Elucidating TOR Function in Kalanchoe Daigremontiana Plantlet

Home , Kalanchoe daigremontiana

Elucidating TOR Function in Kalanchoë

daigremontiana Plantlet Formation

A thesis submitted to the University of Manchester for the

degree of Master of Philosophy in the Faculty of Biology,

Medicine and Health

2019

Kirsty McCready

School of Biological Sciences

TABLE OF CONTENTS LIST OF FIGURES AND TABLES ...... 4 LIST OF ABBREVIATIONS ...... 5 ABSTRACT ...... 6 DECLARATION...... 7 COPYRIGHT STATEMENT ...... 8 ACKNOWLEDGMENTS ...... 9 1 Introduction ...... 10 1.1 Plant growth and development in response to nutrient availability ...... 10 1.2 The Plant TOR kinase complex ...... 11 1.2.1 The Arabidopsis TOR gene ...... 11 1.2.2 The TOR protein...... 11 1.2.3 The TOR complex ...... 12 1.2.4 REGULATORY ASSOCIATED PROTEIN OF AtTOR ...... 13 1.2.5 LETHAL WITH SEC THIRTEEN 8 ...... 14 1.3 TOR inhibition in plants ...... 14 1.3.1 Genetic alteration of TOR expression ...... 14 1.3.2 Chemical inhibition of TOR expression ...... 15 1.4 TOR function during plant development ...... 15 1.4.1 Embryogenesis ...... 15 1.4.2 Germination and Seedling Growth ...... 16 1.4.3 Meristem development ...... 19 1.4.4 Plant and Leaf size ...... 21 1.4.5 Flowering ...... 21 1.5 Triggered pluripotency in the species Kalanchoë daigremontiana ...... 23 1.5.1 The Kalanchoë genus ...... 23 1.5.2 Plantlet formation in the species Kalanchoë daigremontiana ...... 24 1.5.3 K. daigremontiana plantlet formation shares features of organogenesis and embryogenesis ...... 25 1.5.4 Plantlet initiation requires organogenesis meristem maintenance genes ...... 25 1.5.5 Plantlet formation follows an embryo-like developmental program ...... 26 1.5.6 Plantlet formation and the TOR kinase ...... 28 1.6 MPhil Aims...... 30 2 Materials and Methods ...... 31

2.1 Plant Growth Conditions ...... 31 2.2 Treatment Conditions ...... 31 2.3 Cloning and Plant Transformation ...... 31 2.3.1 Primer Design ...... 31 2.3.2 DNA Extraction ...... 32 2.3.3 Vector Construction ...... 32 2.3.4 Golden Gate Assembly ...... 33 2.3.5 Kalanchoë daigremontiana Transformation ...... 34 2.4 Genotyping ...... 35 2.5 RNA Extraction ...... 35 2.6 cDNA Synthesis, RT-PCR and qRT-PCR...... 36 2.7 ß-Glucoronidase (GUS) Staining ...... 37 2.8 Phylogenetic Tree Construction ...... 37 2.9 Image acquisition and Data Analysis ...... 38 3 Results ...... 39 3.1 TOR, RAPTOR and LST8 phylogenies are congruent with the species trees ...... 39 3.2 KdTOR expression varied across plantlet developmental stages ...... 40 3.2 TOR::GUS expression mirrors qPCR pattern across developmental stages ...... 43 3.2.1 promoterTOR::GUS lines were generated ...... 43 3.3 Torin2 application to the leaf margin inhibited plantlet formation ...... 46 3.4 Genetic suppression of TOR abolished plantlet formation...... 47 3.4.1 35S::TORantisense construct was assembled ...... 47 ...... 48 3.4.2 35S::KdTORantisense transgenic lines were produced ...... 49 3.4.3 Repression of tor severely restricted plantlet formation ...... 49 3.4.4 35S::KdTORantisense displayed changes in phyllotaxy and meristem growth ...... 53 3.5 35S::KdSnRK1antisense construct was assembled ...... 54 4 Discussion ...... 55 5 Conclusion ...... 63 6 References ...... 65

Final Word Count: 16480

LIST OF FIGURES AND TABLES

Figure 1: The plant TOR Complex (TORC)……………………………………………………….…………..13 Figure 2: Upstream and downstream targets of the TOR Complex (TORC) in plants…..18 Figure 3: Phylogeny of the Kalanchoë genus………………………………………………………………24 Table 1: Golden Gate Modules amplified by PCR and ligated into pGEM®T-Easy……….34 Table 2: Primer pairs designed for qRT-PCR…………………………………………………..…………..36

Figure 4: Phylogenetic trees of the plant TORC1 components, constructed from amino acid sequence alignments of the a) TOR, b) RAPTOR1 and c) LST8 genes.…………………………………..40 Figure 5: Stages of Kalanchoë daigremontiana plantlet formation and expression of KdTOR and KdRAPTOR1……………………………………………………………………………….…………….42 Figure 6: Cloning of the KdTORpromoter region ~1.5 kb upstream of KdTOR and Agrobacterium tumefaciens transformation……………………………………………………..……….43 Figure 7: KdpTORpromoter::GUS expression during plantlet development…………....….45 Figure 8: Torin2 and mock treatment of Kalanchoë daigremontiana plantlet formation…………………………………………………………………………………………………………………..46 Figure 9: Generation of 35S::KdTORantisense…………………………………………………………….48 Figure 10: Genotypic and phenotypic analysis of 35S::KdTORantisense………………..……50 Figure 11: Microscope images of 35S::KdTORantisense plantlet development stages..52 Figure 12: Phyllotaxy and meristem phenotypes in 35S::KdTORantisense………………….53 Figure 13: Generation of 35S::KdSnRK1antisense……………………………………………………….54

LIST OF ABBREVIATIONS 5’TOP 5’ Terminal Oligopyrimidine tract 5’ uORF 5’ upstream Open Reading Frame ABA Abscisic Acid ABI3 ABSCISIC ACID INSENSITIVE 3 AP1/2/3 APETALA 1/2/3 ARF AUXIN RESPONSE FACTOR asTORi active site TOR inhibitor bHLH basic Helix-Loop-Helix BIN2 BRASSINOSTEROID INSENSITVE 2 BPEp BIGPETALp bZIP11 Bovine Serum Albumen CAM Crassulacean Acid Metabolism CDK CYCLIN DEPENDENT KINASE CLV1/2/3 CLAVATA 1/2/3 COP1 CONSTITUTIVE PHOTOMORPHOGENESIS 1 DMSO Dimethyl Sulfoxide EIF3H EUKARYOTIC TRANSLATION INITIATION FACTOR 3H FATC FRAP, ATM and TRRAP C FKBP12 FK506 BINDING PROTEIN 12 FRB FKBP Rapamycin-Binding Domain FUS3 FUSCA 3 GA Gibberellic Acid GAPDH GLYCERALDEHYDE-3 PHOSPHATE DEHYDROGENASE GFP Green Fluorescent Protein GUS ß-Glucuronidase HEAT found in Huntingtin, Elongation factor 3, A subunit of protein phosphatase 2A, and TOR1 IFM Inflorescence Meristem KNOX KNOTTED1-LIKE HOMEOBOX KOG1 KONTROLLER OF GROWTH 1 L1L LEC1-LIKE LEC1/2 LEAFY COTYLEDON 1/2 LFY LEAFY LST8 LETHAL WITH SEC THIRTEEN 8 MS Murashige and Skoog NAP NAC-LIKE, ACTIVATED BY AP3/PI OFP OVATE FAMILY PROTEIN PI PISTILLATA PIN1 PIN-FORMED 1 PP2A PROTEIN PHOSPHATASE 2A qRT-PCR quantitative Reverse Transcriptase-Polymerase Chain Reaction RAM Root Apical Meristem RAPTOR1/2 REGULATORY ASSOCIATED PROTEIN OF TOR 1/2 RICTOR RAPTOR INDEPENDENT COMPANION OF TOR RPS6 RIBOSOMAL PROTEIN S6 S6K S6 KINASE SAM Shoot Apical Meristem SMR SIAMESE-RELATED SNF1 Suc NON-FERMENTING 1 SnRK1 SNF1-RELATED KINASE 1 SOC1 SUPRESSOR OF OVEREXPRESSION OF CONSTANS 1 STM SHOOTMERISTEMLESS TAP46 TYPE 2A-PHOSPHATASE-ASSOCIATED PROTEIN 46KD TOR TARGET OF RAPAMYCIN WOX5 WUSCHEL-RELATED HOMEOBOX 5 WT Wild Type WUS WUSCHEL YAK1 YET ANOTHER KINASE YUC YUCCA

ABSTRACT

Plants retain the remarkable ability to continually adapt their development in response to nutrient availability, to maximise survival and fitness in a changing environment. The TARGET OF RAPAMYCIN (TOR) kinase has been characterised as a master regulator of the nutrition sensing pathway in all eukaryotes. TOR kinase senses nutrient and energy status to direct growth and development. Understanding of plant TOR complex components (TOR, LST8 and RAPTOR) and upstream and downstream targets has advanced tremendously in the past decade and TOR signalling has been implicated in a myriad of plant developmental processes. However, there is a significant lack of TOR characterisation beyond model plant species. Kalanchoë daigremontiana has evolved the ability to regain totipotency in the serrations of the leaf margin, from which develop plantlets. The mechanism behind the unique phenomenon of plantlet formation remains elusive but is known to recruit key regulators of organogenesis and embryogenesis and respond to favourable environmental conditions. We propose TOR kinase as the central regulator of plantlet formation, integrating external signals and developmental pathways to first trigger pluripotency, then direct plantlet development. Here we show that KdTOR inhibition both chemically (Torin2 application) and genetically (generation of 35S::KdTORantisense lines) significantly reduces plantlet formation. Downregulation of KdTOR also produces mutant phenotypes in the plantlets. We show, by qRT-PCR analysis and generation of KdTORp::GUS lines, that KdTOR is expressed during plantlet development. Taken together, expression and functional analyses suggest that KdTOR signalling is most crucial at the earliest stages of plantlet formation (to trigger pluripotency) and again at the initiation of the plantlet at the leaf pedestal. This work reveals the recruitment of the conserved TOR signalling pathway to a novel plant developmental process.

DECLARATION

I declare the that work referred to in this thesis has not been submitted in support of an application for another degree or qualification of this or any other university or other institute of learning, excluding: Kalanchoë Torin2 treatment and generating 35S::KdTORantisense and pKdTOR::GUS have been submitted both by myself for a Bachelor of Science (BSc) degree and by Victoria Spencer for a Doctor of Philosophy (PhD) degree in the Faculty of Biology, Medicine and Health, University of Manchester, 2018.

i. The author of this thesis (including any appendices and/or schedules to this thesis) owns certain copyright or related rights in it (the “Copyright”) and s/he has given The University of Manchester certain rights to use such Copyright, including for administrative purposes. ii. Copies of this thesis, either in full or in extracts and whether it is in hard or electronic copy, may be made only in accordance with the Copyright, Designs and Patents Act 1988 (as amended) and regulations issued under it or, where appropriate, in accordance with licensing agreements which the University has from time to time. This page must form part of any such copies made. iii. The ownership of certain Copyright, patents, designs, trademarks and other intellectual property (the “Intellectual Property”) and any reproductions of copyright works in the thesis, for example graphs and tables (“Reproductions”), which may be described in this thesis, may not be owned by the author and may be owned by third parties. Such Intellectual Property and Reproductions cannot and must not be made available for use without the prior written permission of the owner(s) of the relevant Intellectual Property and/or Reproductions. iv. Further information on the conditions under which disclosure, publication and commercialisation of this thesis, the Copyright and any Intellectual Property and/or Reproductions described in it may take place is available in the University IP Policy (see http://documents.manchester.ac.uk/DocuInfo.aspx?DocID=2442 0), in any relevant Thesis restriction declarations deposited in the University Library, The University Library’s regulations (see http://www.library.manchester.ac.uk/about/regulations/) and in The University’s policy on Presentation of Theses.

ACKNOWLEDGMENTS

I have been supported by numerous people throughout this MPhil, both academically and personally, whom without it would not have been possible. First and foremost, I would like to thank my supervisor, Dr. Minsung Kim, for his support and advice throughout my MPhil and undergraduate final year project. During my undergraduate project, I was fortunate enough to be supervised by Victoria Spencer, and without her constant support, advice and enthusiasm, I would never have been inspired to continue the project as an MPhil. I would like to thank Vicky for first, inspiring me to explore this area of plant science, and secondly for her continued support during its undertaking.

From the Kim Lab, I am thankful to Joo Phin Ooi and Paco Jácome Blásquez for teaching me many of the molecular biology techniques I have used throughout this project. I would specifically like to thank Joo Phin for giving me her Kalanchoë daigremontiana DNA and RNA at the outset of this project, and Paco for his help with tissue culture. I would also like to thank the students and staff of the plant science department for their continued advice and support, particularly Armida-Irene Gjindali and Holly Allen, for their encouragement inside and outside of the lab.

1 Introduction

1.1 Plant growth and development in response to nutrient availability As a result of their sessile nature, plants cannot escape hostile environments unfavourable to their growth and development. Yet, despite their immobility, they have succeeded in the colonization of a considerable diversity of habitats. Such resilience can be attributed to their remarkable ability to continually adapt (Palmer et al., 2012). While retaining the same basic plan, module shape and physiology is continually adapted in response to developmental and environmental cues, made possible by the presence of populations of stem cells at the Shoot and Root Apical Meristems (SAM and RAM, respectively) (de Jong and Leyser, 2012; Pfeiffer et al., 2016). Through tightly controlled cell division and expansion, a complex molecular network of internal developmental cues, nutrients and hormones coordinate to direct organ and tissue growth across the whole organism (Gonzalez et al., 2012; Powell and Lenhard, 2012). This long distance signal transduction is vital to allow communication between organs with distinct metabolic and developmental functions that are subject to different light, temperature, nutrient and humidity conditions [See Review: (Vanstraelen and Benkova, 2012)].

Our understanding of development in controlled environments is advanced, however our knowledge of how developmental pathways respond to changes in environmental conditions is less clear. Nevertheless, sugars have been highlighted as key signalling molecules for environmental conditions, allowing energy status to be monitored using several signalling systems [See Review: (Lastdrager et al., 2014)]. Crucially, shoot meristem-derived glucose has been reported to drive TARGET OF RAPAMYCIN (TOR) signalling relays to control root meristem activation, independently of direct glucose signalling (Xiong et al., 2013). Furthermore, it has been reported that glucose is sufficient to induce TOR activity in four-day-old Arabidopsis seedlings, and that TOR is necessary for the glucose-dependent exit from mitotic quiescence in young seedlings (Xiong and Sheen, 2012, 2014). It is therefore perhaps unsurprising that TOR has been suggested as a central regulator, responsible for integrating nutrient and energy signals to control development.

The TOR protein kinase is gaining traction as a key developmental regulator, integrating extracellular signals (hormones, growth factors, biotic and abiotic stresses) together with intracellular nutrient and energy status to direct growth via control of protein synthesis and anabolic and catabolic processes (Albert and Hall, 2015). The past decade has seen a significant increase in the understanding of plant TOR complex components, upstream signals and downstream targets, and the relationship between TOR signalling and stress (Schepetilnikov and Ryabova, 2018). Thus, we are finally beginning to understand how TOR is activated, and how it goes on to control cellular processes.

1.2 The Plant TOR kinase complex

1.2.1 The Arabidopsis TOR gene Since the discovery of the TOR inhibitor rapamycin from the soil bacterium Streptomyces hygroscopicus (Sehgal et al., 1975), and its use to identify and isolate TOR in yeast (Heitman et al., 1991; Kunz et al., 1993), mammals (Sabatini et al., 1994) and plants (Menand et al., 2002), our knowledge and understanding of TOR signalling mechanisms and function has progressed immensely. Nevertheless, the study of plant TOR has been largely limited to the model plant Arabidopsis thaliana [and select few other plant species, such as rice and bean (De Vleesschauwer et al., 2018; Nanjareddy et al., 2016)] and further investigation is crucial if we are to fully elucidate TOR function in diverse developmental processes across the plant kingdom.

A single TOR gene exists in Arabidopsis, Chlamydomonas reinhardtii, most animals and humans (Xiong and Sheen, 2014). The A. thaliana TOR (AtTOR) gene spans at least 17 kb of genomic DNA, composed of 56 exons and 55 introns (Menand et al., 2002). Human and AtTOR share 75% amino acid sequence similarity in the protein kinase domain, suggesting conservation of the protein kinase properties and substrates between the two species (Menand et al., 2002).

1.2.2 The TOR protein TOR encodes a highly conserved Ser/Thr kinase, belonging to the phosphatidylinositol 3-kinase-related kinase family (Heitman et al., 1991). The complex domain organisation of AtTOR is conserved amongst metazoans, mammals, and plants [Figure 1; (Menand et al., 2002; Zoncu et al., 2011)]. The TOR N-terminus consists of HEAT (Huntingtin,

Elongation Factor 3, protein phosphatase 2A, Tor1) repeat motifs, followed by FAT (FRAP, ATM, TRRAP), FRB (FKBP12-Rapamycin Binding), Ser/Thr kinase catalytic and FATC (FAT Carboxy terminus) domains (Fig. 1). The HEAT repeats are necessary for substrate recruitment and membrane association, the Ser/Thr kinase domain mediates phosphate transfer, the FRB domain is responsible for binding rapamycin in other eukaryotes, whilst together the FATC and FAT domains contribute to kinase activation (Menand et al., 2002).

1.2.3 The TOR complex In mammals, TOR exists in two structurally and functionally distinct complexes, mammalian TARGET OF RAPAMYCIN COMPLEX1 (mTORC1) and mTORC2 (Jacinto et al., 2004; Wullschleger et al., 2006). The two complexes differ in their components and have distinct substrates and functions. The core components of mTORC1 are mTOR, REGULATORY-ASSOCIATED PROTEIN OF mTOR (RAPTOR) and SMALL LETHAL WITH SEC THIRTEEN 8 (mLST8) (Gonzalez and Hall, 2017; Saxton and Sabatini, 2017). The second complex, mTORC2, likewise contains mLST8, but RAPTOR is not present. Instead, RAPAMYCIN-INSENSITIVE COMPANION OF mTOR (RICTOR) is the distinctive component of mTORC2. Whilst the FKBP12-rapamycin complex is capable of direct binding and inhibition of the mTORC1 complex, it cannot bind and inhibit mTORC2, except when exposed to prolonged rapamycin treatment (Jacinto et al., 2004; Sarbassov et al., 2004; Sarbassov et al., 2006).

