Regulation of Plant Defense Genes Against Bacterial Pathogens

(1)

IN

DEGREE PROJECT BIOTECHNOLOGY, SECOND CYCLE, 30 CREDITS

STOCKHOLM SWEDEN 2021,

Regulation of Plant Defense Genes Against Bacterial Pathogens

JENNY SJÖSTRÖM

KTH ROYAL INSTITUTE OF TECHNOLOGY

SCHOOL OF ENGINEERING SCIENCES IN CHEMISTRY, BIOTECHNOLOGY AND HEALTH

(2)

i

Abstract

Several factors contribute to the demand of new, sustainable solutions to bring food security to the world population. The United Nations predicts, with a confidence of 95%, that the world population will be between 9.9 and 12.7 billion by year 2100. At the same time plant agriculture as seen today is threatened by climate changes e.g., rising temperatures and more extreme weather conditions. In addition, plant bacterial pathogens reduce yields, and cause losses of over $1 billion dollars worldwide every year to the food production chain. The currently most used and effective treatment against bacterial infections on crops is antibiotics, but this is not a viable alternative for most growers due to increasing antibiotic resistance and the high development, production, and distribution cost. During the upcoming years development of new approaches against bacterial infections on crops is of high importance but currently there are information gaps in the field of plant defense regulation systems.

This study was aimed to provide knowledge about the transcriptional regulation of genes that are included in plant immune system towards bacteria. To investigate this, conserved regulatory elements of the upstream sequences of two defense-related plant receptor kinases, FLS2 and SERK1, was searched for in different species. FLS2 is a surface receptor that recognizes a peptide derived from the bacterial flagellin protein, and is part of the pathogen-triggered immunity response of most of higher plants. In FLS2 no conserved module was found but a single motif, CAACTTG, is conserved in all chosen species. In SERK1 a strikingly long and conserved module was found. Both the FLS2 motif and two motifs in the SERK1 module are recognition motifs with MYC2, a transcription factor involved in different plant mechanisms and the regulation of phytohormones like abscisic acid and auxin. To address whether MYC2 is involved in the transcriptional regulation of FLS2, an experimental approach is described, involving transactivation by MYC2 of FLS2 reporter constructs, studies using agroinfiltration in Nicotiana benthamiana. An increased knowledge about the different components and mechanisms of plant defense regulation will help the research towards new bactericides, transgenic plants, and other ways to secure food for upcoming generations.

KEYWORDS: PLANT DEFENSE, PTI, FLS2, SERK1, MYC2, REGULATION, TRANSCRIPTION FACTOR

(3)

ii

Sammanfattning

Sjukdom på grödor orsakad av bakterier kan bidra till ekonomiska förluster för bönder samt brist på mat, därför är det viktigt att utveckla nya hållbara sätt att motverka och behandla grödor mot bakterier. Idag är det mest vanliga tillvägagångssättet antibiotika men detta är inte hållbart p.g.a uppkomst av antibiotikaresistens. Antibiotika är inte heller tillgängligt för alla bönder och grödor då kostnaden blir för hög. Världsbefolkningen växer och om 80 år beräknas det bo mellan 9.9 och 12.7 biljoner (95% konfidens) människor på jorden. Växande befolkning samt ökande klimatförändringar, som torka och höjda temperaturer kräver nya bekämpningsmetoder mot bakterier för att tillgodose behoven i framtiden. Det saknas information om hur växter hanterar och reglerar bakteriella hot, därför är målet med denna studie att bidra med kunskap kring den transkriptionella regleringen av växters immunsystem mot bakterier. För att göra detta har promotorsekvenser hos gener som är förknippade med immunförsvaret i växter undersökts efter konserverade regulatoriska element. En känd receptor, FLS2 har en stor roll i växters försvar mot bakterier och känner igen en peptid från bakteriers flagell. Denna studie har undersökt FLS2 och den sammankopplade receptorn SERK1. Hos FLS2 kunde ingen konserverad modul hittas i uppströmsekvensen, däremot observerades ett 8 bp långt motiv, CAACTTG, i alla undersökta arter. I SERK1 hittades en lång konserverad modul bestående av flera motiv. Både FLS2-motifet och två motiv i SERK1-modulen binds av transkriptionsfaktorn MYC2.

För att testa hypotesen att MYC2 bidrar till den transkriptionella regleringen av FLS2 och SERK1 har en experimentell plan utformats, där Nicotiana benthamiana transfekteras av Agrobacterium tumefaciens innehållandes promotorsekvenserna samt generna till transkriptionsfaktorn MYC2. En ökad förståelse kring de olika delarna och mekanismerna som medverkar inom växters immunförsvar kan bidra till den fortsatta forskningen mot hållbara lösningar till att säkra mat i framtiden.

NYCKELORD: VÄXTFÖRSVAR, PTI, FLS2, SERK1, REGULERING, TRANSKRIPTIONFAKTOR

(4)

Preface

This study was performed as my degree project of 30 credits to complete my Master´s degree in Industrial and environmental Biotechnology and thereby also my engineering degree at the Royal Institute of Technology. The project was performed at the department of Industrial Biotechnology at the CBH-school.

I would like to thank my supervisor Ines Ezcurra for her guidance and support during the project with an exceptional expertise in the area, always able to answer my questions and for introducing me to this previously unknown topic which I have found very interesting. I would also like to thank my co- supervisor for her patience and encouragement during the whole project, with presentations, writing and everything else in between. I would like to direct special thanks to my partner, friends, and family for always being there during these months.

(6)

2

Abbreviations

ABA – abscisic acid

BAK1 – BRI-associated kinase BRI1- brassinosteroid insensitive 1

DAMP – damage associated molecular pattern EBF1 - EIN3-binding F-box protein 1

EIN3 – ethylene insensitive 1 ET - ethylene

ETI – effector triggered immunity FLS2 – flagellin sensitive 2 GUS – β-glucuronidase HR – hypersensitive response JA – jasmonic acid

LRR – leucine rich repeats

LRR-RK – leucine rich repeats receptor kinase

MAMP – microbe associated molecular pattern

MAPK – mitogen-activated protein kinase NB-LRR – nucleotide binding leucine rich repeats

OE – overlap extension

PAMP – pathogen associated molecular pattern

PR1 – phytogenesis related protein 1 PTI – PAMP triggered immunity RK – receptor kinase

SA – salicylic acid

SAR – systemic acquired resistance SERK – somatic embryonic receptor kinase Sm^R – streptomycin resistance

TF – transcription factor

(7)

3

Introduction

Plants must endure both abiotic and biotic stresses, the latter consisting also of fungi, viruses, and bacteria. For the most part the plant stress defense system can withstand infections and thereby stay healthy but as the pathogens evolve and create mechanisms to breach the defense, they can cause disease which is a threat to crop agriculture (Dou & Zhou, 2012).

Infection of crops is a threat to agriculture and bacterial infections contribute to large annual losses of

$1 billion dollars around the globe (Martins, et al., 2018). Climate change is imposing other threats such as higher temperatures, drought and risen sea levels, forcing the need of environmentally and financially sustainable approaches to increase crop yield and to obtain food security for a fast growing world population. Residents of our planet are expected to reacha level between 9.9 and 12.7 billion by year 2100 (United Nations, 2019). By increasing our knowledge about the regulatory networks involved in bacterial defense responses, it is possible to engineer plants with enhanced immune system and thereby increase yield and also minimize the use of antibacterial and other harmful compounds in the fields. For example, the expression of a given gene involved in plant defense could be modified to enhance its function, by editing its cis-regulatory motifs.

