Additional arguments for the key role of "smart" autolysins in the enlargement of the wall of gram-negative bacteria.

Because the wall of Gram-negative bacteria is thin, the mechanism for safe enlargement of the cell is subject to strong constraints. Several models for wall growth have been proposed; in the order that they have been proposed, these include: 1) an "allosteric" model in which the critical autolysin is only functional if the bond to be cleaved is near a covalently cross-linked, but unstretched oligopeptide; 2) a model in which the cell wall is thick enough to enlarge by the "inside-to-outside" mode characteristic of Gram-positive rods; 3) a "patches" model, recently proposed by Höltje, in which only parts of the cell wall are thickened at any one time; 4) a new multienzyme model in which the transpeptidase/autolysin complex cleaves one cross-linked oligopeptidoglycan chain for every two nascent chains covalently polymerized to the sacculus. These models are considered and contrasted. While none can be rigourously excluded, no. 4 is favoured. All models as applied to the Gram-negative rod-shaped bacteria require special, extraordinary features for their autolysins. These features have not been found with any other class of enzymes, but are essential to permit safe cell expansion.