Real space refinement of acto-myosin structures from sectioned muscle

Area of Science:

• Structural Biology
• Biophysics
• Molecular Motors

Background:

• Fitting atomic models to low-resolution 3-D electron microscopy (EM) data is challenging.
• Manual fitting can introduce geometric errors into high-resolution atomic models.
• Optimizing model fit to EM density is crucial for understanding molecular mechanisms.

Purpose of the Study:

• To adapt a real space refinement protocol for fitting atomic models to 3-D EM reconstructions of muscle.
• To improve the accuracy and stereochemical quality of atomic models of actin-myosin interactions.
• To correct for geometric distortions in manually fitted models.

Main Methods:

• Adapted a crystallographic real space refinement protocol.
• Applied the protocol to fit atomic models of actin and myosin subfragment 1 (S1) to 3-D EM data of insect flight muscle.
• Utilized electron tomography and image averaging to obtain 7 nm resolution 3-D reconstructions.

Main Results:

• Achieved modest improvements in model-reconstruction agreement.
• Reduced conflicting atomic contacts by 70% without compromising fit to 3-D density.
• Generated stereochemically reasonable atomic models consistent with experimental data.

Conclusions:

• The adapted real space refinement protocol effectively improves atomic model fitting to low-resolution 3-D EM reconstructions.
• The method enhances the reliability of structural models derived from cryo-electron microscopy.
• This approach is valuable for studying the structure and function of molecular machines like actin-myosin.