© 1999 Robert A. Freitas Jr. All Rights Reserved.

Robert A. Freitas Jr., Nanomedicine, Volume I: Basic Capabilities, Landes Bioscience, Georgetown, TX, 1999

8.5.3.2 Cell Membrane

The cell and most organelles are individually surrounded by their own thin envelope. The envelope surrounding the entire cell is called the plasma membrane.¹⁹⁶⁸ According to the fluid mosaic model,¹¹⁸⁹ all membranes are composed of a double layer of lipid molecules, called the lipid bilayer, in which proteins are embedded (Fig. 8.37). The lipid bilayer acts as a barrier to the diffusion of polar solutes, whereas the embedded proteins provide the pathways for

1. the selective transfer of certain molecular substances through the lipid barrier, and

2. the mechanical transfer of information from the ECM into the interior of the cell.

The plasma membrane actually represents only a tiny fraction of the total membrane surface in the cell. For example, the combined membrane surface area of the endoplasmic reticulum (Section 8.5.3.5) is 44 times larger than the plasma membrane surface area for a typical human cell (Table 8.17).

Lipid bilayer plasma membranes are 6-10 nm thick. The major membrane lipids are phospholipids, fatty acid chains in the range of 16-18 carbons long; chains with fewer than 12 carbons cannot form a stable bilayer.⁹³⁹ Phospholipid chains are amphipathic molecules -- one end, the head, has a negatively-charged (polar) region, while the remainder of the molecule, the tail, consists of two (nonpolar) long fatty acid chains. The phospholipids in cell membranes self-organize into a bimolecular layer, with the nonpolar fatty acid chains in the middle. The polar regions are oriented toward the membrane surfaces due to their attraction to the polar water molecules in the extracellular and cytosolic fluids (Fig. 8.37).

The plasma membrane also contains other lipids (Table 8.18). For example, cholesterol, a steroid lipid, acts as a "mortar" that fills in small gaps in the phospholipid structure, thus improving membrane impermeability to small water-soluble molecules like glucose by a factor of ten. Cholesterol also acts as a membrane antifreeze agent, decreasing bilayer fluidity at higher temperatures (e.g., raising lipid bilayer "melting point") and preventing hydrocarbon chains of phospholipids from aggregating at lower temperatures (e.g., lowering membrane "freezing point"). Plasma membranes may contain up to ~1 cholesterol molecule for each phospholipid molecule. The precise lipid composition of plasma membranes varies from one cell type to another, and also varies among the membranes of organelles within each cell type (Table 8.18). Medical nanorobots equipped with suitable chemosensors may access this information, both for cell type identification during extracellular navigation and for organelle type identification during intracellular navigation.

There are ~5 x 10⁶ lipid molecules in a 1 micron² area of lipid bilayer⁵³¹ or ~2.5 bilayer lipid pairs/nm² of cell membrane surface. Thus the plasma membrane of a typical 20-micron human tissue cell contains ~10 billion lipid molecules. Phospholipids are not covalently bound to each other, so each lipid molecule is free to move independently, resulting in considerable random lateral movement parallel to the bilayer surfaces. The long fatty acid chains each include one unsaturated bond, producing a kink in the otherwise straight chain that prevents close packing (and solidification). The chains also wiggle back and forth, so the lipid bilayer has fluidlike characteristics much like a layer of oil on a water surface. Movement of hydrophilic head groups through the hydrophobic interior of the membrane is thermodynamically unfavorable. Such flip-flopping, or transverse diffusion, does occur in membrane lipids but is relatively slow. For instance, a typical phospholipid molecule undergoes transverse diffusion (one flip flop between monolayers) once every several hours in a lipid bilayer. By contrast, lateral diffusion of phospholipids (movement within each monolayer) is so rapid that a lipid molecule can move 10 microns (the equivalent of ~12% of cell circumference) in a few seconds.⁹³⁹ (But see Section 9.4.3.3 regarding the membrane-skeleton fence model.)

Membrane proteins are embedded in the lipid bilayer plasma membrane. Indeed, it has been said that the lipid bilayer serves as a "solvent" for membrane proteins.⁵³¹ The plasma membrane contains roughly equal masses of lipid and protein (Table 8.18). However, the mass of an individual protein molecule is much larger than the mass of any lipid molecule, so there are 10-100 times more lipid molecules than protein molecules.⁹⁹⁷ The plasma membrane of a typical 20-micron human tissue cell contains ~0.1 billion protein molecules.

There are two classes of membrane proteins: Integral (intrinsic) membrane proteins and peripheral (extrinsic) membrane proteins.

Integral membrane proteins are closely associated with membrane lipids and cannot be extracted from the membrane without disrupting the lipid bilayer. Like phospholipids, integral proteins are amphipathic. Polar amino acid side chains lie in one region of the molecule and nonpolar side chains are in a separate region. Thus integral proteins vertically align with the amphipathic lipids in the plasma membrane -- protein polar regions position themselves at the surfaces in association with polar water molecules, while the protein nonpolar regions are attracted to the interior in association with the nonpolar fatty acid chains at the center of the lipid bilayer membrane (Fig. 8.37; see also Figures 8.33 and 8.34). Many integral proteins can move laterally in the membrane; others are immobilized by links to a network of peripheral proteins located near the cytoplasmic surface of the membrane. How fast do embedded proteins laterally diffuse? If a single hybrid cell is created by fusing two cells having radiochemically-tagged membrane protein molecules, ~1 hour is needed for the two populations of transmembrane protein molecules to become thoroughly randomly intermixed.⁹³⁹

