Essay: Characterizing Plant Regeneration

ABSTRACT
The Polycomb repressive complex (PRC) is a conserved mechanism for generating trimethylation of histone 3 at lysine 27 (H3K27me3), a mark that is associated with repressed chromatin. The repression of key regulatory genes by Polycomb group mediated H3K27me3 is critical for maintaining cell states and cellular reprogramming in both plants and animals. The aim of this study is to characterize gene reactivation and epigenetic reprogramming in the model plant Arabidopsis, in order to elucidate the changes in the chromatin landscape between callus (transdifferentiating adult stem cells/pluripotent cells) and leaf (differentiated cells). As chromatin reprogramming involves the removal of H3K27me3 mark, which is ‘erased’ by histone demethylases, the effect of H3K27me3 specific histone demethylase mutants on the regenerative capacity of Arabidopsis was tested. There are 3 homologues in Arabidopsis that have H3K27me3 specific demethylase activity- Relative of Early Flowering (REF6), Early Flowering (ELF6) and JMJ13. A classical genetic knockout approach was used to test the effect of H3K27me3 specific demethylase knockout mutants on callus formation. Following genotyping and expression analysis of ref6-1, elf6-3 and jmj13 mutants delivered from the Nottingham Arabidopsis Stock Centre, the mutation was confirmed in ref6-1 (line2-homozygous for T-DNA insert) and elf6-3 (line1-heterozygous for the T-DNA insert, line2- homozygous for T-DNA insert). Phenotypic analysis on ref6-1 and elf6-3 indicated that mutations in H3K27me3 specific demethylases REF6 and ELF6 resulted in a loss of their ability to form callus. To conclude, H3K27me3 demethylases are essential for removing PRC2 mediated chromatin silencing and hence, H3K27me3 plays a fundamental role in regenerating callus (transdifferentiating adult stem cells/pluripotent cells) from leaves (differentiated cells).
3
INDEX
Table of Contents
ACKNOWLEDGEMENTS 2
ABSTRACT 3
1.ABBREVIATIONS 5
2.INTRODUCTION 6
2.1 BACKGROUND 6
2.2 AIM 9
3. METHODS 11
3.1 PLANT MATERIALS 11 3.2 PLANT GROWTH MEDIA 11 3.3 GROWING ARABIDOPSIS SEEDS ON COMPOST 12 3.4 DNA EXTRACTION FROM ARABIDOPSIS THALIANA 12 3.5 GENOTYPING OF T-DNA INSERTION LINES 12 3.6 RNA EXTRACTION 14 3.7 REVERSE TRANSCRIPTION AND QPCR 14 3.8 ACCESSION NUMBERS 16
4 RESULTS 17
4.1 GENOTYPIC ANALYSIS FOR N654971, N666435 AND N604032 DELIVERED AS REF6-1 (SALK_001018C), ELF6-3 (SALK_074694C) AND JMJ13 (SALK_104032): 17 4.2 EXPRESSION ANALYSIS FOR REF6 AND ELF6: 20 4.3PHENOTYPESOFREF6-1,ELF6-3ANDJMJ13/N604032MUTANTSCOMPAREDWITHCOL.0WILD-TYPE. 23 4.4 PHENOTYPES OF COL.0 WILD TYPE, REF6-1 AND ELF6-3 MUTANTS DURING CALLUS FORMATION. 24
5.DISCUSSION 26
6. FUTURE DIRECTIONS 29 7. REFERENCES 31
””””4
1. ABBREVIATIONS
ELF6- Early Flowering
HATs- Histone acetyltransferase
HDAC- Histone deacetylases
HDM- Histone demethylases
HMT-Histone methyltransferases
iPSC- Induced pluripotent stem cells
NASC- Nottingham Arabidopsis Stock Centre ncRNA- Non-coding RNAs
PcG- Polycomb
PCR1- Polycomb repressive complex 1 PCR2- Polycomb repressive complex 2
PRE- Polycomb/Triothorax response elements QPCR- Quantitative PCR
REF6- Relative of Early Flowering
TrxG- Triothorax
5
2. INTRODUCTION
2.1 BACKGROUND
Conrad Waddington illustrated the concept of cellular differentiation in the form of an ‘epigenetic landscape’ model, where a cell moving from an undifferentiated/ progenitor state to a ‘terminally’ differentiated state was depicted as a marble rolling down the landscape, into deeper inescapable valleys, irreversibly locking cells into a ‘committed’ state [1,2]. Seminal work by Gurdon (1962), Takashaki and Yamanaka (2006) proved that this dogma could be reversed, as differentiated somatic cells can be reprogramed to a pluripotent state, or in case of plants, a totipotent state and other differentiated states by complete loss of developmentally acquired epigenetic information[1]. The stability of the differentiated cellular state is achieved by a combination of transcription factor networks and epigenetic modifications such as DNA methylation, modification of histone tales, chromatin remodelers and non-coding RNAs [1,3]. The generation of induced pluripotent stem (iPS) cells by direct reprogramming of somatic cells on ectopic expression of defined transcription factors, resurrects interest in the epigenetic stability of the differentiated cell state [4] [5]. It is noteworthy that current approaches involving the creation of iPSCs are inefficient owing to lack of understanding of the complex epigenetic remodeling preceding iPSC formation [6].
As stated by Birnbaum and Alvardo, the diverse array of regeneration processes in one kingdom have counterparts with the other, comparative studies on plant and animal regeneration have revealed that the intermediate structure i.e. plant callus and animal blastema are derived from a specialized population of adult stem cells, which are capable of transdifferentiating or irreversibly fate switching between distinct cell types [7]. Evidence from cell lineage analysis studies focusing on animal regeneration revealed that certain regenerating cells may maintain memory of their earlier cell identity and thus, can only give rise to tissues within their own lineage. This supports the concept that regeneration may not necessarily be achieved from the earliest pluripotent state and questions the superiority of the pluripotent state in Waddington’s ‘epigenetic landscape’ model[5]. Analogous studies on plant regeneration have revealed that callus formation occurs via a root development pathway irrespective of its generation from aerial organs [4]. This is because similar patterns of gene expression and multicellular organization with the lateral root
6
development program were observed during callus formation [4,8]. This implies that that the induction of shoots from callus through application of the hormone cytokinin, is also a transdifferentiation process, occurring from a pre-existing population of adult stem cells [9]. The ability of a differentiated plant cell to initiate a program required for the generation of an entire new plant involves the stable repression of genes inducing differentiation, together with transcriptional activation promoting root pluripotent cells [4]. In order to characterize regeneration in plants, it is crucial to acknowledge the epigenetic mechanisms contributing to the regulatory networks that respond to environmental signals and regulate plant plasticity [10].
Figure 1: Regulation of structure and function of chromatin by histone modifications. Genomic DNA tightly packed in the chromatin. Nucleosome, the fundamental
unit of chromatin is composed of ~146bp of DNA wrapped on a histone octomer [10]. Histone modifying enzymes such as (‘writers’)- histone acetyltransferases (HATs) & histone lysine methyltransferases (HMT) transduce histone modifications and also, remove antagonizing activities by ‘erasers’ HDAC- histone deacetylases, HDM- histone lysine demethylase to modulate between euchromatin-catalyzed by Triothorax proteins (TrxG) & heterochromatin- catalyzed by Polycomb proteins (PcG) [11]. (Created using Microsoft PowerPoint)
‘7
Epigenetic memory is fundamental in maintaining cell identity by regulating gene expression through cell generations without changing genetic information [12,13]. Epigenetic modifications such as histone lysine methylation, methylated DNA regions, ncRNAs and nucleosome structures are chromatin signatures that can be stable through numerous cell division and thus, contribute to cellular memory [11]. Histone methylation is a hallmark of epigenetic inheritance which is fundamental in regulating developmental processes and maintaining epigenomic stability [11]. It is a post-translational modification that occurs at many residues at distinct sites mono, di or tri- methylating them [10]. In order to establish (‘write’) and remove (‘erase’) histone modifications, chromatin-modifying enzymes residing in large multi-subunit complexes catalyze incorporation and removal of covalent modifications with remarkable specificity (Figure 1) [10,11]. Lysine methylation in Arabidopsis occurs at K4, K9, K27 and K36 of histone H3. The ‘writers’ to these modifications are histone lysine methyltransferases (HMTs) and the ‘erasers’ are histone lysine demethylases (HDM) (Figure 1) [10,11].
