Developmental Traits Affecting Low‐temperature Tolerance Response in Near‐isogenic Lines for the Vernalization Locus Vrn‐A1 in Wheat (Triticum aestivum L. em Thell)

Abstract

Investigation of low‐temperature (LT) tolerance in cereals has commonly led to the region of the vrn‐A1 vernalization gene or its homologue in related genomes. Two cultivars, one a non‐hardy spring wheat and one a very cold‐hardy winter wheat, whose growth habits are determined by the Vrn‐A1 (spring habit) and vrn‐A1 (winter habit) alleles, were chosen to produce reciprocal near‐isogenic lines (NILs). These lines were then used to determine the relationship between rate of phenological development and the degree and duration of LT tolerance gene expression. Each allele was isolated in the genetic backgrounds of the non‐hardy spring wheat ‘Manitou’ and the very cold‐hardy winter wheat ‘Norstar’. The effects of each allele on phenological development and low‐temperature tolerance (LT₅₀) were determined at regular intervals over a 4 °C acclimation period of 0–98 d. The vegetative/reproductive transition, as determined by final leaf number (FLN), was found to be a major developmental factor influencing LT tolerance. Possession of a vernalization requirement increased both the length of the vegetative growth phase and LT tolerance. Similarly, increased FLN in spring Norstar and winter Manitou NILs delayed their vegetative/reproductive transition and increased their LT tolerance relative to Manitou. Although the winter Manitou NILs had a lower FLN than the spring Norstar NILs, they were able to extend their vegetative stage to a similar length by increasing the phyllochron (interval between the appearance of successive leaves). Cereal plants have four ways of increasing the length of the vegetative phase, all of which extend the time that low‐temperature tolerance genes are more highly expressed: (1) vernalization; (2) photoperiod responses; (3) increased leaf number; and (4) increased length of the phyllochron.

INTRODUCTION

Chromosomes 5A and 5D of wheat are most frequently found to have the largest influence on low-temperature (LT) tolerance (Limin et al., 1997). Numerous studies (Brule‐Babel and Fowler, 1988; Sutka and Snape, 1989; Roberts, 1990; Storlie et al., 1998) point to the location of the vernalization gene Vrn‐A1 on wheat chromosome 5A, or homologous loci, as being the region where a major LT tolerance gene is located. Characters such as sucrose accumulation (Galiba et al., 1997), unsaturated phospholipid synthesis (DeSilva, 1978) and cell size (Limin and Fowler, 2001), which have been associated with LT tolerance, have also been localized to the Vrn‐A1 locus. Other characters associated with LT tolerance or genes considered to be linked to Vrn‐A1 are a gene for frost tolerance designated Fr1 (Galiba et al., 1995) and rosette growth habit (Roberts, 1990). Regulation of the LT tolerance‐associated Wcs120 and Wcor410 gene families on the group 6 chromosomes has also been associated with chromosome 5A (Limin et al., 1997; Danyluk et al., 1998). The possibility of pleiotropic action of the Vrn‐A1 alleles affecting both growth habit and LT tolerance has been suggested (Brule‐Babel and Fowler, 1988; Roberts, 1990).

Quantitative trait analysis for traits associated with winter hardiness in barley (Hordeum vulgare L.) has localized genes affecting growth habit, fructan content, LT₅₀ and field survival to the chromosomal region homologous to Vrn-1 in wheat (Hayes et al., 1993). The largest factor affecting field survival appeared to be a single Mendelian gene that delayed flowering (Blake et al., 1993), although several lines of evidence appeared to favour linkage of these characters (Hayes et al., 1993).

Through their determination of the spring/winter growth habit in wheat, the vernalization genes play important roles as developmental regulators. In general, plants with spring habit proceed rapidly to the reproductive stage following germination, whereas plants with winter habit have a vernalization requirement. In plants requiring vernalization, ex posure to temperatures in the vernalization range (0–10 °C) shortens the vegetative phase and reduces the leaf number relative to an unvernalized plant (Wang et al., 1995). The spring habit Vrn‐A1 allele is dominant to the winter habit vrn‐A1 allele and also epistatic to other homologous vrn1 loci (Pugsley, 1971, 1972). Therefore, all vernalization alleles at the homologous Vrn1 loci in common wheat must be recessive to produce a plant type with winter habit.

