| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Biological Sciences, Florida International University, Miami, FL 33199, USA 2Department of Clinical Microbiology, Rigshospitalet, Copenhagen, Denmark 3Bartholin Institute, Kommunehospitalet, Copenhagen, Denmark
Correspondence Kalai Mathee matheek{at}fiu.edu
Received November 15, 2002
Accepted April 4, 2003
Pseudomonas aeruginosa is an opportunistic respiratory pathogen that accounts for most of the morbidity and mortality in cystic fibrosis (CF) patients. In CF-affected lungs, the bacteria undergo conversion from a non-mucoid to a non-tractable mucoid phenotype, due to overproduction of alginate. The effect of alginate production on pathogenicity was investigated by using an acute lung infection mouse model that compared a non-mucoid P. aeruginosa strain, PAO1, to its constitutive alginate-overproducing derivative, Alg+ PAOmucA22, and an alginate-defective strain, Alg- PAOalgD. Bacterial suspensions were instilled into the left bronchus and examined 24 and 48 h post-infection. The highest bacterial loads and the most severe lung pathology were observed with strain Alg- PAOalgD at 24 h post-infection, which may have been due to an increase in expression of bacterial elastase by the mutant. Significantly lower lung and spleen bacterial loads were found in the two non-mucoid (PAO1 and Alg- PAOalgD) groups, compared to the mucoid Alg+ PAOmucA22 group, between 24 and 48 h post-infection. The positive correlation between lung bacteriology and lung macroscopic pathology in the Alg+ PAOmucA22 group suggests that alginate production not only impedes pulmonary clearing, but also results in severe lung damage. Positive correlations between IL12 levels and lung macroscopic pathology, and between IL12 and IFN-
levels in the Alg+ PAOmucA22 group, suggested a possible contribution of these pro-inflammatory cytokines to tissue damage. No significant differences were found between the three groups in lung cytokine responses at 24 or 48 h post-infection. However, on comparison within each group at 24 and 48 h post-infection, a significant increase in the pro-inflammatory cytokine IFN- was observed. Higher ratios of IFN-/IL4 and IFN-/IL10, but lower IL10 levels, were also found in all three groups. These results indicate a Th1-predominated immune response in these animals. Such cytokine responses could have aided the clearance of non-mucoid P. aeruginosa, but were not sufficient to alleviate infection by the mucoid variants. Alginate production may promote survival and persistence of this pathogenic micro-organism in the lung.
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Alginate production in vitro can be triggered by nutrient starvation, antibiotic treatment, slow growth rate, ethanol dehydration, high osmotic pressure or high ionic strength (Evans & Linker, 1973; Govan & Fyfe, 1978; DeVault et al., 1990; Terry et al., 1991). We have demonstrated that local accumulation of activated polymorphonuclear cells (PMNs) exerts selective pressures on the bacteria from phagocytosis and oxygen free radicals, resulting in conversion from the non-mucoid to the mucoid phenotype (Mathee et al., 1999). Approximately 84 % of mucoid CF isolates have mutations in the mucA gene, which encodes a negative regulator of an alternative sigma factor, AlgT/U (Boucher et al., 1997). MucA interacts directly with AlgT/U, inhibiting its activity (Hughes & Mathee, 1998). On inactivation of MucA, AlgT/U triggers biosynthesis of alginate via a complex regulatory circuit (Mathee et al., 2002). This culminates in activation of the 18 kb alginate biosynthesis (algD) operon (Mathee et al., 2002).
There is a strong humoral immune response against chronic P. aeruginosa infection in most CF patients; this is dominated by a T helper type 2 (Th2) immune response (Høiby et al., 2001). This is characterized by production of cytokines IL4, IL5, IL6 and IL10 and antibodies IgG1 and IgE. The Th1 response is associated with an increase in levels of IFN-, IL2, IL12 and IgG2a (Banchereau, 1995; Germann et al., 1995). The pronounced antibody response in CF patients leads to the formation of immune complexes in the lung that activate complement and result in an infiltration of PMNs, which is a hallmark of chronic infection in CF-affected lungs (Baltimore et al., 1989; Høiby et al., 2001). Bronchoalveolar lavage fluid of young CF patients has normal levels of IL10 (Noah et al., 1997), whereas patients with chronic infections have significantly lower levels of this cytokine (Bonfield et al., 1995b). Furthermore, normal epithelial cells produce IL10 constitutively, but epithelial cells from chronically infected CF patients are defective in IL10 production (Bonfield et al., 1995a).
