As protein synthesis continues, the signal peptide is transiently immobilized in the membrane by virtue of its hydrophobic nature or its binding to a putative signal peptide receptor.41,42 Although the nascent protein chain is transferred to the cisterna by way of an unknown, energy-dependent translocation process, a luminal surface enzyme, signal peptidase, rapidly performs proteolytic cleavage to remove the signal peptide. The transmembrane transport of the protein does not require signal peptide cleavage and may take place by way of a protein channel or, less likely, through lipid. If the protein is to be N-glycosylated (i.e., to contain asparagine-linked carbohydrate moieties), other enzymes and the dolichol-lipid oligosaccharide carrier provide core glycosylation in this cotranslational process. Moreover, protein folding and oxidation of cysteine residues in disulfide formation occur. At the completion of protein synthesis and complete transfer of the protein to the luminal space, the SRP complex dissociates from its receptor and is recycled into the cytoplasm. Polyribosomes also are disaggregated to form free ribosomes and RNA. The translation of the polypeptide hormone causes the synthesis of a polypeptide core derived from the initial protein precursor, which is already modified, in some instances, by the addition of carbohydrate moieties and by folding and formation of intramolecular disulfide linkages. The precursor polypeptide encoded by mRNA is not found in vivo, because the signal peptide is removed before the completion of the polypeptide chain. The exit from the ER probably depends on appropriate protein assembly or conformation, but glycosylation is not required.
Up to the translational and cotranslational steps in the ER, all secretory, membrane, lysosome, endogenous ER, and Golgi proteins have traversed the same biosynthetic path. After this point, the major task of sorting and transferring the proteins to the correct intracellular destinations must be completed. This complex process occurs in the Golgi stack and requires sorting signals among the proteins and sorting mechanisms in this organelle. A polypeptide hormone destined for regulated secretion must exit the ER, traverse the Golgi stack, and arrive properly in the secretory granule (Fig. 3-12).
FIGURE 3-12. The polypeptide hormone highway. Protein hormone synthesis is initiated in the cytoplasm on polyribosomes. The partially processed hormone, with the signal peptide removed and N-linked carbohydrate moieties attached and with appropriate folding, enters the lumen of the rough endoplasmic reticulum (RER). By way of transport vesicle–transitional elements, these partially processed products are transferred to the Golgi stack on fusion and release. In a serial process of budding formation of secretory vesicles and fusion, processed products are transferred through the Golgi stack, from which they exit as secretory vesicles or granules after sorting in the trans and trans-Golgi network compartments of the Golgi. Materials are then released from granules by the fusion of vesicles or granules with the plasma membrane.
The Golgi stack comprises a series of flattened, saccular membranous compartments that encompass four histologically and functionally distinct regions: the cis, medial, and trans regions of the Golgi complex and the trans-Golgi network (TGN)43,44 (Fig. 3-13). The cis-Golgi region is most proximal to the transitional elements of the RER, and the TGN is most distal. The maintenance of distinct Golgi-specific antigens, unique enzyme markers, and different lectin-binding characteristics suggest that the compartments are not contiguous.
FIGURE 3-13. The Golgi stack. The Golgi stack consists of numerous membranous compartments, including cis, medial, and trans-Golgi elements. These compartments may be differentiated by the presence of specific enzymes. Partially processed protein hormones traverse this system by way of intermediate secretory vesicles in a budding-fusion reiterative process. In addition to transport, protein processing occurs. Sorting with routing to ultimate destinations in cellular sites is accomplished in the trans-Golgi network (TGN). Secretory peptides may be sorted to constitutive or regulated secretory pathways. Constitutive secretory pathways are equivalent to the pathways taken by membrane proteins, whereby non–clathrin-coated membrane segments are used. The regulated secretory-secretory granule pathway involves a clathrin-coated pit among membrane segments. This is similar to the pathway taken by lysosomal components. (Adapted from Griffiths G, Simons K. The trans Golgi network: sorting at the exit site of the Golgi complex. Science 1986; 234:438.)
A vesicle transfer model has been proposed to account for transport of materials from the RER to the TGN. In this model, membrane vesicles form from the upstream compartment by budding at the rims of the Golgi plates and rejoin the adjacent downstream compartment by vesicle fusion and the interaction of microfilaments. The reiterative process of budding and fusion of secretory or transport vesicles causes vectorial transfer of proteins from the RER to the TGN in a unidirectional and energy-dependent process.