The central role of Polycomb (PcG) and Triothorax (TrxG) group of proteins in stabilizing cell identities is discernible in both plants and animals, as the two kingdoms use remarkably similar ways for maintaining differentiated states, despite evolving multicellularity independently [12,14,15]. PcG proteins not only provide the transcriptional memory unique to each cell identity [15], but also silence gene expression programs that can be detrimental to the present cell state [16].
In order to establish a repressed chromatin which is associated with differentiated cell states, the PcG containing complexes i.e. Polycomb repressive complex 1 (PRC1) and Polycomb repressive complex 2 (PRC2) catalyze a variety of biochemical reactions including covalent modifications of histones, such as Lys27 methylation at histone 3 (PRC2), and Lys 119 (Lys 121 in plants) ubiquitylation of histone H2A (PRC1) [15,16] (Figure 1). The PcG and TrxG group of proteins act via the Polycomb/Triothorax response elements (PRE) by maintaining a dynamic balance in order to reinforce active and silent gene expression states [12]. It is well documented that PRC2 catalyzes the trimethylation of lysine 27 of histone 3 [17] via 4 subunits. In Arabidopsis these subunits are three homologs of Enhancer of Zeste E (z) subunit (first found in Drosophila) CURLY LEAF (CLF), MEDEA (MEA/FIS1), SWINGER (SWN) which have a conserved SET domain and mediate the H3K27me3 histone
8
methyltransferase activity of PRC2 [15,17]. A Suppressor of Zeste-12 S(z)12 subunit (first found in Drosophila) which has 3 homologs of EMBRYONIC FLOWER 2 (EMF2), VERNALISATION 2 (VRN2) and FERTILISATION INDEPENDENT SEED (FIS2) , a conserved Zn finger and VEFS box, and function to stimulate H3K27me3 HMT activity of PCR 2. The third component contains five p55 homologs (MSI1), which have a conserved histone-binding domain and WD40 repeats and function to binds to histones and SU (Z). A single homologue of Extra Sex Combs (Esc first found in Drosophila) called the FERTILISATION INDEPENDENTENT ENDOSPERM (FIE) domain in Arabidopsis has conserved WD40 repeats and binds to H3K27me3 mark [15,17-20]. It is noteworthy that recent reports have indicated that PRC1 may take lead in PcG mediated gene repression in plants. This is because PRC1 activity may be required for PRC2 recruitment and PcG marking [21]. TrxG proteins counteract the silencing effects of PRC2, mediated by the histone modification H3K27me3 by catalyzing H3 lysine 4 trimethylation (H3K4me3) in order to maintain the embryonic cell state [12]. Thus, in order to maintain gene expression of both, stem cells and differentiated cells, the erasure and establishment of epigenetic marks is crucial [22].
2.2 AIM
The aim of this project is to characterize the gene reactivation and epigenetic reprogramming in the model plant Arabidopsis in order to understand the changes in the chromatin landscape between callus (transdifferentiating adult stem cells/pluripotent cells) and leaf (differentiated cells). The repression of key regulatory genes by Polycomb group mediated H3 lysine 27 trimethylation (H3K27me3) is critical for stem cell maintenance and cellular reprogramming in both plants and animals [4]. In Arabidopsis ~17% coding genes are marked by the H3K27me3 modification, indicating that H3K27me3 is a major gene silencing mechanism in Arabidopsis [10,23]. In order to appreciate the role of H3K27me3 in regulation of developmental patterning and response to external stimulus, the effect of H3K27me3 specific histone demethylase mutants on the regenerative capacity of Arabidopsis was tested [24,25].
There are 3 homologues in Arabidopsis that have H3K27me3 specific demethylase activity i.e. Relative of Early Flowering (REF6), Early Flowering (ELF6) and JMJ13.
9
Knock-out mutations in each of the H3K27me3 specific demethylases ref6-1, elf6-3 and jmj13 would interfere with gene reactivation due to failure of removing H3K27me3 mediated chromatin silencing [23]. After genotyping N654971, N666435, N604032 (NASC ID) seeds delivered as ref6-1 (SALK_001018C), elf6-3 (SALK_074694C) and jmj13 (SALK_104032) to confirm the presence of the T-DNA insert, their expression level (compared to wild-type) was determined by quantitative PCR (QPCR). As more than 15% of transcribed genes in Arabidopsis are marked by the H3K27me3 modification, characterising regeneration from leaf to callus in H3K27me3 specific demethylases mutants may also insinuate if defects in H3K27me3 mediated silencing antagonizes other marks. This is because the function of several thousands of coding genes that are involved in responses to external stimulus could be compromised [23]. It is noteworthy that mounting evidence from animal studies has indicated that defects in histone lysine demethylase expression leads to progression of several diseases and cancer. This suggests that histone lysine demethylases could be promising druggable targets [26]. As PRC are evolutionary conserved as key regulators of epigenetic states catalyzing H3K27me3 in both plants and animals, Arabidopsis could be an ideal candidate for studying the effects of histone lysine demethylases on regulatory mechanisms [10].
10
3. METHODS
In order to genotype the demethylase mutants in Arabidopsis, i.e. ref6-1 (SALK_001018C), elf-6 (SALK_074694C) and jmj13 (SALK_104032), the mutants were grown on compost along with Columbia wild type (Col.0) as a control. After DNA extraction and PCR purification from Arabidopsis mutants, RNA extraction of samples with confirmed genotypes was performed with SpectrumTM Plant Total RNA Kit (Sigma-Aldrich Co). After RNA extraction and DNAse treatment for RT-QPCR, reverse transcription (RT) was carried out with the SuperScript IV Reverse Transcriptase kit and results were analyzed with quantitative PCR (QPCR). The phenotypes of ref6-1, elf6-3, jmj13 were compared to Col.0 wild-type and results are shown in section 4.3 and 4.4.
3.1 Plant materials
Seeds of Arabidopsis T-DNA insertion mutants i.e. ref6-1 (At3g48430),
elf6-3 (At5g04240) and jmj13 (At5g46910) were ordered from the Nottingham
Arabidopsis Stock Centre (NASC). As wild type control, Columbia (Col.0) plants were used. Morphological changes in Col.0 callus induction were captured at various stages i.e. leaf, cotyledon, callus along a 21 day timeline. The housekeeping gene for positive control was ACTIN2 (AT3G18780). Three biological replicates were grown for experimental investigation.
3.2 Plant growth media
For callus induction of Col.0 plants, seeds were grown on media containing- Murashige and Skoog (MS) basal salt mixture: for 1l: 1X MS salts (4.33g), Sucrose (3g), Phytoagar (8g) at pH 5.8 and autoclaved [9,27]. Seeds were sterilized by adding 5 ml of 20% bleach, 0.01% Triton X-100 (for dispersing seeds), and rotated on a rotar for 8 minutes, after which they were washed with sterile H2O in a laminar flow hood. The growth media was diluted with sterile H2O 1:2 to 1:3 and added to the seeds for plating. Following stratification (breaking seed dormancy) of seeds (for 3 days at 4��C), the sterile seeds were pipetted onto plates containing MS media. After 10 days, individual leaf discs were extracted from each batch of Col.0 seedlings, and transferred on to the callus inducing media (CIM) for dedifferentiation. CIM (for 1l): 1 X Gamborg B5 salts (3.1g), Glucose (20g), MES (0.5g) (2-(N-Morpholino) ethanesulfronic acid), Gamborg Vitamins (1 ml) at pH 5.8 followed by 8g of phytoagar. After autoclaving, 50 ��l of 20,000 X 2,4-D (10mg/ml in 100% ethanol
11
stock solution) and 20,000X kinetin (1mg/ml in 1N KOH stock solution) was added to the CIM in laminar flow hood. After 14 days, 0.1g of sample was collected and flash frozen in liquid nitrogen and stored at -80��C for RNA extraction.
