Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The innate arm of the immune system, which comprises macrophages, neutrophils, natural killer cells, and the complement system, is generally regarded as the body's "first line of defense" against foreign antigens. Recently, our understanding of the role of the innate immune system has been expanded such that it is now recognized not only as a front-line defense mechanism but also as an important bridge to the adaptive immune system (1, 2). This renewed attention to innate immunity has focused our interests on the role of alveolar macrophages (AM). Specifically, we are interested in the possible differences between newborn and adult AM in terms of their responses to bacteria and bacterial products. This interest is based on the now well-established notion that despite the differentiation of the immune system early in fetal life, full functional capacity is only obtained after a maturation process. The consequence of a delay in this maturation process may be manifested in the increased susceptibility to infections seen in early life (3).

In the respiratory tract, AM may respond to pathogens by two important and effective means. First, AM directly bind, phagocytose, and kill pathogens, and whereas some earlier studies have demonstrated deficient phagocytic capacity of AM in the neonates compared with adults (4, 5), other contradictory reports suggest that neonatal and adult macrophages are equivalent in their capacity to phagocytose bacteria (6, 7). Second, AM are able to secrete a large range of mediators, some acting directly on the pathogens while others such as chemokines exert their effects indirectly by recruiting other components of the immune system. Previous studies have established the critical role of AM in the process of recruitment of neutrophils in response to bacterial infections (8) and have shown that in neonates this process is relatively inefficient (11). Thus, we have attempted to clarify these functions in newborn animals.

We have compared the phagocytic capacity of AM from newborn and adult animals for selected gram-negative and gram-positive bacteria. In addition, we have also investigated the cellular response of neonates to Moraxalla catarrhalis, an airway bacterium that we have shown is capable of eliciting a rapid neutrophil influx into the tissues and airspaces of the respiratory tract (12). In particular, we were interested in comparing the response of both the lower and the upper respiratory tract to bacteria, and the ability of neonatal AM to express chemotactic chemokines in response to lipopolysaccharide (LPS) challenge. Finally, we have also investigated the expression of CD14 and LPS binding protein (LBP) by AM and discussed the possible significance of LBP production by macrophages from animals of different ages.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

PVG rats were used throughout all experiments and were specific pathogen free derived and barrier maintained. Animals were housed on low-dust bedding to minimize background airway inflammation and were serologically free of Sendai infection and other known pathogens. Animals were fed ad libitum on autoclaved rat chow. All animal experimentation was conducted with prior approval of the TVW Telethon Institute for Child Health Research Animal Ethics Committee, which operates under guidelines set down by the Australian National Health and Medical Research Council.

Bronchoalveolar Lavage

Rat AM were obtained by bronchoalveolar lavage (BAL) of animals within 24 h of birth (newborn, 1 d old), and 7-d-old, 21-d-old (weanling), and adult animals (9 to 11 wk old). Rats were anesthetized with ether before administration of a lethal dose (0.5 ml/100 g body weight) of Lethobarb (pentobarbitone sodium; Virbac Pty. Ltd. NSW, Sydney, Australia) delivered intraperitoneally via a 21-gauge needle. Tracheas were exposed and catheterized. For adult animals, lungs were lavaged with 10 ml of ice-cold, phosphate-buffered saline (pH 7.4) (PBS) that was instilled gently into the lungs and withdrawn immediately. In newborn animals, 200 µl of PBS was instilled in the lungs in a similar fashion. Lavage fluid from several animals from the same age group was pooled and centrifuged at 4°C for 7 min at 600 × g. Cells were resuspended in macrophage serum-free medium (MSFM) (GIBCO BRL, Life Technologies Inc., Grand Island, NY) containing antibiotic and antimycotic (100 µg/ml penicillin, 100 µg/ml streptomycin, and 250 ng/ml amphotericin B) (Sigma-Aldrich Pty. Ltd. NSW, Sydney, Australia).

Peritoneal Lavage

Peritoneal macrophages were obtained by lavage of the peritoneal cavity with ice-cold PBS. Rats were killed using CO₂ (adults) or anesthetized with ether (newborn) before they were given a lethal dose (0.1 ml/100 g) of Lethobarb administered into the heart via the thoracic cavity. For adult animals, 30 ml of ice-cold PBS was instilled gently into the peritoneal cavity with a 21-gauge needle and withdrawn after gentle massage of the peritoneum. This was repeated three to four times to collect a total volume of approximately 50 to 80 ml of lavage fluid per animal. In the case of newborns, 300 µl of PBS was instilled into the peritoneal cavity in a similar fashion. A total of 1 to 2 ml of lavage fluid was collected from each newborn animal. Lavage fluid was pooled from several animals and centrifuged at 4°C for 7 min at 600 × g. Cells were resuspended in MSFM with antibiotic and antimycotic.