The newly synthesized polypeptide in the lumen of the RER is first translocated to the cis-Golgi region (see Fig. 3-12). From this point, the protein is transported and processed in the Golgi stack. This organelle may be appropriately considered an assembly line for posttranslational processing. It is here that N-linked carbohydrate cores are further modified among glycoproteins45 (Fig. 3-14). This process involves digestion of the high-mannose peripheral sugars in the N-linked carbohydrate cores by multiple glycosidases and subsequent addition of distal or terminal sugars by way of numerous glucosyltrans-ferases. The steps in this process of carbohydrate maturation occur in different Golgi compartments. Other processes also occur, including phosphorylation, acetylation, sulfation, acylation, -amidation of COOH termini, addition of ubiquitin, other modifications, and degradation.46
FIGURE 3-14. Proximal and distal glycosylation. The pathway of glycosylation in the rough endoplasmic reticulum (RER) and Golgi is shown. Core carbohydrate moieties are added cotranslationally by way of a dolichol-sugar intermediate (Dol-) to Asn residues in the protein backbone in the RER. Several glycosidases (steps 1–4) remove distal sugars in this compartment. Distal glycosylation occurs by the actions of mannosidases (steps 5–7) and glycosyl transferases (steps 6, 8–1) in the Golgi. Phosphorylation (I, II) of N-acetyl glucosamines in carbohydrate moieties in the cis Golgi occurs in proteins destined for lysosome localization. (From Kornfeld R, Kornfeld S. Assembly of asparagine-linked oligosaccharides. Annu Rev Biochem 1985; 54:631.)
Another important function of the Golgi stack is the delivery of nascent polypeptides to the appropriate targets within the cell, which occurs in the trans-Golgi region or TGN.47 The proteins destined for lysosomal sites are targeted to those organelles by way of the mannose-6-phosphate receptor.48 In a similar manner, receptor and secretory proteins are targeted to membrane and secretory granule sites, respectively.49,50,51,52 and 53 The nearly mature polypeptide emerges from the Golgi stack in the TGN, where transport organelles, known as secretory vesicles or granules, are formed. These vesicles allow the exit of the nearly mature protein hormone from the Golgi stack.
Secretory proteins are released from a cell by way of two pathways: the constitutive pathway and the regulated pathway.54,55 The constitutive pathway is thought to be mediated by a passive aggregation sorting mechanism whereby peptide hormones form aggregates in the TGN, an action that is facilitated by acidic pH and high calcium concentrations in this compartment. The polarity of the secretory faces of epithelial cells enables proteins that are released in a nonregulated or constitutive manner to be released on the apical surface and regulated release to be performed at the basolateral surface. Whether such polarity of secretion exists in endocrine cells is unknown. Constitutive release generally involves the rapid exocytosis of newly synthesized peptides, but regulated secretion involves the classic secretory granule and signaled degranulation, causing hormone- or factor-regulated release of hormones. Secretory peptides must be segregated into one pathway or the other. Regulated secretion involves the formation of secretory residues and granules composed of clathrin-containing membrane segments, as found in lysosomes. Proteins destined for regulated secretion must end up in a reservoir known as the secretory granule, where the polypeptide hormones are concentrated and stored. This pathway is now considered to operate by active sorting via a signal ligand receptor. Proteins destined for secretion in this manner clearly contain sorting signals in their precursor molecules. For instance, the precursors to proopiomelanocortin (POMC) and proenkephalin have a stretch of aliphatic hydrophobic and acidic amino-acid residues at the N termini that are necessary and sufficient for efficient sorting into secretory granules. Further, carboxypeptidase E (Cpe) appears to serve as a sorting receptor for these peptide signals as determined by biochemical and genetic approaches. In particular, the Cpefat, which harbors a mutant and ineffective Cpe, is obese, diabetic, and infertile. It has elevated levels of proinsulin in pancreatic B cells and of POMC in the anterior pituitary, and decreased insulin and ACTH release.56
Three types of vesicles are formed in the TGN. One is the secretory vesicle, which is not clathrin coated and mediates non–receptor-dependent transport of membrane proteins and protein to be secreted in the constitutive pathway. The other two are the secretory granule, which is partially clathrin coated and mediates the receptor-dependent transfer of regulated secretory peptides, and the lysosome, which is predominantly clathrin coated and mediates transport of lysosomal enzymes and proteins.57 The secretory vesicle participates in the default, bulk-flow sorting system, but the others require the presence of “sorting patches” or sorting signals based on secondary and tertiary, but not primary, structures.55 Although the secretory granules are derived from immature granules with clathrin-coated pits, the precise nature of the receptor-mediated sorting of peptide hormones is unknown.
