Molecular Mechanisms in Signaling by Target of Rapamycin Complex 2 in Saccharomyces Cerevisiae

Molecular Mechanisms in Signaling by Target of Rapamycin Complex 2 in Saccharomyces cerevisiae

Melissa Nicole Locke

A dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy

Molecular and Cell Biology

in the

Graduate Division

of the

University of California, Berkeley

Committee in charge:

Professor Jeremy W. Thorner, Chair Professor Douglas E. Koshland Assistant Professor Roberto Zoncu Assistant Professor James A. Olzmann

Spring 2019

Abstract

Molecular Mechanisms in Signaling by Target of Rapamycin Complex 2 in Saccharomyces cerevisiae by Melissa Nicole Locke Doctor of Philosophy in Molecular and Cell Biology University of California, Berkeley Professor Jeremy W. Thorner, Chair

Protein kinases are enzymes that transfer the γ-phosphate of ATP to acceptor residues, most often serine, threonine or tyrosine, in proteins. Phosphorylation by protein kinases are integral to the signaling processes that control virtually all aspects of cellular physiology and phosphorylation is the most abundant post-translational modification in signal transduction networks. Covalent attachment of a phosphate group to a protein can cause a conformational change, introduce a new epitope that enhances or prevents interaction with another protein, increase or decrease the rate of turnover, and/or affect subcellular localization. In these ways, phosphorylation-dependent events can influence cell metabolism, morphology, and differentiation.

Target of rapamycin (TOR) complex 2 (TORC2) is an evolutionarily conserved multi-subunit protein kinase and indispensable regulator of plasma membrane homeostasis. Genetic studies in yeast have demonstrated that the sole essential downstream target and effector of TORC2 is the protein kinase Ypk1 (and its paralog Ypk2), whose mammalian ortholog is SGK1. In humans, mutations in the TORC2 signaling network have been implicated in inflammatory disorders, metabolic diseases, and multiple cancers.

In this doctoral dissertation, I describe the studies I undertook to explore novel aspects of TORC2-Ypk1 signaling. Because a Rab5-specific guanine nucleotide exchange factor (GEF), Muk1, was identified in a prior global screen for candidate Ypk1 targets, I investigated the role of Rab5-type GTPases in the TORC2 signaling network. I confirmed first that Muk1 is a substrate of Ypk1 and demonstrated that Ypk1-mediated phosphorylation stimulates Muk1 function in vivo. Second, and strikingly, I found that yeast lacking its two Rab5 GEFs (Muk1 and Vps9), or its three Rab5 paralogs (Vps21/Ypt51, Ypt52 and Ypt53), or overexpressing Msb3 (a Rab5-directed GTPase-activating protein), all exhibited pronounced reduction in TORC2- mediated phosphorylation and activation of Ypk1. Finally, I showed that Vps21 co- immunoprecipitates with TORC2, and furthermore, that purified GTP-bound Vps21 is able to stimulate in vitro the activity of TORC2 immuno-enriched from Rab5-deficient cells. My findings suggest that TORC2-dependent and Ypk1-mediated activation of Muk1 provides a control circuit for positive (self-reinforcing) up-regulation to sustain TORC2-Ypk1 signaling.

For Ben, Your courage and strength can move mountains.

i Table of Contents

Acknowledgements ...... iii List of Tables ...... iv List of Figures ...... v List of Abbreviations ...... vii Chapter 1. Introduction— The TORC2-dependent signaling network in the yeast Saccharomyces cerevisiae...... 1 Chapter 2. Materials and Methods ...... 13 Chapter 3. Rab5 GEF Muk1 is a physiologically important substrate of TORC2- activated protein kinase Ypk1 ...... 21 Chapter 4. Rab5 GTPases are necessary for efficient TORC2-mediated Ypk1 activation ...28 Chapter 5. Perspective— Regulation of TORC2 function and localization by Rab5 GTPases in Saccharomyces cerevisiae ...... 44 Literature Cited ...... 54 Appendix A: Interrogating substrate recognition by Ypk1— does Ypk1 bind to a specific docking motif distinct from its consensus phospho-acceptor site ...... 68 Appendix B: Identification of ubiquitylation enzymes that regulate cell cycle progression ...... 75

ii Acknowledgments

Thank you to my mentor, Professor Jeremy Thorner, for providing support and guidance during the most difficult times of grad school. I am grateful for your generosity, wisdom, and mentorship. Thank you for taking me on as your last graduate student, and for giving me the freedom to pursue my interests. I continue to be inspired by your passion for learning, your thorough approach to scientific inquiry, and the way you treat your lab like family.

Thank you to the members of the Thorner Lab who welcomed me as a member of the family, especially Françoise Roelants, for encouraging me when experiments were difficult, and for being a wonderful in-lab mentor.

Thank you to the members of my thesis committee, Doug Koshland, James Olzmann, and Roberto Zoncu, as well as Susan Marqusee and Georjana Barnes, for providing support and mentorship during my entire PhD.

Thank you to the wonderful friends I made in Berkeley, Colleen McGourty, Elijah Mena, Eugene Oh, Hermann Meyer, Joel Gussman, Franziska Lorbeer, Kunitoshi Chiba, Jared Bard, Alison Altman, Charlene Bashore, Matt Davis, Ellen Goodall, and Liam Mader, for countless coffee runs, gourmet meals, and trips to the beach.

Thank you to my parents, Steve and Elaine, for your infinite love, support, and encouragement of all of my endeavors. Thank you for cooking me meals when I needed them, for taking me on vacation when I needed a break, and always making sure I have a piece of home with me wherever I am. Thank you to my incredibly talented brother, Tim, for always pushing me to do my best, laughing with me when I need it, laughing at me when I deserve it, and for being the one who always knows what I need. Thank you to Gung Gung, Po Po, Yeh Yeh, and Nin Nin for your foundation of love and hard work. Thank you to Kate for always being just a phone call away.

Thank you to Ben for your steadfast support and love. This dissertation would not have been possible without all of the strength and positivity you bring to our life. Thank you for your patience and for learning about my research to help me analyze data, proofread manuscripts, and prepare presentations. I‘m grateful for your sense of adventure, and thankful for our travels to explore the beauty of Northern California and discover our magical spots in Point Reyes, Jenner, and Hawaii. Thank you for your unwavering support during this journey. I am so excited for our next adventure together.

iii List of Tables

Table 1. S. cerevisiae strains used in this study

Table 2. Plasmids used in this study

iv List of Figures

Figure Location Description of Contents Page

Figure 1.1 Schematic depiction of the primary structure of Ypk1. 3

Figure 1.2 Selected sites of Ypk1 phosphorylation. 7

Figure 3.1 Muk1 is phosphorylated by Ypk1 in vivo. 24

Figure 3.2 Muk1 is phosphorylated at its Ypk1 consensus phospho- 23 acceptor site motifs. Figure 3.3 Ypk1-dependent phosphorylation of Muk1 is stimulated by 24 sphingolipid depletion. Figure 3.4 Muk1 is phosphorylated by Ypk1 in vitro. 24

Figure 3.5 Ypk1-dependent phosphorylation of Muk1 promotes 25 endocytosis of permeases. Figure 3.6 Ypk1-mediated phosphorylation stimulates the ability of Muk1 26 to support growth under heat stress. Figure 4.1 Rab5 GTPases and their GTP loading are necessary for 29 efficient phosphorylation and activation of Ypk1 by TORC2. Figure 4.2 GTP-bound Vps21 stimulates TORC2 phosphorylation of 31 Ypk1. Figure 4.3 Absence of Rab5 GEFs does not alter TORC2 localization at 32 the cell cortex. Figure 4.4 Absence of Rab5 GEFs does not alter Ypk1 accessibility to 33 either Pkh1 or TORC2. Figure 4.5 Vps21 physically interacts with TORC2, but not Ypk1. 35

Figure 4.6 Interaction of Vps21 with Tor2 in TORC2 is specific. 36

Figure 4.7 Establishment of a reliable in vitro assay for TORC2-mediated 38 phosphorylation of a Ypk1-derived substrate. Figure 4.8 Vps21 stimulates TORC2 in vitro. 39

Figure 4.9 Sequence alignment of the indicated small GTPases. 41

v Figure 4.10 Muk1-dependent Rab5-mediated mechanism for sustained 42 activation of TORC2-Ypk1 signaling. Figure 5.1 Saccharomyces cerevisiae Tor2 modeled onto human mTOR 49 in the RHEB-binding pocket in mTORC1. Figure 5.2 Comparison of Saccharomyces cerevisiae Vps21 to Homo 50 sapiens RHEB. Appendix A Figure 1 Recombinant Fps1 interacts with Ypk1 in vitro 70

Appendix A Figure 2 A minimal Fps1 peptide is sufficient for binding Ypk1 in vitro 71

Appendix A Figure 3 Conservative mutation of the consensus Ypk1 phospho- 73 acceptor site on Fps1 does not diminish interaction with Ypk1 Appendix A Figure 4 PKA in the absence and presence of a PKI-derived peptide. 74

Appendix B Figure 1 Two-color cell cycle progression reporter 76

Appendix B Figure 2 How FUCCI fluorescent reporter system enables visualization 77 of cell cycle progression Appendix B Figure 3 Characterization of K562-FUCCI* cell lines 78

Appendix B Figure 4 Schematic of cell cycle screen 79

Appendix B Figure 5 Genes targeted by shRNAs that were enriched in G1 compared 81 to S/G2/M in the cell cycle screen Appendix B Figure 6 Genes targeted by shRNAs that were enriched in S/G2/M 82 compared to G1 in the cell cycle screen.

vi List of Abbreviations

3MB-PP1 1-(tert-butyl)-3-(3-methylbenzyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine

A600 nm Absorbance at 600 nm

AEBSF 4-(2-aminoethyl)benzenesulfonyl fluoride

AbA Aureobasidin A

ATP Adenosine triphosphate

BSA Bovine serum albumin

CIP Calf intestinal phosphatase

C-terminus Carboxy terminus

DDM n-dodecyl β-D-maltoside

DHAP Dihydroxyacetone-phosphate

DMF Dimethyl formamide

DMSO Dimethyl sulfoxide

EDTA Ethylenediaminetetraacetic acid

FACS Fluorescence-activated cell sorting

GAP GTPase-activating protein

GEF Guanine nucleotide exchange factor

GFP Green fluorescent protein

Glo3P sn-glycerol-3-phosphate

GST Glutathione S-transferase

HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid

IPTG Isopropyl β-D-1-thiogalactopyranoside

MAPK Mitogen- (or messenger-) activated protein kinase

M&M Materials and Methods

vii MBP Maltose-binding protein (E. coli malE gene product)

MIPC mannosyl-inositolphosphoryl-ceramide mNG mNeonGreen

Myr Myriocin

N-terminus Amino terminus

P3C Rhinovirus protease 3C

PM Plasma membrane

PtdIns4,5P2 phosphatidylinositol-4,5-bisphosphate

Rab Ras GTPase super-family member abundant in brain

SDS-PAGE Sodium dodecyl sulfate-polyacrylamide gel electrophoresis shRNA Short hairpin RNA

SPT L-serine:palmitoyl-CoA C-palmitoyltransferase (decarboxylating)

TCA Trichloroacetic acid

TOR Target of Rapamycin

TORC1 TOR Complex 1

TORC2 TOR Complex 2

viii Chapter 1. Introduction— the TORC2-dependent signaling network in the yeast Saccharomyces cerevisiae

1. Background

This chapter of my dissertation is based on a review of the current literature on yeast TORC2 signaling that was published in the journal Biomolecules in 2017 (Roelants et al. 2017a). It was co-written by me with Françoise M. Roelants, Kristin L. Leskoske, Maria Nieves Martinez Marshall, and Jeremy Thorner. The citation is: Roelants FM, Leskoske KL, Martinez Marshall MN, Locke MN, Thorner J (2017) The TORC2-dependent signaling network in the yeast Saccharomyces cerevisiae. Biomolecules 7: E66.

Growth and division of a eukaryotic cell are complex processes coupled to various stimuli, the availability of nutrients, and additional external and internal cues, all of which are tightly regulated at many levels, especially when challenged by suboptimal conditions and other stresses. When it was found that the antibiotic rapamycin was a potent inhibitor of the proliferation of virtually every eukaryotic cell type examined (e.g., fungi, T cells, and tumor cells) (Vezina et al. 1975; Martel et al. 1977; Eng et al. 1984), it became clear that the molecular target of rapamycin must be highly conserved and its function critical for cell survival. Indeed, ever since the authentic target of rapamycin (TOR) was first discovered using an elegant genetic approach in budding yeast (Saccharomyces cerevisiae) (Heitman et al. 1991; Loewith and Hall 2011), TOR has emerged as a universal, centrally important sensor, integrator, and controller of eukaryotic cell growth (Gonzalez and Hall 2017; Saxton and Sabatini 2017).

As first demonstrated in yeast (Kunz et al. 1993; Helliwell et al. 1994; Loewith et al. 2002; Wedaman et al. 2003), TOR is found in all eukaryotic cells in two distinct macromolecular complexes, TOR complex 1 (TORC1) and TOR complex 2 (TORC2). TORC1 is sensitive to inhibition by rapamycin, whereas TORC2 is normally insensitive (Jacinto et al. 2004; Huang and Manning 2009; Betz and Hall 2013; Gaubitz et al. 2015). The catalytic subunit in both TORC1 and TORC2 is the very large TOR polypeptide; metazoans possess a single TOR-encoding gene (human mTOR, 2549 residues), whereas budding yeast (Heitman et al. 1991), fission yeast (Ikai et al. 2011), and other fungi (Eltschinger and Loewith 2016) encode two TOR proteins, Tor1 and Tor2 (2470 and 2474 residues, respectively, in S. cerevisiae).

2. TORC2 Structure and Function

Only Tor2 can serve as the catalytic subunit in yeast TORC2, whereas TORC1 is functional when its catalytic subunit is either Tor1 or Tor2 (Loewith et al. 2002; Wedaman et al. 2003). The small β-propeller (―WD40 repeat‖) protein Lst8 is present in both TORC1 and TORC2, presumably because, as shown for its mammalian ortholog (mLST8), it binds tightly to and greatly stabilizes the TOR kinase fold (Yang et al. 2013; Aylett et al. 2016; Baretic et al. 2016). Aside from Lst8, however, the other known subunits in yeast TORC2, namely Avo1 (mammalian ortholog is mSIN1), Avo2 (appears to be fungal-specific), Avo3/Tsc11 (mammalian ortholog is Rictor), Bit2 and Bit61 (mammalian counterparts are Protor1/PRR5 and Protor2/PRR5L), and Slm1 and Slm2 (also fungal-specific), are all separate and distinct from those in TORC1 (Loewith and Hall 2011; Eltschinger and Loewith 2016; Gaubitz et al. 2016).

1 Structural, genetic, and biochemical analyses have revealed that TORC2 is only insensitive to rapamycin because the C-terminus of Avo3 blocks the ability of rapamycin-bound FKBP12 (Fpr1 in S. cerevisiae) to bind to the FKBP12-rapamycin binding (FRB) domain of Tor2; deleting a portion of the Avo3 C-terminus renders TORC2 sensitive to rapamycin inhibition (Gaubitz et al. 2015). In a yeast cell where such an avo3 truncation (avo3∆C) is combined with a dominant point mutation (TOR1-1) in the FRB domain of Tor1 that blocks its association with rapamycin-Fpr1 (Heitman et al. 1991), TORC2 can be uniquely inhibited by addition of rapamycin (Gaubitz et al. 2015).

Using budding yeast (S. cerevisiae) as the experimental organism, it has been shown by the Thorner lab (Roelants et al. 2010; Roelants et al. 2011; Muir et al. 2014; Muir et al. 2015; Alvaro et al. 2016; Roelants et al. 2017b) and others (Berchtold and Walther 2009; Berchtold et al. 2012; Niles et al. 2012; Sun et al. 2012; Frohlich et al. 2016) that TORC2 plays an essential role in sensing the status of the plasma membrane (PM) and controlling the reactions that ensure PM homeostasis. In addition to influencing reactions that affect PM lipid and protein composition, TORC2 action also modulates assembly and function of the actin cytoskeleton and actin-driven endocytosis (deHart et al. 2002; Bartlett and Kim 2014; Rispal et al. 2015; Roelants et al. 2017b).

3. TORC2 Effectors

Yeast TORC2 is localized at the PM (Kunz et al. 2000; Sturgill et al. 2008; Berchtold and Walther 2009; Niles et al. 2012; Leskoske et al 2018; Martinez Marshall et al 2019) and responds to certain perturbations and stresses by directly phosphorylating two distinct types of protein kinases: Ypk1 (and its paralog Ypk2/Ykr2) (Chen et al. 1993) (Figure 1.1); and, Pkc1 (Levin et al. 1990), the yeast ortholog of mammalian PKN2/PKR2.

TORC2 phosphorylates Ypk1 at multiple sites at its C-terminal end, thereby stimulating its activity (Roelants et al. 2011; Muir et al. 2015; Leskoske et al. 2017). Ypk1 is a member of the sub-family of eukaryotic protein kinases (AGC-family of protein kinases) first defined by the related protein kinases cyclic-3‘5‘-AMP-dependent protein kinase (PKA), cyclic-3‘5‘-GMP- dependent protein kinase (PKG), and the conventional Ca2+-, DAG-, and PtdSer-activated protein kinases (PKC) (Pearce et al. 2010; Leroux et al. 2018). Several criteria indicate that the human orthologue of Ypk1 is SGK1. The catalytic domain of Ypk1 shares a greater degree of amino acid sequence identity (57%) with that of SGK1 than any other human protein kinase (Hunter and Plowman 1997; Manning et al. 2002; Rubenstein and Schmidt 2007). In addition, the optimal site for phosphorylation of substrates (phospho-acceptor site specificity) of Ypk1 and SGK1 are virtually identical (Casamayor et al. 1999). Moreover, expression of SGK1 (and no other, even closely related, mammalian protein kinase) rescues the inviability of a ypk1∆ ypk2∆ double mutant (Casamayor et al. 1999). Lastly, as a prerequisite for its TORC2 stimulation, basal Ypk1 activity requires phosphorylation of its activation loop (T-loop) by the PM-localized protein kinases Pkh1 and Pkh2 (Roelants et al. 2002; Roelants et al. 2004). Pkh1 and Pkh2 are the yeast sequence homologues of the mammalian, PM-associated, 3-phosphoinositide- dependent protein kinase 1 (PDK1) and expression of PDK1 rescues the inviability of a pkh1∆ pkh2∆ double mutant (Casamayor et al. 1999). As for Ypk1, full SGK1 activity requires its

2 phosphorylation by both PDK1 (Mora et al. 2004) and mTORC2 (Garcia-Martinez and Alessi 2008).

4. Structure, Function, and Regulation of Ypk1

Ypk1 is a 680-residue polypeptide with its catalytic domain located within its C-terminal portion, preceded by a long N-terminal segment and followed by a much shorter C-terminal extension (Figure 1.1). Ypk2 (677 residues) shows extensive homology to Ypk1 across its entire length and their catalytic domains share 90% sequence identity (Figure 1.1 legend). Aside from its homology to that in Ypk2, the N-terminal portion of Ypk1 does not exhibit detectable similarity to any characterized sequence motif or structural domain. By contrast, the C-terminal extension shares readily detectable relatedness (30% identity) to that in SGK1. FIGURE 1 FIGURE 1FFIFGIIGGUUURRREEE 1 11

A A BAAA B BBB Fpk1 Pkh1 Fpk1 FFpFpkpk1k11 YPpkh115A-mycPPPkkhkh1h11 Ypk15A-mycYYpYpkpk1k151A55-AAm--mmyycycc Fpk2 Pkh2 TORC2 FPphko2sphataFsFpFepkp:k 2k 2 2 P- k h 2 +TOPPPkRkhkCh2h22 TTOTOORRRCCC222PhosphataPsPPheho:h os o sp s php hah -ata at ta sa s es e :e : : + - - - + ++

Ypk1 phospho-Ypk1 phospho-Yppphpkhoh1osospsphphoho-oY--YpYpkpk1k11 S51 S71 T504 S644 T662 S51 S71 S S 5S 51 5 1 1 S S7 S 71 7 1 1 T 5 0 4 S 6 4T T45T 50 5 04T04 46 6 2 S S6S646444 4 T T6T6662622 Ypk1 - Ypk1 - YYpYpkpk1k1 1- -- YpkL1X LX LX Ypk1 YYpYpkpk1k11 AL TM HM 1 N-termin98al Activation Turn Hydrophob341ic N-terminal NN-Nt-e-tteremrrmminiAniancalatlilvatioIBn: mTAAuycArctncitvtivaiv602ataHiottioyinodnn roTpTuThuruonrrbnnicHH680yHydydrdorr opophphohobobibcicic IB: myc IBIIBB: :m: mmyycycc regulatory domain loop motif motif regulatory Ndo-tmearmiin nal reg luoloaptory mdoomatif in motif regulatory doremrgeaugKiinluanltaoatroyser yd o ddmoo malomaianoipnin mo tliof loopop mCmom-ttoifteoiftrmiif naml motoiftif regulatory C C CCC domain

Figure 1 MYSWKSKFKF 1.1. Schematic GKSKEEKEAK depiction HSGF of theFHS primarySKK EEQQNNQ1 MYSWKSKFKF structure ATA 1 1 1 MYSWKSKFKF GEHDASITRSMYSWKSKFKF MYSWKSKFKFof GKSKEEKEAK Ypk1. Catalytic GKSKEEKEAK SGKSKEEKEAKGKSKEEKEAKLDRKG HSGF TdomainFHINPS S HSGF KKHSGFHSGF (black) EEQQNNQFHFHFHSSSS KKSandKKKK ATA EEQQNNQ EEQQNNQEEQQNNQ GEHDASITRSATAATAATA GEHDASITRS GEHDASITRSGEHDASITRS SLDRKGT INP SSLDRKGSLDRKG LDRKGTTINPTINPINP N- 61(light SN Sblue)SVVPVRV and CS-YterminalDASSSTST (dark VRDSNGGNSE blue) regulatory 61NTNSSQNLDE SN SelementsS VVPVRV 61 6161 S TANIGSTGTPSNS Nare SNSSSVVPVRVYS VVPVRVindicated.DAVVPVRVSSS TST S NDATSSSGMMSYSY DAYShadingDAVRDSNGGNSEDASSSSSSSSSTSTTSTTST reflects VRDSNGGNSE VRDSNGGNSEVRDSNGGNSE NTNSSQNLDE percent NTNSSQNLDE NTNSSQNLDENTNSSQNLDE TANIGSTGTP TANIGSTGTP TANIGSTGTPTANIGSTGTP NDATSSSGMM NDATSSSGMM NDATSSSGMMNDATSSSGMM sequence121 TIKVYNGDDF identity between ILPFPITSSE Ypk1 (680QILNKLLASG residues) 121VPPPHKEISK and TIKVYNGDDF the 121corresponding121121 EV TIKVYNGDDF TIKVYNGDDFTIKVYNGDDF DALIAQLILPFPITSSE segmentS ILPFPITSSE RVQILPFPITSSEILPFPITSSE QILNKLLASGIKNQGPA in its paralog QILNKLLASG QILNKLLASGQILNKLLASG VPPPHKEISKue Ypk2 VPPPHKEISK VPPPHKEISKVPPPHKEISK EV DALIAQL EV EVEVSDALIAQL DALIAQLRVQDALIAQLIKNQGPASS S RVQ RVQRVQ IKNQGPAIKNQGPAIKNQGPA (677181 residues): DEDLISSESA 1–98, AKFIPSTIML 22% (faint PGSSTLNPLL blue); 99–341, 181YFTIEFDNTV 62%DEDLISSESA 181(medium181181 ATIEAEYGTI DEDLISSESA DEDLISSESADEDLISSESA AKFIPSTIML blue); 342 AKPGFNKIST AKFIPSTIML AKFIPSTIMLAKFIPSTIML –PGSSTLNPLL602, 90% PGSSTLNPLL PGSSTLNPLL(black);PGSSTLNPLL YFTIEFDNTV and, YFTIEFDNTV YFTIEFDNTVYFTIEFDNTV ATIEAEYGTI ATIEAEYGTI ATIEAEYGTIATIEAEYGTI AKPGFNKIST AKPGFNKIST AKPGFNKISTAKPGFNKIST 603241–680, FDVTRKLPYL 73% (dark KIDVFARIPS blue). Abbreviations: ILL PSKTWQQ LX,241 EMGLQDEKLQ lowFDVTRKLPYL-complexity241241241 FDVTRKLPYLTIFDKINSNQ FDVTRKLPYLFDVTRKLPYL KIDVFARIPS sequences KIDVFARIPSDIHLDSFHLP KIDVFARIPSKIDVFARIPS ILL predictedPSKT WQQ ILL ILLILL by PSKEMGLQDEKLQ PSKPSK UnitProtTTWQQTWQQWQQ EMGLQDEKLQ EMGLQDEKLQEMGLQDEKLQ TIFDKINSNQ TIFDKINSNQ TIFDKINSNQTIFDKINSNQ DIHLDSFHLP DIHLDSFHLP DIHLDSFHLPDIHLDSFHLP 301 INLSFDSAAS IRLYNHHWIT LDNGLGKINI SIDYKPSR NK PLSIDDFDLL KVIGKGSFGK (Consortium301 INLSFDSAAS 2017) and/orIRLYNHHWIT SMART LDNGLGKINI (Letunic et301SIDYKPSR al. INLSFDSAAS2015)301301NK databases; INLSFDSAAS PLINLSFDSAAS SIRLYNHHWITIDDFDLL AL, activationIRLYNHHWIT KVIGKGSFGKIRLYNHHWIT LDNGLGKINI loop LDNGLGKINI LDNGLGKINI Thr SIDYKPSR (T504), SIDYKPSR NK SIDYKPSR PLSIDDNKFDLLNK PL PL SKVIGKGSFGKIDDSIDDFDLLFDLL KVIGKGSFGK KVIGKGSFGK 361 VMQVRKKDTQ KVYALKAIRK SYIVSKSEVT361 HTLAERTVLA VMQVRKKDTQ361361361 VMQVRKKDTQRVDCPFIVPL VMQVRKKDTQVMQVRKKDTQ KVYALKAIRK KVYALKAIRKKFSFQSPEKL KVYALKAIRKKVYALKAIRK SYIVSKS EVT SSYSYIVSKS YIVSKSHTLAERTVLAIVSKSEVTEVTEVT HTLAERTVLA HTLAERTVLAHTLAERTVLA RVDCPFIVPL RVDCPFIVPL RVDCPFIVPLRVDCPFIVPL KFSFQSPEKL KFSFQSPEKL KFSFQSPEKLKFSFQSPEKL phosphorylated421 YFVLAFINGG by Pkh1ELFYHLQKEG and, les RFDLSRARFYs efficiently, 421 TAELLCALDNby YFVLAFINGGPkh2421 421421(Roelants LHKLDVVYRD YFVLAFINGG YFVLAFINGGYFVLAFINGG ELFYHLQKEG et al. LKPENILLDY 2002)ELFYHLQKEG ELFYHLQKEGELFYHLQKEG RFDLSRARFY; TM, turn RFDLSRARFY RFDLSRARFYRFDLSRARFY TAELLCALDNmotif Ser TAELLCALDN TAELLCALDNTAELLCALDN LHKLDVVYRD LHKLDVVYRD LHKLDVVYRDLHKLDVVYRD LKPENILLDY LKPENILLDY LKPENILLDYLKPENILLDY (S644),481 QGHIALCDFG phosphorylated LCK LNMKDDD by TORC2; KTD T and,FCG T HM,PE481 Y LA hydrophobic QGHIALCDFGPELLLGL481481481 QGHIALCDFG GYTKAVDWWTQGHIALCDFG QGHIALCDFG motif LCKLNMKDDD Thr (T662),LCK LGVLLYEMLTLCKLCK KLNMKDDDLNMKDDDTLNMKDDDD T phosphorylatedFCG T K PEKKTTD TDTYDTFCGLATFCGFCGPELLLGLTTPET PEPE by Y YLAYLA LAGYTKAVDWWTPELLLGLPELLLGLPELLLGL GYTKAVDWWT GYTKAVDWWTGYTKAVDWWT LGVLLYEMLT LGVLLYEMLT LGVLLYEMLTLGVLLYEMLT TORC2.541 GLPPYYDEDV Phosphorylation PKMYKKILQE at both PLVFPDGFDR T504 and 541DAKDLLIGLL S644GLPPYYDEDV 541 are541541 SRDPTRRLGY GLPPYYDEDV essentialGLPPYYDEDVGLPPYYDEDV PKMYKKILQE for NGADEIRNHP PKMYKKILQEPKMYKKILQEPKMYKKILQE Ypk1 PLVFPDGFDR function, PLVFPDGFDR PLVFPDGFDRPLVFPDGFDR DAKDLLIGLL whereas DAKDLLIGLL DAKDLLIGLLDAKDLLIGLL SRDPTRRLGY SRDPTRRLGY SRDPTRRLGYSRDPTRRLGY NGADEIRNHP NGADEIRNHP NGADEIRNHPNGADEIRNHP * * *** 601 FFSQLSWKRL LMKGYIPPYK PAV SNSMDTS601 NFDEEFTR FFSQLSWKRL601601601EK FFSQLSWKRL PIDFFSQLSWKRLFFSQLSWKRL LMKGYIPPYKSVVDEYL LMKGYIPPYK SELMKGYIPPYKLMKGYIPPYK SPAVVQKQFGGSNSMDT PAV PAVPAVS SNSMDTNFDEEFTRSNSMDTSNSMDTSS S NFDEEFTREK NFDEEFTRNFDEEFTR PIDSVVDEYLEKEKEK PID PIDPID SESSVVDEYLSVVDEYLSVVDEYLVQKQFGG SE SESES SVQKQFGGSVQKQFGGVQKQFGG phosphorylation at T662** and* additional conserved C-terminal sites** (S653,** **S672** and*** S678) mediated by661 TORC2 WTYVGNEQLG only occursSSMVQGR underSIR those stressful661 conditionsWTYVGNEQLG661661661 W WTW thatT YVGNEQLGTYVGNEQLGSYVGNEQLGS MVQGR activate S SIRS SS TORC2MVQGRSMVQGRMVQGRSSIRS-IRIRYpk1 signaling Regions covered by mass spectrometry Regions coveredRegionsRegionsRegions by covered coveredcovered mass spectrometry by byby mass massmass spectrometry spectrometryspectrometry (Leskoske et al. 2017). Two IdentifiedN-terminal residuesas phosphorylated phosphorylated by massby Fpk1 spectrometryIdentified and, less Identified IdentifiedIdentified a efficiently,s phosphorylated a aa ssby s phosphorylated phosphorylated phosphorylated by mass spectrometry by byby mass massmass spectrometry spectrometryspectrometry Fpk2 (Roelants et al. 2010) are also shown. Sites mutated in Ypk1 12A Sites mutatedSitesSitesSites in mutated mutatedmutated Ypk1 12A in inin Ypk1 Ypk1Ypk112A12A12A

4.1. Activation Loop Phosphorylation by Pkh1 and Pkh2 As observedD for other AGC family proteinD kinasesE D DD(Pearce et al. 2010; Leroux et al. E 2018) , EEE activation of Ypk1 is regulated by phosphorylation on residues situated within three conserved Ypk15A Ypk15A YYpYpkpk1k151A55AA sequences. First, Ypk1 must be phosphorylatedYpk15A T57A on its activation loopY p(Tk11-1Aloop)YpkY1p11 kAat15A Thr504T57AYYpYpkpk1k 15within1A55AATT5T5757A7AA a Ypk111A Ypk1Y11YpYApkpk1k111A11AYAYpYpkp1kk111A11AA 504 conserved T FCGTPEY motif, a T modification57A S653A installed, asY pmentionedk111A 671S 6 7 above,2TS57A S 65 by3AT T5T Pkh15757A7AASS6S6 5653 and53A3AA Ypk111A 671SYYpYpkpk16k17112A11ASA66767171S1SS 66767272S2SS 5A Pkh2 (Roelants et al. 2002; RoelantsYpk15A S 6et53 al.A S 2004)63A . Pkh1 and Pkh2 are tetheredYpk15A S6 5to3YA YptheYpkSpk1k6151 3APM5ASA S6S65653 53viaA3AASS 6Stheir6363A3AA T57A S63A S64A Ypk1- P T57A S63ATT5TS57576A7A4AASS6S6363A3AASS6S6464A4YAApk1- P YYpYpkpk1k1-1 -P- PP tight association with Y Pil1,pk15A S6 a5 3A phosphatidylinositolS64A S671A -4,5-bisphosphateYpk15A S65 3YA (PtdInYpYpkSpk1k61514A55ASAS6Ss4,5P2)S656653573A31AAASS6S6464A-4AbindingASS6S6767171A1AA Ypk15A T57A S63A S671A S672A Ypk1 Y- pk15A T57AYYpYpkSpk1k61513A55AATT5ST57567A7AA1ASS6S6363A37AA2SAS6S6767171A1AASS6S6767272A2AAYpk1 - YYpYpkpk1k1 1- -- protein (KarotkiYp ket15 Aal.T 52011)7A S65,3 AcontainingS64A S672A anS6 7N8A-terminal regionYpk15A T 5of7AY homologyYpYSpkp1k6k151A535AATT5TS57576A7A4AAS Sto6SS65665 35the73A32AAAS S6Shuman6464A47AA8SAS6S676727 2Aproteins2AASS6S6767878A8AA BIN1 and Amphiphysin and the yeast protein Rvs167 (BAR domain) thatIB: ismy mostc similar to that IB: myc IBIIBB: :m: mmyycycc Ypk1- P Ypk1- P YYpYpkpk1k1-1 -P- PP Ypk1 - 3Y pk1 - YYpYpkpk1k1 1- -- IB: myc IB: myc IBIIBB: :m: mmyycycc in human proteins FER and CIP4 (F-BAR domain) (McDonald and Gould 2016). Both Pil1 and its paralogue Lsp1 (Karotki et al. 2011), are primary subunits of peripherally PM-bound protein complexes, dubbed eisosomes (Walther et al. 2006; Douglas and Konopka 2014), that bear certain similarities to mammalian caveoli (Moreira et al. 2012).