Many animal models have been used successfully to recreate both acute and chronic P. aeruginosa infections (Stotland et al., 2000). However, only a few studies have addressed the role of alginate in pulmonary clearance and these have produced conflicting data. The earliest reported study, which used guinea pigs as an animal model of chronic infection, a CF P. aeruginosa isolate and genetically undefined non-mucoid strains, concluded that the Alg+ phenotype did not selectively impair pulmonary clearing (Blackwood & Pennington, 1981). In a more recent study that used a neutropenic mouse model, no difference was observed between the wild-type strain PAO1 and the mutant strain algT/U : : TcR (Yu et al., 1996). Interestingly, the null mutant showed a reduced mean time of death in the normal C57BL/6 mouse and the endotoxin-resistant C3H/HeJ mouse. The authors concluded that inactivation of algT/U resulted in increased virulence (Yu et al., 1996). Another study looked at clearance of CF isolates from the murine lung in an aerosol infection model with C57BL/6J, BALB/c and DBA/2NHsd mice, using a wild-type strain and a number of isogenic derivatives (Martin et al., 1993; Boucher et al., 1997). In these experiments, the Alg+ strain survived better in the lungs than the parent non-mucoid strain and the algD mutant (Boucher et al., 1997). Similar results were observed following repeated aerosol exposure of C57BL/6J mice to other PAO1 derivatives (Yu et al., 1998). The increased host survival could not be attributed to an increase in the pro-inflammatory cytokine TNF-
(Boucher et al., 1997; Yu et al., 1998). There was no difference in lung histopathology or in MIP-2 (the equivalent of IL8) between the two groups of animals at either the 4 or 18 h time-points (Yu et al., 1998).
Although we know much about the organism, regulation of the alginate genes and changes in the patient's immune response from onset of infection to death, no advances have been made in using this information to control P. aeruginosa successfully in CF patients. Animal studies clearly support the view that alginate production impedes bacterial clearance and favours their survival in the lungs; however, survival could not be correlated with host immune response. In the current study, we present a detailed analysis of the role of alginate in the pathogenesis of lung infection by using a mouse model, in which we compare the prototype strain of P. aeruginosa, PAO1, with a constitutive alginate-overproducing derivative and an alginate-defective strain.
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Bacterial strains.
The prototypic non-mucoid P. aeruginosa strain PAO1 (Holloway & Morgan, 1986) and its isogenic derivatives were used. These derivatives were the mucoid strain Alg+ PAOmucA22 (PDO300), which carries the mucA22 allele that results in constitutive production of alginate (Mathee et al., 1999), and PAOalgD, a non-mucoid strain that carries a deletion in algD (Garrett et al., 1999). Strains were cultured in LB broth (l-1: 10 g tryptone, 5 g yeast extract and 5 g NaCl) overnight at 37 °C and bacterial cells were harvested by centrifugation and resuspended in saline to approximately 1.5 x 108 c.f.u. ml-1.
Animal model.
Before challenge, all mice were anaesthetized subcutaneously with a 1 : 1 (v/v) mixture of etomidate (Janssen) and midazolam (Roche) at a dose of 10 ml (kg body wt)-1 and then tracheotomized (Moser et al., 1997). Each mouse was challenged intra-tracheally with 0.04 ml bacterial suspension as described previously (Moser et al., 1997). Animals were killed by administering 2 ml 20 % pentobarbital (DAK) kg-1.
Macroscopic pathology.
Lung index of macroscopic pathology (LIMP) was calculated by using the following modified formula: LIMP = lung area with pathological changes divided by area of the whole lung (Song et al., 1998). Changes in lung pathology are consolidation, haemorrhage and oedema. Gross pathological changes were grouped according to severity of inflammation as previously described (Song et al., 1998): I, normal; II, swollen lungs, hyperaemia and small atelectasis (< 10 mm2); III, pleural adhesions and atelectasis (< 40 mm2); IV, abscesses, large atelectasis and haemorrhages.