This study investigates the regulation of one receptor involved in the bacterial defense system, FLAGELLIN-SENSING 2 (FLS2) and an associated co-receptor called SOMATIC EMBRYOGENESIS- RELATED KINASE 1 (SERK1). With bioinformatical approaches in silico the regulatory upstream flanks of the receptor genes are analyzed with the aim to find conserved motif modules. In the upstream regulatory sequences of FLS2 and SERK1 a conserved motif and a conserved motif module is identified, respectively. Additionally, an approach is described to experimentally verify the functional purpose of the identified elements. The results from the experimental verification could provide helpful information in the future research towards engineering plants with rewired regulatory networks for enhances resistance towards bacterial pathogens.

(8)

4

Background and theory

The plant defense is a two-tier system

Plants have a constitutive and an inducible antimicrobial defense. The pathogen-inducible defense can be described by a two-level immune system. The first level, known as PAMP-triggered immunity (PTI), is when pathogen/microbe associated molecular patterns (PAMP/MAMP) are recognized by pattern recognizing receptors (PRRs) in the plasma membrane. To circumvent this protection pathogens have developed so called effectors. Effectors are proteins that enhance virulence by direct interaction with the host (Dou & Zhou, 2012) and are secreted either directly into the cell or to the membrane to suppress the PTI (Tsuda & Somssich, 2015). The effectors are typically recognized by the plant´s nucleotide binding leucine-rich repeats receptors (NB-LRRs) leading to activation of the second level of the immune system: effector triggered immunity (ETI). The activation triggers several stress responses, such as the hypersensitive response (HR). The hypersensitive response is characterized by local and rapid cell death, to confine and prevent spreading of the pathogen. The HR also alters the metabolic reactions in nearby cells and triggers systemic acquired resistance (SAR) which is a non- specific response to protect the plant against a variety of pathogens (Fritig, et al., 1998). Together, these mechanisms are known as the plant-pathogen molecular arms race (Tsuda & Somssich, 2015).

The PTI and ETI utilize different immune responses. For instance, MAPK-cascades, reactive oxygen species, phytohormones production and transcriptional reprogramming are reactions activated by the PTI, whereas the ETI activates the HR and SAR. These defense responses are regulated by complex transcription factor (TF) networks, where intricate regulation is needed for the plant to set up a defense system that is sufficient and at the same time sustain growth to maximize fitness (Tsuda &

Somssich, 2015).

Antibiotics are not a sustainable control of infections

Increasing world population does not only imply more people to feed but also decreased areas available for agriculture. An effective bacterial disease management is required to ensure food safety but this is difficult with the approaches available today. Two examples of bacterial infections that are causing yield losses are fire blight caused by the bacteria Erwinia amylovora which attacks pome fruit trees (Figure 1) and Candidatus Liberibacter spp that cause citrus greening disease in citrus plants.

(9)

5

Figure 1. A. Pear shoot with fire blight. B. Electron micrograph of an Erwinia cell showing peritrichous flagella. C. Fire blight of apple and pear disease cycle. Adapted from The American Phytopathological Society (APS) portal https://www.apsnet.org/.

The high growth rate in the optimal environment for E. amylovora creates high population sizes resulting in the disease spreading to new infection sites increasing the risk of an epidemic. Further, a rapid rise in population size favors an intensified level of infection between individual hosts which also contribute to disease epidemics. Both E. amylovora and C. liberibacter spp, like many other bacterial infections, mostly target internal locations of the plant (Figure 1) making it unfeasible to pursue control with agents that are applied with spray (Sundin & Wang, 2018).

Looking at the two examples above it is clear that novel approaches are needed to tackle the challenges that bacterial infections impose. Historically, using antibiotics has been the main approach against bacterial infections on plants although almost none of them have been developed for that purpose.

Antibiotics used in agriculture were in principle initially developed for clinical use (Sundin , et al., 2016).

Copper compounds was introduced as a bactericide in 1880´s and streptomycin was introduced in 1950´s and these two compounds have been the most used against bacterial infections, e.g. towards E. amylovora. Today streptomycin together with oxytetracycline is still the most used antibiotic against bacterial infections on high-value fruit, such as apples and pears (McManus, et al., 2002). Although antibiotics has proven most efficient towards bacterial infection they come with several challenges, in low doses they are not efficient enough and in high doses they may become phytotoxic. The most discussed challenge during recent years has been the prevalence of antibiotic resistance, mostly seen as streptomycin resistance (Sm^R) (Sundin & Wang, 2018). The connection between extensive use of

(10)

6 antibiotics and antibiotic resistance is now well established, both in terms of clinical use and utilization in plant agriculture (Sundin & Wang, 2018). Since antibiotics have been used for 70 years in crops, especially in high value fruits, antibiotics resistance was predictable and the control of infections by antibiotics is decreasing (McManus, et al., 2002). Horizontal gene transfer, the mechanism driving antibiotic resistance occur between phylogenetically distinctive bacteria (Sundin , et al., 2016). There are several mechanisms used by bacteria to obtain Sm^R, one of them is utilizing strAB genes. The strAB genes encode for aminoglycoside phosphotransferases which are enzymes that alter the antibiotic to a non-toxic form. The strAB genes are not only prevalent on E. amylovora but also in other plant pathogens like Pseudomonas syringae and Xanthomonas campestris and are always present on the same transposon, Tn5393, this was shown by isolation of the pathogens mentioned above on different continents (Förster, et al., 2015). Strains of E. amylovora with Sm^R genes were detected on pear trees in 1971 in California for the first time, and the following year they were found in Washington and Oregon, but has since then they have spread to a large portion of the country and to other countries e.g., Lebanon and New Zealand as well as regions where streptomycin was never used.

As an alternative to streptomycin, kasugamycin has been introduced as a bactericide in apple and pear orchards where E. amylovora is streptomycin resistant. This antibioticum is currently used in USA and Japan, and in European Union countries it was used until antibiotics were banned in crop agriculture, due to the possible effect on human health (horizontal transfer of resistance genes to clinical pathogens) (Sundin & Wang, 2018) (Sundin , et al., 2016). However, in Switzerland, Austria, and Germany streptomycin is still used to control fire blight in apples and pear (Sundin & Wang, 2018).

At Michigan State University scientists expected that without antibiotics several orchards and common apple cultivars would be abandoned. In the state of Michigan, where Sm^R was first detected, during three years (1997 to 2000) 18% of the apple cultivation areas was lost to fire blight, and 15% of growers left the apple industry for the same reason. Antibiotics are expensive and growers without bacterial infections in their orchards can set lower produce prices which puts them in a pre-eminent position in a competitive market (Sundin & Wang, 2018).

Phytohormones regulate plant defense in complex networks

Phytohormones are molecules synthesized by plants to regulate resistance towards biotic and abiotic stress. The different layers of plant immunity require a fine-tuned network of phytohormones for it to

(11)

7 be effective. The hormones can be either directly part of the immunity signaling, but some signaling molecules could interact with or affect immunity mechanisms indirectly. Most of the phytohormones are involved in various parts of the plant and most of the processes and connections are only partially, or not at all, identified. Ethylene (ET) is a gaseous phytohormone contributing to many aspects of plant signaling and is mostly connected to pathogen resistance response (Anver & Tsuda, 2015). Another phytohormone produced to acquire pathogen resistance is salicylic acid (SA), that is the most important signaling and regulation component of the SAR, a long-lasting defense towards a broad range of biotrophic pathogens (Yasuda, et al., 2008). Further, an important hormone in pathogen resistance is jasmonic acid (JA). JAs are plant hormones that are essential in a vast number of processes, including but not limited to: necrotrophic pathogen/pest defense, growth, and development by negative or positive altering of gene expression (Wasternack, 2007). JA and SA have an antagonistic relationship where the production of one of them repress the other, a system that is fine-tuned by other hormones such as ET (although, positive regulatory interactions is possible as well (Loake & Grant, 2007)). Further ET has shown to have a synergistic relationship towards JA, leading to an antagonistic relationship between ET and SA (Bürger & Chory, 2019). The antagonistic/synergistic roles between the main hormones in phytopathogenic defense is exploited by a variety of pathogens e.g., a fungi Fusarium oxysporum suppress the SA response by producing an enzyme that upregulate the biosynthesis of JA (Bürger & Chory, 2019). Another example is how pathogens utilize the synergistic role of ET to JA to suppress SA signaling, by producing their own ethylene or making the plant produce an excess of ET. There are hormones traditionally associated with abiotic stress and it has been recently shown interactions in the pathogen signaling networks, one example is abscisic acid (ABA). It has been suggested that ABA has an antagonistic connection to SA signaling since aba mutants showed increased susceptibility towards pathogens (Bürger & Chory, 2019) (Yasuda, et al., 2008).

