Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2018 Nov 27;221(4):1950–1965. doi: 10.1111/nph.15548

Interaction between row‐type genes in barley controls meristem determinacy and reveals novel routes to improved grain

Monika Zwirek 1, Robbie Waugh 1,2, Sarah M McKim 2,✉
PMCID: PMC6492131  PMID: 30339269

Summary

  • Hordeum species develop a central spikelet flanked by two lateral spikelets at each inflorescence node. In ‘two‐rowed’ spikes, the central spikelet alone is fertile and sets grain, while in ‘six‐rowed’ spikes, lateral spikelets can also produce grain. Induced loss‐of‐function alleles of any of five Six‐rowed spike (VRS) genes (VRS1‐5) cause complete to intermediate gains of lateral spikelet fertility. Current six‐row cultivars contain natural defective vrs1 and vrs5 alleles. Little information is known about VRS mechanism(s).

  • We used comparative developmental, expression and genetic analyses on single and double vrs mutants to learn more about how VRS genes control development and assess their agronomic potential.

  • We show that all VRS genes repress fertility at carpel and awn emergence in developing lateral spikelets. VRS4, VRS3 and VRS5 work through VRS1 to suppress fertility, probably by inducing VRS1 expression. Pairing vrs3, vrs4 or vrs5 alleles increased lateral spikelet fertility, despite the presence of a functional VRS1 allele. The vrs3 allele caused loss of spikelet identity and determinacy, improved grain homogeneity and increased tillering in a vrs4 background, while with vrs5, decreased tiller number and increased grain weight.

  • Interactions amongst VRS genes control spikelet infertility, determinacy and outgrowth, and novel routes to improving six‐row grain.

Keywords: barley (Hordeum vulgare), branching, grain number, meristem, spikelet fertility, yield potential

Introduction

Grass inflorescence architecture is defined by the number, nature and arrangement of spikelets, the basic flowering unit, along single or branched reproductive axes (Kellogg, 2015). Spikelet number, fertility and survival determine grain yield, all of which are influenced by competition for finite resources. The Triticeae cereals usually form compact ‘spike’ inflorescences, characterised by sessile spikelets produced directly from alternating, opposing nodes on the branchless main axis or rachis. At each rachis node, Hordeum species develop a central spikelet flanked by two lateral spikelets. This distinctive triple spikelet arrangement arises when each inflorescence node forms a triple spikelet meristem (TSM) that cleaves into a large central spikelet meristem (CSM) bordered by two smaller lateral spikelet meristems (LSMs). Spikelet identity is marked by the initiation of two basal framing glumes, after which the remaining SM forms a single floret meristem (FM) that, if fertile, can develop a single grain. Lateral spikelet sterility is an ancestral trait of barley, Hordeum vulgare ssp. spontaneum, where the arrowhead shape formed by sterile lateral spikelets flanking a central grain is presumed to aid grain dispersal after rachis disarticulation (Harlan, 1968). However, during domestication, the selection of nonbrittle rachises supported the cultivation of variants with lateral spikelet fertility and six rows of grain on the spike (Helbaek, 1959; Harlan, 1968).

Despite their potential, cultivated six‐rowed barleys do not produce three‐fold more grain due to reduced tillering (basal branching) and rachis node number (Kirby & Riggs, 1978; Lundqvist et al., 1997). Moreover, lateral rows produce smaller and lighter grains compared with their central equivalents, making the size and weight of six‐rowed spike grain less uniform (Gupta et al., 2010; Lang et al., 2013). This is an undesirable trait for the premium end‐use malting sector, leaving six‐rowed varieties mostly grown for animal feed. Long‐standing interest in developing improved six‐row varieties has revealed that at least 11 loci regulate lateral spikelet fertility (Lundqvist & Lundqvist, 1988). To date, all alleles associated with lateral spikelet fertility are predicted to be impaired or loss‐of‐function alleles, many of which cause an intermediate phenotype between the two and six‐row form (Lundqvist & Lundqvist, 1988). Induced recessive alleles at five major row‐type loci can independently convert spikes from two‐rowed to six‐rowed: SIX‐ROWED SPIKE 1 (VRS1), VRS2, VRS3 (syn. INTERMEDIUM‐A), VRS4 (syn. INTERMEDIUM‐E) and VRS5 (syn. INTERMEDIUM‐C). Six‐rowed barley originated with multiple natural recessive vrs1 alleles (Tanno et al., 2002), which are now accompanied by a natural allele of vrs5 (Int‐c.a) that confers improved lateral grain fill in six‐row cultivars (Harlan, 1968; Lundqvist et al., 1997). The three other major recessive row‐type alleles are not prevalent in current six‐row cultivars (Koppolu et al., 2013; Bull et al., 2017; Youssef et al., 2017).

Recent efforts have revealed the molecular identity of the five major row‐type genes. VRS5 encodes a class II TCP transcription factor whose homologues in maize, rice, wheat and Arabidopsis repress tiller bud outgrowth to promote apical dominance (Doebley et al., 1995, 1997; Takeda et al., 2003; Lewis et al., 2008; Ramsay et al., 2011; Dixon et al., 2018). VRS1 encodes a homeodomain‐leucine zipper class I protein (HD‐ZIP1; Komatsuda et al., 2007) whose homologue in maize, grassy tillers1 (gt1) inhibits tiller bud outgrowth downstream of tb1 (Whipple et al., 2011). VRS4 encodes a LATERAL ORGAN BOUNDARIES (LOB) transcription factor orthologous to the maize ramosa2 gene that prevents ectopic branching in maize ears and tassels (Bortiri et al., 2006; Koppolu et al., 2013). In addition to controlling row‐type, VRS4 promotes spikelet and floret determinacy (Koppolu et al., 2013). VRS2 encodes a homologue of the Arabidopsis SHORT INTERNODES gene, whose loss of function is associated with hormonal imbalances between auxin and cytokinin along the spike (Youssef et al., 2017). VRS3 encodes a putative Jumonji C‐type (JMJC) H3K9me2/3 histone demethylase (Bull et al., 2017; van Esse et al., 2017), orthologous to the rice gene OsJMJ706 that regulates rice spikelet morphology (Sun & Zhou, 2008). Comparative transcriptomics suggests that VRS3 promotes the expression of other VRS genes (Bull et al., 2017; van Esse et al., 2017).

Understanding the molecular mechanisms underlying row‐type and its relationship to tillering and grain parameters may help to overcome the limitations of six‐rowed varieties. However, studies on row‐type are often confounded by the sensitivity of the phenotype to both genetic background and the environment. To address this, we assembled a panel of near‐isogenic vrs1−vrs5 alleles in a common two‐row cultivar (cv) Bowman recurrent parent, and used crossing to generate a series of near‐isogenic double mutants. By comparative analyses in single and double mutants, we asked several questions:Does lateral spikelet fertility develop in a similar way in different vrs mutants? How does variation in row‐type genes influence each other's expression and function?What is the genetic hierarchy of VRS gene function in the control of spike architecture? Can novel vrs allelic combinations increase tillering and grain homogeneity?

Materials and Methods

Germplasm and growth conditions

Barley (Hordeum vulgare ssp. vulgare) near‐isogenic lines for vrs1−vrs5 row‐type genes in cv Bowman (Druka et al., 2011) were obtained from The James Hutton Institute seed store. Allele designations, mutagen and molecular nature, backcross generation and common names used for clarity in this manuscript are given in Table 1. Double mutants vrs1vrs3, vrs1vrs4, vrs1vrs5, vrs3vrs4, vrs3,vrs5, and vrs4vrs5 were generated by crossing the relevant nearly isogenic lines and selecting homozygous progenies using allele‐specific molecular diagnostics (Supporting Information Method S1). Plants were grown in glasshouses under long‐day conditions (light : dark, 16 h : 8 h, 18°C : 14°C) in 15 cm (crossing and phenotyping) and 7 cm (microscopy and molecular analyses) diameter pots filled with a standard cereal compost mix.

Table 1.

Characteristics and nomenclature of alleles used throughout this article

Allele Mutagen Molecular nature Bowman near‐isogenic line (backcross generation) Names used here Comments
vrs1.a natural 1 bp deletion in exon 2 leading to frameshift (E152<)1 BW898 (BC=<7S3) vrs1 common 6‐row allele
vrs1.c natural unknown BW899 (BC6S3) vrs1.c Awnless/awnletted fertile lateral spikelets2
vrs2.e X‐ray 1 bp deletion leading to frameshift (C18<)3 BW901 (BC5S3) vrs2 limited fertility in basal lateral spikelets
int‐a.1 X‐ray 2 bp deletion leading to frameshift (R529<)4 , 5 BW419 (BC6S3) vrs3 weak vrs3 allele
vrs3.f X‐ray 1 bp deletion leading to frameshift (C133)4 , 5 BW902 (BC6S3) vrs3.f strong vrs3 allele
int‐e.58 EMS NS SNP in the coding region for LOB‐domain leading to A83V6 BW423 (BC6S3) vrs4 weak vrs4 allele
vrs4.k Gamma ray 1 bp deletion leading to frameshift (E17<)6 BW903 (BC6S3) vrs4.k strong vrs4 allele
int‐c.5 X‐ray 1 bp deletion causing frameshift (T279<)7 BW421 (BC6S3) vrs5 recessive
Open in a new tab

Alleles are in Hordeum vulgare ssp. vulgare. EMS, ethyl methanesulfonate; NS, nonsynonymous; SNP, single nucleotide polymorphism. 1Komatsuda et al., 2007; 2Konishi & Franckowiak, 1997; 3Youssef et al., 2017; 4Bull et al., 2017; 5van Esse et al., 2017; 6Koppolu et al., 2013; 7Ramsay et al., 2011.

Scanning electron microscopy

Main culm spikes harvested from plants between 9 and 28 d after germination (dag) were processed as described (Houston et al., 2013) and imaged using an environmental scanning electron microscope under a 10–15 kV acceleration voltage.

