases PIP3 is an important second messenger that interacts with many downstream proteins to regulate a variety of cellular processes including cell survival, polarization, and proliferation In addition, PIP2 plays an important role

in endocytosis as well as recycling receptors back to the plasma membrane (Jost et al., 1998; Haucke, 2005; Sun et al., 2007; Abe et al., 2008) PIP2 is required for

transferrin receptor endocytosis (Abe et al., 2008; Jost et al., 1998) as well as clathrin- and actin-dependent endocytic internalization (Sun et al., 2007) PIP2

recruits and activates adaptor proteins such as AP2 and their binding partners such

as epsin, AP180, and amphyphisin which are important components of coated vesicles for endocytosis (Haucke, 2005)

clathrin-Phosphatidylinositol phosphate (PIP) is the immediate substrate for PIP2 synthesis (Haucke, 2005) and PIP2 synthesis is mediated by either phosphatidylinositol 5-phosphate 4-kinases (PIP4Ks) or PIP5Ks, however; the major PIP2 synthesis pathway is mediated by PIP5Ks (van den Bout and Divecha, 2009) There are three isoforms of PIP5Ks that have been characterized including PIP5Kα, PIP5Kβ, and PIP5Kγ (Ishihara et al., 1996; Loijens and Anderson, 1996)

PIP5Ks are activated by small G proteins Rho, Rac, Arfs and PLD (Brown et al.,

1993; Donaldson, 2009; van den Bout and Divecha, 2009) Active Arf1 and Arf6 are important for PIP5Ks activation and PIP2 synthesis (Hernandez-Deviez et al., 2004)

In fact, Arf1 and Arf6 can interact with and activate PIP5Ks (Martin et al., 1996; Honda et al., 1999) PA, which is generated by the activation of PLD, is an important activator of PIP5Ks as well (Moritz et al., 1992) Therefore, The small G protein-PLD-PIP5K loop can lead to clathrin-mediated endocytosis (Arneson et al., 1999, Brown et al., 2001)

11 Arfs and their role in endocytosis

ADP-ribosylation factors (Arfs) are a family of Ras-related GTP-binding proteins (~20 kDa, small G proteins) that function in the regulation of many cellular activities, including cell-cycle regulation, differentiation, cell-cell interactions, cell migration and endocytosis (Donaldson and Honda, 2005; Donaldson, 2008) There are three classes of Arfs based on amino acid sequence in which Class I includes Arfs 1, 2, 3; Class II includes Arfs 4 & 5; and Class III comprises only Arf6 Arfs are active when they are in the GTP-bound form and are inactive in the GDP-bound form The activation of Arfs is regulated by guanine-nucleotide exchange factors (GEFs) and

GTPases-activating proteins (GAPs) There are 14 GEFs and 24 GAPs that contribute to specific Arfs function by activating and inactivating Arfs, respectively, at specific locations (D’Souza-Schorey and Chavrier, 2006; Gillingham and Munro, 2007; Donaldson, 2008) Among Arfs, Arf1 and Arf6 are best characterized in which Arf1 acts at the Golgi apparatus and Arf6 functions at the plasma membrane to stimulate PIP2 synthesis through the activation of PLD and PIP5K (D’Souza-Schorey and Chavrier, 2006; Gillingham and Munro, 2007; Donaldson, 2008) Arf6 plays an important role in the internalization process of most GPCRs regardless of the endocytosis pathway (Houndolo et al., 2005) The GDP-GTP exchange of Arf6 takes place at the plasma membrane (Marcia et al., 2004; Donaldson and Honda, 2005)

In mu-opioid receptor, Arf6 activation is required for receptor endocytosis and recycling and this endocytosis pathway is mediated by PLD2 (Rankovic et al., 2009)

In angiotensin II type 1 receptor, Arf6 recruits AP2 and clathrin to receptor to activate the receptor endocytosis (Poupart et al., 2007) In addition, Arf6 is associated with clathrin-independent endocytosis in neuroblastoma cell (Kang et al., 2009) and beta

1 integrins (Dunphy et al., 2006) However; there are no publications regarding the involvement of Arfs in SSTRs endocytosis In this dissertation, I explored the role of Arf6 in the SS-induced endocytic signaling cascade for SSTR2A