Histopathological scores.
Cellular alterations in the lungs were classified as acute or chronic inflammation by a scoring system based on the proportion of neutrophils or PMNs and mononuclear leukocytes (MNs) in the inflammatory foci (Johansen et al., 1993). Acute inflammation was defined as cellular infiltration in which PMNs were predominant (
90 %) with 
10 % MNs, whereas chronic inflammation was defined as a prevalence of MNs (MNs 90 %; PMNs 10 %), which included lymphocytes, plasma cells and the presence of granulomas (Johansen et al., 1993). Both macroscopic and histopathological evaluations were performed as a double-blind study.
Lung and spleen bacteriology.
Quantitative bacteriological examination was done as described previously (Johansen et al., 1993). In brief, lungs or spleens were removed aseptically from mice and homogenized in 5 ml PBS. Serially diluted samples were plated onto LB agar plates to determine bacterial count (c.f.u.). The remaining lung homogenate was centrifuged and the supernatant was retained for cytokine determination.
Cytokine determination.
Concentrations of IFN-, IL4, IL10 and IL12 in lung homogenate supernatants were determined by using ELISA kits (Nordic BioSite). Standard curves were constructed for IL4, range 7.8–500 pg ml-1 (lower detection limit, 0.96 pg ml-1); IFN-, 40–5000 pg ml-1 (10 pg ml-1); IL10, 1.2–80 pg ml-1 (1 pg ml-1); and IL12, 5–1000 pg ml-1 (< 5 pg ml-1). Optical density of each sample was plotted on the standard curve of the respective cytokine ELISA plate to obtain cytokine concentrations.
Protease activity and alginate determination.
Bacteria were incubated at 37 °C in LB broth with rapid shaking for 18 h. Cells were collected by centrifugation and LasB elastase activity was determined spectrophotometrically by using elastin–Congo red (Sigma) as substrate (Ohman et al., 1980). Protease activity was expressed as A595 units min-1 (g protein)-1. Concentrations of alginate, expressed as mg (l culture supernatant)-1, were measured by the carbazole–borate method (Knutson & Jeanes, 1968).
Statistical analysis.
Data were analysed by using the statistical software package SPSS (version 10.0 for Windows; SPSS, Chicago, IL, USA; http://www.spss.com). Unpaired differences in continuous data were analysed by the Mann–Whitney U-test and categorical data were compared by using a 
-squared test. Simple regression tests were used for correlation analysis between parameters.
|
RESULTS AND DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Alginate production results in severe lung pathology
Twenty-four hours post-infection, pulmonary histopathological changes in the PAO1 group were milder and less PMN infiltration was seen than in the other strains (Fig. 1). Significant pulmonary oedema with milder haemorrhage was seen in the Alg- PAOalgD group, but marked PMN infiltration was observed in the Alg+ PAOmucA22 group (Fig. 1). All three groups showed clear infiltration of PMNs in the lung foci at 48 h post-infection, although it was more profound in the Alg+ PAOmucA22 group (Fig. 1).
|
Judging by the lung scoring, significantly milder changes in lung pathology were found in the PAO1 group (P < 0.05) at 24 h post-infection than in the Alg- PAOalgD group (Table 1). LIMP analysis of all three groups at 24 h post-infection showed that the scores of the Alg- PAOalgD and Alg+ PAOmucA22 groups were significantly higher than the PAO1 group (P 0.001 and P 0.05, respectively; Table 1). Furthermore, the Alg- PAOalgD group also had significantly higher LIMP scores than the Alg+ PAOmucA22 group (P < 0.05; Table 1). At 48 h post-infection, the lung pathology of the Alg+ PAOmucA22 group was significantly more severe than that of the PAO1 group (P 0.02, Table 1). In the Alg- PAOalgD group, LIMP score at 48 h post-infection was obviously lower than that at 24 h (P < 0.01).