4.2. C-Terminal Phosphorylation by TORC2 For cell survival in response to certain stresses (e.g., sphingolipid depletion, heat shock, hypotonic conditions, high exogenous acetic acid), Ypk1 activity must be upregulated further by phosphorylation at Thr662 within its conserved hydrophobic motif sequence near its C-terminus (Roelants et al. 2004; Kamada et al. 2005; Tanoue et al. 2005; Roelants et al. 2011; Berchtold et al. 2012; Niles et al. 2012; Sun et al. 2012; Guerreiro et al. 2016) (Figure 1.1). As first revealed by analysis of Ypk2 (Kamada et al. 2005), phosphorylation at the hydrophobic motif is mediated by TORC2, which also phosphorylates another C-terminal site (Ser644 in Ypk1) within another conserved sequence, dubbed the turn motif (Roelants et al. 2004; Kamada et al. 2005).

A former Thorner lab graduate student, Kristin L. Leskoske, showed that several additional sites in Ypk1, aside from the turn and hydrophobic motif, are also phosphorylated by TORC2 and that these sites are as important for Ypk1 activity, stability, and biological function as Ser644 and Thr662 (Leskoske et al. 2017). Moreover, phosphorylation at these sites is a prerequisite for TORC2 phosphorylation of Thr662 in the hydrophobic motif (Leskoske et al. 2017). Similarly, the mammalian AGC kinase AKT1 is phosphorylated in an mTORC2-dependent manner at other C-terminal sites that are necessary for efficient phosphorylation of its hydrophobic motif (Liu et al. 2014). Furthermore, phosphorylation at the turn motif has been shown to be especially important for proper C-terminal folding and stability of mammalian AKT and PKC (Facchinetti et al. 2008; Ikenoue et al. 2008). On this basis, it seems likely that phosphorylation of the newly identified TORC2 sites in Ypk1 is also a prelude to phosphorylation of Ser644 in its turn motif because absence of phosphorylation at the new sites markedly reduces Ypk1 stability. These observations make it clear that TORC2 plays a key role in stimulating Ypk1 activity, up to the level needed by the cell to cope with the stresses of sphingolipid depletion, heat shock, hypotonicity, and other stressful conditions known to perturb PM structure and/or function.

In both animal cells (Yang et al. 2006; Lu et al. 2011) and fission yeast (Ikeda et al. 2008; Shiozaki et al. 2013), there is compelling evidence that the Avo1 orthologues (mSIN1 and Sin1, respectively) in TORC2 bind and are required for phosphorylation of the Ypk1 orthologues in these organisms (SGK1 and Gad8, respectively). Likewise, others have shown that S. cerevisiae Ypk2 also seems to interact with TORC2 via binding to Avo1 (Liao and Chen 2012). Moreover, based on cryo-EM, cross-linking MS, and other approaches, the current model of yeast TORC2 structure (Gaubitz et al. 2015; Gaubitz et al. 2016; Karuppasamy et al. 2017) suggests that Avo1 is located in close proximity to the active site of the Tor2-Lst8 complex. Furthermore, convincing data show that a sequence shared by Avo1 with both S. pombe Sin1 and mammalian mSIN1, designated the ―conserved region in the middle‖ (CRIM), is a discrete domain that adopts a stable ubiquitin-like fold with a prominent acidic loop and is necessary and sufficient for binding of Gad8 and SGK1, respectively (Tatebe et al. 2017). For example, CRIMSin1 fused to a different TORC2 subunit permits Gad8 hydrophobic motif phosphorylation in an S. pombe sin1∆ mutant. Therefore, it is likely that Ypk1 is recognized by TORC2 by binding to the CRIM element in Avo1.

Normally, two other proteins, Slm1 and Slm2 are required for TORC2 function, including Ypk1 phosphorylation at its C-terminal sites (Berchtold et al. 2012; Niles et al. 2012; Roelants et al. 2017b). However, Slm1 and Slm2 are peripheral subunits that dock quite far from the catalytic center of the complex (via their interaction with the non-essential subunits Bit61 and/or its paralogue Bit2, as well as Avo2 (Fadri et al. 2005; Wullschleger et al. 2005). So, another essential role, aside from substrate delivery, needs to be invoked to explain how these proteins contribute to Ypk1 stimulation. Aside from its CRIM domain, Avo1, which is tightly Tor2- bound, has a C-terminal PtdIns4,5P2-specific PH domain (Gallego et al. 2010; Vonkova et al. 2015) that is reportedly necessary for efficient PM localization of TORC2 (Berchtold and Walther 2009). Likewise, both Slm1 (and Slm2) have C-terminal PH domains (Yu et al. 2004; Daquinag et al. 2007), which have been shown to be specific for binding PtdIns4,5P2 (Yu et al. 2004; Gallego et al. 2010; Vonkova et al. 2015). It is possible, therefore, that Slm1 and Slm2 help reinforce stable PM recruitment of TORC2 and its positioning near Pkh1- and Pkh2-bound eisosomes, which, as discussed above, are located at PM sites enriched for PtdIns4,5P2. Alternatively, or in addition, Slm1 and Slm2 might help localize TORC2 nearby ER-PM junctions, which also control phosphoinositide metabolism and vice-versa (Omnus et al. 2016); in fact, several of the demonstrated substrates of Ypk1 are endoplasmic reticulum (ER)-localized proteins, as discussed below. The importance of these PH-domain containing proteins is highlighted by the finding that the level of PtdIns4,5P2 is important for TORC2 function; elevating PM PtdIns4,5P2 levels either by deleting the PM-localized synaptojanin Inp51 or overproducing the PM-localized PtdIns4P 5-kinase Mss4 rescues the temperature-sensitive lethality of the tor2-21ts allele (Morales-Johansson et al. 2004).

In this same regard, however, our laboratory has shown that truncated Avo1 (lacking its C- terminal PH domain) is still recruited to the PM and supports growth and, moreover, that even in cells lacking Avo1, Slm1 and Slm2, all of the remaining TORC2 subunits are still recruited to the PM (Martinez Marshall et al. 2019). An interpretation of this recent findings is that the PH domain of Avo1 exerts a negative constraint on its CRIM domain that prevents efficient Ypk1 recruitment and phosphorylation and, furthermore, that binding of PtdIns4,5P2 to the PH domain of Avo1 alleviates that inhibitory conformation, thereby promoting TORC2-mediated Ypk1 phosphorylation (Martinez Marshall et al. 2019).

4.3. N-Terminal Phosphorylation by Fpk1 and Fpk2 When examined by standard SDS-PAGE, especially under conditions known to improve the resolution of phospho-isoforms (75:1 or 100:1 acrylamide: N,N'-methylene-bis-acrylamide), Ypk1 migrates as a set of three or four distinct bands; however, none of these species are attributable to its Pkh1- or TORC2-dependent modification, based on mutation of the corresponding sites or on inactivation of the cognate protein kinases (Roelants et al. 2010). A genome-wide screen of a protein kinase mutant collection revealed that the protein kinases responsible for these observed Ypk1 mobility shifts are Fpk1 and its paralogue Fpk2/Kin82 (Roelants et al. 2010). Fpk1 phosphorylates Ser51 and Ser71 within a defined sequence context (-R-x-S>T-L/V/I/M/A-D/E-) (Mok et al. 2010; Roelants et al. 2010; Roelants et al. 2017b) in the N-terminal regulatory domain of Ypk1 (Figure 1.1). In contrast to the clear-cut experimental evidence both in vitro and in vivo that Ypk1-mediated phosphorylation of Fpk1 inhibits its

5 catalytic function, the effect of Fpk1-mediated phosphorylation on Ypk1 has been more difficult to discern.

5. Substrates of Ypk1

5.1. Control of Aminoglycerophospholipid Asymmetry in the Plasma Membrane Bilayer The first bona fide cellular targets of Ypk1 discovered were Fpk1 and Fpk2, protein kinases that phosphorylate and stimulate at least two PM-localized aminoglycerophospholipid flippases (Nakano et al. 2008). Flippases are P-type ATPases (class IV) that translocate lipid clients from the outer leaflet to the inner leaflet of the PM bilayer (Roland and Graham 2016). It was demonstrated that Fpk1 (and Fpk2) are phosphorylated robustly by Ypk1 in vitro and at the predicted sites and, likewise, that Fpk1 is phosphorylated in a Ypk1-dependent manner in vivo at the expected sites (Roelants et al. 2010), based on earlier studies of the phospho-acceptor site specificity of Ypk1 using selected synthetic peptide substrates (Casamayor et al 1999). The Ypk1 sites in Fpk1 and Fpk2, and in the other well-documented substrates discussed below, are compiled in Figure 1.2.

Moreover, it was shown that an Fpk1(S37A T244A S481A) mutant that is immune to Ypk1 phosphorylation is hyperactive both in vitro and in vivo and, conversely, that the absence of Fpk1 and Fpk2 suppresses the temperature-sensitive growth of ypk1ts ypk2Δ cells (Roelants et al. 2010). These results and other more recent findings (Roelants et al. 2017b) support the conclusion that TORC2-Ypk1 signaling down regulates the rate of inward aminoglycerophospholipid translocation (mainly PtdEth) by inhibiting the ability of Fpk1 (and Fpk2) to phosphorylate and thereby stimulate the flippases.

In addition to its negative regulation by Ypk1-mediated phosphorylation, Fpk1 activity in vivo depends on the presence of mannosyl-inositolphosphoryl-ceramide (MIPC), one of the major, endogenous complex sphingolipids that S. cerevisiae produces (although it is still not clear whether MIPC is a direct allosteric activator of Fpk1 or promotes Fpk1 activity via a more indirect mechanism) (Roelants et al. 2010). Overall, this regulatory circuit, with its interplay between Ypk1 and Fpk1, provides a mechanism to couple the rate of aminophospholipid flipping within the PM to the rate of synthesis of a complex sphingolipid. The more MIPC available, the faster the internalization of outer leaflet PtdEth. Maintaining this balance in PM lipid composition and leaflet distribution appears to be crucial for many aspects of membrane function because defects in flippase activity have been reported to cause problems in the vesicle-mediated transport of proteins in both the endocytic and exocytic pathways (Sebastian et al. 2012), in the non-vesicular trafficking of sterols (Muthusamy et al. 2009), and in maintaining the axis of polarized growth through effects on PM recruitment of Cdc42 (Saito et al. 2007; Das et al. 2012).

6 Fpk1(a) 32 RTRSISF Fpk1(b) 239 RERAGTV Fpk1(c) 476 RLRTKSF Fpk2 304 RLRTKSF Orm1(a) 48 RRRSSSI Orm1(b) 47 RRRRSSS Orm2(a) 43 RRRSSSV Orm2(b) 42 RRRRSSS Lac1(a) 18 RPRRKSS Lag1(a) 19 RRRNSSV Gpd1 19 RKRSSSS Fps1(a) 176 RRRSRSR Fps1(b) 180 RSRATSN Fps1(c) 565 RARRTSD Rod1(a) 133 RKRALST Rod1(b) 802 RSRSSSV Ysp2(a) 39 RTRKEST Ysp2(b) 513 RARSKTL Ysp2(c) 1232 RRRGKTV Lam4(a) 396 RRRAKSM

Lam4(b) 662 RFRSNSN

Figure 1.2. Selected sites of Ypk1 phosphorylation. The phospho-acceptor sites in the indicated gene products (position of first residue in the sequence shown indicated on the left) were validated as authentic Ypk1 phosphorylation sites both in vitro and in vivo. Documentation for these assignments can be found in the following publications: Fpk1 and Fpk2 (Roelants et al. 2010); Orm1 and Orm2 (Roelants et al. 2011); Lac1 and Lag1 (Muir et al. 2014); Gpd1 (Lee et al. 2012); Fps1 (Muir et al. 2015); Rod1 (Alvaro et al. 2016); and, Ysp2 and Lam4 (Roelants et al. 2018). These data illustrate the strict dependence on Arg residues at the -5 and -3 positions, the marked preference for Ser over Thr at the phospho-acceptor residue, and some preference for a hydrophobic (in bold) or an uncharged residue at the +1 position.

5.2. Control of Sphingolipid Biosynthesis The first clue that Ypk1 action was centrally involved in regulating sphingolipid production came from the fact that SLI2, one of the genes isolated in a screen for dosage suppressors of

7 ―SphingoLipid inhibition by ISP-1‖, was isogenic to YPK1 (Sun et al. 2000). It had been established previously that the antibiotic ISP-1, more commonly known as myriocin (Myr) (Miyake et al. 1995; Ikushiro et al. 2004; Yeung 2011), inhibits eukaryotic cell growth because it is a transition state mimic that potently blocks L-serine:palmitoyl-CoA C-palmitoyltransferase (decarboxylating) (SPT; Lcb1-Lcb2-Tsc3 heterotrimer in yeast), the first enzyme unique to the sphingolipid biosynthetic pathway in all eukaryotes (Dunn et al. 2004; Dickson et al. 2006; Megyeri et al. 2016; Olson et al. 2016). Overproduction of a Ypk1 mutant lacking catalytic activity did not suppress growth inhibition by Myr. These findings suggested that elevating the amount of Ypk1 activity was able to somehow overcome the limitation for sphingolipids caused by reducing the rate of their synthesis with the inhibitor; however, no targets of Ypk1 were identified. The next evidence that a protein kinase downstream of TORC2 was important for controlling the rate of sphingolipid production came from the observations, first, that an avo3- 30ts mutation, which cripples TORC2 function at the non-permissive temperature, also greatly diminished the amount of ceramides in the cell under the same conditions and, second, that expression of the hyperactive D239A allele of the Ypk1 paralogue Ypk2 was able to suppress the growth defect of avo3-30ts cells at the restrictive temperature (Aronova et al. 2008). It had already been established that Ypk2 was a downstream target of TORC2 (Kamada et al. 2005). These findings suggested that sphingolipid depletion stimulated TORC2-dependent phosphorylation of Ypk2, which, in turn, somehow promoted more sphingolipid production; however, no direct substrate of Ypk2 was pinpointed.

In 2002, the Thorner lab carried out a comprehensive screen for both dosage (gain-of-function) suppressors and transposon insertion (loss-of-function) suppressors of the temperature-sensitive lethality of a ypk1ts ypk2∆ strain (Roelants et al. 2002). The former potentially represents substrates whose function requires stimulation by Ypk1-mediated phosphorylation and the latter potentially represents substrates whose function requires inhibition by Ypk1-mediated phosphorylation. One of the best suppressors was a transposon insertion in open-reading-frame YLR350w (now designated ORM2). Orm2 and its paralogue Orm1 are small, ER-localized tetraspanins that physically associate with and act as negative regulators of SPT and, upon sphingolipid depletion, both Orm1 and Orm2 become heavily phosphorylated, which alleviates their inhibitory effect on SPT (Breslow et al. 2010). In 2011, the Thorner lab established that Ypk1 is the protein kinase responsible for phosphorylating Orm1 and Orm2 when sphingolipids become limiting, and the primary physiological role of Ypk1-mediated phosphorylation is to negatively regulate their function (Roelants et al. 2011); and, these conclusions were corroborated by others (Berchtold et al. 2012). Further support for the conclusion that regulation of sphingolipid biosynthesis is one of the most physiological important functions under TORC2 control were the findings that: (i) Ypk1 activation upon sphingolipid depletion required TORC2- mediated phosphorylation of Thr662 in its hydrophobic motif (Roelants et al. 2011); and, (ii) otherwise inviable tor2∆ spores could be recovered if they also carried orm1∆ orm2∆ mutations (Rispal et al. 2015). Thus, when under the stress of sphingolipid limitation, the cell responds by increasing metabolic flux into the sphingolipid biosynthetic pathway through the increase in SPT activity that is brought about via the TORC2-Ypk1-mediated phosphorylation and inhibition of the SPT inhibitors, Orm1 and Orm2.

In a subsequent global screen for additional candidate Ypk1 substrates using a combined bioinformatic, genetic, and biochemical screening procedure (Muir et al. 2014), the two catalytic

8 subunits (Lac1 and its paralogue Lag1) of the ceramide synthase complex (Lac1-Lag1-Lip1 heterotrimer in yeast), another ER-localized pacemaker enzyme in sphingolipid synthesis, were identified as direct substrates of Ypk1. In this instance, however, phosphorylation by Ypk1 stimulates the function of this enzyme. Furthermore, Muir et al. (2014) showed that the TORC2- Ypk1-driven increase in ceramide synthase activity ensures that the long-chain base precursors made by the TORC2-Ypk1-driven increase in SPT activity are efficiently channeled into the production of complex sphingolipid end-products of the yeast sphingolipid biosynthetic pathway (Muir et al. 2014). Exerting control at this step of the pathway also prevents accumulation of pathway intermediates that would otherwise compromise cell growth by stimulating autophagy (Zimmermann et al. 2013).

5.3. Control of Intracellular Glycerol Concentration Generation of high internal glycerol is the strategy yeast has evolved to provide a sufficient concentration of an innocuous intracellular osmolyte to combat the loss of water (Hohmann 2015). Gpd1, found in the cytosol and inside peroxisomes, is a dehydrogenase that catalyzes the reduction of dihydroxyacetone-phosphate (DHAP) to sn-glycerol-3-phosphate (Glo3P). Gpd1 is kept inactive via phosphorylation by Ypk1; however, TORC2-Ypk1 signaling is dramatically and rapidly decreased when cells are subjected to hyperosmotic shock (Lee et al. 2012). Thus, under conditions were Glo3P is needed for glycerol production, Ypk1-mediated inhibition of Gpd1 function is rapidly alleviated. The effector of a signaling pathway, the HOG (high osmolarity glycerol) response, that controls the processes needed to cope with hyperosmotic stress, including induction of the expression of appropriate genes, is the MAPK Hog1 (Saito and Posas 2012). Under hyperosmotic stress, GPD1 mRNA and protein expression are markedly upregulated in a Hog1-dependent manner (Albertyn et al. 1994), allowing the level of Glo3P for glycerol production to ramp up quickly. Glycerol is produced from Glo3P by the action of two dedicated phosphatases, Gpp1/Rhr2 and Gpp2/Hor2 (Hohmann 2015).

However, glycerol will only accumulate inside the cell if a PM-localized channel, the aquaglyceroporin Fps1, closes. Strikingly, in a proteome-wide screen (Muir et al. 2014), Fps1 was identified as a likely target of Ypk1. Subsequent characterization (Muir et al. 2015) found that Fps1 is an authentic Ypk1 substrate in vitro and in vivo, and that the open channel state of Fps1 requires its phosphorylation by Ypk1 at three sites (Muir et al. 2015). Under hyperosmotic conditions, where TORC2-Ypk1 signaling is rapidly decreased, Fps1 phosphorylation is lost, causing channel closure, glycerol accumulation, and enhanced survival under hyperosmotic stress (Muir et al. 2015). Thus, inactivation of TORC2-Ypk1 signaling upon hyperosmotic shock has two coordinated consequences that work synergistically to cause glycerol accumulation and promote cell survival, outcomes that work in conjunction with, but in a much more rapid and mechanistically distinct manner from, the processes evoked by activated Hog1. First, in less than one minute, loss of TORC2-Ypk1 signaling alleviates inhibition of Gpd1— which, combined with transcriptional induction of the genes for GPD1 and the phosphatases GPP1 and GPP2 — greatly increases the rate of glycerol production. Second, loss of TORC2-Ypk1 signaling also rapidly closes the Fps1 channel, thereby allowing the glycerol produced to be retained. These findings defined the underlying molecular basis of a previously uncharacterized mechanism for responding to hypertonic conditions (Muir et al. 2015).

9 Gpd1 has a closely related paralogue, Gpd2, which is located in the cytosol and inside the mitochondrion. Gpd2 is phosphorylated at the analogous site as that which Ypk1 phosphorylates in Gpd1 (Lee et al. 2012); however, a minor sequence change found in Gpd2 converts this sequence motif into one that is the target for Snf1, the yeast orthologue of mammalian 5'-AMP- activated protein kinase (Lee et al. 2012), and not Ypk1. Thus, phosphorylation site divergence places two closely-related metabolic enzymes, Gpd1 and Gpd2, under the control of two distinct classes of stress-activated protein kinases (Ypk1 and Snf1, respectively), demonstrating how even a minor sequence change can be a contributory driving force for functional specialization of the products of paralogous genes during evolution.

5.4. Control of Integral Plasma Membrane Protein Endocytosis In addition to modulating the levels and leaflet distribution of glycerophospholipids and the rate of sphingolipid biosynthesis, Ypk1 also influences PM homeostasis by down regulating the rate of endocytosis of integral membrane proteins in at least two ways. Using the α-factor pheromone receptor Ste2 as a model polytypic membrane protein, which is internalized in response to both its constitutive and agonist-induced ubiquitylation (Alvaro et al. 2014), the Thorner lab found that Ypk1-mediated phosphorylation at two sites blocked the ability of the endocytic adaptor (α- arrestin) Rod1/Art4 to mediate ligand-induced internalization of Ste2 (Alvaro et al. 2016). The α- arrestins promote ubiquitylation of their respective cargo molecules and their engagement by the clathrin-dependent endocytic machinery because these adaptors recruit the membrane-associated HECT-domain ubiquitin ligase Rsp5. The S. cerevisiae genome encodes 14 recognized - arrestins, most of which have been implicated in endocytosis and trafficking of a wide variety of nutrient permeases (Lin et al. 2008; Nikko and Pelham 2009; Becuwe et al. 2012). In addition to Rod1/Art4, the majority of these adaptors, including Ldb19/Art1 (1), Ecm21/Art2 (4), Aly2/Art3 (3), Ygr068/Art5 (1), Aly1/Art6 (1), Rog3/Art7 (2), Csr2/Art8 (3), Rim8/Art9 (1), and Ylr392c/Art10 (1), contain consensus Ypk1 phospho-acceptor motifs (number indicated in parentheses) that are highly conserved among the sensu stricto species or have been detected as phosphorylated in phospho-proteomic studies, or both and, thus, may be substrates of Ypk1.

The second mechanism by which Ypk1-mediated phosphorylation impedes endocytosis is via inhibition of Fpk1 and Fpk2. Fpk1 phosphorylates the actin patch protein Akl1 at two conserved C-terminal sites, and these modifications inhibit Akl1 (Roelants et al. 2017b). Under normal growth conditions, Fpk1-mediated inhibition of Akl1 prevents it from interfering with endocytosis. However, under conditions that upregulate TORC2-Ypk1 signaling, Fpk1 is inhibited by Ypk1-dependent phosphorylation, alleviating the Fpk1-mediated inhibition of Akl1, allowing Akl1 to phosphorylate and block the action of proteins needed for endocytic patch function (Roelants et al. 2017b).

So, overall, under stressful conditions that threaten PM integrity and activate TORC2-Ypk1 signaling (including sphingolipid limitation, heat shock, hypotonic conditions), the cell will decrease the rate of endocytic removal of some PM proteins because Ypk1 action blocks α- arrestin function and prevents Fpk1 from inhibiting Akl1. Conversely, upon hypertonic shock, where TORC2-Ypk1 signaling is shut off rapidly, PM proteins can be removed as part of the clearance of the ―excess‖ membrane created by cell shrinkage because α-arrestins remain fully functional and Fpk1 is able to prevent Akl1 from acting. Thus, by these mechanisms, TORC2- Ypk1 signaling adjusts the rate of endocytosis to meet the needs of the cell.

5.5. Control of Plasma Membrane Sterol Level In the same proteome-wide screen for Ypk1 substrates conducted in our laboratory by Muir et al. (2014), Ysp2 (also known as Lam2/Ltc4) was identified as another likely physiologically relevant target of Ypk1. Ysp2 was subsequently shown to be one of a new family of sterol- binding proteins located at plasma membrane (PM)–endoplasmic reticulum (ER) contact sites (Gatta et al. 2015). Our laboratory then documented that Ysp2 and its paralogue Lam4/Ltc3 are authentic Ypk1 substrates in vivo and showed using genetic and biochemical criteria that Ypk1- mediated phosphorylation inhibits the ability of these proteins to promote retrograde transport of ergosterol (the major sterol in S. cerevisiae) from the PM to the ER (Roelants et al. 2018). Furthermore, the results obtained also indicated that retention of ergosterol in the PM promotes cell survival under membrane-perturbing conditions known to activate TORC2-Ypk1 signaling (Roelants et al. 2018). These observations thus defined the underlying molecular basis of a new regulatory mechanism for cellular response to plasma membrane stress.

9. Rationale and Prospectus

For the reasons summarized above, I surmised that further investigation of the network of biochemical processes under the control of TORC2 and its effector protein kinase Ypk1 would continue to serve as an informative avenue for understanding molecular mechanisms in signal transduction and in the cell biology of stimulus-induced metabolic regulation.

Hence, in this dissertation, I first pursued the question of whether a suspected Ypk1 substrate the emerged from the global screen conducted in our laboratory (Muir et al. 2014), but that was not characterized any further, is an authentic target of Ypk1. This candidate substrate was Muk1, a guanine nucleotide exchange factor for the Rab5 class of small Ras-related GTPases. As I document here and have published (Locke and Thorner 2019a), I used various approaches and appropriate reagents (Chapter 2) to validate both in vitro and in vivo that Muk1 is indeed a physiologically relevant substrate of Ypk1 (Chapter 3). Moreover, my studies (Locke and Thorner 2019a) also demonstrated that Ypk1-mediated phosphorylation stimulates Muk1 function in vivo (Chapter 3). Unexpectedly, but importantly, I found further (Locke and Thorner 2019a) that yeast defective in generating active Rab5 have markedly reduced TORC2 activity, as judged by the decrease in TORC2-mediated phosphorylation of Ypk1 and other criteria (Chapter 4). Thus, my dissertation research uncovered a previously unappreciated role of Rab5 in a control circuit for positive regulation of TORC2-Ypk1 signaling (Chapter 4). While the importance of small GTPases (RAGs and RHEB) for the activity of TORC1 had been documented previously (Gonzalez and Hall 2017; Saxton and Sabatini 2017; Tatebe and Shiozaki 2017), my finding that Rab5 GTPases are important for TORC2 function provided an unanticipated and novel insight about the control of TORC2 activity. In the last section (Chapter 5), I discuss the implications of my findings and the additional mechanistic questions raised by my discovery that still remain to be addressed, which have also been published (Locke and Thorner, 2019b).

Finally, I provide two Appendices that describe other research that I accomplished. Appendix A presents studies I performed to delineate whether the determinants that dictate Ypk1 substrate selectivity are confined solely to its consensus phospho-acceptor site motif or may also involve

11 additional recognition of a more extended sequence element or even a completely different high- affinity binding site ("docking site"). Appendix B presents the results of a study I performed in animal cells in the laboratory of Prof. Michael Rape.

12 Chapter 2. Materials and Methods

Strains and Plasmids Saccharomyces cerevisiae strains (Table 1) were constructed using standard genetic methods (Amberg et al. 2005). For integrations into the genome, proper insertion was confirmed by PCR amplification and sequencing. Plasmids (Table 2) were constructed using standard molecular biology techniques (Green and Sambrook 2012) and verified by sequencing. Strains yMLT78, yMLT80, and yMLT85 were constructed using CRISPR-Cas9 (Finnigan and Thorner 2016) to integrate the coding sequence for 3xHA in-frame in place of the fourth and fifth amino acids (Tyr4 and Ile5) of the TOR2 open reading frame on chromosome XI using plasmid template pRS314-3xHA-TOR2 obtained from y2470 (Jiang and Broach 1999).