The growth conditions for all the plants during this project was long days (LD) of 16h light and 8h darkness period in the greenhouse at 22��C.
3.3 Growing Arabidopsis seeds on compost
Arabidopsis T-DNA insertion mutants i.e. ref6-1 (SALK_001018C, At3g48430), elf6-3 (SALK_074694C, At5g04240) and jmj13 (SALK_104032, At5g46910) were grown on compost following the procedure recommended by Nottingham Arabidopsis Stock Centre (NASC) [28]. Seeds were sown in 8 rows for each mutant and Col.0 wild-type.
3.4 DNA extraction from Arabidopsis Thaliana
In order to extract DNA from T-DNA insertion mutants to be used as a PCR template for genotyping, two leaves were placed in a 1.5ml eppendorf tube and 200 ��l of Extraction buffer (Edwards Solution (-0.2M Tris-HCl (pH 7.5), 0.25M NaCl, 0.025M EDTA and 5% SDS) diluted by 10-fold with TE buffer (10mM Tris HCl (pH 8) and 1mM EDTA) and crushed with the plastic micro tube pestles against the wall of tube, until the solution turned transparent green, with visible tissue residue in the solution. To remove the tissue residue, the sample was centrifuged at 14,000 rpm for 5 min and the supernatant was recovered. Then, the DNA precipitation protocol was followed to purify DNA. After this, 40 ��l of sodium acetate was added to the 200 ��l of DNA solution along with 500 ��l of ethanol to precipitate out the DNA from contaminants. The solution was then placed in -20��C freezer for 15 minutes, after which it was centrifuged at 14000 rpm for 20 minutes. The supernatant was discarded and the pellet was washed with 75% ethanol. The centrifugation step was repeated to remove salts from interfering with further manipulations. DNA pellet was then re-suspended in 30 ��l H20. DNA was quantified via NanoDrop prior to proceeding with Polymerized chain reaction (PCR).
3.5 Genotyping of T-DNA insertion lines
The primer pairs suggested by the web tool ‘T-DNA Primer Design’ on the Salk Institute website were used for genotyping [28].To detect the T-DNA insert, a left border primer (LBb1.3) was used. [28]. The intron-exon boundaries along with the
12
location of the T-DNA insert and primers were also marked with SnapGene software (Figure 2)
Table 1: Genotyping primers for H3K27me3 specific demethylases SALK Knock out mutants ref6-1 (SALK_001018C, At3g48430), elf-6 (SALK_074694C, At5g04240) and jmj13 (SALK_104032, At5g46910).
PCR was done separately for each primer set in 25 ��l of the following reaction mix: 10X DreamTaq (Thermo Scientific) green buffer (2.5 ��l), dNTP Mix (0.5 ��l of 10mM dNTP), Forward, Reverse and LbB1.3 for T-DNA primer (0.25 ��l of 10mM stock), H2O (18.125 ��l), DreamTaq DNA Polymerase (0.125 ��l) and genomic DNA as template (1 ��l).
Amplification was done using following cycles:
95��C (Initial denaturation) 95��C (Denaturation) 56��C (Annealing)
72��C (Extension)
72��C (Final Extension) 16��C Pause
3 min 30 sec 30 sec 40 sec 4 min
35 cycles
””’H3K27me3 Specific Demethylase Mutant
””SALK Line Number & NASC ID
”””Gene & Insert Location
”””Genotyping Primers
””ref6-1
N654971 SALK_001018C
”At3g48430 Insert location- exon
‘Forward Primer- 5′-ACAGGGAACACAGCTTCTGG-3′ Reverse Primer- 5’GCTGCACATTCCTCTTCCTC-3’
””’elf6-3
‘N666435 SALK_074694C
””’At5g04240 Insert location- exon
”’Forward Primer- 5′-TGGCTTAAAGCTTTGCCATT-3′ Reverse Primer- 5′-AGGGGTGCCTTTTCTACCTC-3′
””’jmj13
””N604032 SALK_109121C
”””At5g46910 Insert location- intron
”””Forward Primer- 5′-GTGAACACAGAAAACGGAAACA-3′ Reverse Primer- 5′-CGCCTGACAAAACCAAGTGT-3’
””””’Twenty-five ��l of PCR reaction was loaded on to 1% agarose gel for electrophoresis.
13
3.6 RNA Extraction
RNA Extraction was performed using SpectrumTM Plant Total RNA Kit (Sigma- Aldrich) according to manufacturers instructions [29]. For sample collection, 0.1g of leaf samples was collected for RNA extraction and ground to a fine powder in dry ice. Then, 500 ��l of Lysis Solution supplemented with 2-mercaptoethanol (2ME) was added to 100 mg of tissue powder, and vortexed for 30 seconds after which it was incubated at 56��C for 5 minutes. To pellet the cellular debris, the sample was centrifuged for 3 minutes and the lysate supernatant was pipetted through the filtration column seated in a 2 ml collection tube without any pellet contamination. To bind RNA to the column, 500 ��l of binding solution (provided in the SpectrumTM Plant Total RNA Kit) was pipetted into the cleared lysate and mixed by inverting the tube 5 times. Then, 700 ��l of the mixture was added to the binding column seated in a 2 ml collection tube. The solution was then centrifuged at maximum speed for 1 minute and the flow through was discarded. Any extra droplets were tapped by placing the collection tube upside down on a sterile absorbent paper. For the first column wash, 500 ��l of Wash Solution 1 (provided in the SpectrumTM Plant Total RNA Kit) was pipetted into the column and as previously, the flow through was discarded. After diluting the Wash Solution 2 Concentrate as instructed by the manufacturer, 10ml of 100% ethanol and 500 ��l of Wash Solution 2 was added into the column and centrifuged at maximum speed for 30 seconds. The flow through was discarded as previously and the previous step was repeated as a third column wash, with Wash Solution 2. The column was then centrifuged dry for 30 seconds prior to proceeding with the elution step. Then, the column was transferred to a 2 ml collection-tube and 50 ��l of Elution Solution was added directly into the center of the filter inside the column. The samples were left at room temperature for 1 minute after which the column was centrifuged for 1 minute to elute. RNA was quantified with the NanoDrop and the samples were stored at -20��C (Table 3).
3.7 Reverse Transcription and QPCR
In order to remove genomic DNA contamination, the DNAse treatment for RT-PCR was followed (Promega). Then, 500 ng RNA sample was added to 1 ��l RQ1 RNase- Free DNase 10X Reaction buffer along with RQ1 RNase-Free DNase (1u/��gRNA) and 10 ��l of nuclease free H20 to make up a final volume of 20 ��l. The reaction mixture was then incubated at 37��C for 30 minutes. After this, 1 ��l of RQ1 DNase
14
stop solution was added in order to terminate the reaction. This was then followed by incubation at 65��C for 10 minutes for DNase inactivation. The samples were then incubated at 42��C for 30 minutes for reverse transcription, and 85��C for 10 minutes to denature the RTase.
