In silico analysis reveals 75 members of mitogen-activated protein kinase kinase kinase gene family in rice.

Mitogen-Activated Protein Kinase Kinase Kinases (MAPKKKs) are important components of MAPK cascades, which are universal signal transduction modules and play important role in plant growth and development. In the sequenced Arabidopsis genome 80 MAPKKKs were identified and currently being analysed for its role in different stress. In rice, economically important monocot cereal crop only five MAPKKKs were identified so far. In this study using computational analysis of sequenced rice genome we have identified 75 MAPKKKs. EST hits and full-length cDNA sequences (from KOME or Genbank database) of 75 MAPKKKs supported their existence. Phylogenetic analyses of MAPKKKs from rice and Arabidopsis have classified them into three subgroups, which include Raf, ZIK and MEKK. Conserved motifs in the deduced amino acid sequences of rice MAPKKKs strongly supported their identity as members of Raf, ZIK and MEKK subfamilies. Further expression analysis of the MAPKKKs in MPSS database revealed that their transcripts were differentially regulated in various stress and tissue-specific libraries.

Introduction

Mitogen-Activated Protein Kinase (MAPK) cascade plays an important role in plant growth and development, transferring the extracellular stimuli into intracellular response. MAPK cascades are evolutionarily conserved signalling modules in eukaryotes including animals, yeasts and plants.[1,2] MAPK cascades are composed of three protein kinases: MAPKs, MAPK Kinases (MAPKKs/MKKs) and MAPKK Kinases (MAPKKKs/MEKKs). MAPKs are activated when both tyrosine and threonine residues in the TXY motif are phosphorylated by MAPKKs. MAPKKs are activated when serine and serine/threonine residues in the S/TXXXXXS/T motif are phosphorylated by MAPKKKs.[3]

By sequence comparison and signature motif searches, putative orthologue of MAPK cascade members have been identified in rice, Medicago sativa, Zea mays, tobacco and tomato. Southern blot analysis have also revealed that genomes of monocotyledonous plants such as rice and maize possesses sequences that are homologous to the NPKl gene (MAPKKK related gene in tobacco), an indication that NPKl-related genes are present in a number of plant species.[4] During the past decade, incredible progress has been made towards the functional understanding of all genes in the model dicot Arabidopsis. In the sequenced Arabidopsis genome, 20 MAPKs, 10 MAPKKs and 80 MAPKKKs were identified.[5,6] However, little is known about the MAPK gene family and their function and regulation in rice (Oryza sativa) and other economically important cereal crops.[7] After the completion rice genome project 16 MAPKs and 8 MAPKKs are reported in rice.[8] However, reports on the presence of MAPKKKs in rice have been very slow and so far a total of only five MAPKKKs have been reported. MAPKKK, which gets activated by upstream signals, forms a very important component of MAPK cascade. In plants the roles MAPKKK has been identified in various stresses,[9,10] plant cytokinesis,[11,12] ethylene signalling,[13] innate immunity[14] and defence responses,[15,16,17] among many others.

However, the poor information about MAPKKK gene family in rice is proving to be bottleneck in elucidating MAPK cascade in this very important monocot crop. In the present study, an in silico search of rice genome databases was conducted to identify members of the rice MAPKKK gene family. A total of 75 genes were identified and among which 70 were novel. A phylogenetic tree was constructed and MAPKKKs were grouped into three different subfamilies. Conserved consensus motifs were analysed in all the subfamilies to support their association. To further validate the findings, expression analysis of all the 75 MAPKKKs in different stresses and tissue-specific libraries were carried out using MPSS database.

