Identification of genes from the ICE–CBF–COR pathway under cold stress in Aegilops–Triticum composite group and the evolution analysis with those from Triticeae

Abstract

Adverse environmental conditions limit various aspects of plant growth, productivity, and ecological distribution. To get more insights into the signaling pathways under low temperature, we identified 10 C-repeat binding factors (CBFs), 9 inducer of CBF expression (ICEs) and 10 cold-responsive (CORs) genes from Aegilops–Triticum composite group under cold stress. Conserved amino acids analysis revealed that all CBF, ICE, COR contained specific and typical functional domains. Phylogenetic analysis of CBF proteins from Triticeae showed that these CBF homologs were divided into 11 groups. CBFs from Triticum were found in every group, which shows that these CBFs generated prior to the divergence of the subfamilies of Triticeae. The evolutionary relationship among the ICE and COR proteins in Poaceae were divided into four groups with high multispecies specificity, respectively. Moreover, expression analysis revealed that mRNA accumulation was altered by cold treatment and the genes of three types involved in the ICE–CBF–COR signaling pathway were induced by cold stress. Together, the results make CBF, ICE, COR genes family in Triticeae more abundant, and provide a starting point for future studies on transcriptional regulatory network for improvement of chilling tolerance in crop.

Electronic supplementary material

The online version of this article (10.1007/s12298-017-0495-y) contains supplementary material, which is available to authorized users.

Introduction

Low temperature is a vital environmental factor that restricts growth, productivity, and ecological distribution of plants (Zhuang et al. 2015). Upon exposure to low non-freezing temperatures, many plants have evolved sophisticated mechanisms with the acquisition of freezing tolerance, a process known as cold acclimation (Thomashow 1999; Zhao et al. 2016). During this complex process, extensive changes are involved, including gene expression and regulation, biochemical and metabolic changes (Chinnusamy et al. 2007; Shi et al. 2015). Identifying genes involved in freezing tolerance is crucial for understanding how plants regulate growth at low temperature. Transcription factors and effector genes are precisely regulated in plant cells suffering from cold stress (Ma et al. 2014; Zhou et al. 2011). These regulation cascades or signaling pathways play pivotal roles in adaptation to cold stress (Wingler 2014). C-repeat binding factor/dehydration responsive element binding factor (CBF/DREB) are critical transcription factors that specifically interplay with the C-repeat (CRT)/dehydration-responsive element (DRE) and positively regulated the expression of downstream cold-responsive (COR) genes during cold acclimation (An et al. 2017; Liu et al. 1998; Park et al. 2015; Sakuma et al. 2002; Stockinger et al. 1997; Yamaguchi-Shinozaki and Shinozaki 1994). At the same time, CBF/DREB genes are activated by inducers of CBF expression (ICEs) through specific binding to the MYC recognition cis-elements (CANNTG) in the promoter (Chinnusamy et al. 2003). These entities constitute the signal regulatory pathway, ICE–CBF–COR transcriptional cascade, which mediates the cold acclimation process (Chinnusamy et al. 2007; Lissarre et al. 2010; Lu et al. 2017; Ryu et al. 2014; Zhu et al. 2007).

CBF/DREBs which belong to the AP2/ERF (APETALA2/Ethylene-Responsive Factor) superfamily have characteristic signature sequence motifs flanking the AP2 domain, PKK/RPAGRxKFxETRHP and DSAWR (An et al. 2017; Gilmour et al. 1998; Jaglo et al. 2001; Jia et al. 2016; Monroe et al. 2016; Park et al. 2015; Skinner et al. 2005). ICE proteins, member of the MYC family transcription factor (Lee et al. 2015; Lu et al. 2017), have a MYC basic helix–loop–helix (bHLH) domain and localize in the nucleus (Peng et al. 2014). CORs generally refer to the proteins encoded by all cold regulated genes (Lee et al. 1999; Uemura et al. 1996), including late embryogenesis abundant protein (LEA) (Tsuda et al. 2000), stress responsive protein (SRP) (Unpublished, from GenBank), cold induced (KIN), low temperature induced (LTI) (Dong and Pei 2014; Nordin et al. 1993; Welin et al. 1995), and so on. In Arabidopsis, three CBF genes are cold stress-inducible genes (Liu et al. 1998; Ryu et al. 2014; Stockinger et al. 1997), and the three CBFs are not redundant (Zhou et al. 2011). CBF1 and CBF3 are regulated in a different way from CBF2, and CBF2 is a negative regulator of CBF1 and CBF3 expression during cold acclimation (Novillo et al. 2004, 2007; Qin et al. 2011). Furthermore, the recent research using the CRISPR/Cas9 technology reveals the broad mechanism of CBF genes that CBF1 and CBF3 negatively regulate CBF2 expression (Zhao and Zhu 2016). Coordinately, ICE proteins have distinct functions rather than redundant functions (Rahman et al. 2014). In Arabidopsis, ICE1 can activate the expression of CBF3/DREB1A during cold acclimation (Chinnusamy et al. 2003), and ICE2 triggers the expression of CBF1 (Fursova et al. 2009). Finally, many COR genes are induced to resist ambient cold temperature. These elaborate signaling modules constitute a regulatory pathway, which enhances tolerance to cold stress in plants.

Many close wild relatives of common wheat in Triticeae carry abundant and valuable stress resistant genes to facilitate genetic improvement, and particularly in species from Aegilops with strong resistance to cold stress (Jia et al. 2013). Many CBF genes have been characterized from some Triticeae species, including 37 from hexaploid wheat (Triticum aestivum) (Badawi et al. 2007), 13 from Triticum monococcum (Miller et al. 2006), 10 from durum wheat (Triticum durum) (Leonardis et al. 2007), 20 from barley (Hordeum vulgare) (Skinner et al. 2005), 11 from rye (Secale cereale) (Siddiqua and Nassuth 2011), 4 from Brachypodium distachyon (Li et al. 2012; Ryu et al. 2014), 1 from Aegilops tauschii (Badawi et al. 2007) and 9 from Aegilops biuncialis (Unpublished, from GenBank). The previous research indicated that CBF14 gene in A. tauschii associated with resistance to freezing stress (Masoomi-Aladizgeh et al. 2015), however, the lack of essential functional domains of the CBF14 results in many uncertainties. In addition, the functions of CBF genes from A. biuncialis remain largely unknown. Some ICE, COR genes also have been identified from these Triticeae species. In Triticeae, 2 ICE genes and 5 COR genes were obtained from wheat (T. aestivum), including TaICE41, TaICE87, WCS120, Wcor410, Wcor14, WRAB15 and WRAB18 (Badawi et al. 2008; Ganeshan et al. 2008; Soltesz et al. 2013; Talanova et al. 2013), as well as 4 COR genes from barley (H. vulgare) including HVA1, DHN5, DHN8 and COR14b (Dal Bosco et al. 2003; Jeknic et al. 2014; Koag et al. 2009; Kosova et al. 2013). However, the deeper researches on ICE, COR genes in Triticeae are less. To get better understanding of ICE–CBF–COR pathway and its involvement in the chilling tolerance, more genes related to this regulatory network await identification.

