1. Simple epithelia (Figure 1-2), formed by only one layer of cells. They are subdivided into simple squamous, simple cuboidal, and simple columnar, according to the height and width of the cells. The specific name endothelium is used for the simple epithelium lining the blood and lymphatic vessels. Mesothelium is the simple epithelium lining all body cavities (peritoneum, pericardium, and pleura).

2. Stratified epithelia (Figure 1-3) are composed of two or more cell layers. Stratified epithelia are subclassified according to the shapes of the cells at the superficial or outer layer into stratified squamous, stratified cuboidal, and stratified columnar. Stratified squamous is the epithelium most frequently found and can be subdivided into moderately keratinized (also known as nonkeratinized) or highly keratinized types. The cells of the outer layer of a moderately keratinized squamous epithelium can display nuclei (for example, esophagus and vagina). Nuclei are absent in the outer layer of the highly keratinized stratified squamous epithelium (for example, the epidermis of the skin). The basal cells aligned along the basal lamina are mitotically active and replace the differentiating cells of the upper layers.

3. Two special categories are the pseudostratified epithelium and the transitional epithelium (Figure 1-4). The pseudostratified epithelium consists of basal and columnar cells resting on the basal lamina. Only the columnar cells reach the luminal surface, however. Because the nuclei of the basal and columnar cells are seen at different levels, one has the impression of a stratified epithelial organization. Within this category are the pseudostratified columnar ciliated epithelium of the trachea and the pseudostratified columnar epithelium with stereocilia of the epididymis. The transitional epithelium of the urinary passages is also referred to as urothelium. The urothelium also consists of basal and columnar dome-shaped cells. An important feature of this epithelium is that its height varies with distention and contraction of the organ (see Chapter 14, Urinary System).

Figure 1-2 Simple epithelium

Figure 1-3 Stratified epithelium

Figure 1-4 Pseudostratified and transitional epithelia

An important aspect of epithelia is its polarity. Most epithelial cells line surfaces and cavities and have three domains (Figure 1-5):

1. The apical domain is exposed to the lumen or external environment.

2. The lateral domain faces neighboring epithelial cells linked to each other by cell adhesion molecules and junctional complexes.

3. The basal domain is associated with a basal lamina that separates the epithelium from underlying connective tissue. The basal lamina is reinforced by components of the connective tissue. The basal lamina–connective tissue complex is designated the basement membrane.

Figure 1-5 Domains of a polarized epithelial cell

Epithelial cells are attached to each other by junctional complexes and adhesion molecules. Epithelial cells are specialized to fulfill important roles, such as absorption and secretion or to act as a water or gas barrier. Several cell barriers and their functional significance are studied.

EPITHELIAL CELL POLARITY

Epithelial cells have two major domains (Figure 1-5):

1. An apical domain

2. A basolateral domain

Each domain is defined by specific structural and functional characteristics. For example, the apical domain has structures important for the protection of the epithelial surface (such as cilia in the respiratory tract) or for the absorption of substances (such as microvilli in the intestinal epithelium).

Junctional complexes and cell adhesion molecules are present at the basolateral domain to anchor epithelial cells to each other and to the basement membrane.

Apical differentiations

The apical domain of some epithelial cells can display three types of differentiation:

1. Cilia

2. Microvilli

3. Stereocilia

Cilia (singular, cilium; Figure 1-6) are motile cell projections originating from basal bodies anchored by rootlets to the apical portion of the cytoplasm. A basal body contains nine triplet microtubules in a helicoid array without a central microtubular component. By contrast, a cilium consists of an assembly called an axoneme, formed by a central pair of microtubules surrounded by nine concentrically arranged microtubular pairs. This assembly is known as the 9 + 2 microtubular doublet arrangement. The axoneme is also a component of the sperm tail, or flagellum.

Figure 1-6 Apical differentiations of epithelial cells: cilia and primary cilium

The trachea and the oviduct are lined by ciliated epithelial cells. In these epithelia, ciliary activity is important for the local defense of the respiratory system and for the transport of the fertilized egg to the uterine cavity.

Some cells have a primary cilium. The importance of primary cilia emerges from rare recessive human disorders known as ciliopathies caused by structural or functional abnormalities of cilia. The structure and mechanism of assembly of primary cilia are shown in Figure 1-6. The significant aspects of the primary cilium are: (1) they are non-motile; (2) they participate in the early stages of embryonic patterning leading to organogenesis; (3) many components of the hedgehog signaling pathway, essential at least in early development, are present in primary cilia; and (4) the position of the primary cilium, called kinocilium, of the hair cell of the organ of Corti in the inner ear determines the correct polarity of the actin-containing stereocilia (see Chapter 9, Sensory Organs: Vision and Hearing).

Microvilli (singular, microvillus; Figure 1-7) are finger-like cell projections of the apical epithelial cell surface containing a core of cross-linked microfilaments (a polymer of G-actin monomers). At the cytoplasmic end of the microvillus, bundles of actin and other proteins extend into the terminal web, a filamentous network of cytoskeletal proteins running parallel to the apical domain of the epithelial cell.

Figure 1-7 Apical differentiations of epithelial cells: microvilli and stereocilia

The intestinal epithelium and portions of the nephron in the kidney are lined by epithelial cells with microvilli forming a brush border. In general, a brush border indicates the absorptive function of the cell.

Stereocilia (singular, stereocilium; see Figure 1-7) are long and branching finger-like projections of the apical epithelial cell surface. Similar to microvilli, stereocilia contain a core of cross-linked actin with other proteins. Stereocilia do not have axonemes. Stereocilia are typical of the epithelial lining of the epididymis and contribute to the process of sperm maturation occurring in this organ.

CELL ADHESION MOLECULES

A sheet of epithelial cells results from the tight attachment of similar cells to each other and to the basal lamina, a component of the extracellular matrix. Cell adhesion molecules enable interepithelial cell contact, and this contact is stabilized by specialized cell junctions. A consequence of this arrangement is the apical and basolateral domain polarity of an epithelial sheet.

Although cell adhesion molecules and cell junctions are considered here within the framework of epithelia, nonepithelial cells also can use cell adhesion molecules and junctions to establish contact with each other, enabling cell-cell communication. A typical example of nonepithelial cells connected by specialized junctions is the cardiac muscle (see Chapter 7, Muscle Tissue).

There are two major classes of cell adhesion molecules (see Box 1-B):

1. Ca²⁺-dependent molecules, including cadherins and selectins

2. Ca²⁺-independent molecules, which compose the immunoglobulin super-family and integrins

Box 1-B Cell adhesion molecules

• Cell adhesion molecules can be classified as Ca²⁺-dependent and Ca²⁺-independent.

• Ca²⁺-dependent adhesion molecules include cadherins and selectins.

• Ca²⁺-independent adhesion molecules include cell adhesion molecules of the immunoglobulin superfamily (CAMs) and integrins.

• Cadherins and CAMs display trans-homophilic interaction across the intercellular space.

• Integrins are the only cell adhesion molecules consisting of two subunits: a and β.

• Cadherins and integrins interact with F-actin through adapters (catenins for cadherins, and vinculin, talin, and α-actinin for integrins).

Many cells can use different cell adhesion molecules to mediate cell-cell attachment. Integrins are mainly involved in cell–extracellular matrix interactions. Cadherins and integrins establish a link between the internal cytoskeleton of a cell and the exterior of another cell (cadherins) or the extracellular matrix (integrins).

Cadherins (Figure 1-8) are a family of Ca²⁺-dependent molecules with a major role in cell adhesion and morphogenesis. A loss of cadherins is associated with the acquisition of invasive behavior by tumor cells (metastasis) (see Chapter 4, Connective Tissue).

Figure 1-8 Cadherins

There are more than 40 different cadherins. E-cadherin is an epithelial cadherin found along the lateral cell surfaces and is responsible for the maintenance of most epithelial layers. The removal of calcium or the use of a blocking antibody to E-cadherin in epithelial cell cultures breaks down cell-cell attachment, and the formation of stabilizing junctions is disrupted. E-cadherin molecules form cis-homophilic dimers (“like-to-like”), which bind to dimers of the same or different class of cadherins in the opposite cell membrane (trans-homophilic or heterophilic [“like-to-unlike”] interaction). These forms of binding require the presence of calcium and result in a specialized zipper-like cell-cell adhesion pattern.

N-cadherin is found in the central nervous system, the lens of the eye, and in skeletal and cardiac muscle. P-cadherin is observed in placenta (trophoblast).

The cytoplasmic domain of cadherins is linked to actin through intermediate proteins known collectively as the catenin (Latin catena, chain) complex. The complex includes catenins (α, β, and γ) and actin-binding proteins, such as α-actinin, vinculin, and formin-1, among others.

The catenin complex has at least three distinct roles in the function of cadherins: (1) catenins mediate a direct link to filamentous actin; (2) they interact with regulatory molecules of the actin cytoskeleton; and (3) they control the adhesive state of the extracellular domain of cadherins. The association of actin to the cadherin-catenin complex is essential for cell morphogenesis, changes in cell shape, and the establishment of cell polarity.

Members of the cadherin family also are present between cytoplasmic plaques of the zonula and the macula adherens. β-catenin plays a significant role in colorectal carcinogenesis (see Chapter 16, Lower Digestive Segment).

Selectins (Figure 1-9), similar to cadherins, are Ca²⁺-dependent cell adhesion molecules. In contrast to cadherins, selectins bind to carbohydrates and belong to the group of lectins (Latin lectum, to select). Each selectin has a carbohydrate-recognition domain (CRD) with binding affinity to a specific oligosaccharide attached to a protein (glycoprotein) or a lipid (glycolipid). The molecular configuration of the CRD is controlled by calcium.

Figure 1-9 Selectins

Selectins participate in the movement of leukocytes (Greek leukos, white, kytos, cell) circulating in blood (neutrophils, monocytes, B and T cells) toward tissues by extravasation. Extravasation is the essence of homing, a mechanism that enables leukocytes to escape from blood circulation and reach the sites of inflammation (see Figure 1-12). Homing also permits thymus-derived T cells to home in on peripheral lymph nodes (see Chapter 10, Immune-Lymphatic System).

The three major classes of cell surface selectins are as follows:

1. P-selectin, found in platelets and activated endothelial cells lining blood vessels

2. E-selectin, found on activated endothelial cells

3. L-selectin, found on leukocytes

P-selectin is stored in cytoplasmic vesicles in endothelial cells. When endothelial cells are activated by inflammatory signaling, P-selectin appears on the cell surface. On their surface, leukocytes contain sialyl Lewis-x antigen, a specific oligosaccharide ligand for P-selectin. P-selectin binding to the antigen slows down streaming leukocytes in blood, and they begin to roll along the endothelial cell surfaces. P-selectins get additional help from members of the immunoglobulin (Ig) superfamily and integrins to stabilize leukocyte attachment, leading to extravasation (see Figure 1-12).

N-CAM (for neural cell adhesion molecule) belongs to the Ig superfamily and mediates homophilic and heterophilic interactions. In contrast to cadherins and selectins, members of the Ig superfamily are Ca²⁺-independent cell adhesion molecules and are encoded by a single gene. Members of the Ig superfamily are generated by the alternative messenger RNA (mRNA) splicing and have differences in glycosylation.

A conserved feature shared by all members of the Ig superfamily is an extracellular segment with one or more folded domains characteristic of immunoglobulins (Figure 1-10). Of particular interest is CD4, a member of the Ig superfamily and the receptor for the human immunodeficiency virus type 1 (HIV-1) in a subclass of lymphocytes known as T cells or helper cells. We discuss the significance of several members of the Ig superfamily in Chapter 10, Immune-Lymphatic System.

Figure 1-10 Immunoglobulin superfamily

Other members of the Ig superfamily play important roles in the homing process during inflammation. Examples include intercellular adhesion molecules 1 and 2 (ICAM-1 and ICAM-2) on endothelial cell surfaces. ICAM-1 is expressed when an inflammation is in progress to facilitate the transendothelial migration of leukocytes (see Chapter 6, Blood and Hematopoiesis).