All sequenced plant species possess orthologs of the RAPTOR and LST8 genes [Fig. 1; (Anderson et al., 2005; Deprost et al., 2005; Diaz-Troya et al., 2008; Duan et al., 2006; Mahfouz et al., 2006; Moreau et al., 2012)]. In contrast, no RICTOR ortholog has been found in plant genomes (Tatebe and Shiozaki, 2017). This suggests that TORC1, but not TORC2, is conserved in plants. Furthermore, many of the interactions between TOR complex components in animals appear to be conserved in plants [See review: (Dobrenel et al., 2016)]. It will be interesting to determine whether complexes form with as yet unidentified plant components, and whether complex formation is dependent on tissue type or developmental stage; raptor mutant embryos for example abort at a later stage than tor mutants (Deprost et al., 2005; Menand et al., 2002), suggesting RAPTOR- independent function in the embryo.

TOR

Figure 1: The plant TOR Complex (TORC). TOR (TARGET OF RAPAMYCIN) protein organisation and association with complex components RAPTOR (REGULATORY ASSOCIATED PROTEIN OF AtTOR) and LST8 (LETHAL WITH SEC THIRTEEN 8). TOR consists of HEAT (Huntingtin, Elongation Factor 3, protein phosphatase 2A, TOR1) repeat motifs, FAT (FRAP, ATP, TRRAP), FRB (FKBP12- Rapamycin binding), Ser/Thr Kinase catalytic and FATC (FAT Carboxy terminus) domains. The HEAT repeats interact directly with RAPTOR, whilst LST8 has been shown to interact with the kinase domain.

1.2.4 REGULATORY ASSOCIATED PROTEIN OF AtTOR

Two homologs of the mammalian RAPTOR gene exist in Arabidopsis, RAPTOR1 (or RAPTOR1B, AT3G08850) and RAPTOR2 (or RAPTOR1A, AT5G01770) (Anderson et al., 2005; Deprost et al., 2005). Both sequences are well conserved among eukaryotes and show 80% protein similarity throughout their length (Anderson et al., 2005). In silico analyses reveal that RAPTOR1 is highly expressed throughout development, whereas RAPTOR2 expression is markedly lower. It has been suggested that RAPTOR2 arose as a result of a duplication of the ancestral RAPTOR gene in the higher plant lineage and is as such a redundant copy (Deprost et al., 2005). raptor2 mutants display no phenotypic defects (Deprost et al., 2005), further supporting redundancy.

As described in other eukaryotic species, plant RAPTOR proteins contain three conserved blocks in the N-terminal region, three highly conserved heat motifs in the C- terminal, followed by seven WD-40 repeats (Deprost et al., 2005). Co- immunoprecipitation experiments in tobacco leaves confirmed that RAPTOR1 binds TOR via its three heat motifs (Mahfouz et al., 2006). Furthermore, transmission electron microscopy has confirmed RAPTOR WD-40 repeats interact with the TOR N-terminal (Adami et al., 2007). Future work into plant-specific RAPTOR interactions would prove informative for elucidating any direct interactions with the TORC1 complex in plant development.

1.2.5 LETHAL WITH SEC THIRTEEN 8 All sequenced plant genomes contain a functional LST8 gene (Moreau et al., 2012). As in other eukaryotes, the Arabidopsis LST8 (AtLST8) protein contains seven WD-40 repeats. GUS reporter expression analyses reveal that LST8 is expressed throughout plant development, particularly in the aerial parts of the plant. AtLST8 shares 45% sequence identity with mammalian LST8, and 51% with yeast LST8. Yeast lst8 mutants expressing the AtLST8 coding sequence were able to grow normally, demonstrating that despite low sequence similarity, AtLST8 is a functional homolog of yeast LST8 with conserved function. As with RAPTOR, the interaction of LST8 with plant-specific components will reveal potential pathways by which developmental phenotypes arise.

1.3 TOR inhibition in plants 1.3.1 Genetic alteration of TOR expression Historically, the study of TOR in photosynthetic organisms has been hampered by the embryo lethality of Arabidopsis null tor mutants, the inefficiency of inducible tor mutants, and the relative insensitivity of plant TOR to rapamycin inhibition (Baena- Gonzalez and Sheen, 2008; Moreau et al., 2012). Embryonic arrest of Arabidopsis upon disruption of the TOR gene was first reported in 2002, indicating a crucial role for TOR in early plant development (Menand et al., 2002). The issue of embryo lethality was subsequently circumvented via the generation of T-DNA insertion lines to give overexpression of the AtTOR gene, and RNA interference (RNAi) lines with reduced AtTOR expression (Deprost et al., 2007). The morphology of Arabidopsis plants with increased or decreased levels of AtTOR messenger RNA (mRNA) were subsequently investigated. They demonstrated a positive correlation between the level of AtTOR expression and Arabidopsis growth, seed yield, osmotic stress resistance, abscisic acid (ABA) and polysome accumulation. Several publications have since confirmed that reduced shoot and root growth and decrease in cell size results from a reduction in TOR activity (Caldana et al., 2013; Liu and Bassham, 2010; Ren et al., 2012; Xiong and Sheen, 2012), and that a converse increase in TOR transcript levels gives rise to an increase in growth and cell size (Ren et al., 2012).

1.3.2 Chemical inhibition of TOR expression Alongside advances in the genetic study of plant TOR, chemical study has also progressed. Overexpression of either yeast or human FK506 BINDING PROTEIN 12 (FKBP12) enhanced sensitivity of AtTOR to rapamycin treatment by up to 100-fold (Ren et al., 2012; Sormani et al., 2007; Xiong and Sheen, 2012). Previously, sequence differences in protein residues of plant FKBP12 prevented the formation of a TOR- rapamycin-FKBP12 ternary complex, hindering the ability of rapamycin to inhibit TOR (Mahfouz et al., 2006; Sormani et al., 2007). Furthermore, active site TOR inhibitors (asTORis) developed in animal and medical research, such as Torin2 and AZD-8055 are capable of successful plant TOR inhibition (Schepetilnikov et al., 2013). Nevertheless, most plant TOR studies to date have been restricted to early seedling growth in the model plant Arabidopsis (Pfeiffer et al., 2016; Xiong et al., 2013). There is therefore a requirement for the study of plant TOR across developmental stages, and across diverse plant species, in order to fully elucidate TOR function in the plant kingdom.

1.4 TOR function during plant development 1.4.1 Embryogenesis In higher plants, seed formation is characterized by double fertilisation of the female gametophyte, giving rise to two distinct tissues: the zygote and the endosperm. The zygote will develop into the embryo, whilst the endosperm surrounds and nourishes the embryo as it develops (Dumas and Rogowsky, 2008). Disruption of AtTOR has been shown to interfere with the development of both tissues. The endosperm typically grows as a syncytium until it reaches around 200 nuclei, before cellularisation. The Arabidopsis tor mutant endosperm reaches approximately 48 (±13) cells and cellularisation does not occur (Menand et al., 2002). Embryos of Arabidopsis null tor mutants arrest early at the dermatogen stage, with cells in metaphase still present. While cell division itself is thus not inhibited by the disruption of AtTOR in the embryo, cell expansion is supressed (Menand et al., 2002). This is consistent with wide-scale downregulation of translation machinery and cell wall modifying enzymes such as CELLULASE SYNTHASE 6 (CESA6) and EXPANSIN B1 (EXPB1) after AtTOR inhibition (Xiong et al., 2013).

Evidence to suggest the involvement of RAPTOR1 in embryogenesis in the literature is conflicted. Mutation of raptor1 via T-DNA insertion gave rise to viable embryos, suggesting that AtTOR function in embryogenesis is independent of RAPTOR1 (Anderson et al., 2005). However, further work found the same line (SALK_078159) to be embryo lethal (Deprost et al., 2005). These differences may be explained by incomplete penetrance due to environmental effects. The varying light and temperature conditions in which they were grown could affect the phenotypic severity. A detailed phenotypic analysis of two independent raptor1b mutants (SALK_101990 and SALK_022096) during seed development and germination found that raptor1b mutant seeds displayed delayed germination and reduced stress resistance, resulting in reduced viability (Salem et al., 2017). Furthermore, morphological changes were observed, including reduced seed-coat pigmentation and mucilage production, accompanied by significant changes in metabolic content, such as an increase in free amino acids, and a decrease in protective secondary metabolites and storage proteins. This is consistent with the transcriptional reprogramming of gene sets involved in central and secondary metabolism in response to glucose-AtTOR signalling in seedlings (Xiong et al., 2013). Alongside these changes, significant alterations in phytohormone levels of raptor1b mutant seeds were generally observed. Of particular interest were the concurrent increases in abscisic acid, auxin and jasmonic acid, all known to inhibit germination (Salem et al., 2017). Together these observations imply that AtTOR and its components are central to the regulation of embryogenesis, seed metabolism and maturation.

1.4.2 Germination and Seedling Growth For germination to succeed, seeds must integrate and respond to signals such as light, temperature, water, nutrients and hormones. Whilst the endosperm provides nutrition initially, seedlings must switch to photoautotrophic growth, as photosynthetically- derived glucose is a major nutrient for cellular and organismal development in plants (Xiong et al., 2013). The sequential events controlling seed germination are well characterized (Bentsink and Koornneef, 2008), however AtTOR signalling has been implicated as a key mediator of the process (Xiong et al., 2013). To drive this transition from heterotrophic to photoautotrophic growth, glucose‐AtTOR signalling activates broad gene sets involved in the cell cycle and anabolic processes, and suppresses gene

16 sets controlling catabolic processes (Xiong et al., 2013). This in turn activates root growth via glycolysis-mitochondria-ETC (electron transport chain) relays (Xiong et al., 2013). Furthermore, photosynthesis-derived sugars are necessary for hormones (auxin, brassinosteroid, cytokinin and gibberellin) to promote rapid root elongation and reactivate the quiescent root during this transition to photoautotrophy (Xiong et al., 2013).

Glucose can also activate AtTOR indirectly via the inactivation of SUC NON-FERMENTING 1-RELATED KINASE 1 (SnRK1), a conserved glucose/energy sensor protein kinase [Fig. 2; (Baena-Gonzalez and Sheen, 2008)]. Arabidopsis SnRK1 (AtSnRK1) forms a heterotrimeric complex with the catalytic subunits KIN10 and KIN11 (Baena-Gonzalez and Sheen, 2008). In contrast to TORC1, the AtSnRK1-KIN10/11 complex is suppressed by glucose, but activated by starvation, energy deprivation, and abiotic stresses (Baena- Gonzalez and Sheen, 2008). Furthermore, KIN10 has been shown to directly interact with and phosphorylate RAPTOR (Nukarinen et al., 2016). Thus, TORC1 and AtSnRK1 dominate a complex network, acting antagonistically to direct growth according to nutrient and energy availability in the organism.

Nutrients such as nitrogen (N), phosphate (Pi) and sulphur (S) play crucial roles in the promotion of plant growth and development and recent studies suggest that AtTOR functions in these processes. For example, S availability has been shown to coordinate glucose signalling to activate AtTOR (Dong et al., 2015). Furthermore, nitrate, a major N source, behaves as a nutrient signal to promote system-wide shoot and root growth in Arabidopsis (Liu et al., 2017). Notably, Arabidopsis seedlings modified to overexpress AtTOR show hypersensitivity to high nitrate inhibition of roots (Deprost et al., 2007). By sensing the nutrient content in the cell, the AtTOR kinase is able to act as a molecular switch, initiating growth at a time when sufficient resources are available for healthy plant development. Whether these processes operate in the autotrophic to photoautotrophic transition is yet to be determined.

The inhibition of AtTOR expression in germinating seedlings, via either asTORis application or genetic suppression, reduces cotyledon greening, chloroplast development and seedling growth (Deprost et al., 2007; Dong et al., 2015; Li et al., 2015; Xiong et al., 2017). AtTOR and ABA signalling have been shown to interact

Figure 2: Upstream and downstream targets of the TOR Complex (TORC) in plants. Upstream regulators of plant TORC1: Light, glucose and nutrients are known activators of the TOR pathway. Light activates the TOR pathway via the inactivation of the negative regulator COP1, triggering the activation of the auxin pathway, leading to TOR activation during seed de-etiolation. Light is also known to trigger GTP-ROP2/auxin activation of TOR signalling in the shoot apex. Light and glucose coordinate to inactivate the TOR antagonist SnRK1, leading to the indirect activation of TORC. Downstream targets of plant TORC1: Direct phosphorylation targets of TORC1 include PP2A (via the subunit TAP46), E2FA/B and S6K, leading to the activation of cellular processes throughout development. elF3H and RPS6 are direct phosphorylation targets of S6K-P. Plant TORC also interacts with 45SrRNA to promote ribosome biogenesis and polysome loading. The YAK1 kinase is inhibited by active TOR, relieving the inhibition of CYC/CDKs to allow cell proliferation in the meristem. Solid arrows indicate direct interaction, dashed arrows indicate indirect interaction.

18 antagonistically in the promotion of seedling development, and the Arabidopsis 40S ribosomal PROTEIN S6 KINASE (S6K) is thought to further promote chloroplast development and seedling growth via the regulation of BR INSENSITIVE 2 (BIN2) [See review: (Shi et al., 2018)].

An exciting link has been made between light and the activation of AtTOR-RPS6 (RIBOSOMAL PROTEIN S6) in de-etiolating seedlings. In a sophisticated pathway, light is first perceived by photoreceptors such as phytochrome A and cryptochromes, leading to the inactivation of the negative regulator CONSTITUTIVE PHOTOMORPHOGENESIS 1 (COP1), which triggers the activation of the auxin pathway and thus AtTOR-dependent phosphorylation of RPS6 [Figure 2; (Chen et al., 2018)]. Suppression of the auxin-AtTOR- RPS6 pathway can then promote enhanced de novo protein synthesis. Mutant seedlings lacking functional AtTOR, RPS6A or RPS6B displayed delayed cotyledon opening, a characteristic vital to the de-etiolation process. This study reveals the complex layers of regulation to ensure a timely switch from strict skotomorphogenic development in the dark to photomorphogenic development upon light detection in de-etiolating seedlings.

1.4.3 Meristem development Plants are capable of the continuous production of new organs throughout their lifetime in a plastic, environment-dependent manner, made possible by the presence of the SAM and the RAM. Whilst it is clear that meristem activity is controlled by environmental conditions, the exact mechanism remains uncertain. For example, the meristem will terminate during leaf initiation in the absence of light, even when photosynthate is provided (Xiong et al., 2013). The meristem is not in direct contact with environmental triggers such as light, and therefore it is likely that such signals are perceived in distant organs and relayed to direct meristem activity.

AtTOR is a known essential regulator of cell proliferation in both the SAM and the RAM, as supported by AtTOR localisation in these tissues (Menand et al., 2002). Consistently, delayed shoot growth has been observed in tor knockdown and raptor1 mutant lines, which could be a result of disrupted meristem activity (Li et al., 2017; Menand et al., 2002; Pfeiffer et al., 2016; Xiong et al., 2013). Recent data shows that AtTOR-E2FA phosphorylation activates the RAM by activating S phase, whilst AtTOR, activated by light-Auxin-Rho-like small GTPase (ROP) signalling, phosphorylates both E2FA and E2FB

19 to activate S phase in the SAM [Fig. 2; See Review: (Shi et al., 2018)]. AtTOR is thus a likely candidate for the integrator of environmental signals to direct meristem activity in both roots and shoots.

YET ANOTHER KINASE 1 (YAK1) has recently been reported as a downstream target of the AtTOR pathway and a major regulator of meristem activity (Barrada et al., 2019).YAK1 was discovered through a pharmaco-genetic screen of mutants with resistance to the asTORis AZD-8055. yak1 loss-of-function mutants are resistant to AZD- 8055, whilst Arabidopsis overexpressing YAK1 is hypersensitive to AtTOR inhibition. In the WT plant, RAM cells first proliferate to increase root length, before they differentiate and subsequently expand. Treatment of WT plants with AZD-8055 reduces primary root length and induces early differentiation. Conversely, in the three yak1 loss-of-function mutants identified through screening, root length was not reduced by treatment with AZD-8055. Furthermore, treatment of WT plants with inhibitors of the mammalian ortholog of YAK1 (DYRK1) phenocopies yak1 loss-of-function. The generation of YAK1 reporter lines revealed that, in the absence of AtTOR activity, YAK1 induces the expression of SIAMESE-RELATED (SMR) genes, which in turn repress CYCLIN-DEPENDENT KINASES (CDKs) to promote differentiation. Contrarily, YAK1 inhibition by AtTOR kinase promotes growth, by lifting the repression of CDKs and cyclin, to maintain proliferation (Fig. 2). Thus, AtTOR-YAK1 interaction plays a crucial role in the regulation of RAM activity and maintenance.

Patterning of the SAM has also been connected to AtTOR activity. A population of stem cells is maintained at the SAM via the interaction of WUSCHEL (WUS) and CLAVATA (CLV) genes [See review: (Somssich et al., 2016)]. Interestingly, when grown in the asTORis AZD-8055, the expression of WUS in WUS::GFP seedlings decreased (Pfeiffer et al., 2016). This suggests that AtTOR may activate WUS expression in the meristem, thus promoting meristem activity in favourable conditions. However, the seedlings were grown for three days in liquid culture containing the inhibitor before WUS expression was measured. It is therefore difficult to determine to what extent AtTOR is controlling WUS expression, or whether long term metabolic changes are responsible for WUS activation. Conversely, there is little evidence to suggest that AtTOR regulates meristem patterning in the RAM. No expression changes of the root meristem patterning gene,

WUSCHEL RELATED HOMEOBOX 5 (WOX5) were observed when treated with asTORis (Xiong et al., 2013). Further determining the exact role of AtTOR in the SAM and RAM will prove critical for understanding environment dependent meristem activity.