Most integral proteins are transmembrane proteins with polar regions at each end and a nonpolar region in the middle, spanning the entire membrane. These polar regions may extend up to 10-20 nm beyond the surface of the lipid bilayer, forming channels through which water, ions, or chemical signals can pass into the cell (Section 3.3.3). A few integral proteins do not cross the entire membrane and are found only in the outer or inner layer, performing functions localized to only one membrane surface. These proteins are also amphipathic and oriented parallel to the lipid molecules. Some are anchored to the membrane by covalent bonds with phospholipids. For example, in the red blood cell membrane, glycophorin spans the entire membrane, all glycolipids and most of the phosphatidylcholine are in the outer monolayer, and the majority of the phosphatidylethanolamine and phosphatidylserine molecules are in the inner monolayer where most of the proteins reside. (Cholesterol is distributed about equally between the two layers.) The number of different integral proteins in a membrane ranges from 6-8 in the sarcoplasmic reticulum to over 100 in the plasma membrane (including enzymes, transport and structural proteins, antigens and receptors), many of which are present in only a few copies per cell, although the 135 micron² red cell surface has ~1 x 10⁶ copies of the glycophorin A molecule and ~1 x 10⁵ copies of glycophorin B.¹⁰⁹¹

Peripheral membrane proteins are bound to the hydrophilic regions of integral membrane proteins or to the hydrophilic heads of membrane lipids by weak electrostatic forces.⁹³⁹ Most peripheral proteins are located near the cytoplasmic surface of the plasma membrane rather than on the extracellular surface and mediate such properties as cell shape and motility. Peripheral proteins are not amphipathic and do not associate with the hydrophobic regions of the lipids in the membrane interior.

Both lipid and protein components of the plasma membrane are continually removed and replaced. Turnover allows the cell to continuously change out damaged components. This is a highly selective process, since the rate of turnover varies for different proteins and lipids. For instance, the half-life of some phospholipids in membranes is ~10,000 sec;⁹³⁹ the "off-rate" (half-life) for cholesterol from a lipid bilayer (e.g., the red cell surface) into the cytoplasm is ~7200 sec at 310 K.¹¹¹³ Protein turnover half-lives may range from several minutes to several years, but the "typical" protein has a turnover half-life of ~200,000 sec^531,939 or ~2 days. Protein replacement is carried out by protease enzymes located in the cytoplasm and in lysosomes. Replacement rates also depend upon cell type. For example, the plasma membrane surface of the macrophage has an unusually fast mean turnover time, ~1800 sec, vs. ~5400 sec for fibroblasts.⁹⁹⁶

The plasma membrane also contains small amounts of carbohydrate. This carbohydrate is covalently linked to some of the membrane lipids and proteins. Carbohydrate portions of the membrane glycoproteins (e.g., Section 8.5.2) are always located at the extracellular surface, forming the glycocalyx (together with collagen proteins and glycosaminoglycans, aka "mucopolysaccharides"). The red cell membrane, for instance, contains 52% protein, 40% lipid, and 8% carbohydrate by weight.⁹³⁹ A small proportion of membrane carbohydrate is glycolipids, but most is in the form of glycoproteins. The sugar units are usually short oligosaccharide chains attached to serine, threonine, or asparagine side chains.

The glycocalyx, or fuzzy coat, lies exterior to the plasma membrane. In most cell types, the glycocalyx is 10-100 nm thick consisting of tangled strands of up to ~10,000-atom glycoproteins each measuring 5-8 nm thick and up to 100-200 nm in length.^531,998 The experimentally-measured thickness of the glycocalyx of various cells ranges from ~6 nm for human blood-group A erythrocytes,³¹⁶³ to 13 nm in Eimeria microgametes,³⁵⁸⁸ 20-30 nm for chick fibroblasts,³⁵⁸⁹ 30-60 nm for human bladder cells,³⁵⁹⁰ 40-70 nm for human lymphocytes,³⁵⁹¹ ~50 nm for human myocardial cells,³⁵⁹² 56 nm for frog mesenteric microvessels,³⁵⁹³ >70 nm for rat vasculature,³⁵⁹⁴ ~81 nm for rabbit endothelial cells of the systemic arteries (e.g., carotid),³¹⁶⁴ and 90 nm for human cochlear hair cells.³⁵⁹⁵ The most prominent glycocali are found in intestinal epithelial cells, where the fuzzy coat may reach 150 nm in thickness and consists primarily of oligosaccharide chains 1.2-2.5 nm in diameter.⁹³⁹ (There is one report of rat venule endothelial cells with glycocali up to 870 nm thick,³⁵⁹⁴ and a few macroscopic parasites such as the fork-tailed cercariae of the blood fluke Schistosoma mansomi have glycocali 500-2000 nm thick.³⁵⁹⁶)

The glycoproteins of the glycocalyx provide a set of highly specific biological markers that are readily recognizable by suitably equipped medical nanorobots. These markers assist normal cellular interactions by allowing blood group recognition, bacterial and toxin binding sites, egg recognition by sperm, immune responses, guidance of embryonic development, and cellular lifespan determination (e.g., the red cell coat thins with age which may serve as an RBC removal signal for phagocytes and in the liver).⁹⁴⁰

The cell's surface is also strewn with numerous pits and indentations. For example, one class of these is the "coated pits" whose inner surfaces are covered by a dense layer of the protein clathrin, important in receptor-mediated endocytosis wherein proteins and other large molecules are imported into the cytoplasm (Section 8.5.3.7). Another class of membrane indentation is the ~50 nm caveolae ("tiny caves") that serve to draw substances such as vitamins and signal transduction molecules into the cell's interior.^{1137,3376-3379} Caveolae are coated with a unique membrane marker protein called caveolin, making them easy for nanorobots to identify.

Last updated on 20 February 2003