Expression and Activity of -Defensins and LL-37 in the Developing Human Lung
http://www.100md.com
免疫学杂志 2005年第3期
Abstract
Immaturity of innate immunity contributes to the increased susceptibility of human neonates to infection. The lung is a major portal of entry for potential pathogens in the neonate, and human -defensins (HBDs) and LL-37 participate in pulmonary innate immunity. We hypothesized that these antimicrobial factors would be developmentally regulated, expressed by neonatal pulmonary tissues, and participate in neonatal innate immunity. We found HBD-2 to be the predominant -defensin in human neonatal lung. HBD-2 mRNA expression was developmentally regulated, induced by the proinflammatory factor IL-1, and decreased by dexamethasone. Additionally, HBD-2 abundance in neonatal tracheal aspirates increased as a function of gestational age. HBD-1 had a lower level of expression compared with HBD-2 and was induced by dexamethasone. HBD-3 and LL-37 messages were not detected in airway epithelial cultures. Additionally, each antimicrobial peptide exhibited a unique spectrum of antimicrobial activity and salt sensitivity against bacteria commonly causing sepsis in the neonate. Lower levels of HBD-2 may be one factor contributing to the increased susceptibility of premature infants to pulmonary infections.
Introduction
The morbidity and mortality rates for infectious diseases are higher during the neonatal period. This increased susceptibility to infection has been related to the immaturity of the neonatal immune system. At birth, the immune system exhibits deficiencies in Th-1 immune responses (1), Ig synthesis (for a review, see Ref.2), and selected cytokine production (1, 3). Immaturity of antimicrobial innate immune factors may also be involved in the increased susceptibility of neonates to infection. For example, neutrophils of newborn infants are selectively deficient in bactericidal/permeability-increasing protein (4). Additionally, human -defensins 5 (HD-5)3 and HD-6 have significantly lower levels of expression in fetal life compared with that in term infants (5, 6). As a major portal of entry for potential pathogens during the neonatal period, the lung has the second largest epithelial surface area exposed to the outside environment. Components of the innate immune repertoire of airway epithelia include human -defensins (HBDs) (7, 8) and the only known human cathelicidin, LL-37 (9, 10). Defensins and LL-37 are also present in the skin and vernix and contribute to the innate immunity of neonates (11, 12). Recently, HBDs and LL-37 were shown to be increased in neonatal tracheal aspirates during infections (13); however, little is known about the development or regulation of human -defensins and cathelicidins in neonatal lung.
In this study we investigated the ontogeny, expression, and regulation of three well-characterized -defensins, HBD-1, -2, and -3, and the cathelicidin, LL-37, in the neonatal lung and determined their spectrum of antimicrobial activity. We hypothesized that these antimicrobial factors would be developmentally regulated, expressed by neonatal pulmonary tissues, and participate in innate immunity. We found HBD-2 to be developmentally regulated, inducible, and the predominant -defensin in the neonatal lung. Additionally, the -defensins and LL-37 exhibited unique spectrums of antimicrobial activity and salt sensitivity against neonatal pathogens.
Materials and Methods
Reagents
Recombinant HBD-1 was purified from a baculoviral production system as previously described (14). LL-37 was produced synthetically as previously described (15). All peptides, including recombinant HBD-2 and HBD-3 (PeproTech, Rocky Hill, NJ), were analyzed for purity and concentration using mass spectroscopy and amino acid composition. Peptides were stored in 0.02% acetic acid at –80°C. Before use in antimicrobial assays, peptides were lyophilized and then reconstituted in 0.02% acetic acid with 0.1% human serum albumin.
Culture methods and human samples
All tracheal aspirate specimens and culture samples used in these studies were obtained using protocols approved by the institutional review board at the University of Iowa.
Human fetal lung explant cultures
Midgestational fetal lung explants were cultured as previously described (16). Briefly, 18–22 wk gestation lung tissues were dissected free of major airways and blood vessels, cut into 1- to 2-mm pieces, and cultured at 37°C in serum- and hormone-free Weymouth’s MB752/1 medium (Invitrogen Life Technologies) supplemented with 1% penicillin/streptomycin (Invitrogen Life Technologies; 15140-122) and 0.1% amphotericin B (Cellgro; 30-003-C1) in a humidified atmosphere with 5% CO2. Tissues were incubated in either control medium or medium containing 100 ng/ml recombinant human IL-1 (Sigma-Aldrich; I-9401), 100 ng/ml IFN- (Sigma-Aldrich; I-3265), 100 ng/ml both IL-1 and IFN-, 10–7 M dexamethasone (Roxanne), 106 CFU heat-killed Escherichia coli DH5 bacteria, or 100 ng/ml PMA (Sigma-Aldrich; P-1585). Explants were maintained in culture for 24 h. The tissue samples were then frozen and stored at –80°C before analysis by RT-PCR and real-time PCR. For each experimental condition, replicates were performed using tissues from different donors (n = 4–5).
Primary cultures of human airway epithelia
Primary cultures of well-differentiated human airway epithelia were prepared and grown at the air-liquid interface as previously described (17). Epithelial cultures were incubated at 37°C in 5% CO2 for 24 h either with 100 μl of PBS apically and 500 μl of medium basolaterally or with the same volumes of apical PBS and basolateral medium containing 100 ng/ml IL-1, 100 ng/ml IFN-, 100 ng/ml both IL-1 and IFN-, 10–7 M dexamethasone, or 100 ng/ml TNF- (Sigma-Aldrich; T-6674). For each experimental condition, replicates were performed using primary cultures from different donors (n = 3).
Neonatal tracheal aspirate samples and lung tissue
After obtaining informed consent from the parents, 14 mechanically ventilated newborn infants in the neonatal intensive care unit at the University of Iowa were enrolled during an 18-mo study period (May 2000 to November 2002). Tracheal aspirates were obtained using a standardized procedure for routine suctioning by General Clinical Research Center staff with a catheter inserted through the endotracheal tube and instillation of 1 ml of normal saline. When possible, samples were obtained on postnatal days 0, 1, 2, 7, 10, and 14 and weekly thereafter for as long as the infant remained intubated (n = 5 for <28 wk gestation, n = 6 for 28–37 wk gestation, and n = 3 for >37 wk gestation). Samples were stored at –20°C before analysis by ELISA. Cadaveric lung samples were obtained at autopsy, immediately flash-frozen, and stored at –80°C before analysis with RT-PCR.
RNA isolation, semiquantitative PCR, and real-time PCR
Total RNA was extracted from primary cultures of airway epithelia and fetal lung explant cultures using TriReagent (Molecular Research Center) according to the manufacturer’s recommendations. The mRNA samples were treated with DNase (Promega) according to the manufacturer’s specifications. RT was performed using 1 μg of RNA with the Superscript II reverse transcriptase kit (Invitrogen Life Technologies) and random sequence hexamers. After treatment with Biolase DNA polymerase (Bioline), RT reaction products were subjected to PCR using an automated DNA thermal cycler for 30 cycles. PCR conditions included denaturation for 30 s at 94°C, annealing for 30 s at 60°C, extension for 30 s at 72°C, and a final extension at 72°C for 5 min. The following primers were used: HBD-1 forward, 5'-GAT CAT TAC AAT TGC GTC AGC AGT GG-3'; HBD-1 reverse, 5'-CTC ACT TGCAGC ACT TGG CCT TC-3'; HBD-2 forward, 5'-GTT ATA GGC GAT CCT GTT ACC TGCATC AGC CAT GAG GGT CTT GT-3'; HBD-2 reverse, 5'-TCA TGG CTT TTT GCA GCA TTT TGT TC-3'; HBD-3 forward, 5'-TGT TTG CTT TCG TCT TCC TG-3'; HBD-3 reverse, 5'-CTT TCT TCG GCA GCA TTT TC-3'; LL-37 forward, 5'-GTG ACT TCA AGA AGG ACG GG-3'; LL-37 reverse, 5'-GGG TAG GGC ACA CAC TAG GA-3'; GAPDH forward, 5'-GTC AGT GGT GGA CCT GAC CT-3'; and GAPDH reverse, 5'-AGG GGT CTA CAT GGC AAC TG-3'. Products were visualized after electrophoresis in 2% agarose gels and staining with ethidium bromide. To quantify the abundance of -defensin expression, densitometry values were standardized to their corresponding GAPDH control using a ChemiImager 4400 (AlphaInnotech). Real-time PCR was performed using TaqMan Universal PCR Master Mix (Applied Biosystems; 4304437) and TaqMan GAPDH control reagents (Applied Biosystems; 402869). The following primers and probe were used: HBD-1 forward, 5'-AAC AGG TGC CTT GAA TTT TGG T-3'; HBD-1 reverse, 5'-TTG CGT CAG CAG TGG AGG-3'; HBD-1 probe, 56-FAM-CAA TGT CTC TAT TCT GCC TGC CCG ATC TT-36 T; HBD-2 forward, 5'-CCT CTT CAT ATT CCT GAT GCC TCT-3'; HBD-2 reverse, 5'-GGC TCC ACT CTT AAG GCA GGT-3'; HBD-2 probe, 56-FAM-CCA GGT GTT TTT GGT GGT ATA GGC GAT CC-36 T; HBD-3 forward, 5'-TGA GGA TCC ATT ATC TTC TGT TTGC-3'; HBD-3 reverse, 5'-TGT GTT TAT GAT TCC TCC ATG ACC-3'; and HBD-3 probe, 56-FAM-TTG CTC TTC CTG TTT TTG GTG CCT GTT C-36 T (Integrated DNA Technologies). Real-time PCR analysis was performed on an Applied Biosystems sequence detection system (model 7700).