12 GEFs and their role in endocytosis

Arfs require GEFs to catalyze the exchange of GDP for GTP This group of proteins includes Gea/Gnom/GBF (Geal/2p, GBF1, Emb30/Gnom), Sec7/BIG (Sec7p, p200/BIG1/2, AL022604), ARNO/cytohesin/GRP (ARNO, Cytohesin-1, GRP1/ARNO3), and EFA6 subfamilies in which Gea/Gnom/GBF and Sec7/BIG subfamilies are high-molecular-weight GEFs, whereas ARNO/cytohesin/GRP and EFA6 subfamilies are low-molecular-weight GEFs (Jackson and Casanova, 2000) Figure 4 illustrates these GEFs They all contain conserved, catalytic Sec7 domain -

a region of roughly 200 amino acids with strong homology (Jackson and Casanova, 2000; Cohen et al., 2007) In yeast, two high-molecular-weight GEFs including

Gea1p and Gea2p had been shown to catalyze nucleotide exchange on Arf1 in vitro

(Peyroche et al., 1996)

The low-molecular-weight GEFs contain the coiled-coil (CC) and

pleckstrin-homology (PH) domains, whereas the proline-rich regions are only in EFA6 (Jackson and Casanova, 2000) The proteins in this group are ARNO, Cytohesin-1, GRP1/ARNO3, and EFA6 They are ~45-50kDa in size and share ~77% homology

In contrast to high-molecular-weight GEFs, ARNO/cytohesin1 family functions to

catalyze nucleotide exchange on Arf6 rather than Arf 1 in vivo (Frank et al., 1998)

Claing et al (2001) demonstrated that ARNO was involved in the β-adrenergic receptor endocytosis in HEK cells and ARNO forms a complex with β-arrestins and Arf6 upon agonist stimulation (Claing et al., 2001) These findings elucidated the interactions between signaling proteins in the GPCR endocytosis In contrast, Cohen

et al (2007) reported that Arf6 was activated less by ARNO than EFA6 in HeLa and COS-7 cells EFA6 family includes EFA6A, EFA6B, EFA6C, and EFA6D EFA6A, EFA6C, and EFA6D are primary expressed in the brain and intestine, whereas EFA6B is broadly expressed, but is not detected in the brain (Derrien et al., 2002) The work from our lab determined that EFA6A existed in β-cells (Grodnitzky et al., 2007) Because the findings from our lab indicate that EFA6A mediates SS-induced PLD activity through Arf6-PLD cascade (Grodnitzky et al., 2007), we hypothesize that EFA6A-Arf6 mediates SS-induced SSTR2A endocytosis in β-cells

13 Arf GTPase activating proteins (Arf GAPs)

As discussed above, GEFs turn on the signaling for small G proteins by catalyzing the exchange of GDP for GTP, whereas GAPs turn off the signaling by promoting GTP hydrolysis These two groups of proteins together regulate cellular events The Arf GAPs have a common domain, Arf GAP domain, comprising a zinc-binding motif

At least 24 genes that encode proteins with Arf GAP domains have been cloned in the human and are classified into two major groups including ArfGAP1 and AZAP types (Inoue and Randazzo, 2007) The Arf GAP domain of ArfGAP1 is at the N-terminus of the protein, whereas that in AZAP type is between PH and ankyrin (ANK) repeat domain (Figure 5) Each group is further divided into subgroups based

on additional domains Six genes encoding ArfGAP1 type are divided into three subtypes: ArfGAP, SMAP, and GIT (Figure 5) ArfGAP subtypes including ArfGAP1

and ArfGAP3 function at the Golgi apparatus to inactivate Arf1 (Lanoix et al., 1999; Tanigawa et al., 1993) SMAP1 and SMAP2 of SMAP subtype function as GAPs of Arf6 and Arf1, respectively Two members of GIT (GIT1 and GIT2) have been shown

to function with Arf6 in GPCR endocytosis events (Claing et al., 2001; Hoefen and Berk, 2006)

