Immunomodulatory role for membrane vesicles released by THP-1 macrophages and respiratory pathogens during macrophage infection.

BACKGROUND: During infection, inflammation is partially driven by the release of mediators which facilitate intercellular communication. Amongst these mediators are small membrane vesicles (MVs) that can be released by both host cells and Gram-negative and -positive bacteria. Bacterial membrane vesicles are known to exert immuno-modulatory and -stimulatory actions. Moreover, it has been proposed that host cell-derived vesicles, released during infection, also have immunostimulatory properties. In this study, we assessed the release and activity of host cell-derived and bacterial MVs during the first hours following infection of THP-1 macrophages with the common respiratory pathogens non-typeable Haemophilus influenzae, Moraxella catarrhalis, Streptococcus pneumoniae, and Pseudomonas aeruginosa. RESULTS: Using a combination of flow cytometry, tunable resistive pulse sensing (TRPS)-based analysis and electron microscopy, we demonstrated that the release of MVs occurs by both host cells and bacteria during infection. MVs released during infection and bacterial culture were found to induce a strong pro-inflammatory response by naive THP-1 macrophages. Yet, these MVs were also found to induce tolerance of host cells to secondary immunogenic stimuli and to enhance bacterial adherence and the number of intracellular bacteria. CONCLUSIONS: Bacterial MVs may play a dual role during infection, as they can both trigger and dampen immune responses thereby contributing to immune defence and bacterial survival.

Bead-based flow-cytometry for semi-quantitative analysis of complex membrane vesicle populations released by bacteria and host cells.

During infection, the release of nano-sized membrane vesicle is a process which is common both for bacteria and host cells. Host cell-derived membrane vesicles can be involved in innate and adaptive immunity whereas bacterial membrane vesicles can contribute to bacterial pathogenicity. To study the contribution of both membrane vesicle populations during infection is highly complicated as most vesicles fall within a similar size range of 30-300nm. Specialized techniques for purification are required and often no single technique complies on its own. Moreover, techniques for vesicle quantification are either complicated to use or do not distinguish between host cell-derived and bacterial membrane vesicle subpopulations. Here we demonstrate a bead-based platform that allows a semi-quantitatively analysis by flow-cytometry of bacterial and host-cell derived membrane vesicles. We show this method can be used to study heterogeneous and complex vesicle populations composed of bacterial and host-cell membrane vesicles. The easy accessible design of the protocol makes it also highly suitable for screening procedures to assess how intrinsic and environmental factors affect vesicle release.

Budesonide, fluticasone propionate, and azithromycin do not modulate the membrane vesicle release by THP-1 macrophages and respiratory pathogens during macrophage infection.

Patients with more severe chronic obstructive pulmonary disease frequently experience exacerbations and it is estimated that up to 50% of these exacerbations are associated with bacterial infections. The mainstay treatment for these infection-related exacerbations constitutes the administration of glucocorticoids, alone or in combination with antibiotics. A recent line of evidence demonstrates that many hormones including the steroid beclomethasone can also directly affect bacterial growth, virulence, and antibiotic resistance. The effect of these regimens on the release of potentially virulent and toxic membrane vesicles (MVs) is at present unclear. In this study, we determined the effect of several pharmacological agents on MVs release by and bacterial growth of common respiratory pathogens. We found that neither the release of MVs nor the bacterial growth was affected by the glucocorticoids budesonide and fluticasone. The macrolide antibiotic azithromycin only inhibited the growth of Moraxella catarrhalis but no effects were observed on bacterial MV release at a concentration that is achieved locally in the epithelial lining on administration. The macrophage pro-inflammatory response to MVs was significantly reduced after treatment with budesonide and fluticasone but not by azithromycin treatment. Our findings suggest that these glucocorticoids may have a positive effect on infection-related inflammation although the bacterial growth and MV release remained unaffected.

Effects of N-acetyl-L-cysteine on the membrane vesicle release and growth of respiratory pathogens.

