Characterization of Pharyngeal Gland Development and Function in Caenorhabditis

UNIVERSITY OF CALGARY

Characterization of pharyngeal gland development and function in Caenorhabditis

elegans

Ryan B. Smit

A THESIS

SUBMITTED TO THE FACULTY OF GRADUATE STUDIES

IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE

DEGREE OF DOCTOR OF PHILOSOPHY

DEPARTMENT OF BIOCHEMISTRY AND MOLECULAR BIOLOGY

CALGARY, ALBERTA

SEPTEMBER 2010

Library and Archives Bibliothèque et Canada Archives Canada

Published Heritage Direction du Branch Patrimoine de l’édition

395 Wellington Street 395, rue Wellington Ottawa ON K1A 0N4 Ottawa ON K1A 0N4 Canada Canada

Your file Votre référence ISBN: 978-0-494-69500-5 Our file Notre référence ISBN: 978-0-494-69500-5

NOTICE: AVIS:

The author has granted a non- L’auteur a accordé une licence non exclusive exclusive license allowing Library and permettant à la Bibliothèque et Archives Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par télécommunication ou par l’Internet, prêter, telecommunication or on the Internet, distribuer et vendre des thèses partout dans le loan, distribute and sell theses monde, à des fins commerciales ou autres, sur worldwide, for commercial or non- support microforme, papier, électronique et/ou commercial purposes, in microform, autres formats. paper, electronic and/or any other formats. . The author retains copyright L’auteur conserve la propriété du droit d’auteur ownership and moral rights in this et des droits moraux qui protège cette thèse. Ni thesis. Neither the thesis nor la thèse ni des extraits substantiels de celle-ci substantial extracts from it may be ne doivent être imprimés ou autrement printed or otherwise reproduced reproduits sans son autorisation. without the author’s permission.

In compliance with the Canadian Conformément à la loi canadienne sur la Privacy Act some supporting forms protection de la vie privée, quelques may have been removed from this formulaires secondaires ont été enlevés de thesis. cette thèse.

While these forms may be included Bien que ces formulaires aient inclus dans in the document page count, their la pagination, il n’y aura aucun contenu removal does not represent any loss manquant. of content from the thesis.

Abstract

The present thesis is an investigation into the specification, development and function of the pharyngeal gland cells in C. elegans. Although the role of PHA-4 in the specification of pharyngeal identity has been well established (Gaudet and Mango, 2002; Horner et al.,

1998; Kalb et al., 1998; Mango et al., 1994a), little is known about how the cell types within the pharynx are specified. I therefore sought to elucidate the mechanisms responsible for specification of the gland cells. I have thus explored three novel aspects of gland development.

The first is the characterization of the only known gland-specific transcription factor. Through bioinformatic analysis, I was able to identify a biologically active cis- regulatory element (called PGM1 for pharyngeal gland motif 1) in the promoters of fourteen gland specific genes. I found that this site is likely a site for the binding of at least two transcription factors and one of those is the gland specific basic-Helix-Loop-

Helix transcription factor HLH-6. All PGM1 dependent genes are also HLH-6 dependent, but at least two genes are partially independent of HLH-6 and gland cells are still present in hlh-6 mutants, suggesting HLH-6 does not specify gland fate.

Second, I not only characterize the role of HLH-6 in activating gland gene expression, I also investigate its role in the function of the glands. hlh-6 mutants occasionally arrest as larvae, are slow growing and have small brood sizes suggestive of a defect in the ability of the animals to feed properly. hlh-6 mutants are rescued by feeding on bacteria that are less sticky, suggesting the glands may play a role in lubricating the pharyngeal lumen. In support of this model I show that an HLH-6 dependent gene product is normally found lining the pharyngeal lumen.

Lastly, I identify gland genes whose expression does not require HLH-6. Analysis of two such genes identified a promoter element originally described in the hlh-6 promoter (HRL3 for hlh-6 regulatory element 3) as necessary for HLH-6-independent gland expression. From these studies I propose a model of gland development and compared it to known regulatory networks in the specification of other C. elegans cell types.

iii

Acknowledgements

Some of the plasmids used in this dissertation were created by Dr. Jeb Gaudet, Indra

Raharjo, Leanne Sayles and Shoubin Wen (see Appendix B for details). The initial runs of the Improbizer program were completed by Dr. Jeb Gaudet.

Table of Contents

Abstract ...... ii Acknowledgements ...... iv Table of Contents ...... v List of Tables ...... vii List of Figures and Illustrations ...... viii List of Symbols, Abbreviations and Nomenclature ...... x

CHAPTER ONE: INTRODUCTION ...... 1 1.1 Transcriptional Control of Cell Type Specification ...... 1 1.2 The Pharynx of C. elegans ...... 3 1.2.1 The Pharyngeal Glands ...... 8 1.3 Pharyngeal Specification ...... 12 1.3.1 Maternal Contributions ...... 12 1.3.2 Zygotic Control of Specification ...... 16 1.3.3 The Role of PHA-4 ...... 18 1.3.4 Cell Type Specification ...... 19 1.4 Experimental Objectives ...... 21

CHAPTER TWO: MATERIALS AND METHODS ...... 23 2.1 Nematode Handling Conditions ...... 23 2.2 Construction of Plasmids and Reporters ...... 24 2.3 Strain Constructions ...... 26 2.4 Motif Searches using Improbizer ...... 28 2.5 Electrophoretic Mobility Shift Assay ...... 28 2.6 Ectopic Expression of HLH-6 ...... 30 2.7 Lifespan Assays ...... 30 2.8 Growth Assays ...... 31 2.9 Pharyngeal Cell Counts in WT vs. HLH-6 ...... 32

CHAPTER THREE: IDENTIFICATION OF THE GLAND SPECIFIC CIS- REGULATORY ELEMENT PGM1 AND THE TRANS-ACTING FACTOR HLH-6 ...... 33 3.1 Identification of Gland Specific Genes ...... 33 3.2 Identification of Pharyngeal Gland Motif 1 ...... 38 3.3 PGM1 is Necessary for Gland Expression ...... 42 3.4 PGM1 is Sufficient for Gland Expression ...... 52 3.5 PGM1 Contains Two Important Regulatory Sequences ...... 52 3.6 HLH-6 is a Gland Specific Transcription Factor ...... 60 3.7 HLH-6 Acts Through PGM1 ...... 61 3.8 Summary ...... 76

CHAPTER FOUR: CHARACTERIZATION OF THE HLH-6 MUTANT PHENOTYPE ...... 79 4.1 Characterization of a Lethality Closely Linked to hlh-6 ...... 79 v

4.2 The g2 Gland Cells are Missing in hlh-6 Mutants ...... 85 4.3 hlh-6 Mutants are Feeding Defective ...... 93 4.4 Ablation of the Gland Cells Phenocopies the hlh-6 Mutant ...... 101 4.5 A PHAT Protein Secreted by the Glands Binds to the Pharyngeal Cuticle ...... 101 4.6 Summary ...... 108

CHAPTER FIVE: IDENTIFICATION OF HLH-6-INDEPENDENT CIS- REGULATORY ELEMENTS...... 110 5.1 Identification of Gland Expressed Genes Without PGM1 ...... 111 5.2 HLH-6 Dependent Genes Without PGM1 ...... 118 5.3 Regulation of nas-12 ...... 124 5.4 Regulation of Y8A9A.2 ...... 134 5.5 Why do These Multiple Pathways Exist? ...... 143 5.6 Summary ...... 147

CHAPTER SIX: DISCUSSION ...... 149 6.1 Thesis Summary and Significance ...... 149 6.2 Role of HLH-6 in Gland Development ...... 151 6.3 Role of PHA-4 in Gland Development ...... 153 6.4 Gland Function ...... 155 6.5 Evolutionary Conservation ...... 157 6.6 Future Directions ...... 159

CHAPTER SEVEN: REFERENCES ...... 160

APPENDIX A: LIST OF OLIGONUCLEOTIDES ...... 176

APPENDIX B: LIST OF PLASMIDS ...... 185

List of Tables

Table 3.1 Gland-specific genes identified by microarray ...... 34

Table 3.2 List of gland and pharyngeal (non-gland) genes and their associated Motif Matcher score ...... 39

Table 5.1 Gland expressed genes ...... 112

vii

List of Figures and Illustrations

Figure 1.1 The pharynx of C. elegans...... 5

Figure 1.2 The pharyngeal glands ...... 9

Figure 1.3 The early blastomeres and maternal contributions to pharyngeal fate ...... 13

Figure 3.1 PGM1 is required for expression of some pharyngeal gland genes ...... 36

Figure 3.2 Occurrences of PGM1 are conserved among nematodes ...... 43

Figure 3.3 The region between -212 and -102 bp is required for phat-3 expression ...... 45

Figure 3.4 The PGM1 at -69 bp is required for phat-1 expression ...... 48

Figure 3.5 The PGM1 at -581 bp is required for lys-8 expression ...... 50

Figure 3.6 The extended PGM1 is sufficient for gland expression ...... 53

Figure 3.7 Alignment of functional PGM1 sites ...... 55

Figure 3.8 The E-box and YMAAY components of PGM1 are necessary for phat-1 expression ...... 58

Figure 3.9 hlh-6 is expressed in pharyngeal glands ...... 62

Figure 3.10 hlh-6 is required for PGM1 activity ...... 65

Figure 3.11 hlh-6 is required for expression of some gland genes ...... 67

Figure 3.12 HLH-2 can bind directly to PGM1 in vitro ...... 71

Figure 3.13 Ectopic HLH-6 (+HLH-2) is not sufficient to activate ectopic gland reporter expression ...... 73

Figure 3.14 Model of gland gene regulation by HLH-6 and PHA-4 ...... 77

Figure 4.1 let-x genetically lies between 0.95 and 1.29 on Linkage Group II ...... 81

Figure 4.2 The lethality of let-x is due to the presence of live bacteria ...... 83

Figure 4.3 The g2 glands in hlh-6 mutants do not express any pharyngeal markers ...... 88

Figure 4.4 The g2 glands are not generated in hlh-6 mutants ...... 91

Figure 4.5 hlh-6 mutants are slow growing ...... 94

viii

Figure 4.6 Nile red staining is increased in hlh-6 mutants ...... 98

Figure 4.7 C. elegans PHAT-1 is similar to T. canis MUC-5 ...... 103

Figure 4.8 PHAT-5::mCherry localizes to the pharyngeal lumen ...... 105

Figure 5.1 Some gland expressed genes without PGM1 are dependent on hlh-6 ...... 116

Figure 5.2 Expression of gland reporters in the hlh-6 mutant ...... 119

Figure 5.3 E-boxes are required for the gland expression of T05B4.13 ...... 122

Figure 5.4 A HRL3-like site is necessary and sufficient for the gland expression of nas-12 ...... 126

Figure 5.5 The HRL3-like site in the nas-12 promoter is conserved among nematodes 128

Figure 5.6 Alignment of the HRL3-like sequences ...... 131

Figure 5.7 Y8A9A.2 expression changes during larval development ...... 135

Figure 5.8 A HRL3-like site is necessary for Y8A9A.2 expression ...... 137

Figure 5.9 The HRL3-like site in the Y8A9A.2 promoter is conserved among nematodes ...... 140

Figure 5.10 Model of gland development ...... 144

List of Symbols, Abbreviations and Nomenclature

Symbol Definition ATPase Adenosine Triphosphatase Ash Acheate Scute Homolog bHLH Basic Helix Loop Helix BMP Bone Morphogenetic Protein Brn Brain Specific Homeobox bZIP Basic Leucine Zipper Cbre Caenorhabditis brenneri Cbri Caenorhabditis briggsae Cele Caenorhabditis elegans Cjap Caenorhabditis japonica Crem Caenorhabditis remanei DBD DNA Binding Domain DNA Deoxyribonucleic Acid EDTA Ethylenediaminetetraacetic Acid EGF Epidermal Growth Factor EMSA Electrophoretic Mobility Shift Assay Fox Forkhead Box GFP Green Fluorescent Protein GST Glutathione S-transferase HNF Hepatocyte Nuclear Factor HRL3 hlh-6 regulatory element 3 IPTG Isopropyl β-D-1-thiogalactopyranoside Kan Kanamycin Lef Lymphoid enhancer factor LG Linkage Group Mash Murine Acheate Scute Homolog MAPK Mitogen Activated Protein Kinase mins Minutes MyoD Myogenic Differentiation NDIC Nomarski Differential Interference Contrast NEXTDB Nematode Expression Pattern Database NGM Nematode Growth Medium Nkx NK Homeobox Oct Octamer Binding PAGE Polyacrylamide Gel Electrophoresis PBS Phosphate Buffered Saline PBS PHA-4 binding Site PCR Polymerase Chain Reaction PDX Pancreas Duodenum Homeobox PGM1 Pharyngeal Gland Motif 1 POU Pit Oct Unc PTF1 Pancreas Transcription Factor 1

RBP-J Recombining Binding Protein-J RT-PCR Reverse Transcriptase Polymerase Chain Reaction SDS Sodium Dodecyl Sulfate Ser Serine ShK Stichodactyla Toxin Su(H) Suppressor of Hairless SXC Six Cystein Domain TAL Activated in T Cell Acute Lymphocytic Leukemia Tcf T Cell Factor Thr Threonine WT Wild Type YFP Yellow Fluorescent Protein

xi 1

Chapter One: INTRODUCTION

1.1 Transcriptional Control of Cell Type Specification

An important question in the study of organ development is how different cells are instructed to become part of a common structure and yet are also specified to have a distinct identity within that structure. Recent studies using C. elegans have begun to elucidate some of the different ways cells become specified by examining how terminal differentiation genes are regulated (see Gaudet and McGhee, 2010 for a review). From these studies a few models have been proposed for how cell type specification occurs.

One model is the “terminal selector” model proposed for controlling specification of neuronal sub-types (Hobert, 2008). In this model each cell type has one factor or a complex of factors that activates all of the genes expressed in that cell, thus specifying the identity of that cell. For example, all genes expressed in the AIY interneuron are activated by the combination of the two homeodomain transcription factors TTX-3 and

CEH-10 (Wenick and Hobert, 2004) and all genes expressed in the ASE sensory neuron are activated by the zinc finger transcription factor CHE-1 (Uchida et al., 2003).

Specification of the C. elegans intestine also seems to fit the terminal selector model. One transcription factor, zinc finger transcription factor ELT-2, activates all genes expressed in the intestine (McGhee et al., 2009; McGhee et al., 2007). However, in contrast to expression of neuronal genes, there are additional transcription factors that act with ELT-

2 to modulate the expression of intestinal genes. For example, expression of yolk genes is activated by ELT-2 but repressed by MAB-3 in males (MacMorris et al., 1992; Yi and

Zarkower, 1999). Thus ELT-2 is not a terminal selector gene because it requires the

2 function of other transcription factors in order to specify the intestine. Lastly, the excretory cell in C. elegans seems to have multiple transcription factors regulating multiple terminal differentiation gene batteries. CEH-6 (a POU homeodomain transcription factor), DCP-66 (a zinc finger transcription factor) and NHR-31 (a nuclear receptor) each activate distinct gene batteries within the excretory cell (Hahn-Windgassen and Van Gilst, 2009; Mah et al., 2007; Zhao et al., 2005). For example, NHR-31 seems to only activate expression of vacuolar ATPase subunits raising the possibility that each of these transcription factors activate terminal differentiation gene batteries with distinct functions in the excretory cell.

An interesting observation from these three models and many other cases is that many transcription factors seem to work in complexes: CEH-10 and TTX-3 bind to an

AIY specific motif cooperatively and both transcription factors are required to induce ectopic expression of AIY genes (Wenick and Hobert, 2004). ELT-2 synergizes with the

Notch transcriptional effector LAG-1 to activate the endodermal expression of the basic-

Helix-Loop-Helix transcription factor ref-1 (Neves et al., 2007). The Ex-1 motif, which is bound by DCP-66 in the excretory cell, alone is not sufficient to activate ectopic expression but requires additional motifs (and presumably additional transcription factors, Zhao et al., 2005). It is important to note that in all three of these examples, the transcription factor complexes act through motifs that contain multiple transcription factor binding sites, suggesting that many transcription factors require complexes of multiple factors to function.

In each of these three examples, the cells studied are either isolated or part of a relatively uniform organ. Therefore I wanted to investigate how cell fates are specified

3 within a more “complex” organ. For this reason I chose to examine cell type specification in the pharynx of C. elegans, which contains multiple cell types derived from multiple lineages.

1.2 The Pharynx of C. elegans

I chose to use the C. elegans pharynx (foregut) as a model for understanding organ development and the cell type specification that occurs within the organ. In comparison to organs of other higher metazoans, the pharynx of C. elegans is relatively simple being comprised of only 80 cells. Yet, in contrast to the C. elegans intestine which only has one cell type, there are five different cell types in the pharynx allowing the investigation of cell type specification within the organ. Other groups argue that the

“pharynx” includes the arcade and valve cells found at the anterior and posterior ends of the organ respectively (Kiefer et al., 2007) because they express the pharyngeal organ identity factor PHA-4 (see Chapter 1.3.3). Thus by this definition the pharynx has 95 cells and seven cell types.

The entire cell lineage of C. elegans is known (Sulston et al., 1983) and with the availability of many molecular tools, such as Green Fluorescent Protein (GFP) markers, individual cells can be visualized throughout development. These tools together with the well established genetics of C. elegans (genetic screens, ease of RNAi, availability of gene knock-out strains and the sequenced genome) make the pharynx an excellent model system for the study of organogenesis. As well, of the few genes implicated in development of the C. elegans pharynx, most have homologs that are also involved in the development of homologous organs in mammalian systems (see below for examples).

The pharynx is a bilobed linear tube organized into three sections: the corpus, isthmus and terminal bulb (Figure 1.1A). Muscular contractions of the corpus result in widening of the pharyngeal lumen resulting in an inflow of bacteria (food) suspended in liquid. Bacteria are concentrated at the posterior end of the corpus and excess water is expelled during corpus relaxation (Fang-Yen et al., 2009). Waves of peristalsis transport bacteria through the isthmus and into the terminal bulb. The entire lumen of the pharynx is lined with cuticle that is distinct from but continuous with the epidermal cuticle. In the posterior bulb, the lumen forms a specialized structure called the grinder, a set of chitinous projections that physically digest the bacteria. The remains are then passed into the intestine where chemical digestion occurs. To properly perform these functions of pumping, filtering and digesting, the pharynx requires the concerted actions of multiple cell types.

The pharynx contains five (or seven) different cell types (muscles, epithelia, neurons, marginal cells, glands) that are not restricted by their lineal origins (Figure 1.1B,

C, Albertson and Thomson, 1976). The epithelial and marginal cells are thought to mainly provide support and structure to the organ. Nine epithelial cells (three groups of e1-3) support the opening of the pharynx where it connects to the buccal cavity (along with the arcade cells), while the nine marginal cells (three cells each of mc1-3) support the entire length. The marginal cells (as well as the muscle cells) are organized radially with three-fold symmetry around the pharyngeal lumen (Figure 1.1D).

The 37 muscle nuclei of the pharynx are responsible for the rhythmic contractions that transport food through the lumen. There are eight sections of muscles (pm1-8) that are positioned as rings along the longitudinal axis. pm1, 3, 4 and 5 cells fuse together

Figure 1.1 The pharynx of C. elegans

(A) NDIC photomicrograph of the pharynx highlighting the three sections. (B) The lineage of C. elegans (adapted from Sulston, et al., 1983) highlighting the pharyngeal cells in black. (C) Diagrams of the five pharyngeal cell types (adapted from Albertson and Thomson, 1976). (D) A diagram of a representative cross section through the pharynx highlighting the muscles in green and marginal cells in blue.

7 during development, thus becoming syncytial. The pm1 and pm2 cells are thin sheets but the rest of the muscle groups are wedge shaped and are arranged radially with three-fold symmetry. The pharyngeal muscles have characteristics of epithelia: they have an apical surface that faces the lumen and adherens junctions between the apical and basal regions of the cells. Muscle fibres within muscle groups 3-5 are organized radially so that contractions result in widening of the lumen. Interestingly, the pm6 cells have regions without muscle fibres and instead have many large filled vesicles. Since there are no hypodermal cells to secrete the pharyngeal cuticle (as there are for the epidermal cuticle) it is thought that the pm6 cells secrete the pharyngeal cuticle (including components of the grinder, Albertson and Thomson, 1976). Muscle fibres in pm7 are oriented both radially and longitudinally which is proposed to provide the unique motions of the grinder, which is attached to pm7. The single pm8 cell has radially oriented muscle fibres but does not receive any innervation by the pharyngeal neurons. The rest of the muscle cells are all innervated by pharyngeal neurons.

The pharyngeal nervous system is the key regulator of the muscular contractions.

There are twenty neurons: seven motor neurons, eight interneurons, two marginal cell neurons, one motor-interneuron and two neurosecretory-motor neurons. Intriguingly, all of the pharyngeal neurons make synapses en passant and do not have the typical synaptic terminals of vertebrate neurons. Only one set of neurons seems responsible for controlling the rate of pharyngeal pumping, the MC neurons (Avery and Horvitz, 1989), yet none of the neurons are critically dependent for pumping to occur. Ablation of M4 does result in the loss of isthmus peristalsis but ablation of all pharyngeal neurons does not stop pharyngeal pumping (Avery and Horvitz, 1989). However, animals with all

8 pharyngeal neurons ablated were slow growing and sick, suggesting that neurons have subtle effects on pumping. One explanation is that the pharyngeal neurons control the rate of pumping in response to environmental cues and that the pharyngeal muscles have the intrinsic ability to pump without neural input.

1.2.1 The Pharyngeal Glands

The pharyngeal glands are five cells with their bodies in the posterior bulb of the pharynx. Each gland has a cellular projection that extends anteriorly where it connects to the pharyngeal lumen (Figure 1.2, Albertson and Thomson, 1976). The glands are further divided into two sub-groups, g1 and g2: the three g1 glands have a lamellar cytoplasm while the two g2 cells have a clear cytoplasm (Albertson and Thomson, 1976), though the significance of these sub-types is not known. The g1P cell body lies the most posteriorly in the terminal bulb and has an extension that opens just behind the buccal cavity. The two g1A cell (g1AL and g1AR, for left and right respectively) bodies lie in the ventral half of the most anterior portion of the terminal bulb and extend their projections to the back of the corpus. The two g2 cells (g2L and g2R for left and right respectively) have cell bodies in the ventral posterior half of the terminal bulb (just below the g1P cell body) and open into the pharyngeal lumen just anterior to the grinder. Interestingly, these morphologies of the three g1 glands seem to be present in many other nematodes, including many parasitic ones (Chitwood and Chitwood, 1974).

Based on the morphologies of the glands in electron micrograph sections, it is not possible to deduce the function of these cells, besides their ability to secrete. However, observation of the glands during the development of the animal does suggest one possible

Figure 1.2 The pharyngeal glands

Diagram of the pharynx, highlighting the pharyngeal glands, modified from Albertson and Thomson, 1976.

11 function. Singh and Sulston (1978) noticed that the glands seemed to secrete large vesicles just prior to each molt. C. elegans undergoes four molts where both the epidermal and pharyngeal cuticles are shed. This observation suggests a few possibilities: first, the glands could be signalling or preparing for each molt. Second, the glands might be responsible for breakdown of the old cuticle during each molt. Lastly, the glands might either be secreting the new cuticle or some product(s) that attach to the cuticle.

However, little is known about the composition of the pharyngeal cuticle or how it is secreted or shed, leaving the role of the glands in this process unclear.

A second potential function for the pharyngeal glands is aiding in digestion or some other aspect of the feeding process. This function is suggested based on the fact that secretory cells in the foreguts of other animals often secrete digestive enzymes. This possibility is also supported by the observation that some of the glands are innervated by pharyngeal neurons (Albertson and Thomson, 1976). Since the pharyngeal neurons modulate the mechanical digestive properties of the pharynx it is possible they might also affect the potential chemical digestive properties of the glands. Indeed, at least one potential digestive enzyme is known to be expressed in the gland cells: lys-8, a lysozyme encoding gene (Mallo et al., 2002). However, the exact function of this gene has yet to be demonstrated. Another known gland expressed gene, kel-1 (Drosophila kelch related), when lost results in larval lethality suggestive of a feeding-related function for the glands

(Ohmachi et al., 1999). However, by in situ hybridization kel-1 is expressed broadly during embryogenesis (Kohara, 2001a; Kohara, 2001b) and therefore it could be functioning in other cells.

The third potential function of the glands comes from the mutant phenotype of another gland expressed gene, srf-3. The srf-3 (surface antigenicity abnormal) gene encodes a nucleotide sugar transporter that is expressed in the glands and epidermal cells. srf-3 mutants are resistant to infections by multiple bacterial species (Hoflich et al., 2004) suggesting that the glands might secrete some product that is involved in bacterial infection. The evidence for this possible function (like the previous two potential gland functions) is largely circumstantial, since srf-3 is also expressed in the hypodermis. srf-3 function will remain unclear until evidence links gene function with a gland phenotype.

The question I intend to address in this thesis is how these five different cells are each specified and come together as part of an organ.

1.3 Pharyngeal Specification

Much work has been done over the past thirty years that has elucidated how the pharynx is specified, providing a framework to begin to understand how the cell types within the pharynx are specified. It started with the publishing of the embryonic lineage of C. elegans (Sulston et al., 1983). This work allowed many groups to analyze the disrupted lineages in embryos that were either experimentally or genetically manipulated.

How the pharynx becomes specified ultimately goes back to the single cell embryo and the first few divisions that result in the specific localization of key maternal factors.

1.3.1 Maternal Contributions

During embryogenesis the pharyngeal precursors arise from two of the six founder cells, ABa and MS (Figure 1.3). The ABa derived pharyngeal cells are specified

Figure 1.3 The early blastomeres and maternal contributions to pharyngeal fate

The early C. elegans embryo is diagrammed from the one-cell stage to the four-cell stage.

At the four-cell stage, signalling from the P2 cell to the ABp cell results in the inhibition of tbx-37/38. In the ABa cell, tbx-37/38 are activated by an unknown mechanism and in combination with Notch signalling from the MS cell at the twelve-cell stage (not shown) activate PHA-4 in the ABa lineage either directly or indirectly. Signalling from the P2 cell to the EMS cell results in a low POP-1/SYS-1 ratio in E and a high POP-1/SYS-1 ratio in MS. These ratios activate and repress end-1/3 expression (and thus intestinal fate) respectively. SKN-1 in the MS cell activates med-1/2 expression, which in turn results in tbx-35 and ceh-51 activation (not shown). TBX-35 and CEH-51 act together and activate

PHA-4 expression in the MS lineage either directly or indirectly. See text for full details.

15 by two rounds of Notch signalling involving the Notch receptor GLP-1 (Hutter and

Schnabel, 1994). The very first cell division produces two daughter cells, AB and P1.

Maternally supplied glp-1 mRNA is repressed in the P1 cell but only translated in the AB cell (Evans and Hunter, 2005). The Notch ligand apx-1 is selectively translated in the P2 cell (a daughter of the P1 cell, Mickey et al., 1996), which only contacts the posterior daughter of the AB cell (ABp, Figure 1.3). Notch signalling from the P2 cell to the ABp cell (Hutter and Schnabel, 1994; Mango et al., 1994b; Mickey et al., 1996) activates the ref-1 family of basic-Helix-Loop-Helix (bHLH) transcriptional repressors in ABp (Good et al., 2004; Neves and Priess, 2005). These repressors inhibit the expression of the closely related T-box homeodomain transcription factors tbx-37 and tbx-38 (two redundant activators of ABa-derived pharyngeal fate; (Good et al., 2004) in the ABp lineage. How tbx-37 and tbx-38 are activated in the ABa lineage is not known, but these two genes are necessary but not sufficient for the pharyngeal fate in these cells (Good et al., 2004). A second Notch signalling event is required to induce the ABa pharyngeal fate. This signal involves an unknown Notch ligand from the MS blastomere that comes into contact with two ABa granddaughter cells and requires glp-1 expression in these

ABa derived cells (Hutter and Schnabel, 1994; Moskowitz et al., 1994). This signal, in combination with tbx-37/tbx-38, activates pharyngeal fate in the ABa cells presumably by activating the organ identity factor PHA-4 (see below), either directly or indirectly.

SKN-1, a basic leucine zipper (bZIP) family transcription factor specifies both intestinal (E cell) and pharyngeal (MS cell) fate (Bowerman et al., 1992). SKN-1 localization depends on numerous factors including mex-1 and par-1 (Bowerman et al.,

1993; Mello et al., 1992), which results in SKN-1 protein only in EMS and its sister P2.