Table 1. S. cerevisiae strains used in this study Strain Genotype Source BY4741 MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0 Research Genetics yAM123A BY4741 Ypk1(L424A)::URA3 ypk2Δ::KanMX4 (Muir et al. 2014) JTY6142 BY4741 ypk1Δ::KanMX4 Research Genetics yMLT42 BY4741 vps21Δ::NatMX ypt52Δ::KanMX This study ypt53Δ::HphMX yMLT60 BY4741 ypt7Δ::KanMX This study yMLT9 BY4741 vps9Δ::HphMX muk1Δ::NatMX This study yMLT3 BY4741 vps9Δ::HphMX This study yMLT5 BY4741 muk1Δ::NatMX This study yMLT78 BY4741 3xHA-TOR2 This study yMLT86 BY4741 mup1Δ::HisMX MET15 This study yMLT87 BY4741 MET15 This study yAEA343 BY4741 vps9Δ::HphMX muk1Δ::NatMX MET15 This study yAEA344 BY4741 vps9Δ::HphMX MET15 This study yMLT64 BY4741 AVO3-GFP::KAN This study yMLT66 BY4741 AVO3-GFP::KAN vps9Δ::HphMX This study muk1Δ::NatMX y2470 BY4741 tor1::HIS3 tor2::LEU2 [pRS314(CEN4 (Jiang and Broach TRP1)-3xHA-TOR2] 1999) yMLT80 BY4741 AVO3-3xFLAG::KanMX 3xHA-TOR2 This study yMLT85 BY4741 AVO3-3xFLAG::KanMX 3xHA-TOR2 This study vps21Δ::HphMX ypt52Δ::NatMX

13 yMLT69 BY4741 AVO1-6xHA::HIS3MX6 This study yKL7 BY4741 AVO3-3xFLAG::KanMX K. Leskoske, this lab TOR2(L2178A)::Hyg yMLT70 BY4741 AVO1-6xHA::HIS3MX6 vps9Δ::HphMX This study muk1Δ::NatMX CGA84 MATa leu2Δ1::GEV::NatMX pep4∆:HIS3 prb1∆1.6R (Alvaro et al. 2014) ura3-52 trp1-1 lys2-801a leu2∆1 his3∆200 can1 GAL met15Δ0 ura3Δ0

Table 2. Plasmids used in this study Plasmid Description Source/reference

BG1805 2 μm, URA3, GAL1prom, C-terminal tandem affinity Open Biosystems (TAP) tag vector pMAL-c5X Bacterial expression vector for production of MBP New England (MalE) fusion proteins BioLabs pESC-Leu 2 μm, GAL1/10prom, LEU2 Agilent

YCpLG CEN, LEU2, GAL1/10prom vector (Bardwell et al. 1998) pRS425 2 μm, LEU2 vector (Sikorski and Hieter 1989) pRS426 2 μm, URA3 vector (Sikorski and Hieter 1989) pRS315 CEN, LEU2 vector (Sikorski and Hieter 1989) pET30a Bacterial expression vector Novagen pMLT35 pMAL-c5X MBP-Muk1-(His)6 This study pMLT74 pMAL-c5X MBP-Muk1(S171A S172A S173A S183A This study S184A S185A)-(His)6 pMLT22 pESC-LEU GAL1prom -Muk1-myc This study pMLT56 pESC-LEU GAL1prom -Muk1(1-220)-myc This study pMLT57 pESC-LEU GAL1prom -Muk1(1-220; S171A S172A This study S173A S183A S184A S185A)-myc pAX50 BG1805 Ypk1(L424A)-TAP (Muir et al., 2014) pFR252 pRS315 Ypk1(S51A T57A S71A T504A S644A (Muir et al., 2015) S653A T662A)-myc

14 pAM20 pRS315 YPK1-myc (Roelants et al., 2011) pMLT101 pRS426 GAL1prom -FLAG-Vps21 This study pMLT102 pRS426 GAL1 prom -FLAG-Vps21(S21L) This study pMLT103 pRS426 GAL1 prom -FLAG-vps21(Q66L) This study pMLT110 pRS426 GAL1 prom -FLAG-Ypt7 This study pMLT114 pRS426 GAL1 prom -FLAG-mNeonGreen This study pMLT111 pRS426 Msb3-FLAG This study pMLT83 pRS315 Muk1-3xFLAG This study pMLT84 pRS315 Muk1(S171A S172A S173A S183A S184A This study S185A)-3xFLAG pMLT85 pRS315 Muk1(D353A)-3xFLAG This study pMLT92 pRS315 Muk1(S171E S172E S173E S183E S184E This study S185E)-3xFLAG pMLT132 pET30a Ypk1(D242 D470A)-(His)6 This study pMLT135 pMAL-c5X MBP-Ypk1(603-680) [MBP-Ypk1CT] This study pMLT138 pMAL-c5X MBP-Ypk1(603-680; S644 S653A T662A This study S671A S672A S678) [MBP-Ypk1CT (6A)] pMLT88 pET30a Vps21-(His)6 This study pMLT141 pET30a Ypt7-(His)6 This study pPL495 pRS315 PHSlm1-Ypk1-3xHA (Niles et al., 2012) pMLT115 pRS425 GAL1prom-GST This study pMLT116 pRS425 GAL1 prom -GST-Ypk1 This study pMLT19 PET30a (His)6-MBP This study pMLT2 pMAL-c5X MBP-Fps1(1-255)-(His)6 This study pMLT3 pMAL-c5X MBP-Fps1(531-669)-(His)6 This study pMLT37 pMAL-c5X MBP-Fps1(531-559)-(His)6 This study pMLT38 pMAL-c5X MBP-Fps1(531-586)-(His)6 This study pMLT40 pMAL-c5X MBP-Fps1(607-669)-(His)6 This study pMLT48 pMAL-c5X MBP-Fps1(531-573)-(His)6 This study

15 pMLT60 pMAL-c5X MBP-Fps1(531-573; K561A, K562A, This study R563A, R565A)-(His)6 pMLT61 pMAL-c5X MBP-Fps1(554-573)-(His)6 This study pMLT53 pMAL-c5X MBP-Fps1(531-586; R565K R567K This study S570T)-(His)6 pMLT23 BG1805 YPK1(D242A, D470A)-TAP This study

Cell Culture Conditions Yeast strains were propagated in either standard rich medium (YP) containing 2% dextrose/glucose or a defined synthetic medium (SC) (Sherman, 2002) with the same carbon source and supplemented with appropriate nutrients to maintain selection for plasmids. For galactose induction of gene expression, cells were pre-grown in SC medium containing 2% raffinose-0.2% sucrose, and induced by addition of galactose (2% final concentration) for 3 h. To inhibit complex sphingolipid synthesis, a stock of AbA (Cat. # 630466; Takara Bio USA, Inc., Mountain View, CA) dissolved in ethanol was added at the indicated final concentrations and an equivalent volume of the same solvent was added to control cultures. Stocks of the amino acid analogs L-canavanine (Cat.# C1625; Sigma-Aldrich Chemical Company, St. Louis, MO) and L- ethionine (Cat.# E1260;Sigma) dissolved in water were added to plates of SD-containing agar medium at the indicated concentrations. For inhibition of analog- sensitive Ypk1(L424A), a stock of the adenine analog 1-(tert-butyl)-3-(3-methylbenzyl)-1H- pyrazolo[3,4-d]pyrimidin-4- amine (3MB-PP1; Cat.# 17860, Cayman Chemical Inc., Ann Arbor, MI) dissolved in dimethyl sulfoxide (DMSO) was added at the indicated final concentration and an equivalent volume of the same solvent was added to control cultures.

Protein Purification from Escherichia coli Expression from the corresponding plasmids (Table 2) of MBP-Muk1-(His)6, MBP-Muk1(6A)- (His)6, Vps21-(His)6, Ypt7-(His)6, and MBP-Fps1-(His)6 fragments in the E. coli derivative LOBSTR-BL21(DE3) (Andersen et al., 2013) was induced by addition of IPTG (0.75 mM final concentration). After incubation of the cultures at 16˚C for 16 h, the cells were harvested by centrifugation, lysed by sonication, and the corresponding proteins purified by immobilized metal ion affinity chromatography using Ni2+- charged nitrilotriacetate-derivatized agarose beads (Qiagen, Inc., Germantown, MD) following standard procedures. After elution with 200 mM imidazole, pooled eluates were dialyzed into PBS containing 2 mM DTT. For Vps21-(His)6, the dialyzed protein was further purified by ion exchange chromatography on a MonoQ 10/100GL column (Sigma-Aldrich) using a AKTA Fast Protein Liquid Chromatography system (GE Healthcare Life Science, Inc., Chicago, IL). To load purified Vps21-(His)6 or purified Ypt7- (His)6 with the desired guanine nucleotide, samples were incubated for 10 min in 5 mM EDTA to strip any bound Mg2+-nucleotide complexes, then incubated for 30 min with 2 mM GTP S (Roche Diagnostics, Inc, Indianapolis, IN), followed by addition of MgCl2 (20 mM final concentration) and incubation for an additional 30 min. MBP-Ypk1(603-680) [denoted MBP- Ypk1CT] and MBP-Ypk1CT(6A) were purified by affinity chromatography on amylose resin (New England BioLabs, Inc., Ipswich, MA), following the manufacturer's protocol.

16 Ypk1 Protein Purification from S. cerevisiae Yeast cells expressing the analog-sensitive Ypk1-as [pAX50 Ypk1(L424A)-TAP] were grown to mid-exponential phase, harvested, lysed and the Ypk1(L424A)-TAP protein purified by affinity purification using IgG-SepharoseTM and elution by cleavage with Pre-ScissionTM protease (GE Healthcare), as described previously (Muir et al., 2014) and in a more recent, highly detailed KD procedure (Locke et al. 2019). To purify Ypk1 , pMLT23 (2 µm DNA, GAL1prom- Ypk1(D242A,D470)-TAP) was introduced into yeast strain CGA84 by DNA-mediated transformation. Transformed cells were grown in 3L of SD-Leu + 2% dextrose medium and expression induced by addition of 1 µM β-estradiol (dissolved in ethanol) for 16 h. Cells were harvested by centrifugation, washed once in buffer containing 50 mM Tris-Cl pH 7.4, 125 mM KOAc, 0.5 mM EDTA, and frozen in liquid N2 in small pellets. Cells were lysed cryogenically using a Mixer Mill MM400 (Retsch). The resulting frozen yeast powder (typically, 5 g) was thawed in 10 mL TAP lysis buffer [50 mM Tris-Cl pH 7.4, 200 mM NaCl, 1.5 mM MgOAc, 1 mM DTT, 2 mM NaVO4, 10 mM NaF, 10 mM Na-PPi, 10 mM β-glycerol phosphate, 1x cOmplete protease inhibitorsTM (Roche), 0.15% Triton X-100]. The thawed lysate was first clarified by centrifugation at 15,000xg for 20 min at 4˚C. The resulting supernatant fraction was subjected to centrifugation at 100,000xg at 4˚C for 1 h. The resulting clarified extract was then filtered through a 0.45 µM polyethersulfone (PES) filter. The Ypk1-TAP fusion was affinity purified from the final filtered extract by incubation for 1 h with 300 µl IgG Sepharose (GE Healthcare) at 4˚C. Resin was washed extensively 5 times with TAP lysis buffer, and 1 time with Rhinovirus protease 3C (P3C) cleavage buffer [50 mM Tris-Cl pH 7.4, 200 mM NaCl, 1.5 mM MgOAc, 1 mM DTT, 10 mM NaF, 10 mM Na-PPi, 10 mM β-glycerol phosphate, 0.01% Triton X-100]. To elute proteins, resin was resuspended in 300 µl P3C cleavage buffer containing 8 units Prescission ProteaseTM [commercially-produced glutathione S-transferase (GST)-human rhinovirus 3C protease fusion protein] (GE Healthcare) for 5 h at 4˚C. The GST-protease 3C fusion protein was removed by the addition of 15 µl glutathione-sepharose beads (GE Healthcare), followed by centrifugation, and the supernatant fraction containing the purified Ypk1 was frozen as aliquots in liquid N2.

Ypk1 Protein Kinase Assays A detailed protocol for measurement of Ypk1 activity in vitro is available elsewhere (Locke et al. 2019). In brief, to monitor the protein kinase activity of Ypk1, samples of purified Ypk1-as (0.15 μg) were mixed with, typically, 1 μg of the purified substrate of interest in kinase assay buffer [125 mM potassium acetate, 12 mM MgCl2, 0.5 mM EDTA, 2 mM DTT, 1% glycerol, 0.2 mg/mL BSA, 12.5 mM β-glycerol phosphate, 1 mM NaVO4, 50 mM HEPES (pH 7.4) containing cOmpleteTM EDTA-free protease inhibitor cocktail (1 mini-tablet per 10 ml; Roche)] in the absence or presence of 1-(tert-butyl)-3-(3-methylbenzyl)-1H-pyrazolo[3,4-d]pyrimidin-4- amine (3MB-PP1; 10 μM final concentration), and reaction was initiated by addition of ATP (10 μM final concentration) containing 2 μCi [ -32P]ATP. After incubation at 30˚C for 30 min, reactions were terminated by the addition of an equal volume of 6x SDS gel sample buffer followed by boiling for 5 min and the resulting products resolved by SDS-PAGE and analyzed by both protein staining (0.05% Coomassie Blue R-250, 10% acetic acid, 25% isopropanol) and by autoradiography of the same fixed and stained gel performed using a Phosphorimager screen (Molecular Dynamics, Sunnyvale, CA) and a TyphoonTM imaging system (GE Healthcare).

17 Immuno-isolation of TORC2 and other FLAG-tagged Proteins For immunoprecipitation of FLAG-tagged proteins expressed in yeast, included TORC2 marked with Avo3-3xFLAG, frozen cell pellets were resuspended in 16 µl / A600 nm ice-cold lysis buffer [200 mM KCl, 10 mM magnesium acetate, 10 mM β-glycerol-phosphate, 10 mM NaF, 10% glycerol, 0.1 mM EDTA, 50 mM HEPES (pH 7.4) containing 2 mM 4-(2-aminoethyl) benzenesulfonyl fluoride (AEBSF) and cOmpleteTM EDTA-free protease inhibitor cocktail (1 mini-tablet per 10 ml; Roche)], mixed with an equivalent volume of chilled glass beads, and lysed by vigorous vortexing for 10 min at 4˚C. The resulting lysate was separated from the beads by draining after puncturing the conical tip of the plastic centrifuge tube with a 25-gauge hypodermic needle. The recovered liquid was diluted with 0.5 volume of lysis buffer containing 2% n-dodecyl β-D-maltoside (DDM; Cat.# D4641, Sigma) and incubated with gentle agitation on a rollerdrum for 20 min at 4˚C. The detergent-treated extract was clarified by centrifugation at maximum rpm in a microfuge (Model 5424, Eppendorf AG, Hamburg, Germany) and the resulting supernatant fraction was withdrawn and mixed with 15 µl of a slurry of magnetic affinity resin decorated with anti-FLAG M2 antibodies (Cat.# M8823, Sigma) that had been pre- equilibrated in lysis buffer containing 0.5% DDM. After incubation with resin for 2 h with gentle agitation on a rollerdrum at 4˚C, the resin was removed from the bulk solution and moved to the side of the tube using a magnet tube rack (Dynal, Thermo Fisher Scientific Inc., Waltham, MA), resuspended in and washed six times with 1 ml lysis buffer containing 0.5% DDM. For TORC2 purification, the remaining bound protein was eluted with 3xFLAG peptide (Sigma) in 1x kinase assay buffer and samples of the resulting eluate used immediately thereafter for in vitro kinase assays. For other FLAG-tagged proteins, the remaining bound protein was eluted with urea-SDS gel sample buffer [6 M urea, 6% SDS, 25% glycerol, 0.1% bromophenol blue, 150 mM Tris-HCl (pH 6.8)].

TORC2 Protein Kinase Assays A detailed protocol for measurement of TORC2 activity in vitro is available elsewhere (Locke and Thorner 2019c). In brief, to monitor the protein kinase activity of TORC2, samples of TORC2 (each derived by immuno-isolation from the equivalent of 50 µg of clarified lysate, as described above) were incubated with purified MBP-Ypk1CT or MBP-Ypk1CT(6A), in kinase assay buffer in the absence or presence of the indicated amounts of purified recombinant GTP S- loaded Vps21 or Ypt7, and reaction initiated by the addition of ATP (10 μM final concentration) containing 2 μCi [ -32P]ATP. Reactions were incubated at 30˚C for 30 min, unless otherwise noted, and the products analyzed in the same manner as for the Ypk1 protein kinase assays (see above), except SYPRO Ruby (ThermoFisher Scientific Inc.) was used for protein staining. For specific inhibition of WT TORC2, the enzyme was incubated with QL-IX-55 (gift of Nathanael Gray, Harvard Med. Sch.) dissolved in DMSO or, as a control, with an equivalent volume of the same solvent at 30˚C for 5 min prior to addition of the other reaction components. For specific inhibition of TORC2 containing Tor2(L2178A) (Kliegman et al. 2013), the enzyme was incubated with NVP- BEZ235 (Cayman Chemical) dissolved in DMF or, as a control, with an equivalent volume of the same solvent.

Preparation of Cell Extracts and Immunoblotting Whole-cell lysates of yeast were prepared by alkaline lysis with 1.85 M NaOH and 7.4% 2- mercaptoethanol followed by trichloroacetic acid (TCA) precipitation, as described in detail

18 previously (Westfall et al., 2008). For samples treated with calf-intestinal phosphatase (CIP), the TCA precipitates of whole-cell lysates were washed with ice-cold acetone, resuspended in 100 μl CIP resuspension buffer (126 mM sorbitol, 42 mM NaCl, 10.5 mM MgCl2, 0.24 mM EDTA, 2% 2-mercaptoethanol, 180 mM Tris Base), diluted with 900 μl 50 mM Tris-HCl (pH 8.0) and incubated with 45 units of CIP (New England Biolabs) for 2 h at 30˚C. Reactions were terminated by addition of TCA (25% final concentration) and the precipitated protein was resuspended in a buffer containing 0.1 M Tris Base and 5% SDS prior to addition of an appropriate volume of 5X SDS sample buffer.

In most cases, protein samples were separated by standard SDS-PAGE in gels containing the appropriate concentration of acrylamide to resolve the species of interest, transferred electrophoretically to nitrocellulose filter paper, and analyzed by immunoblotting. To resolve the spectrum of phospho-isoforms of particular phosphorylated proteins (such as Muk1 and Ypk1), resuspended whole-cell lysates were separated using Phos-tagTM SDS-PAGE (Kinoshita et al. 2009) in 8% acrylamide gels containing 35 μM Phos-Tag reagent (Fujifilm Wako Chemicals USA, Corp., Richmond, VA), according to manufacturer‘s instructions, and, after transfer to nitrocellulose, analyzed by immunoblotting.

Primary antibodies used in this study (at the dilution indicated) were: mouse anti-FLAG (1:5,000; Sigma, Product No. F1804); rabbit anti-DYKDDDDK (1:2,500; Cell Signaling Technologies, Inc. Danvers, MA, Product No. 2368); mouse anti-c-myc mAb 9E10 (1:100; Monoclonal Antibody Facility, Cancer Research Laboratory, University of California, Berkeley); mouse anti-HA 6E2 (1:5,000; Cell Signaling Technologies, Product No, 2367); rabbit anti-GST (1:10,000; Santa Cruz Biotechnology, Dallas, TX, Product No. SC-459); rabbit anti-Pgk1 (1:200,000) (Baum et al., 1978); rabbit anti-SGK1(pT256) (1:1,000; Santa Cruz Biotechnology, Product No. SC-16744); mouse anti-Pma1 (1:5,000; Abcam, Cambridge, UK, Product No. ab4645); and, rabbit anti-Ypk1(pT662) (1:20,000; gift from Ted Powers, Univ. of California, Davis, CA). Infrared dye-labeled secondary antibodies were: CF770-derivatized goat anti- mouse IgG (1:20,000; Biotium, Fremont, CA, Product No. 20077) and CF680-derivatized goat anti- rabbit IgG (1:20,000; Biotium, Product No. 20067). The resulting filter-bound immune complexes were visualized using an infrared imaging system (Model OdysseyTM CLx, LI-COR Biosciences, Inc., Lincoln, NE).

GST Pull-down Analysis Lysates were prepared in the same manner as for immunoprecipitation of FLAG-tagged proteins (see above). The resulting extracts were mixed with a slurry (10 μl) of glutathione-SepharoseTM 4B beads (GE Healthcare) pre-equilibrated in lysis buffer containing 0.5% DDM. After incubation for 2 h at 4˚C, the beads were separated from the protein solution by centrifugation, washed 6X with 1 ml lysis buffer, and the bound proteins eluted with urea-SDS gel sample buffer and analyzed.

Fluorescence Microscopy Yeast cells expressing Avo3-GFP (excitation max 489 nm; emission max 508) were grown to mid-exponential phase in synthetic medium lacking Trp and mounted onto 1% agarose pads and viewed immediately at 23˚C using a confocal microscope (Zeiss, Model LSM 710) equipped with a 100x oil immersion objective (1.40 NA) and excited with an argon laser at 488 nm at 25%

19 power and emission monitored using a bandpass filter (495-550 nm window). Images were acquired with an iXon3 EM-CCD camera (Andor, Belfast, UK) using Metamorph software (Molecular Devices, Sunnyvale, CA) and processed using ImageJ (Collins 2007).

MBP Pull-down Analysis MBP-Fps1-(His)6 proteins, purified by immobilized metal ion affinity chromatography using Ni2+-charged nitrilotriacetate-derivatized agarose resin, were adsorbed onto 30 µl amylose beads (New England Biolabs) by incubation in 300 µl binding buffer [150 mM NaCl, 5 mM KCl, 1.5 mM MgCl2, 20 mM HEPES pH 7.4, 0.1% Triton X-100, 0.1 µg/ml BSA] for 2 h rotating at 4˚C. The protein-decorated amylose beads were washed 2 times with 600 µl binding buffer without BSA. For pull-downs, 20 ng purified Ypk1KD was added to the protein-decorated beads in 300 µl binding buffer. After incubation with the beads for 2 h with gentle agitation on a rollerdrum at 4˚C, samples were washed 5 times with 400 µl binding buffer without BSA. Bound proteins were eluted by addition of 40 µl 2x urea-SDS gel sample buffer and analyzed by SDS-PAGE followed by immunoblotting.

20 Chapter 3. Rab5 GEF Muk1 is a physiologically important substrate of TORC2-activated protein kinase Ypk1

Background This chapter of my thesis is based on a primary research article that was published in Journal of Cell Biology in 2019 (Locke and Thorner, 2019a), for which I designed, executed, and analyzed all the experiments under the guidance of my mentor, Prof. Jeremy Thorner.

Among presumptive, but as yet uncharacterized, targets of Ypk1 that were identified by our group was Muk1 (Muir et al. 2014). Muk1 and its paralog Vps9 are guanine nucleotide exchange factors (GEFs) that catalyze GDP-GTP exchange in the Rab5 class of GTPases in S. cerevisiae, namely Vps21/Ypt51 and its paralogs Ypt52 and Ypt53 (Burd et al. 1996; Hama et al. 1999; Paulsel et al. 2013; Bean et al. 2015). As in other eukaryotes, the function of Rab5 GTPases was thought to be confined to promoting vesicle formation, trafficking and sorting in the early stages of the endocytic pathway (Singer-Kruger et al. 1994; Singer-Kruger et al. 1995; Cabrera and Ungermann 2013) and disruption of Rab5 function in yeast leads to accumulation of cargo in endosomes and disruption of their delivery to the vacuole (Horazdovsky et al. 1994; Gerrard et al. 2000). Muk1 and Vps9 have overlapping roles as Rab5 GEFs because a muk1∆ vps9∆ double mutant has a more severe phenotype than either single mutant and, tellingly, exhibits a phenotype closely resembling a vps21∆ ypt52∆ ypt53∆ triple mutant (Cabrera and Ungermann 2013; Paulsel et al. 2013). Although Muk1 and Vps9 both act on Rab5 GTPases, the presence of two distinct proteins suggests that they may be differentially regulated. Hence, I sought to determine whether Muk1 is indeed an authentic and physiologically relevant target of Ypk1 and, if so, the consequences of its Ypk1-mediated phosphorylation.

Results

Muk1 is a substrate of protein kinase Ypk1 A ~260-residue segment of Vps9 (451 residues) is necessary and sufficient for its Rab5 GEF activity (Carney et al. 2006; Barr and Lambright 2010; Bean et al. 2015). Compared to Vps9 itself (Gough et al. 2001), the catalytic (Vps9 homology) domain of Muk1 (612 residues) is split by an 83-residue insert that contains, in tandem, two matches (168RSRSSSG174 and 179RPRRSSS185) to the consensus phospho-acceptor site motif of Ypk1 (Mok et al. 2010; Muir et al. 2014) (Fig. 3.1A). These same sites are completely conserved among S. cerevisiae and all its sensu stricto relatives, as well as in more divergent yeast species (e.g. Candida glabrata, Kluyveromyces lactis and Ashbya gossypii) [see: portals.broadinstitute.org/cgi-bin/regev/ orthogroups/show_orthogroup.cgi?orf=YPL070W] and phosphorylation at these same sites in vivo has been detected in several global phosphoproteomic analyses of S. cerevisiae (Albuquerque et al. 2008; Holt et al. 2009; Swaney et al. 2013).

Figure 3.1. Muk1 is phosphorylated by Ypk1 in vivo. (A) Schematic depiction of Muk1. Dark purple, split catalytic (Vps9 homology) domain; yellow and underlined, consensus Ypk1 phospho- acceptor site; red, phosphorylated residues. (B) WT cells (BY4741) expressing Muk1-myc from the GAL promoter on a multi-copy (2 μm DNA) vector (pMLT22) were grown to mid-exponential phase, harvested, and lysed, and equivalent samples of the resulting extract protein were incubated in the absence (−) or presence (+) of CIP and, after treatment, resolved by SDS-PAGE in an 8% acrylamide gel and analyzed by immunoblotting (IB), all as described in Materials and methods. (C) WT (BY4741) or otherwise isogenic ypk1-as ypk2Δ (yAM123-A) cells expressing Muk1-myc as in B were grown to mid-exponential phase, treated with vehicle (DMSO) or 3MB-PP1 (10 μM final concentration) in the same solvent for 90 min, harvested, lysed, and examined as in B, except SDS- PAGE was conducted using a 7% acrylamide gel.

I confirmed that Muk1 is a phosphoprotein, first, by showing that an epitope-tagged derivative (Muk1-myc) migrates even in standard SDS-PAGE as a series of multiple, slower mobility isoforms that are collapsed to a faster mobility species by treatment with calf intestinal phosphatase (CIP) (Fig. 3.1B). To determine whether any isoforms were attributable to phosphorylation by Ypk1, I expressed Muk1-myc in either WT or otherwise isogenic cells lacking Ypk2 and carrying an allele of Ypk1, Ypk1(L424A) [hereafter Ypk1-as], sensitive to acute inhibition by the adenine analog 3MB-PP1 (Lopez et al. 2014; Muir et al. 2014). I found that only in the analog-sensitive cells and only in the presence of inhibitor were the slowest mobility species greatly reduced and the most hypophosphorylated species markedly increased (Fig. 3.1C), demonstrating that Muk1 is indeed phosphorylated in a Ypk1-dependent manner in the cell.

The Thorner lab has shown previously that in other Ypk1 substrates containing Ser residues juxtaposed (immediately upstream, immediately downstream, or both) to the primary phospho- acceptor Ser, such residues can also be phosphorylated by Ypk1 both in vivo and in vitro (Roelants et al. 2011; Muir et al. 2014). Hence, to test whether Ypk1-mediated phosphorylation of Muk1 was occurring at the expected sites, I examined the mobility pattern of WT Muk1 tagged with a 3xFLAG epitope or a derivative in which all six potential Ser residues at the Ypk1 consensus motifs were mutated to Ala. For this purpose, I used Phos-tag SDS-PAGE, a technique which enhances retardation of a protein depending on the extent of its phosphorylation (Kinoshita et al. 2009). I found that lack of the six Ser residues resulted in a marked reduction in

22 the slowest mobility isoforms and a pronounced increase in the fastest mobility (hypophosphorylated) isoform (Fig. 3.2), indicating that these sites are indeed phosphorylated in vivo.

Figure 3.2. Muk1 is phosphorylated at its Ypk1 consensus phospho-acceptor site motifs. WT (BY4741) cells expressing from the native MUK1 promoter on a low-copy (CEN) vector either Muk1-3xFLAG (pMLT83) or Muk1(6A)-3xFLAG (pMLT84) were grown to mid-exponential phase, harvested, and lysed, and equivalent amounts of protein from the resulting extracts were resolved by Phos-tag SDS-PAGE and analyzed by immunoblotting (IB; top) and by staining with Ponceau S dye (bottom) to confirm equal sample loading, all as described in Materials and Methods.

To more readily visualize Ypk1-specific phosphorylation of Muk1, I expressed a smaller fragment (residues 1-220) of WT Muk1-myc or an otherwise identical fragment in which the six potential Ser residues at the Ypk1 consensus motifs were mutated to Ala. I also took advantage of the fact that TORC2-dependent activation of Ypk1 is stimulated in response to inhibition of sphingolipid biosynthesis by antibiotics such as myriocin (Myr; which inactivates L- Ser:palmitoyl-CoA acyltransferase) (Miyake et al. 1995) or aureobasidin A (AbA; which inactivates inositolphosphate-ceramide synthase) (Sugimoto et al. 2004), as the Thorner lab and others have shown (Roelants et al. 2011; Berchtold et al. 2012; Khakhina et al. 2015). Indeed, when cells were treated with AbA, I found that the Muk1(1-220) fragment displayed a readily detectable slower mobility isoform that was eliminated by the 6A mutations or by absence of just Ypk1 itself (Fig. 3.3A) because under normal growth conditions very little Ypk2 is expressed (Gasch et al. 2000; Roelants et al. 2002). Moreover, as expected, this species appeared when cells were treated with aureobasidin, but not in its absence (Fig. 3.3B).

23 A

Supplemental material

Locke and Thorner, https:// doi .org/ 10 .1083/ jcb .201807154

Figure 3.3. Ypk1-dependent phosphorylation of Muk1 is stimulated by sphingolipid depletion. (A) WT (BY4741) or otherwise isogenic ypk1Δ (JTY6142) cells expressing from the GAL expression vector pMLT22 either Muk1(1–220; pMLT56) or Muk1(1–220 6A; pMLT57) were grown to mid-exponential phase, treated with AbA (1.8 µM concentration) for 2 h, harvested, and lysed, and the resulting extracts were resolved by SDS-PAGE using a 13.5% acrylamide gel and analyzed by immunoblotting. Pgk1 served as a loading control. (B) WT (BY4741) cells expressing either Muk1(1– 220; pMLT56) or Muk1(1–220 6A; pMLT57) were grown to mid-exponential phase, treated with 1.8 μM AbA or vehicle for 2 h, harvested, and lysed, and the resulting extracts were analyzed by SDS-PAGE on a 75:1 12% acrylamide gel. Pgk1 served as a loading control. Asterisk (red) indicates phosphorylated species.

Finally, purified recombinant Muk1, but not Muk1(6A), was robustly phosphorylated in vitro by purified Ypk1-as in the absence of 3-MB-PP1, but not in its presence (Fig. 3.4). Thus, Muk1 is a bona fide substrate of Ypk1.