The ‘positive no RT’ control was not required for this setup as the primers span introns or exon-exon junctions and thus, intron containing genomic sequences would not be amplified. The ‘negative no RT’ control was H20. For reverse transcription of the extracted RNA to cDNA, the SuperScript IV Reverse Transcriptase kit was used. In order to anneal the primer to template DNA, 1 ��l of 2 ��M gene specific reverse primer was added with 1 ��l Oligo d(T)20 (50��M), 1 ��l dNTP (10mM), 500 ng of RNA and 13 ��l of nuclease-free water. The components were mixed and centrifuged briefly and the RNA-primer mix was then heated at 65��C for 5 minutes following which, it was incubated on ice for 1 minute. In order to prepare the RT reaction mix, the 5X SSIV buffer was vortexed and centrifuged briefly. After this, 4 ��l of 5X SSIV Buffer was added to 1 ��l DTT (100mM) and 1 ��l RNaseOUT Recombinant RNase Inhibitor, along with 1 ��l of SuperScript IV Reverse Transcriptase (200 U/ ��l) in a
1 mL eppendorf tube and centrifuged briefly. The RT reaction mix was then added to annealed RNA and incubated at 55��C for 10 minutes. The reaction was then inactivated by incubating at 80��C for 10 minutes.
For the amplification and detection of the cDNA samples, further analysis was carried out by quantitative PCR (QPCR [30]).
The LightCycler 480 SYBR Green I Master Mix (Roche) kit was used to prepare 10 ��l standard reaction. Then, 2 ��l of H2O was added to 5 ��l 2X SYBR Green I Master Mix. Following this, 0.5 ��l of each gene specific forward and reverse primer (Table 2) was added. The mix was prepared without the cDNA, as the mix and cDNA were pipetted into the PCR plate on the liquid handling robot. For the negative control, the template DNA was replaced with 2��l PCR-grade H2O.
The parameters programmed for the LightCycler 480 system PCR run were: Pre- incubation for 1 cycle at 95��C for 5 minutes, amplification for quantification for 45 cycles at 95��C for 10 seconds and 72��C for 35 seconds, melting curve for 1 cycle at 95��C for 5 seconds and 65��C for 1 minute, cooling for 1 cycle at 40��C for 10 minutes for a 384 well plate.
In order to analyze results from QPCR, the comparative Ct method was used. The Ct value average of triplicate repeats from REF6, ELF6 and JMJ13 was taken to
15
calculate 2-ct for each sample. The resulting values for REF6, ELF6 and JMJ13 were normalized with ACTIN2.
””Genes encoding H3K27me3 demethylase
”’Gene specific primers for RT-QPCR
””REF6
”’REF6_UPL_F: 5′-TTAATGTTACTGTGCCATCCAGA-3′ REF6_UPL_R: 5′-CGAGCTCTTCGGCAACTATC-3′
””ELF6
”’ELF6_F: 5′-CAAAGCCTTCAGCAGAACAA-3′
ELF6_R: 5′-CAAAATCTTTTGATTTAGCTAGGGTAG-3′
””Control- ACTIN2
”””ACT2_ F: 5′-CCGCTCTTTCTTTCCAAGC-3′ ACT2 _R: 5′-CCGGTACCATTGTCACACAC-3
”Table 2: Gene specific primers for QPCR.
3.8 Accession Numbers
1. ACTIN2(AT3G18780) 2. elf6-3(At5g04240)
3. jmj13(At5g46910)
4. ref6-1(At3g48430).
16
4 RESULTS
4.1 Genotypic analysis for N654971, N666435 and N604032 delivered as ref6-1 (SALK_001018C), elf6-3 (SALK_074694C) and jmj13 (SALK_104032):
In order to analyze the SALK insertion lines, At3g48430, At5g04240 and At5g46910 were mapped in the SnapGene software. The location of the T-DNA insert is shown by an inverted triangle in Figure 2. The T-DNA is located at the 8th exon in ref6-1 (At3g48430), the 1st exon in elf6-3 (At5g04240) and the 1st intron in jmj13 (At5g46910). The location of genomic primers along with the T-DNA left border primer (LBb1.3) is also marked (red). Different sized PCR products are expected for wild-type and mutant allele.
a) REF6
‘b) ELF6
’17
c) JMJ13
‘Figure 2: Genotyping of H3K27me3 specific demethylase knock out mutants.
Thick lines, exons, Thin lines-introns. T-DNA insertion site is marked by an inverted triangle. The primers were designed left and right of the T-DNA insertion site. A third primer (LBb1.3) is located on the left border of the T-DNA insert at 110bp (marked in red). The LBb1.3 primer pairs with one of the gene specific primers to amplify a band with a different size than the gene-specific primers do with wild-type. a) REF6: Genomic Primers are located 3748 bp and 4990 bp from the transcription start site. Without the T-DNA insert, the gene specific left and right primers will prime to produce an amplicon size of 743 bp. In the mutant, an amplicon size of 374 bp is expected. Lane-1, 2, 3, 4, ref6-1 lines; Lane 5, Col.0 (positive Control). Sample 3 and 4 shown are homozygous for the T-DNA insert as only a band at 374bp is seen.1 % agarose gel was run for 35 minutes at 100V. Marker-1Kb Hyperladder I b) ELF6: genomic primers are located 25 bp and 689 bp from the transcription start site. Without the T-DNA insert, the gene specific left and right primers will prime to produce an amplicon size of 665 bp. In the mutant, an amplicon size of 383bp is expected. Lane-1, 2, 3, and 4, elf6-1 samples; Lane 5 and 6, Col.0 (positive control). Sample 2 & 3 are heterozygous for the T- DNA insert as a band at 665 bp and 383 bp is present. Sample 4 is homozygous for the T- DNA insert as a band at 665 bp only is present. 1 % agarose gel was run for 35 minutes at 100V c) JMJ13: Genomic primers are located at 34 bp and 515 bp from the transcription start site. Without the T-DNA insert, the gene-specific left and right primers will prime to produce an amplicon size of 481 bp. In the mutant, an amplicon size of 341 bp is expected. Lane-1, 2, 3, 4- jmj13 samples, Lane 5, 6 is Col.0 (positive control). T-DNA insert is not present for sample 1 & 2 as a 481 bp band (same as Col.0) is seen. 1 % agarose gel was run for 35 minutes at 100V. HyperLadder 100 bp (Bioline) was used as the marker.
18
After DNA extraction from Arabidopsis (section 3.4), genotyping results were visualized following PCR on agarose gel electrophoresis stained with SYBERsafe. Figure 2 shows the cartoon of T-DNA insertion sites for ref6-1, elf6-3 and jmj13. The gel images show the genotyping results for T-DNA insertion mutation in REF6, ELF6 and JMJ13. Col.0 wild-type was used as positive control.
As N654971 was delivered as homozygous for the ref6-1 mutation, a 374 bp mutant band is expected. Figure 2a, lane 1 & lane 2 indicate a failed PCR for the loaded samples. This could be due to low concentration or poor quality of genomic DNA with contaminants inhibiting the PCR reactions. The samples loaded in lane 3 and 4 produce an amplicon at 374bp confirming that they are homozygous for the T-DNA insertion ref6-1 mutation. Col.0 positive control has produced an amplicon of 743 bp as expected in lane 5.
As N666435 was delivered as homozygous for the elf6-3 mutation, a 383 bp mutant band is expected. Figure 2b, lane 1 indicates a failed PCR for the loaded sample. This could be due to low concentration of genomic DNA present or poor quality of genomic DNA with contaminants inhibiting the PCR reaction. However, the samples loaded in lane 2 and 3 produce an amplicon at 383 bp (mutant) and 665 bp (wild- type) confirming that they are heterozygous for the T-DNA insertion elf6-3 mutation. This is because if N666435 is heterozygous for the mutation, a 665bp wild type band and a 383 bp mutant band is expected. The sample loaded in lane 4 produces an amplicon at 383 bp confirming that it is homozygous for the T-DNA insertion elf6-3 mutation. The Col.0 wild-type samples loaded in Lane 5 & 6 produce the expected 665 bp amplicon size.