|
Both micro- and macroscopic pathological analyses revealed that wild-type PAO1 infection resulted in the mildest pathology (Fig. 1). In terms of infective severity at 24 h, the order of the strains was PAO1 < Alg+ PAOmucA22 < Alg- PAOalgD, whereas at 48 h post-infection it was PAO1/Alg- PAOalgD < Alg+ PAOmucA22. Of the three strains used, only the Alg+ strain produced copious amounts of alginate [255 mg (l culture supernatant)-1]. No alginate was detected in broth culture supernatants of the non-mucoid strains. Alginate production has been inversely correlated with protease activity (Mohr et al., 1990); elastase activity in Alg+ PAOmucA22 (17.8 U) was reduced approximately four- and sixfold compared to its parent strain (PAO1; 69.6 U) and Alg- PAOalgD (107.9 U), respectively. The Alg- PAOalgD strain produced 1.5-fold more elastase than PAO1. The severity of Alg- PAOalgD at 24 h post-infection may be explained by the ability of this strain to produce higher amounts of elastase.
Similarly, in another study that used a neutropenic mouse model of fatal P. aeruginosa sepsis, the authors concluded that loss of ability to make alginate (inactivation of algT/U) resulted in increased virulence (Yu et al., 1996). They did not observe any differences in LD50 between wild-type PAO1 and the algT/U : : TcR mutant. However, the null mutant showed a reduced mean time of death in two of the animal strains tested: the normal C57BL/6 mouse and the endotoxin-resistant C3H/HeJ mouse (Yu et al., 1996). Although it appeared initially that the loss of alginate production resulted in increased virulence, it was temporary. At 48 h post-infection, we did not observe any difference between the non-mucoid strains.
Despite significant PMN infiltration in the lung foci at 48 h post-infection in Alg+ PAOmucA22-infected lungs, this did not appear to influence bacterial clearance. This supports in vitro studies that demonstrated that alginate protected bacteria from phagocytosis, opsonization, antibodies, complement and PMN infiltration (Laharrague et al., 1984; Oliver & Weir, 1985; Eftekhar & Speert, 1988; Krieg et al., 1988; Stiver et al., 1988; Jensen et al., 1990; Pedersen et al., 1990; Mathee et al., 1999). Conversion of bacteria to the non-tractable, alginate-producing phenotype is partly due to PMN infiltration (Mathee et al., 1999). Alginate has been demonstrated to interfere with PMN functions such as adherence, oxygen metabolism, degranulation and bactericidal activity (Pedersen et al., 1990; Mai et al., 1993). At the molecular level, alginate also alters the signal transduction pathway by modulating PMN protein kinase C and G-protein activity (König et al., 1992).
Excessive infiltration of PMNs contributes to lung tissue damage, due to the release of reactive oxygen species, and the severity of histopathological changes correlates with the ability of the bacteria to persist in the lung and evade host clearance. We observed a positive correlation (r = 0.66) between quantitative lung bacteriology and score of lung pathology in the Alg+ PAOmucA22 group at 24 h post-infection (Table 2). Thus, alginate production not only impedes pulmonary clearance, but also results in severe tissue damage. This is reminiscent of a late-stage chronic infection that exhibits marked local destruction of lung tissues of CF patients with Alg+ P. aeruginosa strains; for some such patients, lung transplantation is the only hope for survival (Pedersen et al., 1990).
|
Alginate production and bacterial persistence in the lung
Persistence of invading bacteria and ability of the host immune system to clear bacteria can be inferred from lung bacteriology. At 24 h post-infection, significantly higher bacterial counts were found in the Alg- PAOalgD group than in the PAO1 (P 0.05) and the Alg+ PAOmucA22 (P 0.05) groups (Fig. 2). Bacterial load at 48 h post-infection in the Alg+ PAOmucA22 group was 57-fold higher than that in the Alg- PAOalgD group (P < 0.02) and 158 times higher than that in the PAO1 group (P < 0.04) (Fig. 2). Comparing the 24 and 48 h time-points, the wild-type PAO1 and Alg- PAOalgD groups showed significantly reduced bacterial loads (P < 0.002 and P = 0.0002, respectively), but there was no significant change in the Alg+ PAOmucA22 group (P > 0.4) (Fig. 2).