Peptide quantification by ELISA
HBD-2 protein abundance in neonatal tracheal aspirates was measured by ELISA. A 96-well Immulon-4 plate (MTX Lab Systems) was coated with rabbit anti-human HBD-2 polyclonal affinity-purified antiserum (provided by Dr. D. Proud, University of Calgary, Calgary, Canada) (18) diluted 1/2000 in BupH Carbonate-Bicarbonate Buffer Pack (Pierce). Wells were blocked with 0.05% Tween 80 and 1% BSA before incubation with 100 μl of tracheal aspirate diluted in PBS and N-acetyl cysteine (final concentration, 2%; Sigma-Aldrich; A7250). Biotinylated rabbit anti-human HBD-2 polyclonal affinity-purified antisera (gift from Dr. D. Proud) (18) was diluted 1/2000 in PBS with 0.05% Tween 20 before labeling with streptavidin peroxidase (Pierce; 21126), followed by ABTS 1-Step (Pierce; 37615). Samples were then analyzed at 405 nm using a VERSAmax microplate reader (Molecular Devices). Samples were stored at –80°C before analysis by ELISA and were run in duplicate to ensure reproducibility.
LL-37 protein abundance in neonatal tracheal aspirates was measured by ELISA. A 96-well Maxisorp plate (Nunc) was coated with 5 μg/ml LL-37 mAb (11) in coating buffer (1.59 g of Na2CO3, 2.93 g of NaHCO3, and 0.2 g of Na N3 in 1000 ml of MilliQ-water, pH 9.6). Wells were blocked with 0.1% gelatin in PBS before incubation with 100 μl of tracheal aspirate diluted in PBS and N-acetyl cysteine (final concentration, 2%) or LL-37 (Innovagen) standard (0.01–10.0 ng/ml). Samples were then incubated with biotinylated R-anti-LL-37 (19) (20 μg/ml in 0.1% gelatin in PBS buffer) before labeling with streptavidin-AP (DACO D 0396) diluted 1/2000 in 0.1% gelatin in PBS buffer. Samples were subsequently developed with 1 mg/ml p-nitrophenylphosphate (Sigma-Aldrich; N-9389) in diethanolamine buffer according to the manufacturer’s recommendations. Samples were analyzed at 405 nm using a ThermoMax microplate reader (Molecular Devices), and results were compared with the standard curve. Samples were stored at –80°C before analysis by ELISA and were run in duplicate to ensure reproducibility.
IL-1 protein abundance in neonatal tracheal aspirates was measured by ELISA using primary and secondary Abs from a DuoKit (DY201; R&D Systems) according to the manufacturer’s recommendations with 100 μl of tracheal aspirate diluted in PBS and N-acetyl cysteine (final concentration, 2%) on Immulon-4 96-well plates. Reactions were developed using ImmunoPure streptavidin, alkaline phosphatase-conjugated (Pierce; 21324) and p-nitrophenyl phosphate 20-mg tablets (Sigma-Aldrich; N2765). Samples were run in duplicate to ensure reproducibility.
Antimicrobial assays
We used a modified radial diffusion assay to investigate bacterial sensitivity to the HBDs and LL-37 as previously described (16, 20). Briefly, 4 x 106 bacteria in mid-log phase were suspended in an underlay gel. Wells (2 mm in diameter) were punched into the gel and filled with peptide at concentrations ranging from 0.25–250 μg/ml before incubation for 3 h at 37°C. Nutrient-rich gels were then overlaid, and the plates were incubated overnight at 37°C. Zones of clearance were manually measured and plotted on a semilog graph, where the x-intercept represents the minimal effective concentration. Low salt agarose underlay gels contained 10 mM sodium phosphate (pH 7.4), whereas high salt underlay gels contained 150 mM sodium chloride (pH 7.4). Test organisms included Staphylococcus aureus ATCC 29213 and 25923 (American Type Culture Collection), Staphylococcus epidermidis ATCC 12228 and 35547, Candida albicans ATCC 1023, E. coli DH5, E. coli ATCC 25922, Klebsiella pneumoniae ATCC 13883, Moraxella catarrhalis ATCC 25238, Pseudomonas aeruginosa PA01, and clinical strains of Enterococcus species, group B Streptococcus, Listeria monocytogenes, methicillin-sensitive Staphylococcus aureus, methicillin-resistant Staphylococcus aureus, nonmucoid Pseudomonas aeruginosa, and mucoid Pseudomonas aeruginosa.
Statistics
All statistics were calculated in Microsoft Excel and used one-tailed Student’s t test with unequal variance, except for real-time PCR results. A value of p < 0.05 was considered significant. For real-time PCR, because the data were not normally distributed, the log transformation was applied to the data points to normalize the data distribution. Using the log-transformed values, the treated groups were compared with the medium control group using the one-tailed, two-sample t test. The fold effect of the treatment vs the control value was computed by taking the anti-log of the difference of the log means between control and treated groups.
Results
Term and postnatal lung tissues express HBD-2 message
As shown in Fig. 1, semiquantitative RT-PCR on lung tissue at various ages revealed no detectable message for HBD-2 in two prenatal tissues (18 and 22 wk gestation). However, HBD-2 message was detected at term (42 wk gestation) and both postnatal tissues (7 mo and 13 years of age). Similarly, faint HBD-1 signals were detected in the term and postnatal tissues, but message abundance was less evident compared with HBD-2. No HBD-3 message was detected in any of the lung tissues. In contrast, tissues from all developmental ages expressed LL-37.
FIGURE 1. HBD-2 expression is developmentally regulated in lung tissue from various ages. RT-PCR of RNA extracted from whole lung tissue from different ages; 18, 22 and 42 refer to weeks of gestation and 7 mo and 13 yr refer to postgestational age. +CTRL, positive control plasmid DNA.
HBD-2 is induced by IL-1 and is the predominant defensin in fetal lung explants
Of the three defensins studied, HBD-2 message was the most abundant and was significantly increased by IL-1 and IL-1/IFN-. As shown in Fig. 2, we detected no defensin message in the start tissues before culture. After 1 day of growth in serum- and hormone-free medium, the HBD-2 message significantly increased (p < 0.001). Tissues stimulated with 100 ng/ml IL-1 or a combination of IL-1 and IFN- demonstrated a 2.3-fold greater signal than the unstimulated controls (p < 0.05). However, there was no significant additive increase in the combined IL-1 and IFN- signal over that of IL-1 alone. Similarly, compared with medium controls, there was no significant difference in response to IFN- stimulation alone. These results suggest that IFN- does not regulate HBD-2 expression. Treatment with the anti-inflammatory corticosteroid dexamethasone resulted in a significant decrease in HBD-2 message (p < 0.05). Treatment with heat-killed E. coli or PMA caused no significant change in HBD-2 message. HBD-1 and HBD-3 were not detected in any sample after 30 cycles of PCR. LL-37 message was only detected in some of the starting tissues, but in none of the cultured explants (data not shown). Using real-time PCR, we detected a similar pattern of HBD-2 induction, with significantly increased mRNA detected in the IL-1 (p < 0.0001) and IL-1/IFN- conditions (p < 0.001; see Fig. 3). HBD-1 message significantly increased with dexamethasone stimulation (p < 0.05). HBD-2 was detected after 25 cycles in the unstimulated tissues and after 20 cycles in the IL-1-stimulated state, whereas HBD-1 amplification increased in the low to mid-30-cycle range, and HBD-3 generally had no signal by 40 cycles (see inset, Fig. 3). This indicates a low message abundance of HBD-2 in the unstimulated state, a significantly higher expression with IL-1 stimulation, a very low abundance of HBD-1 message, and virtually no HBD-3 message. Although not directly comparable because of the different amplification efficiency of each defensin primer set, these results also support the idea that HBD-2 message is the most abundant of the three -defensins in fetal lung.
FIGURE 2. Developmental and cytokine regulation of HBD-2 expression. RNA was extracted from fetal lung explant samples. Tissues were cultured overnight with and without stimulation as indicated. Upper panel, Average densitometry measurements for HBD-2 mRNA standardized to GAPDH were compared with the medium control condition (n = 4–5). Lower panel, Representative RT-PCR results for the corresponding GAPDH and HBD-2. Start, tissue before culture; CTRL, medium without cytokines; IL, IL-1; IFN, IFN-; IL/IFN, IL-1 and IFN-; Dex, dexamethasone; E. coli, heat-killed E. coli. *, p < 0.05.
FIGURE 3. Fold increase in defensin message in fetal lung explants. RNA was extracted from fetal lung explants with and without stimulation. as indicated. Real-time PCR was performed for HBD-1, HBD-2, and HBD-3. HBD-3 is not shown because it had no measurable signal (n = 3–8). The inset in the upper left corner shows the relative cycles at which the -defensins were detected by real-time PCR under IL-1-stimulated conditions. CTRL, medium without cytokines; Dex, dexamethasone; E. coli, heat-killed E. coli. *, p < 0.05.
HBD-2 is induced by IL-1 and is the predominant defensin in primary cultures of airway epithelia
To further verify the regulation of the -defensins, we used primary cultures of human airway epithelia grown at the air-fluid interface as a second model of expression in the lung. We detected a similar pattern of defensin message regulation using this model. HBD-2 was significantly increased by IL-1 and IL-1/IFN- stimulation (p < 0.05) (see Fig. 4A). HBD-2 message was increased with TNF- stimulation, but did not reach statistical significance. Again, real-time PCR confirmed an almost identical pattern of HBD-2 message induction, with a significant increase in the IL-1 (p < 0.005) and IL-1/IFN- conditions (p < 0.001; see Fig. 4B). The messages for HBD-1, HBD-3, and LL-37 were not detected after 30 cycles of semiquantitative PCR and were therefore not assayed using real-time PCR (data not shown).
FIGURE 4. Inflammatory mediators regulate HBD-2 expression. RNA was extracted from stimulated primary cultures of airway epithelia. Tissues were cultured overnight with and without stimulation as indicated. A, Average densitometry measurements for HBD-2 mRNA standardized to GAPDH were compared with the medium control condition (n = 4–5). Representative RT-PCR results for the corresponding GAPDH and HBD-2. B, RNA was extracted from primary cultures of airway epithelia with and without stimulation, as indicated. Real-time PCR was performed for HBD-2. CTRL, medium without cytokines; IL, IL-1; IFN, IFN-; IL/IFN, IL-1 and IFN-; Dex, dexamethasone; TNF, TNF-; +CTRL, plasmid HBD-2 DNA. *, p < 0.05.