In this study, we identified and characterized 10 CBFs, 9 ICEs and 10 CORs in Aegilops–Triticum composite group. The results suggest that all CBF, ICE, COR amino acid sequences contained specific and representative conserved domains. We analyzed phylogenetic relationship of different CBF genes in Triticeae, as well as ICE and COR genes in Poaceae. Moreover, expression analysis revealed that mRNA accumulation was altered by cold treatment and the genes of three types involved in the ICE–CBF–COR signaling pathway were induced by cold stress. This study concluded that CBF genes may be activated by ICE, and may induce the expression of COR genes to enhance plant tolerance to cold. Consequently, the results provide candidate genes and transcriptional regulatory network for improvement of chilling tolerance in Triticeae.

Materials and methods

Plant materials, growth condition and RNA isolation

Four Aegilops species together with one winter-hardy wheat cultivar (T. aestivum) were selected as research materials in this study, which are presented in Table 1. The winter-hardy wheat cultivar Mironovska 808 (abbreviated as M808) was originally screened for its strong cold tolerance in the Ukraine (Tsvetanov et al. 2000). All seeds were germinatied in a petri dish covered with distilled water at 28 °C, then were planted in plastic boxes filled with soil substrates and placed in a controlled growth chamber with the temperature set at 25 °C, a 16 h light/8 h dark photoperiod, and a humidity of 45%. Ten centimeter seedlings were treated with 4 °C cold stress for 6 h before being frozen in liquid nitrogen for total RNA isolation. Total RNA of samples was extracted by TaKaRa MiniBEST Plant RNA Extraction Kit (TaKaRa, Japan) according to the manufacturer’s instruction. First strand cDNA synthesis was performed using PrimeScript™ RT reagent Kit with gDNA Eraser (Takara), and stored at − 20 °C until use.

Table 1.

Research materials in this study

Species	Cultivar	Ploidy	Source	Origin
Triticum aestivum	Mironovska 808	6X	Shenyang Agriculture university, China	Ukraine
Triticum aestivum	China Spring	6X	Shenyang Agriculture university, China	China
Aegilops umbellulata	–	2X	National Clonal Germplasm Repository, USA	Turkey
Aegilops tauschii	–	2X	National Clonal Germplasm Repository, USA	Unknown
Aegilops speltoides	–	2X	National Clonal Germplasm Repository, USA	Turkey
Aegilops neglecta	–	4X	National Clonal Germplasm Repository, USA	France

Gene isolation and bioinformatic analyses

To isolate the Aegilops and T. aestivum CBF, ICE and COR genes, homologous sequences from other species in the GenBank were aligned with the Basic Local Alignment Search Tool (BLAST). A set of primers for gene amplification were designed by Primer 5 software (http://www.Premierbiosoft.com) (Table 2). The cDNA clone of all CBF and COR genes were amplified via Polymerase Chain Reaction (PCR) and then were sequenced to test whether they contain identical sequence or not. To obtain the whole sequence of ICE genes, 5′-RACE and 3′-RACE experiments were performed using a Full RACE Kit (TaKaRa) after the acquisition of the conserved sequence. To validate the whole coding sequence of ICE genes, primers were designed on the basis of the assembled results, and several clones of two independent PCR reactions were sequenced to correct errors introduced during PCR.

Table 2.

Primers in this study

Primer name	Forward primer sequence (5′–3′)	Rerverse primer sequence (5′–3′)	Purpose
CBF1	ATGGACGTCGCCGACG	TTAGTCAAACAAATAGCTCCATAGC	Amplification of CBFs
CBF2	GATGGACGTCGCCGACGC	TTAGTCGAACAAGTAGCTC
ICE-C	GAAGAA(A/G)GG(T/A/G)ATGCC(T/G)GC(T/C)AAGAA	GG(A/G)CACA(A/G)CTC(T/C)TCCTTGA(C/T)(A/G)CG	Amplification of the conserved sequence of ICEs
ICE-3F	CCAATAGCTTCCTCCCGTCGA	TCGCCTACATCGCGTTCTGGA	3′ RACE of ICEs	1st round
		CTAGACCGCTGGATGGAACCCC
ICE-3S	GCCGCATCAAGGAGGAGTTGTG	CTGGAGACCGGCGCAGTGCA		2nd round
		CCTACATCGTGGGATGGAACC
ICE-5F	AGGGCGTGATGGAGAACTCGG	TGTGCAGGAGCTCCTTGAGGTA	5′ RACE of ICEs	1st round
	CGATGCTTGATGACGACAGCT
ICE-5S	GGGGGCGACTGGGGGTACCT	AAGCCCTGTCCATTTTGCTGATC		2nd round
	CGATGCTTCCCGCTCCTCAACC
ICE-IF1	GCAGAGATAAGTTGCAGTAACTGA	GCGAGCCTCATCCCATCA	Verification of the whole coding sequence of ICEs	1st round
ICE-IS1	GAACTTGCCAAGAATCCGT	AAAGGAGCGGAACAGCACA		2nd round
ICE-IF2	GCTATGCTGTCGCAGTTCAATAG	AAACCATGGGTACTTATTGCAAAT		1st round
ICE-IS2	TTGGGTGCAGGAAGGCGA	TTCCTTTGGTATGCTGTCCTCT		2nd round
COR1	ATGTCTGGTTGGTTTCAAGAGAAG	TCACCCTCCCATCCCGAG	Amplification of CORs
COR2	ATGTCGGGTTGGTTCGGGGGTAAC
18S	TGAGAAACGGCTACCACATC	TCGGCATCGTTTATGGTTGA	qRT-PCR analysis
CBF4	AGCCCGAGCAACACTCTTT	GCGCCAAGCTGTCGTAGTA
CBF9	GCAGGGTATGCTCGTGTCA	GTCGAACAAGCAGCTCCATA
ICE41	AGGGGCAGGCGGTGAACA	GGCGAAGCCATCGAAGCAG
ICE87	AAGGCCTTGGTCTTGACGTG	CGCCTTGATTTCCTCAGCCT
COR1	TGAGAAAAGCGAGGCTGTCA	TGATCCTTACCGGCTTCCAC
COR2	GACCAGACGTTCGGCTTCTT	CGGTCTCCTCGACACACTTT

Multiple sequence alignments were done using DNAMAN 8 (http://www.lynnon.com) and MEGA 5.10 (http://www.megasoftware.net/mega.php). Conserved motifs were identified using MEME Suite 4.10.1 (http://meme-suite.org/) and WebLogo 3.4 (http://weblogo.threeplusone.com/). To compare evolutionary relationship of the isolated Triticeae CBF, ICE, COR family members, the phylogenetic trees were constructed by MEGA 5.10 (http://www.megasoftware.net/mega.php) based on the neighbor-joining (NJ) method and bootstrap analysis (1000 replicates). Highly similar homologous genes were downloaded from NCBI (http://www.ncbi.nlm.nih.gov/) and presented in Tables 3, 4, and 5.

