hemidesmosome

Picture a carpet that must be tacked down to the floor so foot traffic does not bunch it up or peel it away. A sheet of cells, such as the outer skin, sits on a thin mat of matrix beneath it and faces the same problem: it must be pinned firmly to that mat, or it will slide off. A hemidesmosome is one of the tacks that does this pinning — a small, button-like patch that bolts the bottom of a cell down to the matrix layer underneath it.

The name means half a desmosome, and under the microscope it looks like one half of a desmosome facing not another cell but the basement membrane below. The key difference is in the proteins. Instead of using cadherins to grab a neighboring cell, a hemidesmosome uses a different family of anchoring proteins called integrins, which reach down and grip the matrix protein laminin in the basement membrane. On the inside, the integrins connect, through a plaque, to the cell's intermediate filaments (keratin in skin), so the cell's internal ropes are tied straight down into the supporting mat outside.

Hemidesmosomes are what keep your skin attached to the tissue beneath it, and their failure is again most visible when they break. In epidermolysis bullosa, a group of inherited diseases, the proteins of the hemidesmosome or its laminin anchor are faulty; the skin then separates from its underlying layer at the slightest friction, causing severe blistering from ordinary touch. A common mix-up is calling any cell-to-matrix bolt a desmosome — but desmosomes join cell to cell, while hemidesmosomes join cell to matrix and use integrins, not cadherins.