ORIGINAL RESEARCH
published: 24 August 2016
doi: 10.3389/fpls.2016.01284
Edited by:
Shabir Hussain Wani,
Sher-e-Kashmir University
of Agricultural Sciences
and Technology of Kashmir, India
Reviewed by:
Biswapriya Biswavas Misra,
University of Florida, USA
Guotian Li,
University of California, Davis, USA
*Correspondence:
P. B. Kirti
pbkirti@uohyd.ac.in
Specialty section:
This article was submitted to
Plant Biotechnology,
a section of the journal
Frontiers in Plant Science
Received: 09 June 2016
Accepted: 11 August 2016
Published: 24 August 2016
Citation:
Moin M, Bakshi A, Saha A, Dutta M,
Madhav SM and Kirti PB (2016) Rice
Ribosomal Protein Large Subunit
Genes and Their Spatio-temporal
and Stress Regulation.
Front. Plant Sci. 7:1284.
doi: 10.3389/fpls.2016.01284
Rice Ribosomal Protein Large
Subunit Genes and Their
Spatio-temporal and Stress
Regulation
Mazahar Moin
1
, Achala Bakshi
1
, Anusree Saha
1
, Mouboni Dutta
1
, Sheshu M. Madhav
2
and P. B. Kirti
1
*
1
Department of Plant Sciences, University of Hyderabad, Hyderabad, India,
2
Department of Biotechnology, Indian Institute
of Rice Research, Hyderabad, India
Ribosomal proteins (RPs) are well-known for their role in mediating protein synthesis
and maintaining the stability of the ribosomal complex, which includes small and
large subunits. In the present investigation, in a genome-wide survey, we predicted
that the large subunit of rice ribosomes is encoded by at least 123 genes including
individual gene copies, distributed throughout the 12 chromosomes. We selected 34
candidate genes, each having 2–3 identical copies, for a detailed characterization of
their gene structures, protein properties, cis-regulatory elements and comprehensive
expression analysis. RPL proteins appear to be involved in interactions with other
RP and non-RP proteins and their encoded RNAs have a higher content of alpha-
helices in their predicted secondary structures. The majority of RPs have binding sites
for metal and non-metal ligands. Native expression profiling of 34 ribosomal protein
large (RPL) subunit genes in tissues covering the major stages of rice growth shows
that they are predominantly expressed in vegetative tissues and seedlings followed by
meiotically active tissues like flowers. The putative promoter regions of these genes
also carry cis-elements that respond specifically to stress and signaling molecules. All
the 34 genes responded differentially to the abiotic stress treatments. Phytohormone
and cold treatments induced significant up-regulation of several RPL genes, while
heat and H
2
O
2
treatments down-regulated a majority of them. Furthermore, infection
with a bacterial pathogen, Xanthomonas oryzae, which causes leaf blight also induced
the expression of 80% of the RPL genes in leaves. Although the expression of RPL
genes was detected in all the tissues studied, they are highly responsive to stress
and signaling molecules indicating that their encoded proteins appear to have roles
in stress amelioration besides house-keeping. This shows that the RPL gene family is
a valuable resource for manipulation of stress tolerance in rice and other crops, which
may be achieved by overexpressing and raising independent transgenic plants carrying
the genes that became up-regulated significantly and instantaneously.
Keywords: ribosomal proteins, abiotic stress, biotic stress, gene expression, rice
Abbreviations: H
2
O
2
, hydrogen peroxide; MeJa, methyl jasmonate; RP, ribosomal protein; RPL, ribosomal protein large
subunit; SA, salicylic acid.
