Euphytica (2008) 164:779–787 DOI 10.1007/s10681-008-9727-0

Genetic mapping of a new Xowering time gene on chromosome 3B of wheat Katelina Pánková · Zbyn5k Milec · James Simmonds · Michelle Leverington-Waite · Lesley Fish · John W. Snape

Received: 22 January 2008 / Accepted: 19 May 2008 / Published online: 8 June 2008 © Springer Science+Business Media B.V. 2008

Abstract The single chromosome substitution lines of chromosome 3B of the Czech alternative wheat variety Beská Plesívka (CP 3B) into two spring varieties Zlatka and Sandra, revealed clear diVerences in Xowering time compared to the recipient varieties. To map this gene(s), recombinant substitution lines for chromosome 3B were produced from crosses of the substitution lines with their recipient parents and genetic maps developed using SSR markers. Two populations were mapped, Sandra//Sandra 3B/Sandra (CP 3B) and Zlatka//Zlatka/Zlatka (CP 3B). Combining the genotype data with phenotype data on Xowering time in Wve independent experiments under natural long day or controlled short day conditions revealed a single Xowering time QTL. This gene had an additive eVect of 1– 6 days, depending on environment and genetic background, and was mapped in both populations to a position in the region of marker Xbarc164 near the centromere on the long arm of 3B. Comparisons of the genetic maps with other 3B maps developed by the authors indicated that the QTL may be homologous to a QTL segregating in UK germplasm.

K. Pánková (&) · Z. Milec Crop Research Institute, Drnovská 507, Prague 6 161 06, Czech Republic e-mail: [email protected] J. Simmonds · M. Leverington-Waite · L. Fish · J. W. Snape John Innes Centre, Norwich Research Park, Colney, Norwich NR4 7UH, UK

Keywords Chromosome 3B · Earliness per se · Genetic mapping · Photoperiod response · Vernalization · Wheat Abbreviations Eps Earliness per se FT Flowering time QTL Quantitative trait locus Vrn Vernalization Ppd Photoperiod RSL Recombinant substitution line SL Substitution line

Introduction Flowering time (FT) is an important adaptive trait of all plants. In agricultural terms, optimising yield performance depends, to a great extent, on understanding the mechanisms controlling the transition from the vegetative to the reproductive stage, so that an optimal FT, tuned to the environment, can be obtained. Cereal species were the Wrst to be used for the study of the genetic control of Xowering because of their importance to agriculture. The genetic basis of winter/spring habit in wheat was Wrst described by Vavilov and Kuznetsova (1921), and since then understanding of the processes involved in Xowering has advanced considerably, most recently using molecular techniques combined with physiological analyses.

123

780

Probably all of the wheat homoeologous groups carry genes involved in the control of FT (Snape et al. 2001). Vernalization response is mostly controlled by the homoeoloci Vrn-A1, Vrn-B1 and Vrn-D1 mapped to the long arms of chromosomes 5A, 5B and 5D, respectively (Sarma et al. 1998; Leonova et al. 2003; Shindo et al. 2003). More Vrn loci have been found, or are expected from comparative studies, on homoeologous groups 1, 3, 6 and 7 (Cockram et al. 2007). The homoeoloci Ppd-A1, Ppd-B1 and Ppd-D1 are responsible for most of the photoperiod response and are located on the short arms of chromosomes 2A, 2B and 2D, respectively (Welsh et al. 1973; Law et al. 1978; Mohler et al. 2004). Earliness per se (called Eps) genes have been detected as quantitative trait loci (QTL) on almost all homoeologous groups of wheat (Snape et al. 2001). These inXuence FT independently of environmental conditions (Law and Worland 1997). Evidence for about twenty-Wve Eps loci comes from comparative genetic studies (Snape et al. 2001). Some, but not many have been mapped. For example, a major Eps locus was mapped distal to RFLP marker locus Xcdo549 on the short arm of chromosome 3A (Shah et al. 1999). Flowering time QTL were mapped on 3AL and other chromosomes by Börner et al. (2002). Genes causing delays in FT and higher photoperiod sensitivity were detected by monosomic analysis in the Czech wheat landrace Beská Plesívka (Konner 1987, 1992). Substitution of chromosome 3B from Beská Plesívka (CP 3B) into the genetic background of the spring variety Zlatka revealed the action of a gene(s) that inXuenced FT on that chromosome; the substitution line Zlatka (CP 3B) showed a delay in Xowering and better over-wintering (Konner 1987). Comparisons between several recipient varieties and their respective CP 3B substitution lines (SLs) revealed no eVect on vernalization response due to the substitutions, although there were suggestions of a genotype x vernalization interaction in the parental background. The response of 3B substitution lines in the spring wheat backgrounds, compared to the parental cultivars, showed a considerable eVect on FT under both short and long days, suggesting the eVect of an Eps gene(s) (Konner and Pánková 2002; Pánková et al. 2006). This paper describes the intrachromosomal mapping of this Eps gene on chromosome 3B through the use of recombinant substitution lines.