1.4.4 Plant and Leaf size AtTOR is also involved in size regulation. ß-estradiol inducible and ethanol inducible AtTOR silencing plants show a reduction in plant biomass, including reduced cell size and ultimately reduced leaf size (Deprost et al., 2007; Xiong and Sheen, 2012), consistent with a T-DNA raptor1 SALK line (Anderson et al., 2005). Accordingly, lst8 mutants show similar reductions in growth, as well as increased shoot branching (Moreau et al., 2012). Ser-Thr PROTEIN PHOSPHATASE 2A (PP2A) contains a conserved regulatory subunit TAP46 (TYPE 2A-PHOSPHATASE-ASSOCIATED PROTEIN 46KD) (TAP42 in yeast), which is directly phosphorylated by AtTOR [Fig. 2; (Ahn et al., 2011)]. Disruption of TAP46 expression results in global translation defects, decreased polysome accumulation and methionine incorporation, and in turn smaller plants as above (Ahn et al., 2015). Furthermore, a recent study has confirmed that TOR inhibition with asTORis prevents leaf primordia initiation, causing a reduction in leaf number (Mohammed et al., 2018).

Conversely, overexpression of both TAP46 and AtTOR results in larger seeds and plants (Ahn et al., 2015; Deprost et al., 2007), with bigger leaves due to larger epidermal cells and longer petioles. AtTOR domain overexpression lines possess twisted leaves and siliques (Ahn et al., 2011; Deprost et al., 2007). Together these studies clearly indicate the involvement of AtTOR in leaf development, however it is unclear whether AtTOR only directly controls global cell cycle regulators and cell growth machinery (Li et al., 2017; Xiong et al., 2013), or affects leaf development genes such as the OVATE FAMILY PROTEINS (OFPs) (Wang et al., 2011) to target specific leaf development pathways.

1.4.5 Flowering Disruption to TOR, RAPTOR1 and LST8 delays flowering time in Arabidopsis (Anderson et al., 2005; Deprost et al., 2007; Moreau et al., 2012). Myriad external and internal factors, such as plant age, sugar availability, photoperiod and temperature, control the transition to flowering time [See review: (Cho et al., 2017)]. These signals converge on factors such as LFY to convert the SAM into an Inflorescence Meristem (IFM) (Blazquez

21 et al., 1997). Clearly, there are multiple pathways in play to ensure that flowering time is regulated robustly. There is therefore scope for the involvement of AtTOR as a regulator in this known process. Alternatively, AtTOR could regulate flowering independently, perhaps indirectly through metabolic changes. Future work is necessary to determine where AtTOR fits in to the flowering time pathway, thus allowing a link to be established between the observed AtTOR delayed flowering phenotypes and its role in the complex network controlling flowering time.

In the Floral Meristem (FM), each floral organ primordium is patterned according to the classic ABC model [See review: (Irish, 2017)]. Mutation of LST81 produces flowers with smaller floral organs, but no changes in organ patterning or number have been reported (Moreau et al., 2012), suggesting that LST81 may be independent from the ABC gene patterning, but affect later development and organ growth. Abnormal flower phenotypes have also been recorded in raptor1 SALK lines, but not described fully (Anderson et al., 2005). Furthermore, tor knockdown flowers have yet to be investigated, and are necessary to determine whether flower development is under the control of AtTOR as well as LST8 and RAPTOR1. The study of the relationship of flower specific genes to AtTOR will reveal whether the phenotypes are due to direct changes to cell cycle/growth genes, and/or interactions with genes specific to floral development. For example, through AUXIN RESPONSE FACTOR 8 (ARF8), auxin inhibits cell expansion in the distal part of the petal by interacting with the basic helix-loop-helix (bHLH) gene, BIGPETALp (BPEp) (Szecsi et al., 2006; Varaud et al., 2011). Whether such pathways interact with AtTOR signalling remains to be determined.

Interestingly, ectopic expression of Lily S6K (LIS6K) in A. thaliana produces flowers with shortened petals and stamens, due to reduced cell expansion and normal cell division (Tzeng et al., 2009). S6K is a conserved target of TOR in plants, which binds to RAPTOR and in turn activates EIF3H [Fig. 2; (Mahfouz et al., 2006; Schepetilnikov et al., 2013)]. EIF3H phosphorylation promotes loading of mRNAs with untranslated 5’ upstream Open Reading Frames (uORFs) into the ribosome for translation re-initiation (Fig. 2). Plant specific genes such as AUXIN RESPONSE FACTORS (ARFs) and BASIC LEUCINE ZIPPER 11 (bZIP11) are both known targets of this process (Fig. 2). Similarly, S6K activation

22 upregulates translation of transcripts with 5’ Terminal Oligopyrimidine Tracts (5’ TOPS) [Fig. 2; (Schepetilnikov et al., 2013)]. The ABC genes APETALA3 (AP3) and PISTILLATA (PI) have been shown by bioinformatic analysis to contain 5’ TOPS, leading to their increased translation following activation of S6K (Tzeng et al., 2009). AP3 and PI then activate transcription of NAC-LIKE ACTIVATED BY AP3/PI (NAP). Arabidopsis flowers overexpressing NAP have short petals and stamens (Sablowski and Meyerowitz, 1998), mirroring Lily S6K overexpression. It is therefore plausible that AtTOR acts through S6K to promote flower development via the regulation of AP3, PI and NAP. However, a direct link has yet to be proven.

1.5 Triggered pluripotency in the species Kalanchoë daigremontiana 1.5.1 The Kalanchoë genus In order to maximise fitness in diverse environments, plants have evolved a wide range of reproductive strategies. Of these, the evolution of the flower in the angiosperm lineage is perhaps the most striking, an innovation that facilitated the rapid dispersal and divergence of the angiosperms (Moyroud and Glover, 2017). Whilst the evolution of sexual reproduction is highly advantageous, the development of an alternative asexual reproductive strategy in some lineages has also proved successful. Species in the genus Kalanchoë represent one such lineage (Garces and Sinha, 2009b). Whilst many of the species within this genus are still able to flower, they can also produce clonal individuals (termed plantlets) from differentiated leaf tissue, which detach and grow independently from the mother leaf. The mechanisms and triggers for this process vary amongst Kalanchoë species, perhaps as a result of the different ecological contexts in which they evolved. Some species cannot produce plantlets (e.g. K. marmorata); some produce plantlets only upon stress induction (e.g. K. pinnata); and others are capable of constitutive plantlet formation (e.g. K. daigremontiana). Phylogenetic analyses suggest that species lacking plantlets represent the ancestral state, followed by stress-induced species, whilst constitutive-plantlet-forming species with non-viable seed are the most recent innovation [Fig. 3; (Garces et al., 2007)].

Figure 3: Phylogeny of the Kalanchoë genus. Species in black represent the ancestral seed- producing lineage that do not make plantlets, species in red are those who produce plantlets only upon stress induction, whereas species in blue represent the most recent innovation: constitutive plantlet formation and loss of viable seed. Images to the right depict K. marmorata (top), a species that does not make plantlets, K. pinnata (middle), a stress-induced plantlet forming species, and K. daigremontiana (bottom), a constitutive plantlet forming species. (Phylogenetic tree taken from Garcês (unpublished data), images taken from Google).

1.5.2 Plantlet formation in the species Kalanchoë daigremontiana The phenomenon of plantlet formation represents an incredibly unique reproductive strategy, requiring the reversion of mature, differentiated leaf cells into undifferentiated stem cells. Studies of the molecular mechanisms governing such cell fate change have largely focused on K. daigremontiana (Garces et al., 2007; Garces et al., 2014; Liu et al., 2016; Zhu et al., 2017). As a constitutive-plantlet-forming species, K. daigremontiana plantlets will always form within the plant’s lifetime. The timing of plantlet formation, however, is influenced by a combination of factors, such as the age of the plant, the maturity of the leaves, and environmental conditions such as day length and water availability (Liu et al., 2016). Plantlet formation in K. daigremontiana is promoted by long days and drought. Expression of the circadian clock gene SUPRESSOR OF OVEREXPRESSION OF CONSTANS 1 (SOC1) was found to correlate with such conditions, suggesting a role for SOC1 in the plantlet initiation mechanism. Furthermore, KdSOC1

24 overexpression leads to a reduction in plantlet formation, and an increase in PIN1 expression and auxin content (Zhu et al., 2017).

1.5.3 K. daigremontiana plantlet formation shares features of organogenesis and embryogenesis Once conditions are favourable, plantlet formation proceeds sequentially from the tip to the base of the leaf, within the serrations of the leaf margin [Fig. 4; (Johnson, 1934)]. Plantlets proceed through globular, heart-shaped and torpedo-shaped developmental stages, before forming starch-containing cotyledons with differentiated vascular system (Batygina et al., 1996). As such, plantlet formation can be said to follow an embryo-like developmental program. However, unlike embryos, which form distinct root and shoot apical poles, plantlets produce adventitious roots from the basal hypocotyl, resembling shoots (Garces et al., 2007). Plantlets detach from the mother leaf once the root system has developed, fall to the ground and grow into new plants. Via confocal imaging, Garcês and colleagues (2007) showed that plantlets retain a vascular system independent of the parent leaf throughout all stages of development, paralleling embryo development. Therefore, due to resemblance to both shoot and embryo formation, plantlet formation in K. daigremontiana is said to share features of both organogenesis and embryogenesis.

1.5.4 Plantlet initiation requires organogenesis meristem maintenance genes The initiation of plantlet formation requires the reversion of differentiated mature leaf cells to an undifferentiated stem cell fate. Organogenesis meristem maintenance genes are known to be involved in this process (Garces et al., 2007). SHOOTMERISTEMLESS (STM) is a class 1 KNOTTED1-LIKE HOMEOBOX (KNOX) gene responsible for maintaining stem cells in the SAM of Arabidopsis and preventing their differentiation into specific cell types (Endrizzi et al., 1996). A loss-of-function mutation in the Arabidopsis stm gene results in the loss of the SAM and arrest at the seedling stage (Long et al., 1996; Vollbrecht et al., 2000). Conversely, constitutive overexpression of KNOX1 results in the formation of ectopic shoots on leaves (Chuck et al., 1996; Sentoku et al., 2000; Sinha et al., 1993). In situ hybridisation analysis of K. daigremontiana during plantlet development showed that KdSTM is highly expressed in the plant SAM and axillary buds. KdSTM mRNA was also detected in the small group of cells at the leaf margin that will undergo plantlet formation, and, later in development, KdSTM expression increased in

25 the vascular bundles at the heart-like embryo stage, and in the upper half of the developing plantlet into the cotyledon-like leaves (Garces et al., 2007). In contrast, STM expression in Arabidopsis zygotic embryos is restricted to a few cells at the globular stage, and never in cotyledon primordia. KdSTM localization is instead reminiscent of maize somatic embryogenesis, where the expression domain of KNOX1 genes appears to be broader than in maize zygotic embryogenesis (Long and Barton, 1998; Smith et al., 1995; Zhang et al., 2002). Furthermore, down-regulation of KdSTM via RNAi suppression abolished plantlet formation in seven out of eight transgenic lines (Garces et al., 2007). In the one line where plantlet formation could still occur, KdSTM mRNA was detected in the leaf margins, suggesting incomplete suppression. Notably, STM expression is consistently absent in the leaf margins of species that do not produce plantlets. Together, these observations strongly suggest that KdSTM is necessary for plantlet formation in K. daigremontiana.

Expression of WUS, another meristem maintenance gene, was found to increase as leaves of K. daigremontiana matured (Guo et al., 2015). However, a link between WUS function and plantlet formation was not investigated. Interestingly, KdSOC1 overexpression reduced plantlet formation, but showed no change in KdWUS expression, providing indirect evidence that KdWUS is not involved in plantlet formation(Zhu et al., 2017). Thus, in order to understand meristem formation in the K. daigremontiana leaf margin, more detailed functional analyses of KdWUS and other meristem identity genes (e.g. CLV genes) is required.

1.5.5 Plantlet formation follows an embryo-like developmental program Following organogenesis initiation at the leaf margin, constitutive plantlet development closely resembles embryogenesis. Owing to this, the genetic pathways underlying Arabidopsis embryogenesis have been explored in K. daigremontiana (Garces et al., 2007). In Arabidopsis, LEAFY COTYLEDON1 (LEC1), LEC2 and FUSCA3 (FUS3) are expressed during embryo morphogenesis and maturation and are known to control reserve accumulation and seed desiccation tolerance (Meinke, 1992; Meinke et al., 1994; Parcy et al., 1997; Vicient et al., 2000; West et al., 1994). Ectopically expressing LEC1 and LEC2 in Arabidopsis can induce somatic embryo formation in vegetative cells, not unlike plantlet formation in Kalanchoë (Garces et al., 2007; Lotan et al., 1998; Stone

26 et al., 2001). Thus, the vital role of LEC1 in Arabidopsis embryo development suggests that it may likewise be necessary for K. daigremontiana plantlet formation.

Alike Arabidopsis, expression of the KdLEC1 homolog is detected at the zygotic torpedo- like stage of plantlet formation, as well as in the early and heart-like stages (Garces et al., 2007). Furthermore, KdLEC1 is not detected in the SAM, demonstrating that KdLEC1 is a marker for embryogenesis in K. daigremontiana. However, downregulation of Kdlec1 does not affect plantlet formation, suggesting that KdLEC1 is not required for plantlet formation. Furthermore, when compared with Arabidopsis, the KdLEC1 protein is truncated. In the C-terminal region of the B domain, 11 unique amino acids take the place of a 20-nucleotide deletion, introducing a premature stop codon in the KdLEC1 gene. KdLEC1 cannot rescue the lec1 mutant when expressed in Arabidopsis. The Atlec1 mutant embryo is desiccation-intolerant and cannot produce viable seed. However, restoration of the KdLEC1 B region with 20 nucleotides from the Arabidopsis LEC1-LIKE (L1L) restores the ability of Arabidopsis to produce viable seed. The data suggest that the truncated form of LEC1 is necessary for constitutive plantlet formation, to bypass plantlet dormancy. In accordance with this, stress-induced plantlet-forming species such as K. pinnata, do not undergo typical embryogenesis-like development and possess a functional LEC1. Furthermore, expression of a functional chimeric LEC1 gene in K. daigremontiana resulted in developmental defects in leaf embryos, even inducing characteristics of seed dormancy (Garces et al., 2014). Truncated LEC1 is therefore essential for K. daigremontiana leaf embryos to bypass dormancy and develop into viviparous plantlets on the leaf margin. LEC2 is known to play a similar role to LEC1 in Arabidopsis (Lotan et al., 1998; Stone et al., 2001), so an investigation into KdLEC2 function in plantlet formation would be interesting.

Alike KdLEC1, KdFUS3 is expressed in plantlets, but not in the Kalanchoë SAM (Garces et al., 2007; Luerßen et al., 1998) and could therefore control plantlet formation. KdFUS3 silencing lines produce fewer plantlets per leaf, supporting a role for KdFUS3 in control of plantlet formation (Kim Lab, Unpublished data). LEC1 and FUS3 are both controlled by and control the plant hormones Gibberellic Acid (GA) and ABA [See review: (Braybrook and Harada, 2008)]. FUS3 inhibits GA biosynthesis (Gazzarrini et al., 2004), whilst ABA indirectly activates FUS3 (Kagaya et al., 2005a) and LEC1 expression induces

ABSCISIC ACID INSENSITIVE 3 (ABI3) and FUS3 expression (Kagaya et al., 2005b). Such a complexity of interactions suggests that LEC1, LEC2, FUS3, GA and ABA form a highly connected network to control embryo maturation. How this network has been co-opted to K. daigremontiana plantlet formation has yet to be elucidated.

LEC2 and FUS3 are known to bind the promoter of YUCCA 4 (YUC4), an auxin biosynthesis gene, during lateral root development to stimulate auxin production (Tang et al., 2017). Furthermore, growth on exogenous auxin leads to increased FUS3 expression (Gazzarrini et al., 2004). It has therefore been suggested that LEC2 and FUS3 control K. daigremontiana plantlet formation in an auxin-dependent manner. Since KdSOC1 overexpression leads to upregulation of KdPIN1 (Zhu et al., 2017), LEC2 and FUS3 perhaps act in response to SOC1 activation.

1.5.6 Plantlet formation and the TOR kinase Although there remains much to elucidate with regards to the molecular mechanisms governing K. daigremontiana plantlet initiation and development, it is clear that complex signalling pathways integrating photoperiod, hormones, organogenesis and embryogenesis are required. The TOR kinase provides an excellent candidate for the central regulator of this process. TOR kinase is a key developmental regulator, competent to integrate nutrient and energy availability to regulate cell growth, via control of embryogenesis and organogenesis in plants (Deprost et al., 2005; Menand et al., 2002; Moreau et al., 2012; Pfeiffer et al., 2016; Schepetilnikov et al., 2013; Wang et al., 2018), thus it is sensible to assume it plays a role transducing environmental signals to promote plantlet formation. Whilst significant progress has been made in the characterization of TORC1 signalling in animals and yeast, knowledge of plant TOR is comparatively lacking. Nevertheless, our understanding of plant TOR has boomed over the past years, with studies beginning to expand beyond the model plant Arabidopsis (De Vleesschauwer et al., 2018; Menand et al., 2002; Nanjareddy et al., 2016; Xiong et al., 2016). Evidence is clearly emerging that TOR has a conserved regulatory role in photosynthetic organisms, acting in conjunction with the antagonist SnRK1 to adapt growth and metabolism according to nutrient and hormone signals. Developmental pathways are highly interconnected, and it will be interesting to determine how these interact with TOR signalling. The synthesis of such processes will require bioinformatic

28 pathway analysis to build networks at the RNA expression, protein expression, and protein modification levels, for a complete understanding of TOR activity. Crucially, these signalling pathways may be even more critical in plants than in animals and yeast, since plant immobility prevents their escape from hostile environments or nutrient scarcity, placing increased importance on their developmental plasticity in response to the environment.