Frontiers in Plant Science | www.frontiersin.org 1 August 2016 | Volume 7 | Article 1284
Moin et al. RPL Genes Are Developmental and Stress-Responsive
INTRODUCTION
Ribosomes are tiny (200–300Å) ribonucleoprotein complexes
typically existing as two unequal sized subunits in all organisms
and constituting 25–30% of total cell mass (
Alberts et al.,
2002). The ribosome complex, as a whole, performs mRNA-
directed protein synthesis. Specific interaction of RPs and rRNA
with mRNA, tRNA, and other non-ribosomal protein cofactors
ensure that the process of initiation of protein synthesis, amino
acid assembly and termination occurs appropriately in the cells
(
Maguire and Zimmermann, 2001). Eukaryotic ribosomes have a
sedimentation coefficient of 80S with the large 60S subunit having
25S, 5.8S, 5S rRNA, and the small 40S subunit consisting of 18S
rRNA (Ben-Shem et al., 2010). The number of RPs in ribosomes
varies between organisms, with eukaryotes having up to 80 RPs
and prokaryotes possess a total of only 54 RPs in both the subunits
(Doudna and Rath, 2002).
The ribosomal gene family has more than 200 genes, but less
than 100 corresponding RPs are incorporated into the ribosomes
in all organisms including yeast, animals, and plants (
Ban et al.,
2000
; Barakat et al., 2001; Hanson et al., 2004). This supports
the fact that each RP-gene exists as 2–5 identical members with
95–100% nucleotide and predicted protein similarity. An RP
synthesized from only one gene copy of a group incorporates into
a ribosome under a given condition/tissue (Guarinos et al., 2003;
Schuwirth et al., 2005). For example, the Arabidopsis genome
has 249 genes for 80 RPs (48-large subunit proteins, 32-small
subunit proteins) with each gene having 3–4 expressed copies
and none exists as a single gene copy (Wool et al., 1996). RPs,
in addition to their universal roles of stabilizing the ribosomal
complex and mediating polypeptide synthesis also have extra-
ribosomal functions such as their involvement in response to the
environmental stresses (
Warner and McIntosh, 2009; Sormani
et al., 2011
).
Mutations in plant RP genes have been implicated in
perturbed phenotypes as has been seen in animal systems
including humans. Earlier studies with Arabidopsis showed
that mutations in many RP genes (RPS18A, RPL24B, RPS5B,
RPS13B, and RPL27A) resulted in a ‘pointed first leaf’ phenotype
characterized by reduced cell division and growth, and genotoxic
sensitive plants (
Lijsebettens et al., 1994; Revenkova et al.,
1999; Ito et al., 2000; Szakonyi and Byrne, 2011). A T-DNA
insertion mutation in the Arabidopsis AtRPL10 gene caused
lethal female gametophytes, while overexpression complemented
the same with the re covery of the severe dwarf phenotype
that resulted from the disruption of the ACL5 gene (Imai
et al., 2008). A transposon insertion mutation in one of
the three copies of the AtRPS13A gene resulted in reduced
cell division, late flowering, retarded root and leaf growth
(Ito et al., 2000). Similar effects of plant growt h retardation
and reduced fertility were observed after knockdown of
AtRPL23aA resulting in reduced synthesis of the RPL23aA
protein, while knockout of its paralog, RPL23aB, had no effect
on growth (
Degenhardt and Bonham-Smith, 2008a). RPL23aB
is the only RP paralog that did not produce any visible
phenotypic defects upon knockout (
Degenhardt and Bonham-
Smith, 2008b).
These RP-gene knockout studies clearly show that although
RP genes exist as multiple gene copies, the maximum possible
expression of all the gene copies is required for them to be
incorporated into the ribosomes during spe cific stages of growth
and development and under certain stress conditions (
Schmid
et al., 2005; Byrne, 2009). The variation in the composition of
ribosomes by the incorporation of RPs derived from identical
members could be a major factor in the translational regulation
of transcripts in different cell types and under various specific
conditions (
Giavalisco et al., 2005; Carroll et al., 2008). The
change in the composition of RPs upon feeding of Arabidopsis
leaves with sucrose further supports the heterogeneity of
ribosomes in response to external stimuli (
Hummel et al.,
2012).