Low‐temperature acclimation is a cumulative process (Fowler et al., 1999), resulting in increased LT tolerance with continued exposure to acclimation conditions. Once the vernalization requirement has been met (vernalization saturation), accumulation of LT tolerance stops even at temperatures in the acclimation range (Fowler et al., 1996a) and the plant is capable of an almost immediate transition to the reproductive phase when exposed to floral induction conditions (Mahfoozi et al., 2001a). This delay in development until the vernalization requirement is met is the primary adaptation mechanism of winter habit plants that allows them to remain in the vegetative growth phase and accumulate LT tolerance (Fowler et al., 1996b) prior to the onset of winter. Another primary mechanism regulating development and flowering time is response to photoperiod (Thomas and Vince‐Prue, 1997). It has been demonstrated that a lengthening of the vegetative phase in cereals by short days also allows for greater accumulation and longer retention of LT tolerance (Mahfoozi et al., 2000, 2001b). In all of these studies, the vegetative/reproductive transition has been shown to be the critical point, after which reduced expression of LT tolerance genes begins (Fowler et al., 1996a, b; Mahfoozi et al., 2001a).

In the present study, the effect of phenological development on the degree and duration of LT tolerance gene expression was investigated using reciprocal near‐isogenic lines (NILs) of two cultivars whose spring/winter growth habit is determined by the Vrn‐A1 (spring habit) and vrn‐A1 (winter habit) alleles. Each allele was isolated in the genetic backgrounds of the non‐hardy spring wheat ‘Manitou’ and the very cold‐hardy winter wheat ‘Norstar’. Time sequence curves were established for 0–98 d of LT acclimation to identify the effect of each allele on the degree and duration of LT tolerance gene expression, and final leaf number (FLN) estimates were used to determine the vegetative/reproductive transition.

MATERIALS AND METHODS

Genetic stocks

Reciprocal near‐isogenic lines were produced for this experiment using the non‐hardy spring habit (Vrn‐A1) cultivar ‘Manitou’ and the very cold‐hardy winter habit (vrn‐A1) cultivar ‘Norstar’ to determine the effect of spring/winter habit‐determining alleles in each genetic background. Vrn‐1 genes on homologous chromosomes have previously been shown to be recessive (Brule‐Babel and Fowler, 1988), making Vrn‐A1 allelic differences responsible for the determination of growth habit in these parents. The parent cultivars were crossed to produce a hybrid that was then backcrossed to each parental cultivar. In subsequent generations, each cultivar was crossed to the backcross F₁ (BCF₁) of the previous generation. The BCF₁ plants were selected for heterozygosity (Vrn‐1/vrn‐1) at the Vrn‐A1 locus. When Norstar was the recurrent parent, heterozygosity at the Vrn‐A1 locus was determined by the spring habit type (Vrn‐1/vrn‐1), because all other progeny would have winter habit due to the dominance of the spring habit allele. When Manitou was the recurrent parent, heterozygosity (Vrn‐1/vrn‐1) at the Vrn‐A1 locus was determined by the heterozygotes’ flowering time, which was approx. 2 weeks later than that of plants with the homozygous (Vrn‐1/Vrn‐1) spring habit. This phenotype‐based selection ensured that the donor parent allele was incorporated into the genetic background of the recurrent parent. Heterozygous plants (BC⁴F₁) of each reciprocal line were self‐pollinated and the progeny grown out after four backcrosses to each recurrent parent. Two plants with homozygous winter growth habit (1W and 2W) and two with homozygous spring growth habit (3S and 4S) were selected from the selfed progeny of each reciprocal line. The ‘recovered’ lines had gone through the crossing procedure for production of isogenic lines but, rather than selection for the donor allele, the parental allele was recovered. These lines were used to demonstrate the level of recovery of the recurrent parent in isogenic lines produced in each series.

