The Role of the Human Growth Hormone Gene Family in Pregnancy

2.1. Human growth hormone (HGH)

HGH can exert metabolic effects either directly or indirectly through, in the latter case, the increase in hepatic production of the insulin-like growth factor 1 (IGF-1). Regarding their direct actions, the administration of HGH has been reported to antagonize the action of insulin, thus producing intolerance to carbohydrates in spite of elevated plasma insulin levels. In contrast, insulin sensitivity increases during the administration of IGF-1, which exerts hypoglycemic effects even with concomitant suppression of insulin secretion. Another important direct metabolic effect of HGH is to increase the mobilization and oxidation of fat and therefore to reduce total body fat; there is no evidence that IGF-1 acts directly on adipose tissue in vivo. The administration of HGH results in the retention of sodium through the stimulation of Na-K-ATPase. It is suggested that part of the effects of HGH on tubular function (e.g., phosphate reabsorption) is mediated through IGF-1 [6].

The administration of HGH increases the circulating levels of IGF-1 through the stimulation of its hepatic synthesis and secretion; it can also improve the synthesis of local IGF-1 in peripheral tissues, where it exerts autocrine or paracrine effects [6]. While HGH increases lipolysis, as a direct effect on the adipocyte, as well as the oxidation of lipids by increasing substrate availability, IGF-1 increases lipid oxidation only when administered chronically, most likely as a result of chronic insulinopenia. These metabolic regulators have been tested in a variety of catabolic conditions in man, and both hormones have been effective in reducing protein loss caused by glucocorticosteroids and in mitigating some of the catabolic effects of severe hypogonadism in men [7].

Another function of IGF-1 is to increase myelination by increasing the number of myelinated axons and the thickness of myelin sheaths. The latter is by mechanisms involving the stimulation of myelin protein gene expression and by increasing the number of oligodendrocytes [8]. For this reason, people with IGF-1 deficiency, caused by a homozygous mutation in the IGF-1gene at the 12q22 locus, present, in addition to intra- and extrauterine growth retardation, mental retardation and sensorineural deafness [9].

Some studies have compared the phenotypes of dwarfism manifested in mutant mice lacking the HGH receptor, or IGF-1, or both, and have provided conclusive evidence that HGH and IGF-1 promote postnatal growth, both by independent and by common functions, given that the delay in the growth of double nullizygotes HGHR/IGF-1 is more severe than that observed with any of the mutant classes separately [10].

2.2. The growth hormone receptor (GHR)

The cloning of the HGH receptor (HGHR) gene in 1987 opened the door for the study of HGH signaling at the molecular level. Its mRNA encodes a protein of 638 amino acids (aa) with single extracellular, transmembrane, and cytoplasmic domains [19]. HGHR belongs to the transmembrane superfamily of proteins that includes the PRL receptor (PRLR) and a number of cytokine receptors [20].

The determination of the structure of HGH bound to the extracellular domain of HGHR has led to the model where a single molecule of HGH binds to two molecules of HGHR. This binding of HGH leading to the HGHR (2)-HGH complex is thought to be sequential. The initial step is the binding of HGH to a high-affinity HGHR monomer, whereby a different face of HGH is contacted by a second HGHR monomer, stabilizing the HGHR dimer. This binding of HGH to its receptor dimer is thought to be an initial and crucial event in the HGH signaling [19]. Subsequent to this, a conformational change in the extracellular domain of the receptor is important for signaling. At present, virtually nothing is known about the structure of the cytoplasmic domain of the HGHR; a clue to the mystery of the signaling mechanism came with the discovery that HGH promotes the tyrosine phosphorylation of receptors and other cellular proteins. The current working model of the signal transduction of HGH action is that its binding to two HGHR monomers increases the affinity of each receptor for JAK2. The dimerization brings two JAK2 molecules in proximity so that each JAK2 can phosphorylate the activation tyrosine of the other JAK2 molecule, blocking it in an active conformation. This allows the activated JAK2 to phosphorylate itself and the cytoplasmic domain of the HGHR in the tyrosine residues, in order to activate the signaling pathway required for the specific function of the HGH that needs to develop: regulation of gene transcription, metabolic actions, etc. [19].