Figure S1. Muk1 is phosphorylated at its Ypk1 consensus phospho-acceptor site motifs. (A) WT (BY4741) cells expressing from the native MUK1 promoter on a low-copy (CEN) vector either Muk1-3xFLAG (pMLT83) or Muk1(6A)-3xFLAG (pMLT84) were grown to mid-exponential phase, harvested, and lysed, and equivalent amounts of protein from the resulting extracts were resolved by Phos-tag SDS-PAGE and analyzed by immunoblotting (IB; top) and by staining with Ponceau S dye (bottom) to conf rm equal sample loading, all as described in Materials and methods. (B) WT (BY4741) cells expressing either Muk1(1– 220; pMLT56) or Muk1(1–220 6A; pMLT57) were grown to mid-exponential phase, treated with 1.8 µM AbA or vehicle for 2 h, harvested, and lysed, and the resulting extracts were analyzed by SDS-PAGE on a 75:1 12% acrylamide gel. Pgk1 served as a loading control. Asterisk (red) indicates phosphorylated species. (C) Ypk1-mediated phosphorylation stimulates the ability of Muk1 to support growth under heat stress. Samples of exponentially growing cultures of otherwise isogenic strains of the indicated genotype were plated in f vefold serial dilutions on appropriate selective growth medium and incubated at either 30° (lef ) or 37°C (right) for 3 d and then imaged.

Figure 3.4. Muk1 is phosphorylated by Ypk1 in vitro. Equivalent amounts (∼0.5 μg) of MBP- Muk1-(His)6 or MBP- Muk1(6A)-(His)6, expressed in and purified from E. coli, were incubated with

Locke and Thorner Jo urnal of Cell Biology S17 Stimulation of yeast TORC2 by Rab5 GTPases https://doi.org/10.1083/jcb.201807154 [γ-32P]ATP and Ypk1-as purified from S. cerevisiae in the absence (–) or presence (+) of 3MB-PP1, as described in Materials and Methods, and the resulting products were analyzed by both autoradiography (top) and staining with Coomassie Blue dye (bottom).

Muk1 function is stimulated by Ypk1-mediated phosphorylation The three yeast Rab5 isoforms (Vps21/Ypt51, Ypt52 and Ypt53) are involved in the early stages of the endocytic pathway, which is responsible for clearing integral PM membrane proteins, including amino acid permeases (Singer-Kruger et al. 1994; Cabrera and Ungermann 2013; Paulsel et al. 2013). Hence, defects in that process will increase the steady-state level of such permeases, enhancing uptake of cognate amino acid analogs, thus conferring on the cell greater sensitivity to the toxic effect of such compounds (Grenson et al. 1966; Lin et al. 2008) (Fig. 3.5A). Indeed, I found that compared to otherwise isogenic WT cells, cells lacking both Rab5 GEFs or all three Rab5 isoforms were markedly more sensitive to the growth-inhibitory action of canavanine, an Arg analog taken up by the Arg permease (Can1), whereas cells lacking Ypt7, the sole yeast ortholog of Rab7, which acts later in the endocytic pathway (Schimmoller and Riezman 1993; Rink et al. 2005) were not (Fig. 3.5B), indicating that the observed defect was specific for Rab5.

Figure 3.5. Ypk1-dependent phosphorylation of Muk1 promotes endocytosis of permeases. (A) Schematic depiction of how inefficient internalization of amino acid permeases that reside in the PM 25 results in increased sensitivity to the growth-inhibitory effect of a toxic amino acid analogue (small green circle). (B) Samples of exponentially growing cultures of otherwise isogenic strains of the indicated genotype were plated in five-fold serial dilutions on growth medium lacking Arg in the absence (–) or presence (+) of canavanine (5.7 μM final concentration). (C) Strains of the indicated genotype carrying an empty vector (EV; pRS315) or expressing from the same vector either Muk1- 3xFLAG (pMLT83) or the indicated Muk1 variants Muk1(6A)-3xFLAG (pMLT84), Muk1(6E)- 3xFLAG (pMLT92), or catalytically inactive Muk1(D353A)-3xFLAG (pMLT85), were grown and plated as in B. (D) Strains of the indicated genotype, as in C, were plated in five-fold serial dilutions on growth medium lacking Met in the absence (−) or presence (+) of ethionine (10 μM final concentration). (E) To confirm equivalent levels of expression, WT cells (BY4741) expressing the indicated Muk1-3xFLAG constructs, as in C and D, were grown to mid-exponential phase, harvested, and lysed, and proteins in the resulting extracts were treated with CIP, as in Materials and methods, resolved by SDS-PAGE on an 8% acrylamide gel, and analyzed by immunoblotting (IB).

This phenotype provided a means to assess the effect of Ypk1-mediated phosphorylation on Muk1 function. Indeed, I found that a muk1∆ vps9∆ double mutant conferred a greater degree of canavanine sensitivity than a vps9∆ mutation alone and that re-expression of Muk1 from its endogenous promoter on a CEN plasmid reproducibly restored readily detectable canavanine resistance to the muk1∆ vps9∆ cells (Fig. 3.5C). Moreover, a site-directed Muk1 mutant (D353A) that substituted a residue in its Vps9 homology domain known to be essential for catalytic function (Bean et al. 2015) was unable to confer canavanine resistance to the muk1∆ vps9∆ cells, demonstrating that the GEF function of Muk1 is required (Fig. 3.5C). Most telling, Muk1(6A), which cannot be phosphorylated on any of its Ypk1 sites was also unable to confer canavanine resistance to muk1∆ vps9∆ cells, whereas the phosphomimetic mutant Muk1(6E), which simulates full Ypk1-dependent phosphorylation at these same sites, was at least as potent as WT Muk1 at restoring canavanine resistance to the muk1∆ vps9∆ cells (Fig. 3.5C). I found the same trends when I examined the sensitivity of the same set of strains to the growth-inhibitory action of ethionine, a Met analog taken up by the Met permease (Mup1) (Isnard et al. 1996; Guiney et al. 2016) (Fig. 3.5D). In addition, I found exactly the same trends monitoring rescue of a completely independent phenotype, namely the poor growth of muk1∆ vps9∆ cells at an elevated temperature (37˚C) (Paulsel et al. 2013; Shideler et al. 2015) (Fig. 3.6). Furthermore, I confirmed that the difference in behavior between Muk1(6A) and Muk1(6E) could not be attributed to any difference in their relative level of expression, compared to each other or WT Muk1 (Fig. 3.5E). Taken together, these findings indicated that Ypk1-mediated phosphorylation of Muk1 is required for the function of this Rab5 GEF.

26 Figure 3.6. Ypk1-mediated phosphorylation stimulates the ability of Muk1 to support growth under heat stress. Samples of exponentially-growing cultures of otherwise isogenic strains of the indicated genotype were plated in five-fold serial dilutions on appropriate selective growth medium and incubated at either 30° (left) or 37°C (right) for 3 d and then imaged.

Discussion This study was initiated to determine whether I could extend our understanding of TORC2- Ypk1 signaling by establishing whether a candidate substrate for Ypk1 that emerged in a global screen (Muir et al., 2014) was authentic and physiologically relevant. As I have documented here, the Rab5-specific GEF Muk1 is indeed a target of Ypk1. Even though Muk1 is clearly phosphorylated by other protein kinases, which may influence other aspects of Muk1 function, our genetic evidence demonstrates that Ypk1-mediated phosphorylation is required for the GEF activity of Muk1 in vivo. S. cerevisiae contains just two, functional Rab5-specific GEFs that possess the Vps9 homology domain, Vps9 itself and Muk1 (Paulsel et al., 2013; Bean et al., 2015). The human genome encodes nine proteins with readily discernible Vps9 homology domains (Carney et al., 2006; Barr and Lambright, 2010); however, none of them are the obvious ortholog of either Vps9 or Muk1. It has been reported that SGK1, the mammalian homolog of Ypk1 (Casamayor et al., 1999), facilitates plasma membrane recycling of KCNQ1 channels in a Rab5-dependent manner, but whether that behavior was dependent on SGK1-mediated phosphorylation of any Rab5 GEF was not explored (Seebohm et al., 2008). Likewise, it has been reported that SGK1 function promotes association of a Rho- and Rac-specific GEF (ARHGEF2/GEF-H1) with a component (Sec5/EXOC2) of the mammalian exocyst complex, but again whether that behavior was dependent on SGK1-mediated phosphorylation of the GEF was not determined (Wang et al., 2015).

My finding that Muk1, an activator of Rab5 GTPases, is a substrate of the TORC2-Ypk1 pathway raises intriguing possibilities about the role of Muk1 phosphorylation. Cells lacking all three of this class of GTPases (Vps21/Ypt51, Ypt52 and Ypt53) are characterized by the accumulation of transport vesicles, defects in endosome maturation, and enlarged vacuoles (Horazdovsky et al. 1994; Singer-Kruger et al. 1994; Singer-Kruger et al. 1995; Gerrard et al. 2000; Cabrera and Ungermann 2013). Ypk1-mediated activation of the Rab5 GTPases via Muk1 phosphorylation may be a mechanism to adjust the flow of intracellular lipids and proteins to respond to cellular needs. As discussed in the next chapter, small GTPases have been identified as essential positive regulators of mTORC1. If the GTPases under Muk1 control are activators of TORC2, then an alternative, but not mutually exclusive, role for Muk1 phosphorylation by Ypk1 may be to act as a self-reinforcing positive feedback mechanism to sustain the level of TORC2 activity. As described in Chapter 4, I obtained compelling evidence that Vps21 serves as a direct positive effector that stimulates TORC2 activity.

27 Chapter 4. Rab5 GTPases are necessary for efficient TORC2-mediated Ypk1 activation

Background This chapter of my thesis is based on the primary research article that was published in Journal of Cell Biology in 2019 (Locke and Thorner, 2019a).

TORC1 function requires its interaction with both the Rheb (Avruch et al. 2009; Yang et al. 2017) and Rag (Sancak and Sabatini 2009; Nicastro et al. 2017) classes of small GTPases. In addition, Ryh1, a Schizosaccharomyces pombe GTPase that most closely resembles human Rab6 and its S. cerevisiae ortholog (Ypt6), was reported to stimulate TORC2 in fission yeast (Tatebe et al. 2010). Based on these precedents, my finding that Muk1, a guanine nucleotide exchange factor for the Rab5 small GTPases, is a substrate of TORC2-Ypk1 signaling pathway raised the intriguing possibility that Rab5 might have a role in modulating TORC2 activity in yeast. As demonstrated by the evidence presented in this chapter, I found that GTP-bound Vps21 is indeed required for optimal TORC2 function because it physically associates with TORC2 and directly activate its kinase activity.

Results

Rab5 GTPases are necessary for efficient TORC2-mediated Ypk1 activation I first sought to examine whether Rab5 function influenced TORC2 activity by monitoring its phosphorylation and activation of Ypk1. For that purpose, I utilized, first, a version of Ypk1 (Ypk17A) that contains only TORC2-dependent phosphorylation sites, which when examined by Phos-tag SDS-PAGE allows for convenient and reliable detection of the isoforms produced by its TORC2-mediated phosphorylation, as the Thorner lab has extensively documented elsewhere (Muir et al. 2015; Leskoske et al. 2017). Hence, I analyzed the phosphorylation pattern of the Ypk17A reporter in cells before and after treatment with AbA. TORC2-dependent phosphorylation of Ypk17A could be readily visualized in WT cells as a series of isoforms well resolved by Phos-tag SDS-PAGE, and inhibition of complex sphingolipid biosynthesis with AbA stimulated the appearance of the slowest-mobility (most highly phosphorylated) isoform (Fig. 4.1A, left), in accord with our prior observations with Myr (Muir et al. 2015; Leskoske et al. 2017). Strikingly, however, in the absence of both Rab5 GEFs, basal and AbA-induced TORC2 phosphorylation of Ypk17A were both markedly reduced, and loss of Vps9 (the major Rab5 GEF) alone had a stronger effect than loss of Muk1 (Fig. 4.1A). Likewise, compared to WT control cells, basal and AbA-induced TORC2 phosphorylation of Ypk17A was dramatically decreased in cells lacking all three Rab5 paralogs (Fig. 4.1B). Thus, in cells that cannot generate active Rab5, TORC2-mediated phosphorylation of Ypk1 was very inefficient.

Figure 4.1 Rab5 GTPases and their GTP loading are necessary for efficient phosphorylation and activation of Ypk1 by TORC2. (A) WT (BY4741) or otherwise isogenic yMLT9 (muk1Δ vps9Δ), yMLT3 (vps9Δ), or yMLT5 (muk1Δ) cells, all expressing Ypk17A-myc from its native promoter on a CEN plasmid (pFR252), were grown to mid-exponential phase and treated with ethanol or an equivalent volume of the same solvent containing AbA (1.8 µM final concentration). After incubation for 2 h, the cells were harvested, and extracts were prepared, resolved by Phos-tag SDS-PAGE, transferred to a nitrocellulose filter, stained with Ponceau S, and then analyzed by immunoblotting (IB), as in Materials and methods. (B) As in A, except WT (BY4741) and yMLT42 (vps21Δ ypt52Δ ypt53Δ) cells were compared. (C) Cultures of WT (BY4741) and JTY6142 (ypk1Δ) cells were plated in fivefold serial dilutions on growth medium in the absence (–) or presence (+) of AbA (40 nM final concentration). (D) As in C, for WT (BY4741), yMLT9 (muk1Δ vps9Δ), yMLT42 (vps21Δ ypt52Δ ypt53Δ), and yMLT42 (ypt7Δ) cells. (E) As in C, for WT (BY4741) cells transformed with either empty multicopy vector (EV; pRS426) or the same vector expressing from its native promoter Msb3-FLAG (pMLT111).

Phosphorylation by TORC2-activated Ypk1 of multiple targets upregulates flux through the sphingolipid biosynthetic pathway and is required for the cell to survive inhibition of ceramide synthesis caused by the presence of Myr (Roelants et al. 2011; Berchtold et al. 2012; Muir et al. 2014; Leskoske et al. 2017). Similarly, as expected, I also found that Ypk1 function is required for the cell to survive the challenge of the depletion of complex sphingolipids caused by AbA treatment (Fig. 4.1C). Thus, the degree of resistance that cells display to agents like Myr and AbA provides a convenient phenotypic read-out for the efficiency of Ypk1 activation by

29 TORC2. By this criterion, and in agreement with our biochemical analysis (Fig. 4.1A and B), cells lacking both Rab5 GEFs and cells lacking all three Rab5 paralogs were much more sensitive to the growth-inhibitory action of AbA than otherwise isogenic WT control cells (Fig. 4.1D), whereas cells devoid of Ypt7 (as an additional control for specificity) were not. Thus, cells that cannot generate active Rab5 exhibit physiological behavior consistent with inefficient TORC2-mediated phosphorylation of Ypk1. As a further test of this conclusion, I examined cells overexpressing Msb3, one of two, apparent Rab5-specific GAPs encoded in the S. cerevisiae genome. Although Msb3 has a paralog (Msb4), only Msb3 had demonstrable Rab5 GAP activity in vivo and in vitro in two independent studies (Lachmann et al. 2012; Nickerson et al. 2012). I reasoned, therefore, that overexpressed Msb3 should lower the steady-state pool of activated (GTP-bound) Rab5 in the cell. Indeed, I found that such cells grew significantly less robustly on AbA than the control cells harboring the empty vector (Fig. 4.1E). Collectively, these data indicate that Rab5 in its GTP-bound state is necessary to support TORC2 phosphorylation and activation of Ypk1.

Active Rab5 stimulates TORC2 phosphorylation of Ypk1 As a further test of the conclusion that activated Rab5 promotes TORC2 phosphorylation of Ypk1, I examined the effect of re-expressing in the Rab5-deficient vps21∆ ypt52∆ ypt53∆ triple mutant either WT Vps21, or Vps21(Q66L) [a GTP hydrolysis-defective mutant (so-called "GTP- locked" variant)], or Vps21(S21L) [a mutant that preferentially binds GDP (so-called "GDP- locked" variant)] (Tall et al. 1999; Plemel et al. 2011; Lo et al. 2012). I focused on Vps21 because it is constitutively expressed and the most abundant yeast Rab5 isoform (Schmidt et al. 2017), has been shown to play the major role in endocytic vesicle trafficking (Horazdovsky et al. 1994), and, when absent, causes more pronounced phenotypes compared to loss of the other two Rab5 GTPases (Singer-Kruger et al. 1994; Nickerson et al. 2012; Nakatsukasa et al. 2014). I found that, when present as the sole Rab5, WT Vps21 was able to rescue AbA-induced TORC2- phosphorylation of Ypk17A (Fig. 4.2). This rescue was reproducibly more efficacious in cells expressing GTP-locked Vps21, but undetectable in cells expressing GDP-locked Vps21 (Fig. 4.2), consistent with the need for a Rab5 GTPase to be in its GTP-bound state to promote TORC2 phosphorylation of Ypk1. The weaker rescue conferred by WT Vps21 in the vps21∆ ypt52∆ ypt53∆ mutant cells may be attributable to its reduced expression relative to GTP-locked Vps21. Alternatively, the high level of GAL promoter-driven expression could overwhelm the capacity of the Rab5 GEFs, thereby lowering the fraction of total WT Vps21 in the GTP-bound state. By contrast, once loaded, GTP-locked Vps21 persists in its active state. In any event, because GTP-locked Vps21 cannot cycle and is defective in endocytic trafficking (Tall et al. 1999; Lachmann et al. 2012), its ability to stimulate TORC2 phosphorylation of Ypk1 appears to be a direct effect of Vps21-GTP and not some secondary consequence of its roles in vesicle trafficking per se. Nonetheless, I explored whether the reduction in TORC2 phosphorylation of Ypk1 in cells deficient in GTP-Rab5 might be attributable to such indirect effects.

Fig. 4.2. GTP-bound Vps21 stimulates TORC2 phosphorylation of Ypk1. Equivalent samples of cultures of WT (BY4741) or yMLT42 (vps21∆ ypt52∆ ypt53∆) cells expressing Ypk17A-myc from its native promoter on a CEN plasmid (pFR252), and carrying either empty vector ( - ; pRS426) or co-expressing FLAG-Vps21 (pMLT101), FLAG-Vps21Q66L (pMLT103), or FLAG-Vps21S21L (pMLT102) from the GAL promoter on the same vector, were grown to mid-exponential phase in 2% raffinose-0.2% sucrose medium, shifted for 1 h to galactose medium (2% final concentration), then treated for 2 h with ethanol (-) or an equivalent volume of the same solvent containing AbA (1.8 μM final concentration; +), harvested, lysed, resolved by Phos-tag SDS-PAGE and analyzed by immunoblotting (IB).

Lack of GTP-Rab5 does not cause mislocalization of either TORC2 or Ypk1 Deletion of the C-terminal PH domain in Avo1, an essential subunit of TORC2, reportedly rendered cells inviable, but they could be rescued by expression of an Avo1 construct in which a CaaX box was substituted for its PH domain (Berchtold and Walther 2009), suggesting that association with the PM is crucial for TORC2 function. A more recent study from our laboratory has shown that Avo3, not Avo1, is pivotal for TORC2 assembly and its association with the PM (Martinez Marshall et al. 2019). Nonetheless, the conclusion that TORC2 must associate with the PM to be functional is still valid. Given that Rab5 GTPases are important in endocytic vesicle internalization, and also have been implicated in recycling of endosomes back to the PM (Jovic et al. 2010; Feyder et al. 2015; MacDonald and Piper 2016), it seemed possible that cells deficient in GTP-bound Rab5 might have decreased TORC2 activity simply owing to lack of a sufficient pool of TORC2 at the PM. However, when I analyzed the subcellular distribution of TORC2, monitored by fluorescence microscopy in cells marked with Avo3-GFP, an essential TORC2 subunit very tightly bound to the catalytic subunit Tor2 (Wullschleger et al. 2005; Karuppasamy et al. 2017; Martinez Marshall et al. 2019), I saw no difference in PM localization of TORC2 between WT and muk1Δ vps9Δ mutant cells (Fig. 4.3A). Thus, TORC2 does not appear to be mislocalized in the absence of active Rab5.

Figure 4.3. Absence of Rab5 GEFs does not alter TORC2 localization at the cell cortex. Avo3 is an essential and very tightly bound subunit of TORC2. Hence, to visualize TORC2, either WT cells (yMLT64) or an otherwise isogenic strain lacking both Muk1 and Vps9 (yMLT66), as indicated, expressing Avo3-GFP from its native promoter at its endogenous locus on chromosome V, were examined. After growing the cultures to mid-exponential phase, samples were mounted on agarose pads and imaged using a confocal fluorescence microscope, as described in Materials and methods. Scale bar, 2 μm.

Alternatively, it seemed possible that the decrease in TORC2 phosphorylation of Ypk1 observed in cells deficient in GTP-bound Rab5 might be due to diminished accessibility of Ypk1 at the PM. For its proper folding and function, Ypk1 is obligatorily phosphorylated on T504 in its activation loop by Pkh1 (and Pkh2); (Roelants et al. 2002; Roelants et al. 2004), which are protein kinases very tightly bound to PM structures called eisosomes (Walther et al. 2007; Douglas and Konopka 2014). However, I saw no difference in Ypk1 activation loop phosphorylation between WT and muk1Δ vps9Δ mutant cells (Fig. 4.4A), suggesting that there is no impairment in Ypk1 availability at the PM. Similarly, substantial data support the conclusion that interaction with the CRIM element in the Avo1 subunit of TORC2 is responsible for substrate presentation to the active site in the Tor2 catalytic subunit of TORC2 (Liao and Chen 2012; Tatebe et al. 2017; Yao et al. 2017), and I saw no significant difference in Ypk1

32 association with Avo1 between WT and muk1Δ vps9Δ mutant cells in GST pull-down experiments (Fig. 4.4B). Thus, lack of efficient phosphorylation by TORC2 in cells deficient in GTP-Rab5 cannot be explained by the inability of Ypk1 to encounter TORC2 at the PM. As a further test, I expressed PHSlm1-Ypk1-3xHA, which others have demonstrated is efficiently targeted to the PM via its PtdIns4,5P2-binding PH domain and serves as an effective substrate of TORC2 (Niles et al. 2012). I found, however, that TORC2-dependent phosphorylation of the C- terminal regulatory tail in PHSlm1-Ypk1-3xHA was still poor in cells deficient in GTP-Rab5 (Fig. 4.4C). Therefore, I explored whether there might be a more direct, physical interaction between Vps21 and TORC2, or between Vps21 and Ypk1.

Figure 4.4. Absence of Rab5 GEFs does not alter Ypk1 accessibility to either Pkh1 or TORC2. (A) Absence of Rab5 GEFs does not impair activation loop phosphorylation of Ypk1 by eisosome- associated Pkh1/2. Either WT (BY4741) cells or otherwise isogenic muk1Δ vps9Δ (yMLT9) cells were transformed with Ypk1-myc (pAM20; Roelants et al. 2011), grown to mid-exponential phase, harvested, and lysed, and equivalent amounts of protein in the resulting extracts were resolved by SDS-PAGE and analyzed by immunoblotting (IB) with the indicated antibodies. Our lab has documented before (Roelants et al., 2010) that commercial antibodies raised against the activation loop residue in mammalian SGK1 (anti-pT256 SGK1, sc-16744; Santa Cruz Biotechnology) phosphorylated by PDK1 specifically detect the homologous site (pT504) in the activation loop of Ypk1 phosphorylated by Pkh1 (or Pkh2; Roelants et al. 2002). (B) Absence of Rab5 GEFs does not 33 impair Ypk1-Avo1 association. Otherwise WT (yMLT69) or isogenic muk1Δ vps9Δ (yMLT70) cells expressing Avo1-6xHA expressed from its endogenous locus and coexpressing from the GAL promoter on a plasmid either GST alone (pMLT115) or GST- Ypk1 (pMLT116) were grown to mid- exponential phase, induced with galactose (2% final concentration) for 3 h, harvested, and lysed; the GST proteins in the extracts were isolated by adsorption to glutathione-agarose (see Materials and methods); and the bound proteins were resolved by SDS-PAGE and analyzed by immunoblotting. (C) Forced PM association mediated by a PtdIns4,5P2-specifc PH domain does not restore TORC2- dependent phosphorylation of the Ypk1 C-terminal regulatory tail in cells deficient in Rab5 GEFs. As in A, except the cells were transformed with a plasmid (pPL495) expressing PHSlm1-Ypk1-3xHA (pPL495; Niles et al., 2012), treated with AbA (1.8 μM), and blotted with an antiserum that detects phosphorylation at one of the primary C-terminal sites (T662) in Ypk1 that is phosphorylated specifically by TORC2 (Niles et al. 2012; Leskoske et al. 2017).

Vps21 associates with Tor2 in TORC2 As one means to test whether Vps21 and TORC2 are able to physically associate, I examined whether they were able to coimmunoprecipitate. For this purpose, I first expressed either FLAG- Vps21 or FLAG-mNeonGreen (Shaner et al. 2013), in cells coexpressing a 3xHA epitope-tagged version of the Tor2 catalytic subunit of TORC2, either from a CEN plasmid in tor1Δ tor2Δ cells (Fig. 4.5A) or integrated at the TOR2 locus on chromosome XI (Fig. 4.5B). I chose mNeonGreen as an initial control because mNeonGreen (237 residues) is quite comparable in size to Vps21 (210 residues), as well as quite similar in overall surface charge (pI = 5.7 for mNeonGreen versus pI = 5.0 for Vps21]. In multiple trials, when extracts of the cells were immunoprecipitated with anti-FLAG antibodies, I reproducibly found readily detectable coimmunoprecipitation of 3xHA-tagged Tor2 from the cells expressing FLAG-Vps21, but not from the cells expressing FLAG-mNeonGreen (Fig. 4.5A and B). I conducted similar immunoprecipitation experiments using cells expressing either FLAG-Vps21 or FLAG-mNeonGreen and coexpressing Ypk1-myc, because phosphorylation and activation of certain other AGC family protein kinases requires association of their regulatory domains with other classes of small GTPases (Pearce et al. 2010; Leroux et al. 2018). However, only an equivalent level of nonspecific background interaction was observed for both FLAG-Vps21 and FLAG- mNeonGreen with Ypk1-myc (Fig. 4.5C). Thus, these initial experiments indicated that Vps21 associates with TORC2, not Ypk1. I found that both WT and GTP-locked Vps21 were efficacious in pulling down Tor2, but I also observed detectable association with GDP-locked Vps21 (Fig. 4.5D), similar to prior reports showing that both GTP-locked Rheb and GDP-locked Rheb bound to mTORC1 (Long et al. 2005).

Figure 4.5. Vps21 physically interacts with TORC2, but not Ypk1. (A) Strain y2470 (tor1Δ tor2Δ) expressing 3xHA-Tor2 from a CEN plasmid was transformed with plasmids expressing either FLAG-mNG (pMLT114) or FLAG-Vps21 (pMLT101). The resulting transformants were grown to mid-exponential phase, and expression of the FLAG-tagged proteins was induced with 2% galactose for 4 h. The induced cells were harvested and lysed, and FLAG-tagged proteins were immuno- isolated from the extracts using anti-FLAG resin as described in Materials and methods. Samples of the bound proteins were resolved by SDS-PAGE on 8% gels and analyzed by immunoblotting (IB) with anti-HA antibodies and on 13% gels and analyzed by immunoblotting with anti-FLAG antibodies. IP, immunoprecipitation. (B) Strain yMLT78 expressing 3xHA-Tor2 from its endogenous locus was transformed with the same plasmids and treated as in A, except that proteins bound to the anti-FLAG resin were resolved by SDS-PAGE on a 4–20% gradient gel. (C) WT cells (BY4741) expressing Ypk1-myc (pAM20) from a LEU2-marked CEN plasmid and coexpressing either FLAG-Vps21 (pMLT101) or FLAG-mNEON (pMLT114) from a URA3-marked 2 μm vector were examined as in A. (D) Strain yMLT78 was transformed with plasmids expressing either FLAG- Vps21 (pMLT101), FLAG-Vps21(S21L; pMLT102), or FLAG-Vps21(Q66L; pMLT103), and proteins were immunoprecipitated, resolved, and analyzed as in B.

35 Finally, as a critical control for specificity, I repeated these same experiments using both FLAG- Vps21 and FLAG-Ypt7, and found again that only Vps21 exhibited a readily detectable association with Tor2 (Fig. 4.6A), indicating that the observed interaction does not occur with an unrelated small GTPase. Because both Vps21 and Ypt7 are membrane-anchored proteins, this specificity made it highly unlikely that Vps21-TORC2 interaction resulted from the fact that both Vps21 and TORC2 are membrane associated. To rule out this possibility, I directly documented that the level of detergent present in the cell lysis and immunoprecipitation buffers, as described in Materials and Methods, was sufficient to fully solubilize both proteins (Fig. 4.6B).

Figure 4.6. Interaction of Vps21 with Tor2 in TORC2 is specific. (A) Strain yMLT78 expressing 3xHA-Tor2 from its endogenous locus was transformed with plasmids expressing either FLAG-Ypt7 (pMLT110) or FLAG-Vps21 (pMLT101). The resulting transformants were grown to mid- exponential phase, and expression of the FLAG-tagged proteins was induced with 2% galactose for 4 h. The induced cells were harvested and lysed, and FLAG-tagged proteins were immuno-isolated from the extracts using anti-FLAG resin as described in Materials and methods. Samples of the bound proteins were resolved by SDS-PAGE on a 4–20% gradient gel analyzed by immunoblotting (IB) with anti-HA and anti-FLAG antibodies. (B) Strain yMLT78 expressing 3xHA-Tor2 from its endogenous locus was transformed with a plasmid expressing FLAG-Vps21 (pMLT101). The resulting transformants were cultured and harvested as in A. Cells were lysed without detergent. Samples were split and treated with buffer containing 0.67% DDM or no detergent to solubilize proteins for 30 min before clarifying lysates by centrifugation, as described in Materials and methods. IP, immunoprecipitation; T, total lysate; P, pellet fraction; S, supernatant fraction. Protein extracts were resolved and analyzed as in A, with the indicated antibodies.

Rab5 GTPase stimulates TORC2 activity Given the interaction between Vps21 and Tor2 that I detected by coimmunoprecipitation, I explored the possibility that this association, whether stable or transient, is able to stimulate the catalytic function of TORC2, analogous to the stimulation of mTORC1 activity by Rheb in mammalian cells (Avruch et al. 2009; Yang et al. 2017). In vegetatively growing and unstressed

36 cells, Ypt53 expression is repressed (Schmidt et al. 2017). Therefore, to obtain TORC2 from cells depleted of Rab5 GTPases, I used anti-FLAG antibodies to immuno-isolate TORC2 from extracts of vps21Δ ypt52Δ mutant cells expressing Avo3-3xFLAG, which I was able to demonstrate also successfully enriched for the Tor2 catalytic subunit of TORC2 (Fig. 4.7A). I used the identical approach to isolate TORC2 from otherwise isogenic WT cells (Fig. 4.7A). Regardless of the strain background, our procedure using FLAG-tagged Avo3 for immuno- isolation recovered TORC2 complexes that contain equivalent amounts of HA-tagged Tor2. As the substrate, I used a purified recombinant protein in which the 78 C-terminal residues of Ypk1 (680 residues), containing all of its TORC2 phosphorylation sites (Leskoske et al. 2017), was fused to the C-terminus of maltose-binding protein (MBP) (Duplay et al. 1984), hereafter referred to as MBP-Ypk1CT.