Materials and methods

Sequence and database search for OsMAPKKKs

In an attempt to obtain all the MAPKKKs, rice protein sequences available in Rice Genome Annotation Project (ftp://ftp.plantbiology.msu.edu/pub/data/Eukaryotic_Projects/ O_sativa/annotation_dbs/pseudomolecules/version_6.0) were downloaded to construct a local protein database. It comprised of 67393 sequences. This database was searched with MAPKKK query sequences downloaded from NCBI (ncbi.nlm.nih.gov), TAIR (arabidopsis.org), Royal Holloway University of London (www.rhul.ac.uk./Biological-Sciences/AcademicStaff/Bogre/mapkkk.htm), MIPS (mips.gsf.de), Uniprot (uniprot.org), BrassicaDB (brassica.bbsrc.ac.uk/BrassicaDB), MPSS (mpss.udel.edu/rice) and Maizegdb (maizegdb.org). The query consisted of 100 sequences from 13 different plant species which included Arabidopsis thaliana, Brassica napus, Cucumis melo, Cucumis sativus, Lotus japonicus, Lycopersicon esculentum, M. sativa, Nicotiana tabacum, O. sativa, Solanum chacoense, Triticum aestivum, Vitis vinifera and Z. mays. The search was carried out using BLASTP and 50% identity was taken as the threshold for the sequences obtained from BLAST analysis. Further all the MAPKKK sequences were aligned using CLUSTAL 2.0.3 (http://www.ebi.ac.uk/clustalw/) multiple sequence alignment and used to construct HMM profile. Using HMMER 2.3.2 (ftp://ftp.genetics.wustl.edu/pub/eddy/hmmer), the local protein database was searched on the basis of the HMM profile. HMMER and BLAST hits were compared and parsed by manual editing as well as using small PERL scripts. In addition self BLAST of the sequences was carried out to remove the redundancy and then they were analysed for the presence of domains. Subcellular localization prediction of each of the rice MAPKK kinases was carried out using the CELLO v2.5 server (http://cello.life.nctu.edu.tw/).

Phylogenetic tree construction

The Kinase domain of all the sequences (including both query and hits obtained) were analysed using ScanProsite program from Expasy server (au.expasy.org/prosite/). All the kinase domains were aligned using ClustalW. To investigate the evolutionary relationship among MAPKKK proteins, a phylogenetic tree was constructed by employing the minimal evolution (ME) method and the neighbour-joining (NJ) method wrapped in MEGA4 software suite.

Multiple alignment and conserved signature detection

Multiple sequence alignments of related proteins belonging to each group from both Arabidopsis and rice were performed using Multalin, which creates a multiple sequence alignment from a group of related sequences using progressive pairwise alignments (http://bioinfo.genotoul.fr/multalin/multalin). Programs INTERPROSCAN, SMART and MOTIF SCAN were employed to detect conserved domains.

Analysis of MPSS database for expression profiles

To gain insight into expression profiles of OsMAPKKK members in O. sativa in different stress and tissues, the MPSS database (http://mpss.udel.edu/rice/) was searched (opting 20-nt signature sequences) using the locus ID given in the TIGR database. The data thus obtained have been analysed and grouped based on stress and tissue specificity.

Results and discussion

In silico search and identification of novel MAPKKK genes

Availability of complete rice genome sequences[18] has made it possible for the first time to identify all the MAPKKK family members in this plant species. In order to identify the MAPKKK genes, 100 query sequences of MAPKKKs from different plant species were analysed by BLASTP against 67393 sequences of the local protein database of rice from RGP, which resulted in 122 hits as subject sequences. These hits were passed through several layers of filters which primarily included 50% identity with the query sequence which further reduced to 107 hits as subject sequences. This list included all the five MAPKKKs from rice, which were already reported in NCBI and MPSS database. Simultaneously an HMM profile (HMMER version 2.3.2) was created with the 100 query sequences and a profile search was carried out against the local rice proteome database which resulted in 1574 hits. After the comparison of the sequences obtained from BLAST hits and HMM searches, the number of MAPKKKs was further reduced to 102. A self BLAST of these sequences followed by manual editing to remove the redundancy finally resulted in identification of 75 MAPKKK genes. This analysis has revealed that the O. sativa genome has 75 putative MAPKKK genes including five already known MAPKKKs. However, in Arabidopsis 80 putative MAPKKKs were predicted.[6] Most of the MAPKKKs existence was supported by EST hits and full-length cDNA sequences from KOME or GenBank database (Table 1) except in 14 MAPKKKs where no such data were available. It indicates either all 14 are not expressed in all the conditions used for analysis or that they are expressed in very low quantity, which cannot be detected. Expression of other members was found in either of the two conditions mentioned above. MPSS database measures the absolute expression level of most genes in the sample and provide information about potentially novel transcripts.[19] Out of the 14 MAPKKKs which were not reported in EST or cDNA database, eight of them were represented in MPSS database with their expression levels in different tissues and stresses and in the remaining six MAPKKKs, five were without any expression in the tested conditions additionally OsMAPKKK53 was not represented in MPSS database. Since there was no standard nomenclature followed for MAPKKKs neither in Arabidopsis nor in rice, we named the MAPKKKs sequentially based on the HMM search output. All the 75 MAPKKKs were having conserved protein kinase domain which is backbone for MAPK family. The predicted protein localization of most of the MAPKKKs varied from cytoplasm, mitochondria, chloroplast to nucleus except in MAPKKK33 and MAPKKK47 where it was present in cytoskeleton and peroxisomes, respectively.