SKN-1 is specifically active only in the EMS because its activity in P2 is inhibited by

PIE-1, which globally represses transcription in the P2 cell (Seydoux et al., 1996), including the transcription of SKN-1 targets. It is also known how the MS and E cells take on completely different fates despite having equal amounts of SKN-1. Non- canonical Wnt and MAP kinase signalling from the P2 cell to the E cell results in low levels of POP-1 (a Tcf/Lef-1-like transcription factor) and high levels of SYS-1 (a β-

Catenin), which activates end-1 and end-3 in combination with SKN-1 (Broitman-

Maduro et al., 2005; Huang et al., 2007; Shetty et al., 2005). end-1 and end-3 are a pair of redundant GATA family transcription factors that specify the intestinal fate (Zhu et al.,

1997). Conversely, in the MS cell, high levels of POP-1 and low levels of SYS-1 inhibit the expression of end-1/3 (Huang et al., 2007; Maduro et al., 2002). Although both MS and E have equal levels of SKN-1, end-1/3 are only activated in E and thus only E descendents become intestinal. SKN-1 also activates a pair of redundant GATA transcription factors med-1 and med-2 in both E and MS cells. In the E lineage, the role of MED-1/2 has been debated (Captan et al., 2007; Goszczynski and McGhee, 2005;

Maduro et al., 2007; Maduro et al., 2001), but it may help to activate expression of end-

1/3. In the MS lineage, the POP-1/SYS-1 ratio is high and effectively blocks MED-1/2 from activating end-1/3 in this lineage. The transcriptional activation of med-1 and med-2 mark the first zygotic events in MS pharyngeal specification.

1.3.2 Zygotic Control of Specification

The switch from maternal to zygotic control of pharyngeal specification occurs around the 28 cell stage. The first zygotic genes in the specification of the pharynx

17 include the activation of tbx-37/38, med-1/2 and the inhibition of end-1/3. In the ABa lineage, TBX-37/38 also require Notch signalling from MS in order to correctly specify the ABa pharyngeal fate, but how these factors do so (i.e. which genes they activate) remains unknown. In the MS lineage POP-1’s inhibition of end-1/3 prevents the MS cell from becoming intestinal. This regulation, in addition to the activation of med-1/2 by

SKN-1 instructs the MS cell to become pharyngeal. One important function of MED-1/2 is to activate the T-box transcription factor tbx-35 which also specifies MS fate (both

MS-pharyngeal and non-pharyngeal fates, Broitman-Maduro et al., 2006). TBX-35 itself activates another transcription factor, the homeodomain transcription factor ceh-51

(Broitman-Maduro et al., 2009), which acts together with TBX-35 to specify MS.

Intriguing is the common role for T-box transcription factors in activating the pharyngeal fate of AB (by tbx-37/38) and MS (by tbx-35) lineages. In fact, the role for T-box transcription factors in the specification of mesodermal tissues is common throughout evolution. The DNA binding domains of tbx-37/38 are most similar to murine Tbx6, which is involved in somatic mesoderm development (Chapman and Papaioannou, 1998), and T-box transcription factors are involved in heart development (also a mesodermal structure) in Drosophila, Xenopus and mouse (Horb and Thomsen, 1999; Lyons et al.,

1995; Miskolczi-McCallum et al., 2005). How both TBX-37/38 and TBX-35/CEH-51 are both capable of specifying pharyngeal fate is unclear. One possibility is that all of these factors converge on one gene that activates all of the genes required for formation of the pharynx, such as pha-4.

1.3.3 The Role of PHA-4

Although it is not known how, all of the maternal and zygotic factors must converge on the forkhead family transcription factor PHA-4. pha-4 was initially identified because mutant larvae completely lack pharyngeal tissues (Mango et al.,

1994a). No other genes yet identified have a complete and specific loss of pharynx (pop-

1 and glp-1 mutants only lose MS and ABa derived pharynx respectively, skn-1 mutants completely lack a pharynx but also have other EMS defects). pha-4 lies genetically downstream from both the ABa and MS maternal specification pathways (Mango et al.,

1994a) suggesting that it is the point of convergence for pharyngeal specification. pha-4 encodes a transcription factor related to mammalian FoxA/HNF-3 (Horner et al., 1998;

Kalb et al., 1998), which is also expressed very early in foregut development and results in severe foregut defects in mutants (Ang and Rossant, 1994; Ang et al., 1993). PHA-4 is also very similar to Drosophila forkhead which again is expressed very early and is involved in foregut development (Weigel et al., 1989a; Weigel et al., 1989b). PHA-4 was proposed to be an organ identity gene based on four criteria: (1) pha-4 mutants completely lack a pharynx. (2) pha-4 lies genetically downstream of the maternal specification pathways. (3) PHA-4 is expressed in all pharyngeal cells from the time they are specified and (4) ectopic expression of PHA-4 activates ectopic pharyngeal genes.

Interestingly, PHA-4 was able to bind a promoter element of the pharyngeal muscle specific myosin myo-2 (Kalb et al., 1998) that matched the predicted binding site consensus for FoxA/HNF-3 (Overdier et al., 1994). This binding consensus was found to be enriched in pharyngeal expressed genes identified by microarray (Gaudet and Mango,

2002) and Chromatin Immunoprecipitation (Zhong et al., 2010) and was found to be

19 required for expression in every pharyngeal promoter examined (Gaudet and Mango,

2002). Thus a model was proposed where PHA-4 activates all (or most) pharyngeal genes. However, PHA-4 alone cannot be responsible for all aspects of organ development and must function with other factors to control the various sub-programs of pharyngeal organogenesis, such as specification of the distinct cell types. Aside from the involvement of PHA-4, little is known about the specification and development of any of the distinct pharyngeal cell types, though regulators of pharyngeal muscle development have been identified.

1.3.4 Cell Type Specification

Although the exact mechanism controlling specification of the pharyngeal muscles is not known, there are many factors involved in muscle development that are known (eg. ceh-22, tbx-2 and pha-2). All pharyngeal muscle cells express the myosin heavy chains myo-1 and myo-2 (Okkema et al., 1993). Specific promoter elements of myo-2 lead to the discovery of ceh-22, an Nkx homeobox transcription factor that binds to and activates myo-2 (Okkema and Fire, 1994; Okkema et al., 1997). However ceh-22 is not expressed in all muscle cells, mutants in ceh-22 do not lose myo-2 expression nor are any pharyngeal muscles absent (Okkema et al., 1997) thus ceh-22 is not necessary for the muscle fate. Interestingly an ortholog of ceh-22 in vertebrates is Nkx2.5, which is involved in cardiac muscle differentiation (Lints et al., 1993) and can functionally replace ceh-22 (Haun et al., 1998). pha-2 is a homeodomain transcription factor that is expressed only in a subset of pharyngeal muscles (particularly pm5) whose mutant only affects specific muscle cells (the pm5 cells) resulting in a thickened isthmus (Morck et al.,

2004). However, this defect is likely due to a lack of elongation and not a defect in pm5 specification. The T-box domain transcription factor tbx-2, also only affects specific muscle cells. However, tbx-2 mutants lack all ABa derived pharyngeal muscles (Roy

Chowdhuri et al., 2006; Smith and Mango, 2006), suggesting that it might specify the

ABa pharyngeal muscle fate. All three of these muscle transcription factors are downstream targets of PHA-4, consistent with the model that PHA-4 activates all pharyngeal genes. Lastly, Notch signalling has been shown to be involved in pm8 cell morphogenesis by activating a fusogen that ensures that it fuses with itself and not with other cells nearby (Rasmussen et al., 2008)

The development of some pharyngeal neurons has also been previously investigated. The homeobox transcription factor ceh-28 is thought to inhibit synaptogenesis in the M4 neuron (Ray et al., 2008). Mutations in another homeobox transcription factor, ceh-2, result in the loss of function of the M3 neuron, although its morphology is unaffected (Aspock et al., 2003). A zinc finger protein mnm-2 also regulates the proper function of the M3 neuron and indirectly affects the guidance of the axon of M2 which seems to use the M3 neuron as a guide (Rauthan et al., 2007). There are also many genes that have been shown to affect the morphology and function of the two NSM neurons (Axang et al., 2008), including the POU homeodomain transcription factor UNC-86 which regulates the genes necessary for serotonergic neurotransmission in

NSM (Sze et al., 2002). Despite this work, little is known about how other pharyngeal neurons are specified. Thus, for my thesis I decided to examine the development of the uncharacterized gland cells to elucidate the mechanisms by which cell fates in the pharynx are specified.

In this work, I chose to examine development of the pharyngeal glands. I chose this cell type for three reasons: first, nothing is known about regulation of gland gene expression nor about the specification of the glands (aside from the general involvement of PHA-4). Second, the function of the glands is poorly understood, although proposed roles include initiation of digestion, molting of the pharyngeal cuticle and resistance to pathogenic bacteria and the digestive tract glands of parasitic nematodes are known to play crucial roles in host-parasite interactions (reviewed in Jasmer et al., 2003). Third, several genes with gland-specific expression have been identified, based on a combination of microarray and in situ hybridization data (Ao et al., 2004). Lastly, all five of the gland cells derive from the MS blastomere, so the regulation of their specification is probably simpler than the muscle or marginal cells (which derive from ABa and MS lineages) and thus is more tractable.

1.4 Experimental Objectives

In this thesis, I investigate how the pharyngeal gland cells are specified in C. elegans by determining how gland expressed genes are regulated. Are all gland cell genes regulated by one central “terminal selector gene?” Is there a central regulator with modifiers that can act on multiple gene batteries? Or are there multiple transcription factors with multiple sub-groups of gland genes and are they organized with any discernible logic? I also examined the function of the glands based on phenotypic analysis of gland mutants. To do this I used the following three aims.

In the first aim, I identified a gland specific cis-regulatory motif (called PGM1) and its probable trans-acting factor (HLH-6). I accomplished this by computationally

22 analyzing a set of previously identified gland specific genes (Ao et al., 2004) for shared promoter elements. I then identified the transcription factor that acts through this cis- acting element based on the sequence of the motif.

In the second aim I determined the role of the trans-acting factor in gland development and function. I analyzed a loss-of-function mutant for this factor and found that it is involved in the development of the g2 cells. I also determined that the glands may play a role in aiding the feeding process.

In the last aim, I identified additional cis-acting elements in the promoters of

HLH-6 independent gland genes. I also found that there are some genes whose regulation cannot be explained by these two known gland motifs, suggesting that there are additional gland specific motifs and corresponding transcription factors.

Chapter Two: MATERIALS AND METHODS

2.1 Nematode Handling Conditions

Standard nematode handling conditions are described in Brenner, 1974. All animals were grown on Nematode Growth Medium (NGM) agar seeded with the E. coli strain OP50, except for the OP50-GFP, HB101 and B. subtilis strains used to analyze the hlh-6 mutant. OP50-GFP was grown on NGM plates containing 100 g/mL ampicillin,

HB101 was grown on NGM plates containing 200 g/mL streptomycin and B. subtilis was grown on standard NGM plates without antibiotics. All bacterial strains were provided by the Caenorhabditis Genetics Center. All worm strains were grown at either room temperature or placed at 15 degrees Celsius to stage them appropriately, except for strains analyzed in the growth, egg laying, Nile Red or pumping assays which were all grown at 25 degrees. For feeding with kanamycin-killed bacteria, OP50 cultures were grown for 5 hr at 37°C before addition of 100 g/mL kanamycin, then incubated for an additional hour. Bacterial growth was measured by standard light absorption at 600 nm.

Only cultures showing cessation of log-phase growth were used for feeding. Antibiotic- treated OP50 were plated on NGM agar plates containing 50 g/mL kanamycin and incubated at 25°C overnight. Bacteria from these plates were streaked onto LB plates

(with no antibiotic) to ensure that bacteria had been killed. The abiotic media was a provided as a gift from Dr. R. Gravel (University of Calgary, Canada), which was made as described previously (Fahlen et al., 2005).

2.2 Construction of Plasmids and Reporters

All transcriptional reporters were made by one of two methods, transgenic plasmids were constructed by PCR amplification of promoter fragments from genomic

DNA, followed by cloning into either the pPD95.77 or pPD95.77-YFP vectors (gifts from A. Fire), which contain the coding sequences for gfp and yfp, respectively. PCR transgenic reporters were made using the GFP stitching protocol described in Hobert,

2002. The first round of PCR consists of PCR amplification of promoter fragments from genomic DNA using 3’oligonucleotides that contain 22 base pairs of the 5’ gfp coding sequence, and in a second reaction PCR amplification of the gfp coding sequence from pPD95.77 (which includes synthetic introns). The second round of PCR amplification uses the products of the first two reactions as templates and “stitches” them together.

Mutations in specific sequences of the promoters were subsequently made by PCR-based site-directed mutagenesis (Ho et al., 1989). Two oligonucleotides were designed encompassing the sequence of interest but including the specific mutation desired. These oligonucleotides are then used in a two step PCR amplification similar to that described for “GFP stitching” above. All constructs containing site specific mutations were cloned into pPD95.77 and sequenced to confirm the desired mutation was created. A list of all oligonucleotides is found in Appendix A and a list of all plasmids made is found in

Appendix B.

Enhancer constructs were built using synthetic oligonucleotides that were cloned into pPD95.77. Use of this vector for enhancer assays was established previously

(Wenick and Hobert, 2004).

For rescue of hlh-6 mutants, a 3398 bp PstI-XbaI fragment of fosmid

WRM066cG05 that contains hlh-6(+) was subcloned into pBlueScriptII(SK+) (“genomic rescue”). The "minigene" construct was created by amplification and subcloning of the hlh-6 cDNA from a library provided by R. Barstead using oGD102 and oGD106.

Sequencing of the cDNA insert revealed a single silent mutation (nt.618 A→G), but no other errors. The cDNA was ligated to a 568 bp fragment of the hlh-6 promoter that is active in pharyngeal glands (Raharjo and Gaudet, 2007). A synthetic intron was cloned into a blunt-ended KpnI site of the hlh-6 cDNA using annealed primers oGD198 and oGD199.

The hs::HLH-6 plasmid was constructed as follows. The 807 base pair hlh-6 cDNA was amplified from a cDNA Library provided by R. Barstead (same source as for the “minigene” construct above). The product was digested and cloned into the heat- shock vector pPD49.83 (a gift from A. Fire), which includes a hsp16-41 heat-shock promoter (Stringham et al., 1992).

The hlh-6::EGL-1 plasmid was constructing by PCR amplification of egl-1 from genomic N2 DNA using oGD531 and oGD532. The amplified product was digested and cloned downstream of a 747 bp fragment of the hlh-6 promoter. Design of this construct was based on previous work (Rauthan et al., 2007).

The 750 base pair phat-5 cDNA was amplified from a cDNA library provided by

R. Barstead using oGD258 and oGD259. The product was digested and cloned in-frame to YFP or mCherry (Shu et al., 2006). The PHAT-5::YFP fusion was placed under the control of the lys-8 promoter, while the hlh-6 minimal promoter was sub-cloned from min-hlh-6::YFP (Raharjo and Gaudet, 2007) in front of the PHAT-5::mCherry fusion to

26 create the hlh-6::PHAT-5::mCherry construct. The myo-2::PHAT-5::mCherry plasmid was cloned using the myo-2 promoter from plasmid pSEM474 (Gaudet and Mango,

2002). All clones were verified by restriction digests and sequencing.

2.3 Strain Constructions

The hlh-6(tm299) II allele was kindly provided by S. Mitani (Gengyo-Ando and

Mitani, 2000). Presence of the tm299 deletion was followed by genomic PCR with oligonucleotides oGD65 and oGD97. The original hlh-6(tm299)-bearing chromosome contains a linked larval lethal mutation (let-x) to the left of hlh-6. hlh-6 was outcrossed five times and the arms of LG II were replaced by selecting appropriate recombinants tested for the presence of hlh-6(tm299) by PCR. First, we placed unc-4(e129) in cis with let-x hlh-6 and then selected Rol non-Daf recombinants from let hlh-6 unc-4(e120)/rol-

6(e187) daf-19(m86) to obtain + rol-6 hlh-6 unc-4 (GD211). Because this strain is Rol

Unc, in all subsequent functional assays a rol-6 unc-4 strain was used as a control.

Reporter DNA was injected at 5-30 ng/μL together with 50 ng/μL pRF4 (rol-

6(su1006)), which confers a dominant Roller phenotype (Mello et al., 1991), and 20-45 ng/μL pBS II (SK+) to a total DNA concentration of 100ng/μL. For some analyses, including all transgenes introduced into the GD211 background, we included 20 ng/μL of an intestine specific reporter (elt-2::GFP::LacZ, ges-1::mRFP::His2B or elt-

2::mTomato::HIS2B) that served as an independent marker for transgenic arrays when scoring expression (Fukushige et al., 1998). For injections with enhancer constructs, 50 ng/μL of the construct was injected with 50 ng/μL pRF4 into N2 animals. For hlh-

6::PHAT-5::mCherry, 40 ng/μL was injected while myo-2::PHAT-5::mCherry was

27 injected at 5 ng/μL, because the myo-2 promoter is very strong and can be toxic at higher concentrations. Except where noted, a minimum of two independent transgenic lines were analyzed for each construct.

The integrated hlh-6 reporter ivIs10 [hlh-6::YFP ges-1::mRFP::His2B rol-

6(su1006)] and integrated phat-1::YFP reporter ivIs12 [phat-1::YFP elt-2::GFP::LacZ rol-6(su10060] were generated by gamma-ray-induced integration of extrachromosomal arrays carried in a wild-type background (Mello and Fire, 1995).

To induce cell death in glands, the hlh-6::egl-1 construct was injected at 20 ng/μL with 30 ng/μL elt-2::mTomato::HIS2B and 50 ng/μL pBS II (SK+) into a strain carrying an integrated phat-1::YFP reporter (GD139 ivIs12, see above). Doubly transgenic animals were identified based on the Rol phenotype of GD139 (100%) and the presence of red intestinal fluorescence. Animals lacking visible YFP expression (indicating a loss of glands) were then analyzed for survival and growth.

For rescue of hlh-6, both the genomic fragment and the minigene were injected at

50 ng/µL with 30 ng/µL of phat-1::YFP and 20 ng/μL elt-2::GFP::LacZ into N2 animals. These arrays were subsequently crossed into GD211.

RNAi was performed as described previously (Timmons et al., 2001; Timmons and Fire, 1998). “Feeding” RNAi was preformed by growing available dsRNA- expressing bacteria and seeding them onto NGM agar containing 12.5 g/mL tetracycline and 0.4 mM isopropyl-β-D-thiogalactopyranoside (IPTG) and grown at room temperature overnight. Young adult worms were placed on the RNAi plates and the progeny were analyzed. dsRNA for “injection” RNAi was created by in vitro transcription using T7

RNA polymerase (Promega) and injected at 1 g/ L.

2.4 Motif Searches using Improbizer

I used the Improbizer program (Ao et al., 2004, available at http://www.soe.ucsc.edu/~kent/improbizer/ ) available at http://www.soe.ucsc.edu/~kent/improbizer/ to search for possible gland-specific regulatory elements. I initially searched for motifs occurring once per sequence, using the input sequence as background. The motif presented here (PGM1) was obtained with a search for a motif size of six. Searches for motifs of larger sizes (8–20 bases) recurrently found variations of PGM1. Other parameters of Improbizer were used at their default settings. I also performed control runs in which the input gene sequence was randomized and searched and found that only PGM1 obtained an Improbizer score greater than the scores of ten or more control runs.

To find probable occurrences of PGM1 in other promoters (as in pqn-8 and lys-8),

I used the Improbizer sister program, Motif Matcher (www. http://www.cse.ucsc.edu/

~kent/improbizer/motifMatcher.html), which searches for top-scoring matches to the

Improbizer-generated position weight matrix.

2.5 Electrophoretic Mobility Shift Assay

Recombinant GST-HLH-6 and His-HLH-2 proteins were purified using standard techniques. The His-HLH-2 plasmid was provided from Dr. M. Krause (Krause et al.,

1997). The GST-HLH-6 construct was created by cloning the hlh-6 cDNA from the hs::HLH-6 plasmid in frame into the pGEX-4T2 plasmid (Addgene). Both His-HLH-2 and GST-HLH-6 plasmids were transformed into BL21(DE3)pLysS cells and grown

29 overnight at 37 degrees. Induction cultures of His-HLH-2 containing bacteria were grown at 37 degrees for 4 hours before the addition of IPTG to a concentration of 1mM. Induced

His-HLH-2 cultures were further incubated for 2 hours before purified. Cultures of GST-

HLH-6 containing bacteria were grown at room temperature for 6 hours and induced with

10mM IPTG for 1 hour and then purified. Cells were collected by centrifugation at 3000 rpm for 10 min at 4°C, washed in PBS and pelleted again. The cells were then sonicated on ice until no longer viscous and centrifuged at 5000 rpm for 10 min at 4°C.

Supernatants were collected and applied to glutathione-sepharose beads or Ni-NTA agarose beads (Qiagen) respectively. After incubations the GST fusion protein was eluted with 20 mM Glutathione and the His fusion protein with 250 mM Imidazole. Samples were run on SDS-PAGE to confirm the presence of fusion protein and quantify the concentration.

Probe DNA oligonucleotides were annealed and then 32P-labeled with the Klenow fragment enzyme. Oligonucleotides used were oGD137 and oGD138 taken from the

PGM1 site in the promoter of phat-3, oGD47 and oGD48 for cold 3xPGM1, oGD364 and oGD365 for cold 3xPGM1Δflank and oGD384 and oGD385 for cold 3xPGM1ΔEbox.

Binding conditions used were previously described (Krause et al., 1997). 20 μL of protein, probe, 0.15M KCL, 25 mM Hepes-HCl (pH 7.9), 2.5 mM MgCl2, 20mM dithiothreitol, 2.5 % (v/v) glycerol, 0.1% (v/v) Nonidet P-40, 37 ng/μg poly (dI-dC) and

50 ng/μL bovine serum albumin were incubated at 37 degrees for 15 mins. Samples were run on 5% (w/v) native polyacrylamide gels at 100V for one hour at room temperature in a solution of 45 mM Tris, 45 mM Boric acid and 1 mM EDTA pH 8.

2.6 Ectopic Expression of HLH-6

The transgenic line ivEx43 [hs::hlh-6 rol-6(su1006)] was created from injections of

50 ng/μL hs::hlh-6 and 50 ng/μL pRF4 into N2. ivEx93[hs::hlh-2 rol-6(su1006)] was created from injections of 50 ng/μL pKM1035 (hs::hlh-2) (a gift from M. Krause) and 50 ng/μL pRF4 into N2. ivEx65 [hs::hlh-6 hs::hlh-2 rol-6(su1006)] was created from injections of 50 ng/μL hs::hlh-6, 50 ng/μL pKM1035 (hs::hlh-2) and 50 ng/μL pRF4 into

N2. The reporter strain ivEx44 [phat-1::YFP] was made by injecting phat-1::YFP into N2 animals, using the reporter as its own marker. Males carrying ivEx44 were crossed to ivEx43, ivEx93 or ivEx65 animals to generate the double transgenic strains ivEx43; ivEx44, ivEx44;ivEx93 and ivEx44; ivEx65. Double transgenics were followed by phat-

1::YFP expression and the Rol phenotype. Embryos from these strains (containing both transgenic arrays) were collected for 1 hr at 25°C. These embryos were then placed at

33°C for 1 hr and incubated at 20°C for 5 to 7 hr before scoring phat-1::YFP expression.

2.7 Lifespan Assays

Homozygous hlh-6 segregants from a let-x hlh-6 unc-4/mC6g strain were identified by the absence of the myo-2::GFP marker carried on mG6g. Gravid adults were allowed to lay eggs for 1 hr at 25°C. Eleven hours later (when embryos should have hatched), larvae were scored for GFP expression to determine the genotype as either let-x or let-x(+), and transferred to new plates individually. Larvae were scored for viability at several intervals and were considered dead when they no longer responded to gentle poking by a platinum wire. OP50 was the standard E. coli strain used for all experiments, except where indicated. For the lifespan of animals grown on plates without bacteria,

31 plates with kanamycin-killed bacteria or carbenicillin-killed bacteria, embryos were first collected on normal OP50-containing plates, transferred to the corresponding plates and treated with a bleach solution (2% sodium hypochlorite and 1.3 M NaOH) to ensure no living bacteria were transferred with the embryos.

2.8 Growth Assays

For measurement of body length, embryos laid over a one hour period by gravid adults were collected from and grown at 25º. Larvae were removed from plates and transferred to slides at the indicated times. Pictures were taken at 400X magnification and the lengths of the animals were measured using ImageJ (http://rsb.info.nih.gov/ij/) as described previously (Morck and Pilon, 2006). Greater than twenty animals were analyzed for each genotype at each time-point. For measuring time to reach adulthood, single eggs were placed on plates and followed at 24 hour intervals until the animal reached adulthood. For brood sizes the number of eggs laid was counted throughout the lifetime of each animal.

hlh-6 mutants were stained using the dye Nile Red (Sigma N-3013) as described

(Ashrafi et al., 2003). Briefly, L4 animals of the indicated genotype were transferred to plates with 0.05 μg/mL Nile Red and allowed to grow for 24 hours before being scored using conventional fluorescence microscopy. At least fifteen animals were observed for each genotype and one animal that represents the average level of fluorescence per each genotype is shown.

For pharyngeal pumping assays, L4 animals were transferred to fresh plates and grown for 24 hours at room temperature before scoring. Pumping was counted under a dissecting microscope at 100x magnification.

2.9 Pharyngeal Cell Counts in WT vs. HLH-6

Pharyngeal cell type-specific markers were examined in hlh-6 mutants. Only cells in the back half of the posterior bulb were counted, (except for the neuronal marker for which all posterior bulb neurons were counted). I used the pan-neuronal rgef-1::GFP (a gift from Dr. Oliver Hobert) marker to count pharyngeal neurons (expect 7 in wild type), myo-2::GFP::HIS2B (a gift from Dr. Susan Mango) to count pharyngeal muscle nuclei

(expect 4 in wild type), pax-1::GFP::HIS2B (a gift from Dr. Susan Mango) to count pharyngeal marginal cell nuclei (expect 4 in wild type, as pax-1::GFP is also expressed in the pm8 muscle) and the pha-4::GFP::HIS2B reporter (a gift from Dr. Susan Mango) to count all pharyngeal nuclei (expect 11). All four markers were crossed into the hlh-6 mutant strain GD211.

Chapter Three: IDENTIFICATION OF THE GLAND SPECIFIC CIS- REGULATORY ELEMENT PGM1 AND THE TRANS-ACTING FACTOR HLH-6

I have chosen to examine the specification of the pharyngeal gland cells, one of five cell types within the pharynx as a model for cell-type specification within an organ.

To do this I set out to initially identify a gland specific cis-regulatory element in the hopes of subsequently identifying a gland transcription factor. I hypothesised that genes that are expressed in the glands are co-regulated and therefore share a common cis- regulatory element. In this chapter, I describe the identification of such an element, named PGM1 (for Pharyngeal Gland Motif 1), the demonstration that this element is both necessary and sufficient for gland expression, and the identification of the probable transcription factor acting through it. In this way, I not only identified a pharyngeal gland transcription factor, but also some of its downstream targets. This approach is advantageous over traditional forward genetic approaches in that both a transcription factor and a number of its downstream targets are identified. This approach also clearly defines the transcription factor’s binding site, which would be difficult to determine otherwise.

3.1 Identification of Gland Specific Genes

To investigate regulation of pharyngeal gland development, I began with a list of fourteen confirmed and probable gland-specific genes, based on previous work (Table

3.1, Ao et al., 2004). Eight genes identified by microarray analysis as expressed in the pharynx had in situ hybridisation patterns that were gland specific. Seven of these eight genes are predicted to encode proteins whose only recognizable features are a signal

Table 3.1 Gland-specific genes identified by microarray

Supporting in situ hybridization data is from NEXTDB (Kohara, 2001a; Kohara, 2001b).

Microarray data (Gaudet et al., 2004) indicates probable pharyngeal expression, though not necessarily gland-specific. * = reporter expression previously described (Ao et al.,

2004).

Gene Name Expression Supporting Prominent Motifs Evidence B0507.1 All pharyngeal Reporter, in situ EGF-like (x2), glands hybridization Worm-specific repeat type 1 (x2) C46H11.8 phat-1 All pharyngeal Reporter*, in situ ShK (x4) glands hybridization C46H11.9 phat-2 All pharyngeal in situ ShK (x3) glands hybridization C49G7.4 phat-3 All pharyngeal Reporter*, in situ ShK (x3) glands hybridization F07C4.11 Probable gland Microarray ShK (x2) F41G3.10 Probable gland Microarray ShK (x3) M153.3 Probable gland Microarray ShK (x2) T05B4.3 phat-4 All pharyngeal in situ ShK (x3) glands hybridization T05B4.8 Probable gland Microarray ShK (x3) T05B4.11 phat-5 Anterior-most Reporter, in situ ShK (x3) pharyngeal hybridization glands T05B4.12 Probable gland Microarray ShK (x3) T05B4.13 Probable gland Microarray ShK (x3) T10B10.6 phat-6 Anterior-most in situ ShK (x1) pharyngeal hybridization glands T20G5.7 dod-6 All pharyngeal in situ ShK (x1) glands hybridization

35 peptide and multiple copies of the ShK motif, a cysteine-rich sequence first described in metridin toxin from the sea anemone (Castaneda et al., 1995). Proteins containing only

ShK motifs appear to be gland-specific, while proteins containing ShK motifs in the presence of other recognizable domains (such as astacin in NAS-14 or tyrosinase in

TYR-1) are not gland-specific (Ao et al., 2004). Thus I predicted that microarray positives that encode ShK proteins are gland expressed, adding six more genes to the list of probable gland expressed genes (Table 3.1). Interestingly the majority of these genes are very close to each other, sometimes right beside one another, which could increase the probability that they are co-regulated. Since not all of these fourteen genes have supporting in situ hybridization data, I will refer to ShK-encoding genes with in situ hybridization patterns that are gland-specific as phat genes, for pharyngeal gland toxin- related.

