Gene and protein expression profiling analysis of young spike development in large spike wheat germplasms

Journal of Integrative Agriculture 2016, 15(4): 744–754 Available online at www.sciencedirect.com

ScienceDirect

RESEARCH ARTICLE

Gene and protein expression profiling analysis of young spike development in large spike wheat germplasms CHEN Dan1, 2*, ZHANG Jin-peng1*, LIU Wei-hua1, WU Xiao-yang1, YANG Xin-ming1, LI Xiu-quan1, LU Yu-qing1, LI Li-hui1 1

National Key Facility for Crop Gene Resources and Genetic Improvement/Institute of Crop Sciences, Chinese Academy of Agricultural Sciences, Beijing 100081, P.R.China 2 Biotechnology and Germplasm Resources Institute, Yunnan Academy of Agricultural Sciences/Key Laboratory of Southwestern Crop Gene Resources and Germplasm Innovation, Ministry of Agriculture/Scientific Observation for Rice Germplasm Resources of Yunnan, Ministry of Agriculture/Yunnan Provincial Key Laboratory of Agricultural Biotechnology, Kunming 650223, P.R.China

Abstract The wheat grain number per spike (GNPS) is a major yield-limiting factor in wheat-breeding programs. Germplasms with a high GNPS are therefore valuable for increasing wheat yield potential. To investigate the molecular characteristics of young spike development in large-spike wheat germplasms with high GNPS, we performed gene and protein expression profiling analysis with three high-GNPS wheat lines (Pubing 3228, Pubing 3504 and 4844-12) and one low-GNPS control variety (Fukuho). The phenotypic data for the spikes in two growth seasons showed that the GNPS of the three large-spike wheat lines were significantly higher than that of the Fukuho control line. The Affymetrix wheat chip and isobaric tags for relative and absolute quantitation-tandam mass spectrometry (iTRAQ-MS/MS) technology were employed for gene and protein expression profiling analyses of young spike development, respectively, at the floret primordia differentiation stage. A total of 598 differentially expressed transcripts (270 up-regulated and 328 down-regulated) and 280 proteins (122 upregulated and 158 down-regulated) were identified in the three high-GNPS lines compared with the control line. We found that the expression of some floral development-related genes, including Wknox1b, the AP2 domain protein kinase and the transcription factor HUA2, were up-regulated in the high-GNPS lines. The expression of the SHEPHERD (SHD) gene was up-regulated at both the transcript and protein levels. Overall, these results suggest that multiple regulatory pathways, including the CLAVATA pathway and the meristem-maintaining KNOX protein pathway, take part in the development of the high-GNPS phenotype in our wheat germplasms. Keywords: wheat, high grain number per spike, spike development, expression profiling

Received 3 March, 2015 Accepted 28 August, 2015 CHEN Dan, Mobile: +86-13668714841, E-mail: xiaoyezi09@163. com; ZHANG Jin-peng, Mobile: +86-13269833221, E-mail: [email protected]; Correspondance LI Li-hui, Tel: +8610-62186670, Fax: +86-10-62189650, E-mail: [email protected] * These authors contributed equally to this study. © 2016, CAAS. All rights reserved. Published by Elsevier Ltd. doi: 10.1016/S2095-3119(15)61179-0

1. Introduction Wheat (Triticum aestivum L.) is one of the most important cereal crops throughout the world. Enhanced sustainable wheat production is therefore essential for meeting the increasing food demand caused by global population growth and climate change. Wheat yield increases depend on

CHEN Dan et al. Journal of Integrative Agriculture 2016, 15(4): 744–754

three factors, namely the spike number per square (SNPS), grain number per spike (GNPS), and the thousand-kernel weight (TKW). In recent decades, increases in the SNPS and TKW have approached saturation, such that future wheat yield increases should focus on increased GNPS (Zhuang 2003). Good genetic resources with high GNPS levels have significant applications for potential yield increases in crop breeding. Our previous studies have shown that 4844-12, Pubing 3504 and Pubing 3228, were derived from the same cross between wheat and its wild relative Agropyron cristatum (A. cristatum), represent three high-GNPS wheat lines (Li et al. 1998; Wu et al. 2006; Zhang et al. 2011). Cytogenetic analysis has demonstrated that the high-GNPS trait observed in the addition line 4844-12 is derived from the alien 6P chromosome of A. cristatum (Wu et al. 2006). Pubing 3228 and Pubing 3504 are both derived from the addition line 4844-12 and show high-GNPS, narrow-sense heritability along with wide environmental adaptability and genetic stability (Chen et al. 2012). Multiple important quantitative trait loci (QTLs) that control GNPS and yield-related traits in the Pubing 3228 line have been mapped to chromosomes 1A, 4A, 4B, 5A, and 6A (Wang et al. 2011). These germplasms with high GNPS values represent valuable materials for research on the molecular mechanisms of large-spike wheat formation. Many efforts have been made to understand the genetic and molecular mechanisms underlying spike development in cereals. Some key genes and proteins that take part in spike development have been identified (Chuck et al. 2002; Komatsu et al. 2003). Phytohormones also play a crucial role in meristem determination and differentiation (Barazesh and McSteen 2008). Moreover, the CLAVATA (CLV) pathway plays a critical role in the differentiation of stem cells by restricting the expression of WUSCHEL (WUS); the CLV/ WUS pathway is also involved in CTK-regulated signaling circuits and cellular processes in Arabidopsis (Brand et al. 2000; Leibfried et al. 2005; Sablowski 2009; Yadav et al. 2011). In monocot plants, the key genetic and molecular switches have been reviewed in two model grass species,

745

namely rice and maize; such switches include meristem identity, meristem size and maintenance, the initiation and outgrowth of axillary meristems, and organogenesis (Zhang and Zheng 2014). However, the molecular mechanisms of high-GNPS development in wheat remain unclear. The purpose of this study was to analyze the molecular mechanisms of spike development in our high-GNPS lines. We attempted to identify interesting genes or proteins that were involved in the formation of the high-GNPS phenotype by analyzing differentially expressed genes at both the RNA and protein levels during the spike differentiation stage, thereby providing a molecular basis for cloning candidate genes and for future breeding applications.