This procedure results in reciprocal near‐isogenic lines in which, theoretically, 96·9 % of the recurrent parent DNA is recovered. These reciprocal isogenic lines, centred on the Vrn‐A1 locus, produced a winter habit type of non‐hardy Manitou with the vrn‐1 allele of Norstar in its genetic background, and a spring habit type of the hardy winter Norstar genotype with Vrn‐1 from Manitou in its genetic background. All genetic stocks are listed in Table 1.

Table 1.

Wheat genotypes studied

Spring habit	Winter habit
Manitou	Norstar
Recovered Manitou	Recovered Norstar
Spring Norstar 3S	Winter Manitou 1W
Spring Norstar 4S	Winter Manitou 2W

LT₅₀ and vernalization determination

The final leaf number (total number of leaves produced on the main stem) and LT₅₀ (temperature at which 50 % of the plants are killed by low‐temperature stress) were determined for all plant lines. The experimental design for these studies was an 8 (genotype) × 11 (acclimation period) factorial in a two replicate randomized complete block design. The genotypes that were considered are listed in Table 1. All near‐isogenic lines and parental material were evaluated after 11 LT (4 °C) acclimation periods (0, 2, 7, 14, 21, 28, 35, 42, 49, 77 and 98 d). Light was provided by a combination of cool white fluorescent lamps and incandescent bulbs at the intensities indicated below. ANOVAs were conducted to determine the significance of treatment differences. Replicates were separated over time.

Imbibed seeds were held at 5 °C for 3 d, and then transferred to an incubator and held at 22 °C for 2 d. Seedlings were grown for 13 d in hydroponic tanks containing half‐strength Hoaglands solution (Brule‐Babel and Fowler, 1988) at 20 °C with a 20 h day and photosynthetic photon flux density (PPFD) of 320 µmol m^–2 s^–1 until they reached the three to four leaf stage. They were then transferred to a 4 °C chamber with a 20 h photoperiod and a PPFD of 220 µmol m^–2 s^–1 for vernalization/LT acclimation. The procedure outlined by Limin and Fowler (1988) was used to determine the LT₅₀ of each genotype at the end of each vernalization/LT acclimation period. Briefly, plant crowns were covered with moist sand in an aluminium weighing can, placed in a programmable freezer and held at –3 °C for 12 h. After 12 h they were cooled at rate of 2 °C h^–1 down to –17 °C, and thereafter cooled at 8 °C h^–1. Five crowns were removed at 2 °C intervals for each of five test temperatures selected for each genotype in each treatment. Samples were then thawed overnight at 4 °C. Thawed crowns were transplanted into trays containing potting medium in a growth room maintained at 20 °C with a 20 h day. Plant recovery was rated (alive vs. dead) after 3 weeks and LT₅₀ was calculated for each treatment.

Germinated seeds for FLN measurements were grown at 20 °C with a 20 h day at a PPFD of 320 µmol m^–2 s^–1 for 13 d in pots (two plants per pot) before being exposed to 4 °C. Plants grown in pots at 4 °C were transferred to 20 °C chambers with 20 h photoperiods (floral induction conditions) and a PPFD of 320 µmol m^–2 s^–1 at the end of each LT acclimation period (0, 2, 7, 14, 21, 28, 35, 42, 49, 77 and 98 d) to determine the stage of phenological development. Leaves were numbered and plants were grown until the flag leaf emerged and the FLN on the main shoot could be determined (Wang et al., 1995). Vernalization saturation was considered to be complete for each genotype once the cold treatment no longer affected FLN. At this point, plant apical primordia can be transformed into potential reproductive structures that become morphologically visible as ‘double ridges’ (Kirby and Appleyard, 1987; Delécolle et al., 1989). Degree‐days from planting to appearance of the flag leaf collar (FLN) were calculated using a base temperature of 0 °C.

