Published ahead of print on August 6, 2003, doi:10.1164/rccm.200304-479OC
© 2003 American Thoracic Society Surfactant Protein Profile of Pulmonary Surfactant in Premature InfantsNeonatology, Department of Pediatrics, and Department of Anesthesiology and Critical Care Medicine, University of Pennsylvania School of Medicine, Children's Hospital of Philadelphia, Philadelphia, Pennsylvania; and Neonatology, Department of Pediatrics, University of Missouri-Kansas City School of Medicine, Children's Mercy Hospital, Kansas City, Missouri Correspondence and requests for reprints should be addressed to Philip L. Ballard, M.D., Ph.D., Division of Neonatology, Room 416, Abramson Research Center, 3517 Civic Center Boulevard, Philadelphia, PA 19104. E-mail: ballardp{at}email.chop.edu
Although premature infants are known to be deficient in pulmonary surfactant, there is limited information regarding surfactant protein (SP) composition. To assess the postnatal profile of SPs, tracheal aspirate samples were collected from 35 intubated infants of 23–31 weeks of gestation between 8 and 80 days of age. In 71 large aggregate surfactant samples that had normal in vitro function (minimum surface tension of less than 1 mN/m by pulsating bubble surfactometry), mean ± SEM contents of SP-A, SP-B, and SP-C (3.7 kD) were 7.1 ± 1.4%, 1.8 ± 0.2%, and 4.6 ± 0.6%, respectively, of phospholipid. To assess SPs in the 1st week of life, we analyzed samples from additional infants receiving only synthetic replacement surfactant. On the 2nd day of life, contents of SP-A, SP-B, and SP-C were 13.4%, 8.4%, and 0.1%, respectively, of the mean levels for Day 8–80 samples. The major postnatal increases for SP-A, SP-B, and SP-C occurred during the 1st, 2nd, and 3rd weeks, respectively. We conclude that surfactant of newborn premature infants is markedly deficient in SPs, in particular SP-C. Despite continuing lung disease, some infants who are more than 1 week of age have surfactant with normal in vitro function that contains SPs at levels comparable to adult surfactant.
Key Words: surface tension • tracheal aspirate • bronchopulmonary dysplasia Pulmonary surfactant is a mixture of lipids and proteins that is synthesized and secreted by lung alveolar type II cells. The major phospholipid (PL) component, saturated phosphatidylcholine, forms a highly surface active film at the air–aqueous interface in the alveolus and maintains alveolar expansion at end expiration. The surfactant-associated proteins, in particular surfactant protein (SP)-A, SP-B, and SP-C, interact with surfactant PLs to facilitate film formation, stability, and recycling. SP-A is required for formation of tubular myelin, an intermediate structural form of secreted surfactant, participates in formation of the surface active film, and recycles PL into the type II cell. Pulmonary function in mice with SP-A gene ablation is essentially normal, indicating that this protein is not required for the formation of a functional surfactant film. However, this protein has an important role in lung immune defense and modulation of the inflammatory response to infection and other stimuli (1). SP-B is synthesized in pulmonary type II and Clara cells as a 40-kD precursor peptide. In type II cells, this protein undergoes several proteolytic processing steps to produce mature SP-B of 8 kD, whereas processing in Clara cells is incomplete and mature SP-B is not produced (2). SP-B has a critical role in both lamellar body genesis and surfactant function. Newborn mice with a genetic absence of SP-B die of respiratory failure, and full-term human infants with inherited SP-B deficiency have severe, intractable respiratory distress (3, 4). In both of these situations, there is also deficiency of mature SP-C because of impaired formation of lamellar bodies where SP-C undergoes final processing (5). SP-C is a hydrophobic peptide that is synthesized as a 21-kD propeptide that is processed to a 3.7-kD mature protein. Like SP-B, SP-C is intimately associated with pulmonary surfactant and participates in both film formation and stability (5). Mice genetically deficient in SP-C have no respiratory difficulties at birth but exhibit surfactant instability at low lung volume and later develop pneumonia and emphysema (6, 7). Infants born prematurely at less than 32 weeks of gestation are deficient in pulmonary surfactant and often develop respiratory distress syndrome. Previous studies have used tracheal aspirate (TA) lavage fluid to examine components of surfactant and other constituents from airways of premature infants (8–12). This procedure is used clinically