To verify the in situ hybridization patterns, I constructed GFP or YFP reporters for four of the genes (two of which, phat-1/C46H11.8 and phat-3/C49G7.4, were previously reported, Ao et al., 2004) and found that all four were expressed specifically in pharyngeal glands (Figure 3.1 and Table 3.1). Of the four genes, three (B0507.1, phat-

1, and phat-3) were expressed in all five glands (Figure 3.1B, D, H), while phat-5 was only expressed in the two anterior-most glands, the left and right g1A cells (g1AR and g1AL, Figure 3.1F). Previous reports have suggested that the g1AR and g1P cells are fused (Albertson and Thomson, 1976), yet I saw no passage of phat-5-expressed YFP from g1AR to g1P, suggesting either that YFP is restricted from diffusing between these cells or that the two cells are not fused. Although my reporter for B0507.1 was exclusively expressed in the pharyngeal glands, a shorter reporter created by

Figure 3.1 PGM1 is required for expression of some pharyngeal gland genes

(A) WebLogo (Crooks et al., 2004) of the computationally identified PGM1. (B-M)

Fluorescence micrographs of gland-expressed GFP or YFP reporters with wild-type promoter sequence (left column) or promoter sequence in which PGM1 is mutated (right column). In wild-type sequences (left) the E-box is underlined and in mutant sequences

(right) the mutation is underlined. Anterior is at left and the pharynx is outlined. Scale bars represent 10 m.

Reece-Hoyes et al. (2009) was also expressed in the pharyngeal-intestinal valve cells, spermatheca, rectal gland cells and distal tip cell. However by in situ, B0507.1 appears to be gland specific (Kohara, 2001a; Kohara, 2001b).

3.2 Identification of Pharyngeal Gland Motif 1

Having identified a set of confirmed gland specific genes, I proceeded to search for a shared cis-regulatory element in their promoters. By searching the upstream 500 bp

(relative to the ATG) of the fourteen gland genes using the Improbizer program (Ao et al., 2004) for shared sequence motifs, I identified one candidate gland-specific cis-acting element, which I named PGM1 (for Pharyngeal Gland Motif 1, Figure 3.1A). This size of promoter was justified because many of the gland genes have neighbouring genes within

500 bp upstream, consistent with the observation that C. elegans promoters are generally small (Okkema and Krause, 2005; Spieth and Lawson, 2006). The Improbizer program searches for sequences found in promoters that are present more often than expected by chance. Thus the Improbizer “score” is only meaningful in relation to control runs which randomize the input sequence. PGM1 was the only motif identified by Improbizer that had a position weight matrix score higher than any of the motifs generated in control runs

(See Materials and Methods), suggesting that it might be a functional regulatory element.

In addition, PGM1 appeared to be enriched in the promoters of gland-expressed genes, as these promoters were four times more likely to contain significant occurrences of PGM1

(12/14 = 86%) than a control set of promoters from pharyngeal (but not gland-specific) genes (20/96 = 21%, Table 3.2). Scores over 7.00 were considered to be good matches to

PGM1, consistent with my functional characterization of the motif (see below). Given

Table 3.2 List of gland and pharyngeal (non-gland) genes and their associated Motif Matcher score

The gland list is as in Table 3.1, the pharyngeal (non-gland) list is a list of previously identified microarray positives with supporting expression data (Gaudet and Mango,

2002). Motif Matcher scores were generated using the computationally identified PGM1 run against 500 bp of sequence upstream sequence (relative to the ATG) for each of the indicated genes. Motif Matcher is the sister program to Improbizer and is available at http://www.soe.ucsc.edu/~kent/improbizer/motifMatcher.html

Gene Set Gene Score Gland T05B4.12 5.09

Gland T05B4.11 10.85 Pharyngeal T06E4.11 10.32

Gland C49G7.4 10.85 Pharyngeal F35A5.3 9.84

Gland T20G5.7 10.83 Pharyngeal C10G8.5a 9.16

Gland F07C4.11 10.83 Pharyngeal F08B12.1 8.97

Gland B0507.1 10.83 Pharyngeal K08F8.2 8.90

Gland C46H11.9 9.27 Pharyngeal T05E11.3.1 8.47

Gland T05B4.3 8.45 Pharyngeal M03D4.4a 8.45

Gland T05B4.8 8.43 Pharyngeal T22B2.6 8.42

Gland C46H11.8 8.43 Pharyngeal C01B10.5a 8.29

Gland F41G3.10 7.96 Pharyngeal Y76A2B.2 8.24

Gland M153.3 7.57 Pharyngeal DY3.5 8.02

Gland T05B4.13 7.50 Pharyngeal T06E4.9 7.93

Gland T10B10.6 6.30 Pharyngeal F49E10.2a 7.81

Pharyngeal R07B1.9 7.80 Pharyngeal T21C9.9 6.28

Pharyngeal T23F6.1 7.64 Pharyngeal C14B9.2.1 6.11

Pharyngeal T10E10.4 7.45 Pharyngeal F22A3.1 5.98

Pharyngeal C23H3.9a 7.35 Pharyngeal F21D5.9 5.98

Pharyngeal C03A7.14 7.26 Pharyngeal F07H5.8 5.96

Pharyngeal F13H8.4 7.09 Pharyngeal ZK816.4 5.93

Pharyngeal F25G6.6 7.07 Pharyngeal T06E4.7 5.85

Pharyngeal F54F3.1 7.00 Pharyngeal F12F3.1a.1 5.83

Pharyngeal R03C1.1 6.89 Pharyngeal T06E4.8 5.76

Pharyngeal C14C11.8 6.89 Pharyngeal F21H11.3.1 5.75

Pharyngeal C06G1.2 6.84 Pharyngeal CD4.9 5.66

Pharyngeal T27C5.10 6.78 Pharyngeal M02G9.1 5.64

Pharyngeal M01D1.2a 6.78 Pharyngeal R09E10.5 5.59

Pharyngeal K06A1.3 6.78 Pharyngeal F54E2.3a 5.53

Pharyngeal F16B4.8 6.78 Pharyngeal F58G4.1 5.39

Pharyngeal F19G12.7 6.73 Pharyngeal T04C9.6a.1 5.29

Pharyngeal T04H1.6 6.68 Pharyngeal F53H4.5 5.27

Pharyngeal ZK892.7 6.61 Pharyngeal F11E6.8 5.25

Pharyngeal F10G8.8 6.51 Pharyngeal ZK418.3 5.23

Pharyngeal C03A7.7 6.42 Pharyngeal C03A7.8 5.10

Pharyngeal T04C9.4a 6.34 Pharyngeal F48E3.8a 4.99

Pharyngeal R09B5.5 6.31 Pharyngeal F40E10.5 4.98

Pharyngeal C04E6.12 4.97 Pharyngeal ZK1025.7 3.68

Pharyngeal F26A10.2 4.94 Pharyngeal D2024.4 3.66

Pharyngeal D1054.9a 4.68 Pharyngeal T18D3.4 3.62

Pharyngeal T01D1.6 4.67 Pharyngeal F45G2.2a 3.24

Pharyngeal C03A7.4 4.58 Pharyngeal F14B4.1 3.20

Pharyngeal R07E3.2 4.53 Pharyngeal K02F6.4 3.04

Pharyngeal F54E2.2 4.51 Pharyngeal F53B3.3 3.02

Pharyngeal T23F4.4 4.48 Pharyngeal F09F7.8 2.93

Pharyngeal T23F1.6 4.37 Pharyngeal K11H12.5 2.86

Pharyngeal F53A9.3 4.30 Pharyngeal R07C3.5 2.78

Pharyngeal R09F10.7 4.02 Pharyngeal F38B6.1 2.78

Pharyngeal R09F10.2 4.02 Pharyngeal F35C5.10 2.65

Pharyngeal ZC250.1 4.00 Pharyngeal W10D9.1 2.53

Pharyngeal F57B1.6 3.97 Pharyngeal C32H11.5 2.49

Pharyngeal R02F11.1 3.92 Pharyngeal W01C9.1 2.34

Pharyngeal T06E4.10 3.91 Pharyngeal M195.2 2.15

Pharyngeal C44H4.1 3.88 Pharyngeal F20B10.3 2.07

Pharyngeal T06D8.3 3.82 Pharyngeal ZK1067.7 1.44

Pharyngeal E01G6.1 3.81 Pharyngeal C27A2.5 1.19

Pharyngeal D1009.5 3.72

42 that this threshold score is somewhat arbitrary, I also examined the difference between the scores for the two gene sets using the Mann-Whitney U test and found that gland genes had a significantly higher PGM1 score than did non-gland genes (P<0.001). Lastly, some occurrences of PGM1 are very highly conserved across other nematode species

(Figure 3.2), suggesting they are important in the proper regulation of the genes.

3.3 PGM1 is Necessary for Gland Expression

Analysis of PGM1 in the context of pharyngeal gland-specific promoters demonstrated that PGM1 was required for expression. Deletion analysis of the phat-3 promoter supported the potential for PGM1 in activating gland expression. Three 5’ deletion reporters were constructed (Figure 3.3). The first two deletion constructs removed one or both of the PHA-4 consensus binding sites and resulted in decreased levels of expression, consistent with a role for PHA-4 in the activation of all pharyngeal genes (Gaudet and Mango, 2002). The third deletion construct resulted in a complete loss of expression, suggesting the sequence between –212 and –102 bp relative to the ATG is necessary for activation of the gland expression. The PGM1 site identified in the phat-3 promoter is within this necessary region, suggesting it could be required for expression.

Consistent with the deletion construct results, a site-directed mutation in the PGM1 sequence of the phat-3 reporter completely eliminated expression (Figure 3.1E).

Since the PGM1 site in the phat-3 promoter was required for expression, I tested the remaining reporters for PGM1-dependence. The promoter of phat-1 had two matches to the PGM1 position weight matrix, one at -560 and another at -69 bp relative to the start codon. Mutation analysis of both suggested only the one at -69 bp was required for

Figure 3.2 Occurrences of PGM1 are conserved among nematodes

The PGM1 occurrences in the promoters of phat-3, phat-4, phat-5 and B0507.1 and the corresponding sequences from their homologs in C. briggsae, C. brenneri and C. remanei are shown. Sequences from the homologs were identified using the UCSC Genome

Browser (http://genome.ucsc.edu). The computationally defined PGM1 is highlighted in yellow.

Figure 3.3 The region between -212 and -102 bp is required for phat-3 expression

A representative diagram of the phat-3 promoter (right column) and fluorescence micrographs of the corresponding expression patterns (left column) are shown. Wild type expression is shown in (A). The reporter expression is decreased in the -326 (B) and -212 bp (C) reporters but still expressed in all five gland cells. The -102 bp (C) reporter is not expressed. Open triangles represent candidate PHA-4 binding sites, black rectangles represent occurrences of PGM1. Anterior is at left and the pharynx is outlined. Scale bars represent 10 m.

47 expression (Figure 3.4). Mutations in the PGM1 sequence of B0507.1 and phat-5 greatly reduced both the intensity and frequency of expression (Figure 3.1I, G). The promoter of phat-5 has one other potential occurrence of PGM1 that could account for its residual activity (at -118 bp). The B0507.1 promoter has no other apparent PGM1 sequences, suggesting that the remainder of its gland expression is dependent on an as yet unidentified cis-regulatory motif. The B0507.1 reporter created by Reece-Hoyes et al. also displayed only partial dependence on the PGM1 sequence for gland expression

(Reece-Hoyes et al., 2009). Together, these results suggested that PGM1 is necessary for expression of many, but not all, genes in pharyngeal glands.

I queried other gland-expressed genes to determine whether they also required

PGM1 for expression. Through a literature search I identified two genes that were not part of the original data set, but that were reported to be expressed in glands: pqn-8 and lys-8 (Hope et al., 1996; Mallo et al., 2002). Transcriptional reporters made for the genes confirmed gland expression (Figure 3.1J, L). The pqn-8 reporter was expressed exclusively in pharyngeal glands whereas the lys-8 reporter was expressed in pharyngeal glands and the intestine, as reported. The pqn-8 promoter sequence only contained one potential PGM1 and mutation of this completely abolished expression (Figure 3.1K). The lys-8 promoter had three potential PGM1 sites at -180, -452 and -581 bp relative to the

ATG (Figure 3.5). Two of these sequences (at -180 and -452) are not required for expression in pharyngeal glands, while mutation of the third site (-581 bp) resulted in a loss of expression (Figure 3.5).

Figure 3.4 The PGM1 at -69 bp is required for phat-1 expression

A representative diagram of the phat-1 promoter (right column) and fluorescence micrographs of the corresponding expression patterns (left column) are shown. Wild type expression is shown in (A). The mutation of the PGM1 consensus at -560 bp (in the reverse orientation) is underlined in (B) and does not affect expression. The mutation of the PGM1 consensus at -69 bp (in the reverse orientation) is underlined in (C) and completely abolishes expression. Open triangles represent candidate PHA-4 binding sites, black rectangles represent occurrences of PGM1. Anterior is at left and the pharynx is outlined. Scale bars represent 10 m.

Figure 3.5 The PGM1 at -581 bp is required for lys-8 expression

A representative diagram of the lys-8 promoter (right column) and fluorescence micrographs of the corresponding expression patterns (left column) are shown. Wild type expression is shown in (A). The mutation of the PGM1 consensus at -581 bp (in the reverse orientation) is underlined in (B) and results in the complete loss of gland expression. The mutations in the PGM1 sites at -452 (in the reverse orientation) and -180 bp (in the forward direction) are underlined in (C) and (D) respectively and do not affect reporter expression. Open triangles represent candidate PHA-4 binding sites, black rectangles represent occurrences of PGM1. Anterior is at left and the pharynx is outlined.

Scale bars represent 10 m.

3.4 PGM1 is Sufficient for Gland Expression

Given that PGM1 is necessary for expression of many genes in pharyngeal glands, I next asked whether PGM1 was also sufficient for gland expression. To do this I performed an in vivo enhancer assay where I placed multiple copies of the motif in front of a promoter-less GFP. This technique has been well established and works in multiple contexts (Okkema and Fire, 1994; Wenick and Hobert, 2004). Indeed, three tandem copies of the PGM1 sequence from phat-3 placed upstream of a "promoter-less" reporter

(to make the “3xPGM1” construct) was sufficient to activate pharyngeal gland expression in 78% (31/40) of transgenic animals (Figure 3.6B). A fraction of these animals (7/31) also showed weak expression in the I3 pharyngeal neuron, a sister cell of the g1P gland

(Sulston et al., 1983). A control reporter that contained no insert did not show any expression (n= 44, Figure 3.6A). These results indicate that PGM1 is a pharyngeal gland- specific enhancer element, and further suggests that PGM1 is a binding site for one or more transcription factors that function in pharyngeal glands.

3.5 PGM1 Contains Two Important Regulatory Sequences

Computational predictions of binding sites can be inaccurate or imperfect. Indeed the computationally predicted PGM1 at -560 bp of the phat-1 promoter was not functionally required for expression. Thus I aimed to redefine the PGM1 motif in hopes of being able to identify the transcription factor(s) that binds to it. Manual alignment of the functionally defined PGM1 sequences revealed an extended consensus of

CAnvTGhdYMAAY (where V = A, C or G, H = A, C or T, D = A, G or T, M = A or C, and Y = C or T, Figure 3.7). This extended consensus is present in 13 of the 14 genes

Figure 3.6 The extended PGM1 is sufficient for gland expression

Fluorescence micrographs of GFP enhancer constructs containing (A) no insert, (B) three tandem copies of the extended PGM1, (C) three tandem copies of the extended

PGM1 in which the E-box has been mutated , (D) three tandem copies of the extended

PGM1 in which sequence flanking the E-box has been altered and (E) three tandem copies of the extended PGM1 in which two additional bases are inserted between the E- box and flanking sequence. Mutations are underlined. Anterior is at left and the pharynx is outlined. Scale bars represent 10 m.

Figure 3.7 Alignment of functional PGM1 sites

Alignment of PGM1 occurrences in the promoters of gland-expressed genes. Expression of genes in bold is experimentally verified to be both PGM1 and HLH-6 dependent.

57 in the initial list that contained PGM1. The promoter of T05B4.13 does not contain this consensus (it does however contain sequences that differ by one base pair) and it is analyzed in Chapter 5. The functionally defined consensus may represent either an extended binding preference for the relevant trans-acting factor or the juxtaposition of binding sites for two (or more) distinct factors. Two sequences are perfectly conserved in this functional consensus, CAnnTG and YMAAY. Importantly, CAnnTG is an E-box, the consensus binding site for basic helix-loop-helix (bHLH) transcription factors (Ephrussi et al., 1985). Mutations that specifically disrupt the E-box sequence in the phat-5,

B0507.1, lys-8 (Figure 3.1G, I, M) and phat-1 (Figure 3.8C) promoters eliminate PGM1 activity, suggesting that the E-box is required for expression. However, the E-box is not sufficient for PGM1 activity: mutation of sequence flanking the E-box in the phat-1 reporter also resulted in a significant loss of expression (Figure 3.8D), suggesting that the extended sequence is required for activity.

Given the apparent extended consensus sequence for PGM1, I performed additional enhancer tests to determine what portions of PGM1 were required for its activity. I first tested a version of the 3xPGM1 plasmid in which all three copies of the E- box were changed from CAnnTG to AAnnTG. This construct (3xPGM1ΔE) showed no expression in transgenics, indicating (as above) that the E-box was required for PGM1 activity (Figure 3.6C). I next tested an enhancer in which sequence flanking the E-box was altered (3xPGM1Δflank) and found that this sequence was also required for PGM1 activity (Figure 3.6D), demonstrating that the E-box is not sufficient for PGM1 activity.

Additionally, I created an enhancer that inserted two extra bases (TT) after the E-box

(3xPGM1-TT) and found that this eliminated PGM1 activity (Figure 3.6E). These two

Figure 3.8 The E-box and YMAAY components of PGM1 are necessary for phat-1 expression

The functionally relevant PGM1 occurrence of the phat-1 promoter (right column) and fluorescence micrographs of the corresponding expression patterns (left column) are shown. Wild type expression (A) and full PGM1 mutation (B) are identical to those shown in Figure 3.4. The mutation of only the E-box (C) or only the YMAAY (D) results in the complete loss of gland expression. Mutations are underlined. Anterior is at left and the pharynx is outlined. Scale bars represent 10 m.

60 extra bases either disrupt the binding sequence for a relevant transcription factor or it suggests that the spacing between the E-box and YMAAY sequences is important. Thus there are two models that fit these results. The PGM1 consensus might be an extended consensus for a bHLH protein that is disrupted by the addition of extra bases. Or the extended consensus is a binding site for a second transcription factor. Solved bHLH-

DNA structures indicate contact of bHLH proteins up to but not beyond three bases outside of the E-box (Ma et al., 1994; Shimizu et al., 1997). Since the YMAAY sequence extends beyond this limit it is likely that PGM1 contains binding sites to two transcription factors (discussed further in Chapter 3.7).

3.6 HLH-6 is a Gland Specific Transcription Factor

Since PGM1 activity is dependent on an E-box sequence, my search for the relevant trans-acting factor(s) began with bHLH proteins. bHLH proteins typically bind to DNA as heterodimers, composed of a ubiquitous “Class I” subunit and a tissue- restricted “Class II” partner (Massari and Murre, 2000). In C. elegans, the sole Class I bHLH is encoded by hlh-2 (Krause et al., 1997), which is expressed in many cells throughout development, including the glands. To identify the relevant Class II bHLH, I examined data from microarray experiments that identified candidate pharynx-expressed genes (Gaudet and Mango, 2002; Gaudet et al., 2004), including three Class II bHLHs: hlh-3, hlh-6 and hlh-8. Both hlh-3 and hlh-8 are expressed exclusively in non-pharyngeal tissue (in neurons and muscles, respectively, Harfe et al., 1998; Krause et al., 1997) suggesting that they are false positives with respect to the microarray data and are thus

61 unlikely to function through PGM1. At the time of my analysis, hlh-6 was uncharacterized and was therefore a candidate PGM1 trans-acting factor.

To examine the involvement of hlh-6 in PGM1 activity, I first determined the expression of a transcriptional reporter that included almost all intergenic sequence (1175 bp of 1190 bp) between hlh-6 and its nearest upstream neighbour, T15H9.2. I found that hlh-6::YFP was expressed strongly and specifically in the pharyngeal glands (98% of transgenics), with occasional (12%), weak expression in the pharyngeal neuron I3 (Figure

3.9B). Expression was first detectable shortly after the terminal cell division that gives rise to pharyngeal glands (bean stage embryos) and persisted throughout the life cycle in all five pharyngeal glands. Because PGM1 and hlh-6 both appear to be active in pharyngeal glands and because PGM1 contains a bHLH binding site, I hypothesized that

HLH-6 (±HLH-2) is the cognate trans-acting factor for PGM1. Indeed, HLH-2 is the only HLH protein that was found to interact with HLH-6 in a yeast 2-hybrid assay of all

C. elegans HLH proteins (Grove et al., 2009).

3.7 HLH-6 Acts Through PGM1

I determined that HLH-6 is required for PGM1 activity by demonstrating that

PGM1-dependent reporters were not expressed in hlh-6 mutants. I initially performed

RNAi against hlh-6 by both “feeding” and “injection” methods (Kamath et al., 2003;

Timmons et al., 2001). Injection of double-stranded hlh-6 RNA had no observable effect on either wild type or RNAi-sensitive worms (rrf-3, Simmer et al., 2002) as previously reported (Thellmann et al., 2003), but also had no effect on expression of a PGM1- dependent reporter using a strain that contained the phat-1::YFP reporter. Some tissues

Figure 3.9 hlh-6 is expressed in pharyngeal glands

(A) Schematic of the genomic region containing hlh-6. The position of the deletion allele tm299 is indicated. The portion of hlh-6 encoding the DNA Binding Domain (DBD) is shown as is the hlh-6 "minigene", which rescues all aspects of the hlh-6 mutant phenotype. (B) Expression of the hlh-6::YFP reporter (indicated in (A)), containing 1175 bp (of 1190 bp) of intergenic sequence from the ATG of hlh-6 to just downstream of the stop codon of the next upstream gene, T15H9.2 Anterior is at left and the pharynx is outlined. Scale bar represents 10 m.

(e.g. neurons) in C. elegans are known to be RNAi insensitive (Timmons et al., 2001) and other studies in the Gaudet Lab have demonstrated that the pharyngeal glands are in fact resistant to RNAi (S. Wen personal communication and data not shown). I therefore decided to examine an hlh-6 deletion mutant that was generated by a component of the C. elegans Knockout Consortium using Ultra-Violet mutagenesis in the presence of trimethylpsoralen (Gengyo-Ando and Mitani, 2000). The deletion mutant hlh-6(tm299) is a probable null, as it removes 595 bp from hlh-6, including the splice donor from the second intron, resulting in a frameshift and premature stop codon (Figure 3.9A).

The original hlh-6(tm299)-bearing chromosome contained a linked larval lethal mutation (let-x) to the left of hlh-6 (see Materials and Methods and Chapter 4). However the hlh-6(tm299) mutation is homozygous viable, which allowed me to examine gland reporter expression in these mutants. I found that expression of 6/6 gland reporters (phat-

1, phat-3, phat-5, B0507.1, pqn-8 and lys-8) were significantly reduced in hlh-6 animals

(Figure 3.10, 3.11). For example, only 26% of hlh-6 mutants had visible phat-1::YFP expression (n=65), and this expression was significantly weaker than the expression in wild type animals. It is possible that in the absence of hlh-6 some other HLH protein

(such as HLH-2 possibly acting as a homodimer) activates expression of phat-1, thus explaining why this reporter is expressed in 26% of hlh-6 mutants, but not expressed when the E-box in the promoter is mutated (Figure, 3.8C). Four of the other gland reporters showed a similar loss of expression in hlh-6 mutants. Expression of the B0507.1 reporter was less affected than the others, consistent with it being only partially PGM1 dependent (Figure 3.11E). This result was also confirmed using the B0507.1 reporter

Figure 3.10 hlh-6 is required for PGM1 activity

Quantitation of the number of animals expressing each reporter in hlh-6 mutants. For the phat-1::YFP reporter in wild type and hlh-6 mutants, only one transgenic line was scored but the same array was used in both genotypes. Two lines of the genomic rescue were scored for phat-1::YFP expression (lines 5 and 2). An integrated hlh-6 reporter was scored for hlh-6 expression. Only one line of “minigene” rescue was scored. Number of animals scored is indicated.

Figure 3.11 hlh-6 is required for expression of some gland genes

Representative images of the (A) hlh-6, (B) phat-1, (C) phat-3, (D) phat-5, (E) B0507.1,

(F) pqn-8 and (G) lys-8 reporters in hlh-6 mutant animals. hlh-6, B0507.1 and lys-8 reporters are visible only in g1 cells. The absence of the g2 cells is indicated by arrows.

Anterior is at left and the pharynx is outlined. Scale bars represent 10 m.

69 created by Reece-Hoyes et al. (2009). There is thus a strong correlation between PGM1- dependent gene expression and hlh-6-dependent gene expression.

To confirm that loss of reporter expression was due to the hlh-6 mutation, I performed transgenic rescue with either genomic hlh-6 or an hlh-6 “minigene”. The genomic fragment contains hlh-6 and 2030 bp upstream of the ATG (including 840 bp of the upstream neighbour, T15H9.2) and 60 bp downstream of the predicted stop codon.

The minigene construct consists of 568 bp of promoter sequence fused to hlh-6 cDNA containing a synthetic intron, added to increase the expressivity of the construct (Figure

3.10). The 568 bp promoter fragment is only active in pharyngeal glands (Raharjo and

Gaudet, 2007), so the hlh-6 minigene is expressed only in pharyngeal glands. Both genomic and minigene versions of hlh-6 rescued phat-1 reporter expression in hlh-6 mutants to greater than 90% (Figure 3.10).

Together, the above three lines of evidence indicate that the bHLH transcription factor encoded by hlh-6 functions through PGM1. First, PGM1 activity depends on an E- box, the canonical binding site for bHLH transcription factors. Second, the expression patterns of hlh-6 and the PGM1 enhancer are identical. Third, hlh-6 is required for

PGM1-dependent reporter activity. Thus I proceeded to test whether this interaction is direct or indirect.

As with other bHLH proteins, HLH-6 probably functions as a dimer, most likely with the broadly-expressed Class I protein HLH-2 (Krause et al., 1997) since they are capable of interacting in vitro (Grove et al., 2009). I therefore tested whether HLH-6 (±

HLH-2) could bind to PGM1 in vitro using electrophoretic mobility shift assays (EMSA).

I found that HLH-2 alone could bind to PGM1 and that this binding was specific and

70 could be competed by wild type PGM1 but not PGM1 with a mutant E-box (Figure 3.12).

Since hlh-2 is broadly expressed and PGM1 is only active in the pharyngeal glands, it is unlikely that HLH-2 is the only trans-acting factor acting through PGM1. HLH-6 alone was not capable of binding to PGM1, but it could not be determined if HLH-6+HLH-2 could bind since an HLH-2 homodimer may not be distinguishable from an HLH-

6+HLH-2 heterodimer in this assay (Figure 3.12).

To determine if HLH-6 is sufficient in vivo for the activation of PGM1-dependent gene expression, I induced ectopic expression of HLH-6. hlh-6 cDNA was placed under the control of a heatshock promoter and used to create a transgenic line that also contained the PGM1/HLH-6 dependent reporter phat-1::YFP. Embryos from doubly transgenic adults were heat-shocked and then examined up to 7 hr later. No ectopic expression of the reporter was observed (Figure 3.13C). However, recent work indicated that ectopic expression of both members of a C. elegans bHLH dimer (HLH-8 and HLH-

2) is required for ectopic expression of target genes (Wang et al., 2006). I therefore created a transgenic line containing both hlh-6 and hlh-2 cDNAs under heat-shock inducible promoters and the phat-1::YFP reporter. After heat-shock of embryos from doubly transgenic adults, I observed occasional ectopic expression of the reporter (Figure

3.13D). However I observed a similar level of ectopic expression when hlh-2 alone was over-expressed (Figure 3.13B). Therefore, either ectopic expression of HLH-6 (± HLH-2) is not sufficient to activate ectopic expression of a gland-expressed marker or the heatshock constructs do not efficiently activate HLH-6 expression. This also suggests that an additional factor may be required to induce target gene expression and is supported by the evidence that PGM1 likely contains two transcription factor binding sites, one for a

Figure 3.12 HLH-2 can bind directly to PGM1 in vitro

Electrophoretic mobility shift assay of PGM1. (A) Recombinant HLH-2 and HLH-6 protein incubated in the presence of a probe from the phat-3 promoter. Lanes contain added protein as indicated. The band in lane 4 cannot be distinguished from the band in lane 2, therefore it is not conclusive whether or not HLH-6 can bind to PGM1 with HLH-

2, but HLH-2 can bind alone. (B) Recombinant HLH-2 protein incubated in the presence of a probe from the phat-3 promoter with the addition of the cold probes 3xPGM1,

3xPGM1Δflank and 3xPGM1ΔEbox. Addition of 10x (+) 3xPGM1 cold probe (lane 3) competes the band shift. Addition of 10x (+), 100x (++) or 1000x (+++) of

3xPGM1ΔEbox cold probe (lanes 4-6) cannot compete the binding of HLH-2 to PGM1 but addition of 10x (+), 100x (++) or 1000x (+++) of 3xPGM1Δflank cold probe (lanes

4-6) can.