2. Results 2.1. Investigation of spike morphological features in the high-GNPS germplasm Morphological observation of the young spikes showed that each developmental stage was retarded in the high-GNPS lines compared with the control Fukuho cultivar (Appendix A). The differentiation stage for Fukuho occurred on March 22th, whereas that of the three high GNPS lines occurred 5–7 d later. However, the length of the floret primordia differentiation stage in the young spikes did not significantly differ between the three high-GNPS lines and the Fukuho control. Investigation of the spike traits at the mature stage indicated that the three high-GNPS lines showed significant differences (P<0.01) compared to the Fukuho cultivar in GNPS, spilelet number per spike (SPN), kernel number per spikelet (KNS), and spike length (SL) (Table 1, Fig. 1-D and E). In two growing seasons, the average GNPS in the three high-GNPS lines ranged from 76.51 to 117.36, whereas that of Fukuho was 59.23. In the high-GNPS lines, the average SPN ranged from 22.31 to 25.29, and the average KNS ranged from 5.92 to 7.19. The SPN and KNS values for the Fukuho cultivar were 18.68 and 4.35, respectively. The spike feature data suggested that both SPN and KNS

Table 1 Investigation of mature spike traits in three high-GNPS wheat lines1) Material Pubing 3504 Pubing 3228 4844-12 Fukuho 1)

GNPS SPN KNS SL Chromosome composition2) 2007–2008 2008–2009 2007–2008 2008–2009 2007–2008 2008–2009 2007–2008 2008–2009 42W 118.90 A 116.88 A 24.80 A 25.44 A 7.00 A 7.25 A 13.95 A 11.48 A 42W 88.10 B 105.83 B 24.20 A 23.43 B 5.30 B 6.83 B 12.28 B 10.31 B 42W+2P 92.30 B 71.07 C 23.00 B 22.07 C 6.30 A 5.79 C 11.28 C 8.47 D 42W 59.70 C 59.07 D 19.20 C 18.50 D 4.10 C 4.43 D 10.19 D 9.56 C

GNPS, grain number per spike; SPN, spikelet number per spike; KNS, kernel number per spikelet; SL, spike length. W and P denote wheat and Agropyron cristatum chromosomes, respectively. Values followed by the same letters were not significantly different at P<0.01 (Duncan’s multiple range test, n=30). The data are means.

2)

746

CHEN Dan et al. Journal of Integrative Agriculture 2016, 15(4): 744–754

D Pubing 3504 Pubing 3228 4844-12 Fukuho

A B

B

C Fukuho

150 μm

1.5 mm

E Pubing 3504 Pubing 3228 4844-12

Fig. 1 Morphological observation of young spike development and mature spikes in high-grain number per spike (GNPS) wheat lines. A, plant morphology during the floret primordia differentiation stage. B and C, the morphology of young spike samples during the floret primordia differentiation stage. D, morphology of plant at maturity. E, morphology of spike at maturity.

contribute to the high-GNPS phenotype in 4844-12, Pubing 3228 and Pubing 3504.

2.2. Transcriptional profiling analysis of the highGNPS germplasm In total, 598 transcripts were differentially expressed between the three high-GNPS lines and the Fukuho cultivar. Among these, 270 transcripts were up-regulated (increased more than two folds) and 328 were down-regulated (decreased more than two folds). Annotation was successfully completed for 162 of the up-regulated genes and 215 of the down-regulated genes (Appendix B). According to their Kyoto encyclopedia of genes and genomes (KEGG) categories, we classified the differentially expressed genes into seven classes (Fig. 2). Metabolism-related genes made up the largest proportion of the classified genes, accounting for 17.68% of the up-regulated and 22.57% of the down-regulated genes. The expression levels of several genes known to be functionally involved in floral development were significantly changed. SHD, encoding a protein kinase with

an AP2 domain, HUA2 and Wknox1b were up-regulated (Table 2). Meanwhile, some genes related to yield, such as erect panicle 2 (EP2), were also up-regulated. We selected 19 up-regulated and 40 down-regulated genes to confirm the reproducibility of the GeneChip data. The results of a quantitative real-time PCR (qRT-PCR) analysis validated the GeneChip results (Appendix C).