Table 3.

Description of CBF genes used in the study

Proposed gene name	Gene name	Gene accession	Protein accession	Gene fragment	Protein fragment	Species	References
TaCBFIVd-4A	TaCBFIVd-4A	KY931522		672	223	Triticum aestivum	This study
AeuCBFIVd-4	AeuCBFIVd-4	KY931528		672	223	Aegilops umbellulata	This study
AetCBFIVd-4	AetCBFIVd-4	KY931534		672	223	Aegilops tauschii	This study
AesCBFIVd-4	AesCBFIVd-4	KY931540		672	223	Aegilops speltoides	This study
AenCBFIVd-4	AenCBFIVd-4	KY931546		678	225	Aegilops neglecta	This study
TaCBFIVd-9D	TaCBFIVd-9D	KY931521		810	269	Triticum aestivum	This study
AeuCBFIVd-9	AeuCBFIVd-9	KY931527		765	254	Aegilops umbellulata	This study
AetCBFIVd-9	AetCBFIVd-9	KY931533		765	254	Aegilops tauschii	This study
AesCBFIVd-9	AesCBFIVd-9	KY931539		765	254	Aegilops speltoides	This study
AenCBFIVd-9	AenCBFIVd-9	KY931545		765	254	Aegilops neglecta	This study
AebCBFIVd-4	AebCBF7	FR719739	CBX87021	555	184	Aegilops biuncialis	Unpublished, 2010
AebCBFIVd-9	AebCBF9	FR719741	CBX87023	768	255	Aegilops biuncialis	Unpublished, 2010
AebCBFIIIa-6	AebCBF1	FR719733	CBX87015	729	242	Aegilops biuncialis	Unpublished, 2010
AebCBFIIIc-3	AebCBF2	FR719734	CBX87016	741	246	Aegilops biuncialis	Unpublished, 2010
AebCBFIIIc-10	AebCBF3	FR719735	CBX87017	714	237	Aegilops biuncialis	Unpublished, 2010
AebCBFIIId-12	AebCBF4	FR719736	CBX87018	720	239	Aegilops biuncialis	Unpublished, 2010
AebCBFIIId-19	AebCBF5	FR719737	CBX87019	711	236	Aegilops biuncialis	Unpublished, 2010
AebCBFIVa-2	AebCBF6	FR719738	CBX87020	678	225	Aegilops biuncialis	Unpublished, 2010
AebCBFIVd-22	AebCBF8	FR719740	CBX87022	855	284	Aegilops biuncialis	Unpublished, 2010
AtCBFV-1	AtCBF1	U77378	AAC49662	642	213	Arabidopsis thaliana (thale cress)	Stockinger et al. (1997)
AtCBFV-2	AtCBF2	AF074601	AAD15976	651	216	Arabidopsis thaliana (thale cress)	Gilmour et al. (1998)
AtCBFV-3	AtCBF3	AF074602	AAD15977	651	216	Arabidopsis thaliana (thale cress)	Gilmour et al. (1998)
BdCBFIIId-1	BdCBF1	JQ180470	AFD96407	735	244	Brachypodium distachyon (stiff brome)	Li et al. (2012)
BdCBFIIId-2	BdCBF2	JQ180470	AFD96408	777	258	Brachypodium distachyon (stiff brome)	Li et al. (2012)
BdCBFIIId-3	BdCBF3	JQ180470	AFD96409	753	250	Brachypodium distachyon (stiff brome)	Li et al. (2012)
BdCBFIIId-4	BdCBF4	JQ180470	AFD96410	747	248	Brachypodium distachyon (stiff brome)	Li et al. (2012)
HvCBFIa-1	HvCBF1	AF418204	AAL84170	654	217	Hordeum vulgare subsp. vulgare (domesticated barley)	Xue (2002)
HvCBFIVa-2	HvCBF2	AF442489	AAM13419	666	221	Hordeum vulgare subsp. vulgare (domesticated barley)	Xue (2003)
HvCBFIIIc-3	HvCBF3	AF239616	AAG59618	750	249	Hordeum vulgare subsp. vulgare (domesticated barley)	Choi et al. (2002)
HvCBFIVd-4	HvCBF4B	AY785848	AAX28948	678	225	Hordeum vulgare subsp. vulgare (domesticated barley)	Skinner et al. (2006)
HvCBFII-5	HvCBF5	AY785857	AAX23700	645	214	Hordeum vulgare subsp. vulgare (domesticated barley)	Skinner et al. (2006)
HvCBFIIIa-6	HvCBF6	EU331996	ACA29480	735	244	Hordeum vulgare subsp. vulgare (domesticated barley)	251
HvCBFV-7	HvCBF7	AY785866	AAX23706	660	219	Hordeum vulgare subsp. vulgare (domesticated barley)	Skinner et al. (2006)
HvCBFIVd-9	HvCBF9	AY785880	AAX23709	876	291	Hordeum vulgare subsp. vulgare (domesticated barley)	Skinner et al. (2006)
HvCBFIIIc-10A	HvCBF10A	AY785884	AAX23713	684	227	Hordeum vulgare subsp. vulgare (domesticated barley)	Skinner et al. (2006)
HvCBFIIIc-10B	HvCBF10B	AY785887	AAX23716	723	240	Hordeum vulgare subsp. vulgare (domesticated barley)	Skinner et al. (2006)
HvCBFIa-11	HvCBF11	AY785892	AAX23720	657	218	Hordeum vulgare subsp. vulgare (domesticated barley)	Skinner et al. (2006)
HvCBFIIId-12	HvCBF12	DQ095157	ABA01491	735	244	Hordeum vulgare subsp. vulgare (domesticated barley)	Skinner et al. (2005)
HvCBFIIIc-13	HvCBF13	DQ095158	ABA01492	759	252	Hordeum vulgare subsp. vulgare (domesticated barley)	Skinner et al. (2005)
HvCBFIVc-14	HvCBF14	DQ095159	ABA01493	645	214	Hordeum vulgare subsp. vulgare (domesticated barley)	Skinner et al. (2005)
ScCBFIa-11	ScCBFIa-11	EU194240	ABY59777	615	204	Secale cereale (rye)	Campoli et al. (2009)
ScCBFII-5	ScCBFII-5	EU194241	ABY59778	642	213	Secale cereale (rye)	Campoli et al. (2009)
ScCBFIIIa-6	ScCBFIIIa-6	EU194242	ABY59779	690	229	Secale cereale (rye)	Campoli et al. (2009)
ScCBFIIIc-10	ScCBFIIIc-10	EU194243	ABY59780	666	221	Secale cereale (rye)	Campoli et al. (2009)
ScCBFIIIc-3A	ScCBFIIIc-3A	EU194244	ABY59781	687	228	Secale cereale (rye)	Campoli et al. (2009)
ScCBFIIIc-3B	ScCBFIIIc-3B	EU194245	ABY59782	696	231	Secale cereale (rye)	Campoli et al. (2009)
ScCBFIIId-12	ScCBFIIId-12	EU194246	ABY59783	699	232	Secale cereale (rye)	Campoli et al. (2009)
ScCBFIIId-15	ScCBFIIId-15	EU194247	ABY59784	663	220	Secale cereale (rye)	Campoli et al. (2009)
ScCBFIIId-19	ScCBFIIId-19	EU194248	ABY59785	666	221	Secale cereale (rye)	Campoli et al. (2009)