ANOVAs were conducted to determine the level of significance of differences due to genotypes and acclimation periods, and the genotype × acclimation period interaction, in each experiment. Where significant differences were identified, regression analyses of treatment means were used to plot curves that best described the shape and behaviour of the responses. The sigmoid four‐parameter equation was employed to describe the relationship between FLN and days of vernalization and degree‐days to FLN and days of vernalization:

The peak four parameter Weibull equation was used to describe the relationship between LT₅₀ and days of acclimation:

Non‐linear regression procedures outlined by SigmaPlot (2000) were used to provide least squares estimates of the regression coefficients in these equations.

RESULTS AND DISCUSSION

Production of near‐isogenic lines

Reciprocal near‐isogenic lines were produced in which the dominant spring habit allele Vrn‐A1 of the non‐hardy cultivar Manitou was isolated in the genetic background of the very LT tolerant cultivar Norstar. The vrn‐A1 winter habit allele of Norstar was inserted into the genetic background of the very non‐hardy spring habit cultivar Manitou to produce the reciprocal near‐isogenic line. It had previously been established (Brule‐Babel and Fowler, 1988) that spring type growth habit in Manitou was determined by the Vrn‐A1 allele. The backcrossing procedure used to produce these lines results in a theoretical recovery of 96·9 % of the recurrent parent DNA. Following the entire crossing procedure, both alleles were selected for homozygosity in the genetic background of each recurrent parent. Recovery of the original allele in each recurrent parent provided a measure of the recovery of its genetic background. In both instances, FLN, degree‐days to FLN and LT₅₀ of ‘recovered’ lines were not significantly different (P < 0·05) from those of the parental cultivar, indicating that the isogenic lines had recovered all of the measured characters of the recurrent parent, excluding the effect of the introduced gene. Similarly, differences between sibling near‐isogenic lines were non‐significant (P < 0·05). Therefore, the results were pooled and annotated as Norstar (Norstar and recovered Norstar), spring Norstar (spring Norstar 3S and 4S), Manitou (Manitou and recovered Manitou) and winter Manitou (winter Manitou 1W and 2W).

Effect of the Vrn‐1/vrn‐1 alleles on plant development

Among environmental factors, temperature is considered to be the primary determinant of the rate of plant development (Bauer et al., 1984; McMaster, 1997). For example, many more days are required to produce a leaf at 4 °C than at 20 °C. This makes accumulated thermal units, such as growing degree‐days (GDD), a better measure of plant development than calendar days. Plant development can, in large part, also be characterized by the number of leaves produced and by the number of GDD required to produce each leaf (Klepper et al., 1982; Frank and Bauer, 1995; McMaster, 1997). In the present study, ANOVA of FLN, and degree‐days required to reach FLN, of plants acclimated at 4 °C for 0–98 d and then transferred to 20 °C indicated that genotype, acclimation period and the genotype × acclimation period interaction were significant factors (P < 0·0001).

When exposed to LT, a wheat plant simultaneously undergoes changes in several physiological processes. The process of LT acclimation begins in all genotypes but plant development is slowed in the spring types, reducing their rate of leaf production. Paradoxically, although leaf production is also slowed in winter types, development is accelerated by reduction in the minimum FLN in response to the vernalization process. Replacing the Vrn‐A1 allele of Manitou with the vrn‐A1 allele from Norstar converted Manitou into a vernalization‐responsive winter habit type (winter Manitou). In genotypes with a winter habit, the vegetative phase can clearly be seen in time‐series plots of the number of leaves produced on plants that have been systematically removed from vernalization conditions (Fig. 1). Transfer of unvernalized or partially vernalized plants to growing conditions that were inductive for flowering resulted in a greater number of leaves being produced relative to those produced when the transfer took place after vernalization saturation.

Fig. 1. Final leaf number (FLN) of Manitou, Norstar, winter Manitou and spring Norstar near‐isogenic lines (NILs) vernalized at 4 °C for 0–98 d (s.e. of data points = 0·28). See Table 2 for regression coefficients.