As N604032 was delivered as heterozygous for the jmj13 mutation, both, a 341 bp mutant band and/or a 481bp wild type band can be expected from this segregating population. Figure 2c, lane 2, lane 4, and lane 5 produce and amplicon size of 481bp confirming that they do not have the T-DNA insertion jmj13 mutation. This is because if N666435 does not have a mutation, a single 410 bp wild-type band is expected. Figure 2c, lane 1 and 3 show faint bands at 481 bp for the loaded samples. This could be due to low concentration of genomic DNA present in those samples. The positive control Col.0 produces a defined band in lane 5 & 6 at 481bp.
For RNA extraction, two samples (one heterozygous for insertion in elf6-3) from each of the two lines i.e. ref6 and elf6, confirmed for T-DNA insertion were chosen.
19
4.2 Expression analysis for REF6 and ELF6:
In order to confirm the gene knock out, expression analysis was performed on ref6-1 and elf6-3 mutants (Section 3.7). Following RNA extraction, RNA was quantified with the NanoDrop (Table 3). To remove any genomic DNA contamination, the DNAse treatment for RT-PCR was followed (Promega). After this RNA was reverse transcribed to cDNA using the SuperScipt IV Reverse Transcription kit (Section 3.7).
”’H3K27me3 Specific Demethylase mutants
””’Concentration of RNA sample (ng/��l)
”””’Purity (260/280)
””Ref6-1
Sample 1 (Homozygous) Sample 2 (Heterozygous)
”’96.33 67.30
””2.13 2.17
””Elf6-3
Sample 1 (Heterozygous) Sample 2 (Homozygous)
‘82.99 67.30
”2.13 2.07
”’Col.0 Wild-Type
Sample 1
””’132.27
”””’2.08
”’Table 3: Concentration and purity determination of RNA samples by NanoDrop.
As the primers designed for RT-QPCR span introns, the melting curve would confirm the presence of genomic DNA contamination, as the melting temperature of genomic DNA would be different from cDNA. This would be seen as a different band size on regular PCR gel electrophoresis or as different peaks on melting curve of QPCR. Figure 3a shows ref6-1 line 1 sample without DNAse digest (shown in red). As the presence of genomic DNA contamination was noted (sample melting at different temperature) the next step of experiments were performed after DNAse treatment prior to reverse transcription and proceeding to QPCR (Section 3.7). The ref6-1 samples following DNAse digest (Fig 3a, b) show a homozygous defined peak confirming that there is no genomic DNA contamination. To conclude, the melting curve genotyping analysis indicated a single peak for ACTIN2 transcripts, ELF6 transcripts and REF6 transcripts confirming that there is no genomic cDNA contamination (Figure 3 b,c,d).
20
Figure 3: Melting curve graph for ref6-1, elf6-3 sample fluorescence versus temperature for RT-QPCR. Results were normalized with ACTIN2. Col.0 was used as
positive control. (a) ref6-1 line 1 sample with DNAse digest(shown in blue) and without DNAse digest (shown in red). In order to remove genomic DNA contamination, the DNAse digest was performed before the RT reaction (section 3.7). (b) Following DNAse digest, a homozygous defined peak was observed for ref6-1 line 1 and 2 along with Col.0, indicating that all PCR products melt at the same temperature and there is no contamination. (c) elf6-3 along with Col.0 samples after DNAse digest. A single peak is seen for all elf6-3 samples confirming that there is no contamination from genomic DNA. (d) All samples were normalized with ACTIN2. A defined peak was observed for the ACTIN2 control in elf6-3 and ref6-1 mutants along with Col.0 wt.
Figure 4 illustrates RT-QPCR results for REF6 and ELF6 gene expression. Col.0 cDNA was used as positive control. The results were normalized with ACTIN2 (housekeeping gene). The mutant ref6-1 (line 1 & 2), elf6-3 (line 1 & 2) is expected to show lower expression compared to wild-type if the mutation is present.
’21
”a)
0.035 0.030 0.025 0.020 0.015 0.010 0.005 0.000
Ref6
”ref6-1 (line 1)
ref6-1 (line 2)
Col.0 (WT)
”b) 0.035
0.030 0.025 0.020 0.015 0.010 0.005 0.000
Elf6
”elf6-3 (line 1)
elf6-3 (line 2)
Col. 0 (WT)
Figure 4: Expression level of genes ELF6 and REF6 determined by QPCR. Expression levels were normalized with ACTIN2. To remove any genomic DNA
contamination, the DNAse treatment for RT-PCR was followed (Promega). Col.0 cDNA is used as positive control. Each sample was extracted from a different line. The sample from ref6-1 (line 1), ref6-1 (line 2) and elf-3 (line 2) are homozygous for the T-DNA insertion (Fig 2). elf6-3 (line 1) is heterozygous for the T-DNA insertion. (a) QPCR analyses on ref6-1 mutants indicate that the expression level for ref6-1 is similar to Col.0, therefore it is not a true knock out. ref6-1 (line 2) however has only 4% of Col.0 WT expression confirming that the mutation is present in this line. elf6-3 (line1 & line2) both show the presence of the T- DNA insertion (expression level reduced to 10% of Col.0 WT).
The sample from ref6-1 (line1) showed similar expression to Col.0 wild-type indicating that the mutation is not present (Figure 4a), which is contradictory with genotyping results (figure 2a). This is because if the T-DNA insertion is present in
22
Transcript level relative to ACTIN2 Transcript level relative to ACTIN2
ref6-1 line 1, it is unlikely that the gene is still expressed. These results indicate that a wild-type copy was present in ref6-1, line 1, however, PCR failed to detect it. This could be because of poor quality of genomic DNA inhibiting the PCR reaction. Moreover, ref6-1 (line 2) showed only 4% of Col.0 wild-type expression level confirming that it is a true knockout. Also, elf6-3 (line 1 & line 2) have only 10% of Col.0 wild-type expression confirming that it is a true knockout.
To conclude, ref6-1 (line 2,homozygous), elf6-3 (line 1, heterozygous) and elf6-3 (line 2, homozygous) have the T-DNA insertion mutation as seen by decrease in expression levels compared to wild-type Col.0.
4.3 Phenotypes of ref6-1, elf6-3 and jmj13/N604032 mutants compared with Col.0 wild-type.
‘Figure 5: Phenotypes of Col.0, elf6-3, ref6-1 and jmj13/N604032. (A), (B), (C), and (D): Col.0, elf6-3, ref6-1 and N604032 (delivered as jmj13 however, mutation not confirmed by genotyping) 15 days after sowing seeds on compost (t=15). N604032 show similar phenotype to Col.0. ref6-1 has slower growth compared to Col.0 wild type. White scale bar= 3cm (E), (F), (G), and (H): Col.0, elf6-3, ref6-1 and N604032 (phenotype and genotype analysis confirm mutation is absent) 28 days after sowing seeds on compost. Mutations in ELF6 cause early flowering compared to mutations in REF6 (no flowering after t=28). Plants were observed for 28 days for Long Days (LD) of 16h light and 8h darkness period in the greenhouse at 22��C.
The phenotypes of Col.0, ref6-1, elf6-3, and N604032 (delivered as jmj13 however, genotyping confirms that they do not have the T-DNA insert, figure 2c) is shown in
23
figure 5, 15 days and 28 days after sowing SALK_001018C, SALK_074694C and SALK_104032 seeds on compost. After 15 days, N604032 / jmj13 showed similar phenotype to Col.0, as expected, because genotyping and expression analysis confirmed that the T-DNA insertion mutation is not present for N604032 / jmj13. However ref6-1 (Figure 5c) has slower growth compared to Col.0 (Figure 5a) and elf6-3 (Figure 5b). After 28 days, Col.0, elf6-3 and N604032 mutants (Figure 5e, f, h) showed flowering. Mutations in REF6 cause delay in flowering however, mutations in ELF6 show normal development until 15 days however, flower before Col.0 wild type.