ScCBFIVa-2A	ScCBFIVa-2A	EU194249	ABY59786	582	193	Secale cereale (rye)	Campoli et al. (2009)
ScCBFIVa-2B	ScCBFIVa-2B	EU194250	ABY59787	582	193	Secale cereale (rye)	Campoli et al. (2009)
TaCBFIa-11	TaCBFIa-A11	EF028751	ABK55354	657	218	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFII-5A	TaCBFII-5.1	EF028752	ABK55355	648	215	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFII-5B	TaCBFII-5.2	EF028753	ABK55356	660	219	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFII-5C	TaCBFII-5.3	EF028754	ABK55357	657	218	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIIIa-6B	TaCBFIIIa-6.1	EF028755	ABK55358	711	236	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIIIa-6C	TaCBFIIIa-6.2	EF028756	ABK55359	729	242	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIIIa-6A	TaCBFIIIa-D6	EF028757	ABK55360	717	238	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIIIc-3B	TaCBFIIIc-3.1	EF028758	ABK55361	708	235	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIIIc-3C	TaCBFIIIc-3.2	EF028759	ABK55362	741	246	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIIIc-10	TaCBFIIIc-B10	EF028761	ABK55364	723	240	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIIIc-3A	TaCBFIIIc-D3	EF028760	ABK55363	738	245	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIIId-12B	TaCBFIIId-12.1	EF028762	ABK55365	738	245	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIIId-15B	TaCBFIIId-15.2	EF028765	ABK55368	726	241	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIIId-15A	TaCBFIIId-A15	EF028764	ABK55367	720	239	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIIId-19B	TaCBFIIId-A19	EF028766	ABK55369	705	234	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIIId-12A	TaCBFIIId-B12	EF028763	ABK55366	738	245	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIIId-19C	TaCBFIIId-B19	EF028767	ABK55370	705	234	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIIId-19A	TaCBFIIId-D19	EF028768	ABK55371	705	234	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVa-2C	TaCBFIVa-2.2	EF028770	ABK55373	693	230	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVa-2A	TaCBFIVa-2.3	EF028771	ABK55374	579	192	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVa-2B	TaCBFIVa-A2	EF028769	ABK55372	678	225	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVb-21A	TaCBFIVb-21.1	EF028775	ABK55378	609	202	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVb-20A	TaCBFIVb-A20	EF028772	ABK55375	654	217	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVb-20B	TaCBFIVb-B20	EF028773	ABK55376	639	212	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVb-20C	TaCBFIVb-D20	EF028774	ABK55377	639	212	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVb-21B	TaCBFIVb-D21	EF028776	ABK55379	609	202	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVc-14C	TaCBFIVc-14.1	EF028777	ABK55380	639	212	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVc-14A	TaCBFIVc-14.3	EF028779	ABK55382	645	214	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVc-14 B	TaCBFIVc-B14	EF028778	ABK55381	645	214	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVd-4C	TaCBFIVd-4.1	EF028780	ABK55383	669	222	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVd-9B	TaCBFIVd-9.1	EF028782	ABK55385	810	269	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVd-22 A	TaCBFIVd-A22	EF028785	ABK55388	828	275	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVd-22 B	TaCBFIVd-B22	EF028786	ABK55389	873	290	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVd-4B	TaCBFIVd-B4	EF028781	ABK55384	669	222	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVd-9A	TaCBFIVd-B9	EF028783	ABK55386	810	269	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVd-22 C	TaCBFIVd-D22	EF028787	ABK55390	828	275	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TaCBFIVd-9C	TaCBFIVd-D9	EF028784	ABK55387	810	269	Triticum aestivum (bread wheat)	Badawi et al. (2007)
TmCBFIIIc-3	TmCBF3	AY951949	AAY32553	741	246	Triticum monococcum	Miller et al. (2006)
TmCBFII-5	TmCBF5	AY951947	AAY32551	633	210	Triticum monococcum	Miller et al. (2006)
TmCBFV-7	TmCBF7	AY785904	AAX28965	828	275	Triticum monococcum	Skinner et al. (2005)
TmCBFIIIc-10	TmCBF10	AY951950	AAY32554	720	239	Triticum monococcum	Miller et al. (2006)
TmCBFIIId-12	TmCBF12	EU076381	ABW87011	744	247	Triticum monococcum	Knox et al. (2008)
TmCBFIIIc-13	TmCBF13	AY951951	AAY32555	720	239	Triticum monococcum	Miller et al. (2006)
TmCBFIVc-14A	TmCBF14	EU076382	ABW87012	639	212	Triticum monococcum	Knox et al. (2008)
TmCBFIVc-14B	TmCBF14	AY951948	AAY32552	639	212	Triticum monococcum	Miller et al. (2006)
TmCBFIIId-15	TmCBF15	EU076383	ABW87013	726	241	Triticum monococcum	Knox et al. (2008)
TmCBFIIId-16	TmCBF16	EU076384	ABW87014	864	287	Triticum monococcum	Knox et al. (2008)
TmCBFIIIb-18	TmCBF18	AY951946	AAY32550	738	245	Triticum monococcum	Miller et al. (2006)