Integrins (Figure 1-11) differ from cadherins, selectins, and members of the Ig superfamily in that integrins are heterodimers formed by two associated α and β subunits encoded by different genes. There are about 22 integrin heterodimers consisting of 17 forms of α subunits and 8 forms of β subunits.

Figure 1-11 Integrins

Figure 1-12 Homing, a process involving selectins and integrins

Almost every cell expresses one or several integrins. Similar to cadherins, the cytoplasmic domain of β integrins is linked to actin filaments through connecting proteins (talin, vinculin, and α-actinin).

The extracellular domain of integrins binds to the tripeptide RGD (Arg-Gly-Asp) sequence present in laminin and fibronectin, two major components of the basement membrane, a specific type of extracellular matrix. Laminin and fibronectin interact with various collagen types (including type IV collagen), heparan sulfate proteoglycan perlecan, and entactin (also called nidogen).

The integrin–extracellular matrix relationship is critical for cell migration to precise sites during embryogenesis and can be regulated when cell motility is required. In addition to their role in cell-matrix interactions, integrins also mediate cell-cell interaction. Integrins containing β₂ subunits are expressed on the surface of leukocytes and mediate cell-cell binding. An example is the α₁β₂ integrin heterodimer that binds to ligands on endothelial cell surfaces during the integrin phase (extravasation) of homing (Figure 1-12).

Integrins respond to intercellular events by changing their adhesive conformation with respect to molecules of the extracellular matrix. This response is known as inside-out signaling. In addition, integrins mediate a complex intracellular cascade in response to extracellular events.

ADAM proteins

The reversal of integrin-mediated cell binding to the extracellular matrix can be disrupted by proteins called ADAM (for a disintegrin and metalloprotease). ADAMs have pivotal roles in fertilization, angiogenesis, neurogenesis, heart development, cancer and Alzheimer’s disease (see Chapter 8, Nervous Tissue).

A typical ADAM protein (Figure 1-13) contains an extracellular domain and an intracellular domain. The extracellular domain consists of several portions including a disintegrin domain and a metalloprotease domain.

Figure 1-13 ADAM protein, a sheddase

1. A disintegrin domain binds to integrins and competitively prevents integrin-mediated binding of cells to laminin, fibronectin, and other extracellular matrix proteins.

2. A metalloprotease domain degrades matrix components and enables cell migration.

A significant function of ADAMs is protein ectodomain shedding, consisting of the proteolytic release of the ectodomain of a membrane protein cleaved adjacent to the plasma membrane. ADAMs are members of the family of sheddases. Ectodomain shedding targets for cleavage the proinflammatory cytokine tumor necrosis factor–α (TNF-α) and all ligands of the epidermal growth factor receptor. A released soluble ectodomain of a cytokine or growth factor can function at a distance from the site of cleavage (paracrine signaling). Ectodomain shedding of a receptor can inactivate the receptor by functioning as a decoy sequestering soluble ligands away from the plasma membrane-bound unoccupied receptor.

A defect in TNF receptor 1 (TNFR1) shedding, determined by a mutation in the receptor cleavage site, causes aperiodic febrile syndrome because of continuous availability of TNFR1 for TNF-α binding. Consequently, recurring fever occurs by increased inflammatory responses.

CELL JUNCTIONS

Although cell adhesion molecules are responsible for cell-cell adhesion, cell junctions are necessary for providing stronger stability. In addition, the movement of solutes, ions, and water through an epithelial layer occurs across and between individual cell components. The transcellular pathway is controlled by numerous channels and transporters. The paracellular pathway is regulated by a continuous intercellular contact or cell junctions. A deficiency in the cell junctions accounts for acquired and inherited diseases caused by inefficient epithelial barriers.

Cell junctions are symmetrical structures formed between two adjacent cells. There are three major classes of symmetrical cell junctions (Figure 1-14; see Box 1-C):

1. Tight junctions

2. Anchoring junctions

3. Gap or communicating junctions

Figure 1-14 Anchoring and communicating junctions

Box 1-C Cell junctions

• Cell junctions can be classified as symmetrical and asymmetrical. Symmetrical junctions include the tight junctions, the belt desmosome (zonula adherens), desmosomes (macula adherens), and gap junctions. The hemidesmosome is an asymmetrical junction

• Tight junctions contain occludin and claudin, belonging to the protein family of tetraspanins because four segments of each protein span the plasma membrane. An additional component is the afadin-nectin protein complex.

Junctional adhesion molecules (JAMs), zonula occludens (ZO) proteins ZO-1, ZO-2, and ZO-3 and F-actin are additional protein components. Tight junctions form a circumferential gasket that controls the paracellular pathway of molecules.

• Zonula adherens (belt desmosome) consists of a plaque that contains desmoplakin, plakoglobin (γ catenin), and plakophilin. Cadherins, mainly desmocollins and desmogleins dimers, and the afadin-nectin complex extend from the plaque to the extracellular space. A catenin complex links actin filaments to the plaque. Similar to tight junctions, the belt desmosome forms a circumferential gasket at the apical region of epithelial cells.

• Macula adherens (spot desmosome) are structurally comparable with the zonula adherens except that the afadin-nectin and catenin complexes are absent and intermediate filaments (tonofilaments), instead of actin filaments, are attached to the plaque.

• Hemidesmosomes consist of an inner membrane plate—to which tonofilaments attach—and an outer membrane plaque, linked by integrin α₆β₄ and laminin 5 to the basal lamina.

• Tight junctions, the belt desmosome, spot desmosomes, and hemidesmosomes are anchoring junctions. Gap junctions are not anchoring junctions. Instead, gap junctions are communicating junctions linking adjacent cells. The basic unit of a gap junction is the connexon, formed by 6 connexin molecules surrounding a central channel.

Tight junctions (also called occluding junctions) (Figure 1-15) have two major functions:

1. They determine epithelial cell polarity by separating the apical domain from the basolateral domain and preventing the free diffusion of lipids and proteins between them.

2. They prevent the free passage of substances across an epithelial cell layer (paracellular pathway barrier).

Figure 1-15 Molecular organization of tight junctions

Cell membranes of two adjacent cells come together at regular intervals to seal the apical intercellular space. These areas of close contact continue around the entire surface of the cell like a belt, forming anastomosing strips of the trans-membrane proteins occludin and claudin. Occludin and claudin belong to the family of tetraspanins with four transmembrane domains, two outer loops, and two short cytoplasmic tails.

Occludin interacts with four major zonula occludin (ZO) proteins: ZO-1, ZO-2, ZO-3, and afadin. Claudin (Latin claudere, to close), a family of 16 proteins forming linear fibrils in the tight junctions, confers barrier properties on the paracellular pathway. A mutation in the gene encoding claudin 16 is the cause of a rare human renal magnesium wasting syndrome characterized by hypomagnesemia and seizures.

Two members of the Ig superfamily, nectins and junctional adhesion molecules (JAMs), are present in tight junctions. Both form homodimers (cis homodimers) and then trans homodimers across the intercellular space. Nectins are connected to actin filaments through the protein afadin. The targeted deletion of the afadin gene in mice results in embryonic lethality. A mutation in the nectin-1 gene is responsible for cleft lip/palate and ectodermal dysplasia (CLEPD1) of skin, hairs, nails, and teeth in humans. Nectin-2–deficient male mice are sterile.

Tight junctions can be visualized by freeze-fracturing a network of branching and anastomosing sealing strands. We discuss in Chapter 2, Epithelial Glands, the procedure of freeze-fracturing for the study of cell membranes.

Anchoring junctions are found below the tight junctions, usually near the apical surface of an epithelium. There are three classes of anchoring junctions (see Figures 1-14, 1-16, 1-18, and 1-19):

1. The zonula adherens or belt desmosome

2. The macula adherens or spot desmosome

3. The hemidesmosome

Figure 1-16 Zonula adherens (belt desmosome)

Figure 1-17 Desmogleins in skin disease: Pemphigus foliaceus

Figure 1-18 Macula adherens (spot desmosome)

Figure 1-19 Hemidesmosome

Similar to the tight junctions, the zonula adherens is a beltlike junction. The zonula adherens (Figure 1-16) is associated with actin microfilaments. This association is mediated by the interaction of cadherins (desmocollins and desmogleins) with catenins (α, β, and γ). The main desmogleins expressed in the epidermis of the skin are desmoglein 1 and desmoglein 3 (Figure 1-17).

The macula adherens (also called desmosome) is a spotlike junction associated with keratin intermediate filaments (also known as tonofilaments) extending from one spot to another on the lateral and basal cell surfaces of epithelial cells (Figure 1-18). Spot desmosomes provide strength and rigidity to an epithelial cell layer. Spot desmosomes are also present in the intercalated disks linking adjacent cardiocytes in heart (see Chapter 7, Muscle Tissue) and in the meninges lining the outer surfaces of the brain and spinal cord.

In contrast to occluding junctions, adjacent cell membranes linked by zonula and macula adherens are separated by a relatively wide intercellular space. This space is occupied by the glycosylated portion of proteins of the cadherin family, desmogleins and desmocollins, anchored to cytoplasmic plaques containing desmoplakin, plakoglobin (γ-catenin), and plakophilin. The cytoplasmic plaques are attached to the cytosolic face of the plasma membrane. The interlocking of similar cadherins binds two cells together by Ca²⁺-dependent homophilic or heterophilic interaction, as we have already seen. Inherited disorders of some of the desmosomal components are indicated in Figure 1-18.

The human desmosomal cadherins genes include four desmogleins and three desmocollins. Their cytoplasmic regions interact with plakoglobin and plakophilin. Desmoplakin interacts with the intermediate filaments keratin in epidermis, desmin in the intercalated disks, and vimentin in the meninges. Desmoglein 1 and desmoglein 3 maintain the cohesiveness of the epidermis, a stratified squamous epithelium. Autoantibodies to desmoglein 1 cause a blistering disease (disruption of cell adhesion) of the skin called pemphigus foliaceus (see Figure 1-17).

Hemidesmosomes are asymmetrical structures anchoring the basal domain of an epithelial cell to the underlying basal lamina (Figure 1-19).

Hemidesmosomes have a different organization compared with a macula adherens or desmosome. A hemidesmosome consists of the following:

1. An inner cytoplasmic plate associated with intermediate filaments (also called keratins or tonofilaments)

2. An outer membrane plaque linking the hemidesmosome to the basal lamina by anchoring filaments (composed of laminin 5) and integrin α₆β₄

Although hemidesmosomes look like half-desmosomes, none of the biochemical components present in the desmosome is found in hemidesmosomes. Hemidesmosomes increase the overall stability of epithelial tissues by linking intermediate filaments of the cytoskeleton with components of the basal lamina. We consider additional details of the hemidesmosomes and their role in autoimmune diseases of the skin when we discuss the structure of intermediate filaments in the cytoskeleton section.

Gap junctions are symmetrical communicating junctions formed by integral membrane proteins called connexins. Six connexin monomers associate to form a connexon, a hollow cylindrical structure that spans the plasma membrane. The end-to-end alignment of connexons in adjacent cells provides a direct channel of communication (1.5 to 2 nm in diameter) between the cytoplasm of two adjacent cells (Figure 1-20). Connexons have a clustering tendency and can form patches about 0.3 mm in diameter.

Figure 1-20 Gap junctions

These junctions facilitate the movement of molecules 1.2 nm in diameter (for example, Ca²⁺ and cyclic adenosine monophosphate [cAMP]) between cells. The connexon axial channels close when the concentration of Ca²⁺ is high. This junction is responsible for the chemical and electrical “coupling” between adjacent cells. A typical example is cardiac muscle cells connected by gap junctions to enable the transmission of electrical signals.

Clinical significance: Connexin mutations in human disease

Several diseases occur when genes encoding connexins are mutated. Mutations in the connexin 26 (Cx26) gene, highly expressed in cells of the cochlea, are associated with deafness.