The expression of RP genes has also been shown to be
differentially regulated by signaling molecules and environmental
stresses. The transcript levels of Arabidopsis RPS15a (RPS15aA,
C, D and F) were up-regulated in response to phytohormone
and heat treatments (
Hulm et al., 2005). Similar transcript
abundance under BAP treatment was detected for Arabidopsis
RPS14, RPL13, and RPL30 genes (Cherepneva et al., 2003). Low
temperature induced the expression of three RP genes; RPS6,
RPS13, and RPL37 in soybean (Kim et al., 2004) and a homolog
of RPL13, BnC24 in Brassica and E. coli (Sáez-Vásquez et al.,
2000; Tanaka et al., 2001). The overexpression of RPL13 also
resulted in tolerance against a fungal pathogen, Verticillium
dahliae in transgenic pot ato with coordinated up-regulation of
genes coding for defense and antioxidant enzymes (Yang et al.,
2013) implying that RPs function in stress-response/tolerance
through a network of multiple stress-related genes. In maize and
Arabidopsis, RPL10A and RPL10C were shown to be significantly
up-regulated under UV-B stress (
Casati and Virginia, 2003;
Ferreyra et al., 2010, 2013). RPL44 was found to be up-regulated
under osmotic stresses, and the overexpression of Aspergillus
glaucus RPL44 in yeast and tobacco ensured increased tolerance
to salt and drought stresses (Liu et al., 2014). The majority of
studies on RPs were undertaken in Arabidopsis largely bec ause
of the availability of insertion mutant lines and smaller genome
size.
Until now, not much emphasis has been placed on the
differential expression patterns of RP genes of rice in response to
external stimuli. We had generated a large-scale enhancer based
activation-tagged gain-of-function mutant population in indica
rice, which was screened for water-use efficiency. Among the
potential mutants with sustained productivity under prolonged
water-limiting conditions, two of them were found to have
enhanced expression of large subunit ribosomal genes because of
their being t agged by the enhancers (
Moin et al., 2016). This has
prompted us to investigate the other rice RPL genes in the context
of stress-responsiveness.
In the present study, we describe the genome-wide
organization of predicted 123 RPL genes in rice including
the individual gene copies. We investigated their overall
expression pattern in selected tissues covering the major growth
stages of rice. Also, we have provided an overview of their
differential expression pattern under biotic and abiotic stress
conditions that limit rice productivity. We identified specific RP
Frontiers in Plant Science | www.frontiersin.org 2 August 2016 | Volume 7 | Article 1284
Moin et al. RPL Genes Are Developmental and Stress-Responsive
genes, whose expression is unique or overlapping under native
and treated conditions. In summary, the information presented
in t his study provides a resource for subsequent exploitation of
RPL genes to ameliorate abiotic and biotic stress conditions in
rice and also other crop plants in future.
MATERIALS AND METHODS
Nucleotide Sequence Retrieval of RPL
Genes
To identify the total members of the large subunit ribosomal
gene family, a keyword search using “ribosomal” was performed
under the putative function search of Rice Genome Annotation
Project Data Base (RGAP-DB v7)
1
and Phytozome v11
2
. The
large subunit members were shortlisted by selecting the genes
starting with prefix ‘L,’ for large subunit as opposed to ‘S’ that
specifies small subunit genes. A total of 123 RPL genes were
identified, and since the number of RPL genes in both the
databases was same, the gene sequences were downloaded from
RGAP-DB. When further looked for the presence of identical
members or copies of each gene in RGAP-DB, we observed
that each RPL gene has an average of 2–3 gene copies in the
genome. From these 123 genes, we selected 34 candid ate genes
each representing one orthologous group excluding the identical
copies for expression studies. All the identified 123 sequences
were also confirmed through nucleotide and protein BLAST
search in the NCBI
3
and Hidden Markov Model (HMM) of
Pfam
4
databases, respectively. The predicted protein sequences
of all the 123 RPL genes were also verified in NCBI conserved
domain database
5
. To minimize the missing of potential RPL
genes and to ensure that all the identified sequences belong to
the ribosomal large subunit gene family, multiple dat abases were
employed.
Chromosomal Distribution of RPL Genes
To determine the chromosomal distribution, the locus number
of each of the 123 RPL genes obtained from RGAP-DB was
submitted to the OryGenesDB
6
. Based on the output generated
in OryGenesDB, the position of each gene at its corresponding
locus on the chromosome was located manually.
RPL Gene Structures
The structure of each of the 34 RPL genes was determined to
study the number and position of introns and exons, GC-content,
gene orientation in the genome and alternative splice forms. The
full-length sequences of each gene and cDNA were submitted
to t he Gene Structure Display Server (GSDSv2)
7
to predict the
structure.