I first incubated TORC2 from WT cells with MBP-Ypk1CT and [γ-32P]ATP and found robust incorporation, as expected. I found that this incorporation was almost completely eliminated when I used the mutant substrate MBP-Ypk1CT(6A), in which all six of the TORC2-specific phospho-acceptor residues were mutated to Ala, confirming that the observed phosphorylation was occurring at the previously defined TORC2-specific sites in Ypk1 (Fig. 4.7B). TORC2 itself, and not some other protein kinase present in the preparation, was responsible for the observed phosphorylation because incubation with a compound (QL-IX- 55) specifically designed and demonstrated by others to selectively inhibit yeast TORC2 (Liu et al. 2012) markedly reduced the incorporation into MBP-Ypk1CT (Fig. 4.7C). Likewise, when I prepared TORC2 in the same way from cells expressing a Tor2 allele (Tor2(L2178A)) designed (Kliegman et al. 2013) to be specifically sensitive to an orthogonal inhibitor (NVP-BEZ235), incorporation into MBP-Ypk1CT was almost completely eliminated (Fig. 4.7D). These controls established that the observed phosphorylation of MBP-Ypk1CT was occurring in a strictly TORC2-dependent manner. I further established that, under our assay conditions, incorporation was roughly linear with time of incubation (Fig. 4.7E and F).

Figure 4.7. Establishment of a reliable in vitro assay for TORC2-mediated phosphorylation of a Ypk1-derived substrate. (A) WT (yMLT80) and yMLT85 (vps21Δ ypt52Δ) strains expressing both Avo3-3xFLAG and 3xHA-Tor2 from their endogenous loci were grown to mid-exponential phase, harvested, and lysed, and TORC2 was immuno-enriched from the resulting extracts as described in Materials and methods. The immuno-isolated TORC2 was analyzed by SDS-PAGE and immunoblotting (IB). (B) TORC2 isolated from WT cells was incubated with [γ-32P]ATP and recombinant MBP-Ypk1CT(603–680) containing its TORC2 phosphorylation sites or MBP- Ypk1CT(6A) with the TORC2 sites mutated to Ala. Products were separated on SDS-PAGE and analyzed by Coomassie blue staining and autoradiography. (C) TORC2 isolated from WT cells was

38 incubated with [γ-32P]ATP and recombinant MBP-Ypk1CT in the presence or absence of the TORC2 inhibitor QL-IX-55 (500 μM) or vehicle (DMSO). (D) TORC2 isolated from TORC2-as (yKL7) cells was incubated with [γ-32P]ATP and recombinant MBP-Ypk1CT in the presence or absence of the NVP-BEZ235 (500 μM) or vehicle (DMF). (E) Equivalent amounts of TORC2 immunopurified from WT (yMLT80) or yMLT85 (vps21Δ ypt52Δ) strains were incubated with [γ-32P]ATP and recombinant MBP-Ypk1CT. Reactions were terminated at the indicated time points. (F) Quantitation of the 32P autoradiogram in E.

Having thoroughly validated the specificity of this in vitro reaction, I then compared TORC2 isolated from WT cells to TORC2 isolated from the vps21Δ ypt52Δ cells and found, in multiple trials, that the specific activity of the WT TORC2 was reproducibly higher (Fig. 4.8A and B). Moreover, addition of GTP-loaded Vps21 to the TORC2 isolated from vps21Δ ypt52Δ mutant cells reproducibly stimulated incorporation into MBP-Ypk1CT (Fig. 4.8C), whereas under the exact same assay conditions, GTP-loaded Ypt7 did not (Fig. 4.8D). The most parsimonious model to explain these in vitro results is that GTP-bound Rab5 associates with Tor2 and causes a conformational change in TORC2 that allows it to access the C-terminal end of Ypk1 more efficiently, fully consistent with all of the in vivo evidence presented in this study that active (GTP-bound) Rab5 GTPases promote TORC2 activity toward its substrate Ypk1 (Fig. 4.5).

Figure 4.8. Vps21 stimulates TORC2 activity in vitro. (A) Equivalent amounts of TORC2 immunopurified from WT (yMLT80) or yMLT85 (vps21Δ ypt52Δ) strains (see Fig. 4.7 A) were added in increasing amounts and incubated with [γ-32P]ATP and MBP-Ypk1CT. The reaction products were resolved by SDS-PAGE and analyzed by autoradiography and staining with SYPRO Ruby, as described in Materials and methods. Results of one representative experiment are shown. (B) The results of multiple experiments (n = 3) as in A are plotted as the mean 32P 39 incorporation into MBP-Ypk1CT versus the amount of TORC2 complex present. Error bars, SEM. 32P incorporation was measured by quantifying the corresponding autoradiograms using ImageJ and estimating the amount of TORC2 present from the staining intensity of the SYPRO Ruby-stained gel and accompanying standards (not depicted). The values were then normalized to the mean 32P incorporation into MBP-Ypk1CTcatalyzed by the lowest amount of TORC2 from the vps21Δ ypt52Δ cells, which was set at [1]. (C) TORC2 immunopurified from yMLT85 (vps21Δ ypt52), denoted TORC2ΔΔ, was incubated with [γ-32P]ATP and MBP-Ypk1CT in the absence or presence of Vps21-GTPγS (0.2 µg and 0.4 µg). (D) Equivalent amounts of TORC2 immunopurified from yMLT85 (vps21Δ ypt52Δ) as in A, denoted TORC2ΔΔ, were incubated with [γ-32P]ATP, MBP- Ypk1CT, and either Vps21-GTPγS (2 µg) or Ypt7-GTPγS (4 µg), as indicated. Reactions were terminated at the indicated time points.

Discussion Having established TORC2-Ypk1–mediated activation of the Rab5-specific GEF Muk1, I explored what the role of that pathway might be. It is now well established that TORC1 localization and activity are influenced by small GTPases of the Rag and Rheb classes, respectively. The Rag GTPases tether TORC1 at the surface of the lysosome/vacuole through interaction with the RAPTOR (yeast counterpart Kog1/Las24) subunit (Sancak and Sabatini 2009; Nicastro et al. 2017). By contrast, recent structural work (Yang et al. 2017) has shown that TORC2 activation by Rheb (Avruch et al. 2009) is mediated by its direct interaction with a pocket constituted by the M- and N-HEAT repeats and FAT domain of mTOR. Whether small GTPases modulate TORC2 in any similar fashion has been more elusive. It has been reported that mammalian Rab35 affects TORC2 signaling to the AGC kinase AKT, but only indirectly through the effect of Rab35 on intracellular trafficking of PtdIns 3-kinase (Wheeler et al. 2015). The closest S. cerevisiae orthologue of mammalian Rab35 is Ypt1, which clearly functions in the ER-to-Golgi step of the yeast secretory pathway (Novick 2016), not at the PM. Evidence for a more direct involvement of a small GTPase in TORC2 regulation was provided for fission yeast Ryh1 (Tatebe et al. 2010). However, the closest S. cerevisiae orthologue of Ryh1 is Ypt6 (mammalian counterpart Rab6), which functions in cis- to trans-Golgi vesicle trafficking (Suda et al. 2013), not at the PM. I was quite surprised, therefore, when I observed that deficiencies of active GTP-bound Rab5— due to absence of the Vps9 and Muk1 GEFs, to absence of all three Rab5 isoforms (Vps21, Ypt52, and Ypt53), or to elevation of the Rab5-specific GAP Msb3 —all lowered both basal and stress-induced TORC2-mediated phosphorylation of Ypk1, as assessed biochemically, genetically, or both. Thus, in S. cerevisiae, Rab5 GTPases are an important positive effector of TORC2 function.

Yeast TORC2 components are located at the cell cortex in close apposition to the PM, but are highly dynamic (Berchtold and Walther 2009; Leskoske et al. 2018; Martinez Marshall et al 2019). Muk1 and its target Vps21 have both been localized to endosomes (Cabrera and Ungermann 2013; Paulsel et al. 2013; Bean et al. 2015; Shideler et al. 2015). My findings that Muk1 is a substrate of the TORC2-Ypk1 kinase cascade and that Vps21 is required for full TORC2 activity raise the possibilities that the observed dynamic behavior of yeast TORC2 reflects, first, that a fraction of the population of active TORC2 is on endosomes and, second, that a fraction of the GTP- bound Vps21 is at the PM. Similarly, by monitoring a fluorescent mSIN1 reporter (S. cerevisiae counterpart Avo1), it was observed in mammalian cells that active mTORC2 could be found at both the PM and on endosomes (Ebner et al. 2017).

40 On the other hand, despite the fact that Rab5 function is involved in endosome internalization, trafficking, and recycling, I could find no impairment in TORC2 localization in cells deficient in active Rab5. So, unlike Rag GTPases, Rab5 does not seem to have a role in tethering TORC2 to the PM per se. Rather, our data, especially the ability of Vps21 to co-immunoprecipitate with Tor2, even in the presence of detergent, and the ability of GTP-loaded Vps21 to stimulate TORC2-mediated phosphorylation of Ypk1 in vitro, indicate a more direct role for GTP-bound Rab5 in TORC2 function, as suggested for Ryh1 in fission yeast (Tatebe et al. 2010). Even though S. pombe Ryh1 has greater overall relatedness to S. cerevisiae Ypt6 (63% identity; 76% similarity) than to S. cerevisiae Vps21 (37% identity; 61% similarity), it is the case that mammalian Rab5B, Vps21, and Ryh1 all share with each other greater resemblance in their Switch I, Switch II, and hypervariable loop regions than they do with other classes of Ras superfamily GTPases (Fig. 4.9). In the case of Rheb, it is residues mainly in its Switch I and Switch II regions that make the main contacts with the M- and N-heat repeat regions and FAT domain in mTOR (Yang et al. 2017). Thus, it is possible that Rab5 binds to the Tor2 subunit of yeast TORC2 in the same binding pocket as that occupied by Rheb in the mTOR subunit in mTORC1. In both yeast and mammals, aside from Lst8, the ancillary subunits in TORC2 are all completely different from those in TORC1 (Eltschinger and Loewith 2016; Tatebe and Shiozaki 2017), and interaction of these other components with the catalytic subunit likely modifies the geometry of the binding pocket, so that it accommodates the right GTPase effector and excludes spurious ones. Furthermore, yeast and other fungi encode two highly related, but not identical, TOR isoforms; Tor2 is the primary catalytic subunit in TORC2, whereas Tor1 is the primary catalytic subunit in TORC1 (Helliwell et al. 1994; Roelants et al. 2017a). Hence, subtle sequence differences between Tor2 and Tor1 in their M- and N-heat repeat regions and FAT domain (Hill et al. 2018) may further fine-tune the pocket in Tor2 to be selective for Rab5.

Figure 4.9. Sequence alignment of the indicated small GTPases. Sequence identities shared among related classes of GTPase are indicated by a white letter on a black box, and standard conservative substitutions are indicated by a bold letter on a gray box. Hs, Homo sapiens; Sc, S. cerevisiae; Sp, S. pombe. The similarities in the Switch I, Switch II, and hypervariable loop regions of SpRyh1 and ScVps21 and HsRab5B are indicated by green boxes.

Regardless of the detailed molecular mechanism by which Rab5 stimulates the function of TORC2, Muk1 is activated in a TORC2-dependent and Ypk1-mediated manner, presumably generating more GTP-bound Rab5 (Fig. 4.10). Given that I found that active Rab5 promotes TORC2 function, this control circuit provides self-reinforcing positive feedback that would lead to sustained TORC2-Ypk1 signaling, once it has been initiated. In this same regard, using the split-YFP (Venus) system to tag Vps21 and Msb3 for bimolecular fluorescence complementation, which traps protein–protein complexes (Sung and Huh 2007; Kerppola 2009), a substantial fraction of this Rab5-GAP complex was found at the PM, not on endosomes (Lachmann et al. 2012), indicating that Msb3 may be responsible for deactivation of the PM- associated pool of GTP-bound Vps21, thus providing a mechanism for down-regulation of Rab5- mediated stimulation of TORC2.

Figure 4.10. Muk1-dependent Rab5-mediated mechanism for sustained activation of TORC2- Ypk1 signaling. Basal Ypk1 (orange) activity requires phosphorylation of T504 in its activation loop (indicated by the left-most P) and TORC2-mediated phosphorylation at S644 near its C- terminus. However, under many stress conditions, TORC2 (aqua) up-regulates Ypk1 by 42 phosphorylating T662 and additional multiple sites at its C-terminal end (indicated by the cluster of Ps to the right). As documented in this study, activated Ypk1 phosphorylates the Rab5-specific GEF Muk1 (purple), which in turn is activated by its Ypk1-mediated phosphorylation. Activated Muk1 will generate more GTP-bound Rab5 (pink), and as I also have demonstrated here, activated Rab5 promotes the ability of TORC2 to phosphorylate Ypk1. Based on the observed coimmunoprecipitation of the Rab5 Vps21 with the Tor2 catalytic subunit of TORC2, the stimulation of TORC by Rab5 may be direct.

Chapter 5. Perspective— Regulation of TORC2 function and localization by Rab5 GTPases in Saccharomyces cerevisiae

Background This chapter of my dissertation is based on a review article that has been accepted for publication in the journal Cell Cycle (Locke and Thorner, 2019b), for which I designed, executed, and analyzed all the experiments under the guidance of my mentor, Prof. Jeremy Thorner.

In brief, the evolutionarily conserved Target of Rapamycin (TOR) complex-2 (TORC2) is an essential regulator of plasma membrane homeostasis in budding yeast (Saccharomyces cerevisiae). In this yeast, TORC2 phosphorylates and activates the effector protein kinase Ypk1 and its paralog Ypk2. These protein kinases, in turn, carry out all the crucial functions of TORC2 by phosphorylating and thereby controlling the activity of at least a dozen downstream substrates. A previously uncharacterized interplay between the Rab5 GTPases and TORC2 signaling was uncovered through my analysis of a new suspected Ypk1 target. I found that Muk1, one of two guanine nucleotide exchange factors for the Rab5 GTPases, is a physiologically relevant Ypk1 substrate; and, my genetic analysis indicated that Ypk1-mediated phosphorylation activates the guanine nucleotide exchange activity of Muk1. Second, I was able to demonstrate both in vivo and in vitro that the GTP-bound state of the Rab5 GTPase Vps21/Ypt51 physically associates with TORC2 and acts as a direct positive effector required for full TORC2 activity. These interrelationships provide a self-reinforcing control circuit for sustained up-regulation of TORC2-Ypk1 signaling. In this chapter, I summarize the experimental basis of these findings, their implications, and speculate as to the molecular basis for Rab5- mediated TORC2 activation.

Overview In organisms from yeasts to humans, essential aspects of cellular physiology are coordinated by the actions of two, evolutionarily conserved, multi-component protein kinase complexes in which the catalytic subunit is the large Target of Rapamycin (TOR) polypeptide (Eltschinger and Loewith 2016; Gonzalez and Hall 2017; Saxton and Sabatini 2017; Tatebe and Shiozaki 2017). These two TOR-containing complexes— TOR complex 1 (TORC1) and TOR complex 2 (TORC2)— are spatially and functionally distinct. Active TORC1 resides on the cytosolic surface of the lysosome in animal cells and on the surface of its counterpart, the vacuole, in yeast (Saccharomyces cerevisiae), whereas active TORC2 resides mainly on the plasma membrane (PM). Animal cells express a single TOR protein that populates both complexes; in yeast, however, there are two TOR paralogs, Tor1 and Tor2. TORC1 is functional when it contains either Tor1 or Tor2 as its catalytic subunit; but to be functional, TORC2 must contain Tor2 as its catalytic subunit (Loewith et al. 2002; Wedaman et al. 2003). Hence, a tor1∆ mutant is viable, but a tor2∆ mutant is not (Kunz et al. 1993; Helliwell et al. 1994). The activity of TORC1 can be acutely inhibited by rapamycin and related compounds (Dowling et al. 2010), whereas TORC2 is largely immune to these inhibitors (Gaubitz et al. 2015), which has made analysis of TORC1 function more readily accessible than analysis of TORC2 function. However, the demonstration that the primary essential function of TORC2 is to phosphorylate two, paralogous, AGC-family protein kinases, Ypk1 and Ypk2, which then execute all of the critical downstream functions, has

44 dramatically increased our understanding of the physiological roles of TORC2 (Roelants et al. 2017a).

As with other AGC kinases, the basal activity of Ypk1 requires phosphorylation of a conserved Thr residue (T504) in its activation loop within its catalytic domain. This modification is installed by two paralogous protein kinases, Pkh1 and Pkh2 (Roelants et al. 2002; Roelants et al. 2004), which are stably associated components of the protein coats of PM invaginations called eisosomes (Walther et al. 2006; Foderaro et al. 2017). Basal activity and stability of Ypk1 also requires its phosphorylation at S644, which lies within a conserved sequence (turn motif) located downstream of its kinase homology domain within a C-terminal regulatory domain. This modification is installed by TORC2, which is also largely PM-associated (Kunz et al. 2000; Sturgill et al. 2008; Berchtold and Walther 2009; Leskoske et al. 2018). Under certain stressful conditions that stimulate TORC2-mediated phosphorylation of Ypk1, such as sphingolipid limitation (Roelants et al. 2011), heat stress (Sun et al. 2012), hypertonic conditions (Niles et al. 2012), and acetic acid stress (Guerreiro et al. 2016), TORC2 further elevates Ypk1 activity by phosphorylating four additional sites in its C-terminal regulatory domain, paramount among them is T662, which lies within another conserved sequence (hydrophobic motif) in the C- terminal domain (Roelants et al. 2004; Leskoske et al. 2017). Phosphorylation of Ypk1 at these locations further enhances both its activity and stability. Under other stressful conditions, such as hypertonic shock (Lee et al. 2012; Muir et al. 2015), treatments that damage the cell wall (Leskoske et al. 2018), and treatments that decrease "membrane tension" (Riggi et al. 2018), TORC2-mediated phosphorylation of Ypk1 is dramatically reduced.

S. cerevisiae TORC2 comprises four essential core subunits (Avo1, Avo3, Lst8 and Tor2) (Loewith and Hall 2011), two classes of non-essential peripherally-associated subunits (Avo2 and Bit61 and its paralog Bit2) (Eltschinger and Loewith 2016; Gaubitz et al. 2016; Karuppasamy et al. 2017), and two, essential ancillary subunits (Slm1 and Slm2) that undergo dynamic shuttling between the eisosomes and TORC2 (Berchtold et al. 2012; Niles et al. 2012). The tertiary fold of the kinase domain of the catalytic subunit Tor2 is stabilized by its tight association with the β-propeller protein Lst8 (which also binds to Tor1). Tor2 is also intimately entwined with Avo1 and Avo3 (Wullschleger et al. 2005; Karuppasamy et al. 2017), to form a dimeric rhombohedral complex (Wullschleger et al. 2005; Karuppasamy et al. 2017), creating the scaffold onto which the other TORC2 components dock.

Based on a cryo-EM-derived structure of S. cerevisiae TORC2 (Karuppasamy et al. 2017), Avo1 appears to be located in close proximity to the active site of the Tor2-Lst8 complex. Furthermore, convincing biochemical evidence shows that a sequence in Avo1 shared with its Schizosaccharomyces pombe ortholog Sin1 and its mammalian counterpart (mSIN1), designated the ―conserved region in the middle‖ (CRIM), is the sequence element that binds the corresponding Ypk1 orthologs in these organisms, Gad8 (Tatebe et al. 2017) and both SGK1 (Casamayor et al. 1999) and AKT1 (Yang et al. 2006), and presents them to the TOR kinase for phosphorylation. Therefore, by analogy, Ypk1 is likely to be recognized as a substrate for TORC2 by its binding to the CRIM element in Avo1. In this regard, although slm1∆ slm2∆ cells are inviable, fusion of the PtdIns4,5P2-binding PH domain of Slm1 (Fadri et al. 2005; Tabuchi et al. 2006) to Ypk1 restores viability to slm1∆ slm2∆ cells (Niles et al. 2012), suggesting that,

45 normally, one function of the Slm1 proteins is to promote, somehow, the Avo1-mediated recognition of Ypk1 by TORC2 at the PM.

Muk1 emerges as a substrate for Ypk1 Various approaches have been used to identify physiologically relevant substrates of the TORC2-Ypk1 signaling axis, including genetic methods (Roelants et al. 2002), biochemical analysis (Roelants et al. 2010; Roelants et al. 2011), chemogenetic strategies (Kliegman et al. 2013; Rispal et al. 2015; Bourgoint et al. 2018), a genome-wide candidate screen (Muir et al. 2014), and global phosphoproteomics (Frohlich et al. 2016). As summarized in a recent comprehensive review (Roelants et al. 2017a), among the thoroughly validated direct substrates of Ypk1 identified from these studies are: (a) two protein kinases (Fpk1 and Fpk2) whose role is to phosphorylate and thereby stimulate both PM- (Dnf1 and Dnf2) and trans-Golgi- (Dnf3) localized aminoglycerophospholipid flippases; Ypk1-mediated phosphorylation inhibits Fpk1 and Fpk2; (b) one of the two glycerol-3P dehydrogenase isoforms (Gpd1) whose role is to supply sn-glycerol-3P for both glycerophospholipid synthesis and glycerol production; Ypk1-mediated phosphorylation inhibits Gpd1; (c) the major aquaglycerolporin (Fps1), which is a channel required for glycerol efflux; Ypk1-mediated phosphorylation of Fps1 is required to keep this channel in its open state; (d) two, endoplasmic reticulum (ER)-localized tetraspanins (Orm1 and Orm2) whose role is to inhibit the enzyme that catalyzes the first committed step in sphingolipid biosynthesis; Ypk1-mediated phosphorylation blocks the inhibitory functions of Orm1 and Orm2, thereby stimulating metabolic flux into the sphingolipid pathway; (e) the heterodimeric ceramide synthase (Lac1-Lag1) required for generation of sphingolipids; Ypk1-mediated phosphorylation stimulates the activity of Lac1-Lag1, thereby further promoting production of sphingolipids; (f) at least two α-arrestins (Art3/Aly2 and Art4/Rod1), which are adaptors necessary to target the HECT domain ubiquitin ligase Rsp3 to specific integral membrane protein clients to promote their endocytosis; Ypk1-mediated phosphorylation inhibits the function of these proteins; and (g) two paralogous StARkin domain–containing proteins (Ysp2/Lam2/Ltc4 and its paralog Lam4/Ltc3) located at PM-endoplasmic reticulum (ER) contact sites, which participate in the retrograde movement of ergosterol from the PM to the ER; Ypk1- mediated phosphorylation inhibits the ability of these proteins to promote this retrograde sterol transport.

This cohort of substrates demonstrates that TORC2-Ypk1 signaling is essential for growth and viability because it controls virtually every aspect of PM homeostasis— sphingolipid and glycerolipid production, PM lipid asymmetry, the concentration of an intracellular osmolyte (glycerol), PM protein composition, and sterol content. In a global screen for candidate Ypk1 targets (Muir et al. 2014), Muk1, one of two Rab5-specific guanine nucleotide exchange factors (GEFs) encoded in the S. cerevisiae genome (the other is Vps9), was identified among potential targets of Ypk1. I confirmed that Muk1 is indeed a bona fide substrate of Ypk1 (Locke and Thorner 2019a). I demonstrated that Muk1 is phosphorylated in a Ypk1-dependent manner both in vivo and in vitro and, under either condition, is phosphorylated by Ypk1 at its two consensus Ypk1 phospho-acceptor motifs (RSRSSSG and RPRRSSS). Moreover, using three different phenotypic screens in vivo, I found that a corresponding Muk1(6A) mutant was a total loss-of- function allele, whereas the cognate phosphomimetic allele, Muk1(6E), was somewhat hyperactive compared to wild-type Muk1.

46 These results predicted that stimulation of TORC2-Ypk1 signaling should result in a Muk1- dependent increase in the pool of active (GTP-bound) Rab5 in the cells. In this regard, I was intrigued by the possibility that one or more of the three Rab5 GTPases encoded in S. cerevisiae genome (Ypt51/Vps21, Ypt52 and Ypt53) might serve as a direct modulator of TORC2 function, based on two precedents. First, the function of the other TOR-containing complex, TORC1, requires its interaction with two other classes of small GTPases, both RHEB (Avruch et al. 2009; Yang et al. 2017) and RAGs (Sancak and Sabatini 2009; Nicastro et al. 2017). Second, it was reported, largely on the basis of genetic findings, that GTP-bound Ryh1 (a small GTPase that most closely resembles human Rab6 and its S. cerevisiae ortholog Ypt6) stimulates TORC2 in fission yeast (Tatebe et al. 2010).

Rab5 GTPases emerge as regulators of TORC2 For the reasons above, I examined whether Rab5 function influenced TORC2 activity, as assessed by monitoring the TORC2-mediated phosphorylation and activation of Ypk1. Strikingly, I found, first, that yeast lacking its two Rab5 GEFs (Muk1 and Vps9) or its three Rab5 paralogs (Vps21/Ypt51, Ypt52 and Ypt53) or overexpressing Msb3, a Rab5-directed GTPase-activating protein (GAP), all exhibited pronounced reduction in TORC2-mediated phosphorylation and activation of Ypk1 (Locke and Thorner 2019a). Second, I found that cells lacking both Rab5 GEFs and cells lacking all three Rab5 paralogs were much more sensitive than otherwise isogenic WT control cells to the growth-inhibitory action of aureobasidin A (AbA), an antibiotic that blocks sphingolipid biosynthesis (Locke and Thorner 2019a). It was demonstrated previously that upregulation of Ypk1 activity via its TORC2-dependent phosphorylation is required for cells to survive inhibition of sphingolipid biosynthesis by another antibiotic (myriocin) (Roelants et al. 2011; Leskoske et al. 2017). Collectively, these results demonstrated that cells that cannot generate active Rab5 exhibit both biochemical and physiological behavior consistent with inefficient TORC2-mediated phosphorylation of Ypk1.

In further support of the conclusion that active Rab5 stimulates TORC2 phosphorylation of Ypk1, I found that expression of Vps21(Q66L), a GTP hydrolysis-defective ( ―GTP-locked‖) variant, restored readily detectable TORC2-mediated phosphorylation of Ypk1 in vps21∆ ypt52∆ ypt53∆ triple mutant cells, whereas expression of Vps21(S21L), a variant ( ―GDP-locked‖) that preferentially binds GDP, did not (Locke and Thorner 2019a). I focused on Vps21 because it is constitutively expressed, is the most abundant yeast Rab5 isoform, has been shown to play the major role in endocytic vesicle trafficking, and, when absent, causes more pronounced phenotypes as compared to loss of either of the other two Rab5 GTPases. Once loaded with GTP, Vps21(Q66L) persists in its active state and has been shown to block endocytosis in yeast cells (Plemel et al. 2011; Lo et al. 2012), suggesting that the ability of Vps21(Q66L) to promote TORC2-dependent phosphorylation of Ypk1 is distinct from its role in promoting early endosome formation and recycling.

To begin to explore the mechanism by which GTP-bound Rab5 supports robust TORC2- mediated phosphorylation of Ypk1, I first examined whether there might be some more indirect reason for the observed impairment in TORC2-Ypk1 signaling in cells that cannot generate active Rab5.

47 Given that Rab5 GTPases are important in endocytic vesicle internalization, and have also been implicated in recycling of endosomes back to the PM, it seemed possible that cells deficient in GTP-bound Rab5 might have decreased TORC2 activity simply due to altered TORC2 localization. However, when I analyzed the subcellular distribution of TORC2 by fluorescence microscopy, I saw no marked change in the amount of TORC2 located at the cell cortex between wild-type cells and muk1Δ vps9Δ double mutant cells. Thus, TORC2 did not appear to be grossly mislocalized in the absence of active Rab5. It also seemed possible that, in cells deficient in GTP-bound Rab5, Ypk1 might not be accessible to the cell cortex at all. However, in contrast to its poor phosphorylation by TORC2, there was no difference in activation loop phosphorylation of Ypk1 by eisosome-associated Pkh1 in WT and muk1Δ vps9Δ double mutant cells. Also, using pull-down assays, there was no difference in the amount of Ypk1 associated with Avo1 between wild-type and muk1Δ vps9Δ double mutant cells, indicating that the lack of efficient phosphorylation by TORC2 in cells deficient in GTP-Rab5 could not be attributed to lack of encounter between TORC2 and Ypk1. Another possibility, given that phosphorylation and activation of certain other AGC family protein kinases requires association of their regulatory domains with other classes of small GTPases (Pearce et al. 2010; Leroux et al. 2018), it seemed possible that a conformation change induced in Ypk1 by its binding of GTP-loaded Rab5 might be necessary for it to be competent for phosphorylation by TORC2. However, as judged by coimmunoprecipitation, I could detect no interaction between Ypk1 and GTP-bound Vps21. Taken together, these findings led us to test whether there is a direct physical interaction between Vps21 and TORC2.

Indeed, in marked contrast to the lack of interaction between Vps21 and Ypk1, association of Vps21 with Tor2 was readily detectable by co-immunoprecipitation under a variety of different expression conditions and strain backgrounds. Moreover, consistent with our in vivo results, in in vitro kinase assays using a purified recombinant substrate [E. coli MalE (maltose-binding protein)-Ypk1(603-680)] that contains all of the TORC2 sites in Ypk1, I found that the specific activity of TORC2 isolated from vsp21∆ ypt52∆ double mutant cells was reproducibly lower than that of an equivalent amount of TORC2 purified from wild-type cells. Most strikingly, the ability of TORC2 isolated from vsp21∆ ypt52∆ cells to phosphorylate MalE-Ypk1(603-680) could be markedly stimulated by the addition of purified recombinant Vps21 loaded with GTPγS, but not by yeast Rab7 (Ypt7) loaded with GTPγS. Thus, a GTP-bound Rab5 appears to be a direct activator of TORC2 kinase function.

Given the evidence that GTP-bound Rab5 stimulates TORC2 function, and given that I showed that the Rab5-specific GEF Muk1 is activated in a TORC2-dependent and Ypk1-mediated manner, one expected outcome of these interrelationships is that, once TORC2-Ypk1 signaling has been initiated, more GTP-bound Rab5 should be generated. In essence, this control circuit imposes self-reinforcing positive feedback to sustain TORC2-Ypk1 signaling. Our unexpected discovery that Rab5 function is necessary to support maximal TORC2 function reveals a previously unappreciated new connection between the vesicle trafficking machinery and the control of PM homeostasis by TORC2-Ypk1 signaling.

Implications of Rab5-dependent regulation of TORC2 The demonstration of the link between Rab5 GTPases and TORC2 raises numerous important mechanistic questions about how and where Rab5 GTPases influence TORC2 function.

How? One intriguing possibility for how Vps21 GTPase could serve as a direct positive activator of the kinase activity of Tor2 in yeast TORC2 is that it could act in a manner analogous to how the RHEB GTPase activates mTOR in mTORC1 (Avruch et al. 2009). Structural analysis (Yang et al. 2017) has revealed that, in its GTP-bound state, RHEB occupies a pocket constituted by the M- and N-HEAT repeats and FAT domain of mTOR, thereby inducing rearrangements of key elements important for stabilizing the catalytically-competent state of mTOR. As one means to assess whether yeast Tor2 in TORC2 might contain a similar binding pocket, we used SWISS- MODEL (Waterhouse et al. 2018) to build a homology map (Fig. 5.1) in which we threaded the sequence of yeast Tor2 onto that of mTOR in the structure of RHEB-activated mTORC1 (PDB: 6BCU). Superimposition of the homology model of Tor2 onto mTOR in RHEB-bound mTORC1 structure reveals a very similar cavity that could readily accommodate a small GTPase. Additionally, many of the residues in mTOR that make contact with RHEB are conserved in Tor2 and adopt a similar orientation in the binding pocket in the model (Fig. 5.1).