Table 1

List of MAPKKKs from rice

Sl. No.	MAPKKKs	Score	TIGR/MSU ID	Amino acid length	Total number of mapped ESTs	ID's of cDNAs	Subcellular localization	TIGR predicted function (from MPSS Db.)
1	MAPKKK1	529.3	LOC_Os03g06410	1017	74	AY167575 AK111595	Nuclear	EDR1, putative, expressed
2	MAPKKK2	522	LOC_Os10g29540	972	30	AK121718	Nuclear	EDR1, putative, expressed
3	MAPKKK3*	515.1	LOC_Os02g32610	781	25	CT835420	Chloroplast	Protein kinase domain containing protein, expressed
4	MAPKKK4	513.1	LOC_Os02g12810	864	31	N/A	Chloroplast	MAP3K delta-1 protein kinase, putative, expressed
5	MAPKKK5	512.9	LOC_Os12g37570	758	24	N/A	Nuclear	ATP binding protein, putative, expressed
6	MAPKKK6	497.4	LOC_Os02g50970	1111	42	AK102767	Nuclear	Protein kinase domain containing protein, expressed
7	MAPKKK7	491.2	LOC_Os06g12590	1078	17	AK099500 AK105681	Plastid	ATP binding protein, putative, expressed
8	MAPKKK8	489.3	LOC_Os11g10100	653	35	N/A	Nuclear	Mitogen-activated kinase kinase kinase alpha, putative, expressed
9	MAPKKK9	489.1	LOC_Os02g44642	894	22	AK073040	Nuclear	YDA, putative, expressed
10	MAPKKK10	485.3	LOC_Os04g47240	894	38	N/A	Nuclear	YDA, putative, expressed
11	MAPKKK11#	482.4	LOC_Os07g02 780	753	7	AK069889	Chloroplast	MAPKKK5, putative, expressed
12	MAPKKK12	482	LOC_Os09g39320	1220	6	N/A	Cytoplasm	CTR1-like protein kinase, putative, expressed
13	MAPKKK13	479.6	LOC_Os09g21510	848	0	N/A	Nuclear	Mitogen-activated protein kinase kinase kinase 2, putative, expressed
14	MAPKKK14	464.8	LOC_Os04g52 140	778	48	AK120898	Chloroplast	Protein kinase domain containing protein, expressed
15	MAPKKK15	463.4	LOC_Os08g32600	690	9	AK103087	Nuclear	Protein kinase domain containing protein, expressed
16	MAPKKK16	462.9	LOC_Os04g35700	708	38	AK061622 AK068725	Nuclear	Mitogen-activated kinase kinase kinase alpha, putative, expressed
17	MAPKKK17	461.8	LOC_Os09g37230	603	57	AK072690	Cytoplasmic	ATP binding protein, putative, expressed
18	MAPKKK18	461.6	LOC_Os03g55560	777	112	N/A	Chloroplast	MAPKKK5, putative, expressed
19	MAPKKK19	457.5	LOC_Os02g35010	690	9	AK287889	Mitochondria	MAP3KA, putative, expressed
20	MAPKKK20	453.9	LOC_Os07g38530	704	32	DQ837532 AK100426 AF080436	Nuclear	Protein kinase domain containing protein, expressed
21	MAPKKK21	446.7	LOC_Os07g25680	1219	6	AK241123	Nuclear	Protein kinase domain containing protein, expressed
22	MAPKKK22*	443.1	LOC_Os03g49640	654	15	N/A	Nuclear	Mitogen-activated protein kinase, putative, expressed
23	MAPKKK23	440.3	LOC_Os12g40279	4262	1	N/A	Plastid	Protein kinase domain containing protein, expressed
24	MAPKKK24#	440.1	LOC_Os04g56530	1357	54	AK099839	Cytoplasm	MAPKKK7, putative, expressed
25	MAPKKK25	438	LOC_Os02g38080	352	4	N/A	Cytoplasm	Mitogen-activated protein kinase kinase kinase 7, putative, expressed
26	MAPKKK26	437.6	LOC_Os07g29330	439	27	AK066198	Cytoplasm	Serine/threonine-protein kinase CTR1, putative, expressed
27	MAPKKK27	436.5	LOC_Os03g43760	379	98	AK243690	Cytoplasm	ATP binding protein, putative, expressed
28	MAPKKK28	436.1	LOC_Os03g15570	597	51	AK242766 CT835152 CT835203	Nuclear	Mitogen-activated protein kinase 1, putative, expressed
29	MAPKKK29	434.4	LOC_Os02g45130	612	42	AK073845	Chloroplast	WNK6, putative, expressed
30	MAPKKK30	434.3	LOC_Os02g02780	583	49	AK060220 AK103704	Cytoplasm	ATP binding protein, putative, expressed
31	MAPKKK31	429.9	LOC_Os01g45380	388	1	AK108130	Cytoplasm	ATP binding protein, putative, expressed
32	MAPKKK32	426.2	LOC_Os08g12750	418	103	AY156510 AK099003 AK111601 AK111800 AK121704 AY646225	Cytoplasm	Serine/threonine-protein kinase, putative, expressed
33	MAPKKK33	423.9	LOC_Os02g07790	421	83	AY156512 AK101327	Cytoskeleton	Serine/threonine-protein kinase CTR1, putative, expressed
34	MAPKKK34	423.4	LOC_Os05g50190	381	23	AY224453 AK107217	Nuclear	ATP binding protein, putative, expressed
35	MAPKKK35	419.3	LOC_Os02g54510	1083	2	N/A	Nuclear	Protein kinase domain containing protein, expressed
36	MAPKKK36	418.6	LOC_Os05g01780	621	40	AK070061	Nuclear	ZIK1 protein, putative, expressed
37	MAPKKK37	418.5	LOC_Os04g51950	422	54	AY156511 AK111698	Cytoplasm	Serine/threonine-protein kinase, putative, expressed