AZAP type Arf GAPs contains PH, Arf GAP, ANK repeat domain structural motif Twelve genes encoding AZAPs are subdivided into four subtypes (ASAPs, ACAPs, ARAPs and AGAPs) (Figure 5) (Inoue and Randazzo, 2007) ASAP1 and ASAP2 prefer Arf1 and Arf5 other than Arf6 in vitro (Brown et al., 1998) Activated ASAP1 decreases Arf1-GTP, but not Arf6-GTP formation (Furman et al., 2002; Liu et al., 2005b) In contrast to ASAPs, activated ACAPs prefer Arf6 over Arf1 and Arf5 in vivo and in vitro (Jackson et al., 2000)

ARAPs are the largest proteins of ArfGAP family ARAPs have a RhoGAP domain, a sterile α-motif (SAM) domain, five PH domains, and a Ras association (RA) domain Two of five PH domains contain PIP3-binding consensus sequences; therefore PIP3

is more potent than any other phosphoinositides to stimulate ARAPs (Inoue and Randazzo, 2007) There are three members of ARAPs including ARAP1, ARAP2, and ARAP3 ARAP1 functions with Arf1 and Arf5 (Miura et al., 2002), whereas ARAP2 and ARAP3 prefer Arf6 over Arf1 and Arf5 to regulate Arf6-dependent events including endocytosis (Krugmann et al., 2002; Yoon et al., 2006)

There are two members of AGAPs: AGAP1 and AGAP2 These AGAPs have a GTP-binding protein-like domain (GLD) at the N-terminus AGAPs activity prefers Arf1 and Arf5 to Arf6 The GAP activity of AGAPs is stimulated by PIP2 and PA (Nie

et al., 2002; Nie et al., 2005)

Figure 5 Domain structures of Arf GAP subfamilies Abbreviations are: ALPS, Arf GAP1 lipid-packing

sensor; Arf GAP, Arf GAP domain; ANK, ankyrin repeat; BAR, Bin/Amphiphysin/Rvs; BoCCS, binding

of coatomer, cargo and SNARE; CALM, CALM binding domain; CB, clathrin-box; CC, coiled-coil; FG repeats, multiple copies of the XXFG motif; GLD, GTP-binding protein-like domain; GRM, Glo3 regulatory motif; PBS, Paxillin binding site; PH, pleckstrin homology domain; Pro(PxxP)3, cluster of three Proline-rich (PxxP) motifs; Pro(D/ELPPKP)8, eight tandem Proline-rich (D/ELPPKP) motifs; RA, Ras association motif; RhoGAP, RhoGAP domain; SAM, sterile a-motif; SH3, Src homology 3 domain; SHD, Spa-homology domain Adapted from Kahn et al (2008) and Spang et al (2010)

In this dissertation, I hypothesize that SS-induced SSTR2A endocytosis in β-cells is mediated by the β-arrestins-EFA6A-Arf6-PLD cascade Therefore, the objective of this dissertation is to elucidate the role of four important signal proteins (β-arrestins, EFA6A, Arf6, and PLD) and their interrelationships in the SS-induced SSTR2A endocytosis pathway I used RINm5F cells transiently overexpressed with SSTR2A

as a model Cells were overexpressed with dn-constructs of Arfs, ARNO and EFA6

to generate mutant signaling proteins or knockdown β-arrestins, Arf6, EFA6 and PLDs by siRNAs SS-14 was used to stimulate endocytosis We then used radioligand binding technique to determine endocytosis In addition, we used immunoprecipitation technique to investigate the protein-protein interactions

CHAPTER 2: MATERIALS AND METHODS

1 Antibodies and reagents

Somatostatin (SS)-14 was purchased from American Peptide Company, Inc (Sunnyvale, CA) Mouse monoclonal anti-Arf 6 (3A-1; sc-7971), rabbit polyclonal anti-PLD1 (sc-25512), rabbit polyclonal anti-β-arrestin1 (sc-9182), and mouse monoclonal anti-β-actin (ACTBD11B7; sc-81178) antibodies were purchased from Santa Cruz Biotechnolgy, Inc (Santa Cruz, CA) Rabbit polyclonal anti-β-arrestin2 antibody was purchased from Biorbyt Ltd (Riverside, UK) Rabbit polyclonal anti-PLD2 antibody was a kind gift from Dr Sylvain Bourgoin, Université Laval, Quebec, Canada Rabbit polyclonal anti-EFA6A, rabbit polyclonal anti-EFA6B, and goat polyclonal anti-ARNO antibodies were gifts from Dr Claing, Université de Montréal, Montréal, Canada Horseradish peroxidase (HRP) conjugated goat anti-mouse IgG (H+L, SJ 22238) and HRP conjugated goat anti-Rabbit IgG (H+L, SJ 29096) second antibodies were purchased from Biomeda Corp (Burlingame, CA) HRP conjugated rabbit anti-goat (H+L) was purchased from Kirkegaard and Perry Laboratory, Inc (Gaithersburg, MD) Arf6 siRNA (sc-77367), β-arrestin1 siRNA (sc-63298), β-arrestin

