º THE TRANSCRIPTOME OF QUIESCENCE AND DORMANCY IN SUBTROPICAL AND MEDITERRANEAN GRAPEVINE Presented by: Sandra Patricia Agudelo Romero, PhD. MASTER OF SCIENCE IN BIOINFORMATICS AND COMPUTATIONAL BIOLOGY NATIONAL HEALTH RESEARCH INSTITUTE INSTITUTE OF HEALTH CARLOS III (ISCIII) 2014-2015 UNIVERSITY OF WESTERN AUSTRALIA Professor Dr. Michael Considine 2nd of February of 2015
37
Embed
THE TRANSCRIPTOME OF QUIESCENCE AND DORMANCY IN ... · the transcriptome of quiescence and dormancy in subtropical and mediterranean grapevine presented by: sandra patricia agudelo
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
º
THE TRANSCRIPTOME OF
QUIESCENCE AND DORMANCY IN
SUBTROPICAL AND MEDITERRANEAN
GRAPEVINE
Presented by:
Sandra Patricia Agudelo Romero, PhD.
MASTER OF SCIENCE IN BIOINFORMATICS
AND COMPUTATIONAL BIOLOGY
NATIONAL HEALTH RESEARCH INSTITUTE
INSTITUTE OF HEALTH CARLOS III (ISCIII)
2014-2015
UNIVERSITY OF WESTERN AUSTRALIA
Professor Dr. Michael Considine
2nd of February of 2015
º
i
CONTENTS
DEDICATION ii
ACKNOWLEDGEMENT iii
1. ABSTRACT 1
2. THE APPROACH TO THE PROBLEM 2
3. OBJECTIVES 3
4. INTRODUCTION 4
5. MATERIALS AND METHODS 6
5.1. Sample collection 6
5.2. RNA extraction, Illumina library construction and sequencing 6
5.3. Data processing analysis 6
5.4. Functional enrichment analysis 7
6. RESULTS AND DISCUSSION 8
6.1. Dormancy and Quiescent Gene Expression Profiles 8
6.2. Transcriptional Bases for Bud Dormancy and Quiescent
Differentiation Between Climates. 12
6.2.1. Pre-chilling 12
6.2.2. Post-chilling 18
6.3. Finding Potential Biomarkers 22
7. CONCLUSIONS 26
8. BIBLIOGRAPHY 27
9. ANEXOS 33
9.1. FastQC command line 33
9.2. Trimmomatic command line 33
9.3. Kallisto on hiseq command line 33
º
ii
DEDICATION
For the two great loves of my life,
thank you so much for existing.
º
iii
ACKNOWLEDGEMENT
I was supported by an Australian Research Council grant (ARC: LP130100347). The
research was supported by the ARC grant: LP0990355. This project was carried out at
the University of Western Australia in the ARC Center of Excellence in Plant Energy
Biology.
º
1
1. ABSTRACT
Vine physiology is dependent on climate for orderly transitions between vegetative and
reproductive growth. Productive viticulture requires a temperate/Mediterranean climate,
while in warmer, low latitude climates. Low latitude viticulture is only viable for table
grapes if development is intensively managed with chemical and physical stress
treatments, such as deficit irrigation to force a ‘rest’ period and cyanamide to force bud
burst. Even with intensive management, vine growth is disorderly and yields are
considerably lower and more variable, as the seasonal cues that grapevine relies on for
developmental transitions are lacking. To gain insight into how differences in the
temperature due to climate features can modify grapevine bud dormancy, a RNA-seq
study was performed to investigate differences between subtropical and Mediterranean
climates in table grapes (Flame Seedless).
For this, gene expression changes in buds from two adjacent vineyards in subtropical
Western Australia (25°S latitude) were compared against one vineyard in a
Mediterranean climate (32°S). Buds were collected for differential expression analysis
at the end of summer (March; henceforth termed pre-chilling) and in mid-winter (June
henceforth termed post-chilling), over two successive years (2012 and 2013). Principal
Components Analysis (PCA) of RNA-seq data revealed that the main factor explaining
the global gene expression differences was between consecutive years.
