During transformation, a bacterium can take up DNA from its environment. A small fraction of bacterial species are known to be naturally competent, meaning that they can engage in this sort of horizontal gene transfer in their natural environments.
As I understand it, naturally competent bacteria carry genes that encode the machinery necessary to facilitate the transport of this DNA across the membrane and into the cell. The exogenous DNA binds to a receptor located on the surface of competent cells, gets fragmented into small pieces, and then one strand is degraded while the other strand is transported across the membrane into the cytoplasm with the aid of a DNA translocase enzyme. Ultimately, assuming a sufficient degree of homology, it incorporates into the host genome via recombination.
During this process, however, the DNA has to be protected from degradative nucleases. I remember reading something about a transformation-specific DNA binding protein that coats the entire length of the DNA strand and prevents degradation. Is it the same translocase enzyme that facilitates movement across the membrane that is responsible for this? Or is there another protein involved?
And then, regardless of which protein is responsible in the case of natural competence, how does that protein get expressed and activated in the case of artificial competence? When we induce competence in the laboratory through one of various methods (e.g., electroporation), the DNA is essentially forced into the cell through openings in the membrane, so no transferase protein is needed. What signals the bacterium, then, to express the protective binding protein? What keeps the exogenous DNA from being readily degraded by endogenous nucleases? Whatever signals induced the expression of this protein in the case of natural competence would seem to be absent when competence is induced artificially, and it doesn't make any sense that the bacterium would constitutively express such a protein.