2 siRNA (sc-63299), PLD1 siRNA and PLD2 siRNA were purchased from Santa Cruz Biotechnolgy, Inc (Santa Cruz, CA) EFA6A siRNA was from Dharmacon, Inc (Lafayette, CO) LipofectaminTM2000 was from Invitrogen Corp (Carlsbad, CA) PureYield® Plasmid kits (MiniPrep, MidiPrep, and MaxiPrep systems) were purchased from Promega Corp (Fitchburg, WI) Protein G agarose was from Pierce® (Rockford, IL) Protease inhibitors were purchased from Sigma-Aldrich® Arf1 dominant negative mutant (dn) [Arf1(T31N)], dn-Arf5 [Arf5(T31N)], and dn-Arf6 [Arf6(T27N)], dn-EFA6A [EFA6A(E242K)], dn-EFA6B [EFA6B(E651K)], and dn-ARNO [ARNO(E156K)] constructs were kind gifts from Dr Julie Donaldson, National Institutes of Health (Bethesda, MD) Cy-3 streptavidin from Jackson Immuno Research Laboratories, Inc (West Grove, PA) and 4’,6-diamidino-2-phenylindole

dilactate (DAPI dilactate; 0.5 % stock) from Molecular Probes, Inc (Eugene, OR) were donated from Dr Row’s laboratory, Iowa State University (Ames, IA)

2 Cell culture

The rat insulinoma cell line (RINm5F) was maintained at 37 oC in RPMI 1640 medium (GIBCO®) with 10% fetal bovine serum (FBS) and aerated with 5% CO2

and 95% air

3 Cloning of the rat SSTR2A

Total RNAs were isolated from RINm5F cells using RNAqueous® Kit (Ambion, Austin, Texas) according to the manufacturer’s protocol SSTR2 primers were designed by using VECTOR® NTI (Invitrogen®) software The sense sequence is

5’ATGGAGTTGACCTCTGAGCAGTTC3’and the antisense sequence is

5’TCAGATACTGGTTTGGAGGTCTCC3’ according to the known sequence of the rat

SSTR2 (GenBank accession No 54305) Polymerase Chain Reaction (PCR)

product was obtained by using SuperScriptTM III One-Step RT-PCR System (Invitrogen®) following the manufacturer’s protocol PCR product was testified by

running with 1.2 % agarose gel (40 ml) at 80 V for 45 min Whole PCR product was

run with 60 ml of agarose gel at 60 V for 1.5 h The DNA bands were collected by cutting under the UV light and purified by using PurelinkTM Quick gel Extraction Kit (Invitrogen®) according to the manufacturer’s protocol The genes were ligated and

transformed to E coli (JM109) and then grew on petri dishes Ten different white

colonies were chosen from the dishes and put into 10 of 50-ml centrifuge tubes containing 10 ml of broth agar and 13 µl of ampicillin (100 mg/ml) Those tubes were shaken overnight at 37 oC and 225 rpm The DNA –contained E coli in each tube

was lysated and purified by using PROMEGA® KIT The DNAs were collected and submitted to ISU DNA Facility for sequencing The data were analyzed using Invitrogen® NTI software and the expected SSTR2A gene with 1110 bp were confirmed The full-length of rSSTR2A cDNA was subcloned into correspondent restriction sites of pcDNA3.1+ vector (Invitrogen, Barcelona, Spain) This step was

performed in Spain by Dr Justo Castaños laboratory (University of Co ́rdoba, Co

́rdoba, Spain)