2.3. The IGF receptor

The components of the IGF system include IGFs (IGF-1 and IGF-2), IGF type 1 (IGF-1R), and type 2 (IGF-2R) receptors, a family of six secreted IGF binding proteins (IGFBP) and IGFBP proteases [21]. The two IGF receptors are structurally and functionally related. The signaling of the IGF ligand is mediated by IGF-1R, which is a transmembrane glycoprotein with tyrosine kinase activity. IGF-2R is a single-chain protein with no kinase activity. IGF-1R binds to IGF-1 with up to 20 times higher affinity than with IGF-2, while IGF-2R binds strongly to IGF-2 but hardly recognizes IGF- 1. The genes for IGF2 and IGF2R have imprinting, expressing themselves in a monoallelic way depending on the parental origin [21].

IGF-1R is activated by two ligands, IGF-1 and IGF-2, and by insulin at supraphysiological concentrations [22]. After the binding of IGF-1 to its receptor, it undergoes a conformational change that unleashes its tyrosine activity kinase (TK). This autophosphorylates tyrosines which act as coupling sites for Shc (a signaling adaptor protein) and IRS-1 or IRS-2 signaling proteins. The IRS proteins are phosphorylated by the TK activity of the receptor and then binds to the p85 subunit of phosphatidylinositol 3-kinase (PI3K), which is the type A regulatory/subunit of the isoforms of class 1 of the PI3K, leading to the activation of protein kinase B (PKB) and the stimulation of protein synthesis, as well as the inhibition of apoptosis. The Shc signaling protein is also phosphorylated, which allows it to bind to growth factor receptor-bound protein 2 (Grb-2), leading to the activation of MAP kinase and the stimulation of DNA synthesis and cell growth (Figure 2).

The critical importance of this receptor for normal development and physiology is underlined by neonatal lethality as a result of its complete absence. Conversely, IGF-2R, also called mannose-6-phosphate receptor independent of cations, is less important for growth stimulation but is important for the regulation of IGF-1 and IGF-2 activities, both by sequestration of hormones, as by promoting their degradation [23].

IGF-2 is a key regulator of cell growth, survival, migration, and differentiation. Its fundamental role in these processes requires strict regulation, both of expression and activity. The IGF-1R mediates the actions of IGF-2, and a family of six high-affinity binding proteins of IGF (IGFBP-1 to IGFBP-6) regulates the circulating half-life of IGF-2 and its availability to bind to IGF-1R. In addition, IGF2-R modulates circulating and tissue levels of IGF-2, directing it to lysosomes for degradation [23].

An example of the growth regulation function of IGF-2 is the Beckwith-Wiedemann syndrome (BWS), which is a pediatric overgrowth disorder with predisposition to form embryonic tumors. Individuals with BWS can grow at a higher rate during the second half of pregnancy and in the first years of life [24]. BWS results from alterations in the imprint control region 1 (ICR1), either by deletion or by DNA methylation. The ICR1 region controls the genomic imprint of the H19gene, a gene that, unlike many others, does not code for a protein but rather a noncoding RNA molecule whose function is unknown, although it is suspected that it acts as a suppressor of the tumor and of the IGF-2gene, which has already been mentioned previously. This anomaly alters the regulation of both genes; specifically, it leads to a loss of the activity of the H19gene and an increase in the activity of the IGF-2gene in many tissues [25].

3.1. Growth hormone-releasing hormone

HGH releasing hormone (HGHRH) is a 44 aa peptide. Its concentration throughout pregnancy is similar to that in nonpregnant women despite fluctuations in HGH values, which are always higher than in nonpregnant levels [31], thus supporting the idea that HGH values are higher during pregnancy due to the placental secretion of HGH-V not regulated by HGHRH [29].