HBD-2, but not LL-37, is developmentally regulated in neonatal tracheal aspirates
We quantified HBD-2 and LL-37 protein abundance in neonatal tracheal aspirates to further verify their developmental regulation in vivo. We exclusively investigated HBD-2 and LL-37 in these samples because of the availability of an ELISA for these peptides and because HBD-2 was the most prevalent transcript in the surveyed lung tissue, fetal lung explants, and primary cultures of human airway epithelia. HBD-2 was detected in almost all samples. As shown in Fig. 5, the abundance of HBD-2 was significantly higher in samples from infants born after 37 wk gestation compared with those born at <28 wk and those between 28 and 37 wk gestation (p < 0.05). The average peptide abundance in samples obtained from infants born at 37 wk or earlier did not significantly change during their first 2 wk of life. No infants >37 wk clinically required intubation for >2 days, limiting the number of days that samples could be obtained. In contrast to HBD-2, LL-37 abundance did not significantly change in the >37 wk samples and the only significant change from the initial measurement was on day 10 for the >28 wk gestation samples (see Fig. 6). Additionally, there were no significant differences between groups in IL-1 abundance (data not shown). Most samples contained IL-1 at the lower limit of detection for the ELISA in the 10–50 pg/ml range.
FIGURE 5. Increase in HBD-2 from tracheal aspirates from term infants. Equal volumes of tracheal aspirates were assayed by ELISA for the abundance of HBD-2. Samples were run in duplicate to ensure reproducibility. Day 2 samples from neonates >37 wk gestation were significantly increased compared with tracheal aspirates from earlier gestations. n = 5 for <28 wk gestation, n = 6 for 28–37 wk gestation, and n = 3 for >37 wk gestation. *, p < 0.05.
FIGURE 6. Abundance of LL-37 in neonatal tracheal aspirates. Equal volumes of the same tracheal aspirates used for the HBD-2 ELISA (Fig. 5) were assayed for LL-37 protein abundance by ELISA. Samples were run in duplicate to ensure reproducibility. n = 5 for <28 wk gestation, n = 6 for 28–37 wk gestation, and n = 3 for >37 wk gestation. *, p < 0.05.
Each antimicrobial has a unique spectrum of antimicrobial activity and salt sensitivity
The minimal effective concentrations of each peptide against a range of bacteria and against a fungus are shown in Table I. HBD-3 was generally the most potent peptide, displaying activity against both Gram-positive and -negative bacteria in low microgram per milliliter concentrations. HBD-2 mainly had antimicrobial activity against Gram-negative bacteria, with a slightly lower potency compared with HBD-3. Notable exceptions to this trend were HBD-2, which exhibited the most potent antimicrobial activity against M. catarrhalis and K. pneumoniae, two of the more common bacteria causing pneumonia after the neonatal period (21, 22). HBD-1 generally had the least activity, but was active against both Gram-positive and -negative bacteria. Due to limited quantities of HBD-1, this peptide was tested against a smaller number of bacteria. HBD-1 and HBD-2 both displayed salt-sensitive antimicrobial activity, whereas HBD-3 showed salt-insensitive activity against specific Gram-positive and -negative organisms. In contrast, LL-37, a human neutrophil cathelicidin, also showed potent Gram-positive and -negative microbicidal activities, but exhibited salt-insensitive activity only against Gram-negative bacteria. Thus, each peptide has a unique spectrum of activity and salt sensitivity.
Table I. Unique spectrums of antimicrobial activity for each -defensin and cathelicidin1
Discussion
Of the three defensins studied, HBD-2 was the predominant -defensin expressed in developing human lung. HBD-2 message was developmentally regulated, induced by the proinflammatory factor IL-1, and decreased by dexamethasone. HBD-2 is known to be regulated by IL-1 in human skin (23), gingival mucosa (24), intestinal epithelia (25), and respiratory epithelia (7, 26, 27), but to our knowledge, these are the first data to describe developmental regulation of the IL-1-induced -defensin response in human airway epithelia. HBD-2 protein abundance in neonatal tracheal aspirates increased as a function of gestational age. HBD-1 message was detected to a lesser degree and was only induced by dexamethasone (Fig. 3). Although HBD-1 is generally considered constitutive rather than inducible, there are precedents for the possibility of steroidal or hormonal regulation of HBD-1, because HBD-1 is highly expressed in breast milk (28) and in the urine of pregnant women (14). LL-37 and HBD-3 mRNAs were generally not detected in airway epithelial culture samples. Additionally, each antimicrobial peptide exhibited a unique spectrum of antimicrobial activity and salt sensitivity against the most common bacteria causing sepsis in the neonate.
These data show that in addition to known deficiencies in acquired immunity (2, 3) and other innate immune factors (4, 5, 6), the lungs of premature infants have a decreased abundance of human -defensins. Because tissues quickly mature in the fetal lung explant model, the significant increase in HBD-2 message after fetal tissue has been cultured for 24 h (Fig. 2) supports the idea that HBD-2 expression is diminished at midgestation and is developmentally regulated; however, this result could also have been due to a variety of other factors, including stress or trauma to tissues or lack of serum or hormone factors in the culture model. Developmental regulation of HBD-2 was corroborated by the increase in HBD-2 message in the survey of lung tissues at various developmental ages (Fig. 1) and the in vivo protein abundance in term neonatal tracheal aspirates after 37 wk gestation (Fig. 5). The observation that LL-37 and IL-1 peptide abundance was not elevated in the tracheal aspirate samples from infants born at >37 wk gestation also supports the developmental regulation of HBD-2. Because the samples were standardized by volume, if the increase in HBD-2 on day 2 was simply due to dilutional or protein concentration effects, we would have expected LL-37 and IL-1 to be similarly increased. IL-1 abundance in tracheal aspirates was not significantly different among the groups. This suggests that although a stimulus for HBD-2 induction was present at all time points, other factors necessary for the expression of HBD-2 were not matured until near term.
The expression of other defensins and antimicrobial peptides is developmentally regulated in humans as well as other animals. In animal studies, defensin or defensin-like peptides increase during gestation in the intestinal tissues of sheep (29, 30) and mice (31). Similarly, in humans, HD-5 and HD-6 mRNAs were lower in midgestation fetal intestinal samples compared with adult samples (5, 6). Neonatal skin in mice and humans expresses increased cathelicidin message during the perinatal period (32). Developmental regulation of defensins in the lung has been shown with bovine tracheal antimicrobial peptide (33) and HBD-1 (34). In contrast to our data showing increased HBD-2 abundance in term infants, other recent investigations of -defensins in neonatal tracheal aspirates showed increased abundance after infection, but no developmental regulation (13). This may reflect maturational differences in the age ranges in each study, because the most developed group in the study by Schaller-Bals et al. (13) included tracheal aspirates from more preterm time points (wk 31–40 gestation), and the present study had a more narrow range at more developed time points (>37 wk gestation).
Each of the four innate host factors studied exhibited unique spectrums of antimicrobial activity against the most common bacterial pathogens in neonatal infections, including E. coli, Enterococcus species, group B Streptococcus, H. influenzae, Klebsiella species, L. monocytogenes, P. aeruginosa, coagulase-negative Staphylococcus, and Staphylococcus aureus (35, 36, 37, 38, 39). These antimicrobial factors have complimentary activities, because each peptide has different bacterial spectrum and salt sensitivity. The combined expression of many innate immune factors will have a wider range of antimicrobial activity. Although the abundance of the defensins and LL-37 in neonatal tracheal aspirates appears greatly lower than the ranges at which they exert antimicrobial activity, several factors need to be kept in mind. Determination of the correct dilution factor for saline instilled to obtain fluid lining the epithelial has been problematic (40, 41). However, dilutions for obtaining bronchoalveolar lavage fluid or tracheal aspirates have been estimated to be 1 ml of airway fluid/10–100 ml of bronchoalveolar lavage fluid based on urea concentration measurements (42). Assuming up to a 100-fold dilution estimate, the HBD-2 protein levels in the >37 wk gestation day 2 tracheal aspirates samples would average 75 ng/ml. Defensins could act in concert with numerous other antimicrobial factors in airway surface liquid, such as lactoferrin, lysozyme, secretory leukocyte protease inhibitor (43), and CCL20 (16). Furthermore, defensins might exhibit synergistic antimicrobial activity in a manner similar to that seen in lysozyme, lactoferrin, and secretory leukocyte protease inhibitor (44). Also, there may be microenvironments that have much higher concentrations of defensins, such as airway surface liquid, the paracellular space, or directly under the basolateral membrane. Defensins may be in high enough concentrations in these microenvironments to exert direct antimicrobialactivity. Alternatively, defensins and LL-37 may be acting by modulating local immunity (45, 46, 47, 48).
Two unexpected findings were the selective salt insensitivity of HBD-3 against specific strains of bacteria and the salt insensitivity of LL-37 only against Gram-negative bacteria. HBD-3 was identified by Schroeder et al. (8) using isolation methods designed to identify peptides that bind to Gram-positive bacteria. Others identified HBD-3 using bioinformatic approaches (8, 49). Schroeder et al. (8) showed that HBD-3 had salt-insensitive antimicrobial activity, but this salt sensitivity was only tested against a single strain of S. aureus. We found similar HBD-3 salt insensitivity only against specific strains of S. aureus, other staphylococcal organisms, and some Gram-negative organisms (see Table I). We found that HBD-3 displayed salt insensitivity that was species and even strain specific. LL-37 is known to have salt-insensitive microbicidal activity (50), but we found salt insensitivity only against Gram-negative bacteria. LL-37 also displayed species and strain variability in a manner similar to that reported previously (50).