Mutations in the connexin 32 (Cx32) gene are found in X-linked Charcot-Marie-Tooth demyelinating neuropathy resulting in progressive degeneration of peripheral nerves, characterized by distal muscle weakness and atrophy and impairment of deep tendon reflexes. Connexin 32 protein is expressed in Schwann cells, which are involved in the production of rolled myelin tubes around the axons in the peripheral nervous system (see Chapter 8, Nervous Tissue). Gap junctions couple different parts of the rolled myelin tubes of the same Schwann cell, rather than different cells. A loss of the functional axial channels in myelin leads to the demyelinating disorder. Mutations in the connexin 50 (Cx50) gene are associated with congenital cataracts, leading to blindness.

Bone cells (osteoblasts/osteocytes) are connected by gap junctions and express connexin 43 (Cx43) and connexin 45 (Cx45) proteins. A deletion of the Cx43 gene determines skeletal defects and delays in mineralization.

BASEMENT MEMBRANE

Integrins mediate cell-matrix interactions by their binding affinity to the RGD domain in laminin and fibronectin (see Figure 1-11). Laminin and fibronectin are distinct proteins of the extracellular matrix and are associated with collagens, proteoglycans, and other proteins to organize a basement membrane, the supporting sheet of most epithelia.

The basement membrane consists of two components (Figure 1-21):

1. The basal lamina, a sheetlike extracellular matrix in direct contact with epithelial cell surfaces. The basal lamina results from the self-assembly of laminin molecules with type IV collagen, entactin, and proteoglycans.

2. A reticular lamina—formed by collagen fibers—supports the basal lamina and is continuous with the connective tissue.

Figure 1-21 Basement membrane

The basal and reticular laminae can be distinguished by electron microscopy. Under the light microscope, the combined basal and reticular laminae receive the name of basement membrane, which can be recognized by the periodic acid–Schiff (PAS) stain (see Figure 1-21; see Box 1-D).

Box 1-D Periodic acid–Schiff (PAS) reaction

• PAS is a widely used histochemical technique to show 1,2-glycol or 1,2-aminoalcohol groups, such as those present in glycogen, mucus, and glycoproteins.

• Periodic acid, an oxidant, converts these groups to aldehydes. The Schiff reagent, a colorless fuchsin, reacts with the aldehydes to form a characteristic red-purple (magenta) product.

• Some important PAS-positive structures are the basement membrane, glycocalyx, mucus produced by goblet cells, stored glycoprotein hormones in cells of the pituitary gland, and collagens.

The basal lamina has specific functions in different tissues. The double basal lamina of the renal corpuscle constitutes the most important element of the glomerular filtration barrier during the initial step in the formation of urine (see Chapter 14, Urinary System).

In skeletal muscle, the basal lamina maintains the integrity of the tissue, and its disruption gives rise to muscular dystrophies (see Chapter 7, Muscle Tissue).

During the migration of primordial germinal cells, basal lamina components guide the migrating cells toward the gonadal ridge in preparation for the development of the gonads. The basal lamina not only provides support to epithelia, but also participates in other non–epithelial cell functions.

Laminin (Figure 1-22) is a cross-shaped protein consisting of three chains: the α chain, the β chain, and the γ chain. Laminin molecules can associate with each other to form a meshlike polymer. Laminin and type IV collagen are the major components of the basal lamina, and both are synthesized by epithelial cells resting on the lamina.

Figure 1-22 Laminin and fibronectin

Laminin has binding sites for nidogen (also called entactin), proteoglycans (in particular, heparan sulfate perlecan), α-dystroglycan (see Chapter 7, Muscle Tissue), and integrins.

Fibronectin (see Figure 1-22) consists of two protein chains cross-linked by disulfide bonds. Fibronectin is the main adhesion molecule of the extracellular matrix of the connective tissue and is produced by fibroblasts. Fibronectin has binding sites for heparin present in proteoglycans, several types of collagens (types I, II, III, and V), and fibrin (derived from fibrinogen during blood coagulation).

Fibronectin circulating in blood is synthesized in the liver by hepatocytes. It differs from fibronectin produced by fibroblasts in that it lacks one or two repeats (designated EDA and EDB for extra domain A and extra domain B) as a result of alternative mRNA splicing. Circulating fibronectin binds to fibrin, a component of the blood clot formed at the site of blood vessel damage. The RGD domain of immobilized fibronectin binds to integrin expressed on the surface of activated platelets, and the blood clot enlarges. We return to the topic of blood coagulation or hemostasis in Chapter 6, Blood and Hematopoiesis.

How cells interact with one another and with the basal lamina

Figure 1-23 summarizes the highlights of cell adhesion molecules and cell junctions. An epithelium is a continuous sheet of polarized cells supported by a basement membrane. The polarized nature of an epithelium depends on the tight junctions that separate the polarized cells into apical and basolateral regions. Tight junctions control the paracellular pathway of solutes, ions, and water. Tight junctions form a belt around the circumference of each cell.

Figure 1-23 Summary of cell junctions and cell adhesion molecules

Endothelial cells, the constituents of a simple squamous epithelium, are linked by tight and spot desmosomes tightly regulated to maintain the integrity of the endothelium and protect the vessels from unregulated permeability, inflammation, and reactions leading to blood coagulation in the lumen (see Chapter 12, Cardiovascular System). Leukocytes reach the site of infection by attaching to endothelial cell surfaces and migrate across the endothelium into the underlying tissues by a mechanism called diapedesis. Leukocytes find their way through endothelial cell-cell junctions after docking to activated or resting endothelial cells by the endothelial cell adhesion molecules ICAM-1 and VCAM-1 (see Figure 1-10). ICAM-1 and VCAM-1 bind to β₂ and β₁ integrin subunits in leukocytes (see Figure 1-12).

The cohesive nature of the epithelium depends on three factors: cell junctions, cell adhesive molecules in general, and the interaction of integrins with the extracellular matrix, produced to a large extent by fibroblasts. The basal lamina is essential for the differentiation of epithelial cells during embryogenesis.

Note in Figure 1-23 that:

1. The basal domain of epithelial cells interacts with the basal lamina through hemidesmosomes and integrins. Hemidesmosomes, so called because of their appearance as half-desmosomes in electron micrographs, are anchored to the basal lamina ouside the cell and to a network of keratin intermediate filaments inside the cell through a plate-plaque complex. Mutations in hemidesmosome components cause severe skin blistering as a result of a rupture of the anchoring molecular integrity.

2. Integrins interact directly with laminin and fibronectin, in particular the RGD domain to which integrins bind. Inside the cell, integrins interact with actin micro filaments. Integrins connect the extracellular environment to the intracellular space. We have seen that some ADAM proteins can use their disintegrin domain to prevent integrin binding to extracellular matrix ligands.

3. Collagens and proteoglycans do not interact directly with the basal domain of epithelial cells. Instead, this interaction is mediated by laminin and fibronectin, which contain specific binding sites for collagens, the proteoglycan perlecan, and nidogen.

4. The lateral domains of adjacent epithelial cells communicate by gap junctions (not shown in Figure 1-23). In contrast to tight junctions and belt and spot desmosomes, gap junctions are not anchoring devices. They consist of intercellular channels connecting the cytoplasm of adjacent cells. They are communicating junctions.

5. Cadherins and the afadin-nectin complex are present in tight junctions and zonula adherens. Actin microfilaments are associated with these two junctions.

CYTOSKELETON

Cytoskeleton is a three-dimensional network of proteins distributed throughout the cytoplasm of eukaryotic cells.

The cytoskeleton has roles in:

1. Cell movement (crawling of blood cells along blood-vessel walls, migration of fibroblasts during wound healing, and movement of cells during embryonic development)

2. Support and strength for the cell

3. Phagocytosis

4. Cytokinesis

5. Cell-cell and cell–extracellular matrix adherence

6. Changes in cell shape

The components of the cytoskeleton were originally identified by electron microscopy. These early studies described a system of cytoplasmic “cables” that fell into three size groups, as follows:

1. Microfilaments (7 nm thick)

2. Intermediate filaments (10 nm thick)

3. Microtubules (25 nm in diameter)

Biochemical studies, involving the extraction of cytoskeletal proteins from cells with detergents and salts and in vitro translation of specific mRNA, showed that each class of filaments has a unique protein organization. When cytoskeletal proteins were purified, they were used as antigens for the production of antibodies. Antibodies are used as tools for the localization of the various cytoskeletal proteins in the cell. The immunocytochemical localization of cytoskeletal proteins (Figure 1-24) and cell treatment with various chemical agents disrupting the normal organization of the cytoskeleton have been instrumental in understanding the organization and function of the cytoskeleton.

Figure 1-24 Immunocytochemistry

Microfilaments

The main component of microfilaments is actin. Actin filaments are composed of globular monomers (G-actin, 42 kd), which polymerize to form helical and asymmetrical filaments (F-actin).

Actin is a versatile and abundant cytoskeletal component forming static and contractile bundles and filamentous networks specified by actin-binding proteins and their distinctive location and function in a cell. F-actin bundles are present in the microvilli of the intestinal (Figure 1-25) and renal epithelial cells (brush border) and the stereocilia from the hair cells of the inner ear.

Figure 1-25 F-actin bundles form the core of intestinal microvilli

We have already seen that the intracellular portion of the cell adhesion molecules cadherins and integrin β₁ interacts with F-actin through linker proteins (see Figures 1-8 and 1-11). As discussed in Chapter 6, Blood and Hematopoiesis, actin—together with spectrin—forms a filamentous network on the inner face of the red blood cell membrane that is crucial for maintaining the shape and integrity of red blood cells. Spectrin is a tetramer consisting of two distinct polypeptide chains (α and β).

Growth of actin filaments may occur at both ends; however, one end (the “barbed end” or plus end) grows faster than the other end (the “pointed end” or minus end). The names correspond to the arrowhead appearance of myosin head bound at an angle to actin. Actin filaments can branch in the leading edge (lamellipodia) of cells involved in either motility or interaction with other cell types. F-actin branching is initiated from the side of a preexisting actin filament by Arp2/3 (for actin-related protein), an actin nucleating complex of seven proteins (Figure 1-26). Formin regulates the assembly of unbranched actin in cell protrusions such as the intestinal microvilli (see Figure 1-25).

Figure 1-26 Role of actin binding proteins in the assembly and disassembly of F-actin

Actin monomers have a binding site for adenosine triphosphate (ATP), which is hydrolyzed to adenosine diphosphate (ADP) as polymerization proceeds. Actin polymerization is ATP-dependent (see Box 1-E).

Box 1-E Microfilaments

• Microfilaments consist of G-actin, globular monomers, which polymerize in the presence of ATP into a long filamentous polymer, F-actin, that is 7 nm thick.

• F-actin has a distinct polarity: a barbed or polymerization end, and a pointed or depolymerizing end. Profilin has two roles: it severs F-actin and regulates F-actin assembly by catalyzing the exchange of G-actin bound ADP for ATP. Cofilin is a depolymerizing factor. The Arp2/3 complex initiates the branching of F-actin.

• Treadmilling, the dynamic balance between the polymerizing and depolymerizing ends of F-actin.

The kinetics of actin polymerization involves a mechanism known as treadmilling: G-actin monomers assembled at one end of the filament concurrently disassemble at the other end (see Figure 1-26). Four types of proteins control treadmilling (see Figure 1-26), as follows:

1. Thymosin sequesters pools of G-actin monomers within cells.

2. Profilin suppresses nucleation of G-actin and promotes F-actin growth at the barbed end. Profilin can favor the assembly of monomeric G-actin into filaments by facilitating the exchange of bound ADP for ATP. Only ATP-actin monomers can be assembled into filaments.

3. Cofilin (also known as actin depolymerizing factor) triggers depolymerization of ADP-bound actin at the pointed end. Similar to profilin and thymosin, cofilin forms a dimeric complex with G-actin.

4. Gelsolin has a dual role: it is a capping protein and prevents the loss and addition of actin monomers, and it is a severing protein. In the presence of Ca²⁺, gelsolin fragments actin filaments and remains bound to the barbed end, forming a cap that prevents further filament growth.

