Although many people have attempted to crystallize or otherwise solve the structure of the entire protein, these attempts have not been successful. There are several reasons for this lack of success, including the fact that full-length INs can be significantly less soluble than the catalytic core, and that the three domains are connected by flexible regions. Flexibility is undoubtedly important for enzyme function, but this very flexibility may result in more conformational variability than the conditions for protein crystallization can handle. It appears that a more rigid protein unit, such as the catalytic core domain, can pack better into a crystalline form than can the full-length protein. The structures of the two other domains, the N- and C-termini, have been solved in solution by nuclear magnetic resonance (NMR) spectroscopy. However, having all three domain structures solved separately does not allow a determination of how the entire protein fits or works together. Additional work on solving the structure of the entire protein or separate domains with inhibitors and DNA is in progress. *Core domain diagram adapted from work done for the MSL by Richard Frederickson, Publications Dept., NCI-FCRDC.