|
The order of these three strains in terms of their lung bacterial load at 24 h post-infection was Alg+ PAOmucA22 < PAO1 < Alg- PAOalgD, and at 48 h post-infection it was PAO1 < Alg- PAOalgD < Alg+ PAOmucA22 (Table 1). At 24 –h post-infection, the Alg- PAOalgD group had the highest bacterial load; however, at 48 h post-infection, the Alg+ PAOmucA22 strain prevailed, suggesting significant microbial protection that was probably due to the overproduction of alginate. These data contradict an earlier reported animal study that used guinea pigs, which concluded that the mucoid phenotype had no selective effect to impede pulmonary clearing (Blackwood & Pennington, 1981). The negative control non-mucoid strain used in the guinea pig study was isolated by repeated passages of the mucoid CF strain that is likely to have a defective mucA allele, encoding the anti-sigma factor (Martin et al., 1993; Boucher et al., 1997). Based on our current understanding of alginate genes, it is likely that these non-mucoid strains have second site suppressor mutations, possibly in algT/U, which encodes the sigma factor (DeVries & Ohman, 1994; Schurr et al., 1994). The use of two strains with undefined mutations may have affected the outcome of their experiment.
The most detailed study to date looked at the effect of alginate on bacterial clearance by using an aerosol infection model in which C57BL/6J, BALB/c and DBA/2Nhsd mice (Boucher et al., 1997) were infected with P. aeruginosa PAO381 (Fyfe & Govan, 1980) and its derivatives, PAO578I (mucA22), PAO578II (mucA22, sup-2; salt-dependent Alg+ phenotype) and PAO578IIalgD : : Gm (Martin et al., 1993; Boucher et al., 1997). In this 4 h post-infection study, the alginate- producing strain had greater survival in the lungs than the parent strain and its algD mutant (Boucher et al., 1997). Similar results were observed in a repeated aerosol exposure study on C57BL/6J mice that used PAO381 and PAO578I (Yu et al., 1998). Although the experimental strains and animal infection models used here are different, the parallels in our conclusions are clear: alginate production promotes bacterial survival and persistence in lung infection.
Alginate production and persistence in the spleen
The presence of bacteria in the spleen reflects the ability of invading bacteria to spread from the initial point of inoculation. At 24 h post-infection, the spleen bacterial count for the Alg+ PAOmucA22 group was significantly lower than that for the PAO1 group (P < 0.002) (Fig. 2). However, at 48 h post-infection, higher counts were found for the Alg+ PAOmucA22 group than for the Alg- PAOalgD group (P < 0.006), with bacterial counts approximately 5.6- and 40-fold higher for the groups with the non-mucoid strains PAO1 and Alg- PAOalgD, respectively. Comparing the two time points at 24 and 48 h post-infection, significant differences were found for the PAO1 group (P = 0.01, Fig. 2) and the Alg- PAOalgD group (P < 0.0008), but no significant difference was confirmed for the mucoid Alg+ PAOmucA22 group (P > 0.6), indicating that the Alg+ strain is more resistant to immune clearance associated with its mucoid phenotype (Fig. 2).
The order of strains in terms of spleen bacterial load at 24 h post-infection was Alg+ PAOmucA22 < PAO1/Alg- PAOalgD; whereas at 48 h, it was PAO1/Alg- PAOalgD < Alg+ PAOmucA22 (Fig. 2). Each strain produced significant septicaemia, but at 24 h post-infection, a significantly lower spleen bacterial count for the Alg+ PAOmucA22 group compared to the parent stain was evident. This difference may be partly associated with the reduced motility of the strain, compared to the non-mucoid strain PAO1 (data not shown). In addition, alginate production has been negatively correlated with synthesis of flagella, which may reduce the number of bacteria entering the bloodstream (Garrett et al., 1999). At 24 h post-infection, a positive correlation (r = 0.75) between lung and spleen bacteriology was evident for the Alg- PAOalgD group (Table 2). Both the PAO1 (r = 1.00) and Alg- PAOalgD (r = 0.99) groups showed positive correlation at 48 h post-infection, whereas no significant correlation was seen in the Alg+ PAOmucA22 group. The lower spleen bacterial count in the non-mucoid groups at 48 h post-infection suggests that the host was able to clear these non-mucoid organisms more efficiently than Alg+ PAOmucA22.