123

Euphytica (2008) 164:779–787

Materials and methods Mapping populations Beská Plesívka (CP) is a Czech alternative wheat variety (a landrace), which Xowers from a spring sowing, but is outstanding in its high frost tolerance. Populations of recombinant substitution lines (RSL) were developed following the techniques described by Law and Worland (1973) from crosses of substitution lines with chromosome 3B of Beská Plesívka in the backgrounds of spring wheat varieties, Zlatka and Sandra, developed at the Crop Research Institute (CRI), Prague. In the case of Sandra (CP 3B), RSLs were developed by initially backcrossing seven diVerent F1 plants to plants of the Sandra monosomic 3B line. However, the Sandra monosomic plants were not cytologically characterised before the backcrosses, so the progeny would be both monosomic and disomic. Each of the individual monosomic/disomic RSL plants was selfed to produce seed for mapping and phenotypic analysis. Because cytological screening was not carried out on the initial monosomics or on the progenies, disomic recombinants (expected at a 25% frequency) were identiWed by marker analysis as heterozygotes in the selfed generation. The Zlatka RSL population was produced directly by selWng Wve diVerent F1 plants from the Zlatka/Zlatka (CP 3B) cross to produce a pooled population of individual F2 recombinant plants. Development of genetic maps Genetic maps of chromosome 3B were developed using the RSL mapping populations of Sandra//Sandra 3B/Sandra (CP 3B) and Zlatka//Zlatka/Zlatka (CP 3B). For the Sandra lines, DNA was extracted from pooled leaf samples from Wve seedlings of each individual line using QiaGen and DNeasy kits. For the Zlatka population, initially, DNA was extracted from the individual F2 plants, and this was repeated on a sample of Wve random plants of each of their F3 families. Mapping was carried out using publicly available single sequence repeat (SSR) primer sets available at the JIC. Sources were the JIC (psp), IPK Gatersleben (gwm/gdm), Wheat Microsatellite Consortium (wmc), Beltsville Agricultural Research Station (barc) and INRA (cfd/cfa) collections, see GrainGenes website (http://www.wheat.pw.usda.gov/). Primers for markers previously located on chromosome 3B by the

Euphytica (2008) 164:779–787

authors and other groups were Wrst tested for polymorphism on the recipient parents and 3B substitution lines. Primers giving clearly scorable polymorphisms were then genotyped on the entire mapping populations. DNA fragments were ampliWed with PCR and run on 6% polyacrylamide gels for separation. The silver staining technique was used to visualize fragments (Bassam et al. 1991). Data for each population were accumulated into a genotype Wle, and genetic maps were developed using JoinMap software. Phenotypic assessments of Xowering time Five experiments were carried out to measure Xowering times on the two RSL populations. Experiment 1 In autumn 2005, 99 RSL lines of the Sandra//Sandra 3B/Sandra (CP 3B) population and the parental controls were sown in a controlled environment room (at 20°C, 14 h days/10 h nights) with four replicate plants of each line and 12 of the parents, in an individual plant randomisation experimental design. Seeds of the mapping population and parents were pre-germinated in petri-dishes for 3 days at 5°C to synchronise germination, then individually potted into 10 cm pots before being placed in the controlled environment room. Some lines had poor germination and two lines did not germinate leaving 97 lines for measurement of Xowering time. The date of Xowering was recorded when the ear on the leading tiller had half emerged. The marker analysis showed that only 37 of the original 116 backcross plants were fully homozygous. The remainder were heterozygous at one or more loci, either having been derived from backcrosses to disomic, rather than monosomic Sandra 3B plants, or were disomics from backcrosses to monosomics. Nevertheless, such plants were useful for QTL analysis, although slightly less informative than fully homozygous ones. Experiment 2 Over the winter of 2005/06, 117 F2 RSL plants of Zlatka//Zlatka/Zlatka (CP 3B) were grown for DNA extraction and assessment of Xowering times in a controlled environment room (20°C, 14 h days/10 h nights), together with 12 replicate plants each of the