Figure 3.13 Ectopic HLH-6 (+HLH-2) is not sufficient to activate ectopic gland reporter expression

Expression of phat-1::YFP in embryos subjected to heat shock treatment. phat-1::YFP is not normally expressed until late embryogenesis and is not normally visible in these earlier embryos. Relevant genotypes and conditions are: (A) ivEx44 [phat-1::YFP], heat shock; (B) ivEx44[phat-1::YFP]; ivEx93[hs::hlh-2], heat shock; (C) ivEx44[phat-

1::YFP]; ivEx43[hs::hlh-6], heat shock; (D) ivEx44[phat-1::YFP]; ivEx65[hs::hlh-2 hs::hlh-6], heat shock. Arrows indicate ectopic expression of phat-1::YFP. Anterior is at left, the pharynx and embryo are outlined. Scale bars represent 10 m.

75 bHLH transcription factor (the E-box sequence) and one for an unknown factor (the

YMAAY sequence). HLH-6 appears to require this additional non-bHLH factor that functions through the YMAAY sequence in order to activate gland genes in vivo and bind to PGM1 in vitro.

Given the success of the identification of the factor that acts through the E-box component of PGM1, I wanted to identify the factor that acts through the YMAAY component as well. The YMAAY consensus matches the known binding site for the mammalian POU domain-containing homeobox transcription factors Oct-1 and Oct-2

(Figure 3.15, Fletcher et al., 1987; Scheidereit et al., 1987). I therefore examined the expression of phat-1 in mutants for the C. elegans POU domain containing transcription factors unc-86, ceh-6 and ceh-18 (Latchman, 1999). However the expression of phat-1 was unaffected in any of the three mutant backgrounds (data not shown, S. Wen and B.

Grintuch personal communication). The YMAAY consensus also weakly matches the consensus binding site for PHA-4 (Gaudet and Mango, 2002). To test the possibility that

PHA-4 could act with HLH-6 to activate PGM1 expression, an enhancer that combined a generic PHA-4 binding site (3xPBSmix) with a PGM1 site that could not respond to HLH-

6 (3xPGM1ΔE) was made. Each of these enhancers alone is not sufficient to drive expression, but in combination they are capable of driving expression in the gland cells

(Ghai and Gaudet, 2008). This suggests that HLH-6 (+HLH-2) and PHA-4 could act together through PGM1 to activate gland expression. However, if this was the case, I would have expected the heat-shock overexpression of HLH-6 (+HLH-2) to activate expression in pharyngeal cells. It is possible that heat-shock inducible expression is not efficient in pharyngeal cells or that PHA-4 is not the factor acting through YMAAY.

Testing whether or not in vivo PHA-4 acts with HLH-6 would be complicated since

PHA-4 is thought to activate all pharyngeal genes and would therefore be indirectly involved in activating PGM1-dependent expression (Figure 3.14).

3.8 Summary

In this chapter I successfully used a computational approach to identify an regulatory element that is both necessary and sufficient for gland expression and further identified a transcription factor that acts through it, HLH-6, probably directly. Since

B0507.1 is only partially dependent on PGM1 and HLH-6, there must be other gland transcription factors and this possibility is addressed in Chapter 5. The success of this approach is likely based on the specificity of the set of genes that were computationally analyzed. All of the genes analyzed were expressed exclusively in the gland cells and all encode members of the same protein family (ShK domain containing proteins). However,

HLH-6 does regulate genes that do not encode ShK containing proteins, such as lys-8 and pqn-8. I also present evidence that HLH-6 (likely as a dimer with HLH-2) requires another factor that acts through the YMAAY sequence of PGM1, in order to activate gland expression. While this factor might be PHA-4, there is precedence for HLH proteins requiring additional factors for activity (see Chapter 6 for full details).

Mammalian Mash1, PTF1 and TAL1 require additional factors in order to bind to and activate gene expression (Beres et al., 2006; Castro et al., 2006; Cockell et al., 1989; Xu et al., 2003). My original question was, how are gland genes specified? I found that

HLH-6 regulates gland genes, but it appears the glands are still present in hlh-6 mutants suggesting hlh-6 does not specify the gland cell fate.

Figure 3.14 Model of gland gene regulation by HLH-6 and PHA-4

HLH-6 regulates the expression of the phat genes, likely directly, through the PGM1 binding site. PHA-4 activates expression of hlh-6 (Raharjo and Gaudet, 2007) and the phat genes (this work) also likely directly. I propose that PHA-4 might also act through

PGM1 to activate gland specific expression, rather than broad pharyngeal activation.

Since PHA-4 can activate PGM1 expression indirectly by activating HLH-6 the only way to determine if PHA-4 can act through PGM1 is by demonstrating direct binding.

Chapter Four: CHARACTERIZATION OF THE HLH-6 MUTANT PHENOTYPE

In the previous chapter, I identified a set of gland specific genes and found that they all (at least partially) require the PGM1 motif for expression. I demonstrated that the gland specific transcription factor hlh-6 was required for the expression of PGM1- dependent genes. However, some gland cells were still present in hlh-6 mutants. I therefore wanted to determine what had happened to the missing glands and also what effect the loss of hlh-6 would have on pharyngeal function. Previous work suggests three potential functions for the glands. First, Ohmachi, et al. proposed that the glands play a role in the feeding process because loss of expression of the gland-expressed gene kel-1 results in a starvation phenotype (Ohmachi et al., 1999). Second, during the four larval molts, the glands were observed to be secreting larger vesicles, suggesting a role in the molting of the pharyngeal cuticle (Singh and Sulston, 1978). Third, loss of function of the gland gene srf-3 results in resistance to bacterial infections that involve binding of bacteria to the cuticle (Hoflich et al., 2004), arguing for a role of the glands in bacterial infectious processes. HLH-6 and its targets may be involved in some or all of these functions.

4.1 Characterization of a Lethality Closely Linked to hlh-6

The hlh-6(tm299) strain I received from the Knock-out Consortium had a lethal phenotype, but data suggested that this lethality was not due to loss of hlh-6. The genomic rescue described in the previous chapter did not rescue the lethality suggesting the lethality was caused by a secondary mutation. I was able to map this lethality, using deficiencies in the region, to the left of hlh-6. I found that animals of the genotype let-x

80 hlh-6/mnDf85 were viable but let-x hlh-6/mnDf 67 animals were not. This places the lethality in a region of approximately 0.34 map units (from 0.95 to 1.29 m.u., Figure 4.1).

This lethality had to be outcrossed in order to examine the hlh-6(tm299) phenotype.

However I found that let-x had an interesting phenotype. I found that over 50% of the let- x mutants died by 16 hr after hatching and over 90% died by 36 hours after hatching

(n=41) (Figure 4.2). This lethality resembled a phenotype often called “rod-like larval lethal,” characterized by animals having a stiff appearance, which is often accompanied by the presence of vacuoles throughout the body (Figure 4.2). To test if this lethality was conditional, I grew let-x animals on plates without food. Normally, wild-type worms hatched in the absence of food arrest as L1 larvae and eventually starve to death after several days. Although not as healthy as wild type, let-x mutants grown in the absence of food survived much longer than those grown in the presence of food (Figure 4.2). The majority of let-x animals grown without food survived for >3 days and similar results were obtained when kanamycin-treated E. coli (Garigan et al., 2002) were provided as a food source. This lethality is not a specific response to E. coli, as let-x mutants exhibited the same early lethal phenotype when grown on B. subtilis, a non-pathogenic, gram- positive bacterial strain that is a potential food source of C. elegans in the wild.

I hypothesised that the bacteria might kill let-x mutants either by colonization/infection or by toxicity. To examine the possibility that E. coli infects or proliferates in let-x mutants, I fed these animals GFP-labeled E. coli (Mallo et al., 2002).

However, I saw no evidence for bacterial infection as the GFP+ bacteria were only visible in the pharyngeal lumen in numbers comparable to that seen in wild-type animals

(Figure 4.2). This suggested that some bacterial product was toxic to let-x animals.

Figure 4.1 let-x genetically lies between 0.95 and 1.29 on Linkage Group II

A scale diagram of Linkage Group II showing the extent of various deficiencies (Df) is shown. Genetically, hlh-6 is found at 1.6 m.u. and thus its mutant phenotype was not present in any of the deficiency heterozygotes shown. The extent of mnDf68 between

1.29 and 1.61 is not known (dashed line). Animals of the genotype mnDf67/let-x hlh-6 unc-4 are lethal but mnDf85/let-x hlh-6 unc-4 animals are not (nor do they display the hlh-6 phenotype described later), demonstrating that let-x lies within the extents of mnDf67 and mnDf85, approximately 0.34 m.u.

Figure 4.2 The lethality of let-x is due to the presence of live bacteria

(A) Viability of wild type and let-x mutants under various feeding conditions. The “WT on No Food” line (light blue) is masked by the “WT on Kan-OP50” line. (B-C) GFP- expressing E. coli in (B) wild type and (C) let-x mutants. Bacteria are only found in the pharyngeal lumen anterior to the grinder. let-x mutants exhibit vacuoles outside of the pharynx, prior to death. Scale bars represent 10 m.

Consistent with the involvement of a bacterially-produced toxin, the kanamycin treatment

(above) did not completely suppress lethality, as let-x mutants fed these dead bacteria eventually display some vacuoles prior to death and fail to reach adulthood. I therefore grew let-x/mC6g animals in an abiotic food source (Fahlen et al., 2005) and was able to recover viable, self-fertile let-x adults. The recovered let-x adults were transferred to standard E. coli-containing plates and quickly became sick. Progeny of these adults died as rod-like early larvae with vacuoles, verifying the parental genotype as let-x/let-x. The recovery of viable let-x adults from the abiotic media, but not from the other feeding experiments, supports the hypothesis that some bacterial product(s) are responsible for the death of let-x animals.

4.2 The g2 Gland Cells are Missing in hlh-6 Mutants

I initially examined hlh-6 mutants using the hlh-6 reporter to determine the effect of loss of hlh-6 on pharyngeal gland development. I used the hlh-6 reporter because in order to determine the fate of the glands in hlh-6 mutants I needed to use an HLH-6 independent reporter. Expression of hlh-6 is not critically dependent on hlh-6, though hlh-6 shows weak autoactivation (Raharjo and Gaudet, 2007). Importantly, during outcrossing, the hlh-6(tm299) mutation became linked to both rol-6(e187) and unc-

4(e120). Thus in all subsequent phenotypic analyses, the genotype of all hlh-6 mutants is rol-6(e187) hlh-6(tm299) unc-4(e120) and all control strains used were rol-6 unc-4. I found that an integrated hlh-6::YFP reporter is expressed in 100% of hlh-6 mutants

(Figure 3.10). However, in 84% of hlh-6 mutants (n=90), expression was observed in only three gland cells, rather than the expected five (Figure 3.11A). Based on the position

86 and morphology of expressing cells, it appeared that the three g1 glands (g1AR, g1AL and g1P) were present, while the two g2 cells were either missing or failed to express all gland reporters that were tested (hlh-6::YFP, B0507.1::GFP, lys-8::YFP, Figure 3.11A,

E, G).

The apparent absence of g2 glands in hlh-6 mutants could be explained by three possibilities: first, the g2 glands may undergo apoptosis, since during normal development the sister of g2 does (Sulston et al., 1983); second, the cells may be mis- specified and adopt an alternate fate, because pharyngeal cells in the absence of pha-4 adopt an ectodermal fate (Horner et al., 1998); third, the cells may persist as undifferentiated cells.

To test whether the g2 glands undergo apoptosis in hlh-6 mutants, I tested whether blocking apoptosis with a mutation in ced-3 would restore g2 glands. ced-3 is the key death promoting caspase in C. elegans and strong loss-of-function mutations in ced-3 result in the survival of all cells that normally undergo programmed cell death (Ellis and

Horvitz, 1986; Yuan and Horvitz, 1990). However, only 9% of hlh-6; ced-3 double mutants (n=32) expressed the hlh-6::YFP reporter in g2 cells, comparable to the expression in hlh-6 mutants (16%, see above) indicating that g2 glands are not restored by preventing apoptosis.

To address the possibility that g2 glands adopt an alternate cell fate, I performed nuclear counts in the back half of the posterior pharyngeal bulb, where the g2 cells are normally located. Nuclei were marked with a pha-4::GFP::HIS2B reporter, which is expressed in all pharyngeal nuclei (except for some pharyngeal neurons (Kalb et al.,

1998). There are 11 pharyngeal cells in this region (four muscles, three marginal cells,

87 three glands and one neuron), 10-11 of which express pha-4 post-embryonically

(expression in the pharyngeal neuron in the posterior bulb is variable). I expected that hlh-6 mutants would either have a wild type number of PHA-4-expressing cells or an average loss of ~1.6 such cells (because ~80% of hlh-6 mutants do not have visible g2 cells). There was a significant decrease in pha-4::GFP::HIS2B expressing cells between wild type and hlh-6 mutants (9.1 vs. 7.8, respectively, p<0.05, Figure 4.3), suggesting that the presumptive g2 cells do not express pha-4::GFP::HIS2B. I was not able to clearly visualize the expected number of nuclei (10-11), likely reflecting the error associated with counting nuclei that are close together. In order to confirm the result that the g2 cells lose their pharyngeal identity in hlh-6 mutants I also counted nuclei that express markers for the muscle, marginal and neural cells in this region. The myo-

2::GFP::HIS2B transgene (a gift from Dr. Susan Mango), which labels pharyngeal muscle cell nuclei (Okkema et al., 1993), labeled the same number of nuclei in wild type and hlh-6 mutants (3.6 vs 3.7, respectively, Figure 4.3). A pax-1::GFP::HIS2B transgene

(a gift from Dr. Susan Mango) is known to label pharyngeal marginal cells, but is also expressed in pharyngeal muscle cell 8 (Portereiko and Mango, 2001); thus I expected this transgene to label four nuclei in this region. I found that in both wild type and hlh-6 mutants, an average of approximately four nuclei were labeled by this marker (3.7 vs 3.6, respectively, Figure 4.3). Lastly, to visualize the pharyngeal neurons (seven of which are in the entire posterior bulb), I used the pan-neuronal marker rgef-1::GFP (Altun-Gultekin et al., 2001). Again, I found the number of cells labeled by this marker was unchanged in wild type versus hlh-6 mutants (6.6 vs 7.0, respectively, Figure 4.3). These results demonstrate that the presumptive g2 cells have not adopted an alternate pharyngeal

Figure 4.3 The g2 glands in hlh-6 mutants do not express any pharyngeal markers

Pharyngeal cell type-specific markers were examined to determine if the g2 cells had adopted an alternate pharyngeal cell fate. I used the pan-neuronal rgef-1::GFP marker to count pharyngeal neurons (expect 7 in wild type), myo-2::GFP::HIS2B to count pharyngeal muscle nuclei (expect 4 in the back half of the posterior bulb in wild type) and pax-1::GFP::HIS2B to count pharyngeal marginal cell nuclei (expect 4 in the back half of the posterior bulb in wild type, as pax-1::GFP is also expressed in the pm8 muscle) (Altun-Gultekin et al., 2001; Okkema et al., 1993; Portereiko and Mango, 2001).

I saw no change in the number of cells expressing these three markers in wild-type animals and hlh-6 mutants (6.6 vs 7.0 neurons, 3.6 vs. 3.7 muscles and 3.7 vs. 3.6 marginal cells, respectively). I also used the pan-pharyngeal marker pha-4::GFP::HIS2B count pharyngeal nuclei in the back half of the posterior bulb. I found a statistically significant difference between WT and hlh-6 mutants (9.1 vs. 7.8, respectively, p<0.05

Student’s t-test). Error bars are standard deviation. Numbers of animals score is indicated.

90 identity, consistent with these cells not expressing pha-4. Other types of pharyngeal nuclei were not affected in hlh-6 mutants (in particular, pm6 cells, which are lineally- related to the g2 glands), suggesting that the hlh-6 mutation specifically affects glands and does not act in the differentiation of other pharyngeal cell types, as expected given the gland-specific expression pattern of hlh-6. Interestingly, the morphology of the three g1 cells appeared unaffected by the loss of hlh-6.

The failure of the presumptive g2 cells to express any tested pharyngeal reporters implies that these cells may not be present in hlh-6 mutants. To explore this possibility,

Dr. Ralf Schnabel at the Technische Universität Braunschweig, Germany followed the lineages that give rise to g2 in hlh-6 mutant animals using a 4D microscope (Schnabel et al., 1997). In eight out of eleven cases (73%), the immediate precursor to the g2 cell

(MSnapapa, where n = a or p) failed to undergo its terminal division, but remained in its usual position within the embryo (Figure 4.4). In one case, the grandmother of g2 failed to divide. Such a lineage defect would prevent formation of one of the pm6 muscles, though I did not see a loss of pm6 cells in hlh-6 mutants. In the remaining two cases

(18%), the g2 precursor underwent its normal division. Thus, in 82% of cases, the g2 cell failed to be generated, consistent with the observation that 84% of hlh-6 mutants do not express hlh-6::YFP in g2 cells. Interestingly, PHA-4 expression is lost in the arrested g2 precursors, based on the counts of pha-4::GFP::HIS2B nuclei, yet PHA-4 must be normally expressed earlier in this lineage (i.e., in the g2 grandmother MSnapap), as no other pharyngeal cells (e.g., pm6 cells, which are cousins of the g2’s) were missing.

Formally, this result might indicate that hlh-6 is required for maintenance of pha-4 expression in g2 cells. However, the 10-20% of hlh-6 mutants that have g2 cells

Figure 4.4 The g2 glands are not generated in hlh-6 mutants

The lineages of the g2 glands in wild-type and hlh-6 mutants. MSn is used because both the MSa and MSp cell give rise to a g2 cell. If n = a, the g2L cell is made (as well as pm6VL and vpi2DL) and if n = p, the g2R cell is made (and pm6VR and vpi2DR). The sister cell of g2 cell undergoes apoptosis (X) in wild-type animals.

93 presumably express pha-4 in these cells, suggesting that pha-4 maintenance depends on a differentiated state. This hypothesis is corroborated by analysis of the tbx-2 mutant, where the loss of tbx-2 results in pharyngeal muscle cells arresting during their development. These cells initially express PHA-4 early, but eventually lose expression of pha-4 and do not adopt an alternate fate (Smith and Mango, 2007) similar to the g2 glands in hlh-6 mutants.

4.3 hlh-6 Mutants are Feeding Defective

In addition to a loss of gland gene expression and defects in g2 gland development, hlh-6 animals display a variety of characteristics that indicate a starvation phenotype: partially penetrant larval arrest and slow growth, smaller body size and decreased brood size among those surviving to adulthood. On average, 32% (n=105) of hlh-6 mutants arrest as L1 larvae. The anterior pharyngeal lumen of arrested larvae is stuffed with bacteria (Figure 4.5A, B), indicating a failure of these animals to properly transport food along the pharyngeal lumen. Animals that develop beyond the L1 stage also exhibit signs indicative of starvation. To examine the post-embryonic phenotypes of hlh-6 mutants I compared rol-6 hlh-6 unc-4 animals to the control strain rol-6 unc-4 (see

Materials and Methods for details). First, hlh-6 mutants are consistently smaller than wild-type worms of the same chronological age (Figure 4.5B), even when comparing gravid adults. Embryos laid over a one hour period by gravid adults were grown at 25ºC and scored every 24 hours. Upon hatching, hlh-6 mutants had approximately the same length as control strains, but at every other time point hlh-6 mutants were smaller. hlh-6 mutants also grow more slowly than control strains, taking more than twice as long to

Figure 4.5 hlh-6 mutants are slow growing

(A-B) The stuffed pharynx phenotype of hlh-6 mutants grown on OP50-GFP bacteria.

(A) NDIC image, (B) merged NDIC and fluorescence image. Anterior is at left and scale bars represent 10 μm. (C-E) Assays for growth defects in wild-type, hlh-6 mutants and hlh-6 mutants rescued by either the hlh-6 genomic fragment, the hlh-6 minigene or by using the HB101 strain of E. coli. (C) Graph of body length versus time, (D) time to reach adulthood and (E) brood sizes. For the hlh-6 mutants the L1 arrested animals are omitted. Error bars represent one standard deviation.

96 reach sexual maturity compared to controls. Embryos were collected as above, separated to individual plates and were scored as sexually mature when embryos were observed on the plate. hlh-6 mutants took 6.6 ± 1.7 days to reach adulthood (n=22) and the control strain took 3.1 ± 0.4 days (n = 22, Figure 4.5). As adults, hlh-6 mutants also have dramatically smaller broods, laying an average of 11.9 ± 15.4 eggs throughout their lifetime (n=21) compared to the control strain (116.5 ± 25.7 eggs, n = 22, Figure 4.5).

The control strain in all of the above cases is a rol-6 unc-4, which is not as healthy as the

N2 strain (explaining why the brood sizes are so low). All aspects of the hlh-6 mutant phenotype were rescued by either the hlh-6 genomic fragment or the “minigene” constructs described previously (Chapter 3.7). The hlh-6 genomic fragment rescued body length at all time points during development, decreased the time to adulthood to an average of 4.1 ± 0.5 days and increased brood sizes to an average of 75.5 ± 20.5 eggs

(Figure 4.5C-E). The minigene construct also rescued body length, time to sexual maturity and brood sizes (Figure 4.5 and data not shown). These results indicate that these phenotypes result from a loss of hlh-6 activity specifically in the pharyngeal glands.

The larval arrest, small size, slow growth and low brood size are all characteristic of starvation and are observed in other mutants that are feeding defective, such as the eat mutants and animals with abnormal pharynx morphology (Avery, 1993; Morck and

Pilon, 2006).

To further verify that hlh-6 mutants are starved, I stained animals with the lipophilic dye Nile Red. Although it was originally reported that Nile Red detects intestinal fat stores (Ashrafi et al., 2003) it has recently been demonstrated that Nile Red instead stains lysosomes within the intestine (O'Rourke et al., 2009; Schroeder et al.,

2007). Regardless, disrupted Nile Red staining is a feature common to many feeding defective strains (e.g., tph-1 (Ashrafi et al., 2003; Sze et al., 2000), pha-2 and pha-3

(Morck and Pilon, 2006)), validating the use of this stain as an indicator of feeding defects. I consistently observed increased Nile Red staining in hlh-6 mutants compared to control strains (Figure 4.6). These results suggest that hlh-6 mutants show a response to disrupted feeding by increasing the Nile Red staining lysosomes. However, because Nile

Red staining does not correlate with fat levels, the precise meaning of the increased staining remains unclear.

The starvation of hlh-6 mutants can be rescued by providing an alternate food source, suggesting that hlh-6 animal do not have impaired digestion of food or nutrient absorption. I hypothesized two explanations for the feeding defect in hlh-6 mutants.

Either there is a defect in the digestive process or there is a defect in the mechanical ability of hlh-6 mutants to move bacteria into the intestine for digestion. The stuffed pharynx phenotype of hlh-6 mutants suggests the latter. If the defect was digestive (i.e. the mutants are not capable of extracting nutrients from the bacteria), the hlh-6 mutants would be slow growing no matter what bacteria they were fed. Conversely if the defect was mechanical, feeding the mutants bacteria that more readily pass through the digestive tract would rescue the phenotype. C. elegans are usually grown by feeding with the E. coli strain OP50, but feeding with the strain HB101 can rescue the starvation phenotype of some eat mutants as it appears to be easier for C. elegans to eat (Avery and Shtonda,

2003). I found that hlh-6 mutants grown on HB101 were not starved, exhibiting wild type growth rates and a suppression of larval arrest (Figure 4.5 and data not shown). This suggests that the defect in hlh-6 mutants is not digestive since the two food sources

Figure 4.6 Nile red staining is increased in hlh-6 mutants

Fluorescence images of (A) wild type and (B) hlh-6 animals grown in the presence of

Nile Red. At least fifteen animals were observed for each genotype and one animal that represents the average level of fluorescence (as judged by exposure times) is shown.

Anterior is at left and the pharynx is outlined.

100 should pose the same problems with respect to digestion and nutrient absorption. Two factors that affect the ability of different food sources to rescue eat mutants are bacterial cell size and the relative “stickiness” of the cells (Avery and Shtonda, 2003). HB101 and

OP50 cells are approximately the same size (2.8 0.7 µm and 3.0 0.4 µm, respectively), but OP50 are more adhesive when manipulated with a worm pick compared to HB101 (as noted previously, Avery and Shtonda, 2003).

Most eat mutants, as well as other mutations that affect feeding, generally affect the rhythmic contractions of pharyngeal muscle, resulting in decreased or arrhythmic pharyngeal pumping and therefore “inefficient” feeding. Such mutations affect either pharyngeal muscle morphology and/or function (e.g. pha-2 and eat-2, Avery, 1993;

McKay et al., 2004) or the neurons that innervate the muscles (e.g. eat-4 and ceh-28, Lee et al., 1999; Ray et al., 2008). hlh-6 differs from other genes involved in feeding as hlh-6 functions in pharyngeal glands. Consistent with hlh-6 not acting in either pharyngeal muscle or neurons, I found that hlh-6 mutants had normal pharyngeal pumping with respect to both rate and rhythm of the muscle. L4 animals were collected and allowed to grow at room temperature overnight before scoring. Control animals (rol-6 unc-4) had an average of 169 ± 39 pumps per minute (n =20) and hlh-6 mutants (rol-6 hlh-6 unc-4) had an average of 156 ± 42 pumps per minute (n = 19). Likewise, peristaltic contractions of the pharyngeal isthmus (Fang-Yen et al., 2009) were also normal, with both control and mutant strains showing an average of one isthmus contraction per four pharyngeal pumps. These findings indicate that hlh-6 mutants are defective in some other aspect of food transport for which the glands are required.

101

4.4 Ablation of the Gland Cells Phenocopies the hlh-6 Mutant

Because some gland cells are still present in hlh-6 mutants, hlh-6 mutants might be only partially impaired with respect to gland activity. To examine the effect of complete loss of pharyngeal glands, I genetically ablated the glands using an hlh-6::egl-1 transgene, which activates expression of the pro-apoptotic gene egl-1 in pharyngeal glands (Conradt and Horvitz, 1998). Induction of egl-1 alone is sufficient to induce apoptosis in other cells, such as pharyngeal neurons (Rauthan et al., 2007). I assayed the presence or absence of glands using an integrated phat-1::YFP reporter and followed the presence of the hlh-6::egl-1 transgene with an intestine-specific mTomato marker (Shu et al., 2006). Preliminary results suggest that transgenic animals that lacked pharyngeal glands were viable but showed delayed growth and development, with 39% (n = 23) larval arrest, comparable to hlh-6 mutants. These results suggest that the pharyngeal glands of C. elegans are primarily involved in efficient feeding and that in the absence of hlh-6, glands are largely nonfunctional with respect to growth and fecundity.

4.5 A PHAT Protein Secreted by the Glands Binds to the Pharyngeal Cuticle

By analogy to foregut glands in other organisms, I postulated that pharyngeal glands could function in feeding by one of three ways: first, glands may secrete digestive enzymes required for efficient feeding; second, glands may produce secretions that coat food to ensure its passage along the lumen; third, glands may produce secretions that line the lumen and prevent adhesion of food. The first possibility, that the glands produce digestive enzymes, was suggested in part by the fact that the gland-expressed gene lys-8 is predicted to encode a lysozyme (Mallo et al., 2002). However, the ability of HB101

102 bacteria to rescue the starvation phenotype of hlh-6 animals suggests that glands are not required for digestion of food.

The other two possibilities, in which the glands lubricate the pharyngeal lumen, were suggested by the ability of a less sticky food source (HB101) to rescue hlh-6 starvation. As noted, the majority of gland-expressed genes are predicted to encode secreted proteins that contain multiple copies of the ShK domain. Interestingly, this family of proteins is similar to a group of gland-secreted mucins from the parasitic nematode Toxocara canis (Doedens et al., 2001; Gems and Maizels, 1996; Loukas et al.,

2000). The T. canis mucins are defined by a signal sequence and multiple copies of the

ShK domain (sometimes referred to as the SXC domain for six-cysteine motif) separated by Ser/Thr-rich stretches, which are probable sites of glycosylation. I found that, like the

T. canis proteins, the PHAT proteins contain stretches of Ser/Thr-rich sequence between their ShK domains (Figure 4.7) and many of these Ser/Thr sites are predicted to be sites for O-linked glycosylation (Julenius et al., 2005). The PHAT proteins may therefore function as mucin-like proteins.

