Response to divergent selection on meiotic recombination in Saccharomyces cerevisiae

This study demonstrates that the meiotic recombination rate in *Saccharomyces cerevisiae* is highly evolvable under divergent selection, achieving significant local changes within ten generations through the selection of specific haplotypes and the reduction of heterozygosity, though genome-wide effects varied across independent lineages.

The Gist

Imagine you have a giant library of books (the yeast genome). Every time the yeast reproduces sexually, it doesn't just copy a book page-for-page; it takes two different books, rips out chapters, and swaps them around to create a brand-new, unique story. This swapping process is called recombination.

Scientists have long wondered: Can we train yeast to swap chapters more often, or less often? And if we do, what happens to the rest of the library?

This paper describes a fascinating experiment where researchers acted like "genetic trainers" for yeast, trying to teach them to be either "super-swappers" or "lazy-swap-ers."

The Setup: The Fluorescent Library

The researchers started with a very diverse group of yeast (a mix of four different wild strains). To see how much swapping was happening, they inserted two special "fluorescent tags" (like glowing stickers) onto a specific chromosome.

• Tag A (Blue): Glows blue.
• Tag B (Yellow): Glows yellow.

If a yeast cell has both tags, it glows Green. If it has neither, it's Dark. If it has only one, it glows Blue or Yellow.

The Training Regimen: Sorting the Swappers

The scientists set up a high-tech sorting machine (a flow cytometer) that acts like a bouncer at a club. They wanted to see if they could force the yeast to evolve. They ran three different "training camps" for 10 generations:

1. The "Swap-It-Up" Camp (Sel+): The bouncer only let in the Blue and Yellow spores. These are the "swappers" because they must have exchanged genetic material to lose one tag and keep the other. The "Green" (no swap) and "Dark" (no swap) spores were kicked out.
2. The "Keep-It-Original" Camp (Sel-): The bouncer only let in the Green and Dark spores. These are the "non-swappers." The "Blue" and "Yellow" (swappers) were kicked out.
3. The "Chill" Camp (Sel=): The bouncer let everyone in, regardless of color. This was the control group to see what happens naturally.

The Results: The Yeast Listened!

After just 10 generations (which is a blink of an eye in evolutionary time), the results were dramatic:

• The "Swap-It-Up" Camp: The yeast evolved to swap genes 28% more often than the original group.
• The "Keep-It-Original" Camp: The yeast evolved to swap genes 24% less often.

The Analogy: Imagine a classroom where the teacher only lets students who change seats every day stay in class. After a few weeks, the whole class becomes experts at changing seats. Conversely, if the teacher only lets students who never change seats stay, the class becomes very rigid. The yeast did exactly this.

The Twist: The Ripple Effect

Here is where it gets interesting. The scientists only asked the yeast to swap genes in one specific neighborhood (Chromosome VI).

• In that neighborhood: The training worked perfectly.
• In the next-door neighborhood: The yeast did the opposite! If they were trained to swap more in the first area, they swapped less in the next area.
• In the rest of the city (other chromosomes): Nothing happened.

The Metaphor: Think of the chromosome as a long highway. The scientists put a "Construction Zone" (high swapping) in one lane. Because of a rule called "Interference" (which is like a traffic law saying "don't put two construction zones too close together"), the lane right next to it became a "No-Construction Zone." But the highways in other cities (other chromosomes) were unaffected.

The Deep Dive: How Did They Do It?

The researchers took the "Super-Swapper" yeast and sequenced their entire DNA to see how they changed. They found two types of changes:

1. Local Fixes (Cis): In the specific area where they were trained, the yeast swapped their messy, mixed-up DNA for a clean, uniform version that matched the "tester" strain. It's like they realized, "Hey, if we all wear the same uniform, it's easier to swap chapters!" They removed the genetic "noise" that was blocking the swaps.
2. Global Changes (Trans): Surprisingly, in two of the four groups, the yeast didn't just swap more in the training zone; they swapped more everywhere in the genome. It's as if they installed a new "engine" that made the whole car go faster, not just one wheel.

