Main

PTHrP was discovered as an agent responsible for the syndrome of humoral hypercalcemia of malignancy. After its purification and cloning in 1987, it has been found in a variety of normal mammalian tissues, including kidney, bone, skin, brain, liver, pancreas, intestine, heart, and lung (seeRefs. 1 and 2 for recent reviews). Located on the short arm of chromosome 12, the PTHrP gene is highly conserved among animal species. In fact, disruption of the PTHrP gene is neonatally lethal(3). As its name suggests, PTHrP shares many structural and functional characteristics with PTH. The first 34 residues of each protein are similar in primary and secondary structures, and comprise the binding region for a common receptor, the PTHrP/PTH receptor. Thus, PTHrP can mimic many of the effects of PTH.

However, PTHrP most likely acts locally because it is undetectable in the circulation using current immunoassays, except in malignancy, in lactating women, and in fetuses(4, 5). PTHrP may regulate smooth muscle relaxation and transepithelial calcium transport in various tissues. Importantly, there is considerable evidence that PTHrP regulates cellular growth and differentiation of various cell and tissue types in a paracrine and/or autocrine manner(6–13). Of note, proliferation and maturation of adult alveolar type II cells may be regulated in this manner(14, 15).

PTHrP and its receptor have been found early in gestation in most types of rat and human embryonic tissues(4, 16–19). PTHrP expression can be detected as early as 7 to 8 wk of gestation in the human lung(17). Lung levels of PTHrP and its mRNA peak shortly before birth in rats(16–19). Interestingly, PTHrP expression is largely restricted to epithelial cells, whereas its receptor is expressed in underlying mesenchymal cells. During fetal lung development, PTHrP is largely expressed in bronchiolar epithelia and alveolar type II cells, while its receptor is expressed in the subepithelial mesenchyme. PTHrP induces time- and dose-dependent increases in cAMP and inositol phosphate in pulmonary fibroblasts, but not in type II pneumocytes from rat fetal lung(20). PTHrP stimulates surfactant phospholipid synthesis (a marker of differentiated type II cell function) in fetal lung explants and co-cultures of fibroblasts and pneumocytes by 75-100%, but not in cultures of pneumocytes alone(20). Thus, mesenchymal regulation of epithelial differentiation in fetal lung tissue is dependent on epithelial derived PTHrP.

The effects on surfactant synthesis suggest that PTHrP may play a role in lung maturation. In recent abstracts, Rubin et al.(21, 22) examined lung development in PTHrP knockout mice, and noted that PTHrP deficiency is associated with lung hypoplasia, arrested canalicular development, and impaired epithelial cytodifferentiation. In addition, PTHrP may be a marker of type II cell maturity. There are no current studies on the recovery of PTHrP from tracheal washings in human newborns. Our goals were to identify and quantitate PTHrP in tracheal aspirates from newborn infants, and to correlate PTHrP levels recovered from the washings with various indices of neonatal lung maturation and severity of lung disease.

METHODS

This study was approved by the University of California at San Diego Human Subjects Committee, and parental consent was obtained before enrollment. Infants (preterm and term) admitted to the Infant Special Care Center at the University of California at San Diego from January through June 1996 who required mechanical ventilation were eligible. Infants with meconium aspiration syndrome, pulmonary hemorrhage, and congenital anomalies (including congenital heart disease and diaphragmatic hernia) were excluded. Tracheal aspirates were collected and handled with a consistent, standardized protocol for all enrolled infants. Specimens were encoded to maintain patient confidentiality. Clinical data were gathered from patient medical records.

Collection of tracheal washings. All tracheal aspirates were obtained within the first 24 h of life. Specimens were collected when routine suctioning of the airway was deemed necessary by the medical team. Approximately 1 mL of sterile normal saline was instilled into the endotracheal tube. After a 10-15-s period of ventilation, the trachea was suctioned, and washings were collected in a mucus trap. The suction catheter was rinsed with approximately 0.5 mL of sterile normal saline into the same mucus trap. Specimens were placed in a freezer at -20°C until analysis. Aspirates were collected before or at least 4 h after surfactant administration.