Peritoneal lavage from adult rats contains a mixture of macrophages, mast cells, and lymphocytes. Peritoneal macrophages (as determined by monoclonal antibody [mAb] ED1 and ED2 staining) were purified from the lavage mixture using a system of discontinuous Percoll gradient (Pharmacia LKB, Uppsala, Sweden).

A 30%/50%/80% (5 ml of each fraction for separation of approximately 10⁷ cells) discontinuous Percoll gradient was prepared in a universal tube. Peritoneal lavage cells resuspended in a small volume (1 ml) of MSFM were then layered onto the gradient and centrifuged at 600 × g for 20 min at room temperature (~ 24°C). Cells at the 30%/50% interface consisted of greater than 95% ED1- and ED2-positive macrophages with small amounts (< 1%) of mast cells as determined by morphology from Leishman stain (BDH Laboratory Supplies, Merck Pty. Ltd., Victoria, Australia) of cytospins. Cells at the interface were collected and washed twice with PBS.

Phagocytosis Assay

AM were incubated in Teflon vessels (Savillex Corp., Minnetonka, MN) with MSFM. Fluorescein-labeled Escherichia coli (50 µg) from a commercial phagocytosis assay kit (F-6694; Molecular Probes Inc., Eugene, OR) was incubated with AM for 4 h in a 37°C humidified incubator with 5% (vol/vol) CO₂. AM were recovered and after washing with ice-cold PBS were divided into two equal portions. PBS (100 µl) was added to one portion (Part A) and 100 µl of trypan blue (1.25 mg/ml in citrate-balanced salt solution, pH 7.4) were added to the other (Part B). Part A represented total bound and ingested E. coli, whereas Part B represented ingested E. coli only. Phagocytosis was then assessed by flow cytometry (Coulter Epics XL; Beckman Coulter, Inc., Fullerton, CA). In a series of preliminary experiments, trypan blue was shown to effectively quench the fluorescence of fluorescein-labeled E. coli as previously described (13).

M. catarrhalis Aerosol

Rats were exposed to an aerosol of heat-killed M. catarrhalis for 1 h. The aerosol was generated with a Tri R nebulizer (Glas-Col Pty., Ltd., Terre Haute, IN). According to the manufacturer's specification, the nebulizer generated an aerosol with mean droplet size between 1 and 2 µm. Control animals were exposed to nebulized PBS or were left untouched.

Neutrophil Counts

Cells from the BAL of M. catarrhalis aerosolized animals were cytocentrifuged onto glass microscope slides and stained with Leishman stain (BDH Laboratory Supplies). Differential cell counts were performed and the number of neutrophils (determined by morphology) in the BAL was expressed as a percentage of total cells.

Immunohistochemistry

At various times after M. catarrhalis aerosolization, animals were killed by intraperitoneal injection of Lethobarb. After the aorta was severed, lungs were perfused with 20 ml of PBS containing 0.2% (wt/vol) bovine serum albumin and 1% (wt/vol) heparin by injection into the right ventricle. Tracheas were removed and immediately fixed in cold ethanol for 30 min, rehydrated in PBS, embedded in 100% OCT (Tissue Tek; Miles, Elkhart, IN), and snap-frozen in liquid nitrogen-cooled isopentane. Cryostat sections (8 to 10 µm thick) were cut and allowed to air dry at room temperature. Immunostaining was performed with primary antibody (RP3), immunoglobulin (Ig) M antirat neutrophil (14) for 1 h at room temperature. Slides were washed three times (10 min each) with PBS and incubated with biotinylated sheep antimouse Ig (Amersham Australia Pty. Ltd., Amersham Pharmacia Biotech, Sydney, Australia) containing 10% (vol/vol) normal rat serum (NRS) and streptavidin-conjugated horseradish peroxidase (Amersham Australia Pty. Ltd.). The reaction was visualized with 3,3'-diaminobenzidene and 0.015% (vol/vol) hydrogen peroxide in PBS. Sections were counterstained with hematoxylin, dehydrated, mounted, and analyzed. Tissue sections were counted with the help of a graticule under the microscope and at least 300 positive cells were counted.