Evidence exists for pH-regulated, receptor-dependent sorting in the trans-Golgi and TGN. The pH of the compartments decreases as the Golgi stack is traversed from cis to trans regions. Such gradients in pH may participate in the molecular aggregation of polypeptide hormones. Possibly, these aggregates formed in the process of hormone concentration may initiate the budding of secretory granules. Chloroquine, which prevents Golgi acidification, may inhibit granule formation by preventing aggregation in neutralized Golgi stacks.
Proteolytic processing of protein precursors (i.e., proproteins or polyproteins) to yield smaller bioactive peptides (see Table 3-1) also occurs in acidic Golgi and secretory vesicles.58 Such proteolysis, however, is not required for packaging.
Much has been learned about the nature of polypeptide hormones and secretory granules.55,59,60 The hormones in this organelle are highly concentrated. In particular, a number of polypeptide hormones are condensed in a crystal lattice formation to increase the amount of hormone (up to 200-fold) in this organelle. Secretory granules allow cells to store enough hormone to be released on demand by extracellular signals at a level not possible by de novo synthesis. The t1/2 of stored hormones may be days, whereas the t1/2 of similar proteins in secretory vesicles may be minutes.
The size of secretory granules varies greatly, depending on the nature of the stored hormone. The condensation of hormone is demonstrated by the presence of electron-opaque or “dense” cores. The granule core is quite stable and is often visible even after exocytosis or in vitro enzymatic digestion of the granule membrane. It is osmotically inert yet sensitive to pH levels higher than 7.0.55
The formation of the secretory granule proceeds in stages, beginning in the trans-Golgi, where the initial hormone concentration may be observed. This aggregation process60a is facilitated by changes in pH, calcium concentration, and possible presence of other proteins such as secretogranins, chromogranins, and sulfated proteoglycans. Aggregates may form in different regions of the secretory granule. The colocalization of two or more polypeptide hormones in a granule may be observed. Within a cell, the relative distribution of two hormones is constant from granule to granule; however, variability in overall distribution is achieved from cell to cell. The mechanism by which the gonadotrope, a cell that generally produces luteinizing hormone (LH) and follicle-stimulating hormone (FSH), may be regulated to release LH and FSH differentially remains unclear.61
Secretory granules release their contents by cytoskeletal protein-mediated movement of the granule toward the cellular surface.61a There, secretory granule membranes fuse with the plasma membrane and allow eversion or exocytosis of stored hormone.62 This process of emiocytosis causes secretion of hormone. The mechanisms involved in stimulus-secretion coupling are not well known, although responses to cellular signals causing changes in intracellular calcium, ion currents, or intra-cellular pH may lead to these events.
In the unstimulated cell, a web of actin-associated microfilaments on the cytoplasmic face of the plasma membrane may act as a physical barrier to secretory granule fusion. However, changes in intracellular calcium, ion currents, or intracellular pH may cause differences in actin-binding protein interactions and alterations in the “secretory barrier” and permit exocytosis to occur.
Secretion and rapid membrane fusion of multiple secretory granules require an endocytotic pathway to retrieve the extra membranes resulting from exocytosis in the plasma membrane and to return them to the Golgi stack and lysosome.
REGULATION OF POLYPEPTIDE HORMONE SYNTHESIS
Regulation of the biosynthesis of polypeptides may occur at any of the biosynthetic levels in the pathway (Table 3-2 and Fig. 3-4). Of major interest is the regulation of peptide hormone synthesis at the transcriptional level.
TABLE 3-2. Loci of Genetic Regulation of Polypeptide Hormone Synthesis*
Studies using gene transfer and structure-function analysis have established that specific DNA elements in the regulatory region of the transcriptional unit are critical for determining transcriptional rates of various structural regions.63,64 In particular, hormone-regulatory elements (HREs) have been characterized for glucocorticoid, estrogen, androgen progesterone, vitamin D, mineralocorticoid, retinoic acid, and thyroid hormone receptors. In each case, a DNA element 8 to 20 nucleotides long may be necessary and sufficient for conferring hormonal regulation to its associated structural region. Several factors, including the steroid and thyroid hormones, interact with nuclear receptor proteins, which interact with DNA elements directly to modulate gene transcription.65,66,67 and 68 For the glucocorticoid receptor, the glucocorticoid ligand binds to the inactive glucocorticoid receptor in the cytoplasm, present in a complex with heat shock proteins, hsp 90 and hsp 70, and others.
The activated receptor-ligand complex interacts as a transacting factor to bind the DNA element corresponding to the glucocorticoid regulatory element (GRE). Studies have been performed on GREs in genes for mouse mammary tumor virus (MMTV) and murine sarcoma viruses, human metallothionein IIa, tyrosine aminotransferase, tryptophan oxygenase, and growth hormone, and in other genes. The long terminal repeat region of MMTV contains five GREs.64,69 A consensus sequence for the putative GRE is shown by the sequence 5′-GGTA-CANNNTGTTCT-3′, inwhich N = A, C, G, or T.
