Main

In animals, TRH, when administered to the mother or the fetus, alone or in combination with corticosteroid, accelerates biochemical, functional, and morphologic fetal lung maturation(1–18). Results of all clinical studies, with the exception of the ACTOBAT trial, also suggest that maternal TRH plus corticosteroid treatment is better than corticosteroid alone in enhancing fetal lung maturation(19–24).

The mechanisms of TRH-mediated fetal lung maturation remain unclear; either the neuroendocrine or the neurotransmitter effects are involved(1–3). The neuroendocrine effect of TRH is believed to occur through placental transfer of TRH and stimulation of the fetal pituitary-thyroid axis, leading to an increase in fetal plasma TH concentration(1, 8, 19–24). TH are known to enhance fetal lung maturity via direct receptor-mediated mechanisms(25, 26). This has been the basis for clinical trials with TRH(19–24). The neurotransmitter effect of TRH is mediated by stimulation of fetal sympathetic or parasympathetic ANS(1, 6, 18, 25). The production and release of lung surfactant is influenced by the ANS(1, 11, 15, 25). Thus acceleration of fetal lung maturation from TRH stimulation could result from either its neuroendocrine or neurotransmitter effects.

The hyt/hyt mouse is a model of primary hypothyroidism due to a mutation (proline to leucine, amino acid 556) of TM4 domain of the β-subunit of the TSH receptor of the euthyroid (Balb-c) mouse(27, 28). Because of this mutation in the hyt/hyt mouse, TSH does not bind to the TSH receptor of the thyroid gland, resulting in primary hypothyroidism(27, 28). In the hyt/hyt mouse fetus and in 1-d-old hyt/hyt pup lungs, ultrastructural maturation is delayed(29). In the euthyroid mouse, fetal lung ultrastructural maturation is accelerated by maternal TRH treatment(18). Therefore this investigation was undertaken to study the effect of maternal TRH treatment on hyt/hyt fetal mouse lung ultrastructural maturation. If TRH effect is mediated through extrathyroidal pathway(s), treatment of the hyt/hyt mouse fetus should enhance lung maturity, but this should not occur if the TRH effect is mediated solely via the fetal pituitary-thyroid axis.

To further characterize primary hypothyroidism in the hyt/hyt mice, we quantitated brain TRH concentrations in the fetus and adult mice, and serum TRH in adult mice. If the hyt/hyt fetus already has a high level of steady endogenous TRH, it may not respond to exogenous TRH treatment.

METHODS

Care of hyt/hyt and Balb-c mice. The protocol for animal experimentation using the hyt/hyt and the Balb-c (+/+) mice was approved by the Animal Care Committee of Saint Louis University Health Sciences Center. Care of these mice was according to the guidelines described by the National Institutes of Health. The colony of genetically hypothyroid (hyt/hyt) mice was established from homozygous (hyt/hyt) males and heterozygous (hyt/+) females (Jackson Laboratory, Bar Harbor, ME)(29). Mice were housed individually and allowed to eat food and drink water ad libitum. T3 was added to the drinking water (0.2 μg/mL) of hyt/hyt mice to restore fertility. Homozygous pups could be easily distinguished from heterozygous litter mates at ≈1 mo of age by low activity, small size and weight, and reduced amount of fur(27, 29). To restore growth and fertility, T3 was added to the drinking water of hyt/hyt pups (0.2 μg/mL). After ≈3 mo of T3 replacement, hyt/hyt pups grew and became fertile. To verify hypothyroidism in these mice, T3 was then stopped for 1 mo. Blood from the orbital sinus of these mice was collected under light methoxyflurane (Metofane, Pitman-Moore Co., Mundelein, IL) inhalational anesthesia(29). A similar protocol was followed for the collection of blood from hyt/+ and Balb-c (+/+) mice. Homozygous hyt/hyt mice were identified by a high serum TSH and a low free T4 concentration (Table 1).

Table 1 Serum-free T4 and TSH concentration in adult Balb-c, hyt/+, and hyt/hyt mice
Full size table

Quantitation of serum-free T4 and TSH concentration in hyt/hyt, hyt/+, and Balb-c mice. Serum-free T4 (Incstar Corp., Stillwater, MN, free T4 two-step γ-coat RIA) and TSH (Amersham Corp., Arlington Heights, IL) concentrations were determined using commercially available RIA kits. The polyclonal primary antibodies used in these RIA kits are raised in the rabbit and have minimal cross-reactivity with other compounds. For both these RIAs the within and between assay coefficient of variation is less than 12%. Homozygous hyt/hyt mice with high serum TSH concentrations had very low free T4 concentrations when compared with the hyt/+ or the Balb-c (+/+) mice (Table 1).