Differential expression analyzes of subtropical and Mediterranean climates comparison
in pre-chilling and post-chilling conditions were carried out (1% FDR and FC |3-fold|).
Cluster and functional enrichment analyzes were then performed to each condition. In
the comparison performed during pre-chilling, WRKY family transcription and
oxidative stress (Glutathione S-transferase) categories showed differences between
climates. Whereas in the post-chilling condition was detected ethylene-mediated
signaling pathway and C2C2-YABBY family transcription factor categories.
This work provides a global view of major transcriptional changes taking place in
Australian subtropical and Mediterranean climates, highlighting those molecular and
biological functions that showed differences between climates, suggesting a main role
of those functional categories during regulation of bud dormancy.
º
2
2. THE APPROACH TO THE PROBLEM
Grapevine cultivation in subtropical climates is a fragile system requiring intensive
management, for which there are no biological markers. Miss-timing intervention can
result in phytotoxic effects, as in the case of the dormancy-releasing hydrogen
cyanamide. In temperate-grown grapevine, the reproductive and metabolic cycles are
regulated by environmental signals, particularly the induction of dormancy during
autumn and the subsequent re-activation of growth during spring (Lavee et al., 1997).
Like many woody perennials, energy reserves accumulate in the perennial tissues prior
to the onset of dormancy and their mobilisation entirely supports the initial stages of
vegetative and reproductive growth in spring (Lebon et al., 2005). Hence, the dormant
phase is necessary for coordinated, productive and sustainable growth (Lavee et al.,
1997). In grapevine, molecular investigations of dormant axillary buds have also
revealed coordinated profiles, including reprogramming of carbohydrate metabolism,
but these have been under temperate conditions and confined to dormancy release (bud
break), a single event in a complex reproductive cycle (Mathiason et al., 2008). Here,
the reprogramming of buds during dormancy and quiescent (latent) stages can be
studied using RNAseq approach. For this, buds sampled from a subtropical Western
Australian climate (Carnarvon) were compared to buds from a Mediterranean climate
(Swan Valley, Perth) (Figure 1).
Figure 1. Map of Australian grape growing regions and temperature differences. A. Subtropical Western
Australian climate (Carnarvon) and Mediterranean climate (Swan Valley, Perth) are highlight.
Subtropical climate is represented by Bumbak (B) and Condo (C) sites from Carnarvon. Mediterranean
climate is represented by Nuich (N) from Swan Valley (Perth). B. Graphics showed the average of the
differences in temperature of both climates provided by the airports (Carnarvon and Perth respectively) in
the last decades.
º
3
3. OBJECTIVES
The aim of this project is to determine the relationship between climates in grapevine
buds grown in a subtropical region (Carnarvon, WA) and a Mediterranean climate
(Swan Valley) in Australia during two consecutive seasons (2012 and 2013) by using of
RNAseq technology.
Specifically:
To dissect the effects and interactions between climate conditions through their
gene expression profiles.
To finding potential gene candidates in order to be used as biomarkers for
anticipate actions in warm-temperate or stressed conditions (i.e. water stress).
º
4
4. INTRODUCTION
Grapes (Vitis spp.) are economically the most important fruit crop worldwide with a
global production of around 67 million t in 2012 (Food and Agriculture Organization
Corporate Statistical Database (FAOSTAT, 2014,
http://faostat.fao.org/site/567/default.aspx#ancor). Table grapes rape and processed
products such as: wine, juice, jam and dried fruit, represent an important sector in the
global market. Moreover, their consumption has an important added value in a healthy
diet by the polyphenolic compounds with antioxidant and anticarcinogenic properties
found in them (Ali et al., 2010).
Therefore, it is important to understand how improve the production of grapes in
Australian subtropical climate since theoretically, this region lacks sufficient
temperature to regulate dormancy correctly. This fact is a critical environmental
requirement for sustainable table grape production of cultivars as Flame Seedless.