781

Zlatka and Zlatka (CP 3B) parental lines. Seeds of the mapping population and parents were pre-germinated in petri-dishes for 3 days at 5°C to synchronise germination, then individually potted into 10 cm pots before being placed in the controlled environment room. The date of Xowering was recorded when the ear on the leading tiller had half emerged. Experiment 3 In Experiment 2, only 85 plants of the Zlatka//Zlatka/ Zlatka (CP 3B) F2 RSL plants produced suYcient seed for further experiments; of these, 68 F3 families were randomly chosen for further Xowering time experiments (due to space limitations) in the summer of 2006, and 63 were used for QTL analysis. These 68 F3 families and parental controls were sown in the same controlled environment room but under longer days than the previous experiment (at 20°C, 16 h days/8 h nights) with 5 replicate plants per family, and again an individual plant randomisation experimental design was used. As above, seeds of the mapping population and parents were pre-germinated in petri-dishes for 3 days at 5°C to synchronise germination, then individually potted into 10 cm pots before being placed in the controlled environment room. The date of Xowering was recorded when the ear on the leading tiller had half emerged. For Experiments 1, 2 and 3, all data for individual plants was converted to days from the start of the experiment for analysis. Experiment 4 The 37 homozygous RSL lines of the Sandra//Sandra 3B/Sandra (CP 3B) population were re-sown in a soil glasshouse experiment in the Spring/Summer of 2007. Each line was hand sown directly into soil in two replicate randomised blocks where each line was represented by a single 1m dibbed row, 11 spaced plants per row, but with 4 rows of each parental control. Plants were sown in February and grown under natural conditions until heading over the summer. Flowering time was scored as the time when ears of half the leading tillers of plants in a row had emerged. Experiment 5 Fifty of the Zlatka//Zlatka/Zlatka (CP 3B) F3 RSL families were grown in a soil glasshouse in the

123

782

Euphytica (2008) 164:779–787

Spring/Summer of 2007 in two replicate blocks of 1m rows as in Experiment 4, but with 20 plants sown/row and with 4 rows of each parental control. Plants were sown in February and grown under natural conditions until heading over the summer. Flowering time was scored as the time when ears of half the leading tillers of plants in a row had emerged.

sons with other available maps. For Zlatka//Zlatka/ Zlatka (CP 3B), a similar but slightly longer map at 159 cM covered both arms.

Statistical and QTL analyses

In Experiment 1, Xowering times for individual plants spanned 49 days, ranging from 71 to 120 days from planting, and reXected the retardant eVect of shorter days. Means varied from 72 to 115 days. Analyses of variance for the individual observations revealed signiWcant genetic variation for FT between the RSL lines (Table 1). Sandra (CP3B) and the RSL mean were signiWcantly earlier than the Sandra parent, by 6 days, contrary to previous Weld observations. In Experiment 4, mean Xowering times of the 37 homozygous Sandra//Sandra 3B/Sandra (CP 3B) RSL lines grown had a range of only 7 days, much smaller than in the controlled environment room. These data were subjected to ANOVA (not shown) and again the substitution line was earlier than the Sandra parent, but only by 1 day, as all the Xowering times were much more truncated under these natural, rather than shorter and generally cooler day conditions. QTL analysis was carried out on the line means for each of the two experiments. In the Wrst controlled environment experiment, QTL analysis was carried out on the whole data, and on the sub-divided data, using the appropriate algorithms to account for the mixed population structure. Figure 1 presents the QTL Café output for marker regression analysis of the whole data. Single marker ANOVA using the F2 algorithm on the whole population showed a skewed distribution of genotypes as expected from the way the population was derived as a mixture of F2 and backcross derivatives (data not shown). However, sub-dividing the population into the two components, F2 and backcross, on genotype gave the same results for each, with a QTL detected in the region distal to Xbarc164 (p < 0.05) with an additive eVect of around 1.5 days, largest at Xbarc164. There was also a strong dominance eVect of the CP 3B allele for earliness when the heterozygous lines were considered (not shown). With respect to the reduced homozygous RSL set grown in the glasshouse in 2007, again a single QTL associated with markers Xbarc164 and Xcfa2170 was detected in the same region (Fig. 1) using the