Clinical definitions. A diagnosis of RDS was suspected based on clinical signs, oxygen requirement, and typical radiographic findings. The severity of RDS was graded arbitrarily by the number of doses of surfactant(i.e. Survanta, Ross Products Division, Abbott Laboratories, Columbus, OH) administered: no RDS/minimal disease-no surfactant treatment or a single dose, and moderate/severe disease-two to four doses. The decision to administer surfactant was at the discretion of the neonatal attending or fellow based on the patient's clinical examination, ventilatory support, and response to previous doses. Maternal steroid exposure was classified as complete, partial, minimal, or none. A complete course was defined as two doses (12 mg each) of β-methasone administered to the mother 24 h apart, with the birth of the infant occurring at least 48 h and up to 1 wk after the initial dose. A second course was administered if delivery was delayed beyond 1 wk of the initial course. A partial course was defined as at least one dose of steroids administered a minimum of 12 h before delivery, and included those infants who did not fulfill the criteria of a complete course. A minimal course was defined as a single dose of steroid that did not fulfill the criteria of a partial course. Diagnoses of small for gestation age or large for gestational age were made if the birth weight was 2 SD or more below or above the gestational age-specific mean, respectively. Chronic lung disease was diagnosed if supplemental oxygen was needed at 36 wk of adjusted gestational age or greater. Air leak was defined as pneumothorax or pulmonary interstitial emphysema observed within the first 24 h of life. Infection was documented on the basis of positive tracheal or blood culture within the first 24 h of life. Intraventricular hemorrhage was graded based on the method of Papile(23).

Detection of PTHrP. Samples were thawed and centrifuged at 120× g for 20 min. RIA was performed on the supernatants using a rabbit polyclonal antibody to PTHrP peptide 38-64, as previously described(24). This antibody does not cross-react with at least a 100-fold excess of human calcitonin, human calcitonin gene-related peptide, human PTH, and PTHrP-derived peptides other than the target peptide. A synthetic human PTHrP fragment, PTHrP 38-64, was used as assay standard. The tracer, human PTHrP 1-86, was prepared by radioiodination of the tyrosine residues using the chloramine-T method. All samples were assayed in multiple dilutions that paralleled the corresponding PTHrP peptide standard. The assay incubations were conducted under standard, nonequilibrium conditions. The intra- and interassay variations were approximately 7 and 12%, respectively. The assay detection limit is 5 pg/tube. The TP per specimen was measured by the Coomassie Blue binding method(25). Values are reported for both PTHrP/TP (pg/mg) and PTHrP (pg/mL) so as to minimize the contribution of protein transudates and fluid shifts that can occur in the newborn lung.

Statistical analysis. A single specimen was obtained in 26 patients. Duplicate samples were obtained within the first 24 h of life in 14 patients. PTHrP concentrations in duplicate specimens from the same patient were averaged. The distribution of PTHrP/TP in the population is displayed in Figure 1. A similar profile was demonstrated for PTHrP(data not shown). Comparative analysis of PTHrP data used a two-tailed Mann-Whitney rank sum test, implemented with the statistical software package SPSS Base 7.0 for Windows (Chicago, IL). PTHrP concentrations were below the limit of detection in 5 of 40 samples. The PTHrP/TP and PTHrP values for these samples were assigned the same rank for statistical purposes. PTHrP/TP and PTHrP levels in tracheal aspirates were compared between neonates divided on the basis of gestational age (≤34 wk versus >34 wk). Other indices or factors associated with lung maturity were also studied, including gender, exposure or lack of exposure to antenatal steroids, and presence of mild versus moderate/severe RDS, as described in the section on clinical definitions above. The t test was used to examine whether gestational age was a confounding factor in comparisons of groups of patients divided by gender, steroid exposure, or RDS. Fisher's exact test was used to compare the incidence of RDS in premature male versus female infants, the proportion of male and female infants receiving antenatal steroids, and the incidence of RDS in infants who were or were not exposed to antenatal steroids. PTHrP/TP and PTHrP data are expressed as median (25th percentile; 75th percentile). Gestational age and birth weight are expressed as the mean (range). Significance was accepted at p < 0.05.

Figure 1
figure 1

Distribution of PTHrP. The number of patients(n = 40) is plotted for each increment of 100 in PTHrP/TP (pg/mg). The symbols are placed at the midpoint of each increment. The line was visually fitted. The data do not follow a normal distribution.