2.3. Proteomic analysis of the high-GNPS germplasm After Mascot software analysis, 12 761 peptides matched to wheat expressed sequence tags (ESTs), and a total of 1 308 unique proteins exceeded the 95% confidence intervals. There were 280 differentially expressed proteins between the high-GNPS lines and the Fukuho line, including 122 up-regulated and 158 down-regulated proteins (Appendix D). The largest proportion of differentially expressed genes encoded genetic information processing proteins; they were both up- and down-regulated and were mainly involved in DNA replication, transcription and translation (Fig. 3, Table 3). The second-most-common protein type

747

CHEN Dan et al. Journal of Integrative Agriculture 2016, 15(4): 744–754

17.68% metabolism

A 27.81% unknown protein

0.61% cellular processes and signaling

19.40% others

4.88% environment information processing 14.02% genetic information processing

15.60% un-classified

B

22.57% metabolism

24.33% unknown protein

6.64% cellular processes and signaling 20.80% others 4.42% environment information processing 7.52% genetic information processing

13.72% un-classified

Fig. 2 Functional classification of the differentially expressed genes during the floret primordia differentiation stage. A, classification of up-regulated genes in high-GNPS wheat lines. B, classification of down-regulated genes in high-GNPS wheat lines. Table 2 Differentially expressed transcripts involved in floral development in the high-GNPS wheat lines Probe set ID Ta.5164.1.S1_at

Functional annotation Blast

Protein kinase with AP2 domain Ta.9460.3.S1_s_at HUA2 (ENHANCER OF AG-4 2) TaAffx.77836.1.S1_at MYB3R1

Blastx

Function catalog

Transcription factors Blastx Transcription factors Blastn Transcription factors Ta.28235.1.A1_s_at SHD (SHEPHERD), Blastx Folding, sorting ATP binding/unfolded and degradation protein binding Ta.26820.1.A1_at ATUBC2-1, ubiquitin- Blastx Ubiquitinprotein ligase mediated proteolysis or signal transduction TaAffx.56883.1.S1_at Phosphatidylserine Blastn Lipid metabolism decarboxylase, ZCCT2, ZCCT1, and SNF2P genes Ta.2927.1.S1_s_at GASA5 (GAST1 Blastn GA signal PROTEIN pathway HOMOLOG 5) Ta.19608.2.S1_at UDP-glucoronosyl/ Blastx Glycan UDP-glucosyl biosynthesis and transferase family metabolism protein

Signal log ratio1) Accession no. E-score Pubing 3228: Pubing 3504: 4844-12: Fukuho Fukuho Fukuho LOC_ 1.00E-05 2.0 2.3 2.7 Os03g08470.3 AT5G23150.1 7.00E-06 1.2 1.2 1.8 HQ236494.1

0

1.4

2.8

2.4

AT4G24190.2

6.00E-76

5.3

5.1

4.9

AT1G45050.1

4.00E-60

3.1

3.0

2.8

AY485644.1

2.00E-40

2.7

3.3

1.4

AT3G02885.1

2.00E-29

1.7

1.3

1.5

AT3G55700.1

3.00E-13

1.0

1.1

1.3

(Continued on next page)

748

CHEN Dan et al. Journal of Integrative Agriculture 2016, 15(4): 744–754

Table 2 (Continued from preceding page) Probe set ID

Functional annotation Blast

Ta.13280.3.S1_at

Phosphoenolpyruvate Blastn Carbohydrate carboxykinase metabolism (PCK1) Erect panicle 2 (EP2) Blastn Others

Ta.2016.3.A1_at

Function catalog

Signal log ratio1) Accession no. E-score Pubing 3228: Pubing 3504: 4844-12: Fukuho Fukuho Fukuho U09241.1 1.00E-06 1.9 1.6 1.2

GQ449684.1

9.00E-08

3.5

3.8

3.1

TaAffx.38542.1.A1_at Wknox1b gene for KN1 homeobox protein TaAffx.29451.1.S1_at ETTIN-like auxin response factor TaAffx.93139.1.A1_at Protein binding/ ubiquitin-protein ligase/zinc ion binding

Blastn Others

AB182944.1

1.00E-43

3.0

3.1

1.4

Blastx

Others

9.00E-18

5.7

5.2

5.2

Blastx

5.00E-19

5.6

6.3

7.1

TaAffx.113564.1.A1_ F-box domain at containing protein Ta.20002.1.A1_at HSP70

Blastx

BB (BIG BROTHER); protein binding/ ubiquitin-protein ligase/zinc ion binding Un-classified

UniRef90_ Q6U8C8 AT3G63530.2

1.7

2.8

2.1

–2.8

–3.1

–3.0

Ta.25611.1.S1_at

Blastx

–2.2

–2.0

–1.8

–1.7

–1.3

–1.0

–1.3

–2.8

–1.1

–2.1

–2.6

–1.8

–1.8

–5.6

–2.4

–4.3

–1.7

–1.4

–2.2

–1.3

–1.5

Blastx

LOC_ 2.00E-12 Os06g49930.2 Chaperones and AT3G12580.1 7.00E-57 folding catalysts Transcription AT5G39610.1 6.00E-61 factors Transcription AT5G59780.3 5.00E-57 factors Transcription AT3G15030.2 6.00E-22 factors Glycan AT5G39990.1 3.00E-82 biosynthesis and metabolism Carbohydrate AT3G21720.1 6.00E-06 metabolism Carbohydrate AT1G43670.1 1.00E-67 metabolism Biosynthesis of AT5G22020.1 5.00E-35 other secondary metabolites Un-classified AT5G51560.1 9.00E-13

–1.4

–1.1

–1.1

Blastx

Un-classified

Ta.25919.1.S1_at

NAC-domain transcription factor MYB59

Blastx

Blastx

TaAffx.12823.1.S1_at TCP family transcription factors Ta.12174.1.A1_at Glycosyl hydrolase family protein

