The ankyrin repeat gene family in rice: genome-wide identification, classification and expression profiling.

Abstract

Ankyrin repeat (ANK) containing proteins comprise a large protein family. Although many members of this family have been implicated in plant growth, development and signal transduction, only a few ANK genes have been reported in rice. In this study, we analyzed the structures, phylogenetic relationship, genome localizations and expression profiles of 175 ankyrin repeat genes identified in rice (OsANK). Domain composition analysis suggested OsANK proteins can be classified into ten subfamilies. Chromosomal localizations of OsANK genes indicated nine segmental duplication events involving 17 genes and 65 OsANK genes were involved in tandem duplications. The expression profiles of 158 OsANK genes were analyzed in 24 tissues covering the whole life cycle of two rice genotypes, Minghui 63 and Zhenshan 97. Sixteen genes showed preferential expression in given tissues compared to all the other tissues in Minghui 63 and Zhenshan 97. Nine genes were preferentially expressed in stamen of 1 day before flowering, suggesting that these genes may play important roles in pollination and fertilization. Expression data of OsANK genes were also obtained with tissues of seedlings subjected to three phytohormone (NAA, GA3 and KT) and light/dark treatments. Eighteen genes showed differential expression with at least one phytohormone treatment while under light/dark treatments, 13 OsANK genes showed differential expression. Our data provided a very useful reference for cloning and functional analysis of members of this gene family in rice.

Full text

Page 1

The ankyrin repeat gene family in rice: genome-wide
identification, classification and expression profiling
Jianyan Huang Æ Æ Xiaobo Zhao Æ Æ Huihui Yu Æ Æ
Yidan Ouyang Æ Æ Lei Wang Æ Æ Qifa Zhang
Received: 15 March 2009/Accepted: 12 June 2009
? Springer Science+Business Media B.V. 2009
Abstract
comprise a large protein family. Although many members
of this family have been implicated in plant growth,
development and signal transduction, only a few ANK
genes have been reported in rice. In this study, we analyzed
the structures, phylogenetic relationship, genome local-
izations and expression profiles of 175 ankyrin repeat genes
identified in rice (OsANK). Domain composition analysis
suggested OsANK proteins can be classified into ten sub-
families. Chromosomal localizations of OsANK genes
indicated nine segmental duplication events involving 17
genes and 65 OsANK genes were involved in tandem
duplications. The expression profiles of 158 OsANK genes
were analyzed in 24 tissues covering the whole life cycle of
two rice genotypes, Minghui 63 and Zhenshan 97. Sixteen
genes showed preferential expression in given tissues
compared to all the other tissues in Minghui 63 and
Zhenshan 97. Nine genes were preferentially expressed in
stamen of 1 day before flowering, suggesting that these
genes may play important roles in pollination and fertil-
ization. Expression data of OsANK genes were also
obtained with tissues of seedlings subjected to three phy-
tohormone (NAA, GA3 and KT) and light/dark treatments.
Eighteen genes showed differential expression with at least
one phytohormone treatment while under light/dark
Ankyrin repeat (ANK) containing proteins
treatments, 13 OsANK genes showed differential expres-
sion. Our data provided a very useful reference for cloning
and functional analysis of members of this gene family in
rice.
Keywords
Microarray ? Expression profiles
Rice ? Ankyrin repeat ? OsANK gene family ?
Abbreviations
ANK
Os
At
ANK-M
TM
TPR
RF
BTB
Ankyrin repeat
Oryza sativa
Arabidopsis thaliana
Only contain ANK domain
Transmembrane
Tetratricopeptide repeat
Ring finger
Broad-complex, tramtrack and bric a brac
domains
Zinc-finger
BAR, PH and ArfGap domains
Calmodulin-binding domain
S_TKc domain or Pkinase-tyr domain
ZnF
BPA
IQ
PK
Introduction
Ankyrin repeat (ANK) is one of the most common protein
domain widely distributed in organisms ranging from virus
to human (Sedgwick and Smerdon 1999). This domain,
composed of 33 amino acids, was initially found in two
yeast cell-cycle regulators Swi6/Cdc10 and the Drosophila
signaling protein Notch (Breeden and Nasmyth 1987), and
was named after the discovery of 24 copies of this
sequence in the cytoskeletal protein ankyrin (Lux et al.
Electronic supplementary material
article (doi:10.1007/s11103-009-9518-6) contains supplementary
material, which is available to authorized users.
The online version of this
J. Huang (&) ? X. Zhao ? H. Yu ? Y. Ouyang ? L. Wang ?
Q. Zhang
National Key Laboratory of Crop Genetic Improvement,
National Centre of Plant Gene Research (Wuhan), Huazhong
Agricultural University, 430070 Wuhan, China
e-mail: hjy020@webmail.hzau.edu.cn
123
Plant Mol Biol
DOI 10.1007/s11103-009-9518-6