Full size image

RESULTS

We collected tracheal washings within the first 24 h of life from 40 intubated newborn infants. The characteristics of the infants are summarized in Table 1. The study enrolled no infants who were large for gestational age. Two infants, both born greater than 34 wk of gestation, were small for gestational age. Infection was documented in two infants, a term infant with group B streptococcal sepsis and a premature infant with ureaplasma pneumonia. Thirteen infants born less than 35 wk of gestational age received a partial/complete course of steroids. Seven of these neonates received a complete course, and six received a partial course (one dose administered at least 12 h before delivery). Among the other infants, eight had no steroid exposure and two received a minimal course (a single dose administered within 12 h of delivery). Fourteen infants born less than 35 wk of gestational age had severe RDS requiring two or more doses of surfactant. Five premature infants with mild RDS received a single dose of surfactant and another with mild RDS received none. Three premature infants had no signs or symptoms of RDS.

Table 1 Infant characteristics
Full size table

The median levels of PTHrP/TP and PTHrP (25th percentile; 75th percentile) recovered from tracheal aspirates were 168 pg/mg (78; 449) and 107 pg/mL (33; 268), respectively. Because PTHrP may have a role in lung maturation, we compared levels of PTHrP/TP and PTHrP in infants born less than or equal to 34 wk of gestation (n = 23) to those born greater than 34 wk of gestation (n = 17). The gestationally older group had higher PTHrP/TP (p = 0.007) and PTHrP (p = 0.01) than the younger group (Fig. 2). In addition, infants whose birth weights were less than or equal to 2 kg (n = 21), approximately the median weight at 34 wk of gestation, had lower levels than those whose birth weights were greater than 2 kg (n = 19). PTHrP/TP levels were 113 pg/mg (78; 207) versus 246 pg/mg (125; 525) in the lower and higher birth weight groups (p = 0.01), respectively. PTHrP concentrations in the aspirates were 87 pg/mL (20; 218) versus 213 pg/mL (59; 596)(p = 0.04), respectively, in the same groups.

Figure 2
figure 2

Gestational age. The levels of PTHrP/TP (pg/mg) and PTHrP (pg/mL) are presented as the median (center line), upper and lower quartiles (bars), and 10th and 90th percentiles (error bars) for infants born at ≤34 wk of gestation (n = 23,hatched bars) vs those born at >34 wk of gestation(n = 17, stippled bars). Infants at earlier gestational age had lower PTHrP/TP and PTHrP levels than older infants. +p = 0.007; *p = 0.01.

Full size image

Due to the association of RDS and lung immaturity, we restricted our analysis of the effects of varying degrees of RDS to infants born less than or equal to 34 wk of gestation. Patients were divided into two groups based on the number of surfactant doses administered (see “Methods”). Fourteen infants received two or more doses of surfactant, described as moderate/severe disease, whereas nine infants received no surfactant or only one dose and were described as having no RDS/mild RDS. The infants with moderate/severe RDS had significantly lower levels of PTHrP/TP and PTHrP than infants with no RDS/mild disease (Fig. 3). There was no significant difference in the mean gestational ages between those infants with moderate/severe RDS and those with no RDS/mild disease [27(24, 32) wk versus 28 wk(23, 34), p = 0.75].

Figure 3
figure 3

RDS. The levels of PTHrP/TP (pg/mg) and PTHrP (pg/mL) are plotted as the medians, quartiles, and percentiles for infants with moderate/severe RDS (n = 14, hatched bars) vs those with no RDS/minimal disease (n = 9, stippled bars). PTHrP/TP levels and PTHrP concentrations in tracheal aspirates were inversely related to the severity of RDS. The data are restricted to infants born ≤34 wk of gestation (n = 23). +p = 0.005; *p = 0.009

Full size image

We compared PTHrP/TP and PTHrP levels among premature infants born less than or equal to 34 wk of gestational age whose mothers did and did not receive antenatal steroids for accelerated lung maturation in anticipation of premature delivery (Fig. 4). The use of antenatal steroids was categorized as none/minimal (n = 10) versus partial/complete (n = 13) (see “Methods”). Patients receiving none/minimal steroids had significantly lower PTHrP/TP levels and PTHrP concentrations in tracheal aspirates than patients exposed to a partial/full course of steroids (p = 0.01). There was no significant difference in the mean gestational ages between those infants whose steroid exposure was none/minimal and those who received a partial/complete course of antenatal steroids [27(24, 34) wk compared with 28(23, 34) wk, respectively, p = 0.63]. Four of the 10 male infants received a partial/complete course of antenatal steroids versus 9 of the 13 female infants (p = 0.22). Six of the 13 infants whose mothers received a partial/complete course of antenatal steroids developed moderate/severe RDS versus 8 of the 10 infants whose mothers received none/minimal course (p = 0.15).

