Main

CLD is a major problem in the care of very low birth weight infants (1). Small preterm neonates with immature lungs who are treated with mechanical ventilation and oxygen have a high risk of experiencing CLD (2–4). This disease is typified by an initial increase in the number of macrophages and polymorphonuclear lymphocytes and high levels of proinflammatory cytokines (TNF-α, IL-1β, IL-6, and IL-8) in TAF (5, 6). These cytokines induce the inflammatory response seen in CLD. We found that the TAF level of IL-6 and TNF-α already at d 2 and 3 could predict which infants would later develop CLD (7). In later stages, lung fibrosis often follows, leading to death or long-term oxygen supplementation (7, 8).

Ureaplasma urealyticum can be found in the genital tract of 40–80% of women of child-bearing ages (9). There is some evidence supporting the hypothesis that vertically transmitted colonization and infection with U. urealyticum is an important risk factor for CLD (8, 10, 11), possibly by inducing a cytokine-mediated inflammation. However, the contribution of U. urealyticum to the development of CLD is still controversial (9). U. urealyticum has been isolated from blood, cerebrospinal fluid, TAF, and lung tissue (12), and evidence exists that it can cause invasive diseases in preterm neonates. Respiratory tract colonization has been associated with higher incidence of pneumonia (13), respiratory distress syndrome, intraventricular hemorrhage, and CLD (14). Experimental colonization of premature baboons with U. urealyticum caused an acute bronchiolitis with epithelial ulceration and polymorphonuclear infiltration (15). Newborn mice inoculated with U. urealyticum developed an acute interstitial pneumonia (16). These findings suggest that U. urealyticum may elicit an inflammatory response also in preterm infants.

We have investigated the potential pulmonary pathogenicity of U. urealyticum in vitro. We studied whether a human monocytic cell line, differentiated to macrophages, a rat alveolar macrophage cell line, and premature human infant lung macrophages could produce TNF-α and IL-6 in direct response to U. urealyticum at the mRNA and protein levels.

METHODS

Cell culture

Cell line.

The human monocytic cell line (THP-1; American Type Culture Collection, Rockville, MD, U.S.A.) was maintained in RPMI 1640 medium (Sigma Chemical Co, St. Louis, MO, U.S.A.) supplemented with 2 mM l-glutamine (Sigma), 5 mM mercaptoethanol, 10 mM HEPES (GibcoBRL, Gaithersburg, MD, U.S.A), and 10% heat-inactivated fetal bovine serum Myclone (GibcoBRL) at 37°C, 5% CO2. Before cells were used, they were split and seeded in fresh medium for 48 h and then differentiated with PMA (Sigma) at 10−8 M with 5 × 105 cells/mL in Teflon-coated bottles for 48 h. To confirm the population of macrophages, flow cytometry analysis was performed. The expression of macrophage-specific cell surface antigens showed that 96.2% of cells expressed CD14 and CD45, and 99.9% of cells expressed HLA-DR, indicating a full differentiation of macrophages. The rat alveolar macrophage cell line (Nr8383) (American Type Culture Collection) was maintained in Ham's F-12 medium (GibcoBRL) and supplemented with 15% heat-inactivated fetal bovine serum Myclone (GibcoBRL).

Human TAF macrophages.

Premature infant lung macrophages were obtained from four patients who were receiving mechanical ventilation. The clinical data for the patients are shown in Table 1. The aspirates were obtained at times when routine endotracheal suctioning was performed as follows: 0.5 mL of 0.9% saline solution was instilled into the endotracheal tube. The infant was then reconnected to the ventilator. After the ventilator was disconnected, the suction catheter was inserted past the tip of the endotracheal tube. Aspiration was then performed during the withdrawal of the catheter. After this, the catheter was flushed with a further 0.5 mL of 0.9% saline solution into a sterile specimen trap. Then the TAF was filtered through a special cell strainer (Falcon, Franklin Lakes, NJ, U.S.A.) and centrifuged, and the supernatant was discarded. The cells were then cultured in 96-microwell plates (Costar, Cambridge, MA, U.S.A.) at 5 × 105 cells/mL in RPMI 1640 medium with 2 mM l-glutamine, 10% heat-inactivated fetal bovine serum Myclone (GibcoBRL), and penicillin (100 U/mL)–streptomycin (100 μg/mL; Sigma). Macrophages were left to adhere to the plastic during a 2-h preincubation at 37°C, 5% CO2, and the supernatant together with nonadhesive cells was discarded. This part of the study was approved by the ethical committee of the Karolinska Hospital, and the parents gave their informed consent.