for pulmonary toilet of intubated infants and samples the epithelial lining fluid of conducting airways. Surfactant isolated from TA of term infants is surface active, and the composition of both PLs and phosphatidylcholine molecular species is comparable to that of human alveolar surfactant (13). Studies of unfractionated TA from newborn premature infants have found markedly reduced levels of SP-A and SP-B as well as of surfactant PL in infants with respiratory distress syndrome compared with infants without respiratory distress syndrome (8–11). Currently, there are no data for SP-C concentrations in surfactant of premature infants and relatively little information regarding the SP composition beyond the first week of postnatal life. The goal of this study was to determine the profile of SP-A, SP-B, and SP-C in large aggregate surfactant isolated from TA of premature infants. We wished to establish both the postnatal developmental pattern and the concentrations of SPs in infant large aggregate surfactant that had normal function in vitro (minimum surface tension [STmin] of less than 1 mN/m). The findings establish that some premature infants with continuing lung disease have highly surface active surfactant that contains SPs at concentrations comparable to normal adult surfactant. These data provide reference values for studies of SP composition in infants who display symptoms of surfactant insufficiency or dysfunction. Some of these results have been previously reported in the form of an abstract (14).
Patient Population Between July 1997 and December 2001, infants of 23–31 weeks of gestation without congenital anomalies and requiring intubation at birth for respiratory support were enrolled in clinical protocols for studies of infant lung disease. Infants born at the hospital at the University of Pennsylvania were enrolled on the first day of life, and additional infants from the other collaborating nurseries were enrolled at 1–3 weeks of age in a clinical study of inhaled nitric oxide. All protocols included parental consent and were approved by institutional review boards. TA was collected from clinically stable infants on Days 2, 4, and 7 of life and weekly thereafter until extubation. In this sampling procedure, a catheter was placed at the end of the endotracheal tube, and three separate 0.5-ml aliquots of saline were instilled and collected by gentle suction. TA samples were also collected from eight newborn (Days 2–5) term infants without lung disease who were intubated for surgical procedures. Residual TA that was available after other studies was pooled to provide a term sample.
Processing of TA Samples
Surface Tension Measurements
Antibodies
Immunoblotting
Statistical Analysis
SP Profile of Infant Surfactant with Normal Function The first goal of this study was to define the protein composition of large aggregate surfactant with normal in vitro function, defined as a STmin value less than 1.0 mN/m in the pulsating bubble surfactometer. To avoid interference by replacement surfactant given shortly after birth, we limited our sample selection for this study to TA samples collected after the first week of life. We identified 71 samples that had normal surfactant function; Table 1 summarizes the surface tension values obtained for these samples in surfactometer studies.
The 35 infants ranged between 23 and 30 weeks of gestation, and samples were obtained between 8- and 80-days postnatal age (Table 2) . Most of the infants received both antenatal betamethasone and postnatal surfactant treatment, and 26% received a course of postnatal dexamethasone. As expected in view of the requirement for intubation beyond 7 days of age, most of the infants (66%) had both respiratory distress syndrome and adverse outcome defined as bronchopulmonary dysplasia or death at 36 weeks of postmenstrual age.
Western blot analysis was performed on randomly selected TA samples, before determining surface tension, to examine the immunoreactive forms of SPs in large aggregate surfactant (Figure 1) and supernatant (data not shown). SP-A was detected in both supernatant and surfactant pellet, in variable amounts, as singlet or doublet bands at approximately 35 kD as expected. For SP-B, surfactant pellets contained only 8-kD or 8- and 9-kD bands representing mature SP-B. Low amounts of mature SP-B were present in the supernatant with the major forms detected at 20–25 kD and representing precursor SP-B forms. With the SP-C antibody, only mature SP-C (approximately 4 kD) was detected in the surfactant pellet, and no immunoreactive bands were seen in supernatants.