Phenotyping

Spikelet fertility of the main culm spike per individual was scored using grain and awn formation as a proxy at a whole spike scale for n = 10 spikes/genotype and along each rachis node for n = 5 spikes/genotype. The numbers of rachis nodes, and any additional branches and/or spikelets on the main spike were recorded. Tiller number was recorded at 5 d after germination (dag) and repeated at 5 d intervals until plants ceased to produce new tillers (n = 6–14 individuals/genotype). Tiller number included tillers with visible spikes as well as ‘vegetative tillers’ lacking visible spikes. Grain width, length, area and thousand grain weight (TGW), along with grain number per spike, were measured using a Marvin Grain Analyser (GTA Sensorik, Neubrandenburg, Germany) from grain harvested from the main culm and tallest tiller per individual (n = 10 individuals/genotype). Heading date was defined as the emergence of the main culm spike at 5 cm above the sheath of the upper leaf (Zadoks stage 55; Zadoks et al., 1974). Following Shapiro−Wilk tests for normality, pairwise comparisons were analysed with two‐tailed Student's t‐test and multiple comparisons with one‐way analysis of variance (ANOVA) combined with Tukey's test.

RNA extraction and cDNA synthesis

Total RNA was extracted from five to seven main culm spikes per biological replicate (n = 3 replicates/genotype) at double ridge (DR)/TSM, glume primordium (GP)/lemma primordium (LP), stamen primordium (SP)/awn primordium (AP) and white anther (WA) stages (according to Kirby and Appleyard, 1984) from Bowman and single vrs mutants (Fig. 2) and at GP, LP, SP and AP stages for double mutants, parents and Bowman (Fig. 5), using the RNeasy Plant Mini Kit (Qiagen) followed by on‐column DNA removal using a RNase‐Free DNase set (Qiagen). RNA integrity was determined by gel electrophoresis. cDNA was synthesised from 1 μg of RNA with oligo(dT)23VN primers (New England Biolabs (NEB), Ipswich, MA, USA) using ProtoScript® II Reverse Transcriptase (NEB).

Figure 2.

Figure 2

Open in a new tab

Hordeum vulgare ssp. vulgare single vrs mutants and Bowman (Bw) lateral spikelet ontogeny. Scanning electron microscopy of spikes at (a, b, e–h) early awn primordium (AP) and (c, d) later AP stage. (a′–h′) Close‐up images of lateral spikelets. (b′, d′, h′) White arrowheads indicate carpel primordia in lateral florets. (e′) Green arrowhead shows carpel primordium in vrs4.k additional spikelet. (c′, g′) Arrow shows the absence of carpel primordium in lateral florets; *, awn primordium. Bars: (a–h), 250 μm; (a′–h′), 50 μm. Staging according to Kirby & Appleyard (1984). cs, central spikelet; ls, lateral spikelet; fm, floret meristem.

Figure 5.

Figure 5

Open in a new tab

Defective VRS3 combined with either impaired VRS4 or VRS5 alleles in Hordeum vulgare ssp. vulgare leads to aberrant meristem identity and determinacy. (a–g) vrs3vrs4: (a) extended rachilla bearing a triplet; (b) multi‐noded branch from middle spike node; (c) glume to spikelet conversion (arrow); (d, e) scanning electron micrographs (SEMs) of spikes at lemma primordium (LP) stage. (f, g) SEM at stamen primordium (SP) stage. (h) SEM of Bowman (Bw) at SP stage. (i–j) SEM of vrs3vrs4 late awn primordium (AP) stage. ^ in (d) shows the central glume axillary meristem. White arrows in (d–g) show ectopic spike branch meristem formation, pink arrowheads show additional spikelet meristems formed adjacent to/in place of lateral spikelet glumes and white arrowhead shows the nascent spike branch. The blue arrowhead in (f) shows additional spikelet meristem in place of central glume. (k–o) vrs3vrs5: (k) basal node bearing a cluster of awned spikelets; (l) triplet from middle node of spike base; (m) SEM of late SP stage; (n) base of AP spike and (o) WA spike. Orange shading in (j) marks early multi‐floreted spikelet. Yellow shading (f, j, m–o) shows basal additional band and subsequent spikelet meristems. (p–r) Mean normalised expression (mne) determined by qPCR on spikes from glume primordium (GP), LP and SP‐staged spikes for: (p) Barley Meristem 3 (BM3) and Barley Meristem 8 (BM8), ± SE; (q) HvLONELY GUY1 (HvLOG1) and HvCYTOKININ OXIDASE2 (HvCKX2), ± SD; and (r) VRS5 and VRS1, ± SD; (n = 3). Significant differences to Bowman indicated by *, P < 0.05; **, P < 0.01; ***, P < 0.001 (Student's t‐test). Scales (a–c, k, l), 0.5 cm; (d–j, m–o), 100 μm.

Quantitative real‐time polymerase chain reaction (qRT‐PCR)

qRT‐PCRs were conducted according to Bustin et al. (2009), with primers, primer efficiency and R 2, as described in Method S1. Expression was considered differential if P‐values were < 0.05 with Student's t‐test (two‐tailed) and > 50% altered. Bar graphs represent mean normalised expression (mne; Simon, 2003).

Results

Variation in spike morphology across a row‐type near‐isogenic line panel

To compare vrs spike morphology directly, we grew and phenotyped Bowman and a set of Bowman near‐isogenic lines, each containing one of the five major recessive vrs alleles (Druka et al., 2011; Table 1). We used the induced int‐c.5 allele, not the natural Int‐c.a allele for two reasons: (1) int‐c.5 and not Int‐c.a had been introgressed into Bowman; and (2) we wanted to explore int‐c.5's genetic potential. We measured spikelet traits from 10 spikes/genotype (combined spikelets, Table 2; central, lateral and additional spikelets separated, Table S1) and trends along the rachis of five spikes/genotype (Fig. 1a,b; Table S2). As expected, Bowman lateral spikelets were rudimentary and infertile, such that each node presented a single grain derived from the central spikelet (Fig. 1a,b; Table 2). All row‐type alleles increased lateral spikelet development with distinct trends along the rachis (Fig. 1a,b; Tables S1, S2) that altered grain set per node (Table 2): vrs1 and the strong vrs4.k allele spikes showed the most robust lateral spikelet fertility, only lacking grain in lateral spikelets at the top and basal nodes, and increased grain set per node to 2.6 (± 0.2) and 2.9 (± 0.2) respectively; vrs2 spikes showed extensive sterility (0.6 ± 0.1 grain per node) and increased development of lateral spikelets only at basal nodes; while vrs3, vrs3.f (strong allele), vrs4 and vrs5 mutants showed lateral spikelet fertility in the middle to upper spike, leading to increases in grain set per node of between 1.6 (± 0.3) and 2.0 (± 0.9). While Bowman lateral spikelets were awnless, row‐type mutants’ lateral spikelets setting grain always developed awns (long needle‐like projections) off the lemma (awned spikes in Fig. S1), as reported in other cultivars (Gustafsson & Lundqvist, 1980; Lundqvist & Lundqvist, 1988), although awns on vrs5 lateral spikelets were small (< 2 cm long; Figs 1a, S1). However, many lateral spikelets were both empty and awned in vrs3 and vrs4 (Table S1). All row‐type mutants, except vrs2, also had a reduced number of rachis nodes (Fig. 1b; Table 2).

Table 2.

Spike parameters of Bowman, single vrs mutant and double vrs mutants

Line Rachis nodes Spikelets/spike (/node) Spikelets setting grain/spike (/node) Empty awned spikelets/spike (/node) Grain area (mm2)^ TGW (g)
Bw 22.4 ± 1.3 66.9 ± 3.4
(3.0 ± 0.0) 		19.1 ± 2.5
(0.9 ± 0.1) 		3.0 ± 1.5
(0.1 ± 0.1) 		27.2 ± 1.5 56.5 ± 4.6
vrs1 20.4 ± 0.5*** 60.3 ± 2.1*
(3.0 ± 0.1) 		52.7 ± 4.2***
(2.6 ± 0.2***) 		7.7 ± 4.2***
(0.4 ± 0.2***) 		22.2 ± 3.1 *** 37.7 ± 11 ***
vrs2 23.6 ± 0.5 73.6 ± 3.1*
(3.1 ± 0.2) 		14.6 ± 2.9*
(0.6 ± 0.1*) 		20 ± 1.4 d
(0.8 ± 0.0***) 		24.2 ± 3.1** 48.1 ± 7.6 ***
vrs3.f 20.6 ± 0.5 a 62.5 ± 1.6**
(3.0 ± 0.1) 		35.8 ± 7.4***
(1.7 ± 0.4***) 		8.75 ± 4.3**
(0.4 ± 0.2***) 		23.2 ± 4.1 39.6 ± 13 *
vrs3 19.5 ± 1.2*** 58.5 ± 3.7**
(3.0 ± 0.0) 		31.2 ± 4.7***
(1.6 ± 0.3***) 		13.5 ± 2.1***
(0.7 ± 0.1***) 		22.6 ± 0.9*** 36.8 ± 7.1***
vrs4.k 19.6 ± 1.1*** 78.4 ± 6.6***
(4.0 ± 0.3***) 		56.4 ± 4.9***
(2.9 ± 0.2***) 		13.4 ± 5.0***
(0.7 ± 0.3***) 		20.7 ± 1.7*** 34.9 ± 6.5***
vrs4 19.5 ± 1.2*** 58.5 ± 3.7**
(3.0 ± 0.0) 		37.3 ± 4.8***
(2.0 ± 0.9***) 		19.5 ± 6.1***
(0.9 ± 0.3***) 		22.2 ± 0.5*** 38.2 ± 2.3***
vrs5 19.5 ± 1.6*** 58.5 ± 4.9**
(3.0 ± 0.1) 		34.5 ± 3.2***
(1.8 ± 0.3***) 		8.5 ± 4**
(0.4 ± 0.2***) 		22.3 ± 4.4*** 43.1 ± 16***
vrs1vrs3 22.4 ± 2.7 70.8 ± 7.7
(3.1 ± 0.0) 		56.4 ± 9.5***
(2.5 ± 0.6***) 		11.0 ± 8.2*
(0.5 ± 0.3*) 		19.83 ± 1.5*** 20.7 ± 3.2***
vrs1vrs4 20.4 ± 0.4*** 61.2 ± 2.5*
(3.0 ± 0.1) 		49.4 ± 5.3***
(2.4 ± 0.2***) 		9.3 ± 3.3***
(0.5 ± 0.2***) 		18.1 ± 1.1*** 24.2 ± 5.8***
vrs1vrs5 18.2 ± 1.3*** 57.4 ± 5.5*
(3.2 ± 0.6) 		48.4 ± 7.8***
(2.7 ± 0.5***) 		3.20 ± 2.8
(0.2 ± 0.1) 		25.3 ± 1.42* 49 ± 4.5**
vrs3vrs4 21.6 ± 2.7 84 ± 8.6***
(3.9 ± 0.5***) 		43.8 ± 14.9***
(2.1 ± 0.8***) 		35.8 ± 16**
(1.7 ± 0.7***) 		21.4 ± 1.3*** 41.2 ± 2.4***
vrs3vrs5 17.4 ± 1.3*** 59.7 ± 4.5**
(3.4 ± 0.2) 		41.4 ± 2.6***
(2.4 ± 0.2***) 		14.7 ± 4.1***
(0.8 ± 0.2***) 		28.1 ± 1.4 54.1 ± 2.3
vrs4vrs5 17.4 ± 1.4*** 53.4 ± 4.1*
(3.1 ± 0.1) 		46.7 ± 5.6***
(2.7 ± 0.2***) 		3.8 ± 2.6(0.2 ± 0.2) 24.9 ± 0.9*** 46.1 ± 4.5***
Open in a new tab