Because LL-37 is present in airway epithelia and neutrophils, it is difficult to determine the relative epithelial contribution of LL-37 in some of these studies. For example, the LL-37 message detected in whole lung tissue at various ages (Fig. 1) may have been derived from neutrophils in the tissue. Similarly, we only detected LL-37 message in the start tissue from the fetal lung explants, not after incubation for 24 h (data not shown). Because neutrophils do not survive in fetal lung explant culture, any message detected after 24 h in the cultured samples should originate from lung tissue. However, the signal in the start tissues could again have been from nonepithelial sources, such as neutrophils present in freshly isolated tissue. One potential problem with the hypothesis of LL-37 transcripts originating from neutrophils is that peripherally circulating neutrophils have little protein synthesis. In neutrophils derived from adults, transcription of many proteins primarily occurs in bone marrow. However, it is not known whether this is also the case in the developing fetus. Thus, we found little evidence of LL-37 expression in human airway epithelia under the culture conditions used in these studies.
As with all models, there are limitations to the in vitro cultures used in these studies. The failure to see the expected induction of the -defensins by heat-killed E. coli in fetal lung explants may reflect the inability of bacterial products to adequately penetrate into cultured tissue in this model or the absence of myeloid cells, which may be the predominant bacterial sensors for epithelia (51) and are lost in culture models (52). In human epithelial models, HBD-2 expression has been shown to be up-regulated by some bacterial strains (25, 26, 53). In contrast to the fetal lung explant model, we did not see a significant inhibition of HBD-2 message by dexamethasone in the primary cultures of human airway epithelia. However, the low basal expression of HBD-2 mRNA in the unstimulated airway epithelia may preclude determination of inhibition with the primary culture model. Failure to detect HBD-3 may also reflect shortcomings of the model or the stimuli used in these studies. We previously amplified HBD-3 cDNA from human fetal lung after 30 cycles of PCR (49). Harder et al. (8) detected HBD-3 in tracheal primary cultures stimulated with a clinical strain of mucoid P. aeruginosa. Similarly, Duits et al. (54) found that rhinovirus-16-stimulated cultures of primary bronchial epithelial cells exhibited increased HBD-2 and HBD-3 mRNAs.
HBD-2 may play a significant role in pulmonary innate immunity. Because it is developmentally regulated, decreased abundance of HBD-2 may contribute to the increased incidence of pulmonary infections in premature neonates. The prevalence of HBD-2 message in the lung, its inducibility, and its potent antimicrobial activity against common respiratory pathogens lends credence to this assertion. We speculate that premature infants may produce less HBD-2 in response to a bacterial challenge. This could allow specific HBD-2-sensitive bacteria to more easily colonize and infect the respiratory tract. Interestingly, HBD-2 has poor antimicrobial activity against H. influenzae (20), which is a not only a common respiratory pathogen, but is also a common commensal in the human respiratory tract. The abilities of H. influenzae to colonize and cause disease in the human airway and to evade killing by innate immune factors may be partly due to its resistance to a prevalent -defensin in the airway.
Because defensins and cathelicidins have properties other than their antimicrobial abilities, these data may also have implications for adaptive immunity and tissue development. Defensins signal adaptive immunity by inducing chemotaxis of immature dendritic cells and memory T cells (45). Lower levels of these peptides may contribute to the decreased adaptive immune responses in neonates and premature infants. This role that defensins play in recruiting APCs may be especially important early in the neonatal period because adaptive immune signaling is inhibited more than innate immune signaling (55). Because LL-37 induces angiogenesis (56) and other defensins are mitogenic for epithelial cells and fibroblasts (57), these data may also have implications for tissue development and wound repair.
In summary, we found that HBD-2 was the most prevalent -defensin in human lung cultures and neonatal tissues. HBD-2 message and protein were developmentally regulated and were less abundant in premature compared with term or post-term samples. Each defensin displayed a unique spectrum of antimicrobial activity and salt sensitivity against a wide variety of neonatal and respiratory pathogens. Lower levels of HBD-2 in the lungs of premature infants may be one factor leading to an immature innate immune system and contributing to the increased susceptibility of neonates to respiratory infections.
Acknowledgments
We acknowledge the support of the Cell Culture Core in preparing airway epithelial cultures. We thank Hong Peng Jia for his support in producing HBD-1. We thank Dr. David Proud for providing the rabbit anti-HBD2 antisera and for technical assistance with the HBD-2 ELISA, and Dr. Michael Acarregui for assistance with the fetal lung explant cultures. We thank Monica Lindh for technical assistance with the LL-37 ELISA. We also thank the nurses and the General Clinical Research Center at University of Iowa for their help in obtaining the neonatal tracheal aspirate samples.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grants HL61234, HL07638, HL67992, 2K1227748 (Child Health Research Career Development Award), P30DK54759, and RR00059 (General Clinical Research Center), and the Swedish Foundation for International Cooperation in Research and Higher Education.
2 Address correspondence and reprint requests to Dr. Timothy D. Starner, Department of Pediatrics, University of Iowa, 2JCP, 200 Hawkins Drive, Iowa City, IA 52242. E-mail address: timothy-starner@uiowa.edu
3 Abbreviations used in this paper: HD-5, human -defensin 5; HBD, human -defensin.
Received for publication March 10, 2004. Accepted for publication November 16, 2004.
References
Langrish, C. L., J. C. Buddle, A. J. Thrasher, D. Goldblatt. 2002. Neonatal dendritic cells are intrinsically biased against Th-1 immune responses. Clin. Exp. Immunol. 128:118.
Lawton, A. R.. 1986. Ontogeny of B cells and pathogenesis of humoral immunodeficiencies. Clin. Immunol. Immunopathol. 40:5.
Suen, Y., S. M. Lee, J. Qian, C. van de Ven, M. S. Cairo. 1998. Dysregulation of lymphokine production in the neonate and its impact on neonatal cell mediated immunity. Vaccine 16:1369.
Levy, O., S. Martin, E. Eichenwald, T. Ganz, E. Valore, S. F. Carroll, K. Lee, D. Goldmann, G. M. Thorne. 1999. Impaired innate immunity in the newborn: newborn neutrophils are deficient in bactericidal/permeability-increasing protein. Pediatrics 104:1327.
Salzman, N. H., R. A. Polin, M. C. Harris, E. Ruchelli, A. Hebra, S. Zirin-Butler, A. Jawad, E. Martin Porter, C. L. Bevins. 1998. Enteric defensin expression in necrotizing enterocolitis. Pediatr. Res. 44:20.
Mallow, E. B., A. Harris, N. Salzman, J. P. Russell, R. J. DeBerardinis, E. Ruchelli, C. L. Bevins. 1996. Human enteric defensins: gene structure and developmental expression. J. Biol. Chem. 271:4038.
Singh, P. K., H. P. Jia, K. Wiles, J. Hesselberth, L. Liu, B. A. Conway, E. P. Greenberg, E. V. Valore, M. J. Welsh, T. Ganz, et al 1998. Production of -defensins by human airway epithelia. Proc. Natl. Acad. Sci. USA 95:14961.
Harder, J., J. Bartels, E. Christophers, J. M. Schroder. 2001. Isolation and characterization of human -defensin-3, a novel human inducible peptide antibiotic. J. Biol. Chem. 276:5707.
Bals, R., X. Wang, M. Zasloff, J. M. Wilson. 1998. The peptide antibiotic LL-37/hCAP-18 is expressed in epithelia of the human lung where it has broad antimicrobial activity at the airway surface. Proc. Natl. Acad. Sci. USA 95:9541.
Agerberth, B., J. Grunewald, E. Castanos-Velez, B. Olsson, H. Jornvall, H. Wigzell, A. Eklund, G. H. Gudmundsson. 1999. Antibacterial components in bronchoalveolar lavage fluid from healthy individuals and sarcoidosis patients. Am. J. Respir. Crit. Care Med. 160:283.
Yoshio, H., M. Tollin, G. H. Gudmundsson, H. Lagercrantz, H. Jornvall, G. Marchini, B. Agerberth. 2003. Antimicrobial polypeptides of human vernix caseosa and amniotic fluid: implications for newborn innate defense. Pediatr. Res. 53:211.
Marchini, G., S. Lindow, H. Brismar, B. Stabi, V. Berggren, A. K. Ulfgren, S. Lonne-Rahm, B. Agerberth, G. H. Gudmundsson. 2002. The newborn infant is protected by an innate antimicrobial barrier: peptide antibiotics are present in the skin and vernix caseosa. Br. J. Dermatol. 147:1127.
Schaller-Bals, S., A. Schulze, R. Bals. 2002. Increased levels of antimicrobial peptides in tracheal aspirates of newborn infants during infection. Am. J. Respir. Crit. Care Med. 165:992.
Valore, E. V., C. H. Park, A. J. Quayle, K. R. Wiles, P. B. McCray, Jr, T. Ganz. 1998. Human -defensin-1: an antimicrobial peptide of urogenital tissues. J. Clin. Invest. 101:1633.
Saiman, L., S. Tabibi, T. D. Starner, P. San Gabriel, P. L. Winokur, H. P. Jia, P. B. McCray, Jr, B. F. Tack. 2001. Cathelicidin peptides inhibit multiply antibiotic-resistant pathogens from patients with cystic fibrosis. Antimicrob. Agents Chemother. 45:2838.
Starner, T. D., C. K. Barker, H. P. Jia, Y. Kang, P. B. McCray, Jr. 2003. CCL20 is an inducible product of human airway epithelia with innate immune properties. Am. J. Respir. Cell Mol. Biol. 29:627.
Karp, P. H., T. O. Moninger, S. P. Weber, T. S. Nesselhauf, J. L. Launspach, J. Zabner, M. J. Welsh. 2002. An in vitro model of differentiated human airway epithelia: methods for establishing primary cultures. Methods Mol. Biol. 188:115.
Proud, D., S. P. Sanders, S. Wiehler. 2004. Human rhinovirus infection induces airway epithelial cell production of human -defensin 2 both in vitro and in vivo. J. Immunol. 172:4637.