Analysis of Gene Expression by Reverse Transcription/ Polymerase Chain Reaction

AM were suspended in MSFM with penicillin/streptomycin and cultured in 96-well plates (Falcon 3072; Becton Dickinson Labware, Franklin Lakes, NJ) for 48 h. For the kinetics experiments, E. coli LPS (Sigma-Aldrich Pty. Ltd.) at 100 ng/ml was added to the culture together with 1% (vol/vol) NRS. Cytokine messenger RNA (mRNA) levels were analyzed in freshly isolated AM samples and in AM exposed for varying periods to LPS in vitro after an initial 48-h preculture period in medium alone. For dose- response experiments, various concentrations of LPS were added to cultured AM for 3 h. Total RNA was prepared by lysis in RNAzol B (BIOTECX Laboratories Inc., Houston, TX) according to the manufacturer's instructions. The recovered RNA was resuspended in 10 µl of sterile water followed by complementary DNA (cDNA) synthesis with the total recovered RNA primed with 250 ng of oligo dT (Biotech International, Sydney, Australia). Reverse transcription (RT) was performed with the total oligo dT primed RNA in a final volume of 50 µl containing 7.5 mM MgCl₂, 0.4 mM each deoxynucleotide triphosphate (dNTP), 10 U RNase inhibitor (all from Biotech International), and 5 U reverse transcriptase (Promega, Madison, WI) with reaction buffer consisting of 50 mM Tris-HCl, pH 8.3 (42°C), 50 mM KCl, 10 mM MgCl₂, 10 mM dithiothreitol (DTT), and 0.5 mM spermidine. This mixture was incubated at 42°C for 60 min. The cDNA mixture was then inactivated at 95°C for 10 min in a heating block before storage at 4°C.

Polymerase chain reactions (PCR) were performed with 1 µl of cDNA in a total volume of 12.5 µl containing 1.5 mM MgCl₂; 0.2 mM each dNTP; 2.5 pmol each of 3' and 5' primers; sterile H₂O; 0.25 U platinum Taq DNA polymerase (GIBCO BRL, Life Technologies) in storage buffer consisting of 20 mM Tris-HCl (pH 8.0), 40 mM NaCl, 2 mM Na₃PO₄, 0.1 mM ethylenediaminetetraacetic acid, 1 mM DDT, stabilizers, and 50% (vol/vol) glycerol, and reaction buffer consisting of 200 mM Tris-HCl (pH 8.4) and 500 mM KCl. The mixture was capped with 50 µl of DNA grade mineral oil (Sigma-Aldrich Pty. Ltd.). PCR was performed for 22 to 38 cycles depending on primers in a solid block thermal cycler (Perkin Elmer DNA Thermal Cycler 480; Perkin Elmer). The number of PCR cycles used was first determined by amplifying cDNA through 15 to 40 cycles to obtain a standard curve. The cycle number chosen satisfies easy visualization and amplification along the linear part of the standard curve. The sequences of PCR primer pairs used are listed in Table 1.

PCR products were electrophoresed on an ethidium bromide- stained, 1.5% (wt/vol) agarose gel (Progen Industries Ltd., Brisbane, Qld., Australia). Gel photographs were scanned with a UMAX Vista S-6 scanner using Photoshop software (Adobe System, Inc.) on a Macintosh computer followed by densitometry using Scan Analysis software (Biosoft, Ferguson, MO). The data obtained was expressed as a ratio relative to the density of the -actin band for each sample.

Tumor Necrosis Factor- Bioassay

Tumor necrosis factor (TNF)- in the supernatant of AM culture was measured using the L-929 bioassay. Briefly, L-929 cells were grown in RPMI-1640 with 10% (vol/vol) heat-inactivated fetal calf serum (R10). When cells attained 80% confluency, they were removed and seeded into 96-well plates at 5.5 × 10⁵ cells per well in R10. After 20 h of culture, the medium was aspirated from the wells and replaced with R10 containing 100 µg/ml of actinomycin D. Standard, sample, or medium alone (100 µl) was added to triplicate wells. Triton X-100 (10 µl) was added to one duplicate as a kill blank, and this value was subtracted from all others. Plates were reincubated for 20 h after which the medium was removed and the plate stained with 0.1% (wt/vol) crystal violet in 1% (vol/ vol) acetic acid for 15 min. Plates were washed three times in distilled water and allowed to dry. Remaining cells were solubilized with 100 µl of 1% (wt/vol) sodium dodecyl sulfate in water, the plate gently mixed, and absorbance at 590 nm was determined. Recombinant rat TNF- (107 U/mg; Serotec, Oxford, UK) was used as a standard.

Statistical Analysis

Statistical analyses were performed using standard InStat software package (GraphPad Software, San Diego, CA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

AM from Neonatal and Young Animals Are More Phagocytic Than Those from Adults

To investigate the phagocytic activity of AM from animals of different ages, AM were cultured in Teflon vessels for 4 h at 37°C in the presence of fluorescein isothiocyanate (FITC)-labeled E. coli and phagocytosis assessed by flow cytometry. The percentage of cells that were phagocytic (Figures 1A and 1C) and the mean fluorescence intensity (MFI) of the phagocytic population (Figures 1B and 1D) were determined. A comparison of AM from animals of different ages showed that there was no significant difference in the number of AM that were able to ingest FITC-labeled E. coli (Figure 1A) (53 ± 11% in newborn animals compared with 50 ± 8% in adult animals). There was, however, a significant difference in the MFI of those cells that had phagocytosed the E. coli (Figure 1B), ranging from 33 ± 9, P < 0.05 in newborn animals to 10 ± 0.3 in adult animals. Because the MFI is a direct indicator of the number of fluorescent bacteria that has been phagocytosed by each AM, this suggests that AM from newborn animals have a significantly greater phagocytic capacity than do their adult counterparts. Comparable results were obtained with FITC-labeled gram-positive Streptococcus pyogenes (Figures 1C and 1D); however, in this case phagocytosis was greatest in the 7-d-old animals.