Blastx

Ta.23989.1.A1_at

Isocitrate lyase

Blastx

Ta.28890.1.A1_s_at

Fructose-1,6Blastx bisphosphatase Strictosidine synthase Blastx family protein

Ta.7999.2.S1_a_at

TaAffx.119353.1.A1_ Leucine-rich repeat at transmembrane protein kinase TaAffx.20089.1.A1_at Leucine rich repeat family protein Ta.13056.1.S1_at G protein-coupled receptor 54 Ta.11544.2.S1_a_at Protein serine/ threonine kinase TaAffx.26508.1.S1_at CLASP (CLIPASSOCIATED PROTEIN) Ta.11179.1.A1_at cDNA ulp1 protease family, C-terminal catalytic domain containing protein Ta.18122.1.S1_at Protease inhibitor Ta.4544.1.A1_a_at FCA-like protein Ta.9379.3.S1_a_at PHD-finger domain containing protein Ta.9223.1.A1_at Protein kinase family protein 1)

Blastx

–2.0

–1.6

–1.8

Blastn Receptor family

LOC_ 1.00E-08 Os06g03970.1 FR667382.1 3.00E-08

–6.2

–6.5

–4.1

Blastx

Un-classified

AT1G68690.1

5.00E-07

–1.1

–1.1

–1.2

Blastx

Un-classified

AT2G20190.1

2.00E-22

–3.1

–2.6

–2.0

Blastx

Others

LOC_ 3.00E-06 Os01g16730.1

–6.1

–4.4

–2.6

AT3G18280.1 AY235464.1 LOC_ Os06g10690.2 AT5G38560.1

5.00E-22 7.00E-29 5.00E-30

–2.0 –4.7 –2.3

–3.0 –4.5 –1.7

–1.9 –5.5 –1.7

3.00E-05

–5.8

–3.6

–5.7

Blastx Others Blastn Un-classified Blastx Un-classified Blastx

Un-classified

Signal log ratio denotes a mean of the log ratios of probe pair intensities across the high-GNPS lines and the control Fukuho.

CHEN Dan et al. Journal of Integrative Agriculture 2016, 15(4): 744–754

A 17.89% unknown protein

749

26.83% metabolism

6.50% cellular processes and signaling

8.13% environment information processing

B 22.64% unknown protein

40.65% genetic information processing

13.21% metabolism

37.11% genetic information processing 16.35% cellular processes and signaling

10.69% environment information processing

Fig. 3 Functional classification of differentially expressed proteins during the floret primordia differentiation stage in high-GNPS wheat lines. A, classification of up-regulated proteins in high-GNPS wheat lines. B, classification of down-regulated proteins in high-GNPS wheat lines.

was metabolic proteins, which included glucose metabolism enzymes, ATP synthase and peroxidases. This finding suggests that the protein biosynthesis pathway is more active during the spike phase in the high-GNPS lines than in the control Fukuho line.

2.4. Common differentially expressed genes at the transcriptional and translational levels Combining the results of the gene and protein expression profiling, two types of differentially expressed genes were consistent both at the RNA and protein levels. The SHD gene was up-regulated at both the RNA and protein levels. At the RNA level, the SHD genes in Pubing 3228, Pubing 3504 and 4844-12 were up-regulated by a factor of 5.3, 5.1 and 4.9, respectively, over the Fukuho line (Table 2). Meanwhile, the SHD protein levels were increased by factors of 1.107, 1.498 and 1.298 in Pubing 3228, Pubing 3504 and 4844-12, respectively (Table 3). In addition, a few ribosomal proteins were up-regulated both at the RNA and protein levels. Taken together, our results show that the expression differences at the protein level are mainly limited to high-abundance proteins involved in the metabolism path-

way, while at the mRNA level, low-abundance transcription factors with important biological functions were detected.

3. Discussion In grass species such as wheat and rice, spike development contributes to the ultimate grain yield. The spike or inflorescence of these cereal crops consists of a unique reproductive structure called the spikelet, which is comprised of several florets. Studies on the genetic regulation of spike differentiation provide vital information for crop breeding. In this study, we found that the genes with roles in regulating the cell proliferation process and controlling the numbers of floral organs are important for enhancing meristematic activity and take part in forming increased numbers of spikelets and florets during spike development. KNOTTED1 (KN1)-like homeobox (KNOX) transcription factors function in plant meristems, which are self-renewing structures consisting of stem cells that have important roles in shoot apical meristem maintenance (Scofield and Murray 2006; Shani et al. 2006; Zhang and Zheng 2014). The KNOX family of genes can activate the ATP/ADP ISOPENTYLTRANSFERASE (IPT) genes to promote Cy-

750

CHEN Dan et al. Journal of Integrative Agriculture 2016, 15(4): 744–754

Table 3 Differentially expressed proteins involved in the floral development of high-GNPS wheat lines GenBank accession CJ867268 CJ801174 CJ921456 CJ939710 CD888214 CK164178 CK166001 CK171755 CV762944 DR735768 CJ552903 CJ631404 CJ626873 CJ730539 CD922131 CF134161

Protein name 40S ribosomal protein S12 (Zea mays) Glyceraldehyde-3-phosphate dehydrogenase, cytosolic (Zea mays) Enolase, putative, expressed (Oryza sativa (japonica cultivar-group)) Hypothetical protein OsI_14101, HSP70 (Oryza sativa indica Group) Heat shock protein 80 (Triticum aestivum) 60S ribosomal protein L12 (Zea mays) 40S ribosomal protein S26 (Zea mays) Small Ras-related GTP-binding protein (Triticum aestivum) Leucine aminopeptidase 2, chloroplastic SHD protein, endoplasmin homolog, GRP94 homolog Phosphoglycerate kinase, cytosolic Chromomethylase 1 (Hordeum vulgare subsp. vulgare) Hypothetical protein OsJ_030550 (Oryza sativa (japonica cultivar-group)), groEL, chaperonin GroEL Mitochondrial chaperonin 60 (Zea mays)