Page 2

1990). The ANK, known to mediate protein–protein
interactions (Michaely and Bennett 1992; Bork 1993), was
found in numerous proteins with diverse functions (Mosavi
et al. 2004). In animals and yeast, some ANK proteins
played important roles in cell-cycle control, transcriptional
regulation, cytoskeleton integrity and signal transduction
(Sedgwick and Smerdon 1999).
In plants, ANK proteins have been demonstrated to be
involved in a number of important processes. AKR was the
first reported ANK protein in higher plant Arabidopsis
(Zhang et al. 1992). It was regulated by light and played a
regulatory role in cell differentiation and development.
Overexpressing antisense or sense RNA of the AKR gene
exhibited a chlorotic phenotype. AKR2A was found to play a
role in the biogenesis of the chloroplast outer envelope
membrane (OEM) proteins (Bae et al. 2008). BOP1, an
ANK gene from Arabidopsis was required for leaf mor-
phogenesis (Ha et al. 2004). EMB506 containing five ANK
repeats organized in tandem within the C terminal moiety
wasessentialforembryogenesisinArabidopsis(Albertetal.
1999). Further study revealed that EMB506 interacted with
AKRP encoded by AKR through their ANK domains. Both
EMB506 and AKRP were essential for the plastid differ-
entiation linked to cell differentiation, morphogenesis and
organogenesis during the plant life cycle and development
(Garcion et al. 2006). TIP1, also an ANK gene in Arabid-
opsis encoding an S-acyl transferase, affected cell growth
throughout the whole life of plant (Hemsley et al. 2005).
It was shown that overexpression of TIP1 led to longer
root hairs suggesting it functions in root hair formation.
Another ANK gene XBAT32 also affected the lateral root
initiation and its expression was induced by auxin (Nodzon
et al. 2004). LIANK, an ANK gene in Lily, was essential for
pollen germination and pollen tube growth. Overexpression
of LIANK caused abnormal pollen tube growth while
knockout of this gene decreased the growth of the polarized
tip of the pollen tube (Huang et al. 2006).
Ankyrin repeat proteins were also found to play impor-
tant roles in responses to biotic and abiotic stresses in plant.
The Arabidopsis ANK protein AKR2 might be involved in
the regulation of antioxidation metabolism in both disease
resistance and stress responses (Yan et al. 2002). In pepper,
CaKR1 was found to play roles in both biotic and abiotic
stress responses (Seong et al. 2007b). Transgenic tomato
plants expressing CaKR1 showed enhanced resistance to
Phytophthora infestans (Seong et al. 2007a). ACD6, acting
as a plasma membrane localized positive regulator of sali-
cylic acid signaling, controlled defense responses against
virulent bacteria (Lu et al. 2003, 2005). ZFAR1 was a gene
encoding a putative zinc-finger protein with ANK domains,
and the ZFARI mutants showed increased local suscepti-
bility to Botrytis and sensitivity to germination in the
presence of abscisic acid (AbuQamar et al. 2006). ITN1
encoding an ANK-transmembrane protein was implicated
in diverse cellular processes such as signal transduction.
The itn1 mutation partially impaired ABA signaling path-
ways (Sakamoto et al. 2008). NPR1, a positive regulator of
acquired resistance responses, was a central activator of
SA-regulated gene expression (Cao et al. 1997).
Ankyrin repeat proteins were also found to perform
other functions. Examples included ACBP2 that may play a
role in mediation of AtEBP movement between cells
(Li and Chye 2004), MjXB3, which was involved in the
coordination of the senescence program (Xu et al. 2007),
AKT3, which was a functional transport protein (Ketchum
and Slayman 1996), and IGN1, which was shown to be
required for the maintenance of nitrogen-fixing symbiosis
in root nodules (Kumagai et al. 2007).
To our knowledge, only two ANK proteins have been
reported in rice. One was XB3 containing an ANK domain
interacting with the kinase domain of XA21, a receptor-like
kinase protein in rice having a major role in resistance
against Xanthomonas oryzae pv oryzae (Wang et al. 2006).
The other was OsCBT identified as a transcriptional acti-
vator modulated by CaM binding (Choi et al. 2005).
Rice is the main staple food for a large segment of the
world population and is an ideal model plant to analyze
gene expression and function (Zhang 2007; Zhang et al.
2008). The finished high quality sequences of the rice
genome (International Rice Genome Sequencing Project,
2005) and data generated from high-throughput expression
analysis provided an excellent opportunity for genome-
wide analysis of all the genes belonging to a specified gene
family. In the study reported in this paper, we identified
175 OsANK genes in rice by database searches, which
genes were classified by protein domains. We analyzed the
phylogenetic relationship of the ANK genes in rice and
Arabidopsis as well as segmental and tandem duplications
of OsANK genes. We also surveyed the expression patterns
of OsANK genes in the whole life developmental stages of
rice and their responses to three representative phytohor-
mone (NAA, KT and GA3) and light/dark treatments. The
data generated will be very helpful for studies on the bio-
logical functions of each OsANK gene.
Materials and methods
Collection and classification of OsANK members
First of all, we download the Hidden Markov Model
(HMM) profile of ANK from Pfam (http://pfam.sanger.
ac.uk/). The consensus protein sequence of ANK was
generated by hmmemit (Eddy 1998) (http://mobyle.pasteur.
fr/cgi-bin/MobylePortal/portal.py?form=hmmemit).
BLAST search tools BLASTP and TBLASTN (Altschul
Then
Plant Mol Biol
123

Page 3

et al. 1997) were used to identify putative OsANK with the
ANK consensus protein sequence as a query against three
databases: TIGR rice genome annotation (http://rice.plant
biology.msu.edu/), National Centre for Biotechnology
Information (http://www.ncbi.nlm.nih.gov/)
Knowledge-Based Oryza Molecular Biological Encyclo-
pedia (KOME) (http://www.cdna01.dna.affrc.go.jp/cDNA).
The BLASTP and TBLASTN search parameters of the
three databases were set as follows: max target sequences-
500 and expect value less than ten. In addition, aimed at a
more complete collection of putative ANK genes in rice,
keywords (ankyrin, ankyrin repeat) and domain (PF00023)
search against TIGR were also performed. The SMART
(http://smart.embl-heidelberg.de/) and Pfam searches were
used to confirm and make classification of each predicted
OsANK gene or protein. Based on additional conserved
motifs or domains besides ANK, we classified the OsANK
proteins into subfamilies and the sample protein structures
of each subfamily were drawn manually. Information about
the gene structures, transcripts, chromosomal localization,
full-length cDNA, BAC accessions for each gene and
characteristics of corresponding proteins were procured
from TIGR, KOME and GRAMENE (Liang et al. 2008).
and the
Phylogenetic analysis and sequence alignment
By using OsANK protein sequences, an unrooted tree was
generated by ClustalX version1.83 (Thompson et al. 1997)
with neighbor-joining method (Saitou and Nei 1987) and
bootstrap analysis (1,000 replicates). The tree was analyzed
anddisplayedusingMEGAsoftwareversion4(Tamuraetal.
2007).Anotherunrootedtreewasconstructedusingthesame
method with the alignment of OsANK and AtANK protein
sequences. We defined two proteins with 100% support as
homologous proteins in the same species while those from
differentspecieswith100%supportasorthologousproteins.
Multiple sequence alignments were performed with Alignx
in Vector NTI 9.0 (Lu and Moriyama 2004) and were con-
firmed by ClustalX version1.83 (Thompson et al. 1997).
Localization on chromosomes and duplications
OsANK genes were placed on rice chromosomes according
to their positions given in the TIGR rice database. The
distribution of OsANK genes on the rice chromosomes was
drawn by MapInspect (http://www.plantbreeding.wur.nl/
UK/software_mapinspect.html) and modified manually
with annotation. For detection of large segmental dupli-
cations, we referred to the segmental genome duplication
of rice on TIGR with a maximum length distance permitted
between collinear gene pairs of 100 kb as well as
500 kb (http://rice.plantbiology.msu.edu/segmental_dup/index.
shtml). We designated tandem duplicated genes if two OsANK
genes were separated by five or fewer gene loci according to
Rice Genome Annotation Release 6 of TIGR. The software
MegAlign 4.03 (Clewley and Arnold 1997) was used to ana-
lyze the homology of duplicated OsANK genes.
Phytohormone and light/dark treatments
For phytohormone treatments, 7-day-old light-grown rice
seedlings (trefoil stage) of two elite hybrid rice parents
Minghui 63 and Zhenshan 97 were both transferred to
solutions of 0.1 mM NAA (a member of the auxin family),
GA3 (first isolated and identificated GA) and KT (a cyto-
kinin) respectively. Samples were harvested at the time
points of 5, 15, 30 and 60 min after treatments and the
samples under the same phytohormone treatment of dif-
ferent time points were mixed together.
For the light/dark treatments, seedlings of Minghui 63
and Zhenshan 97 at the plumule and radicle stages were
placed under 48 h continuous light or darkness and har-
vested at the indicated time respectively.
Expression profile analysis
Expression profile data of OsANK genes in Minghui 63 and
Zhenshan 97 were extracted from database CREP (http://
crep.ncpgr.cn/) which composed of hybridization of RNA
samples from 39 tissues that covered the whole life cycle of
ricewiththeAffymetrixricemicroarray.Thirteenvegetative
and 11 reproductive tissues of different developmental
stages covering the whole life cycle of rice were used for
OsANK expression profile analysis in this study: (1) germi-
nating seed at 72 h of imbibitions (Seed); (2) seedlings of
3 daysaftersowing(Seedling1);(3)seedlingsattrefoilstage
(Seedling2); (4)shoots ofseedlings with twotillers(Shoot);
(5) roots of seedlings with two tillers (Root); (6) leaves at
secondary branch primordial stage of young panicle devel-
opment(Leaf1);(7)leavesatmeiosisstageofyoungpanicle
development(Leaf2)(8)flagleavesat5 daysbeforeheading
(Flag leaf 1); (9) flag leaves at 14 days after flowering (Flag
leaf 2); (10) sheath at secondary branch primordial stage of
young panicle development (Sheath 1); (11) sheath at mei-
osis stage of young panicle development (Sheath 2); (12)
stemat5 daysbeforeheading(Stem1);(13)stematheading
stage (Stem 2); (14) panicle at secondary branch primordial
stage (Panicle 1); (15) panicle at pistil/stamen primordial
differentiation stage (Panicle 2); (16) panicle at pollen-
mother cell formation stage (Panicle 3); (17) panicle at
meiosisstageofyoungpanicledevelopment(Panicle4);(18)
panicleatheadingstage(Panicle5);(19)hullat1 daybefore
flowering (Hull); (20) stamen at 1 day before flowering
(Stamen); (21) spikelet at 3 days after flowering (Spikelet);
(22) endosperm at 7 days after pollination (Endosperm 1);
(23) endosperm at 14 days after pollination (Endosperm 2);
Plant Mol Biol
123

