Chapter 18

I. HYPOTHALAMUS-PITUITARY SYSTEM

A. Anatomical Considerations

B. Regulation of Pituitary Functions

II. ANTERIOR LOBE AND INTERMEDIATE LOBE

A. Proopiomelanocortin-Derived Peptides

B. Glycoprotein Hormones

C. Somatomammotropic Hormones

III. NEUROHYPOPHYSIS

A. Vasopressin

B. Oxytocin

IV. ASSESSMENT OF PITUITARY FUNCTION

A. Adenohypophysis

B. Neurohypophysis

REFERENCES

I. HYPOTHALAMUS-PITUITARY SYSTEM

The hypothalamic-pituitary system is a preeminent example of the integration of neural and endocrine control. It consists of three major systems: (1) a neuroendocrine system connected to an endocrine system by a portal circulation, (2) a neurosecretory pathway, and (3) a direct neural regulation of endocrine secretion (Fig. 18-1).

The neuroendocrine system involves clusters of peptide- and monoamine-secreting cells in the anterior and midportion of the ventral hypothalamus. Their products reach the median eminence by axonal transport. From there they are released into the capillary vessels of the hypothalamic-pituitary portal system and transported to the pituitary to regulate the secretion of hormones from the anterior lobe (AL) of the adenohypophysis (Amar and Weiss, 2003).

The neurosecretory pathway runs from the anterior hypothalamus, traverses the floor of the ventral hypothalamus, and terminates in the neural lobe (NL) of the neurohypophysis on fenestrated blood vessels (Page, 1986). The system is involved in osmoregulation through the production and release of vasopressin, and in parturition and nursing through the secretion of oxytocin.

A. Anatomical Considerations

The portal capillaries coalesce into a series of vessels that descend through the pituitary stalk and form a second capillary plexus that surrounds the AL cells (Fig. 18-1).

Inferior hypophyseal arteries supply the neurohypophysis. From the primary plexus of the neural lobe (NL), blood flows not only to the systemic circulation but also to the AL and the hypothalamus. There is evidence for some degree of circulatory flow within the pituitary, that is, from the AL to the NL, from there to the infundibulum, and then back to the AL. The primary capillary plexus of the NL appears to be well positioned in the minicirculatory system, controlling all of the afferent vascular events and many of the efferent vascular events in the pituitary (Page, 1986).

During embryogenesis the adenohypophysis develops from Rathke’s pouch, which arises from the primitive roof of the mouth in contact with the base of the brain. Rathke’s pouch subsequently separates by constriction from the oral cavity. The anterior wall thickens and forms the anterior lobe of the adenohypophysis. This largest portion of the adenohypophysis remains separated from the intermediate lobe by the hypophyseal cleft, which is the residual lumen of Rathke’s pouch. In several species (Fig. 18-2), the adenohypophysis also extends into a pars tuberalis that forms a cuff or collar around the proximal neurohypophysis and may even envelop part of the median eminence (Batten and Ingleton, 1987; Hullinger, 1993).

In the embryonic development, the anterior lobe undergoes major cellular proliferation and differentiation (Dubois et al. 1997). Totipotent pituitary stem cells give rise to two main cell lineages, the acidophilic (mammasomatotropic, somatotropic, lactotropic) and the basophilic (corticotropic, thyrotropic, and gonadotropic) differentiated pituitary cell types. Determination of AL cell-type lineages results from a temporally regulated cascade of homeodomain transcription factors that are being dissected by genetic and molecular approaches at a rapid pace (Treier et al. 1998). Although most pituitary developmental information has been acquired from murine models, histological and pathogenetic observations in others mammals and human subjects have corroborated these developmental mechanisms. The transcription factor Rpx (Rathke’s pouch homeobox) is the earliest known specific marker for the adenohypophyseal primordium. Rathke’s pouch expresses several transcription factors of the LIM homeodomain family, including Lhx3 and Lhx4 (Sheng et al. 1997). Lhx3 is one of the earliest markers for cells that are destined to form the AL and the IL and is required for Pit-1 expression (Sheng et al. 1996). The homeodomain transcription factor Ptx1 behaves as a universal pituitary regulator and activates transcription of α-GSU (the α-subunit of gonadotroph hormones) and POMC (Drouin et al. 1998; Lamonerie et al., 1996). Transcription factors Tpit (T box transcription factor) (Lamolet et al., 2001) and NeuroD1 (Lamolet et al., 2004) appear to be a prerequisite for POMC expression and determine the development ofthe corticomelanotrope cell lineage. TSH and gonadotropin (LH and FSH)-expressing cells share the common α-GSU expression under developmental control of GATA-2. Prop-1, the prophet of Pit-1, stimulates the Pit-1 gene. Pit-1 (or POU1F1), a POU-homeodomain transcription factor, determines the development and appropriate temporal and spatial expression of cells committed to GH, PRL, and TSH. Corticotroph cell commitment, although occurring earliest during fetal development, is independent of Pit-1-determined cell lineages. Mutations arising within these transcription factors may result in isolated or combined pituitary hormone failure syndromes. German shepherd dogs with pituitary dwarfism have a combined deficiency of GH, TSH, and PRL together with impaired release of gonadotropins, whereas ACTH secretion is intact (Kooistra et al., 2000c). The transcription factors Lhx4 (van Oost et al., 2002), Prop-1 (Lantinga-van Leeuwen et al., 2000a), Pit-1 (Lantinga-van Leeuwen et al., 2000b), and the LIF receptor gene (Hanson et al., 2006b) have been excluded as candidates for pituitary dwarfism in German shepherds and the cause remains to be elucidated.