View larger version (68K): [in this window] [in a new window]   Figure 1.   Phagocytic ability of AM from rats of different ages. AM from animals of different ages were incubated with FITC- labeled E. coli (A and B) and FITC-labeled S. pyogenes (C and D) for 4 h at 37°C. Cells were recovered and phagocytic capacity was determined by flow cytometry. Fluorescence as a result of bacteria adherent to the outside of the AM was quenched with trypan blue. The number of cells that were significantly phagocytic (above background fluorescence) is expressed as a percentage of the whole population (A and C). The phagocytic capacity of this population is represented as MFI (B and D), which is indicative of the number of bacteria each cell has ingested. Asterisks indicate differences from adult AM. *P < 0.05; **P < 0.005; ***P < 0.001; and ****P < 0.0001.

Because CR3 (CD11b/CD18) is an important surface receptor on AM that mediates phagocytosis of bacteria, its expression was determined immunohistochemically using the rat CR3-specific mAb OX42. We found (data not shown) that a significantly higher number of BAL cells from newborn animals stained positively for OX42 compared with adults (82 ± 2.4% versus 29 ± 7.0%, P < 0.0001). Interestingly, AM from 1-wk-old animals had an even higher number of OX42-positive-staining cells (91 ± 3.3%) compared with newborns. However, by 21 d of age, the number of OX42-positive-staining cells (33 ± 3.6%) were not significantly different than that of adults.

Delayed and Decreased Neutrophil Recruitment into the Airways of Neonates after Inhalation of a Bacterial Aerosol

Although AM provide the first line of defense against invading bacteria, recruitment of large numbers of neutrophils, which are in turn able to assist in fighting this infection, is the characteristic mucosal response to bacterial infection. To compare the efficiency of neutrophil recruitment into the respiratory tract of newborn and adult rats in response to bacteria, animals were exposed to aerosols of M. catarrhalis. This organism was chosen because we have previously demonstrated that it induces a strong neutrophil response in the airways of rats (15).

The kinetics of neutrophil influx was monitored by determining the percentage of neutrophils within the BAL fluid from these animals and by immunohistochemical determination of neutrophils in tracheal sections. We found that neutrophil influx in newborn animals was significantly delayed compared with that in adults (Figure 2). In addition, although the number of AM from BAL did not change significantly (data not shown), the number of neutrophils determined as a percentage of total BAL cells was significantly less in newborn animals compared with adults (Figure 2). Thus, by 3 h postaerosol, greater than 90% of the lavagable cells from adult rats were neutrophils as compared with less than 5% in newborns. Neutrophil influx in newborns reached a peak at 6 h postaerosol with less than 60% neutrophils in the BAL (Figure 2) before gradually declining (data not shown). In contrast, the adult response peaked at 3 h and this was maintained for at least 12 h (data not shown) with greater than 75% of the BAL cell count being neutrophils after 24 h.

View larger version (23K): [in this window] [in a new window]   Figure 2.   Neutrophil responses in alveolar lavage after exposure to M. catarrhalis aerosol. Newborn (1 d old) and adult (9 to 11 wk old) rats were exposed to heat-killed M. catarrhalis aerosols for 1 h in a Tri-R nebulizer. Cells were obtained by BAL at 1, 3, and 6 h after exposure to the bacteria. Cells were cytocentrifuged onto glass slides and stained with Leishman stain. The percentage of neutrophils in the BAL was determined from counts based on distinct nuclear morphology characteristic of neutrophils. Percentages of neutrophils in BAL were calculated for newborn animals (shaded columns) and adult animals (solid columns). Asterisks indicate differences from adult AM. *P < 0.05; **P < 0.005; ***P < 0.001; and ****P < 0.0001.

Immunohistochemical staining with the neutrophil-specific mAb RP3 demonstrated that tracheas from newborn animals exposed to a bacterial aerosol were completely devoid of neutrophils (Figure 3A). In contrast, by 4 h after aerosol exposure there were a large number of RP3-positive neutrophils within the tracheal epithelium of adult animals (Figure 3B) with a density of 923 ± 145 RP3-positive cells/mm² of tissue section.