Th1 cytokines are up-regulated at 48 h post-infection
The immune response is modulated by T helper cells via changes in cytokine levels. In the present study, at 24 and 48 h post-infection, no significant differences were found in IFN-, IL4, IL10 or IL12 between the three infection groups (Table 3). However, a comparison of the 24 and 48 h post-infection data indicated a significant increase in IFN- and decrease in IL10 in all three groups (Table 3). Higher IL12 production, however, was detected in the PAO1 group compared to the Alg- PAOalgD group at 48 h post-infection (P < 0.05; Table 3). Significant reduction of IL4 levels was seen only in the PAO1 group (P < 0.02; Table 3) at 48 h post-infection.
|
Positive correlations between IFN- and IL12 were found in the PAO1 (r = 0.97, P < 0.0001) and Alg+ PAOmucA22 (r = 0.64, P < 0.05) groups at 24 h post-infection, and in the PAO1 (r = 0.86, P < 0.002) and Alg- PAOalgD (r = 0.76, P = 0.01) groups at 48 h post-infection (Table 2). This was not unexpected, as both of these cytokines are Th1 indicators. IL12 stimulates production of IFN- by NK and T cells (Belardelli, 1995). The increase of IFN- and decrease of IL10 in lung tissues that were seen at 48 h post-infection, compared to the results at 24 h post-infection, are predicted as IL10, if produced, suppresses the synthesis of pro- inflammatory cytokines by inhibiting IL12 and IFN- production (Belardelli, 1995). In addition, levels of the pleiotropic cytokine IL4, a product of Th2 lymphocytes that inhibits the differentiation of Th1 cells (Hart et al., 1989; Belardelli, 1995), were also reduced in all three bacterial groups at 48 h post-infection, even though this difference was not statistically significant. These results suggest that the immune response to acute infection with P. aeruginosa at 48 h post-infection was predominantly a Th1-type response, compared to the results at 24 h post-infection. This was further demonstrated by analysing the ratios of IFN-/IL4 and IFN-/IL10 in all three groups. These ratios can be used as an indicator of Th1 versus Th2 responses; a higher ratio is an indicator of a Th1-dominated response and a lower ratio suggests a Th2-dominated response. The significant increase in these ratios between the 24 and 48 h time-points is indicative of a Th1-based immune response (Table 3). The immune response in CF patients with chronic P. aeruginosa infections is predominantly a Th2 response and is correlated with a poor prognosis (Moser et al., 2000). It has been proposed that shifting the immune response to Th1 may help to combat P. aeruginosa infection (Høiby et al., 2001). However, our results suggest that although the animals mount a Th1 response, it is not sufficient to alleviate mucoid P. aeruginosa infection.
Diagnostic value of IL4 levels
Early intermittent colonization of CF-affected lungs by non-mucoid P. aeruginosa, often treatable with minimal clinical symptoms, precedes obdurate chronic infection with mucoid variants. In our study, infection by the prototypic, non-mucoid strain PAO1 exhibited the mildest pathology and was cleared more quickly than the other strains. A negative correlation was found between the level of IL4 and lung (r = -0.72, P < 0.02) and spleen (r = -0.71, P = 0.02) bacteriology in the PAO1 group at 48 h post-infection. IL4 levels were positively correlated with lung pathological scoring in the PAO1 group at 48 h post-infection (r = 0.72, P < 0.02). In addition, the Alg- PAOalgD group showed a negative correlation between IL4 levels and lung (r = -0.63, P = 0.05) and spleen (r = -0.67, P < 0.04) bacteriology at 24 h post-infection (Table 2). It appears that the immune response mounted was successful in lowering bacterial load, but the inflammatory defence mechanisms did lead to tissue damage. It is possible that the outcome in the non-mucoid groups was determined by the significant reduction in IL4 levels and relatively higher IFN- levels. However, Jain-Vora et al. (1998) reported enhanced pulmonary clearance of P. aeruginosa in acute infection in transgenic mice by IL4. Thus, the role of IL4 levels in infection remains unclear with non-mucoid P. aeruginosa.