The posterior wall of Rathke’s pouch is closely apposed to the neural tissue of the NL, thereby forming the intermediate lobe (IL), which is well developed in most mammals, but not in humans and birds (Fig. 18.2). In humans only during fetal life is a distinct IL found (Amar and Weiss, 2003), whereas there is some debate about the view that in adults “invading” cells of the posterior lobe are homologous with the IL cells of lower vertebrates (McNicol, 1986). Unlike reptiles and mammals, birds have no IL (Batten and Ingleton, 1987). Nevertheless the typical “IL hormone” MSH is present in birds, having been found in the adenohypophysis, where it coexists with ACTH in cells named corticomelanotrophs (Iturriza et al., 1986).

Identification of the adenohypophyseal cells has been developed by histochemical staining techniques, immunohistochemical methods, and ultrastructural studies. Staining and immune reactions depend on the chemical nature of the hormones that are stored in granules within the cytoplasm. A general identification of adenohypophyseal cells (Batten and Ingleton, 1987) is given in Table 18-1, although it must be realized that the various cell types do not always give exactly the same reaction in different animal species. Having undergone a number of modifications over the years to improve sensitivity and specificity, immunohistochemistry remains the single most often used method in the identification of normal and diseased AL cell types. Apart from staining for the full spectrum of hormones (GH, PRL, ACTH, FSH, LH, TSH, and α-subunit), special stains investigate the diseased AL cell, such as staining with MIB-1, a cell proliferation marker that recognizes the KI-67 antigen, or demonstration of the protein p53, the product of a tumor suppressor gene, as parameters for aggressive behavior of cells. Development of sophisticated tools of molecular biology during the 1990s has enabled researchers to study chromosomal defects, genetic abnormalities such as specific mutations, the presence of oncogenes or of tumor suppressor genes, cell membrane receptors, signal transductions mechanisms, and intracellular changes that affect hormone synthesis and release. Pituitary cell cultures (normal and diseased, e.g., murine AtT-20 corticotrope cell line) allow various in vitro studies to be undertaken. In situ hybridization demonstrates specific messenger RNAs (mRNAs) at the subcellular level, which allows the reliable assessment of gene expression. In situ hybridization can be used to demonstrate the mRNA of receptors, growth factors, transcription factors, genes, and other regulatory peptides. Whereas immunocytochemistry provides information regarding cell hormone content, in situ hybridization confirms the presence of ongoing hormone synthesis.

Transmission electron microscopy allows the study of AL cells at the ultrastructural level. Applications of immunohistochemical methods at the ultrastructural level have also contributed to the rapid progress in the field. Ultrastructural studies of AL cells demonstrate the presence of distinct varieties of secretory granules and typical organization of cytoplasmic organelles for peptide synthesis and release. Characteristics such as the form, size, and location of the various organelles and the size and shape of the cells and their nuclei allow identification of adenohypophyseal cells at the ultrastructural level (Batten and Ingleton, 1987; Mikami, 1986).

B. Regulation of Pituitary Functions

The main hypothalamic neurohormones may stimulate or inhibit the release of a single hormone, or it may affect several hormone-producing cells (Fig. 18.3). The predominant influence of the hypothalamic hormones on the pituitary is stimulatory, and these peptides are known as releasing hormones (Guillemin, 2005). Interference with the integrity of the hypothalamic-pituitary connections results in decreased secretion of pituitary hormones. The exception is PRL, the secretion of which is increased when hypothalamic influence is removed.

As the complexity of the peptide structures of the hypophysiotropic hormones increases, species variation in sequence may occur. Whereas the structures of TRH, GnRH, and somatostatin (SS) (3, 10, and 14 amino acids, respectively) are identical for mammals, those of GHRH and CRH (44 and 41 amino acids) exhibit species specificity. The one nonpeptide hypophysiotropic hormone is dopamine. In addition to its major role as a neurotransmitter, dopamine is the most important inhibitor of PRL release.

The developments in recombinant DNA technology have enabled an increased knowledge on the genes encoding pituitary hormones and the regulatory elements involved in gene transcription. In short, the main elements regulating the gene transcription will be mentioned (Fig. 18.4). Eukaryotic genes encoding peptide hormones consist of a promoter (regulatory) unit and a transcription unit encoding the primary transcript that after appropriate processing will form the messenger RNA (mRNA).

The promoter unit is the upstream part (5′) of the gene. The promoter has specific DNA sequences (response elements) permitting the binding of transcription factors that enhance or inhibit gene expression by changes in the stability of the RNA polymerase-TATA box complex at the constitutive promoter. Also, with a higher distance to the gene-specific enhancer elements, or locus control regions as seen with GH expression (Ho et al., 2006), may regulate gene transcription. Tissue-specific silencers and enhancers that are mandatory for gene expression regulate the highly specialized hormone production. Methylation patterns of the promoter may also strongly influence the overall expression level (Newell-Price, 2003).