Immunodot assay was performed to quantitate SP content in large aggregate surfactant samples. Content was normalized to both PL and total protein, and results are summarized in Figure 2 . The mean content of SP-A, SP-B, and SP-C was 7.1%, 1.8%, and 4.6% of PL by weight, respectively. Content of each SP was slightly higher when expressed as the percentage of total protein in large aggregate surfactant, and collectively, the three SPs represented approximately 20% of the protein that was present in the large aggregate fraction.
Table 3 compares these data for SP content of premature infant surfactant with data for a pooled term infant sample, results reported in the literature for normal adults, and three commercially available bovine surfactant preparations. The SP-A content of premature infant surfactant beyond the first week of postnatal life is similar to that in adult bronchoalveolar lavage and in the term sample. SP-B content is at the lower end of the range of values reported for adults and is comparable to that found for term infants. Infant SP-C concentration is approximately twofold to threefold higher than the single values reported for adults and term infants. In contrast, commercial surfactants do not contain SP-A and have a lower level of SP-B compared with values in surfactant of adults and term infants.
Postnatal Ontogeny of SP Content The second goal of this study was to define the postnatal pattern of SPs in infant large aggregate surfactant. The 71 TA samples with normal in vitro surfactant function were collected between 2 and 11 weeks of age, and these data were examined for age-related correlations. Over this time interval, there was no significant correlation of SP-A and SP-B contents (% PL) with postnatal age by regression analysis using log-transformed data. In contrast, content of SP-C increased between the second and 11th week of life (r = 0.43, p = 0.01). There were two potential difficulties in examining SP content during the first week of life. Because most infants received one or more doses of animal-derived replacement surfactant in the first day of life (Survanta), TA samples from these infants would contain variable amounts of exogenous PL, SP-B, and SP-C derived from surfactant treatment. To circumvent these confounders, we identified a separate group of infants who had received synthetic replacement surfactant (Exosurf) that does not contain SPs. Because PL from Exosurf would be present in the large aggregate surfactant isolated from these infants, we elected to express SP content as percent of total protein in the large aggregate fraction. Thirty-one TA samples from 22 Exosurf-treated premature infants were available from the first week of life. This group of infants ranged from 24 to 31 weeks of gestation (mean ± SEM, 27.8 ± 0.3); 79% had received antenatal corticosteroid treatment, and 63.6% had an adverse outcome (death or bronchopulmonary dysplasia at 36-weeks postmenstrual age). The results for SP content of surfactant from birth to 9 weeks of age, grouped by postnatal age intervals, are shown in Figure 3 . SP-A was detected in all samples collected on the second day of life, with the mean value of approximately 15% of that found at 2–9 weeks. SP-A content was several-fold higher by the fourth day of life, comparable to levels found at Weeks 2–9. SP-B content (Figure 2B) on the second day of life was approximately 10% of the mean value at Weeks 2–9, and content increased markedly between Days 5 and 14. SP-C content (Figure 3C) was barely detectable in samples collected on the second and fourth day of life and increased significantly by regression analysis (r = 0.55, p = 0.001) over subsequent weeks of life in agreement with data expressed as percentage PL.
For comparison, the protein/PL ratio was determined for samples of large aggregate surfactant with normal function (Figure 3D). There was no change in total protein content from 2–9 weeks with a mean value for protein/PL of 0.7. Thus, large aggregate surfactant from premature infants as isolated by a single centrifugation step is approximately 40% protein by weight. To assess whether the relatively low concentration of SPs on Days 2 and 4 might reflect in part relatively high amounts of airway protein, we calculated total protein recovered in TA of Exosurf-treated and Day 8–80 infants. Protein yield on Days 2 and 4 was 30.6 ± 11.3 and 29.0 ± 5.1 µg/ml of TA compared with 59.6 ± 5.6 µg/ml for the TA samples at Days 8–80 (p < 0.01 vs. both Day 2 and Day 4 by unpaired t test).