4 Plasmids and siRNAs Transfections

Cells (2X105/well) were plated in 24-well plates and were transfected with SSTR2A (1.5 µg/well), dn-constructs (1.5 µg/well), and/or siRNAs (40 pmol/well) on the next day of plating using Lipofectamine®2000 following the manufacturer’s instructions Briefly, 1-1.5 µg of SSTR2 or dn-constructs and/or 4 µl of 10 µM siRNAs solution were diluted in 50 µl of OptiMEM® and 2 µl of Lipofectamine® 2000 was diluted in

another tube in 50 µl of OptiMEM® with gentle mixing The OptiMEM® used for

transfection had no FBS After allowing them to stand for 5 min both the mixtures were combined and mixed gently The complex mixture was incubated for 20 min and then 100 µl of transfection mixture was added to each well containing fresh 150

µl OptiMEM® After 5 h of starvation, 250 µl of RPMI 1640 medium with 20% FBS were added The regular media were replaced the next day The experiments were performed 72 h after transfections

5 Radio-ligand binding assays

125I-SS tracer was prepared using chloramine T and purified by Sephadex G25

column The determination of SSTR endocytosis using reduction of cell surface

receptors and uptake of 125I-Tyr-SS has been described by Roosterman et al (2008) Briefly, RINm5F cells (2x105 cells/well) were plated in 24-well plates 24 h before the transient transfection with SSTR2A and/or dn-constructs/siRNAs The experiments were performed after 72 h of transfections using RPMI 1640 medium without FBS Cells were washed twice with 0.5 ml RPMI 1640 medium, and then equilibrated with 250 µl RPMI 1640 medium for 10 min at 37 oC SS-14 (1 µM) was applied for 30 min to induce endocytosis The reaction was terminated by putting the 24-well plate on ice and excess SS-14 was washed 3X with 500µl ice cold RPMI

1640 (pH 7.4) medium followed by 2X of 500 µl RPMI acid wash (pH 5) to remove

cell membrane bound SS 125I-SS (~100,000 cpm) was added to each well and incubated for 10 min at 37 oC The nonspecific binding group contained 1 µM SS during 125I-SS incubation Cells were washed 4X with cold RPMI 1640 medium (pH 7.4) and 2X of 500 µl ice cold RPMI (pH=5) Cells were then collected using 1 N NaOH and counted by a γ-counter (Packard®, A Canberra Company) In the

membrane bound assays, 125I-SS (~100,000 cpm) was added and incubated for 30 min at 4 oC Cells were only washed 5X with cold RPMI1640 medium then were collected using 1 N NaOH and counted by a γ-counter The selection of 125I-SS at

~100,000 cpm was based on the results showed in the Figure 1 The results showed that at 100,000 cpm we observed the highest percentage of binding (bound/free) At

a lower number of tracer (10,000; 30,000; 50,000 cpm) or higher (at 300,000; 1,000,000 cpm) the percentage of binding was lower (Figure 1) These results justified that use of 100,000 cpm in our previous experiments, which is in agreement with Roosterman et al (Roosterman et al., 2008)

Figure 1 Percentage of specific binding of 125I-SS at different total quantity of 125I-SS 125I-SS at

~10,000; 30,000; 50,000; 100,000, 300,000; and 1,000,000 cpm was added to each well with 0.25 ml RPMI 1640 medium and incubated for 5 min at 37 oC The nonspecific binding group contained 1 µM

SS during 125I-SS incubation Cells were washed 4X with cold RPMI 1640 medium (pH 7.4) Cells were then collected using 1 N NaOH and counted to determine the total binding Data are mean ± SEM, n=3 independent experiments

6 Immunoprecipation (IP)

The IP technique was used to study the interactions among 4 proteins including arrestin, Arf6, EFA6A, and PLD The technique has been described previously by Claing et al (2001) RINm5F cells (3X106) were washed with 1 ml Krebs Ringer Buffer (KRB) buffer and re-suspended in 1 ml KRB Cell suspension was then treated with 100 nM SS-14 for indicated times The reactions were stopped by

β-freezing cells at -140 oC feezer for 90 s Cells were collected by centrifugation at

