Volume 25, number 1 I-EBS LETTERS September 1972 zyxwvutsrq THE CAUSE OF IRREVERSIBLE POLYMERISATION OF TOBACCO MOSAIC VIRUS PROTEIN A.C.H. DURHAM zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPO MRC I,aboratorj3 of‘hfolccrtlar Bioiog),. Hills Road. Cambri&c. CB-7 ?QH. Enplaud Received 9 June 1972 1. Introduction Protein from tobacco mosaic virus (TMV) can polymerise into many different organised structures. Some of these polymers form reproducibly and revers- ibly in conditions which have been described recent- ly [I-S], but others, of which the best known is the stacked disk rod, form irreproducibly in conditions which have been only partially characterised [4.6]. The present paper shows that the stacked disk rod. and several other polymers, result from proteolytic cleavage of the TMV polypeptide chain, thus explain- ing many strange observations upon the protein and providing an unusually detailed illustration of a spoil- age mechanism by which a protein can lose its biolo- gical activity. Stacked disk rods of TMV protein have been ob- served in many laboratories and their structure has been characterised by Finch and Klug [7]. It was originally thought that they differed only slightly from the virus-like single helix, since both types of rod have very similar intersubunit bonds and overall dimensions (see fig. l), but it is now clear that there is a fundamental difference, since stacked disk rods are much more difficult to dissociate than single helices or any other reproducible polymers. Carpenter [6] observed that stacked disk rods formed when TMV protein was kept for weeks in alkaline solution, and that they were not in reversible equilibrium with their precursors, but required extremes of pH for dis- sociation. lndependently [8], 1 had also observed that stack- ed disk rods generally were the ultimate aggregate North-Hollarrd Publishitlg Compatry Arustcrdam formed when alkaline TMV protein solutions were left to stand for a long time. Observation of a num- ber of other irreproducible polymerisation states of TMV protein - “fish” [ 1] , a single peak sediment- ing at about 8 S [‘_I. irreversibly associated short stacks of disks [3] , abnormally high molecular weight: in sedimentation equilibrium experiments [3], cloudiness appearing in solutions stored at alkaline pH [3] , open helical rods [4] , rods with an unusual- ly wide central hole 14) , and double helical rods [ 71 led to the working hypothesis that some slow change (such as deamidation. cross-linking, loss of the C- terminal residue [‘9], or a slow change to another con- formation) could occur to TMV protein in any state of reversible association, causing it to become “locked in that structure and unable to dissociate easily. Thus, fish would be locked two-layer aggregates (A-protein). stacked disk rods would be a locked and longer vel-s- ion of reversible stacks of disks, the single 8 S sedi- mentation peak would be due to a locked form of the 8 S structure normally found in reversible equilibrium with A-protein, and so on. Other authors have also observed behaviour which is inconsistent with the normal reversible polymerisation behaviour. For ex- ample. McCarthy [lo] separated by gel clectrophor- esis several TMV protein polymers which could not have been in a state of rapid association equilibrium. and Kleczkowski [ II] observed rods, whose electron micrographs look suggestive of stacked disk rods, when he irradiated TMV protein with ultraviolet light, at a pH where single helix is not stable. Stacked disk rods have been useful experimentally. being regular polymers in which the structural details 147