outer surface, similar protrusion inwards might be expected with isolated walls, whereas a difference in the gel structure of the wall could explain the morphological difference, and also the finding (Burger, 1966) that isolated walls bind little more antibody than do whole bacteria. The inacessibility of new teichoic acid to concanavalin A at the bacterial surface shows that the teichoic acid chains do not extend through the thickness of the wall; it also shows that concanavalin A molecules cannot penetrate the gel structure of the interior region of the wall (Lang & Archibald, 1983). This is consistent with previous work showing that walls are of low porosity (Scherrer et al., 1977). It follows that large proteins exported from the membrane are unlikely to be freely diffusible through the wall both because of pore size and because of the possibilities of ionic and specific binding interactions between proteins and wall polymers. In a preliminary study (with Dr. C. R. Harwood) we have found that there is a delay between synthesis and export of proteins in B. subtilis (see Fig. 2). This delay is of the same order as the time taken for inside to outside movement of the cell wall, suggesting that a dynamic interaction between wall assembly and macromolecular export is possible. Whether export can be added to the list of functionally important interactions between proteins and teichoic acids remains to be demonstrated. However, the importance of teichoic acids in microbial physiology and in medical and industrial microbiology ensures that the next 25 years, like the last, will see increasing attention paid to the properties and functions of these most interesting polymers.
Biochemical Society transactions