1
http://rice.plantbiology.msu.edu/index.shtml
2
https://phytozome.jgi.doe.gov/pz/portal.html
3
http://blast.ncbi.nlm.nih.gov/Blast.cgi
4
http://pfam.xfam.org/
5
http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi
6
http://orygenesdb.cirad.fr/tools.html
7
http://gsds.cbi.pku.edu.cn/
Protein Properties, Secondary Structure,
Homology Modeling, and Phylogenetic
Analysis
The predicted sequences of 34 RPL proteins were obtained from
the RGAP-DB and analyzed using an online tool, PSORT
8
to
predict the protein properties such as size, molecular weight
and isoelectric point (pI). The amino acid sequences of these
proteins were aligned in ClustalW
9
and submitted to the
Molecular Evolutionary Genetic Analysis (MEGAv6)
10
program
for constructing an unrooted phylogenetic tree to identify the
protein similarities in the RPL family in rice. The domains
and motifs in proteins were identified using SMART
11
(Simple
Modular Architecture Research Tool). The GRAVY (Grand
average of hydropathicity) indices of RPL proteins, which are
the determinants of the hydrophobicity of whole protein was
calculated using ExPASy ProtParam
12
. The GRAVY values of
most of the proteins are usually in the range of +2 to −2,
and values in negative range or less than zero indic ate that the
proteins are hydrophilic in nature (
Song et al., 2015).
Although the detailed crystal structure of ribosomal complex
has been well-characterized (
Ben-Shem et al., 2010), we tried
to study the properties of individual RPs. To gain an insight
into the secondary structure of RPL proteins and to characterize
the presence of metal–ligand/protein/RNA interacting sites, the
three-dimensional secondary structures of 34 RPL proteins were
predicted using Phyre2
13
program (Protein Homology/AnalogY
Recognition Engine v2;
Kelley et al., 2015). Individual protein
sequences were submitted in Phyre2 in FASTA format and after
studying the properties such as α-helices and β-strands, they
were directed to 3DLigandSite
14
(
Wass et al., 2010) to predict the
metal/non-metal ligands and their binding sites in each protein.
In silico Putative Promoter Analysis of 34
RPL Genes
To determine the presence of stress-responsive cis-regulatory
elements, the nucleotide sequence ≤1 kb upstream of each RPL
gene was retrieved from RGAP-DB and submitted to the Plant
Cis-Acting Re gulatory Elements
15
database. The location and
number of repeats of each cis-regulatory sequence in the putative
promoter regions of each RPL gene were identified.
Plant Material and Growth Conditions
The seeds of Oryza sativa L. sp. indica var. Samba Mahsuri
(BPT-5204) maintained in greenhouse conditions were surface
sterilized with 70% ethanol for 50–60 s followed by 4% sodium
hypochlorite for 20 min. Seeds were then washed thrice with
sterile double-distilled water, blot dried and cultured on solid MS
medium at 28 ± 2
◦
C and 16 h light/8 h dark photoperiods.
8
http://psort.hgc.jp/form.html
9
http://www.genome.jp/tools/clustalw/
10
http://www.megasoftware.net/
11
http://smart.embl-heidelberg.de/
12
http://web.expasy.org/cgi-bin/protparam/protparam
13
http://www.sbg.bio.ic.ac.uk/phyre2/html/page.cgi?id=index
14
http://www.sbg.bio.ic.ac.uk/3dligandsite/
15
http://bioinformatics.psb.ugent.be/webtools/plantcare/html/
Frontiers in Plant Science | www.frontiersin.org 3 August 2016 | Volume 7 | Article 1284
Moin et al. RPL Genes Are Developmental and Stress-Responsive
To analyze the native tissue-specific expression pattern of 34
RPL genes at different stages of rice development, samples were
collected from 13 different tissues covering major stages of rice
growth. After sterilization, seeds were soaked in water in a rotary
shaker and after 16 h of incubation, the embryonic portion was
manually cut under a stereo-microscope to collect the embryos
and endosperm. Some of these seeds were allowed to continue to
germinate on MS medium. After 3 and 6 days of germination,
the plumules, radicles, shoot, and leaf tissues were collected
separately. After 2 weeks of growth on MS medium, some of
the seedlings were transferred to pots containing alluvial soil and
grown under greenhouse conditions (30 ± 2
◦
C, 16 h light/8 h
dark photoperiods). Plants were amply watered with RO (Reverse
Osmosis) purified water up to 3 cm overlay in the pots as required
for normal growth of rice. About 45 days after transfer (DAT) to
the greenhouse, rice plants were uprooted to collect shoot, root,
flag leaf, and root–shoot transition tissues. After 60 DAT, flowers,
partially filled grains and spikes were collected.