Vernalization saturation, the point at which a plant no longer requires exposure to vernalization temperatures to enter the reproductive phase under warm inductive conditions (Hay and Kirby, 1991; Mahfoozi et al., 2001a), is considered to be complete once a cold treatment no longer reduces the FLN of a genotype (Wang et al., 1995). At this point, primordia on the shoot apex no longer develop into leaves but into reproductive structures of the spike. Consequently, the transition from the vegetative to the reproductive phase can be identified by determining the time at which minimum FLN is reached. When vernalized at 4 °C, the average leaf number for Norstar and winter Manitou was reduced from 22·5 to 13·1 and from 15·4 to 9·9, respectively (Fig. 1), at vernalization saturation. The winter Manitou reached its minimum leaf number after approx. 35 d of vernalization while Norstar required an extra week or so at 4 °C. Although both genotypes have the Norstar vrn‐A1 alleles, Norstar produced 3·2 more leaves and remained in the vegetative growth phase for an additional 120 degree‐days (Fig. 2; Table 3) compared with winter Manitou. These differences illustrate that, while determining growth habit, the vrn‐A1 locus does not act alone to establish FLN or the time to vernalization saturation.

Fig. 2. Degree‐days (planting to flag leaf collar emergence) required to reach FLN for Manitou, Norstar, winter Manitou and spring Norstar near‐isogenic lines (NILs) vernalized at 4 °C for 0–98 d (s.e. of data points = 59·3). See Table 3 for regression coefficients.

Table 3.

Estimated regression coefficients (sigmoid four‐parameter equation) for degree‐days to final leaf number (FLN) of Manitou, Norstar, winter Manitou and spring Norstar acclimated at 4 °C for 0–98 d (see Fig. 2)

	Regression coefficient
Cultivar	a	b	x 0	y 0	R 2
Winter Manitou	974	–2·06	22·7	901	0·999
Norstar	2125	–6·70	20·8	1021	0·994
Spring Norstar	0			887	1·000
Manitou	0			799	1·000

Replacing the vrn‐A1 allele of Norstar with the Vrn‐A1 allele from Manitou converted Norstar into a spring habit type (spring Norstar) without a vernalization response to LT, in contrast to the strong vernalization requirement in the Norstar parent (Fig. 1). However, spring Norstar (Vrn‐A1) produced 11·0 leaves compared with 8·9 leaves for Manitou, indicating that the Norstar genetic background had a significant (P < 0·001) influence on FLN. Although the spring habit Vrn‐A1 allele of Manitou is dominant to the vrn‐A1 allele of Norstar and epistatic to other homologous recessive winter vrn‐1 alleles on the B and D genomes (Pugsley, 1971, 1972), it alone does not determine FLN. The FLN of spring Norstar with the Vrn‐A1 allele from Manitou may be determined by its interaction with the other homologous vrn‐1 loci or possibly by other genes that differ from the Manitou background. The 2·1 additional leaves produced by spring Norstar also increased the degree‐days required to reach FLN, thereby delaying the vegetative/reproductive transition and lengthening the vegetative growth phase (Fig. 2).

The effect of Vrn‐1/vrn‐1 alleles on LT tolerance

ANOVA of LT₅₀ indicated that genotype, acclimation period and the genotype × acclimation period interaction were highly significant factors (P < 0·001). In winter habit genotypes exposed to continuous LT acclimation, the reproductive transition (Figs 1 and 2), or vernalization saturation, coincided closely with maximum LT tolerance (Fig. 3). Vernalization saturation signals a reduced level of expression of LT tolerance associated genes (Fowler et al., 1996b) and the consequent loss of LT tolerance (Fowler et al., 1996a), which is clearly illustrated in the Norstar LT₅₀ response curve. A longer time to vernalization saturation and the requirement for 3·2 more leaves in Norstar as compared with winter Manitou (Fig. 1; Table 2) lengthened the vegetative phase by 120 degree‐days (Fig. 2; Table 3). This allowed more time for rapid accumulation of LT tolerance (Fig. 3) before the acclimation rate slowed as the plants approached the reproductive transition.