The bacterial peptide flg22 is perceived by FLS2 and a defense response is activated

One of the most studied MAMP/PRR-pair is the receptor FLS2 and the 22-amino-acid long bacterial flagellin derivate flg22 (Boller & Felix , 2009). Flg22 is a short region located on the conserved part of the flagellin protein N-terminus (Felix, et al., 1999) and when bound to FLS2 it induces inhibition of seedling growth, accumulation of phytogenesis related protein 1 (PR1, a defense protein) and callose deposition and formation (Gómez-Gómez, et al., 1999).

(12)

8 Highly conserved orthologues to the FLS2 receptor are present in all higher plants (Sun, et al., 2013) and so is responsiveness to the flg22 peptide (Boller & Felix , 2009). The wide distribution of flg22- sensitivity and FLS2 orthologues suggest ancient origin (Felix, et al., 1999).

FLS2 is a receptor kinase with leucine rich repeats and oligomerize with BAK1 upon activation

FLS2 belongs to the family receptor kinases (RK) which make up 60% of the transmembrane proteins in Arabidopsis thaliana (annotated as Arabidopsis here and after) and is thereby qualified as one of the largest families in mentioned species. RKs have extracellular N-terminals and intracellular C-terminal kinases. FLS2 has an extracellular part consisting of 28 leucine rich repeats (LRR) that is flanked by pairs of conservatively spaced cysteines (Shiu & Bleecker, 2003) (Boller & Felix , 2009). The LRR region is characteristically horseshoe-shaped consisting of multiple units made of a β-sheet and an α-helix connected by a loop (Figure 2A). The conserved pattern LxxLxLxxN present on the inner side of the curved solenoid, surrounding the β-sheets is suggested to be responsible for ligand interaction (Sun, et al., 2013) (Kobe & Kajava, 2001). Subsequent to ligand binding, FLS2 oligomerize with a leucine-rich repeat receptor-like kinase (LRR-RLK) called BRASSINOSTEROID INSENSITIVE 1-associated kinase (BAK1) (Sun, et al., 2013). The interaction between the two receptors was established as functionally important when a bak1 mutant Arabidopsis showed normal ligand bounding with abnormal, lower but not completely absent, flagellin-triggered response when treated with flg22. Because of this BAK1 is indicated to be a positive regulator (Chinchilla, et al., 2007). BAK1 is part of the SERK-family and is therefore also called SERK3 (Boller & Felix , 2009)(Chinchilla, et al., 2009). SERK3 was renamed to BAK1 when the interaction with the BRASSINOSTEROID INSENSITIVE 1 (BRI1), yet another LRR-RK that show high similarity to FLS2, was discovered. Together, the BAK1/BRI1 pair is part of brassinosteroids signaling and is a well-studied plant receptor model. It has been proposed that BAK1/SERK3 is of high importance in a stimulus-dependent manner in the regulation of several ligand bounding LRR-RKs (Chinchilla, et al., 2009). The structure and interactions of FLS2, BAK1/SERKs and BRI1 are summarized in Figure 2B.

(13)

9

Figure 2. A. Structure of the LRR-repeats, here shown by the porcine ribonuclease inhibitor, and displaying the curved solenoid, or horseshoe, shape (Adapted from Wikipedia). B. Immunogenic molecular patterns such as peptide epitopes flg22 are perceived by the Arabidopsis immune receptor kinases such as FLS2 and its corresponding co-receptors BAK1/SERKs.

Brassinosteroid (BR) regulates Arabidopsis growth and development through the receptor kinase BRI1 and the co-receptors BAK1/SERKs. Receptor-like cytoplasmic kinases (green circles) act downstream of receptor kinases to regulate the defense responses and/or other biological responses. FLS2, FLAGELLIN SENSING2; BAK1/SERKs, BRI1-ASSOCIATED KINASE1/SOMATIC EMBRYOGENESIS RECEPTOR KINASEs; BRI1, BRASSINOSTEROID INSENSITIVE1. Oval shapes indicate the leucine-rich repeats, LRRs. Adapted from (Liang & Zhou, 2018)

Another member of the SERK-family is SERK1 and it has high resemblance towards SERK3/BAK1. SERK1 is expressed in somatic and embryogenic cells during somatic embryogenesis (Hecht, et al., 2001).

Since bak1 mutants still show some sensitivity to flg22 it has been proposed that SERK-family homologues to SERK3 can substitute in its absence. SERK1 can form a complex with BRI1 (Karlova, et al., 2006), suggesting that it could also substitute BAK1 in the FLS2 complex. A network of SERK1 and SERK3/BAK1 and their interactions is described schematically in Figure 3.

(14)

10

Figure 3. SERK3/BAK1 influence a number of processes such as flagellin responses, BR responses, cell death responses, PAMP responses and light responses, the three latter via unknown receptors (indicated with ???). SERK1 influence somatic embryogenesis by unknown receptor/receptors, also shown with ???. SERK1 also influence the BRI1-receptor.

PAMP-responses

When treated with flg22, Arabidopsis show induction of almost 1000 genes and downregulation of around 200 genes in the following 5-30 minutes. The gene coding for the FLS2 receptor is part of the induced gene, indicating that this early response is increasing the receptor capability via positive feedback. Arabidopsis response to flg22 is almost identical to the response of treatment with elf26 (bacterial protein derivative, elongation factor) and also fungal chitin is suggested to produce a similar gene response (Boller & Felix , 2009).

A late response of Arabidopsis when treated with flg22 is seedling growth inhibition indicating that as part of the immunity, the plant redirect focus from growth to defense. After treatment in Arabidopsis the bacterial flagellin peptide induce a microRNA that down-regulates three auxin receptor transcripts resulting in inhibited auxin signaling and thereby also inhibited growth (Navarro, et al., 2006). Auxin is a plant hormone/regulator and helps plants to grow upwards towards the light, and the roots downwards. It also has a function in flowering and fruit production (Alberts , et al., 2014).

(15)

11

FLS2 regulation

Ethylene has a central role in the positive feedback loop of FLS2 regulation

Endogenous levels of ethylene support an integral expression level of FLS2; upon flg22 binding to the FLS2/BAK1 complex activation of the mitogen-activated protein (MAP) kinases MPK3 and MPK6 occur.

MPK6 phosphorylates the transcription factor ethylene insensitive 3 (EIN3) as well as ethylene synthases (ACS2/6). EIN3 is a transcription factor involved in ethylene signaling but has also shown direct interaction with the FLS2 promotor (Boutrot, et al., 2010).

Transcription factor MYC2 may have an indirect role in FLS2 regulation

MYC2 is a basic-helix-loop-helix TF mostly associated with JA signaling, acting as a pathway hub (Kazan

& Manners, 2013). It has been suggested that MYC2 is a positive regulator of JA-responsive genes, including genes coding for TFs involved in e. g. insect defense and wound response. On the other side, MYC2 is indicated to negatively regulate pathogen defense genes that are JA-dependent (Kazan &

Manners, 2008). MYC2 is acting as both an activator and a repressor but how repression is obtained is unclear, it could either repress passively by blocking the positive regulating TFs of the defense genes or actively by direct interaction with the promotors of the genes of the TFs and with a co-repressor.