A representative PHAT protein, PHAT-5, lines the pharyngeal lumen, consistent with the protein having a mucin-like function. I examined the subcellular location of

PHAT-5 using a phat-5::mCherry fusion expressed under the control of the hlh-6 promoter. The PHAT-5::MCHERRY fusion protein was visible in discrete puncta throughout the cell bodies of the glands, as well as along their extensions (Figure 4.8A).

In live animals, these puncta could be seen to traffic along the extensions, suggesting that the protein had been packaged into secretory vesicles. More importantly, the PHAT-

5::MCHERRY fusion protein was found along the lumen of the pharynx, indicating that

103

Figure 4.7 C. elegans PHAT-1 is similar to T. canis MUC-5

(A) C. elegans PHAT-1 and T. canis MUC-5 (Loukas et al., 2000) protein sequences showing predicted signal sequence (highlighted in yellow), ShK motifs (red) and Ser/Thr- rich tracts predicted to contain O-glycosylation sites (underlined). Signal sequences predicted using SignalP 3.0 (Emanuelsson et al., 2007). (B) PHAT-1 contains numerous predicted O-glycosylation sites that lie between the ShK motifs. Generated using the

NetOGlyc 3.1 server (Julenius et al., 2005).

104

105

Figure 4.8 PHAT-5::mCherry localizes to the pharyngeal lumen

Fluorescence and NDIC images of (A-C) wild-type and (D-F) hlh-6 animals expressing the hlh-6::PHAT-5::mCherry translational fusion construct. (B) and (C) are close-ups of animal shown in (A). (E) and (F) are close-ups of (D). Fluorescence and NDIC images of

(G) wild-type and (K) hlh-6 animals expressing the myo-2::PHAT-5::mCherry translational fusion construct with corresponding close-ups in (H) and (I) and (L) and

(M). (J) and (N) are the intestines of the animals from (G) and (K) respectively. (O) and

(P) are NDIC and fluorescence images of wild type animals expressing lys-8::PHAT-

5::YFP observed during the L1 to L2 molt. The asterisk indicates fluorescence attached to the L1 cuticle as it is being expelled out of the buccal cavity. Arrowheads indicate the pharyngeal lumen, arrows mark the processes of the g1 glands and triangles mark the boundary of the pharyngeal cuticle. Anterior is at left and the pharynx is outlined. Scale bars represent 10 m.

106

107 the protein had been secreted from the glands (Figure 4.8A-C). The fusion protein had a discrete anterior boundary, extending as far as the cheilostom groove in the buccal cavity

(Figure 4.8B-C), the boundary between the epidermal cuticle and the pharyngeal cuticle

(Wright and Thomson, 1981), suggesting that PHAT-5 is specifically associated with pharyngeal cuticle. In addition, PHAT-5 fusion protein remained associated with shed pharyngeal cuticle, arguing that the protein forms part of the lining of the pharyngeal lumen (Figure 4.8O-P). No protein was seen to co-localize with bacteria in the pharynx lumen, suggesting that PHAT-5 does not coat food particles.

To investigate whether the glands of hlh-6 mutants are functionally impaired, I examined whether PHAT-5::MCHERRY could be secreted by the glands of hlh-6 mutants. phat gene expression is absent from hlh-6 animals, so I expressed the PHAT-5 fusion under the control of the hlh-6 promoter, which remains active in hlh-6 mutants.

The hlh-6::phat-5::mCherry construct was expressed in pharyngeal glands, but no protein was seen at the pharyngeal lumen, likely reflecting a functional defect in the hlh-6 glands (Figure 4.8D-F). Punctate signal was observed in the gland ducts and in live animals these puncta appeared to migrate along the ducts as in wild type, suggesting that vesicles were still present and capable of being transported within the glands. The hlh-6 mutants are therefore defective either in secretion of the PHAT-5 protein or in retention of this protein at the pharyngeal lumen or both. To distinguish between these possibilities

I expressed PHAT-5::MCHERRY in pharyngeal muscles (using the myo-2 promoter,

Okkema et al., 1993) to investigate the localization of PHAT-5 independent of gland function. In wild type animals, pharyngeal muscle could secrete PHAT-5::MCHERRY.

Signal was seen lining the pharyngeal cuticle in addition to puncta throughout the

108 muscles (Figure 4.8G-J). In hlh-6 mutants, some signal was visible on the luminal surface, but I also observed significant signal in the intestinal lumen (though not associated with cell surfaces), which was not observed in wild type animals (Figure 4.8K-

N). This result suggests that while PHAT-5::MCHERRY can associate with the pharyngeal cuticle in hlh-6 animals, this association is less stable, resulting in the movement of the fusion protein along the digestive tract. This observation is consistent with the hypothesis that the pharyngeal lining is defective in hlh-6 mutants, possibly due to the absence of other gland-secreted proteins, including the other PHAT proteins. In addition, in hlh-6 mutants the glands may not be capable of properly secreting.

Interestingly, no rescue of the hlh-6 phenotype by either hlh-6::phat-5::mCherry or myo-

2::phat-5::mCherry was observed. These results suggests that the major defect in hlh-6 mutants is not rescued by the addition of PHAT-5 protein. However given that the ectopic PHAT-5 protein in hlh-6 mutants does not properly adhere to pharyngeal cuticle this hypothesis cannot be confirmed. It is also possible that more than one PHAT protein needs to be present in order for functionality to be restored, or that some other hlh-6 dependent protein is required.

4.6 Summary

The results described in this chapter demonstrate a clear role for hlh-6 in the development and function of the pharyngeal glands. Although hlh-6 does not specify the fate of the glands as a whole, it does play a role in the development of the g2 gland cells.

Without hlh-6, the g2 cells do not divide properly and arrest as presumably undifferentiated cells that have lost their pharyngeal identity (as indicated by the loss of

109

PHA-4 expression). I additionally found that hlh-6 plays a role in the feeding process.

One interesting possibility is that the normal function of PHAT proteins is to line the lumen and lubricate it. Supporting this model, I demonstrated that hlh-6 mutants have a defective pharyngeal lumen (either because of non-functional PHAT proteins or the loss of some other proteins) and are rescued by HB101 bacteria. This model is also supported by the observation that the PHAT proteins are similar to nematode mucins, which are known to act as lubricants. The observation that the glands increase their secretions during molting (Singh and Sulston, 1978) likely reflects the need for the newly synthesized cuticle to be properly lined before molting has completed. Because hlh-6 is not required for the development or specification of all the glands, how these cells are specified remains unclear. One way to investigate this question is to identify other regulators of gland gene expression, which is the subject of the next chapter, and/or to identify regulators of hlh-6 expression (Ghai and Gaudet, 2008; Raharjo and Gaudet,

2007).

110

Chapter Five: IDENTIFICATION OF HLH-6-INDEPENDENT CIS- REGULATORY ELEMENTS

In the previous two chapters I showed that hlh-6 was responsible for multiple aspects of gland development and function. HLH-6 activates the expression of many gland genes, specifies the g2 cells fate and possibly aids in the transportation of food through the pharyngeal lumen. However, hlh-6 mutants still possess the three g1 gland cells, suggesting there are other transcription factors involved in gland specification and development. Additionally, the reporter for B0507.1 is only partially dependent on PGM1 and hlh-6 so there is some other transcription factor that activates B0507.1 in the gland cells. Thus I planned to identify other gland transcription factors using a similar approach as in Chapter Three. I first identified gland genes that were potentially independent of hlh-6 based on the absence of PGM1 in their promoters. After confirming that the gland expression of these genes was completely independent of hlh-6, I then analyzed the promoters using a deletion series for regions of necessity. Surprisingly I found that two hlh-6-independent gland genes are regulated by a motif that activates hlh-6 expression:

HRL3 (hlh-6 regulatory element 3). This approach has provided some insight into the gland transcription factor network. These findings also raised the question why different gland genes are regulated differently. That is, why are all gland genes not regulated by one factor? Is there any logic to the separation of gland genes regulated by hlh-6 and those regulated by other factors?

111

5.1 Identification of Gland Expressed Genes Without PGM1

As a first step to identify new gland gene regulators, I searched for more genes expressed in the glands. Extensive searches of previously published papers identified 14 genes with reported expression in the glands (including kel-1 and srf-3, Table 5.1).

Searches of the BC C. elegans Gene Expression Consortium database

(http://elegans.bcgsc.ca/home/ge_consortium.html) identified 14 gland expressed genes and searches of Ian Hope Laboratory’s expression database

(http://bgypc059.leeds.ac.uk/~web/databaseintro.htm) identified 5 genes. Genes reported as expressed broadly in the pharynx in addition to the glands were omitted (unless the gland expression was stronger than the surrounding tissue). Lastly, searches through Yuji

Kohara’s in situ expression database (performed by Dr. J. Gaudet, Kohara, 2001a;

Kohara, 2001b) identified 16 genes whose in situ hybridization pattern appeared to be gland specific. Three genes (mig-23, klc-2 and pqn-75) had expression data available from more than one of these four sources, yielding a total of 46 genes.

These reported gland expressed genes were then sorted based on the presence or absence of the hlh-6 dependent motif PGM1. I searched the 5’ sequences of these 46 genes from the ATG to the next upstream gene. If a PGM1 sequence was not identified then the introns of the gene were also searched (two genes had a PGM1 in an intron, sams-4 and kel-1). These searches were performed by inspection for a

CAnnTGnnYMAAY consensus, which is a less stringent version of the PGM1 consensus sequence (see Figure 3.7), in order to cover all possible occurrences of functional PGM1 sites. However other deviations from this consensus might be functional PGM1 sites. For example, four promoters (H27A22.1, npp-15, Y110A2AL.2 and ztf-11) carry the

112

Table 5.1 Gland expressed genes

Gland expressed genes were identified from one of four sources: a previously published paper (as indicated), the BC C. elegans Gene Expression Consortium database

(BC), Ian Hope Laboratory’s database (Hope), or Yuji Kohara’s in situ database (YK, see text for details). Reporters for genes in bold were either obtained or created by myself and the expression pattern found by my reporter is also indicated in bold. T05B5.13 not included in this list but was analyzed further (see text). Searches for PGM1 include all sequence 5’ from the ATG to the next upstream gene. Abbreviations for expressions are as follows: pha = pharynx, pha mc = marginal cells, int = intestine, bwm = body wall muscles, neurons = any non-pharyngeal neurons, vpi = pharyngeal-intestinal valve, hyp = hypodermis and/or seam cells, dtc = distal tip cell, exc = excretory cell, “?” indicates unclear data.

Gene Protein Expression Data PGM1?

B0280.7 unknown gland (BC), gland no chaperonin complex cct-4 gland, pha, int (BC) CAnnTGnnYMAAG component Ca2+ binding EGF- gland? (YK), pha C08G9.2 no like domains muscle (Hope) gland, pha, bwm, C17G10.2 heat shock protein no vulva (BC) C47D12.8 endonuclease gland, int (BC) no neurons (BC), neuron, int, rectal, pha neuron, dop-4 dopamine receptor yes g1P? (Sugiura et al., 2005) gland, maybe pha D2085.5 unknown yes muscle (BC) gland (Zheng et al., F08A8.5 fucosyltransferase yes 2008) F15A4.6 ShK domain gland (YK) yes

113

cysteine proteinase F15D4.4 gland (YK) yes cathepsin L gland, vpi, int, hyp, F26A1.8 proteinase inhibitor CAnnTGnnYMAAA (BC) gland, pha, exc, F48E3.3 glucosyltransferase no neurons (BC) F54F11.1 similar to hydrolase gland? (YK) yes gland, neurons (Wang F56B6.6 unknown yes et al., 2006) N-acetylglucosaminyl g2 glands (Warren et gly-15 no transferase al., 2001) H27A22.1 guanylate cyclase gland, pha, hyp (BC) CAnnTGnYMAAY BTB/POZ- and Kelch g1 gland cells kel-1 yes, intron motif (Ohmachi et al., 1999) nucleoside gland, pha, int, hyp mig-23 no phosphatase (BC) M04G7.1 ShK domain gland? (YK) yes molybdenum cofactor gland, int, bwm, dtc moc-1 CAnnTGnnYMAAA biosynthesis (BC) ShK domain, gland, int (BC), gland, nas-12 CAnnTGnnYMAAA metalloprotease int, spermathecal valve, nex-1 annexin gland? (Daigle and yes Creutz, 1999) gland, spematheca nuclear hormone nhr-48 (Gissendanner et al., no receptor 2004), weak pha gland, seam, vulva nuclear hormone nhr-60 (Simeckova et al., no receptor 2007) nuclear hormone gland, neuron, int nhr-108 yes receptor (Hope) nuclear pore complex gland or pha (BC), npp-15 CAnnTGnYMAAY Protein neuron pqn-25 kinase gland, int (YK) yes prion-like-(Q/N-rich)- gland, weak int, gonad pqn-75 yes domain (YK), gland (Hope) hyp, vulva, neuron, rcn-1 calcineurin inhibitor pha mc, gland? (Lee et yes al., 2003) S- gland?, other body sams-4 adenosylmethionine yes, intron (YK) synthetase scl-3 SCP protein, gland (YK), gland no

114

extracellular SCP protein, scl-5 gland (YK) yes extracellular gland, hyp, UDP-galactose srf-3 spermatheca (Hoflich no transporter et al., 2004) gland, spematheca tag-312 Ras GTPase yes (Hope) T22B11.4 unknown pha/gland (YK) yes ubiquinol cytochrome ucr-2.3 gland? (YK), neuron no c reductase g1 glands, hyp wrt-3 warthog-like (Aspock et al., 1999), CAnnTGnYMAAA gland, hyp CAnnTGnYMAAY, Y110A2AL.2 unknown g2 glands (YK) CAnnTGnnYMAAA tRNA deaminase Y47D3A.14 gland? (YK), pm7 no subunit Y54G2A.4 sugar transporter gland, int (BC) CAnnTGnnYMAAA similar to Y62F5A.9 gland? (YK) yes Helicobacter antigen gland, int (BC), gland, Y76B12C.3 selenoprotein no weak pha, weak int Y8A9A.2 thrombospondin-like gland (YK), g1P no everywhere (embryo), zinc finger CAnnTGnYMAAY x3 ztf-11 gland and 2 neurons transcription factor CAnnTGnnYMAAR x4 (larva, Hope) CLIP associating ZC84.3 neuron, gland? (paper) yes protein gland (YK), gland, ZK596.1 unknown no weak int, neuron

115 consensus CAnnTGnYMAAY and four (F26A1.8, moc-1, Y110A2AL.2 and Y54G2A.4) contain CAnnTGnnYMAAA. Out of the 46 genes reported as gland expressed, 21 had a recognizable PGM1 and 25 did not. I also re-evaluated my original list of gland genes

(Table 3.1) and found that one gene, T05B4.13, did not have an identifiable PGM1 site and so was included for further analysis (below). Genes whose promoters lack PGM1 are candidate hlh-6-independent genes and promoter analysis of these genes could therefore identify additional gland specific cis-acting motifs.

I next verified whether the reported gland expression was accurate for some of these genes. Out of the 25 genes that did not have a recognizable PGM1 within their 5’ upstream region or introns, twelve were randomly chosen for validation. I obtained reporters for two of the twelve genes, nhr-48 and wrt-3, which were previously published

(Aspock et al., 1999; Gissendanner et al., 2004), while I made transcriptional reporters for the other ten genes by “PCR stitching” (see Materials and Methods). Four of the twelve tested reporters had either very weak expression in the glands (nhr-48) or were not expressed in the glands at all (but were expressed elsewhere, npp-15, ucr-2.3 and

Y47D3A.14) and were therefore excluded from further analysis. The lack of gland expression for these three genes likely reflects misidentification of expression from in situ (ucr-2.3 and Y47D3A.14) or incorrect annotation in the databases (npp-15). The remaining eight reporters had strong, consistent expression in the pharyngeal glands

(Figure 5.1). B0280.7, scl-3 and Y8A9A.2 were expressed exclusively in the pharyngeal glands. Interestingly, Y8A9A.2 seemed to be expressed only in the g1P cell (see Chapter

5.3 below). The other five genes (nas-12, T05B4.13, wrt-3, Y76B12C.3 and ZK596.1)

116

Figure 5.1 Some gland expressed genes without PGM1 are dependent on hlh-6

Expression of gland reporters in wild type (left column) and hlh-6 mutants (Right column). B0280.7 (A, B) and T05B4.13 (G, H) reporters are completely inactive in hlh-6 mutants. scl-3 (E, F) and ZK596.1 (O, P) reporters were partially affected by the loss of hlh-6 while nas-12 (C, D), wrt-3 (I, J), Y76B12C.3 (K, L) and Y8A9A.2 (M, N) reporters are completely unaffected in the hlh-6 mutant.

117

118 were expressed in glands as well as other non-pharyngeal tissues (the intestine for example, see Table 5.1 for details).

I then tested whether the expression of these eight reporters is independent of hlh-

6 by examining their activity in hlh-6 mutants. If these genes do not require hlh-6, then I would expect that the reporters would still be expressed in the hlh-6 mutants.

Surprisingly, two of the eight reporters (B0280.7 and T05B4.13) were not expressed in the gland cells of hlh-6 mutants (Figure 5.1A,B, G, H and Figure 5.2) but still retained expression in other tissues, such as vulva muscles and the distal tip cell for T05B4.13

(data not shown), suggesting that only the gland expression of the reporters was affected.

The other six reporters showed only partial loss of activity (scl-3 Figure 5.1E,F and

ZK596.1 Figure 5.1O,P) or appeared unaffected (nas-12 Figure 5.1C,D, Y76B12C.3

Figure 5.1K.L, Y8A9A.2 Figure 5.1M,N and wrt-3 Figure 5.1I,J) in hlh-6 mutants. The finding that four of the reporters were partially or wholly dependent on hlh-6 for gland expression, despite not having a PGM1 site, suggests either that my current definition of the hlh-6 response element is not completely accurate or that hlh-6 indirectly regulates the expression of these genes through a second transcription factor.

5.2 HLH-6 Dependent Genes Without PGM1

In order to determine whether or not hlh-6 might directly regulate the expression of the four gland genes that do not have PGM1 in their promoters but are still hlh-6 dependent (B0280.7, T05B4.13, scl-3, ZK596.1), I analyzed the promoter of one:

T05B4.13. A reporter for T05B4.13 is expressed in all five pharyngeal gland cells throughout all larval and adult stages (Figure 5.1G), as well as in vulval muscles during

119

Figure 5.2 Expression of gland reporters in the hlh-6 mutant

Quantitation of the data presented in Figure 5.1 showing the percentage of transgenics with gland expression in wild type and hlh-6 mutant backgrounds for the reporters indicated. Numbers of animals scored are as follows (wild type and hlh-6 respectively):

B0280.7 (n = 30, 41), nas-12 (n = 62, 58), scl-3 (n = 56, 53), T05B4.13 (n = 46, 49), wrt-

3 (n = 32, 40), Y76B12C.3 (n = 56, 21), Y8A9A.2 (n = 51, 37) and ZK596.1 (n = 48, 54).

120

121 the L4 and adult stages and the distal tip cell during the third and fourth larval stages. The reporter includes all intergenic sequence between T05B4.13 and T05B4.4, the next upstream gene (338 bp) but does not contain any matches to PGM1

(CAnnTGnnYMAAY). However its gland expression is dependent on hlh-6. One model to explain the dependence of T05B4.13 (and others) on hlh-6 despite not having a known hlh-6 binding site in its promoter is that it actually has a binding site and the

CAnnTGnnYMAAY consensus is too strict. In fact, the promoter of T05B4.13 contains one close match to PGM1, differing by only one base pair: CAtcTGttCCAGC (Figure

5.3). Given the hypothesis that the factor acting through YMAAY might be PHA-4, another explanation for the dependence of these genes on hlh-6 is that the HLH-6

(CAnnTG) and PHA-4 (YMAAY) sites are separated in these promoters. This hypothesis is supported by the fact that these two sites can be separated in an enhancer assay and still have activity in the glands (see Chapter 3.7). A third possibility is that HLH-6 is capable of acting with factors other than the factor acting through YMAAY. Lastly, HLH-6 could be activating the expression of these genes indirectly.

In order to test whether HLH-6 might act directly on T05B4.13, I mutated the E- box sequences in its promoter. This promoter contains 2 E-boxes (CAnnTG) separated by six base pairs (Figure 5.3). Site directed mutagenesis of both E-boxes resulted in a complete loss of gland expression (Figure 5.3), with 0% (n = 35) of transgenic animals showing gland expression. Given that this reporter is hlh-6 dependent, this result argues that hlh-6 may directly activate gland expression of T05B4.13. If so, then either the mismatched PGM1 is functional or hlh-6 can function through non-PGM1 E-boxes.

Mutating the sequences flanking the E-boxes (especially the CCAGC flanking sequence)

122

Figure 5.3 E-boxes are required for the gland expression of T05B4.13

A diagram of the promoter of T05B4.13 is on the right and expression of wild type (A) and mutant E-box promoters (B) is on the left. The sequences mutated are underlined.

The sequence that differs from PGM1 at one base pair is highlighted (in the reverse orientation). The black rectangles represent the E-boxes. Anterior is at left and the pharynx is outlined. Scale bars represent 10 m.

123

124 would test whether the deviant PGM1 is functional or whether a more distal element acts together with the E-box to drive gland expression. One possibility is that HLH-6 acts with a PHA-4 binding site in order to activate gland expression, since PHA-4 might be the factor acting through the YMAAY component of PGM1. A sequence matching the in vivo defined PHA-4 binding site (TGTBTSY, discussed further in Chapter 5.3, Zhong et al., 2010) is present in the T05B4.13 promoter (TGTGTGC). The B0280.7, scl-3 and

ZK596.1 promoters were not functionally analyzed but it is possible that they also are directly activated by HLH-6. Additionally these three promoters also have consensus

PHA-4 binding sites in their promoters.

5.3 Regulation of nas-12

Identification of cis-acting elements in the promoters of the four hlh-6 independent gland genes (nas-12, Y76B12C.3, Y8A9A.2 and wrt-3) would aid in the identification of new gland transcription factors. The first hlh-6 independent gene I analyzed was nas-12. The protein encoded by nas-12 is predicted to be a metalloprotease similar to astacin (Mohrlen et al., 2003), which also contains two ShK domains. The astacin family of proteases contains members that generally act as either digestive enzymes (such as astacin, Titani et al., 1987) or peptide processing enzymes (such as

BMP1, Reddi, 1996). In C. elegans, astacin family members have been shown to be collagen processing enzymes (Novelli et al., 2004), hatching enzymes (Hishida et al.,

1996) and enzymes involved in molting (Harris et al., 2004; Suzuki et al., 2004). A sub- group of C. elegans astacins all contain ShK domains (nas-6 to nas-15, Mohrlen et al.,

2003) and recently mutations in nas-6 and nas-7 were shown to result in slow larval

125 growth and pharyngeal cuticle defects (both are expressed in pharyngeal muscle and marginal cells as well as the intestine, hypodermis and body wall muscles, Park et al.,

2010). nas-12 was discovered as a gland expressed gene from the BC C. elegans Gene

Expression database (also recently published in Park et al., 2010). By both their reporter and mine, nas-12 is expressed in all five gland cells as well as in the intestine and hindgut

(Figure 5.1C).

Deletion analysis of the nas-12 promoter identified a region of 24 base pairs that is required for gland expression. The initial promoter fragment used for nas-12::GFP reporters was 406 bp of sequence 5’ to the start codon (Figure 5.4A). This included all sequence upstream to the next coding gene. Deletions up to -132 bp have no effect on the pattern, or relative levels of gland expression (Figure 5.4A and data not shown). In order to identify sequences that might be important for expression of this gene, I analyzed the conservation of this 132 bp region with other closely related Caenorhabditis species.

Using the UCSC Genome Browser (http://genome.ucsc.edu) I was able to identify the sequences of the nas-12 homologs in the genomes of four other sequenced

Caenorhabditis species (briggsae, brenneri, remanei and japonica). Alignment of these sequences showed multiple regions that were conserved across all five species (Figure

5.5). I therefore created two more deletion constructs (-118 and -93 bp) to determine which conserved region was responsible for expression. The -118 construct had normal expression, while a promoter fragment that only contained 93 bp of upstream sequence had expression in only 6% (n=34) of transgenic animals (Figure 5.4A). This suggests that there is an important cis-acting element within the -118 to -93 bp region.

126

Figure 5.4 A HRL3-like site is necessary and sufficient for the gland expression of nas-12

(A) Scale diagram of the nas-12 promoter and the various deletion constructs.

Percentages of transgenic animals with gland expression are indicated to the right of the deletions scored. The sequence between -118 and -93 bp is shown and the candidate

PHA-4 binding site (PBS) is highlighted. The mutations in the PBS and its flanking sequence (underlined) result in the complete loss of reporter activity. (B) A fluorescence micrograph of the enhancer (sequence indicated) displaying gland specific expression.

Anterior is at left and the pharynx is outlined. Scale bar represents 10 m.

127

128

Figure 5.5 The HRL3-like site in the nas-12 promoter is conserved among nematodes

The sequence of nas-12 from -132 to -1 bp and the corresponding sequences from C. remanei (Crem), C. briggsae (Cbri), C. brenneri (Cbre) and C. japonica (Cjap).

Sequences were obtained from the UCSC Genome Browser (http://genome.ucsc.edu).

The necessary candidate PHA-4 binding site is highlighted.

129

130

Within this region is a conserved sequence that resembles a DAF-12 binding site and a previously described cis-element from the hlh-6 promoter that contains a PHA-4 binding site and is called HRL3 (Ghai and Gaudet, 2008). Within the 24 bp necessary region of nas-12, there was a perfectly conserved site that matched a published binding site for the transcription factor DAF-12 (TGTGTG, Ao et al., 2004). DAF-12 is a steroid hormone receptor that is mainly involved in dauer formation, a long lived larval stage in C. elegans

(Antebi, 2006). In order to determine if nas-12 responds to DAF-12, I examined the nas-

12 reporter in daf-12 mutants. I found that the -118 bp nas-12 reporter was still expressed in the glands of daf-12 mutants (data not shown). The conserved sequence also resembles the PHA-4 binding consensus TRTTKRY, differing by only one base pair (TGTGTGC).

Recently published data using Chromatin Immunoprecipitation assays on PHA-4 protein actually redefined the consensus site bound by PHA-4 in embryos in vivo to TGTBTSY

(Zhong et al., 2010), which matches the TGTGTGC site perfectly. Consequently I found that a mutation in this putative PHA-4 site of the -118 bp nas-12 promoter completely eliminated expression (Figure 5.4A). The sequence flanking the PHA-4 site in nas-12 also looked similar to the sequence flanking a PHA-4 binding site in the promoter of hlh-

6 (Figure 5.6). Mutations in the sequence immediately 3’ to a PHA-4 binding site in the hlh-6 promoter were capable of abolishing hlh-6 reporter expression (Ghai and Gaudet,

2008). This result suggested there is a cis-acting element right next to or overlapping the

PHA-4 site and it was called HRL3 (hlh-6 regulatory element 3). Ghai and Gaudet demonstrated that this was a distinct cis-acting element and not simply an extended binding site for PHA-4 by separating the PHA-4 site and the HRL3 site in an enhancer assay. I mutated the sequence flanking the conserved PHA-4 binding site in nas-12 and

131

Figure 5.6 Alignment of the HRL3-like sequences

The HRL3 like sequences from hlh-6, nas-12, Y8A9A.2 and Y76B12C.3 are aligned. The core PHA-4 binding site is highlighted. N = A, G, T or C, K = G or T, R = A or G, Y = T or C, W = A or T, M = A or C.

132

133 found that this too completely abolished gland expression (Figure 5.4A), suggesting that the gland expression of nas-12 is regulated by the HRL3 sequence. Recently, the

TGTGTGC PHA-4 consensus binding site was shown to have minimal activity on its own (Raharjo et al., 2010) suggesting that in the nas-12 promoter there must be synergy between PHA-4 and the HRL3 factor.

Since the HRL3 site in nas-12 is necessary for its gland expression, I next asked whether this occurrence of HRL3 was also sufficient for gland expression. Ghai and

Gaudet found that in an enhancer assay, the HRL3 site from hlh-6 was capable of activating expression in the glands, but also weakly throughout the entire pharynx (Ghai and Gaudet, 2008). Indeed, I found that three tandem copies of the HRL3 sequence from nas-12 placed upstream of the "promoter-less" reporter was sufficient to activate expression exclusively in the pharyngeal glands of 100% (n=40) of transgenic animals

(Figure 5.4B). Interestingly, this enhancer did not activate expression broadly in the pharynx as might be expected for a PHA-4 binding site. Therefore this site is likely a weak PHA-4 binding site (as mentioned previously, Raharjo et al., 2010) that must act cooperatively with the HRL3 binding site in order for strong activity. These results indicate that the conserved element in the nas-12 promoter is in fact a functional HRL3 site. Thus, although nas-12 is not regulated by hlh-6, both hlh-6 and nas-12 are regulated by the same factor through their HRL3 sites. This raises the question of why nas-12 is not regulated by hlh-6. Or conversely, why does hlh-6 regulate some gland genes but not others?