Table 4.

Description of ICE genes used in the study

Proposed gene name	Gene name	Gene accession	Protein accession	Gene fragment	Protein fragment	Species	References
AeuICE1	AeuICE1	KY931524		1149	382	Aegilops umbellulata	This study
AetICE1	AetICE1	KY931530		1151	383	Aegilops tauschii	This study
AesICE1	AesICE1	KY931536		1149	382	Aegilops speltoides	This study
AenICE1	AenICE1	KY931542		1151	382	Aegilops neglecta	This study
TaICE2A	TaICE2A	KY931518		1329	442	Triticum aestivum	This study
AeuICE2	AeuICE2	KY931523		1331	443	Aegilops umbellulata	This study
AetICE2A	AetICE2A	KY931529		1331	443	Aegilops tauschii	This study
AesICE2	AesICE2	KY931535		1331	443	Aegilops speltoides	This study
AenICE2	AenICE2	KY931541		1331	443	Aegilops neglecta	This study
AetICE2B	AetICE1-like	XM_020312906	XP_020168495	1503	500	Aegilops tauschii	RefSeq, 2017
AtICE1	AtICE1	AY195621	AAP14668	1485	494	Arabidopsis thaliana	Chinnusamy et al. (2003)
AtICE2	AtICE2	KM210288	AIU34717	1335	444	Arabidopsis thaliana	Unpublished, 2014
BdICE1	BdICE1	XM_003567379	XP_003567427	1116	371	Brachypodium distachyon (stiff brome)	RefSeq, 2015
HvICE2	HvICE2	DQ151537	ABA25897	579	192	Hordeum vulgare subsp. vulgare (domesticated barley)	Skinner et al. (2006)
OsICE1	OsICE1	XM_015795517	XP_015651003	1239	412	Oryza sativa Japonica Group (Japanese rice)	RefSeq, 2016
SbICE1	SbICE1-like	XM_002458977	XP_002459022	1131	376	Sorghum bicolor (sorghum)	Paterson et al. (2009)
SiICE1	SiICE1-like	XM_004971094	XP_004971151	1131	376	Setaria italica (foxtail millet)	RefSeq, 2015
TaICE1	TaICE41	EU562183	ACB69501	1146	381	Triticum aestivum (bread wheat)	Badawi et al. (2008)
TaICE2B	TaICE87	EU562184	ACB69502	1332	443	Triticum aestivum (bread wheat)	Badawi et al. (2008)
ZmICE2	ZmICE2	EU974475	ACG46593	1131	376	Zea mays	Alexandrov et al. (2009)

Table 5.

Description of COR genes used in the study

Proposed gene name	Gene name	Gene accession	Protein accession	Gene fragment	Protein fragment	Species	References
TaCOR1	TaCOR1	KY931520		515	163	Triticum aestivum	This study
AeuCOR1	AeuCOR1	KY931526		518	164	Aegilops umbellulata	This study
AetCOR1	AetCOR1	KY931532		518	164	Aegilops tauschii	This study
AesCOR1	AesCOR1	KY931538		518	164	Aegilops speltoides	This study
AenCOR1	AenCOR1	KY931544		518	164	Aegilops neglecta	This study
TaCOR2	TaCOR2	KY931519		528	166	Triticum aestivum	This study
AeuCOR2	AeuCOR2	KY931525		525	165	Aegilops umbellulata	This study
AetCOR2	AetCOR2	KY931531		528	166	Aegilops tauschii	This study
AesCOR2	AesCOR2	KY931537		528	166	Aegilops speltoides	This study
AenCOR2	AenCOR2	KY931543		528	166	Aegilops neglecta	This study
AcLEA	AcLEA	JN007028	AEJ88291	630	209	Agropyron cristatum	Unpublished, 2011
AetPHV A1-like	AetPHV A1-like	XM_020294635	XP_020150224	501	166	Aegilops tauschii	RefSeq, 2017
AtCOR15a	AtCOR15a	U01377	AAB87705	420	139	Arabidopsis thaliana	Baker et al. (1994)
AtCOR6.6	AtCOR6.6	X55053	CAA38894	201	66	Arabidopsis thaliana	Gilmour et al. (1992)
EtLEA3	EtLEA3	KJ123698	AIZ11400	600	199	Eremopyrum triticeum	Sun et al. (2014)
HvES2A	HvES2A	X79466	CAA55976	534	177	Hordeum vulgare	Speulman and Salamini (1995)
OsLEA-like	OsLEA-like	NM_001062730	NP_001056195	603	200	Oryza sativa Japonica Group (Japanese rice)	Rice Annotation et al. (2008)
PpLEA3	PpLEA3	GU947647	ADF36679	552	183	Pogonatherum paniceum	Wang et al. (2012)
TaLEA	TaLEA	HQ718763	ADW41578	570	189	Triticum aestivum (bread wheat)	Min et al. (2012)
TaSRP3	TaSRP3	JQ923472	AFN10738	570	189	Triticum aestivum (bread wheat)	Unpublished, 2012
WCOR615	WCOR615	U73217	AAB18208	528	175	Triticum aestivum (bread wheat)	Unpublished, 1996
Wrab17	Wrab17	AF255053	AAF68628	501	166	Triticum aestivum (bread wheat)	Tsuda et al. (2000)
Wrab19	Wrab19	AF255052	AAF68627	540	179	Triticum aestivum (bread wheat)	Tsuda et al. (2000)

Expression analysis of ICE–CBF–COR genes

To detect the expression of ICE, CBF, COR genes under cold stress, A. tauschii, Aegilops neglecta, winter-hardy wheat M808 and spring wheat T. aestivum L. cultivar China Spring (CS) were placed in growth chambers under long-day conditions (16 h light, 8 h dark) at 25 °C before cold stress. Then, the plants were removed from the growth chambers to a 4 °C cold room, and the leaves were harvested at intervals of 0, 2, 4, 8, 12, 24 h after cold treatment. Total RNA was extracted by TaKaRa MiniBEST Plant RNA Extraction Kit (TaKaRa). First strand cDNA synthesis was derived using PrimeScript™ RT reagent Kit with gDNA Eraser (Takara). Gene-specific primers (Table 2) were designed according to the 3′-untranslated regions (UTR) for qPCR amplification. 18S (AJ315041) was selected as the reference gene from the GenBank database. According to the instructions of the SYBR R Premix Ex Taq™ (Takara), qRT-PCR was performed using an iQ5 Real-Time PCR System (Bio-Rad, USA). The relative expression level of genes were calculated using iQ5 software and the Microsoft Excel program. Mean values and standard errors of three biological replicates were presented in this analysis.

Results

Identification of CBF genes and nomenclature

To enrich the gene family of Triticeae cereal CBF, a total of 10 distinct CBF genes have been isolated from five materials (Aegilops species and T. aestivum) (Table 3) using PCR amplification with degenerate primers (CBF1 and CBF 2, Table 2). The gene names and GenBank accession numbers are summarized in Table 3. Comparative analysis of the cDNA sequence with genomic sequence confirmed that CBF genes are intronless. The identified CBFs have amino acid residue numbers ranging from 223 to 269. BLAST searches using the reported CBF sequences showed that all CBFs displayed strong similarity to the Triticeae cereal CBFs. Following homology comparison and phylogenetic analysis (CBFIa, CBFII, CBFIIIa, CBFIIIb, CBFIIIc, CBFIIId, CBFIVa, CBFIVb, CBFIVc, and CBFIVd) with the established nomenclature of CBFs from H. vulgare, S. cereale, T. aestivum, and T. monococcum (Badawi et al. 2007; Miller et al. 2006; Skinner et al. 2005), we assigned consistent gene numbers to these identical orthologous genes which belong to the same gene subfamily (Table 3). When the same gene number of several genes arose, a letter designating its order trails CBF gene number (e.g. TaCBFIVd-4A).