The Cost of Speed

Usually, when you force an organism to change a fundamental trait, it gets weaker (like a racehorse that runs fast but gets tired easily). The scientists checked if the "Super-Swappers" were weaker.

• Growth: They grew just as fast as the others.
• Survival: Surprisingly, the trained yeast actually had higher survival rates for their spores than the untrained group. It seems that by optimizing their swapping, they accidentally fixed some other genetic glitches.

The Big Picture

This paper proves that recombination is not a fixed setting; it's a dial that can be turned up or down very quickly by natural selection.

• Why it matters: In nature, this ability allows species to adapt quickly to new environments. If a virus attacks, a species that can shuffle its genetic deck faster might create a "winner" combination that survives.
• The Takeaway: Evolution is flexible. By simply sorting for a specific trait (swapping genes), the yeast didn't just change that one trait; they rewired their entire genetic strategy, sometimes locally and sometimes globally, proving that the "rules" of how we shuffle our DNA are written in pencil, not stone.

Technical Summary

1. Problem Statement

Meiotic recombination is a fundamental evolutionary driver that reshapes genetic diversity. While it is known that recombination rates (RR) vary naturally within and between species due to cis-acting (local DNA sequences) and trans-acting (genome-wide factors) elements, the evolutionary mechanisms governing this variation remain poorly understood.

• The Gap: Previous attempts to study the evolvability of recombination rates in yeast were limited. While artificial selection experiments have been conducted in insects (e.g., Drosophila) and plants, there is a lack of long-term experimental evolution studies in fungi using direct, high-throughput selection on recombination phenotypes.
• The Challenge: Measuring recombination phenotypes is typically destructive (requiring tetrad dissection or cytology) and low-throughput, making it difficult to apply the large population sizes required for experimental evolution.

2. Methodology

The authors employed an experimental evolution approach using a genetically diverse yeast population and fluorescence-activated cell sorting (FACS) to select for specific recombination phenotypes over ten generations.

• Biological Material:
  • Base Population: The SGRP-4X population, a highly admixed strain derived from four divergent S. cerevisiae founders (North American, West African, Sake, and Wine/European clades), ensuring high standing genetic variation.
  • Reporter System: A bi-fluorescent tester strain (SK1-VI_C1Y2) containing two fluorescent markers on Chromosome VI: yECerulean (C) and Venus (Y).
• Experimental Design:
  • Crossing: The SGRP-4X population was crossed with the bi-fluorescent tester to create a diploid population (Generation 0, G0) heterozygous for the markers.
  • Selection Regimes (10 Generations): Four independent lineages were evolved under three distinct modes:
    1. Sel+ (Positive Selection): FACS sorting selected only recombinant spores (those with fluorescence patterns indicating a crossover between markers C and Y).
    2. Sel- (Negative Selection): FACS sorting selected only non-recombinant spores.
    3. Sel= (Control): All spores were kept regardless of recombination status.
  • Replicates: Four biological replicates (A, B, C, D) were maintained for each mode.
• Phenotyping:
  • Recombination rates were measured at each generation by crossing sorted spores with tri-fluorescent tester strains and analyzing segregation patterns via flow cytometry.
  • Genomic Analysis: At Generation 8 (G8), individual diploids were isolated and phenotyped. The highest and lowest recombining individuals from the Sel+ lineages were selected for whole-genome sequencing (WGS) of their tetrads (segregants) to map crossover (CO) and gene conversion (GC) events genome-wide.
  • Fitness Assays: Growth kinetics, sporulation efficiency, and spore viability were measured to detect fitness costs associated with the selection.