134

5.4 Regulation of Y8A9A.2

The second hlh-6 independent gene I analyzed was Y8A9A.2. The protein encoded by Y8A9A.2 contains multiple thrombospondin type repeats. Thrombospondins are multimeric glycoproteins that generally function as regulators of cell-cell interactions

(Bork and Beckmann, 1993). Some thrombospondin family members are expressed in the extracellular matrix and promote neural adhesion (Feinstein et al., 1999). This is intriguing given that the gland extensions come into contact with pharyngeal neuron axons suggesting that the two processes might interact with each other (discussed in

Chapter 6.3, Albertson and Thomson, 1976). Y8A9A.2 was initially discovered as a gene whose in situ hybridisation pattern appeared to be gland specific (Kohara, 2001a; Kohara,

2001b). Interestingly, both by in situ and reporters, Y8A9A.2 appeared to be expressed only in the g1P cell (Figure 5.1M). Closer examination revealed that the Y8A9A.2 reporter was exclusively expressed in the g1P cell in young larvae: in 62% of young larvae (L1 and L2, n = 45), the Y8A9A.2 reporter is expressed only in the g1P cell (and is expressed in additional gland cells in the other 38%). By adulthood, only 21% (n=63) of animals show g1P-specific expression (Figure 5.7). These results suggest that Y8A9A.2 is both spatially and temporally regulated.

Deletion analysis of the Y8A9A.2 promoter identified a region of 26 base pairs that is required for expression. The initial promoter fragment used for Y8A9A.2::GFP reporters was 2.2 kb of sequence 5’ to the start codon (Figure 5.8), including all sequence up to the next coding gene (fbxb-99). This promoter is expressed specifically in the g1P cell from the 3-fold stage to late larval stages, when it becomes expressed in all five gland cells. Deletions up to -276 bp of this promoter have no effect on the pattern, levels

135

Figure 5.7 Y8A9A.2 expression changes during larval development

Fluorescence photomicrographs of Y8A9A.2 reporter expression in wild type (A) L1 and

(B) adult animals showing the change from g1P specific expression to broad gland expression. Anterior is at left and the pharynx is outlined. Scale bar represents 10 m.

136

137

Figure 5.8 A HRL3-like site is necessary for Y8A9A.2 expression

Scale diagram of the Y8A9A.2 promoter and the various deletion constructs. Percentages of transgenic animals with gland expression are indicated to the right of the deletions scored. The sequence between -196 and -170 bp is shown and the candidate PHA-4 binding site (PBS) is highlighted. The mutations in the PBS and its flanking sequence

(underlined) result in the complete loss of reporter activity.

138

139 or timing of expression (Figure 5.8). Again I analyzed the conservation of this 276 bp region with homologs of Y8A9A.2 in other closely related Caenorhabditis species and found there were multiple regions that were conserved across all five species (Figure

5.9). Thus, I created three more deletion constructs (-254, -196 and -170 bp) to determine which conserved region was responsible for proper expression. The -254 and -196 constructs had normal expression patterns and levels. However a promoter fragment that only contained -170 bp of upstream sequence was not expressed at all (Figure 5.8). This suggests that there is a cis-acting element within the -196 to -170 bp region but does not exclude the possibility that there are additional elements within the -170 to -1 bp region.

Similar to nas-12, a HRL3 promoter element is also necessary for the gland expression of Y8A9A.2. Within the -196 to -170 bp region there is a conserved match to the PHA-4 binding site consensus of TRTTKRY (TGTTGGT, Figure 5.9). A mutation specifically in this consensus completely abolished expression (Figure 5.8). Since the nas-12 promoter is dependent on a conserved PHA-4 binding site and its flanking sequence, I also tested the possibility that the PHA-4 binding site in Y8A9A.2 might also be a HRL3 binding site. The sequence flanking the PHA-4 site is not identical to that found in the nas-12 and hlh-6 HRL3 sites, but there are additional sequences that do seem to be conserved (Figure 5.6). Thus, I mutated the sequence flanking the conserved PHA-4 binding site in Y8A9A.2 and found that this too completely eliminated expression (Figure

5.8), suggesting that Y8A9A.2 (like nas-12 and hlh-6) requires a HRL3 site for activation in the glands.

Although the HRL3 element in the Y8A9A.2 promoter is required for expression, it is not sufficient for activation by my enhancer assay. Unlike the enhancer constructs for

140

Figure 5.9 The HRL3-like site in the Y8A9A.2 promoter is conserved among nematodes

The sequence of Y8A9A.2 from -276 to -1 bp and the corresponding sequences from C. remanei (Crem), C. briggsae (Cbri), C. brenneri (Cbre) and C. japonica (Cjap).

Sequences were obtained from the UCSC Genome Browser (http://genome.ucsc.edu).

The necessary candidate PHA-4 binding site is highlighted.

141

142 the hlh-6 and nas-12 occurrences of HRL3, three tandem copies of the HRL3 sequence from Y8A9A.2 was not sufficient to activate expression in any cells (data not shown). I did observe rare expression of the enhancer in all of the gland cells of F1 transgenic animals. Expression of transgenes in the F1 generation is known to be stronger and/or broader than in subsequent generations or in stable transgenic lines (Kuchenthal et al.,

2001; Okkema et al., 1993; Raharjo et al., 2010). Thus expression in F1s but not subsequent generations can be an indicator of weak activity. Indeed, the PHA-4 consensus from the Y8A9A.2 promoter was recently found to have almost no activity in an in vivo enhancer study (Raharjo et al., 2010). Since these enhancer constructs are artificial, the combination of a weak PHA-4 binding site with an imperfect HRL3 flanking sequence may not be sufficient to fully activate gland expression. Nonetheless, the activity in F1 transgenics together with the site-specific mutation data suggests that expression of Y8A9A.2 is dependent on a HRL3-like sequence. Both the PHA-4 and flanking site mutations resulted in the complete loss of expression, suggesting that they may not be responsible for restricting Y8A9A.2 expression to g1P in early larvae. A simple model is that Y8A9A.2 is activated in all gland cells by HRL3 but repressed in all non-g1P glands.

Given the repeated involvement of HRL3 sequences in gland expression, I searched the promoters of Y76B12C.3 and wrt-3 (the two other hlh-6 independent genes) for potential HRL3 sites. I found that Y76B12C.3 has a strong match to the HRL3 sites of hlh-6, nas-12 and Y8A9A.2 and this sequence is conserved between other nematode species (Figure 5.6 and data not shown). However, the promoter of wrt-3 does not

143 contain a good match to HRL3, suggesting that there may be yet another gland transcription factor that activates it, warranting further investigation of wrt-3 expression.

5.5 Why do These Multiple Pathways Exist?

One model for gland specification is that all gland genes are either directly or indirectly regulated by the factor acting through HRL3 (Figure 5.10). To test this hypothesis I analyzed the promoters of gland genes that do not have PGM1 for the presence of HRL3. However, I found that there are gland genes that do not have good matches to PGM1 or HRL3 in their promoters (for example, wrt-3). This suggests that they may not be regulated by the factor acting through HRL3, although it is not possible to know if these genes are indirectly regulated by the HRL3 transcription factor. Another unknown with regards to this model is how certain gland genes are expressed only in specific gland sub-types. As mentioned previously with Y8A9A.2, it seems that sub-type specific gland genes are activated broadly in all glands (by HRL3 or hlh-6), and inhibited in a subset of glands. This mode of regulation is also true for phat-5, which is expressed only in the two g1A cells, but activated by hlh-6. Deletions of the phat-5 promoter result in expression in all five glands (Dr. J. Gaudet, personal communication). However there must ultimately be some other mechanism for sub-type-specific expression, as in these examples the inhibiting transcription factor is itself sub-type-specific. Additionally this model of all gland genes being dependent on HRL3 does not answer the question of why some gland genes require hlh-6 but others do not.

One possible explanation for the “division of labour” between HRL3 and hlh-6 is that hlh-6 is subject to regulation relevant to the function of its target genes. For example,

144

Figure 5.10 Model of gland development

I identified a HRL3-like sequence in the promoters of nas-12 and Y8A9A.2 that is necessary for their expression in the glands and wrt-3 does not have a recognizable HRL3 or PGM1 site in its promoter. HRL3 was previously shown to regulate the expression of hlh-6 (Ghai and Gaudet, 2008), suggesting that gland genes that aren’t directly regulated by HRL3 might be indirect targets. Thus the cis-acting element and corresponding transcription factor that regulates wrt-3 might be regulated by HRL3 and the factor acting through HRL3 might be a terminal selector gene. Also interesting is the fact that gland sub-type specific expression patterns occur by repression of broad gland activators (as in phat-5 and Y8A9A.2), which has not been observed in neuronal specification.

145

146 hlh-6 might be able to respond to the availability of food and thus regulate only gland genes that are involved in the feeding process. In order to understand why only certain gland genes are regulated by HLH-6, it is important to understand how hlh-6 is regulated.

One observation of interest is that hlh-6 may be regulated by DAF-16. The FOXO transcription factor daf-16 has been widely implicated in dauer formation, feeding/starvation response, reproduction and aging (Fielenbach and Antebi, 2008;

Henderson and Johnson, 2001). Importantly, nuclear localization of daf-16 is responsive to nutritional status (Henderson and Johnson, 2001). By both yeast one-hybrid and

Chromatin Immunoprecipitation assays, daf-16 was also able to bind to the promoter of hlh-6 (Deplancke et al., 2006). Quantitative RT-PCR for hlh-6 in daf-16 mutants showed a 2-fold increase in expression compared to controls, suggesting that DAF-16 inhibits the expression of hlh-6. There is a consensus DAF-16 binding site (TATTTAC) in the promoter of hlh-6 though it is not conserved. As well, the functional effects of DAF-16 on hlh-6 reporter expression was not shown and the 2-fold decrease in hlh-6 expression was from mixed stage animals leaving it unclear if DAF-16 does repress hlh-6 expression in vivo. Nonetheless, these results suggest the possibility that hlh-6 might be able to respond to nutritional status via DAF-16. For example, starved animals that are not actively feeding may down-regulate genes required for feeding, like hlh-6 and its targets.

To test this hypothesis, I compared the expression of hlh-6, phat-1, nas-12 and Y8A9A.2 in fed versus starved animals using GFP transgenics. nas-12 and Y8A9A.2 were included as control genes whose expression was not expected to be affected by nutritional status, given their independence from hlh-6. I found that the levels of expression of all four genes were unchanged after more than five days of starvation (including in Dauer larvae,

147 data not shown). It is possible that this assay was not sensitive enough to observe the changes in gene expression that occur. As suggested by the quantitative PCR, the changes in expression may be very small (2-fold) and perdurance of GFP may also obscure any changes in reporter expression. Thus, in order to verify that hlh-6 and the phat genes do not respond to starvation, as my data suggests, a quantitative RT-PCR experiment would be needed.

5.6 Summary

In this chapter I was able to demonstrate that at least two hlh-6 independent gland genes require a cis-acting motif (HRL3) previously described as required for hlh-6 expression. I also demonstrated that at least one other (confirmed) hlh-6 independent gland gene (Y76B12C.3) contains a HRL3-like element that could account for its expression. In addition, there is at least one gland gene (wrt-3) whose expression cannot at present be accounted for by either hlh-6/PGM1 or HRL3, implying that there are additional gland cis-acting elements and transcription factors. It will be interesting to see how many other transcription factors are involved in gland development, why these multiple transcription factor batteries exist and whether or not all gland genes are activated by HRL3 (directly or indirectly). The data from this chapter opens the door for many different studies that could answer these questions. The identification of the transcription factor that binds to HRL3 has the potential to be the most illuminating of the proposed studies. Analysis of the HRL3 transcription factor mutant would not only answer whether all gland genes are downstream of HRL3 but it would also provide clues as to the other functions of the glands. For example what is the function of nas-12? Is

148 nas-12 a collagen processing enzyme involved in the formation of the pharyngeal cuticle or is it a digestive enzyme? Subsequent to this are the questions relating to hlh-6: why is nas-12 independent of hlh-6 and yet dependent on the hlh-6 activator HRL3? Similar questions arise with regard to the function of Y8A9A.2. The promoter of Y8A9A.2 also raises the interesting idea that sub-type-specific gland genes arise by the use of activation in all glands and subsequent repression in specific sub-types. The model of gland development that arises from this chapter fits very well with the idea of terminal differentiation genes and terminal selectors (Gaudet and McGhee, 2010). Most gland genes seem to be regulated in a very simple manner, usually involving activation by only one cis-acting element broadly in all five gland cells. The few gland genes with

“complex” expression patterns (i.e. sub-type-specific) only involve the addition of one more cis-acting element. The possibility that HRL3 might activate all gland genes also fits the idea that it might be a “terminal selector” factor. Another interesting possibility is that co-regulated genes are functionally related. All hlh-6 dependent genes might be involved in feeding, Y8A9A.2 (the only known g1P specific gene) might be involved in guiding the M1 neuron and nas-12 (which is broadly activated at all times) might be important for proper pharyngeal cuticle formation, though these hypotheses remain to be tested.

149

Chapter Six: DISCUSSION

6.1 Thesis Summary and Significance

In this thesis I describe the work I have done that demonstrates a role for the gland specific transcription factor HLH-6 in gland development and function. I identified both a cis-regulatory element (PGM1) and trans-acting factor (HLH-6) that are required for expression of certain genes in the pharyngeal glands, though it is presently not known whether the two components interact directly. There are two lines of evidence that support the hypothesis that HLH-6 interacts directly with PGM1. First, the PGM1 motif contains a functional E-box and bHLH proteins (like HLH-6) bind to E-boxes. Second,

PGM1 activity requires HLH-6. A formal possibility is that HLH-6 acts upon a second bHLH that in turn binds to PGM1, as seen with the cascades of neurogenic and myogenic bHLH factors (Schuurmans et al., 2004; Tapscott, 2005). However, no other C. elegans class II bHLH is known to be expressed in pharyngeal glands, though some bHLH genes remain uncharacterized.

Based on my findings, I propose that HLH-6 regulates a battery of pharyngeal gland-expressed genes in C. elegans and is required for both differentiation and function of the glands. While some glands are present in hlh-6 mutants, they are non-functional, as the removal of pharyngeal glands phenocopies the loss of hlh-6. The pharyngeal glands are essential for efficient feeding and appear to play a role in facilitating the transport of bacteria along the pharyngeal lumen, though they are not involved in regulation of pharyngeal pumping. These findings illustrate a previously unknown role for the

150 pharyngeal glands in efficient feeding and demonstrate that aspects of both foregut gland development and function are evolutionarily conserved.

In the last chapter of my thesis, I identified an hlh-6-independent cis-regulatory element (HRL3) necessary for the gland expression of nas-12 and Y8A9A.2. I identified eight genes that were expressed in the glands but did not have a recognizable PGM1 site in their promoters. Surprisingly four of these genes were partially or wholly dependent on hlh-6 for expression. At least one of the genes (T05B4.13) completely dependent on hlh-6 is likely directly regulated by hlh-6 as it is also completely dependent on E-boxes within its promoter. This implies that either HLH-6 can act through non-PGM1 E-boxes or that a

PGM1 site exists and is too divergent to be recognized. I found the previously identified gland element HRL3 in the promoters of three of the four genes whose gland expression was completely independent of hlh-6. This motif was initially discovered in the hlh-6 promoter suggesting a possible model where all gland genes are regulated by the factor acting through HRL3, either directly (like hlh-6 and nas-12) or indirectly (like the phat genes). In this case, the factor acting through HRL3 could be called a terminal selector gene like the genes responsible for specification of some C. elegans neurons. However, one gland gene that did not have PGM1 and was completely independent of hlh-6 (wrt-3) does not have a HRL3 site in its promoter. Further characterization of the wrt-3 promoter and the identification of the HRL3 trans-acting factor would test this model of gland development.

151

6.2 Role of HLH-6 in Gland Development

Characterization of the hlh-6 mutant suggests that hlh-6 is involved in gland differentiation. Although hlh-6 is necessary for the expression of the phat genes and is expressed in all of the glands, the defects in hlh-6 mutants are consistent with HLH-6 playing a role in differentiation of gland cells rather than their specification. The g1 cells are still present in hlh-6 mutants and still have several “gland-like” features: the cell bodies are located in the terminal bulb, they express some gland-specific markers and the cells send projections to the appropriate positions within the pharynx. However, these cells are not fully functional as ablation of all the gland cells is no more severe than loss of hlh-6 alone, indicating that the residual glands in hlh-6 mutants contribute little, if any, wild type function. The g2 gland defect is more pronounced as these cells fail to differentiate in hlh-6 mutants, apparently arresting as precursor cells with an uncertain identity, suggesting that hlh-6 is required for the division of the g2 precursors.

Interestingly, this loss of g2 cells seems to accompany many gland mutants identified in genetic screens (Dr. J. Gaudet, personal communication). Also interesting is the observation that many parasitic nematodes do not have g2-like cells, they only have g1- like cells, three cells with the same position and morphologies of C. elegans g1 glands .

This suggests that the g2 cells might have only recently evolved and thus their development might be more susceptible to perturbations.

The approach that I used in Chapter 3, of taking a set of known gland expressed genes and identifying a common regulator, could have identified a factor that specifies the glands. However the loss of a gland specifying factor would result in the loss of all glands and hlh-6 mutants still have gland cells. One way to identify such a factor would

152 be to identify the regulators of hlh-6 expression. But the expression of hlh-6 is dependent on the combinatorial action of multiple transcription factors (Ghai and Gaudet, 2008;

Raharjo and Gaudet, 2007) suggesting that gland specification might not be as simple as the activation of one master regulator. Nonetheless there is one cis-acting element in the hlh-6 promoter that activates expression in all the glands, HRL3. Until the factor acting through HRL3 is identified, it will remain open whether it does indeed activate all gland genes thus specifying the gland fate.

As with other bHLH proteins, HLH-6 probably functions as a dimer, most likely with the broadly-expressed Class I protein HLH-2. However, HLH-6 appears to require an additional non-bHLH factor that functions through the YMAAY sequence found in

PGM1. Three lines of evidence indicate that HLH-6 requires additional factor(s) to activate gland gene expression. First, the YMAAY sequence is required for PGM1 activity, but is unlikely to represent an extended binding sequence for HLH-6. Second, ectopic expression of HLH-6 (± HLH-2) is not sufficient to activate ectopic expression of a gland-expressed marker suggesting that an additional factor is required to induce target gene expression. Third, HLH-6 (± HLH-2) was not able to bind to PGM1 in vitro using electrophoretic mobility shift assays. Thus, the YMAAY sequence likely represents a binding site for an additional factor. This factor may be limiting with respect to activation of gland genes in vivo and binding to PGM1 in vitro. A precedent for such a model comes from studies of mammalian Mash1, which must form a complex with the POU domain transcription factor Brn2 in order to bind to specific target sequences (Castro et al., 2006). Similarly, the pancreatic determinant PTF1 is a complex of the bHLH Ptf1a with a ubiquitous Class I bHLH and the mammalian Su(H) ortholog RBP-J (Beres et al.,

153

2006); the PTF1 complex binds to a composite DNA sequence consisting of an E-box and a Su(H) site (Cockell et al., 1989).

Involvement of an additional factor may explain the specificity of PGM1 activity.

A general question in transcription factor biology is how specificity of response is achieved. For example, the E-box of PGM1 could be recognized by any of the numerous bHLH factors expressed in the various tissues of C. elegans, yet it is only activated in pharyngeal glands. One solution to this problem is that related transcription factors distinguish between different binding sites based on subtle differences within the core

DNA sequence. For example, different MyoD-containing bHLH dimers have well- characterized binding site preferences (Blackwell and Weintraub, 1990; Huang et al.,

1996), as do the C. elegans bHLH factor Twist/HLH-8 (Harfe et al., 1998; Wang et al.,

2006) and the Drosophila bHLHs atonal and scute (Powell et al., 2004). However, given that binding of bHLH factors to E-boxes may be somewhat promiscuous in vitro, an additional approach to ensure specific response is the involvement of spatially restricted co-factors. Tertiary interactions between bHLH dimers and non-bHLH co-factors are known to affect dimerization and activity (Acar et al., 2006; Hill and Riley, 2004). In our case, a cofactor may recognize the YMAAY portion of PGM1 and be required for transcriptional activation of target genes.

6.3 Role of PHA-4 in Gland Development

The FoxA transcription factor PHA-4 is required for specification of all pharyngeal cells, including glands (Mango et al., 1994a). One question, then, is the regulatory relationship between PHA-4, the HLH-6 gene battery and the HLH-6-

154 independent gland genes. Other work has found that HLH-6 is a probable direct target of

PHA-4, so PGM1-dependent genes are at least indirectly regulated by PHA-4 (Raharjo and Gaudet, 2007). However, previous work suggested that most or all pharyngeal genes are directly regulated by PHA-4 (Gaudet and Mango, 2002). Consistent with this idea, I found candidate PHA-4 binding sites in the regulatory regions of all gland genes analyzed. Furthermore, deletions of the phat-1 promoter suggested PHA-4 was involved in strengthening the gland expression of the gene (see Chapter 3.1). Similar results are seen with the PHA-4 sites in other promoters (e.g. myo-2, Gaudet and Mango, 2002).

PHA-4 may regulate gland-specific gene expression both directly and indirectly, consistent with the proposed model of PHA-4 action. This type of feed-forward transcriptional regulation is also observed in other developmental pathways, such as the myogenic cascade of bHLH transcription factors (Tapscott, 2005).

I have also found that PHA-4 may act directly on PGM1 dependent promoters through this motif. As described in Chapter 3.7, the YMAAY component of PGM1 weakly matches the PHA-4 consensus binding site and can be substituted for a PHA-4 binding site in an enhancer assay. This suggests that HLH-6 (+HLH-2) acts in concert with PHA-4 to activate gland expression and may do so in one of two ways: HLH-6 and

PHA-4 might form a complex in vivo in order to function. Or they might act sequentially.

Forkhead family members in mammalian development are capable of binding to compacted chromatin and opening it, thus allowing transcription to occur (Cirillo et al.,

2002). It is possible that the same occurs in C. elegans, that PHA-4 opens the chromatin surrounding the PGM1 site thus allowing HLH-6 to activate transcription. Chromatin

Immunoprecipitation of PHA-4 bound to a PGM1 consensus would corroborate this

155 hypothesis, and this experiment is currently underway (Dr. J. Gaudet, personal communication).

The maintenance of pha-4 expression in the glands may depend on the cells retaining their differentiated state. Evidence for this hypothesis comes from the observation that in hlh-6 mutants, the g2 gland precursors presumably express pha-4 early in development but when these cells arrest~80% of the time, they lose expression of pha-4::GFP::HIS2B and the other ~20% of the time they retain pha-4 expression (see

Chapter 4.2) A similar loss of pha-4 reporter expression in seen in tbx-2 mutants, which fail to produce anterior pharyngeal muscle (Smith and Mango, 2007). An interesting possibility is that successful differentiation of pharyngeal cells (into specific cell types) is required for maintenance of pha-4 expression and pharyngeal identity. A similar loss of cell identity may occur in unc-120 ; hlh-1; hnd-1 triple mutants, in which presumptive body muscles are found in their normal position within the embryo yet do not adopt a muscle identity nor do they adopt an alternate (non-muscle) fate (Fukushige et al., 2006).

In contrast, C. elegans neurons that lose specific sub-type identities retain their neuronal identity (Wenick and Hobert, 2004).

6.4 Gland Function

In this thesis I have shown that the pharyngeal glands are required for efficient feeding (see Chapter 4.3). A compelling model is that the glands secrete material that coats the pharyngeal lumen to prevent food from adhering to the pharyngeal cuticle.

Support for this model comes from three lines of evidence. First, hlh-6 mutants are feeding defective yet have normal pharyngeal pumping. Second, the starvation phenotype

156 of hlh-6 mutants is rescued by feeding with a different (less viscous) food source. Third, the lining of the pharynx differs in hlh-6 mutants as shown by the inability of the secreted

PHAT-5::MCHERRY protein to adhere tightly to the pharyngeal lumen as it does in wild type (Figure 4.8). Many of the HLH-6-dependent gland genes encode mucin-like proteins, at least one of which (PHAT-5) lines the pharyngeal cuticle. Although I did not demonstrate that PHAT’s are responsible for lubrication, I propose a speculative model in which gland secretion of the mucin-related PHAT proteins act to lubricate the pharyngeal lining, comparable to some aspects of mucin function in other organisms (Kaplan and

Baum, 1993). On the contrary, there may be other HLH-6-dependent gland genes that are responsible for properly lining the pharyngeal lumen to ensure efficient feeding and the primary defect in hlh-6 mutants is the loss of these genes.

Previous observations suggested that the pharyngeal glands might be involved in the molting process based on the observation that they secrete more actively during the molting process (Singh and Sulston, 1978). I have shown that the glands secrete PHAT proteins which are normally bound specifically to the pharyngeal cuticle (see Chapter

4.5). Although the glands may be involved in the molting process, I propose that the reason they are more active during molting is to ensure that the newly synthesized cuticle is properly lined. In the wild, it is likely that an improperly lined cuticle (such as that found in the hlh-6 mutants) would be lethal. Thus it seems logical that the glands would be active during molting so that the newly synthesized cuticle is lined properly before it is needed to function.

Given that the glands express many other types of proteins, it is possible that they have functions in addition to feeding. The gland specific gene Y8A9A.2 might aid in the

157 guidance of pharyngeal neurons. Previous studies have found that ablation of the glands results in guidance defects of the pharyngeal neuron M1 (Dr. J. Gaudet, personal communication). In the procorpus the g1P gland extension comes into direct contact with the M1 axon and it is specifically in this region where the M1 neural defects occur in gland ablated animals. Importantly, hlh-6 mutants have normal M1 neurons (E. Rollins, personal communication), thus whatever proteins are involved in aiding the M1 neuron are hlh-6 independent. This places Y8A9A.2 as a perfect candidate for M1 guidance: it is expressed (early on) specifically in the g1P cells and it encodes a thrombospondin protein which may be involved in promoting neural adhesion (Feinstein et al., 1999). However, since RNAi is ineffective in the glands (see Chapter 3.7) and there are no mutations in

Y8A9A.2, this hypothesis remains untested.

6.5 Evolutionary Conservation

An interesting finding is that both the regulation (by bHLH factors) and function

(feeding) of foregut glands appears to be evolutionarily conserved. The closest mammalian homolog of HLH-6 is Sgn1, a bHLH required for normal salivary gland development in the mammalian foregut (Yoshida et al., 2001). In addition, development of salivary glands in the Drosophila foregut depends on sage (a salivary gland expressed bHLH) (Abrams et al., 2006), although sage is not the closest homolog to hlh-6. Database searches have found other genes encoding proteins with high similarity to HLH-6, including the Ash2 gene, which is expressed in the digestive tract glands of the jellyfish

P. carnea (Seipel et al., 2004) and SGSF which is expressed in the silk glands of black widow spiders (Kohler et al., 2005). I also identified HLH-6-related sequences in the

158 genomes of parasitic nematodes. Gland function in parasitic nematodes is critical for parasitism, suggesting a conserved function of foregut glands in the processing or passage of food (Davis et al., 2000; Loukas et al., 2000). Targeting gland development or function may offer a new strategy for controlling these parasitic species.

Another interesting finding is the combinatorial role of HLH and forkhead proteins in foregut gland development. In Drosophila, forkhead (the ortholog of PHA-4 and founding member of the forkhead family) regulates both foregut and salivary gland fate (Kuo et al., 1996; Weigel et al., 1989a; Weigel et al., 1989b). In fact, development of the salivary glands in Drosophila depends on the combined activity of forkhead and sage, which can act together on the same promoter (Abrams et al., 2006). As well, Forkhead was shown to activate sage, which is identical to the feed-forward hierarchy of PHA-4, hlh-6 and the phat genes. There are many glands found in the mammalian foregut including the very well studied pancreas. Much like in Drosophila, mammalian foregut and pancreas precursors express and require FoxA/HNF3 the mammalian ortholog of

PHA-4 (Ang and Rossant, 1994; Ang et al., 1993; Monaghan et al., 1993). FoxA/HNF3 was shown to directly activate the critical pancreas determinant PDX1 (pancreas duodenum homeobox 1) suggesting it sits at the top of the pancreas specification hierarchy (Samaras et al., 2002), much like PHA-4 in C. elegans pharyngeal gland development. FoxA/HNF3 was also shown to act with Neurogenin 3 (NGN3) a bHLH, to activate transcription of a pancreas specific gene (Watada et al., 2003). Although there are additional bHLH’s involved in pancreas development (such as PTF1, Beres et al.,

2006) and there is no evidence for the role of FoxA/HNF3 in mammalian salivary gland development, the fact that FoxA/HNF3 and bHLH proteins specify foregut gland fates in

159 mammals, Drosophila and C. elegans suggests this combination of transcription factors was highly conserved through evolution.

6.6 Future Directions

Because of the pioneering nature of this thesis, there are many more questions left unanswered. The main question left unanswered is how are the glands specified? I have elucidated a major portion of the gland cell developmental hierarchy, but until there is a mutant (or combination of mutants) that completely lacks gland cells this question will remain unanswered. It is possible that gland fate requires the combinatorial actions of multiple factors, as does the regulation of hlh-6 (Ghai and Gaudet, 2008; Raharjo and

Gaudet, 2007), or that it requires only one factor, which could be the factor acting through HRL3. Future studies are aimed at identifying these unknown factors. This begs the question how many gland transcription factors and gene batteries are there? We now know of at least 3: HLH-6 dependent genes (phat genes), HRL3 directly dependent genes

(nas-12 and Y8A9A.2) and HLH-6/HRL-3 independent genes (wrt-3 and possibly others).