The assembly of G-actin monomers into filaments and the organization of these filaments into thick bundles are controlled by various types of actin-binding or actin-related proteins. A bundle of parallel nonbranching actin filaments, forming the core of the microvillus, is held together by actin-linking proteins, villin and fimbrin. Side arms of myosin-I and the Ca²⁺-binding protein calmodulin anchor the bundle to the plasma membrane (see Figure 1-25).

Arp2/3 and additional regulatory proteins form a nucleation complex for the assembly of branching actin filaments. Branching actin filaments assemble at the leading edge of a cell during cell motility. In the microvillus, formins (proteins with highly conserved formin-homology domains, FH1 and FH2), instead of the Arp2/3 complex, seem to regulate the elongation of nonbranching actin filaments, while remaining attached to the barbed end (see Box 1-E). Formins are located at the tip of the microvillus, the cap region (see Figure 1-25).

Male patients with defects in proteins that activate the Arp2/3 complex—in particular a protein of the Wiskott-Aldrich syndrome protein (WASP) family—display recurrent respiratory infections because of hereditary immunodeficiency, thrombocytopenia (low platelet count) present from birth on and eczema of the skin after the first month of life (see Box 1-F). The mutation is inherited from the mother, a healthy carrier of the defective gene.

Box 1-F Wiskott-Aldrich syndrome

• The Arp2/3 complex is necessary for nucleating the assembly of branched networks of actin filaments. The function of phagocytic cells and platelets depends on a functional actin cytoskeleton.

• Numerous proteins activate the Arp2/3 complex. Without these proteins, the Arp2/3 complex is inactive.

• Two major proteins that bind and activate the Arp2/3 complex include the Wiskott-Aldrich syndrome protein (WASP) family, which consists of several members (WASP, neuronal WASP [N-WASP] and SCAR/WAVE1-3 (suppressor of cAMP receptor/WASP family verprolin-homologous protein 1-3). Additional members belong to the cortactin family, which includes cortactin and hematopoietic-specific protein.

• Mutations in the WASP gene, present in the X chromosome, are characterized by recurrent respiratory infections (defective function of T and B cells), a reduction in the number of platelets (thrombocytopenia) leading to increased susceptibility to bleeding, and eczema of the skin. Males but not females are affected by the Wiskott-Aldrich syndrome.

Microvilli and stereocilia are comparable structures, although they differ in length and the number of actin filaments: intestinal microvilli are 1 to 2 μm long, 0.1 μm wide, and consist of 20 to 30 bundled actin filaments; stereocilia in hair cells of the inner ear have a tapered shape at their base, the length range is 1.5 to 5.5 μm, and each actin bundle contains up to 900 actin filaments. Hair cells are extremely sensitive to mechanical displacement, and a slight movement of the stereocilium is amplified into changes in electric potential transmitted to the brain. We study hair cells of the inner ear in Chapter 9, Sensory Organs: Vision and Hearing.

Microtubules

Microtubules are composed of tubulin dimers (Figure 1-27; see Box 1-G). Each tubulin dimer consists of two tightly bound tubulin molecules: α-tubulin and β-tubulin. Tubulin subunits are arranged in longitudinal rows called protofilaments. Thirteen protofilaments associate side by side with each other to form a cylinder of microtubules with a hollow core. The diameter of a microtubule is 25 nm.

Figure 1-27 Assembly of a microtubule

Box 1-G Microtubules

• Microtubules consist of tubulin dimers, α and β, which polymerize in the presence of GTP into longitudinal rows of protofilaments. Thirteen protofilaments form a cylinder or microtubule 25 nm in diameter.

• Microtubules have a distinct polarity: a plus or polymerizing end and a minus or depolymerizing end.

• Microtubules undergo alternate phases of slow growth and rapid depolymerization, a process known as dynamic instability.

• Centrioles, basal bodies, and axonemes of cilia and flagella contain a precise array of microtubules.

• Kinesin and cytoplasmic dynein, two molecular motor proteins, use microtubules as tracks for the transport of vesicle and nonvesicle cargos.

In contrast to actin filaments, most individual microtubules seem to undergo alternate phases of slow growth and rapid depolymerization. This process, called dynamic instability, consists of three major steps: (1) a polymerization phase, in which GTP-tubulin subunits add to the plus end of the microtubule and a GTP cap is assembled to facilitate further growth; (2) release of hydrolyzed phosphate (Pi) from tubulin-bound GTP; and (3) a depolymerization phase, in which GDP-tubulin subunits are released from the minus end at a fast rate. The polymerization-to-depolymerization transition frequency is known as catastrophe; the depolymerization-to-polymerization transition frequency is known as rescue.

The stability of microtubules can be modified by microtubule-associated proteins (MAPs). MAPs are classified into two groups: (1) classical MAPs, such as MAP1A, MAP1B, MAP2, and tau, and (2) nonclassical MAPs, including Lis1 and DCX family members. MAPs stabilize microtubules by phosphorylation/dephosphorylation. In Chapter 7, Nervous Tissue, we discuss the significance of tau phosphorylation and dephosphorylation in Alzheimer’s disease. A lack of expression of Lis1 causes a sever brain developmental disorder called lissencephaly.

Centrosome: A microtubule-organizing center

The centrosome has three major functions: (1) it nucleates the polymerization of tubulin subunits into microtubules, (2) it organizes microtubules into functional units, and (3) it duplicates once every cell cycle.

Centrosomes consist of a pair of centrioles surrounded by pericentriolar material, an amorphous, electron-dense substance rich in proteins such as pericentrin and γ-tubulin.

Centrosomes are part of the mitotic center, which, together with the mitotic spindle, constitutes the mitotic (or meiotic) apparatus (Figure 1-28). A centriole is a small cylinder (0.2 μm wide and 0.4 μm long) composed of nine microtubule triplets in a helicoid array. In contrast to most cytoplasmic microtubules, which display dynamic instability, the centriolar microtubules are very stable.

Figure 1-28 Mitotic apparatus

During interphase, centrioles are oriented at right angles to each other. Before mitosis, centrioles replicate and form two pairs. During mitosis, each pair can be found at opposite poles of the cell, where they direct the formation of the mitotic or meiotic spindle.

There are three types of microtubules extending from the centrosomes: radiating or astral microtubules, anchoring each centrosome to the plasma membrane; kinetochore microtubules, attaching the chromosome-associated kinetochore to the centrosomes; and polar microtubules, extending from the two poles of the spindle where opposite centrosomes are located (Figure 1-28). If kinetochores fail to assemble, chromosomes cannot segregate properly (see Box 1-H).

Box 1-H Differences between centromeres and kinetochores

• The terms centromere and kinetochore are often used synonymously, but do not mean the same thing.

• The centromere (not the centrosome) is the chromosomal site associated with microtubules of the spindle. Centromeres can be recognized cytologically as a narrow chromatin region on metaphase chromosomes known as primary constriction where centromeric DNA is present.

• The kinetochore consists of proteins assembled on the centromeric chromatin on sister chromatids. The assembly of the kinetochore depends exclusively on the presence of centromeric DNA sequences. The centromere and the kinetochore mediate attachment of the kinetochore microtubules of the spindle.

The pericentriolar material contains the γ-tubulin ring complex and numerous proteins, including pericentrin. Each γ-tubulin ring complex is the nucleation site or template for the assembly and growth of one microtubule. The centrioles do not have a direct role in the nucleation of microtubules in the centrosome. Tubulin dimers associate to the γ-tubulin ring by the α-tubulin subunit. Consequently, the minus end of each microtubule points to the centrosome; the plus end, the growing end, is oriented outward, free in the cytoplasm.

Microtubules in cilia and flagella

Centrioles give rise to structurally similar basal bodies, which are the outgrowth origin of cilia (see Figure 1-5) and flagella. A defect in the assembly of the basal body and cilia, caused by abnormal transport of ciliary proteins, results in the Bardet-Biedl syndrome (see Box 1-I). Cilia and flagella are motile cytoplasmic extensions containing a core of microtubules called the axoneme (Figure 1-29). The axoneme consists of nine peripheral microtubule doublets surrounding a central pair of microtubules. This arrangement is known as the 9 + 2 configuration (see Box 1-J).

Box 1-I Bardet-Biedl syndrome

• Bardet-Biedl syndrome (BBS) is a pleiotropic (multisystemic) disorder that includes age-related retinal dystrophy, obesity, polydactyly, renal dysplasia, reproductive tract abnormalities, and learning disabilities.

• BBS is a disorder of basal bodies and cilia resulting from a defective microtubule-based transport function (intraciliary transport) required for the assembly, maintenance, and function of basal bodies, cilia, and flagella (intraflagellar transport).

• Eight BBS genes (BBS1-8) have been identified. The degree of clinical variability in BBS is not fully explained.

Figure 1-29 Axoneme

Box 1-J Major components of the ciliary and flagellar axonemes

• Microtubules: Major component of the axoneme. Motor proteins use microtubules of the axoneme as tracks for intraciliary or intraflagellar cargo transport. Microtubule-based axonal transport also depends on motor proteins.

• Tektins: Intermediate filament-like proteins extending along the length of axonemal microtubules and, presumably, adding mechanical strength to the axoneme.

• Dynein arms: ATPase responsible for ciliary and flagellar movement. The heads are in contact with the adjacent outer microtubules at a periodic distance and move along them.

• Nexin links: A beltlike arrangement stabilizing the nine outer concentric pairs of microtubules.

• Radial spokes: Project from each of the nine outer microtubule doublets to the inner sheath surrounding the central pair.

• Inner sheath: A structure surrounding the central pair of microtubules, in contact with the globular end of the radial spokes.

Each peripheral doublet consists of a complete microtubule (called an A tubule, with 13 protofilaments), sharing its wall with a second, partially completed microtubule (called a B tubule, with 10 to 11 protofilaments). Extending inward from the A tubule are radial spokes that insert into an amorphous inner sheath surrounding the central microtubule pair. Adjacent peripheral doublets are linked by the protein nexin.

Projecting from the sides of the A tubule are sets of protein arms: the inner and outer arms of dynein, a microtubule-associated adenosine triphosphatase (ATPase). In the presence of ATP, the sliding of peripheral doublets relative to each other bends cilia and flagella. Sliding and bending of microtubules are the basic events of their motility.

Clinical significance: Microtubule-targeted drugs and sterility

Two groups of antimitotic drugs act on microtubules: microtubule-destabilizing agents, which inhibit microtubule polymerization, and microtubule-stabilizing agents, which affect microtubule function by suppressing dynamic instability.

The first group includes colchicine, colcemid, vincristine, and vinblastine, which bind to tubulin and inhibit microtubule polymerization, blocking mitosis. Colchicine is used clinically in the treatment of gout. Vincristine and vinblastine, from Vinca alkaloids isolated from the leaves of the periwinkle plant, have been successfully used in childhood hematologic malignancies (leukemias). Neurotoxicity—resulting from the disruption of the microtubule-dependent axonal flow (loss of microtubules and binding of motor proteins to microtubules)–and myelosuppression are two side effects of microtubule-targeted drugs.

The second group includes taxol (isolated from the bark of the yew tree) with an opposite effect: It stabilizes microtubules instead of inhibiting their assembly (Figure 1-30). Paclitaxel (taxol) has been used widely to treat breast and ovarian cancers. Similar to Vinca alkaloids, its main side effects are neurotoxicity and suppression of hematopoiesis.

Figure 1-30 Agents that prevent microtubular function

Kartagener’s syndrome is an autosomal recessive disorder frequently associated with bronchiectasis (permanent dilation of bronchi and bronchioles) and sterility in men.

Kartagener’s syndrome is the result of structural abnormalities in the axoneme (defective or absent dynein) that prevent mucociliary clearance in the airways (leading to persistent infections) and reduce sperm motility and egg transport in the oviduct (leading to sterility).

Microtubules–cytoskeletal tracks for cargo transport powered by motor proteins

The transport of vesicles and nonvesicle cargos occurs along microtubules and F-actin. Specific molecular motors associate to microtubules and F-actin to mobilize cargos to specific intracellular sites. Microtubule-based molecular motors include kinesin and cytoplasmic dynein for the long-range transport of cargos. F-actin–based molecular motors include unconventional myosin Va and VIla for the short-range transport of cargos. We discuss additional aspects of the F-actin–based cargo transport mechanism during the transport of melanosomes in Chapter 11, Integumentary System.