Figure 5.1. Saccharomyces cerevisiae Tor2 modeled onto human mTOR in the RHEB-binding pocket in mTORC1. A homology model of ScTor2 (tan) built with SWISS-MODEL (Waterhouse et al., 2018) based on the conformation of mTOR (blue) in the cryo-EM-derived structure of

49 mTORC1 occupied by GTPγS-bound RHEB (faint fuschia) (PDB: 6BCU) (Yang et al., 2017). Side chains in human mTOR that make intermolecular contacts with RHEB (blue) and the corresponding side chains in yeast Tor2 (tan) are shown in stick representation.

To further explore this possibility, we compared the crystal structure of GTP-bound Vps21 to the structure of GTP-bound RHEB. Despite the divergence in the amino acid sequences between these two Ras super-family members, I found that their GTPase folds adopt a nearly identical shape (Fig. 5.2). Hence, GTP-bound Vps21 is clearly sufficiently compact that it could occupy the analogous pocket in Tor2 in TORC2 as GTP-bound RHEB occupies in mTOR in mTORC1. However, how could Tor2 in TORC2 distinguish GTP-bound Rab5 from any other small GTPase? In this regard, and as with other effectors modulated by small GTPases, it is primarily residues in the Switch I and Switch II loops of RHEB that make contact with the RHEB-binding pocket in mTOR in mTORC1. Notably, the superimposition of Vps21 and RHEB reveals that the most major structural differences between these two GTPases lie in their respective Switch II regions (Fig. 5.2). It is not hard to imagine, therefore, that the distinct subunits that define TORC2 force Tor2 into a conformation that is optimal for making specific contacts with the unique geometry of the Switch II residues in Vps21, thereby stabilizing Tor2 in its most catalytically-competent state.

Figure 5.2. Comparison of Saccharomyces cerevisiae Vps21 to Homo sapiens RHEB. Crystal structure (1.48 Å resolution) of yeast Vps21 (green) bound to the non-hydrolyzable GTP analog GppNHp (PDB: 1EK0) (Esters et al., 2000) was aligned with the crystal structure (2.65 Å resolution) of GppNHp-bound RHEB (fuschia) (PDB: 1XTR) (Yu et al., 2005) by superimposing the GTP- binding fold in each of these two GTPase, revealing the distinct conformation of the Switch II regions (top) in these proteins. Side chains in the Switch I and Switch II loops in both GTPases are shown in stick representation.

In this same regard, in a screen of Drosophila small GTPases (Li et al. 2010), it was found that overexpression of activated Rab5 strongly inhibited amino acid- or Rag GTPase-stimulated mTORC1 activity, a process that requires RHEB (Bar-Peled and Sabatini 2014), but did not inhibit mTORC1 stimulation by activated RHEB itself (Li et al. 2010). At the time, it was not known how to interpret these observations. In light of our findings, however, a likely explanation for these prior results is that Rab5, if overexpressed, can associate with the RHEB-binding

50 pocket in mTOR in mTORC1, but does so non-productively (i.e. it cannot stimulate mTOR in this context), and thus under these conditions Rab5 serves, in essence, as a simple competitive inhibitor of RHEB-mediated mTORC1 activation. By contrast, when activated RHEB is overexpressed, it is able to out-compete Rab5.

Where? Localization of TORC2 in S. cerevisiae to discrete puncta residing at the cell cortex has been well documented, based on analysis of fixed cells by indirect immunofluorescence (Kunz et al. 2000) or immuno-gold EM (Wedaman et al. 2003), or examination of live cells expressing fluorescently-tagged derivatives of Tor2 (Sturgill et al. 2008; Leskoske et al. 2018) or TORC2- specific subunits that associate tightly with Tor2 (Berchtold and Walther 2009; Leskoske et al. 2018). In addition, the effector protein kinases (Ypk1 and Ypk2) directly phosphorylated by TORC2 are, as mentioned earlier, also obligatorily phosphorylated by other protein kinases (Pkh1 and Pkh2) that are tightly associated with the PM-anchored eisosomes. Moreover, the PM in yeast is highly enriched in PtdIns4,5P2 compared to any other cellular membrane (Garrenton et al. 2010; Riggi et al. 2018) and both the core TORC2 subunit Avo1 (Berchtold and Walther 2009; Gallego et al. 2010; Vonkova et al. 2015) and the more dynamically associated TORC2 subunits Slm1 and Slm2 (Fadri et al. 2005; Tabuchi et al. 2006; Niles et al. 2012) contain PtdIns4,5P2-binding PH domains. Furthermore, maintenance of an adequate PtdIns4,5P2 level is necessary for TORC2 activity (Morales-Johansson et al. 2004). These observations are consistent with the conclusion that, to a large extent, the TORC2 cortical puncta are docked on the PM.

The three yeast Rab5 GTPases (Vps21/Ypt51, Ypt52 and Ypt53) (Schmidt et al. 2017) are involved at an early stage of the endocytic pathway and, akin to their mammalian Rab5 counterparts (Barr 2013; Wandinger-Ness and Zerial 2014), are therefore located primarily on intracellular endosomal vesicles prior to their fusion and delivery to the multivesicular body (Horazdovsky et al. 1994; Gerrard et al. 2000; Toshima et al. 2014). Correspondingly, fluorescently-tagged derivatives of the three yeast Rab5 members exhibit punctate staining on vesicular structures within the cell that co-localize with well-documented endocytic cargo (Peplowska et al. 2007; Cabrera and Ungermann 2013). Interestingly, however, for GFP-Vps21, a significant fraction of the fluorescent structures are found at the cell periphery on, or in very close apposition to, the PM, whereas GFP-Ypt52 and GFP-Ypt53 generally occupy more interior locations (Buvelot Frei et al. 2006).

There is additional evidence that Rab5 GTPases can be associated with the PM. For example, experiments to localize Rab5 in mammalian cells also exhibited some staining on the PM (Chavrier et al. 1990). When heterologously expressed in animal cells, Vps21 colocalizes with the endogenous Rab5 and can stimulate Rab5-specific endocytic events (Singer-Kruger et al. 1995), indicating that, in addition to their C-terminal prenylation (Pfeffer and Aivazian 2004), the yeast and vertebrate proteins must share other conserved localization determinants. In addition, using a bimolecular fluorescence complementation assay, interaction between Vps21 and its specific GAP Msb3 was detected only at patches on the PM (Lachmann et al. 2012), indicating again that at least a fraction of the GTP-bound pool of Vps21 resides at the PM. In this context, it is important to note that the phenotypes of cells devoid of each of the three yeast Rab5 isoforms differ significantly. In addition to severe defects in endocytic vesicle trafficking, cells lacking Vps21 also exhibit an obvious defect in growth rate, whereas cells lacking either Ypt52

51 or Ypt53 (or both) have much milder phenotypes. Although the relative levels of each of these proteins may contribute to the differential effects observed in their absence (Schmidt et al. 2017), these results are also consistent with Vps21 having a role in promoting the function of TORC2, a central growth regulator, distinct from its function in vesicle trafficking. In agreement with that proposal, in my study (Locke and Thorner 2019a), I observed that in cells devoid of all three Rab5 paralogs TORC2-mediated phosphorylation of Ypk1 could be stimulated by expression of a GTP-locked Vps21 mutant, which (as already mentioned above) is unable to support vesicle trafficking. It will be important to test GTP-locked variants of Ypt52 and Ypt53 in the same way to determine whether it is exclusively Vps21 that associates with and acts to activate TORC2.

It is also possible that TORC2 association with and activation by Vps21 might occur on endosomes. In subcellular fractionation studies showing that TORC2 is a protein complex peripherally associated with membranes, Tor2 co-fractionated with PM markers, but also with a second fraction that appeared to represent vesicular structures (Kunz et al. 2000). Similarly, localization of endogenous Tor2 at the ultrastructural level using immuno-gold EM labeling revealed gold particle clusters at the PM, but also gold particles associated with internal membranous structures that resembled vesicles of the endocytic pathway and were distinct from the vacuole (Wedaman et al. 2003). In this regard, I found that in cells lacking the two Rab5 GEFs (Vps9 and Muk1) that the amount of TORC2 localized at the PM, as judged by monitoring a core TORC2 subunit (Avo3-GFP), is not reduced compared to otherwise isogenic wild-type cells. A similar conclusion was reached using Tor2-mNeonGreen in cells lacking all three Rab5 GTPases (F.M. Roelants, this laboratory, unpubished data). These observations indicate that PM association of TORC2 does not require Rab5 function. However, I have not carefully analyzed whether the absence of the Rab5 GEFs or the Rab5 GTPases themselves markedly diminishes the minor pool of TORC2 that appears to be associated with internal vesicles. In this regard, there is other evidence that TORC2 may be distributed among spatially distinct populations.

It has been inferred from recent studies in mammalian cells of mTORC2-mediated phosphorylation of one of its effector protein kinases, AKT/PKB, in mammalian cells that mTORC2 can act at discrete subcellular locations. AKT has a PH domain specific for binding PtdIns(3,4,5)P3 generated by growth factor-activated PtdIns 3-kinase (PI3K); however, there is some controversey about where and whether lipid binding is necessary to induce a conformational change in AKT that makes it competent to be phosphorylated on its activation loop by the upstream activating kinase PDK1 before it is competent to be subject to further stimulation via its mTORC2-mediated phosphorylation (Vanhaesebroeck and Alessi 2000; Agarwala 2018; Lučić et al. 2018). In any event, using a chemically-inducible dimerization system to recruit AKT to endosomes, it was observed that AKT becomes phosphorylated at its mTORC2 sites when tethered in this fashion, arguing that there must be mTORC present on the same endosomes (Ebner et al. 2017). Further, it was purported that mTORC2-dependent activation of AKT at the PM can be uncoupled from the need for PI3K-generated PtdIns(3,4,5)P3, whereas the endosomally-associated mTORC2, it is claimed, requires PI3K generation of PtdIns(3,4,5)P3 (Jethwa et al. 2015), suggesting, if true, that these spatially distinct pools may be differentially regulated. The role of Rab5 in this putative endosomal pool of mTORC2 was not examined in these studies. However, in a recent study of mouse hippocampal pyramidal neurons, AKT was found in association with Rab5-positive endosomes, but whether TORC2 was also present was not addressed (Goto-Silva et al. 2019). To further complicate

52 matters as to which GTPase(s) may regulate TORC2, and where, it was recently reported that in mammalian cells, oncogenic variants of RAS, a PM-anchored GTPase (Simanshu et al. 2017), associates with mTORC2, as assessed by proximity-dependent biotin labeling (Kovalski et al. 2019).

Hypothetically, if there are distinct pools of TORC2 with differing requirements for activation, such a scenario could provide a means to adjust the level of TORC2 activity to meet the physiological needs of the cell in the most effective manner. In yeast, for Ypk1 that has been phosphorylated by Pkh1 on its activation loop, all that is required for cell survival under steady- state growth conditions in rich medium with glucose as the carbon source is TORC2-dependent phosphorylation of Ypk1 on a single site (S644 in the turn motif). However, if the cells are subjected to any significant stress, especially limitation for sphingolipids, TORC2-mediated phosphorylation of T662 and additional residues in the C-terminal tail of Ypk1 occurs and modification of these residues is essential for yeast cell survival under these stressful conditions (Roelants et al. 2004; Roelants et al. 2011; Leskoske et al. 2017). That phosphorylation at these additional resides is not required under basal conditions, suggests that TORC2 activity provides a graded mechanism to fine-tune the level of Ypk1 activation. Furthermore, C-terminal phosphorylation of Ypk1 by TORC2 not only enhances its specific activity (Roelants et al. 2011), but also stabilizes Ypk1 against degradation (Leskoske et al., 2017), thereby maintaining both its active conformation and prolonging the duration of its activated state. Thus, the demonstration (Locke and Thorner 2019a) that TORC2-Ypk1 signaling also stimulates the generation of GTP-bound Vps21 and that this Rab5 is, in turn, a positive effector of TORC2, provides a feed forward mechanism to further sustain TORC2-Ypk1 signaling, once initiated. This control circuit also provides a means for TORC2 to serve as both a sensor and a regulator of the rate of endocytic vesicle trafficking. Moreover, this Rab5-mediated stimulation could exist to up-regulate TORC2 activity and hence Ypk1 signaling at the PM or, potentially, to provide the means for localized TORC2-dependent Ypk1-mediated phosphorylation of specific substrates located on or nearby early endosomes.

53 Literature Cited Agarwala AK. 2018. How to explain the AKT phosphorylation of downstream targets in the wake of recent findings. Proc Natl Acad Sci USA 115: E6099–E6100. Albertyn J, Hohmann S, Thevelein JM, Prior BA. 1994. GPD1, which encodes glycerol-3- phosphate dehydrogenase, is essential for growth under osmotic stress in Saccharomyces cerevisiae, and its expression is regulated by the high-osmolarity glycerol response pathway. Mol Cell Biol 14: 4135-4144. Albuquerque CP, Smolka MB, Payne SH, Bafna V, Eng J, Zhou H. 2008. A multidimensional chromatography technology for in-depth phosphoproteome analysis. Mol Cell Proteomics 7: 1389-1396. Alvaro CG, Aindow A, Thorner J. 2016. Differential phosphorylation provides a switch to control how -arrestin Rod1 down-regulates mating pheromone response in Saccharomyces cerevisiae. Genetics 203: 299-317. Alvaro CG, O'Donnell AF, Prosser DC, Augustine AA, Goldman A, Brodsky JL, Cyert MS, Wendland B, Thorner J. 2014. Specific alpha-arrestins negatively regulate Saccharomyces cerevisiae pheromone response by down-modulating the G-protein- coupled receptor Ste2. Mol Cell Biol 34: 2660-2681. Amberg DC, Burke DJ, Strathern JN. 2005. Methods in Yeast Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Aronova S, Wedaman K, Aronov PA, Fontes K, Ramos K, Hammock BD, Powers T. 2008. Regulation of ceramide biosynthesis by TOR complex 2. Cell Metab 7: 148-158. Avruch J, Long X, Lin Y, Ortiz-Vega S, Rapley J, Papageorgiou A, Oshiro N, Kikkawa U. 2009. Activation of mTORC1 in two steps: Rheb-GTP activation of catalytic function and increased binding of substrates to raptor. Biochem Soc Trans 37: 223-226. Aylett CH, Sauer E, Imseng S, Boehringer D, Hall MN, Ban N, Maier T. 2016. Architecture of human mTOR complex 1. Science 351: 48-52. Ballabeni A, Zamponi R, Moore JK, Helin K, Kirschner MW. 2013. Geminin deploys multiple mechanisms to regulate Cdt1 before cell division thus ensuring the proper execution of DNA replication. Proc Natl Acad Sci USA 110: E2848-2853. Bar-Peled L, Sabatini DM. 2014. Regulation of mTORC1 by amino acids. Trends Cell Biol 24: 400-406. Bardwell L. 2006. Mechanisms of MAPK signalling specificity. Biochem Soc Trans 34: 837- 841. Bardwell AJ, Bardwell L. 2015. Two hydrophobic residues can determine the specificity of mitogen-activated protein kinase docking interactions. J Biol Chem 290: 26661-26674. Bardwell L, Cook JG, Zhu-Shimoni JX, Voora D, Thorner J. 1998. Differential regulation of transcription: repression by unactivated mitogen-activated protein kinase Kss1 requires the Dig1 and Dig2 proteins. Proc Natl Acad Sci USA 95: 15400-15405. Bardwell L, Thorner J. 1996. A conserved motif at the amino termini of MEKs might mediate high-affinity interaction with the cognate MAPKs. Trends Biochemical Sci 21: 373-374. Baretic D, Berndt A, Ohashi Y, Johnson CM, Williams RL. 2016. Tor forms a dimer through an N-terminal helical solenoid with a complex topology. Nat Commun 7: 11016. Barr F, Lambright DG. 2010. Rab GEFs and GAPs. Curr Opin Cell Biol 22: 461-470. Barr FA. 2013. Rab GTPases and membrane identity: causal or inconsequential? J Cell Biol 202: 191-199.

54 Bartlett K, Kim K. 2014. Insight into Tor2, a budding yeast microdomain protein. Eur J Cell Biol 93: 87-97. Bean BD, Davey M, Snider J, Jessulat M, Deineko V, Tinney M, Stagljar I, Babu M, Conibear E. 2015. Rab5-family guanine nucleotide exchange factors bind retromer and promote its recruitment to endosomes. Mol Biol Cell 26: 1119-1128. Becuwe M, Herrador A, Haguenauer-Tsapis R, Vincent O, Leon S. 2012. Ubiquitin-mediated regulation of endocytosis by proteins of the arrestin family. Biochem Res Int 2012: 242764. Beese-Sims SE, Lee J, Levin DE. 2011. Yeast Fps1 glycerol facilitator functions as a homotetramer. Yeast 28: 815-819. Berchtold D, Piccolis M, Chiaruttini N, Riezman I, Riezman H, Roux A, Walther TC, Loewith R. 2012. Plasma membrane stress induces relocalization of Slm proteins and activation of TORC2 to promote sphingolipid synthesis. Nat Cell Biol 14: 542-547. Berchtold D, Walther TC. 2009. TORC2 plasma membrane localization is essential for cell viability and restricted to a distinct domain. Mol Biol Cell 20: 1565-1575. Betz C, Hall MN. 2013. Where is mTOR and what is it doing there? J Cell Biol 203: 563-574. Biondi RM, Nebreda AR. 2003. Signalling specificity of Ser/Thr protein kinases through docking-site-mediated interactions. Biochem J 372: 1-13. Bourgoint C, Rispal D, Berti M, Filipuzzi I, Helliwell SB, Prouteau M, Loewith R. 2018. Target of rapamycin complex 2-dependent phosphorylation of the coat protein Pan1 by Akl1 controls endocytosis dynamics in Saccharomyces cerevisiae. J Biol Chem 293: 12043- 12053. Breslow DK, Collins SR, Bodenmiller B, Aebersold R, Simons K, Shevchenko A, Ejsing CS, Weissman JS. 2010. Orm family proteins mediate sphingolipid homeostasis. Nature 463: 1048-1053. Burd CG, Mustol PA, Schu PV, Emr SD. 1996. A yeast protein related to a mammalian Ras- binding protein, Vps9p, is required for localization of vacuolar proteins. Mol Cell Biol 16: 2369-2377. Busino L, Donzelli M, Chiesa M, Guardavaccaro D, Ganoth D, Dorrello NV, Hershko A, Pagano M, Draetta GF. 2003. Degradation of Cdc25A by beta-TrCP during S phase and in response to DNA damage. Nature 426: 87-91. Buvelot Frei S, Rahl PB, Nussbaum M, Briggs BJ, Calero M, Janeczko S, Regan AD, Chen CZ, Barral Y, Whittaker GR et al. 2006. Bioinformatic and comparative localization of Rab proteins reveals functional insights into the uncharacterized GTPases Ypt10 and Ypt11. . Mol Cell Biol 26: 7299-7317. Cabrera M, Ungermann C. 2013. Guanine nucleotide exchange factors (GEFs) have a critical but not exclusive role in organelle localization of Rab GTPases. J Biol Chem 288: 28704- 28712. Carney DS, Davies BA, Horazdovsky BF. 2006. Vps9 domain-containing proteins: activators of Rab5 GTPases from yeast to neurons. Trends Cell Biol 16: 27-35. Casamayor A, Torrance PD, Kobayashi T, Thorner J, Alessi DR. 1999. Functional counterparts of mammalian protein kinases PDK1 and SGK in budding yeast. Curr Biol 9: 186-197. Chavrier P, Parton RG, Hauri HP, Simons K, Zerial M. 1990. Localization of low molecular weight GTP binding proteins to exocytic and endocytic compartments. Cell 62: 317-329. Chen P, Lee KS, Levin DE. 1993. A pair of putative protein kinase genes (Ypk1 and Ypk2) is required for cell growth in Saccharomyces cerevisiae. Mol Gen Genet 236: 443-447.

55 Chen RE, Thorner J. 2007. Function and regulation in MAPK signaling pathways: lessons learned from the yeast Saccharomyces cerevisiae. Biochim Biophys Acta 1773: 1311- 1340. Collins TJ. 2007. ImageJ for microscopy. Biotechniques 43: 25-30. Consortium TU. 2017. UniProt: the universal protein knowledgebase. Nucleic Acids Res 45: D158-D169. Daquinag A, Fadri M, Jung SY, Qin J, Kunz J. 2007. The yeast PH domain proteins Slm1 and Slm2 are targets of sphingolipid signaling during the response to heat stress. Mol Cell Biol 27: 633-650. Das A, Slaughter BD, Unruh JR, Bradford WD, Alexander R, Rubinstein B, Li R. 2012. Flippase-mediated phospholipid asymmetry promotes fast Cdc42 recycling in dynamic maintenance of cell polarity. Nat Cell Biol 14: 304-310. Date DA, Burrows AC, Venere M, Jackson MW, Summers MK. 2013. Coordinated regulation of p31(Comet) and Mad2 expression is required for cellular proliferation. Cell Cycle 12: 3824-3832. deHart AKA, Schnell JD, Allen DA, Hicke L. 2002. The conserved Pkh-Ypk kinase cascade is required for endocytosis in yeast. J Cell Biol 156: 241-248. de Felipe P, Luke GA, Hughes LE, Gani D, Halpin C, Ryan MD. 2006. E unum pluribus: multiple proteins from a self-processing polyprotein. Trends Biotechnol 24: 68-75. Demaison C, Parsley K, Brouns G, Scherr M, Battmer K, Kinnon C, Grez M, Thrasher AJ. 2002. High-level transduction and gene expression in hematopoietic repopulating cells using a human immunodeficiency [correction of imunodeficiency] virus type 1-based lentiviral vector containing an internal spleen focus forming virus promoter. Hum Gene Ther 13: 803-813. Dickson RC, Sumanasekera C, Lester RL. 2006. Functions and metabolism of sphingolipids in Saccharomyces cerevisiae. Prog Lipid Res 45: 447-465. Douglas LM, Konopka JB. 2014. Fungal membrane organization: the eisosome concept. Annu Rev Microbiol 68: 377-393. Dowling RJ, Topisirovic I, Fonseca BD, Sonenberg N. 2010. Dissecting the role of mTOR: lessons from mTOR inhibitors. Biochim Biophys Acta 1804: 433-439 Dunn TM, Lynch DV, Michaelson LV, Napier JA. 2004. A post-genomic approach to understanding sphingolipid metabolism in Arabidopsis thaliana. Ann Bot 93: 483-497. Duplay P, Bedouelle H, Fowler A, Zabin I, Saurin W, Hofnung M. 1984. Sequences of the malE gene and of its product, the maltose-binding protein of Escherichia coli K12. J Biol Chem 259: 10606-10613. Ebner M, Sinkovics B, Szczygiel M, Ribeiro DW, Yudushkin I. 2017. Localization of mTORC2 activity inside cells. J Cell Biol 216: 343-353. Eltschinger S, Loewith R. 2016. TOR Complexes and the Maintenance of Cellular Homeostasis. Trends Cell Biol 26: 148-159. Eng CP, Sehgal SN, Vezina C. 1984. Activity of rapamycin (AY-22,989) against transplanted tumors. J Antibiot (Tokyo) 37: 1231-1237. Facchinetti V, Ouyang W, Wei H, Soto N, Lazorchak A, Gould C, Lowry C, Newton AC, Mao Y, Miao RQ et al. 2008. The mammalian target of rapamycin complex 2 controls folding and stability of Akt and protein kinase C. EMBO J 27: 1932-1943.

56 Fadri M, Daquinag A, Wang S, Xue T, Kunz J. 2005. The pleckstrin homology domain proteins Slm1 and Slm2 are required for actin cytoskeleton organization in yeast and bind phosphatidylinositol-4,5-bisphosphate and TORC2. Mol Biol Cell 16: 1883-1900. Fay MP, Proschan MA. 2010. Wilcoxon-Mann-Whitney or t-test? On assumptions for hypothesis tests and multiple interpretations of decision rules. Stat Surv 4: 1-39. Feyder S, De Craene JO, Bar S, Bertazzi DL, Friant S. 2015. Membrane trafficking in the yeast Saccharomyces cerevisiae model. Int J Mol Sci 16: 1509-1525. Finnigan GC, Thorner J. 2016. mCAL: A new approach for versatile multiplex action of Cas9 using one sgRNA and loci flanked by a programmed target sequence. G3 (Bethesda) 6: 2147-2156. Foderaro JE, Douglas LM, Konopka JB. 2017. MCC/eisosomes regulate cell wall synthesis and stress responses in fungi. J Fungi (Basel) 3: E61. Frescas D, Pagano M. 2008. Deregulated proteolysis by the F-box proteins SKP2 and β-TrCP: tipping the scales of cancer. Nat Rev Cancer 8: 438-449. Frohlich F, Olson DK, Christiano R, Farese RV, Jr., Walther TC. 2016. Proteomic and phosphoproteomic analyses of yeast reveal the global cellular response to sphingolipid depletion. Proteomics 16: 2759-2763. Gallego O, Betts MJ, Gvozdenovic-Jeremic J, Maeda K, Matetzki C, Aguilar-Gurrieri C, Beltran-Alvarez P, Bonn S, Fernandez-Tornero C, Jensen LJ et al. 2010. A systematic screen for protein-lipid interactions in Saccharomyces cerevisiae. Mol Syst Biol 6: 430. Garcia-Martinez JM, Alessi DR. 2008. mTOR complex 2 (mTORC2) controls hydrophobic motif phosphorylation and activation of serum- and glucocorticoid-induced protein kinase 1 (SGK1). Biochem J 416: 375-385. Garrenton LS, Stefan CJ, McMurray MA, Emr SD, Thorner J. 2010. Pheromone-induced anisotropy in yeast plasma membrane phosphatidylinositol-4,5-bisphosphate distribution is required for MAPK signaling. Proc Natl Acad Sci USA 107: 11805-111810. Gasch AP, Spellman PT, Kao CM, Carmel-Harel O, Eisen MB, Storz G, Botstein D, Brown PO. 2000. Genomic expression programs in the response of yeast cells to environmental changes. Mol Biol Cell 11: 4241-4257. - (2015). A new family of StART domain proteins at membrane contact sites has a role in ER-PM sterol transport. Elife 4: e07235. Gaubitz C, Oliveira TM, Prouteau M, Leitner A, Karuppasamy M, Konstantinidou G, Rispal D, Eltschinger S, Robinson GC, Thore S et al. 2015. Molecular basis of the rapamycin insensitivity of Target of Rapamycin Complex 2. Mol Cell 58: 977-988. Gaubitz C, Prouteau M, Kusmider B, Loewith R. 2016. TORC2 Structure and Function. Trends Biochem Sci 41: 532-545. Gerrard SR, Bryant NJ, Stevens TH. 2000. VPS21 controls entry of endocytosed and biosynthetic proteins into the yeast prevacuolar compartment. Mol Biol Cell 11: 613-626. Goldsmith EJ, Akella R, Min X, Zhou T, Humphreys JM. 2007. Substrate and docking interactions in serine/threonine protein kinases. Chem Rev 107: 5065-5081. Gonzalez A, Hall MN. 2017. Nutrient sensing and TOR signaling in yeast and mammals. EMBO J 36: 397-408. Goto-Silva L, McShane MP, Salinas S, Kalaidzidis Y, Schiavo G, Zerial M. 2019. Retrograde transport of Akt by a neuronal Rab5-APPL1 endosome. Sci Rep 9: 2433.

57 Gough J, Karplus K, Hughey R, Chothia C. 2001. Assignment of homology to genome sequences using a library of hidden Markov models that represent all proteins of known structure. J Mol Biol 313: 903-919. Green MR, Sambrook J. 2012. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Grenson M, Mousset M, Wiame JM, Bechet J. 1966. Multiplicity of the amino acid permeases in Saccharomyces cerevisiae. I. Evidence for a specific arginine-transporting system. Biochim Biophys Acta 127: 325-338. Guerreiro JF, Muir A, Ramachandran S, Thorner J, Sa-Correia I. 2016. Sphingolipid biosynthesis upregulation by TOR complex 2-Ypk1 signaling during yeast adaptive response to acetic acid stress. Biochem J 473: 4311-4325. Guiney EL, Klecker T, Emr SD. 2016. Identification of the endocytic sorting signal recognized by the Art1-Rsp5 ubiquitin ligase complex. Mol Biol Cell 27: 4043-4054. Habu T, Kim SH, Weinstein J, Matsumoto T. 2002. Identification of a MAD2-binding protein, CMT2, and its role in mitosis. EMBO J 21: 6419-6428. Hagan RS, Manak MS, Buch HK, Meier MG, Meraldi P, Shah JV, Sorger PK. 2011. p31(comet) acts to ensure timely spindle checkpoint silencing subsequent to kinetochore attachment. Mol Biol Cell 22: 4236-4246. Hama H, Tall GG, Horazdovsky BF. 1999. Vps9p is a guanine nucleotide exchange factor involved in vesicle-mediated vacuolar protein transport. J Biol Chem 274: 15284-15291. Harper JW, Tan MK. 2012. Understanding cullin-RING E3 biology through proteomics-based substrate identification. Mol Cell Proteomics 11: 1541-1550. Heitman J, Movva NR, Hall MN. 1991. Targets for cell cycle arrest by the immunosuppressant rapamycin in yeast. Science 253: 905-909. Helliwell SB, Wagner P, Kunz J, M. D-R, Henriquez R, Hall MN. 1994. TOR1 and TOR2 are structurally and functionally similar but not identical phosphatidylinositol kinase homologues in yeast. Mol Cell Biol 5 105-118. Hill A, Niles B, Cuyegkeng A, Powers T. 2018. Redesigning TOR Kinase to Explore the Structural Basis for TORC1 and TORC2 Assembly. Biomolecules 8: E36. Hockemeyer D, Wang H, Kiani S, Lai CS, Gao Q, Cassady JP, Cost GJ, Zhang L, Santiago Y, Miller JC et al. 2011. Genetic engineering of human pluripotent cells using TALE nucleases. Nat Biotechnol 29: 731-734. Hohmann S. 2015. An integrated view on a eukaryotic osmoregulation system. Curr Genet 61: 373-382. Holland PM, Cooper JA. 1999. Protein modification: docking sites for kinases. Curr Biol 9: R329-331. Holt LJ, Tuch BB, Villen J, Johnson AD, Gygi SP, Morgan DO. 2009. Global analysis of Cdk1 substrate phosphorylation sites provides insights into evolution. Science 325: 1682-1686. Horazdovsky BF, Busch GR, Emr SD. 1994. VPS21 encodes a rab5-like GTP binding protein that is required for the sorting of yeast vacuolar proteins. EMBO J 13: 1297-1309. Huang JX, Manning BD. 2009. A complex interplay between Akt, TSC2 and the two mTOR complexes. Biochem Soc Trans 37: 217-222. Hunter T, Plowman GD. 1997. The protein kinases of budding yeast: six score and more. Trends Biochem Sci 22: 18-22. Husnjak K, Dikic I. 2012. Ubiquitin-binding proteins: decoders of ubiquitin-mediated cellular functions. Annu Rev Biochem 81: 291-322.