View larger version (111K): [in this window] [in a new window]   Figure 3.   Immunohistochemical analysis of neutrophil influx into tracheal epithelium of newborn and adult animals. Animals were exposed to an aerosol of heat-killed M. catarrhalis for 1 h. Tracheas from (A) newborn animals and (B) adult animals were then removed and immunohistochemically stained for neutrophils with RP3 mAb. Cells staining positively for RP3 are indicated by arrows. AL = airway lumen; Ep = epithelium; Sm = submucosa. Original magnification: ×40.

CC and CXC Chemokine Expression in AM of Both Newborn and Adult Animals Is Upregulated in a Dose-Dependent Manner in Response to In Vitro LPS Challenge

Previous reports have shown that the secretion of chemokines by AM is important in the process of neutrophil migration to sites of inflammation (8). We, therefore, looked at the expression of mRNA for various chemokines by AM in the newborn and compared this with that found in adult animals. Figure 4 shows that AM taken from both newborn and adult animals upregulated expression of CC (macrophage inflammatory protein [MIP]-1, MIP-1, and monocyte chemotactic protein [MCP]-1) and CXC (gro/cytokine-induced neutrophil chemoattractant [CINC] and MIP-2) chemokines in a dose-dependent manner in response to increasing concentrations of LPS. Interestingly, LPS appears to induce higher levels of both CC and CXC mRNA expression in newborns compared with adults. To better understand the responses of AM from different age groups to LPS stimulation, we have also investigated the kinetics of chemokine mRNA induction. We have found that AM from animals of different age groups were equally capable of inducing chemokine expression over time (data not shown).

View larger version (79K): [in this window] [in a new window]   Figure 4.   Expression of mRNA for (A) CC chemokines MIP-1, MIP-1, and MCP-1, and (B) CXC chemokines gro/CINC and MIP-2 in response to increasing concentrations of LPS. AM from 1-d-old and adult animals were obtained by BAL. Cells were cultured with various concentrations of LPS in the presence of 1% (vol/vol) NRS for 3 h. RT-PCR was performed using primers specific for the respective rat CC and CXC chemokines.

TNF- Expression in AM of Both Newborn and Adult Animals Is Upregulated in a Dose-Dependent Manner in Response to In Vitro LPS Challenge

Apart from being the archetypal macrophage proinflammatory cytokine produced in response to LPS, TNF- is also known to directly and indirectly induce the influx of neutrophils (16). Thus, we have investigated the expression of TNF- in the context of both a neutrophil chemoattractant and as a typical proinflammatory cytokine of AM to LPS stimulation. In response to LPS, AM from both newborn and adult animals were able to increase their expression of TNF- mRNA in a dose-dependent manner (Figure 5). Interestingly, we have found that expression of mRNA for TNF- was more sensitive to LPS in newborns compared with adults. This is evident in the photographs of the ethidium bromide-stained gels in Figure 5A and further illustrated by the bar graphs. However, when the level of biologically active TNF- in the supernatant of cultured neonatal and adult AM was measured using the L-929 bioassay, we found that the difference between levels of secreted TNF- in the two groups of animals was not statistically significant (Figure 5B).

View larger version (42K): [in this window] [in a new window]   Figure 5.   Expression of TNF- in response to increasing concentration of LPS. AM from 1-d-old and adult rats were obtained by BAL. Cells were cultured with various concentrations of LPS in the presence of 1% (vol/vol) NRS for 3 h. RT-PCR was performed using primers specific for rat TNF- (A). Bioactive TNF- from cell culture supernatants was measured using the L-929 cytotoxic bioassay (B).

Kinetics of TNF- mRNA Expression and the Secretion of Bioactive TNF- Are Not Different in AM from Newborn Animals Compared with Those from Adult Animals

Because there appeared to be discrepancies between mRNA expression and protein secretion of TNF- by AM from newborn and adult animals, the kinetics of TNF- expression was examined. We have found that the level of TNF- mRNA in newborn AM appears to persist longer, whereas in older animals this tends to wane to basal level by 12 h postchallenge (data not shown). When the secretion of bioactive TNF- was investigated (Figure 6), significant differences between AM from newborn animals compared with those from adults were observed. Thus, newborn AM (Figure 6A) showed significantly higher (P < 0.05) levels of TNF- secretion at 6 h post-LPS challenge, whereas AM from 7-d-old (Figure 6B) and 21-d-old (Figure 6C) animals were significantly lower (P < 0.05) than those from the adults (Figure 6D). Secretion of TNF- by newborn AM remained significantly higher than the secretion of TNF- by adult AM at 12 h post-LPS challenge but was comparable to adult AM secretion by 24 h.

View larger version (19K): [in this window] [in a new window]   Figure 6.   Kinetics of TNF- protein secretion in response to LPS. AM were cultured in 96-well plates for 48 h and then challenged with 100 ng/ml LPS in the presence of 1% (vol/vol) NRS for 3, 6, 12, and 24 h before the supernatant was used in L929 bioassay. Amounts of bioactive TNF- in the supernatant over time is shown in the graphs for animals at (A) 1 d old, (B) 7 d old, (C) 21 d old, and (D) 9 to 11 wk old (adults) and normalized to concentration of TNF- per 106 cells. Asterisks indicate differences from adult AM. *P < 0.05.