The process of peptide synthesis occurs principally on the rough endoplasmic reticulum (RER). Messenger RNA (mRNA), encoded by nuclear DNA, passes to cytosolic ribosomes, whereby sequential processing of transfer RNAs (tRNA), with their attached amino acids, the translation process of mRNA to a peptide starts (Fig. 18.4). The beginning of the growing peptide forms a specific signal peptide that facilitates the attachment of the translation complex to the RER and enables the passage into the lumen of the RER. The signal sequence of the preprohormone is cleaved, and the remaining prohormone undergoes several modifications as disulfide formation and glycosylation. The peptide passes along the RER lumen into the Golgi complex, where peptides are packaged and released into the cytoplasm as membrane-bound granules. During storage of these granules, the prohormone is further processed by specific proteolytic cleavage, C-terminal amidation, or N-terminal carboxylation. Characteristic for proteolytic cleavage sites are pairs of the basic amino acids arginine and lysine. The granules are stored until the hormone is released by exocytosis. This process involves fusion of the granule membrane with the cell membrane.

II. ANTERIOR LOBE AND INTERMEDIATE LOBE

A. Proopiomelanocortin-Derived Peptides

The corticotropic cells of the AL and the melanotropic cells of the IL are both able to synthesize proopiomelanocortin (POMC), the common prohormone for ACTH, α-MSH, β-lipotropin, and a family of β-endorphin-related peptides (Fig. 18-5).

POMC prohormone sequences are known for a variety of species and share a common structure (Fig. 18.5). Among species a high sequence homology is found between the N-terminal site (N-POC) of the prohormone (including γ-MSH), the ACTH region, and the β-endorphin region. In contrast, the regions between γ-MSH and ACTH and the first part of β-lipotropin are rather heterogeneous (Mol et al., 1991; Numa and Imura, 1985). Species differences in the sequence of ACTH occur only in the N-terminal part (Fig. 18.6), whereas the ACTH(1-24) sequence, necessary for full biological activity, is identical among mammals. The specificity of immunoassays to measure ACTH concentrations is achieved by directing antibodies toward the ACTH (13-18) epitope, which is absent in MSH and corticotropin-like intermediate lobe peptide (CLIP). These antibodies do, however, cross-react with intact POMC and may result in apparent elevated ACTH concentrations (Goossens et al., 1995). Two-sided assays are more sensitive to variation in the carboxy-terminal part and show insufficient cross-reactivity among the species.

The proteolytic enzymes PC1/PC3 and PC2 are involved in the processing of prohormones in neuroendocrine cells. They cleave the precursor at pairs of basic amino acids, resulting in the formation of biologically active hormones. The PC1/PC3 enzyme present in the AL cleaves POMC to ACTH and β-LPH. PC2 generates the ACTH(1-13) fragment in both the AL and IL. The combination of both enzymes is present in the melanotropes of the IL. In the hypothalamus, these enzymes determine also the availability of POMC-derived melanocortins, which play an essential role in the regulation of energy balance (Bertile and Raclot, 2006; Helwig et al., 2006).

In the melanotrope α-MSH is formed by C-terminal amidation and N-terminal acetylation of ACTH(1-13). Various degrees of acetylation of MSH result in the storage of desacetyl-, monoacetyl-, and diacetyl-α MSH in secretory granules. The canine IL contains predominantly the monoacetyl-α MSH (Young et al., 1992).

Many factors influence the secretion of ACTH by the AL (Table 18-2). A number of these factors modulatethe release of CRH and AVP, which are considered to be thepredominant stimulating neurohormones in vivo (Antoni, 1986; Keller-Wood and Dallman, 1984). The relative contribution of CRH and AVP to ACTH release varies among species and circumstances. In dogs (van Wijk et al., 1994) and pigs (Minton and Parsons, 1993), both exogenously administered CRH and LVP induce comparably high plasma ACTH concentrations. In sheep, AVP is the predominant ACTH-releasing factor. In horses, AVP is the immediate stimulus for ACTH release, and even ACTH micropulses appear to be regulated by AVP set point (Alexander et al., 1996). On the other hand, CRH secretion and pituitary responsiveness to CRH rise when cortisol falls, suggesting that in horses a major role for CRH is to fix the cortisol set point (Alexander et al., 1996). In the rat, the stimulation of ACTH release by CRH and AVP is synergistic, meaning that the response to CRH + AVP is greater than the sum of the reactions to CRH and AVP separately (Buckingham, 1987).

The basal release of ACTH is regulated by the occupancy of type I or mineralocorticoid receptors (MR) in the hippocampus (Jacobson and Sapolsky, 1991). Decreases in brain MR binding capacity of the dog during aging results in enhanced basal activity of the hypothalamus-pituitary-adrenal axis (Rothuizen et al., 1993). In sheep, basal ACTH release is inhibited by active immunization against AVP (Guillaume et al., 1992a) but not by immunization against AVP (Guillaume et al., 1992b).

Exercise and insulin-induced hypoglycemia stimulate ACTH release. In the horse, exercise-induced ACTH release and mild insulin-induced hypoglycemia are accompanied by increased AVP concentration in pituitary venous blood without changes in CRH concentrations (Alexander et al., 1991). However, when hypoglycemia becomes severe, CRH is released, which augments the ACTH response (Alexander et al., 1996, 1997). In sheep, mild hypoglycemia is accompanied by increases in AVP and CRH concentrations in portal blood, but in severe hypoglycemia the AVP secretion is relatively much higher than the CRH release (Caraty et al., 1990). AVP also regulates the ACTH response on hypoglycemia in the neonatal rat (Muret et al., 1992).