The structures of the steroid and thyroid hormone receptors are better known. These hormone receptors are encoded by genes related to a viral oncogene, v-erbA.70,71 The thyroid hormone receptor is encoded by the protooncogene c-erbA. Each receptor contains a stereotypic structure, including a protein that is ~45 to 60 kDa, with a central DNA-binding domain and a carboxyl-terminal ligand-binding domain. These and other regions mediate trans-activation, dimerization, and nuclear localization. The DNA-binding region consists of multiple cysteine and histidine residues that are critical for the formation of Zn2+ fingers first described in the DNA-binding protein TFIIIa, a transcription regulatory factor for the 5S ribosomal gene in Xenopus.72 This Zn2+finger interaction is a common motif for the binding of many eukaryotic proteins to DNA.73,74 and 75
The steroid–thyroid hormone receptors represent the first major examples of trans-acting factors well described in mammalian systems. The motif found in prokaryotic systems, particularly the interactions of cro and lambda repressor proteins with their target DNA elements in bacteriophage lambda, occurs with a homopolymeric dimer of subunits containing alpha helix–turn–alpha helix structure. The binding generally involves protein dimers; it requires a twofold axis of symmetry in the DNA sequence and involves the major groove of the target DNA over several helical turns. Data indicate that the thyroid hormone, retinoic acid, and vitamin D receptors are active only in the heterodimeric state, with other nuclear factors such as retinoid X receptors as their partners.
Hormones that act by way of surface membrane receptors may induce the production of second messengers that may directly or indirectly interact with DNA elements within the gene.76,77,78 and 79 HREs may not be restricted to interactions observed with steroid–thyroid hormone receptor complexes, but they may involve other protein-DNA interactions. Advances in the isolation of such trans-acting factors and the identification of cis-acting HREs will probably speed an understanding of the molecular mechanisms of the hormonal regulation of gene expression at the transcriptional level.80
The presence of multiple enhancer elements or HREs in the regulatory regions of genes allows fine tuning of transcriptional efficiency and influences the rate of production of the initial RNA transcript81,82,83,84 and 85 (Fig. 3-15).
FIGURE 3-15. Thyroid hormone action. This diagram depicts the mechanism of action of thyroid hormones in the regulation of a thyroid hormone–responsive gene. Thyroxine (T4) or triiodothyronine (T3) enters the cell. T4 is converted to T3 intracellularly in many cells by means of 5′-deiodinase activity. T3 then enters the nucleus, where it binds to the nuclear thyroid hormone receptor, which is encoded by c-erbA. This hormone nuclear receptor complex then serves as a transacting factor for binding to a thyroid hormone regulatory element (TRE), which may then positively or negatively regulate gene expression, with resultant production of RNA and protein derived from the thyroid hormone–regulated gene. (mRNA, messenger RNA.)
Other loci for regulation in this biosynthetic pathway include elongation and termination of transcription86 (see Fig. 3-9). The various steps of RNA maturation, most notably RNA splicing, may also change mRNA levels encoding a particular polypeptide hormone, which ultimately determines the amount of polypeptide produced. The nuclear stability of the hnRNA and transport of the RNA from the nucleus to the cytoplasm also may be regulated. A major determinant of the steady-state levels of mRNA is cytoplasmic mRNA stability. Examples include the estrogen regulation of chicken liver vitellogenin mRNA, prolactin regulation of breast casein mRNA stabilities, and thyroid hormone control of the TSH b subunit.87,88 and 89
The interaction of mRNA with the protein synthetic machinery in the process of translation may be regulated. Several examples of translational control have been observed, including glucose regulation of the translational efficiency of insulin mRNA. Moreover, a number of the posttranslational processing events that occur in the RER and Golgi stack and the control of secretory granule formation and release may also be loci for regulation.
Even after proteins are released from the secretory cell, the bioactive peptide may be further acted on by degradative processes and proteolytic events that may activate proteins in extracellular steps to determine the bioactivity of a particular polypeptide hormone. A major example involves the cascade of the extracellular enzymatic conversion of the precursors of angiotensin II (see Chap. 79). Another example of postsecretion proteolytic processing of precursor polypeptides involves the conversion of iodinated thyroglobulin to the iodinated thyronines, thyroxine and triiodothyronine, in the follicular cell of the thyroid. Plasma stability of a polypeptide is a major determinant of the activity of the hormone in its eventual interaction with target cells.