38	MAPKKK38	416.7	LOC_Os06g45300	428	49	AY224431 AK112024	Cytoplasm	Serine/threonine-protein kinase, putative, expressed
39	MAPKKK39	416.2	LOC_Os06g08280	1273	27	AK067771	Nuclear	Protein kinase domain containing protein, expressed
40	MAPKKK40	415.7	LOC_Os01g48330	801	66	AK070808 AK102209 AK111983	Nuclear	ATP binding protein, putative, expressed
41	MAPKKK41	408.5	LOC_Os06g43840	1112	49	AK243413	Nuclear	Protein kinase domain containing protein, expressed
42	MAPKKK42	405.8	LOC_Os03g60150	383	67	CT830832	Cytoplasm	Tyrosine-protein kinase 2, putative, expressed
43	MAPKKK43	400	LOC_Os06g50920	564	16	AK073747	Cytoplasm	ATP binding protein, putative, expressed
44	MAPKKK44	399.5	LOC_Os02g14530	790	50	AK111618	Nuclear	ATP binding protein, putative, expressed
45	MAPKKK45	399	LOC_Os06g43030	398	0	N/A	Chloroplast	HT1 protein kinase, putative, expressed
46	MAPKKK46	394.2	LOC_Os11g06140	439	3	AK070490	Nucleus	Serine/threonine-protein kinase WNK4, putative, expressed
47	MAPKKK47	380.6	LOC_Os07g08750	601	37	AK060552 AK100930	Peroxisome	WNK1, putative, expressed
48	MAPKKK48	371.7	LOC_Os01g01740	376	32	AK059460	Nuclear	HT1 protein kinase, putative, expressed
49	MAPKKK49	366.1	LOC_Os05g44290	604	3	N/A	Nuclear	Protein kinase, putative, expressed
50	MAPKKK50	363.3	LOC_Os12g02250	619	48	AK067447 AK072172	Nuclear	Mitogen-activated protein kinase, putative, expressed
51	MAPKKK51	361.1	LOC_Os01g54350	637	38	AK102467	Nuclear	Protein kinase, putative, expressed
52	MAPKKK52	357.4	LOC_Os12g06490	418	71	AK062812 AK073772	Cytoplasm	Serine/threonine-protein kinase WNK3, putative, expressed
53	MAPKKK53	355.8	LOC_Os11g02305	622	0	N/A	Nuclear	–
54	MAPKKK54	355.5	LOC_Os03g28300	859	63	CT828499	Nuclear	ATP binding protein, putative, expressed
55	MAPKKK55	354.1	LOC_Os01g50400	418	0	N/A	Chloroplast	Mitogen-activated protein kinase kinase kinase 1, putative, expressed
56	MAPKKK56	349.8	LOC_Os05g01780	621	20	AY336987	Mitochondria	ZIK1 protein, putative, expressed
57	MAPKKK57#	349.4	LOC_Os05g46750	591	0	N/A	Chloroplast	MAPKKK16, putative, expressed
58	MAPKKK58	348.8	LOC_Os03g39150	351	0	N/A	Cytoplasm	Serine/threonine-protein kinase CTR1, putative
59	MAPKKK59	346.4	LOC_Os12g41260	400	1	AK109696	Cytoplasm	ATMRK1, putative, expressed
60	MAPKKK60	341.8	LOC_Os03g53410	407	11	N/A	Chloroplast	ATMRK1, putative, expressed
61	MAPKKK61	334.7	LOC_Os01g10450	563	25	AK069537	Nuclear	Protein kinase, putative, expressed
62	MAPKKK62	329.7	LOC_Os01g50420	541	0	N/A	Mitochondria	Mitogen-activated protein kinase kinase kinase 2, putative, expressed
63	MAPKKK63	328.8	LOC_Os01g50370	484	0	N/A	Chloroplast	Mitogen-activated protein kinase kinase kinase 1, putative, expressed
64	MAPKKK64	325.1	LOC_Os07g39520	327	42	CT832200 AK072014 AK072176 AK105268	Mitochondria	Serine/threonine-protein kinase WNK2, putative, expressed
65	MAPKKK65	324	LOC_Os07g43900	321	12	AK241519	Cytoplasmic	Mitogen-activated protein kinase kinase kinase 10, putative, expressed
66	MAPKKK66	320.8	LOC_Os10g04010	525	0	N/A	Chloroplast	Mitogen-activated protein kinase kinase kinase 1, putative
67	MAPKKK67	318.7	LOC_Os10g04000	526	0	N/A	Cytoplasm	Mitogen-activated protein kinase kinase kinase 1, putative
68	MAPKKK68	315.8	LOC_Os12g30570	384	0	N/A	Nuclear	ATMRK1, putative
69	MAPKKK69	305.8	LOC_Os05g46760	441	8	AK105946	Chloroplast	Mitogen-activated protein kinase kinase kinase 1, putative, expressed
70	MAPKKK70	300.2	LOC_Os01g50410	451	0	N/A	Mitochondria	Mitogen-activated protein kinase kinase kinase 1, putative, expressed
71	MAPKKK71	278.4	LOC_Os02g21700	413	0	N/A	Cytoplasm	Mitogen-activated protein kinase kinase kinase 2, putative, expressed
72	MAPKKK72	265.4	LOC_Os01g54480	468	29	AK104674	Mitochondria	Serine/threonine-protein kinase TNNI3K, putative, expressed
73	MAPKKK73	258.8	LOC_Os03g18170	511	0	N/A	Chloroplast	Mitogen-activated protein kinase kinase kinase 1, putative, expressed
74	MAPKKK74	225	LOC_Os01g66860	500	13	AK070097	Chloroplast	Ankyrin-kinase, putative, expressed
75	MAPKKK75	212.5	LOC_Os02g39560	502	13	AK071922 AK099756	Chloroplast	Non-receptor tyrosine kinase spore lysis A, putative, expressed