Gudmundsson, G. H., B. Agerberth, J. Odeberg, T. Bergman, B. Olsson, R. Salcedo. 1996. The human gene FALL39 and processing of the cathelin precursor to the antibacterial peptide LL-37 in granulocytes. Eur. J. Biochem. 238:325.(Timothy D. Starner, Birgi)
Immaturity of innate immunity contributes to the increased susceptibility of human neonates to infection. The lung is a major portal of entry for potential pathogens in the neonate, and human -defensins (HBDs) and LL-37 participate in pulmonary innate immunity. We hypothesized that these antimicrobial factors would be developmentally regulated, expressed by neonatal pulmonary tissues, and participate in neonatal innate immunity. We found HBD-2 to be the predominant -defensin in human neonatal lung. HBD-2 mRNA expression was developmentally regulated, induced by the proinflammatory factor IL-1, and decreased by dexamethasone. Additionally, HBD-2 abundance in neonatal tracheal aspirates increased as a function of gestational age. HBD-1 had a lower level of expression compared with HBD-2 and was induced by dexamethasone. HBD-3 and LL-37 messages were not detected in airway epithelial cultures. Additionally, each antimicrobial peptide exhibited a unique spectrum of antimicrobial activity and salt sensitivity against bacteria commonly causing sepsis in the neonate. Lower levels of HBD-2 may be one factor contributing to the increased susceptibility of premature infants to pulmonary infections.
Introduction
The morbidity and mortality rates for infectious diseases are higher during the neonatal period. This increased susceptibility to infection has been related to the immaturity of the neonatal immune system. At birth, the immune system exhibits deficiencies in Th-1 immune responses (1), Ig synthesis (for a review, see Ref.2), and selected cytokine production (1, 3). Immaturity of antimicrobial innate immune factors may also be involved in the increased susceptibility of neonates to infection. For example, neutrophils of newborn infants are selectively deficient in bactericidal/permeability-increasing protein (4). Additionally, human -defensins 5 (HD-5)3 and HD-6 have significantly lower levels of expression in fetal life compared with that in term infants (5, 6). As a major portal of entry for potential pathogens during the neonatal period, the lung has the second largest epithelial surface area exposed to the outside environment. Components of the innate immune repertoire of airway epithelia include human -defensins (HBDs) (7, 8) and the only known human cathelicidin, LL-37 (9, 10). Defensins and LL-37 are also present in the skin and vernix and contribute to the innate immunity of neonates (11, 12). Recently, HBDs and LL-37 were shown to be increased in neonatal tracheal aspirates during infections (13); however, little is known about the development or regulation of human -defensins and cathelicidins in neonatal lung.
In this study we investigated the ontogeny, expression, and regulation of three well-characterized -defensins, HBD-1, -2, and -3, and the cathelicidin, LL-37, in the neonatal lung and determined their spectrum of antimicrobial activity. We hypothesized that these antimicrobial factors would be developmentally regulated, expressed by neonatal pulmonary tissues, and participate in innate immunity. We found HBD-2 to be developmentally regulated, inducible, and the predominant -defensin in the neonatal lung. Additionally, the -defensins and LL-37 exhibited unique spectrums of antimicrobial activity and salt sensitivity against neonatal pathogens.
Materials and Methods
Reagents
Recombinant HBD-1 was purified from a baculoviral production system as previously described (14). LL-37 was produced synthetically as previously described (15). All peptides, including recombinant HBD-2 and HBD-3 (PeproTech, Rocky Hill, NJ), were analyzed for purity and concentration using mass spectroscopy and amino acid composition. Peptides were stored in 0.02% acetic acid at –80°C. Before use in antimicrobial assays, peptides were lyophilized and then reconstituted in 0.02% acetic acid with 0.1% human serum albumin.
Culture methods and human samples
All tracheal aspirate specimens and culture samples used in these studies were obtained using protocols approved by the institutional review board at the University of Iowa.
Human fetal lung explant cultures
Midgestational fetal lung explants were cultured as previously described (16). Briefly, 18–22 wk gestation lung tissues were dissected free of major airways and blood vessels, cut into 1- to 2-mm pieces, and cultured at 37°C in serum- and hormone-free Weymouth’s MB752/1 medium (Invitrogen Life Technologies) supplemented with 1% penicillin/streptomycin (Invitrogen Life Technologies; 15140-122) and 0.1% amphotericin B (Cellgro; 30-003-C1) in a humidified atmosphere with 5% CO2. Tissues were incubated in either control medium or medium containing 100 ng/ml recombinant human IL-1 (Sigma-Aldrich; I-9401), 100 ng/ml IFN- (Sigma-Aldrich; I-3265), 100 ng/ml both IL-1 and IFN-, 10–7 M dexamethasone (Roxanne), 106 CFU heat-killed Escherichia coli DH5 bacteria, or 100 ng/ml PMA (Sigma-Aldrich; P-1585). Explants were maintained in culture for 24 h. The tissue samples were then frozen and stored at –80°C before analysis by RT-PCR and real-time PCR. For each experimental condition, replicates were performed using tissues from different donors (n = 4–5).
Primary cultures of human airway epithelia
Primary cultures of well-differentiated human airway epithelia were prepared and grown at the air-liquid interface as previously described (17). Epithelial cultures were incubated at 37°C in 5% CO2 for 24 h either with 100 μl of PBS apically and 500 μl of medium basolaterally or with the same volumes of apical PBS and basolateral medium containing 100 ng/ml IL-1, 100 ng/ml IFN-, 100 ng/ml both IL-1 and IFN-, 10–7 M dexamethasone, or 100 ng/ml TNF- (Sigma-Aldrich; T-6674). For each experimental condition, replicates were performed using primary cultures from different donors (n = 3).
Neonatal tracheal aspirate samples and lung tissue
After obtaining informed consent from the parents, 14 mechanically ventilated newborn infants in the neonatal intensive care unit at the University of Iowa were enrolled during an 18-mo study period (May 2000 to November 2002). Tracheal aspirates were obtained using a standardized procedure for routine suctioning by General Clinical Research Center staff with a catheter inserted through the endotracheal tube and instillation of 1 ml of normal saline. When possible, samples were obtained on postnatal days 0, 1, 2, 7, 10, and 14 and weekly thereafter for as long as the infant remained intubated (n = 5 for <28 wk gestation, n = 6 for 28–37 wk gestation, and n = 3 for >37 wk gestation). Samples were stored at –20°C before analysis by ELISA. Cadaveric lung samples were obtained at autopsy, immediately flash-frozen, and stored at –80°C before analysis with RT-PCR.
RNA isolation, semiquantitative PCR, and real-time PCR
Total RNA was extracted from primary cultures of airway epithelia and fetal lung explant cultures using TriReagent (Molecular Research Center) according to the manufacturer’s recommendations. The mRNA samples were treated with DNase (Promega) according to the manufacturer’s specifications. RT was performed using 1 μg of RNA with the Superscript II reverse transcriptase kit (Invitrogen Life Technologies) and random sequence hexamers. After treatment with Biolase DNA polymerase (Bioline), RT reaction products were subjected to PCR using an automated DNA thermal cycler for 30 cycles. PCR conditions included denaturation for 30 s at 94°C, annealing for 30 s at 60°C, extension for 30 s at 72°C, and a final extension at 72°C for 5 min. The following primers were used: HBD-1 forward, 5'-GAT CAT TAC AAT TGC GTC AGC AGT GG-3'; HBD-1 reverse, 5'-CTC ACT TGCAGC ACT TGG CCT TC-3'; HBD-2 forward, 5'-GTT ATA GGC GAT CCT GTT ACC TGCATC AGC CAT GAG GGT CTT GT-3'; HBD-2 reverse, 5'-TCA TGG CTT TTT GCA GCA TTT TGT TC-3'; HBD-3 forward, 5'-TGT TTG CTT TCG TCT TCC TG-3'; HBD-3 reverse, 5'-CTT TCT TCG GCA GCA TTT TC-3'; LL-37 forward, 5'-GTG ACT TCA AGA AGG ACG GG-3'; LL-37 reverse, 5'-GGG TAG GGC ACA CAC TAG GA-3'; GAPDH forward, 5'-GTC AGT GGT GGA CCT GAC CT-3'; and GAPDH reverse, 5'-AGG GGT CTA CAT GGC AAC TG-3'. Products were visualized after electrophoresis in 2% agarose gels and staining with ethidium bromide. To quantify the abundance of -defensin expression, densitometry values were standardized to their corresponding GAPDH control using a ChemiImager 4400 (AlphaInnotech). Real-time PCR was performed using TaqMan Universal PCR Master Mix (Applied Biosystems; 4304437) and TaqMan GAPDH control reagents (Applied Biosystems; 402869). The following primers and probe were used: HBD-1 forward, 5'-AAC AGG TGC CTT GAA TTT TGG T-3'; HBD-1 reverse, 5'-TTG CGT CAG CAG TGG AGG-3'; HBD-1 probe, 56-FAM-CAA TGT CTC TAT TCT GCC TGC CCG ATC TT-36 T; HBD-2 forward, 5'-CCT CTT CAT ATT CCT GAT GCC TCT-3'; HBD-2 reverse, 5'-GGC TCC ACT CTT AAG GCA GGT-3'; HBD-2 probe, 56-FAM-CCA GGT GTT TTT GGT GGT ATA GGC GAT CC-36 T; HBD-3 forward, 5'-TGA GGA TCC ATT ATC TTC TGT TTGC-3'; HBD-3 reverse, 5'-TGT GTT TAT GAT TCC TCC ATG ACC-3'; and HBD-3 probe, 56-FAM-TTG CTC TTC CTG TTT TTG GTG CCT GTT C-36 T (Integrated DNA Technologies). Real-time PCR analysis was performed on an Applied Biosystems sequence detection system (model 7700).