Flowering time results from the controlled environment and glasshouse experiments were Wrstly analysed statistically by analysis of variance (ANOVA) using the MINITAB v15 statistical software package. Means were used for QTL analysis. The program QTL Café (http://www.biosciences.bham.ac.uk/labs/ kearsey) was used for QTL discovery and location, based on single marker ANOVA, marker regression and interval mapping approaches.

Results Development of genetic maps In the initial primer screen, 14 of 38 chromosome 3B speciWc markers were polymorphic. These markers provided good coverage of most of the chromosome, but not all were suitable for QTL mapping of both populations since Sandra or Zlatka have null alleles at some loci, such that when the CP allele was present, heterozygotes and homozygotes could not be distinguished. This was the case for markers Xgwm389, Xgwm131, Xcfd283, Xgwm108, Xgwm533, and Xbarc229. Nevertheless, these markers were used for genetic mapping, and marker and map alignment. The maps (Figs. 1– 3) based on 97 RSL lines for Sandra// Sandra 3B/Sandra (CP 3B) and 117 individual F2 plants for Zlatka//Zlatka/Zlatka (CP 3B) correspond closely with the publicly available SSR consensus map (Somers et al. 2004). Seven co-dominant polymorphic markers for the Sandra population and six for the Zlatka population were subsequently used for QTL analysis. For Sandra//Sandra 3B/Sandra (CP 3B), the map length was 130 cM, covering both the short and long arms, but in two linkage groups. JoinMap was unable to link the two linkage groups with statistical signiWcance, but distances could be estimated quite accurately from these data and compari-

123

Flowering time variation and QTL analysis Sandra//Sandra 3B/Sandra (CP 3B) RSL population

Euphytica (2008) 164:779–787 250

QFt.CRI-3B.1 QFt.cri-3B.1 location Location 200 Sandra 05

Residual MS

Fig. 1 QTL proWles from marker regression output from ‘QTL Café’ analysis of the Sandra//Sandra 3B/Sandra (CP 3B) recombinant substitution lines from two experiments, (i) JIC controlled environment room, short days, Winter 2005/06 (Experiment 1), (ii) JIC glasshouse Summer 2007 (Experiment 4) (see text for details). Marker regression position of signiWcant QTL (P < 0.05) arrowed

783

Sandra 07

150

100

05

centromere 07 05

07

50

0

Xgwm566

Xgwm389

100

120

140 Xbarc77

80

Xcfa2170

60

Xbarc164

40

Xgwm285

20 Xgwm493

0

Distance cM 400

QFt.cri-3B.1 locations QFt.CRI-3B.1 Location 350 300

05

Zlatka 05 Zlatka 06 Zlatka 07

250 200

07

150

07

06 06

centromere

100 50 0 60

80

100

120

140

Distance cM

appropriate doubled haploid algorithm. The largest eVect was associated with Xbarc164, with an additive eVect of 1.5 days. Although the signiWcance levels for the QTL were not high, they were highly consistent. From all analyses, interval mapping (data not shown) and the marker regression analysis concurred in showing the presence of a single QTL located in the region 96–104 cM, between Xbarc164 and Xcfa2170, depending on the population and QTL method (Fig. 1).