Spike parameters of Hordeum vulgare ssp. vulgare single vrs mutants. Values are mean (± SD) from awned spikelets from n = 10 spikes per line. ^Averaged values for grains deriving from central, lateral and additional spikelets. Significant differences between Bowman and mutants indicated by *, P < 0.05; **, P < 0.01; ***, P < 0.001 (Student's t‐test). Bold text indicates no significant difference to Bowman. Bw, Bowman; TGW, thousand grain weight.

Figure 1.

Figure 1

Open in a new tab

Hordeum vulgare ssp. vulgare single vrs mutants show gradients in lateral spikelet fertility, awn development and additional spikelet formation along the rachis. (a) Spikes of Bowman (Bw) and vrs1, vrs4.k, vrs2, vrs5, vrs3.f, vrs3 and vrs4 mutants. Awns were removed from whole spikes for clarity. Representative spikelets from basal, central and apical nodes are displayed to the right of each spike. The scale bar in Bw (a) applies to all, except vrs4.k middle and basal spikelets for which the scale bar underneath vrs4.k basal spikelet applies. An asterisk indicates the small awn of the vrs5 lateral spikelet. Bar, 0.5 cm. (b) Distribution of grain‐setting fertile (green bars), and empty (grey bars) spikelets at individual rachis nodes along the main culm spike rachis. Bars represent mean values (± SD) calculated from main culm spikes (n = 5).

Consistent with Koppolu et al. (2013), vrs4.k spikes had 19.5 (± 5.6) fertile, awned, additional spikelets leading to an average of 4.0 (± 0.3) spikelets per node (Table 2), and occasionally developed short ectopic spike branches from middle nodes (Fig. 1a,b). The vrs2 spikes developed 2.8 (± 1.5) rudimentary, awnless additional spikelets in each of the two basal nodes (Fig. 1a,b; Tables S1, S2), similar to that found by Youssef et al. (2017). While, rarely, the middle nodes of vrs3.f developed one and the basal rachis nodes above the collar in vrs5 developed one to three awnless, infertile additional spikelets (Tables S1, S2).

Gain of rapid, coordinated emergence of carpel and awn primordia is associated with lateral spikelet fertility

We first explored whether lateral spikelet fertility across the panel initiated in the same way. Clear differences emerged by the early AP stage. Bowman lateral spikelets were small with short glumes, had limited lemma growth and no awn differentiation (Fig. 2a, a′). By contrast, vrs1 lateral spikelets were larger with well developed glumes and enlarged lemmas that were hooded with triangular awn primordia (Fig. 2b,b′). Similar to findings for central spikelets, vrs1 lateral spikelets had already initiated stamen and carpel primordia, while Bowman only initiated stamen primordia in late AP stage without accompanying carpel or awn primordia (Fig. 2c,c′), by which time vrs1 awns enclosed the developing floret organs (Fig. 2d,d′). Carpel and awn primordia in vrs4.k developed on both lateral and additional spikelets (Fig. 2e,e′). Lateral spikelets in vrs3.f, vrs4 and vrs5 also initiated carpel primordia and awns at the AP stage, except at some basal nodes, with more variable awn growth than in vrs1 (Figs 2f–h,f′–h′, S2). vrs5 uniquely showed butterfly‐shaped anther lobes in basal lateral spikelets compared with the less‐developed stamen primordia in Bowman and other row‐type alleles (Fig. S2). Therefore, lemma growth, awn initiation and carpel primordia emergence represent a suite of a morphological events normally repressed by row‐type gene function.

VRS gene expression is dependent on VRS gene function

To establish how VRS gene expression is related to lateral spikelet in/fertility, we constructed a gene expression by genotype profile by developmental stage matrix (Fig. 3; plotted values and statistics in Table S3). In Bowman, VRS1 was highly expressed at GP/LP, SP/AP and WA. VRS1 expression was reduced in vrs4.k (by 97% at GP/LP, 89% at SP/AP, and 99% at WA, P < 0.005), and in vrs3.f (87% at GP/LP, 69% at SP/AP, and 76% at WA, P < 0.005), showing that VRS4, and to a lesser extent VRS3, function is important to induce VRS1. However, VRS1 levels were only 55% reduced at GP/LP, and 47% at SP/AP and WA (P < 0.05) in vrs5, suggesting that VRS5 contributes less to VRS1 expression compared with VRS3 and VRS4.

Figure 3.

Figure 3

Open in a new tab

VRS gene expression in Hordeum vulgare ssp. vulgare single vrs mutants and Bowman (Bw) during spike development. Scanning electron micrographs in the top left panel show Bowman spikes at the double ridge/triple spikelet mersitem (DR/TSM), glume primordium/lemma primordium (GP/LP) and stamen primordium/awn primordium (SP/AP) and white anther (WA) stages. Bar graphs show the mean normalised expression (mne) (± SD) of VRS1‐5 genes determined by qPCR on DR/TSM, GP/LP, SP or WA‐staged spikes (n = 3). Significant differences to Bowman indicated by *, P < 0 0.05; **, P < 0 0.01; ***, P < 0.001 (Student's t‐test).

Consistent with inducing VRS1 expression, VRS4 levels were highest at DR/TSM and VRS3 at GP/LP. VRS5 transcripts were lowly expressed at all stages and VRS2 expressed most highly at WA. VRS3 expression was specifically reduced at GP/LP across all lines, suggesting that VRS3 expression is sensitive to the row‐type switch, potentially in a feed‐forward loop dependent on VRS1 activation. By contrast, VRS4 expression is only reduced in vrs1 and vrs2 at DR/TSM, although as this precedes significant VRS1 and VRS2 expression, this may be a pleiotropic effect. Rather, VRS4 transcript abundance increased later on, especially in vrs5 (117% at SP/AP, P < 0.05; 400% at WA), suggesting that VRS5 may be important to fine tune VRS4 expression during spikelet differentiation. VRS5 levels were lower in vrs3.f at GP/LP (50%, P < 0.05), and WA in vrs4.k (75%, P < 0.05). VRS2 expression was little altered across the lines.

Genetic analyses of double vrs mutants

To test regulatory associations amongst VRS genes in the control of row‐type and assess the potential of novel vrs combinations, we generated six double mutants: vrs1vrs3, vrs1vrs4, vrs1vrs5, vrs3vrs4, vrs3vrs5, and vrs4vrs5; and phenotyped as before, comparing with Bowman (Figs 4, S3; Tables 2, S1, S2) and comparing with parents (Table S4). Gains in vrs1vrs3, vrs1vrs4 and vrs1vrs5 lateral spikelet fertility increased grain number per spike greater than the vrs3/4/5 parents and equal those in vrs1 (Tables S1, S4), suggesting that VRS3, VRS4 and VRS5 work through VRS1 to control row‐type. However, grain distribution was noticeably more even along vrs1vrs5 spikes (Figs 1, 4), suggesting a more complete gain of lateral fertility. Strikingly, double vrs3, 4 and 5 mutants showed an increased level of lateral spikelet fertility similar to vrs1, despite containing the functional Vrs1 allele. vrs4vrs5 spikes increased grain number and showed a gain of lateral fertility in basal nodes compared with their parents, while vrs3vrs4 and vrs3vrs5 double mutants also generated awned lateral spikelets throughout their spikes (Figs 4, S3; Tables S1, S2, S4), although many of these spikelets were empty (Table S1). Rachis node number increased in vrs1vrs3 and vrs3vrs4 to be equivalent to Bowman, in contrast with other double mutants (Tables 2, S4), removing the rachis node penalty associated with lateral spikelet fertility.

Figure 4.

Figure 4

Open in a new tab

Hordeum vulgare ssp. vulgare double vrs mutants show novel routes to six rows. (a) Spikes of vrs1vrs3, vrs1vrs4, vrs1vrs5, vrs3vrs4, vrs3vrs5, and vrs4vrs5. The awns were removed from the whole spikes for clarity. Representative spikelets from basal, central and apical nodes are displayed to the right. The middle node of vrs3vrs4 contains a spike branch. Scale bar in vrs1vrs3 applies to all images. Bar, 0.5 cm. (b) Distribution of grain‐setting fertile (green bars), and infertile (grey bars) spikelets at individual rachis nodes along the main culm spike rachis. Bars represent mean values (± SD) calculated from the main culm spikes (n = 5).