Three examples of microtubule-based cargo transport in mammalian systems are as follows (see Box 1-K):

1. Axonemal transport, including flagella (intraflagellar transport) and cilia (intraciliary transport) (Figure 1-31). During axonemal transport, particles are mobilized by kinesin and cytoplasmic dynein along the microtubule doublets of the axoneme. Defective axonemal transport results in the abnormal assembly of cilia and flagella, including polycystic kidney disease, retinal degeneration, respiratory ciliary dysfunction, and lack of sperm tail development. As indicated before (see Box 1-H), the Bardet-Biedl syndrome is a disorder caused by basal body/ciliary dysfunction secondary to a defective microtubule-based transport function.

2. Axonal transport, along the axon of neurons (see Figure 1-31).

3. Intramanchette transport, along microtubules of a transient microtubular structure, the manchette, assembled during the elongation of the spermatid head (see Chapter 20, Spermatogenesis).

Box 1-K Microtubule-based cargo transport by molecular motors

• Microtubules participate in the intracellular traffic of vesicles and nonvesicle materials or cargos.

• Molecular motor proteins, such as kinesin and cytoplasmic dynein, mediate the long-range transport of cargos, whereas short-range transport occurs on actin filaments.

• There are three major microtubule-based transport systems: (1) axonemal transport, including intraciliary transport and intraflagellar transport; (2) axonal transport; and (3) intramanchette transport.

• Axonemal transport is essential for the delivery of tubulin dimers and other molecules to the distal polymerization end of microtubules of cilia and flagella. Axonemes originate from basal bodies, centriole-derived, microtubule-containing structures.

• Axonal transport is crucial for the traffic of neurotransmitter-containing vesicles to neuronal synapses.

• Intramanchette transport is required for morphogenetic events during sperm development. The manchette is a transient microtubule-containing structure associated to a perinuclear ring that assembles during the elongation of the spermatid head and then disassembles.

Figure 1-31 Intraciliary and axonal cargo transport

Axonal transport

Axons are cytoplasmic extensions of neurons responsible for the conduction of neuronal impulses. Membrane-bound vesicles containing neurotransmitters produced in the cell body of the neuron travel to the terminal portion of the axon, where the content of the vesicle is released at the synapse.

Bundles of microtubules form tracks within the axon to carry these vesicles. Vesicles are transported by two motor proteins (see Figure 1-31):

1. Kinesin

2. Cytoplasmic dynein

Kinesins and cytoplasmic dyneins participate in two types of intracellular transport movements:

1. Saltatory movement, defined by the continuous and random movement of mitochondria and vesicles.

2. Axonal transport, a more direct intracellular movement of membrane-bound structures.

Kinesins and cytoplasmic dyneins have two ATP-binding heads and a tail. Energy derives from continuous ATP hydrolysis by ATPases present in the heads. The head domains interact with microtubules, and the tail binds to specific receptor binding sites on the surface of vesicles and organelles.

Kinesin uses energy from ATP hydrolysis to move vesicles from the cell body of the neuron toward the end portion of the axon (anterograde transport). Cytoplasmic dynein also uses ATP as an energy source to move vesicles in the opposite direction (retrograde transport).

Myosin family associates with F-actin to form contractile structures

Members of the myosin family of proteins bind and hydrolyze ATP to provide energy for their movement along actin filaments from the pointed (minus) end to the barbed (plus) end. Myosin I and myosin II are the predominant members of the myosin family (Figure 1-32; see Box 1-L).

Figure 1-32 Classes of myosin molecules and how they work

Box 1-L Types of myosins

• Myosins are members of a large family of motor proteins that generate movement along actin filaments using energy from the hydrolysis of ATP.

• Two groups of myosins exist: conventional myosin, myosin II, which drives muscle contraction and contractile processes in nonmuscle cells, and unconventional (nonmuscle) myosins—myosin I and myosin V among others—involved in the movement of vesicle cargos inside cells.

• Myosin II consists of two polypeptides, each displaying a globular head attached to a tail coiled around the tail of its partner. The tails can self-assemble into bipolar filaments. Each head, which also contains a light chain, has an actin-binding site with ATPase activity stimulated by actin binding and regulated by the light chain.

• Myosin I is single-headed and has a tail shorter than myosin II. Myosin II is involved in vesicle transport along F-actin.

• Myosin V is double-headed with coiled double tails. The heads contain ATP and actin binding sites. The distal end of the tails is recruited by vesicles. The recruitment is mediated by the vesicle receptor Rab27a.

• Myosin V-Rab27a interaction plays a role in the transfer of melanosomes from melanocytes to keratinocytes. Defective transfer of melanosome from melanocytes to keratinocytes of the hair shaft by a mutation of Rab27a or myosin Va genes is the cause of Griscelli syndrome type I and II. Patients with Griscelli syndrome have silvery hair, partial albinism, occasional neurologic defects, and immunodeficiency.

Myosin I, regarded as an unconventional myosin, is found in all cell types and has only one head domain and a tail. The head is associated with a single light chain. The head interacts with actin filaments and contains ATPase, which enables myosin I to move along the filaments by binding, detaching, and rebinding. The tail binds to vesicles or organelles. When myosin I moves along an actin filament, the vesicle or organelle is transported. Myosin I molecules are smaller than myosin II molecules, lack a long tail, and do not form dimers.

Myosin II, a conventional myosin, is present in muscle and nonmuscle cells. Myosin II consists of a pair of identical molecules. Each molecule consists of an ATPase-containing head domain and a long rodlike tail. The tails of the dimer link to each other along their entire length to form a two-stranded coiled rod. The tail of myosin II self-assembles into dimers, tetramers, and a bipolar filament with the heads pointing away from the midline.

The two heads—linked together but pointing in opposite directions—bind to adjacent actin filaments of opposite polarity. Each myosin head bound to F-actin moves toward the barbed (positive) end. Consequently, the two actin filaments are moved against each other, and contraction occurs (see Figure 1-32).

Heads and tails of myosin II can be cleaved by enzymes (trypsin or papain) into light meromyosin (LMM) and heavy meromyosin (HMM). LMM forms filaments, but lacks ATPase activity and does not bind to actin. HMM binds to actin, is capable of ATP hydrolysis, and does not form filaments. HMM is responsible for generating force during muscle contraction. HMM can be cleaved further into two subfragments called S1. Each S1 fragment contains ATPase and light chains and binds actin.

Myosin V, an unconventional myosin, is double-headed with a coiled double tail. The head region binds to F-actin; the distal globular ends of the tails bind to Rab27a, a receptor on vesicle membranes. Myosin Va mediates vesicular transport along F-actin tracks. A specific example is the transport of melanosomes from melanocytes to keratinocytes, first along microtubules and later along F-actin.

Mutations in the Rab27a and myosin Va genes disrupt the F-actin transport of melanosomes. An example in humans is Griscelli syndrome, a rare autosomal recessive disorder characterized by pigment dilution of the hair caused by defects in melanosome transport and associated with disrupted T cell cytotoxic activity and neurologic complications.

Figure 1-33 summarizes the structural and functional characteristics of motor proteins.

Figure 1-33 Comparison of motor proteins

Light-chain phosphorylation by myosin light-chain kinase

The self-assembly of myosin II and interaction with actin filaments in nonmuscle cells takes place in certain sites according to functional needs. These events are controlled by the enzyme myosin light-chain kinase (MLCK), which phosphorylates one of the myosin light chains (called the regulatory light chain) present on the myosin head. The activity of MLCK is regulated by the Ca²⁺-binding protein calmodulin (Figure 1-34).

Figure 1-34 Light-chain phosphorylation of myosin II in nonmuscle cells

MLCK has a catalytic domain and a regulatory domain. When calmodulin and Ca²⁺ bind to the regulatory domain, the catalytic activity of the kinase is released. The MLCK–calmodulin–Ca²⁺ complex catalyzes the transfer of a phosphate group from ATP to the myosin light chain, and myosin cycles along F-actin to generate force and muscle contraction.

Phosphorylation of one of the myosin light chains results in two effects:

1. It exposes the actin-binding site on the myosin head. This step is essential for an interaction of the myosin head with the F-actin bundle.

2. It releases the myosin tail from its sticky attachment site near the myosin head. This step also is critical because only myosin II stretched tails can self-assemble and generate bipolar filaments, a requirement for muscle contraction (see Figure 1-33).

In smooth muscle cells, a phosphatase removes the phosphate group from myosin light chains. Skeletal muscle contraction does not require phosphorylation of the myosin light chains. We discuss additional details of muscle contraction when we study the muscle tissue (see Chapter 7, Muscle Tissue).

Intermediate filaments

Intermediate filaments (Figure 1-35) represent a heterogeneous group of structures so named because their diameter (10 nm) is intermediate between those of micro tubules (25 nm) and microfilaments (7 nm). Intermediate filaments are the most stable cytoskeletal structures.

Figure 1-35 Fine structure of the major components of the cytoskeleton

Detergent and salt treatments extract microfilament and microtubule components and leave intermediate filaments insoluble. All intermediate filaments have a common monomer consisting of a central α-helical rod flanked by head and tail domains (Figure 1-36).

Figure 1-36 Assembly of an intermediate filament

The structure of the intermediate filament does not fluctuate between assembly and disassembly states similar to microtubules and microfilaments. In contrast to actin and tubulin, the assembly and disassembly of intermediate filament monomers are regulated by phosphorylation.

Intermediate filament protein monomers consist of three domains (see Figure 1-36): A central α-helical rod domain is flanked by a nonhelical N-terminal head domain and a C-terminal tail domain. During assembly, pairs of dimers—formed by the parallel alignment of monomers—associate into tetramers in a side-by-side but antiparallel orientation. About eight tetramers align end-to-end to form a protofilament. Pairs of protofilaments associate laterally to form a protofibril, and four protofibrils—a total of eight protofilaments—wind up to form a ropelike intermediate filament (see Figure 1-36). Intermediate filaments do not have the structural polarity seen in F-actin and microtubules. One end of an intermediate filament cannot be distinguished from another. Molecular motors associated to an intermediate filament would find it difficult to identify one direction from another.

The major function of intermediate filaments is to provide mechanical support for the cell. Five major types of intermediate filament proteins have been identified on the basis of sequence similarities in the rod domain. They are referred to as types I through V (see Box 1-M). About 50 intermediate filament proteins have been reported so far.

Box 1-M Types of intermediate filament proteins

• Type I (acidic) and type II (basic) Keratins (40-70 kd): Keratins assemble as type I and type II heteropolymers. Different keratin types are coexpressed in epithelial cells, hair, and nails. Keratin gene mutations occur in several skin diseases (blistering and epidermolysis diseases).

• Type III (can self-assemble as homopolymers) Vimentin (54 kd): Present in mesenchymal-de-rived cells.

Desmin (53 kd): A component of Z disks of striated muscle and smooth muscle cells.

Glial fibrillary acidic protein (GFAP 51 kd): Present in astrocytes.

Peripherin (57 kd): A component of axons in the peripheral nervous system.

• Type IV

Neurofilaments (NF): Three forms coexpressed and forming heteropolymers in neurons: NF-L (light, 60 to 70 kd), NF-M (medium, 105 to 110 kd) and NF-H (heavy, 135 to 150 kd).

α-lnternexin (66 kd): A component of developing neurons.

• Type V

Lamin A and lamin B (60 to 70 kd, 63-68 kd): Present in the nuclear lamina associated to the inner layer of the nuclear envelope. Maintain the integrity of the nuclear envelope. A group of human diseases—laminopathies—is associated with lamin A gene (LMNA) mutations (see Box 1-N).

