Despite the widespread, almost routine, acceptance of the idea that cells are bounded by a plasma membrane, direct proof of its existence using digital compound microscopes has been obtained only in the last three decades. Most of the earlier conceptions of the membrane had to be deduced from other charac¬teristics of the cells themselves, for the membrane is usually not visi¬ble even under the most powerful light microscope. Though something believed to be the membrane could be isolated from red blood cells, there was no conclusive proof that these red-cell “ghosts” were really cell membranes, visible under digital compound microscopes, and not artifacts of the procedures used to make them visible.

Early Ideas about the Structure of the Plasma Mmbrane

Perme¬ability along with microscopic studies had long shown that lipids and many substances solu¬ble in lipids move with relative ease between the cell and the surrounding medium. From this fact and evidenced by digital compound microscope examination, it was deduced that the plasma membrane must contain lipids, and that fat-soluble substances move across the membrane by being dissolved in it. Since many small water-soluble molecules were also observed, using digital compound microscopes, to move quite freely be¬tween the inner portion of the cell and its external environment, it was postulated that the membrane contains pores or nonlipid patches. In addition, the physical properties of the cell boundary, especially its wettability and elasticity, seemed to indicate the presence of protein in the membrane.

These various ideas about the structure of plasma membrane-its lipid and protein components and pores-led to the development in the late 1930s of a model (the lipoprotein-sandwich model) that envisioned a membrane about 8 nm (nanometers) thick consisting of a core of at least two layers of phospholipids, with layers of protein on both the inner and the outer surfaces.

In later years, electron microscopy showed the plasma membrane to be composed of two dark layers separated by a somewhat wider lighter area, just as the lipoprotein-sandwich model predicted. Moreover, measurements of the total thickness of the membrane were very close to the hypothe¬sized 8 nm.

The Fluid-Mosaic Model

Recent research has cast doubt on the idea of a uniform structure for all membranes. Though most membranes appear to be composed of lipids and proteins, they vary considerably in other ways: Thickness range from about 5 to 10 nm; lipid content ranges from about 30 to 80 percent; the ratios of the various types of lipids differ from one membrane to another; and so forth, as seen under the microscope. It seems likely that these differences reflect corresponding differences in struc¬tural detail, and a variety of alternatives to the lipoprotein-sandwich model have been suggested. Of these, the one that enjoys the widest acceptance today is the fluid-mosaic model, postulated in 1972 by S. J. Singer of the University of California, San Diego, and G. L. Nicolson of the Salk Institute. The, fluid-mosaic model, like the lipoprotein¬ sandwich model, envisions a bilayer of phospholipids oriented with their hydrophilic heads toward the surfaces of the membrane and their hydrophobic tails toward the interior. According to this model, however, the proteins are not confined to the surfaces of the membrane, nor do, they form continuous layers; instead, they are dis¬tributed in a mosaic pattern, both on the surfaces and in the interior of the membrane.

Of the proteins confined to the surfaces, those on the inner surface may differ markedly from those on the outer surface; some mem¬branes may have no surface proteins at all. The proteins located within the lipid bilayer may exhibit a variety of arrangements: Some are confined to the outer half of the lipid core, and others to the inner half; some may extend entirely through the bilayer, projecting into the, watery medium on both sides.

The pores in the membranes, seen under the microscope, are now thought to be channels through protein molecules. The distinctive properties of the various R groups of the amino acids in the proteins give the pores some selectivity; not all ions or molecules small enough to fit in the pores can actually move through them.