1.6 MPhil Aims This thesis aimed to determine whether the TARGET OF RAPAMYCIN (TOR) kinase complex plays the role of regulator in the initiation of plantlet formation in Kalanchoë daigremontiana. This will show if the conserved TOR kinase has been recruited to the phenomenon of triggered pluripotency in K. daigremontiana plantlet formation. Investigation of TOR function in a process unique to the Kalanchoë genus will provide insight into how the adaptation of TOR throughout plant evolution has accommodated developmental plasticity in response to nutrient availability.

TOR expression throughout plantlet development was visualised and genetic and chemical inhibition were used to locally supress TOR. Subsequent phenotypic and molecular changes were analysed. This project uncovers a unique role for TOR kinase in plantlet formation in a novel species, that has not been investigated before now.

In brief, this was achieved by:

i. Expression analyses of TOR in WT plantlet developmental stages ii. Chemical TOR inhibition at the site of plantlet formation iii. Generation of 35S::KdTORantisense and promoterTOR::GUS transgenic lines iv. Generation of 35S::KdSnRK1antisense transgenic lines

2 Materials and Methods

2.1 Plant Growth Conditions Wild-type (WT) and transgenic K. daigremontiana plantlets were individually potted in a 6:1:1 (v/v) mixture of Levington’s F2 compost (Scott’s Miracle Gro, Surrey, UK), vermiculite and perlite (Sinclair Horticulture Ltd., Lincoln, UK) and grown in a MLR-350 Versatile Environmental Test Chamber (Sanyo, Japan) at 23°C in long day conditions (16 hours light, 8 hours dark, 680 LUX).

2.2 Treatment Conditions

Six weeks after planting, the margins of the central pair of Plastochron 2 (P2) leaves were treated with Torin2 solutions [100 µM Torin2 (Sigma-Aldrich, UK), 0.5 % Dimethyl Sulfoxide (DMSO) (Sigma-Aldrich, UK) and 0.5 % Tween-20 (Applichem, USA)] and mock solutions [0 µM Torin2, 0.5% DMSO and 0.5 % Tween-20]. Solutions were brushed onto the leaf margins before plantlet formation, and leaf size recorded before application. Plantlet number was recorded every three days for 27 days. Plantlet number was recorded for one margin per leaf per plant, as the opposite leaf margin is identical. The results were separated into leaves greater than 3cm and less than 3cm for analysis.

2.3 Cloning and Plant Transformation 2.3.1 Primer Design All primers were designed with NCBI Primer BLAST (https://www.ncbi.nlm.nih.gov/tools/primer-blast/). Primers were designed using K. laxiflora and K. fedtschenkoi sequences (obtained from Phytozome v12.1, JGI, DOE). Primer quality was checked with the online IDT OligoAnalyzer software (https://www.idtdna.com/calc/analyzer); those with a Tm close to 60°C and self-dimer and hetero-dimer ∆G values above -5 kcal/mol were favoured.

Three primer pairs were designed to amplify the KdTORantisense and KdSnRK1antisense fragments, and the KdTORpromoter region from K. daigremontiana genomic DNA (gDNA). For antisense constructs, a 375 bp fragment of KdTOR exon 8 and a 156bp fragment of KdSnRK1 exon 3 were cloned using primers synthesized by Eurofins Genomics (Germany) (Table 1). The promoter fragment was designed to include 1466 bp upstream of the start codon and covered the entire non-coding region (Table 1).

Primers were designed to amplify the target sequence and add a Bsal Site and linker sequence to specify the order of insertion (Table 1).

2.3.2 DNA Extraction Genomic DNA (gDNA) was extracted from young leaves 1-3 cm in length according to the protocol in (Garces and Sinha, 2009a). gDNA was stored at -80°C until needed for Polymerase Chain Reaction (PCR).

2.3.3 Vector Construction Fragments were amplified by PCR with Q5® High-Fidelity DNA Polymerase (New England Biolabs, USA) in a T100TM Thermal Cycler (Bio-Rad UK). Reaction mixture consisted of 0.8 µM forward and reverse primer, 800 µM dNTP (Bioline, UK), 0.8 µM gDNA, 1X Q5® Buffer (New England Biolabs, USA) and 0.028 U/µL Q5® High-Fidelity DNA polymerase. Thermal cycling conditions were set according to the manufacturer’s instructions (New England Biolabs, USA), with specific modifications for each fragment. The KdTORantisense fragment was amplified with an annealing temperature of 60°C for 30 seconds. KdSnRK1antisense amplification was achieved at an annealing temperature of 52.6°C (15 seconds) and with an extension time of 30 seconds. The KdTORpromoter fragment was amplified with an annealing temperature of 55.1°C (30 seconds) and an extension time of 2 minutes.

For visualisation of PCR products, samples were run on an Agarose gel [Agarose (Bioline, UK), Tris-Acetate EDTA (TAE) (Millipore, UK)] in a Gel Electrophoresis Tank (Wolf Laboratories, UK) with a PowerPacTM HC (Bio-Rad, UK). 5 µL SafeView Nucleic Acid Stain (NBS Biologicals, UK) was added per 100 mL of agarose gel to allow DNA visualisation under UV light. Images were acquired using a Gel DocTM XR+ Machine (Bio-Rad, UK) with Image LabTM Software. HyperladderTM 100 bp and (Bioline, UK) and HyperladderTM 1 kb (Bioline, UK) were run in conjunction to allow size comparison. Antisense PCR products were run on a 2 % agarose gel for 25 minutes (100 V) and the promoter amplicon was run on a 0.8 % agarose gel for 35 minutes (70 V).

The three PCR products were purified by gel extraction (due to non-specific amplification of the PCR products) using the Macherey-Nagel Nucleospin® gel and PCR Clean-up Kit (GmbH & Co., Germany). The purified DNA product was then incubated with

BioTaqTM DNA Polymerase (Bioline, UK) and dATP (Bioline, UK) at 68°C for 30 minutes to add a single 3’ A overhang. The adenylated DNA was then ligated with the pGEM®T-Easy Vector (Promega, USA) following the pGEM®T-Easy Vector System I protocol (Promega, USA), and incubated overnight at 16°C. pGEM®T-Easy constructs were checked by digestion with EcoRI-HF (New England Biolabs, USA) for 1 hour at 37°C. Samples were sequenced by GATC Biotech (Eurofins, Germany).

2.3.4 Golden Gate Assembly Constructs were assembled via the Golden Gate Assembly Reaction (New England Biolabs, USA). For assembly of the antisense constructs, 100 ng/µL of the required Golden Gate pGEM®T-Easy modules [KdTORantisense/KdSnRK1antisense, 35S RNAi Promoter and Terminator and a modified pBI121 vector (referred to as pBI128)] were mixed with 1X Bovine Serum Albumen (BSA) (New England Biolabs, USA), 1X T4 Ligase Buffer (New England Biolabs, USA), 0.4 U/µL BsaI-HF (New England Biolabs, USA) and T4 Ligase (New England Biolabs, USA). The reaction mix was incubated at 37°C for 3 minutes, followed by 16°C for 4 minutes, repeated 40 times. The reaction was terminated at 50°C for 5 minutes, then 80°C for 5 minutes. For assembly of the KdTORpromoter construct, the Golden Gate pGEM®T-Easy modules KdTORpromoter, GUS and the Nopaline Synthase (Nos) Terminator were added to the reaction mix described above and assembled via the same reaction conditions. Constructs were then transformed into Escherichia coli, strain DH5α for selection.

Digestion checks of the assembled constructs were performed with EcoRI-HF and HindIII-HF for 1 hour at 37 °C. Further confirmation was achieved by culture PCR with M13 Forward and M13 Reverse primers (M13Forw GTTTTCCCAGTCACGAC; M13Rev CAGGAAACAGCTATGAC), followed by sequencing. Correct constructs were then transformed into Agrobacterium tumefaciens strain LBA4404 by electroporation.

Table 1: Golden Gate Modules amplified by PCR and ligated into pGEM®T-Easy. Primers are shown 5’ to 3’. The BsaI site (ggtctc) is highlighted in bold. The native sequence is in capital letters, with the linker sequence also highlighted in bold. Added spacer sequences are lower case. The product size includes the cloned region and one linker sequence.

Module Insert Primer Size p35SKd 35S CaMV promoter Forw gtggtctca GGAG GCTAGAGCAGCTTGCCAAC 833 Rev gtggtctca CACC GGTCGATCGACAGATCTGCG pKdTOR ~1.5 kb upstream of K. Forw gtggtctca GGAG 1467 daigremontiana TOR AGGTGAATTTCAAAGCAATCATGC

Pro Rev gtggtctca CATt CCGATAAACCCTAACTCGATTCC GUS ß-Glucuronidase (GUS) Forw gtggtctca aATG 1810 Reporter TTACGTCCTGTAGAAACCCCAA Rev gtggtctca AAGC TCATTGTTTGCCTCCCTGCT

KdTORa K. daigremontiana TOR Forw gtggtctct AAGC GGCAATGGGACCTACGGTAG 360 exon 8 (antisense) Rev gtggtctct GGTG GTTGTATGAGGGCAGGACCA C KdSnRK K. daigremontiana Forw gtggtctct GGTGCCACAGGAATATGGTCGCYC 160 1a SnRK1 exon 3 Rev gtggtctct AAGC GGTGCAGCGTAGTTTGGAC Coding Seq Coding (antisense) Nos Nopaline Synthase Forw gtggtctct GCTT 279 Term Terminator GATGATCCCCGATCGTTCAAAC Rev gtggtctct AGCG GACAGGAGGCCCCGATCTAG 35S 35S CaMV Terminator Forw gtggtctct GCTT GGGACTCTGGGGTTCGGATC 237

Term Rev gtggtctct AGCG Term GGTGATCTGGATTTTAGTACTGG

2.3.5 Kalanchoë daigremontiana Transformation Wild-type K. daigremontiana plants were transformed with 35S::TORantisense, 35S::SnRK1antisense and KdTORpromoter::GUS according to the protocol previously used in the lab (Garces and Sinha, 2009c). Mature leaves greater than 5cm in length were washed in 70% Ethanol for 30 seconds on a shaker (New Brunswick Scientific, USA), then 10 % Bleach (Chlorox, USA) + Triton-X 100 (Fisher Scientific, UK) for ten minutes on a shaker. Leaves were then washed 6 times in water in sterile conditions. Leaves were cut along the midvein and margins were removed, and the remaining tissue was cut into approximately 1 cm2 pieces. Following a 2-hour dark incubation in Agrobacterium tumefaciens resuspended in MSO [1X Murashige Skoog Including Vitamins and MES Buffer (Duchefa Biochemie, UK] + 100µM Acetosyringone (Aldrich, UK), leaf pieces were transferred to Shoot Inducing Medium (SIM) [1x MS, 1 mg/L Thidiazuron (TDZ, Fluka, Germany), 0.1 mg/L Indole-3-Acetic Acid (IAA, Sigma, UK), 3 % sucrose (Fischer Scientific, UK), 0.75 % Agar (Melford, UK)], and incubated for 3 days in the dark at 23°C. Tissue was then transferred to SIM + 40 mg/L Kanamycin Monosulfate (Alfa Aesar, UK) + 500 mg/L

Carbenicillin Disodium (Melford, UK) and incubated in long day conditions at 23°C. Two weeks later, the kanamycin concentration was increased to 100 mg/L. Tissue was sub- cultured on this medium until callus formed, then transferred to Shoot Elongation Medium (SEM) [1x MS, 2 mg/L Gibberellic Acid (GA, Sigma, UK), 0.2 mg/L IAA, 3 % sucrose, 0.75 % agar] and returned to long day conditions at 23°C. Once shoots had elongated and leaves formed, tissue was transferred to Root Inducing Medium (RIM) [0.5X MS, 3 % sucrose, 0.75 % agar]. Upon production of roots, individuals were transferred to soil. Subculture of SnRK1a terminated at this step.

2.4 Genotyping To confirm successful transformation of transgenic lines, DNA was extracted, and PCR performed. DNA was extracted according to the ‘Quick DNA prep for PCR’ protocol outlined in (Weigel and Glazebrook, 2002). In brief, leaf discs were ground in 400 µL extraction buffer [200mM Tris-Cl, 250mM NaCl, 25mM EDTA, 0.5% SDS] and centrifuged at 13000 rpm for 5 minutes. 300 µL of 2-propanol was added to the supernatant and centrifugation repeated. The resulting pellet was rinsed with 70 % ethanol and dissolved in 100 µL TE [10 mM Tric-Cl, 1mM EDTA].

PCR was performed with the same reaction mix as described in section 2.3.3 but with 0.014 U/µL Q5® High-Fidelity DNA polymerase and 0.014 U/µL BioTaqTM polymerase. KdTORa and 35STerm reverse primers (Table 1.) were used for amplification. Additional PCR checks were carried out with NPTII forward and reverse primers (NPTII FORW: 5’CACAACAGACAATCGGCTGC; NPTII REV: 5’GCACGAAGCGGTCAG3’). Cycling conditions were set according to the Q5® protocol, with annealing temperature of 58° and extension for 30 seconds. Products were run on a 1 % agarose gel at 90 V for 25 minutes.

2.5 RNA Extraction The indented region of K. daigremontiana leaves at each plantlet formation stage (Stage 0-4, Fig. 5B-F) were harvested and frozen immediately in liquid nitrogen. Notches of several leaves were pooled to have enough tissue for each sample. Indentations of young leaves with pedestal but without evidence of plantlet (Stage 2) were harvested from several plants per individual lines of 35S::KdTORantisense and similarly pooled. For both experiments, total RNA was extracted using the RNeasy Plant Mini Kit (Qiagen, USA). 10mg Polyvinylpyrrolidone (PVP, MW=40000) dissolved in 600 µL RLC buffer was

35 added per 100 mg ground tissue. Resulting RNA was immediately frozen in liquid nitrogen until ready for cDNA synthesis.

2.6 cDNA Synthesis, RT-PCR and qRT-PCR RQ1 DNase (Promega, USA) and Tetro cDNA Synthesis (Bioline, UK) reactions were carried out according to the manufacturers’ protocols. A mixture of Random Hexamer and Oligo d(T) primers were used for cDNA synthesis reactions, and the reaction proceeded for 1 hour at 45°C.

To measure KdTOR expression levels across plantlet developmental stages, a StepOnePlusTM Real-Time PCR machine with StepOneTM software v2.3 was used, with a SensiFASTTM SYBR Hi ROX Kit (Bioline, UK). GLYCERALDEYHDE-3-PHOSPHATE DEHYDROGENASE (KdGAPDH) was used as a control gene, with an annealing temperature of 60°C. Three biological and three technical replicates were used.

To confirm that KdTOR expression was downregulated in 35S::KdTORantisense lines,

Reverse Transcription (RT)-PCR was performed. 100 µl/ml NH4 Reaction Buffer (10X), 1.5 mM MgCl2, 1 mM dNTPs, 1 mM forward and reverse primers, 2.5 ng/µl cDNA, 10 µl/ml BioTaqTM Polymerase and 10 µl/ml Q5® High-Fidelity Polymerase were mixed in a final volume of 20 µl. A thermal cycling reaction was run according to settings recommended by the BioTaqTM protocol, with annealing at 58°C and extension for 30 seconds, for 39 cycles. GAPDH was used as a loading control, identical settings were used, for 35 cycles.

Table 2: Primer pairs designed for qRT-PCR. Primers had a Tm of 60°C and were designed to amplify no more than 200 bp where possible.

Target Product Size Primer Name Primer Sequence (bp) TOR exons 73 qKdTORForw 5’CGACGCTGGTGAAACTCTTG3’ 35-36 qKdTORRev 5’CTTGAGGCAGTCCGTGAAAC3’ RAPTOR1 178 QKdRAPTOR1Forw 5’CCTCTCCATCCAGGGGCATA3’ QRAPTOR1Rev 5’TGTCAGCGAACCCTTGTGAG3’ GAPDH 290 QKdGAPDH1 5’GGAGCAGAGATAACAACCTTC3’ QKdGAPDH2 5’TCCATTCATCAACACAGACTAC3’

2.7 ß-Glucoronidase (GUS) Staining KdTORp::GUS leaf margins were incubated in GUS staining solution [100 mM Sodium Phosphate Buffer pH 7.2 (BDH Chemicals, UK), 10 mM EDTA pH 8 (Promega, USA), 0.1 % (v/v) Triton X-100 (Fisher Scientific, UK), 1 mM Potassium Ferrocyanide (III) (Sigma- Aldrich, UK), 2 mM X-GlcA (Melford, UK)]. Tissues were incubated for 24 hours in the dark, and then cleared with 100 % Ethanol.

2.8 Phylogenetic Tree Construction Phylogenetic amino acid trees were produced from alignments of the complete TOR, RAPTOR1 and LST8 peptide sequences from 21 divergent plant species. The sequence with the highest Dual Affine Smith-Waterman alignment score of each species was chosen for analysis. Multiple alignment was performed with the ClustalX2.1 program. Trees were constructed using RAxML Blackbox; an open access Maximum-likelihood based phylogenetic interference software, according to (Kozlov et al., 2019). The resulting trees were visualised and edited using Dendroscope (v3.6.3). In all cases, Mus musculus was used as the outgroup. Species include (listing TOR, RAPTOR and LsT8 accessions respectively): Ananas comosus (Aco002506.1, Aco007098.1, Aco005260.1), Arabidopsis halleri (Araha.1806s0004.1, Araha.1514s0002.1, Araha.3948s0008.1), Amaranthus hypochondriacus (AHYPO_005215-RA, AHYPO_009059-RA, AHYPO_000093-RA), Arabidopsis thaliana (AT1G50030.1, AT5G01770.1, AT3G18140.1), Amborella trichopoda (evm_27.model.AmTr_v1.0_scaffold00060.8, evm_27.model.AmTr_v1.0_scaffold00007.183, evm_27.model.AmTr_v1.0_scaffold00012.141), Chlamydomonas reinhardtii (Cre09.g400553.t1.1, Cre08.g371957.t1.1, Cre17.g713900.t1.2), Citrus sinensis (orange1.1g000194m, orange1.1g000697m, orange1.1g042260m), Glycine max (Glyma.11G002600.1, Glyma.09G278500.1, Glyma.13G227200.1), Gossypium raimondii (Gorai.008G019100.1, Gorai.011G130300.1, Gorai.002G229900.2), Kalanchoë fedtschenkoi (Kaladp0047s0074.1, Kaladp0020s0076.1, Kaladp0007s0077.2), Kalanchoë laxiflora (Kalax.0075s0076.1, Kalax.0576s0012.1, Kalax.0052s0084.3), Marchantia polymorpha (Mapoly0047s0119.1, Mapoly0002s0262.1, Mapoly0012s0149.1), Medicago trunculata (Medtr5g005380.1, Medtr7g072330.1, Medtr2g016690.1), Oryza

37 sativa (LOC_Os03g52794.1, LOC_Os12g01922.1, LOC_Os03g47780.1), Physcomitrella patens (Pp3c6_4910V3.1, Pp3c9_4770V3.4, Pp3c11_25160V3.1), Populus trichocarpa (Potri.001G289200.1, Potri.006G106600.1, Potri.016G052800.1), Sorghum bicolor (Sobic.009G109200.1, Sobic.005G008800.3, Sobic.005G172500.1), Solanum lycopersicum (Solyc01g106770.2.1, Solyc09g014780.2.1, Solyc03g059310.2.1). Selaginella moellendorffii (413509, 171199, 233900), Volvox carterii (Vocar.0014s0198.1, Vocar.0065s0020.2, Vocar.0026s0114.1), Zea mays (GRMZM2G049342_T01, GRMZM2G048067_T01, GRMZM2G094959_T05). Bootstrap of values of greater than 50 out of 100 are displayed at the nodes.