Fig. 3. Low‐temperature tolerance of Manitou, Norstar, winter Manitou and spring Norstar near‐isogenic lines (NILs) acclimated at 4 °C for 0–98 d (s.e. of data points = 0·52). See Table 4 for regression coefficients.

Table 2.

Estimated regression coefficients (sigmoid four‐parameter equation) for final leaf number (FLN) of Manitou, Norstar, winter Manitou, and spring Norstar acclimated at 4 °C for 0–98 d (see Fig. 1)

	Regression coefficient
Cultivar	a	b	x 0	y 0	R 2
Winter Manitou	5·44	–2·44	23·4	9·9	0·999
Norstar	9·28	–4·74	23·3	13·1	0·993
Spring Norstar	0·00			11·0	1·000
Manitou	0·00			8·9	1·000

Delay in development in spring habit genotypes has an effect that is similar to vernalization. Final leaf numbers in plants are determined by the interactions of temperature and photoperiod with genotype (Hay and Kirby, 1991). There was no response to vernalization treatment in spring Norstar (Fig. 1) and the 20 h photoperiod would not be expected to increase FLN. Therefore, differences in FLN between Manitou and spring Norstar must be considered to be an effect of genotypic differences controlling leaf number. Spring Norstar produced 2·1 more leaves than Manitou (Fig. 1; Table 2) and the total length of the vegetative phase in spring Norstar was extended by 88 degree‐days (Fig. 2; Table 3) over that of Manitou. Because the growth rate is slow under LT acclimation (4 °C), the additional degree‐days required to reach the vegetative/reproductive transition in spring Norstar allowed it considerably more time to accumulate LT tolerance than Manitou (Fig. 3).

The above observations indicate that wheat plants simultaneously undergo several physiological‐developmental processes upon exposure to LT. The process of LT acclimation begins in all genotypes but plant development is slowed in the spring types, reducing the rate of leaf production. If the FLN has been reached when the plant is subjected to LT acclimation, very little LT tolerance will be accumulated. In wheat, by the time the first leaf has emerged under field conditions, five or six leaf primordia may have already been formed on the shoot apex (Hay and Kirby, 1991; Brooking, 1996). However, in genotypes where more leaves are produced, for example spring Norstar vs. Manitou, then the vegetative phase is lengthened and greater LT tolerance is accumulated. Paradoxically, although leaf production is also slowed in winter types, development is actually accelerated by reduction in the minimum FLN in response to the vernalization process. The length of the vernalization requirement and the minimum FLN are the primary factors that determine the length of the vegetative phase in winter habit types. As previously demonstrated, lengthening the vegetative growth phase (Fowler et al., 1996b; Mahfoozi et al., 2000) increases the duration of expression of LT tolerance genes (Fowler et al., 1996a) and delays the vegetative/reproductive transition (Fowler et al., 1996a; Mahfoozi et al., 2001a), thereby affecting LT tolerance.

Winter Manitou and spring Norstar had similar (P > 0·05) LT acclimation curves. The coincidence of these acclimation curves resulted from three different developmental mechanisms. The increased number of leaves produced by spring Norstar relative to Manitou increased the length of the vegetative growth phase (Fig. 2) and resulted in greater LT tolerance (Fig. 3). An earlier vernalization saturation point and lower minimum FLN also shortened the vegetative growth stage in winter Manitou resulting in less LT tolerance relative to Norstar. Winter Manitou had a minimum FLN that was 1·1 leaves less than spring Norstar (Fig. 1; Table 2). However, the average phyllochron was 10 degree‐days less for spring Norstar than for winter Manitou, with the result that these two genotypes reached the vegetative/reproductive transition at similar times (Fig. 2) and produced similar LT acclimation curves at 4 °C (Fig. 3). Consequently, the winter Manitou and spring Norstar near‐isogenic lines had transitions later than Manitou and earlier than Norstar, which translated into LT acclimation curves that were correspondingly intermediate to the parents.