VRS3 promotes spikelet identity and determinacy

In addition to gains in lateral spikelet fertility, novel vrs combinations also increased additional spikelet production. While vrs1vrs3, vrs1vrs5 and vrs4vrs5 spikes showed modest increases in additional spikelet formation from between one to four (Table S1), vrs3vrs4 spikes generated 19 (± 9.1) additional spikelets and one to two ectopic spike branches from middle nodes, similar to vrs4.k (Figs 4a,b, 5a–c; Table S1). These phenotypes may reflect a delayed or weakened progression from TSM to determinate spikelet and floret in vrs3vrs4 spikes, manifested by altered meristem identity and determinacy. First, the CSMs at GP‐LP transition elongated to form an abaxial boundary ridge and subsequent axillary meristem which could adopt branch identity (Fig. 5d–g). Second, axillary meristems formed within the glume axil of central and lateral spikelets (Fig. 5d,f), sometimes consuming the glume into a spikelet (Fig. 5c). Third, secondary spikelet meristems developed on the basal border of the lateral spikelets’ outer glume which could also consume the glume into an ectopic spikelet (Fig. 5f,g,i). Fourth, floret organ differentiation was noticeably delayed in LP to SP spikelet meristems compared with Bowman (Fig. 5f–h). Last, both canonical and additional spikelets could generate more than one floret and extra floral organs (Fig. 5j). The vrs3vrs4 double mutant also formed a continuous indeterminate lateral meristematic ridge between glume primordia of the central spikelet on the basal node which then split into additional spikelet meristems (Fig. 5f,i,j), fuelling the large increase in spikelet number at the basal node (Fig. 4; Table S2), a phenotype not observed in vrs4.k. Basal nodes of vrs3vrs5 spikes also developed clusters of additional spikelets and mutlifloreted spikelet bouquets with extra glumes (Figs 4a,b, 5h,i), similarly developing from an enlarged indeterminate basal meristem (Fig. 5m–o). Thus, both vrs3vrs4 and vrs3vrs5 spikes displayed meristem identity and determinacy defects.

To explore this further, we examined the expression of genes involved in determining meristem fate. In agreement with the vrs3vrs4 and vrs3vrs5 phenotypes, we found that transcripts of APETALA1/FRUITFULL homologue spikelet/FM identity genes, BM3 (HvMADS18) and BM8 (HvMADS15) (Schmitz et al., 2000; Trevaskis et al., 2007) were almost completely undetectable in vrs3vrs5 at LP (P < 0.05), while BM3 was 67% reduced in vrs3vrs4 (P < 0.05; 61% decline in BM8 was P > 0.05 for vrs3vrs4; Fig. 5p; Table S5). Furthermore, elevated transcripts of HvLONELY GUY1, which encodes a cytokinin biosynthetic enzyme, in vrs3vrs4 (P < 0.05) and at later stages for vrs3vrs5 (P < 0.005) which, coupled with lower HvCYTOKININ OXIDASE levels at most stages examined (P < 0.05), suggested a misregulation of cytokinin metabolism (Fig. 5q; Table S5), previously associated with additional spikelet and FMs in rice and barley (Han et al., 2014; Youssef et al., 2017). Expression of the barley orthologue of RAMOSA ENHANCER LIKE2 (REL2) was 50% reduced at LP (P < 0.05) in both double mutants (Fig. S4; Table S5), consistent with enhanced indeterminacy of maize rel2 mutants (Gallavotti et al., 2010). Notably, HvINDETERMINATE SPIKELET1 (HvIDS1) expression doubled at LP in vrs3vrs4 (P < 0.05; Fig. S4; Table S5). Ectopic expression of ZmIDS1 in maize is associated with extra FMs (Chuck et al., 2008). Taken together, these changes in gene expression are consistent with altered meristem fate in vrs3vrs4 and vrs3vrs5.

As many features of vrs3vrs4 resembled vrs4.k (Fig. 1), we also examined whether loss of VRS4 expression in vrs3vrs4 may explain its severe phenotypes. While reduced in vrs3vrs5 at GP, VRS4 expression was unaltered in vrs3vrs4 (Fig. S4; Table S5). This may reflect the increase in meristematic tissue load on these spikes, such that VRS4 level per meristem is actually reduced, or that these phenotypes are independent of effects on VRS4 expression. Both vrs3vrs4 and vrs3vrs5 had basal spikelet clusters, so we also hypothesised that downregulation of VRS5 in vrs3vrs4 may contribute to this phenotype. VRS5 in vrs3vrs4 was reduced by 91% at SP (P < 0.005) and 97% at AP (P < 0.005; Fig. 5r), far greater than either parent (Table S5), agreeing with a role for both VRS3 and VRS4 in VRS5 induction. Reduced VRS2 expression at LP in vrs3vrs5 (50%, P < 0.005) is also consistent with extra spikelet production in basal nodes (Fig. S4; Table S1). VRS1 expression was extremely low in both double mutants, in agreement with their awned, fertile spikelets (Fig. 5r, P < 0.0005).

Combined variation in VRS3, VRS4 and VRS5 associated with improved grain parameters

We next assessed whether novel vrs allelic combinations improve either grain weight and/or homogeneity compared with parents. We first measured grain parameters across the single mutant lines which showed decreased average grain area and reduced TGW compared to Bowman, highlighting a well known trade‐off between grain size and number (Liller et al., 2015; Table 2). However, vrs3, vrs3.f, vrs4 and vrs5 alleles had lower grain number per spike compared with vrs1 yet similar TGW (Tables 2, S4, S6), due to decreased lateral grain weight (Fig. 6; Table S7), suggesting that the relationship between TGW and grain number/spike is influenced by size differences between central and lateral grain, which varied amongst row‐type alleles. For instance, vrs1vrs3 and vrs1vrs4 set more or equal grain per spike as vrs1, and showed a lower or equivalent TGW compared with parents (Tables 2, S8). By contrast, despite increases in grain number per spike, vrs5 in combination with vrs1, vrs3 or vrs4, either increased grain area and/or TGW (Table S4) due to increased area, length, width and TGW of lateral grain to be equivalent to central grain of the single vrs5 parent (Fig. 6; Table S8). Central grain area and width of vrs3vrs5 double mutants exceeded the vrs5 parent (Fig. 6; Table S8), and most strikingly, significantly increased compared with Bowman (Fig. 6; Table S8). Additional spikelet grain from vrs3vrs5 was smaller compared with central and lateral grain (Table S8), but they contributed little to total grain (3.4%), such that overall grain area and TGW in vrs3vrs5 were equivalent to Bowman (Table 2). By contrast, pairing vrs3 with vrs4 decreased the area, length and weight of central grain to equal that of lateral and additional grain (Fig. 6; Table S8), potentially due to the very high spikelet load per spike. As grain from additional spikelets of vrs3vrs4, which contributed 18% of total grain (Table S1), showed equivalent weight to other spikelet grain resources may be more evenly distributed across vrs3vrs4 spikelets than other vrs mutants with additional spikelets which showed lower weight compared to central and lateral grain (Table S8). Decreased central grain weight in vrs3vrs4 reduced the central: lateral area ratio to 1.01 (± 0.1) while increased additional spikelet grain weight reduced the central: additional spikelet grain ratio to 1.11 (± 0.04) (Tables 3, S8). Taken together, entire grain lots from vrs3vrs4 showed superior grain homogeneity compared with all other lines (Fig. S5).

Figure 6.

Figure 6

Open in a new tab

Grain area for grain from lateral spikelets, central spikelets and additional spikelets from Hordeum vulgare ssp. vulgare Bowman (Bw), single vrs mutants and double vrs mutants. Violin plots show grain area distribution. Central spikelets shown in green, lateral spikelets shown in blue and additional spikelets shown in orange. CS, central spikelet; LS, lateral spikelet; AS, additional spikelet.

Table 3.

Grain area uniformity in double vrs mutants and their parents

Line Central : lateral Central : additional
vrs1 1.26 ± 0.12 na
vrs3 1.27 ± 0.09 na
vrs4 1.23 ± 0.10 na
vrs5 1.45 ± 0.08 na
vrs1vrs3 1.12 ± 0.04***(vrs1)/***(vrs3) 1.49 ± 0.32
vrs1vrs4 1.22 ± 0.05 na
vrs1vrs5 1.14 ± 0.04**(vrs1)/***(vrs5) 1.45 ± 0.17
vrs3vrs4 1.01 ± 0.07***(vrs3)/***(vrs4) 1.11 ± 0.04
vrs3vrs5 1.10 ± 0.07**(vrs3)/***(vrs5) 1.34 ± 0.09
vrs4vrs5 1.13 ± 0.05**(vrs4)/***(vrs5) 1.41 ± 0.25
Open in a new tab

Grain area uniformity calculated as a ratio of central grain area to lateral or additional grain area in Hordeum vulgare ssp. vulgare double vrs mutants and their parents. Values are mean (± SD) from grain harvested from n = 10 spikes per line. Significant differences between double vrs mutants and single vrs mutant parents by **, P < 0.01; ***, P < 0.001 (Student's t‐test). na, not applicable.

Row‐type allelic variation shows differential effects on tillering

In addition to reducing TGW, we noted that row‐type alleles also increased the proportion of fully‐awned, empty spikelets compared with Bowman (Table 2), consistent with a reduction in spikelet survival observed for six‐rowed varieties (Alqudah & Schnurbusch, 2014). The proportion of empty awned spikelets was particularly high in vrs3vrs4 spikes, potentially due to resources diverted to ectopic branches and many additional spikelets, as well as additional rachis nodes (Table 2). However, other architectural changes influenced by row‐type alleles such as the outgrowth of axillary basal buds or tillers, could be acting as competing energy sinks (Elalaoui et al., 1988; Gu & Marshall, 1988).