2.9 Image acquisition and Data Analysis Photographs of K. daigremontiana were taken using a Huawei P smart (FIG-LX1) with a 10X camera attachment. A GXCAM Eclipse (0654) Wi-Fi camera was attached to a S8AP0 Stereo Microscope (Leica, USA) for all microscope photographs. Scale bars were added using Fiji ImageJ (https://imagej.nih.gov/ij/). All graphs were produced using Microsoft Office 365 Excel or GraphPad Prism Version 7.00. Statistics were produced using GraphPad Prism Version 7.00.

3 Results

3.1 TOR, RAPTOR and LST8 phylogenies are congruent with the species trees Phylogenetic amino acid trees were produced from amino acid sequence alignments of the TOR, RAPTOR1 and LST8 peptide sequences from 21 divergent plant species (Fig. 4). Overall, the resulting trees show congruence with each other and with the species tree. The algal species V. carteri and C. reinhardtii are consistently the most basal, grouping together in >97 of the replicate trees for all genes (Fig. 4A-C). The phylogeny for bryophytes (P. patens and M. polymorpha) and lycophytes (S. moellendorffii) is identical in the TOR and LST8 trees (Fig. 4A, C, respectively) but differs in the RAPTOR1 phylogeny (Fig. 4B). Furthermore, neither arrangement recapitulates the early divergence of, first, bryophytes from the vascular plants and then lycophytes from the euphyllophytes (Morris et al., 2018). However, their positioning with regards to the chlorophyta and angiosperms broadly reflects land plant evolution. Within the angiosperms, A. trichopoda is consistently placed as the sister group to all remaining flowering plants, thus the early diverging angiosperm species resolve according to evolutionary history.

With the exception of the placement of O. sativa within the eudicot lineage of the TOR phylogeny (Fig. 4A), the monocot and eudicot groups are well resolved for all three genes. Within these groups however, the true species order cannot be resolved, and differs amongst the three trees, and against true species phylogenies, particularly within the eudicots. However, closely related species, such as species of Arabidopsis and Kalanchoë, consistently group together with bootstrap values of greater than 90, supporting the resolution of the trees. Overall phylogenetic analyses suggest high conservation of the TOR complex across the plant kingdom, consistent with the known evolutionary trajectory of most species.

A B C

Eudicots

Monocots

TOR RAPTOR1 LST8

Figure 4: Phylogenetic trees of the plant TORC1 components, constructed from amino acid sequence alignments of the a) TOR, b) RAPTOR1 and c) LST8 genes. Trees are displayed as the best-scoring maximum likelihood tree, with bootstrap values out of 100 replicates shown at each node. Nodes with a bootstrap value of lower than 50 are excluded from the final tree.

3.2 KdTOR expression varied across plantlet developmental stages Plantlet formation in K. daigremontiana occurs in the indentations along the margins of the leaf, sequentially from the distal leaf tip to the base (Fig. 5A). Within the indentations, termed nodes, four stages of plantlet development can be distinguished. Stage 1 is identified as a flat node at the indented region, with no protrusion or plantlet visible (Figure 5C). From the node emerges a structure that will hold the developing plantlet, known as the pedestal. Stage 2 is defined as a pedestal with no visible plantlet

(Fig 5D). Next, a visible, pin-shaped plantlet will protrude from the pedestal (Stage 3, Fig. 5E). The pin will then develop into rounded plantlet cotyledons (Stage 4, Figure 5F). Indentations of young leaves with no evidence of node formation were used as Stage 0 (Figure 5B), before the onset of plantlet formation. These five distinct developmental stages were harvested and KdTOR expression in stages 1-3 was quantified by qRT-PCR.

KdTOR expression was detected at all stages of plantlet formation (Figure 5G). Compared with stage 0, KdTOR expression did not markedly increase during stage 1 of plantlet formation. However, KdTOR expression in stages 2 and 3 increased significantly compared with stages 0 and 1. The highest expression was detected in stage 2, and expression was 6.3-fold higher than expression in stage 0. Expression then slightly decreased to 5.5-fold higher than stage 0 in stage 3, however this difference was not significant between stages 2 and 3. This pattern of expression suggests that KdTOR plays a role in plantlet formation throughout the developmental stages, but is most crucial for the initiation of plantlet formation at the pedestal (Stage 2), and is subsequently expressed as the pin develops into cotyledons (Stage 3).

KdRAPTOR expression in stages 0-3 was also analysed (Fig. 5H). As with KdTOR expression, KdRAPTOR is expressed at a low level in stage 1, with no significant difference compared with Stage 0. KdRPATOR expression is then markedly increased in stage 2, with a fold change of 37, before decreasing again in stage 3, with no significant difference relative to stage 0. Whilst the trend therefore closely mirrors the expression pattern of KdTOR, the differences are not statistically significant.

A B

Stage 0 C D

Stage 1 Stage 2 E F

Stage 3 Stage 4 G H

ns ns

Relative Expression Relative Relative Expression Relative ns

KdTOR KdRAPTOR1

Figure 5: Stages of Kalanchoë daigremontiana plantlet formation and expression of KdTOR and KdRAPTOR1. (A) Plantlets develop from the tip to the base of the leaf margin (indicated by arrow). (B-F) Stages of plantlet development harvested for qRT-PCR. (B)Stage 0. Young leaf indentation with no evidence of pedestal or plantlet. (C) Stage 1. No pedestal has formed and the node is visible at the indentation boundary (white arrow). (D) Stage 2. A pedestal has formed. (E) Stage 3. Thin, pin-shaped plantlet emerges from pedestal. (F) Stage 4. Plantlet cotyledon begins to round. Scale bars: 200 µm. (G) qRT-PCR of KdTOR expression in Stages 0-3, relative to Stage 0. (H) qRT-PCR of KdRAPTOR expression in Stages 0-3, relative to Stage 0. A One Way ANOVA with Dunnet’s Multiple comparison was performed, n=3.

3.2 TOR::GUS expression mirrors qPCR pattern across developmental stages 3.2.1 promoterTOR::GUS lines were generated qRT-PCR shows overall expression patterns in different tissue types, but it cannot show how KdTOR expression levels change at the cell level within a tissue. Promoter::reporter transgenic lines were therefore generated. The ~1.5 kb region upstream of KdTOR was cloned, ligated into pGEM-T Easy® and checked by digestion with EcoRI (Fig. 6A and B). The fragment was then assembled into a modified pBI121 vector alongside GUS and the Nos terminator, using Golden Gate Cloning. Constructs were checked by digestion, culture PCR and sequencing, before transformation into A. tumefaciens strain LBA4404, which was again verified by culture PCR (Fig. 6C). Correct constructs were then transformed into K. daigremontiana leaf tissue and incubated on Kanamycin plates to

A B

pKdTOR 1 2 3

3 kb -

1.5 kb - 1.5 kb -

C 1 2 3 + -

1.5 kb -

Figure 6: Cloning of the KdTORpromoter region ~1.5 kb upstream of KdTOR and Agrobacterium tumefaciens transformation. (A) PCR of ~1.5 kb KdTORpromoter. Band was gel extracted and ligated into pGEM®-T Easy. (B) pKdTOR::pGEM digestion. Colonies 1, 2 and 3 had the correct sized bands (2997 bp and 1514 bp). (C) Agrobacterium tumefaciens culture PCR for pKdTOR::GUS::pBI128. Colonies 1, 2 and 3 show the correct band size (~1500 bp).

43 select for transgenic tissue. Transformants were cultured for ~9 months on SIM, SEM and RIM media, before transfer into soil. Mature plants were harvested, and GUS expression analysed. Of the seven independent lines generated, three showed strong GUS expression in the leaf margins, these were subsequently propagated and harvested for analysis.

The expression pattern of KdTORp::GUS was analysed, from before evidence of plantlet formation, through to completion of plantlet development. GUS accumulation was detected across many stages of plantlet development (Fig. 7). Interestingly, GUS expression was also detected in the tips of leaf protrusions of developing leaves (hydathodes), in a similar pattern to auxin [Fig. 7A; (Bilsborough et al., 2011)]. GUS accumulated in the young leaf indentation before the formation of the node (Fig. 7B; Stage 0) and was later detected at the node (Fig. 7C, D; Stage 1). As the pedestal forms (Fig. 7E; Stage 2), GUS expression was less visible, but returns strongly at the centre of the pedestal in the transition from stage 2 to 3 (as the pin-like plantlet structure develops) (Fig. 7F and G). Strong GUS expression was continually detected until the cotyledons began to round. As the plantlet matures, and cotyledon size increased, no GUS expression was detected (Fig. 7H-J). GUS expression then returned in the root primordia of the developing plantlet (Fig. 7K) and was later detected at the root tips of mature plantlets (Fig. 7L).

A B C D

E F G H

I J K L

Figure 7: KdTORpromoter::GUS expression during plantlet development. (A and B) GUS accumulated at the hydathode (A) and the indentation (B) of the leaf margin. (C-G) GUS accumulation was detected throughout stages 1-3 of plantlet formation. GUS expression was visible at the node (C and D), in the developing pedestal (E), at the initiation of plantlet development at the pedestal (F) and in the pin-like plantlet structure (G). (H-J) GUS was not expressed in the plantlet cotyledons. (K and L) GUS accumulated at the root primordia (K) of the developing plantlet, and at the root tips (L) of mature plantlets. Scale bars: 1 mm (A, K, L); 200 µm (B-J).

3.3 Torin2 application to the leaf margin inhibited plantlet formation In order to restrict TOR inhibition to the site of plantlet formation, individually potted K. daigremontiana plants were brushed with Torin2 solution at the leaf margins of one leaf pair per plant. For leaves greater than 3 cm at the time of treatment, the percentage of treated leaves showing evidence of plantlet formation was similar at all time points regardless of treatment with mock or Torin2 solution (Fig. 8A). Furthermore, there was no significant difference in final plantlet number per treated leaf on day 28 (Fig. 8C; t=1, P value=0.3229). In contrast, Torin2-treated leaves of less than 3cm had a consistently lower percentage of leaves with plantlets between days 3 and 24 compared to mock (Fig. 8B). Although all plants (both Torin2 and mock treated) had produced plantlets by the final measurement on day 28, there were significantly fewer plantlets per leaf when A B

Leaves > 3 cm Leaves < 3 cm 100

100

e t

l 80

e t

l 80

l P

h 60

w s

40 t

s 40

l a

Mock l P

20 P Mock 20 % Torin2 % Torin2 0 0 0 3 6 9 12 15 18 21 24 27 30 0 3 6 9 12 15 18 21 24 27 30 Days After Treatment Days After Treatment

C D E ns ** **

Figure 8: Torin2 and mock treatment of Kalanchoë daigremontiana plantlet formation. (A and C) Percentage of plantlet formation was similar in leaves longer than 3 cm after Mock (0.5 % DMSO and 0.5 % Tween-20) or Torin2 (100 µM Torin2, 0.5 % DMSO and 0.5 % Tween-20) treatment (A). The final number of plantlets per leaf margin was not significantly different after Mock or Torin2 treatment (recorded on day 28) (C; t=1, P value= 0.3229). (B and D) Leaves less than 3 cm in length had delayed plantlet formation after Torin2 treatment (B). The final number of plantlets per leaf was significantly lower in Torin2-treated plants compared with mock (recorded on day 28) (D; t=3.198, P value= 0.0025). (E) Number of plantlets per leaf was significantly lower on Torin2-treated leaves on day 15 of recording (t=3.263, P value=0.0021). Error bars show SEM. Torin2-treated plants, n = 54; Mock-treated plants, n = 48.

46 treated with Torin2 (Fig. 8D; t=3.198, P value=0.0025). Furthermore, at earlier time points, such as on day 15 (Fig 8E; t=3.263, P value=0.0021), there were significantly fewer plantlets on Torin2-treated leaves. This suggests that smaller leaves are more susceptible to treatment, supporting the qRT-PCR and KdTORp::GUS expression data that shows TOR functions early in plantlet formation.

3.4 Genetic suppression of TOR abolished plantlet formation

3.4.1 35S::TORantisense construct was assembled The KdTOR exon 8 sequence had not been previously reported, and therefore an alignment was performed with TOR homologs of divergent eukaryotic species (Fig. 9A). Residues were highly conserved across plant species, particularly the closely related Kalanchoë species. K. daigremontiana and K. laxiflora shared 91/92 residues (98.9 % sequence similarity). Amongst the 23 species included in the alignment, including humans and mice, 21/95 amino acid residues were shared, demonstrating high TOR sequence conservation across diverse kingdoms.

Whilst informative, chemical inhibition is only transient, and therefore to fully investigate the result of an endogenous KdTOR knockout, 35S::KdTORantisense transgenic lines were generated. A 375 bp fragment of a conserved region of TOR exon 8 was amplified by PCR and ligated in pGEM-T Easy. The plasmids were checked by EcoRI digestion; two bands of size 2997 bp and 324 bp were expected (Fig. 9B). The KdTOR sequence in antisense orientation was assembled into a modified pB1121 vector with the 35S promoter and 35S terminator by Golden Gate Cloning. Constructs were checked by digestion with EcoRI and HindIII (Fig. 9C) and correct constructs were transformed into A. tumefaciens strain LBA4404. Constructs were checked by culture PCR and transformed into K. daigremontiana leaf tissue. Tissue was plated onto Kanamycin media for selection, and sub-cultured on SIM, SEM and RIM media for ~9 months before transfer to soil (Fig. 9D-F).

B C 1 2 3 4 5 1 2

3.0 kb - -10 kb

0.4 kb - -1.5 kb

D E F

Figure 9: Generation of 35S::KdTORantisense. (A) Amino acid alignment of the 375 bp sequenced fragment of K. daigremontiana (red box) with divergent eukaryotic species. (B) KdTORa::pGEM digestion. Colonies 1,2,3 and 5 show the correct band sizes (2997 bp and 324 bp). (C) 35S::KdTORa::pBI128 digestion. Colony 1 showed the correct bands (10579 bp and 1367 bp). (D) ~1 cm2 leaf pieces of K. daigremontiana were transformed with 35S::KdTORa in A. tumefaciens. (E) Kanamycin resistant callus formed on SIM media. (F) Transgenic plants were transferred to soil after ~9 months of sub-culture.

3.4.2 35S::KdTORantisense transgenic lines were produced A total of 14 independent transgenic lines were generated, of which eight were confirmed transgenic by PCR (Fig. 10A and B). Down-regulation of TOR expression in the transgenics was confirmed by RT-PCR (Fig. 10C and D). Plants were chosen for analysis by the severity of the phenotype observed; two ‘severe’ lines (lines A1 and D2) with little evidence of plantlet formation, three ‘mild’ lines (lines B3, C2 and C3) with phenotypic plantlet formation and one ‘weak’ line (line D1) without an observable phenotype. All lines except C3 showed a moderate decrease in TOR expression (Fig. 10C), suggesting moderate suppression of KdTOR had been achieved.

Phenotypes varied across lines, but a general reduction in plant size, leaf size and plantlet number, alongside an increase in the thickness of leaves and a reduction in the number of indentations at the margins was observed. Reduced plant and leaf size is consistent with TOR repression in published data (Deprost et al., 2007). Variation in phenotype severity was observed across lines, likely due to differences in TOR repression, perhaps because of variation in the insertion position of the construct in the genome. Due to the embryo lethality of complete KdTOR knockout, it is likely that the strongest silencing lines died early in development, and only mild knockdown lines were recovered.

3.4.3 Repression of tor severely restricted plantlet formation All eight confirmed transgenic lines showed a significant reduction in plantlet number compared to WT (Fig. 10E; P value<0.0001). This is consistent with Torin2 treatment data (Fig. 9; Section 3.3). Even the least phenotypically severe line, E2, which closely resembled WT plants, had significantly fewer plantlets than WT. The most severe lines (A1, C3 and D1) had severely restricted growth, with thick, discoloured leaves and very little evidence of plantlet formation. Interestingly, even when plantlet number was severely reduced, leaf margins and indentations remained intact, though indentation number was significantly reduced relative to WT (Fig. 10F; P value<0.0001). Even when taking into consideration an overall reduction in indentation number, the percentage of indentations occupied by plantlets in transgenic lines was markedly lower than WT (Fig. 10G).