Page 4

(24) endosperm at 21 days after pollination (Endosperm 3).
In addition, expression profiles of OsANK genes under three
different phytohormone treatments (NAA, KT and GA3) of
seedlings at trefoil stage as well as light/dark treatments of
the seedlings (plumule and radicle tissues) were also ana-
lyzed. The detailed information of samples used in micro-
array analysis was listed in Supplemental Table 1.
After normalization, variance stabilization, the average
signal value of two biological replicates for each tissue was
used for analysis, except for five tissues (2, 3, 14, 15 and
16) which had three biological replicates and two technical
replicates. Wherever more than one probe set was available
for one gene, the average signal value of these probe sets
belonging to the same gene was used for analysis. To
identify preferentially expressed genes, a student-t test was
performed. A gene in a given tissue was defined as pref-
erentially expressed only if the expression value of the
gene in this tissue was more than two-fold and had a P
value less than 0.05 compared to all other 23 tissues. Under
phytohormone and light/dark treatments, genes that were
up- or down-regulated more than two-fold and with P value
less than 0.05 compared to control were considered as
differentially expressed.
Real-time PCR analysis
Real-time PCR reactions were carried out using the same
RNA samples which were used for microarrays. Gene-
specific primers were designed using PRIMER EXPRESS
version 2.0 (Applied Biosystems) with default parameters.
For real-time PCR analysis, the first-strand cDNA was
synthesized from total RNA using Superscript III reverse
transcriptase (Invitrogen). Real-time PCR was carried out
using ABI PRISM 7500 Real-Time PCR system (Applied
Biosystems). Each reaction of a volume of 25 ll contained
2.0 ll transcription product, 0.5 ll primers, 12.5 ll
2 9 SYBR Premix?Ex TaqTMand 0.5 ll 50 9 ROX ref-
erence dye II (TaKaRa). The thermal cycle was set as: 95?C
for 10 s; 45 cycles of 95?C for 5 s, 60?C for 34 s. Rice
Actin1 gene (Accession number X16280) was used as
internal control. The relative expression levels were ana-
lyzed as described previously (Livak and Schmittgen 2001).
Results
Identification of ANK genes in rice
To identify ANK genes in the rice genome, three public
databases: TIGR, NCBI and KOME were used. The con-
sensus protein sequence generated by hmmemit from ANK
HMM profile (PF00023) is DGFTPLHLAALRGNLEV
VKLLLSQGADLNAQDD. Using this sequence as a
query, searching by BLASTP and TBLASTN against rice
(Oryza sativa japonica subsp. cv Nipponbare) genome was
proceeded. By removal of the same sequences from the
three databases and different transcripts of the same gene,
we identified 172 putative OsANK genes including two
genes (AK110327 and AK119826) only found in KOME.
Keywords and domain searches against TIGR were also
performed, resulting in one and eight new members
respectively. All the protein sequences of the 181 putative
OsANK genes were confirmed by SMART and Pfam
searching for the presence of ANK domain. Six of them
(Os01g28220, Os05g39750, Os07g09160, Os08g28220,
Os11g43690 and Os12g13170) had no ANK domain and
were excluded in further analysis. Therefore, there were at
least 175 OsANK genes in the rice genome, of which 173
had corresponding locus IDs in TIGR database. The
number of ANK proteins in rice (175 members) is greater
than that in Arabidopsis (105 members) (Becerra et al.
2004). The detailed information and the gene structure of
representative OsANK genes could be found in Supple-
mental Table 2 and Supplemental Fig. 1. For convenience,
all the ‘‘LOC_’’ prefix of TIGR locus IDs were omitted in
the rest of this paper.
Classification of OsANK proteins
According to the detailed results of SMART and Pfam
searches, the 175 OsANK proteins were classified into ten
subfamilies based on their domain compositions (Fig. 1).
Seventy-three members merely with ANK domain belon-
ged to subfamily ANK-M. Besides ANK domain, OsANK
proteins containing several other known functional domains
were classified into the following subfamilies. Thirty-seven
members containing the transmembrane domain were
identified as subfamily ANK-TM; 22 members containing
the tetratricopeptide repeat domain were identified as sub-
family ANK-TPR; 9 members with the Ring Finger domain
were identified as ANK-RF subfamily; ANK-BTB sub-
family (6 members) had broad-complex, tramtrack and bric
a brac domains; ANK-ZnF subfamily (7 members) con-
tained zinc-finger domain; ANK-BPA subfamily (3 mem-
bers) had BAR, PH and ArfGap domains; ANK-IQ
subfamily (4 members) contained the Calmodulin-binding
domain; ANK-PK subfamily (4 members) had S_TKc
domain or Pkinase-tyr domain; ANK-O subfamily (10
members) contained other domains including CHROMO,
ACBP, RCC1, AAA, PPR, Motile-sperm, HDAC, rve,
RVT, CNMP, Ion-trans and RRM (Supplemental Fig. 2).
The differences and comparison of classification of OsANK
and ANK gene family in Arabidopsis can be seen in
Table 1. The numbers of members in ANK-TM, ANK-ZnF,
ANK-BPA, ANK-IQ, ANK-BTB subfamilies are not very
Plant Mol Biol
123