The promotors of the MYC2-regulated TFs all contain variants of a palindromic motif called the G-box (CACGTG). These conserved sequences are associated with MYC2 binding, implying binding of MYC2 to the promotors. There is also a possibility of MYC2 regulating plant defense genes without direct DNA interaction, by direct transcriptional regulation of their TF regulators (Kazan & Manners, 2013).

MYC2 regulates the TF EIN3-binding F-box protein 1 (EBF1) and EBF1 regulates EIN3 which is a positive regulator for a number of genes (Zhu, et al., 2011), FLS2 included. Thus, MYC2 regulated FLS2 indirectly, by regulating EBF1, which regulates EIN3, in a scheme as MYC2 →EBF1→EIN3→FLS2. Transcriptional repression by MYC2 is not fully understood but a few mechanisms have been discussed, i.e., MYC2 could bind to the G-box and integrate a co-repressor complex or MYC2 could dimerize with another TF that would affect binding to the G-box (Kazan & Manners, 2013).

Methodology

Conserved cis-regulatory modules

It is possible to investigate the transcriptional regulation of plant genes by searching for cis-regulatory elements (CREs) (Peters, et al., 2017). It has been concluded that during domestication of crops most of the selection occurs through the cis-regulatory elements (Swinnen, et al., 2016), which is why the

(16)

12 non-coding regulatory elements are in focus rather than the coding genes. The mutations in the CREs effect both spatial and temporal expression (Swinnen, et al., 2016). CREs are developed to maintain complex expression patterns, by assembling the TF-binding CREs (motifs) in different orders and modules in the upstream regulatory sequence of the plant gene. This is based on the observation that modules of CREs, are more conserved across species than coding regions and they contain high concentration of TF binding sites compared to other regions (Wilczynski, et al., 2009). FLS2 has been shown to be directly regulated by the transcription factor EIN3 via a conserved mot if in the FLS2 promotor (Boutrot, et al., 2010). Due to this fact and the fact that FLS2 is an essential receptor in the plant defense system and is present in all higher plants it was reasoned that there could be additional transcription factors regulating the transcription of the gene and thereby additional conserved regulatory modules in the promotor. To investigate this, the upstream regulatory sequences of FLS2 orthologues in eudicots are searched for conserved motifs using the program MEME. The same approach is utilized for the gene SERK1, where there is little known about the transcriptional regulation. This hypothesis and approach are supported by previous studies (Peters, et al., 2017) (Ezcurra, et al., manuscript to be published), where searching for conserved regulatory motifs in orthologues of a gene of interest revealed strikingly conserved cis-regulatory modules, consisting of clusters of conserved motifs, across all Eudicots species. This approach is in contrast to searching for conserved motifs in the upstream regulatory sequences of co-expressed genes, a frequently used approach.

Agrobacterium and Agrobacterium vectors

Agrobacterium tumefaciens (annotated as Agrobacterium here and after) was isolated over 100 years ago when searching for the pathogen causing crown galls disease, which is characterized by large tumors on tree stems (Figure 4A). It was later revealed that the pathogenic soil bacteria Agrobacterium utilizes an interkingdom gene transfer, by inserting its DNA into the plant’s genome. The bacterium has a megaplasmid, the Ti-plasmid (tumor inducing), which contains vir genes that enable the insertion of transfer DNA (T-DNA) (Hwang, et al., 2015) into the plant chromosomic material in the nucleus, where it is expressed (Adnan, et al., 2013). The T-DNA includes regions that code for enzymes mediating synthesis of plant growth hormones (auxin and cytokinin), which induce tumor formation in the plant, plus a region that codes for enzymes mediating synthesis of opines, which provide nutrients to the bacteria (Figure 4B). Agrobacterium can transfer DNA independently of the tumor inducing genes, making it possible to replace the oncogenic genes by inserting genes of interest to be expressed in transgenic plants (Figure 4C) (Hwang, et al., 2015) (Subramoni, et al., 2014). For this project, no

(17)

13 transgenic plants will be made but Agrobacterium will be used as a tool to transiently introduce sequences of interest into leaves on 4-week-old plants.

Figure 4. A. A crown gall tumor induced by Agrobacterium. B. The Ti-plasmid, a natural plant vector. T-DNA region containing genes for Auxin, Cytokinin and Opine synthesis promote tumor formation and opine secretion. Opines are used as C-and N- source by Agrobacterium. Border sequences (left and right) are required for T-DNA integration. vir-genes are required for transfer and integration of T-DNA. C. Plant transformation vectors, also called binary (able to replicate in both E. coli and Agrobacterium) vectors, are modified Ti plasmids. The vir genes necessary for DNA transfer are kept in the Agrobacterium strain by a modified Ti-plasmid without T-DNA, called helper plasmid (not shown). Adapted from (Glick, et al., 2010).

Transactivation studies in plants with reporter and effector constructs

Transfecting the leaves with Agrobacterium will result in a local insertion of the genes, allowing us to perform transactivation studies (Figure 5). The transactivation studies are based reporter and effector constructs where the reporter construct is the E. coli iudA gene coding for -glucuronidase (GUS, absent in wild-type plants) and the sequence of a promotor sequence placed upstream of the GUS gene, and the effector construct is the gene coding for a given transcription factor regulated by a strong constitutive plant promotor of viral origin, called 35S. When introduced into the plant the transcription factor is highly expressed and if it regulates the promotor sequence, there will be change in the expression level of the GUS reporter gene. Five days after transfection, the leaves are harvested, and the enzymatic GUS activity is measured. If the activity level is altered when the effector construct is co-transfected compared to a negative control lacking the effector, it can be concluded that the effector regulates the promotor by activation or inhibition.

(18)

14

Figure 5. The agroinfiltration method. A. Agroinfiltration is produced by infiltrating the lower side of the leaf, which is porous because of stomata. B. Leaf stomata are openings on the lower leaf surface. C. Depiction of a cross-section of a leaf stoma.

The mixed liquid cultures enter the leaf’s internal air space through the stomata, transfecting adjacent cells. D. The agrobacterium strains dock on a plant cell and deliver the T-DNA to the cell nucleus. E. Reporter and effector constructs (not to scale). GOI, gene of interest; urs, upstream regulatory sequence; TF, transcription factor.

(19)

15

Materials and Methods

Data analysis

Two target genes associated with PTI, FLS2 and SERK1, were chosen and the upstream regions of the orthologues of these chosen genes was retrieved from the Phytozome v13 portal (https://phytozome.jgi.doe.gov/pz/portal.html#), a database where 93 plant genomes are available and offer tools to find orthologues of the gene of interest (Goodstein, et al., 2012). The size and location of the upstream regions was set to 1500 base pairs upstream of the translational start codon ATG, and promotor sequences of 29 orthologue sequences from 27 species of FLS2 were downloaded.

The upstream region size of 26 SERK1 orthologues were obtained in the same manner. In the above- mentioned research by the Ezcurra lab it was shown that regulatory elements are usually found within 1000 bp from the start codon, for this project 1500 bp was chosen to make sure no regulatory motifs were excluded. Conserved motifs between the sequences were searched for with the program MEME, a program used to discover novel motifs from given sequences. MEME use statistical modeling techniques to find recurring patterns of fixed length, motifs between related genes (Bailey, et al., 2009). In MEME the orthologue sequences were used as input, and the settings were altered for each search, starting with setting the length of the motif to 50 bp and the total number of different motifs to 15 (this number was stationary for the process). For each search an assessment of the significance of the output motifs was done in discussions with the supervisor and by literature mining. The setting of the motif length was lowered for each search until a motif of interest was found at 7 bp for FLS2 and 40 bp for SERK1.