1,200 x g at 4 oC The lysates were collected using 500 µl of modified radioimmunoprecipitation assay (RIPA) buffer containing 50 mM Tris-HCL; 150 mM NaCl; 0.5% Triton X-100; 2 mM EDTA; and 1 % protease inhibitors 1-1.5 mg of the lysate was used for each IP assay The lysates were pre-cleaned by adding 25 µl of 50% slurry washed protein-G beads and incubated for 30 min at 4 oC The lysates were then collected by centrifugation at 10,000 x g for 10 min at 4 oC 5 µg of first antibodies were added to each lysate vial and the lysate was incubated overnight at

4 oC 50 µl of 50% slurry washed protein-G beads were added and incubated for 2 h

to form the antibody-immunogen complexes The antibody-immunogen complexes were precipitated by washing 5X with 1 ml of phosphate buffer saline (PBS) (10,000

X g at 4 oC, 10 s each) 60 µl of 2X loading buffer containing 10% SDS and 20 mM dithiothreitol was added and boiled for 10 min to release the proteins from the beads The eluted proteins were analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting We used first antibody against β-arrestin2 to test immunoprecipitation of Arf6 and Arf6 antibody to test the

IP of EFA6A, PLD1, PLD2, and ARNO

incubation times were 1 h at room temperature) The membranes were developed using diaminobenzidine

8 Immunocytochemistry (ICC)

We used ICC to determine the expression of dn-constructs in the transfected cells Cells were transfected with dn constructs 72 h before the experiments were performed Cells were washed 2X with PBS and fixed in paraformaldehyde 4% for

15 min at 4 oC After 2X of 0.1M PBS and 3X of 50mM KPBS washes, blocking buffer (KPBS buffer containing 1% BSA, 0.4% Triton-X-100, 1.5% FBS) was added for 2 h before adding 1st Ab overnight The protein expression of constructs with tags (HA-tag for Arf1, Arf5, Arf6, whereas that with FLAG-tag for EFA6A, EFA6B, and ARNO) were determined by using antibodies against HA and FLAG tags (1:200 dilution), respectively Cells then were washed 3X with KPBS containing 0.02% of Triton-X-100 Biotinylated secondary antibodies were added and incubated for 2 h at room temperature in the humid chamber The cells were rinsed 4X over 10 min with KPBS Cells were then incubated for 30 min in the humid chamber with Cy-3 streptavidin (1:3000) The cells were washed 5X over 5 min with KPBA DAPI (1:333) was used to stain the nucleus The expressions of the genes were measured under fluorescent microscope (Nikon Microscope, Japan) by counting of 100 individual cells of each group to compare the expression of the HA of FLAG-tagged dominant constructs

9 Cyclic AMP (cAMP) assay

RINm5F cells (2x105 cells/well) were plated in 24-well plates 24 h before the transient transfection with SSTR2A and dominant negative (dn) construct for Arf6 [Arf6(T27N)] On the day of experiment, cells were washed twice with 500 µl of RPMI Cells then preincubated with 0.25 ml of 1 mM of phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine (IBMX) The adenylyl cyclase activator forskolin 1 µM was added 5 min, which was followed by 3 nM SS for 4 min In the SS-induced endocytosis groups, cells were incubated with 100 nM SS for 30 min, then subjected

to 4X acid wash (pH=5) to remove membrane bound and free SS The cAMP reaction was stopped by adding 0.3 ml of 0.01 N HCl Cells were then collected for cAMP radioimmunoassay (RIA) using a previously published method (Kato et al., 1993)

10 Data analysis

Data are expressed as mean ± SEM from at least 3 independent experiments (n=3) Data were subjected to one-way ANOVA analysis using Prism 5.0 software (Graphpad Software, Inc, CA) Tukey’s test was used to compare means with P<0.05 being considered as statistically significant

CHAPTER 3: RESULTS

1 Total binding of 125 I-SS to RINm5F cells

The highest total binding was observed at 5 min incubation of the tracer When we increased the incubation time, the total binding was decreased At 90 min, the binding was lowest with ~ 7 % compared to the 5 min time point These results were consistent with the results from the tracer uptake experiments in the dissertation (Figure 1) These results suggest that by increase 125I-SS incubation time SSTR2A receptors are either desensitized or inactivated

Figure 1 Total binding of 125 I-SS RINm5F cells were transiently transfected with SSTR2A