Page 5

different between the two species. However, the number of
protein kinases in OsANK is slightly smaller than that in
Arabidopsis. The numbers of ANK-M and ANK-TPR in
rice are larger than that in Arabidopsis, likely because those
proteins having only one ANK or more than two ANK
domains but separated by more than 20 amino acids were
excluded in the previous study of ANK gene family in
Arabidopsis (Becerra et al. 2004).
Conserved motifs investigation in ANK-M subfamily
proteins
Almost half of the OsANK proteins (41.71%) had ANK as
the only recognizable domain. In order to find more
conserved motifs or domains among the 73 members of
ANK-M subfamily, the motif investigation software
MEME (Multiple EM for Motif Elicitation) version 4.0.0
(Bailey et al. 2006) (http://meme.sdsc.edu/meme4/cgi-bin/
meme.cgi) was employed. The parameters of this analysis
were set up as below: number of repetitions—any, maxi-
mum number of motifs—20, optimum motif width set to
C6 and B200. The E values of 20 putative motifs identified
were less than 1.00E-30 (Fig. 2). The motifs identified by
MEME were between 15 and 119 amino acids in length
and exist in at least two of predicted ANK-M proteins.
Motif 8 existed in as many as 60 ANK-M proteins with E
value less than 1.00E-100. Motif 17 composed of 119
amino acids was only found in two ANK-M proteins.
ANK
CHROMO
BARBAR
PH PH
ArfGap
RING
ACBP
AAA
Pkinase_Tyr
Pkinase_Tyr
PPR
PPRPPR
BTB
Motile_Sperm
ZnF
TPR
TM
IQ
IQ
FYVE FYVERCC1 RCC1
HDAC
interact
rve rve
RVT
cNMP
Ion_trans
RRM
ubiquitin
Os12g43940
ANK-TPR
Os07g26490
ANK-RF
Os11g24670
ANK-TM
Os05g03760
ANK-ZnF
Os03g46440
ANK-BTB
Os03g27080
ANK-IQ
Os01g54480
ANK-PK
Os02g42134
ANK-BPA
Os02g56530
ANK-M
Os04g58550
Os07g07080
Os03g03990
Os03g03990
Os02g16660
Os02g16660
Os04g28300
Os01g61330
Os07g42640
Os07g42640
Os02g29030
Os02g29030
Os05g35410
Os12g39700
Os12g39700
ANK-O
Fig. 1 Structure of representative OsANK protein from each sub-
family. Subfamily name of each corresponding protein belonged to
and TIGR locus ID are given on the left. Domain abbreviations are:
ANK ankyrin repeat domain; TM transmembrane; TPR tetratricopep-
tide repeat domain; RING ring finger domain; ZnF zinc finger; BTB
broad-complex, tramtrack and bric a brac domains; IQ calmodulin-
binding domain; BAR BAR domain; PH pleckstrin homology domain;
ArfGap putative GTP-ase activating proteins for the small GTPase,
ARF; Motile_Sperm MSP (major sperm protein) domain; AAA
ATPase family associated with various cellular activities; CHROMO
CHRromatin organisation MOdifier domain; ACBP acyl CoA binding
protein; PPR pentatricopeptide repeats; Ion_trans ion transport
protein; cNMP cyclic nucleotide-binding domain; RCC1 regulator
of chromosome condensation; FYVE FYVE zinc finger; rve integrase
core domain; RVT reverse transcriptase (RNA-dependent DNA
polymerase); RRM RNA recognition motif; ubiquitin ubiquitin
family. The length and order of domains represent actual situation
in each protein
Plant Mol Biol
123

Page 6

These putative motifs obtained from MEME were anno-
tated by searching SMART and InterProScan (Hunter et al.
2009), which showed that most motifs had unknown
functions except motif 1, 2, 3, 8, 9, 11 and 17. All the
motifs with known functions were annotated as ANK. The
log likelihood ratio, E value, consensus amino acid
sequence and length of each motif are given in Supple-
mental Table 3. The phylogenetic analysis of OsANK gene
family (below) showed that those ANK-M genes had very
distant evolutionary relationships. Therefore, we inferred
that maybe these motifs identified by MEME were
important for the function of the proteins although most of
their functions were unknown.
Phylogenetic analysis and multiple sequence
alignments
To clarify the phylogenetic relationship among the OsANK
genes and infer the evolutionary history of this gene fam-
ily, a combined phylogenetic tree was constructed with the
aligned OsANK protein sequences (Fig. 3), from which it
can be seen that the OsANK proteins fall into eight major
groups (group A to group H). Group B, E, F and G were
further divided into 4, 3, 2 and 2 subgroups, respectively.
In addition, we observed that most members in the same
groups or subgroups shared one or more domains outside
the ANK domain, further supporting the subfamily defini-
tion above. For instance, all the members of ANK-TPR
subfamily belonged to group A. Seven out of nine of ANK-
RF members were assigned to subgroup B1. Similarly,
subgroup B2 contained all the seven members of ANK-RF.
Group C consisted of 21 members of ANK-M and 20
members of ANK-TM. All the members of ANK-BTB,
ANK-BPA, and three of four members of ANK-IQ sub-
family were classified into group E and all the ANK-PK
proteins were assigned to subgroup G1. Group F1 consisted
of 12 proteins while 10 of them belonged to ANK-M.
However, the members of ANK-M were distributed in
almost all the groups. It reflected that the functions of
ANK-M subfamily genes were diversified. In general, most
of the closely related members in the phylogenetic tree had
the same or very similar domain composition. However,
the classification based on domain composition did not
completely match the phylogenetic classification.
With the development of comparative genomics, it is
possible to analyze proteins of the same gene family
among different species. In order to evaluate the phylo-
genetic relationship among the ANK proteins in rice and
Arabidopsis, another unrooted tree was constructed from
alignments of all the 280 ANK protein sequences of the
two species using the same method mentioned above
(Supplemental Fig. 3). The AtANK protein sequences
were obtained fromThe
Resources (TAIR) (http://www.arabidopsis.org/) accord-
ing to the IDs provided in the previous article (Becerra
et al. 2004). This phylogenetic analysis suggested that
most OsANK and AtANK proteins clustered in specie-
specific clades with very high bootstrap supporting,
exceptfor17pairs of
Os05g01310 and AT3G04710. We also identified 17 pairs
of paralogous proteins in Arabidopsis and 20 pairs in rice,
such as AT3G23280 and AT4G14365, Os02g29040 and
Os02g29130. This result indicated that only a few mem-
bers of OsANK and AtANK gene families possibly origi-
nated from the same ancestral genes before divergence of
dicots and monocots.
Arabidopsis
Information
orthologous proteins,i.e.
Table 1 Numbers of each subfamily of ANK proteins in rice and Arabidopsis
Subfamily DescriptionRice
Arabidopsis
ANK-M Proteins with only ankyrin repeats73 18
ANK-TM Ankyrin-transmembrane proteins 3740
ANK-TPRProtein with tetratricopeptide repeats 221
ANK-OProteins with other domains1013
ANK-RFRing finger proteins 95
ANK-ZnF Zinc-finger proteins76
ANK-BTBProteins with BTB domain 67
ANK-IQCalmodulin binding motif-containing protein44
ANK-PKProtein kinases47
ANK-BPAARF GTPase-activating domain-containing protein34
Total175 105
Fig. 2 Distribution of conserved motifs in the members of ANK-M
subfamily identified by MEME. The name of each member and
combined P value are shown on the left side of the figure and length
of each motif is indicated at the bottom of the figure. Different motifs
are indicated with different color boxes numbered 1–20. The same
number in different proteins refers to the same motif. The length and
order of motifs represent the actual situation in each protein. The
detailed information of the motifs is given in Supplemental Table 3
c
Plant Mol Biol
123