Figure 4
figure 4

Antenatal steroids. For infants born <34 wk of gestation (n = 23), the levels of PTHrP/TP (pg/mg) and PTHrP (pg/mL) are plotted as the medians, quartiles, and percentiles for infants receiving none or minimal antenatal steroids (n = 10, hatched bars)vs those receiving a partial or complete course (n = 13,stippled bars). Steroid administration was associated with increased levels of PTHrP/TP and PTHrP. The data are restricted to infants born ≤34 wk of gestation (n = 23). +p = 0.04; *p = 0.01.

Full size image

Finally, we compared PTHrP/TP and PTHrP levels between male and female infants born less than or equal to 34 wk of gestation and greater than 34 wk of gestation. We found no difference for those born at greater than 34 wk of gestation, but levels of PTHrP/TP levels and PTHrP were lower in male infants(n = 10) than in female infants (n = 13) born less than or equal to 34 wk of gestation. PTHrP/TP and PTHrP levels were 57 pg/mg (37; 133) and 25 pg/mL (18; 91), respectively, in male infants compared with 199 pg/mg(88; 249) and 114 pg/mL (52; 247), respectively, in female infants. There was no significant difference in the mean gestational ages between male and female infants born less than or equal to 34 wk of gestation [28(24, 34) wk for male infants, 28(23, 34) wk for female infants, p = 0.74]. Nine of the 10 male infants had moderate/severe RDS, based on surfactant dosing, versus 5 of the 13 female infants (p = 0.03).

DISCUSSION

We have shown that PTHrP levels in tracheal washings collected from intubated newborn infants within the first 24 h of life are significantly lower in infants born less than 35 wk of gestation and with a birth weight less than 2 kg. This is consistent with earlier observations that expression of PTHrP in fetal lung epithelium is gestationally dependent, peaking shortly before birth(4, 16–18). We also found significantly lower PTHrP levels in preterm infants (<35 wk of gestation) who required multiple doses of surfactant. Insofar as the need for surfactant reasonably reflects the severity of RDS in the preterm infant, it appears that preterm infants under 35 wk of gestation who have lower PTHrP levels are at greater risk for RDS. Thus, PTHrP may be a marker of lung maturation in the neonate. We found no association between PTHrP levels and the incidence of chronic lung disease, possibly reflecting the multiple etiologies for this disorder.

A relationship between PTHrP and lung maturation is not surprising because PTHrP has known effects on epithelial growth and function in fetal and adult lung. Rubin et al.(20) demonstrated that PTHrP stimulated surfactant phospholipid synthesis in cultured rat fetal lung epithelium. Their observations are consistent with our results that demonstrate lower PTHrP levels in the surfactant-deficient lungs of preterm infants with moderate-to-severe RDS. Interestingly, they also noted that PTHrP-dependent stimulation of surfactant production in rat fetal epithelium required co-culture with fetal lung fibroblasts. Apparently, epithelial-derived PTHrP acts on underlying fibroblasts, which in turn regulate epithelial surfactant production through as yet unknown mechanisms. This conclusion is supported by evidence from in situ hybridization studies of the localization of the PTHrP receptor in underlying mesenchymal tissue(19). Mesenchymal-epithelial interactions are important in fetal lung development(26). Thus PTHrP may trigger maturational changes in the neonatal lung. PTHrP appears to have direct effects on pneumocytes in adult lung. Hastings et al.(14) demonstrated that type II cells express a receptor for PTHrP and that endogenous PTHrP increases the number of lamellar bodies per cell and stimulates production and secretion of phosphatidylcholine in cultured adult type II epithelial cells. These properties are characteristic of the differentiated type II cell. In addition, this group(15) showed that PTHrP inhibited the proliferation of type II cells. Thus, PTHrP has been identified as a paracrine growth regulator in fetal lung tissue and an autocrine factor for adult type II cells, but additional mechanisms of action have not been excluded.