Peptide quantification by ELISA
HBD-2 protein abundance in neonatal tracheal aspirates was measured by ELISA. A 96-well Immulon-4 plate (MTX Lab Systems) was coated with rabbit anti-human HBD-2 polyclonal affinity-purified antiserum (provided by Dr. D. Proud, University of Calgary, Calgary, Canada) (18) diluted 1/2000 in BupH Carbonate-Bicarbonate Buffer Pack (Pierce). Wells were blocked with 0.05% Tween 80 and 1% BSA before incubation with 100 μl of tracheal aspirate diluted in PBS and N-acetyl cysteine (final concentration, 2%; Sigma-Aldrich; A7250). Biotinylated rabbit anti-human HBD-2 polyclonal affinity-purified antisera (gift from Dr. D. Proud) (18) was diluted 1/2000 in PBS with 0.05% Tween 20 before labeling with streptavidin peroxidase (Pierce; 21126), followed by ABTS 1-Step (Pierce; 37615). Samples were then analyzed at 405 nm using a VERSAmax microplate reader (Molecular Devices). Samples were stored at –80°C before analysis by ELISA and were run in duplicate to ensure reproducibility.
LL-37 protein abundance in neonatal tracheal aspirates was measured by ELISA. A 96-well Maxisorp plate (Nunc) was coated with 5 μg/ml LL-37 mAb (11) in coating buffer (1.59 g of Na2CO3, 2.93 g of NaHCO3, and 0.2 g of Na N3 in 1000 ml of MilliQ-water, pH 9.6). Wells were blocked with 0.1% gelatin in PBS before incubation with 100 μl of tracheal aspirate diluted in PBS and N-acetyl cysteine (final concentration, 2%) or LL-37 (Innovagen) standard (0.01–10.0 ng/ml). Samples were then incubated with biotinylated R-anti-LL-37 (19) (20 μg/ml in 0.1% gelatin in PBS buffer) before labeling with streptavidin-AP (DACO D 0396) diluted 1/2000 in 0.1% gelatin in PBS buffer. Samples were subsequently developed with 1 mg/ml p-nitrophenylphosphate (Sigma-Aldrich; N-9389) in diethanolamine buffer according to the manufacturer’s recommendations. Samples were analyzed at 405 nm using a ThermoMax microplate reader (Molecular Devices), and results were compared with the standard curve. Samples were stored at –80°C before analysis by ELISA and were run in duplicate to ensure reproducibility.
IL-1 protein abundance in neonatal tracheal aspirates was measured by ELISA using primary and secondary Abs from a DuoKit (DY201; R&D Systems) according to the manufacturer’s recommendations with 100 μl of tracheal aspirate diluted in PBS and N-acetyl cysteine (final concentration, 2%) on Immulon-4 96-well plates. Reactions were developed using ImmunoPure streptavidin, alkaline phosphatase-conjugated (Pierce; 21324) and p-nitrophenyl phosphate 20-mg tablets (Sigma-Aldrich; N2765). Samples were run in duplicate to ensure reproducibility.
Antimicrobial assays
We used a modified radial diffusion assay to investigate bacterial sensitivity to the HBDs and LL-37 as previously described (16, 20). Briefly, 4 x 106 bacteria in mid-log phase were suspended in an underlay gel. Wells (2 mm in diameter) were punched into the gel and filled with peptide at concentrations ranging from 0.25–250 μg/ml before incubation for 3 h at 37°C. Nutrient-rich gels were then overlaid, and the plates were incubated overnight at 37°C. Zones of clearance were manually measured and plotted on a semilog graph, where the x-intercept represents the minimal effective concentration. Low salt agarose underlay gels contained 10 mM sodium phosphate (pH 7.4), whereas high salt underlay gels contained 150 mM sodium chloride (pH 7.4). Test organisms included Staphylococcus aureus ATCC 29213 and 25923 (American Type Culture Collection), Staphylococcus epidermidis ATCC 12228 and 35547, Candida albicans ATCC 1023, E. coli DH5, E. coli ATCC 25922, Klebsiella pneumoniae ATCC 13883, Moraxella catarrhalis ATCC 25238, Pseudomonas aeruginosa PA01, and clinical strains of Enterococcus species, group B Streptococcus, Listeria monocytogenes, methicillin-sensitive Staphylococcus aureus, methicillin-resistant Staphylococcus aureus, nonmucoid Pseudomonas aeruginosa, and mucoid Pseudomonas aeruginosa.
Statistics
All statistics were calculated in Microsoft Excel and used one-tailed Student’s t test with unequal variance, except for real-time PCR results. A value of p < 0.05 was considered significant. For real-time PCR, because the data were not normally distributed, the log transformation was applied to the data points to normalize the data distribution. Using the log-transformed values, the treated groups were compared with the medium control group using the one-tailed, two-sample t test. The fold effect of the treatment vs the control value was computed by taking the anti-log of the difference of the log means between control and treated groups.
Results
Term and postnatal lung tissues express HBD-2 message
As shown in Fig. 1, semiquantitative RT-PCR on lung tissue at various ages revealed no detectable message for HBD-2 in two prenatal tissues (18 and 22 wk gestation). However, HBD-2 message was detected at term (42 wk gestation) and both postnatal tissues (7 mo and 13 years of age). Similarly, faint HBD-1 signals were detected in the term and postnatal tissues, but message abundance was less evident compared with HBD-2. No HBD-3 message was detected in any of the lung tissues. In contrast, tissues from all developmental ages expressed LL-37.
FIGURE 1. HBD-2 expression is developmentally regulated in lung tissue from various ages. RT-PCR of RNA extracted from whole lung tissue from different ages; 18, 22 and 42 refer to weeks of gestation and 7 mo and 13 yr refer to postgestational age. +CTRL, positive control plasmid DNA.
HBD-2 is induced by IL-1 and is the predominant defensin in fetal lung explants
Of the three defensins studied, HBD-2 message was the most abundant and was significantly increased by IL-1 and IL-1/IFN-. As shown in Fig. 2, we detected no defensin message in the start tissues before culture. After 1 day of growth in serum- and hormone-free medium, the HBD-2 message significantly increased (p < 0.001). Tissues stimulated with 100 ng/ml IL-1 or a combination of IL-1 and IFN- demonstrated a 2.3-fold greater signal than the unstimulated controls (p < 0.05). However, there was no significant additive increase in the combined IL-1 and IFN- signal over that of IL-1 alone. Similarly, compared with medium controls, there was no significant difference in response to IFN- stimulation alone. These results suggest that IFN- does not regulate HBD-2 expression. Treatment with the anti-inflammatory corticosteroid dexamethasone resulted in a significant decrease in HBD-2 message (p < 0.05). Treatment with heat-killed E. coli or PMA caused no significant change in HBD-2 message. HBD-1 and HBD-3 were not detected in any sample after 30 cycles of PCR. LL-37 message was only detected in some of the starting tissues, but in none of the cultured explants (data not shown). Using real-time PCR, we detected a similar pattern of HBD-2 induction, with significantly increased mRNA detected in the IL-1 (p < 0.0001) and IL-1/IFN- conditions (p < 0.001; see Fig. 3). HBD-1 message significantly increased with dexamethasone stimulation (p < 0.05). HBD-2 was detected after 25 cycles in the unstimulated tissues and after 20 cycles in the IL-1-stimulated state, whereas HBD-1 amplification increased in the low to mid-30-cycle range, and HBD-3 generally had no signal by 40 cycles (see inset, Fig. 3). This indicates a low message abundance of HBD-2 in the unstimulated state, a significantly higher expression with IL-1 stimulation, a very low abundance of HBD-1 message, and virtually no HBD-3 message. Although not directly comparable because of the different amplification efficiency of each defensin primer set, these results also support the idea that HBD-2 message is the most abundant of the three -defensins in fetal lung.
FIGURE 2. Developmental and cytokine regulation of HBD-2 expression. RNA was extracted from fetal lung explant samples. Tissues were cultured overnight with and without stimulation as indicated. Upper panel, Average densitometry measurements for HBD-2 mRNA standardized to GAPDH were compared with the medium control condition (n = 4–5). Lower panel, Representative RT-PCR results for the corresponding GAPDH and HBD-2. Start, tissue before culture; CTRL, medium without cytokines; IL, IL-1; IFN, IFN-; IL/IFN, IL-1 and IFN-; Dex, dexamethasone; E. coli, heat-killed E. coli. *, p < 0.05.
FIGURE 3. Fold increase in defensin message in fetal lung explants. RNA was extracted from fetal lung explants with and without stimulation. as indicated. Real-time PCR was performed for HBD-1, HBD-2, and HBD-3. HBD-3 is not shown because it had no measurable signal (n = 3–8). The inset in the upper left corner shows the relative cycles at which the -defensins were detected by real-time PCR under IL-1-stimulated conditions. CTRL, medium without cytokines; Dex, dexamethasone; E. coli, heat-killed E. coli. *, p < 0.05.
HBD-2 is induced by IL-1 and is the predominant defensin in primary cultures of airway epithelia
To further verify the regulation of the -defensins, we used primary cultures of human airway epithelia grown at the air-fluid interface as a second model of expression in the lung. We detected a similar pattern of defensin message regulation using this model. HBD-2 was significantly increased by IL-1 and IL-1/IFN- stimulation (p < 0.05) (see Fig. 4A). HBD-2 message was increased with TNF- stimulation, but did not reach statistical significance. Again, real-time PCR confirmed an almost identical pattern of HBD-2 message induction, with a significant increase in the IL-1 (p < 0.005) and IL-1/IFN- conditions (p < 0.001; see Fig. 4B). The messages for HBD-1, HBD-3, and LL-37 were not detected after 30 cycles of semiquantitative PCR and were therefore not assayed using real-time PCR (data not shown).
FIGURE 4. Inflammatory mediators regulate HBD-2 expression. RNA was extracted from stimulated primary cultures of airway epithelia. Tissues were cultured overnight with and without stimulation as indicated. A, Average densitometry measurements for HBD-2 mRNA standardized to GAPDH were compared with the medium control condition (n = 4–5). Representative RT-PCR results for the corresponding GAPDH and HBD-2. B, RNA was extracted from primary cultures of airway epithelia with and without stimulation, as indicated. Real-time PCR was performed for HBD-2. CTRL, medium without cytokines; IL, IL-1; IFN, IFN-; IL/IFN, IL-1 and IFN-; Dex, dexamethasone; TNF, TNF-; +CTRL, plasmid HBD-2 DNA. *, p < 0.05.