160 Xwmc326

Xbarc164

40

Xcfd4 Xwmc307

20 Xgwm493

0 Xbarc147

Residual MS

Fig. 2 QTL proWles from marker regression output from ‘QTL Café’ analysis of the Zlatka//Zlatka/Zlatka (CP 3B) recombinant substitution lines from three experiments, (i) F2 JIC controlled environment room, short days, Winter 2005/06 (Experiment 2), (ii) F3 JIC controlled environment room, long days, Autumn 06 (Experiment 3, (iii) F3 JIC glasshouse Summer 2007 (Experiment 5) (see text for details). Marker regression position of signiWcant QTL (P < 0.05) arrowed

Zlatka//Zlatka/Zlatka (CP 3B) RSL variation and QTL location In Experiment 2, there was clear segregation for Xowering time between the 117 F2 RSL plants, spanning 64 days, much larger than in the Sandra population. This long period indicated the eVect of shorter days, but may also have indicated further background genetic variation in the Zlatka (CP 3B) substitution line despite the eight backcrosses carried as is recom-

123

784

Euphytica (2008) 164:779–787

Sandra (CP 3B) Xgwm389

Spark x

Zlatka

Charger x

Rialto 3B

(CP 3B)

Badger 3B

0 Xgwm389

Xgwm389

Xbarc133 Xgwm389

Xgwm533

Xgwm493

Xgwm493

Xgwm493

30

Xgwm566

72

Xgwm285

79

Xbarc164

88

Xbarc147 Xbarc229 Xwmc56 stm5.2gag Xgwm376 Xgwm285

Xgwm493

QFt.cri-3B.1

Xbarc164 Xgwm131 stm40tcac

Xwmc54

Xgwm566

QFt.cri-3B.1 Xgwm285 Xcfd4

stm12tgag

0 10

Xwmc307

59 63

Xbarc164

76

Xgwm566 Xgwm285 Xpsp3144 Xbarc164 Xwmc56 Xbarc229

QFt.cri-3B.1

QFt.cri-3B.1 Xgwm131 Xcfa2170

115 Xcfd283 Xgwm108 Xwmc326

Xbarc77 159

Xgwm547 Xbarc77

130 Xgwm299 Xpsp3001 Xgwm547

Fig. 3 Comparative genetic maps and QTL locations for the Sandra//Sandra 3B/Sandra (CP 3B) and Zlatka//Zlatka/Zlatka (CP 3B) recombinant substitution line populations, and compar-

isons with the 3B maps for the Spark/Rialto and Charger/Badger doubled haploid populations

mended for the development of a substitution line (Law and Worland 1973). The means of the parents and substitution line controls were not signiWcantly diVerent, and the mean of the RSL population was also not signiWcantly diVerent from the parents (data not shown). But there was larger than expected segregation between plants within the substitution line parent families, which may also indicate background segregation. The ANOVA for Experiment 3 (Table 2) showed signiWcant genetic variation between F3 families. FT for individual plants varied by 27 days, and means by 11 days, all earlier than when grown under shorter days. There was also a signiWcant diVerence between the parents (Table 2) with the CP 3B substitution line Xowering nearly 3 days earlier than Zlatka. In Experiment 5 containing Wfty of the Zlatka// Zlatka/Zlatka (CP 3B) F3 RSLs, again Xowering times were much more truncated in these summer

conditions with long daylengths and higher temperatures, with no signiWcant diVerences between the parents, although the CP 3B substitution line Xowered 1 day earlier on average, consistent with the previous experiment. There was also only a 7 day range of family means, and ANOVA for diVerences between families failed to reach statistical signiWcance (data not shown). For mapping and QTL analysis, all 117 F2 plants were genotyped, and QTL analysis of Xowering times of the surviving F2 plants under shorter day conditions in Experiment 2 was Wrst carried out (Fig. 2). In this experiment a QTL peak was observed, but this failed to reach statistical signiWcance. The peak was in the region Xgwm493–Xcdf4, or distal to Xbarc164. The latter suggests that the same Eps gene was segregating here as in the Sandra RSL lines. Data from the replicated controlled environment room experiment in 2006 was analysed for QTL in two

123

Euphytica (2008) 164:779–787 Table 1 Flowering time means (days from sowing to heading) and ANOVA for Sandra//Sandra 3B/Sandra (CP 3B) SRL population and their parental controls grown under continuous 20°C, 14 h day/10 h night conditions

Table 2 Flowering time means (days from sowing to heading) and ANOVA for the set of F3 RLS of Zlatka// Zlatka/Zlatka (CP 3B) and their parental controls grown under continuous 20°C, 14 h day/10 h night conditions