A A1 A2 B2 B3 C2 C3 D1 D2 D2 E2 - - + 400 bp - H I

B + A2 B2 B3 C2 C3 D1 D2 E2 - - 300 bp -

Severe Mild Weak Wild-type Wild-type A1 D2 B3 C2 C3 D1 WT J K C 100 bp - TOR

D 300 bp - GAPDH

E **** A1 D2 L M

F **** A2 B3 N O

G Wild-type A2 P Q

B2 A2

Figure 10: Genotypic and phenotypic analysis of 35S::KdTORantisense. (A and B) Amplification of the KdTORa::35Sterm region (A) and the NPTII region (B) confirms presence of the 35S::KdTORa transgene. Negative control: wild-type; Positive control: KdTORa::pBI128 plasmid. (C and D) RT-PCR of TOR expression in 35S::KdTORa showing severe, mild or weak phenotypes. (C) 76 bp region of TOR. (D) GAPDH control gene as loading control (600 bp). (E-G) Plantlet and indentation number is reduced in transgenics compared to WT. All transgenic lines have significantly fewer plantlets (E; P Value <0.0001) and indentations (F; P value <0.0001) than WT. One-Way ANOVA with Dunnett’s Multiple comparison. (G) Percentage of indentations with and without evidence of plantlet formation. (H-Q) Phenotypes observed across 35S::KdTORa lines. Scale bars: 1 cm.

Alongside a reduction in plantlet number, changes to plantlet development were also observed. Along a single leaf margin, the stage at which plantlet formation was affected varied. In the most phenotypically severe lines (A1 and D2), plantlet formation was completely abolished (Fig. 10J and K). In lines with a milder phenotype, plantlet formation occurred at some indentations. Plantlets had smaller, thicker and misshapen leaves, resembling the mother plants, and lacked a meristem (Fig. 10L, M and Q). Plantlet meristem growth was restricted late into plantlet development. In the least severe line, E2, plantlets were phenotypically similar to WT, however meristem growth was slowed. Typically, however, plantlet formation was terminated before plantlets reached maturity. At the indentations across several lines, plantlet formation was not initiated, the node did not form, and notches resembled Stage 0 (Fig. 5B). At other indentations, plantlet formation terminated following the development of the node, and tissue at the node subsequently died (Fig. 10O). At most indentations, however, plantlet formation appeared to progress beyond node formation, and terminated following the formation of the pedestal (Stage 2), or shortly thereafter (Fig. 10P).

Observation under the microscope revealed that whilst appearing to have terminated before plantlet emergence at the pedestal, structures phenotypically dissimilar to WT often protruded from the pedestal of 35S::KdTORa lines (Fig. 11). In many transgenic lines, the pedestal was bleached, misshapen and lacked the central meristematic node from which the plantlet emerges (Fig. 11G). In other lines, however, microscopic observation revealed a globular structure in place of the stage 3 plantlet (Fig. 11H) and bleached, misshapen structures in place of stage 4 plantlets (Fig. 11I). Likewise, mature 35S::KdTORa plantlets displayed mutant phenotypes; cotyledons were discoloured and bilobed (Fig. 11J).

A2D2A1

A F

B G

C H

D I

E J

Figure 11: Microscope images of 35S::KdTORantisense plantlet development stages. (A-E) WT plantlet formation. (A) A node forms at the indentation (Stage 1). (B) The pedestal develops from the node; a plantlet will emerge from the centre (Stage 2). (C) Pin-like plantlet emerges from the pedestal (Stage 3). (D) Two cotyledons can be distinguished (Stage 4). (E) The mature plantlet. (F-J) Transgenic plantlets display mutant phenotypes dissimilar to WT throughout the developmental stages. (F) Plantlet formation terminated at the development of the node. (G) The pedestal is misshapen and small. (H) Globular structure emerges from the pedestal. (I) Plantlet cotyledons are bleached and misshapen. (J) The mature plantlet has bilobed cotyledons. Scale bars: 200 µm (A-D and F-I); 1 cm (E and J).

3.4.4 35S::KdTORantisense displayed changes in phyllotaxy and meristem growth WT K. daigremontiana produce leaves in an opposite formation (decussate); two leaves emerge in a pair from the node simultaneously (Fig. 12A and D). In two independent 35S::KdTORantisense lines, leaves instead emerged three at a time from the main meristem (Fig. 12B). Furthermore, leaves were produced alternately (one leaf emerged from the meristem at a time) in 6 independent lines (Fig. 12C). These changes to phyllotaxy suggest TOR may play a role in meristem patterning. This is further supported by phenotypic changes to the main meristem in 2 independent lines. The meristem grew larger and appeared to terminate before a new leaf emerged from the side in line B2 (Fig. 12E). New leaves emerged from the meristem differently to WT in line C2 and were thicker and misshapen (Fig. 12F). A B C

Wild-type C2 A1 D E F

Wild-type B2 C2

Figure 12: Phyllotaxy and meristem phenotypes in 35S::KdTORantisense. (A-C) Changes to phyllotaxy. (A) WT K. daigremontiana leaves emerge in pairs in the opposite formation. (B) An example of leaves emerging 3 at a time from the same node in 35S::KdTORa. (C) An example of leaves emerging one at a time from the node – alternate leaf patterning. (D-F) Changes to meristem development. (D) Close-up of new leaf pair emerging from the WT meristem. (E) 35S::KdTORa meristem has terminated and a new leaf is emerging from the side. (F) 35S::KdTORa leaf pair is mutated. Scale bars: 1 cm.

3.5 35S::KdSnRK1antisense construct was assembled A 157 bp fragment of the KdSnRK1 gene was amplified by PCR and ligated in pGEM®-T Easy. The construct was checked by digestion with EcoRI (data not shown) and sequencing, before assembly into a modified pB1121 vector in the antisense orientation, with 35S promoter and terminator by Golden Gate Cloning. Constructs were checked by digestion with EcoRI and HindIII; two bands of size 11244 bp and 1245 bp were expected (Fig. 13A). Correct constructs were transformed into A. tumefaciens strain LBA4404 and checked by culture PCR with M13 forward and reverse primers; a 1245 bp band was expected (Fig. 13B). Correct constructs were transformed into K. daigremontiana tissue and plated onto kanamycin plates for transgenic selection (Fig. 13C and D). Tissue is currently being sub-cultured in SEM media and will be ready for transfer to soil in ~1 month (Fig. 13E).

A B

1 2 3 1 2 3 4 10 kb -

1.2 kb - 1.2 kb -

C D E

Figure 13: Generation of 35S::KdSnRK1antisense. (A) KdSnRK1a::pBI128 digestion. Colonies 1 and 3 show the correct digestion pattern (11244 bp and 1245 bp). (B) Agrobacterium tumefaciens culture PCR for 35S::KdSnRK1a::pBI128. Colonies 2, 3 and 4 show the correct band size (1245 bp). (C-E) 35S::KdSnRK1a tissue culture. (C) ~1 cm2 leaf pieces of K. daigremontiana were transformed with 35S::KdSnRK1a in A. tumefaciens. (D) Kanamycin resistant callus formed on SIM media. (E) Calli were transferred to SEM media to induce shoot growth.

4 Discussion The phenomenon of plantlet formation in K. daigremontiana represents a unique reproductive strategy in which mature, differentiated leaf cells acquire the ability to regain totipotency, allowing the asexual reproduction of clones (Garces and Sinha, 2009b). Whilst K. daigremontiana is described as a constitutive plantlet-forming species, plantlet initiation is dependent upon environmental conditions, and in this species is promoted by long days and drought (Liu et al., 2016). How these signals are transduced to trigger pluripotency in the leaf, however, remains unclear. Following initiation, plantlet development recruits key regulators of both embryogenesis (LEC1, LEC2 and FUS3) and organogenesis (STM) pathways (Garces et al., 2007; Garces et al., 2014). Recently, TOR has gained traction as a key developmental regulator, responsible for sensing energy and nutrient availability and activating plant developmental pathways (See Review: Shi et al., 2018). Environmental conditions represent one such signal able to activate TOR. Light, glucose and auxin signalling are known to promote TOR in the shoot apex (Li et al., 2017; Xiong et al., 2013) and TOR is in turn responsible for the promotion of auxin signalling through ARF translation (Schepetilnikov et al., 2013). TOR has also been shown to control meristem patterning through WUS and CLV3 (Pfeiffer et al., 2016), and meristem activity through YAK1 (Barrada et al., 2019). Furthermore, the embryo lethality of Attor knockouts suggests a crucial role for TOR during embryogenesis (Menand et al., 2002). TOR’s role in several developmental processes common to K. daigremontiana plantlet formation, and its ability to integrate environmental signals and developmental responses, suggests huge potential for the kinase as a key regulator controlling plantlet initiation and development. Therefore, the link between the TOR kinase and plantlet formation was investigated for the first time.

The high conservation of the TOR Kinase Complex amongst eukaryotes highlights its importance in growth and development across the plant and animal kingdoms. TORC1 phylogeny in the plant kingdom has not been previously investigated, and therefore maximum likelihood trees of the three TOR complex components (TOR, RAPTOR1 and LST8) were constructed. The K. daigremontiana genome sequence has not been reported, and therefore the TOR, RAPTOR1 and LST8 protein sequences of the sequenced Kalanchoë species K. laxiflora and K. fedtschenkoi were included in the

55 alignment. As expected, the resulting trees showed general congruence with each other and with the species tree; the chlorophyta were consistently placed most basally, followed by the bryophytes and lycophytes and then the angiosperms. Whilst the phylogeny for the bryophytes and lycophytes varied amongst genes and did not recapitulate the early divergence of, first, the bryophytes from the vascular plants, followed by the lycophytes from the euphyllophytes; their positioning relative to the chlorophyta and angiosperms broadly reflected land plant evolution. Within the higher plant lineage, A. trichopoda was consistently positioned as the earliest diverging angiosperm, and the monocot and eudicot groups were generally well resolved. Within the monocot and eudicot groups, however, the true species order could not be resolved. Resolving the relationships between the clades of the core eudicots is challenging, due to the rapid diversification of the major lineages after their first appearance (Magallón et al., 2015; Moore et al., 2010). Incomplete lineage sorting has thus been implicated as an important factor contributing to discordance amongst gene histories in phylogenetic analyses (Maddison and Knowles, 2006). Nevertheless, overall, phylogenetic analyses suggest high conservation of the TOR complex across the plant kingdom, implying that TOR function is conserved. Most importantly, Kalanchoë species consistently group together with a bootstrap value of greater than 90, suggesting that the gene sequences have not diverged and are functionally conserved. TORC1 is thus likely functional in K. daigremontiana, validating the need for further investigation.

The importance of TOR in eukaryotic growth and development is highlighted by the conservation of TOR expression in all eukaryotes. This extends to K. daigremontiana plantlet formation. Nodes at stages 0-3 all had detectable KdTOR expression (Fig. 6G). KdTOR was expressed at the earliest stage of plantlet formation (Stage 1), however there was negligible difference in expression between stage 1 and the stage 0 control. Perhaps this is because KdTOR is expressed at stage 0, before the formation of the node at the indentation, in order to trigger cell fate change. It is therefore worth repeating the analysis with a further, earlier, developmental stage, in order to also include stage 0 in the comparison.

Highest KdTOR expression was detected at stages 2 and 3; the development of the pedestal and emergence of the plantlet from the pedestal. This suggests that KdTOR is

56 most crucial for pedestal development, and the initiation of plantlet formation at the pedestal. The increase in KdTOR expression at stages 2 and 3 is not statistically significant, however this could be due to technical errors. It is difficult to harvest uniform stages, as the plantlet-forming region is small, and the stages, particularly stages 2 and 3, can be difficult to distinguish. There is therefore likely variation within the pooled samples. Furthermore, the increase in KdTOR expression at stage 2 and 3 could be due to an increase in the size of the structure expressing KdTOR, relative to the smaller stages 0 and 1. Nevertheless, this analysis indicates that KdTOR is indeed expressed throughout stage 0 to 3 of plantlet formation and is perhaps most crucial following the formation of the pedestal. Furthermore, expression of the TORC1 component KdRAPTOR1 throughout stages 0-3 supported the KdTOR expression pattern. KdRAPTOR1 expression was highest at stage 2, reflecting the expression of KdTOR. However, ∆∆CT values representing relative differences in KdRAPTOR1 expression are very high, suggesting that the analysis may not be representative of true KdRAPTOR1 expression.

To further investigate KdTOR expression during stages 1-4 of plantlet formation at the cellular level, KdTORp::GUS constructs were made and transformed into K. daigremontiana. Successful integration of the construct was checked by GUS staining of the plantlets from the seven independent lines generated. The leaf margins of the three lines showing strong expression were subsequently stained for analysis. Interestingly, the pattern of KdTORp::GUS expression in plantlet developmental stages broadly reflected the expression pattern obtained through qRT-PCR, further verifying the validity of the result.

GUS expression was detected at the stage 0 indentation (Fig. 7B), supporting the argument that stage 0 tissue may not be a good control stage for qRT-PCR analyses. KdTOR is likely involved very early in plantlet formation, perhaps to trigger pluripotency, and stage 0 should thus be included as a measurable stage of plantlet formation. GUS expression was detected through stages 0 to 3 (Fig. 7C-G), only disappearing when the cotyledons began to expand (Fig. 7H-J; Stage 4). Perhaps at this stage, the developmental pathway of the plantlet is determined, all necessary downstream genes have been activated, and TOR expression is no longer necessary. It would be interesting

57 to repeat the qRT-PCR analysis of TOR expression in developmental stages and include stages 0 to 4, in order to directly compare the expression pattern with KdTORp::GUS accumulation.

KdTORp::GUS expression returned in the root primordia of the developing plantlet (Fig. 7K). Roots represent a new structure, and therefore TOR is likely again necessary to activate a new set of downstream processes specific to root growth. Later, GUS was consistently detected at the tip of the growing root, in a similar pattern to auxin [See review: (Overvoorde et al., 2010)]. Further to this, GUS is detected at the hydathode of the leaf, located beneath the point of a serration, before evidence of plantlet formation. This same pattern has been previously observed in DR5::GUS Arabidopsis lines showing auxin expression (Bilsborough et al., 2011). TOR may therefore be promoting auxin accumulation in the hydathode (or vice versa), similar to its role in leaf primordia initiation in A. thaliana (Mohammed et al., 2018). This suggests that TOR and auxin could interact to activate downstream pathways.

In Arabidopsis, light-inactivated COP1 triggers auxin activation of AtTOR, driving the phosphorylation of RPS6 to activate a timely switch from skotomorphogenic to photomorphogenic development in seedlings (Chen et al., 2018). It is therefore not unreasonable to suggest that auxin and KdTOR interact to drive plantlet initiation at the leaf margin. However, expression of DR5::GUS is not detected in stages 1 and 2 of plantlet formation (Unpublished data), therefore perhaps this interaction only occurs in the root. Interestingly, whilst auxin-ROP signalling has been shown to activate TOR to trigger S phase in the SAM, auxin has not been implicated in the activation of AtTOR- E2FA signalling to activate the RAM [See review: (Shi et al., 2018)]. The identical expression patterns of KdTORp::GUS and DR5::GUS in the plantlet root could therefore mark an innovation in auxin-TOR signalling at the root primordia. Furthermore, TOR has been shown to play a crucial role in auxin signalling transduction in two independent studies (Deng et al., 2016; Schepetilnikov et al., 2013). Torin1 can interfere with auxin redistribution in root tips and root gravitropic responses (Dinkova et al., 2000; Schepetilnikov et al., 2013). This could explain the identical expression pattern of KdTORp::GUS and DR5::GUS at the plantlet root tips. A TOR-Auxin relationship during plantlet formation is thus worthy of investigation. Treatment of KdTORp::GUS with auxin

58 inhibitor NPA, and DR5::GUS K. daigremontiana plants with TOR inhibitor could reveal which of the two is upstream, or whether the interaction is more complex. Auxin is known to activate TOR in Arabidopsis to promote selective translation of ARF genes (Li et al., 2017; Schepetilnikov et al., 2013), and so determining if this signalling module is conserved in Kalanchoë may inform the elucidation of a possible auxin-TOR-plantlet signalling pathway.

Whilst GUS expression was visualised in KdTORp::GUS lines, quantification was not possible. Traditionally GUS expression is quantified by 4-Methylumbelliferyl-beta-D- glucuronide (MUG) assay. GUS could not be successfully extracted from K. daigremontiana leaf tissue using this method (Data not shown). This is likely because, as a Crassulacean Acid Metabolism (CAM) plant, K. daigremontiana produces a set of specific compounds such as complex polysaccharides, pigments, waxes and terpenoids that interfere with the analysis of proteins by electrophoretic techniques (Carpentier et al., 2005; Wang et al., 2008). One way to circumvent this would be to extract total protein and quantify GUS by western blot analysis with a GUS-specific antibody. Once again, however, traditional protein extraction from K. daigremontiana tissue is not possible (Data not shown). Lledias et al. (2017) proposed a method for total protein extraction from succulent plants for proteomic analysis, and therefore provide an avenue worthy of exploration, perhaps allowing GUS expression to be quantified in K. daigremontiana leaf tissue.

With the confirmation that KdTOR is expressed during plantlet development, TOR inhibitor Torin2 was applied once directly to the margins of WT plants to determine whether KdTOR was also functioning where expression was detected. No difference in plantlet number was observed in leaves greater than 3 cm (Fig. 9A and C), perhaps because plantlet initiation had already occurred. This is supported by KdTORp::GUS expression patterns showing that GUS is expressed as early as stage 0 (Fig. 8B). Furthermore, leaves less than 3 cm had significantly fewer plantlets when KdTOR was inhibited (Fig. 9B, D and E), suggesting that KdTOR indeed promotes plantlet formation at the very early stages.