Page 7

Plant Mol Biol
123

Page 8

Os01g07640
Os01g07980
Os01g09370
Os01g08000
Os01g09384
Os03g47686
Os03g47702
Os02g29030
Os02g29040
Os02g29130
Os02g29140
Os02g29160
Os09g03630
Os09g03680
Os09g03750
Os12g40770
Os12g40780
Os12g43840
Os12g43940
Os08g13640
Os02g29190
Os02g29210
Os03g42350
Os05g01310
Os03g47620
Os03g47640
Os03g47650
Os03g47693
Os03g47720
Os07g46500
Os10g43040
Os03g47670
743
Os05g07020
Os10g13870
Os02g29110
Os01g61330
Os03g55110
Os02g25960
Os04g41280
Os09g11120
Os03g03990
Os01g74320
Os05g02130
Os03g16780
Os02g54860
Os08g15840
Os06g03800
Os10g37730
Os02g01240
Os08g23590
(A)
B1
Os02g38970
Os05g23320
Os09g36690
Os11g24770
Os11g07980
Os11g08050
Os11g24690
Os11g24730
Os11g24850
Os12g12810
Os11g08020
Os11g08070
Os11g24670
Os11g24720
Os11g24750
Os11g24780
Os11g24840
Os09g16550
Os03g05260
Os03g45290
Os11g09260
Os07g22390
Os07g09550
Os07g34820
Os07g34760
Os07g34780
Os07g34830
Os05g35310
Os06g13120
Os07g34390
B2
B3
B4
(B)
Os07g29860
Os07g29940
Os07g31070
Os08g31000
Os09g16050
Os11g09230
Os09g15370
Os09g16160
Os09g17120
Os09g17329
Os09g15950
Os09g12350
Os09g16240
Os11g09190
Os11g09220
(C)
782
1000
782
1000
1000
1000
1000
986
1000
1000
1000
712
455
1000
458
1000
749
796
538
629
1000
1000
483
447
475
593
1000
755
398
102
879
952
164
684
1000
588
126
17
14
926
1000
1000
980
1000
992
429
105
980
534
496
988
996
679
559
565
349
1000
956
953
563
1000
925
1000
232
34
16
369
3
7
727
531
588
977
564
1000
914
334
799
742
789
890
582
1000
960
925
774
853
718
813
434
642
280
251
43
Fig. 3 Evolutionary relationship among the rice OsANK proteins.
The unrooted tree was generated using ClustalX program by
neighbor-joining method. Bootstrap values from 1,000 replicates are
indicated at each node. OsANK proteins are divided into eight distinct
groups (A–H). Group B, E, F and G are further divided into 11
subgroups (B1–B4, E1–E3, F1–F2, and G1–G2)
Plant Mol Biol
123

Page 9

To gain understanding of the connection between the
domain composition and phylogenetic relations of the
members in the same subfamily, multiple sequence align-
ments of each subfamily were proceeded. As an example,
the alignment result of ANK-TPR subfamily was list in
Supplemental Fig. 4. The amino acid residues corre-
sponding to the ANK and TPR domain could be easily
identified from the alignments. As all the members of
ANK-TPR subfamily were in the same group of the phy-
logenetic tree, the result further supported the relationship
analysis between subfamily and phylogenetic group above.
Chromosomal locations of the OsANK genes
To determine the genomic distribution of OsANK genes,
we used the DNA sequences of OsANK genes to search
g
Os11g09220
Os11g09270
Os09g16200
Os12g18590
Os01g61990
Os03g17240
Os08g42960
Os09g34280
Os03g17250
Os02g50270
Os07g09530
Os11g14544
Os11g14570
Os09g14870
Os11g14630
Os11g14260
Os11g15460
Os09g12480
Os02g09130
Os06g43680
Os11g34860
Os12g38460
Os12g39700
Os03g55330
Os01g56200
Os03g46440
Os01g09800
Os11g04600
Os12g04410
Os01g72020
Os03g04300
Os06g13000
Os09g17520
(D)
E1
Os01g69910
Os04g31900
Os03g27080
Os08g42530
Os09g33600
Os02g42134
Os08g42690
Os09g33810
Os03g63480
Os07g30774
Os07g26490
Os04g58550
Os01g71540
Os01g71590
Os02g15790
Os10g01274
Os07g33200
Os07g33210
Os05g03320
Os05g03390
Os05g03410
Os11g01720
Os12g02660
Os11g11240
Os11g04910
Os07g42640
Os02g16660
Os01g66860
Os02g39560
Os01g54480
Os07g43900
Os04g48520
Os04g28300
Os07g07080
AK119826
Os01g45990
Os07g07910
Os05g35410
Os04g36740
Os06g14030
Os02g56530
Os07g25460
Os03g49170
Os12g33090
Os07g38090
Os05g03760
Os09g12420
AK110327
E2
E3
(E)
F1
F2
(F)
G1
G2
(G)
(H)
886
813
937
1000
1000
997
387
1000
993
619
790
760
991
163
424
432
140
1000
1000
1000
997
48
2
1000
1000
1000
1000
1000
1000
753
103
989
1000
1000
1000
523
1000
690
366
3
235
98
3
0
1000
817
1000
1000
884
1000
462
913
1000
1000
1000
351
35
7
0
1000
926
1000
112
28
6
278
1000
1000
1000
1000
6
1000
1000
1000
1000
854
278
58
0.1
Fig. 3 continued
Plant Mol Biol
123

