My answer would be No, a protein localed in the periplasm compartment cannot be considered as extracellular.
We have exploited this phenomenon in M. tuberculosis to direct a secretion of recombinant cathepsin S to improve its secretion outside the bacterium (M. bovis BCG).
In this organism, the Antigen 85 kDa is secreted upon trancation of the peptide signal. we constructed a chimera of this peptid signal with an human Cathespsin S, and cloned it in M. bovis BCG, We have confirmed the presence of two form of CatS: One that is secreted extracellularaly and have an enzymatic activity, the other one is still in bacterial pellet and contains the signal peptid (high molecular weight) and is not active.
This stategy was used in several works to induce secretion of foreign protein by mycobacteria.
In general the signal peptid is located up-or downstream a ribosomal binding site, or an hydrophobic region. All this domain is beleived to act as propetid to direct its localization and/or activity or inhibition Inside the cell..
a protein is considered secreted when it can be found extracellularly. This means outside the outer membrane for a Gram negative bacterium and outside the cytoplamic membrane for a Gram positive bacterium.
Cytoplasmic proteins do not have a signal peptide. Proteins that will be present in the inner membrane, in the periplasm or in the outer membrane have a signal peptide sequence. Some secreted proteins have a signal peptide but not all. It depends how they are secreted. For example proteins secreted by a type III secretion system do not have a signal peptide sequence as they go directly from the cytoplasm to the extracellular milieu. The signal peptide allows the protein to go through the inner membrane using the Sec system. It is cleaved once the protein reaches the outer face of the inner membrane. The Tat system is a second system that allow the proteins to go through the inner membrane.