Severity of infection is correlated with the imbalance between pro- and anti-inflammatory cytokine levels
The Alg- PAOalgD group had the highest bacterial load, the most severe pathology and showed a negative correlation between lung scores and IL10 levels (r = -0.68, P < 0.032) and a positive correlation between IL12 levels and LIMP (r = 0.8, P < 0.006) at 24 h post-infection. At 48 h post-infection, the Alg+ PAOmucA22 strain prevailed, suggesting significant microbial protection that is probably due to overproduction of alginate. The severity of lung pathology and PMN infiltration in the Alg+ PAOmucA22 group may be due to an increase in pro-inflammatory cytokines IL12 and IFN-. This is supported by positive correlations between IL12 levels and lung pathology (r = 0.83, P < 0.003) and between IL12 and IFN- levels (r = 0.64, P < 0.05), and negative correlations between IL10 levels and bacterial counts in the lung (r = -0.78, P < 0.009) and spleen (r = -0.7, P < 0.025) for the Alg+ PAOmucA22 group at 48 h post-infection (Table 2).
The imbalance of pro- versus anti-inflammatory cytokines and the correlation with the severity of infection are similar to those reported in CF patients (Kronborg et al., 1993). It has been shown that bronchoalveolar lavage fluid of CF patients with chronic infection by mucoid strains has significantly lower levels of IL10 (Bonfield et al., 1995b). These authors further demonstrated that normal epithelial cells produce IL10 constitutively, but epithelial cells from chronically infected CF patients are defective for production of IL10 (Bonfield et al., 1995a). The IL10 knockout mice studies of Yu et al. (1998) provided supportive evidence for the role of IL10 in CF infection: following repeated aerosol exposure, they observed higher mortality and severe lung pathology in the IL10-deficient mice compared to controls (Yu et al., 1998). In another study, using a chronic endobronchial infection by P. aeruginosa, it was demonstrated that IL10 knockout mice had severe weight loss and an increased area of lung inflammation, but no alterations in bacterial burden, compared to wild-type mice (Chmiel et al., 1999, 2002). Interestingly, in the current study, no correlation was seen between the PAO1 group and IL10 levels. This is in agreement with Noah et al. (1997), who showed that bronchoalveolar lavage fluid of young CF patients, whose early onset of infection was caused by non-mucoid P. aeruginosa, had normal levels of IL10.
Constitutive production of IL10 may help to prevent local tissue destruction by pro-inflammatory cytokines (Moore et al., 1993). It is possible that severity of lung pathology could be reduced by increasing IL10 levels, which would suppress synthesis of the pro-inflammatory cytokines IL12 and IFN- (Belardelli, 1995). This idea is also supported by studies in which IL10 treatment increased the survival rate of mice infected with P. aeruginosa. The treatment enhanced bacterial clearance (Sawa et al., 1997; Matsumoto et al., 1998), improved survival, lessened severe weight loss, lowered the number of bronchoalveolar lavage neutrophils and decreased the area of lung inflammation (Chmiel et al., 1999, 2002).
Currently, most of the knowledge that concerns mucoid strains of P. aeruginosa, especially alginate production and its gene regulation, has been achieved from studies performed in vitro. Investigations in vivo are still quite limited and have focused on pulmonary clearance of the pathogen. As far as we know, the present study is the most comprehensive analysis to date of the role of alginate in the pathogenicity of P. aeruginosa, in terms of its bacterial survival in the lung and spleen and the host cytokine response during an early acute infection. Overproduction of alginate could be an important contributory factor for Alg+ PAOmucA22 persistence and virulence. This study supports previous inferences that alginate is a virulence factor that contributes to bacterial adherence and persistence (Marcus & Baker, 1985; Ramphal & Pier, 1985; Doig et al., 1987), formation of microcolony and biofilm growth (Høiby, 1975; Lam et al., 1980; Pedersen, 1992), inhibition of PMN chemotaxis (Stiver et al., 1988; Pedersen et al., 1990), suppression of lymphocyte and PMN function (Pedersen et al., 1990; Mai et al., 1993), formation of a physical barrier to the immune system and antibiotics (Nichols et al., 1989; Evans et al., 1991; Hoyle & Costerton, 1991; Mathee et al., 1994) and resistance to opsonic killing by PMNs and macrophages (Meshulam et al., 1984; Cabral et al., 1987; Jensen et al., 1990; Pedersen et al., 1992).