Page 10

against the rice genome database with BLASTN at TIGR.
The position of each gene could be found in Supplemental
Table 2. Totally, 173 of the 175 OsANK genes could be
localized on the 12 chromosomes with obviously uneven
distribution. Within a single chromosome, OsANK genes
could be found in all regions: at the telomeric ends, near
the centromere, dispersed all over, individually or in
clusters (Fig. 4). Chromosome 11 had the largest number
of 28 OsANK genes followed by 22 on chromosome 3. In
contrast, only four OsANK genes were found on chromo-
some 10 and six OsANK genes were on the long arm of
chromosome 4. Seven out of 10 were located in the short
arm of chromosome 5 and seven genes were located in the
long arm of chromosome 12. Less than 10 OsANK genes
were found on chromosome 4, 6, 8 and 10. Two OsANK
genes (Os07g22390 and Os10g13870) encoding proteins
having only the ANK domain were positioned around the
centromere, which were considered to contain actively
transcribed genes (Cooke 2004; Nagaki et al. 2004). Addi-
tionally, six OsANK genes, Os01g74320, Os02g01240,
Os05g01310, Os10g01274, Os10g43040 and Os12g43940,
were located near the telomeric regions.
During the evolution of plant gene family, segmental
duplication and tandem duplication play a part in retaining
the large number of gene family (Cannon et al. 2004). With
the purpose of elucidating the potential mechanism of
evolution of OsANK gene family, both segmental and
tandem duplication events were analyzed. We found nine
segmental duplication events (Fig. 4). Although the ANK-M
subfamily contained the maximum members of OsANK
Fig. 4 Distribution of OsANK genes on the rice chromosomes. The
scale is in megabases (Mb). The chromosome numbers are indicated
at the top of each bar. The white circle on each chromosome (vertical
bars) shows the rough position of the centromere. The markers in
front of the OsANK genes indicate the subfamily of each protein
belonged. The ten markers represent the subfamily are mentioned at
the bottom of the diagram. The genes with open reading frames in
opposite orientations are marked on different sides of the chromo-
some, i.e. on the left side represent downward while on the right
indicate upward. Straight lines connect the OsANK genes presented
on duplicated chromosomal segments and tandem duplicated gene
clusters are marked by the red rectangle
Plant Mol Biol
123

Page 11

genes, only one pair of OsANK genes belonging to this
subfamily were on duplicated chromosomal segments on
chromosome 8 and 9 (Os08g35930 and Os09g27090).
Similarly, the second large subfamily ANK-TM held three
pairs of segmental duplicated genes. Os05g02130 partici-
pated in two duplication events with Os01g74320 and
Fig. 5 Expression analysis of
OsANK genes in the whole life
cycle of Minghui 63.
Hierarchical cluster displaying
expression profiles for 158
OsANK genes based on average
log2 signal values in 24 tissues.
The tissues are indicated at the
bottom. The color key
representing log2 signal values
is shown at the bottom. On the
left side of expression map, the
cluster dendrogram is shown.
On the right side, five groups
have been made for the genes
showing discrete expression
patterns. Different color lines
representing different groups
are mentioned at the bottom of
the diagram
Plant Mol Biol
123

Page 12

Os03g16780 and these three genes all belonged to the
ANK-RF. Besides, no duplication events occurred among
the chromosome 4, 7 and 10.
Totally 65 OsANK genes were involved in tandem
duplications consisting of 19 clusters or 47 pairs (Fig. 4).
The number of OsANK genes arranged in the tandem
repeats varied from 2 to 9 and most of these genes were
assigned in the same orientation on the chromosomes
excepttwopairs(Os09g16200
Os11g09230 and Os11g09260). Moreover, the homology
of protein sequences of these genes ranged from 21.5 to
99.4% and 17 pairs were juxtaposed with no intervening
gene. The detailed information about the duplicated genes
could be found in Supplemental Table 4.
andOs09g16240;
(a)
Os02g15790
0
5
10
15
20
25
30
35
40
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4
P5
Hull
Stamen
Spikelet
E1
E2
E3
Os02g16660
0
400
800
1200
1600
2000
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4
P5
Hull
Stamen
Spikelet
E1
E2E3
Os03g17250
0
50
100
150
200
250
Seed
Sl1Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4P5
Hull
Stamen
Spikelet
E1
E2
E3
Os07g07080
0
500
1000
1500
2000
2500
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4
P5
Hull
Stamen
Spikelet
E1
E2
E3
Os07g07910
0
100
200
300
400
500
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4
P5
Hull
Stamen
Spikelet
E1
E2
E3
Os04g36740
0
50
100
150
200
250
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1P2
P3
P4
P5
Hull
Stamen
Spikelet
E1
E2E3
Os07g31070
0
50
100
150
200
250
300
350
400
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4
P5
Hull
Stamen
Spikelet
E1
E2
E3
Os08g42530
0
500
1000
1500
2000
2500
3000
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4P5
Hull
Stamen
Spikelet
E1
E2
E3
Os09g16050
0
50
100
150
200
250
Seed
Sl1Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3P4
P5
Hull
Stamen
Spikelet
E1
E2
E3
Os11g14544
0
10
20
30
40
50
60
70
80
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4
P5
Hull
Stamen
Spikelet
E1
E2
E3
Os11g24780
0
50
100
150
200
250
300
350
400
450
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1FL2
She1
She2
Stem1
Stem2
P1
P2P3
P4
P5
Hull
Stamen
Spikelet
E1
E2
E3
Os11g24840
0
50
100
150
200
250
300
350
400
450
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4
P5
Hull
Stamen
Spikelet
E1
E2
E3
Os07g34830
0
10
20
30
40
50
60
70
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4
P5
Hull
Stamen
Spikelet
E1
E2
E3
Fig. 6 Expression patterns of
13 preferentially expressed
OsANK genes in Minghui 63
and 5 genes in Zhenshan 97. a
13 preferentially expressed
genes in Minghui 63. b 5
preferentially expressed genes
in Zhenshan 97. The X-axis is
for the selected tissues and
Y-axis represents the signal
values. Tissue names: Seed
germinating seed at 72 h of
imbibitions; Sl 1 seedlings of
3 days after sowing; Sl 2
seedlings at trefoil stage; Shoot
shoots of seedlings with two
tillers; Root roots of seedlings
with two tillers; Leaf 1 leaves at
secondary branch primordial
stage of young panicle
development; Leaf 2 leaves at
meiosis stage of young panicle
development; FL 1 flag leaves at
5 days before heading; FL 2 flag
leaves at 14 days after
flowering; She 1 sheath at
secondary branch primordial
stage of young panicle
development; She 2 sheath at
meiosis stage of young panicle
development; Stem 1 stem at
5 days before heading; Stem 2
stem at heading stage; P1
panicle at secondary branch
primordial stage; P2 panicle at
pistil/stamen primordial
differentiation stage; P3 panicle
at pollen-mother cell formation
stage; P4 panicle at meiosis
stage of young panicle
development; P5 panicle at
heading stage; Hull hull at 1 day
before flowering; Stamen
stamen at 1 day before
flowering; Spikelet spikelet at
3 days after flowering;
E1 endosperm at 7 days after
pollination; E2 endosperm at
14 days after pollination;
E3 endosperm at 21 days after
pollination
Plant Mol Biol
123