Phylogenetic analysis of CBF genes in Triticeae

To establish the evolutionary relationship among the different CBFs in Triticeae, phylogenetic analysis was performed using Triticeae CBF homologs (Table 3). The results revealed that these CBF homologs were divided into 11 monophyletic groups (CBFIa, CBFII, CBFIIIa, CBFIIIb, CBFIIIc, CBFIIId, CBFIVa, CBFIVb, CBFIVc, CBFIVd, and CBFV) (Fig. 1). All 10 CBF members isolated in this study belonged to Group IVd, with another 8 CBFs from T. aestivum, 3 CBFs from A. biuncialis and 2 CBFs from H. vulgare. In Group IVd, three subgroups were presented (CBFIVd-9, CBFIVd-22 and CBFIVd-4), and 5 genes isolated in this study belonged to Subgroup CBFIVd-9, while another 5 belonged to CBFIVd-4.

Fig. 1.

Phylogenetic analysis of CBF proteins in Triticeae. The tree was constructed using the neighbor-joining (NJ) method with a Poisson correction model and 1000 bootstrap replicates. The tree was divided into 11 groups, and the solid points before the sequence names indicate the gene isolated in this study

In the phylogenetic analysis, only one group contains a single member (TmCBFIIIb-18), which was designated Group IIIb. Group IVb is the only one that contains CBFs only from T. aestivum. Group IIId is the only one that contains CBFs from the whole five Triticeae subfamilies used in this study. Groups IIIa, IIIc and IVa are the ones that contain CBFs from four Triticeae subfamilies (Aegilops, Hordeum, Triticum, Secale), suggesting that the ancestral CBFIIIa, CBFIIIc and CBFIVa genes were already present before divergence of these subfamilies. Similarly, Groups Ia and II are the ones that contain CBFs from three Triticeae subfamilies (Hordeum, Triticum, Secale), manifesting that the ancestral CBFIa and CBFII genes were already present before divergence of these subfamilies. In addition, 5 CBF homologues belonging to Group V are somewhat distant from those belonging to the rest of the groups, and it may reveal that sequence discrepancies of amino acids play a critical role in the classification reflecting their evolutionary relationships. CBFs from Triticum were found in every group, showing that these CBFs generated prior to the divergence of Triticum and other members of Triticeae. At present, all four genes from B. distachyon were found in the CBFIIId group, indicating that BdCBFs had appeared inferior to the radiation of these subfamilies.

Sequence structure analysis of CBF proteins

To identify some of the structural differences associated with the CBF genes, the genes involved in the subgroup CBFIVd-9 and CBFIVd-4 were selected to perform multiple sequences alignments on the basis of the evolutionary relationship of Triticeae CBFs. Conserved amino acids analysis was performed to highlight conservation of domains by MEME and WebLogo. It was showed that all CBF amino acid sequences contained an AP2 DNA binding domain in the N-terminal and an acidic C-terminal domain (Fig. 2a). The AP2 DNA binding domain directly flanked by two special signature motifs, which were used to distinguish the CBF subfamily from other members of the AP2/ERF superfamily (Fig. 2b). The former is a leader sequence of 16 amino acids (PKRPAGRTKFxExRHP) and the latter is the DSAWR motif, with the exception of AenCBFIVd-4 possessing the first amino acid position of an alanine (A) in place of the more common aspartic acid (D). Analysis of all placed CBFs revealed that three distinct group CBF genes were present with the C-terminal sequence discrepancies (Fig. 2a, c). As shown in the Fig. 2a, c, the Cluster I includes AenCBFIVd-9, AesCBFIVd-9, AetCBFIVd-9 and AeuCBFIVd-9 with the presence of fragment 2, 3, but the absence of fragment 4. The rest of members from the subgroup CBFIVd-9 belonged to the Cluster II with fragment 1, 2, 3 but no fragment 4. All CBF amino acid sequences from the subgroup CBFIVd-4 fall into the Cluster III with the unique fragment 4. Another conserved domain is typically present in the C-terminal, the LWSC/Y motif (Fig. 2c). Members of the subgroup CBFIVd-9 (Cluster I and Cluster II) have a preference for cysteine (C) at the end of this motif, compared to the tyrosine (Y) of the subgroup CBFIVd-4 (Cluster III). The phylogenetic relationship, structural conservation and consistent function of these CBF proteins would reflect the close genetic relationship among Triticeae.

Fig. 2.

Sequence structure analysis of CBF proteins. a Structure schematic of CBF proteins. All CBF amino acid sequences were divided into two parts including the N-terminal and the C-terminal domain. In the N-terminal, all CBF proteins contained an AP2 DNA binding domain, signature 1 (PKRPAGRTKFxExRHP) and signature 2 (DSAWR). In the acidic C-terminal, three distinct groups were present depending on the C-terminal sequence discrepancies (fragment 1, 2, 3, 4), and all CBF proteins contained another conserved domain (LWSC/Y). The length of schemes does not reflect protein size. b Details of the N-terminal structure of CBF proteins. All characteristics were marked in the figture. c Details of the C-terminal structure of CBF proteins. All characteristics were marked in the figture

Identification of ICE genes and phylogeny of ICE genes in Poaceae

As master regulators of CBF expression, ICEs bind to the promoter of CBF genes, and positively induce the expression of CBF genes. We therefore isolated ICEs from the above five contextual materials. Firstly, a partial conserved cDNA fragment of 230 bp was produced. Then 5′ end (650 bp) and 3′ end (360 bp) of ICE were obtained. The assembled whole coding sequences of ICE genes were verified by several clones and sequencing of two independent PCR reactions. We identified 9 potential ICE homologues, and the gene names and GenBank accession numbers were summarized in Table 4. Nomenclature of ICEs was similar to the previous description, but group information of phylogenetic analysis was not displayed in the name. These ICEs encode proteins ranging from 382 to 443. ICE1s shared higher homology with the published TaICE1 (ACB69501), but ICE2s shared lower homology with the TaICE1. Similarly, ICE2s shared higher homology with the published TaICE2B (ACB69502), but ICE1s shared lower homology with the TaICE87. By further analysis, we found that ICE1s and ICE2s had low homology with the full amino acid sequences for less conservative N-terminal sequences. However, these two groups of proteins are characterized by the presence of the identically C-terminal conserved regions. To get rid of the influence of less conservative N-terminal sequences, we retained the conserved C-terminal domain to perform the phylogenetic analysis.