Table 1 Clinical data of patients from whom TAF was obtained Abbreviations: GA, gestational age; BW, body weight; RDS, respiratory distress syndrome.
Full size table

Preparation of U.

urealyticum antigen

U. urealyticum serotype standard strain 8 (T960; ATCC) was cultured at 37°C in 1.5 L of Ureaplasma broth medium containing 22.5 g/L of trypticase soy broth; 16.5% horse serum; 7.5% of 25% fresh yeast extract; 0.36% urea; 380,000 U/L penicillin G; and phenol red. Cells were harvested in late log phase by centrifugation at 30,000 ×g for 90 min at 4°C. The pellet was washed three times by resuspension in PBS and centrifuged as above for 30 min. After the final wash, the Ureaplasma was resuspended in a total volume of 2 mL of PBS. The number of CCUs of the concentrated suspension was determined in duplicate by 10-fold titration in Ureaplasma broth. The remaining U. urealyticum suspension was heat-killed by incubation in a waterbath at 56°C for 20 min. Complete killing was assured by incubating 25 μL of the suspension on Ureaplasma agar as well as incubation in Ureaplasma broth without urea supplement and subsequent subculture on agar on d 1, 3, and 7. The latter procedure was used as the heat-killed suspension of U. urealyticum produced a prompt color change in Ureaplasma broth because of urease activity. The U. urealyticum antigen was stored at −80°C in 0.1-mL aliquots until used.

Antigen stimulation

Human and rat macrophages from cell lines were distributed into 24-microwell plates at a concentration of 1 × 106 cells/mL and stimulated with 100 ng/mL of LPS (Sigma) or 4 × 106 to 4 × 108 CCU/mL of U. urealyticum antigen for 24 h at 37°C and 5% CO2. For rat IL-6 and TNF-α mRNA analysis, the stimulation was continued for 4 h. As controls, the same cells were incubated with media alone. All experiments were repeated four to six times. Stimulation in human TAF macrophages was performed with U. urealyticum antigen (4 × 108 CCU/mL) and medium alone for 24 h at 37°C and 5% CO2.

Cytokine analysis

All the supernatants were collected after stimulation and stored at −70°C for analysis of cytokines. ELISA kits for human TNF-α, IL-6, and rat TNF-α were obtained from R & D System (Abingdon, Oxon, U.K.). Supernatants were used for cytokine determination by the quantitative “sandwich” ELISA technique, using MAbs specific for the tested cytokines. The limits of detection of human IL-6 and TNF-α and rat TNF-α are 0.7, 4.4, and 10 pg/mL, respectively.

Specific bioassay

Rat IL-6 bioactivity analysis was performed by a specific bioassay using an IL-6–dependent cell line–7TD1 hybridoma line as previously described (17). In brief, the cells grew in an IL-6 medium, and then the cells were starved with a lower amount of IL-6 before the experiment. The viability of the cells was 70–95%. Then cells were resuspended in the medium without IL-6 to 20000 cells/mL, and 100 μL of this suspension was added to serial dilutions of samples in the same volume. A recombinant mouse IL-6 (Genzyme; Novakemi AB, Stockholm, Sweden) was used as a standard. After 72 h at 37°C in 5% CO2 incubation, the cultures were pulsed with 1 μCi of [3H]thymidine in 10 μL of saline solution, and further incubated for 6 h. The culture was then harvested onto fiberglass strips. Filter strips were dried and punched into vials, and scintillation liquid was added. The radioactivity was determined. Calculation of the IL-6 concentration in the samples was performed referring to the standard curve.

RT-PCR

Total RNA was extracted from 4 × 106 cells with RNAzolB (Biotecx Laboratories; Houston, TX, U.S.A.) after the different treatments. The cells were suspended in 1 mL of ice-cold RNAzolB, and total RNA was extracted according to the manufacturer's instructions. The final RNA pellet was air-dried, and the RNA was resuspended in 20 μL of diethyl-pyrocarbonate–treated water; 10 μL was used for spectrophotometric determination of the RNA concentration at 260 nm, and quality was determined by assessing the 230- to 320-nm profile. The remaining 10 μL was stored at −70°C.

First-strand cDNA synthesis of total RNA was performed using SuperScript RNase H− Reverse Transcriptase (GibcoBRL) and random hexamer primers [pd(N)6; Amersham Pharmacia Biotech, Uppsala, Sweden]. In brief, 1 μg of RNA together with 100 pmol of pd(N)6 was heated at 70°C for 10 min; then 0.1 mmol of DTT, 10 μmol of dNTP, 60 U of RNasin (Boehringer Mannheim, Mannheim, Germany), and 300 U of superscript were added, and the mixture was incubated at 45°C for 2 h.

Specific oligonucleotide primers were synthesized for human IL-6, human TNF-α (Scandinavian Gene Synthesis, Köping, Sweden), and the housekeeping gene human G3PDH (CyberGene AB, Huddinge, Sweden). The rat IL-6, TNF-α, and G3PDH primers were obtained from Innovagen (Lund, Sweden). The sequences of the 3′- and 5′-primers used are shown in Table 2.