Hypotension induced by nitroprusside (Kemppainen and Sartin, 1987) or by hemorrhage (Lilly et al., 1983) causes ACTH release in the dog, together with large increases in plasma AVP derived from the NL (Raff et al., 1989). Selective neurohypophysectomy results in greatly attenuated ACTH and AVP responses to hypotension and angiotensin II, whereas the ACTH response to CRH injection remains unchanged. By substitution with adequate AVP infusion, the ACTH response to hypotension can be restored (Raff et al., 1992). In sheep, chronic absence of ovarian hormones after ovariectomy reduces the ACTH response to hypotension also, but not to CRH, AVP, or hypoglycemia (Pecins-Thompson and Keller-Wood, 1994).

In the dog, the release of ACTH is stimulated by β-adrenergic agonists (isoproterenol), dopaminergic antagonists (haloperidol), and serotoninergic agonists (quipazine maleate). TRH and GnRH stimulate the release of cortisol, probably by stimulating ACTH release (Stolp et al., 1982). In the horse, the α2-adrenergic agonist clonidine lowers ACTH secretion primarily by reducing the secretion of AVP and possibly CRH (Alexander and Irvine, 2000). Although direct stimulatory effects of catecholamine on the in vitro release of ACTH from ovine pituitary cells have been found, central stimulation of predominantly noradrenergic, but also adrenergic, pathways evokes the highest ACTH response (Liu et al., 1991).

Endogenous opiates (metenkephalin, dynorphin, and β-endorphin) inhibit the release of ACTH in humans (Besser et al., 1987). Conflicting results have been reported on the effect of metenkephalin in the rat, but β-endorphin and dynorphin may exert tonic inhibition of CRH release (Plotsky, 1986). The metenkephalin agonist DAMME stimulates the release of ACTH in the dog (Meij et al., 1990).

Activation of the immune system by infections results in the enhanced production of the cytokine interleukin-1 (IL-1β), which has the ability to stimulate CRH secretion from the hypothalamus and thus activates the hypothalamus-pituitary-adrenal axis. The biosynthesis and release of IL-6 has been found in the folliculostellate cells of the AL. IL-6 also stimulates the HPA-axis, at both the hypothalamic and the pituitary level (Sweep et al., 1991).

The predominant action of ACTH is stimulation of steroidogenesis and corticosteroid release from the adrenals (see Chapter 19 on adrenal function). ACTH also exerts a growth-stimulating effect on the adrenal cortex. Moreover, non-ACTH portions of POMC—that is, N-terminal POMC peptides—are involved in adrenocortical growth (Lowry et al., 1987). In pharmacological dosages, ACTH may promote lipolysis in fat cells and amino acid uptake in muscle. The role of intact ACTH produced in hypothalamic neurons projecting to higher brain centers remains to be elucidated. Distribution patterns have been reported for the cat (Rao et al., 1986). The C-terminal fragment of ACTH (18-39), called corticotropin-like intermediate lobe peptide (CLIP), has been found a potent stimulator of in vitro adrenal DNA synthesis (Wulffraat et al., 1987).

For details on clinical manifestations, laboratory findings, diagnostics, and treatment of these diseases, including iatrogenic hypoadrenocorticism resulting from corticosteroid therapy, refer to textbooks by Feldman and Nelson (2004) and Rijnberk (1996) and Chapter 19 on adrenocortical function in this volume.

g. Tests

Basal levels of circulating POMC-derived peptides are measured for diagnostic purposes in situations of suspected hypo- as well as hypersecretion. ACTH values below or just within the reference range (Table 18-3) may be found in cases of hypothalamus-pituitary disease, as well as in situations where endogenous or exogenous glucocorticoid excess suppresses hormone synthesis in the corticotropic cells. This makes the measurement of basal ACTH levels a useful tool in the differentiation between pituitary-dependent hyperadrenocorticism and hyperadrenocorticism resulting from adrenocortical tumor (Feldman, 1983; Peterson, 1986). In pituitary-dependent hyperadrenocorticism, ACTH values exceeding the reference range may be found, but there is considerable overlap (Meij et al., 1997a). Basal MSH concentrations have been found to be elevated in horses (Orth et al., 1982) and dogs (Meij et al., 1997b; Peterson et al., 1986; Rijnberk et al., 1987) with pituitary-dependent hyperadrenocorticism of IL origin. In the horse, basal plasma ACTH concentration predicts a pituitary adenoma of the IL with 100% sensitivity (van der Kolk et al., 1995).

Of the dynamic tests, the dexamethasone suppression tests (see Chapter 19 on adrenocortical function) are still the best for the diagnosis and differential diagnosis of excessive ACTH and glucocorticoid secretion. In animals suspected of having pituitary-adrenocortical insufficiency, the secretory capacity for POMC-derived peptides can be measured by provocative testing.

CRH and AVP are also used in the differentiation between pituitary-dependent hyperadrenocorticism and hyperadrenocorticism resulting from adrenocortical tumor. The chronically suppressed corticotropic cells are presumed to be unresponsive to these stimuli. However, Meijer et al., (1978) found considerable overlap in plasma cortisol value in the AVP test, whereas much less is claimed to occur in the CRH test (Peterson, 1986). It has been demonstrated that LVP stimulates cortisol release by adrenal tumors in a direct way (van Wijk et al., 1994).