3. Key Contributions

• First High-Throughput Yeast Selection: This study establishes a robust protocol for experimental evolution of recombination rates in yeast using FACS, overcoming the throughput limitations of traditional tetrad dissection.
• Dissection of Cis vs. Trans Effects: By combining population-level selection with individual whole-genome sequencing, the study successfully distinguishes between local (cis) structural/sequence changes and genome-wide (trans) regulatory changes.
• Quantification of Evolvability: It provides empirical evidence that recombination rates can evolve rapidly (within 10 generations) in response to direct selection, driven by standing genetic variation rather than new mutations.

4. Key Results

A. Response to Divergent Selection

• Local Response: After 10 generations, the Sel+ populations showed a +28% increase in recombination rate within the selected interval (VI_C1Y2), while Sel- populations showed a -24% decrease. The Sel= control remained stable.
• Adjacent Response: In the immediately adjacent interval (VI_Y2R3), an inverted response was observed (Sel+ decreased, Sel- increased), likely due to crossover interference mechanisms shifting CO distribution.
• Genome-Wide Response:
  • In two of the four Sel+ lineages (A and B), there was a significant genome-wide increase in crossover rates, indicating the selection of trans-acting modifiers.
  • In the other two lineages (C and D), the increase was restricted to the selected interval, suggesting cis-only adaptation.
• Rapid Reversibility: When selection was relaxed (Sel=) and then re-applied (Sel+), recombination rates in some lineages dropped and recovered within a single generation, suggesting the presence of strong, rapidly segregating genetic factors (e.g., structural variants or specific haplotypes).

B. Genetic Mechanisms (WGS Analysis)

• Cis-Selection: In high-recombining [high] segregants, the selection interval on Chromosome VI showed strong enrichment for the WA (West African) and SK1 haplotypes. These haplotypes possess high structural similarity and low sequence heterozygosity compared to other founders (like SA or WE), which are known to suppress recombination due to mismatch repair systems. Selection favored the removal of heterozygosity in this region.
• Trans-Selection: The [high] lineages (specifically A and B) exhibited a global increase in COs across multiple chromosomes (III, IX, IV) without a corresponding increase in gene conversions (GCs). This suggests selection acted on pathways regulating the CO:NCO choice (deciding whether a double-strand break becomes a crossover or non-crossover).
• Outlier Analysis: An individual with an extremely high recombination rate (near 0.5) was sequenced. WGS confirmed the markers were in their original positions with no large insertions or duplications, confirming a true biological increase in recombination frequency.

C. Fitness Effects

• No Growth Cost: There were no significant differences in vegetative growth rates or sporulation efficiency between the evolved populations (Sel+, Sel-, Sel=) and the G0 ancestor.
• Increased Viability: Both Sel+ and Sel- populations showed significantly higher spore viability than the Sel= control and G0.
  • Hypothesis: The Sel+ increase may be due to a reduction in achiasmatic bivalents (chromosomes failing to pair), reducing aneuploidy. The Sel- increase is less clear but may be linked to the specific fitness effects of the fluorescent markers themselves (where the Venus marker confers a viability advantage).

5. Significance

• Evolutionary Dynamics: The study demonstrates that recombination rates are highly evolvable traits controlled by both local sequence context and global regulatory networks. The rapid response suggests that standing genetic variation contains "modifier" alleles with large effects that can be quickly fixed by selection.
• Mechanistic Insight: The finding that selection can act on trans-factors to increase COs genome-wide (without increasing GCs) provides direct evidence for the existence of selectable variation in the CO:NCO decision pathway.
• Methodological Advancement: The successful use of FACS for recurrent selection in yeast opens new avenues for studying the genetics of complex quantitative traits that are difficult to phenotype in bulk.
• Implications for Adaptation: By showing that recombination rates can be manipulated to increase or decrease, this work provides tools to study how recombination rates influence the speed of adaptation in changing environments, a central question in evolutionary biology.

In conclusion, this paper provides a comprehensive view of how meiotic recombination evolves, revealing that it is a plastic trait shaped by a combination of local structural constraints and global regulatory mechanisms, all of which can be rapidly altered by artificial selection without incurring immediate fitness costs.