Strong Expression of LBP mRNA in Neonatal AM Is Not Observed in Adult AM

Because it appeared that AM from newborn animals were exhibiting an increased sensitivity to LPS, we reasoned that this may in part be explained by a difference in expression of CD14. As there are no mAbs available to rat CD14, we used RT-PCR to investigate the relative expression of CD14 on AM from newborn and adult animals. The results demonstrate that CD14 mRNA expression increased in a dose-dependent manner in response to LPS in both neonatal and adult AM (Figure 7). Furthermore, the level of CD14 mRNA again appeared to be higher in the neonates than in the adults. Because CD14 in AM has been reported to upregulate in response to LPS (17) and LBP is known to have a catalytic effect on the binding of LPS to CD14 (18), we decided to look at the possibility of local production of LBP by AM. Using PCR primers specific for rat LBP (19), we were able to demonstrate strong LBP mRNA expression by neonatal AM (Figure 8A). Furthermore, the expression of LBP was inducible by LPS and upregulated in a dose-dependent manner. In contrast, LBP expression by AM from adult animals varied from no expression to detectable albeit small amounts of LBP mRNA after incubation with higher doses of LPS.

View larger version (76K): [in this window] [in a new window]   Figure 7.   Expression of mRNA for CD14 in response to increasing concentrations of LPS. AM from newborn and adult animals were obtained by BAL. Cells were precultured for 48 h before the addition of various concentrations of LPS in the presence of 1% (vol/vol) NRS for 6 h. After extraction of mRNA, RT-PCR was performed using primers specific for rat CD14.

View larger version (54K): [in this window] [in a new window]   Figure 8.   Expression of mRNA for LBP in response to increasing concentrations of LPS. (A) AM from newborn, 7-d-old, and adult animals were obtained by BAL and cultured for 48 h before the addition of various concentrations of LPS in the presence of 1% (vol/vol) NRS for 6 h. RT-PCR was then performed using primers specific for rat LBP. (B) Similar experiments were performed using rat PM.

An examination of peritoneal macrophages (PM) from both newborn and adult animals (Figure 8B) clearly demonstrated that newborn PM were unable to produce mRNA for LBP in response to LPS. PM from adult animals produce small amounts of LBP mRNA when removed from the peritoneal cavity; however, LPS was not able to maintain or may have downmodulated this expression of LBP. To investigate the possibility that local factors may be influencing LBP production, AM from newborn, 7-d-old, and adult animals were cultured in vitro for 7 d before exposure to varying doses of LPS. These data are shown in Figure 9 where AM from newborn and 7-d-old animals are clearly able to maintain their ability to produce LBP after this period of culture out of the lung environment, albeit at a significantly lower level. In similar experiments (Figure 9C) with AM from adult animals, we found that these cells were unable to produce mRNA for LBP.

View larger version (46K): [in this window] [in a new window]   Figure 9.   Expression of LBP in response to increasing concentrations of LPS after 7 d preculture. AM from 1-d-old (A), 7-d-old (B), and adult animals (C) were obtained by BAL. Cells were precultured for 7 d before the addition of various concentrations of LPS in the presence of 1% (vol/vol) NRS for 3 h. RT-PCR was performed using primers specific for rat LBP. Negative controls consisted of RT-PCR performed in the absence of any cDNA.

The discrepancies between newborn and adult animals appeared to be a developmental phenomenon because AM from 1-wk-old and weanling animals appear to express LBP mRNA and they also retained this expression after 7 d in culture. Perhaps importantly, the expression of LBP in AM from 7-d-old and weanling animals showed significantly lower expression when compared with the neonates.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

It is becoming increasingly clear that many factors contribute to the age-related susceptibility of neonates to bacterial infections. These deficiencies in neonates and young animals involve both humoral immunity (such as specific antibodies and complement factors) and cellular immunity (such as chemotaxis and cell receptors) (20). Here, we have shown that a major component of the innate arm of the immune system in the neonatal lungs, i.e., the AM population, is not completely inefficient in function when compared with adult animals. In particular, several key functions of AM such as phagocytosis and secretion of mediators in response to bacterial products appear to be more efficient in young animals than in adults.

The validity of the phagocytosis assays described here has been rigorously tested. To measure phagocytosis, it was important to exclude the fluorescence due to FITC- labeled bacteria bound to the surface of the cells but not ingested. This was done by quenching the fluorescence of externally bound bacteria with trypan blue (13). Because internalized bacteria were not in contact with trypan blue, their fluorescence was not affected. This method was shown to be extremely effective because bacteria in contact with trypan blue were effectively quenched (data not shown). Furthermore, fluorescence detected as a result of bacteria binding to AM at 4°C (i.e, no phagocytosis) were effectively abolished in the presence of trypan blue (data not shown).