To explore, we first assessed tillering in single vrs mutants compared Bowman. Other than vrs2, all single mutants generated fewer tillers at maturity and tillered for a shorter period of time (Figs 7a,b, S6; Table 4). The tillering window was especially curtailed in vrs4.k and vrs5 which ceased tillering 20 d earlier than Bowman at 45 dag. Although tillers emerged at either that same (vrs1, vrs3 and vrs5, 0.47 tillers per day) or reduced rate (vrs4, 0.4 tillers per day) as Bowman, the vrs3.f mutant developed 0.5 tillers per day and the rate was even higher in vrs2 and vrs4.k at 0.72 and 0.54 tillers per day, respectively (Fig. S6). We analysed this trait further by comparing tiller number in vrs mutants to Bowman at 5 d intervals following germination (Fig. 7c). The vrs3.f, vrs4.k and vrs5 mutants had more tillers until 40 dag, after which tiller number was equivalent, only becoming lower than Bowman as tillering stopped in these lines and continued in Bowman. By contrast, vrs2 showed a higher tiller number only at and after 35 dag, suggesting a later tillering effect. Conversely, vrs1, vrs3 and vrs4 did not lead to increased tiller number at any stage but reduced tiller number at or shortly after the end of their tillering window. Early cessation of tillering was not associated with earlier heading times, which were either equivalent or later than Bowman, or with changes in plant height (Table 4). In summary, row‐type alleles fell into one of three classes of tiller outgrowth compared with the two‐rowed parent: (1) promotion of early tillering followed by early cessation (vrs3.f, vrs4.k and vrs5); (2) early cessation only (vrs1, vrs3, vrs4); or (3) promotion of late tillering (vrs2). Repeating this experiment in smaller pots with more resolution at earlier time points along with apex inspection, we found that tillers emerged earliest in vrs4.k and vrs5, followed by vrs3.f, when the apex was still a vegetative dome (Fig. S7). So, elevated tiller number at early stages may reflect an earlier release of bud dormancy compared with other lines. Notably, row‐type alleles linked to early increases in tillering rate (vrs4.k, vrs3.f, and vrs5) showed a longer TSM duration and were also associated with additional spikelet production (Fig. S7).

Figure 7.

Figure 7

Open in a new tab

Tiller outgrowth across Hordeum vulgare ssp. vulgare Bowman (Bw), single vrs mutants and double vrs mutants. (a) Final tiller number at harvest for Bowman and single vrs mutants. Horizontal bars in each box represent mean values (± SE) calculated from main culm spikes (n = 7–12). (b) Tillering window in single vrs mutants. (c) Tiller outgrowth during development in Bowman and single vrs mutants and selected double mutants. Dots represent mean values (± SE) calculated from main culm spikes (n = 7–12). Significant differences between Bowman and vrs mutants indicated by lower colour panel (P < 0.05). Green indicates statistical increase in tiller number, beige shows no difference and black indicates a statistical decrease in tiller number, compared with Bowman. (d) Pictures of single vrs mutants 28 d after sowing. (e) Final tiller number determined at harvest for Bowman and double vrs mutants. Horizontal bars in each box represent mean values (± SE) calculated from main culm spikes (n = 7–12). (f) Tiller outgrowth during development single vrs mutant parents and selected double mutants. Significant differences to Bowman in (a) and (e) indicated by *, P < 0.05; **, P < 0.01; ***, P < 0.001 (Student's t‐test). Box plot horizontal line denotes the mean, while the box shows the 95% confidence interval. Bars, (d) 1 cm. Bw, Bowman.

Table 4.

Whole plant traits and phenology for Bowman (Bw), single vrs mutants and double vrs mutants

Line Final tiller number Heading (dag) Plant height (cm) Spike length (cm)
Bw 27.0 ± 3.2 38.6 ± 1.8 92.3 ± 9.5 8.90 ± 0.58
vrs1 21.3 ± 2.3** 39.2 ± 3.2 91.3 ± 4.9 8.60 ± 0.45
vrs2 36.8 ± 2.8*** 45.4 ± 3.2 88.7 ± 7.6 9.20 ± 0.6
vrs3.f 22.5 ± 2.6** 43.4 ± 2.9 90.1 ± 4.1 9.00 ± 0.43
vrs3 21.6 ± 2.1*** 41.5 ± 2.2 85.4 ± 10 8.37 ± 0.64
vrs4.k 18.0 ± 1.4*** 39.0 ± 3.2 100 ± 2.0* 7.80 ± 0.48***
vrs4 19.3 ± 5.2*** 48.3 ± 6.0*** 105 ± 8.8** 8.45 ± 0.55
vrs5 17.6 ± 2.9*** 44.7 ± 3.8*** 86.2 ± 7.0 8.03 ± 0.53***
vrs1vrs3 21 ± 3.7** 45.6 ± 4.6*** 101 ± 14 9.22 ± 0.87
vrs1vrs4 22.2 ± 5.5** 39.9 ± 1.8 90.2 ± 6.9 7.95 ± 0.66
vrs1vrs5 11.5 ± 2.8*** 42.8 ± 1.3*** 84.3 ± 4.9* 7.64 ± 0.26***
vrs3vrs4 26.1 ± 6.1 51.9 ± 5.2*** 86.1 ± 7.0 9.66 ± 0.79*
vrs3vrs5 8.95 ± 2.0*** 57.6 ± 11*** 82.8 ± 7.0** 8.12 ± 0.61
vrs4vrs5 13.2 ± 2.0*** 44.8 ± 5.7*** 101 ± 8.3* 7.98 ± 0.55***
Open in a new tab

Whole plant traits and phenology in Hordeum vulgare ssp. vulgare single vrs mutants, double vrs mutants and Bowman. Values are the mean (± SD) from n = 10 individuals per line. Bold text indicates no significant difference with Bowman. Significant differences with Bowman indicated by *, P < 0.05; **, P < 0.01; ***, P < 0.001 (Student's t‐test). dag, days after germination.

We next examined whether combining vrs alleles within and between classes altered tiller number (Fig. 7f; Tables 4, S4). The vrs1vrs3 and vrs1vrs4 mutants generated fewer tillers at harvest compared with Bowman, equivalent to their parental lines. Double mutants with vrs5 showed reduced tiller number compared with both parents. Suppression in tiller number may compensate for increased fertile spikelet load per spike and may explain why vrs1vrs5 and vrs4vrs5 double mutants showed an equivalent number of empty awned spikelets as Bowman (Table 2). However, vrs3vrs5 plants were especially deficient in tillers (Fig. 7e,f; Table S4), reflecting both the shortest tillering window (25 d) and lowest overall rate (0.31 tillers per day; Figs 7f, S6) across all genotypes, despite having the latest heading date, suggesting that resources diverted to tillering may not be linked to the substantial loss of grain set in otherwise fertile vrs3vrs5 spikelets. In striking contrast with vrs3vrs5, vrs3vrs4 plants had the same tiller number as Bowman, increased compared with their parents and vrs4.k (Fig. 7f; Tables 2, S4). As the tillering window of vrs3vrs4 was unchanged compared with their parents, the increased tiller number reflected a higher overall tillering rate (0.53 tillers per d; Figs 7f, S6). Increased tillering coincides with stem elongation to heading (Table 4), when spikelets are particularly susceptible to abortion (Alqudah et al., 2016), which may contribute to absence of grain in otherwise fertile spikelets, in addition to the ectopic awned spikelet and floret production in vrs3vrs4.

Discussion

Developmental basis for awn association with lateral fertility

Our work suggests that VRS1 acts as a downstream integrator of VRS3, VRS4 and VRS5 to repress both lateral spikelet fertility and awn development. Coordinated initiation of carpel and awn primordia in larger lateral spikelets may reflect a linked step in the spikelet developmental programme, the ‘vital connection’ speculated by Brenchley (1920), specifically targeted by VRS1. Carpels initiate when the central FM differentiates into a determinate organ primordium (Smyth et al., 1990; Schoof et al., 2000) and the AP itself may represent a proliferating ‘quasi meristem’ (Girin et al., 2009; Toriba & Hirano, 2014). Therefore, VRS1 may block carpel and awn initiation at AP stage by inhibiting localised meristem proliferation, leading to a dormant‐like state. Indeed, VRS1 was recently shown to repress proliferation in both leaf and lateral spikelet primordia (Sakuma et al., 2017; Thirulogachandar et al., 2017). Lateral spikelets in a different vrs1 mutant to that studied here, vrs1.c, are fully fertile, yet only develop very short lemma extensions (Konishi & Franckowiak, 1997; Ullrich, 2011), suggesting VRS1 control of lateral spikelet fertility and awn development can be uncoupled. As vrs1.c contains no coding sequence mutations (Saisho et al., 2009), potential differential cis‐regulation between the awn and carpel may be relevant.

Lateral spikelet in/fertility shows parallels with sex determination in maize

Lateral spikelets in two‐row barley have rudimentary stamens yet exceptionally reduced ovaries and are sometimes referred to as staminate (Beavan, 1902; Brenchley, 1920; Komatsuda et al., 2007), a state that may originate at the AP stage when infertile lateral spikelets initiate stamen but not carpel primordia (Fig. 2). Stamen growth appeared particularly robust in vrs5 mutants and our previous work showed that grain set in vrs5 (Intermedium‐c) mutants in a functional VRS1 (Vrs1.b) background was correlated with lateral spikelet male fertility (Ramsay et al., 2011). We surmise that VRS5 may be more important to repress male fertility, while initiation of female organs may be more sensitive to VRS1. The possibility of sex‐specialised regulation by VRS1 and VRS5 shows an overlap with their homologues in sex determination of the maize ear and tassel. The maize VRS1 homologue, grassy tillers1 (gt1), promotes carpel abortion to generate the distinctive staminate spikelets of the tassel, while loss of function leads to carpel development and feminised tassels (Whipple et al., 2011). Conversely, reduced expression of the maize VRS5 orthologue teosinte branched 1 (tb1) causes tassel‐tipped ears and conversion of ear spikelets to tassel spikelets, suggesting that tb1 excludes male identity from the ear (Doebley et al., 1995; Hubbard et al., 2002). Differential expression probably contributes to sex determination, with tb1 expression strongly enhanced in stamen primordia of the ear spikelet and gt1 specifically expressed in carpels of the tassel spikelet (Hubbard et al., 2002; Whipple et al., 2011). By contrast, VRS1 is expressed in all lateral spikelet organs, although enriched in developing carpels and lemmas (Sakuma et al., 2013). The VRS5 expression pattern is yet unknown and it would be interesting to determine whether the role of VRS5 in male infertility is associated with increased expression in stamen vs carpel primordia.