Page 13

Expression profiling of OsANK gene family
in the whole life cycle of rice
Microarray analysis was performed using Affymetrix rice
microarray for studying the expression pattern of OsANK
genes. The rice tissues and developmental stages selected
for microarray analysis cover the entire life cycle of rice in
Minghui 63 and Zhenshan 97. Detailed information on
selected stages could be found in methods. The expression
profile of the microarray data was confirmed by RT-PCR in
the previous articles of our lab (Nayidu et al. 2008; Nur-
uzzaman et al. 2008; Ye et al. 2009).
Probes for 158 of the 175 OsANK genes could be iden-
tified in the Affymetrix microarray. Twenty-eight genes had
two probe sets and the average signal value of the probe sets
was used for analysis. Os01g71540 and Os01g71590 shared
the same probe set because of their highsequence homology
(99.4%). Os03g47640 had two probe sets in the array and
one of the probe sets was shared with Os03g47620. The
transcripts of these genes would cross-hybridize in the
microarray analyses, which made it difficult to obtain
the signal values for each of the genes. We thus used real-
time PCR with gene-specific primers to validate the expres-
sion patterns of these four genes (Supplemental Fig. 5).
Average expression values for 158 OsANK genes of
each sample are given in Supplemental Table 5. Based on
the signal values, it was obvious that most of the OsANK
genes were expressed in at least one of the 24 investigated
tissues. A hierarchical cluster displaying the logarithm of
average signal values for the 158 OsANK genes in Minghui
63 is presented in Fig. 5, based on which the expression
patterns of OsANK genes can be classified into five groups.
Twenty-nine genes belonged to group 1, all of them
showed high expression levels in all the tissues analyzed.
Among the 29 genes, gene Os09g33810 had the highest
expression level in the entire life cycle. As the largest
group, group 2 consisted of 57 genes, all the genes in this
group showed low expression in all the analyzed tissues.
Group 3 of 30 genes showed relatively high expression
level during reproductive tissues compared to vegeta-
tive tissues. Notably, the expression signals of four
genes(Os05g03320,Os07g07080,
Os08g42530) observed in stamen was three to 40-fold
higher than that in the vegetative tissues. Five genes
(Os02g29130, Os03g47650, Os12g43940, Os12g43840
and Os03g42350) showed the highest expression in
developing panicles, one gene (Os11g14570) had the
highest expression in hull, and Os12g40780 had high
expression level in panicle and endosperm stages. Inter-
estingly, the expression of two genes (Os03g47650 and
Os12g43840) which was high in panicles was declined
gradually as the panicles matured. Group 4 comprised of 19
genes with predominant expression in vegetative tissues,
most of which showed a similar expression pattern in the
same tissue at different developmental stages. Os03g46440
had a higher expression signal in flag leaf 2 than in flag leaf
1 and in leaf 2 than in leaf 1. Similarly, Os03g03990
expressed lower in flag leaf 2 than in flag leaf 1 and in leaf
2 than in leaf 1. Another gene (Os11g08020) showed more
than 10-fold higher expression in flag leaf 2 than in flag
leaf 1. Os01g74320 showed lower expression level in stem
2 than in stem 1. Group 5 consisted of 23 genes showing
relatively high expression signal in some vegetative and
reproductive tissues compared with others. Os01g74320
Os09g33600and
Os01g07640
0
5
10
15
20
25
30
35
40
45
50
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4
P5
Hull
Stamen
Spikelet
E1
E2
E3
Os09g15950
0
50
100
150
200
250
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4
P5
Hull
Stamen
Spikelet
E1
E2
E3
Os08g42530
0
200
400
600
800
1000
1200
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4 P5
Hull
Stamen
Spikelet
E1
E2
E3
Os07g07910
0
20
40
60
80
100
120
140
160
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4
P5
Hull
Stamen
Spikelet
E1
E2
E3
Os02g29040
0
50
100
150
200
250
Seed
Sl1
Sl2
Shoot
Root
Leaf1
Leaf2
FL1
FL2
She1
She2
Stem1
Stem2
P1
P2
P3
P4
P5
Hull
Stamen
Spikelet
E1
E2
E3
(b)
Fig. 6 continued
Plant Mol Biol
123

Page 14

had high expression level in flag leaf 2 and stamen.
Interestingly this gene showed more than 60-fold higher
expression in flag leaf 2 than in flag leaf 1.
Most genes had the same expression pattern in Zhenshan
97 as in Minghui 63 except 13 genes (Os01g09370,
Os01g07980, Os02g29160, Os02g29190, Os02g29210,
Os03g17250, Os05g01310, Os05g03320, Os09g03630,
Os09g03680, Os09g15950, Os09g16050 and Os11g24850).
The expression patterns ofOsANK genes inZhenshan97 are
giveninSupplementalFig. 6.Notably,Os09g03680hadlow
expressionsignalinMinghui63whilehavingrelativelyhigh
expressionsignalinleaf,flagleafandsheathinZhenshan97.
Similarly, Os09g03630 had relatively high expression level
in Zhenshan 97 but showed very low expression in Minghui
63 in almost all the vegetative tissues.
With the aim of revealing OsANK gene expression
features, an analysis of preferential expression was per-
formed. Finally, 13 and 5 OsANK genes were identified in
Minghui 63 and Zhenshan 97 respectively which showing
preferential expression in a given stage (Figs. 6a, b; 7).
Among these,onlytwo
Os08g42530) showed the same expression pattern in two
genotypes (Fig. 7). Out of the 16 genes, nine showed
preferential expression in stamen at 1 day before flower-
ing; one (Os02g29040) was preferentially expressed in
panicle 4 and two (Os09g15950 and Os11g24840) in root;
Os01g07640, Os04g36740, Os09g16050 and Os07g34830
were preferentially expressed in panicle 1, seed, stem 1 and
flag leaf 1, respectively.
genes(Os07g07910 and
Responses of OsANK genes to NAA, KT, GA3
and light/dark treatments
Phytohormones play a critical role in plant growth and
development. To investigate the OsANK genes in response
to phytohormone treatment, microarray analysis was per-
formed. We identified a total of 18 OsANK genes that were
differentially expressed with treatments of one or more of
the phytohormone NAA, KT, GA3 in the two genotypes
compared with the control (Fig. 8). The fold change values
with respect to control are given in Supplemental Table 6.
Among these, 15 genes were up-regulated while three
genes were down-regulated. The down-regulated genes
were Os05g02130 treated by NAA in Minghui 63 and
Os07g07080 and Os09g34280 treated by GA3 in Zhenshan
97. Seven genes showed differential expression with phy-
tohormone treatments in both genotypes. Interestingly, two
genes (Os04g48520 and Os06g03800) were both up-regu-
lated with all three phytohormone treatments in both
genotypes. However, the expression profiles of the
remaining genes in two genotypes were different. For
instance, Os01g07980 was up-regulated specifically by
GA3 treatment in Zhenshan 97 while it was up-regulated
by all three phytohormone treatments in Minghui 63.
Conversely, Os06g13000 was up-regulated by all three
phytohormone treatments in Zhenshan 97 but up-regulated
by GA3 and KT treatments in Minghui 63. Six genes
(Os02g29210, Os03g04300, Os05g23320, Os07g07080,
Os09g34280 and Os12g43940) were differentially expres-
sed only in Zhenshan 97 while five genes (Os01g08000,
Os01g09370, Os03g03990, Os05g02130 and Os11g08020)
only in Minghui 63 (Figs. 9, 10). The fold change values
with respect to seedlings treated by phytohormone can be
obtained in Supplemental Table 6.
To investigate the light regulation of OsANK genes,
expression profiles of OsANK genes in seedlings (plumule
and radicle tissues) treated with light or dark for 48 h were
also investigated, with the fold changes with respect to
control listed in Supplemental Table 6. Nine OsANK genes
Minghui 63
11
11
Zhenshan 97
3
3
22
Os07g07910
Os08g42530
Minghui 63 Zhenshan 97
Zhenshan 97
0
1
2
Panicle 1 Panicle 4StamenRoot
No.of Regulated Genes
Os01g07640 Os02g29040
Os07g07910
Os08g42530Os09g15950
0
1
2
3
4
5
6
7
8
9
StamenSeedFlag leaf 1Stem 1Root
No. of Regulated Genes
Minghui 63
Os02g15790
Os02g16660
Os03g17250
Os07g07080
Os07g07910
Os07g31070
Os08g42530
Os11g14544
Os11g24840Os04g36740Os07g34830Os09g16050Os11g24780
(B)
(A)
(C)
Fig. 7 Summary of expression analyses of OsANK genes that show
preferential expression in Minghui 63 and Zhenshan 97. a Venn
diagram indicates the numbers of genes showing preferential
expression in Minghui 63, Zhenshan 97 or in both genotypes; b and
c Black bars indicate the numbers of genes showing preferential
expression in Minghui 63 and Zhenshan 97. Corresponding genes are
listed on the right side of this column. Tissues are mentioned at the
bottom of the column
Plant Mol Biol
123

