Dodging the crisis of folding proteins with knots

Abstract

Proteins with nontrivial topology, containing knots and slipknots, have the ability to fold to their native states without any additional external forces invoked. A mechanism is suggested for folding of these proteins, such as YibK and YbeA, that involves an intermediate configuration with a slipknot. It elucidates the role of topological barriers and backtracking during the folding event. It also illustrates that native contacts are sufficient to guarantee folding in Ϸ1-2% of the simulations, and how slipknot intermediates are needed to reduce the topological bottlenecks. As expected, simulations of proteins with similar structure but with knot removed fold much more efficiently, clearly demonstrating the origin of these topological barriers. Although these studies are based on a simple coarse-grained model, they are already able to extract some of the underlying principles governing folding in such complex topologies. molecular dynamics ͉ slipknots ͉ backtracking ͉ topological barriers D uring the past 2 decades, a joint theoretical and experimental effort has largely advanced the quantitative understanding of the protein folding mechanism. Most small-and intermediate-size proteins live on a minimally frustrated funnel-like energy landscape, which allows fast and robust folding (1-3). Because proteins have been able to solve the energy problem, the final challenge is the structural complexity of the protein folding motifs. Most proteins avoid complex topologies, but recent discoveries have shown that some proteins are actually able to fold into nontrivial topologies where the main chain folds into a knotted conformation (4-6). Although these ''knotted'' folding motifs have been observed, we still have to face the challenging question of how the protein overcomes the kinetic barrier associated with the search of the knotted conformation. We suggest a possible mechanism where the knot formation is preceded by a conformation called a ''slipknot.'' A slipknot is topologically similar to a knot, except that an internal knot is effectively undone as the pathway of the backbone folds back on itself. The fact that such slipknots have already been observed in some protein final structures (7) adds support to this suggestion.This folding mechanism is explored in the context of the two most experimentally investigated knotted families of proteins, Haemophilus influenzae YibK and Escherichia coli YbeA, which are homodimeric ␣/␤-knot methyltransferases (MTases). A schematic representation of these proteins is shown in Fig. 1, [see also supporting information (SI) Fig. S1]. It has been shown experimentally that both these proteins unfold spontaneously and reversibly on addition of chemical denaturant (8-11) and they are able to fold even when additional domains are attached to one or both termini (12). In very recent experimental work (13), based on analysis of the effect of mutations in the knotted region of the protein, a folding model for YibK was also proposed. In this model the threading of the polypeptide chain and fo...

Slipknotting upon native-like loop formation in a trefoil knot protein

“…Most of the nonspecific knotted configurations were of types i and iii; however, one iv case was also found. This is consistent with theoretical evidence that folding nucleation by nonspecific knots is entropically unlikely in proteins (11,13). This process should have a barrier with a large entropic contribution because there is little energetic stabilization until the native environment forms around the knot.…”

“…1). We characterize the knot position by monitoring its depth (11), distance along the sequence K N , K C to the knot, respectively, from the N terminal and C terminal. In the case of a slipknot we additionally monitor depth of a slipknot loop (12), which is located between K N and K C .…”

“…The covalently connected nature of the polypeptide backbone makes some of the folding pathways for a knotted protein sterically impossible. Even under this more restricted situation, it has been shown that, just as with traditional proteins, a funnel-like landscape dominated by native interactions manages to fold topologically complex proteins (11). The inherent geometric constraints in the structure are able to guide the necessary preordering of chain crossings.…”

“…Theoretical investigations by Sulkowska et al showed that the native state of YibK is kinetically accessible with a nativebiased coarse-grained model through a knotting mechanism where the protein has significant native structure when the knot is created (11). In this scenario one of the termini threads a native-like loop through a slipknot intermediate, a collapsed configuration where the terminal polypeptide makes native contacts and adopts a hairpin-like configuration while threading (for detailed description of slipknot topology see also ref.…”

Conservation of complex knotting and slipknotting patterns in proteins

“…In general, knots in proteins are orders of magnitude less frequent than would be expected for random polymers with similar length, compactness, and flexibility (3). In principle, the polypeptide chains folding into knotted native protein structures encounter more kinetic difficulties than unknotted proteins (4)(5)(6)(7)(8)(9)(10)(11)(12). Therefore, it is believed that knotted protein structures were, in part, eliminated during evolution because proteins that fold slowly and/or nonreproducibly should be evolutionarily disadvantageous for the hosting organisms.…”