To better understand the regulation of FLS2 and SERK1, their gene expression was analyzed, together with the expression of their known or hypothetical regulators, EIN3 and MYC2. Gene expression patterns and levels were obtained by the BAR ePlant tool (http://bar.utoronto.ca/eplant), a visual analytic tool for exploring gene expression in Arabidopsis. The ePlant tool creates visual depictions of gene expression levels by retrieving transcriptome data from multiple publicly available web for any given gene of interest (Waese, et al., 2017). Gene expression patterns and levels of SERK1, FLS2, EIN3 and MYC2 during Arabidopsis development were obtained by the AtGenExpress eFP tool at ePlant, whereas root elicitation maps for FLS2, EIN3, BAK1 and SERK1 were generated with the Root Immunity Elicitation eFP tool at ePlant. The expression levels were assessed in discussion with supervisor to find possible relationships between the transcription factors and target genes. The data for the different maps were obtained with the Affymetrix GeneChip® Arabidopsis ATH1 Genome Array, in which 24 000 of Arabidopsis genes are available (Waese, et al., 2017).

(20)

16

Description of experimental verification

Cloning

To test the potential function of the identified conserved regulatory motif in the FLS2 promotor and the conserved regulatory motif module in the SERK1 promotor (see Tables 9 and 10 in Results), an experimental approach is described in which the TF MYC2´s potential functionality will be studied, because the identified conserved motif/modules contain MYC2 binding sites. Leaves of N.

benthamiana will be transfected by Agrobacterium containing the promotor sequence and the transcription factor genes that are to be tested. If a positive result is obtained from the transactivation assay when inserting the whole promotors, the experiment will be repeated to investigate the role the specific motifs (Tables 9 and 10) in FLS transcription. In the latter experiment the promotor with mutated motifs will be inserted for a loss-of-function study or only the motif will be inserted without the rest of the promotor sequence for a gain-of-function study (Figure 6).

Figure 6. Loss-of-function and gain-of-function reporter constructs of FLS2, and an effector construct of MYC2, to address regulation of FLS2 by MYC2. CM, conserved motif (CAACTTG, Table 9). Other conserved motifs, such as MYC2 binding motifs, may tested in a similar way. Also, SERK1 reporters and their conserved modules (Table 10) may be studied using this scheme.

For the study FLS2 and SERK1 promotor sequences from Arabidopsis, poplar (Populus trichocarpa) and apple (Malus domestica) will be analyzed and compared, because Arabidopsis and poplar are model species whereas apple is a susceptible crop. However, and for simplicity, the description below shows only production of a A. thaliana FLS2 reporter construct, as an example to be used also for SERK1 reporter constructs. All DNA constructs will be produced by overlap extension (OE) PCR and amplified in Escherichia coli before transformation into Agrobacterium. All template sequences for PCR

(21)

17 amplification of promotors or TFs will be obtained by commercial gene synthesis (Eurofins Genomics) based on the coding/upstream sequences in plant genomes database (Phytozome). The promotors will be designed to encompass 1500 bp starting just upstream of the translational start codon, ATG. The promotor sequence will be inserted upstream of the reporter uidA gene coding for β-glucuronidase (GUS) replacing the 35S promotor in the binary vector pCF201 which was derived from the pGA580 vector used for Agrobacterium transformation (An, 1987) (Figure 7).

Figure 7. A. Map of the pGA580 vector. Restriction endonuclease sites are indicated on the outer circle. The size of 12.1 kbp is indicated by the coordinates shown on the inner circle. B, BamHI; Bg, BglII; BL, the T-DNA left border; BR, the T-DNA right border; C, ClaI; H, HindlII; Hp, HpaI; K, KpnI; npt, neomycin phosphotransferase gene; oriT, RK2 origin of conjugal transfer;

oriV, RK2 origin of replication; P, PstI; P (inside the circle), the lac promotor; R, EcoRI; S, SstII; Sa, SalI; Sc, ScaI; St, SstI (=SacI);

Su, StuI; tet, tetracycline resistance gene of RK2 plasmid; trfA*, a segment code for a replication protein; X, XbaI.. B. Map of the pCF201 vector used as template for cloning by overlap-extension PCR. Adapted from (An, 1987).

The promotor sequence will be amplified by PCR using primers as in Table 1, with template DNA obtained by commercial gene synthesis of the relevant promotor regions, according to the PCR program and reaction in Tables 2 and 3, respectively. For the purpose of OE-PCR cloning, the primers are designed with overlapping segments that match the vector fragments (See Table 1). Since the vector is large it will be split and two vector fragments, A and B, will be produced by PCR with the primers in Table 1, by the PCR reaction and program in Tables 4 and 5. The primers in this report were designed for this project and their functionality is therefore not yet verified, but the general experience is that all standard primers produce the expected PCR product when plasmid DNA is used as template.

The TF constructs are produced in a similar manner as the promotor sequence, although here the TF

(22)

18 sequence will be amplified from a gene synthesis plasmid and inserted downstream of the 35S promotor, replacing the uidA GUS gene.

Table 1. Primers used to create the vector fragments A and B, and the Arabidopsis FLS2 promotor constructs. The segments in the promotor construct primers overlapping the vector fragments are highlighted in grey.

Sequence (from 5´ to 3´)

Frag A – forward CGATATAGAGAACCCAAAGGAAAGGCG Frag A – reverse AAGCTTGGCGTAATCATGGTCATAGCTG Frag B – forward CATAAGGGACTGACCTACCCGGGGATCC Frag B – reverse TGTCGACTACGCCATCATGGCGACAG

Prom – forward CAGCTATGACCATGATTACGCCAAGCTTTAATGTGACAGTTTGGCCG Prom – reverse CATAAGGGACTGACCTACCCGGGGATCCTTTCGACTAATCATTGCCC

Table 2. PCR reaction for production of the promotor fragments.

Reagents µl

dH2O 17.7

5X Phusion buffer 5

dNTP (10 mM) 0.5

Primer-F (100 µM) 0.25

Primer-R (100 µM) 0.25

Gene synthesis plasmid (10 ng/µL) 1

Phusion taq 0.3

Total 25

(23)

19

Table 3. PCR program for amplification of the promotor sequence.

Temperature [°C] Time [min:s] Cycles

Initial denaturation 98 0:30 1x

Denaturation 98 0:10

30x

Annealing 65 0:30

Extension 72 0:45

Final extension 72 10:00 1x

Table 4. PCR reaction for production of the vector backbone fragments.

Reagents Fragment A, µl Fragment B, µl

dH2O 17.7 17.7

5X Phusion buffer 5 5

dNTP (10 mM) 0.5 0.5

Primer-F (100 µM) 0.25 0.25

Primer-R (100 µM) 0.25 0.25

pCF201 (10 ng/µL) 1 1

Phusion taq 0.3 0.3

Total 25 25

Table 5. PCR program for production of the vector backbone fragments.

Temperature [°C] Time [min:s] Cycles

Initial denaturation 98 0:30 1x

Denaturation 98 0:10

30x

Annealing 69 0:30

Extension 72 6:00

Final extension 72 10:00 1x

(24)

20 All PCR products will be verified by electrophoresis agarose gels with the following expected results:

➢ Promotor: Expected band at 1500 bp

➢ Transcription factor: Expected band at 3000 bp

➢ Vector fragments: Expected bands at 6 000 bp for each fragment

If results are not as expected, PCR conditions, specifically the annealing temperature and/or Mg²⁺

concentration, will be optimized by trial and error until bands as expected are obtained.

For promotor cloning, the promotor sequence will replace the 35S promotor upstream from the GUS gene, and for TF cloning the TF genes will replace the GUS gene, downstream from the strong 35S promotor, Figure 8. One part of the tetracycline (tet) resistance gene is in fragment A and the other in fragment B, resulting in any cells with a faulty assembly of the vector will be unviable in selection media. When the vector fragments, promotor fragments and TF gene fragments have been produced they will be assembled by OE-PCR (Bryksin & Matsumura, 2010) as indicated in Figure 8 and using Phusion polymerase. The reaction will be as Table 6 and the PCR program as in Table 7.