*From NCBI database (MAPKKK3 and MAPKKK22).

#From MPSS database (MAPKKK11, MAPKKK24 and MAPKKK57).

Phylogenetic analysis of MAPKKKS

The family of MAPKKKs forms the largest group of MAPK pathway components. Arabidopsis contains 80 MAPKKKs,[6] which can be subdivided into three major subtypes, Raf, MEKK and ZIK.[20] Among these two groups, MEKK group is most similar to animal MEKKs and yeast MAPKKKs. In total, it consists of 21 MEKK-like and 11 ZIK kinases along with Arabidopsis ANP1-3 (Arabidopsis NPK1-like protein kinases). The other group consists of 48 genes encoding Raf-like protein kinases including Arabidopsis CTR1 and EDR1.[6] To further characterize the MAPKKKs from rice and to evaluate the phylogenetic relationships with Arabidopsis MAPKKKs, the kinase domains of rice and Arabidopsis were aligned using ClustalW and analysed using MEGA4. Phylogenetic tree was constructed by employing the NJ method (Fig. 1) and ME method have shown similar topologies, with only minor modifications at deep nodes. On the basis of phylogenetic analysis, MAPKKKs in rice were classified in to three categories, which include Raf, ZIK and MEKK subfamilies. There were 43 MAPKKKs from rice and 48 from Arabidopsis grouped under Raf subfamily, 22 MAPKKKs from rice and 21 from Arabidopsis were grouped in to MEKK subfamily where as only 10 MAPKKKs from rice and 11 from Arabidopsis were grouped under ZIK family. As the results indicate, rice and Arabidopsis show similarity in the number of MAPKKKs in each subgroup.