Table 2 Primers used in the RT-PCR reactions
Full size table

PCR using Taq polymerase (final concentration, 0.025 U/μL; GibcoBRL) was performed in a final volume of 25 μL containing 2 μL of cDNA for TNF-α and G3PDH and 4 μL for IL-6 in a DNA Thermocycler 480 (Perkin Elmer, Norwalk, CT, U.S.A.). The other reaction constituents included MgCl2 (2 mM), dNTP (0.2 mM), PCR buffer, and 3′- and 5′-primers (0.5 μM each). PCR was conducted for 40 cycles for human IL-6, for 35 cycles for human TNF-α, and for 27 cycles for human G3PDH under the following conditions: 1 min denaturation at 94°C; 1 min annealing at 54°C for human IL-6, at 56°C for human TNF-α, and at 60°C for human G3PDH; and 1 min extension at 72°C, followed by a final extension for 5 min at 72°C and cooling down to 4°C. PCR was conducted for 40 cycles for rat IL-6 and rat TNF-α with 1 min at 94°C, 30 s at 54°C for rat IL-6 and at 58°C for rat TNF-α, and 3 min at 72°C; and for 35 cycles for rat G3PDH with 1 min at 94°C, 1 min at 60°C, and 1 min at 72°C, followed by a final extension for 5 min at 72°C and cooling down to 4°C.

The PCR products were separated on a 1.5% agarose gel (GibcoBRL). The ethidium bromide–stained gel was photographed under UV light with the DC120 Digital Zoom Camera (Eastman Kodak Company, Rochester, NY, U.S.A.), and the net intensities of the PCR products were analyzed with the Kodak Digital Science Electrophoresis Documentation and Analysis System 120 (Eastman Kodak).

Data analysis

ELISA and bioassay data from pooled experiments were reported as means and the 95% confidence interval. Data were analyzed by t test or one-way ANOVA. Nonnormality paired data were analyzed by Wilcoxon matched pairs test. A p value of <0.05 was considered to be significant.

RESULTS

TNF-α production.

U. urealyticum antigen stimulated the production of TNF-α in the THP-1 cell line and the rat alveolar macrophage cell line in a dose-dependent manner (Fig. 1, A and B). The highest concentration of U. urealyticum (4 × 108 CCU/mL) produced levels of TNF-α in the THP-1 cell line close to those with 100 ng/mL LPS. In the rat alveolar macrophage cell line, high concentrations (≥4 × 107 CCU/mL) of U. urealyticum stimulated the production of TNF-α (p < 0.05), but to much lower levels than stimulation with 100 ng/mL LPS (mean, 10,545 pg/mL).

Figure 1
figure 1

TNF-α production in the THP-1 cell line (PMA-differentiated monocytes) stimulated by different doses of U. urealyticum antigen or LPS for 24 h (A). TNF-α production in the rat alveolar macrophage cell line stimulated by different doses of U. urealyticum antigen or LPS for 24 h (B).

Full size image

The production of TNF-α from lung macrophages from premature infants stimulated by U. urealyticum antigen at 4 × 108 CCU/mL was studied, and the TNF-α levels with U. urealyticum stimulation were 14.4 to 84.0% higher than the unstimulated (19,554 versus 17,098 pg/mL; 3,847 versus 2,524 pg/mL; and 3,388 versus 1,841 pg/mL, referring to samples 1–3 in Table 1;p < 0.05).

IL-6 production.

U. urealyticum antigen stimulated the production of IL-6 the THP-1 cell line (Fig. 2A) and the rat alveolar macrophage cell line (Fig. 2B) in a dose-dependent way. High concentrations (≥4 × 107 CCU/mL) of U. urealyticum antigen stimulated the production of IL-6 in the THP-1 cell line, but to significantly lower levels than with 100 ng/mL LPS (632 pg/mL;p < 0.05). In the rat alveolar macrophage cell line, high concentrations (≥4 × 106 CCU/mL) of U. urealyticum antigen stimulated the production of IL-6, but to much lower levels than stimulation with 100 ng/mL LPS (mean, 62,400 pg/mL).

Figure 2
figure 2

IL-6 production in the THP-1 cell line (PMA-differentiated monocytes) stimulated by different doses of U. urealyticum antigen or LPS for 24 h (A). IL-6 production in the rat alveolar macrophage cell line stimulated by different doses of U. urealyticum antigen or LPS for 24 h (B).

Full size image

IL-6 production in the premature infant lung macrophages stimulated by U. urealyticum (4 × 108 CCU/mL) was studied, and the IL-6 levels of U. urealyticum stimulation were 45.5 to 268.4% higher than the unstimulated (1,259 versus 861 pg/mL; 596 versus 277 pg/mL; 5,493 versus 3,776 pg/mL; 11,836 versus 3,213 pg/mL; 80,789 versus 50,307 pg/mL, referring to samples 4–8 in Table 1;p < 0.05).