Type I (acidic keratins) and type II (neutral to basic keratins). This class of proteins forms the intermediate filament cytoskeleton of epithelial cells (called cytokeratins to distinguish them from the keratins of hair and nails). Equal amounts of acidic (40 to 60 kd) and neutral-basic (50 to 70 kd) cytokeratins combine to form this type of intermediate filament protein. Type I and type II intermediate filament keratins form tonofilaments associated with molecules present in the cytoplasmic plaques of desmosomes and hemidesmosomes (see Figures 1-18 and 1-19). We come back to intermediate filament–binding proteins, such as filaggrins, when we discuss the differentiation of keratinocytes in the epidermis of the skin (Chapter 11, Integumentary System), and plectin, when we analyze the cytoskeletal protective network of skeletal muscle cells (Chapter 7, Muscle Tissue).

In the epidermis of the skin, the basal cells express keratins K5 and K14. The upper differentiating cells express keratins K1 and K10. In some regions of the epidermis, such as in the palmoplantar region, keratin K9 is found. Mutations in K5 and K14 cause hereditary blistering skin diseases belonging to the clinical type epidermolysis bullosa simplex (see later, Clinical significance: Intermediate filaments and blistering diseases).

Type III. This group includes the following intermediate filament proteins:

Vimentin (54 kd) is generally found in cells of mesenchymal origin. In some cells, vimentin establishes a structural link between the plasma membrane and nuclear lamins.

Desmin (53 kd) is a component of skeletal muscle cells and is localized to the Z disk of the sarcomere (see Chapter 7, Muscle Tissue). This intermediate filament protein keeps individual contractile elements of the sarcomeres attached to the Z disk and plays a role in coordinating muscle cell contraction. Desmin is also found in smooth muscle cells.

Glial fibrillary acidic protein (GFAP) (51 kd) is observed in astrocytes and some Schwann cells (see Chapter 8, Nervous Tissue).

Peripherin (57 kd) is a component of neurons of the peripheral nervous system and is coexpressed with neurofilament proteins (see Chapter 8, Nervous Tissue).

Type IV. Neurofilaments are the main components.

Neurofilaments (NFs) are found in axons and dendrites of neurons. Three types of proteins can be found in a neurofilament: NF-L (60 to 70 kd), NF-M (105 to 110 kd), and NF-H (135 to 150 kd), for low-molecular-weight, middle-molecular-weight, and high-molecular-weight neurofilaments.

α-Internexin (66 kd) is found predominantly in the central nervous system (particularly in the spinal cord and optic nerve).

Type V. Proteins of this group, the nuclear lamins, are encoded by three genes: LMNA, LMNB1, and LMNB2. Lamin A and lamin C arise from the alternative splicing of transcripts encoded by the LMNA gene. The LMNB1 gene encodes lamin B1 expressed in all somatic cells. The LMNB2 gene encodes lamin B2, expressed in all somatic cells, and lamin B3, which is specific for spermatogenic cells.

Nuclear lamins (60 to 75 kd) differ from the other intermediate filament proteins in that they organize an orthogonal meshwork—the nuclear lamina—in association with the inner membrane of the nuclear envelope. Lamins provide mechanical support for the nuclear envelope and bind chromatin. Because of their clinical relevance, we come back to nuclear lamins and associated proteins when we discuss the organization of the nuclear envelope.

A group of human diseases, known as laminopathies, are linked to defects in proteins of the nuclear envelope, including lamins (see Box 1-N). Numerous laminopathies affect cardiac and skeletal muscle, adipose tissue (lipodystrophies), and motor and sensory peripheral nerves.

Box 1-N Clinical features of laminopathies

• Classified into three distinct categories: muscular dystrophy, partial lipodystrophy, and neuropathy. Caused by lamin A or C mutations affecting skeletal and cardiac muscle and fat distribution.

• Emery-Dreifuss muscular dystrophy (phenotype inherited by autosomal dominant and recessive and X-linked mechanisms—the latter caused by mutations in emerin gene): Achilles tendon contractures, slow and progressive muscle weakness and wasting, cardiomyopathy with conduction defects.

• Limb girdle muscular dystrophy: Progressive muscle weakness of hip girdle and proximal arm and muscle of the leg. Dilated cardiomyopathy.

• Charcot-Marie-Tooth disorder type 2B1: Motor and sensory deficit neuropathy distal in the upper limbs and proximal and distal in the lower limbs.

Note: X-linked Charcot-Marie-Tooth type 1 disease also displays motor and sensory neuropathies of the peripheral nervous system, but is caused by a mutation in the connexin32 (Cx32) gene expressed in Schwann cells. It affects myelin.

• Dunnigan-type familial partial lipodystrophy: Becomes evident at puberty with a loss of subcutaneous fat from the trunk and limbs and accumulation of fat in the face and neck.

Two hypotheses concerning the pathogenic mechanism of laminopathies have been considered:

1. The gene expression hypothesis regards lamin A and lamin C as essential for the correct tissue-specific expression of certain genes

2. The mechanical stress hypothesis proposes that a defect in lamin A and lamin C weakens the structural integrity of the nuclear envelope.

During mitosis, the phosphorylation of lamin serine residues causes a transient disassembly of the meshwork, followed by a breakdown of the nuclear envelope into small fragments. At the end of mitosis, lamins are dephosphorylated, and the lamin meshwork and the nuclear envelope reorganize. See the cell nucleus section concerning the mechanism of phosphorylation and dephosphorylation of lamins during the cell cycle.

Hemidesmosomes and intermediate filaments

Hemidesmosomes are specialized junctions observed in basal cells of the stratified squamous epithelium attaching to the basement membrane (Figure 1-37). Inside the cell, the proteins BPAG1 (for bullous pemphigoid antigen 1) and plectin (members of the plakin family of cross-linker proteins) are associated to intermediate filaments (also called tonofilaments). Plectin connects intermediate filaments to the integrin subunit β₄.

Figure 1-37 Structure and composition of a hemidesmosome

On the extracellular side, integrin α₆β₄, BPAG2 (for bullous pemphigoid antigen 2) and laminin 5, a protein present in specialized structures called anchoring filaments, link hemidesmosomes to the basal lamina. The plakin-related protein BPAG1 associates to BPAG2, a transmembrane protein with an extracellular collagenous domain. Putting all things together, BPAG1 constitutes a bridge between the transmembrane protein BPAG2 and intermediate filaments. If this bridge is disrupted, as in bullous pemphigoid, the epidermis becomes detached from the basal lamina anchoring sites. BPAG1 and BPAG2 were discovered in patients with bullous pemphigoid, an autoimmune disease.

Clinical significance: Intermediate filaments and blistering diseases

Bullous pemphigoid is an autoimmune blistering disease similar to pemphigus vulgaris (called “pemphigoid”). Blisters or bullae develop at the epidermis-dermis junction when circulating immunoglobulin G (IgG) cross-reacts with bullous pemphigoid antigen 1 or 2. IgG-antigen complexes lead to the formation of complement complexes (C3, C5b, and C9), which damage the attachment of hemidesmosomes and perturb the synthesis of anchoring proteins by basal cells (Figure 1-38).

Figure 1-38 Pathogenesis of bullous pemphigoid, an autoimmune disease

The production of local toxins causes the degranulation of mast cells and release of chemotactic factors attracting eosinophils. Enzymes released by eosinophils cause blisters or bullae.

Intermediate filaments strengthen the cellular cytoskeleton. The expression of mutant keratin genes results in the abnormal assembly of keratin filaments, which weakens the mechanical strength of cells and causes inherited skin diseases, as shown in Figure 1-39:

1. Epidermolysis bullosa simplex (EBS), characterized by skin blisters after minor trauma. EBS is determined by keratin 5 and 14 mutant genes.

2. Epidermolytic hyperkeratosis (EH), in which patients have excessive keratinization of the epidermis owing to mutations of keratin 1 and 10 genes.

3. Epidermolytic plantopalmar keratoderma (EPPK), a skin disease producing fragmentation of the epidermis of the palms and soles, caused by a mutation of the keratin 9 gene.

Figure 1-39 Examples of skin diseases caused by mutated intermediate filament keratins

CELL NUCLEUS

Nuclear envelope and nuclear pore complex

The mammalian cell nucleus consists of three major components: (1) the nuclear envelope, (2) chromatin, and (3) the nucleolus. The nuclear envelope consists of two concentric membranes separated by a perinuclear space. The inner nuclear membrane is associated with the nuclear lamina (see Box 1-O), chromatin, and ribonucleoproteins. The outer nuclear membrane is continuous with the membranes of the endoplasmic reticulum and can be associated with ribosomes.

Box 1-O Nuclear lamina

• Lamins, type V intermediate filament proteins, are the main components of the nuclear lamina.

• Lamins bind to proteins of the inner nuclear membrane, including emerin (with eight transmembrane spans), lamin B receptor, lamina-associated polypeptides 1 and 2β, and nesprin 1α, a protein with several spectrin-like repeats that binds lamin A and emerin (see Figure 1-40).

• Lamins and their associated proteins have roles in chromatin organization, spacing of nuclear pore complexes, and reassembly of the nucleus after cell division.

• Mutations of lamins and lamin-binding proteins cause various diseases (called laminopathies) (see Box 1-L). Hutchinson-Gilford progeria syndrome (premature aging) is caused by a mutation in lamin A.

The nuclear pore complex has a tripartite structure, composed of a central cylindrical body placed between inner and outer octagonal rings, each consisting of eight protein particles. The central cylinder consists of a central plug and eight radiating spokes (Figure 1-40). The exact role of individual nuclear pore complex proteins in nucleocytoplasmic trafficking is unclear.

Figure 1-40 Nuclear envelope and nuclear pore complex

Nuclear pore complexes embedded in the nuclear envelope establish bidirectional communication gates for the trafficking of macromolecules between the cytoplasm and the nucleus. Small molecules (less than 40 to 60 kd) can diffuse through the nuclear pore complex. Proteins of any size, containing a nuclear localization amino acid sequence (NLS, Pro-Lys-Lys-Lys-Arg-Lys-Val), can be imported into the nucleus, however, by an energy-dependent mechanism (requiring ATP and GTP).

Nucleocytoplasmic transport: Ran-GTPase

Protein nuclear import/export is controlled by Ran (for Ras-like nuclear GTPase), a small GTPase of the Ras superfamily that dictates the directionality of nucleocytoplasmic transport.

Ran shuttles across the nuclear pores and accumulates inside the nucleus by an active transport mechanism (Figure 1-41).

Figure 1-41 Ran GTPase directs nucleocytoplasmic transport

1. In the nucleus, a high concentration of Ran-GTP is achieved by RCC1, a GDP-GTP exchanger protein bound to chromatin. Ran-GTP determines the dissociation of imported proteins containing NLS by binding to importin β, the transporter receptor protein.

2. In the opposite direction, from the nucleus to the cytoplasm, binding of Ran-GTP to the carrier protein exportin/Crm1 facilitates the assembly of complexes containing proteins with nuclear export sequence (NES).

3. In the cytoplasm, Ran-GTP is converted to Ran-GDP by Ran-GTPase, which is activated by two cooperating proteins: Ran-GAP (Ran-GTPase-activating protein) and RanBP (Ran-GTP binding protein). Consequently, the exported protein is dissociated from its transporter receptor protein exportin/Crm1 and Ran-GTP. Importin and exportins are recycled by transport back across the nuclear pore complex.

Ran-GTPase also has a role in the assembly of the mitotic spindle.

Chromatin

Chromatin is defined as particles or “beads” (called nucleosomes) on a double-stranded DNA string (Figure 1-42). Each nucleosome consists of a histone octamer core and about two turns of DNA wound around the histone core. The histone octamer contains two molecules each of H2A, H2B, H3, and H4 histones. H1 histone cross-links the DNA molecule wrapped around the octamer.

Figure 1-42 Structure of the chromatin fiber: the nucleosome

Chromatin is packed in separate chromosomes that can be visualized during mitosis (or meiosis). During interphase (phases G₁, S, and G₂ of the cell cycle), individual chromosomes cannot be identified as such, but are present in a diffuse or noncondensed state.

Diffuse chromatin, called euchromatin (“good chromatin”), is transcriptionally (RNA synthesis) active and represents about 10% of total chromatin. Euchromatin is the site of synthesis on nonribosomal RNAs, including mRNA and transfer RNA (tRNA) precursors. Condensed chromatin, called heterochromatin (“different chromatin”), is transcriptionally inactive and represents about 90% of total chromatin (Figure 1-43).