Why these multiple pathways to gland expression exist and their effects on gland function are also important questions to ask. Related to this is how gland sub-type specific expression is regulated. It seems that sub-type specific genes (for example phat-5 and Y8A9A.2) are activated broadly in all five glands and then repressed in specific sub- types. But what are these sub-type specific repressors and how are they themselves activated? Lastly, and most importantly, do the principles involved in gland specification and development apply to other pharyngeal cell types, or to cell types in other organisms?

This thesis provides the groundwork to be able to answer these questions in the future.

160

Chapter Seven: REFERENCES

Abrams, E. W., Mihoulides, W. K., and Andrew, D. J. (2006). Fork head and Sage maintain a uniform and patent salivary gland lumen through regulation of two downstream target genes, PH4{alpha}SG1 and PH4{alpha}SG2. Development 133, 3517-3527.

Acar, M., Jafar-Nejad, H., Giagtzoglou, N., Yallampalli, S., David, G., He, Y., Delidakis, C., and Bellen, H. J. (2006). Senseless physically interacts with proneural proteins and functions as a transcriptional co-activator. Development 133, 1979-1989.

Albertson, D. G., and Thomson, J. N. (1976). The pharynx of Caenorhabditis elegans. Philos Trans R Soc Lond B Biol Sci 275, 299-325.

Altun-Gultekin, Z., Andachi, Y., Tsalik, E. L., Pilgrim, D., Kohara, Y., and Hobert, O. (2001). A regulatory cascade of three homeobox genes, ceh-10, ttx-3 and ceh-23, controls cell fate specification of a defined interneuron class in C. elegans. Development 128, 1951-1969.

Ang, S. L., and Rossant, J. (1994). HNF-3 beta is essential for node and notochord formation in mouse development. Cell 78, 561-574.

Ang, S.-L., Wierda, A., Wong, D., Stevens, K. A., Cascio, S., Rossant, J., and Zaret, K. S. (1993). The formation and maintenance of the definitive endoderm lineage in the mouse: involvement of HNF-3/forkhead proteins. Development 119, 1301-1315.

Antebi, A. (2006). Nuclear hormone receptors in C. elegans. WormBook, 1-13.

Ao, W., Gaudet, J., Kent, W. J., Muttumu, S., and Mango, S. E. (2004). Environmentally induced foregut remodeling by PHA-4/FoxA and DAF-12/NHR. Science 305, 1743- 1746.

Ashrafi, K., Chang, F. Y., Watts, J. L., Fraser, A. G., Kamath, R. S., Ahringer, J., and Ruvkun, G. (2003). Genome-wide RNAi analysis of Caenorhabditis elegans fat regulatory genes. Nature 421, 268-272.

Aspock, G., Kagoshima, H., Niklaus, G., and Burglin, T. R. (1999). Caenorhabditis elegans has scores of hedgehog-related genes: sequence and expression analysis. Genome Res 9, 909-923.

Aspock, G., Ruvkun, G., and Burglin, T. R. (2003). The Caenorhabditis elegans ems class homeobox gene ceh-2 is required for M3 pharynx motoneuron function. Development 130, 3369-3378.

161

Avery, L. (1993). The genetics of feeding in Caenorhabditis elegans. Genetics 133, 897- 917.

Avery, L., and Horvitz, H. R. (1989). Pharyngeal pumping continues after laser killing of the pharyngeal nervous system of C. elegans. Neuron 3, 473-485.

Avery, L., and Shtonda, B. B. (2003). Food transport in the C. elegans pharynx. J Exp Biol 206, 2441-2457.

Axang, C., Rauthan, M., Hall, D. H., and Pilon, M. (2008). Developmental genetics of the C. elegans pharyngeal neurons NSML and NSMR. BMC Dev Biol 8, 38.

Beres, T. M., Masui, T., Swift, G. H., Shi, L., Henke, R. M., and MacDonald, R. J. (2006). PTF1 is an organ-specific and Notch-independent basic helix-loop-helix complex containing the mammalian Suppressor of Hairless (RBP-J) or its paralogue, RBP-L. Mol Cell Biol 26, 117-130.

Blackwell, T. K., and Weintraub, H. (1990). Differences and similarities in DNA-binding preferences of MyoD and E2A protein complexes revealed by binding site selection. Science 250, 1104-1110.

Bork, P., and Beckmann, G. (1993). The CUB domain. A widespread module in developmentally regulated proteins. J Mol Biol 231, 539-545.

Bowerman, B., Draper, B. W., Mello, C. C., and Priess, J. F. (1993). The maternal gene skn-1 encodes a protein that is distributed unequally in early C. elegans embryos. Cell 74, 443-452.

Bowerman, B., Eaton, B. A., and Priess, J. R. (1992). skn-1, a maternally expressed gene required to specify the fate of ventral blastomeres in the early C. elegans embryo. Cell 68, 1061-1075.

Broitman-Maduro, G., Lin, K. T., Hung, W. W., and Maduro, M. F. (2006). Specification of the C. elegans MS blastomere by the T-box factor TBX-35. Development 133, 3097- 3106.

Broitman-Maduro, G., Maduro, M. F., and Rothman, J. H. (2005). The noncanonical binding site of the MED-1 GATA factor defines differentially regulated target genes in the C. elegans mesendoderm. Dev Cell 8, 427-433.

Broitman-Maduro, G., Owraghi, M., Hung, W. W., Kuntz, S., Sternberg, P. W., and Maduro, M. F. (2009). The NK-2 class homeodomain factor CEH-51 and the T-box factor TBX-35 have overlapping function in C. elegans mesoderm development. Development 136, 2735-2746.

162

Captan, V. V., Goszczynski, B., and McGhee, J. D. (2007). Neither maternal nor zygotic med-1/med-2 genes play a major role in specifying the Caenorhabditis elegans endoderm. Genetics 175, 969-974.

Castaneda, O., Sotolongo, V., Amor, A. M., Stocklin, R., Anderson, A. J., Harvey, A. L., Engstrom, A., Wernstedt, C., and Karlsson, E. (1995). Characterization of a potassium channel toxin from the Caribbean Sea anemone Stichodactyla helianthus. Toxicon 33, 603-613.

Castro, D. S., Skowronska-Krawczyk, D., Armant, O., Donaldson, I. J., Parras, C., Hunt, C., Critchley, J. A., Nguyen, L., Gossler, A., Gottgens, B., et al. (2006). Proneural bHLH and Brn proteins coregulate a neurogenic program through cooperative binding to a conserved DNA motif. Dev Cell 11, 831-844.

Chapman, D. L., and Papaioannou, V. E. (1998). Three neural tubes in mouse embryos with mutations in the T-box gene Tbx6. Nature 391, 695-697.

Chitwood, B. G., and Chitwood, M. B. (1974). Introduction to nematology, University Park Press).

Cirillo, L. A., Lin, F. R., Cuesta, I., Friedman, D., Jarnik, M., and Zaret, K. S. (2002). Opening of compacted chromatin by early developmental transcription factors HNF3 (FoxA) and GATA-4. Mol Cell 9, 279-289.

Cockell, M., Stevenson, B. J., Strubin, M., Hagenbuchle, O., and Wellauer, P. K. (1989). Identification of a cell-specific DNA-binding activity that interacts with a transcriptional activator of genes expressed in the acinar pancreas. Mol Cell Biol 9, 2464-2476.

Conradt, B., and Horvitz, H. R. (1998). The C. elegans protein EGL-1 is required for programmed cell death and interacts with the Bcl-2-like protein CED-9. Cell 93, 519- 529.

Crooks, G. E., Hon, G., Chandonia, J. M., and Brenner, S. E. (2004). WebLogo: a sequence logo generator. Genome Res 14, 1188-1190.

Daigle, S. N., and Creutz, C. E. (1999). Transcription, biochemistry and localization of nematode annexins. J Cell Sci 112 ( Pt 12), 1901-1913.

Davis, E. L., Hussey, R. S., Baum, T. J., Bakker, J., Schots, A., Rosso, M. N., and Abad, P. (2000). Nematode Parasitism Genes. Annu Rev Phytopathol 38, 365-396.

Deplancke, B., Mukhopadhyay, A., Ao, W., Elewa, A. M., Grove, C. A., Martinez, N. J., Sequerra, R., Doucette-Stamm, L., Reece-Hoyes, J. S., Hope, I. A., et al. (2006). A gene- centered C. elegans protein-DNA interaction network. Cell 125, 1193-1205.

163

Doedens, A., Loukas, A., and Maizels, R. M. (2001). A cDNA encoding Tc-MUC-5, a mucin from Toxocara canis larvae identified by expression screening. Acta Trop 79, 211- 217.

Ellis, H. M., and Horvitz, H. R. (1986). Genetic control of programmed cell death in the nematode C. elegans. Cell 44, 817-829.

Emanuelsson, O., Brunak, S., von Heijne, G., and Nielsen, H. (2007). Locating proteins in the cell using TargetP, SignalP and related tools. Nat Protoc 2, 953-971.

Ephrussi, A., Church, G. M., Tonegawa, S., and Gilbert, W. (1985). B lineage--specific interactions of an immunoglobulin enhancer with cellular factors in vivo. Science 227, 134-140.

Evans, T. C., and Hunter, C. P. (2005). Translational control of maternal RNAs. WormBook, 1-11.

Fahlen, T., Sunga, J., Rask, J., Herrera, A., Lam, K., Sing, L., Sato, K., Ramos, R. A., Kirven-Brooks, M., and Reiss-Bubenheim, D. (2005). Methods for the culture of C. elegans and S. cerevisiae in microgravity. Gravit Space Biol Bull 18, 91-92.

Fang-Yen, C., Avery, L., and Samuel, A. D. (2009). Two size-selective mechanisms specifically trap bacteria-sized food particles in Caenorhabditis elegans. Proc Natl Acad Sci U S A 106, 20093-20096.

Feinstein, Y., Borrell, V., Garcia, C., Burstyn-Cohen, T., Tzarfaty, V., Frumkin, A., Nose, A., Okamoto, H., Higashijima, S., Soriano, E., and Klar, A. (1999). F-spondin and mindin: two structurally and functionally related genes expressed in the hippocampus that promote outgrowth of embryonic hippocampal neurons. Development 126, 3637-3648.

Fielenbach, N., and Antebi, A. (2008). C. elegans dauer formation and the molecular basis of plasticity. Genes Dev 22, 2149-2165.

Fletcher, C., Heintz, N., and Roeder, R. G. (1987). Purification and characterization of OTF-1, a transcription factor regulating cell cycle expression of a human histone H2b gene. Cell 51, 773-781.

Fukushige, T., Brodigan, T. M., Schriefer, L. A., Waterston, R. H., and Krause, M. (2006). Defining the transcriptional redundancy of early bodywall muscle development in C. elegans: evidence for a unified theory of animal muscle development. Genes Dev 20, 3395-3406.

Fukushige, T., Hawkins, M. G., and McGhee, J. D. (1998). The GATA-factor elt-2 is essential for formation of the Caenorhabditis elegans intestine. Dev Biol 198, 286-302.

164

Garigan, D., Hsu, A. L., Fraser, A. G., Kamath, R. S., Ahringer, J., and Kenyon, C. (2002). Genetic analysis of tissue aging in Caenorhabditis elegans: a role for heat-shock factor and bacterial proliferation. Genetics 161, 1101-1112.

Gaudet, J., and Mango, S. E. (2002). Regulation of organogenesis by the Caenorhabditis elegans FoxA protein PHA-4. Science 295, 821-825.

Gaudet, J., and McGhee, J. D. (2010). Recent advances in understanding the molecular mechanisms regulating C. elegans transcription. Dev Dyn 239, 1388-1404.

Gaudet, J., Muttumu, S., Horner, M., and Mango, S. E. (2004). Whole-genome analysis of temporal gene expression during foregut development. PLoS Biol 2, e352.

Gems, D., and Maizels, R. M. (1996). An abundantly expressed mucin-like protein from Toxocara canis infective larvae: the precursor of the larval surface coat glycoproteins. Proc Natl Acad Sci U S A 93, 1665-1670.

Gengyo-Ando, K., and Mitani, S. (2000). Characterization of mutations induced by ethyl methanesulfonate, UV, and trimethylpsoralen in the nematode Caenorhabditis elegans. Biochem Biophys Res Commun 269, 64-69.

Ghai, V., and Gaudet, J. (2008). The CSL transcription factor LAG-1 directly represses hlh-6 expression in C. elegans. Dev Biol 322, 334-344.

Gissendanner, C. R., Crossgrove, K., Kraus, K. A., Maina, C. V., and Sluder, A. E. (2004). Expression and function of conserved nuclear receptor genes in Caenorhabditis elegans. Dev Biol 266, 399-416.

Good, K., Ciosk, R., Nance, J., Neves, A., Hill, R. J., and Priess, J. R. (2004). The T-box transcription factors TBX-37 and TBX-38 link GLP-1/Notch signaling to mesoderm induction in C. elegans embryos. Development 131, 1967-1978.

Goszczynski, B., and McGhee, J. D. (2005). Reevaluation of the role of the med-1 and med-2 genes in specifying the Caenorhabditis elegans endoderm. Genetics 171, 545-555.

Grove, C. A., De Masi, F., Barrasa, M. I., Newburger, D. E., Alkema, M. J., Bulyk, M. L., and Walhout, A. J. (2009). A multiparameter network reveals extensive divergence between C. elegans bHLH transcription factors. Cell 138, 314-327.

Hahn-Windgassen, A., and Van Gilst, M. R. (2009). The Caenorhabditis elegans HNF4alpha Homolog, NHR-31, mediates excretory tube growth and function through coordinate regulation of the vacuolar ATPase. PLoS Genet 5, e1000553.

Harfe, B. D., Gomes, A. V., Kenyon, C., Liu, J., Krause, M., and Fire, A. (1998). Analysis of a Caenorhabditis elegans Twist homolog identifies conserved and divergent aspects of mesodermal patterning. Genes Dev 12, 2623-2635.

165

Harris, T. W., Chen, N., Cunningham, F., Tello-Ruiz, M., Antoshechkin, I., Bastiani, C., Bieri, T., Blasiar, D., Bradnam, K., Chan, J., et al. (2004). WormBase: a multi-species resource for nematode biology and genomics. Nucleic Acids Res 32 Database issue, D411-417.

Haun, C., Alexander, J., Stainier, D. Y., and Okkema, P. G. (1998). Rescue of Caenorhabditis elegans pharyngeal development by a vertebrate heart specification gene. Proc Natl Acad Sci U S A 95, 5072-5075.

Henderson, S. T., and Johnson, T. E. (2001). daf-16 integrates developmental and environmental inputs to mediate aging in the nematode Caenorhabditis elegans. Curr Biol 11, 1975-1980.

Hill, A. A., and Riley, P. R. (2004). Differential regulation of Hand1 homodimer and Hand1-E12 heterodimer activity by the cofactor FHL2. Mol Cell Biol 24, 9835-9847.

Hishida, R., takeshi, I., Kondo, K., and Katsura, I. (1996). hch-1, a gene requried fo noml atcing and norl migrtion of a neuoblas nC> elegans, ecodsarteinelated to TOLLOID and BMP-1. EMBO J 15, 4111-4122.

Ho, S. N., Hunt, H. D., Horton, R. M., Pullen, J. K., and Pease, L. R. (1989). Site- directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 77, 51-59.

Hobert, O. (2008). Regulatory logic of neuronal diversity: terminal selector genes and selector motifs. Proc Natl Acad Sci U S A 105, 20067-20071.

Hoflich, J., Berninsone, P., Gobel, C., Gravato-Nobre, M. J., Libby, B. J., Darby, C., Politz, S. M., Hodgkin, J., Hirschberg, C. B., and Baumeister, R. (2004). Loss of srf-3- encoded nucleotide sugar transporter activity in Caenorhabditis elegans alters surface antigenicity and prevents bacterial adherence. J Biol Chem 279, 30440-30448.

Hope, I. A., Albertson, D. G., Martinelli, S. D., Lynch, A. S., Sonnhammer, E., and Durbin, R. (1996). The C. elegans expression pattern database: a beginning. Trends Genet 12, 370-371.

Horb, M. E., and Thomsen, G. H. (1999). Tbx5 is essential for heart development. Development 126, 1739-1751.

Horner, M. A., Quintin, S., Domeier, M. E., Kimble, J., Labouesse, M., and Mango, S. E. (1998). pha-4, an HNF-3 homologue, specifies pharyngeal organ identity in Caenorhabditis elegans. Genes Dev 12, 1947-1952.

Huang, J., Blackwell, T. K., Kedes, L., and Weintraub, H. (1996). Differences between MyoD DNA binding and activation site requirements revealed by functional random sequence selection. Mol Cell Biol 16, 3893-3900.

166

Huang, S., Shetty, P., Robertson, S. M., and Lin, R. (2007). Binary cell fate specification during C. elegans embryogenesis driven by reiterated reciprocal asymmetry of TCF POP- 1 and its coactivator beta-catenin SYS-1. Development 134, 2685-2695.

Hutter, H., and Schnabel, R. (1994). glp-1 and inductions establishing embryonic axes in C. elegans. Development 120, 2051-2064.

Jasmer, D. P., Goverse, A., and Smant, G. (2003). Parasitic nematode interactions with mammals and plants. Annu Rev Phytopathol 41, 245-270.

Julenius, K., Molgaard, A., Gupta, R., and Brunak, S. (2005). Prediction, conservation analysis, and structural characterization of mammalian mucin-type O-glycosylation sites. Glycobiology 15, 153-164.

Kalb, J. M., Lau, K. K., Goszczynski, B., Fukushige, T., Moons, D., Okkema, P. G., and McGhee, J. D. (1998). pha-4 is Ce-fkh-1, a fork head/HNF-3alpha,beta,gamma homolog that functions in organogenesis of the C. elegans pharynx. Development 125, 2171-2180.

Kamath, R. S., Fraser, A. G., Dong, Y., Poulin, G., Durbin, R., Gotta, M., Kanapin, A., Le Bot, N., Moreno, S., Sohrmann, M., et al. (2003). Systematic functional analysis of the Caenorhabditis elegans genome using RNAi. Nature 421, 220-221.

Kaplan, M. D., and Baum, B. J. (1993). The functions of saliva. Dysphagia 8, 225-229.

Kiefer, J. C., Smith, P. A., and Mango, S. E. (2007). PHA-4/FoxA cooperates with TAM- 1/TRIM to regulate cell fate restriction in the C. elegans foregut. Dev Biol 303, 611-624.

Kohara, Y. (2001a). NEXTDB: The Nematode Expression Pattern Database.

Kohara, Y. (2001b). [Systematic analysis of gene expression of the C. elegans genome]. Tanpakushitsu Kakusan Koso 46, 2425-2431.

Kohler, K., Thayer, W., Le, T., Sembhi, A., Vasanthavada, K., Moore, A. M., and Vierra, C. A. (2005). Characterization of a novel class II bHLH transcription factor from the black widow spider, Latrodectus hesperus, with silk-gland restricted patterns of expression. DNA Cell Biol 24, 371-380.

Krause, M., Park, M., Zhang, J. M., Yuan, J., Harfe, B., Xu, S. Q., Greenwald, I., Cole, M., Paterson, B., and Fire, A. (1997). A C. elegans E/Daughterless bHLH protein marks neuronal but not striated muscle development. Development 124, 2179-2189.

Kuchenthal, C. A., Chen, W., and Okkema, P. G. (2001). Multiple enhancers contribute to expression of the NK-2 homeobox gene ceh-22 in C. elegans pharyngeal muscle. Genesis 31, 156-166.

167

Kuo, Y. i. M., Jones, N., Zhou, B., Panzer, S., Larson, V., and Beckendorf, S. K. (1996). Salivary duct determination in Drosophila: roles of the EGF receptor signaling pathway and the transcription factors fork head and trachealess. Development 122, 1909-1917.

Latchman, D. S. (1999). POU family transcription factors in the nervous system. J Cell Physiol 179, 126-133.

Lee, J. I., Dhakal, B. K., Lee, J., Bandyopadhyay, J., Jeong, S. Y., Eom, S. H., Kim, D. H., and Ahnn, J. (2003). The Caenorhabditis elegans homologue of Down syndrome critical region 1, RCN-1, inhibits multiple functions of the phosphatase calcineurin. J Mol Biol 328, 147-156.

Lee, R. Y., Sawin, E. R., Chalfie, M., Horvitz, H. R., and Avery, L. (1999). EAT-4, a homolog of a mammalian sodium-dependent inorganic phosphate cotransporter, is necessary for glutamatergic neurotransmission in caenorhabditis elegans. J Neurosci 19, 159-167.

Lints, T. J., Parsons, L. M., Hartley, L., Lyons, I., and Harvey, R. P. (1993). Nkx-2.5: a novel murine homeobox gene expressed inearly heart progenitor cells and their myogenic descendants. Development 119, 419-431.

Loukas, A., Hintz, M., Linder, D., Mullin, N. P., Parkinson, J., Tetteh, K. K., and Maizels, R. M. (2000). A family of secreted mucins from the parasitic nematode Toxocara canis bears diverse mucin domains but shares similar flanking six-cysteine repeat motifs. J Biol Chem 275, 39600-39607.

Lyons, I., Parsons, L. M., Hartley, L., Li, R., Andrews, J. E., Robb, L., and Harvey, R. P. (1995). Myogenic and morphogenetic defects in the heart tubes of murine embryos lacking the homeo box gene Nkx2-5. Genes Dev 9, 1654-1666.

Ma, P. C., Rould, M. A., Weintraub, H., and Pabo, C. O. (1994). Crystal structure of MyoD bHLH domain-DNA complex: perspectives on DNA recognition and implications for transcriptional activation. Cell 77, 451-459.

MacMorris, M., Broverman, S., Greenspoon, S., Lea, K., Madej, C., Blumenthal, T., and Spieth, J. (1992). Regulation of vitellogenin gene expression in transgenic Caenorhabditis elegans: short sequences required for activation of the vit-2 promoter. Mol Cell Biol 12, 1652-1662.

Maduro, M. F., Broitman-Maduro, G., Mengarelli, I., and Rothman, J. H. (2007). Maternal deployment of the embryonic SKN-1-->MED-1,2 cell specification pathway in C. elegans. Dev Biol 301, 590-601.

Maduro, M. F., Lin, R., and Rothman, J. H. (2002). Dynamics of a Developmental Switch: Recursive Intracellular and Intranuclear Redistribution of Caenorhabditis

168 elegans POP-1 Parallels Wnt-Inhibited Transcriptional Repression. Dev Bio 248, 128- 142.

Maduro, M. F., Meneghini, M. D., Bowerman, B., Broitman-Maduro, G., and Rothman, J. H. (2001). Restriction of mesendoderm to a single blastomere by the combined action of SKN-1 and a GSK-3 beta homolog is mediated by MED-1 and -2 in C. elegans. Mol Cell 7, 475-485.

Mah, A. K., Armstrong, K. R., Chew, D. S., Chu, J. S., Tu, D. K., Johnsen, R. C., Chen, N., Chamberlin, H. M., and Baillie, D. L. (2007). Transcriptional regulation of AQP-8, a Caenorhabditis elegans aquaporin exclusively expressed in the excretory system, by the POU homeobox transcription factor CEH-6. J Biol Chem 282, 28074-28086.

Mallo, G. V., Kurz, C. L., Couillault, C., Pujol, N., Granjeaud, S., Kohara, Y., and Ewbank, J. J. (2002). Inducible antibacterial defense system in C. elegans. Curr Biol 12, 1209-1214.

Mango, S. E., Lambie, E. J., and Kimble, J. (1994a). The pha-4 gene is required to generate the pharyngeal primordium of Caenorhabditis elegans. Development 120, 3019- 3031.

Mango, S. E., Thorpe, C. J., Martin, P. R., Chamberlain, S. H., and Bowerman, B. (1994b). Two maternal genes, apx-1 and pie-1, are required to distinguish the fates of equivalent blastomeres in the early Caenorhabditis embryo. Development 120, 2305- 2315.

Massari, M. E., and Murre, C. (2000). Helix-loop-helix proteins: regulators of transcription in eucaryotic organisms. Mol Cell Biol 20, 429-440.

McGhee, J. D., Fukushige, T., Krause, M. W., Minnema, S. E., Goszczynski, B., Gaudet, J., Kohara, Y., Bossinger, O., Zhao, Y., Khattra, J., et al. (2009). ELT-2 is the predominant transcription factor controlling differentiation and function of the C. elegans intestine, from embryo to adult. Dev Biol 327, 551-565.

McGhee, J. D., Sleumer, M. C., Bilenky, M., Wong, K., McKay, S. J., Goszczynski, B., Tian, H., Krich, N. D., Khattra, J., Holt, R. A., et al. (2007). The ELT-2 GATA-factor and the global regulation of transcription in the C. elegans intestine. Dev Biol 302, 627- 645.

McKay, J. P., Raizen, D. M., Gottschalk, A., Schafer, W. R., and Avery, L. (2004). eat-2 and eat-18 are required for nicotinic neurotransmission in the Caenorhabditis elegans pharynx. Genetics 166, 161-169.

Mello, C., and Fire, A. (1995). DNA transformation. Methods Cell Biol 48, 451-482.

169

Mello, C. C., Draper, B. W., Krause, M., Weintraub, H., and Preiss, J. R. (1992). The pie- 1 and mex-1 genes and maternal control of blastomere identity in early C.elegans embryos. Cell 70, 163-176.

Mello, C. C., Kramer, J. M., Stinchcomb, D., and Ambros, V. (1991). Efficient gene transfer in C. elegans: extrachromosomal maintenance and integration of transforming sequences. EMBO J 10, 3959-3970.

Mickey, K. M., C.C.Mello, Montgomery, M. K., Fire, A., and J.R.Priess (1996). An inductive interaction in 4-cell stage C. elegans embryos involves APX-1 expression in the signalling cell. Development 122, 1791-1798.

Miskolczi-McCallum, C. M., Scavetta, R. J., Svendsen, P. C., Soanes, K. H., and Brook, W. J. (2005). The Drosophila melanogaster T-box genes midline and H15 are conserved regulators of heart development. Dev Biol 278, 459-472.

Mohrlen, F., Hutter, H., and Zwilling, R. (2003). The astacin protein family in Caenorhabditis elegans. Eur J Biochem 270, 4909-4920.

Monaghan, A. P., Kaestner, K. H., Grau, E., and Schutz, G. (1993). Postimplantation expression patterns indicate a role for the mouse forkhead/HNF-3 and genes in determination of the definitive endoderm, chordamesoderm and neuroectoderm. Development 119, 567-578.

Morck, C., and Pilon, M. (2006). C. elegans feeding defective mutants have shorter body lengths and increased autophagy. BMC Dev Biol 6, 39.

Morck, C., Rauthan, M., Wagberg, F., and Pilon, M. (2004). pha-2 encodes the C. elegans ortholog of the homeodomain protein HEX and is required for the formation of the pharyngeal isthmus. Dev Biol 272, 403-418.

Moskowitz, I. P. G., Gendreau, S. B., and Rothman, J. H. (1994). Combinatorial specification of blastomere identity by glp-1 dependent cellular interactions in the nematode Caenorhabditis elegans. Development 120, 3325-3338.

Neves, A., English, K., and Priess, J. R. (2007). Notch-GATA synergy promotes endoderm-specific expression of ref-1 in C. elegans. Development 134, 4459-4468.

Neves, A., and Priess, J. R. (2005). The REF-1 family of bHLH transcription factors pattern C. elegans embryos through Notch-dependent and Notch-independent pathways. Dev Cell 8, 867-879.

Novelli, J., Ahmed, S., and Hodgkin, J. (2004). Gene interactions in Caenorhabditis elegans define DPY-31 as a candidate procollagen C-proteinase and SQT-3/ROL-4 as its predicted major target. Genetics 168, 1259-1273.

170

Ohmachi, M., Sugimoto, A., Iino, Y., and Yamamoto, M. (1999). kel-1, a novel Kelch- related gene in Caenorhabditis elegans, is expressed in pharyngeal gland cells and is required for the feeding process. Genes Cells 4, 325-337.

Okkema, P. G., and Fire, A. (1994). The Caenorhabditis elegans NK-2 class homeoprotein CEH-22 is involved in combinatorial activation of gene expression in pharyngeal muscle. Development 120, 2175-2186.

Okkema, P. G., Ha, E., Haun, C., Chen, W., and Fire, A. (1997). The Caenorhabditis elegans NK-2 homeobox gene ceh-22 activates pharyngeal muscle gene expression in combination with pha-1 and is required for normal pharyngeal development. Development 124, 3965-3973.

Okkema, P. G., Harrison, S. W., Plunger, V., Aryana, A., and Fire, A. (1993). Sequence requirements for myosin gene expression and regulation in Caenorhabditis elegans. Genetics 135, 385-404.

Okkema, P. G., and Krause, M. (2005). Transcriptional regulation. WormBook, 1-40.