Bacterial infections contribute to the disease progression of chronic obstructive pulmonary disease by stimulating mucus production in the airways. This increased mucus production and other symptoms are often alleviated when patients are treated with mucolytics such as N-acetyl-L-cysteine (NAC). Moreover, NAC has been suggested to inhibit bacterial growth. Bacteria can release membrane vesicles (MVs) in response to stress, and recent studies report a role for these proinflammatory MVs in the pathogenesis of airways disease. Yet, until now it is not clear whether NAC also affects the release of these MVs. This study set out to determine whether NAC, at concentrations reached during high-dose nebulization, affects bacterial growth and MV release of the respiratory pathogens non-typeable Haemophilus influenzae (NTHi), Moraxella catarrhalis (Mrc), Streptococcus pneumoniae (Spn) and Pseudomonas aeruginosa (Psa). We observed that NAC exerted a strong bacteriostatic effect, but also induced the release of proinflammatory MVs by NTHi, Mrc and Psa, but not by Spn. Interestingly, NAC also markedly blunted the release of TNF-α by naive macrophages in response to MVs. This suggests that the application of NAC by nebulization at a high dosage may be beneficial for patients with airway conditions associated with bacterial infections.

The host immune response contributes to Haemophilus influenzae virulence.

BACKGROUND: There is compelling evidence that infections with non-typeable Haemophilus influenzae (NTHi) are associated with exacerbations in COPD patients. However, NTHi has also been isolated frequently during clinically stable disease. In this study we tested the hypothesis that genetically distinct NTHi isolates obtained from COPD patients differ in virulence which could account for dissimilarities in the final outcome of an infection (stable vs. exacerbation). RESULTS: NTHi isolates (n = 32) were obtained from stable COPD patients, or during exacerbations. Genetically divergent NTHi isolates were selected and induction of inflammation was assessed as an indicator of virulence using different in vitro models. Despite marked genomic differences among NTHi isolates, in vitro studies could not distinguish between NTHi isolates based on their inflammatory capacities. Alternatively, when using a whole blood assay results demonstrated marked inter-, but not intra-individual differences in cytokine release between healthy volunteers irrespective of the origin of the NTHi isolate used. CONCLUSION: Results suggest that the individual immune reactivity might be an important predictor for the clinical outcome (exacerbation vs. no exacerbation) following NTHi infection.

Interferon-ß induces a long-lasting antiviral state in human respiratory epithelial cells.

OBJECTIVES: Interferon-ß (IFNß) induces strong antiviral effects and is therefore an attractive agent to prevent or reduce the incidence of virus-mediated exacerbations in asthmatic or chronic obstructive pulmonary disease (COPD) patients. We therefore investigated the effects of prophylactic IFNß on respiratory epithelial cells infected with rhinovirus (RV). METHODS: A549 cells and primary bronchial epithelial cells (PBECs) were exposed for 18 h to IFNß. Then, IFNß was either removed or maintained in the supernatant for the rest of the experiment and cells were infected with RV-1B at t = 0 or 72 h after the initial exposure to IFNß. RESULTS: Viral RNA levels were decreased in both cell types. Furthermore, both viral RNA and infectious virus levels in the supernatant of infected A549 cells were still significantly reduced at 72 h after removal of IFNß. This pronounced antiviral pre-treatment effect was associated with increased expression of the antiviral genes IFN-stimulated protein of MR15000 (ISG15) and Myxovirus resistance 1 (Mx1) and the effect was maintained even when IFNß levels in the supernatant of A549 cells were undetectable. CONCLUSIONS: These data show that IFNß has not only a strong, but also a long-lasting protective effect against RV infection of respiratory epithelium.

CpG and poly(I:C) stimulation of dendritic cells and fibroblasts limits herpes simplex virus type 1 infection in an IFNß-dependent and -independent way.