Figure 1-43 X chromosome inactivation

Dosage compensation: Inactivation of one of the X chromosomes

The random inactivation of one of the two X chromosomes in every female somatic cell is known as dosage compensation. Both X chromosomes in oocytes remain active. The inactivation is random because either the paternal or the maternal X chromosome is inactivated. The choice remains nonrandom for all subsequent cell descendants. The transcriptional inactivation of one of the two X chromosomes is observed in the trophoblast on day 12 after fertilization and on day 16 in the embryo.

In humans, the inactivated X chromosome is recognized by the presence of the Barr body, a heterochromatin mass observed adjacent to the nuclear envelope or in the form of a drumstick in polymorphonuclear leukocytes (see Figure 1-43). If a cell has more than two X chromosomes, the extra X chromosomes are inactivated, and more than one Barr body is visualized.

Nucleolus

The nucleolus is the site of synthesis of ribosomal RNA (rRNA) and assembly or ribosomal subunits. The nucleolus houses several proteins, including fibrillarin and nucleolin, required for pre-rRNA processing. In addition, the nucleolus contains nucleostemin, a protein unrelated to ribosomal biogenesis. Nucleolin and nucleostemin are shuttling proteins; they relocalize from the nucleolus to the nucleoplasm where they interact with protein p53, a protector of DNA damage by preventing DNA replication in response to genomic stress. We come back to p53 later (see Figure 1-54).

Essentially, the nucleolus is a multifunctional nuclear structure consisting of stable proteins involved in ribosomal synthesis and molecules shuttling between the nucleolus and nucleoplasm to fulfill non-nucleolar functions.

Structurally, the nucleolus consists of three major components (Figure 1-44; see Box 1-P):

1. A fibrillar center (corresponding to chromatin containing repeated rRNA genes and the presence of RNA polymerase I and signal recognition particle [SRP] RNA).

2. A dense fibrillar component (where nascent rRNA is present and undergoing some of its processing). Fibrillarin and nucleolin are found in the fibrillar dense component.

3. A granular component (where the assembly of ribosomal sub units—containing 18S rRNA [small subunit] and 28S rRNA [large subunit]—is completed). Nucleostemin, a protein unrelated to ribosomal biogenesis, coexists with the granular components.

Figure 1-44 Components of the nucleus and nucleolus

Box 1-P Nucleolus

• The nucleolus is the site of synthesis, processing, and modification of pre-rRNA and initial preribosomal assembly. It also houses proteins unrelated to ribosome synthesis and shuttling between the nucleolus and the nucleoplasm to serve specific functions.

• The nucleolus consists of three components: (1) fibrillar centers; (2) a dense fibrillar component surrounding the fibrillar centers; and (3) a granular component. Pre-rRNA synthesis occurs at the interface between the fibrillar centers and the surrounding dense fibrillar component. Nascent pre-rRNA transcripts extend into the dense fibrillar component and migrate to the granular component where processing, modification, and preribosomal assembly occur.

• The fibrillar centers contain chromatin and transcription factors, including RNA polymerase I. The dense fibrillar component, the site of initial pre-rRNA processing, contains small ribonucleoproteins involved in RNA modification. The granular component accounts for about 75% of the nucleolar mass; the granules correspond to preribosomes.

• The nucleolus disappears during mitotic prophase and reassembles at the end of telophase at specific chromosomal regions named nucleolar-organizer regions (NORs).

The nucleolus dissociates during mitosis, then reappears at the beginning of the G₁ phase. More than one nucleolar mass, each representing the product of a chromosome with a nucleolar organizing region (NOR), can be observed in the nucleus. In some cells with an extended interphase, such as neurons, a single large nucleolus is organized by the fusion of several nucleolar masses.

The active process of rRNA synthesis can be visualized at the electron microscopic level (Figure 1-45) by spreading the contents of nuclei of cells with hundreds of nucleoli (e.g., amphibian oocytes). rRNA genes can be seen as repeating gene units along the chromatin axis, like “Christmas trees,” pointing in the same direction and separated by nontranscribed spacers. The entire rRNA gene region is covered by more than 100 RNA polymerase I molecules synthesizing an equivalent number of fibrils, each with a terminal granule.

Figure 1-45 Processing of ribosomal RNA

Each fibril represents an rRNA precursor (45S) ribonucleoprotein molecule oriented perpendicularly to the chromatin axis similar to the branches of a tree. The 45S rRNA precursor is detached from the chromatin axis and cleaved into 28S, 18S, and 5.8S rRNAs.

The 18S rRNA and associated proteins form the small ribosomal subunit. The 28S and 5.8S, together with 5S rRNA made outside the nucleolus, and associated proteins form the large ribosomal subunit.

The mRNA precursor is transcribed by RNA polymerase II, and the tRNA precursor is transcribed by RNA polymerase III.

Localization of nucleic acids

Cytochemistry and autoradiography (Figure 1-46) provide information about the cellular distribution and synthesis of nucleic acids. The Feulgen reaction is specific for the localization of DNA (see Box 1-Q). Basic dyes, such as toluidine blue, stain DNA and RNA (see Box 1-R). Pretreatment with deoxyribonuclease (DNAse) and ribonuclease (RNAse) defines the distribution sites of DNA and RNA by selective removal of one of the nucleic acids.

Figure 1-46 Localization of nucleic acids

Box 1-Q PAS and Feulgen reactions

• Both reactions use the Schiff reagent.

• In the PAS reaction, periodic acid forms aldehyde groups in sugars of glycoproteins by an oxidation process.

• In the Feulgen reaction, hydrochloric acid forms aldehyde groups in deoxyribose by hydrolysis.

Box 1-R Basophilia and acidophilia

Many cytologie stains use acidic and basic dyes.

• Basic or cationic dyes have positively charged color radicals forming electrostatic linkages with acidic groups (e.g., phosphate groups in nucleic acids). Toluidine blue is a cationic dye that binds to phosphate groups in DNA and RNA to give a blue color. DNA and RNA are considered to be basophilic (having binding affinity for a basic dye).

• Acidic or anionic dyes have negatively charged color radicals establishing electrostatic linkages with basic groups. Eosin is an anionic dye that stains many basic proteins. Basic proteins are considered to be acidophilic (having affinity for an acidic dye).

Autoradiography and radiolabeled precursors for one of the nucleic acids can determine the timing of their synthesis. In this technique, a radioactive precursor of DNA ([³H]thymidine) or RNA ([³H]uridine) is exposed to living cells. As a result of exposure to the radiolabel, any synthesized DNA or RNA contains the precursor. The radioactivity is detected by coating the cells with a thin layer of a photographic emulsion. Silver-containing crystals of the emulsion are exposed to structures of the cell containing radioactive DNA or RNA. After development of the emulsion, silver grains indicate the location of the labeled structures. This approach has been used extensively for determining the duration of several phases of the cell cycle.

CELL CYCLE

The cell cycle is defined as the interval between two successive mitotic divisions resulting in the production of two daughter cells (Figure 1-47). The cell cycle is traditionally divided into two major phases: (1) interphase and (2) mitosis (also known as the M phase).

Figure 1-47 Phases of the cell cycle

The most relevant event of interphase is the S phase, when the DNA in the nucleus is replicated. S phase is preceded by an interval or gap called the G₁ phase. The beginning of mitosis is preceded by the G₂ phase, a phase in which the cell ensures that DNA replication is completed before starting the M phase. Essentially, G₁ and G₂ phases provide time for cell growth before and after DNA synthesis. Cell growth is required for doubling the cell mass in preparation for cell division.

Cells in G₁ can make a commitment to DNA replication and enter the S phase or stop their progression into the following S phase. If a cell does not enter the S phase, it remains in a resting state known as G₀, where it can remain for days, months, or years before reentering the cell cycle.

In a more contemporary view, the cycle is regarded as the coordinated progression and completion of three separate cycles:

1. A cytoplasmic cycle, consisting of the sequential activation of cyclin-dependent protein kinases in the presence of cyclins.

2. A nuclear cycle, in which DNA is replicated and chromosomes condense in preparation for cell division.

3. A centrosome cycle, consisting of the duplication of the two centrioles, called mother and daughter centrioles, and assembly of pericentriolar proteins in preparation for the organization of the mitotic spindle curing mitosis or meiosis (see Figure 1-47). Recall from our previous discussion on the centrosome as a microtubule organizing center that γ-tubulin ring complexes are microtubule-nucleating complexes interacting with the protein pericentrin in the pericentriolar material. If this interaction is disrupted, the cell cycle is arrested during the G₂-M phase transition, and the cell undergoes programmed cell death or apoptosis. Basal bodies, the origin site of cilia and flagella, derive from centrioles.

The activities of cyclin-dependent protein kinases–cyclin complexes coordinate the timed progression of the nuclear and centrosome cycles (see Box 1-S). Figure 1-48 provides additional details.

Box 1-S Cell cycle

• Cell division requires the coordination of three cycles: cytoplasmic cycle, nuclear cycle, and centrosome cycle. The centrosome cycle plays a role in regulating the cytoplasmic and nuclear cycles.

• The cytoplasmic cycle depends on the availability of cyclins activated and deactivated by cyclin-dependent kinases (Cdks). Cdk inhibitors inactivate Cdk-cyclin complexes. Cdk inhibitors are up-regulated at the transcriptional level to arrest, if necessary, the cytoplasmic and nuclear cycle.

• The nuclear cycle involves DNA duplication and chromosomal condensation. Cdk2 phosphorylation of a protein complex bound to the origin of DNA replication recruits DNA polymerase to initiate and complete DNA synthesis in S-phase. Cdk1 phosphorylation triggers chromosomal condensation (mediated by histone H3 phosphorylation) and breakdown of the nuclear envelope (determined by nuclear lamin phosphorylation).

• During the centrosome cycle, the two centrioles of a centrosome duplicate during S-phase after phosphorylation of centrosome substrates by Cdk2. Daughter centrioles derive from each centriole.

• Cdks are involved in the coordination of the cytoplasmic, nuclear, and centrosome cycles.

• Cdk2 activity is required to initiate DNA replication and centriolar duplication.

Figure 1-48 Regulation of the cell cycle

Autoradiography and FACS: Analysis of the dynamics of the cell cycle

The various phases of the cell cycle can be studied by autoradiography. Cells in the S phase can be recognized by detecting the synthesis of DNA using [³H] thymidine as a radiolabeled precursor. Cells can be stained through the developed emulsion layer to determine the precise localization sites of the overlapping silver grains.

The time progression of cells through the different phases of the cell cycle can be estimated using both brief and prolonged [³H] thymidine pulses. The number of cells radiolabeled during interphase (generally about 30%) represent the labeling index of the S phase. The fraction of radiolabeled cells seen in mitosis (mitotic index) indicates that the radiolabeled precursor, which entered the cell during the S phase, progressed through the G₂ phase into M phase.

An alternative to autoradiography is the measurement of DNA content (C value 1.5 pg per haploid cell) using a fluorescence-activated cell sorter (FACS). Cells are stained with a fluorescent dye, which binds to DNA. The amount of fluorescence detected by the FACS is equivalent to the amount of DNA in each cell (for example, 2C in G₁; 4C at the end of S phase; 4C during G₂).

Breakdown and reassembly of the nuclear envelope

The disassembly of the nuclear envelope occurs at the end of the mitotic and meiotic prophase. It involves the fragmentation of the nuclear envelope, the dissociation of the nuclear pore complexes, and the depolymerization of the nuclear lamina (Figure 1-49).

Figure 1-49 Assembly and disassembly of the nuclear envelope

The nuclear lamina is composed of type V intermediate filament proteins, lamins A, B, and C, which associate with each other to form the nuclear lamina. Phosphorylation of lamins—catalyzed first by protein kinase C and later by cyclin A–activated Cdk1 kinase—results in the disassembly of the nuclear lamina. In addition, the components of the nuclear pore complex, the nucleoporins, and the membranous cisternae of the endoplasmic reticulum also disperse. The endoplasmic reticulum is the nuclear membrane reservoir for nuclear envelope reassembly.