72 h prior to experiments 125 I-SS was added and incubated for indicated times (5, 10, 15,

30, 45, 60, 90 min) Cells were applied to ice cold RPMI washes The total binding 125 I-SS was determined by γ-counter Data are mean ± SEM, n=3 independent experiments

γ-~7% compared to the 10-min time point (Figure 2) Based on these results, we chose 10 min tracer incubation for uptake experiments

Figure 2 Uptake of 125I-SS RINm5F cells were transiently transfected with SSTR2A

72 h prior to experiments 125I-SS was added and incubated for indicated time (5, 10,

15, 30, 45, 60, 90 min) Cells were applied to ice cold RPMI and acid washes The uptake of 125I-SS was determined by γ-counter Data are mean ± SEM, n=3

independent experiments, * P≤ 0.05

3 SS-induced SSTR2A endocytosis in RINm5F cells

Upon agonists’ stimulation, SSTRs undergo internalization (Roosterman et al., 2008; Duran-Prado et al., 2007; Roth et al., 1997; Liu et al., 2005) Our results on SSTR2A endocytosis (Figure 3) were consistent with these previous findings (Roosterman et al., 2008; Liu et al., 2005) in which SS stimulated SSTR2 endocytosis SSTR2A-transfected RIN cells were incubated with SS-14 at different concentrations (1 nM,

10 nM, 100 nM, 1µM, and 10 µM) for 30 min to stimulate endocytosis 125I-SS was added and incubated for 10 min at 37 oC Cells were applied to ice cold RPMI and acid washes The results showed that SS-14 evoked SSTR2A endocytosis leading

to the decrease in the membrane receptors Therefore, SS-14 pretreatment lowered the uptake of 125I-SS SS-induced SSTR2A endocytosis was in a concentration-

dependent manner (Figure 3) With 1 and 10 nM SS-14 pretreatment, there was no inhibition of 125I-SS uptake 0.1-10 µM SS-14 caused concentration-dependent inhibition of 125I-SS uptake, which is interpreted as SS-induced SSTR2A endocytosis Based on these results, we used SS-14 100 nM to stimulate SSTR2A endocytosis SS-induced endocytosis can also be observed in RINm5F cells without transfection of SSTR2A However, tracer uptake in non-SSTR2A transfected cells was 4 times lower compared to SSTR2A transfected cells With 100 nM SS stimulation, similar degree of endocytosis was attained when compared to that of the SSTR2A transfected cells However, at higher concentration of 1 and 10 µM of SS there was no greater inhibition of 125I-SS uptake compared to the 100 nM SS pretreatment (Figure 4)

Our next question is: how much time is needed to for SS-14 to induce apparent SSTR2A endocytosis? To examine the time course of SS-induced endocytosis, SSTR2A-transfected RIN cells were stimulated with 100 nM SS-14 for 5, 10, 15, 30,

45, and 60 min The results showed that the highest degree of endocytosis was attained at 30 min of SS treatment (Figure 5)

Since 0.45 M sucrose can block clathrin-mediated endocytosis (Roosterman et al., 1997; 2007) We wanted to investigate if sucrose can block endocytosis in our cell model The results showed that SS-induced endocytosis was significantly blocked by 0.45 M sucrose (Figure 6), which suggested SSTR2A endocytosis is clathrin-dependent

Figure 3 SS-induced SSTR2A endocytosis is dose-dependent SSTR2A

transfected RINm5F cells were incubated with SS-14 at different concentrations (1nM, 10nM, 100nM, 1 µM, and 10 µM) for 30 min to stimulate endocytosis 125I-SS was added and incubated for 10 min at 37 oC Cells were applied to ice cold RPMI and acid washes Data are mean ± SEM, n=4 independent experiments * P≤ 0.05

Figure 4 SS-induced SSTR2A endocytosis is dose-dependent in RINm5F without

SSTR2A tansfection RINm5F cells were incubated with SS-14 at different concentrations (0.1-10 µM) for 30 min to stimulate endocytosis 125I-SS was added and incubated for 10 min at 37 oC Cells were applied to ice cold RPMI and acid washes Data are mean ± SEM, n=4 independent experiments * P≤ 0.05