Viral activation of toll-like receptors (TLRs) on dendritic cells (DCs) leads to production of various cytokines, including antiviral type I interferons (IFNs). Synthetic ligands specific for TLRs are also able to induce the production of type I IFNs (IFNα/ß) by DCs, suggesting that these ligands have potential as antiviral drugs. In this in vitro study we extensively investigated the antiviral activity of various TLR ligands. Mouse bone marrow (BM) cells were differentiated into plasmacytoid and conventional DCs (pDCs and cDCs), stimulated with various TLR ligands and tested the antiviral abilities of collected supernatants in an in vitro herpes simplex virus type 1 (HSV-1) infection model. We observed a significant IFNß-, (but not IFNα-) dependent reduction in HSV-1 infection when a mixed pDC/cDC population was stimulated with the TLR9 ligand CpG. In the absence of pDCs, TLR stimulation resulted in less pronounced antiviral effects. The most pronounced antiviral effect was observed when both DC subsets were stimulated with poly(I:C). A similar noticeable antiviral effect was observed when fibroblasts (L929 cells) were stimulated directly with poly(I:C). These poly(I:C)-mediated antiviral effects were only partially IFNß-mediated and probably TLR independent. These data demonstrate that TLR ligands are not only able to produce type I IFN but can indeed act as antiviral drugs. In particular poly(I:C), which exerts its antiviral effects even in the absence of DCs, may become a promising drug e.g. to prevent respiratory infections by topical intranasal application.

Lack of evidence for the role of human adenovirus-36 in obesity in a European cohort.

Adenovirus infection has been shown to increase adiposity in chickens, mice, and nonhuman primates. Adenovirus type 36 (Ad-36) DNA was detected in adipose tissues in these animal trials. In the United States, Ad-36 significantly correlates with obesity as illustrated by an Ad-36 seroprevalence of 30% in obese individuals and 11% in nonobese individuals. We investigated the possibility of a similar correlation of Ad-36 in Dutch and Belgian persons. In total, 509 serum samples were analyzed for Ad-36 antibodies using a serum neutralization assay. In addition, PCR was used to detect adenoviral DNA in visceral adipose tissue of 31 severely obese surgical patients. Our results indicated an overall Ad-36 seroprevalence of 5.5% increasing with age. BMI of Ad-36 seropositive humans was not significantly different from seronegative humans. No adenoviral DNA could be found using PCR on visceral adipose tissue. In conclusion, this first Ad-36 study in the Netherlands and in Belgium indicates that Ad-36 does not play a role as a direct cause of BMI increase and obesity in humans in Western Europe.

Adenoviral activin A expression prevents vein graft intimal hyperplasia in a rat model.

Autologus vein grafts are used for coronary artery and infra-inguinal bypass procedures. Although initially successful, long-term patency rates are limited by lumen occlusion due to neointima formation by smooth muscle cell hyperplasia. Gene therapy to prevent this smooth muscle cell proliferation has been studied extensively with limited success. Activin A, a member of the transforming growth factor-beta super family, promotes the contractile phenotype of smooth muscle cells. Maintaining the contractile phenotype could be a novel strategy to prevent intimal hyperplasia. In an epigastric vein-to-common femoral artery interposition grafts rat model, activin A over-expression resulted in a significant decrease in intimal cross-sectional area and percentage stenosis as compared to the control group. BrdU staining identified lower proliferation rates of the smooth muscle cells in the group treated with activin A. We report for the first time evidence that activin A can diminish vein graft failure in a rat model supporting a novel strategy to prevent intimal hyperplasia.

Chlamydia pneumoniae aggravates vein graft intimal hyperplasia in a rat model.

BACKGROUND: Along with angioplasty, autologus vein grafts are commonly used for artery bypass grafting in patients with advanced arterial stenosis and drug-resistant angina pectoris. Although initially a successful procedure, long-term functionality is limited due to proliferation and migration of smooth muscle cells. Like in atherosclerosis, common chronic infections caused by viruses and bacteria may contribute to this process of vein graft failure. Here we investigated the possible role of Chlamydia pneumoniae (Cpn) in the pathogenesis of venous graft failure in an experimental animal model. In 2 groups (n = 10 rats/group), an epigastric vein-to-common femoral artery interposition graft was placed. Immediately thereafter, rats were infected with Cpn (5*108 IFU) or injected with control solutions. Rats were sacrificed three weeks after surgery and the grafts were harvested for morphometrical and immunohistochemical analysis. RESULTS: Cpn administration immediately after vein grafting resulted in a significant increase in medial cross-sectional area, wall thickness and total wall area. There were no significant differences in T-cell or macrophage influx. Likewise, although positive immunostaining for both HSP60 and CRP could be detected, no differences were found between groups. Based on the observation that the number of cells/microm2 was also not altered, we conclude that Cpn infection stimulates smooth muscle cell proliferation by hereunto unknown molecular mechanisms, resulting in a significant increase in intimal hyperplasia. CONCLUSION: In conclusion, in a well defined animal model we present here for the first time evidence for a role of Chlamydia pneumoniae in the process of venous graft failure.