During anaphase, nucleoporins and three transmembrane protein components of the inner nuclear membrane—lamina-associated polypeptide 2β, lamin B receptor, and emerin—attach to the surface of the chromosomes (chromatin). Then, cisternae of the endoplasmic reticulum are recruited by nucleoporins and inner nuclear membrane proteins, and the nuclear envelope is rebuilt by the end of telophase.

A final step in the reconstruction of the nuclear envelope is the dephosphorylation of lamin B by protein phosphatase 1. Dephosphorylated lamin B associates with lamins A and C to form the nuclear lamina before cytokinesis. This sequence of events stresses the impact of gene mutations affecting the expression of lamin A or lamin-binding proteins (see Box 1-M) as causes of laminopathies.

Tumor-suppressor genes

Not only Cdk-cyclin complexes control the progresssion and completion of the cell cycle. Tissues use two strategies to restrict cell proliferation:

1. By limiting mitogenic factors, such as platelet-derived growth factor (PDGF) and fibroblast-growth factor (FGF), which stimulate cell growth.

2. By regulatory genes that actively suppress proliferation. These genes, called suppressor genes, control normal cell proliferation.

The retinoblastoma model provides important clues on how suppressor genes work (Figure 1-50). Each cell has duplicate copies of the retinoblastoma (Rb) gene as a safety backup. When the two copies of the Rb gene are mutated, an abnormal Rb protein induces cancerous growth of retinal cells.

Figure 1-50 Rb protein, an inhibitor of cell cycle progression

When a single copy of the Rb gene pair is mutated, the remaining Rb gene copy functions normally and suppresses unregulated cell proliferation unless a second mutation occurs. In children with only a single intact Rb gene copy, all cells of the developing embryo grow normally. Late in gestation, retinal cells may lose the normal copy of the Rb gene, and a retinoblastoma develops.

The Rb gene specifies a nuclear protein involved in regulating the activity of a group of proteins—transcription factors—involved in DNA synthesis and cell cycle progression. When Rb protein is dephosphorylated, it binds to transcription factors. Although the Rb protein–transcription factor complex can bind to target genes, the activity of the transcription factors is repressed.

When Rb protein is phosphorylated by the Cdk4–cyclin D complex, it dissociates from the transcription factor complex, which activates specific gene expression (Figure 1-51). Phosphorylated Rb protein switches transcription factors from suppression to activation required for DNA synthesis and progression of the cell cycle.

Figure 1-51 Dephosphorylated Rb protein, a gene suppressor

Clinical significance: Retinoblastoma gene and other suppressor genes

Retinoblastoma tumors occur early in life and are seldom seen after age 5 or 6 years. The disease often runs in families. In such families, this tumor may affect one-half of the offspring. Children with the familial form of retinoblastoma usually have multiple tumor sites growing in both eyes.

A second type of retinoblastoma, the sporadic form, is seen in children whose parents have no history of the disease. Once cured, these patients, as adults, do not transmit the disease to the next generation. Children with the sporadic retinoblastoma are genetically normal at fertilization, but during embryonic development two somatic mutations occur in a cell lineage, giving rise to the photoreceptors of the retina: the rods and cones. The resulting double-mutated Rb genes induce cells to proliferate into a retinoblastoma.

In familial retinoblastoma, the fertilized egg already carries a single mutant Rb gene, acquired from the sperm or egg. All cells derived from the zygote carry this mutation, including the cells of the retina. The remaining normal Rb gene must undergo a mutation to reach the double-mutated condition required for tumor formation. Each of the retinal cells is primed for tumorigenesis, and a single event triggers the malignant tumor.

Retinoblastoma is only one of several tumors that arise through loss or inactivation of critical genes. Wilms’ tumor of the kidney is caused by the loss of a growth-regulating gene, called WT-1. Similar to the Rb gene, both copies must be mutated before a cell begins to grow out of control.

One suppressor gene that does not fit easily into this model is p53, the most frequently mutated gene in human tumors (leukemias, lymphomas, brain tumors, and breast cancer, among others). The p53 gene encodes the p53 protein, a tetramer that binds to a specific sequence of DNA involved in the transcriptional control of certain genes.

A mutation that affects one of the four subunits of p53 may compromise the function of the remaining three subunits. In contrast to the mutations that affect most other suppressor genes by knocking out gene function completely, the p53 mutations can result in either mild or aggressive growth.

In Chapter 16, Lower Digestive Segment, we study the tumor-suppressor adenomatous polyposis coli (APC) gene responsible for a hereditary form of colon cancer (familial adenomatous polyposis) derived from the malignant transformation of some of the many polyps (benign tumors) observed in individuals affected by this condition.

MITOSIS

Mitosis is preceded by the duplication of a pair of centrioles, each of which moves toward opposite sites of the nucleus to organize a centrosome. The primary function of the centrosome is the formation and maintenance of the mitotic spindle consisting of microtubules. Because of this function, the centrosome is also called the microtubule-organizing center (MOC). About 1000 new microtubules can be generated per minute on each centrosome using a pool of tubulin dimers derived from disassembled cytoplasmic microtubules.

Mitosis is divided into four substages: prophase, metaphase, anaphase, and telophase. The highlights of mitosis are summarized in Figure 1-53.

Telomerase, senescence, and cancer

Somatic cells can undergo a limited number of cell divisions, after which they enter a state of senescence. In contrast, tumor cells have an unlimited life span required for the formation of a tumor. In vitro studies using cultured cells have provided a model for the study of the biological clock of normal somatic cells.

The telomeres are the ends of chromosomes formed by a stretch of repeated nucleotide sequences (see Figure 1-52). Telomeres are responsible for maintaining chromosomal integrity and represent the cellular biological clock. When DNA polymerases fail to copy the chromosomal ends, telomeres decrease in size with every cell division. Cellular senescence occurs when the telomeres shorten to a point at which the integrity of a chromosome cannot be maintained.

Figure 1-52 The telomerase complex

Figure 1-53 Phases of mitosis

The length of the telomeres in male and female germinal cells and hematopoietic stem cells is protected by the enzyme telomerase, a ribonucleoprotein with reverse transcriptase activity that uses an RNA template to maintain the length of the telomeres. Telomerase is not present in somatic cells.

Most tumor cells express high levels of telomerase. The telomerase complex (see Figure 1-52) consists of the catalytic telomerase reverse transcriptase (TERT), the RNA sub unit telomerase template RNA (TR), which provides the template for repeat synthesis of chromosome ends, and dyskerin (DKC1), an auxiliary protein. This complex is assembled in Cajal bodies in the nucleus and is transported to the telomeres by an accessory protein, telomerase Cajal protein 1 (TCAB1). Two ATPases, pontin and reptin, activate the telomerase complex at the chromosomal end and initiate nucleotide addition.

Telomere dysfunction has been directly implicated in two diseases: dyskeratosis congenita and idiopathic pulmonary fibrosis. Dyskeratosis congenita is characterized by bone marrow failure, abnormal skin pigmentation, nail dystrophy, and leukoplakia (patches of keratosis on the tongue and the inside of the cheeks). Idiopathic pulmonary fibrosis leads to the progressive destruction of the lung tissue with a fatal outcome. Short telomeres are observed in both diseases.

Clinical significance: Role of protein p53 in chemotherapy

Chemotherapy and radiotherapy are effective in the treatment of metastatic tumors. Chemotherapeutic agents can:

1. Chemically cross-link DNA (alkylating agents).

2. Inhibit enzymes required for DNA synthesis (nucleotide analogues).

3. Affect microtubules of the mitotic spindle (taxol, vinblastine). These agents are usually administered in combination, during short periods or continuously, depending on the sensitivity of the type of tumor and avoiding toxic effects on highly sensitive organs, such as the bone marrow, intestinal epithelium, kidneys, and nervous system.

There are two kinds of resistance of tumors to chemotherapeutic agents:

1. Intrinsic resistance (tumors typically refractory to many agents—melanoma, liver cancer, renal cell carcinoma).

2. Acquired resistance (tumors becoming resistant to chemotherapy, after initial sensitivity).

One form of acquired resistance is caused by genes of the multidrug-resistance (mdr) gene family (Figure 1-54). These genes encode ATP-dependent pumps involved in the transport of large organic compounds. We see the mdr gene family of proteins again in Chapter 17, Digestive Glands, when we discuss the mechanism of bile secretion by hepatocytes.

The most studied gene involved in resistance to cancer chemotherapy is mdr-1. Repeated exposure to certain chemotherapeutic agents correlates with overexpression of mdr-1 and increased export of antitumoral agents when they enter the tumor cell.

DNA damage induced by chemotherapy and radiotherapy—genotoxic stress—triggers the activation of p53, a transcription factor tetramer that destroys terminally damaged cells through the activation of a cell death program, or apoptosis (see Box 1-T). In normal cells, genotoxic stress leads to the inhibition of Mdm2 (for mouse double minute 2) allowing activation of p53 and continuation of normal growth and development (see Figure 1-54).

Box 1-T p53, a tumor-suppressor protein

• The tumor-suppressor protein p53 protects the integrity of DNA in response to harmful stress, called genotoxic stress.

• The protective function depends on the ability of p53 to induce programmed cell death or apoptosis or arrest cell cycle activities, when a cell undergoes genotoxic stress.

• How does p53 work? As a transcription factor, p53 controls the transcriptional activation of proapoptotic genes and the inactivation of antiapoptotic genes. By this mechanism, a cell affected by genotoxic stress is eliminated.

• What can go wrong? A loss of p53 function may occur by a mutation of the TP53 gene, which encodes p53—or by an abnormal signaling pathway controlling p53 function (see Figure 1-54).

• Why is p53 important? Cancer cells are highly sensitive to apoptotic signals, but can survive if there is a loss of p53 function.

Mdm2 is a ubiquitin ligase that binds to p53 and facilitates its ubiquitin-dependent degradation in the cytoplasm by the 26S proteasome (see Figure 3-14 in Chapter 3, Cell Signaling). Mdm2 inhibition (for example, by ARF—for alternate reading frame—a 14-kd protein) allows p53 to activate its tumor-suppressor functions. Mdm2 exerts a similar inhibitory effect on retinoblastoma tumor-suppressor protein (Rb protein). The protein levels of ARF, Mdm2, and p53 are not abundant in genotoxic stress-free cells. The half-life of p53 is only 10 to 15 minutes.

In cancer cells, three possible mechanisms may prevent apoptotic cell destruction after genotoxic stress:

1. Mdm2 is unable to inhibit p53 tumor-suppressive function.

2. A mutational inactivation affects p53 function.

3. The apoptotic cascade has been disrupted (e.g., a loss in the activation of caspase 9).

Mutations of the TP53 gene, which encodes the p53 protein, are observed in 50% of human cancers. The loss of TP53 gene expression by an autosomal dominant mutation is responsible for a multicancer phenotype known as Li-Fraumeni syndrome (see Box 1-U). p53 is a tumor-suppressor gene. The inactivation of p53 activity is disrupted in drug-resistant cancer cells (see Figure 1-54). Loss of p53 expression is observed in human cancer cells, and clinical studies suggest that inactivation of p53 expression correlates with resistance to chemotherapeutic agents.

Box 1-U Li-Fraumeni syndrome

• Li-Fraumeni syndrome (LFS) is an autosomal dominant condition characterized by a predisposition to cancer.

• Several types of cancer develop in a young individual (younger than 45 years old): brain tumors, breast tumors (40% of the tumors in females), acute leukemia, and soft tissue and bone sarcomas.

• LFS syndrome is caused by a mutation of the tumor-suppressor gene encoding p53, a transcription factor with a cell cycle regulatory function.

• The incidence of LFS is low. Although the initial cancer can be successfully treated in affected children, there is a significant risk in the subsequent development of a second primary malignant tumor.

Pharmacologic agents binding to Mdm2 could stabilize and increase the levels of p53 in cancer cells to exert a tumor-suppressor activity through its death-inducing functions. We discuss in detail the mechanism of programmed cell death or apoptosis in Chapter 3, Cell Signaling.