It has been shown that dogs with pituitary-dependent hyperadrenocorticism usually remain responsive to single CRH administration (Rijnberk et al., 1987) or combined hypophysiotropic stimulation in spite of persisting hypercortisolism and neoplastic transformation of the corticotropes (Meij et al., 1997a). The cells of the pituitary lesions were found to be less responsive to CRH in vitro than were normal corticotropes (Mol et al., -1987) and less sensitive to inhibition by glucocorticoids (van Wijk et al., 1998).

As far as the possibility for manipulation of canine IL function is concerned, of the substances tested only the dopamine agonist haloperidol (Kemppainen and Sartin, 1986; Rijnberk et al., 1987) caused significant increases in circulating α-MSH. More detailed information pertaining to some of these tests is presented in Section IV and in Table 18-4.

2. β-Endorphin/β-Lipotropin

a. Gene Expression and Biosynthesis

Both β-lipotropin and β-endorphin are derived from the precursor protein POMC. The gene expression of POMC has been described in Section II. A.1. β-Lipotropin β-LPH) consists of the 91 C-terminal amino acids of the POMC precursor. It is synthesized in the corticotropic cells of both the AL and IL. The C-terminal sequence (36-91) is remarkably similar in human, porcine, and ovine pituitaries, whereas the N-terminal sequence (1-36) is rather heterogeneous. β-MSH (sequence 37-58) appears to be an extraction artifact and plays no physiological role.

By proteolytic cleavage at amino acid 61, γ-LPH (β-LPH[1-61]) and a family of endorphins are formed. Through specific deletion of C-terminal amino acids and N-terminal acetylation, a variety of β-END-related peptides are formed. The three main compounds are β-END(61-91), γ-END(61-77), and α-END(61-76). The corticotropic cells of the AL synthesize approximately equal or higher concentrations of β-LPH than of β-END(1-31) in most species. The equine AL, however, contains primarily β-END(1-31) and N-acetyl-β-END(1-27) (Millington et al., 1992).

The corticotropic/melanotropic cells of the IL produce relatively more β-END and related peptides. In the IL of the normal dog, the ratio β-LPH/β-END is less than 0.1 (Krieger, 1983). In the dog, β-END(1-27) is the most abundant form in the IL (Young and Kemppainen, 1994).

b Secretion

The stimuli that induce ACTH or α-MSH release from either the AL or IL cells cause also a release of the β-LPH-related peptides from the same cell. The cellular subset of specific proteolytic, amidating, and acetylating enzymes present in the secretory vesicles defines the ultimate composition of POMC-derived peptides that are secreted into the blood. The secretion of the IL is mainly under control of multiple neurotransmitters (Saland, 2001). In fearful dogs, only marginal increases in plasma β-END were measured after a gunshot test (Hydbring-Sandberg et al., 2004). In horses, a critical exercise threshold exists before plasma β-END concentrations increase (Mehl et al., 2000).

c Action

The main action of β-LPH is to mobilize fat from adipose tissues, as demonstrated in the rabbit. Its biological function in humans and other species has not been fully elucidated. Brain tissue can break down β-LPH to form β-END-related peptides. However, it is questionable whether these pituitary peptides are involved in brain function; in conscious sheep, hemorrhagic stress elevates β-END concentrations in plasma but not in cerebrospinal fluid (Smith et al., 1986).

β-END is an endogenous opiate with a potent morphinomimetic action. It is also produced in the hypothalamus where the entire POMC precursor is present. Brain and pituitary endorphin are probably part of separate systems. The role of β-END and other opioid peptides in the secretion of pituitary hormones has been studied with long-acting analogues, opiates, and opioid receptor antagonists (Estienne and Barb, 2005; Molina, 2006).

d. Disease/Tests

Hypersecretion of β-END is associated with the ACTH hypersecretion of IL tumors in both equine and canine pituitary-dependent hyperadrenocorticism (Orth et al., 1982; Peterson et al., 1986; Rijnberk et al., 1987). As the regulation of the secretion of β-END is similar to that of other POMC-derived peptides from the AL and IL, testing procedures can be applied as mentioned in the section on ACTH and α-MSH. Elevation of plasma β-END has been documented in horses following running and shipping (Li and Chen, 1987) and in sheep during electroimmobilization and shearing procedures (Jephcott et al., 1987). Treadmill exercise of Thoroughbred horses induces a variable response in plasma β-END concentrations (Art et al., 1994). Elevated plasma β-END concentrations are found in dogs with congestive heart failure (Himura et al., 1994) and are further elevated after naloxone treatment.

B. Glycoprotein Hormones

FIGURE 18-7 Sequence comparison of the-subunit for glycoprotein hormones (LH, FSH, and TSH). See the legend for Figure 18-6.

b. (Pro)hormone

Translation of the mRNA results in the formation of a precursor peptide with a molecular weight of approximately 13,500. The α-subunit has a high degree of homology among species (Fig. 18-7). The human sequence contains 92 amino acids, rather than the 96 as found in all other species studied so far. The shorter human sequence is due to a deletion of 12 nucleotides at the beginning of exon 3. After cleavage of the signal peptide, five disulfide bridges are formed. Two asparagine residues are prone to N-glycosylation, but there is also a putative O-glycosylation site (Fig. 18-7). The α-subunit is produced in excess of the β-subunit that determines the hormone specificity. The β-subunit formation is rate limiting in the formation of hetero αβ dimer.