Production of near‐isogenic lines for the alleles of the Vrn‐A1 locus has demonstrated that this gene regulates LT tolerance through its control of the developmental programme. In the near‐isogenic lines and their parents, three mechanisms affected the length of the vegetative phase: vernalization requirement, minimum FLN and the length of the phyllochron. The rate of LT acclimation is temperature dependent (Fowler et al., 1999) and when plants were acclimated at 4 °C, 32 % (4·1 °C) of the greater LT tolerance of hardy winter Norstar over that of non‐hardy spring Manitou could be associated with the vrn‐A1 allele of Norstar (Fig 3; Table 4). This allele also maintained winter Manitou in the vegetative phase for an extra 102 degree‐days (Fig 2; Table 3) when grown under the conditions of this experiment. Approx. one‐third more (Spring Norstar NILs – Manitou = 11·1 – 7·0 = 4·1; Fig 3; Table 4) of the 12·8 °C difference between Manitou and Norstar could be attributed to the direct action of genes other than vrn‐A1 in Norstar that delayed the vegetative/reproductive transition of the Spring Norstar NILs by 88 degree‐days (Fig 2; Table 3). Consequently, just over one‐third of the difference in LT tolerance between Manitou and Norstar remains unaccounted for, and is possibly due to gene interactions responsible for the extended vegetative phase of Norstar.

Table 4.

Estimated regression coefficients (peak four‐parameter Weibull equation) for LT₅₀ of Manitou, Norstar, winter Manitou and spring Norstar vernalized at 4 °C for 0–98 d (see Fig. 3)

	Regression coefficient
Cultivar	a	b	c	x 0	R 2
Manitou	–7·0	89·2	1·14	14·2	0·870
NILs	–11·1	156·4	1·15	26·6	0·916
Norstar	–19·8	91·5	1·42	38·6	0·992

The Vrn/vrn and Ppd/ppd alleles have previously been shown to regulate the expression of LT tolerance genes through their effect on the rate of phenological development (Mahfoozi et al., 2001b). Both vernalization requirement (Fowler et al., 1996b) and short‐day photoperiod sensitivity (Mahfoozi et al., 2000, 2001b) function to increase the level and duration of LT tolerance in cereals. The increase in LT tolerance was dependent upon shifting the vegetative/reproductive transition and was brought about by one critical alteration of plant development—extension of the vegetative growth phase. The close relationship between LT tolerance accumulation and the length of the vegetative phase again supports the concept that plant response to LT acclimation is a function of both degree and duration of LT tolerance gene expression (Fowler et al., 1996a, b). In this system the developmental genes are the key factors responsible for the duration of expression of LT tolerance genes.

The large effects on LT tolerance attributed to the major developmental Vrn‐A1 locus demonstrate why homologous group‐5 chromosomes of the Triticeae, which carry the Vrn‐1 loci, are most frequently found to be the major carriers of genes affecting LT tolerance (Limin et al., 1997; Fowler et al., 1999). LT tolerance has also commonly been localized to the chromosomal region of the homologous Vrn‐1 loci (Brule‐Babel and Fowler, 1988; Sutka and Snape, 1989; Hayes et al., 1993; Storlie et al., 1998). A major part of the LT tolerance associated with the Vrn‐1 chromosomal region may, in fact, be a result of this gene’s regulation of LT tolerance expression through the plant’s developmental programme.

ACKNOWLEDGEMENTS

The excellent technical assistance of Garcia Schellhorn is greatly appreciated. Financial support from the Saskatchewan Agriculture and Food Agricultural Development Fund and a National Science and Engineering Research Council of Canada strategic research grant are gratefully acknowledged.

Received: 11 October 2001; Returned for revision: 22 November 2001; Accepted: 5 February 2002.