RosettaRemodel: a generalized framework for flexible backbone protein design

Abstract

We describe RosettaRemodel, a generalized framework for flexible protein design that provides a versatile and convenient interface to the Rosetta modeling suite. RosettaRemodel employs a unified interface, called a blueprint, which allows detailed control over many aspects of flexible backbone protein design calculations. RosettaRemodel allows the construction and elaboration of customized protocols for a wide range of design problems ranging from loop insertion and deletion, disulfide engineering, domain assembly, loop remodeling, motif grafting, symmetrical units, to de novo structure modeling.

Figure 1. Examples of backbone manipulation using RosettaRemodel.

In the center is the crystal structure of protein G (PDB ID: 1PGA), which was used as the starting point for all the different cases. The colored regions highlights changed made with RosettaRemodel.

Figure 2. Blueprint Assignment Examples.

Figure 3. Constraints file and its associated blueprint file.

Each block in the constraints file is represented in the blueprint file with the CST1A, CST1B, CST2A and CST2B notation. The enzdes constraint format is discussed in Richter et al. in detail. In this figure we show the association between cst and blueprint files. Each enzdes constraint block, defined between CST::BEGIN and CST::END statements, always contains two elements, and to interface with blueprint, the first definition is defined as “A” and the second as “B,” and each element requires a corresponding assignment in the blueprint. Each block is also associated with a numerical value from 1 to the total number of blocks defined in the cst file. In this example, the first constraint pair (CST1A/CST1B) is used to restraint one of the residues (residue 17) on the strand being built (residues 16–20) to within a hydrogen bonding distance with a stationary residue (residue 12). The second constraint pair (CST2A/CST2B) operates on the sidechains of residue 9 and 20 for hydrogen bonding between the functional groups. The distinction between a backbone and sidechain definition is the choice of atom types using Rosetta atom type names and a required “is_backbone” statement because enzdes constraint protocol does not automatically treat atoms as backbone by names. In this example, the hydrogen bonding constraint is defined for a pair of atoms within 2.8+/−0.2 Å, with a force constant of 100. The trailing “0” in the distanceAB definition is for non-covalent interaction.

Figure 4. Examples of designs made with RosettaRemodel.

A) A cysteine protease site, with cys-his intermediate shown in green sticks. The loop in cyan was rebuilt from its original position (in green) to introduce sidechain-directed stabilization of the oxyanion in the intermediate. The re-designed model uses an asparagine in direct contact with the oxyanion instead of the wild-type aspartic acid (in pink). B) Two interacting loops in a symmetry arrangement were rebuilt to increase interactions across the subunits. C) Domain localization. A domain assembly designed with a pair of linkers. In this figure, the ensemble of an internally inserted domain is shown moving relative to the stationary structure that hosts the insertion as a result of sampling the loops linking them. With this type of sampling, one could model the localization of the final assembly.