Figure 1

Phylogenetic tree and domain organization of MAPKKKs from rice and Arabidopsis. Kinase domains of putative MAPKKKs from Arabidopsis and rice were aligned using CLUSTALW program. The NJ phylogenetic tree was created using MEGA4 software suite. Bootstrap value above 50% was shown. The domain organization was depicted on the right, gaps in lines were introduced while depicting the protein sequences of bigger size for convenience. Scanning of the protein sequences for the presence of known motifs and domains was performed using PlantsP. To identify the species of origin for each MAPKKK, a species acronym is included before the protein name: At, Arabidopsis thaliana; Os, Oryza sativa.

Domain architecture of MAPKKKs from Arabidopsis and rice revealed that most of the Raf family proteins have a C-terminal kinase domain and a long N-terminal regulatory domain. In contrast, majority of the ZIK family members have N-terminal kinase domain whereas members of MEKK family has less conserved protein structure with kinase domain located either at N- or C-terminal or central part of the protein. Bipartite NLS and transmembrane regions are distributed across the members of all the subfamilies whereas extensin like region is observed in Raf and MEKK members. Ubiquitin-interaction motif and ACT domain which is known to play a role in the regulation of a wide range of metabolic enzymes by responding to amino acid concentration are present only in the members of Raf family from rice and Arabidopsis (Fig. 1).

Extrapolating the phylogenetic tree shown in Fig. 1 along with 14 additional sequences from 11 different plants revealed that most of the known MAPKKK from other plants are grouped with MEKK subfamily (Supplementary Fig. S1). Comparing MPK and MKK gene families in rice, Arabidopsis and Populus genomes it was reported that the recent duplication events in amplifying the respective gene families are more evident in eudicots compared with monocot rice.[8] We also observed comparing the MAPKKK gene family between rice and Arabidopsis that the eudicot has undergone more recent duplication events, more so in the Raf subfamily (Supplementary Fig. 1).

In silico localization of MAPKKKs on chromosomes indicated that all the 75 MAPKKKs are distributed on all 12 chromosomes of rice and half of them were present in first three chromosomes, which include chromosome 1, 2 and 3. Among all, chromosomes 8 and 11 were sharing only two MAPKKKs each, whereas chromosome 2 was having as many as 13 MAPKKKs (Fig. 2). Although similar analysis for other two components of MAPK cascade namely, MAPKK/MKK and MAPK/MPK revealed that eight MKKs are present in four chromosomes and 16 MPKs are distributed on seven chromosomes.

Figure 2

Graphical (scaled) representation of location of MAPK, MAPKK and MAPKKK genes on rice chromosomes.

Analysis of conserved motifs among MAPKKKs

The pattern of amino acid residues found in many subdomains is conserved among the family members. All the rice MAPKKKs that were grouped under Raf, ZIK and MEKK subfamilies were further analysed for the presence of specific signatures. Raf subfamily consists of the largest number of MAPKKKs in both Arabidopsis and rice. Experimental data from Drosophila and Caenorhabditis elegans have provided much evidence that the Rafs sensu stricto stimulate MAP2K and MAPK activation.[21] Human Rafs are involved in a signalling network that controls cell proliferation, cell differentiation and apoptosis. Many of their effects are transmitted through the ERK/MAPK pathway.[22,23]

Raf family has a conserved signature in its kinase domain across the members. Analysis of the rice MAPKKKs along with the Arabidopsis for Raf specific signature GTXX (W/Y) MAPE was carried out by multiple alignments of kinase domains. The data revealed the presence of signature in all the members of Raf family in rice (Fig. 3) and strongly supported their identity as members of Raf subfamily. Around 43 MAPKKKs were grouped under Raf subfamily in rice where as in Arabidopsis this number is higher and consists of 48 members. The kinase domain of human B-Raf is a strong activator of MEK and has a high affinity for MEK.[24,25] In Arabidopsis, members of Raf subfamily CTR1 and EDR1 act as negative regulators in ethylene signalling[13] and in response to powdery mildew attack,[26] respectively. Further CTR1 has been found to interact with the histidine kinase domain of ETR1 and the ethylene response sensor (ERS1) in vitro.[27]

Figure 3

Alignment of MAPKKKs of Raf subfamily from rice and Arabidopsis. The highlighted part shows the conserved signature motif.