Quantitation of fetal and adult brain TRH and adult serum TRH in hyt/hyt and Balb-c mice. TRH concentrations were determined by RIA in brain and serum samples after extraction(30). Synthetic TRH was used for standards and for radioiodination substrate. TRH polyclonal antiserum generated in rabbits does not cross-react with TRH-glycine, TRH-COOH, or the NH2-terminally extended forms of the TRH precursor(30).

Administration of TRH and animal euthanasia. Homozygous hyt/hyt mice were allowed to mate to produce hyt/hyt fetuses. During the breeding period, male and female hyt/hyt mice received T3 water(29). Females were separated from the males after mating had occurred (+ vaginal plug = d 1 of pregnancy; term, ≈20 d). T3 water was stopped at d 12 of pregnancy to ensure fetal hypothyroidism(27, 29). Saline (0.2 mL) or TRH (0.4 or 0.6 mg/kg in 0.2 mL of saline, Sigma Chemical Co., St. Louis, MO) was injected (i.p.) into the pregnant mouse on d 16 and 17 (b.i.d.) and 1 h before killing on d 18 of gestation. We have established that either concentration of TRH, when administered to the Balb-c mouse on this protocol, accelerates fetal lung ultrastructural maturation(18). On d 18 of gestation, pregnant mice were weighed and then killed (pentobarbitone, 100 mg/kg i.p.). Fetuses were delivered after hysterotomy, weighed, and killed by cervical dislocation. Because of the small size of the pups, enough serum was not available to quantitate serum TSH or free T4 concentrations in each of these litters. However, we have established that hyt/hyt fetuses have a higher serum TSH concentration when compared with the euthyroid controls(29). To decrease variability in the degree of lung maturity, the fetus located most laterally in the right uterine horn was selected as the representative pup within each litter. To eliminate interlobar variability, the upper lobe of the right lung from the representative fetus was saved for EM.

Fetal lung ultrastructure. Morphometric analysis of fetal lung ultrastructure was done as described previously(18, 29). Briefly, the right upper lobe of the lung was minced into 1-mm cubes and fixed in 2% glutaraldehyde in PBS for 2 h. Lung samples were processed through graded acetone and embedded in Spurr (Polysciences, Inc., Warrington, PA). Two blocks of lung were randomly selected for 1-μm sectioning, and of these, one was chosen for thin sectioning based on predominance of saccules and absence of preacinar airways. These blocks were thin sectioned, mounted on copper grids, and counterstained with Reynold's lead and uranyl acetate. For each pup, 10 randomly chosen areas in the section were photographed using a JEM-100S electron microscope (JEOL Ltd., Tokyo, Japan). The photomicrographs were printed at a final magnification of 9000×, and morphometric analysis was done using an interactive computerized image analysis program (Optimas, Bioscan Inc., Edmonds, WA). Type II cells were defined as plump, cuboidal cells with microvilli, present on the surface of an air space. The number of type II cells in a unit area, number of type II cells that contained LB, and number of LB in a unit area were counted. From these data, the average number of LB per type II cell was derived. In addition, the volume densities of intracellular glycogen expressed as the ratio of the glycogen to the nucleus, and the alveolar/parenchymal ratios were derived using electron micrographs, a 100-point grid, and the Chalkley point counting system(31). For each pup, the means of these data from all electron micrographs were calculated. The means for each group, saline- or TRH-treated, were derived. The average measurements in saline- or both TRH-treated (low and high dose) hyt/hyt groups were compared. Investigators who performed the morphometric analysis were unaware of the treatment protocol.

Statistics. All data are means ± SEM. Statistical significance within the saline- or the TRH-treated groups was derived by ANOVA.

RESULTS

TRH concentration. The serum or brain TRH concentration in the adult hyt/hyt mice and the fetal brain TRH concentration in the hyt/hyt mice were similar to that of the Balb-c mice (Table 2).

Table 2 Serum and brain TRH concentration in adult and fetal brain TRH concentration in Balb-c and hyt/hyt mice. (All data are mean ± SEM, n = number of animals). QNS = Quantity not sufficient
Full size table

Maternal and fetal data. Maternal and fetal weights and fetal mortality were similar in the three groups (Table 3).