4.4 Phenotypes of Col.0 wild type, ref6-1 and elf6-3 mutants during callus formation.
‘Figure 6: Leaf explants of Col.0 wild type and the stages involved in callus formation. REF6 and ELF6 mutants lose the ability to form callus. For callus
induction of Col.0 plants, sterilized seeds were grown on MS media plates and transferred onto callus inducing media (CIM) plates after 14 days (Section 3.2). For ref6-1 and elf6-3 plants, leaves were dissected from elf6-3 and ref6-1 plants grown on compost (after t=28 days, Fig 5 f, g) and transferred onto CIM. Leaf explants were dissected from Col.0 in a similar way as ref6-1 and elf6-3 plants and grown on compost, and Col.0 was used as positive control. The leaves of ref6-1, elf6-3 and Col.0 were washed with ethanol and 0.01% Triton X-100. (A) Leaf explant from Col.0. The true leaves have hair like structures (indicated by arrows, scale bar=7mm) (B) True leaves were used for wild-type callus regeneration and transferred on to CIM (marked as t=0 days) (scale bar=16.7mm). (C) After 21 days, Col.0 wild type leaves were fully differentiated into callus. (D) Col.0 WT explants from plant grown on compost differentiate into callus after 21 days (E) 21-days old leaf explants from ref6-1 mutants lose ability to form callus (scale bar=20mm). (F) elf6-3 mutants are very prone to fungal infection and lose their ability to form callus (scale bar=20mm).
24
In order to regenerate callus from leaves, sterilized Col.0 wild type seeds were grown on MS media plates. True leaves from Col.0 leaf explants were extracted and transferred on to callus inducing media (figure 6a, 6b; section 3.2). After 21 days the Col.0 wild type leaves regenerated into callus.
In order to test the effect of H3K27me3 specific demethylase mutants on regeneration of callus (transdifferentiating adult stem cells/pluripotent cells) from leaves (differentiated cells), Col.0 explants were dissected in a similar manner as ref6-1 and elf6-3 mutants from plants grown on compost. The leaves were transferred to callus inducing media after washing with ethanol and 0.01% Triton X- 100. After 21 days, Col.0 wild-type (positive control) differentiated into callus (Figure 6d). Figure 6(e) shows that ref6-1 mutants (line 2 confirmed as homozygous, Figure 2a), lose the ability to form callus (only 4% of Col0 wild-type expression confirmed by QPCR, Figure 4a). Figure 6(f) shows that elf6-3 mutants (line 1 confirmed as heterozygous (Figure 2b), line 2 confirmed as homozygous) lose the ability to form callus (only 10% of Col.0 wild-type expression confirmed by QPCR, Figure 4b). These results indicate that mutations in H3K27me3 specific demethylases ref6-1 and elf6-3 resulted in a loss of their ability to form callus.
It is noteworthy that owing to discrepancy between ref6-1 line 1 genotyping results (Figure 2a) and expression analysis (Figure 4a) by QPCR, ideally the experiment should be repeated three times. This would also add statistical significance to the QPCR data, however, there was limited time given for this project.
25
5. DISCUSSION
Histone methylation at lysine residues is associated with activation (H3K4) and repression (H3K27) of transcription by allowing differential regulation of cellular processes (Figure 1 [11]). As the H3K27me3 is associated with gene repression [9], in order to characterize epigenetic reprogramming from leaves (differentiated cells) to callus (transdifferentiating adult stem cells/pluripotent cells), the effect of H3K27me3 specific demethylases on Arabidopsis was tested using the classical genetic knock out approach. Therefore, the three homologues in Arabidopsis that have H3K27me3 specific demetheylase activity i.e. Relative of Early Flowering (REF6), Early Flowering (ELF6) and JMJ13 were chosen for this study.
Genotyping N654971, N666435 and N604032 delivered as ref6-1 (SALK_001018C), elf6-3 (SALK_074694C) and jmj13 (SALK_104032) revealed that the T-DNA insertion mutation was not present in all lines (section 4.1). N654971/ ref6-1 was delivered as homozygous for the T-DNA insertion mutation and ref6-1 was confirmed to be homozygous for the T-DNA insertion in two lines (Figure 2a). N666435/ elf6-3 was delivered as homozygous for the T-DNA insertion however, two lines with both, a wild-type and a mutant copy i.e. heterozygous, were seen along with one homozygous line (Figure 2b). This means that breeding and crossing them could be required if lower expression level owing to the T-DNA insertion is seen. N604032/ jmj13 was delivered as heterozygous for the T-DNA insertion however; genotyping results seen with PCR-gel electrophoresis confirmed that there was no mutation present in any line of the samples delivered as jmj13 (Figure 2c).
As the presence of low quality/concentration of DNA can inhibit the PCR reaction and effect the interpretation of results, in order to confirm the gene knockout, expression analysis was performed on the ref6-1, elf6-3 and jmj13 mutants (section 4.2). Quantifying the abundance of gene-specific transcript should show a decrease in the gene expression of mutants ref6-1 and elf6-3 when compared to Col.0 wild-type. Quantitative reverse transcription PCR was used for expression analysis owing to its sensitivity in detecting the accumulation of PCR product, seen as the amount of florescence emitted from the SYBR Green fluorophore [31]. The ‘no RT’ control was not required for this setup as the primers span introns or exon-exon junctions.
Figure 3 shows the melting curve genotyping analysis indicating the amplification efficiency and replicate consistency. As the melting curve shows the changes in fluorescence when double stranded DNA melts into single stranded DNA, the melting
26
dynamics of in the curve show the presence of genomic DNA contamination as a separate curve with samples melting at different temperatures. The RT-QPCR technique is highly sensitive in detection of genomic DNA contamination if amplicon from genomic DNA is of a different size due to primers spanning introns. This is because, the melting curve output in QPCR groups uniform PCR amplicons, and genomic DNA would be grouped separately from cDNA. Figure 3a shows the melting curve genotyping analysis illustrating genomic DNA contamination in samples without DNAse digest (red) and uniform peaks for PCR amplicon in samples after the DNAse digest prior to reverse transcription (section 3.7). Following the DNAse digest, homogenous peaks for REF6, and ELF6 with Col.0 with gene specific primers, and ACTIN2 primers is seen (Figure 3 b,c,d). As RT-PCR focuses on the exponential phase of PCR, for which in an ideal scenario, the products would double with every cycle, the threshold and CT values provide precise and accurate data for quantitation of gene expression [32]. Figure 4 shows the expression levels of REF6 and ELF6 calculated by the comparative CT method relative to the housekeeping gene, ACTIN2. It can be confirmed that ref6-1 (line 2,homozygous; 4% Col.0 WT expression), elf6-3 (line 1, heterozygous; 10% Col.0 WT expression) and elf6-3 (line 2, homozygous; 10% Col.0 WT expression) have the T-DNA insertion mutation as seen by a drop in expression levels compared to wild-type Col.0. This is because mutations in REF6/ELF6 result in loss of gene expression. This indicates that REF6/ELF6 could potentially affect gene expression of PcG targets owing to changes in H3K27me3 levels on chromatin of PcG target genes [33].
Contrastingly, ref6-1 (line 1) confirmed as homozygous for the T-DNA insertion, has similar expression to wild type. RT-QPCR can be used to quantify cDNA from smaller amount of samples than northern blot and other techniques, and is superior to PCR gel electrophoresis as it provides a quality check for the samples via the melting curve analysis. The discrepancy in the results for ref6-1 genotyping (Figure 2a) and expression analysis (Figure 4a) could be because PCR failed to detect the wild-type copy in genotyping experiments (Section 4.1). This is because it is highly unlikely for the gene to be expressed if the T-DNA insertion mutation is present.