The ZIK subfamily consists of 10 putative MAPKKKs in rice and 11 in Arabidopsis. Recently, the Arabidopsis ZIK protein WNK1 (At3g04910) was demonstrated to phosphorylate a protein involved in the control of circadian rhythms,[28] suggesting a function different from that of other MAPKKKs. The characteristic feature of this family consists of a conserved signature GTPEFMAPE (L/V/M) (Y/F/L) across the members. Rice putative MAPKKKs were analysed for the presence of the above signature, which reassured that 10 MAPKKKs out of 75 have ZIK specific signatures (Fig. 4) and eventually grouped under this family.

Figure 4

Alignment of MAPKKKs of ZIK subfamily from rice and Arabidopsis. The highlighted part shows the conserved signature motif.

Among the three families MEKK subfamily is relatively well characterized. NPK1 gene from tobacco whose role in cytokinesis have been established,[11,29] ScFRK2, a MAPKKK from S. chacoense involved in fertilization and embryogenesis,[30] OMTK1, MAPKKK from M. sativa which channels oxidative stress signalling[31] were some of the members of MEKK subfamily. Further ANP1 in Arabidopsis was found responsive to oxidative stress and is involved in negative regulation of auxin signal transduction pathway.[32,33] It was also reported that Arabidopsis ANP1, ANP2, ANP3 are involved in plant cytokinesis[11,34] and MAPKKKα in relation to defence response.[15] Role of YODA a MAPKKK from Arabidopsis was characterized in stomatal development[35] with its targeting downstream MAPKKs.[36] Another member of this family MEKK1 functions in integrating ROS homeostasis with plant development and hormone signalling.[37,38] Twenty-two MAPKKKs from rice and 21 from Arabidopsis belong to this subfamily. Relationship analysis based on the amino acid sequences of the protein kinase catalytic domain shows that the sequence, G (T/S) PX (F/Y/W) MAPEV forms a conserved signature of this family (Fig. 5). Presence of this signature in 18 putative rice MAPKKKs further confirmed their association with MEKK family. Moreover analysis of the transcript abundance of MAPKKK subfamilies from rice in different tissue and stress specific libraries revealed that they are differentially expressed and regulated.

Figure 5

Alignment of MAPKKKs of MEKK subfamily from rice and Arabidopsis. The highlighted part shows the conserved signature motif.

In silico analysis of expression of MAPKKKs based on MPSS database

MPSS is a valuable tool to have an insight into gene expression.[39] It has been used previously for genome-level expression analysis in several systems including Arabidopsis.[40] To extract information about the relative abundance of transcripts of O. sativa MAPKKK members, we have carried out the analysis in the available MPSS database (http://mpss.udel.edu/rice/). This database is derived from the TIGR O. sativa genome sequence and the search has been performed employing the 20 nucleotides long signatures in tissue-specific and stress-related libraries. Our analysis has revealed that among the different tissue-specific libraries, maximum numbers of MAPKKKs were expressed and their transcript abundance was found in crown vegetative meristematic tissue (NME), ovary and mature stigma (NOS) tissue libraries. Libraries of salinity and cold stress in rice roots have shown maximum transcript abundance of OsMAPKKK64 which is a member of ZIK subfamily (Fig. 6) and with drought stress elevated transcripts of OsMAPKKK28 was observed (Fig. 7). OsMAPKKK28 is a member of MEKK subfamily and an orthologue of AtMEKK1 from Arabidopsis. Similarly in leaves highest Transcripts Per Million (TPM) of OsMAPKKK4 a member of Raf subfamily in salinity stress (Fig. 8), OsMAPKKK63 in cold, OsMAPKKK8 in drought, all members of MEKK subfamily, were observed. This suggested the involvement of MAPKKK members from all the subfamilies and more precisely MEKK subfamily in the regulation of abiotic stress. These observations further strengthen their prediction and the differential regulation in various stress conditions suggests their active participation in stress signalling.