Table 3 Maternal and fetal data in saline or TRH treated hyt/hyt mice (All data are mean ± SEM, n = litters)
Full size table

Lung EM. The lung morphology of saline-treated hyt/hyt mice at 18 d of gestation in this study is similar to the lung morphology of hyt/hyt mice reported previously(29). When compared with the Balb-c mouse(18), the air spaces were smaller and the epithelial cells plumper, giving the lung a more glandular appearance. Table 4 and Figures 1 to 4 show the results of fetal lung ultrastructural morphometric analysis in saline- or TRH-treated hyt/hyt mice. Fetal lungs in the high (0.6 mg/kg) but not the low (0.4 mg/kg) dose of the TRH group contained more LB per type II cell compared with the saline-treated group. There was an increase in the alveolar to parenchymal ratio in the high dose of TRH group when compared with the low dose TRH- or the saline-treated group (Table 4). These results could reflect an acceleration of lung growth, cellular differentiation, development, maturation, or a combination of these processes. The lungs of the high dose TRH-treated group had more air spaces that were larger with thinner alveolar septae compared with the control or low dose TRH-treated lungs, which appeared more solid. More air-blood barriers and tubular myelin were apparent in the high dose TRH-treated lungs, again reflecting acceleration of structural lung maturation (Fig. 4). The glycogen to nucleus ratio within the type II cells was similar in all three groups (Table 4).

Table 4 Analysis of fetal lung ultrastructural features in saline or TRH treated hyt/hyt mice (*P < 0.05 vs saline treated control, LB = Lamellar bodies)
Full size table
Figure 1
figure 1

Light photograph of hyt/hyt fetal mouse lung at 18 d of gestation pretreated with saline (control) or high dose of TRH. Note that in saline-treated fetal lungs the interstitium is thick and air spaces are poorly formed, giving the lung a very glandular appearance. In high dose TRH-treated fetuses the lungs are more mature with thinner interstitium and better formed air spaces.

Full size image
Figure 4
figure 2

Lamellar bodies (lb) and tubular myelin (tm) were observed in the alveoli of TRH- but not saline-treated fetal lungs.

Full size image

DISCUSSION

We have shown that fetal hypothyroidism in the hyt/hyt mouse is associated with a delay in fetal lung ultrastructural maturation(29). This study confirms our previous findings and further supports the contention that TH play an important role in mammalian fetal lung development(25, 26, 29). In addition, we have observed in the hyt/hyt mouse that high dose maternal TRH therapy accelerates fetal lung maturation, as evidenced by an increase in the number of LB per type II cell and an increased alveolar to parenchymal ratio. These findings point to an important role for TRH in enhancing fetal lung maturation through stimulation of extrathyroidal pathways.

TRH-mediated acceleration of fetal lung maturation may occur either through its neuroendocrine or neurotransmitter actions(1–3). The neuroendocrine effect occurs through stimulation of pituitary prolactin and TSH secretion, whereas the neurotransmitter action occurs via stimulation of the sympathetic and parasympathetic ANS outflow(1–3, 6, 32–34). TRH stimulates neural pathways in the CNS-activating efferent vagal fibers and sympathetic nerve terminals in the adrenal medulla leading to increased plasma epinephrine concentration(32–38). Therefore, the possible mechanisms of the TRH effect on fetal lung include stimulation of: 1) fetal pituitary prolactin secretion, 2) fetal pituitary TSH secretion and then TH production, 3) CNS autonomic sympathetic β-adrenergic or parasympathetic system, and 4) fetal adrenal medullary epinephrine secretion leading to enhanced β-adrenergic effects. In sheep, fetal administration of cortisol + T3 + prolactin increases fetal lung distensibility and stability(39). However, treatment with a combination of any of these two hormones or one hormone alone did not increase fetal lamb lung maturity(39). Although Hamosh and Hamosh(40) reported that prolactin increases fetal rabbit pulmonary surfactant(40), other investigators were unable to observe a similar effect of prolactin in the fetal rat, rabbit, or the human lungs(26, 41–43). We have shown that administration of both TRH or DN1417 (TRH analog) to the pregnant rabbit comparably enhances fetal lung maturation(3). DN1417, like TRH, stimulates CNS neurotransmission and release of epinephrine from the adrenal medulla(3). Although DN1417 has ≈5% activity of TRH in enhancing the secretion of TSH, unlike TRH it inhibits the secretion of prolactin(3). Therefore it is unlikely that TRH-mediated fetal rabbit lung maturation is due to increased prolactin secretion(3). Thus in our previous study with the euthyroid fetal rabbit, the lung maturation effect of DN1417 or TRH could have resulted from stimulation of fetal TSH-TH secretion or through extrathyroidal actions(3). In the present study with the hyt/hyt mouse, endogenous serum TSH concentration is already high(29), and TSH fails to bind to the TSH receptor of the thyroid gland(27, 28). Therefore, the acceleration of fetal lung maturation in the hyt/hyt mouse from high dose TRH stimulation must result, at least in part, from stimulation of extrathyroidal pathway(s).