Table 6. Overlap extension PCR reaction for assembly of vector fragments with the promotor sequence.

Reagent (µL)

dH2O 16

5X Phusion buffer 5

dNTP (10 mM) 0.5

A fragment 100 ng/µl 1

B fragment 100 ng/µl 1

Promotor fragment 100 ng/µl 1

Phusion taq 0.3

Total 25

(25)

21

Table 7. Overlap extension PCR program for assembly of vector fragments with the promotor sequence.

Temperature [°C] Time [min:s] Cycles

98 1:00 1x

98 0:15

60 3:00 3x

72 5:00

98 0:15

14x

60 00:30

72 5:00

72 10:00 1x

a.

(26)

22

Figure 8. Schematic representation of the components used to build the cloning constructs A. The vector pCF201 with fragment A and B represented by green and blue/yellow respectively. The 35S promotor is shown in red and the GUS gene is shown in yellow. The transcription factor gene is shown in magenta with the overlapping parts in red (matching the 35S promotor) and blue (matching fragment B). The FLS2 promotor is represented with purple with the overlapping segments in green (matching fragment A) and yellow (matching the GUS gene).

B. Final vector constructs where the left one shows the transcription factor inserted after the 35S promotor and the right one show the FLS2 promotor before the reporter gene GUS.

After the OE PCR reaction, 1 µl of the resulting reaction will be used to transform competent E. coli (Invitrogen™ One Shot™ TOP10 Chemically Competent E. coli, Catalog number: C404010) according to the manufacturer’s instructions. Briefly, the reaction will be transferred into a vial containing 50 µL competent cells, mixed by tapping the vial and incubated on ice for 30 min. After the initial incubation, the cells will be transformed by heat shock by incubation on 42° C for 30 seconds and then placed back on ice. A volume of 250 µL of S.O.C media will be then added to the vial before placing in a 37° C shake incubator for 1 hour, at 225 rpm. As a final step the cells will be spread on tetracycline selection agar plates (10 mg/l) and incubated at 37° C overnight. Obtained colonies will be screened by colony PCR using primers outside of the insertion site. Bacterial clones harboring the expected PCR fragments will be cultured overnight in LB/tetracycline (10 mg/l). The amplified plasmids will be extracted and purified by the QIAGEN QIAprep Spin Miniprep Kit and further analyzed by commercial sequencing (Eurofins Genomics).

Competent Agrobacterium cells will be prepared in the lab since they are not sold commercially. The Agrobacterium celles (strain C58C1-RS with helper plasmid pCH32) will be incubated in 5 mL LB/rif medium (rifampicin 25 mg/L) overnight at 28° C, and then added to 100 mL of LB medium. The inoculum will be incubated at 28° C until it has reached an OD600 between 0.5 and 1.0. The culture will

b.

(27)

23 be placed on ice before allocating it to 50 mL Falcon tubes for centrifugation in at 4000 g for 20 min at 4° C. The supernatant will be discarded, and the pellet will be resuspended in 2 mL cold CaCl2 (20 mM).

The mix will then be distributed to Eppendorf tubes with a volume of 100 µL and will then be placed in storage in -80° C after freezing by liquid nitrogen.

For the transformation of Agrobacterium 0,1-1 µg of the plasmid constructs will be placed on the frozen competent cells on ice, then cells are thawed on ice for 15 min. The mix will be frozen again by liquid nitrogen for 5 min before thawing in a water bath with 37° C for 5 min. 150 µL of LB medium will be added before a 2-hour incubation in 30° C. As a last step of the transformation the cells will be spread on prewarmed (28° C) LB plates, with 20 mg/l kanamycin and 5 mg/l tetracycline, before incubation at 28° C. If the transformation is successful colonies will appear after 2 days. The Agrobacterium strains harboring the different constructs can be used directly or stored at -80°C.

Agroinfiltration and transactivation study

Prior to transactivation assays, N. benthamiana plants will be grown from seeds in commercial soil with vermiculite during 4-5 weeks at 22°C, with a light regime of 16 hours of light and 8 hours of darkness. To perform the transactivation assay, leaves of 4-5 week-old N. benthamiana will be infiltrated by Agrobacterium with the selected combinations of effector and reporter constructs, Table 8. The infiltration will be performed by starting a 15 mL overnight culture (28°C) from the Agrobacterium colonies mentioned above. The cells will be spun down at 4000 rpm for 15 min before resuspension in Agroinfiltration buffer (10 mM MES, pH 5.6, 10 mM MgC12) and Acetosyringone (150 mM), the mix will be placed in incubation for 5 h to overnight at room temperature. After (or during) incubation the OD600 will be adjusted to OD600 0.4 by agroinfiltration buffer and Acetosyringone as previously. After the incubation and dilution, the leaves will be infiltrated by mixing 2 mL of each construct, each mix will then be applied by syringe to three leaves/mix.

Table 8. The combinations of which the reporter and effector constructs will be used for agroinfiltration.

Sample FLS2-GUS 35S-MYC2

1 + +

2 + -

(28)

24 After four days the infiltrated leaves will be analyzed for GUS activity. The total protein extract from 4- day-old leaves will be analyzed by a colorimetric GUS assay for plants (Leborgne-Castel, et al., 1999) using the substrate 4-Nitrophenyl-β-D- glucopyranoside (PNPG, Sigma N1627). To harvest the protein, leaf discs will be cut from the infiltrated leaves which then will homogenized in GUS Extraction buffer and spun down at 10000 rpm for 10 min; the supernatant will be then utilized for the assay. During the assay, the cleared extract is mixed with GUS Assay buffer with and without 5 µM PNPG. Cleavage of PNPG by GUS produces a yellow color, yielded by the product p-Nitrophenol. The absorbance level of PNP will be measured at 405 nm and the absorbance value obtained in the absence of PNPG will be subtracted. The specific activity will be obtained by relating to the total protein content, which is established by BioRad Protein Assay, and by the time length (hours) of the reaction.

Expected results: The expectation is that the effector MYC2 influences the transient transcription of the GUS gene via the FLS2 promotor sequence. Enzyme activity can be both higher or lower with the reporter construct introduced since MYC2 could either inhibit or activate the promotor. If an effect of MYC2 on FLS2-reporter gene expression is obtained, the involved motif will be addressed by constructing loss- and gain-of-function FLS2 reporters as described in Figure 6. Both the mutation and the tetramer construct of selected motifs will be done by gene synthesis and subsequent OE-PCR as described above for the wild-type FLS2 promotor. A similar study will be done on SERK1 and the SERK1 conserved motif module, using the above-described scheme.

(29)

25

Results

Motif search

To find conserved regulatory motifs or motif modules, I searched the upstream regulatory sequences of 27 FLS2 orthologues in eudicots with the program MEME. A conserved module was not found, although a highly conserved motif, CAACTTG (Figure 9) was found at least once in all 27 species, Table 9. Interestingly, an identical motif is also conserved in the genes coding for the repressor TEMPRANILLO 1 (TEM1) in above-mentioned research by the Ezcurra group. In the TEM1 promotor the motif was found in a highly conserved motif module named TMM. In the TMM module the CAACTTG motif was next to another motif, TCATGT, a variant of the G-box that was shown to mediate regulation through the transcription factor MYC2. This observed association of the CAACTTG motif with MYC2 binding elements led to the speculation that MYC2 could bedirectly regulating the FLS2 gene, together with the EIN3 transcription factor. Variants of the MYC2 binding G-box sequence were searched for in the FLS2 promotors, and were found to be present in most species. A literature search revealed that the CAACTTG motif is highly enriched in the promotors of ABA-repressed genes when transcriptomes of ABA-regulated gene sets was analyzed (Wang, et al., 2011) (Choudhury & Lahiri, 2010). The motif is included in a binding sequence of the transcription factor AUXIN RESPONSE FACTOR 1 (ARF1) (Ulmasov, et al., 1997). As a comparison, the CAACTTG motif is mostly absent in upstream regions of BAK1 or SERK1 orthologues, strongly suggesting its role in FLS2 regulation.

