New Arabidopsis recombinant inbred lines (Landsberg erecta x Nossen) reveal natural variation in phytochrome-mediated responses

Abstract

We used 52 Arabidopsis (Arabidopsis thaliana) accessions and developed a new set of 137 recombinant inbred lines between Landsberg erecta (Ler) and Nossen (No-0) to explore the genetic basis of phytochrome-mediated responses during deetiolation. Unexpectedly, most accessions showed weak or moderate hypocotyl growth and cotyledon unfolding responses to pulses of far-red light (FR). Crosses between Columbia and No-0, two accessions with poor response, segregated seedlings with unfolded cotyledons under pulsed FR, suggesting the occurrence of accession-specific loci in the repression of morphological responses to weak light signals. Confirming the latter expectation, mapping of responses to pulsed FR in the Ler x No-0 lines identified novel loci. Despite its weak response to pulsed FR, No-0 showed a response to continuous FR stronger than that observed in Ler. By mapping the differential effect of pulsed versus continuous FR, we identified two high-irradiance response loci that account for the steeper response to continuous FR in No-0. This underscores the potential of the methodology to identify loci involved in the regulation of the shape of signal input-output relationships. Loci specific for a given phytochrome-mediated response were more frequent than pleiotropic loci. Segregation of these specific loci is predicted to yield different combinations of seedling responsivity to light. Such flexibility in combination of responses is observed among accessions and could aid in the adjustment to different microenvironments.

Figure 1.

Variability of Arabidopsis accessions in the VLFR of cotyledon unfolding (A) and hypocotyl growth inhibition (B). The VLFR is calculated as the difference between dark controls and seedlings exposed to hourly FR pulses (cotyledon angle in darkness = 0 for all the accessions). The inset shows the lack of genetic correlation between VLFR and HIR of cotyledon angle of different accessions. Heritability was 68% for the VLFR of cotyledon unfolding, 94% for the VLFR of hypocotyl growth, and 70% for the HIR of cotyledon unfolding.

Figure 2.

Reduced VLFR of hypocotyl growth and cotyledon unfolding in No-0 compared to Ler. A, Representative seedlings of each accession grown in darkness (D) or under hourly pulses of FR. B, Hypocotyl length response to the proportion of Pfr (Pfr/P) established by hourly pulses of R/FR mixtures. C, Cotyledon angle response to Pfr/P. Data are means and se of at least 17 replicate boxes (i.e. 170 seedlings).

Figure 3.

Segregation of cotyledon unfolding under hourly pulses of FR in the F₂ generation of a cross between No-0 and Col. Number of seedlings, F₂ = 216; Col = 245; No-0 = 86.

Figure 4.

Enhanced HIR of hypocotyl growth in No-0 compared to Ler. A, Hypocotyl length response to the fluence rate of continuous FR. The inset shows the average slope across the tested range of fluence rates and the results of Students' t test to compare the slopes. B, Cotyledon angle response to the fluence rate of continuous FR. C, Fluence response curves of hypocotyl length in Ep-0 and Ak-1 compared to Ler. Data are means and se of at least 12 (A and B) or three (C) replicate boxes.

Figure 5.

The Ler × No-0 linkage map. Gray areas indicate regions with higher (P < 0.05) proportion of Ler alleles and black areas indicate regions with higher proportion of No-0. The locations of QTLs affecting leaf lamina and petiole length (LNG), leaf lamina width (WID), hypocotyl length (HYP2), and cotyledon unfolding (UNF1), their 2-LOD support intervals, percent of accounted variability (Var), and additive effect (Add, Ler minus No-0) are indicated.

Figure 6.

VLFR, LFR, and HIR in Ler × No-0 RILs. A, Distribution of mean VLFR, LFR, and HIR of hypocotyl growth inhibition. B, Cotyledon unfolding. Heritability ranged between 33% and 38%. The means of the parental lines are indicated.

Figure 7.

Likelihood plot of the QTLs affecting VLFR, LFR, and HIR of hypocotyl growth and cotyledon unfolding. The horizontal line represents the LOD threshold. The 2-LOD support interval, the percentage of accounted variation (Var), and the additive effect (Add) are indicated for each QTL.

Figure 8.

Identification of HIR2 does not require incorporating hourly FR data in the calculations. Likelihood plot of the QTL affecting the difference in hypocotyl length between seedlings grown in darkness or under 10 μmol m⁻² s⁻¹. The distribution of the data was normal (S = 1.45). The horizontal line represents the LOD threshold. The 2-LOD support interval, the percentage of accounted variation (Var), and the additive effect (Add) are indicated.