785

Genotype

FT mean

SE mean

Sandra parent

101.6

0.25

85.0

0.75

Sandra//Sandra 3B/Sandra (CP 3B) RSL 84.5

0.26

Source

MS

Sandra (CP 3B) sub. line

DF

One-way ANOVA for Xowering time variation between RSL Genotype 96 103.6 Error

238

Total

334

Genotype

FT mean 77.3

0.89

74.1

0.89

Zlatka//Zlatka/Zlatka (CP 3B) RSL 76.8

0.291

Source

MS

One-way ANOVA for Xowering time variation between RSL Genotype 67 36.5 220

Total

287

ways. Firstly, QTL analysis was carried out on the F3 means, using the F2 parent genotype data. Secondly, to increase the power of the analysis, 296 individual F3 plants from the 68 F2 plants were genotyped individually, and the QTL analysis was repeated on the Xowering times of the individual plants. The marker regression proWle for the individual F3 plants again revealed a QTL associated with Xbarc164 (Fig. 2). Marker regression located the QTL at map position 94 cM, distal to Xbarc164. This QTL had an additive eVect of 0.5 days, again less than the allelic diVerences in the Sandra population. However, there was a similar large dominance of earliness eVect of 1.7 days for the CP allele. The QTL analysis of the 2007 glasshouse experiment strongly conWrmed the presence of a QTL, but the RMS (Residual Mean Square) minimised in the region Xgwm49- Xcdf4, although the regression minima were quite Xat over a wide range (Fig. 2) with an additive eVect of 1.0 day and a dominance eVect in favour of the CP 3B allele of 1.3 days. The marker regression location was at 46 cM, proximal rather than distal to Xbarc164. However, within sampling limits there was no signiWcant diVerence in location between experiments, given the very Xat proWle.

2.08

<0.001

SE mean

Zlatka (CP 3B)

Error

P

49.7

Zlatka parent

DF

F

F

P

1.75

<0.001

20.8

Comparative genetic maps of the 3B chromosomes All of the experiments and analyses above indicate that the CP 3B chromosome carries an allele for earliness per se located near the centromere on chromosome 3BL. Using common polymorphic makers it was possible to align the maps for the RSL populations with each other, and also with other chromosome 3B maps available at JIC. Two populations, Spark/Rialto and Charger/Badger, had been extensively genotyped for FT in the Weld (Snape et al. unpublished). In these crosses a Xowering time QTL was distal to the marker locus Xgwm285, which was 16 cM proximal to Xbarc164. The alignments of the Sandra//Sandra 3B/Sandra (CP 3B) and Zlatka// Zlatka/Zlatka (CP 3B) maps with the Spark/Rialto and Charger/Badger 3B maps are shown in Fig. 3. From the diVerent experiments it is highly likely that the Eps gene segregating in each population is located at the same locus. While there was no absolute agreement in the locations of the QTL, being estimated as both proximal and distal to Xbarc164 in diVerent experiments, locations were not signiWcantly diVerent from each other. The consensus from the

123

786

present results suggests that it is slightly distal to Xbarc164. The QTL was designated QFt.cri-3B.1.

Discussion Studies reporting the location of FT genes on the homoeologous group 3 chromosomes of wheat, particularly 3B, are rare. A delayed Xowering due to the substitution of chromosome CP 3B into the background of a Czech spring wheat variety Zlatka was the Wrst described by Konner (1987). Results of experiments under diVerent vernalization and photoperiod conditions suggested an Eps gene(s) on this chromosome because of an interaction between the genetic background and Xowering time. However, there was also a suggestion of a photoperiod interaction (Konner 1987, 1992; Konner and Pánková 2002). Other CP 3B substitutions showed a similar eVect. The results in the present paper show the clear presence of a Xowering time gene, QFt.cri-3B.1, near the centromere on the long arm of chromosome 3B. The presence of two Eps genes, one on the long arm and the other on the short arm of chromosome 3A, and the presence of a vernalization-sensitive gene on chromosome 3B was suggested by Miura and Worland (1994). Moreover, these authors showed the eVects of FT genes on each of the homoeologous group 3 chromosomes. Shah et al. (1999) mapped an Eps gene to the end of the short arm of chromosome 3A, distal to the RFLP locus Xcdo549; the analysis also revealed that it signiWcantly inXuenced variation for plant height, 1000 kernel weight and number of kernels per plant. Other Weld-grown Xowering time experiments at JIC using recombinant doubled haploid populations identiWed a QTL located on the long arm of chromosome 3A near to the centromere (GriYths et al. pers. comm.) perhaps in a homoeologous position to the present locus. Lines carrying the substituted CP 3B chromosome have, in previous experiments at CRI, manifested enhanced photoperiod sensitivity, and it is questionable whether this was the eVect of the Eps gene mapped here. There were diVerences in the additive and dominance eVects of the locus under long and short days, but those could have been scalar eVects of the delay in Xowering time under shorter days. The original substitution lines were therefore checked for the presence of the Ppd-D1a allele