In order to confirm that the results seen from genotyping ref6-1 and elf6-3 were evident in their phenotype, ref6-1, elf6-3 and N604032/ jmj13 were grown on compost along with Col.0 (section 4.3). After 15 days, N604032 / jmj13 showed similar phenotype to Col.0 as expected, because mutation is not present in any line
27
of N604032 / jmj13 (Figure 2a). Conversely ref6-1 (Figure 5c) had slower growth compared to Col.0 and elf6-3. After 28 days, Col.0, elf6-3 and N604032/jmj13 flowered; however, delayed flowering was observed for ref6-1. Similarity searches on BLAST with ELF6 against The Arabidopsis Information Source (TAIR) show that At5g04240/ELF6 gene has the highest similarity to REF6 gene At3g48430. However, different expression patterns were indicated for the two genes. For example, REF6 expression is highest in the shoot apical meristem region, and primary and secondary root tips. However REF6 expression is lower in cotyledons, leaves, and root axis. Contrastingly ELF6 expression is highest in cotyledons and leaves and there is no detectable ELF6 expression in apical meristems or root tips [34]. The difference in the expression patterns of REF6/ELF6 could be because of their different functions. ELF6 acts as a repressor of the photoperiodic flowering pathway whereas REF6 acts as an FLC repressor [34].
Moreover, the delay in flowering time for ref6-1 is because REF6 targets floral organ identity genes such as AP1 (apetala1), AP3 (APETALA3) and PI (PISTILLATA) [23], and loss of REF6 owing to the T-DNA insertion mutation could potentially cause high H3K27me3 mediated silencing on these genes, which resulted in the failure to activate the floral pathway [23].Figure 6F shows that elf6-3 mutants were very prone to fungal infection indicating that elf6-3 mutants are more prone to biotic stress and mutation may have compromised the defence mechanism of this line. This could be because PcG proteins bind to stress responsive genes and the H3K27me3 occupancy changes at these genes upon stress conditions [35]. Mutation in ELF6 which is normally expressed at high levels in leaves and cotyledons, may have caused ectopic H3K27me3 levels, affecting the genes that respond to stress [35].It is noteworthy that H3K27me3 has a dynamic genome-wide pattern that differs between meristematic and leaf tissues, indicating that characterizing PcG target genes to study the changes between repressed and active chromatin states may reveal further insights [13].
In order to study the effects of H3K27me3 specific demethylases on the regenerative capacity of Arabidopsis, leaf explants were dissected from ref6-1, elf6-3 and Col.0 (positive control). The leaves were transferred to callus inducing media after washing with ethanol and 0.01% Triton X-100. After 21 days, Col.0 fully differentiated into callus however, mutations in ref6-1 and elf6-3 resulted in a loss of their ability to form
28
callus. These results indicate that REF6 and ELF6 may contribute to gene reactivation by actively removing the H3K27me3 mark [23].
Polycomb complexes are well conserved developmental regulators in plants and animals, and severe phenotypic changes in H3K27me3 specific demethylase mutants ref6-1 and elf6-3 confirm the essential role of PcG regulated H3K27me3 mark in plant development[25]. Analogous demethylase knockout studies in animals have also revealed defects in developmental patterns. As PcG targets multiple genes, misexpression of PcG target genes has shown to cause uncontrolled proliferation and diseases [14]. To conclude, H3K27me3 demethylases are essential for removing PRC2-mediated chromatin silencing and hence, H3K27me3 plays a fundamental role in regeneration of callus (transdifferentiating adult stem cells/pluripotent cells) from leaves (differentiated cells). This suggests that ‘writers’ for the H3K27me3 modification i.e. histone lysine methylasetransferases (HMT, Figure 1) act in parallel with the ‘erasers’ of the H3K27me3 modification, i.e. HDM (REF6, ELF6 and JMJ13 for Arabidopsis) during the tissue identity transition i.e. leaf- callus transition [9]. It is noteworthy that even though the discovery of histone lysine demethylases seems to impact the epigenetic potential of histone lysine methylation, certain methylation marks residing in restricted regions of the chromatin are very stable [36]. Moreover, as PcG mediated gene repression is evolutionary conserved, it is clear that orchestrated response to multiple signals converging from the chromatin are critical for ensuing growth, differentiation, regeneration and dedifferentiation [14].
6. FUTURE DIRECTIONS
It is noteworthy that in order to study the effect of histone demethylases on the (leaf- callus) regenerative capacity of Arabidopsis, a classical loss-of function mutation approach was used. However, the downside of this approach is that it fails to identify genes that act redundantly [37]. Moreover, targeted expression of H3K27me3 specific demethylase genes in specific tissues can identify the changes in expression levels between leaf and callus [38]. These ‘forward genetics’ based experiments can be carried out by fusing promoters with genes that are specifically expressed in certain tissues, for example, using an STM promoter (STM gene specifically expressed in meristems) to target demethylase gene expression in meristems. Another artefact of genetic knock-out approach is that in certain demethylases, a
29
single gene knock-down may not result in visible phenotypes owing to redundancy, which can affect the interpretation of results [37].
Activation tagging is an approach that can be complemented with the genetic knock out approach used for this study, in order to confirm the effect of demethylases not only by disrupting their function, but also by tissue-specific activation [37]. Therefore, a GAL4-based targeted activation tagging system or an inducible gene expression system can be used for future experiments [37]. In the GAL4/UAS activation tagging mutagenesis approach, random insertion of tandem copies of the enhancer sequence from the CamV35S promoter via Agrobacterium-mediated transformation is carried out. The enhancer sequences will enhance the expression of neighbouring genes and thus provide a ‘gain-of-function’ phenotype [37]. The GAL4: VP16 synthetic transcription activator expressed in host plants in a tissue-specific manner can be used for the transformation of T-DNA, with an upstream activation sequence [37]. Another approach that can allow local changes in gene expression without alterations to whole plant development are the inducible gene expression systems. These systems include a chimeric transcription factor/activator and a targeted gene of interest [38]. They provide a spatial and temporal control as the activator can be fused to tissue specific promoters and a strong and ubiquitous expression can be gained in response to appropriate chemical compounds or treatments. Examples of some inducible systems that can be used for this study are the ethanol, dexamethasone and shock induction [38].
Another area that remains to be addressed in the field is the existence of ‘bivalent domains’ for plants like in the case of animals. As both PcG and TrxG proteins act via the PRE antagonistically [18] (Section 2.1), the presence of demethylases that regulate bivalent domains, which maintain the activating and silencing marks (i.e. H3K27me3 and H3K4me3) and allow spontaneous switching of programs to regulate cell identity need to be identified for plants [39]. Bivalent domains in animals allows stem cells to create ‘poised’ promoters in key developmental genes [22]. Therefore, if bivalent domains are present in plants, a chromatin immunoprecipitation followed by sequencing (ChIP-Seq) approach can be used to map the transcriptional complexes and epigenetic modifications throughout the genome to detect them [40]. On comparing the patterns between H3K27me3 and H3K4me3 the identification of a profile that is enriched in bivalent genes for plants could be possible [40]. It is noteworthy that even though ChIP assays are powerful tools for revealing DNA-
30
protein interactions, highly specific antibodies for the modification will be of critical importance [41].
7. REFERENCES
1. Ladewig J, Koch P, Br��stle O: Leveling Waddington: the emergence of direct programming and the loss of cell fate hierarchies. Nature Publishing Group 2013, 14:225’236.
2. Takahashi K: Cellular Reprogramming. Cold Spring Harbor Perspectives in Biology 2014, 6:a018606’a018606.
3. Takahashi K, Yamanaka S: Induction of Pluripotent Stem Cells from Mouse Embryonic and Adult Fibroblast Cultures by Defined Factors. Cell 2006, 126:663’676.
