Elevated temperature increases meiotic crossover frequency via the interfering (Type I) pathway in Arabidopsis thaliana

Abstract

For most eukaryotes, sexual reproduction is a fundamental process that requires meiosis. In turn, meiosis typically depends on a reciprocal exchange of DNA between each pair of homologous chromosomes, known as a crossover (CO), to ensure proper chromosome segregation. The frequency and distribution of COs are regulated by intrinsic and extrinsic environmental factors, but much more is known about the molecular mechanisms governing the former compared to the latter. Here we show that elevated temperature induces meiotic hyper-recombination in Arabidopsis thaliana and we use genetic analysis with mutants in different recombination pathways to demonstrate that the extra COs are derived from the major Type I interference sensitive pathway. We also show that heat-induced COs are not the result of an increase in DNA double-strand breaks and that the hyper-recombinant phenotype is likely specific to thermal stress rather than a more generalized stress response. Taken together, these findings provide initial mechanistic insight into how environmental cues modulate plant meiotic recombination and may also offer practical applications.

Fig 1. Temperature dependent modulation of meiotic recombination frequency occurs through the Type I CO pathway.

(A, B) Genetic distances measured using FTLs in WT, mus81, and msh4 plants at 20˚C and 28˚C in the I3a (A) and I1a (B) interval. Significantly different values between 20°C and 28°C at α = 0.05 are marked with an asterisk (*). (C, D) MLHI foci counting. (C) Pollen mother cells containing chromosomes (DAPI, blue) and MLHI foci (red). (D) Boxplot of MLH1 foci counts, mean shown as white circle; **** indicates p ≤ 0.0001. Scale bars represent 5 μm.

Fig 2. Temperature dependent modulation of meiotic CO frequency can work in conjunction with anti-CO factors.

Genetic distances measured using FTLs in WT and fancm plants at 20°C and 28°C in the I3a interval. Significantly different values between 20°C and 28°C at α = 0.05 are marked with an asterisk (*).

Fig 3. NaCl treatment does not induce changes in meiotic CO frequency.

(A) Genetic distances measured in the I1a interval using FTL lines, showing SE; neither the 100 mM NaCl or 200 mM NaCl values are different from the control treatment. (B) ΔC_T values of AKR4C9 and BHLH122 for 0, 100, and 200 mM NaCl treatments, using TUB4 as an endogenous control. Adjusted p-values are from Tukey’s honest significant difference test, and are indicated as follows: ns = p > 0.5, **** indicates p ≤ 0.0001.

Fig 4. Temperature-dependent COs are not derived from an increase in DSBs.

(A, B) γH2Ax foci counting. (A) Pollen mother cells containing chromosomes (DAPI, blue) and γH2Ax foci (red). (B) Boxplots of γH2Ax foci at 20˚C and 28˚C; mean shown as white circle. (C, D) RAD51 foci counting. (C) Pollen mother cells containing chromosomes (DAPI, blue) and RAD51 foci (red). (D) Boxplots of RAD51 foci at 20°C and 28°C; mean shown as white circle. p-values are indicated as follows: ns = p > 0.5, * = p ≤ 0.05, ** = p ≤ 0.01. Scale bars represent 5 μm.