Various ICEs from Poaceae and two ICEs in Arabidopsis were selected to establish the relationship, and these candidate proteins were listed in Table 4. The evolutionary relationship among the ICE proteins showed that these ICEs were divided into 4 groups (Group I, Group II, Group III, Group V) (Fig. 3). Four ICE1s isolated in this study belong to Group I with the published TaICE1 and HvICE2, but five ICE2s isolated in this study belong to Group V with the published TuICE1, AetICE2B and TaICE2B. ICEs from B. distachyon, Oryza sativa, Setaria italica, Sorghum bicolor and Zea mays were classified into the same cluster, Group II. Two ICEs in Arabidopsis fell into Group III. Group I and Group V are the ones that contain ICEs from two Triticeae subfamilies (Aegilops and Triticum), suggesting that the ancestral Group I and Group V genes were already present before divergence of these subfamilies. Interestingly, ICE1s were distantly relative to ICE2s, and the results indicated that ICE1s may have large sequences differences in the C-terminal regions with ICE2s.

Fig. 3.

Phylogenetic analysisof ICE proteins in Poaceae. The tree was constructed using the NJ method with a Poisson correction model and 1000 bootstrap replicates. The tree was divided into 4 groups

Further analysis of the conserved domains was performed. As other ICE-like proteins, these proteins contain a highly conserved basic helix-loop-helix (bHLH) signature domain, which contains the ICE-specific sequence KMDRASILGDAID/EYLKELL (Fig. S1). These structurally different ICE proteins all possess a predicted SUMO binding domain, but members of the Group I have a preference for IKEE, compared to VKEE of the other groups.

Identification of COR genes and phylogeny of COR genes in Poaceae

As CBF target genes or effector molecules, COR genes have a decisive role in enhancing crop cold tolerance. To dissect potential impacts on the downstream cold responsive genes of CBFs, we isolated 10 CORs from the above five contextual materials (Table 5). The details of the gene names were summarized in Table 5. These CORs encode proteins ranging from 163 to 166. COR1s shared higher homology with the published AetPHV A1-like (XM_020294635), Wrab17 (AF255053), TaSRP3 (JQ923472), TaLEA (HQ718763) and HvES2A (X79466). COR2s shared higher homology with the published AetPHV A1-like (XM_020294636) and WCOR615 (U73217).

Various kinds of CORs from Poaceae and two CORs in Arabidopsis were chosen to construct the relationship, and 23 total candidate proteins were listed in Table 5. The results indicated that these CORs were assigned into four groups (Group I, Group II, Group III, and Group V) (Fig. 4). Five COR1s deriving from this study belong to Group I with the published Wrab17, TaSRP3, TaLEA, AetPHV A1-like and HvES2A. However, five COR2s deriving from this study and the only WCOR615 fell into the same cluster, Group II. CORs from Pogonatherum paniceum, O. sativa, Agropyron cristatum, Eremopyrum triticeum and Wrab19 from T. aestivum were classified into Group III. The rest of CORs from Arabidopsis belong to Group V. Generally, the evolutionary relationship of CORs showed high multispecies specificity. Exceptionally, unlike other members of COR from T. aestivum which either belong to Group I or Group II, Wrab19 belongs to Group III. The appearance of being distantly related to the core group genes stated that the wheat COR genes continued to evolve following the divergence of species.

Fig. 4.

Phylogenetic analysisof COR proteins in Poaceae. The tree was constructed using the NJ method with a Poisson correction model and 1000 bootstrap replicates. The tree was divided into 4 groups with high multispecies specificity

Sequences variations on proteins from Group I and Group II were performed. Further analysis of amino acid composition showed that these fragments had high hydrophilic amino acid contents, including alanine (A), aspartic acid (D), glutamic acid (E), glycine (G), lysine (K) and threonine (T) (Fig. S2). The maximum of hydrophilic amino acid contents are as high as 70% with the absence of praline (P), cysteine (C) and tryptophan (W).

Expression analysis of ICE, CBF, COR genes

To comprehensively understand the precise mechanism and further details of the ICE–CBF–COR transcriptional cascade in Triticeae, expression analysis of the related genes was assessed. Overall, mRNA accumulation was altered by cold treatment (Fig. 5). The changing pattern of three kinds of genes was similar. Without cold treatment, the basal expression of CBF, ICE, COR genes were flat in all four materials. At the preliminary stage of cold treatment (2 h), these genes were not induced by low temperature. With the continuous cold stress, transcript levels of six genes are slightly higher than that without cold induction. After 4 additional hours of cold treatment, the expression of CBF and ICE genes were significantly elevated in A. tauschii, A. neglecta and winter wheat M808, compared with the spring wheat China Spring (CS). The expression of COR genes after cold treatment for 4 h had no obvious changes in comparison to that for 2 h. Then, the expression of six related genes reached the peak level after 4 °C for 12 h. At this peak point, transcript levels of genes in A. tauschii and A. neglecta are moderately higher than that in winter wheat. Whereas, we noticed that transcript levels of six genes were slightly reduced after longer exposure to cold stress (4 °C for 24 h). In Triticeae, transcript levels of all genes in spring wheat were the lowest, followed by winter wheat and A. neglecta, and transcript levels are the highest in A. tauschii. Together, these results indicate that all genes (ICE, CBF, COR) were cold-induced.

Fig. 5.

Expression analysis of ICE, CBF, COR genes under cold stress. The Y-axis indicates the normalized fold change in expression of each ICE, CBF, COR genes. The X-axis indicates the cold (4 °C) treatments for 0, 2, 4, 8, 12, 24 h. Error bars represented the standard error of the mean, and the asterisks represented the significance

Discussion

Wheat is an interesting model species for studying cold tolerance, however as a typical allohexaploid plant, complex genome has limited the further research on its tolerance. Aegilops, a progenitor of the allohexaploid wheat, is a close wild relative of common wheat with cold hardiness. Aegilops tauschii is D genome donor of common wheat with abundant genetic diversity and many important and excellent genes related to the stress resistance, which has been sequenced with a smaller genome (Jia et al. 2013). Consequently, the Aegilops–Triticum composite group is useful for the study and dissection of molecular, transcriptional, and evolutionary components of cold tolerance in cereals. Low temperature induces the expression of many genes, and identifying and comparing the functional domains of key genes involved in the transcriptional cascade are necessary.