The extrathyroidal pathway for TRH-mediated fetal lung maturation is believed to be through its neurotransmitter actions(1, 3, 6, 11). It has been established that surfactant synthesis and secretion is influenced by the parasympathetic and sympathetic (β-adrenergic) ANS, and by adrenal medullary catecholamines(1, 11, 25). TRH also stimulates the parasympathetic and sympathetic ANS and causes the release of epinephrine from the adrenal medulla(32, 36). These effects of TRH could increase β-adrenergic effects in the lung. The cholinergic, sympathetic, or β-adrenergic effects of TRH can be individually blocked by atropine, 6-OHDA, or propranolol, respectively(11). By using 6-OHDA and propranolol it has been shown that sympathetic and β-adrenergic mechanisms are involved in fetal lamb lung maturation from TRH + cortisol stimulation, supporting our present observations that the TRH effect on the fetal lung is due to its neurotransmitter rather than to its neuroendocrine action(11). The fetal sheep study, however, has certain limitations. In this study TRH + cortisol rather than TRH alone was administered, and only one pathway of TRH + cortisol effect could be blocked at a time(11). In addition, in sheep, anatomic proof of alveolar sympathetic innervation is lacking(11). Therefore, it is difficult to interpret the results of chemical sympathectomy with 6-OHDA(11). In the mouse, however, unmyelineated axons packed with large dense core vesicles have been demonstrated by EM to coexist in close contact with alveolar type II cells(44). In the hyt/hyt mouse the role of these pathways in mediating TRH effects can be effectively investigated because the pituitary-thyroid axis is blocked naturally, and the ANS can be blocked pharmacologically.

In a previous study from our laboratory we observed that, in the euthyroid Balb-c mouse, maternal TRH therapy with the low as well as the high dose enhanced fetal lung ultrastructural maturation as evidenced by depletion of glycogen in the type II cells, an increase in the number of LB per type II cell, and an increase in the alveolar to parenchymal ratio(18). In the present study of the hyt/hyt mouse, we have observed that the higher dose of TRH was associated with an acceleration of fetal lung maturation as evidenced by an increase in the number of LB per type II cells and an increase in the alveolar parenchymal ratio. However, fetal lung type II cell glycogen content was similar in control and TRH-treated groups. These results suggest that an increase in fetal TH activity after TRH therapy prevents accumulation of glycogen in fetal lung type II cells or increases its utilization. This is further supported by the work of Rooney et al.(45), who showed that, in the fetal rat, maternal TH therapy decreases total lung glycogen content.

It is possible that, in the hyt/hyt mouse, there is a delay in the anatomical or functional maturity of ANS or there is down-regulation of TRH receptors within the CNS. This would explain the lack of response to the low dose of TRH. Administration of TRH, either into the cerebral ventricles or high dose parenterally, has been shown to stimulate rapid and deep breathing movements in the adult rat and rabbit, and in term and preterm newborn rabbits and fetal lambs(9, 46). These changes in respiratory rate after TRH treatment were thought to be due to an effect on the central respiratory centers in the brain stem(9, 46). It is possible that stimulation of fetal respiratory activity by TRH plays a role the acceleration of lung ultrastructural maturation in the euthyroid as well as the hypothyroid fetal mouse.

In the adult hypothyroid (hyt/hyt) mouse, serum TSH concentration was higher and free T4 concentration was lower than in the euthyroid mouse. Although serum or brain TRH concentrations in the adult and brain TRH concentrations in the fetal hyt/hyt mice were lower than in the Balb-c mice, these differences were not significant. These findings were corroborated by Noguchi(47), who reported in the adult hyt/hyt mouse a slight reduction in immunoreactivity for TRH in the median eminence of the hypothalamus. TRH immunoreactivity returned to normal after treatment with T4, suggesting that the reduced TRH content was due to TH deprivation(47). The hyt/hyt mouse model provides an opportunity to study the effect of TH deficiency on fetal growth and development without surgical or pharmacologic manipulation of the intrauterine milieu to induce fetal hypothyroidism.

In summary, we have further characterized hypothyroidism in the hyt/hyt mouse. In addition, we have shown that maternal treatment with a high dose of TRH enhances lung ultrastructural maturation in the hyt/hyt mouse fetus with primary hypothyroidism. These results point to an extrathyroidal pathway for TRH-mediated fetal lung maturation.

Figure 2
figure 3

Electron micrograph of hyt/hyt fetal mouse lung treated with saline (control) or a high dose of TRH. Note that the alveolar space in TRH-treated fetal lungs is bigger and better formed compared with saline-treated fetal lungs.

Full size image
Figure 3
figure 4

Electron micrograph of type II cells from hyt/hyt fetal mouse lungs treated with saline (control) or a high dose of TRH. Type II cells from TRH-treated hyt/hyt fetal lungs showed more lamellar bodies (arrows) when compared with controls.

Full size image