2. TSH

The thyrotropic cells of the anterior pituitary produce thyroid-stimulating hormone (TSH), which stimulates both the synthesis and secretion of thyroid hormone.

a. Gene Expression

The rat, human, and canine TSH gene consists of three exons, whereas the mouse gene has five exons. Its expression is restricted to pituitary thyrotropes. For the development of thyrotropes, and also gonadotropes, the same transcription factors are necessary as described for αGSU expressing cells. For the development of specific thyrotropes, the transcription factor Pit-1 is essential (Dasen and Rosenfeld, 2001). Expression of the TSHβ gene is stimulated by the transcription factors Pit-1 and GATA2, whereas TSHβ gene expression is suppressed by the active thyroid hormone T (Nakano et al., 2004; Shupnik, 2000). In contrast with the suppression of TSHβ synthesis by T, which is produced within the pituitary by selective deiodination of T by type 2 deiodinase (D2), TRH plays a dominant role in the stimulation of TSH synthesis (Nikrodhanond et al., 2006).

The feline TSH has also been cloned and brought to expression (Rayalam et al., 2006).

b. (Pro)hormone

The TSHβ chain consists of 118 amino acids and forms six intrachain disulfide bonds (Fig. 18-8). There are no free cysteine residues, consistent with the fact that beta subunits form a heterodimer with the αGSU noncovalently. The TSHβ subunit has an N-glycosylation site at asparagine 23 and has a molecular weight of 18,000 after appropriate glycosylation.

The α-subunit is produced in excess of the β-subunit, which determines the hormone specificity, and its formation is rate limiting in the formation of the α–β dimer.

c. Secretion

The release of TSH is mainly regulated by stimulation by the hypothalamic thyrotropin-releasing hormone (TRH) and a strong negative feedback by thyroid hormone (Pazos-Moura et al., 2003) at the pituitary and hypothalamic level. The prohormone thyroxine (T₄) must therefore be converted locally to the active 3,3′,5-triiodothyronine (T₃) by type 2 iodothyronine deiodinase (D2) before binding to the nuclear thyroid hormone receptor of the thyrotrope (Christoffolete et al., 2006). Superimposed on this regulatory system, hypothalamic factors (somatostatin, dopamine) may inhibit TSH synthesis and release. Intrapituitary growth factors can also stimulate (EGF) or inhibit (Neuromedin B) TSH release (Pazos-Moura et al., 2003). Evidence has been presented on an ultra-short-loop feedback control by TSH through pituitary folliculostellate cells (Prummel et al., 2004).

FIGURE 18-8 Sequence comparison of the β-subunit of TSH. See the legend for Figure 18-6.

In the dog, TRH stimulates plasma TSH concentrations. Single administration of TRH results in slightly higher plasma TSH concentrations than measured after combined stimulation with four releasing hormones (Meij et al., 1996b). In euthyroid dogs, cTSH release is relatively stable with hardly any pulses as measured during a 6-h pulse study, whereas in hypothyroid dogs, an increased basal concentration coincides with higher pulse frequencies (Kooistra et al., 2000b).

d. Action

TSH stimulates both synthesis and secretion of thyroid hormones from the thyroid gland. After receptor binding, TSH stimulates via the Gs alpha protein the production of cAMP, which acts as an intracellular second messenger. In addition, intracellular Ca²⁺ may modulate the biological effect of TSH via the phosphoinositol pathway. As a result of receptor activation T₄, and to a much lesser extent T₃, is secreted into the blood. Prolonged stimulation of the thyroid with TSH results not only in hypersecretion of thyroid hormone but also in enlargement of the thyroid gland. Genetic analysis of DNA from hyperplastic or adenomatous thyroid nodules in cats has revealed that somatic mutations in the TSH-receptor or in the Gs alpha protein are the cause of feline thyroid hyperplasia (Peeters et al., 2002; Watson et al., 2005). Apart from the thyroid, TSH receptors have been found in brain and pituitary where they may have a function in feedback mechanisms, and in lymphocytes, thymus, testis, kidney, adipose tissue, and bone (Davies et al., 2005).

e. Disease/Tests

Since 1995, sensitive TSH assays have become available for the dog (Williams et al., 1996). Since then numerous reports have been published on the usefulness of cTSH measurements in the diagnosis of canine hypothyroidism. Low total plasma T₄ concentrations may be caused by decreased binding to plasma TBG—such as seen after long-term glucocorticoid excess in Cushing’s syndrome or interference of T₄ binding to TBG by, for instance, NSAIDs or antiepileptic drugs—and they are therefore not conclusive for the diagnosis of hypothyroidism. Unfortunately some 25% to 40% of dogs with proven hypothyroidism will have plasma cTSH concentrations within the reference range for healthy dogs (Boretti and Reusch, 2004; Peterson et al., 1997). In dogs with nonthyroidal disease also low serum concentrations of total T₄ are seen with serum TSH concentrations that remain within the reference range (Kantrowitz et al., 2001). In a recent study, quantitative measurement of thyroidal ^{99 m}TcO₄^– showed no overlap between dogs with primary hypothyroidism and nonthyroidal illness (Diaz Espineira et al., 2006) and may together with a thyroid biopsy be the ultimate proof of primary hypothyroidism.