Abiotic Stress Treatments to Seedlings
To analyze t he differential expression pattern of 34 RPL genes
and to distinguish RPL genes that are up/down-regulated under
abiotic conditions, 7-day-old seedlings were exposed to five
different abiotic treatments such as MeJa, SA, cold stress (4
◦
C),
heat stress (42
◦
C), and oxidative stress (H
2
O
2
). The 7-d-old
seedlings were dipped in the solutions of 100 µM MeJa (
Wang
et al., 2007
), 3 mM SA (Mitsuhara et al., 2008), and 10 µM
H
2
O
2
(Fuller, 2007). The root and shoot tissues were collected
separately at 5 min, 3 h, 6 h, 24 h, and 60 h after treatments. For
cold and heat induced stresses, seedlings in water were exposed to
4
◦
C and 42
◦
C (Jami et al., 2012), respectively, and root and shoot
samples were collected at time intervals as described. Since WT
rice seedlings started to wilt after 24 h of exposure to 42
◦
C, heat
stress samples were collected up to 24 h only. Seedlings in water
at corresponding time intervals served as controls to normalize
the expression patterns. Tissue samples were collected as three
biological replicates after each time interval.
Biotic Stress Treatment
To check the expression pattern of rice ribosomal genes in
response to biotic stress, we used the bacterial pathogen
Xanthomonas oryzae pv. oryzae that causes Bacterial Le af Blight
(BLB) of rice, which is one of the most severe yield constraints
of rice worldwide (
Sundaram et al., 2014). At the seedling
stage, the infected leaves start to roll-up, and as the disease
progresses, the leaves turn yellow and wilt, leading to drying
up and death. This drastically reduces the total seed yield of
the plant. The yield loss may be as high as 70% when plants
are grown in conditions favorable to the disease (
Ryan et al.,
2011). The bacterial suspension of Xanthomonas oryzae pv.
oryzae was applied on t he leaves of 2-month-old plants grown
in greenhouse conditions, and leaf samples were colle cted after
11 days of infection. Leaf samples of untreated plants grown
under similar conditions were used as a control to normalize the
expression.
Because the transcript level of RPL10 was significantly up-
regulated 11 days after treatment, we selected this gene in
particular to analyze its expression at progressive time-points
such as 3 h, 6 h, 1 day, 2 days, 3 days, up to 11 days post-infection
of rice leaves with Xanthomonas oryzae pv. oryzae pathogen. The
qRT-PCR was performed with Xanthomonas oryzae pv. oryzae
treated and untreated samples collected as three biological and
three technical replicates. Rice specific act1 and β-tub genes
were used as controls for normalization and the mean of the
fold change was represented as bar diagrams constructed using
SigmaPlot v11.
