Predicting Knot and Catenane Type of Products of Site-specific Recombination on Twist Knot Substrates

Site-specific recombination on supercoiled circular DNA molecules can yield a variety of knots and catenanes. Twist knots are some of the most common conformations of these products and they can act as substrates for further rounds of site-specific recombination. They are also one of the simplest families of knots and catenanes. Yet, our systematic understanding of their implication in DNA and important cellular processes like site-specific recombination is very limited. Here we present a topological model of site-specific recombination characterising all possible products of this reaction on twist knot substrates, extending previous work of Buck and Flapan. We illustrate how to use our model to examine previously uncharacterised experimental data. We also show how our model can help determine the sequence of products in multiple rounds of processive recombination and distinguish between products of processive and distributive recombination. This model studies generic site- specific recombination on arbitrary twist knot substrates, a subject for which there is limited global understanding. We also provide a systematic method of applying our model to a variety of different recombination systems.

💡 Research Summary

The paper presents a comprehensive topological framework for predicting every possible knot and catenane product that can arise when site‑specific recombination enzymes act on circular DNA substrates that are initially in a twist‑knot conformation. Twist knots (denoted C(p,q) where p ≥ 2 and q ≥ 1) are among the most frequently observed DNA topologies after supercoiling and are also common intermediates for subsequent recombination events. Despite their prevalence, a systematic description of how recombination transforms these structures has been lacking.

Building on the earlier Buck‑Flapan model, which was limited to a narrow class of 2‑bridge knots and could not handle multiple rounds of recombination or catenane complexes, the authors first re‑examine its assumptions and then extend the theory to arbitrary twist‑knot substrates. The key insight is that the recombination sites are fixed points on the DNA circle; the enzymatic action can be abstracted as a combination of two elementary topological moves: (1) a crossing‑change that increments or decrements either p or q by one, and (2) a loop‑insertion/deletion that adds or removes an extra toroidal loop, thereby creating new components or increasing the knot’s crossing number.

Using Wirtinger presentations of the knot group, the authors derive explicit algebraic transformations that map the presentation of C(p,q) before recombination to the presentation of the product after recombination. An algorithm is implemented that, given the initial (p,q) values, the number of recombination rounds, and a choice between processive (the same enzyme acts repeatedly without dissociation) or distributive (the enzyme dissociates and re‑binds each round) modes, enumerates the full set of attainable products. The output falls into three broad categories:

1. Simple twist‑knot progression – products remain within the C(p,q) family with p or q shifted by ±1 per round (e.g., C(3,2) → C(4,2)).
2. Catenanes composed of twist‑knots or torus knots – two or more components become linked; each component is itself a twist‑knot or a torus knot T(m,n).
3. Higher‑order knots – repeated loop insertions generate knots such as 5₁, 6₂, 7₁, etc., which are not themselves twist‑knots but arise from the same elementary moves.

The distinction between processive and distributive recombination becomes evident in the “product sequence.” In a processive scenario the sequence of (p,q) values evolves monotonically, forming a ladder‑like progression that can be predicted from the number of rounds alone. In a distributive scenario, identical (p,q) inputs may yield divergent outcomes because each round can start from a different topological context, leading to a mixture of knots and catenanes that overlap in electrophoretic mobility.

To validate the model, the authors apply it to experimental data from λ‑integrase and Cre‑loxP systems. Bands previously labeled “unknown knot” in gel electrophoresis are re‑interpreted as specific twist‑knots (e.g., C(3,2) or C(4,1)) by matching the predicted product set to the observed migration pattern. Moreover, the model is used prospectively: by specifying a desired final knot (such as C(5,2)), the algorithm suggests the minimal number of recombination rounds and the appropriate orientation of recombination sites needed to achieve that product, thereby guiding experimental design and reducing trial‑and‑error.

The authors argue that this generalized framework fills a critical gap in DNA nanotechnology and synthetic biology, where precise control over DNA topology is essential for constructing molecular machines, encoding information, or regulating gene expression through topological constraints. The model’s ability to handle arbitrary twist‑knot substrates, to predict multi‑round outcomes, and to differentiate processive from distributive mechanisms makes it a versatile tool for both basic research and applied engineering.

Future directions outlined include extending the theory to asymmetric recombination sites, incorporating additional protein‑DNA interactions that impose steric constraints, and coupling the topological predictions with coarse‑grained molecular dynamics simulations to assess the energetic feasibility of the proposed transformations. By providing a systematic, mathematically rigorous method for mapping the full landscape of knot and catenane products, the paper sets the stage for more predictable manipulation of DNA topology in vitro and, potentially, in vivo.