This study examined the protein components of pulmonary surfactant in a population of very premature infants who received both antenatal corticosteroid treatment and postnatal replacement surfactant. In surfactant samples beyond the first week of life, selected for normal in vitro function, the SPs were present at levels comparable to that reported for term infants and normal adults. To our knowledge, these results are the first data for concentrations of the three SPs in infant large aggregate surfactant with normal function. As such, they provide useful reference values for studies of infants with symptoms of surfactant insufficiency or dysfunction. We used a rigorous definition of normal surface tension properties (STmin < 1.0 mN/m) in this study; however, SP results were similar with a higher cut-off value of 5 mN/m. Surfactant with abnormal in vitro function (defined as STmin > 5 mN/m) was present in over half of the TA samples from more than 7 days of age that were analyzed and will be described in a separate publication. The current findings indicate that many infants can achieve normal surfactant function and adult levels of SPs despite continuing lung disease. Surfactant samples obtained from premature infants within the first week of life contained low levels of SP-A and SP-B, and SP-C was barely detectable. We found distinct developmental patterns for the three SPs after birth. SP-A concentration increased rapidly over the first few postnatal days, with slower increases for SP-B and particularly SP-C. Each of these postnatal patterns represents accelerated maturation of SP production as compared with the pattern occurring in utero (28). The delayed appearance of the hydrophobic SPs in secreted surfactant may relate to the requirement for extensive posttranslational processing and relatively slow development of processing enzymes. Both SP-B and SP-C are produced as precursor proteins that undergo several proteolytic cleavages at the carboxy and amino termini to produce mature, active proteins. Although SP-B and SP-C gene expression is relatively robust at 24 weeks of gestation (28), little mature protein is detected in lung tissue at this time (19, 20, 29). Similarly, in the developing rabbit fetus, the content of mature SP-C in lavage is low until late in gestation (30). One effect of premature delivery and air breathing may be induction of key proteases for processing of precursor SP-B and SP-C. In contrast, SP-A gene expression is very low in fetal lung at 24 weeks of gestation (28). Thus, the rapid accumulation of SP-A in secreted surfactant after birth apparently results from increased SP-A gene expression. Our finding of low concentrations of SP-A and SP-B in surfactant of the newborn premature infant and subsequent postnatal increases is in agreement with previously published data for these two proteins in amniotic fluid (31) and in unfractionated TA or bronchoalveolar lavage samples (8, 9, 11, 12). It is well recognized that premature infants are deficient in the amount of secreted surfactant, providing the rationale for replacement therapy. In addition, surfactant obtained from newborn premature infants is dysfunctional in vitro with high STmin values (11, 32–34). Our data and previous findings indicate that a low content of SP-B/SP-C likely contributes to surfactant dysfunction after premature birth. It is noteworthy that content of all SPs were low on the second day of life (less than 15% of normal) despite antenatal corticosteroid therapy in most of the cases. In studies of newborn premature lambs, antenatal glucocorticoid treatment maximally increased lavage SP-A and SP-B 10-fold and 15-fold, respectively; however, induced levels were less than 20% of term values (21). Moreover, a relatively brief exposure to antenatal glucocorticoid improved pulmonary function at times when there was no increase in surfactant lipids or proteins (21). An early response to glucocorticoids may be condensation of the mesenchymal tissue, increasing airspace volume and compliance and facilitating surfactant function. We focused our studies on the large aggregate fraction of surfactant that represents the most surface active form isolated from airways. By design, we used large aggregate surfactant without further purification, allowing us to evaluate surfactant activity in the presence of any contaminating and potentially inhibitory substances. In addition to achieving low STmin values, these samples had values of maximal surface tension and adsorption surface tension comparable to purified natural surfactant and commercial surfactant preparations (27). As expected, we found only mature forms of SP-B and SP-C in large aggregate