Cytokine mRNA expression.

TNF-α and IL-6 gene expression stimulated by U. urealyticum antigen (4 × 108 CCU/mL) is shown in Figure 3. In the THP-1 cell line, both IL-6 and TNF-α mRNA were induced by U. urealyticum compared with unstimulated cells (I). The ratio of net intensity of IL-6/G3PDH was 0.43 in stimulated cells and 0.13 in unstimulated cells. TNF-α/G3PDH was 1.35 in stimulated cells and 0.79 in unstimulated cells. Similar results were found in the rat alveolar macrophage cell line (II): IL-6/G3PDH was 0.35 in stimulated cells and 0.03 in unstimulated cells, and TNF-α/G3PDH was 0.65 in stimulated cells compared with 0.04 in unstimulated cells.

Figure 3
figure 3

IL-6 and TNF-α mRNA expression stimulated by U. urealyticum antigen in the THP-1 cell line (B, I) and the rat alveolar macrophage cell line (B, II) showed that both IL-6 and TNF-α mRNA have been induced compared with unstimulated cells. The ratio shows the net intensity of the cytokine mRNA vs G3PDH (A).

Full size image

DISCUSSION

In this study, we found a dose-response expression of IL-6 and TNF-α protein and increased levels of IL-6 and TNF-α mRNA in vitro after stimulation of macrophages with U. urealyticum antigen. It has long been debated whether U. urealyticum is one of the pathogenic factors in the development of CLD. Increasing clinical evidence for the association of this organism with clinical disease has emerged (7, 10, 21–24), and a recent meta-analysis also supported an independent role of U. urealyticum in the development of CLD (21). Our present data support a pathogenic role for U. urealyticum.

The initial pathologic changes of CLD are typified by an increase in the number of macrophages and polymorphonuclear lymphocytes in the lung, and high TAF levels of proinflammatory cytokines (25). Preterm infants who acquire CLD demonstrate increased levels of proinflammatory cytokines and chemokines in lung lavage fluid as early as d 1 of life (7). The proinflammatory cytokine levels increase during the first week of life, reach a peak at approximately 2 wk, and remain elevated for several weeks (6, 26). We have previously found that increased levels of especially IL-6 and TNF-α are associated with the development of CLD (7). The exact mechanism explaining the correlation between the increased IL-6 levels and the development of CLD is not known. These previous findings, together with the present results, do, however, demonstrate an imbalance of the complex network of inflammatory mediators, which may lead to irreversible lung tissue destruction and fibrosis (26). Our findings support the correlation between U. urealyticum colonization and TNF-α levels in TAF previously found. However, in contrast to us, these authors found no correlation between IL-6 and U. urealyticum (27).

The human macrophage cell line was much more susceptible to Ureaplasma than the rat cell line. This interspecies difference needs to be recognized in further studies. The findings in the neonatal alveolar macrophages are of course the most interesting but come from only four different infants. These cells are already stimulated in vivo in the babies. They therefore already produced increased and, in some instances, very high cytokine levels. Still, U. urealyticum antigen could in all cases increase the cytokine production further.

There was a dose-response relationship between the concentrations of U. urealyticum and the cytokine response. It is difficult to compare in vitro bacterial concentrations with those present in vivo in neonates. Very marked differences in bacterial numbers have been found in vivo. The numbers of bacteria adhering to the airway surface might also be very different from those found in the bronchial fluid.

Much information concerning the precise role of U. urealyticum in lung disease in newborns comes from necropsy, but little is from clinical and laboratory studies of surviving babies (28, 29). Differences in host immune response, serotype pathogenicity, genetic predisposition, or other factors could affect the outcome of U. urealyticum infection (16, 30). The pathogenicity of U. urealyticum may be caused in part by its ability to produce phospholipases A and C, which can catalyze the release of arachidonic acid from membrane phospholipid (31). Arachidonic acid and its metabolites are potent lipid mediators that can induce the release of proinflammatory cytokines (32, 33). Further studies should elucidate which part of the U. urealyticum is most important in eliciting the cytokine response.

The inflammatory response caused by U. urealyticum seems to be of a type similar to that of other infectious agents. The response, however, seems to be somewhat less pronounced than that caused by bacteria of higher virulence. This gives less pronounced symptoms. The inflammation persists for a longer time, giving a high risk of developing the chronic changes of CLD.

In conclusion, we have demonstrated that U. urealyticum can stimulate macrophages to produce high levels of proinflammatory cytokines in vitro. This supports the hypothesis that U. urealyticum could be an etiological factor associated with the development of CLD in premature infants.