58 Ikai N, Nakazawa N, Hayashi T, Yanagida M. 2011. The reverse, but coordinated, roles of Tor2 (TORC1) and Tor1 (TORC2) kinases for growth, cell cycle and separase-mediated mitosis in Schizosaccharomyces pombe. Open Biol 1: 110007. Ikeda K, Morigasaki S, Tatebe H, Tamanoi F, Shiozaki K. 2008. Fission yeast TOR complex 2 activates the AGC-family Gad8 kinase essential for stress resistance and cell cycle control. Cell Cycle 7: 358-364. Ikenoue T, Inoki K, Yang Q, Zhou X, Guan KL. 2008. Essential function of TORC2 in PKC and Akt turn motif phosphorylation, maturation and signalling. EMBO J 27: 1919-1931. Ikushiro H, Hayashi H, Kagamiyama H. 2004. Reactions of serine palmitoyltransferase with serine and molecular mechanisms of the actions of serine derivatives as inhibitors. Biochemistry 43: 1082-1092. Isnard AD, Thomas D, Surdin-Kerjan Y. 1996. The study of methionine uptake in Saccharomyces cerevisiae reveals a new family of amino acid permeases. J Mol Biol 262: 473-484. Jacinto E, Loewith R, Schmidt A, Lin S, Ruegg MA, Hall A, Hall MN. 2004. Mammalian TOR complex 2 controls the actin cytoskeleton and is rapamycin insensitive. Nat Cell Biol 6: 1122-1128. Jethwa N, Chung GH, Lete MG, Alonso A, Byrne RD, Calleja V, Larijani B. 2015. Endomembrane PtdIns(3,4,5)P3 activates the PI3K-Akt pathway. J Cell Sci 128: 3456- 3465. Jia L, Li B, Warrington RT, Hao X, Wang S, Yu H. 2011. Defining pathways of spindle checkpoint silencing: functional redundancy between Cdc20 ubiquitination and p31(comet). Mol Biol Cell 22: 4227-4235. Jiang Y, Broach JR. 1999. Tor proteins and protein phosphatase 2A reciprocally regulate Tap42 in controlling cell growth in yeast. EMBO J 18: 2782-2792. Jovic M, Sharma M, Rahajeng J, Caplan S. 2010. The early endosome: a busy sorting station for proteins at the crossroads. Histol Histopathol 25: 99-112. Kamada Y, Fujioka Y, Suzuki NN, Inagaki F, Wullschleger S, Loewith R, Hall MN, Ohsumi Y. 2005. Tor2 directly phosphorylates the AGC kinase Ypk2 to regulate actin polarization. Mol Cell Biol 25: 7239-7248. Kampmann M, Bassik MC, Weissman JS. 2013. Integrated platform for genome-wide screening and construction of high-density genetic interaction maps in mammalian cells. Proc Natl Acad Sci USA 110: E2317-2326. Kampmann M, Bassik MC, Weissman JS. 2014. Functional genomics platform for pooled screening and generation of mammalian genetic interaction maps. Nat Protoc 9: 1825- 1847. Karotki L, Huiskonen JT, Stefan CJ, Ziolkowska NE, Roth R, Surma MA, Krogan NJ, Emr SD, Heuser J, Grunewald K et al. 2011. Eisosome proteins assemble into a membrane scaffold. J Cell Biol 195: 889-902. Karuppasamy M, Kusmider B, Oliveira TM, Gaubitz C, Prouteau M, Loewith R, Schaffitzel C. 2017. Cryo-EM structure of Saccharomyces cerevisiae target of rapamycin complex 2. Nat Commun 8: 1729. Kerppola TK. 2009. Visualization of molecular interactions using bimolecular fluorescence complementation analysis: characteristics of protein fragment complementation. Chem Soc Rev 38: 2876-2886.

59 Khakhina S, Johnson SS, Manoharlal R, Russo SB, Blugeon C, Lemoine S, Sunshine AB, Dunham MJ, Cowart LA, Devaux F et al. 2015. Control of Plasma Membrane Permeability by ABC Transporters. Eukaryot Cell 14: 442-453. Kim JH, Lee SR, Li LH, Park HJ, Park JH, Lee KY, Kim MK, Shin BA, Choi SY. 2011. High cleavage efficiency of a 2A peptide derived from porcine teschovirus-1 in human cell lines, zebrafish and mice. PLoS One 6: e18556. Kim SS, Shago M, Kaustov L, Boutros PC, Clendening JW, Sheng Y, Trentin GA, Barsyte- Lovejoy D, Mao DY, Kay R et al. 2007. CUL7 is a novel antiapoptotic oncogene. Cancer Res 67: 9616-9622. Kinoshita E, Kinoshita-Kikuta E, Koike T. 2009. Separation and detection of large phosphoproteins using Phos-tag SDS-PAGE. Nat Protoc 4: 1513-1521. Kliegman JI, Fiedler D, Ryan CJ, Xu YF, Su XY, Thomas D, Caccese MC, Cheng A, Shales M, Rabinowitz JD et al. 2013. Chemical genetics of rapamycin-insensitive TORC2 in S. cerevisiae. Cell Rep 5: 1725-1736. Knighton DR, Zheng JH, Ten Eyck LF, Xuong NH, Taylor SS, Sowadski JM (1991) Structure of a peptide inhibitor bound to the catalytic subunit of cyclic adenosine monophosphate- dependent protein kinase. Science. 253: 414-420. Koeffler HP, Golde DW. 1980. Human myeloid leukemia cell lines: a review. Blood 56: 344- 350. Komander D, Rape M. 2012. The ubiquitin code. Annu Rev Biochem 81: 203-229. Kovalski JR, Bhaduri A, Zehnder AM, Neela PH, Che Y, Wozniak GG, Khavari PA. 2019. The functional proximal proteome of oncogenic Ras includes mTORC2. Mol Cell 73: 830- 844 e812. Kunz J, Henriquez R, Schneider U, Deuter-Reinhard M, Movva NR, Hall MN. 1993. Target of rapamycin in yeast, TOR2, is an essential phosphatidylinositol kinase homolog required for G1 progression. Cell 73: 585-596. Kunz J, Schneider U, Howald I, Schmidt A, Hall MN. 2000. HEAT repeats mediate plasma membrane localization of Tor2p in yeast. J Biol Chem 275: 37011-37020. Lachmann J, Barr FA, Ungermann C. 2012. The Msb3/Gyp3 GAP controls the activity of the Rab GTPases Vps21 and Ypt7 at endosomes and vacuoles. Mol Biol Cell 23: 2516-2526. Lee YJ, Jeschke GR, Roelants FM, Thorner J, Turk BE. 2012. Reciprocal phosphorylation of yeast glycerol-3-phosphate dehydrogenases in adaptation to distinct types of stress. Mol Cell Biol 32: 4705-4717. Leroux AE, Schulze JO, Biondi RM. 2018. AGC kinases, mechanisms of regulation and innovative drug development. Semin Cancer Biol 48: 1-17. Leskoske KL, Roelants FM, Emmerstorfer-Augustin A, Augustin CM, Si EP, Hill JM, Thorner J. 2018. Phosphorylation by the stress-activated MAPK Slt2 down-regulates the yeast TOR complex 2. Gene Dev 32: 1576-1590. Leskoske KL, Roelants FM, Marshall MNM, Hill JM, Thorner J. 2017. The stress-sensing TORC2 complex activates yeast AGC-family protein kinase Ypk1 at multiple novel sites. Genetics 207: 179-195. Letunic I, Doerks T, Bork P. 2015. SMART: recent updates, new developments and status in 2015. Nucleic Acids Res 43: D257-260. Levin DE, Fields FO, Kunisawa R, Bishop JM, Thorner J. 1990. A candidate protein kinase C gene, PKC1, is required for the S. cerevisiae cell cycle. Cell 62: 213-224.

60 Li L, Kim E, Yuan H, Inoki K, Goraksha-Hicks P, Schiesher RL, Neufeld TP, Guan KL. 2010. Regulation of mTORC1 by the Rab and Arf GTPases. J Biol Chem 285: 19705-19709. Liao HC, Chen MY. 2012. Target of rapamycin complex 2 signals to downstream effector yeast protein kinase 2 (Ypk2) through adheres-voraciously-to-target-of-rapamycin-2 protein 1 (Avo1) in Saccharomyces cerevisiae. J Biol Chem 287: 6089-6099. Lin CH, MacGurn JA, Chu T, Stefan CJ, Emr SD. 2008. Arrestin-related ubiquitin-ligase adaptors regulate endocytosis and protein turnover at the cell surface. Cell 135: 714-725. Liu P, Begley M, Michowski W, Inuzuka H, Ginzberg M, Gao D, Tsou P, Gan W, Papa A, Kim BM et al. 2014. Cell-cycle-regulated activation of Akt kinase by phosphorylation at its carboxyl terminus. Nature 508: 541-545. Liu Q, Ren T, Fresques T, Oppliger W, Niles BJ, Hur W, Sabatini DM, Hall MN, Powers T, Gray NS. 2012. Selective ATP-competitive inhibitors of TOR suppress rapamycin- insensitive function of TORC2 in Saccharomyces cerevisiae. ACS Chem Biol 7: 982-987. Liu Z, Chen O, Wall JBJ, Zheng M, Zhou Y, Wang L, Ruth Vaseghi H, Qian L, Liu J. 2017. Systematic comparison of 2A peptides for cloning multi-genes in a polycistronic vector. Sci Rep 7: 2193. Lo SY, Brett CL, Plemel RL, Vignali M, Fields S, Gonen T, Merz AJ. 2012. Intrinsic tethering activity of endosomal Rab proteins. Nat Struct Mol Biol 19: 40-47. Locke MN, Muir A, Roelants FM, Thorner J. 2019. In vitro protein kinase assay for Ypk1 from Saccharomyces cerevisiae. Bio-protocol, in press. Locke MN, Thorner J. 2019a. Rab5 GTPases are required for optimal TORC2 function. J Cell Biol 218: 961-976. Locke MN, Thorner J. 2019b. Regulation of TORC2 function and localization by Rab5 GTPases in Saccharomyces cerevisiae. Cell Cycle, in press. Locke MN, Thorner J. 2019c. In vitro protein kinase assay for TORC2 from Saccharomyces cerevisiae. Bio-protocol, in press. Loewith R, Hall MN. 2011. Target of rapamycin (TOR) in nutrient signaling and growth control. Genetics 189: 1177-1201. Loewith R, Jacinto E, Wullschleger S, Lorberg A, Crespo JL, Bonenfant D, Oppliger W, Jenoe P, Hall MN. 2002. Two TOR complexes, only one of which is rapamycin sensitive, have distinct roles in cell growth control. Mol Cell 10: 457-468. Long X, Lin Y, Ortiz-Vega S, Yonezawa K, Avruch J. 2005. Rheb binds and regulates the mTOR kinase. Curr Biol 15: 702-713. Lopez MS, Kliegman JI, Shokat KM. 2014. The logic and design of analog-sensitive kinases and their small molecule inhibitors. Methods Enzymol 548: 189-213. Lu M, Wang J, Ives HE, Pearce D. 2011. mSIN1 protein mediates SGK1 protein interaction with mTORC2 protein complex and is required for selective activation of the epithelial sodium channel. J Biol Chem 286: 30647-30654. Lučić I, Rathinaswamy MK, Truebestein L, Hamelin DJ, Burke JE, Leonard TA. 2018. Conformational sampling of membranes by Akt controls its activation and inactivation. Proc Natl Acad Sci USA 115: E3940-E3949. MacDonald C, Piper RC. 2016. Cell surface recycling in yeast: mechanisms and machineries. Biochem Soc Trans 44: 474-478. Mann HB, Whitney DR. 1947. On a test of whether one of two random variables is stochastically larger than the other. Ann Math Stat 18: 50-60.

61 Manning G, Plowman GD, Hunter T, Sudarsanam S. 2002a. Evolution of protein kinase signaling from yeast to man. Trend Biochem. Sci 27: 514-520. Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S. 2002b. The protein kinase complement of the human genome. Science 298: 1912-1934. Martel RR, Klicius J, Galet S. 1977. Inhibition of the immune response by rapamycin, a new antifungal antibiotic. Can J Physiol Pharmacol 55: 48-51. Martinez Marshall MN, Emmerstorfer-Augustin A, Leskoske KL, Zhang LH, Li B, Thorner J (2019) Analysis of the roles of phosphatidylinositol-4,5-bisphosphate and individual subunits in assembly, localization and function of Saccharomyces cerevisiae Target of Rapamycin Complex 2. Mol. Biol. Cell, in press. McDonald NA, Gould KL. 2016. Linking up at the BAR: oligomerization and F-BAR protein function. Cell Cycle 15: 1977-1985. Megyeri M, Riezman H, Schuldiner M, Futerman AH. 2016. Making sense of the yeast sphingolipid pathway. J. Mol. Biol. 428: 4765-4775. Miyake Y, Kozutsumi Y, Nakamura S, Fujita T, Kawasaki T. 1995. Serine palmitoyltransferase is the primary target of a sphingosine-like immunosuppressant, ISP-1/myriocin. Biochem Biophys Res Commun 211: 396-403. Mok J, Kim PM, Lam HY, Piccirillo S, Zhou X, Jeschke GR, Sheridan DL, Parker SA, Desai V, Jwa M et al. 2010. Deciphering protein kinase specificity through large-scale analysis of yeast phosphorylation site motifs. Sci Signal 3: ra12. Mora A, Komander D, van Aalten DM, Alessi DR. 2004. PDK1, the master regulator of AGC kinase signal transduction. Semin Cell Dev Biol 15: 161-170. Morales-Johansson H, Jenoe P, Cooke FT, Hall MN. 2004. Negative regulation of phosphatidylinositol 4,5-bisphosphate levels by the INP51-associated proteins TAX4 and IRS4. J Biol Chem 279: 39604-39610. Moreira KE, Schuck S, Schrul B, Frohlich F, Moseley JB, Walther TC, Walter P. 2012. Seg1 controls eisosome assembly and shape. J Cell Biol 198: 405-420. Muir A, Ramachandran S, Roelants FM, Timmons G, Thorner J. 2014. TORC2-dependent protein kinase Ypk1 phosphorylates ceramide synthase to stimulate synthesis of complex sphingolipids. Elife 3. Muir A, Roelants FM, Timmons G, Leskoske KL, Thorner J. 2015. Down-regulation of TORC2- Ypk1 signaling promotes MAPK-independent survival under hyperosmotic stress. Elife 4. Muthusamy BP, Raychaudhuri S, Natarajan P, Abe F, Liu K, Prinz WA, Graham TR. 2009. Control of protein and sterol trafficking by antagonistic activities of a type IV P-type ATPase and oxysterol binding protein homologue. Mol Biol Cell 20: 2920-2931. Nakano K, Yamamoto T, Kishimoto T, Noji T, Tanaka K. 2008. Protein kinases Fpk1p and Fpk2p are novel regulators of phospholipid asymmetry. Mol Biol Cell 19: 1783-1797. Nakatsukasa K, Kanada A, Matsuzaki M, Byrne SD, Okumura F, Kamura T. 2014. The nutrient stress-induced small GTPase Rab5 contributes to the activation of vesicle trafficking and vacuolar activity. J Biol Chem 289: 20970-20978. Nakayama KI, Nakayama K. 2005. Regulation of the cell cycle by SCF-type ubiquitin ligases. Semin Cell Dev Biol 16: 323-333. Nicastro R, Sardu A, Panchaud N, De Virgilio C. 2017. The Architecture of the Rag GTPase Signaling Network. Biomolecules 7: E48,.

62 Nickerson DP, Russell MRG, Lo SY, Chapin HC, Milnes J, Merz AJ. 2012. Termination of isoform-selective Vps21/Rab5 signaling at endolysosomal organelles by Msb3/Gyp3. Traffic 13: 1411-1428. Nikko E, Pelham HR. 2009. Arrestin-mediated endocytosis of yeast plasma membrane transporters. Traffic 10: 1856-1867. Niles BJ, Mogri H, Hill A, Vlahakis A, Powers T. 2012. Plasma membrane recruitment and activation of the AGC kinase Ypk1 is mediated by target of rapamycin complex 2 (TORC2) and its effector proteins Slm1 and Slm2. Proc Natl Acad Sci U S A 109: 1536- 1541. Nishitani H, Lygerou Z, Nishimoto T, Nurse P. 2000. The Cdt1 protein is required to license DNA for replication in fission yeast. Nature 404: 625-628. Nishitani H, Sugimoto N, Roukos V, Nakanishi Y, Saijo M, Obuse C, Tsurimoto T, Nakayama KI, Nakayama K, Fujita M et al. 2006. Two E3 ubiquitin ligases, SCF-Skp2 and DDB1- Cul4, target human Cdt1 for proteolysis. EMBO J 25: 1126-1136. Novick P. 2016. Regulation of membrane traffic by Rab GEF and GAP cascades. Small GTPases 7: 252-256. Okabe H, Lee SH, Phuchareon J, Albertson DG, McCormick F, Tetsu O. 2006. A critical role for FBXW8 and MAPK in cyclin D1 degradation and cancer cell proliferation. PLoS One 1: e128. Olson DK, Frohlich F, Farese RV, Jr., Walther TC. 2016. Taming the sphinx: Mechanisms of cellular sphingolipid homeostasis. Biochim Biophys Acta 1861: 784-792. Omnus DJ, Manford AG, Bader JM, Emr SD, Stefan CJ. 2016. Phosphoinositide kinase signaling controls ER-PM cross-talk. Mol Biol Cell 27: 1170-1180. Paulsel AL, Merz AJ, Nickerson DP. 2013. Vps9 family protein Muk1 is the second Rab5 guanosine nucleotide exchange factor in budding yeast. J Biol Chem 288: 18162-18171. Pearce LR, Komander D, Alessi DR. 2010. The nuts and bolts of AGC protein kinases. Nat Rev Mol Cell Biol 11: 9-22. Peng J, Schwartz D, Elias JE, Thoreen CC, Cheng D, Marsischky G, Roelofs J, Finley D, Gygi SP. 2003. A proteomics approach to understanding protein ubiquitination. Nat Biotechnol 21: 921-926. Peplowska K, Markgraf DF, Ostrowicz CW, Bange G, Ungermann C. 2007. The CORVET tethering complex interacts with the yeast Rab5 homolog Vps21 and is involved in endolysosomal biogenesis. Developmental cell 12: 739-750. Pfeffer S, Aivazian D. 2004. Targeting Rab GTPases to distinct membrane compartments. Nat Rev Mol Cell Biol 5: 886-896. Plemel RL, Lobingier BT, Brett CL, Angers CG, Nickerson DP, Paulsel A, Sprague D, Merz AJ. 2011. Subunit organization and Rab interactions of Vps-C protein complexes that control endolysosomal membrane traffic. Mol Biol Cell 22: 1353-1363. Remenyi A, Good MC, Lim WA. 2006. Docking interactions in protein kinase and phosphatase networks. Curr Opin Struct Biol 16: 676-685. Riggi M, Niewola-Staszkowska K, Chiaruttini N, Colom A, Kusmider B, Mercier V, Soleimanpour S, Stahl M, Matile S, Roux A et al. 2018. Decrease in plasma membrane tension triggers PtdIns(4,5)P2 phase separation to inactivate TORC2. Nat Cell Biol 20: 1043-1051. Rink J, Ghigo E, Kalaidzidis Y, Zerial M. 2005. Rab conversion as a mechanism of progression from early to late endosomes. Cell 122: 735-749.

63 Rispal D, Eltschinger S, Stahl M, Vaga S, Bodenmiller B, Abraham Y, Filipuzzi I, Movva NR, Aebersold R, Helliwell SB et al. 2015. Target of Rapamycin Complex 2 Regulates Actin Polarization and Endocytosis via Multiple Pathways. J Biol Chem 290: 14963-14978. Roelants FM, Baltz AG, Trott AE, Fereres S, Thorner J. 2010. A protein kinase network regulates the function of aminophospholipid flippases. Proc Natl Acad Sci U S A 107: 34- 39. Roelants FM, Breslow DK, Muir A, Weissman JS, Thorner J. 2011. Protein kinase Ypk1 phosphorylates regulatory proteins Orm1 and Orm2 to control sphingolipid homeostasis in Saccharomyces cerevisiae. Proc Natl Acad Sci U S A 108: 19222-19227. Roelants FM, Leskoske KL, Martinez Marshall MN, Locke MN, Thorner J. 2017a. The TORC2- dependent signaling network in the yeast Saccharomyces cerevisiae. Biomolecules 7: E66. Roelants FM, Leskoske KL, Pedersen RT, Muir A, Liu JM, Finnigan GC, Thorner J. 2017b. TOR Complex 2-regulated protein kinase Fpk1 stimulates endocytosis via inhibition of Ark1/Prk1-related protein kinase Akl1 in Saccharomyces cerevisiae. Mol Cell Biol 37. Roelants FM, Torrance PD, Bezman N, Thorner J. 2002. Pkh1 and Pkh2 differentially phosphorylate and activate Ypk1 and Ykr2 and define protein kinase modules required for maintenance of cell wall integrity. Mol Biol Cell 13: 3005-3028. Roelants FM, Torrance PD, Thorner J. 2004. Differential roles of PDK1- and PDK2- phosphorylation sites in the yeast AGC kinases Ypk1, Pkc1 and Sch9. Microbiology 150: 3289-3304. Roland BP, Graham TR. 2016. Decoding P4-ATPase substrate interactions. Crit Rev Biochem Mol Biol 51: 513-527. Rubenstein EM, Schmidt MC. 2007. Mechanisms regulating the protein kinases of Saccharomyces cerevisiae. Eukaryot Cell 6: 571-583. Saito H, Posas F. 2012. Response to hyperosmotic stress. Genetics 192: 289-318. Saito K, Fujimura-Kamada K, Hanamatsu H, Kato U, Umeda M, Kozminski KG, Tanaka K. 2007. Transbilayer phospholipid flipping regulates Cdc42p signaling during polarized cell growth via Rga GTPase-activating proteins. Developmental cell 13: 743-751. Sakaue-Sawano A, Kurokawa H, Morimura T, Hanyu A, Hama H, Osawa H, Kashiwagi S, Fukami K, Miyata T, Miyoshi H et al. 2008. Visualizing spatiotemporal dynamics of multicellular cell-cycle progression. Cell 132: 487-498. Sancak Y, Sabatini DM. 2009. Rag proteins regulate amino-acid-induced mTORC1 signalling. Biochemical Society transactions 37: 289-290. Saxton RA, Sabatini DM. 2017. mTOR Signaling in Growth, Metabolism, and Disease. Cell 168: 960-976. Schimmoller F, Riezman H. 1993. Involvement of Ypt7p, a small GTPase, in traffic from late endosome to the vacuole in yeast. Journal of cell science 106 ( Pt 3): 823-830. Schmidt O, Weyer Y, Fink MJ, Muller M, Weys S, Bindreither M, Teis D. 2017. Regulation of Rab5 isoforms by transcriptional and post-transcriptional mechanisms in yeast. FEBS Lett 591: 2803-2815. Sebastian TT, Baldridge RD, Xu P, Graham TR. 2012. Phospholipid flippases: building asymmetric membranes and transport vesicles. Biochim Biophys Acta 1821: 1068-1077. Shaner NC, Lambert GG, Chammas A, Ni Y, Cranfill PJ, Baird MA, Sell BR, Allen JR, Day RN, Israelsson M et al. 2013. A bright monomeric green fluorescent protein derived from Branchiostoma lanceolatum. Nat Methods 10: 407-409.

64 Shideler T, Nickerson DP, Merz AJ, Odorizzi G. 2015. Ubiquitin binding by the CUE domain promotes endosomal localization of the Rab5 GEF Vps9. Mol Biol Cell 26: 1345-1356. Shiozaki K, Tatebe H, Murayama S, Kojima C. 2013. Mechanisms that determine the substrate specificity of TOR kinase. in Cell Biology of Yeasts (eds. M Cyert, DJ Lew, K Sawin, Organizers), p. 68 Abstr. Cold Spring Harbor Laboratory Press, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Sikorski RS, Hieter P. 1989. A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics 122: 19-27. Simanshu DK, Nissley DV, McCormick F. 2017. RAS Proteins and Their Regulators in Human Disease. Cell 170: 17-33. Singer-Kruger B, Stenmark H, Dusterhoft A, Philippsen P, Yoo JS, Gallwitz D, Zerial M. 1994. Role of three rab5-like GTPases, Ypt51p, Ypt52p, and Ypt53p, in the endocytic and vacuolar protein sorting pathways of yeast. J Cell Biol 125: 283-298. Singer-Kruger B, Stenmark H, Zerial M. 1995. Yeast Ypt51p and mammalian Rab5: counterparts with similar function in the early endocytic pathway. Journal of cell science 108 ( Pt 11): 3509-3521. Smith JR, Maguire S, Davis LA, Alexander M, Yang F, Chandran S, ffrench-Constant C, Pedersen RA. 2008. Robust, persistent transgene expression in human embryonic stem cells is achieved with AAVS1-targeted integration. Stem Cells 26: 496-504. Stegmeier F, Hu G, Rickles RJ, Hannon GJ, Elledge SJ (2005) A lentiviral microRNA-based system for single-copy polymerase II-regulated RNA interference in mammalian cells. Proc Natl Acad Sci USA 102: 13212-13217. Sturgill TW, Cohen A, Diefenbacher M, Trautwein M, Martin DE, Hall MN. 2008. TOR1 and TOR2 have distinct locations in live cells. Eukaryot Cell 7: 1819-1830. Suda Y, Kurokawa K, Hirata R, Nakano A. 2013. Rab GAP cascade regulates dynamics of Ypt6 in the Golgi traffic. Proc Natl Acad Sci U S A 110: 18976-18981. Sugimoto Y, Sakoh H, Yamada K. 2004. IPC synthase as a useful target for antifungal drugs. Curr Drug Targets Infect Disord 4: 311-322. Sun Y, Miao Y, Yamane Y, Zhang C, Shokat KM, Takematsu H, Kozutsumi Y, Drubin DG. 2012. Orm protein phosphoregulation mediates transient sphingolipid biosynthesis response to heat stress via the Pkh-Ypk and Cdc55-PP2A pathways. Mol Biol Cell 23: 2388-2398. Sun YD, Taniguchi R, Tanoue D, Yamaji T, Takematsu H, Mori K, Fujita T, Kawasaki T, Kozutsumi Y. 2000. Sli2 (Ypk1), a homologue of mammalian protein kinase SGK, is a downstream kinase in the sphingolipid-mediated signaling pathway of yeast. Mol Cell Biol 20: 4411-4419. Sung MK, Huh WK. 2007. Bimolecular fluorescence complementation analysis system for in vivo detection of protein-protein interaction in Saccharomyces cerevisiae. Yeast 24: 767- 775. Swaney DL, Beltrao P, Starita L, Guo A, Rush J, Fields S, Krogan NJ, Villen J. 2013. Global analysis of phosphorylation and ubiquitylation cross-talk in protein degradation. Nat Methods 10: 676-682. Tabuchi M, Audhya A, Parsons AB, Boone C, Emr SD. 2006. The phosphatidylinositol 4,5- biphosphate and TORC2 binding proteins Slm1 and Slm2 function in sphingolipid regulation. Mol Cell Biol 26: 5861-5875.

65 Tada S. 2007. Cdt1 and geminin: role during cell cycle progression and DNA damage in higher eukaryotes. Front Biosci 12: 1629-1641. Tall GG, Hama H, DeWald DB, Horazdovsky BF. 1999. The phosphatidylinositol 3-phosphate binding protein Vac1p interacts with a Rab GTPase and a Sec1p homologue to facilitate vesicle-mediated vacuolar protein sorting. Mol Biol Cell 10: 1873-1889. Tanoue D, Kobayashi T, Sun Y, Fujita T, Takematsu H, Kozutsumi Y. 2005. The requirement for the hydrophobic motif phosphorylation of Ypk1 in yeast differs depending on the downstream events, including endocytosis, cell growth, and resistance to a sphingolipid biosynthesis inhibitor, ISP-1. Arch Biochem Biophys 437: 29-41. Tanoue T, Nishida E. 2003. Molecular recognitions in the MAP kinase cascades. Cellul. Signal. 15: 455-462. Tatebe H, Morigasaki S, Murayama S, Zeng CT, Shiozaki K. 2010. Rab-family GTPase regulates TOR complex 2 signaling in fission yeast. Curr Biol 20: 1975-1982. Tatebe H, Murayama S, Yonekura T, Hatano T, Richter D, Furuya T, Kataoka S, Furuita K, Kojima C, Shiozaki K. 2017. Substrate specificity of TOR complex 2 is determined by a ubiquitin-fold domain of the Sin1 subunit. Elife 6: e19594.19591-e19594.19520. Tatebe H, Shiozaki K. 2017. Evolutionary conservation of the components in the TOR signaling pathways. Biomolecules 7: E77. Teixeira LK, Reed SI. 2013. Ubiquitin ligases and cell cycle control. Annu Rev Biochem 82: 387- 414. Toshima JY, Nishinoaki S, Sato Y, Yamamoto W, Furukawa D, Siekhaus DE, Sawaguchi A, Toshima J. 2014. Bifurcation of the endocytic pathway into Rab5-dependent and - independent transport to the vacuole. Nat Commun 5: 3498. Tzeng YW, Li DY, Chen Y, Yang CH, Chang CY, Juang YL. 2018. LMO7 exerts an effect on mitosis progression and the spindle assembly checkpoint. Int J Biochem Cell Biol 94: 22- 30. Ubersax JA, Ferrell JE, Jr. 2007. Mechanisms of specificity in protein phosphorylation. Nat Rev Mol Cell Biol 8: 530-541. Vanhaesebroeck B, Alessi DR. 2000. The PI3K-PDK1 connection: more than just a road to PKB. Biochem J 346: 561-576. Vezina C, Kudelski A, Sehgal SN. 1975. Rapamycin (AY-22,989), a new antifungal antibiotic. I. Taxonomy of the producing streptomycete and isolation of the active principle. J Antibiot (Tokyo) 28: 721-726. Viadiu H, Gonen T, Walz T. 2007. Projection map of aquaporin-9 at 7 A resolution. J Mol Biol 367: 80-88. Vodermaier HC. 2004. APC/C and SCF: controlling each other and the cell cycle. Curr Biol 14: R787-796. Vonkova I, Saliba AE, Deghou S, Anand K, Ceschia S, Doerks T, Galih A, Kugler KG, Maeda K, Rybin V et al. 2015. Lipid Cooperativity as a General Membrane-Recruitment Principle for PH Domains. Cell Rep 12: 1519-1530. Walther TC, Aguilar PS, Frohlich F, Chu F, Moreira K, Burlingame AL, Walter P. 2007. Pkh- kinases control eisosome assembly and organization. EMBO J 26: 4946-4955. Walther TC, Brickner JH, Aguilar PS, Bernales S, Pantoja C, Walter P. 2006. Eisosomes mark static sites of endocytosis. Nature 439: 998-1003. Wandinger-Ness A, Zerial M. 2014. Rab proteins and the compartmentalization of the endosomal system. Cold Spring Harb Perspect Biol 6: a022616.