Page 15

showed differential expression at the plumule stage under
light compared to dark in Minghui 63 and Zhenshan 97
(Fig. 11).Four genes (Os03g03990,
Os06g03800 and Os06g13000) were in common in the two
genotypes. Amongst the nine genes, eight were up-regu-
lated and one (Os02g29210) was down-regulated in
Zhenshan 97. We also found four genes differentially
expressed at the radicle stage under light compared to dark
in both Minghui 63 and Zhenshan 97 (Fig. 11). Interest-
ingly, three of the four genes (1 in Minghui 63 and 2 in
Zhenshan 97) were down-regulated and each was differ-
entially expressed only in one genotype.
Os03g04300,
Comparisons of expression profiles of the duplicated
OsANK genes
Duplicated genes had three alternative outcomes in the
evolutionary course: nonfunctionalization, neofunctional-
ization and subfunctionalization (Lynch and Conery 2000).
The expression patterns for OsANK genes present in seg-
mental and tandem duplicated regions were examined in
Minghui 63. Probe sets were available for seven of the nine
pairs of genes located in the segmental duplicated regions.
Five pairs of genes showed highly similar expression
patterns in most of the tested tissues, indicating subfunc-
tionalization after the duplication events. Os03g47686 was
not expressed in all the selected tissues, which may indi-
cate that one of the members lost its function during the
course of evolution.
CK
NAA
GA3
KT
Os06g03800−MH63
Os06g13000−ZS97
Os06g03800−ZS97
Os03g04300−ZS97
Os05g23320−ZS97
Os03g55330−ZS97
Os01g09370−MH63
Os06g13000−MH63
Os03g55330−MH63
Os09g34280−ZS97
Os03g03990−MH63
Os01g07980−ZS97
Os12g43940−ZS97
Os11g08020−MH63
Os01g66860−MH63
Os01g66860−ZS97
Os04g48520−ZS97
Os04g48520−MH63
Os01g08000−MH63
Os05g02130−MH63
Os07g07080−ZS97
Os12g40770−ZS97
Os01g07980−MH63
Os12g40770−MH63
Os02g29210−ZS97
−
−
−
−
−
−
−
−
−
−
−
Os01g07980−ZS97
−
−
−
−
−
−
−
−
−
−
−
−
−
Fig. 8 Clustering of expression profiles of 18 OsANK genes showing
differential expression in 7-day-old seedlings with three phytohor-
mone (NAA, GA3 and KT) treatments. The color scale representing
log2 signal values is shown on the left. Cluster dendrogram is shown
on the left. Differentially expressed genes are listed on the right of
each lane. The different treatments are mentioned on the top of each
lane
0
100
200
300
400
500
600
700
800
900
1000
Os01g07980-MH63
Os01g08000-MH63
Os01g09370-MH63
Os03g03990-MH63
Os04g48520-MH63
Os06g03800-MH63
Os06g13000-MH63
Os11g08020-MH63
Os01g07980-ZS97
Os01g66860-ZS97
Os02g29210-ZS97
Os03g04300-ZS97
Os03g55330-ZS97
Os04g48520-ZS97
Os06g03800-ZS97
Os06g13000-ZS97
Os07g07080-ZS97
Os09g34280-ZS97
Os12g40770-ZS97
Os12g43940-ZS97
CKGA3
0
50
100
150
200
250
300
350
400
450
500
Os01g07980-MH63
Os01g08000-MH63Os01g09370-MH63
Os04g48520-MH63
Os05g02130-MH63
Os06g03800-MH63
Os11g08020-MH63
Os12g40770-MH63
Os01g66860-ZS97
Os03g55330-ZS97
Os04g48520-ZS97
Os06g03800-ZS97
Os06g13000-ZS97
Os12g40770-ZS97
Os12g43940-ZS97
CKNAA
0
100
200
300
400
500
600
700
800
900
Os01g07980-MH63
Os01g08000-MH63
Os01g09370-MH63
Os01g66860-MH63
Os03g03990-MH63
Os03g55330-MH63
Os04g48520-MH63
Os06g03800-MH63
Os06g13000-MH63
Os12g40770-MH63
Os01g66860-ZS97
Os03g04300-ZS97
Os03g55330-ZS97
Os04g48520-ZS97
Os05g23320-ZS97
Os06g03800-ZS97
Os06g13000-ZS97
Os12g40770-ZS97
CK KT
Fig. 9 Expression profiles of 18 differentially expressed OsANK
genes in 7-day-old seedlings subjected to three phytohormones (NAA,
GA3 and KT). X-axis represents the differentially expressed genes. Y-
axis represents average expression values. Each treatment is men-
tioned at the top of each diagram
Plant Mol Biol
123