After delivery, placental HGH-V disappears from maternal serum within an hour. Amniotic fluid contains low HGH concentrations; cord serum contains high HGH levels, but not because of HGH-V (we assumed that the material responsible for the GH immunoreactivity in late pregnancy maternal serum was of placental origin, since it rapidly disappeared after delivery); thus, it appears to be secreted selectively into the maternal compartment [27]. HGH-V does not appear to have a direct effect on fetal growth as it is secreted only in the maternal circulation and is not detected in the fetal blood [5]. Secreted continuously by the placenta, it seems to control the synthesis of maternal IGF-1; indeed, maternal IGF-1 levels are correlated with placental HGH levels [30]. Its continuous secretion by syncytiotrophoblast villous into the maternal compartment may alter maternal metabolism during pregnancy. In the maternal liver and other organs, HGH-V strongly stimulates gluconeogenesis, lipolysis, and anabolism, thereby increasing nutrient availability for the fetoplacental unit [5]. HGH-V acts in vivo as an HGH-N agonist sharing most of its biological properties [27]. HGH-N, which is synthesized by the fetal pituitary gland (levels of which rise to a maximum at mid-gestation) [26], has little or no physiological action on the fetus until the end of pregnancy, because of the lack of functional HGH receptors in fetal tissues.

Growth, prenatal development, and size at birth are normal in animal fetuses of specimens subjected to hypophysectomy and HGH deficiency, including human fetuses with mutant genes for this and their receptors. However, this pattern is not maintained in the postnatal period, in which the individual manifests abnormal development and growth [21]. HGH deficiency does not eliminate the normal increase in IGF-1 induced by pregnancy and does not reduce fetal weight [30]. Unlike IGF-2 levels, the levels of HGH-V do not correlate with birth weight or placental weight [32].

In humans and mice, mutations or specific deletions of the genes for IGFs, IGF1, and IGF2, as well as IGF1R and for its main signaling molecule IRS, lead to a restriction in fetal growth [33]. The targeted inactivation of the mouse gene for IGF-2 results in a 40% reduction in fetal growth, but postnatal growth remains normal so that alterations in development occur exclusively in the prenatal period. On the other hand, the interruption of the IGF-1gene leads to a similar decrease in fetal growth than the alteration of the IGF-2gene, but it is also characterized by the lack of persistent postnatal growth [34]. Most surprising, however, are the phenotypic consequences of the deletion of the gene for the IGF-1 receptor (IGF-1R), which as described above, is a transmembrane tyrosine kinase that mediates the growth promotion actions of both IGFs. Mice with this suppression have a birth weight that is only 45% of normal and usually die in a matter of hours after birth due to respiratory failure as a result of muscle hypoplasia [34].

3.2. Growth hormone replacement therapy during pregnancy

A retrospective study of 25 women with HGH-N deficiency (HGHD), who underwent pregnancy without HGH replacement therapy (HGHRT), concluded that unsubstituted HGHD during pregnancy is not detrimental to the fetus [35]. Another publication described four HGHD women who stopped HGHRT immediately after confirmation of pregnancy and remained off treatment throughout the pregnancy while having no pregnancy complications and gave birth to healthy babies of normal height and weight [36]. In a case report, physiologic HGHRT until there was evidence of sufficient HGH-V production also led to normal pregnancy and a healthy fetus [37]. This regimen of maintaining HGHRT during the first trimester, gradually decreasing it during the second trimester, and discontinuing it during the third trimester was reported to lead to successful outcomes in 12 pregnancies [38]. In addition, replacement with HGH-N during pregnancy did not suppress the physiologic increase in HGH-V [32].