VRS3 as an epigenetic modifier of row‐type gene function

Loss of VRS3 function along with defective VRS4 or VRS5 alleles caused synergistic losses in meristem determinacy, potentially by regulating their expression as well as that of shared regulatory targets. Lundqvist & Lundqvist (1988) described the modifier ability of vrs3 as having a ‘rigid’ nature that forced other alleles to ‘show their hands’. Consistent with this type of control, VRS3 encodes a putative Jumonji C‐type H3K9me2/me3 demethylase (Bull et al., 2017; van Esse et al., 2017), predicted to control gene expression through regulating chromatin state. An epigenetic mode of action is also consistent with the phenotypic instability conferred by vrs3 alleles (Bull et al., 2017) and reported for lateral spikelets and awns in general (Engledow, 1921). The vrs3vrs4 and vrs3vrs5 double mutants showed fully six‐rowed phenotypes associated with a complete loss of VRS1 expression. VRS5 contributes modestly to VRS1 induction, so VRS3 probably mediates an additional inductive effect on VRS1. While this mechanism remains unknown, VRS3 could act epigenetically to stabilise local VRS1 chromatin in an active transcriptional state, similar to the spatial regulation of Drosophila Hox genes (Bantignies & Cavalli, 2006; Ringrose & Paro, 2007), that may buffer against impaired VRS4 and VRS5 function. Furthermore, although VRS4 is key to VRS1 induction (Koppolu et al., 2013; this paper), VRS4 transcripts significantly decline, especially at AP and later stages while VRS1 levels increase (Koppolu et al., 2013; Sakuma et al., 2013; Fig. 3), suggesting that another factor, which could involve VRS3, maintains VRS1 expression. In maize, the paramutator Rmr6 locus is required to maintain epigenetic states conferred by long noncoding RNA molecules to prevent carpel emergence in tassels (Parkinson et al., 2007). It will be important to define the chromatin modifications associated with VRS3 activity and whether DNA methylation and long noncoding RNAs are involved in VRS3‐mediated epigenetic regulation.

Lateral spikelet fertility regulators influence meristem identity and determinacy

Alteration in the timing of meristem transitions and meristem maturation is associated with variation in inflorescence branching and floral architecture across plants (McKim & Hay, 2010; Yoshida et al., 2013; Park et al., 2014; Lemmon et al., 2016; Ta et al., 2017). The barley inflorescence forms as the lateral meristems transition from DR to TSM to SM to FM identity. In vrs3vrs4, indeterminate inflorescence branches bud from the CSM, potentially due to disrupted progression of this pathway. Similarly, multiple FMs within the vrs3vrs4 and vrs3vrs5 spikelets may reflect a delayed transition from SM (capable of giving rise to florets) to FM, while extra floral organs suggest a problem in determinate floret identity. Branches do not form at the DR or TSM stage of vrs3vrs4 or vrs4.k, so pathways promoting DR and TSM identity may be less sensitive to defective VRS4 or VRS3 function. Rather, these genes may be more important for promoting the transition to determinate spikelet meristem identity and/or repress branching identity, especially in the central spikelet. Branches also do not develop from LSMs, suggesting that spikelet identity may be more complete by the time of LSM transition, which occurs after CSM determination. However, vrs3vrs4 lateral spikelets initiate extensive flanking secondary spikelets, often off, or at the expense of, the glume (an event also observed in central spikelets), which could represent a shift to indeterminacy after spikelet identity is established (Bommert & Whipple, 2018). In vrs3vrs5 and vrs3vrs4, basal indeterminate lateral meristems develop between central spikelet glumes, leading to additional spikelets with extra floral organs later in spike development, consistent with delayed acquisition of determinate spikelet and floret fate and aberrant glume development. We detected the lower expression of spikelet meristem identity genes BM3 and BM8 in vrs3vrs4 and vrs3vrs5 mutants and we suspect that other genes known to regulate spikelet identity and/or determinacy in barley or other grasses, such as COMPOSITUM2, could also be misregulated, as shown for vrs4.k (Poursarebani et al., 2015). Additional spikelet development observed in basal nodes of vrs2 spikes was associated with disrupted hormone gradients characterised by increased cytokinin and lower auxin levels (Youssef et al., 2017). VRS3 and VRS5 may also be important to establish hormone gradients to properly pattern the spike, as supported by reduced VRS2 expression and misexpression of cytokinin metabolic genes in vrs3vrs5.

While VRS3, VRS4 and VRS5 may work through VRS1 to repress lateral spikelet growth, vrs1vrs5, vrs1vrs3 and vrs1vrs4 double mutants showed only a small increase in additional spikelet formation that, along with VRS1 temporal and spatial expression patterns, suggested a minor role for VRS1 in spikelet meristem identity and determinacy. The paralogue of VRS1, HvHOX2 is strongly expressed in the barley rachis vasculature (Sakuma et al., 2010, 2013), as is VRS3 (Bull et al., 2017), while VRS4 is localised in the rachis and the spikelet−rachis boundary (Koppolu et al., 2013). Increased expression of gt1 (VRS1 orthologue) in the vascular nodal plexus in maize prevents secondary ear formation (Wills et al., 2013). Given these partially overlapping expression patterns, VRS3, VRS4 and/or VRS5 gene function may promote expression of HvHOX2 in rachis node either directly on nonautonomously from the spikelet boundary to promote determinacy and identity of the adjacent meristem, which may involve signalling from the boundary, as suggested for the ramosa pathway in maize, and/or from the suppressed bract as suggested by the plastochron mutants in rice (Bommert & Whipple, 2018).

Other mutants with spike branching and extra spikelets, such as compositum1 and poly‐and‐branched‐row spike also increase lateral spikelet development (Larsson, 1985; Shang et al., 2014) that, taken with the results here, suggest a profound link between control of LSM growth and spikelet and floret identity. Dixon et al. (2018) showed that increased dosage of TB1 causes a paired spikelet phenotype in wheat (where an extra spikelet forms immediately beneath the primary spikelet) associated with reduced spikelet meristem identity gene expression and delayed spikelet meristem maturation. Dixon et al. (2018) speculated that VRS5 (HvTB1) similarly blocks lateral spikelet progression to floral fate, thereby leading to bud dormancy (and lateral spikelet infertility) and predicted that impaired vrs5 alleles release lateral spikelet dormancy by enabling expression of floral genes. Our ontogenetic analysis suggests that lateral spikelets in two‐rowed spikes are blocked at carpel emergence, consistent with an arrest of the floral programme, but also that these LSMs are smaller at the onset of floral differentiation. Acquisition of determinate floral fate must be carefully timed to allow floral meristems to proliferate sufficiently to fuel a full complement of floral organs (McKim & Hay, 2010). Our data suggest that vrs5, as well as vrs4.k and vrs3.f, alleles prolong the TSM phase, potentially allowing increased proliferation before acquisition of spikelet fate. A comparison of the effects of TB1 in wheat and VRS5 in barley on stage‐specific meristem maturation transcriptomes may help to understand the divergence and conservation of the pathways that regulate inflorescence to spikelet to floret identity in the Triticeae.

Tillering and VRS5

In maize, TB1 induction of gt1 is necessary to promote apical dominance, especially under unfavourable conditions (Whipple et al., 2011). In barley, however, VRS5 appears less important for VRS1 expression, and defects in both genes are associated with fewer tillers, due to a shortened tillering window compared with the characteristic ‘bushy’ phenotype of tb1 and gt1 mutants (Doebley et al., 1995; Whipple et al., 2011). It will be interesting in the future to examine whether axils without visible tillers have underdeveloped, multiple or dormant tiller buds. Previous studies in barley showed an early increase in tillering in vrs5 (Ramsay et al., 2011; Liller et al., 2015) that we also detected here, suggesting that weakened VRS5 function is associated with an early release of tiller bud dormancy. Unlike vrs1 and vrs5, mutations in gt1 and tb1 are not reported to increase grain number in the inflorescence.

Novel routes to grain improvement

Grain yield and quality is the cumulative output of the barley life cycle and is reflective of source/sink relationships in the context of developmental time (Slafer, 2003). Altering the balance of these latter relationships could therefore impact components of yield. Indeed, our data show that different single and double vrs alleles introduce perturbations in the numbers of fertile spikelets per spike, spikelet uniformity, grain size and/or the number of spikes per plant through increased, decreased or arrested tillering. As an example, the early tillering arrest we observed in vrs5 plants could reflect resource allocation to spikes with an increased number of fertile spikelets. However, we believe this is unlikely as multiple lines, such as vrs1.a and vrs4.k, vrs1vrs3, vrs3vrs4, had both more grain‐setting spikelets per spike and tillers compared with vrs5, suggesting that changes in tillering due to defective vrs gene function is unlikely to be a simple read out of energy demand. Of relevance here is the deficiens allele of VRS1 (Vrs1.t1, Sakuma et al., 2017) that has recently had a significant effect on barley breeding in NW Europe. Vrs1.t1 almost completely curtails development of lateral spikelets and is associated with increased cell proliferation in the central spikelets that may promote larger grains and higher productive yield potential by allocating more resources directly to the central grain (Sakuma et al., 2017). In vrs3vrs4 and vrs3vrs5 double mutants (containing wild‐type VRS1) we observed improved grain homogeneity, an important quality characteristic of malting barley. From a developmental perspective, this may reflect early equivalency in floral organ size that may have practical value in end‐use quality improvement. Our investigations therefore revealed both shared and independent roles for VRS genes in conferring lateral spikelet infertility, meristem determinacy and divergence in tillering regulation, and may provide novel routes to improving six‐row barley grain quality. Our recent work showing improved grain parameters in vrs1.a Int‐c.a vrs3, as described in Bull et al. (2017), highlights efforts to integrate novel row‐type alleles into the cultivated germplasm.

Author contributions

SMM and RW conceived the research. SMM, RW and MZ designed the research. MZ performed the research and analysed the data. SMM, RW and MZ wrote the manuscript.

Supporting information

Please note: Wiley Blackwell are not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

Fig. S1 Spikes of Bowman and single vrs mutants before awn removal.

Fig. S2 Spikes of vrs3.f, vrs4, vrs5, vrs1 and Bowman at late awn primordium stage.

Fig. S3 Spikes of vrs double mutants before awn removal.

Fig. S4 Differential gene expression of meristem regulators in Bowman, vrs3vrs4 and vrs3vrs5 mutants.

Fig. S5 Combined grain area from Bowman, single vrs mutants and double vrs mutants.

Fig. S6 Tiller outgrowth rate over time in Bowman and single vrs mutants and double vrs mutants.

Fig. S7 Early tillering phase across Bowman and single vrs mutants.

Method S1 Supplemental methods for genotyping and qRT‐PCR.