4. la Paz Sanchez de M, Aceves-Garc��a P, Petrone E, Steckenborn S, Vega-Le��n R, ��lvarez-Buylla ER, Garay-Arroyo A, Garc��a-Ponce B: The impact of Polycomb group (PcG) and Trithorax group (TrxG) epigenetic factors in plant plasticity. New Phytol 2015, doi:10.1111/nph.13486.
5. Hanna JH, Saha K, Jaenisch R: Pluripotency and Cellular Reprogramming: Facts, Hypotheses, Unresolved Issues. Cell 2010, 143:508’525.
6. Vierbuchen T, Wernig M: Molecular Roadblocks for Cellular Reprogramming. MOLCEL 2012, 47:827’838.
7. Sugimoto K, Gordon SP, Meyerowitz EM: Regeneration in plants and animals: dedifferentiation, transdifferentiation, or just differentiation? Trends in Cell Biology 2011, 21:212’218.
8. Southall TD, Davidson CM, Miller C, Carr A, Brand AH: Dedifferentiation of Neurons Precedes Tumor Formation in lola Mutants. Developmental Cell 2014, 28:685’696.
9. He C, Chen X, Huang H, Xu L: Reprogramming of H3K27me3 Is Critical for Acquisition of Pluripotency from Cultured Arabidopsis Tissues. PLoS Genet 2012, 8:e1002911’13.
10. Liu C, Lu F, Cui X, Cao X: Histone Methylation in Higher Plants. Annu. Rev. Plant Biol. 2010, 61:395’420.
11. C. David Allis (2014). Epigenetics. 2nd ed. London: Cold Spring Harbor Laboratory Press. p1-117.
12. Steffen PA, Ringrose L: What are memories made of? How Polycomb and Trithorax proteins mediate epigenetic memory. Nature Publishing Group 2014, 15:340’356.
13. Engelhorn J, Blanvillain R, Carles CC: Gene activation and cell fate control in plants: a chromatin perspective. Cell. Mol. Life Sci. 2014, 71:3119’3137.
31
14. Sawarkar R, Paro R: Interpretation of Developmental Signaling at Chromatin:
The Polycomb Perspective. Developmental Cell 2010, 19:651’661.
15. Kim D-H, Sung S: Polycomb-Mediated Gene Silencing in Arabidopsis thaliana.
Molecules and Cells 2014, 37:841’850.
16. Xiao J, Wagner D: ScienceDirectPolycomb repression in the regulation of growth and development in Arabidopsis. Current Opinion in Plant Biology 2015, 23:15’24.
17. Mozgova I, K��hler C, Hennig L: Keeping the gate closed: functions of the polycomb repressive complex PRC2 in development. Plant J 2015, 83:121′ 132.
18. Bouyer D, Roudier F, Heese M, Andersen ED, Gey D, Nowack MK, Goodrich J, Renou J-P, Grini PE, Colot V, et al.: Polycomb Repressive Complex 2 Controls the Embryo-to-Seedling Phase Transition. PLoS Genet 2011, 7:e1002014’19.
19. Margueron R, Reinberg D: The Polycomb complex PRC2 and its mark in life. Nature 2011, 469:343’349.
20. Angel A, Song J, Dean C, Howard M: A Polycomb-based switch underlying quantitative epigenetic memory. Nature 2011, 476:105’108.
21. Merini W, Calonje M: PRC1 is taking the lead in PcG repression. Plant J 2015, 83:110’120.
22. Popowski M: Repressors of reprogramming. WJSC 2015, 7:541’7.
23. Lu F, Cui X, Zhang S, Jenuwein T, Cao X: Arabidopsis REF6 is a histone H3 lysine
27 demethylase. Nature Publishing Group 2011, 43:715’719.
24. Us-Camas R, Rivera-Sol��s G, Duarte-Ak�� F, De-la-Pe��a C: In vitro culture: an
epigenetic challenge for plants. Plant Cell Tiss Organ Cult 2014, 118:187’201.
25. Lafos M, Kroll P, Hohenstatt ML, Thorpe FL, Clarenz O, Schubert D: Dynamic Regulation of H3K27 Trimethylation during Arabidopsis Differentiation. PLoS Genet 2011, 7:e1002040’20.
26. Labb��, R. M., Holowatyj, A., & Yang, Z. Q. (2014). Histone lysine demethylase (KDM) subfamily 4: structures, functions and therapeutic potential. American journal of translational research, 6(1), 1.
27. Sugimoto K, Meyerowitz EM: Regeneration in Arabidopsis Tissue Culture. In Arabidopsis Protocols. Humana Press; 2012:265’275.
28. Li D, Dreher K, Knee E, Brkljacic J, Grotewold E, Berardini TZ, Lamesch P, Garcia- Hernandez M, Reiser L, Huala E: Arabidopsis Database and Stock Resources. In Arabidopsis Protocols. Humana Press; 2013:65’96.
32
29. Sigma.(2005): Spectrum Plant Total RNA Kit. Available: http://www.sigmaaldrich.com/life-science/molecular-biology/plant- biotechnology/plant-molecular-biology/product-highlights/spectrum-plant-total- rna-kit.html. Last accessed 31st Aug, 2015.
30. Roche. (2011). LightCycler�� 480 SYBR Green I Master Mix. Available:https://lifescience.roche.com/shop/products/lightcycler14301-480- sybr-green-i-master. Last accessed 31st Aug 2015.
31. Herrmann MG: Amplicon DNA Melting Analysis for Mutation Scanning and Genotyping: Cross-Platform Comparison of Instruments and Dyes. Clinical Chemistry 2006, 52:494’503.
32. Montgomery J, Wittwer CT, Palais R, Zhou L: Simultaneous mutation scanning and genotyping by high-resolution DNA melting analysis. Nat Protoc 2007, 2:59’66.
33. Bussmann LH, Schubert A, Manh TPV, De Andres L, Desbordes SC, Parra M, Zimmermann T, Rapino F, Rodriguez-Ubreva J, Ballestar E, et al.: A Robust and Highly Efficient Immune Cell Reprogramming System. Stem Cell 2009, 5:554′ 566.
34. Noh B: Divergent Roles of a Pair of Homologous Jumonji/Zinc-Finger-Class Transcription Factor Proteins in the Regulation of Arabidopsis Flowering Time. The Plant Cell 2004, 16:2601’2613.
35. Kleinmanns JA, Schubert D: Polycomb and Trithorax group protein-mediated control of stress responses in plants. Biological Chemistry 2015, 395:1’10.
36. Trojer P, Reinberg D: Histone Lysine Demethylases and Their Impact on Epigenetics. Cell 2006, 125:213’217.
37. Waki T, Miyashima S, Nakanishi M, Ikeda Y, Hashimoto T, Nakajima K: A GAL4- based targeted activation tagging system in Arabidopsis thaliana. Plant J 2012, 73:357’367.
38. Borghi L: Inducible Gene Expression Systems for Plants. In Arabidopsis Protocols. Humana Press; 2010:65’75.
39. Medvedev, S. P., Pokushalov, E. A., & Zakian, S. M. (2012): Epigenetics of pluripotent cells. Acta naturae, 4(4), 28.
40. Young MD, Willson TA, Wakefield MJ, Trounson E, Hilton DJ, Blewitt ME, Oshlack A, Majewski IJ: ChIP-seq analysis reveals distinct H3K27me3 profiles that correlate with transcriptional activity. Nucleic Acids Research 2011, 39:7415′ 7427.
41. Saleh A, Alvarez-Venegas R, Avramova Z: An efficient chromatin immunoprecipitation (ChIP) protocol for studying histone modifications in Arabidopsis plants. Nat Protoc 2008, 3:1018’1025.