A study describing pregnancies in a large group of patients (173 pregnancies in 144 women with HGHD) in 15 countries using the Pfizer International Metabolic Database (KIMS) demonstrates that most patients conceived while receiving HGHRT. Details of HGHRT during pregnancy were reported in 170 out of 173 pregnancies. HGHRT was stopped at the beginning of the pregnancy (or had already been stopped before conception) in 81 cases (46.7%), partially continued in 42 pregnancies (24.7%), and continued throughout the entire pregnancy in 47 pregnancies (27.6%). The practice of partially continuing HGHRT during pregnancy and stopping it at the end of the second trimester was observed in nearly half of the countries but was more prevalent in Sweden (67% of all pregnancies in that country) and Denmark (55% of Danish pregnancies). Most physicians reported making their decisions about whether to continue HGHRT in agreement with the patient’s wish. Outcome data were available for 139 pregnancies (80.3% of cases). In four cases the pregnancy was electively terminated (in three cases because of the patient’s wish, in one case because of the doctor’s advice given several concomitant diseases and the resulting need for multiple medications); these cases were excluded from further analysis. Live birth was observed in 107 pregnancies (79% of all known cases), with a total of 118 babies born. No congenital malformations were reported. No live births were reported in 28 pregnancies, including two extrauterine pregnancies, one blighted ovum (non-evolutive pregnancy), one malformation (severe cystic hygroma in ultrasound, which determined pregnancy termination in the second trimester), one stillbirth, and 23 nonelective abortions. Patients who partially or fully continued HGHRT during the pregnancy did not report any miscarriages happening after the first trimester of the pregnancy [39].

There are few reported cases in the literature of pregnancies with the absence of CSH or very low concentrations in maternal plasma throughout pregnancy, most of them progressing normally and resulting in delivery of normal babies, but only in a few cases have had the genetic background examined in detail using molecular analysis (Table 1).

4.1. Application of the diagnostic test

4.1.1. A case of absence of CSH and HGH-V

In a pregnancy reported without HCS and HGH-V production, which was complicated by severe growth retardation of the fetus and mild preeclampsia and cardiotocogram abnormalities, the patient was given betamethasone, and Cesarean section was performed in week 35. A male baby was delivered with an Apgar score of 7/1, 10/5 and umbilical artery pH of 7.3, weight was 1270 g, and he measured 40 cm (the 10th percentile for weight for Danish male reference material at this gestational age is 2100 grams). Placenta weight was 250 g and was macroscopically normal. The umbilical cord contained only one artery. A thorough examination of the baby by a neonatologist revealed no physical abnormalities. The boy thrived; the only problem being a tendency to low blood sugar the first days, the lowest being 1.2 mM. He was discharged 26 days after delivery. Pediatric examination at discharge and 6 months later revealed no malformations or other problems [54].

Using the PCR method described before, the genes at the hGHmultigene family were investigated in DNA isolated from the placenta of this case. We found that the placenta, and thus the baby, had two different DNA deletions along the 3′ half of the gene cluster, both of which eliminated the two active hCSHgenes (hCSH-1and hCSH-2) and the placental hGHgene (hGH-V). The locus retained the pituitary hGH-Ngene as well as the placental hCSH-Lpseudogene (see Figure 5). The absence of CSH in this patient was caused by deletion of both copies in each chromosome of the active hCSHgenes (hCSH-1and hCSH-2) in the placenta and, thus, in the child’s genome. Both parents were heterozygous for the gene’s deletion, lacking one copy of the three 3′ cluster genes: hCSH-1, hGH-V, and hCSH-2. In the first instance, the baby appeared to be a homozygote for deletion including the 3′ end of the hGHlocus, but the PCR analysis revealed that the baby was, in fact, a double heterozygote for these deletions. Each chromosome lacks a different portion of the 3′ end of the hGHlocus: one deletion begins between hCSH-Land hCSH-1genes and the other beginning at the first exon of the hCSH-1gene [54]. In the few previous reports where the molecular background for complete absence or very low levels of CSH have been examined [55], the cause has also been the deletion of hCSH-1, hGH-V, and hCSH-2genes or only of the last two genes, and in some cases both types of hCSHgene deletion have been found [49].