HBD-2, but not LL-37, is developmentally regulated in neonatal tracheal aspirates
We quantified HBD-2 and LL-37 protein abundance in neonatal tracheal aspirates to further verify their developmental regulation in vivo. We exclusively investigated HBD-2 and LL-37 in these samples because of the availability of an ELISA for these peptides and because HBD-2 was the most prevalent transcript in the surveyed lung tissue, fetal lung explants, and primary cultures of human airway epithelia. HBD-2 was detected in almost all samples. As shown in Fig. 5, the abundance of HBD-2 was significantly higher in samples from infants born after 37 wk gestation compared with those born at <28 wk and those between 28 and 37 wk gestation (p < 0.05). The average peptide abundance in samples obtained from infants born at 37 wk or earlier did not significantly change during their first 2 wk of life. No infants >37 wk clinically required intubation for >2 days, limiting the number of days that samples could be obtained. In contrast to HBD-2, LL-37 abundance did not significantly change in the >37 wk samples and the only significant change from the initial measurement was on day 10 for the >28 wk gestation samples (see Fig. 6). Additionally, there were no significant differences between groups in IL-1 abundance (data not shown). Most samples contained IL-1 at the lower limit of detection for the ELISA in the 10–50 pg/ml range.
FIGURE 5. Increase in HBD-2 from tracheal aspirates from term infants. Equal volumes of tracheal aspirates were assayed by ELISA for the abundance of HBD-2. Samples were run in duplicate to ensure reproducibility. Day 2 samples from neonates >37 wk gestation were significantly increased compared with tracheal aspirates from earlier gestations. n = 5 for <28 wk gestation, n = 6 for 28–37 wk gestation, and n = 3 for >37 wk gestation. *, p < 0.05.
FIGURE 6. Abundance of LL-37 in neonatal tracheal aspirates. Equal volumes of the same tracheal aspirates used for the HBD-2 ELISA (Fig. 5) were assayed for LL-37 protein abundance by ELISA. Samples were run in duplicate to ensure reproducibility. n = 5 for <28 wk gestation, n = 6 for 28–37 wk gestation, and n = 3 for >37 wk gestation. *, p < 0.05.
Each antimicrobial has a unique spectrum of antimicrobial activity and salt sensitivity
The minimal effective concentrations of each peptide against a range of bacteria and against a fungus are shown in Table I. HBD-3 was generally the most potent peptide, displaying activity against both Gram-positive and -negative bacteria in low microgram per milliliter concentrations. HBD-2 mainly had antimicrobial activity against Gram-negative bacteria, with a slightly lower potency compared with HBD-3. Notable exceptions to this trend were HBD-2, which exhibited the most potent antimicrobial activity against M. catarrhalis and K. pneumoniae, two of the more common bacteria causing pneumonia after the neonatal period (21, 22). HBD-1 generally had the least activity, but was active against both Gram-positive and -negative bacteria. Due to limited quantities of HBD-1, this peptide was tested against a smaller number of bacteria. HBD-1 and HBD-2 both displayed salt-sensitive antimicrobial activity, whereas HBD-3 showed salt-insensitive activity against specific Gram-positive and -negative organisms. In contrast, LL-37, a human neutrophil cathelicidin, also showed potent Gram-positive and -negative microbicidal activities, but exhibited salt-insensitive activity only against Gram-negative bacteria. Thus, each peptide has a unique spectrum of activity and salt sensitivity.
Table I. Unique spectrums of antimicrobial activity for each -defensin and cathelicidin1
Discussion
Of the three defensins studied, HBD-2 was the predominant -defensin expressed in developing human lung. HBD-2 message was developmentally regulated, induced by the proinflammatory factor IL-1, and decreased by dexamethasone. HBD-2 is known to be regulated by IL-1 in human skin (23), gingival mucosa (24), intestinal epithelia (25), and respiratory epithelia (7, 26, 27), but to our knowledge, these are the first data to describe developmental regulation of the IL-1-induced -defensin response in human airway epithelia. HBD-2 protein abundance in neonatal tracheal aspirates increased as a function of gestational age. HBD-1 message was detected to a lesser degree and was only induced by dexamethasone (Fig. 3). Although HBD-1 is generally considered constitutive rather than inducible, there are precedents for the possibility of steroidal or hormonal regulation of HBD-1, because HBD-1 is highly expressed in breast milk (28) and in the urine of pregnant women (14). LL-37 and HBD-3 mRNAs were generally not detected in airway epithelial culture samples. Additionally, each antimicrobial peptide exhibited a unique spectrum of antimicrobial activity and salt sensitivity against the most common bacteria causing sepsis in the neonate.
These data show that in addition to known deficiencies in acquired immunity (2, 3) and other innate immune factors (4, 5, 6), the lungs of premature infants have a decreased abundance of human -defensins. Because tissues quickly mature in the fetal lung explant model, the significant increase in HBD-2 message after fetal tissue has been cultured for 24 h (Fig. 2) supports the idea that HBD-2 expression is diminished at midgestation and is developmentally regulated; however, this result could also have been due to a variety of other factors, including stress or trauma to tissues or lack of serum or hormone factors in the culture model. Developmental regulation of HBD-2 was corroborated by the increase in HBD-2 message in the survey of lung tissues at various developmental ages (Fig. 1) and the in vivo protein abundance in term neonatal tracheal aspirates after 37 wk gestation (Fig. 5). The observation that LL-37 and IL-1 peptide abundance was not elevated in the tracheal aspirate samples from infants born at >37 wk gestation also supports the developmental regulation of HBD-2. Because the samples were standardized by volume, if the increase in HBD-2 on day 2 was simply due to dilutional or protein concentration effects, we would have expected LL-37 and IL-1 to be similarly increased. IL-1 abundance in tracheal aspirates was not significantly different among the groups. This suggests that although a stimulus for HBD-2 induction was present at all time points, other factors necessary for the expression of HBD-2 were not matured until near term.
The expression of other defensins and antimicrobial peptides is developmentally regulated in humans as well as other animals. In animal studies, defensin or defensin-like peptides increase during gestation in the intestinal tissues of sheep (29, 30) and mice (31). Similarly, in humans, HD-5 and HD-6 mRNAs were lower in midgestation fetal intestinal samples compared with adult samples (5, 6). Neonatal skin in mice and humans expresses increased cathelicidin message during the perinatal period (32). Developmental regulation of defensins in the lung has been shown with bovine tracheal antimicrobial peptide (33) and HBD-1 (34). In contrast to our data showing increased HBD-2 abundance in term infants, other recent investigations of -defensins in neonatal tracheal aspirates showed increased abundance after infection, but no developmental regulation (13). This may reflect maturational differences in the age ranges in each study, because the most developed group in the study by Schaller-Bals et al. (13) included tracheal aspirates from more preterm time points (wk 31–40 gestation), and the present study had a more narrow range at more developed time points (>37 wk gestation).
Each of the four innate host factors studied exhibited unique spectrums of antimicrobial activity against the most common bacterial pathogens in neonatal infections, including E. coli, Enterococcus species, group B Streptococcus, H. influenzae, Klebsiella species, L. monocytogenes, P. aeruginosa, coagulase-negative Staphylococcus, and Staphylococcus aureus (35, 36, 37, 38, 39). These antimicrobial factors have complimentary activities, because each peptide has different bacterial spectrum and salt sensitivity. The combined expression of many innate immune factors will have a wider range of antimicrobial activity. Although the abundance of the defensins and LL-37 in neonatal tracheal aspirates appears greatly lower than the ranges at which they exert antimicrobial activity, several factors need to be kept in mind. Determination of the correct dilution factor for saline instilled to obtain fluid lining the epithelial has been problematic (40, 41). However, dilutions for obtaining bronchoalveolar lavage fluid or tracheal aspirates have been estimated to be 1 ml of airway fluid/10–100 ml of bronchoalveolar lavage fluid based on urea concentration measurements (42). Assuming up to a 100-fold dilution estimate, the HBD-2 protein levels in the >37 wk gestation day 2 tracheal aspirates samples would average 75 ng/ml. Defensins could act in concert with numerous other antimicrobial factors in airway surface liquid, such as lactoferrin, lysozyme, secretory leukocyte protease inhibitor (43), and CCL20 (16). Furthermore, defensins might exhibit synergistic antimicrobial activity in a manner similar to that seen in lysozyme, lactoferrin, and secretory leukocyte protease inhibitor (44). Also, there may be microenvironments that have much higher concentrations of defensins, such as airway surface liquid, the paracellular space, or directly under the basolateral membrane. Defensins may be in high enough concentrations in these microenvironments to exert direct antimicrobialactivity. Alternatively, defensins and LL-37 may be acting by modulating local immunity (45, 46, 47, 48).
Two unexpected findings were the selective salt insensitivity of HBD-3 against specific strains of bacteria and the salt insensitivity of LL-37 only against Gram-negative bacteria. HBD-3 was identified by Schroeder et al. (8) using isolation methods designed to identify peptides that bind to Gram-positive bacteria. Others identified HBD-3 using bioinformatic approaches (8, 49). Schroeder et al. (8) showed that HBD-3 had salt-insensitive antimicrobial activity, but this salt sensitivity was only tested against a single strain of S. aureus. We found similar HBD-3 salt insensitivity only against specific strains of S. aureus, other staphylococcal organisms, and some Gram-negative organisms (see Table I). We found that HBD-3 displayed salt insensitivity that was species and even strain specific. LL-37 is known to have salt-insensitive microbicidal activity (50), but we found salt insensitivity only against Gram-negative bacteria. LL-37 also displayed species and strain variability in a manner similar to that reported previously (50).