CJ778335 CJ830670 CJ894130

S28 ribosomal protein (Triticum aestivum) Phosphoenolpyruvate carboxylase (Triticum aestivum) α tubulin-2A (Triticum aestivum) Cytosolic malate dehydrogenase (Triticum aestivum) 14-3-3 protein (Triticum aestivum)

CK202005

Glycosyltransferase (Triticum aestivum)

CK202421

Glycine-rich RNA-binding protein (Triticum aestivum) Histone H4 (Zea mays) α tubulin (Setaria viridis) Mitochondrial F-1-ATPase subunit 2 (Zea mays)

CK212468 CK214378 DN829527 CJ631449 CJ685334 CJ691721

UTP-glucose-1-phosphate uridylyltransferase S-adenosylmethionine synthetase 1 (Triticum monococcum) ATP synthase beta subunit (Triticum aestivum)

CJ523554

Nucleoside diphosphate kinase (Lolium perenne)

1)

Functional classification Ribosome Carbohydrate metabolism, enzymes Carbohydrate metabolism, enzymes Chaperones and folding catalysts Chaperones and folding catalysts Ribosome Ribosome Ras-related small GTPbinding family protein Carbohydrate metabolism, enzymes Folding, sorting and degradation Carbohydrate metabolism, enzymes Cysteine and methionine metabolism Chaperones and folding catalysts Chaperones and folding catalysts Ribosome Carbohydrate metabolism; pyruvate metabolism Cytoskeletal proteins Metabolism Cellular processes; cell growth and death Glycan biosynthesis and metabolism RNA-binding protein Histone Cytoskeletal proteins Energy metabolism; oxidative phosphorylation Carbohydrate metabolism Amino acid metabolism Energy metabolism; oxidative phosphorylation Nucleotide metabolism

E-score 114:117 115:117 116:1171) 1.00E-20 4.00E-120

1.46 1.463  1.084 1.126 1.172 1.353

3.00E-98

1.409 1.842 1.094 

3.00E-19

1.035 1.214 1.055 

3.00E-61

1.383 1.676 0.859

3.00E-71 7.00E-33 2.00E-61

1.241 1.260 1.005 1.352 2.129 2.286  1.004 1.836 1.198

3.00E-93

1.078 2.023 1.072 

3.00E-140

1.107 1.498 1.298 

8.00E-120

1.133 1.271 1.219 

1.00E-138

1.148 1.385 1.270

8.00E-112

1.410 1.664 1.038

2.00E-103

1.402 1.483 1.194

2.00E-21 9.00E-115

0.929 0.966 0.786 0.920 0.929 0.830

3.00E-149 2.00E-67 7.00E-48

0.808 0.943 0.759  0.995 0.902 0.900 0.667 0.627 0.678 

3.00E-06

0.753 0.567 0.686

2.00E-42

 0.882 0.952 0.589 

8.00E-38 1.00E-99 8.00E-171

0.812 0.746 0.272  0.819 0.903 0.832  0.917 0.969 0.568 

1.00E-113 9.00E-152

0.947 0.965 0.894  0.857 0.972 0.697 

9.00E-131

0.837 0.940 0.505 

1.00E-41

0.753 0.689 0.632 

114:117, 115:117 and 116:117 denote the ratios of the high-GNPS wheat lines Pubing3228, Pubing3504 and 4844-12 to Fukuho, respectively.

tokinin (CTK) biosynthesis in Arabidopsis (Kepinski 2006; Sakakibara 2006). In maize, knotted1 loss-of-function mutations in the homeobox gene were defective in shoot meristem maintenance (Kerstetter et al. 1997; Bolduc et al. 2012). Our study showed that the Wknox1b gene for the KN1 homeobox protein was up-regulated in high-GNPS lines compared with control line, suggesting that the cell

proliferation process is activated and promoted to permit a high-GNPS phenotype. The SHD gene encodes an ortholog of GRP94, an ER-resident HSP90-like protein. GRP94 regulates the correct folding and complex formation of CLV proteins to allow the correct differentiation of the shoot apical meristem (SAM, Ishiguro et al. 2002). The CLV (CLV1, CLV2

CHEN Dan et al. Journal of Integrative Agriculture 2016, 15(4): 744–754

and CLV3) pathway operates to regulate the size of the stem cell population in the SAM. In grass species, rice FLORALORGAN NUMBER genes (FON1, FON2, FON3, FON4) and maize tassel dwarf1 (td1) and fascinated ear2 (fea2) are functionally conserved as the CLV-like genes in monocots as well as in eudicots (Taguchi et al. 2001; Suzaki et al. 2004, 2009; Bommert et al. 2005; Jiang et al. 2005; Chu et al. 2006). The shd mutant showed a similar phenotype to clv mutants, eliminating control over the size of the SAM, inducing inflorescence and floral meristem abnormalities and increasing the number of flowers and floral organs. In this study, SHD was co-up-regulated both at the transcriptional and translational levels in the three high-GNPS lines. Moreover, our recent results regarding the A. cristatum transcriptome showed that some CLV pathway genes in A. cristatum were located on 6P, which was the same P chromosome additioned in high-GNPS line 4844-12 (Wu et al. 2006; Zhang et al. 2015). Hence, we deduce that the CLV pathway genes of the three high-GNPS lines in this study may contribute to the high-GNPS phenotype. The APETALA2 (AP2) floral homeotic transcription factors mainly participate in the regulatory network of floral meristem development and floral organ morphogenesis (Aukerman and Sakai 2003). The protein kinase with the AP2 domain in our study was significantly up-regulated in the high-GNPS lines, suggesting that it plays an important role in high GNPS formation in wheat. Moreover, HUA2 is involved in the floral homeotic AGAMOUS pathway. AGAMOUS performs the C functions and controls the third and fourth whorls, including the stamen and carpel (Chen and Meyerowitz 1999). In our study, HUA2 was up-regulated in the high-GNPS lines, which might aid in floral organ identity and increase floret formation. Together, our results suggest that gene expression in our high-GNPS lines is very complex and that multiple regulatory pathways may take part in the development of a high-GNPS phenotype. The up-regulation of genes that are involved in the cell proliferation process and that control the numbers of floral organs might be beneficial for enhancing meristematic activity and forming more spikelets and florets in the high-GNPS germplasms.

