Meiotic recombination between homologous chromosomes initiates via programmed DNA double-strand breaks (DSBs), generated by complexes comprising Spo11 transesterase plus accessory proteins. DSBs arise concomitantly with the development of axial chromosome structures, where the coalescence of axis sites produces linear arrays of chromatin loops. Recombining DNA sequences map to loops, but are ultimately tethered to the underlying axis. How and when such tethering occurs is currently unclear. Using ChIPchip in yeast, we show that Spo11-accessory proteins Rec114, Mer2, and Mei4 stably interact with chromosome axis sequences, upon phosphorylation of Mer2 by S phase Cdk. This axis tethering requires meiotic axis components (Red1/Hop1) and is modulated in a domain-specific fashion by cohesin. Loss of Rec114, Mer2, and Mei4 binding correlates with loss of DSBs. Our results strongly suggest that hotspot sequences become tethered to axis sites by the DSB machinery prior to DSB formation.

Meiotic chromosomes are highly compacted yet remain transcriptionally active. To understand how chromosome folding accommodates transcription, we investigated the assembly of the axial element, the proteinaceous structure that compacts meiotic chromosomes and promotes recombination and fertility. We found that the axial element proteins of budding yeast are flexibly anchored to chromatin by the ring-like cohesin complex. The ubiquitous presence of cohesin at sites of convergent transcription provides well-dispersed points for axis attachment and thus chromosome compaction. Axis protein enrichment at these sites directly correlates with the propensity for recombination initiation nearby. A separate modulating mechanism that requires the conserved axial-element component Hop1 biases axis protein binding towards small chromosomes. Importantly, axis anchoring by cohesin is adjustable and readily displaced in the direction of transcription by the transcriptional machinery. We propose that such robust but flexible tethering allows the axial element to promote recombination while easily adapting to changes in chromosome activity.DOI: http://dx.doi.org/10.7554/eLife.07424.001

Repairing broken chromosomes via joint molecule (JM) intermediates is hazardous and therefore strictly controlled in most organisms. Also in budding yeast meiosis, where production of enough crossovers via JMs is imperative, only a subset of DNA breaks are repaired via JMs, closely regulated by the ZMM pathway. The other breaks are repaired to non-crossovers, avoiding JM formation, through pathways that require the BLM/Sgs1 helicase. “Rogue” JMs that escape the ZMM pathway and BLM/Sgs1 are eliminated before metaphase by resolvases like Mus81-Mms4 to prevent chromosome nondisjunction. Here, we report the requirement of Smc5/6-Mms21 for antagonizing rogue JMs via two mechanisms; destabilizing early intermediates and resolving JMs. Elimination of the Mms21 SUMO E3-ligase domain leads to transient JM accumulation, depending on Mus81-Mms4 for resolution. Absence of Smc6 leads to persistent rogue JMs accumulation, preventing chromatin separation. We propose that the Smc5/6-Mms21 complex antagonizes toxic JMs by coordinating helicases and resolvases at D-Loops and HJs, respectively.