123

Euphytica (2008) 164:779–787

determining photoperiod sensitivity using perfect markers (Beales et al. 2007), and also veriWed using SSRs for the correctness of the CP 3B substitutions (Pánková et al. 2008). This analysis showed that all the substitutions were correct and carried the allele for photoperiod sensitivity in their backgrounds. Further experiments with the RSLs are needed, under deWned diVerences in photoperiod, to assess if QFt.cri-3B.1 interacts speciWcally with photoperiod or is solely an Eps gene. One strange aspect of the present series of experiments was the earlier Xowering times of the CP 3B substitutions relative to their respective donor parents. Previous experiments in the Czech Republic always showed that the SLs Xowered later than the respective donors. It is hard to deWne what diVerences in environment could have caused this change, but it may relate to high temperatures during Xowering or vernalization at the juvenile stage. At the CRI, the SLs and parental lines were vernalized before being grown for assessment of Xowering times, whereas at JIC they were not. Vernalization before growing the spring plants could delay the SL by the eVect of a combination of low temperatures and short days in the way that typical alternative varieties behave in autumn. But without vernalization, and grown only under short day conditions, SLs would be quicker to Xower due to the Eps gene eVect, as found at JIC. Alternatively, the diVerence may relate to temperature, since the previous experiments at CRI were Weld based over the summer, at much higher temperatures than under the controlled conditions in both the glasshouse and controlled environment at JIC. The presence of a temperature-sensitive gene would be very interesting to study if detected. These alternative hypotheses are under investigation by growing plants with and without vernalization, and at normal and high temperatures. In conclusion, these experiments have identiWed and mapped a new Xowering time locus from the Czech landrace Beská Plesívka which could have a quite signiWcant eVect on Xowering when introduced into more modern varieties, making them up to 6 days earlier. This could have implications for the breeding of new genotypes to respond to climate change, as well as being an interesting locus for more detailed genetic and molecular studies on its mode of action and pleiotropic eVects on agronomic traits.

Euphytica (2008) 164:779–787 Acknowledgements The authors gratefully acknowledge the fruitful discussions and contributions of Dr. Ilja T. Prasil, CRI Prague. The work was supported by the grant no. 521/05/0257 of the Grant Agency of the Czech Republic and by the research plan MZe BR 0002700602.