4. Conclusion The spike development in cereals is crucial for the final crop yield formation. Our study analyzed the gene and protein expression profiling of high-GNPS wheat germplasm in young spike development stage. The genes involving the cell proliferation process and controlling the numbers of floral organs including Wknox1b, the AP2 domain protein kinase and the transcription factor HUA2, were up-regulated in the

751

high-GNPS wheat germplasm. Moreover, the expression of the SHEPHERD (SHD) gene was up-regulated at both the transcript and translation levels. In conclusion, we demonstrated that these genes were important for enhancing meristematic activity and increasing numbers of spikelets and florets during spike development, which were beneficial for the formation of the high-GNPS phenotype in our wheat germplasm. These findings provided a new insight into the spike development of high GNPS germplasm, which explained the molecular characteristics of young spike development in high GNPS wheat germplasm.

5. Materials and methods 5.1. Plant materials Wheat (Triticum aestivum L.) germplasm 4844-12 (2n=6x=44, 42W+2P) is an addition line derived from a cross between the common wheat cultivar Fukuho (Triticum aestivum L. 2n=6x=42, AABBDD) and A. cristatum (2n=4x=28, PPPP; China National GenBank accession number Z559) (Wu et al. 2006). Pubing 3228 (2n=6x=42, 42W, AABBDD) is a genetically stable derivative line obtained from the F5 progeny of the addition line 4844-12 (Wang et al. 2011). Pubing 3504 (2n=6x=42, 42W, AABBDD) is the sister line of Pubing 3228 and was derived from the same pedigree (Chen et al. 2012). All of the above three wheat germplasm lines possess large spikes and high GNPS values, with superior numbers of spikelets and florets. The wheat cultivar Fukuho, with its low GNPS value, is the recipient parent of the wide cross between wheat and A. cristatum. Fukuho is used as the control cultivar in this study (Li et al. 1998).

5.2. Morphological observation of young spike development and trait investigation All the materials were grown under field conditions at the Chinese Academy of Agricultural Sciences (CAAS) Experimental Station in Beijing during the 2007–2008 and 2008–2009 sowing seasons. The experimental design was a randomized complete block with four replicates. Three blocks were used for trait investigation, and one block was used for RNA and protein sampling. In each block, each material was grown in 15 rows. The field layouts consisted of arrangements that were 2 m in length, with 25-cm rows and a sowing density of 20 plants per row. Young wheat spike development was divided into several stages, including the apex elongation stage (Gardner stage 1, Zadoks stage 13), the single-ridge stage (Gardner stage 2, Zadoks stage 14), the double-ridge stage (Gardner stages 3 and 4, Zadoks stages 14.2 and 14.6), the glume primordia

752

CHEN Dan et al. Journal of Integrative Agriculture 2016, 15(4): 744–754

differentiation stage (Gardner stage 5, Zadoks stage 15), the floret primordia differentiation stage (Gardner stages 6 and 7, Zadoks stages 15.5 and 16), the stamen and pistil differentiation stage (Gardner stage 7, Zadoks stage 16), the anther separation stage (Gardner stage 7, Zadoks stage 16) and the tetrad stage (Gardner stage 8, Zadoks stages 16.5–17) (Zadoks et al. 1974; Gardner et al. 1985; Cui et al. 2008). In the spring season of 2008 (March to May), the developmental stages of the young spikes were observed by stereomicroscopy (SZX2-ILLT, OLYMPUS, Japan) for two or three days each, and five randomly selected individuals were observed from each replicate at each time (Fig. 1-A–C). The length of each spike developmental stage was recorded (Appendix A). When the young spike was in the early floret primordia differentiation stage (Gardner stages 6 and 7, Zadoks stages 15.5 and 16) of 5–7 leaves, the young spikes were collected. They were immediately flash-frozen in liquid nitrogen in separate tubes and stored at −70°C for RNA and protein extraction (Fig. 1-B and C). After harvesting, the spike length (SL), spikelet number per spike (SPN), kernel number per spikelet (KNS), and GNPS were evaluated. A total of 30 plants from each material were randomly selected from each replicate for trait investigation according to previously reported methods (Wang et al. 2011).