PTHrP and PTH are ligands for several recently cloned receptors. Adult alveolar type II cells express the type I receptor PTHrP/PTH receptor, a member of the family of seven transmembrane domain, G protein-coupled receptors(27). This receptor binds the amino terminal portion of PTHrP, PTHrP 1-34 or 1-36. PTHrP 1-36 appears to be responsible for the growth related effects of PTHrP in adult lung and type II cells(15) and has similar effects on a variety of other cell types, including keratinocytes and osteosarcoma cells(12, 28). Midregion and carboxy-terminal fragments of PTHrP also have physiologic effects in some tissues(29). For example, PTHrP 38-94 stimulates transplacental calcium transport. The assay in this study used an antibody directed against PTHrP 38-64.

We found lower PTHrP levels in preterm infants born under 35 wk of gestation who were not exposed or minimally exposed to antenatal steroids. Antenatal steroids reduce the incidence of RDS in the preterm infant by stimulating surfactant production(30–33). It would appear from our data that maternal steroid administration leads to higher PTHrP levels in the fetal lung. Therefore the reduction in the incidence of RDS following antenatal steroid exposure is accompanied by a rise in PTHrP in the fetal lung. The benefits of antenatal steroid exposure are evident at least 24 h after initial administration(31), and may be conferred to some extent to preterm infants born earlier than 24 h postadministration(34). This is consistent with our observations that preterm infants receiving a partial course of antenatal steroids(i.e. a single dose of steroids administered at least 12 h before delivery) have higher levels of PTHrP in tracheal aspirates. In this limited study we did not demonstrate a significant reduction in the incidence of moderate/severe RDS (as defined in “Methods”) for those infants exposed to antenatal steroids. Contrary to our observations, a few studies suggest that glucocorticoids decrease PTHrP production in cultures of nonpulmonary cells(35, 36). It is unclear how these observations related to our findings. Regulation of PTHrP expression is cell-type specific and has not yet been studied in type II cells.

Finally, we found lower PTHrP levels in male compared with female preterm infants born under 35 wk of gestation. RDS occurs more frequently in premature male than female infants(37, 38). The observation of lower PTHrP levels in males is consistent with our other results. When compared by gestational age, birth weight, severity of RDS, antenatal steroid exposure, or gender, PTHrP levels were lower in infants with less mature lung function. However, we cannot separate the effect of gender from the influence of severity of RDS or antenatal steroid exposure. Nine of the 10 male infants born under 35 wk of gestation had moderate-to-severe RDS compared with only 5 of the 13 female infants (p = 0.03). In addition, more female than male infants received antenatal steroids, although the difference in proportion was not statistically significant. A larger study would be necessary to stratify PTHrP expression by the different measures of lung maturation.

Bronchoalveolar lavage and collection of tracheal aspirates are common methods for sampling the fluid lining the airways in adults and children. However, the techniques have not been standardized, and interpretation of data obtained from these fluids is not straightforward because of potential procedural artifacts(39). For example, lavage with small volumes may not recover samples representative of all the airways while tracheal aspirates may reflect predominantly the upper airways in large infants and small children. In contrast, tracheal aspirates from newborns on the first day of life represent predominantly alveolar liquid, as indicated by the high concentrations of surfactant phospholipids(40). Thus, we believe that the PTHrP measured in tracheal aspirates in our study is derived predominantly from alveolar type II cells, but we cannot exclude a source in the upper airways. Another problem is the selection of an appropriate denominator to correct concentrations for dilutional effects. Albumin and total protein have been used, but the concentration of these substances in air space liquid may vary among disease states. We addressed the dilutional problem by using a consistent protocol for obtaining aspirates in all patients. In addition, we analyzed our data as PTHrP concentrations without normalization and as PTHrP normalized to total protein. Because the conclusions from both analyses were the same, we believe that our measurements accurately reflected the relative levels of PTHrP in air space liquid in the different groups of neonates.

The aim of this study was to investigate clinical correlates of the relation between PTHrP and alveolar maturation that had been previously established with cultured cells and explants. We showed that PTHrP levels in tracheal aspirates collected within 24 h of birth from intubated newborn infants are associated with various indices of lung maturation, including gestational age and birth weight, occurrence of RDS and effects of antenatal steroids. Thus, PTHrP appears to be a marker of neonatal lung maturation. PTHrP is known to regulate alveolar epithelial function and differentiation in fetal and adult lung tissue. Hence, our results are consistent with a potential role for PTHrP as a mediator of differentiation in the fetal and neonatal lung, although this issue is not directly addressed in our study.