Dormancy induction is problematic and the normal reproductive cycle is perturbed, with
a shifted and condensed phenological cycle. It is only management intervention that
sustains this cycle; vines would otherwise continue vegetative growth throughout
winter, limiting resource storage and resulting in variable yields and very short vine life
(Possingham 2004). In subtropical climates, sustainable production thus relies on
management intervention to supplement for environmental signals; e.g. water stress and
chemical application (concentrated nitrate) to force leaf fall, imposing a winter “rest,”
and pruning/ chemical application (hydrogen cyanamide) to stimulate vines to
recommence vegetative and reproductive growth. Even with interventions, disorders are
common, as described in the predominant subtropical viticulture regions; Carnarvon
(WA) and Rockhampton (Qld) in Australia, Coachella Valley (California), Mexico,
Northern Chile and Orange River (South Africa):
• Slow and erratic bud burst and extreme dominance of the apical buds.
• Delayed foliation following bud burst, often characterised by a period of
chlorotic growth.
• Inflorescence disorders, resulting in partial or complete abortion or abscission.
º
5
Each of these disorders may co-occur. The result is an industry faced with high
prospects but equally high input costs and yield and quality variation. To date, R&D
into production of temperate fruit crops in subtropical or tropical climates has focussed
heavily on improving bud burst, principally through optimal use of chemicals or
through manipulating the microclimate during autumn (Possingham 2004). There is a
distinct lack of research on the stages of bud development preceding winter, so-called
dormancy onset.
In this project, to gain insight into how differences in the temperature due to climate
features can modify grapevine bud dormancy, a RNA-seq study was performed to
investigate differences between subtropical and Mediterranean climates in table grapes
(Flame Seedless). Differential expression in pre-chilling and post-chilling conditions
were carried out (1% FDR and FC |3-fold|) accompanied of a cluster and enrichment
analysis.
º
6
5. MATERIALS AND METHODS
5.1. Sample collection
Field collection in Western Australia (WA) of grape table, Flame seedless, was done at
two sub-tropical vineyards (B and C) in Carnarvon (24.9ºS and 113.7ºE) and a
Mediterranean vineyard in the Swan Valley (31.8ºS and 116ºE). Sampling was
performed between the end of March and early April for the pre-chilling condition and
in the middle of June for the post-chilling condition. Every sample was composed of
two buds randomly selected from healthy canes, they were immediately frozen in dry
ice and stored at -80ºC. At the time of sampling, buds were cut from the cane with a
scalpel, visually assessed for indications of necrosis from beneath the bud; buds with
obvious visible signs of necrosis within were discarded, however this assessment cannot
determine less dramatic levels of necrosis within the bud. For each time point and place
three/four biological replicates of buds were sampled for the RNA-seq analyses.
5.2. RNA extraction, Illumina library construction and sequencing
Buds were ground under liquid nitrogen to a fine powder. Total RNA extraction was
performed using the Spectrum Plant Total RNA kit with an on-column DNase treatment
according to the supplier’s instructions (Sigma-Aldrich, Castle Hill, Australia),
followed by an isopropanol/acetate precipitation. The quality and integrity of the
isolated RNA was tested using a NanoDrop 100 spectrophotometer and agarose gel
electrophoresis. Only RNA with an Abs260 nm/Abs280 ratio above 1.95 was used
further. RNA-seq libraries were prepared with the TruSeq Stranded Total RNA with
Ribo-Zero Plant kit according to manufacturer's instructions (Illumina, Scoresby,
Australia). Sequencing was performed on an Illumina HiSeq1500 instrument as 100bp
single-end runs.
5.3. Data processing analysis
Resulting reads were aligned to the whole 12X V1 Vitis vinifera PN40024 reference
genome (Jaillon et al., 2007) with Kallisto (Bray et al., 2015). Gene expression profiling
was carried out using edgeR (Robinson et al., 2010) and limma (Ritchie et al., 2015)
Bioconductor packages. The counts matrix obtained with Kallisto was read using
º
7
edgeR, it was converted into a DGEList data class and a TMM normalization method
was applied to obtained the log2 Counts-Per-Million (logCPM). Then, a Voom
transformation was applied to convert it into a EList object to can use limma pipelines
for differential expression under a linear model. To identify differentially expressed
genes, a multiple testing correction via False Discovery Rate (FDR) was performed
(P<=0.01) along with fold change (FC 3-fold). Principal component analysis (PCA) was
performed with the full TMM dataset using Acuity 4.0 (Axon Molecular Devices,
http://www.moleculardevices.com). FastQC, Trimmomatic and Kallisto command lines
used to generate the count matrix are detailed in Anexos.