Because LL-37 is present in airway epithelia and neutrophils, it is difficult to determine the relative epithelial contribution of LL-37 in some of these studies. For example, the LL-37 message detected in whole lung tissue at various ages (Fig. 1) may have been derived from neutrophils in the tissue. Similarly, we only detected LL-37 message in the start tissue from the fetal lung explants, not after incubation for 24 h (data not shown). Because neutrophils do not survive in fetal lung explant culture, any message detected after 24 h in the cultured samples should originate from lung tissue. However, the signal in the start tissues could again have been from nonepithelial sources, such as neutrophils present in freshly isolated tissue. One potential problem with the hypothesis of LL-37 transcripts originating from neutrophils is that peripherally circulating neutrophils have little protein synthesis. In neutrophils derived from adults, transcription of many proteins primarily occurs in bone marrow. However, it is not known whether this is also the case in the developing fetus. Thus, we found little evidence of LL-37 expression in human airway epithelia under the culture conditions used in these studies.
As with all models, there are limitations to the in vitro cultures used in these studies. The failure to see the expected induction of the -defensins by heat-killed E. coli in fetal lung explants may reflect the inability of bacterial products to adequately penetrate into cultured tissue in this model or the absence of myeloid cells, which may be the predominant bacterial sensors for epithelia (51) and are lost in culture models (52). In human epithelial models, HBD-2 expression has been shown to be up-regulated by some bacterial strains (25, 26, 53). In contrast to the fetal lung explant model, we did not see a significant inhibition of HBD-2 message by dexamethasone in the primary cultures of human airway epithelia. However, the low basal expression of HBD-2 mRNA in the unstimulated airway epithelia may preclude determination of inhibition with the primary culture model. Failure to detect HBD-3 may also reflect shortcomings of the model or the stimuli used in these studies. We previously amplified HBD-3 cDNA from human fetal lung after 30 cycles of PCR (49). Harder et al. (8) detected HBD-3 in tracheal primary cultures stimulated with a clinical strain of mucoid P. aeruginosa. Similarly, Duits et al. (54) found that rhinovirus-16-stimulated cultures of primary bronchial epithelial cells exhibited increased HBD-2 and HBD-3 mRNAs.
HBD-2 may play a significant role in pulmonary innate immunity. Because it is developmentally regulated, decreased abundance of HBD-2 may contribute to the increased incidence of pulmonary infections in premature neonates. The prevalence of HBD-2 message in the lung, its inducibility, and its potent antimicrobial activity against common respiratory pathogens lends credence to this assertion. We speculate that premature infants may produce less HBD-2 in response to a bacterial challenge. This could allow specific HBD-2-sensitive bacteria to more easily colonize and infect the respiratory tract. Interestingly, HBD-2 has poor antimicrobial activity against H. influenzae (20), which is a not only a common respiratory pathogen, but is also a common commensal in the human respiratory tract. The abilities of H. influenzae to colonize and cause disease in the human airway and to evade killing by innate immune factors may be partly due to its resistance to a prevalent -defensin in the airway.
Because defensins and cathelicidins have properties other than their antimicrobial abilities, these data may also have implications for adaptive immunity and tissue development. Defensins signal adaptive immunity by inducing chemotaxis of immature dendritic cells and memory T cells (45). Lower levels of these peptides may contribute to the decreased adaptive immune responses in neonates and premature infants. This role that defensins play in recruiting APCs may be especially important early in the neonatal period because adaptive immune signaling is inhibited more than innate immune signaling (55). Because LL-37 induces angiogenesis (56) and other defensins are mitogenic for epithelial cells and fibroblasts (57), these data may also have implications for tissue development and wound repair.
In summary, we found that HBD-2 was the most prevalent -defensin in human lung cultures and neonatal tissues. HBD-2 message and protein were developmentally regulated and were less abundant in premature compared with term or post-term samples. Each defensin displayed a unique spectrum of antimicrobial activity and salt sensitivity against a wide variety of neonatal and respiratory pathogens. Lower levels of HBD-2 in the lungs of premature infants may be one factor leading to an immature innate immune system and contributing to the increased susceptibility of neonates to respiratory infections.
Acknowledgments
We acknowledge the support of the Cell Culture Core in preparing airway epithelial cultures. We thank Hong Peng Jia for his support in producing HBD-1. We thank Dr. David Proud for providing the rabbit anti-HBD2 antisera and for technical assistance with the HBD-2 ELISA, and Dr. Michael Acarregui for assistance with the fetal lung explant cultures. We thank Monica Lindh for technical assistance with the LL-37 ELISA. We also thank the nurses and the General Clinical Research Center at University of Iowa for their help in obtaining the neonatal tracheal aspirate samples.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grants HL61234, HL07638, HL67992, 2K1227748 (Child Health Research Career Development Award), P30DK54759, and RR00059 (General Clinical Research Center), and the Swedish Foundation for International Cooperation in Research and Higher Education.
2 Address correspondence and reprint requests to Dr. Timothy D. Starner, Department of Pediatrics, University of Iowa, 2JCP, 200 Hawkins Drive, Iowa City, IA 52242. E-mail address: timothy-starner@uiowa.edu
3 Abbreviations used in this paper: HD-5, human -defensin 5; HBD, human -defensin.
Received for publication March 10, 2004. Accepted for publication November 16, 2004.
References
Langrish, C. L., J. C. Buddle, A. J. Thrasher, D. Goldblatt. 2002. Neonatal dendritic cells are intrinsically biased against Th-1 immune responses. Clin. Exp. Immunol. 128:118.
Lawton, A. R.. 1986. Ontogeny of B cells and pathogenesis of humoral immunodeficiencies. Clin. Immunol. Immunopathol. 40:5.
Suen, Y., S. M. Lee, J. Qian, C. van de Ven, M. S. Cairo. 1998. Dysregulation of lymphokine production in the neonate and its impact on neonatal cell mediated immunity. Vaccine 16:1369.
Levy, O., S. Martin, E. Eichenwald, T. Ganz, E. Valore, S. F. Carroll, K. Lee, D. Goldmann, G. M. Thorne. 1999. Impaired innate immunity in the newborn: newborn neutrophils are deficient in bactericidal/permeability-increasing protein. Pediatrics 104:1327.
Salzman, N. H., R. A. Polin, M. C. Harris, E. Ruchelli, A. Hebra, S. Zirin-Butler, A. Jawad, E. Martin Porter, C. L. Bevins. 1998. Enteric defensin expression in necrotizing enterocolitis. Pediatr. Res. 44:20.
Mallow, E. B., A. Harris, N. Salzman, J. P. Russell, R. J. DeBerardinis, E. Ruchelli, C. L. Bevins. 1996. Human enteric defensins: gene structure and developmental expression. J. Biol. Chem. 271:4038.
Singh, P. K., H. P. Jia, K. Wiles, J. Hesselberth, L. Liu, B. A. Conway, E. P. Greenberg, E. V. Valore, M. J. Welsh, T. Ganz, et al 1998. Production of -defensins by human airway epithelia. Proc. Natl. Acad. Sci. USA 95:14961.
Harder, J., J. Bartels, E. Christophers, J. M. Schroder. 2001. Isolation and characterization of human -defensin-3, a novel human inducible peptide antibiotic. J. Biol. Chem. 276:5707.
Bals, R., X. Wang, M. Zasloff, J. M. Wilson. 1998. The peptide antibiotic LL-37/hCAP-18 is expressed in epithelia of the human lung where it has broad antimicrobial activity at the airway surface. Proc. Natl. Acad. Sci. USA 95:9541.
Agerberth, B., J. Grunewald, E. Castanos-Velez, B. Olsson, H. Jornvall, H. Wigzell, A. Eklund, G. H. Gudmundsson. 1999. Antibacterial components in bronchoalveolar lavage fluid from healthy individuals and sarcoidosis patients. Am. J. Respir. Crit. Care Med. 160:283.
Yoshio, H., M. Tollin, G. H. Gudmundsson, H. Lagercrantz, H. Jornvall, G. Marchini, B. Agerberth. 2003. Antimicrobial polypeptides of human vernix caseosa and amniotic fluid: implications for newborn innate defense. Pediatr. Res. 53:211.
Marchini, G., S. Lindow, H. Brismar, B. Stabi, V. Berggren, A. K. Ulfgren, S. Lonne-Rahm, B. Agerberth, G. H. Gudmundsson. 2002. The newborn infant is protected by an innate antimicrobial barrier: peptide antibiotics are present in the skin and vernix caseosa. Br. J. Dermatol. 147:1127.
Schaller-Bals, S., A. Schulze, R. Bals. 2002. Increased levels of antimicrobial peptides in tracheal aspirates of newborn infants during infection. Am. J. Respir. Crit. Care Med. 165:992.
Valore, E. V., C. H. Park, A. J. Quayle, K. R. Wiles, P. B. McCray, Jr, T. Ganz. 1998. Human -defensin-1: an antimicrobial peptide of urogenital tissues. J. Clin. Invest. 101:1633.
Saiman, L., S. Tabibi, T. D. Starner, P. San Gabriel, P. L. Winokur, H. P. Jia, P. B. McCray, Jr, B. F. Tack. 2001. Cathelicidin peptides inhibit multiply antibiotic-resistant pathogens from patients with cystic fibrosis. Antimicrob. Agents Chemother. 45:2838.
Starner, T. D., C. K. Barker, H. P. Jia, Y. Kang, P. B. McCray, Jr. 2003. CCL20 is an inducible product of human airway epithelia with innate immune properties. Am. J. Respir. Cell Mol. Biol. 29:627.
Karp, P. H., T. O. Moninger, S. P. Weber, T. S. Nesselhauf, J. L. Launspach, J. Zabner, M. J. Welsh. 2002. An in vitro model of differentiated human airway epithelia: methods for establishing primary cultures. Methods Mol. Biol. 188:115.
Proud, D., S. P. Sanders, S. Wiehler. 2004. Human rhinovirus infection induces airway epithelial cell production of human -defensin 2 both in vitro and in vivo. J. Immunol. 172:4637.
Gudmundsson, G. H., B. Agerberth, J. Odeberg, T. Bergman, B. Olsson, R. Salcedo. 1996. The human gene FALL39 and processing of the cathelin precursor to the antibacterial peptide LL-37 in granulocytes. Eur. J. Biochem. 238:325.(Timothy D. Starner, Birgi)