Mouse cytomegalovirus antigenic immune stimulation is sufficient to aggravate atherosclerosis in hypercholesterolemic mice.

We have previously demonstrated that mouse cytomegalovirus (MCMV) infection aggravates atherosclerosis by stimulating the ongoing inflammatory process in the vascular wall. Here we investigated whether MCMV antigenic immune stimulation by UV-MCMV injection is sufficient to aggravate atherosclerosis. In addition we analyzed whether low viral doses are sufficient to stimulate atherosclerosis. Therefore, apoE(-/-) mice received a low dose injection with infectious virus (MCMV) or replication-deficient virus (UV-inactivated MCMV, UV-MCMV). Atherosclerosis progression, influx of inflammatory cells in atherosclerotic lesions and internal organs and the number of MCMV DNA copies in various organs were determined at 2 weeks after injection. After injection with infectious virus, MCMV DNA was present in internal organs, while no MCMV DNA could be detected after UV-MCMV injection. Interestingly, both MCMV and UV-MCMV significantly increased mean atherosclerotic lesion area and T cell number in the atherosclerotic lesions, while only MCMV infection increased T cell numbers in the internal organs. These data indicate that in apoE(-/-) mice both low dose infectious MCMV as well as MCMV antigenic injections are sufficient for atherosclerosis aggravation.

The rat cytomegalovirus homologue of parvoviral rep genes, r127, encodes a nuclear protein with single- and double-stranded DNA-binding activity that is dispensable for virus replication.

An intriguing feature of the rat cytomegalovirus (RCMV) genome is open reading frame (ORF) r127, which shows similarity to the rep genes of parvoviruses as well as the U94 genes of human herpesvirus type 6A (HHV-6A) and 6B (HHV-6B). Counterparts of these genes have not been found in other herpesviruses. Here, it is shown that the r127 gene is transcribed during the early and late phases of virus replication in vitro as an unspliced 1.1 kb transcript containing the complete r127 ORF. Transcripts of r127 were also detected in various organs of RCMV-infected rats at 1 week post-infection (p.i.), but only in the salivary gland at 4 months p.i. Using rabbit polyclonal antibodies raised against the r127-encoded protein (pr127), pr127 was found to be expressed as early as 12 h p.i. within the nuclei of RCMV-infected cells in vitro. Expression of pr127 was also observed within the nuclei of cells in various organs of RCMV-infected rats at 3 weeks p.i. Moreover, pr127 was demonstrated to bind single- as well as double-stranded DNA. Finally, an RCMV r127 deletion mutant (RCMVDeltar127) was generated, in which the r127 ORF was disrupted. This deletion mutant, however, was shown to replicate with a similar efficiency as wild-type RCMV (wt RCMV), both in vitro and in vivo. Taken together, it is concluded that the RCMV r127 gene encodes a nuclear protein with single- and double-stranded DNA-binding activity that is dispensable for virus replication, not only in vitro, but also during the acute phase of infection in vivo.

The rat cytomegalovirus R78 G protein-coupled receptor gene is required for production of infectious virus in the spleen.