Figure 9. The conserved motif identified in the FLS2 promotor sequence.

(30)

26

Table 9. A list of the species where the motif (magenta) was found in the FLS2 promotor, the position of the motif in relation to the ATG start codon and the surrounding sequence of the motif. The species highlighted in grey are members in the brassica- family. A. halleri: Araha.10139s0013; A. lyrata: AL8G12090; A. thaliana: AT5G46330; C. grandiflora: Cagra.0569s0008; C.

rubella: Carubv10025764m.g; B. stricta: Bostr.8819s0170; B. rapa FPsc: BraraI01928; B. rapa FPsc: Brara.B02886; B. oleracea capitata: Bol032113; E. salsugineum: Thhalv10000746m.g; A. comosus: Aco014813; M. esculenta: Manes.17G067300; P.

persica: Prupe.4G076500; E. grandis: Eucgr.D00318; S. purpurea: SapurV1A.1394s0040; M. truncatula: Medtr4g094610; G.

max: Glyma.08G083300; G. max: Glyma.05G128200; M. guttatus: Migut.H01377; T. cacao: Thecc1EG029529; F. vesca:

gene30550-v1.0-hybrid; P. trichocarpa: Potri.004G065400; G. raimondii: Gorai.010G165300; L. usitatissimum:

Lus10023323.g; C. clementina: Ciclev10024610m.g; M. domestica: MDP0000254122; C. sativus: Cucsa.047560; A. coerulea:

Aqcoe6G285600

Species Nº of bp to

ATG Sequence

A. halleri 577 CTTTTATATGCAACTTGACCAAATTAT

A. lyrata 192 CCTTTATATGCAACTTGACCAAATTAT

A. thaliana 872 GTACAGAACACAACTTGTTGATAAAAC

C. grandiflora 150 ACATTATATGCAACTTGACCGAATTAT

C. rubella 1149 AGCAAAAACCCAACTTGTTTATAACAA

B. stricta 989 ATACAAAATACAACTTGTTAATAAAAA

B. rapa FPsc 1444 ctgtccaccgCAACTTGatcgacgccg

B. rapa FPsc 99 AGTTTAAATGCAACTTGATCAAATTAT

B. oleracea 225 GCTTTATATGCAACTTGGCCGAATTAT

E. salsugineum 889 ATACGAAACACAACTTGTATATAAAAT

A. comosus 99 TCTTGGTAGCCAACTTGCAGAAAAAGG

M. esculenta 613 ATAAAGCAGCCAACTTGTAGTTTAAAT

P. persica 466 CCATTATGCACAACTTGCATATAAAAA

E. grandis 651 CCTGTCCACCCAACTTGCCTATTCGAA

S. purpurea 1311 CCCTTTTCTGCAACTTGTTTTTTTCCG

M. truncatula 1254 AACAAGTGGCCAACTTGTAATAAATAA

G. max 1061 CTTCATCACACAACTTGTTGCAAACCG

G. max 534 CTTTATCACACAACTTGCTGCAAACCG

M. guttatus 1318 CATTTGCTGCCAACTTGCATTATTACA

T. cacao 1489 TCATGCTCTACAACTTGGGTTAATAAC

F. vesca 462 CCTCTATATGCAACTTGTGCTTTAATG

P. trichocarpa 1424 CGCTTTCTGTCAACTTGGGCCACAGCC

G. raimondii 737 AGTAGTCTGACAACTTGATTCTGGTTT

L. usitatissimum 551 TGGCGGCCGCCAACTTGATTTCATTTT

C. clementina 108 CACTCAATTTCAACTTGCAAGCCATCA

M. domestica 1048 TTGGAGCTTTCAGCTTGCTACACCATA

C. sativus 238 CAATTTCAAACAGCTTGGACTTTTCTA

A. coerulea 894 GTACCAATGTCACCTTGGCACTAGTTG

(31)

27 Using the MEME motif search, analyzes of the upstream regulatory sequences of all SERK1 orthologs found a much longer conserved motif module, Table 10. All the species with a conserved version of the module are non-brassica eudicots meaning that it evolved when after the brassicas split during evolution of eudicots. Two of the motifs in the module, GACGTG and CACGTT, are G-box-like, potential MYC2 binding sites (Dombrecht, et al., 2007).

The apparent connection to MYC2 led to further investigation of the different mechanisms and components related to MYC2 regulation which is why the experimental verification was designed with MYC2 as effector, and also why expression levels are analyzed below.

Table 10. A conserved motif module in SERK1 upstream regulatory sequences of non-brassica eudicots. Gene locus and module’s distance to ATG translational start (in parenthesis): C. papaya: evm .TU.supercontig_66.123 (966 bp); T. cacao:

Thecc1EG036627 (569 bp); G. raimondii: Gorai.010G129300 (575 bp); S. purpurea: SapurV1A.1095s0070 (541 bp); S.

purpurea: SapurV1A.0596s0070 (611 bp); P. trichocarpa: Potri.013G117200 (595 bp), Potri.019G087700 (542 bp); M.

esculenta: Manes.02G187500 (636 bp); M. esculenta: Manes.18G097400(587 bp); C. sativus: Cucsa.365750 (610 bp); M.

truncatula: Medtr1g097160 (613 bp); G. max: Glyma.02G076100 (498 bp); G. max: Glyma.10G218800 (573 bp); Gmax:

Glyma.20G173000 (588 bp); V. vinifera: GSVIVG01001600001 (459 bp); F. vesca: gene27511-v1.0-hybrid (510 bp); P. persica:

Prupe.8G048700 (581 bp); A. coerulea: Aqcoe6G053300 (630 bp)

Species Sequence

C. papaya GGTCACGTCATTTTCCACGATATAGCGCGAAGGCGTTTTGACTGCGTTCG T. cacao AGTCATGTCACTTTCCACGATATAGCGCGAACCCGTTTTGACTGCGTTCG G. raimondii ATTTATGTCACTTTCCATTATATAGCGCGAACCCGTTTTGACTGCGTTCG S. purpurea GGTCACGTCACATTCCACGATAGAGCGCGAATGCGGTTTTTGACTGCGTT S. purpurea GGTCACGTCACTTTCCACGATATAGCGCGAATGCTGTTTTTGACTGCGTT P. trichocarpa GGTCACGTCACTTTCCACGATATAGCGCGAATGCTGTTTTTGACTGCGTT P. trichocarpa GGTCACGTCACTTTCCACGTTATAGCGCGAATGAGGTTTTTGACTGCGCT M. esculenta CCTCACGTCACTTTCCATGTTATAGCGCCACTGCGGTTTTTTTGACTGCG M. esculenta GGTCACGTCACTATCCACGGTATAGAGCGAATGCGTTTCTTTTGACTGCG C. sativus TGTCACGTCACTTTCCACGATGTAGCGCGAAGGCGTTTTTGACTGCGTTT M. truncatula AGTCACGTCACTTTCCACGTTATAGCGCCAATACATTTTTGACTGCGTTT G. max CCTCACGTCGCTTTCCGCGATATAGCACGAAGGCGTTTTGACTGGATTCG G. max AGTCACGTCACTTTCCACGATCTCGCGCGAGAGCGTTTTTGACTGCGTTC G. max AGTCACGTCACTTTCCACGATCTCGCGCGAGAGCGTTTTTGACTGCGTTC V. vinifera GGTCACGTCACTTTCCATGATATAGCGGGAAGGAGTTTTGACTGTGTTGG F. vesca GGTCACGTCCCTTTCCACGTTGTAGCGCGAAGGCGTTTTGACCGCCTCCG P. persica AGTCACGTCCCTTTCCACGTTGTAGCGCCAAGGCGTTTTGACTGCGTCCG A. coerulea GGTTAGGTACCTTTCCACACGCGAACATAGCAAAGGAAAGAATAGAGAG-- AATAGAGAGAATAGAGAGAGAGTGAGTTTAATGAGAGTTTTGACTAATA

(32)

28

Expression analysis

Gene expression maps were used to get a possible insight of how the expression levels of the discussed genes compare to each other, because strict co-regulation between a TF and its target gene indicates transcriptional activation, whereas strict opposing expression indicates repression. However, target genes may be regulated by different TFs during different responses, so expression between TFs and their target genes is frequently only partially overlapping or opposing. In this analysis, we examined whether FLS2 or SERK1 and MYC2 show overlapping or opposing expression. We also analyzed whether the TF EIN3 and FLS2 display overlapping expression, because EIN3 is a known regulator of FLS2.