RNA Isolation, cDNA Synthesis, and
Quantitative-PCR (qRT-PCR)
Total RNA was isolated from stress-treated and untreated tissues
using TriReagent (Takara Bio, UK) following the manufacturer’s
protocol. The quality of extracted RNA was checked on 1.2%
agarose gel prepared in TBE (Tris-borate-EDTA) buffer and
quantified using Nanodrop. Total RNA (2 µg) was used to
synthesize the first strand cDNA using reverse transcriptase
(Takara Bio, UK). The cDNA was diluted in 1:7 proportions
and 2 µl of it was used in qRT-PCR. Primers specific for each
RPL gene sequence retrieved from RGAP-DB was designed using
the primer-3
16
online tool and 10 µM of each was used per
reaction. The reaction conditions for qRT-PCR included an
initial denaturation at 94
◦
C for 2 min followed by 40 cycles
of 94
◦
C for 30 s, an appropriate annealing temperature for
25 s and an extension of 72
◦
C for 30 s. At the end of the
reaction, a melting curve step was inserted to analyze the
specificity of amplification of each gene. Rice specific actin (act1)
and tubulin (β-tub) were used as internal reference genes to
normalize the expression patterns. The mean values of relative
fold change, which was calculated as per 11C
T
method (
Livak
and Schmittgen, 2001) obtained from e ach reference gene was
considered as t he final fold change in the transcript levels. Each
qRT-PCR reaction was performed as three biologic al and three
technical replicates.
The relative fold change of the 34 genes in 13 tissues and
under five abiotic treatments was illustrated in the form of
heat maps. A dendrogram was constructed to represent the
Hierarchical clustering of relative fold change of 34 genes under
each treatment using the GENE-E
17
program.
RESULTS
Genome-Wide Identification and
Chromosomal Distribution of RPL Genes
To explore the cytoplasmic large subunit (60S) ribosomal gene
family members in rice, we used a keyword search “ribosomal”
in the putative function search of RGAP-DB and Phytozome
databases, which resulted in the identification of 428 and
754 genes, respectively, and these included genes belonging to
cytoplasmic 60S and 40S subunits and 50S and 30S subunits of
chloroplast and mitochondrial ribosomes. Keyword search and
homology-based identification through HMM are widely used
16
http://bioinfo.ut.ee/primer3-0.4.0/
17
http://www.broadinstitute.org/cancer/software/GENE-E/
Frontiers in Plant Science | www.frontiersin.org 4 August 2016 | Volume 7 | Article 1284
Moin et al. RPL Genes Are Developmental and Stress-Responsive
practices in identifying genome-wide copies of the annotated
genes (Kapoor et al., 2008; Liang et al., 2016). We then searched
for genes starting with the prefix ‘L’ to select large subunit genes.
This process excluded small subunit genes and identified 215
genes that included large subunit members of cytoplasmic (60S)
and chloroplast and mitochondrial (50S) ribosomal subunits.
We then shortlisted the cytoplasmic 60S subunit genes by their
putative cellular localization using the information available in
RGAP-DB. A similar process was applied in shortlisting the
60S subunit genes from Phytozome. Both these approaches
identified 123 genes belonging to the cytoplasmic 60S subunit.
Each of these genes was then confirmed by a BLAST search of
their nucleotide and predicted amino acid sequences in other
rice databases like RAP-DB
18
and OryGenesDB. BLASTn and
BLASTp results in NCBI and HMM of Pfam and NCBI conserved
domain dat abases, respectively, further confirmed that these
18
http://rapdb.dna.affrc.go.jp/
genes belong to the 60S ribosomal family by the presence of
ribosomal domains.
The locus numbers of 123 genes were submitted in
OryGenesDB and, based on the output generated, the location
of each gene on the corresponding chromosome was mapped
manually using OryGenesDB. The location of these 123
genes was found on all chromosomes, indicating their wide
distribution throughout the rice genome. Chromosome-7 has 19;
chromosome-1, being the largest of rice chromosomes has 18;
chromosome-2 has 16; chromosome-5 has 14; and chromosome-
3 showed 13 genes. Chromosomes-9, 10, and 11 exhibited
four genes each, while chromosomes-4, 8, and 9 evidenced 5,
10, and 9 RPL genes, respectively (Figure 1). The nucleotide
sequence alignment of genes within an orthologous group
exhibited 100% similarity, but their chromosomal locations are
different. We selected the 34 candidate genes, each representing
one orthologous group for a detailed characterization to
understand their gene and protein structures, and comprehensive
FIGURE 1 | Chromosomal organization of RPL genes. The chromosomal number and size is represented at the top and bottom of each chromosome,
respectively. The number of RPL genes is given within brackets at the bottom of each corresponding chromosome.
Frontiers in Plant Science | www.frontiersin.org 5 August 2016 | Volume 7 | Article 1284