We have shown here that AM from young animals were significantly more phagocytic for both gram-negative (E. coli) and gram-positive (S. pyogenes) bacteria when compared with adult animals. There may be several reasons for these observations. First, inefficient phagocytosis is not a general phenomenon among cells of the mononuclear phagocyte lineage in neonates because monocytes from newborns have been shown to be equally effective in phagocytosis when compared with those from adults (21, 22). Thus, it may not be surprising that cells of the same lineage, namely AM, present at a site of high antigenic challenge (the lungs), would be highly efficient in bacterial clearance. Second, because phagocytosis of bacteria is usually receptor-mediated, our data indicating higher expression of the important receptor CR3 on neonatal AM are consistent with the high phagocytic capacity of these cells. However, it is also important to note that the data presented suggest that CR3 was not the sole receptor responsible for the phagocytosis of the bacteria because there appears to be no difference in the expression of CR3 by AM from 21-d-old animals compared with adult animals, despite a significantly higher level of phagocytosis by AM from 21-d-old animals. It is plausible that under physiologic conditions a range of receptors are able to bind the bacteria leading to phagocytosis. This leads to another possible factor contributing to the enhanced level of phagocytosis in AM from young animals. We have shown in this report that AM from newborn animals express high levels of LBP. Because CD14 expressed on AM is a receptor for LPS and LBP can enhance the interaction of LPS on bacteria to CD14, the high level of LBP expression in conjunction with other important factors such as soluble CD14 may thus enhance the binding of bacteria and its subsequent phagocytosis by neonatal AM. Recent observations (23) that LBP may play a role in the binding of peptidoglycan from gram-positive bacteria to cell surface CD14 suggest that it may also be opsonic for gram-positive organisms as seen in the enhanced phagocytosis of S. pyogenes.

In previous studies we showed that 8-d-old rats had a significantly reduced neutrophil response compared with adult animals (15). In our present study we have extended those results to the period immediately after birth. At this time, it appears from immunohistochemical analysis that there is no neutrophil influx into the epithelial lining of the trachea. However, the presence of neutrophils in the BAL fluid suggests that neutrophils enter the lung in response to an inflammatory signal. There is known to be a significant marginal pool of neutrophils within the vascular bed of the lung (24), and this may account for the neutrophil response within the lung tissue. The clear inability of neutrophils to enter the epithelium of the respiratory tract shortly after birth highlights the changes that are taking place in the development of the immuno-inflammatory response at that time. In view of the importance of neutrophils in controlling infection, this delayed and decreased recruitment of neutrophils may be severely detrimental to the host. Failure to generate a neutrophil response may be due to the lack of appropriate adhesion molecules such as CR3 or L-selectin on neonatal neutrophils as has been previously reported (25, 26). Alternatively, failure of the AM to produce appropriate neutrophil chemotactic signals may also contribute to a lowered neutrophil response.

Several studies have demonstrated that signals necessary for the recruitment of neutrophils to the airways arise from AM (8). Thus, we undertook to examine the chemokine response of neonates to bacterial stimuli. Because neither mAb based assays nor bioassays are readily available for most of the rat chemokines, we have used RT-PCR in most of our experiments. Conclusions based on PCR data are always made with the assumption that increased levels of mRNA are indicative of active protein production. Although this may not always be true, RT-PCR nevertheless provides a valuable insight into which chemokines or cytokines may or may not be produced in any situation.

The expression of chemokines responsible for the recruitment of inflammatory cells has not been well documented in young animals, although one report suggests a lower expression of MIP-1 in human neonatal mononuclear cells (27). We, therefore, compared the relative chemokine production by AM during development from birth until maturity. Our studies clearly demonstrated that AM taken from animals ranging in age from birth until adulthood were able to react to stimuli such as LPS by producing mRNA for all of the chemokines examined. This is in contrast with the observed decreased neutrophil recruitment into the airways of newborn animals after exposure to a whole bacterial stimuli and suggests that any defect in neutrophil recruitment within neonatal airways is not a consequence of deficient or absent macrophage signaling but may reflect an underlying immaturity in neutrophil migration, as has been previously documented (25, 28, 29).