Click here for additional data file. (1.5MB, pdf)

Table S1 Spike and spikelet traits separated by central, lateral and additional spikelets in Bowman, single vrs mutants and double vrs mutants.

Table S2 Spikelet by rachis node for Bowman, single vrs mutants and double vrs mutants.

Table S3 VRS gene expression in Bowman and single vrs mutants.

Table S4 Comparison of spikelet traits between double vrs mutants, their parents and Bowman.

Table S5 Gene expression in Bowman, vrs3vrs4 and vrs3vrs5 double mutants and vrs3, vrs4 and vrs5 parents.

Table S6 Spikelet parameters in single vrs mutants and Bowman compared by ANOVA.

Table S7 Grain parameters in vrs single mutants and Bowman compared by ANOVA.

Table S8 Grain parameters in vrs double mutants compared with their parents and Bowman by ANOVA.

Click here for additional data file. (106.1KB, xlsx)

Acknowledgements

We are grateful to three anonymous reviewers for helpful suggestions and comments. We thank Nicola Bonar, Hazel Bull, Cristina Casao, Vrushali Patil, Scott McCrimmon, Selver Babi, Duygu Sari‐Yol, Cristiane Calixto, Jennifer Stephens and Marcel Lafos for experimental assistance and helpful suggestions. We also thank Arnis Druka for producing some double mutant crosses early in the project. MZ was supported by the Biotechnology and Biological Sciences Research Council grant no. BB/K01613X/1 to RW and SMM. RW would also like to acknowledge funding from the Scottish Government's Rural and Environment Science and Analytical Services Division Work Program 2.1 and 2.2 and RW and SMM for support from the University of Dundee. SMM acknowledges support as a Royal Society of Edinburgh Personal Research Fellow.

References

  1. Alqudah AM, Koppolu R, Wolde GM, Graner A, Schnurbusch T. 2016. The genetic architecture of barley plant stature. Frontiers in Genetics 7: 117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alqudah AM, Schnurbusch T. 2014. Awn primordium to tipping is the most decisive developmental phase for spikelet survival in barley. Functional Plant Biology 41: 424–436. [DOI] [PubMed] [Google Scholar]
  3. Bantignies F, Cavalli G. 2006. Cellular memory and dynamic regulation of polycomb group proteins. Current Opinion in Cell Biology 18: 275–283. [DOI] [PubMed] [Google Scholar]
  4. Beavan ES. 1902. Varieties of barley. Journal of the Institute of Brewing and Distilling September–October: 542–600. [Google Scholar]
  5. Bommert P, Whipple C. 2018. Grass inflorescence architecture and meristem determinacy. Seminars in Cell and Developmental Biology 79: 37–47. [DOI] [PubMed] [Google Scholar]
  6. Bortiri E, Chuck G, Vollbrecht E, Rocheford T, Martienssen R, Hake S. 2006. ramosa2 encodes a LATERAL ORGAN BOUNDARY domain protein that determines the fate of stem cells in branch meristems of maize. Plant Cell 18: 574–585. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Brenchley WE. 1920. The development of the flower and grain of barley. Journal of the Institute of Brewing 26: 615–632. [Google Scholar]
  8. Bull H, Casao MC, Zwirek M, Flavell AJ, Thomas WTB, Guo W, Zhang R, Rapazote‐Flores P, Kyriakidis S, Russell J et al 2017. Barley SIX‐ROWED SPIKE3 encodes a putative Jumonji C‐type H3K9me2/me3 demethylase that represses lateral spikelet fertility. Nature Communications 8: 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bustin SA, Benes V, Garson JA, Hellemans J, Huggett J, Kubista M, Mueller R, Nolan T, Pfaffl MW, Shipley GL et al 2009. The MIQE guidelines: minimum information for publication of quantitative real‐time PCR experiments. Clinical Chemistry 55: 611–622. [DOI] [PubMed] [Google Scholar]
  10. Chuck G, Meeley RB, Hake S. 2008. Floral meristem initiation and meristem cell fate are regulated by the maize AP2 genes ids1 and sid1 . Development 135: 3013–3019. [DOI] [PubMed] [Google Scholar]
  11. Chuck G, Meeley R, Irish E, Sakai H, Hake S. 2007. The maize tasselseed4 microRNA controls sex determination and meristem cell fate by targeting Tasselseed6/indeterminate spikelet1. Nature Genetics 39: 1517–1521. [DOI] [PubMed] [Google Scholar]
  12. Dixon L, Greenwood J, Bencivenga S, Zhang P, Cockram J, Mellers G, Ramm K, Cavanagh C, Swain SM, Boden SA. 2018. TEOSINTE BRANCHED1 regulates inflorescence architecture and development in bread wheat (Triticum aestivum). Plant Cell 30: 563–581. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Doebley J, Stec A, Gustus C. 1995. teosinte branched1 and the origin of maize: evidence for epistasis and the evolution of dominance. Genetics 141: 333–346. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Doebley J, Stec A, Hubbard L. 1997. The evolution of apical dominance in maize. Nature 386: 485–488. [DOI] [PubMed] [Google Scholar]
  15. Druka A, Franckowiak J, Lundqvist U, Bonar N, Alexander J, Houston K, Radovic S, Shahinnia F, Vendramin V, Morgante M et al 2011. Genetic dissection of barley morphology and development. Plant Physiology 155: 617–627. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Elalaoui AC, Simmons SR, Crookston RK. 1988. Effects of tiller removal on spring barley. Crop Science 28: 305–307. [Google Scholar]
  17. Engledow FL. 1921. Inheritance in barley. II. The awn and the lateral floret. The Journal of Agricultural Science 11: 159–196. [Google Scholar]
  18. van Esse GW, Walla A, Finke A, Koornneef M, Pecinka A, von Korff M. 2017. Six‐Rowed Spike3 (VRS3) is a histone demethylase that controls lateral spikelet development in barley. Plant Physiology 174: 2397–2408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Gallavotti A, Long JA, Stanfield S, Yang X, Jackson D, Vollbrecht E, Schmidt RJ. 2010. The control of axillary meristem fate in the maize ramosa pathway. Development 137: 2849–2856. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Girin T, Sorefan K, Ostergaard L. 2009. Meristematic sculpting in fruit development. Journal of Experimental Botany 60: 1493–1502. [DOI] [PubMed] [Google Scholar]
  21. Gu J, Marshall C. 1988. The effect of tiller removal and tiller defoliation on competition between the main shoot and tillers of spring barley. Annals of Applied Biology 112: 597–608. [Google Scholar]
  22. Gupta M, Abu‐Ghannam N, Gallaghar E. 2010. Barley for brewing: characteristic changes during malting, brewing and applications of its by‐products. Comprehensive Reviews in Food Science and Food Safety 9: 318–328. [DOI] [PubMed] [Google Scholar]
  23. Gustafsson A, Lundqvist U. 1980. Hexastichon and intermedium mutants in barley. Hereditas 92: 229–236. [Google Scholar]
  24. Han Y, Yang H, Jiao Y. 2014. Regulation of inflorescence architecture by cytokinins. Frontiers in Plant Science 5: 669. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Harlan JR. 1968. On the origin of barley In: Wiebe GA, ed. Barley: origin, botany, culture, winter‐hardiness, genetics, utilization, pests. Agriculture Handbook No. 338, Washington, DC, USA: ARS/USDA, 9–31. [Google Scholar]
  26. Helbaek H. 1959. Domestication of food plants in the Old World. Science 130: 365–372. [DOI] [PubMed] [Google Scholar]
  27. Houston K, McKim SM, Comadran J, Bonar N, Druka I, Uzrek N, Cirillo E, Guzy‐Wrobelska J, Collins NC, Halpin C et al 2013. Variation in the interaction between alleles of HvAPETALA2 and microRNA172 determines the density of grains on the barley inflorescence. Proceedings of the National Academy of Sciences, USA 110: 16675–16680. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Hubbard L, McSteen P, Doebley J, Hake S. 2002. Expression patterns and mutant phenotype of teosinte branched1 correlate with growth suppression in maize and teosinte. Genetics 162: 1927–1935. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kellogg EA. 2015. Flowering plants. Monocots: Poaceae In: Kubitzki K, ed. The families and genera of vascular plants, vol 13 Cham, Switzerland: Springer International Publishing, 25–38. [Google Scholar]
  30. Kirby EJM, Appleyard M. 1984. Cereal development guide, 2nd edn Stoneleigh, UK: NAC Cereal Unit. [Google Scholar]
  31. Kirby EJM, Riggs TJ. 1978. Developmental consequences of two‐row and six‐row ear type in spring barley. II. Shoot apex leaf and tiller development. Journal of Agricultural Sciences 91: 207–216. [Google Scholar]
  32. Komatsuda T, Pourkheirandish M, He C, Azhaguvel P, Kanamori H, Perovic D, Stein N, Graner A, Wicker T, Tagiri A et al 2007. Six‐rowed barley originated from a mutation in a homeodomain‐leucine zipper I‐class homeobox gene. Proceedings of the National Academy of Sciences, USA 104: 1424–1429. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Konishi T, Franckowiak JD. 1997. Six‐rowed spike 1, vrs1.c. Barley Genetics Newsletter 26: 94. [Google Scholar]
  34. Koppolu R, Anwar N, Sakuma S, Tagiri A, Lundqvist U, Pourkheirandish M, Rutten T, Seiler C, Himmelbach A, Ariyadasa R et al 2013. Six‐rowed spike4 (Vrs4) controls spikelet determinacy and row‐type in barley. Proceedings of the National Academy of Sciences, USA 110: 13198–13203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Lang L, Rakszegi M, Bedo Z. 2013. Cereal production and its characteristics. Engineering aspects of cereal and cereal‐based products. Boca Raton, FL, USA: CRC Press. [Google Scholar]
  36. Larsson HE. 1985. Morphological analysis of laxatum barley mutants. Hereditas 103: 239–253. [Google Scholar]
  37. Lemmon ZH, Park SJ, Jiang K, Van Eck J, Schatz M, Lippman ZB. 2016. The evolution of inflorescence diversity in the nightshades and heterochrony during meristem maturation. Genome Research 26: 1676–1686. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Lewis JM, Mackintosh CA, Shin S, Gilding E, Kravchenko S, Baldridge G, Zeyen R, Muehlbauer GJ. 2008. Overexpression of the maize Teosinte Branched1 gene in wheat suppresses tiller development. Plant Cell Reports 27: 1217–1225. [DOI] [PubMed] [Google Scholar]
  39. Liller CB, Neuhaus R, von Korff M, Koornneef M, van Esse W. 2015. Mutations in barley row type genes have pleiotropic effects on shoot branching. PLoS ONE 10: e0140246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Lundqvist U, Franckowiak JD, Konishi T. 1997. New and revised description of barley genes. Barley Genetics Newsletter 26: 22–516. [Google Scholar]
  41. Lundqvist U, Lundqvist A. 1988. Induced intermedium mutants in barley: origin, morphology and inheritance. Hereditas 108: 13–26. [Google Scholar]
  42. McKim SM, Hay A. 2010. Patterning and evolution of floral structures – marking time. Current Opinion in Genetics and Development 20: 448–453. [DOI] [PubMed] [Google Scholar]
  43. Park SJ, Eshed Y, Lippman Z. 2014. Meristem maturation and inflorescence architecture–lessons from the Solanaceae. Current Opinion Plant Biology 17: 70–77. [DOI] [PubMed] [Google Scholar]
  44. Parkinson SE, Gross SM, Hollick JB. 2007. Maize sex determination and abaxial leaf fates are canalized by a factor that maintains repressed epigenetic states. Developmental Biology 308: 462–473. [DOI] [PubMed] [Google Scholar]
  45. Poursarebani N, Seidensticker T, Koppolu R, Trautewig C, Gawro?ski P, Bini F, Govind G, Rutten T, Sakuma S, Tagiri A et al 2015. The genetic basis of composite spike form in barley and ?Miracle‐Wheat’. Genetics 201: 155‐165. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Ramsay L, Comadran J, Druka A, Marshall DF, Thomas WT, Macaulay M, MacKenzie K, Simpson C, Fuller J, Bonar N et al 2011. INTERMEDIUM‐C, a modifier of lateral spikelet fertility in barley, is an ortholog of the maize domestication gene TEOSINTE BRANCHED 1. Nature Genetics 43: 169–172. [DOI] [PubMed] [Google Scholar]
  47. Ringrose L, Paro R. 2007. Polycomb/Trithorax response elements and epigenetic memory of cell identity. Development 134: 223–232. [DOI] [PubMed] [Google Scholar]
  48. Saisho D, Pourkheirandish M, Kanamori H, Matsumoto T, Komatsuda T. 2009. Allelic variation of row type gene Vrs1 in barley and implication of the functional divergence. Breeding Science 59: 621–628. [Google Scholar]
  49. Sakuma S, Lundqvist U, Kakei Y, Thirulogachandar V, Suzuki T, Hori K, Wu J, Tagiri A, Twan R, Koppolu R et al 2017. Extreme suppression of lateral floret development by a single amino acid change in the VRS1 transcription factor. Plant Physiology 175: 1720–1731. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Sakuma S, Pourkheirandish M, Hensel G, Kumlehn J, Stein N, Tagiri A, Yamaji N, Ma JF, Sassa H, Koba T et al 2013. Divergence of expression pattern contributed to neofunctionalization of duplicated HD‐Zip I transcription factor in barley. New Phytologist 197: 939–948. [DOI] [PubMed] [Google Scholar]
  51. Sakuma S, Pourkheirandish M, Matsumoto T, Koba T, Komatsuda T. 2010. Duplication of a well‐conserved homeodomain‐leucine zipper transcription factor gene in barley generates a copy with more specific functions. Functional & Integrative Genomics 10: 123–133. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Schmitz J, Franzen R, Ngyuen TH, Garcia‐Maroto F, Pozzi C, Salamini F, Rohde W. 2000. Cloning, mapping and expression analysis of barley MADS‐box genes. Plant Molecular Biology 42: 899–913. [DOI] [PubMed] [Google Scholar]
  53. Schoof H, Lenhard M, Haecker A, Mayer KF, Jurgens G, Laux T. 2000. The stem cell population of Arabidopsis shoot meristems in maintained by a regulatory loop between the CLAVATA and WUSCHEL genes. Cell 100: 635–644. [DOI] [PubMed] [Google Scholar]
  54. Shang Y, Zhu J, Hua W, Wang J, Jia J, Yang J. 2014. Characterization and mapping of a Prbs gene controlling spike development in Hordeum vulgare L. Genes and Genomics 36: 275–282. [Google Scholar]
  55. Simon P. 2003. Q‐Gene: processing quantitative real‐time RT‐PCR data. Bioinformatics Applications Note 19: 1439–1440. [DOI] [PubMed] [Google Scholar]
  56. Slafer GA. 2003. Genetic basis of yield as viewed from a crop physiologist's perspective. Annals of Applied Biology 142: 117–128. [Google Scholar]
  57. Smyth DR, Bowman JL, Meyerowitz EM. 1990. Early flower development in Arabidopsis. The Plant Cell 2: 755–767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Sun Q, Zhou DX. 2008. Rice jmjC domain‐containing gene JMJ706 encodes H3K9 demethylase required for floral organ development. Proceedings of the National Academy of Sciences, USA 105: 13679–13684. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Ta KN, Adam H, Staedler YM, Schönenberger J, Harrop T, Tregear J, Do NV, Gantet P, Ghesquière A, Jouannic S. 2017. Differences in meristem size and expression of branching genes are associated with variation in panicle phenotype in wild and domesticated African rice. Evodevo 8: 2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Takeda T, Suwa Y, Suzuki M, Kitano H, Ueguchi‐Tanaka M, Ashikari M, Matsuoka M, Ueguchi C. 2003. The OsTB1 gene negatively regulates lateral branching in rice. The Plant Journal 33: 513–520. [DOI] [PubMed] [Google Scholar]
  61. Tanno K, Taketa S, Takeda K, Komatsuda T. 2002. A DNA marker closely linked to the vrs1 locus (row type gene) indicates multiple origins of six‐rowed cultivated barley (Hordeum vulgare L.). Theoretical and Applied Genetics 104: 54–60. [DOI] [PubMed] [Google Scholar]
  62. Thirulogachandar V, Alqudah AM, Koppolu R, Rutten T, Graner A, Hensel G, Kumlehn J, Bräutigam A, Sreenivasulu N, Schnurbusch T et al 2017. Leaf primordium size specifies leaf width and vein number among row‐type classes in barley. The Plant Journal 91: 601–602. [DOI] [PubMed] [Google Scholar]
  63. Toriba T, Hirano HY. 2014. The DROOPING LEAF and OsETTIN2 genes promote awn development in rice. The Plant Journal 77: 616–626. [DOI] [PubMed] [Google Scholar]
  64. Trevaskis B, Tadege M, Hemming M, Peacock WJ, Dennis E, Sheldon C. 2007. Short vegetative phase‐like MADS‐box genes inhibit floral meristem identity in Barley. Plant Physiology 143: 225–235. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Ullrich SE, ed. 2011. Barley: production, improvement, and uses. Ames, IA, USA: Wiley‐Blackwell. [Google Scholar]
  66. Whipple CJ, Kebrom TH, Weber AL, Yang F, Hall D, Meeley R, Schmidt R, Doebley J, Brutnell TP, Jackson DP. 2011. grassy tillers1 promotes apical dominance in maize and responds to shade signals in the grasses. Proceedings of the National Academy of Sciences, USA 108: E506–E512. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Wills DM, Whipple CJ, Takuno S, Kursel LE, Shannon LM, Ross‐Ibarra J, Doebley JF. 2013. From many, one: genetic control of prolificacy during maize domestication. PLoS Genetics 9: e1003604. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Yoshida M, Sasao M, Yasuno N, Takagi K, Daimon Y, Chen R, Yamazaki R, Tokunaga H, Kitaguchi Y, Sato Y et al 2013. TAWAWA1 a regulator of rice inflorescence architecture, functions through the suppression of meristem phase transition. Proceedings of the National Academy of Sciences, USA 110: 667–722. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Youssef HM, Eggert K, Koppolu R, Alqudah AM, Poursarebani N, Fazeli A, Sakuma S, Tagiri A, Rutten T, Govind G et al 2017. VRS2 regulates hormone‐mediated inflorescence patterning in barley. Nature Genetics 49: 157–161. [DOI] [PubMed] [Google Scholar]
  70. Zadoks JC, Chang TT, Konzak CF. 1974. A decimal code for the growth stages of cereals. Weed Research 14: 415–421. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Please note: Wiley Blackwell are not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