surfactant, whereas the supernatant fraction contained predominantly precursor forms of SP-B. These forms could arise from type II cells, secreted via a constitutive pathway independent of lamellar body secretion, and/or from Clara cells where full processing does not occur (35). The possible biological roles of partially processed SP-B are not known. Because of limited amounts of individual TA samples, we did not address PL composition in this study; however, it is known that premature infants with lung disease have low or absent levels of phosphatidylglycerol (32). In the study by Mander and colleagues (36), surfactant from children with lung disease had reduced percentage of disaturated (16:0/16:0) phosphatidylcholine, which was associated with reduced in vitro surfactant function. Despite the likelihood of similar alterations of PL composition in many of our surfactant samples, normal function was observed in terms of surface properties found in vitro. There are several limitations to our study. It is possible that surfactant obtained from TA has a different composition from alveolar surfactant. Although large aggregate surfactant has not been previously studied in preterm infants, several investigators have found similar composition for epithelial lining fluid constituents obtained from TA compared with segmental lung lavage (37–39). In an analysis of surfactant isolated from TA of healthy term infants, compositions of both PL and PC molecular species were comparable with surfactant prepared from bronchoalveolar lavage, and all term samples achieved a low STmin in the pulsating bubble surfactometer (13). Also, the focus on large aggregate surfactant rather than whole TA fluid reduces the possible effects of contamination of TA samples by airway secretions. Our patient population consisted of intubated infants requiring supplemental oxygen and ventilatory support, and it is possible that surfactant composition is altered despite normal function as determined in vitro. We note, however, that SP composition in surfactant of these premature infants was comparable to that found in a pooled sample from term infants (without lung disease) and in normal adults. A separate group of Exosurf-treated infants were chosen for analysis of SP composition in the first week of life. We considered the possibility that the total protein content of large aggregate surfactant was increased in these infants compared with infants who are more than 7 days of age, thereby reducing the apparent SP content. However, we found reduced yield of large aggregate protein in the newborn period compared with the later TA samples. Finally, we used an in vitro assay of surfactant function that may not necessarily reflect surface properties in vivo. In a recent study, however, Mander and colleagues (36) found a correlation between surfactant function in vitro and lung function (FEV1) in the patients. In summary, we found that newborn premature infants of less than 32 weeks of gestation uniformly have low concentrations of SP-A and SP-B and extremely low levels of SP-C, in large aggregate surfactant isolated from TA samples. The three SPs have different postnatal developmental profiles with SP-A increasing rapidly and SP-C requiring several weeks to reach plateau values. The concentrations of SPs in surfactant with normal in vitro surface properties are comparable to levels found in normal term infants and adults. Our findings indicate that premature infants have both surfactant deficiency and dysfunction at birth and that normal surfactant composition and function can occur after the first week despite lung disease.
The authors thank J.R. Wright for purified SP-A, W. Steinhilber for SP-C antibody, and Forest Laboratories for Infasurf. We are grateful for statistical assistance from A. Haque and A. Cnaan and for technical support from Y. Ning and T. McDevitt, and we thank C. Coburn, K. Mooney, and other clinical personnel who participated in collection of clinical data and samples.
Supported by National Institutes of Health grants HL56401, HL62514, and MO-RR00240 and by the Gisela and Dennis Alter Endowed Chair in Pediatric Neonatology (P.L.B.) and the Endowed Chair in Critical Care Medicine (R.I.G.) of the Children's Hospital of Philadelphia. Conflict of Interest Statement: P.L.B. has no declared conflict of interest; J.D.M. has no declared conflict of interest; R.I.G. has no declared conflict of interest; M.H.G. has no declared conflict of interest; W.E.T. has no declared conflict of interest; R.A.B. has no declared conflict of interest. Received in original form April 3, 2003; accepted in final form August 6, 2003
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