Patients with chronic infection should benefit from consistent use of anti-inflammatory treatment. However, to better understand the role of alginate in chronic infection and the host immune response, we need to look at the expression of pro- and anti-inflammatory cytokines at both the molecular and cellular levels. This will allow novel therapeutic measures against chronic P. aeruginosa infection in CF patients to be developed.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. J. Moraes, R. Martin, J. D. Plumb, E. Vachon, C. M. Cameron, A. Danesh, D. J. Kelvin, W. Ruf, and G. P. Downey Role of PAR2 in murine pulmonary pseudomonal infection Am J Physiol Lung Cell Mol Physiol, February 1, 2008; 294(2): L368 - L377. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Jelsbak, H. K. Johansen, A.-L. Frost, R. Thogersen, L. E. Thomsen, O. Ciofu, L. Yang, J. A. J. Haagensen, N. Hoiby, and S. Molin Molecular Epidemiology and Dynamics of Pseudomonas aeruginosa Populations in Lungs of Cystic Fibrosis Patients Infect. Immun., May 1, 2007; 75(5): 2214 - 2224. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Lee, J. A. J. Haagensen, O. Ciofu, J. B. Andersen, N. Hoiby, and S. Molin Heterogeneity of Biofilms Formed by Nonmucoid Pseudomonas aeruginosa Isolates from Patients with Cystic Fibrosis J. Clin. Microbiol., October 1, 2005; 43(10): 5247 - 5255. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. T. Sadikot, T. S. Blackwell, J. W. Christman, and A. S. Prince Pathogen-Host Interactions in Pseudomonas aeruginosa Pneumonia Am. J. Respir. Crit. Care Med., June 1, 2005; 171(11): 1209 - 1223. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Hoffmann, T. B. Rasmussen, P. Jensen, C. Stub, M. Hentzer, S. Molin, O. Ciofu, M. Givskov, H. K. Johansen, and N. Hoiby Novel Mouse Model of Chronic Pseudomonas aeruginosa Lung Infection Mimicking Cystic Fibrosis Infect. Immun., April 1, 2005; 73(4): 2504 - 2514. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Malloy, R. A. W. Veldhuizen, B. A. Thibodeaux, R. J. O'Callaghan, and J. R. Wright Pseudomonas aeruginosa protease IV degrades surfactant proteins and inhibits surfactant host defense and biophysical functions Am J Physiol Lung Cell Mol Physiol, February 1, 2005; 288(2): L409 - L418. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M. Cobb, J. C. Mychaleckyj, D. J. Wozniak, and Y. S. Lopez-Boado Pseudomonas aeruginosa Flagellin and Alginate Elicit Very Distinct Gene Expression Patterns in Airway Epithelial Cells: Implications for Cystic Fibrosis Disease J. Immunol., November 1, 2004; 173(9): 5659 - 5670. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. P. Priebe, C. R. Dean, T. Zaidi, G. J. Meluleni, F. T. Coleman, Y. S. Coutinho, M. J. Noto, T. A. Urban, G. B. Pier, and J. B. Goldberg The galU Gene of Pseudomonas aeruginosa Is Required for Corneal Infection and Efficient Systemic Spread following Pneumonia but Not for Infection Confined to the Lung Infect. Immun., July 1, 2004; 72(7): 4224 - 4232. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. P. Stapper, G. Narasimhan, D. E. Ohman, J. Barakat, M. Hentzer, S. Molin, A. Kharazmi, N. Hoiby, and K. Mathee Alginate production affects Pseudomonas aeruginosa biofilm development and architecture, but is not essential for biofilm formation J. Med. Microbiol., July 1, 2004; 53(7): 679 - 690. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Schwiebert Cystic fibrosis, gene therapy, and lung inflammation: for better or worse? Am J Physiol Lung Cell Mol Physiol, April 1, 2004; 286(4): L715 - L716. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
) and Alg+ PAOmucA22 (
) groups. Thirteen animals per bacterial strain were used.