Secondary hypothyroidism caused by a pituitary tumor (Rijnberk, 1971) or panhypopituitarism caused by a suprasellar tumor (Eigenmann et al., 1983b) is rare. German shepherd dogs with dwarfism have a secondary hypothyroidism, which can be demonstrated by a blunted TSH response on TRH in a combined pituitary function test (Kooistra et al., 2000c). For further details, refer to Chapter 20 on thyroid function.

Chemical induction of hypothyroidism in horses using the antithyroid drug PTU results in a steady increase of plasma TSH concentration and an exaggerated response to TRH administration (Breuhaus, 2002).

FIGURE 18-9 Sequence comparison of the β-subunit of LH. See the legend for Figure 18-6.

3. LH and FSH

Luteinizing hormone (LH), which is identical to the interstitial-cell-stimulating hormone, and follicle-stimulating hormone (FSH) are produced by the gonadotropic cells of the AL.

a. Gene Expression

In the gonadotropes αGSU appears to be the lead protein before the expression of the β-subunits, which determine the hormone specificity (Pope et al., 2006). Next the β-subunits arise in the bihormonal gonadotropes. During development of the pituitary, transcription factors Pitx, Hesx1, Lhx3, SF-1, and Otx1 regulate gonadotropin subunit expression (Brown and McNeilly, 1999). Although LH and FSH are produced within the same gonadotrope, its gene expression and secretion pattern are different.

In primates, multiple copies of LHβ-related genes have been identified, the chorionic gonadotropins (CG), which are expressed in the placenta for the maintenance of early pregnancy. Apart from primates, only in equids has this placental production of CG been well described (Jorgensen et al., 2004). In the horse, placental eCGβ is derived from the same gene that encodes pituitary LHβ (Saneyoshi et al., 2001). For pituitary LHβ expression, three highly conserved response elements have been found in the promoter region that bind the early growth response protein (Egr-1), the orphan nuclear factor SF-1, and the homeodomain protein Pitx1 (Jorgensen et al., 2004; Melamed et al., 2006).

For the regulation of basal FSHβ gene expression Lhx3 plays a prominent role, whereas a direct effect of SF-1 on FSHβ expression is not demonstrated in the ovine pituitary gland (Baratta et al., 2003).

The LHβ promoter is sensitive to stimulation by GnRH via Egr-1 and inhibition by androgens. The activated androgen receptor (AR) binds SF-1 to block communication via Pitx1 (Jorgensen et al., 2004). The AR, however, increases FSHβ gene promoter activity, as also documented for the progesterone receptor (PR) and glucocorticoid receptor (GR) (Thackray et al., 2006). Estrogens activate the LHβ promoter (Melamed et al., 2006). Inhibin, activin, and follistatin, originally found as hormones from the gonads but they may also be produced within the pituitary, are important regulators of LH and FSH synthesis and release. Activin stimulates FSHβ gene expression and mRNA stability, whereas inhibin and follistatin inhibit FSHβ gene expression and accelerate the degradation of FSHβ mRNA (Gregory and Kaiser, 2004). As activin stimulates and inhibin inhibits expression of GnRH receptors, an indirect effect on LHβ expression and release also can be found.

b. (Pro)hormone

After cleavage of the signal peptide, LHβ proteins are formed of 121 amino acids (Fig. 18-9). Exception is the horse sequence, which shows a highly variant C-terminal part that is also some 14 to 28 amino acids longer. The β-subunit contains six intrachain disulfide bridges. In humans, LHβ has one unique N-linked glycosylation at asparagine 30. In the horse, asparagine 13 is glycosylated and in addition 12 O-glycosylation sites have been found (Bousfield et al., 2001). The percentage glycosylation is related to increased plasma half-life and thus bioavailability of the hormones.

The FSHβ chain consists of 111 amino acids, six intrachain disulfide bridges, and two N-linked glycosylation sites at asparagine 7 and 24. The N-linked oligosaccharide chains are critical for bioactivity. Deglycosylated FSH may act as a potent FSH antagonist (Fares, 2006). The sequence of canine FSH shows the lowest homology (approximately 80%) with the sequences of other species (Fig. 18-10), which may cause limitations to the use of heterologous immunoassays for proper FSH measurements in the dog.

FIGURE 18-10 Sequence comparison of the β-subunit of FSH. See the legend for Figure 18-6.

c Secretion

As with other pituitary hormones, the gonadotropins are released in a pulsatile fashion, stimulated by pulses of the hypothalamic gonadotropin-releasing hormone GnRH. The release of GnRH in seasonal breeders is tightly regulated by the length of the photoperiod and the subsequent release of melatonin by the pineal gland (Thiery et al., 2002). Pulsatile GnRH secretion has been demonstrated in sheep (Clarke and Cummins, 1982) and the horse (Irvine and Alexander, 1994). GnRH can be used for estrus induction and prevention in the dog (Gobello, 2007). In contrast to single administration of GnRH, continuous administration of GnRH agonists reversibly suppresses gonadotropin secretion because of down-regulation/desensitization of GnRH receptors. Effect of dose and duration of treatment with a GnRH-agonist resulting in a switch from stimulated LH and FSH release toward down-regulation has been reported recently for the dog (Concannon et al., 2006).