Because TNF- is another neutrophil chemoattractant (16) and we were able to corroborate our mRNA data with bioactive proteins, we have also examined its induction in response to LPS. Dose response of TNF- mRNA induction by AM from newborn animals in response to LPS was found to be greater than that from adults (Figure 5A). However, this difference was not observed when bioactive TNF- was determined (Figure 5B). Thus, we investigated the kinetics of TNF- mRNA induction and protein secretion, and found newborn AM cultures have significantly greater levels of TNF- compared with adults after 6 to 12 h of LPS stimulation. Because this is a measure of accumulated TNF-, we speculated that secretion of TNF- by newborn AM was higher than its consumption compared with adult AM secretion at these time points, therefore our observed significantly elevated level. Secretion of TNF- relative to its consumption by newborn AM remained higher in comparison with adult AM secretion over the first 12 h after LPS challenge. In contrast, the ratio of TNF- secretion to consumption by adult AM at 6 h post-LPS challenge was relatively higher than that of 7-d-old and 21-d-old animals. This suggests that newborns are capable of secreting at least adult levels of TNF- in response to LPS.

LBP is an acute-phase protein that is known to enhance the effects of LPS on CD14-bearing cells by accelerating the transfer of LPS to CD14 (18). Here, we have demonstrated that AM from neonatal animals are able to express similar or higher levels of CD14 compared with those of adult animals. However, to the best of our knowledge, macrophage production of LBP has never been reported. Because it appeared that neonatal AM were in general able to react to lower levels of LPS than AM from adults, we examined these cells for their capacity to produce LBP. This resulted in the observation that neonatal AM expressed a high level of mRNA for LBP in response to LPS, whereas AM from adult animals showed very little if any mRNA for LBP under similar conditions. Furthermore, young animals up to the age of weaning (21 d old) also upregulated their LBP mRNA in response to LPS, although at intermediate levels. Confirmation of the importance of LBP production by rat AM will depend on the development of appropriate antibodies and assays; we are, however, in the process of attempting to confirm these observations in macrophages of human origin.

Previously, LBP has been shown to be of hepatic origin; however, extrahepatic expression from vascular smooth muscle cells has been reported (19). LBP has been shown to opsonize LPS-bearing particles for recognition and uptake by macrophages (30). Furthermore, because CD14 is known to mediate binding of intact bacteria (31, 32), LBP is able to accelerate or enhance this binding. Taken together, our observations of enhanced phagocytosis by AM from neonatal or young animals and AM production of LBP may suggest a mechanism whereby local production of LBP may possibly act as a bacterial opsonin. It is still possible that phagocytic efficiency improves with maturation; however, a reduced phagocytic capacity shortly after birth may be compensated by local LBP during early life. This may explain the high phagocytic capacity in young animals (1 to 3 wk old) despite diminishing LBP mRNA messages in response to LPS. We have also examined LBP expression in neonatal peritoneal macrophages but have found little or no expression in response to LPS. Thus, LBP expression by macrophages in young animals appears to be selectively produced in the pulmonary alveolar compartment.

Recent data using an LBP knockout mouse model have shown that LBP is not important in the in vivo clearance of bacteria but is important in holding infection in check (33). Although systemic clearance of LPS is not dependent on LBP, this does not eliminate the possibility that LBP may aid in the clearance of bacterial load locally. Indeed, Jack and coworkers (33) have shown that peritoneal wash, liver, and spleen of LBP-deficient animals contained approximately tenfold more bacteria than did control animals. Furthermore, binding of LPS to CD14 on neutrophils is enhanced by LBP (34). So, although the number of polymorphonuclear leukocyte migration into the lungs of the neonates was diminished compared with the adults, this may be functionally compensated by the enhanced level of LBP in the neonatal airspace, thereby leading to higher levels of neutrophil activation. Recent work has highlighted the potential of LBP to protect mice from septic shock caused by either LPS or gram-negative bacteria (35). In addition, recent observations that LBP may play a role in the binding of peptidoglycan from gram-positive bacteria to cell surface CD14 (23) suggest that it may also be opsonic for gram-positive organisms.

Local factors in developing neonatal airways are known to influence the functions and biology of AM. We investigated this by culturing AM from neonatal animals in vitro for 7 d before stimulation with LPS. Again, we observed a dose-dependent increase in LBP mRNA in neonatal AM, albeit at a significantly lower level compared with freshly isolated AM. Similarly, AM from adult animals were cultured for 7 d to remove the cells from potential microenvironmental inhibitors; here, LBP expression remained absent or undetectable from these cells. These results suggest that the microenvironment in the lungs may play a role in the expression of LBP by neonatal AM. Because 1- to 3-wk-old animals express LBP at a level intermediate between that of neonates and adults, expression of LBP may also be developmentally regulated.

In summary, although neonatal animals have dramatically reduced capacity to mount a neutrophil response within inflamed airways, this deficiency is not a result of AM failure to produce chemotactic signals. Furthermore, we report that AM are able to express LBP mRNA with greater production in newborn or young animals. In light of the potential role of LBP in contributing to inflammatory responses to bacteria and bacterial products, rapid AM production of LBP may be a significant mechanism in the newborn animal, whereby innate immunity in neonates is compensated and strengthened at a time when humoral immunity is still relatively immature.