5.4. Functional enrichment analysis
Gene lists were analysed further with FatiGO (Al-Shahrour et al., 2004) to identify
significant functional enrichment in Babelomics 5 (http://babelomics.bioinfo.cipf.es/)
following a grapevine-specific functional classification of 12X V1 predicted transcripts
(Grimplet et al., 2012). Fisher’s exact test was carried out in FatiGO to compare each
study list with the list of total non-redundant transcripts housed in the grapevine 12X
V1 gene predictions (Grimplet et al., 2012). Significant enrichment was considered for
P<0.01 after Benjamini and Hochberg correction for multiple testing.
Figure 2. PCA plot of ‘Flame Seedless’ bud samples according to their expression data in 2012 and 2013 seasons, buds were harvested from two adjacent vineyards in
subtropical Western Australia (25°S latitude) and one vineyard in a Mediterranean climate (32°S). A. PCA plot of buds samples according to their TMM normalized
expression data. The first (PC1), the second (PC2) and the third (PC3) principal components are represented. Each season is formed by two stages of dormancy (pre- and post-
chilling) with three or four replicates. Green, 2012 season and orange, 2013 season. B. Stage averaged PC3 loading scores. Color code is the same as in A. Lines represent
standard errors (SE). C. Functional categories over-represented in PC3 (B). Absolute values of log10 transformed P-values were used for the bar diagram representing
statistical signification, only categories with P-values < 0,01 are shown. Clear blue, Primary metabolism and dark blue, secondary metabolism.
º
10
Signaling pathway, respectively. Therefore these transcripts summarize the major
differences between years.
Functional enrichment analysis was carried out using FatiGO (Al-Shahrour et al, 2004)
to assess the biological significance of the transcripts showing a higher contribution to
the expression variability. The lists of transcripts most contributing to PC3 were
compared to the rest of the transcripts represented in the 12X V1 predicted transcripts
(Grimplet et al., 2012), highest scored (LS > |5|) transcripts were analyzed together.
This functional analysis discriminated just two functional categories: Primary
metabolism and Transport overview (Fig. 2C). Four processes associated to Primary
metabolism were found: ‘respiratory-chain phosphorylation’, ‘nucleic acid metabolism’,
‘protein processing in endoplasmic reticulum’ and ‘ribosome’. Two processes were
detected in Transport overview: ‘proton-translocating NADH Dehydrogenase’ and
‘proton-translocating Quinol:Cyt c Red’.
Among all processes, ribosome-related processes had the most highly significant adj. P
value (8,38E-11). The eukaryotic ribosome is a complex structure formed for four
rRNAs and about eighty ribosomal proteins. It represents a crucial piece of the cell
machinery, responsible for protein synthesis, and as such plays a major role in
controlling cell growth, division, and development. Several studies have reported that
genetic defects in ribosomal components can produce deleterious effects on the
development and physiology of drosophila, mice, humans and plants (Barakat et al.,
2001). On the other hand, it was also reported a positive correlation between the level of
r-protein gene transcript accumulation and cell division in suspension culture cells and
tissues such as auxin-treated hypocotyls, apical meristems, young leaves, and lateral
roots (Barakat et al., 2001). Here, two Ribosomal RNAs were detected: 23S
(VIT_01s0010g01260; VIT_11s0037g01180 and VIT_12s0035g02010) and 16S
(VIT_13s0101g00220) along with fifteen ribosomal proteins.
To dissect the influences of consecutive seasons, PCA analysis was performed
separately for each year. Each analysis represents two sites: Subtropical site (B -
Bumbak and C - Condo) from Carnarvon and Mediterranean climate (N - Nuich) from
Swan Valley, they represent subtropical and Mediterranean climates, respectively.
Additionally, two stages of grapevine bud development were compared, pre-chilling