O'Rourke, E. J., Soukas, A. A., Carr, C. E., and Ruvkun, G. (2009). C. elegans major fats are stored in vesicles distinct from lysosome-related organelles. Cell Metab 10, 430-435.

Overdier, D. G., Porcella, A., and Costa, R. H. (1994). The DNA-binding specificity of the hepatocyte nuclear factor 3/forkhead domain is influenced by amino acid residues adjacent to the recognition helix. Molecular and Cellular Biology 14, 2755-2766.

Park, J. O., Pan, J., Mohrlen, F., Schupp, M. O., Johnsen, R., Baillie, D. L., Zapf, R., Moerman, D. G., and Hutter, H. (2010). Characterization of the astacin family of metalloproteases in C. elegans. BMC Dev Biol 10, 14.

Portereiko, M. F., and Mango, S. E. (2001). Early morphogenesis of the Caenorhabditis elegans pharynx. Dev Biol 233, 482-494.

Powell, L. M., Zur Lage, P. I., Prentice, D. R., Senthinathan, B., and Jarman, A. P. (2004). The proneural proteins Atonal and Scute regulate neural target genes through different E-box binding sites. Mol Cell Biol 24, 9517-9526.

Raharjo, I., and Gaudet, J. (2007). Gland-specific expression of C. elegans hlh-6 requires the combinatorial action of three distinct promoter elements. Dev Biol 302, 295-308.

Raharjo, I., Logan, C., Wen, S., Kalb, J., and Gaudet, J. (2010). In vitro and in vivo characterization of C. elegans PHA-4/FoxA response elements. Developmental Biology In press.

Rasmussen, J. P., English, K., Tenlen, J. R., and Priess, J. R. (2008). Notch signaling and morphogenesis of single-cell tubes in the C. elegans digestive tract. Dev Cell 14, 559- 569.

171

Rauthan, M., Morck, C., and Pilon, M. (2007). The C. elegans M3 neuron guides the growth cone of its sister cell M2 via the Kruppel-like zinc finger protein MNM-2. Dev Biol 311, 185-199.

Ray, P., Schnabel, R., and Okkema, P. G. (2008). Behavioral and synaptic defects in C. elegans lacking the NK-2 homeobox gene ceh-28. Dev Neurobiol 68, 421-433.

Reddi, A. H. (1996). BMP-1: resurrection as procollagen C-proteinase. Science 271, 463.

Reece-Hoyes, J. S., Deplancke, B., Barrasa, M. I., Hatzold, J., Smit, R. B., Arda, H. E., Pope, P. A., Gaudet, J., Conradt, B., and Walhout, A. J. (2009). The C. elegans Snail homolog CES-1 can activate gene expression in vivo and share targets with bHLH transcription factors. Nucleic Acids Res 37, 3689-3698.

Roy Chowdhuri, S., Crum, T., Woollard, A., Aslam, S., and Okkema, P. G. (2006). The T-box factor TBX-2 and the SUMO conjugating enzyme UBC-9 are required for ABa- derived pharyngeal muscle in C. elegans. Dev Biol 295, 664-677.

Samaras, S. E., Cissell, M. A., Gerrish, K., Wright, C. V., Gannon, M., and Stein, R. (2002). Conserved sequences in a tissue-specific regulatory region of the pdx-1 gene mediate transcription in Pancreatic beta cells: role for hepatocyte nuclear factor 3 beta and Pax6. Mol Cell Biol 22, 4702-4713.

Scheidereit, C., Heguy, A., and Roeder, R. G. (1987). Identification and purification of a human lymphoid-specific octamer-binding protein (OTF-2) that activates transcription of an immunoglobulin promoter in vitro. Cell 51, 783-793.

Schnabel, R., Hutter, H., Moerman, D., and Schnabel, H. (1997). Assessing normal embryogenesis in Caenorhabditis elegans using a 4D microscope: variability of development and regional specification. Development 184, 234-265.

Schroeder, L. K., Kremer, S., Kramer, M. J., Currie, E., Kwan, E., Watts, J. L., Lawrenson, A. L., and Hermann, G. J. (2007). Function of the Caenorhabditis elegans ABC transporter PGP-2 in the biogenesis of a lysosome-related fat storage organelle. Mol Biol Cell 18, 995-1008.

Schuurmans, C., Armant, O., Nieto, M., Stenman, J. M., Britz, O., Klenin, N., Brown, C., Langevin, L. M., Seibt, J., Tang, H., et al. (2004). Sequential phases of cortical specification involve Neurogenin-dependent and -independent pathways. Embo J 23, 2892-2902.

Seipel, K., Yanze, N., and Schmid, V. (2004). Developmental and evolutionary aspects of the basic helix-loop-helix transcription factors Atonal-like 1 and Achaete-scute homolog 2 in the jellyfish. Dev Biol 269, 331-345.

172

Seydoux, G., Mello, C. C., Pettitt, J., Wood, W. B., Priess, J. R., and Fire, A. (1996). Repression of gene expression in the embryonic germ lineage of C. elegans. Nature 382, 713-716.

Shetty, P., Lo, M. C., Robertson, S. M., and Lin, R. (2005). C. elegans TCF protein, POP- 1, converts from repressor to activator as a result of Wnt-induced lowering of nuclear levels. Dev Biol 285, 584-592.

Shimizu, T., Toumoto, A., Ihara, K., Shimizu, M., Kyogoku, Y., Ogawa, N., Oshima, Y., and Hakoshima, T. (1997). Crystal structure of PHO4 bHLH domain-DNA complex: flanking base recognition. Embo J 16, 4689-4697.

Shu, X., Shaner, N. C., Yarbrough, C. A., Tsien, R. Y., and Remington, S. J. (2006). Novel chromophores and buried charges control color in mFruits. Biochemistry 45, 9639- 9647.

Simeckova, K., Brozova, E., Vohanka, J., Pohludka, M., Kostrouch, Z., Krause, M. W., Rall, J. E., and Kostrouchova, M. (2007). Supplementary nuclear receptor NHR-60 is required for normal embryonic and early larval development of Caenorhabditis elegans. Folia Biol (Praha) 53, 85-96.

Simmer, F., Tijsterman, M., Parrish, S., Koushika, S. P., Nonet, M. L., Fire, A., Ahringer, J., and Plasterk, R. H. (2002). Loss of the putative RNA-directed RNA polymerase RRF- 3 makes C. elegans hypersensitive to RNAi. Curr Biol 12, 1317-1319.

Singh, R. N., and Sulston, J. E. (1978). Some observations on moulting in Caenorhabditis elegans. Nematologica 24, 63-71.

Smith, P. A., and Mango, S. E. (2006). Role of T-box gene tbx-2 for anterior foregut muscle development in C. elegans. Dev Biol.

Smith, P. A., and Mango, S. E. (2007). Role of T-box gene tbx-2 for anterior foregut muscle development in C. elegans. Dev Biol 302, 25-39.

Spieth, J., and Lawson, D. (2006). Overview of gene structure. WormBook, 1-10.

Stringham, E. G., Dixon, D. K., Jones, D., and Candido, E. P. (1992). Temporal and spatial expression patterns of the small heat shock (hsp16) genes in transgenic Caenorhabditis elegans. Mol Biol Cell 3, 221-233.

Sugiura, M., Fuke, S., Suo, S., Sasagawa, N., Van Tol, H. H., and Ishiura, S. (2005). Characterization of a novel D2-like dopamine receptor with a truncated splice variant and a D1-like dopamine receptor unique to invertebrates from Caenorhabditis elegans. J Neurochem 94, 1146-1157.

Sulston, J. E., Schierenberg, E., White, J. G., and Thomson, J. N. (1983). The embryonic cell lineage of the nematode Caenorhabditis elegans. Devel Biol 100, 64-119.

173

Suzuki, M., Sagoh, N., Iwasaki, H., Inoue, H., and Takahashi, K. (2004). Metalloproteases with EGF, CUB, and thrombospondin-1 domains function in molting of Caenorhabditis elegans. Biol Chem 385, 565-568.

Sze, J. Y., Victor, M., Loer, C., Shi, Y., and Ruvkun, G. (2000). Food and metabolic signalling defects in a Caenorhabditis elegans serotonin-synthesis mutant. Nature 403, 560-564.

Sze, J. Y., Zhang, S., Li, J., and Ruvkun, G. (2002). The C. elegans POU-domain transcription factor UNC-86 regulates the tph-1 tryptophan hydroxylase gene and neurite outgrowth in specific serotonergic neurons. Development 129, 3901-3911.

Tapscott, S. J. (2005). The circuitry of a master switch: Myod and the regulation of skeletal muscle gene transcription. Development 132, 2685-2695.

Thellmann, M., Hatzold, J., and Conradt, B. (2003). The Snail-like CES-1 protein of C. elegans can block the expression of the BH3-only cell-death activator gene egl-1 by antagonizing the function of bHLH proteins. Development 130, 4057-4071.

Timmons, L., Court, D. L., and Fire, A. (2001). Ingestion of bacterially expressed dsRNAs can produce specific and potent genetic interference in Caenorhabditis elegans. Gene 263, 103-112.

Timmons, L., and Fire, A. (1998). Specific interference by ingested dsRNA. Nature 395, 854.

Titani, K., Torff, H. J., Hormel, S., Kumar, S., Walsh, K. A., Rodl, J., Neurath, H., and Zwilling, R. (1987). Amino acid sequence of a unique protease from the crayfish Astacus fluviatilis. Biochemistry 26, 222-226.

Uchida, O., Nakano, H., Koga, M., and Ohshima, Y. (2003). The C. elegans che-1 gene encodes a zinc finger transcription factor required for specification of the ASE chemosensory neurons. Development 130, 1215-1224.

Wang, P., Zhao, J., and Corsi, A. K. (2006). Identification of novel target genes of CeTwist and CeE/DA. Dev Biol 293, 486-498.

Warren, C. E., Krizus, A., and Dennis, J. W. (2001). Complementary expression patterns of six nonessential Caenorhabditis elegans core 2/I N-acetylglucosaminyltransferase homologues. Glycobiology 11, 979-988.

Watada, H., Scheel, D. W., Leung, J., and German, M. S. (2003). Distinct gene expression programs function in progenitor and mature islet cells. J Biol Chem 278, 17130-17140.

174

Weigel, D., Bellen, J. J., Jurgens, G., and Jackle, H. (1989a). Primordium specific requirement of the homeotic gene fork head in the developing gut of the Drosophila embryo. Roux's Archives of Developmental Biology 198, 201-210.

Weigel, D., Jurgens, G., Kuttner, F., Seifert, E., and Jackle, H. (1989b). The homeotic gene fork head encodes a nuclear protein and is expressed in the terminal regions of the Drosophila embryo. Cell 57, 645-658.

Wenick, A. S., and Hobert, O. (2004). Genomic cis-regulatory architecture and trans- acting regulators of a single interneuron-specific gene battery in C. elegans. Dev Cell 6, 757-770.

Wright, K. A., and Thomson, J. N. (1981). The buccal capsule of Caenorhabditis elegans (Nematoda: Rhabditoidea): an ultrastructural study. Canadian Journal Zool 59, 1952- 1961.

Xu, Z., Huang, S., Chang, L. S., Agulnick, A. D., and Brandt, S. J. (2003). Identification of a TAL1 target gene reveals a positive role for the LIM domain-binding protein Ldb1 in erythroid gene expression and differentiation. Mol Cell Biol 23, 7585-7599.

Yi, W., and Zarkower, D. (1999). Similarity of DNA binding and transcriptional regulation by Caenorhabditis elegans MAB-3 and Drosophila melanogaster DSX suggests conservation of sex determining mechanisms. Development 126, 873-881.

Yoshida, S., Ohbo, K., Takakura, A., Takebayashi, H., Okada, T., Abe, K., and Nabeshima, Y. (2001). Sgn1, a basic helix-loop-helix transcription factor delineates the salivary gland duct cell lineage in mice. Dev Biol 240, 517-530.

Yuan, J. Y., and Horvitz, H. R. (1990). The Caenorhabditis elegans genes ced-3 and ced- 4 act cell autonomously to cause programmed cell death. Dev Biol 138, 33-41.

Zhao, Z., Fang, L., Chen, N., Johnsen, R. C., Stein, L., and Baillie, D. L. (2005). Distinct regulatory elements mediate similar expression patterns in the excretory cell of Caenorhabditis elegans. J Biol Chem 280, 38787-38794.

Zheng, Q., Van Die, I., and Cummings, R. D. (2008). A novel alpha1,2- fucosyltransferase (CE2FT-2) in Caenorhabditis elegans generates H-type 3 glycan structures. Glycobiology 18, 290-302.

Zhong, M., Niu, W., Lu, Z. J., Sarov, M., Murray, J. I., Janette, J., Raha, D., Sheaffer, K. L., Lam, H. Y., Preston, E., et al. (2010). Genome-wide identification of binding sites defines distinct functions for Caenorhabditis elegans PHA-4/FOXA in development and environmental response. PLoS Genet 6, e1000848.

175

Zhu, J., Hill, R. J., Heid, P. J., Fukuyama, M., Sugimoto, A., Priess, J. R., and Rothman, J. H. (1997). end-1 encodes an apparent GATA factor that specifies the endoderm precursor in Caenorhabditis elegans embryos. Genes and Development 11, 2883-2896.

176

APPENDIX A: LIST OF OLIGONUCLEOTIDES

oGD Name Sequence (5’ to 3’) Description

CTACGGTACCTGCTACTGACTG reverse primer for B0507.1 oGD016 AAACT promoter

forward primer for phat-3 -102 bp oGD017 GTAAGTGGGTGAGAAAC deletion

forward primer for phat-3 -212 bp oGD018 CTCCCAGATTCATCCAC deletion

forward primer for phat-3 -326 bp oGD019 CTGCACAAAGTAACGTT deletion

ACAAGTCGACCACTGACTAACA B0507.1 forward primer, ~2.5 kb oGD025 ACG upstream

CACAAACTTGATGCGAATTCTC forward primer for phat-3 PGM1 oGD026 AAATGTATGCCC mutation

GGGCATACATTTGAGAATTCGC reverse primer for phat-3 PGM1 oGD027 ATCAAGTTTGTG mutation

GCCGGTACCTTTTGTTAGATTG oGD045 reverse primer for phat-5 promoter GTAG

TCGACATTTGAACAGGTGCATA oGD047 CATTTGAACAGGTGCATACATT for 3xPGM1 enhancer TGAACAGGTGCATG

AGAGCTGAAATGTTTAGTACTC forward primer for mutation of oGD048 CTTTAGCTCGTGG PGM1 at -560 bp in phat-1

177

CCACGAGCTAAAGGAGTACTAA reverse primer for mutation of oGD049 ACATTTCAGCTCT PGM1 at -560 bp in phat-1

CGTTAATTGTTTAGTACTATGG forward primer for mutation of oGD050 ACAGTATCTCA PGM1 at -69 bp in phat-1

TGAGATACTGTCCATAGTACTA reverse primer for mutation of oGD051 AACAATTAACG PGM1 at -69 bp in phat-1

ACGCCTGCAGTCGATTGATCTA oGD62 forward primer for phat-5 promoter CTAC

CACCTGTTCAAATGTATGCACC oGD63 TGTTCAAATGTATGCACCTGTT for 3xPGM1 enhancer CAAATG

AACTGCAGCGTTTTGCCCTGTG forward primer for hlh-6::YFP oGD64 ATCTAC reporter

CATAACCGGTATCATAGCATTA reverse primer for hlh-6::YFP oGD65 TTACTCGAAT reporter

TGATACAGTTTGTGCCTGTTCA forward primer for PGM1 mutation oGD67 AATTGTCGAAA in phat-5

TTTCGACAATTTGAACAGGCAC reverse primer for PGM1 mutation oGD68 AAACTGTATCA in phat-5

AACTGCAGAACAATTGGATTGC oGD71 forward primer for lys-8 promoter TCACAGG

CGGGGTACCTCTAGAAAATGTT oGD72 reverse primer for lys-8 promoter CAACTTGAATG

AACTGCAGGACACCAAGTTAAT oGD73 forward primer for pqn-8 promoter TGAAATCATCTC

178

CGGGGTACCTTTGAAACGGAAA oGD74 reverse primer for pqn-8 promoter GGGTAAATC

TGAATCAACCAACAGGCATGCC forward primer for PGM1 mutation oGD81 AACAT in B0507.1

ATGTTGGCATGCCTGTTGGTTG reverse primer for PGM1 mutation oGD82 ATTCA in B0507.1

CGGGATCCAATTTATAATTTCG forward primer for PGM1 mutation oGD83 GAAAACATTATC at -180 bp in lys-8 promoter

CGGGATCCGACAAGCGGTACTT reverse primer for PGM1 mutation oGD84 GCCA at -180 bp in lys-8 promoter

CGGGATCCGAATTCAACGATGA forward primer for PGM1 mutation oGD85 TTCATCG in pqn-8 promoter

CGGGATCCTTCAAACATTGAGC reverse primer for PGM1 mutation oGD86 TGCTTGT in pqn-8 promoter

TTATACATTTGAGAATGGGGTC for detection of tm299 allele in hlh- oGD97 TACTCGAC 6 gene

AAGTTGAACGTAAAAGGATCCC forward primer for PGM1 mutation oGD98 ATAATCGCTTT at -452 bp in lys-8 promoter

AAAGCGATTATGGGATCCTTTT forward primer for PGM1 mutation oGD99 ACGTTCAACTT at -452 bp in lys-8 promoter

for amplification of hlh-6 cDNA CGGGATCCATGTCAATTTCCCA oGD102 starts at 5' ATG of cDNA AAACAACTT (inclusive) for amplification of hlh-6 cDNA GGGATATCTCACATAGCATTAT oGD106 ends at 3' TGA of cDNA TACTCGAAT (inclusive)

179

TGCAAACTTGATGCACCTGTTC for gel shift of PGM1, taken from oGD137 AAATGTATGCCC phat-3 promoter

TGGGGCATACATTTGAACAGGT for gel shift of PGM1, taken from oGD138 GCATCAAGTTTG phat-3 promoter

AAGGGCCCGTACGGCCGACTAG reverse primer for GFP to use in oGD164 TAGG PCR stitching

AGCTTGCATGCCTGCAGGTCGA forward primer for GFP to use in oGD186 CT PCR stitching

GGTAAGTTTAAACAGATATCTA for insertion of intron in hlh-6 oGD198 CTAACTAACCCTGATTATTTAA cDNA; "minigene" ATTTTCAGTAC GTACTGAAAATTTAAATAATCA for insertion of intron in hlh-6 oGD199 GGGTTAGTTAGTAGATATCTGT cDNA; "minigene" TTAAACTTACC

CCATCTGTTCTAGAAATTAACG forward primer for flanking oGD248 TT mutation in PGM1 of phat-1

AACGTTAATTTCTAGAACAGAT reverse primer for flanking oGD249 GG mutation in PGM1 of phat-1

for amplification of phat-5 cDNA TAACCGGTATGATTGCAATCAT oGD258 starts at 5' ATG of cDNA TTCAATTCTCTTC (inclusive)

TAGAATTCTTAGCAAATTCTGC for amplification of phat-5 cDNA oGD259 ATGTGGATCCAC ends at TAA of cDNA (inclusive)

CCCATTTGCAGCCAAATAAACA forward primer for PGM1 mutation oGD275 TT at -580 bp in lys-8 promoter

AATGTTTATTTGGCTGCAAATG reverse primer for PGM1 mutation oGD276 GG at -581 bp in lys-8 promoter

180

AACTGCAGCGGATCCTGAATTT forward primer for ~2 kb Y8A9A.2 oGD362 CCTGTG promoter

ATACCGGTTCTGGGGGTACCCT reverse primer for Y8A9A.2 oGD363 ATAAAAA promoter

CACCTGTTCTCCTGAAAGCACC oGD364 TGTTCTCCTGAAAGCACCTGTT for 3xPGM1 with mutant flank CTCCTG TCGACAGGAGAACAGGTGCTTT oGD365 CAGGAGAACAGGTGCTTTCAGG for 3xPGM1 with mutant flank AGAACAGGTGCATG

TGAGATACTGTCTATCTGTTCA forward primer for E-box mutation oGD366 AACAATTAACG in PGM1 of phat-1

CGTTAATTGTTTGAACAGATAG reverse primer for E-box mutation oGD367 ACAGTATCTCA in PGM1 of phat-1

AACCTGTTCAAATGTATGAACC oGD384 TGTTCAAATGTATGAACCTGTT for 3xPGM1 with mutant E-box CAAATG TCGACATTTGAACAGGTTCATA oGD385 CATTTGAACAGGTTCATACATT for 3xPGM1 with mutant E-box TGAACAGGTTCATG CACCTGTTTTCAAATGTATGCA oGD386 CCTGTTTTCAAATGTATGCACCT for 3xPGM1 with "TT" GTTTTCAAATG TCGACATTTGAAAACAGGTGCA oGD387 TACATTTGAAAACAGGTGCATA for 3xPGM1 with "TT" CATTTGAAAACAGGTGCATG

CACCACCGGTATGCTGGTAAGT forward primer for amplification of oGD531 CTAGAAATTATT genomic egl-1

TTCACGGCCGCACATCTGGTGT reverse primer for amplification of oGD532 TGCAGGC genomic egl-1

181

forward primer for PCR of AAGGTACCCATGGTGAGCAAG oGD570 mCherry, to be in frame with GGCGAG PHAT-5 protein

CCGAATTCTTACTTGTACAGCT reverse primer for PCR of oGD571 CGTCCATGCC mCherry, includes stop codon

CAAAAATTCCAGATATGCTTTT forward primer for T05B4.13 oGD727 CATTTT stitching

AGTCGACCTGCAGGCATGCAAG reverse primer for T05B4.13 oGD728 CTATCCTGAACAATAGCTGC stitching

GTTATGGAAACAGTTTCCATTA forwardprimer for B0280.7 oGD729 TCG stitching

AGTCGACCTGCAGGCATGCAAG reverse primer for B0280.7 oGD730 CTATAGCATCTCATTTCAAT stitching

TTTGAAACCGTTTATTCTGCTCT forward primer for ZK596.1 oGD733 CTG stitching

AGTCGACCTGCAGGCATGCAAG reverse primer for ZK596.1 oGD734 CTATTTTGGTTACTTCCAAT stitching

TTTTGAATCTGGAAATTTATGG forward primer for Y76B12C.3 oGD735 CT stitching

AGTCGACCTGCAGGCATGCAAG reverse primer for Y76B12C.3 oGD736 CTCATTTTGCTGGAAATTTTTA stitching oGD753 CTTTCGCTGGGTTAATATAAAC forward primer for npp-15 stitching

AGTCGACCTGCAGGCATGCAAG oGD754 CTTCTGAAATTTCAGTGATTAG reverse primer for npp-15 stitching AA

182

forward primer for -406 bp nas-12 oGD755 TGGGAAATGTTTCAAATCACCT stitching

AGTCGACCTGCAGGCATGCAAG oGD756 CTTATTATTGTTCGAAATATCG reverse primer for nas-12 stitching AAC oGD844 ACAGGGCTCTGCGTCTAAACTT forward primer for scl-3 stitching

AGTCGACCTGCAGGCATGCAAG oGD845 CTATACAACTAGTAGGTTGGAT reverse primer for scl-3 stitching ATTTATAGG

CTAGAAAGTGTAAAAAATTATC oGD846 forward primer for ucr-2.3 stitching AAAAACA

AGTCGACCTGCAGGCATGCAAG oGD847 CTTAAATAACTAATTTACGGAA reverse primer for ucr-2.3 stitching AAAAACAAT

forward primer for Y47D3A.14 oGD848 ACGAGCCGTCCACTTGGAATTT stitching

AGTCGACCTGCAGGCATGCAAG reverse primer for Y47D3A.14 oGD849 CTTTGCAAATCTGAAAAAAAAC stitching TGTTTAAAAA

forward primer for -201 bp nas-12 oGD866 TTTGATCGTCATTAATCTCGA stitching

AACTGCAGAATGGTTTTTAGAT forward primer for Y8A9A.2 -1005 oGD869 TTTGATAATAA bp promoter cloning

AACTGCAGACCTCTGCGATGTT forward primer for Y8A9A.2 -503 oGD870 TCCT bp promoter cloning

AACTGCAGCAAAAATTCCAGAT forward primer for T05B4.13 oGD908 ATGCTTTTCATTTT reporter plasmid cloning

183

CGGGGTACCATCCTGAACAATA reverse primer for T05B4.13 oGD909 GCTGC reporter plasmid cloning

reverse primer for mutation in both oGD915 TGTCTGTTCCAGTGGATG E-boxes of T05B4.13

CATCCACTGGAACAGACACTGA forward primer for mutation in both oGD952 ACGG E-boxes of T05B4.13

AACTGCAGCAAAACCAACAAA forward primer for Y8A9A.2 -276 oGD953 TATCAAATGGGGTG bp cloning

TTCATTTTGTGTGCTCAGCCGAT forward primer for -118 bp nas-12 oGD980 T stitching

forward primer for -93 bp nas-12 oGD981 CATCTCTTTCTCCTCCTCCAC stitching

AACTGCAGTTCATTTTGTGTGCT forward primer for -118 bp nas-12 oGD1003 CAGCCGATT reporter plasmid

ATACCGGTTATTATTGTTCGAA forward primer for nas-12 reporter oGD1004 ATATCGAAC plasmid

AACTGCAGTTCATTTGGATCCC forward primer for mutation in PBS oGD1005 TCAGCCGATT in -118 bp nas-12 plasmid

AACTGCAGTGTACGTGTTGGTT forward primer for Y8A9A.2 -196 oGD1006 TAGCTGCAT bp cloning

AACTGCAGGAAAAAAACGGGT forward primer for Y8A9A.2 -170 oGD1007 GGTTGGCAG bp cloning

AGCTTATTTTGTGTGCTCAGATT oGD1009 TTGTGTGCTCAGATTTTGTGTGC for 3x PBS from nas-12 TCAGGCATG

184

CCTGAGCACACAAAATCTGAGC oGD1010 ACACAAAATCTGAGCACACAA for 3x PBS from nas-12 AATA forward primer for mutation in AACTGCAGTTCATTTTGTGTGCT oGD1019 flanking sequence in -118 bp nas- TTGCCGATT 12 plasmid

AACTGCAGTGTACGCCTTGGTT forward primer for mutation in PBS oGD1020 TAGCTGCAT in -196 bp Y8A9A.2 plasmid

forward primer for mutation in AACTGCAGTGTACGTGTTGGTT oGD1021 flanking sequence in -196 bp TTTTTGCAT Y8A9A.2 plasmid AGCTTTACGTGTTGGTTTAGCT oGD1026 ACGTGTTGGTTTAGCTACGTGT for 3x PBS from Y8A9A.2 TGGTTTAGCGCATG CGCTAAACCAACACGTAGCTAA oGD1027 ACCAACACGTAGCTAAACCAAC for 3x PBS from Y8A9A.2 ACGTAA

185

APPENDIX B: LIST OF PLASMIDS

pGD Name Plasmid Summary Constructed By

pGD025 phat-3::YFP Jeb Gaudet

pGD026 phat-3::YFP with mutant E-box Leanne Sayles

pGD027 B0507.1::GFP::HIS2B Jeb Gaudet

pGD031 phat-1::YFP Ryan Smit

pGD045 phat-1::YFP mutant PGM1 at -560 bp Ryan Smit

pGD046 phat-1::YFP mutant PGM1 at -69 bp Ryan Smit

pGD048 phat-5::YFP Ryan Smit

pGD049 3xPGM1 Ryan Smit

pGD082 phat-5::YFP with mutant PGM1 Ryan Smit

pGD083 B0507.1::GFP::HIS2B with mutant PGM1 Ryan Smit

186 pGD084 pqn-8::YFP Indra Raharjo

pGD085 pqn-8::YFP with mutant PGM1 Indra Raharjo

pGD086 lys-8::YFP Indra Raharjo

pGD087 lys-8::YFP with mutant PGM1 at -180 bp Indra Raharjo

pGD110 hs::HLH-6 Ryan Smit

pGD159 hlh-6::HLH-6 + synthetic intron, "minigene" Jeb Gaudet

pGD168 GST::HLH-6 Ryan Smit

pGD199 lys-8::YFP with mutant PGM1 at -581 bp Shoubin Wen

pGD246 Y8A9A.2::GFP Shoubin Wen

pGD297 lys-8::PHAT-5::YFP Indra Raharjo

phat-1::YFP with mutation in E-box (CAnnTG to pGD329 Ryan Smit TAnnTG) pGD365 lys-8::YFP with mutation in PGM1 at -452 bp Ryan Smit

187 pGD370 3x PGM1 with mutant flank Ryan Smit

pGD371 3x PGM1 with mutant E-box Ryan Smit

pGD372 3x PGM1 with "TT" between E-box and flank Ryan Smit

pGD378 lys-8::PHAT-5::Cherry Ryan Smit

pGD442 hlh-6::PHAT-5::mCherry Ryan Smit

pGD443 myo-2::PHAT-5::mCherry Ryan Smit

pGD517 T05B4.13::GFP with mutation in both E-boxes Ryan Smit