In the dog, LH and FSH pulses coincide during different phases of the estrus cycle although the longer duration of FSH pulses may be due to the difference in plasma half-life. During progression from early to late anestrus basal plasma FSH increase without a concomitant increase in basal LH concentrations (Kooistra and Okkens, 2001a; Okkens and Kooistra, 2006). During the early follicular phase, the LH secretion has frequent increases of short duration (Kooistra et al., 1999). The preovulatory FSH in female dogs may start a few hours earlier or may coincide with the LH surge, which is associated with the highest plasma estradiol-17β concentrations and the start of increasing plasma progesterone concentrations (de Gier et al., 2006). During the mid- to late-luteal phase, basal LH concentrations in pregnant bitches were higher with lower peak frequency and height in comparison to nonpregnant bitches. Plasma FSH concentrations were higher during pregnancy (Onclin et al., 2002). Chronic administration of the synthetic progestin MPA has no effect on plasma LH concentrations, but, like in pregnancy, increased plasma FSH concentrations are found (Beijerink et al., 2006).

Administration of dopamine-agonists to shorten the anestrus interval results in a rapid rise in basal FSH concentration (Okkens and Kooistra, 2006), whereas the serotonin antagonist metergoline, which lowers also plasma prolactin concentrations, does not affect the LH and FSH secretion patterns in the dog (Beijerink et al., 2004).

Low plasma estrogen concentrations inhibit the GnRH and LH release through a negative feedback system, whereas high concentrations of estrogens may exert a positive feedback by stimulating GnRH release. Kiss1 neurons in the periventricular nucleus are thought to be involved in the estrogen and progesterone-induced LH surge (Smith et al., 2006). In the male, LH stimulates the synthesis and release of testosterone in the Leydig cells of the testis, which may in turn exert a negative feedback on LH secretion. This feedback may depend on aromatization of testosterone to estradiol in the brain. In the male dog, both testosterone and estradiol are the major inhibitors of LH and FSH release (Winter et al., 1982).

Finally, FSH can also be regulated by more specific stimulation through activin or by prevention of activin binding and thus inhibition of downstream signaling by inhibin and follistatin (Gregory and Kaiser, 2004; Gregory et al., 2005). Other members of the TGFβ family such as the bone morphogenic protein-6, BMP-7, and BMP-15 can also exert an effect on FSH secretion.

d. Action

FSH stimulates in the female ovary the folliculogenesis and ripening of the ovaries to the antral follicle stage. LH stimulates changes in the ovarian follicle resulting in ovulation and maintenance of the corpus luteum (Hunter et al., 2004). LH stimulates the production of the androgen precursor androstenedione that is converted, by FSH stimulation, to estradiol-17β and during the luteal phase progesterone secretion. The LH effects are mediated by LHR containing granulosa cells through the release of EGF-like growth factors (Conti et al., 2006).

C. Somatomammotropic Hormones

FIGURE 18-11 Sequence comparison of GH. See the legend for Figure 18-6.

Apart from the mammary gland, extrapituitary GH expression seems to be widespread throughout the body. GH functions already as an early local embryonic growth, and differentiation factor, including the neural tube, before pituitary GH expression is detectable (Sanders and Harvey, 2004). Furthermore, GH expression is found in the postnatal rat lung (Beyea et al., 2005), chicken testis (Harvey et al., 2004) and canine lymphomas (Lantinga van Leeuwen et al., 2000), insulinomas (Robben et al., 2002), and growth plate or osteosarcomas (Kirpensteijn et al., 2002).

c. (Pro)hormone

GH is a single-chain polypeptide. It contains two intrachain disulfide bridges and has an apparent molecular weight of 22,000. In humans, a small fraction of circulating GH has a molecular weight of 20 kDa because of alternative splicing of the pituitary GH gene. The metabolic effects are comparable to the normal 22 kDa form (Hayakawa et al., 2004). The amino acid sequence of GH belongs to the best-known sequences of pituitary hormones among species (Fig. 18-11). The sequences of canine and porcine GH are identical and very similar to the horse. The canine and porcine sequence is suggested to be identical to the GH sequence for the ancestral placental mammal (Forsyth and Wallis, 2002).

In cattle, a polymorphism in the GH gene exists, resulting in four GH variants that can arise from two possible N-terminal amino acids (phenylalanine or alanine) and two amino acids in position 127 (leucine or valine). The transmission of the trait of high milk production was greater for homozygous leucine-127 in Holstein cows and valine-127 in Jersey cows (Lucy et al., 1993). GH exhibits also heterogeneity because of posttranscriptional processing that may vary in binding to the plasma GH binding protein (GHBP) or biological effects (De Palo et al., 2006). The glycine in the third α-helix of bovine GH (G118) and human GH (G120) is crucial for biological functioning of GH. Substitution of this glycine results in a molecule without growth-promoting activity that inhibits the actions of GH in vitro and in vivo (Kopchick, 2003).

d. Secretion

The release of GH by the pituitary is regulated by a variety of factors (Fig. 18-12 and Table 18-5). The integration of all these factors results in a pulsatile release of GH. In the female dog, plasma GH pulse patterns are dependent on the estrus cycle. In the early luteal phase, elevated basal GH concentrations are associated with a low pulse frequency of 2 peaks/12 h, whereas at anestrus lower basal GH concentrations are found with an enhanced pulse frequency of 5 peaks/12 h (Kooistra et al., 2000a). The higher basal concentration and diminished pulsatility in early luteal phase may be caused by a direct or indirect inhibition of pituitary GH release by GH secretion from the mammary glands stimulated by progesterone.

Stay updated, free articles. Join our Telegram channel

Join