4.1.2. Cases of children treated with recombinant HGH to treat their severe growth retardation

Genomic DNA samples of 10 patients clinically diagnosed with isolated HGH deficiency (IGHD) were analyzed with our new diagnostic method, to establish if the hGH-Ngene was absent and thus was the causal factor for this condition. Amplification products of all patients were digested with AccI, which is specific for the hGH-Ngene (0.9 and 0.6 kb) and for the hGH-Vgene (0.8 and 0.7 kb) genes. This indicated to us that the former gene was absent in this child and, thus, the cause of this patient’s disease, classifying her condition as IGHD type IA. Moreover, the pediatrician confirmed to us that this particular patient was not responding to the HGHRT [53] (Figure 6).

4.2. Anti-recombinant growth hormone antibodies

Pituitary HGH has been the preferred treatment for growth retardation in children since its efficacy was first reported 40 years ago. This treatment was discontinued in most countries in 1985 following the deaths from Creutzfeldt-Jakob disease of four patients who received the hormone recovered from cadavers in the period of 1965 through 1975. Application of recombinant (r) DNA technology has made the production of unlimited supplies of proteins possible, including HGH that has important therapeutic uses. But, the immunogenicity of commercially available rHGH is a matter of great concern. The adjuvant effects of unrelated contaminants associated with rHGH by disulfide, ionic and/or hydrophobic links, as well as the changes in the intrinsic primary and secondary structure, may occur during the production and/or recovery of the hormone and lead to potential immunogenicity. The main concern with anti-HGH antibodies could be their ability to neutralize circulating rHGH and inhibiting its growth-promoting effect. In a study evaluating 47 children treated with rHGH for up to 6 months, serum samples were examined for specific antibodies against it by ELISA, resulting in four patients positive for serum antibodies against the hormone [56]. Fortunately, new preparations have shown lower immunogenicity profile with no safety concerns [57].

In a study of four patients with gene defects in the GH axis, the results showed that HGH substitution may be effective at the beginning, but development of HGH-Ab often occurs, resulting in a HGH-resistant state with sequelae similar to GHIS (HG insensitivity syndrome). One patient [58] developed high-affinity and high-avidity blocking of HGH-Ab during the first year of pituitary-derived human GH (pit-GH) treatment. Even plasmapheresis and immune modulating treatment to induce tolerance analogous to previous treatment regimes in hemophiliac patients with blocking antibodies did not result in HGH responsiveness, and HGH-Ab reappeared shortly thereafter [59]. The HGH growth response, even in patients with the identical genetic defect, may differ and is not clearly related to the presence of antibodies [60]. Factors contributing to immune response are complex and cannot be clearly demonstrated. Epidemiology and risk factors of HGH-Ab development have not been studied in depth, but analogous to hemophilia, there are various aspects that might be deduced: On a superior level, immune processes of self−/non-self-discrimination, i.e., the likelihood that antibodies will be formed appears to be influenced by the age at first antigen contact, the HLA haplotype, and other immune response genes. In particular, epitopes giving rise to antibody formation may differ and lead to various effects which are dependent on steric conformation changes or potency of complement activation [61]. Patients with genetic HGH defects who are exposed to exogenous rHGH will therefore generate different amounts of HGH-Ab with different affinities, thus compromising HGH binding to the receptor or HGHBP/HGHR (GH binding protein and GH receptor) dimerization kinetics [62]. In patients with different HGHR mutations, the extent of height deficit varies substantially and may be correlated with the presence or absence of HGHBP in plasma, although clear genotype–phenotype associations do not exist, thus suggesting an influence of additional genes or environmental factors [63, 64, 65, 66]. The clinical outcome of treatment with rHGH in patients with IGHD IA is quite variable. The amount and the affinity of GH-Ab modulated by genetic disposition for immune reactions may determine the overall response to HGH therapy [62].