5.3. RNA expression profile analysis Total RNA was isolated from young spikes at the floret primordia differentiation stage using TRIZOL reagent (Invitrogen, Carlsbad, CA, USA). An Affymetrix GeneChip Wheat Genome Array was used for the array hybridization experiment (Gene Tech-Shanghai Company Ltd., Shanghai, China). The wheat genome array contains 61 127 probe sets representing 55 052 transcripts from all 42 chromosomes of the wheat genome. Differentially expressed transcripts (up- or down-regulated by greater than two folds) were identified by comparing high-GNPS germplasms (Pubing 3504, Pubing 3228 and 4844-12) with the control Fukuho cultivar. During the biological comparison analysis, each probe set on an array of three high-GNPS germplasm lines was compared to the control array, and the P-value changes were calculated, thus revealing increased, decreased or unchanged gene expression levels. P-values of less than 0.01 or more than 0.99 were considered to be significant. The differentially expressed genes were annotated using HarvEST WheatChip ver. 1.57 software (http://harvest.ucr. edu/) or by using a BLASTN search in the NCBI database with an E-value cut-off of 1E-05. Differentially expressed genes were categorized based on the KEGG pathway database (http://www.genome.jp/kegg/pathway.html, Kanehisa et al. 2012).

5.4. Isobaric tags for relative and absolute quantitation The methods for protein sample preparation and isobaric tags for relative and absolute quantitation (iTRAQ) analysis have been described by Luo et al. (2009). Protein sample digestion, iTRAQ isobaric labeling, off-line strong cation exchange chromatography and online Nano-LC ESI QqTOF MS analysis were conducted by the UAlbany Proteomics Facility Center for Functional Genomics at the University of Albany (New York, USA). The iTRAQ labels 114, 115, 116 and 117 represent Pubing 3228, Pubing 3504, 4844-12 and Fukuho, respectively. The tandem mass spectrometry (MS/MS) data (95% peptide confidence values) were processed by a search of the National Center for Biotechnology Information (NCBI) wheat expressed sequence tags (ESTs) database using Mascot software. The software search parameters were as follows: S-acetamido N-terminal and lysine modifications were selected as fixed, and methionine oxidation was set as variable, with two missed cleavages allowed. The precursor error tolerance was set to ±0.3 Da, the MS/MS fragment tolerance was set to ±0.3 Da, and full trypsin specificity was used (N- and C-termini were also applied). Differentially expressed proteins were considered to be significant when the fold change ratio was >1.1 or <0.9 (P<0.05) for 114:117, 115:117 and 116:117. The most-often-matched proteins were annotated using BLASTX in a non-redundant NCBI protein database. Finally, the proteins were categorized based on the KEGG pathway database.

5.5. qRT-PCR Differentially expressed genes were validated by conducting real-time PCR with the SYBR Premix Ex Taq Kit (TaKaRa, Japan) according to the manufacturer’s recommendations. β-Tubulin 3 (NCBI accession number CJ888457) was used as an internal control. PCR primers were designed using the Primer3 website (http://frodo.wi.mit.edu/) (Appendix E). Three biological replicates were tested for each sample. Relative expression levels were calculated using the2–ΔΔCt method (Livak and Schmittgen 2001).

Acknowledgements This work was supported by grants from the National Basic Research Program of China (973 Program, 2011CB100104), the National High-Tech R&D Program of China (2011AA100101), the National Natural Science Foundation of China (31071416), and the National Key Technologies R&D Program of China during the 12th FiveYear Plan period (2013BAD01B02).

CHEN Dan et al. Journal of Integrative Agriculture 2016, 15(4): 744–754

Appendix associated with this paper can be available on http://www.ChinaAgriSci.com/V2/En/appendix.htm

References Aukerman M J, Sakai H. 2003. Regulation of flowering time and floral organ identity by a microRNA and its APETALA2-like target genes. The Plant Cell Online, 15, 2730–2741. Barazesh S, McSteen P. 2008. Hormonal control of grass inflorescence development. Trends in Plant Science, 13, 656–662. Bolduc N, Yilmaz A, Mejia M, Morohashi K, O’Connor D, Grotewold E, Hake S. 2012. Unraveling the KNOTTED1 regulatory network in maize meristems. Genes & Development, 26, 1685–1690. Bommert P, Lunde C, Nardmann J, Vollbrecht E, Running M, Jackson D, Hake S, Werr W. 2005. thick tassel dwarf1 encodes a putative maize ortholog of the Arabidopsis CLAVATA1 leucine-rich repeat receptor-like kinase. Development, 132, 1235–1245. Brand U, Fletcher J C, Hobe M, Meyerowitz E M , Simon R. 2000. Dependence of stem cell fate in Arabidopsis on a feedback loop regulated by CLV3 activity. Science, 289, 617–619. Chen D, Zhang J P, Wang J S, Yang X M, Liu W H, Gao A N, Li X Q, Li L H. 2012. Inheritance and availability of high grain number per spike in two wheat germplasm lines. Journal of Integrative Agriculture, 11, 1409–1416. Chen X M, Meyerowitz E M. 1999. HUA1 and HUA2 are two members of the floral homeotic AGAMOUS pathway. Molecular Cell, 3, 349–360. Chu H, Qian Q, Liang W, Yin C, Tan H,Yao X, Yuan Z, Yang J, Huang H, Luo D, Ma H, Zhang D. 2006. The floral organ number4 gene encoding a putative ortholog of Arabidopsis CLAVATA3 regulates apical meristem size in rice. Plant Physiology, 142, 1039–1052. Chuck G, Muszynski M, Kellogg E, Hake S, Schmidt R J. 2002. The control of spikelet meristem identity by the branched silkless1 gene in maize. Science, 298, 1238–1241. Cui J M, Guo T C, Zhu Y J, Wang C Y, Ma X M. 2008. Spike of Wheat. China Agriculture Press, Beijing. pp. 22–33. (in Chinese) Gardner J S, Hess W M, Trione E J. 1985. Development of the young wheat spike: A SEM study of Chinese spring wheat. American Journal of Botany, 72, 548–559. Ishiguro S, Watanabe Y, Ito N, Nonaka H, Takeda N, Sakai T. 2002. SHEPHERD is the Arabidopsis GRP94 responsible for the formation of functional CLAVATA proteins. The EMBO Journal, 21, 898–908. Jiang L, Qian Q, Mao L, Zhou Q, Zhai W. 2005. Characterization of the rice floral organ number mutant fon3. Journal of Integrative Plant Biology, 47, 100–106. Kanehisa M, Goto S, Sato Y, Furumichi M, Tanabe M. 2012. KEGG for integration and interpretation of large-scale molecular data sets. Nucleic Acids Research, 40, D109– D114.