Figure 6

Transcript abundance of MAPKKK members of ZIK subfamily in salinity, drought, cold and tissue-specific libraries from MPSS database along with their dendrogram. Different libraries in MPSS database were analysed for the expression level of MAPKs. NYR, 14 days young roots; NSR, 14 days young roots stressed in 250 mM NaCl for 24 h; NDR, 14 days young roots stressed in drought for 5 days; NCR, 14 days young roots stressed in 4°C cold for 24 h; NYL, 14 days young leaves; NSL, 14 days young leaves stressed in 250 mM NaCl for 24 h; NDL, 14 days young leaves stressed in drought for 5 days; NCL, 14 days young leaves stressed in 4°C Cold for 24 h; NRA, mature roots (60 days) replicate A; NST, mature stem (60 days); NLA, mature leaves (60 days) replicate A; NME, Crown vegetative meristematic tissue (60 days); NPO, mature pollen; NOS, ovary and mature stigma; NIP, Immature panicle; NGS, 3 days germinating seed.

Figure 7

Transcript abundance of MAPKKK members of MEKK subfamily in salinity, drought, cold and tissue-specific libraries from MPSS database along with their dendrogram. Different libraries in MPSS database were analysed for the expression level of MAPKs. NYR, 14 days young roots; NSR, 14 days young roots stressed in 250 mM NaCl for 24 h; NDR, 14 days young roots stressed in drought for 5 days; NCR, 14 days young roots stressed in 4°C cold for 24 h; NYL, 14 days young leaves; NSL, 14 days young leaves stressed in 250 mM NaCl for 24 h; NDL, 14 days young leaves stressed in drought for 5 days; NCL, 14 days young leaves stressed in 4°C cold for 24 h; NRA, mature roots (60 days) replicate A; NST, mature stem (60 days); NLA, mature leaves (60 days) replicate A; NME, crown vegetative meristematic tissue (60 days); NPO, mature pollen; NOS, ovary and mature stigma; NIP, immature panicle; NGS, 3 days germinating seed.

Figure 8

Transcript abundance of MAPKKK members of Raf subfamily in salinity, drought, cold and tissue-specific libraries from MPSS database along with their dendrogram. Different libraries in MPSS database were analysed for the expression level of MAPKs. NYR, 14 days young roots; NSR, 14 days young roots stressed in 250 mM NaCl for 24 h; NDR, 14 days young roots stressed in drought for 5 days; NCR, 14 days young roots stressed in 4°C cold for 24 h; NYL, 14 days young leaves; NSL, 14 days young leaves stressed in 250 mM NaCl for 24 h; NDL, 14 days young leaves stressed in drought for 5 days; NCL, 14 days young leaves stressed in 4°C cold for 24 h; NRA, mature roots (60 days) replicate A; NST, mature stem (60 days); NLA, mature leaves (60 days) replicate A; NME, crown vegetative meristematic tissue (60 days); NPO, mature pollen; NOS, ovary and mature stigma; NIP, immature panicle; NGS, 3 days germinating seed.

Conclusion

The present study has provided the full list of MAPKKKs present in rice for the first time. In silico search of various rice protein databases using BLASTP and HMM profile resulted in identification of 75 MAPKKK genes from rice among these 70 were novel. EST hits and full-length cDNA sequences (from KOME or Genbank database) of 75 MAPKKKs supported their existence. Phylogenetic analysis of MAPKKKs from rice and Arabidopsis has classified them in to three subgroups that include Raf, ZIK and MEKK. Conserved motifs in the deduced amino acid sequences of rice MAPKKKs strongly supported their identity as members of Raf, ZIK and MEKK subfamilies. Expression analysis of the MAPKKKs in MPSS database revealed that maximum number of MAPKKK transcripts was represented in crown vegetative meristematic tissue (NME), ovary and mature stigma (NOS) tissue libraries. Libraries of salinity and cold stress in rice roots have shown maximum transcript abundance of OsMAPKKK64 which is a member of ZIK subfamily and with drought stress elevated transcripts of OsMAPKKK28 was observed. Similarly in leaves highest TPM of OsMAPKKK4 a member of Raf subfamily in salinity stress, OsMAPKKK63 in cold, OsMAPKKK8 in drought, which are members of MEKK subfamily were observed. The information generated will be very significant for further investigating the regulation mechanism of MAPKKKs and eventually MAPK cascade in response to extracellular stimuli and their central roles in various biological functions. Additionally, the information generated will serve the purpose in elucidating still very poorly characterized MAP kinase cascade in plants in general and in rice in particular.

Supplementary Data

Supplementary Data are available at www.dnaresearch.oxfordjournals.org.

Funding

The work is financially supported by the core grant of National Institute of Plant Genome Research, New Delhi, India.