The rat cytomegalovirus (RCMV) R33 and R78 genes are conserved within members of the subfamily Betaherpesvirinae and encode proteins (pR33 and pR78, respectively) that show sequence similarity with G protein-coupled receptors. Previously, the biological relevance of these genes was demonstrated by the finding that R33- and R78-deleted RCMV strains are severely attenuated in vivo. In addition, R78-deleted strains were found to replicate less efficiently in cell culture. To monitor of the expression of R33- and R78-encoded proteins, recombinant RCMV strains, designated RCMV33G and RCMV78G, were generated. These recombinants expressed enhanced green fluorescent protein (EGFP)-tagged versions of pR33 and pR78 instead of native pR33 and pR78, respectively. Here it is reported that, although RCMV33G replicates as efficiently as wt virus in rat embryo fibroblast cultures, strain RCMV78G produces virus titres that are 3- to 4-fold lower than wt RCMV in the culture medium. This result indicates that the pR78-EGFP protein, as expressed by RCMV78G, does not completely functionally replace its native counterpart (pR78) in vitro. Interestingly, in infected rats, infectious RCMV33G was produced in significantly lower amounts than infectious wt RCMV, as well as RCMV78G, in the salivary glands. Conversely, although RCMV33G replicated to similar levels as wt virus in the spleen, both RCMV78G and an R78 knock-out strain (RCMV Delta R78a) replicated poorly in this organ. Together, these data indicate that R78 is crucial for the production of infectious RCMV in the spleen of infected rats.

Chlamydia pneumoniae serology: comparing a commercial enzyme immunoassay and microimmunofluorescence test in patients with cardiovascular disease.

Chlamydia pneumoniae has been associated with cardiovascular disease and the detection of C. pneumoniae antibodies has subsequently challenged many cardiovascular investigators. The micro-immunofluoresence (MIF) test is considered the gold standard for detection of C. pneumoniae antibodies, but requires a high-level of expertise for adequate interpretation. We compared an enzyme immunoassay (EIA) with a microimmunofluorescence test for the detection of C. pneumoniae IgG- and IgA antibodies in sera of 141 patients with atherosclerosis. The MIF test was read by two independent observers. The interobserver agreement of the MIF test for detection of seropositivity at various cut-off levels was good for IgG and for IgA. The intra-test agreement of the EIA was excellent for IgG and IgA. The agreement between EIA and MIF in detection of IgG- and IgA antibodies was adequate at low but not at high titer levels. At low titer levels, the sensitivity, specificity, positive and negative predictive value of EIA compared to the MIF test was sufficient. The sensitivity of the EIA increased, improving the agreement with the MIF at high titer levels by retesting sera with elevated titers at higher pre-dilutions. In conclusion, the EIA shows sufficient agreement with the MIF test in the detection of C. pneumoniae seropositivity. Therefore, the EIA is a practical alternative to the MIF in the detection of C. pneumoniae antibodies in patients with cardiovascular disease, bearing in mind that the sensitivity of the EIA depends on the antibody titer.

Persistent rat cytomegalovirus (RCMV) infection of the salivary glands contributes to the anti-RCMV humoral immune response.

The salivary glands are the major sites of persistent replication of rat cytomegalovirus (RCMV). At several months post infection (pi), infectious RCMV is usually still produced in the salivary glands but not in any other organ or tissue of the rat. To investigate whether the persistence of RCMV in the salivary glands is crucial to the pathogenesis of viral infection, we monitored the progression of RCMV-induced disease in rats from which the salivary glands had been surgically removed (desalivated) as well as in sham-operated rats, both after a lethal and sublethal challenge with RCMV. Desalivation did not have a significant effect on either RCMV-induced morbidity or mortality. As expected, at 1 year pi, relatively high levels of infectious virus were detected in the salivary glands of sham-operated rats, whereas neither infectious virus nor RCMV DNA could be detected in liver, spleen and lungs of these animals. Infectious virus and viral DNA were also undetectable in organs from desalivated animals at 1 year pi. Surprisingly, a difference was found between desalivated and sham-operated rats in the titers of anti-RCMV IgG antibodies, which were significantly higher in sham-operated rats than in desalivated animals at 183, 295 and 365 days pi. This finding indicates that the persistence of RCMV in the salivary glands may contribute significantly to the anti-RCMV humoral immunity of infected rats.