Currently, the K. daigremontiana genome sequence is not available, and so degenerate primers were used to clone a 375 bp region of KdTOR exon 8. Alignment with divergent

59 plant species, and the mammalian species M. musculus and H. sapiens, revealed high conservation of the region (Fig. 10A). Analysis of the aligned A. thaliana amino acid sequence on UniProt revealed that the region corresponds to the HEAT repeats. These are protein motifs of two alpha helices important for protein interactions. This sequence was then used to make a KdTOR silencing construct, to reveal how endogenous KdTOR knockdown affects plantlet formation (Fig. 10). In A. thaliana, tor mutants are known to be embryo lethal (Menand et al., 2002), thus complete or strong KdTOR knockouts are likely to be non-viable. Consequently, a weak silencing system – antisense – was required to obtain transgenic plants able to survive past embryogenesis. Transgenic calli were generated, and sub-cultured until transfer to soil for phenotypic analysis (Fig. 10D- F). Plants were confirmed transgenic by PCR and RT-PCR (Fig. 10A-D). Whilst RT-PCR did not show the level of down-regulation of KdTOR previously achieved by K. daigremontiana RNAi gene silencing constructs (Garces et al., 2007), this could be explained by the embryo lethality of severe KdTOR knockdowns; only weaker knockdowns survived into adulthood. Transformation of Arabidopsis with AtTOR knockdown constructs, for example, yields lines with a modest decrease in AtTOR mRNA accumulation (Deprost et al., 2007). KdTORantisense lines displaying severe (with no plantlet formation; Figs. 10J, K & O), mild (formation of mutant plantlets; Figs. L, M, P & Q) and weak (indistinguishable from WT) phenotypes were harvested, and their expression quantified by RT-PCR. The gradient of expression generally correlated with the severity of the phenotype (Fig. 10C and D), indicating that a degree of silencing had been achieved. Furthermore, whilst the KdGAPDH amplicon was standardised as a loading control, the WT band is visibly weaker (Fig. 10D). This could be due to low RNA quality in the WT sample, meaning that cDNA synthesis was not as efficient. Therefore, if WT loading was standardised against the transgenic lines, we would expect a much brighter band for WT KdTOR RNA (Fig. 10C). Therefore, to improve the reliability of the result, a larger quantity of WT PCR product needs to be loaded onto the gel. Failing this, RNA should be extracted again from WT K. daigremontiana. Furthermore, RNA was extracted from young leaves less than 1 cm but with the knowledge that KdTOR is expressed highly at the leaf margin at stages 0-3 of plantlet formation, it would be worthwhile to extract from this tissue instead. However, overall, RT-PCR confirms that TOR expression is downregulated in 35S::KdTORantisense transgenic tissue.

Plantlet formation in TOR knockdown lines varied in the degree of phenotypic severity. Nevertheless, all lines displayed a significant reduction in the number of plantlets produced along the leaf margins (Fig. 10E). Whilst there was still some evidence of plantlet formation even in the most phenotypically severe lines, suggesting KdTOR is not critical to plantlet formation, this may be due to the lethality of complete TOR knockout described above. The number of indentations per leaf was also significantly reduced in all lines compared to WT (Fig. 10F). It could therefore be argued that plantlet formation is decreased due to mutation of the leaf margins. However, even taking into account the reduction in indentation number, the percentage of plantlets occupying each indentation is still markedly lower in 35S::KdTORantisense compared with WT (Fig. 10G).

Alongside a reduction in plantlet number, changes to plantlet morphology were observed at all developmental stages. Interestingly, plantlet formation across several lines appeared to terminate following the development of the pedestal, suggesting that KdTOR expression is most essential for the initiation of plantlet development at the pedestal. Pedestal phenotype was also altered, however; they were reduced in size and lacked the centre of rapidly dividing cells (Fig. 11B) from which the plantlet will emerge (Fig. 11G). By observation under the microscope, it was found that many of the indentations that appeared by naked eye to terminate following pedestal formation in fact had small globular structures emerging from the point at which WT plantlets develop (Fig. 11H). This suggests that KdTOR is necessary to guide proper development of the plantlet structure. Furthermore, 35S::KdTORantisense plantlet cotyledons were misshapen and white in colour (Fig. 11I), suggesting a loss of chlorophyll production. This is consistent with previous studies in Arabidopsis, in which TOR activity promoted chlorophyll biosynthesis during cotyledon greening through ABA INSENSITIVE 4 (ABI4) (Li et al., 2015). Furthermore, mature 35S::KdTORantisense plantlets lacked a meristem at the centre of the cotyledons from which true leaves will develop (Fig. 11Q), implying a role for TOR in meristem development. Phenotypes observed in the main meristem of KdTORa lines, and changes to phyllotaxy, reinforce the suggestion that KdTOR may be involved in meristem patterning and function. Scanning Electron Microscopy (SEM) of affected meristems will allow the 3D structures to be visualised at the cellular level, perhaps providing some clues as to how KdTOR is functioning. Overall, the variety of

61 phenotypes observed at all stages of plantlet formation imply that KdTOR plays a conserved role throughout. Alongside KdTORp::GUS expression data, the phenotypes suggest that TOR is involved in plantlet formation from the early stages, and is particularly important for the initiation of the plantlet, as demonstrated by termination of KdTORa plantlets at this stage, and high GUS expression in the pin-like plantlet structure.

Beyond the confirmation that KdTOR is necessary for plantlet formation, the 35S::KdTORantisense transgenics can be used to determine other components of the KdTOR pathway during plantlet formation. KdSTM has previously been shown to be essential for plantlet formation (Garces et al., 2007), so it would be interesting to see whether KdTOR is responsible for its activation. Other meristem genes have also been implicated in plantlet formation (such as KdWUS and truncated KdLEC1) (Garces et al., 2014; Guo et al., 2015), and the recent discovery that YAK1 is negatively regulated by TOR to promote meristem maintenance (Barrada et al., 2019) provides scope for investigation of KdTOR as a regulator of meristem genes in plantlet formation. qRT-PCR analysis of expression of these genes in transgenic tissue could uncover the pathway linking them to KdTOR. The embryogenesis gene FUS3 is also necessary for plantlet formation (Unpublished data). It would be interesting to determine whether KdTOR also plays a role in KdFUS3 activation, and if so, whether the embryogenesis and organogenesis pathways are linked by the central regulator. ABA is known to indirectly activate FUS3 (Kagaya et al., 2005a), and so direct application of the hormone to KdTORantisense leaf margins could reveal whether TOR is upstream of this activation. Further hormones such as GA, known to be inhibited by FUS3 (Gazzarrini et al., 2004), are also worth investigation.

The interactions of TOR, auxin, PIN1, LEC2, FUS3 and YUCCA4 suggest a complex network of activation and inhibition may exist to direct plantlet formation in K. daigremontiana. 35S::KdTORantisense and KdTORp::GUS provide the tools with which to elucidate this potentially complex network.

SnRK1 is an important antagonist of the TOR kinase; the two are thought to dominate a complex network, acting in opposition to direct growth according to nutrient availability. Contrary to TOR, SnRK1 is activated by starvation, energy deprivation and abiotic

62 stresses, and negatively regulates growth and development (Baena-Gonzalez and Sheen, 2008). SnRK1 function in K. daigremontiana plantlet formation was thus investigated via the generation of 35S::KdSnRK1antisense transgenic lines. Degenerate primers were used to clone a 157 bp region KdSnRK1 exon 3. The fragment was ligated in pGEM®-T Easy and assembled in a modified pBI121 vector alongside 35S promoter and terminator by Golden Gate Cloning (Fig. 14A). The 35S::KdSnRK1antisense construct was transformed into K. daigremontiana, and transgenic calli were generated. Calli are currently in sub-culture on SEM and will be ready for transfer to soil in ~1 month.

The yeast and animal orthologs of plant SnRK1, SNF1 (SUCROSE NON-FERMENTING 1) and AMPK (AMP-ACTIVATED PROTEIN KINASE), are known upstream negative regulators of TOR, and it is therefore likely that SnRK1 is capable of TOR inhibition in plants [See review: (Margalha et al., 2019)]. Therefore, we would expect to see plantlet formation in 35S::SnRK1antisense plants regardless of environmental conditions; the opposite phenotype to 35S::KdTORantisense. Silencing of the negative regulator SnRK1 will relieve repression of anabolic processes such as cell division and protein synthesis, driving energy-consuming developmental processes like plantlet formation. Whether inhibition of SnRK1 signalling directly relieves repression of TOR can be investigated by qRT-PCR analysis of TOR expression in 35S::KdSnRK1antisense tissue at the site of plantlet formation.

5 Conclusion

Clearly, the TOR kinase lives up to the name of ‘master regulator’. TOR’s conserved role as central mediator of environmental signals and developmental responses extends to the unique process of K. daigremontiana plantlet formation. The asexual reproduction of clones in Kalanchoë species represents a novel innovation, requiring the reversion of mature, differentiated cells to a totipotent state. As such, the confirmation that TOR signalling plays a key role in this process demonstrates how the recruitment of a conserved signalling pathway has contributed to the diversity and success of land plants. We have demonstrated that TOR is expressed throughout plantlet development, from

63 initiation through to the development of roots, and that silencing of TOR results in a reduction in plantlet number, and changes to plantlet morphology. Furthermore, the generation of 35S::KdTORantisense and KdTORp::GUS transgenic lines provides the platform with which to begin investigating the entire TOR network controlling K. daigremontiana plantlet formation. Chemical and molecular applications will be used to elucidate the network of genes, hormones and nutrients that are closely interlinked to direct plantlet formation. Generation of 35S::KdSnRK1antisense lines will allow investigation of the TOR antagonist SnRK1, to determine whether this kinase is too conserved in the complex signalling network. We hope that this work inspires future studies into the involvement of TOR kinase in unique plant mechanisms, in order to explore how the conserved signalling module has contributed to the remarkable diversity within the plant kingdom.

6 References

Adami, A., Garcia-Alvarez, B., Arias-Palomo, E., Barford, D., and Llorca, O. (2007). Structure of TOR and its complex with KOG1. Molecular Cell 27, 509-516. Ahn, C.S., Ahn, H.K., and Pai, H.S. (2015). Overexpression of the PP2A regulatory subunit Tap46 leads to enhanced plant growth through stimulation of the TOR signalling pathway. Journal of Experimental Botany 66, 827-840. Ahn, C.S., Han, J.A., Lee, H.S., Lee, S., and Pai, H.S. (2011). The PP2A Regulatory Subunit Tap46, a Component of the TOR Signaling Pathway, Modulates Growth and Metabolism in Plants. Plant Cell 23, 185-209. Albert, V., and Hall, M.N. (2015). mTOR signaling in cellular and organismal energetics. Current Opinion in Cell Biology 33, 55-66. Anderson, G.H., Veit, B., and Hanson, M.R. (2005). The Arabidopsis AtRaptor genes are essential for post-embryonic plant growth. BioMed Central Biology 3. Baena-Gonzalez, E., and Sheen, J. (2008). Convergent energy and stress signaling. Trends in Plant Science 13, 474-482. Barrada, A., Djendli, M., Desnos, T., Mercier, R., Robaglia, C., Montane, M.H., and Menand, B. (2019). A TOR-YAK1 signaling axis controls cell cycle, meristem activity and plant growth in Arabidopsis. Development 146. Batygina, T., Bragina, E., and Titova, G. (1996). Morphogenesis of propagules in viviparous species Bryophyllum daigremontianum and B. calycinum. Acta Societatis Botanicorum Poloniae 65, 127-133. Bentsink, L., and Koornneef, M. (2008). Seed dormancy and germination. Arabidopsis Book 6, e0119- e0119. Bilsborough, G.D., Runions, A., Barkoulas, M., Jenkins, H.W., Hasson, A., Galinha, C., Laufs, P., Hay, A., Prusinkiewicz, P., and Tsiantis, M. (2011). Model for the regulation of Arabidopsis thaliana leaf margin development. Proceedings of the National Academy of Sciences 108, 3424-3429. Blazquez, M.A., Soowal, L.N., Lee, I., and Weigel, D. (1997). LEAFY expression and flower initiation in Arabidopsis. Development 124, 3835-3844. Braybrook, S.A., and Harada, J.J. (2008). LECs Go Crazy in Embryo Development. Trends in Plant Science 13, 624-630. Caldana, C., Li, Y., Leisse, A., Zhang, Y., Bartholomaeus, L., Fernie, A.R., Willmitzer, L., and Giavalisco, P. (2013). Systemic analysis of inducible target of rapamycin mutants reveal a general metabolic switch controlling growth in Arabidopsis thaliana. Plant Journal 73, 897-909. Carpentier, S.C., Witters, E., Laukens, K., Deckers, P., Swennen, R., and Panis, B. (2005). Preparation of protein extracts from recalcitrant plant tissues: an evaluation of different methods for two- dimensional gel electrophoresis analysis. Proteomics 5, 2497-2507. Chen, G.H., Liu, M.J., Xiong, Y., Sheen, J., and Wu, S.H. (2018). TOR and RPS6 transmit light signals to enhance protein translation in deetiolating Arabidopsis seedlings. Proc Natl Acad Sci U S A 115, 12823-12828. Cho, L.H., Yoon, J., and An, G. (2017). The control of flowering time by environmental factors. Plant Journal 90, 708-719. Chuck, G., Lincoln, C., and Hake, S. (1996). KNAT1 induces lobed leaves with ectopic meristems when overexpressed in Arabidopsis. The Plant Cell 8, 1277-1289.

65 de Jong, M., and Leyser, O. (2012). Developmental plasticity in plants. Cold Spring Harbor symposia on quantitative biology 77, 63-73. De Vleesschauwer, D., Filipe, O., Hoffman, G., Seifi, H.S., Haeck, A., Canlas, P., Van Bockhaven, J., De Waele, E., Demeestere, K., Ronald, P., et al. (2018). Target of rapamycin signaling orchestrates growth-defense trade-offs in plants. New Phytologist 217, 305-319. Deng, K., Yu, L., Zheng, X., Zhang, K., Wang, W., Dong, P., Zhang, J., and Ren, M. (2016). Target of Rapamycin Is a Key Player for Auxin Signaling Transduction in Arabidopsis. Frontiers in Plant Science 7. Deprost, D., Truong, H.N., Robaglia, C., and Meyer, C. (2005). An Arabidopsis homolog of RAPTOR/KOG1 is essential for early embryo development. Biochemical and Biophysical Research Communications 326, 844-850. Deprost, D., Yao, L., Sormani, R., Moreau, M., Leterreux, G., Nicolai, M., Bedu, M., Robaglia, C., and Meyer, C. (2007). The Arabidopsis TOR kinase links plant growth, yield, stress resistance and mRNA translation. EMBO Reports 8, 864-870. Diaz-Troya, S., Florencio, F.J., and Crespo, J.L. (2008). Target of rapamycin and LST8 proteins associate with membranes from the endoplasmic reticulum in the unicellular green alga Chlamydomonas reinhardtii. Eukaryotic Cell 7, 212-222. Dinkova, T.D., Aguilar, R., and DEe Jimenéz, E.S. (2000). Expression of maize eukaryotic initiation factor (eIF) iso4E is regulated at the translational level. Biochemical Journal 351, 825-831. Dobrenel, T., Caldana, C., Hanson, J., Robaglia, C., Vincentz, M., Veit, B., and Meyer, C. (2016). TOR Signaling and Nutrient Sensing. In Annual Review of Plant Biology, Vol 67, S.S. Merchant, ed., pp. 261-285. Dong, P., Xiong, F.J., Que, Y.M., Wang, K., Yu, L.H., Li, Z.G., and Ren, M.Z. (2015). Expression profiling and functional analysis reveals that TOR is a key player in regulating photosynthesis and phytohormone signaling pathways in Arabidopsis. Frontiers in Plant Science 6. Duan, H.Y., Li, F.G., Wu, X.D., Ma, D.M., Wang, M., and Hou, Y.X. (2006). The cloning and sequencing of a cDNA encoding a WD repeat protein in cotton (Gossypium hirsutum L.). DNA Sequence 17, 49-55. Dumas, C., and Rogowsky, P. (2008). Fertilization and early seed formation. Comptes Rendus Biologies 331, 715-725. Endrizzi, K., Moussian, B., Haecker, A., Levin, J.Z., and Laux, T. (1996). The SHOOT MERISTEMLESS gene is required for maintenance of undifferentiated cells in Arabidopsis shoot and floral meristems and acts at a different regulatory level than the meristem genes WUSCHEL and ZWILLE. The plant journal 10, 967-979. Garces, H., and Sinha, N. (2009a). Extraction of DNA from the Plant Kalanchoe daigremontiana. Cold Spring Harbor protocols 2009, pdb.prot5304. Garces, H., and Sinha, N. (2009b). The 'mother of thousands' (Kalanchoe daigremontiana): a plant model for asexual reproduction and CAM studies. Cold Spring Harbor protocols 2009, pdb.emo133. Garces, H., and Sinha, N. (2009c). Transformation of the plant Kalanchoe daigremontiana using Agrobacterium tumefaciens. Cold Spring Harbor protocols 2009, pdb.prot5303. Garces, H.M., Champagne, C.E., Townsley, B.T., Park, S., Malho, R., Pedroso, M.C., Harada, J.J., and Sinha, N.R. (2007). Evolution of asexual reproduction in leaves of the genus Kalanchoe. Proc Natl Acad Sci U S A 104, 15578-15583.