Fig. S1 Spikes of Bowman and single vrs mutants before awn removal.

Fig. S2 Spikes of vrs3.f, vrs4, vrs5, vrs1 and Bowman at late awn primordium stage.

Fig. S3 Spikes of vrs double mutants before awn removal.

Fig. S4 Differential gene expression of meristem regulators in Bowman, vrs3vrs4 and vrs3vrs5 mutants.

Fig. S5 Combined grain area from Bowman, single vrs mutants and double vrs mutants.

Fig. S6 Tiller outgrowth rate over time in Bowman and single vrs mutants and double vrs mutants.

Fig. S7 Early tillering phase across Bowman and single vrs mutants.

Method S1 Supplemental methods for genotyping and qRT‐PCR.

Click here for additional data file. (1.5MB, pdf)

Table S1 Spike and spikelet traits separated by central, lateral and additional spikelets in Bowman, single vrs mutants and double vrs mutants.

Table S2 Spikelet by rachis node for Bowman, single vrs mutants and double vrs mutants.

Table S3 VRS gene expression in Bowman and single vrs mutants.

Table S4 Comparison of spikelet traits between double vrs mutants, their parents and Bowman.

Table S5 Gene expression in Bowman, vrs3vrs4 and vrs3vrs5 double mutants and vrs3, vrs4 and vrs5 parents.

Table S6 Spikelet parameters in single vrs mutants and Bowman compared by ANOVA.

Table S7 Grain parameters in vrs single mutants and Bowman compared by ANOVA.

Table S8 Grain parameters in vrs double mutants compared with their parents and Bowman by ANOVA.

Click here for additional data file. (106.1KB, xlsx)

Articles from The New Phytologist are provided here courtesy of Wiley

RESOURCES