Expression levels were examined both during development (Figure 10) and in the roots treated with flg22 (Figure 11, although this expression study did not include the MYC2 gene).

Figure 10. Expression levels of FLS2, SERK1, EIN3 and MYC2 in different tissues of Arabidopsis Thaliana during different development stages.

As a general trend, no obvious patterns of completely overlapping or opposing expression were observed between TFs MYC2 or EIN3, and FLS2 or SERK1, not even in the case of FLS2 and EIN3, as would have been expected since EIN3 is an activator of FLS2. On the contrary, the two genes display opposing expression patterns in some tissues. As an example, EIN3 is highest in dry seed, pollen, the

(33)

29 immature embryo and senescing leaves, whereas FLS3 is not, or hardly, expressed in these tissues. In spite, both genes are expressed at similar levels in other tissues, such as leaves, rosettes, stem, and some flower structures. In conclusion, EIN3 activates FLS2 but their expression is only partially overlapping and even opposing in certain tissues. In the case of MYC2 and FLS2, the genes show opposing expression in some tissues, such as the vegetative shoot apex, seedling, vegetative rosette, senescent leaves and 12 days old flowers, but overlapping expression in other tissues. In contrast to this lack of lack of clear-cut positive or negative expression correlations between MYC2 or EIN3, and FLS2 or SERK1, MYC2 and EIN3 display opposing expression patterns to a significant extent, suggesting that MYC2 and EIN3 may have antagonistic roles.

The root immunity elicitation maps show an increased expression in FLS2 when treated with flg22, this can also be observed in BAK1 but not in EIN3 and SERK1. SERK1 shows a lowered level of expression compared to the mock treatment. Next to flg22 the roots were treated with Pep1, a 23-amino acid long peptide produced by the plant itself to enhance disease resistance (Yamaguchi, et al., 2010). This sort of compound is called damage-associated molecular pattern (DAMP) and activate a similar stress response as PAMP when perceived by its corresponding receptor (Boller & Felix , 2009). Regretfully, the expression of MYC2 was not analyzed in this study.

(34)

30

Figure 11. Expression levels of FLS2, EIN3, BAK1 and SERK1 in the roots of Arabidopsis when treated with the elicitors flg22 and Pep1.

Discussion

The search in MEME for conserved sequences in the promotors led to finding the conserved motif, CAACTTG, in FLS2. The same motif is enriched in the upstream of many ABA-related genes and in a conserved module, TMM associated with MYC2 binding. The fact that the motif is well conserved in all the searched FLS2-orthologues but mostly absent in BAK1 and SERK1 orthologues indicates that it is of some importance in the transcriptional regulation of FLS2. Functional studies are needed to confirm and establish what the function is, and an experimental approach is described in Methods to carry out such studies, by using transient transactivation by agroinfiltration in N. benthamiana, combined with a MYC2 effector and FLS2 reporter.

The positive feedback loop of FLS2 regulation by ethylene and EIN3 should require a negative regulator as well to prevent excessive accumulation of the components. MYC2 has been reported to be a

(35)

31 transcriptional repressor of other genes, suggesting the possibility that MYC2 could be repressing the FLS2 gene. If MYC2 would have a negative effect on the FLS2 transcription it would mean that it is antagonistic to EIN3, and this is supported by MYC2 having antagonistic relationships with other transcription factors. The fact that MYC2 indirectly inhibits the transcription of EIN3 (Kazan & Manners, 2013), supports the proposition of an antagonistic relationship further. The potential antagonistic relationship between MYC2 and EIN3 is in line with the expression maps, where MYC2 shows increased expression where EIN3 do not and vice versa. On the other hand, the expression maps of MYC2 and FLS2 show both co-expression and some levels of contrasting expression in the two genes, so MYC2 could still be a positive regulator of FLS2 transcription. Of course, these speculations need further research by practical experiments since the different expression levels could be caused by other reasons. The root immunity elicitation maps indicate that EIN3 is endogenously produced and kept at a constitutive level whilst FLS2 expression is induced by ligand-bounding, and FLS2´s co-receptor BAK1 also seems to be expressed by peptide epitope induction. SERK1 shows the opposite but the explanation for this could be that it is not normally part of the flg22 response.

(36)

32

Conclusion

To conclude this study, I will begin to state that it is needed, as the gap of knowledge in this field restricts the development of new discoveries and solutions. Even though this study is at a novel, basic level it could be used in further research, most likely in the area of increased resistance towards bacteria but it is possible to apply in other areas as well. The complexity and overlapping of the regulatory networks make these discussions of the relationships between FLS2, EIN3, MYC2, BAK1, SERK1, ET, JA, SA etc. relevant in other topics than what this study treats. The findings of the experimental approach could be applied when engineering plants to a stronger bacterial resistance for example using CRISPR/Cas to insert several copies in planta of the (potential) MYC2-binding motifs in the promotor of FLS2 to enhance transcription and thereby induce a greater defense.

(37)

33

Future perspectives

The path to food security for the upcoming generations consist of many tiers and approaches, one of which is mapping the plant-pathogen interactions, and another is to discover and disentangle the mechanisms surrounding the plant immune system. This project aimed to do this at a small scale, and to continue this project the first step is to carry out the experimental approach described under Methods and Materials. The results from that experiment would provide answers to the questions the in-silico part of the project set up. If the experiment would yield a negative result, MYC2 is not accountable for regulation in SERK1 or FLS2, other transcription factors can be analyzed. There are also opportunities to continue this project by using the same bioinformatic approach on other plant defense genes such as the ELONGATION FACTOR RECEPTOR 1 (EFR1), an LRR receptor kinase that binds the bacterial peptide epitope elf18, eliciting a defense response through a mechanism as in FLS2-flg22 (Figure 2) (Liang & Zhou, 2018). Besides this project, the need for further knowledge of the networks in the plant defense against bacteria is an urgent matter. Although there are today several applications of biotechnological solutions to increase yield and lower production costs for plant agriculture, there is still an urgent demand for new and improved methods to increase resistance against bacterial pathogens.

Regulation of Plant Defense Genes Against Bacterial Pathogens

Regulation of Plant Defense Genes Against Bacterial Pathogens

JENNY SJÖSTRÖM

Abstract

Sammanfattning

Table of contents

Preface

Abbreviations

Introduction

Background and theory

The plant defense is a two-tier system

Antibiotics are not a sustainable control of infections

Phytohormones regulate plant defense in complex networks

The bacterial peptide flg22 is perceived by FLS2 and a defense response is activated

FLS2 is a receptor kinase with leucine rich repeats and oligomerize with BAK1 upon activation

PAMP-responses

FLS2 regulation

Methodology

Materials and Methods

Data analysis

Description of experimental verification

a.

b.

Results

Motif search

Expression analysis

Discussion

Conclusion

Future perspectives