Garces, H.M., Koenig, D., Townsley, B.T., Kim, M., and Sinha, N.R. (2014). Truncation of LEAFY COTYLEDON1 protein is required for asexual reproduction in Kalanchoe daigremontiana. Plant Physiology 165, 196-206. Gazzarrini, S., Tsuchiya, Y., Lumba, S., Okamoto, M., and McCourt, P. (2004). The transcription factor FUSCA3 controls developmental timing in Arabidopsis through the hormones gibberellin and abscisic acid. Developmental Cell 7, 373-385. Gonzalez, A., and Hall, M.N. (2017). Nutrient sensing and TOR signaling in yeast and mammals. EMBO Journal 36, 397-408. Gonzalez, N., Vanhaeren, H., and Inze, D. (2012). Leaf size control: complex coordination of cell division and expansion. Trends in Plant Science 17, 332-340. Guo, J., Liu, H., He, Y., Cui, X., Du, X., and Zhu, J. (2015). Origination of asexual plantlets in three species of Crassulaceae. Protoplasma 252, 591-603. Heitman, J., Movva, N.R., and Hall, M.N. (1991). Targets for Cell-Cycle Arrest by the Immunosuppressant Rapamycin in Yeast. Science 253, 905-909. Irish, V. (2017). The ABC model of floral development. Current Biology 27, R887-R890. Jacinto, E., Loewith, R., Schmidt, A., Lin, S., Ruegg, M.A., Hall, A., and Hall, M.N. (2004). Mammalian TOR complex 2 controls the actin cytoskeleton and is rapamycin insensitive. Nature Cell Biology 6, 1122-U1130. Johnson, M.A. (1934). The origin of the foliar pseudo-bulbils in Kalanchoe daigremontiana. Bulletin of the Torrey Botanical Club, 355-366. Kagaya, Y., Okuda, R., Ban, A., Toyoshima, R., Tsutsumida, K., Usui, H., Yamamoto, A., and Hattori, T. (2005a). Indirect ABA-dependent regulation of seed storage protein genes by FUSCA3 transcription factor in Arabidopsis. Plant and Cell Physiology 46, 300-311. Kagaya, Y., Toyoshima, R., Okuda, R., Usui, H., Yamamoto, A., and Hattori, T. (2005b). LEAFY COTYLEDON1 controls seed storage protein genes through its regulation of FUSCA3 and ABSCISIC ACID INSENSITIVE3. Plant and Cell Physiology 46, 399-406. Kozlov, A.M., Darriba, D., Flouri, T., Morel, B., and Stamatakis, A. (2019). RAxML-NG: a fast, scalable and user-friendly tool for maximum likelihood phylogenetic inference. Bioinformatics 35, 4453- 4455. Kunz, J., Henriquez, R., Schneider, U., Deuterreinhard, M., Movva, N.R., and Hall, M.N. (1993). Target of Rapamycin In Yeast, TOR2, is an Essential Phosphatidylinositol Kinase Homolog Required for G(1) Progression. Cell 73, 585-596. Lastdrager, J., Hanson, J., and Smeekens, S. (2014). Sugar signals and the control of plant growth and development. Journal of Experimental Botany 65, 799-807. Li, L.X., Song, Y., Wang, K., Dong, P., Zhang, X.Y., Li, F.G., Li, Z.G., and Ren, M.Z. (2015). TOR-inhibitor insensitive-1 (TRIN1) regulates cotyledons greening in Arabidopsis. Frontiers in Plant Science 6. Li, X.J., Cai, W.G., Liu, Y.L., Li, H., Fu, L.W., Liu, Z.Y., Xu, L., Liu, H.T., Xu, T.D., and Xiong, Y. (2017). Differential TOR activation and cell proliferation in Arabidopsis root and shoot apexes. Proc Natl Acad Sci U S A 114, 2765-2770. Liu, C., Zhu, C., and Zeng, H.M. (2016). Key KdSOC1 gene expression profiles during plantlet morphogenesis under hormone, photoperiod, and drought treatments. Genetics and Molecular Research 15.

Liu, K.H., Niu, Y.J., Konishi, M., Wu, Y., Du, H., Chung, H.S., Li, L., Boudsocq, M., McCormack, M., Maekawa, S., et al. (2017). Discovery of nitrate-CPK-NLP signalling in central nutrient-growth networks. Nature 545, 311-+. Liu, Y.M., and Bassham, D.C. (2010). TOR Is a Negative Regulator of Autophagy in Arabidopsis thaliana. PLoS One 5. Lledias, F., Hernandez, F., Rivas, V., Garcia-Mendoza, A., Cassab, G.I., and Nieto-Sotelo, J. (2017). A Rapid and Reliable Method for Total Protein Extraction from Succulent Plants for Proteomic Analysis. The protein journal 36, 308-321. Long, J.A., and Barton, M.K. (1998). The development of apical embryonic pattern in Arabidopsis. Development 125, 3027-3035. Long, J.A., Moan, E.I., Medford, J.I., and Barton, M.K. (1996). A member of the KNOTTED class of homeodomain proteins encoded by the STM gene of Arabidopsis. Nature 379, 66. Lotan, T., Ohto, M.-a., Yee, K.M., West, M.A., Lo, R., Kwong, R.W., Yamagishi, K., Fischer, R.L., Goldberg, R.B., and Harada, J.J. (1998). Arabidopsis LEAFY COTYLEDON1 is sufficient to induce embryo development in vegetative cells. Cell 93, 1195-1205. Luerßen, H., Kirik, V., Herrmann, P., and Miséra, S. (1998). FUSCA3 encodes a protein with a conserved VP1/ABI3‐like B3 domain which is of functional importance for the regulation of seed maturation in Arabidopsis thaliana. The Plant Journal 15, 755-764. Maddison, W.P., and Knowles, L.L. (2006). Inferring Phylogeny Despite Incomplete Lineage Sorting. Systematic Biology 55, 21-30. Magallón, S., Gómez-Acevedo, S., Sánchez-Reyes, L.L., and Hernández-Hernández, T. (2015). A metacalibrated time-tree documents the early rise of flowering plant phylogenetic diversity. New Phytologist 207, 437-453. Mahfouz, M.M., Kim, S., Delauney, A.J., and Verma, D.P.S. (2006). Arabidopsis TARGET OF RAPAMYCIN interacts with RAPTOR, which regulates the activity of S6 kinase in response to osmotic stress signals. Plant Cell 18, 477-490. Margalha, L., Confraria, A., and Baena-González, E. (2019). SnRK1 and TOR: modulating growth– defense trade-offs in plant stress responses. Journal of Experimental Botany 70, 2261-2274. Meinke, D.W. (1992). A Homoeotic Mutant of Arabidopsis thaliana with Leafy Cotyledons. Science 258, 1647-1650. Meinke, D.W., Franzmann, L.H., Nickle, T.C., and Yeung, E.C. (1994). Leafy Cotyledon Mutants of Arabidopsis. Plant Cell 6, 1049-1064. Menand, B., Desnos, T., Nussaume, L., Berger, F., Bouchez, D., Meyer, C., and Robaglia, C. (2002). Expression and disruption of the Arabidopsis TOR (target of rapamycin) gene. Proc Natl Acad Sci U S A 99, 6422-6427. Mohammed, B., Bilooei, S.F., Doczi, R., Grove, E., Railo, S., Palme, K., Ditengou, F.A., Bogre, L., and Lopez-Juez, E. (2018). Converging Light, Energy and Hormonal Signaling Control Meristem Activity, Leaf Initiation, and Growth. Plant Physiology 176, 1365-1381. Moore, M.J., Soltis, P.S., Bell, C.D., Burleigh, J.G., and Soltis, D.E. (2010). Phylogenetic analysis of 83 plastid genes further resolves the early diversification of eudicots. Proceedings of the National Academy of Sciences 107, 4623-4628. Moreau, M., Azzopardi, M., Clement, G., Dobrenel, T., Marchive, C., Renne, C., Martin-Magniette, M.L., Taconnat, L., Renou, J.P., Robaglia, C., et al. (2012). Mutations in the Arabidopsis Homolog of

LST8/G beta L, a Partner of the Target of Rapamycin Kinase, Impair Plant Growth, Flowering, and Metabolic Adaptation to Long Days. Plant Cell 24, 463-481. Moyroud, E., and Glover, B.J. (2017). The Evolution of Diverse Floral Morphologies. Current Biology 27, R941-R951. Nanjareddy, K., Blanco, L., Arthikala, M.K., Alvarado-Affantranger, X., Quinto, C., Sanchez, F., and Lara, M. (2016). A Legume TOR Protein Kinase Regulates Rhizobium Symbiosis and Is Essential for Infection and Nodule Development. Plant Physiology 172, 2002-2020. Nukarinen, E., Nagele, T., Pedrotti, L., Wurzinger, B., Mair, A., Landgraf, R., Bornke, F., Hanson, J., Teige, M., Baena-Gonzalez, E., et al. (2016). Quantitative phosphoproteomics reveals the role of the AMPK plant ortholog SnRK1 as a metabolic master regulator under energy deprivation. Scientific Reports 6. Overvoorde, P., Fukaki, H., and Beeckman, T. (2010). Auxin control of root development. Cold Spring Harb Perspect Biol 2, a001537-a001537. Palmer, C.M., Bush, S.M., and Maloof, J.N. (2012). Phenotypic and Developmental Plasticity in Plants. In eLS. Parcy, F., Valon, C., Kohara, A., Misera, S., and Giraudat, J. (1997). The ABSCISIC ACID-INSENSITIVE3, FUSCA3, and LEAFY COTYLEDON1 loci act in concert to control multiple aspects of Arabidopsis seed development. Plant Cell 9, 1265-1277. Pfeiffer, A., Janocha, D., Dong, Y.H., Medzihradszky, A., Schone, S., Daum, G., Suzaki, T., Forner, J., Longenecker, T., Rempel, E., et al. (2016). Integration of light and metabolic signals for stem cell activation at the shoot apical meristem. Elife 5. Powell, A.E., and Lenhard, M. (2012). Control of Organ Size in Plants. Current Biology 22, R360-R367. Ren, M.Z., Venglat, P., Qiu, S.Q., Feng, L., Cao, Y.G., Wang, E., Xiang, D.Q., Wang, J.H., Alexander, D., Chalivendra, S., et al. (2012). Target of Rapamycin Signaling Regulates Metabolism, Growth, and Life Span in Arabidopsis. Plant Cell 24, 4850-4874. Sabatini, D.M., Erdjumentbromage, H., Lui, M., Tempst, P., and Snyder, S.H. (1994). RAFT1 - A Mammalian Protein That Binds to FKBP12 in a Rapamycin-Dependent Fashion and is Homologous to Yeast TORS. Cell 78, 35-43. Sablowski, R.W.M., and Meyerowitz, E.M. (1998). A homolog of NO APICAL MERISTEM is an immediate target of the floral homeotic genes APETALA3/PISTILLATA. Cell 92, 93-103. Salem, M.A., Li, Y., Wiszniewski, A., and Giavalisco, P. (2017). Regulatory-associated protein of TOR (RAPTOR) alters the hormonal and metabolic composition of Arabidopsis seeds, controlling seed morphology, viability and germination potential. Plant Journal 92, 525-545. Sarbassov, D.D., Ali, S.M., Kim, D.H., Guertin, D.A., Latek, R.R., Erdjument-Bromage, H., Tempst, P., and Sabatini, D.M. (2004). Rictor, a novel binding partner of mTOR, defines a rapamycin-insensitive and raptor-independent pathway that regulates the cytoskeleton. Current Biology 14, 1296- 1302. Sarbassov, D.D., Ali, S.M., Sengupta, S., Sheen, J.H., Hsu, P.P., Bagley, A.F., Markhard, A.L., and Sabatini, D.M. (2006). Prolonged rapamycin treatment inhibits mTORC2 assembly and Akt/PKB. Molecular Cell 22, 159-168. Saxton, R.A., and Sabatini, D.M. (2017). mTOR Signaling in Growth, Metabolism, and Disease. Cell 169, 362-362.

Schepetilnikov, M., Dimitrova, M., Mancera-Martinez, E., Geldreich, A., Keller, M., and Ryabova, L.A. (2013). TOR and S6K1 promote translation reinitiation of uORF-containing mRNAs via phosphorylation of eIF3h. EMBO Journal 32, 1087-1102. Schepetilnikov, M., and Ryabova, L.A. (2018). Recent Discoveries on the Role of TOR (Target of Rapamycin) Signaling in Translation in Plants. Plant Physiology 176, 1095-1105. Sehgal, S.N., Baker, H., and Vezina, C. (1975). Rapamycin (AY-22,989), a new antifungal antibiotic. II. Fermentation, isolation and characterization. The Journal of antibiotics 28, 727-732. Sentoku, N., Sato, Y., and Matsuoka, M. (2000). Overexpression of rice OSH genes induces ectopic shoots on leaf sheaths of transgenic rice plants. Developmental biology 220, 358-364. Shi, L., Wu, Y., and Sheen, J. (2018). TOR signaling in plants: conservation and innovation. Development 145, dev160887. Sinha, N.R., Williams, R.E., and Hake, S. (1993). Overexpression of the maize homeo box gene, KNOTTED-1, causes a switch from determinate to indeterminate cell fates. Genes & Development 7, 787-795. Smith, L.G., Jackson, D., and Hake, S. (1995). Expression of knotted1 marks shoot meristem formation during maize embryogenesis. Developmental Genetics 16, 344-348. Somssich, M., Je, B.I., Simon, R., and Jackson, D. (2016). CLAVATA-WUSCHEL signaling in the shoot meristem. Development 143, 3238-3248. Sormani, R., Yao, L., Menand, B., Ennar, N., Lecampion, C., Meyer, C., and Robaglia, C. (2007). Saccharomyces cerevisiae FKBP12 binds Arabidopsis thaliana TOR and its expression in plants leads to rapamycin susceptibility. Bmc Plant Biology 7. Stone, S.L., Kwong, L.W., Yee, K.M., Pelletier, J., Lepiniec, L., Fischer, R.L., Goldberg, R.B., and Harada, J.J. (2001). LEAFY COTYLEDON2 encodes a B3 domain transcription factor that induces embryo development. Proceedings of the National Academy of Sciences 98, 11806-11811. Szecsi, J., Joly, C., Bordji, K., Varaud, E., Cock, J.M., Dumas, C., and Bendahmane, M. (2006). BIGPETALp, a bHLH transcription factor is involved in the control of Arabidopsis petal size. EMBO Journal 25, 3912-3920. Tang, L.P., Zhou, C., Wang, S.S., Yuan, J., Zhang, X.S., and Su, Y.H. (2017). FUSCA 3 interacting with LEAFY COTYLEDON 2 controls lateral root formation through regulating YUCCA 4 gene expression in Arabidopsis thaliana. New Phytologist 213, 1740-1754. Tatebe, H., and Shiozaki, K. (2017). Evolutionary Conservation of the Components in the TOR Signaling Pathways. Biomolecules 7. Tzeng, T.Y., Kong, L.R., Chen, C.H., Shaw, C.C., and Yang, C.H. (2009). Overexpression of the Lily p70(s6k) Gene in Arabidopsis Affects Elongation of Flower Organs and Indicates TOR- Dependent Regulation of AP3, PI and SUP Translation. Plant and Cell Physiology 50, 1695-1709. Vanstraelen, M., and Benkova, E. (2012). Hormonal Interactions in the Regulation of Plant Development. In Annual Review of Cell and Developmental Biology, Vol 28, R. Schekman, ed., pp. 463-487. Varaud, E., Brioudes, F., Szecsi, J., Leroux, J., Brown, S., Perrot-Rechenmann, C., and Bendahmane, M. (2011). AUXIN RESPONSE FACTOR8 Regulates Arabidopsis Petal Growth by Interacting with the bHLH Transcription Factor BIGPETALp. Plant Cell 23, 973-983. Vicient, C.M., Bies-Etheve, N., and Delseny, M. (2000). Changes in gene expression in the leafy cotyledon1 (lec1) and fusca3 (fus3) mutants of Arabidopsis thaliana L. Journal of Experimental Botany 51, 995-1003.

Vollbrecht, E., Reiser, L., and Hake, S. (2000). Shoot meristem size is dependent on inbred background and presence of the maize homeobox gene, knotted1. Development 127, 3161-3172. Wang, P., Zhao, Y., Li, Z., Hsu, C.-C., Liu, X., Fu, L., Hou, Y.-J., Du, Y., Xie, S., and Zhang, C. (2018). Reciprocal regulation of the TOR kinase and ABA receptor balances plant growth and stress response. Molecular cell 69, 100-112. e106. Wang, S.C., Chang, Y., Guo, J.J., Zeng, Q.N., Ellis, B.E., and Chen, J.G. (2011). Arabidopsis Ovate Family Proteins, a Novel Transcriptional Repressor Family, Control Multiple Aspects of Plant Growth and Development. PLoS One 6. Wang, W., Tai, F., and Chen, S. (2008). Optimizing protein extraction from plant tissues for enhanced proteomics analysis. Journal of separation science 31, 2032-2039. Weigel, D., and Glazebrook, J. (2002). Arabidopsis : a laboratory manual (Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory Press). West, M., Yee, K.M., Danao, J., Zimmerman, J.L., Fischer, R.L., Goldberg, R.B., and Harada, J.J. (1994). LEAFY COTYLEDON1 Is an Essential Regulator of Late Embryogenesis and Cotyledon Identity in Arabidopsis. Plant Cell 6, 1731-1745. Wullschleger, S., Loewith, R., and Hall, M.N. (2006). TOR signaling in growth and metabolism. Cell 127, 5-19. Xiong, F., Dong, P., Liu, M., Xie, G., Wang, K., Zhuo, F., Feng, L., Yang, L., Li, Z., and Ren, M. (2016). Tomato FK506 binding protein 12KD (FKBP12) mediates the interaction between rapamycin and target of rapamycin (TOR). Frontiers in Plant Science 7, 1746. Xiong, F.J., Zhang, R., Meng, Z.G., Deng, K.X., Que, Y.M., Zhuo, F.P., Feng, L., Guo, S.D., Datla, R., and Ren, M.Z. (2017). Brassinosteriod Insensitive 2 (BIN2) acts as a downstream effector of the Target of Rapamycin (TOR) signaling pathway to regulate photoautotrophic growth in Arabidopsis. New Phytologist 213, 233-249. Xiong, Y., McCormack, M., Li, L., Hall, Q., Xiang, C.B., and Sheen, J. (2013). Glucose-TOR signalling reprograms the transcriptome and activates meristems. Nature 496, 181-+. Xiong, Y., and Sheen, J. (2012). Rapamycin and Glucose-Target of Rapamycin (TOR) Protein Signaling in Plants. Journal of Biological Chemistry 287, 2836-2842. Xiong, Y., and Sheen, J. (2014). The Role of Target of Rapamycin Signaling Networks in Plant Growth and Metabolism. Plant Physiology 164, 499-512. Zhang, S., Wong, L., Meng, L., and Lemaux, P.G. (2002). Similarity of expression patterns of knotted1 and ZmLEC1 during somatic and zygotic embryogenesis in maize (Zea mays L.). Planta 215, 191- 194. Zhu, C., Wang, L., Chen, J., Liu, C., Zeng, H., and Wang, H. (2017). Over-expression of KdSOC1 gene affected plantlet morphogenesis in Kalanchoe daigremontiana. Scientific Reports 7, 5629. Zoncu, R., Bar-Peled, L., Efeyan, A., Wang, S., Sancak, Y., and Sabatini, D.M. (2011). mTORC1 senses lysosomal amino acids through an inside-out mechanism that requires the vacuolar H(+)- ATPase. Science 334, 678-683.