To better understand the exact functions and the evolutionary relationships of gene families involved in ICE–CBF–COR pathway during cold stress in the Poaceae, we initiated a study to identify and characterize CBF, ICE, COR genes from the Aegilops–Triticum composite group and obtained 10 CBFs, 9 ICEs and 10 CORs. Conserved amino acids analysis revealed that all amino acid sequences had typical functional domains. Generally, all CBF amino acid sequences contained an AP2 DNA binding domain and an acidic C-terminal domain (Fig. 2a). The above two features were also generally observed in CBFs from other species, which indicated the possible functional significance. The structural conservation and phylogenetic relationship of the CBF proteins would reflect the close genetic relationship among members of Triticeae. However, there existed differences on one of special signature motifs among different protein members, with ASAWR directly flanked of AP2 DNA binding domain of AenCBFIVd-4 (Fig. 2b). Similarly, sequence discrepancies on the CBFs C-terminal with various combinations of four fragments were obvious (Fig. 2a, c). At the end of C-terminal, members of the subgroup CBFIVd-9 (Cluster I and Cluster II) had a preference for LWSC rather than LWSY of subgroup CBFIVd-4 (Cluster III) (Fig. 2c). As transactivation domain, the structural differences described on C-terminal of CBFs in this study may have significant impacts on transcription activities. With less conservative N-terminal sequences of ICEs, we cloned ICE genes through cloning three fragments respectively (a partial conserved fragment, 5′ end and 3′ end), and obtained 9 assembled ICE sequences. These ICE proteins contain a highly conserved bHLH domain, which includes a DNA binding motif and a dimerization motif. The former contributes to binding to the MYC consensus sequence and activates the expression of target genes (Peng et al. 2014), and the latter functions as the formation of homodimers or heterodimers (Mullen et al. 1994; Xu et al. 2014). The ICE-specific sequence of 19 amino acids is highly conserved, but has only an amino acid difference between orthologous genes (D/E). Sumoylation and ubiquitination motifs were essential elements and present in most of ICE proteins (Miura et al. 2007). Structural differentiation of the predicted SUMO binding domain (IKEE/VKEE) may have impacts on the activity and stability of ICE proteins. As downstream effectors, COR genes in this study putatively encoded hydrophilic proteins, which are different from the hydrophobic protein Wrab19 (pI = 10.3) (Tsuda et al. 2000). It has been revealed that the ICE–CBF–COR transcriptional cascade in cold regulation have differences among various organs in Arabidopsis including pollen, leaves and roots (Chinnusamy et al. 2007; Liao et al. 2014), and we also discovered that structure of the members of the transcriptional networks differ among Triticeae subspecies, as well as notable structural differences are displayed by these proteins may have influences on the functions.

On the basis of previous studies on CBFs in Poaceae, we further established the relationship among the different CBFs in Triticeae, and our results revealed that these CBFs were subdivided into 11 groups with the extra CBFV (Fig. 1). A highly conserved AP2 DNA binding domain were observed among CBFs, which demonstated that members of CBF families in Triticeae should also bind the CRT motif, then promote the expression of target genes. Meanwhile, CBFs belonging to Group V are somewhat distant from other groups with larger sequence discrepancies including AtCBFVa-1, AtCBFVa-2, AtCBFVa-3, HvCBFVa-7, and TmCBFVa-7. CBFs in Arabidopsis belong to Brassicaceae, which is evolutionarily distant from Triticeae and serves as members of outgroup. HvCBFVa-7 and TmCBFVa-7 are highly conserved in AP2 DNA binding domain, but the C-terminal and N-terminal of which are not conserved in comparison to other CBFs. Different ICEs from Poaceae and two ICEs in Arabidopsis were divided into 4 groups (Fig. 3). The analysis on multiple conserved domains of ICE proteins indicated that these ICEs can act on initiating the expression of CBFs through binding the MYC motif. In Arabidopsis, different ICE proteins are not redundant but have distinct functions (Rahman et al. 2014). As it turns out, ICE1s in this study were distantly relative to ICE2s, which could imply that the functions of ICE1s and ICE2s are diverse. Similarly, CORs in Poaceae and two CORs in Arabidopsis were assigned into four groups (Fig. 4). Compared with ICE and CBF proteins, as functional proteins rather than transcription factors, CORs displayed greater sequences diversity, which may shed light on important evolutionary trends coupling the functions. Unlike other CORs with high multispecies specificity, Wrab19 presented a long genetic distance away from the core group members, showing that the wheat COR genes continued to evolve following the divergence of species. Recently, functions of every members involved in this pathway in Arabidopsis were explicit, but size and complexity of these genes families in Triticeae are enormous without further research on concrete function of every member so far, as well as functional redundancy of genes families.

The diversity or complexity of different plants depends not on genes but on various regulatory networks and the research on regulation of gene expression under low temperature have recently received more attention. Cold tolerance of plants, as a complex quantitative trait, is regulated by multi-gene, and improvement of single-gene is difficult to completely enhance cold resistance of plants. As a result, this study implies that the ICE–CBF–COR transcriptional networks are likely present in all higher plants and could be a primary and essential component of cereal cold tolerance. After expose to low temperature, all genes in this study involved in the ICE–CBF–COR pathway were cold induced. Under cold conditions, ICE genes were induced and activated downstream CBF genes through binding the MYC recognition motifs in the promoter, subsequently, the target COR genes were accumulated to a higher level of contents. Previous studies had already noted that expression of the upstream regulatory factors occurred before that of the downstream genes in signaling pathways (Lee et al. 2005). In this study, the expression of COR genes after cold treatment for 4 h had no obvious changes in comparison to that for 2 h, but the expression of ICE and CBF genes were slightly elevated during this period, which is consistent with the aforementioned study.

In conclusion, recent studies focus on urgently seeking to improve plant chilling tolerance, and the ICE–CBF–COR transcriptional cascade play a pivotal role. The results showed that the cold stress signaling pathway is conserved in Triticeae. To be specific, ICE genes were activated under cold stress, then CBF genes are induced, and COR genes are ultimately induced to enhance plants chilling tolerance after the two-step cascade of transcriptional activators (Thomashow et al. 2001). In Arabidopsis, cold-activated OST1 kinase can phosphorylate AtICE1 and modulate chilling tolerance by enhancing ICE1 stability (Ding et al. 2015). PHYA (Benedict et al. 2006) and AtSIZ1 protein (Rahman et al. 2014) also can increase the stability of AtICE1, but JAZ1/4 suppresses transcriptional activity of ICE1 (Hu et al. 2013). Moreover, ICE1/SCRM and ICE2/SCRM2 in Arabidopsis are related to stomatal development (Kanaoka et al. 2008). Thus, as a central component in cold signaling pathway or a convergence point integrating cold and other signaling pathways, how ICE1 elaborately modulated covering couple pathways still remains unclear. Cold tolerance is a complex multigenic trait, we should consider it from many aspects of pathways with an aim of extending our understanding on temperature responses in plant, rather than view it as a single entity. In addition, it is notable that how plants response to cold stimulation is unknown. On the whole, this study enriches members of CBF, ICE, COR genes family in Triticeae, and provide a starting point for future studies on transcriptional regulatory network for improvement of chilling tolerance in crop.

Electronic supplementary material

Below is the link to the electronic supplementary material.