66 Wang H, Chen Y, Lin P, Li L, Zhou G, Liu G, Logsdon C, Jin J, Abbruzzese JL, Tan TH et al. 2014a. The CUL7/F-box and WD repeat domain containing 8 (CUL7/Fbxw8) ubiquitin ligase promotes degradation of hematopoietic progenitor kinase 1. J Biol Chem 289: 4009-4017. Wang J, Dye BT, Rajashankar KR, Kurinov I, Schulman BA. 2009. Insights into Anaphase Promoting Complex TPR subdomain assembly from a CDC26–APC6 structure. Nat Struct Mol Biol 16: 987-989. Wang Z, Liu P, Inuzuka H, Wei W. 2014b. Roles of F-box proteins in cancer. Nat Rev Cancer 14: 233-247. Waterhouse A, Bertoni M, Bienert S, Studer G, Tauriello G, Gumienny R, Heer FT, de Beer TAP, Rempfer C, Bordoli L et al. 2018. SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic acids research 46: W296-w303. Wedaman KP, Reinke A, Anderson S, Yates J, 3rd, McCaffery JM, Powers T. 2003. Tor kinases are in distinct membrane-associated protein complexes in Saccharomyces cerevisiae. Mol Biol Cell 14: 1204-1220. Wheeler DB, Zoncu R, Root DE, Sabatini DM, Sawyers CL. 2015. Identification of an oncogenic RAB protein. Science 350: 211-217. Wullschleger S, Loewith R, Oppliger W, Hall MN. 2005. Molecular organization of target of rapamycin complex 2. J Biol Chem 280: 30697-30704. Xia G, Luo X, Habu T, Rizo J, Matsumoto T, Yu H. 2004. Conformation-specific binding of p31(comet) antagonizes the function of Mad2 in the spindle checkpoint. EMBO J 23: 3133-3143. Yang H, Jiang X, Li B, Yang HJ, Miller M, Yang A, Dhar A, Pavletich NP. 2017. Mechanisms of mTORC1 activation by RHEB and inhibition by PRAS40. Nature 552: 368-373. Yang H, Rudge DG, Koos JD, Vaidialingam B, Yang HJ, Pavletich NP. 2013. mTOR kinase structure, mechanism and regulation. Nature 497: 217-223. Yang M, Li B, Tomchick DR, Machius M, Rizo J, Yu H, Luo X. 2007. p31comet blocks Mad2 activation through structural mimicry. Cell 131: 744-755. Yang Q, Inoki K, Ikenoue T, Guan KL. 2006. Identification of Sin1 as an essential TORC2 component required for complex formation and kinase activity. Gene Dev 20: 2820-2832. Yao CA, Ortiz-Vega S, Sun YY, Chien CT, Chuang JH, Lin Y. 2017. Association of mSin1 with mTORC2 Ras and Akt reveals a crucial domain on mSin1 involved in Akt phosphorylation. Oncotarget 8: 63392-63404. Ye Y, Rape M. 2009. Building ubiquitin chains: E2 enzymes at work. Nat Rev Mol Cell Biol 10: 755-764. Yen HC, Elledge SJ. 2008. Identification of SCF ubiquitin ligase substrates by global protein stability profiling. Science 322: 923-929. Yeung BK. 2011. Natural product drug discovery: the successful optimization of ISP-1 and halichondrin B. Curr Opin Chem Biol 15: 523-528. Yu JW, Mendrola JM, Audhya A, Singh S, Keleti D, DeWald DB, Murray D, Emr SD, Lemmon MA. 2004. Genome-wide analysis of membrane targeting by S. cerevisiae pleckstrin homology domains. Mol Cell 13: 677-688. Zimmermann C, Santos A, Gable K, Epstein S, Gururaj C, Chymkowitch P, Pultz D, Rodkaer SV, Clay L, Bjoras M et al. 2013. TORC1 inhibits GSK3-mediated Elo2 phosphorylation to regulate very long chain fatty acid synthesis and autophagy. Cell Rep 5: 1036-1046.

67 Appendix A

Interrogating substrate recognition by Ypk1— Does Ypk1 bind to a specific docking motif distinct from its consensus phospho-acceptor site?

Background Ypk1 is an evolutionarily conserved protein kinase and a member of the AGC sub-family of protein kinases (Pearce et al. 2010), which make up over 10% of the 518 enzymes that comprise the total human kinome (Manning et al. 2002b). Strikingly, the AGC kinases have been highly conserved throughout eukaryotic evolution, and SGK1 (the human homolog of Ypk1) rescues the inviability of ypk1∆ ypk2∆ cells (Casamayor et al. 1999). Thus, discerning the molecular basis by which Ypk1 selects its substrates, and how Ypk1 interacts with its substrates in a regulated fashion, is important to our understanding of how this vital signaling pathway works. Understanding the molecular logic of these interactions will provide invaluable insight into how to directly target these signaling cascades, many of which are dysregulated in human cancers and disease (Saxton and Sabatini 2017). A biochemical means to manipulate the spatial and temporal regulation of the TORC2-Ypk1 signaling network could lay groundwork for small-molecule screens aimed at enhancing the specificity and effectiveness against therapeutic targets.

To fully understand how the TORC2-Ypk1 signaling network acts to maintain plasma membrane homeostasis, it is necessary to characterize at the molecular level how substrate specificity is achieved to ensure proper signaling outputs. The Thorner lab was the first to show that a mitogen-activated protein kinase (MAPK) achieves exquisite substrate specificity by utilizing evolutionarily conserved high-affinity docking sites distal to the target phosphorylation sites composed of cognate linear sequences with a complementary molecular signatures present on both the kinase and substrate (Bardwell and Thorner 1996; Chen and Thorner 2007). This docking paradigm has since been discovered in a diverse range of other kinases (Tanoue and Nishida 2003; Bardwell 2006; Remenyi et al. 2006; Ubersax and Ferrell 2007)

I hypothesized that there is a molecular signature common to all Ypk1 substrates that enables the kinase to achieve high specificity and prevents aberrant targeting of incorrect substrates. Although the consensus phosphorylation motif RxRxxS/T (where x denotes any amino acid) is found in hundreds of the approximately 6,000 protein encoded in the S. cerevisiae genome to date, our laboratory and others have found that only a select cohort of proteins is actually phosphorylated by Ypk1. Studies of other classes of eukaryotic protein kinases have revealed that distal docking sites provide a high level of specificity by increasing affinity for certain substrates, thereby enhancing precise phosphorylation (Holland and Cooper 1999; Biondi and Nebreda 2003; Goldsmith et al. 2007; Bardwell and Bardwell 2015).

There was preliminary evidence that, in addition to its specific phospho-acceptor site sequence specificity, Ypk1 may interact with prospective substrates via a docking interaction between a region of the kinase different from its active site and a sequence in the substrate different from its phospho-acceptor site. In vivo the aquaglyceroporin Fps1 is a transmembrane protein (Beese- Sims et al. 2011) and it is phosphorylated equivalently on Ypk1 consensus sites in both its cytosolic N- and cytosolic C-termini; in vitro, however, a recombinant C-terminal fragment was reported to be a much more efficient substrate for purified Ypk1 than an amino-terminal 68 fragment (Muir et al. 2015). These results suggested that, in the cell, Ypk1 may dock at the C- terminus of intact Fps1 and, once tethered there, be able to more effectively phosphorylate its own amino-terminus [on the same or neighboring subunits, if Fps1 is a tetrameric protein, like its mammalian counterpart AQP9 (Viadiu et al. 2007)].

By dissecting the sequence determinants necessary and sufficient for the high-affinity binding of Ypk1 to several of its most efficient known substrates, I sought to characterize the molecular requirements for the substrate selectivity of Ypk1, which may aid identification of additional novel substrates in this pathway. To elucidate whether Ypk1 recognizes a docking site in its substrates, I devised and implemented an in vitro MBP-based pull-down assay to assess Ypk1- substrate interaction. As a test substrate, I chose the plasma membrane-localized aquaglyceroporin Fps1 because it is a well characterized direct substrate of Ypk1 (Muir et al. 2015).

Results

The carboxy-terminus of Fps1 is sufficient for Ypk1 binding in vitro To uncover substrate sequence determinants necessary for interaction with Ypk1, I sought first to develop a binding assay to detect a direct interaction in vitro. For this approach, I used purified recombinant MBP-fusions of Fps1 immobilized on amylose resin and incubated with purified D242A D470A KD Ypk1 (hereafter Ypk1C ). Genetic and biochemical evidence suggests that the D242A mutation alleviates inhibition of the kinase domain by its N-terminal negative regulatory domain (Aronova et al. 2008) and, thus, makes the kinase "constitutively" (C) active, thereby ensuring that the kinase is in a conformation accessible to bind substrates. I also included the D470A mutation, which eliminates a residue essential for catalysis ("kinase dead" or KD), to prevent any phosphorylation-induced substrate dissociation. I reasoned that the kinase-dead version of Ypk1 would enable me to specifically assess the contribution of sequence determinants on the ability of Ypk1 to bind without the additional input of increased off-rate induced by phosphorylation.

KD In multiple trials, there was robust interaction between Ypk1C and the Fps1 C-terminus (MBP- Fps1531-669), but not with MBP or the N-terminus of Fps1 (MBP-Fps11-225) (Fig. A1).

Figure A1. Recombinant Fps1 interacts with Ypk1 in vitro. Binding assay of MBP and MBP- Fps1 fragments purified from E. coli immobilized on amylose resin and incubated with purified D242A D470A KD Ypk1 -HA (Ypk1C -HA) from S. cerevisiae. After stringent washes, bound proteins were resolved by SDS-PAGE and specific binding was determined by immunoblotting with anti-HA antibodies to detect Ypk1 and Ponceau S to visualize recombinant MBP proteins.

To narrow down the region in the C-terminus of Fps1 important for the observed interaction, I KD generated a nested set of truncation mutants (Fig. A2A) and tested them for binding of Ypk1C (Figs. A1 and A2B).

70 A"

Figure A2. A minimal Fps1 peptide is sufficient for binding Ypk1 in vitro. (A) Schematic of full- length Fps1 protein and truncated constructs cloned as MBP fusions that were tested and the results of binding experiments. Red asterisk (*) indicates phosphorylated residue. (B) Binding assay of MBP and MBP-Fps1 fragments purified from E. coli immobilized on amylose resin and incubated D242A D470A KD with purified Ypk1 -HA (Ypk1C -HA) from S. cerevisiae. After stringent washes, bound proteins were resolved by SDS-PAGE and specific binding was determined by immunoblotting with anti-HA antibodies to detect Ypk1 and Ponceau S to detect recombinant MBP proteins. Residues mutated in Fps1(531-573; K561A K562A R563A R565A) are indicated as AAANA. (C) Sequence of the minimal Fps1 peptide tested that bound Ypk1, residues 554-573. Consensus Ypk1 phospho- acceptor site bold; red, phosphorylated residue.

71 One conclusion of these binding data is that the minimal epitope in Fps1 required for interaction KD with Ypk1C is confined to a linear sequence element. The minimal region that was sufficient for this interaction was a 20-residue segment of Fps1 encompassing amino acids 554-573 (Fig. A2C). The fragments I tested that lacked the phospho-acceptor motif did not bind to Ypk1, indicating that its presence is required for interaction. Immediately upstream of the C-terminal phospho-acceptor sequence in Fps1 is a striking stretch of positively charged residues (KKRNR residues 561-565). To test whether these positions are important for interaction with Ypk1, I generated a corresponding site-directed mutant, Fps1(531-573; K561A K562A R563A R565A). Unfortunately, this protein did not express well in E. coli and I was unable to recover a sufficient amount of this construct to definitively determine whether its apparently poor binding of KD Ypk1C was due to loss of binding affinity or simply due to the very low level of Fps1(531-573; K561A K562A R563A R565A) present (Fig. A2B). Further optimization will be needed to recover more of this construct to resolve this issue.

Based on previously characterized Ypk1 substrates (Fpk1, Fpk2, Orm1, Orm2, and Gpd1) and the in vitro sequence preference identified for Ypk1 phosphorylation of synthetic peptides (Casamayor et al. 1999; Mok et al. 2010), former Thorner lab graduate student Alex Muir developed a position-weighted consensus sequence logo for Ypk1 substrates (Muir et al. 2014): RxRxxS/T. As I found in an additional validated substrate Muk1 (see Chapter 3), the presence of Arg at positions -5 and -3 is invariant with respect to the phosphorylated residue. To query the stringency of the consensus sequence, I made Arg to Lys mutations in the phospho-motif of Fps1 and, additionally, mutated the Ser to Thr (Fig. A3A), as the consensus logo revealed a preference KD for phosphorylation on a Ser residue. Unexpectedly, I found that Ypk1C bound to the mutant KxKxxT Fps1 construct just as well or perhaps better than it did to the Fps1 construct with the wild-type consensus sequence (Fig. A3B). Thus, as far as binding goes, Lys at these positions is tolerated (as is Thr at the phospho-acceptor site). However, as far as phosphorylation goes, I have not tested whether this construct is a substrate for active Ypk1. In any event, I speculate that even though the Lys (and Thr) changes might promote tighter binding to Ypk1, stronger association might be detrimental for a kinase that requires a high turnover rate. Finally, to determine if this assay approach could be extended to other Ypk1 substrates, I also tested an MBP fusion to full-length Muk1 and found that, like the Fps1 C-terminal fragment, full-length Muk1 associated with Ypk1 in vitro (Fig. A3B).

Fig A3. Conservative mutation of the consensus Ypk1 phospho-acceptor site on Fps1 does not diminish interaction with Ypk1. (A) Sequence of the consensus Ypk1 phospho-acceptor site on the C-terminus of Fps1 (wild-type) and resides mutated (phospho-site mutant) to test sequence stringency. Invariant residues, bold blue; phosphorylated residue, red; mutated residues italicized and underlined. (B) Binding assay of MBP and MBP-Fps1 fragments purified from E. coli D242A D470A KD immobilized on amylose resin and incubated with purified Ypk1 -HA (Ypk1C -HA) from S. cerevisiae. After stringent washes, bound proteins were resolved by SDS-PAGE and specific binding was determined by immunoblotting with anti-HA antibodies to detect Ypk1 and Ponceau S to detect recombinant MBP proteins. Asterisks indicates co-purifying, non-specific species. Residues mutated in Fps1(531-669; R565K R567K S570T) are indicated as KxKxxT.

Discussion These experiments were initiated to elucidate the interaction motifs on substrates that were important for their interaction with Ypk1. I established a robust MBP-based in vitro pull-down assay that reproducibly detected a specific interaction between Ypk1 and its substrates Fps1 and Muk1. Using this assay, I found that the C-terminal cytosolic tail of the known Ypk1 substrate

73 Fps1 binds Ypk1 with high affinity. Moreover, a nested set of truncation mutants suggests that the minimal binding epitopes are confined to a linear sequence that includes the phospho- acceptor motif.

A founding member of the AGC protein kinase sub-family is 3',5'-cyclic-AMP-dependent protein kinase (PKA). High-affinity substrates and inhibitors of PKA bind both in the catalytic pocket (via their phospho-acceptor motif) and within a prominent groove that extends from the active site across the lower lobe of the kinase domain (via residues upstream of the phospho- acceptor motif) (Knighton et al. 1991) (Fig. A4). Based on my finding that a linear segment of

Occupancy of active site and extended docking groove

Active Site

Docking Groove

PKA PKA + PKI (red) Fig. A4. PKA in the absence (left) and presence (right) of a PKI-derived peptide. Surface rendering of crystal structures of PKA alone or bound to a 20-residue segment (residues 6-25) of the 76-residue human PKA inhibitor peptide PKIα (product of the PKIA gene) determined by Knighton et al. (1991).

Fps1 extending 11-residues upstream from the phospho-acceptor motif (Fig. A2C) exhibited KD high-affinity binding to Ypk1C , Ypk1 likely selects its substrates in a manner similar to PKA, rather than utilizing a separate and more distantly located docking site, as do MAPKs. To determine if this element is sufficient for Ypk1 recruitment, it would be informative to install it on a carrier protein with which Ypk1 normally does not interact and test the ability of these fusion proteins to associate with Ypk1.

74 Appendix B

Identification of ubiquitylation enzymes that regulate cell cycle progression

Background Ubiquitylation, the covalent attachment of ubiquitin to a protein, is an essential signaling process. The attachment of a single ubiquitin, monoubiquitylation, regulates nonproteolytic protein-protein interactions (Husnjak and Dikic 2012). Polyubiquitylation, the formation of ubiquitin chains on substrates, can mediate protein interactions, alter protein conformation, or act as a destruction signal for proteasomal degradation (Komander and Rape 2012). Thus, modification with ubiquitin can provide exquisite spatial and temporal regulation of protein localization, interaction, and degradation. Ubiquitylation of a given substrate in human cells is catalyzed successively by a cascade of either of two ubiquitin activating enzymes (E1), one of nearly 40 ubiquitin conjugating enzymes (E2), and one of more than 600 ubiquitin ligases (E3) (Ye and Rape 2009). Of the multitude of E3s identified, fewer than 10% have been well characterized and fewer still have identified substrates (Peng et al. 2003; Yen and Elledge 2008; Harper and Tan 2012).

Like phosphorylation, ubiquitylation plays important roles in virtually all cellular processes. The execution of an accurate cell division program is essential for all living cells, and dynamic signaling by ubiquitin ligases has been shown to regulate cell cycle checkpoints to ensure precise progression through the cell cycle. The Skp1-Cul1-F-box protein (SCF) type and the Anaphase Promoting Complex/Cyclosome (APC/C) type ligases are the best characterized E3 complexes involved in the cell cycle, regulating cell cycle checkpoints, thereby creating controlled, unidirectional progression through the division program (Vodermaier 2004; Teixeira and Reed 2013).

The goal of this study was to uncover previously unrecognized ubiquitylation networks that might influence cell cycle progression by using a high-throughput screening approach in human cells. To identify ubiquitin-mediated regulatory mechanisms that control the cell cycle, I first designed a cell line containing a fluorescent reporter. I based this reporter on the previously published FUCCI (Fluorescent, ubiquitination-based cell cycle indicator) system (Sakaue- Sawano et al. 2008); in this dissertation, this optimized system is referred to as FUCCI*. This cell line expresses fluorescently tagged derivatives of the cell cycle regulated proteins Cdt1 and Geminin. As a result, early G1, late G1, the G1/S transition, and S/G2/M cell cycle stages can be distinguished by fluorescence microscopy or flow cytometry, the duration of time individual cells spend during each stage can be monitored, and defects at specific stages can be identified (Fig. B1).

Figure B1. Two-color cell cycle progression reporter. Diagram adapted from Sakaue-Sawano et al. (2008).

Below I describe the high-throughput knock-down screen I performed in this cell line, using a library of short hairpin RNAs (shRNAs) targeting the transcripts of ubiquitin-related genes to identify genes that, when knocked down, alter cell cycle progression.

Results

Construction of cell cycle reporter strain I sought to create a cell line expressing a reporter that could accurately report the cell cycle stage of individual cells in an asynchronous population and enable the visualization of progression through the cell cycle. For this purpose, I created a modified version of the FUCCI-based system (Sakaue-Sawano et al. 2008) utilizing the two cell cycle regulated proteins Cdt1 and Geminin fused to the fluorescent proteins monomeric Azami Green (mAG1) and monomeric Kusabira Orange (mKO2) (Fig. B2A).

Figure B2. How FUCCI fluorescent reporter system enables visualization of cell cycle progression. (A) Two cell cycle reporters, Cdt1(30-120) fused to the red fluorescent protein mKO2 and Geminin(1-110) fused to the green fluorescent protein mAG1 (Sakaue-Sawano et al. 2008). (B) Diagram of cell labeling by FUCCI fluorescent reporters during progression through the cell cycle. (C) Schematic view of the strategy used to target integration of my FUCCI* construct to the adeno- associated virus integration site AAVS1 in the human genome (a bi-cistronic reporter driven by the EF-1α promoter and flanked by 5‘ and 3‘ arms of homology to the AAVS1 locus). The two reporters are split by a self-cleaving P2A peptide derived from the picornavirus leading to two separate peptides after translation (de Felipe et al. 2006; Kim et al. 2011)

Because these proteins regulate licensing of DNA replication, their abundance is tightly regulated and oscillates inversely during the cell cycle (Nishitani et al. 2000) (Figs. B1 and B2B). Cdt1 accumulates during G1 and is ubiquitylated and targeted for degradation at the start of S phase through mitosis (Nishitani et al. 2006; Tada 2007). Geminin, which inhibits Cdt1 function, accumulates during S phase through mitosis, and is ubiquitylated throughout G1 when Cdt1 levels increase (Ballabeni et al. 2013). The FUCCI system takes advantage of the ubiquitylation of these proteins; truncations of fluorescent mKO2-Cdt1(30-120) and mAG1- Geminin(1-110) are constitutively expressed, and protein abundance is regulated by their ubiquitylation and degradation. Thus, in G1, cells are red as they have a high level of mKO2- Cdt1(30-120) expression while mAG1-Geminin(1-110) levels are low, at G1/S transition cells appear yellow as mKO2-Cdt1(30-120) and mAG1-Geminin(1-110) are both expressed, and during S/G2/M phases cells increase in green fluorescence as mKO2-Cdt1(30-120) is ubiquitylated and mAG1-Geminin(1-110) accumulates (Figs. B1 and B2B). Mitosis can be distinguished from G2 in these cells by fluorescence microscopy, as the mAG1-Geminin(1-110) reporter resides in the nucleus until nuclear envelope breakdown at the start of mitosis (Figs. B2B and B3).

Figure B3. Characterization of K562-FUCCI* cell lines. (A) Whole-cell lysates from untargeted (K562-wild type) and FUCCI* targeted clonal lines were separated on SDS-PAGE followed by immunoblot to detect Geminin isoforms (rabbit anti-Geminin, Cell Signaling Technology). The expected molecular weight of each isoform is indicated on the right. (B) FACS plot of K562- FUCCI* clonal cell line (#5). Cells were excited by an argon laser line 488nm and fluorescence detected by PE (for mKO2) and FITC (for mAG1) channels. Cells were gated for G1 (mKO2+/mAG1-), G1/S (mKO2+/mAG1+), and S/G2/M (mKO2-/mAG1+). (C) Still images from timelapse fluorescence imaging of K562-FUCCI* cells progressing through the cell cycle. Images were acquired every 3 min using a confocal microscope. Cells were excited with an argon laser at 488 nm and 561 nm. Time was set to 0:00 (h:min) for initial nuclear envelope breakdown.

78 In the first publication of their FUCCI system, Sakaue-Sawano et al. (2008) integrated the reporter probes mAG1-Geminin(1-110) and mKO2-Cdt1(30-120) by separate lentiviral vectors using distinct promoters (Sakaue-Sawano et al. 2008). Because these viruses randomly integrate the reporters into the genome separately, this method has the potential drawback that one reporter might be silenced over time, leading to inaccurate reporting of the cell cycle stage. To address this issue, I cloned these reporter genes behind a single promoter separated by a self- cleaving P2A sequence (de Felipe et al. 2006; Kim et al. 2011; Liu et al. 2017), into a gene- targeting vector (Fig. B2C). I refer to this bi-cistronic system as FUCCI*. The benefit of this system is these genes are transcribed as a single RNA transcript, and thus cells produce equimolar amounts of the genes products, avoiding uneven expression due to silencing.

To generate a derivative of the human cell line K562 that can grow in suspension culture and expresses this reporter, I used TALEN-mediated double-strand break formation to stimulate homologous recombination (Hockemeyer et al. 2011) to integrate my reporter construct into the safe harbor AAVS1 locus located on chromosome 19 in the first intron of the PPP1R12C gene. This locus is often targeted because the PPP1R12C gene is ubiquitously expressed in multiple cell types and supports the long-term expression of transgenes (Smith et al. 2008). K562 cells are poorly differentiated erythroblast-like leukemia cells (Koeffler and Golde 1980) that are an ideal cell line to use for screening because they are easily sorted in a flow cytometer. To verify that the P2A cleavage was acting as expected, I used immunoblot analysis to determine if mAG1- Geminin(1-110) migrated at the expected 41 kDa size on SDS-PAGE. If the two reporters remained as a fused protein, it would be expected to run at 78 kDa. Of the 6 clones tested, all correctly showed mAG1-Geminin(1-110) migrating at the expected size (Fig. B3A). I then used Fluorescence Activated Cell Sorting (FACS) to sort and obtain single cells that were both mKO2+ and mAG1+, reasoning that these would likely be correctly targeted (Fig. B3B). Next, I performed live-cell fluorescence microscopy to visualize the progression of these cells through the cell cycle. As expected, these cells show dynamic color changes as they progress through the cell division program (Fig. B3C). From the clones isolated, I chose to use clone #5 for the screen, hereafter referred to as FUCCI*-K562 cells.

Pooled shRNA screen to identify genes involved in cell cycle progression To identify ubiquitin-related genes that are involved in cell cycle progression, I collaborated with Eugene Oh and Martin Kampmann in the lab of Jonathan Weissman at UCSF to perform a FACS-based pooled shRNA screen using a library of barcoded shRNAs that targets genes related to ubiquitin signaling pathways.

79 Figure B4. Schematic of cell cycle screen. K562-FUCCI* cells were infected in bulk with barcoded shRNA lentiviral particles. Cells were cultured with blasticidin to select for shRNA integration. Vectors expressing shRNAs also express blue fluorescent protein (BFP). Flow cytometry was used to determine infection rate. After 6 days, FACS was used to sort cells based on cell cycle stage.

This library consisted of 4 sub-libraries: (i) Sub-library 1 contained F-box proteins, SOCS-box proteins, disease-related genes, and those that have specific localization annotations; (ii) Sub- library 2 contained BTB proteins, DCAF proteins, signaling-specific genes, and chromatin- associated genes; (iii) Sub-library 3 contained HECT, U-box, RING-finger, TRIM, and Zn-finger E3s; and, (iv) Sub-library 4 contained general ubiquitin enzymes, including all of the E2s, APC- related components, proteasome-related genes, and most proteases. The total library consisted of ~25 shRNAs per gene targeting 222 genes in each library with ~1000 control non-targeting shRNAs. Each shRNA was expressed from an individual lentiviral plasmid with a Spleen Focus Forming Virus (SFFV) promoter (Demaison et al. 2002) driving a blasticidin-resistance gene and blue fluorescent protein (BFP) for selection and the shRNA in a minimal miR-30 context (Stegmeier et al. 2005).

The purpose of this screen was to identify genes that, when knocked down by shRNA, caused a change in the cell cycle progression of that particular cell. A phenotypic effect in promoting or preventing cell propagation could be read out, respectively, as either an over- or under- representation of that shRNA species compared to control non-targeting shRNAs in a particular cell cycle stage determined after sorting. I performed the screen essentially as described in (Kampmann et al. 2014). The shRNA sub-libraries were first packaged as lentiviruses and viral supernatant was collected and concentrated. To ensure each reporter cell had a single shRNA integrated, I infected FUCCI* K562 cells in bulk with enough lentivirus so that 50% of cells had virus integration, as verified by a BFP positive signal. On the 6th day post infection, cells were sorted by FACS into two pools: early G1 (mKO2+ mAG1-) and S/G2/M (mKO- mAG1+. After genomic extraction and sequencing of the barcoded shRNAs, sequences were counted for an enrichment or depletion of bar-coded shRNAs in the two pools. Analysis was performed using the ‗analyze_primary_screen.py‘ script (Kampmann et al. 2014). The Mann-Whitney U (MW) statistical test was used to derive a P value for the genes targeted in the screen because it has previously been shown to be able to deal with noisy datasets and has been used successfully in other pooled shRNA screens (Kampmann et al. 2013). The Mann-Whitney U test is a rank sum test that is based on the null hypothesis that all values come from the same population (Mann and Whitney 1947; Fay and Proschan 2010). Pooled analysis of the individual genes that were over- represented in the G1 group compared to the S/G2/M group with a P-value less than 0.05 (Fig. B5) identified multiple genes known to regulate cell cycle progression (i.e. BTRC, MAD2L1BP, SKP2) among other hits with poorly or unannotated functions (Busino et al. 2003; Frescas and Pagano 2008; Date et al. 2013). Similarly, when looking at genes that were over-represented in the S/G2/M group compared to the G1 group with a P-value less than 0.05 (Fig. B6), the screen identified well-characterized cell cycle regulators (i.e., FBXW7, CDC26, LMO7) (Nakayama and Nakayama 2005; Wang et al. 2009; Tzeng et al. 2018).

Figure B5. Genes targeted by shRNAs that were enriched in G1 compared to S/G2/M in the cell cycle screen. Genes targeted with a p-value <0.05 by Mann-Whitney U test are colored in red with gene names listed.

Figure B6. Genes targeted by shRNAs that were enriched in S/G2/M compared to G1 in the cell cycle screen. Genes targeted with a p-value <0.05 by Mann-Whitney U test are colored in green with gene names listed.

82 Discussion This screen performed in K562-FUCCI* cells identified multiple genes with well characterized roles in regulating cell cycle progression among the top hits. The ability of this screen to successfully pull out these hits verified the underlying rationale of this approach and thus provides high confidence that many of the other genes identified have a previously uncharacterized role in regulating the cell cycle program. Many of the top hits have poorly understood cellular roles, but altered expression is implicated in some cancers. FBXW8 was one of the top hits in the screen that, when knocked down, caused cells to increase in the G1 pool compared to the S/G2/M pool. Fbxw8 is a substrate adapter and assembles a ubiquitin ligase complex with Skp1, Cul7, and Rbx1 (Wang et al. 2014a). While the specific substrates of Fbxw8 are only beginning to emerge, there is already a wealth of data linking Fbxw8 and Cul7 to cell proliferation and tumorigenesis (Okabe et al. 2006; Kim et al. 2007; Wang et al. 2014b). Another validation of my screening strategy.

The identification of multiple known cell cycle regulators arising from my screen provides reassurance that my method was an effective a tool to identify candidate genes that are involved in cell cycle progression. In one case, however, the significance of a particular gene in the G1 vs the S/G2/M pool seems confounding. MAD2L1BP, the gene encoding the MAD2L1-binding protein (also known as p31 or Comet) (Xia et al. 2004), is thought to positively regulate mitotic metaphase and the anaphase transition and thus be important for progression through mitosis (Yang et al. 2007; Jia et al. 2011). Indeed, in prior studies, knockdown of MAD2L1BP by RNAi led to activation of the spindle assembly checkpoint and arrest of cells in mitosis (Habu et al. 2002; Hagan et al. 2011) Therefore, it was surprising that, in my screen, the shRNA for MAD2L1BP was over-represented in the pool of G1-arrested cells.

From the list of candidate genes identified as hits from my screen, the next step would be to focus on specific candidates and attempt to validate them individually. This corroboration should be performed by first cloning the shRNAs that showed a significant deviation from the non- targeting control shRNAs, assessing knock-down efficiency by qRT-PCR, and visualizing the FUCCI* cell cycle profile by flow cytometry. To determine the specific cell cycle stage that is affected by gene knock-down, live-cell fluorescence microscopy would be useful to determine if a particular gene knock-down causes an extended delay in any particular stage of the cell cycle. To complement this validation procedure, it would be additionally useful to use siRNA to knock- down the candidate genes, using sequences distinct from those used in the initial screening library.