753

Kepinski S. 2006. Integrating hormone signaling and patterning mechanisms in plant development. Current Opinion in Plant Biology, 9, 28–34. Kerstetter R, Laudencia D, Smith L, Hake S. 1997. Loss-offunction mutations in the maize homeobox gene, knotted1, are defective in shoot meristem maintenance. Development, 124, 3045–3054. Komatsu M, Chujo A, Nagato Y, Shimamoto K, Kyozuka J. 2003. FRIZZY PANICLE is required to prevent the formation of axillary meristems and to establish floral meristem identity in rice spikelets. Development, 130, 3841–3850. Leibfried A, To J, Busch W, Stehling S, Kehle A, Demar M, Kieber J, Lohmann J. 2005. WUSCHEL controls meristem function by direct regulation of cytokinin-inducible response regulators. Nature, 438, 1172–1175. Li L H, Yang X M, Li X Q, Dong Y C, Chen X M. 1998. Introduction of desirable genes from Agropyron cristatum into common wheat by intergeneric hybridization. Scientia Agricultura Sinica, 31, 1–5. (in Chinese) Livak K J, Schmittgen T D. 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔCT method. Methods, 25, 402–408. Luo J L, Ning T T, Sun Y F, Zhu J H, Zhu Y G, Lin Q S, Yang D C. 2009. Proteomic analysis of rice endosperm cells in response to expression of hGM-CSF. Journal of Proteome Research, 8, 829–837. Sablowski R. 2009. Cytokinin and WUSCHEL tie the knot around plant stem cells. Proceedings of the National Academy of Sciences of the United States of America, 106, 16016–16017. Sakakibara H. 2006. Cytokinins: Activity, biosynthesis, and translocation. Annual Review Plant Biology, 57, 431–449. Scofield S, Murray J A. 2006. KNOX gene function in plant stem cell niches. Plant Molecular Biology, 60, 929–946. Shani E, Yanai O, Ori N. 2006. The role of hormones in shoot apical meristem function. Current Opinion in Plant Biology, 9, 484–489. Suzaki T, Ohneda M, Toriba T, Yoshida A, Hirano H. 2009. FON2 SPARE1 redundantly regulates floral meristem maintenance with FLORAL ORGAN NUMBER2 in rice. PLoS Genetics, 5, e1000693. Suzaki T, Sato M, Ashikari M, Miyoshi M, Nagato Y, Hirano H Y. 2004. The gene FLORAL ORGAN NUMBER1 regulates floral meristem size in rice and encodes a leucine-rich repeat kinase orthologous to Arabidopsis CLAVATA1. Development, 131, 5649–5657. Taguchi F, Yuan Z, Hake S, Jackson D. 2001. The fasciated ear2 gene encodes a leucine-rich repeat receptor-like protein that regulates shoot meristem proliferation in maize. Genes & Development, 15, 2755–2766. Wang J S, Liu W H, Wang H, Li L H, Wu J, Yang X M, Li X Q, Gao A N. 2011. QTL mapping of yield-related traits in the wheat germplasm 3228. Euphytica, 177, 277–292. Wu J, Yang X M, Wang H, Li H J, Li L H, Li X Q, Liu W H. 2006. The introgression of chromosome 6P specifying for increased numbers of florets and kernels from Agropyron

754

CHEN Dan et al. Journal of Integrative Agriculture 2016, 15(4): 744–754

cristatum into wheat. Theoretical and Applied Genetics, 114, 13–20. Yadav R K, Perales M, Gruel J, Girke T, Jönsson H, Reddy G V. 2011. WUSCHEL protein movement mediates stem cell homeostasis in the Arabidopsis shoot apex. Genes & Development, 25, 2025–2030. Zadoks J C, Chang T T, Konzak C F. 1974. A decimal code for the growth stages of cereals. Weed Research, 14, 415–421. Zhang D B, Zheng Y. 2014. Molecular control of grass inflorescence development. Annual Review of Plant Biology, 65, 553–578. Zhang J P, Liu W H, Han H M, Song L Q, Bai L, Gao Z H,

Zhang Y, Yang X M, Li X Q, Gao A N, Li L H. 2015. De novo transcriptome sequencing of Agropyron cristatum to identify available gene resources for the enhancement of wheat. Genomics, doi: org/10.1016/j.ygeno.2015.04.003 Zhang J P, Liu W H, Yang X M, Gao A N, Li X Q, Wu X Y, Li L H. 2011. Isolation and characterization of two putative cytokinin oxidase genes related to grain number per spike phenotype in wheat. Molecular Biology Reports, 38, 2337–2347. Zhuang Q S. 2003. Chinese Wheat Improvement and Pedigree Analysis. China Agriculture Press, Beijing. pp. 497–506. (in Chinese) (Managing editor WANG Ning)