Concluding remarks on the special issue dedicated to bacterial secretion systems: function and structural biology.

In Gram-positive monodermic bacteria (which have only one membrane), translocation across this unique membrane, in some cases, drives secretion (Freudl, 2013). In Gram-negative didermic bacteria (which have two membranes delimitating a periplasmic space), translocation across the inner membrane does not, in itself, lead to protein secretion. Chronologically, the mechanisms of protein translocation across the cytoplasmic membrane were discovered before the secretion systems (Beckwith, 2013). Both, translocation system Sec and the more recently characterized Tat system rely on a signal peptide and are universal. This is not the case for secretion in Gram-negative bacteria. At present, six different secretion pathways have been identified, numbered TSS1 to TSS6. The first systems to be defined were T1SS and T2SS, at virtually the same time, in the middle of the 1980s (d’Enfert et al., 1987; Mackman et al., 1985; Wandersman et al., 1987). In each case, the discovery of specific secretion machinery was made possible by expressing genes of different bacterial origins encoding extracellular proteins in Escherichia coli K12, a laboratory strain, which does not innately secrete proteins. The foreign proteins were synthesized in E. coli K12, but were not secreted into extracellular medium. This was the first clue suggesting that the secreted proteins do not contain information sufficient to allow their secretion. Some foreign exoproteins with signal peptides were transported to the periplasmic space when expressed in E. coli K12. This was the case for Klebsiella pneumoniae pullulanase. In other cases,