References Bassam BJ, Caetano-Anollés G, GresshoV PM (1991) Fast and sensitive silver staining of DNA in polyacrylamide gels. Anal Biochem 196:80–83. doi:10.1016/0003-2697(91)90120-I Beales J, Turner A, GriYths S, Snape JW, Laurie DA (2007) A pseudo-response regulator is misexpressed in the photoperiod insensitive Ppd-D1a mutant of wheat (Triticum aestivum L.). Theor Appl Genet 115:721–733. doi:10.1007/ s00122-007-0603-4 Börner A, Schumann E, Fürste A, Cöster H, Leithold B, Röder MS et al (2002) Mapping of quantitative trait loci determining agronomic important characters in hexaploid wheat (Triticum aestivum L.). Theor Appl Genet 105:921–936. doi:10.1007/s00122-002-0994-1 Cockram J, Jones H, Leigh FJ, O’Sullivan D, Powell W, Laurie DA et al (2007) Control of Xowering time in temperate cereals: genes, domestication, and sustainable productivity. J Exp Bot 58:1231–1244. doi:10.1093/jxb/erm042 Konner J (1987) A study of inheritance in the alternative growth habit of the cultivar Beská Plesívka. Sci Agric Bohemoslovaca 19:33–45 Konner J (1992) Fotoperiodická citlivost n5kterých odr4d pnenice a substitubní linie Zlatka (Beská Plesívka 3B) (Photoperiodic sensitivity of some wheat varieties and the substitution line Zlatka (Beská Plesívka 3B)). Genet Slecht 28:195–203 Konner J, Pánková K (2002) The eVect of chromosome 3B gene/ s of Beská Plesívka on vernalization response, photoperiod sensitivity and earliness of wheat. Czech J Genet Plant Breed 38:41–49 Law CN, Worland AJ (1973) Aneuploidy in wheat and its uses in genetic analysis. Plant Breed Inst Camb Annu Rep 1972:25–65 Law CN, Worland AJ (1997) Genetic analysis of some Xowering time and adaptive traits in wheat. New Phytol 137:19– 28. doi:10.1046/j.1469-8137.1997.00814.x Law CN, Sutka J, Worland AJ (1978) A genetic study of daylength response in wheat. Heredity 41:185–191. doi:10.1038/hdy.1978.87 Leonova E, Pestsova E, Salina E, Efremova T, Röder M, Börner A (2003) Mapping of the Vrn-B1 gene in Triticum

787 aestivum using microsatellite markers. Plant Breed 122:209–212. doi:10.1046/j.1439-0523.2003.00818.x Miura H, Worland AJ (1994) Genetic control of vernalization, day-length response, and earliness per se by homoeologous Group 3 chromosomes in wheat. Plant Breed 113:160–169. doi:10.1111/j.1439-0523.1994.tb00718.x Mohler V, Lukman R, Ortiz-Islas S, William M, Worland AJ, van Beem J et al (2004) Genetic and physical mapping of photoperiod insensitive gene Ppd-B1 in common wheat. Euphytica 138:33–40. doi:10.1023/B:EUPH.0000047056.58938.76 Pánková K, Milec Z, Pránil IT, Pránilová P, Snape JW (2006) The eVect of photoperiod and vernalization on Xowering time in four wheat lines with the substituted 3B chromosome of Beská Plesívka. In: Börner A, Pankova K, Snape JW (eds) Proc EWAC workshop, 27th June–1st July 2005. Prague, Czech Republic, pp 108–110 Pánková K, Milec Z, Leverington-Waite M, Chebotar S, Snape JW (2008) Characterization of inter-varietal chromosome substitution lines of wheat using molecular markers. Czech J Genet Breed 44:22–29 Sarma RN, Gill BS, Sasaki T, Galiba G, Sutka J, Laurie DA et al (1998) Comparative mapping of the wheat chromosome 5A Vrn-A1 region with rice and its relationship to QTL for Xowering time. Theor Appl Genet 97:103–109. doi:10.1007/ s001220050872 Shah MM, Gill KS, Baenziger PS, Yen Y, Kaeppler SM, Ariyarathne HM (1999) Molecular mapping of loci for agronomic traits on chromosome 3A of bread wheat. Crop Sci 39:1728–1732 Shindo C, Tsujimoto H, Sasakuma T (2003) Segregation analysis of heading traits in hexaploid wheat utilizing recombinant inbred lines. Heredity 90:56–63. doi:10.1038/ sj.hdy.6800178 Snape JW, Butterworth K, Whitechurch E, Worland AJ (2001) Waiting for Wne times: genetics of Xowering time in wheat. Euphytica 119:185–190. doi:10.1023/A:1017594422176 Somers DJ, Isaac P, Edwards K (2004) A high–density wheat microsatellite consensus map for bread wheat (Triticum aestivum L.). Theor Appl Genet 109:1105–1114. doi:10.1007/ s00122-004-1740-7 Vavilov NI, Kuznetsova ES (1921) On the genetic nature of winter and spring varieties of plants. In: Proc Agric Faculty Saratov University Saratov, pp 1–25 Welsh JR, Keim DL, Pirasteh B, Richards RD (1973) Genetic control of photoperiod response in wheat. In: Sears ER, Sears LMS (eds) Proc 4th int wheat genet symp. University of Missouri Research Station, Columbia, MO, USA, pp 879–884

123