Jolanda P. Vermeulen

THE EFFECT OF PARTICLE SIZE ON THE CYTOTOXICITY, INFLAMMATION, DEVELOPMENTAL TOXICITY AND GENOTOXICITY OF SILVER NANOPARTICLES

Silver nanoparticles are of interest to be used as antimicrobial agents in wound dressings and coatings in medical devices, but potential adverse effects have been reported in the literature. The most pronounced effect of silver nanoparticles and the role of particle size in determining these effects, also in comparison to silver ions, are largely unknown. Effects of silver nanoparticles of different sizes (20, 80, 113 nm) were compared in in vitro assays for cytotoxicity, inflammation, genotoxicity and developmental toxicity. Silver nanoparticles induced effects in all endpoints studied, but effects on cellular metabolic activity and membrane damage were most pronounced. In all toxicity endpoints studied, silver nanoparticles of 20 nm were more toxic than the larger nanoparticles. In L929 fibroblasts, but not in RAW 264.7 macrophages, 20 nm silver nanoparticles were more cytotoxic than silver ions. Collectively, these results indicate that effects of silver nanoparticles on different toxic endpoints may be the consequence of their ability to inflict cell damage. In addition, the potency of silver in the form of nanoparticles to induce cell damage compared to silver ions is cell type and size-dependent.

Association of Bordetella pertussis with host immune cells in the mouse lung.

Mouse models are frequently used to study immunity and pathogenesis to Bordetella pertussis infection. To improve the understanding of the mouse infection model, the influx of host cells and B. pertussis localisation in the lungs were evaluated. Furthermore, the roles of filamentous hemagglutinin (FHA) and fimbriae (Fim) in these processes were determined. B. pertussis infection stimulated the recruitment of polymorphonuclear granulocytes (PMN), alveolar macrophages, and lymphocytes. As determined by double immunofluorescence staining, 2 hr after infection most B. pertussis were free in the alveolar space, some were attached to alveolar epithelia, and some were associated with and phagocytosed by PMN. After 3 days, most bacteria were associated with and phagocytosed by macrophages, some by PMN. B. pertussis was shown not to be ingested by epithelial cells or associated with interstitial macrophages. B. pertussis mutants lacking expression of FHA or Fim were associated with and phagocytosed by the same cell types as parental bacteria. The Fim mutant, however, induced a more severe inflammation, and was cleared faster from the lungs compared to the parental strain and the FHA mutant. These results suggest that Fim does not affect bacterial localisation in the mouse lung, but does influence host immune mechanisms. Possibly, Fim may exert an anti-inflammatory function and thereby inhibit killing by macrophages.

Lipopolysaccharide Analogs Improve Efficacy of Acellular Pertussis Vaccine and Reduce Type I Hypersensitivity in Mice

ABSTRACT Pertussis is an infectious disease of the respiratory tract that is caused by the gram-negative bacterium Bordetella pertussis. Although acellular pertussis (aP) vaccines are safe, they are not fully effective and thus require improvement. In contrast to whole-cell pertussis (wP) vaccines, aP vaccines do not contain lipopolysaccharide (LPS). Monophosphoryl lipid A (MPL) and Neisseria meningitidis LpxL2 LPS have been shown to display immune-stimulating activity while exerting little endotoxin activity. Therefore, we evaluated whether these LPS analogs could increase the efficacy of the aP vaccine. Mice were vaccinated with diphtheria-tetanus-aP vaccine with aluminum, MPL, or LpxL2 LPS adjuvant before intranasal challenge with B. pertussis. Compared to vaccination with the aluminum adjuvant, vaccination with either LPS analog resulted in lower colonization and a higher pertussis toxin-specific serum immunoglobulin G level, indicating increased efficacy. Vaccination with either LPS analog resulted in reduced lung eosinophilia, reduced eosinophil numbers in the bronchoalveolar lavage fluid, and the ex vivo production of interleukin-4 (IL-4) by bronchial lymph node cells and IL-5 by spleen cells, suggesting reduced type I hypersensitivity. Vaccination with either LPS analog increased serum IL-6 levels, although these levels remained well below the level induced by wP, suggesting that supplementation with LPS analogs may induce some reactogenicity but reactogenicity considerably less than that induced by the wP vaccine. In conclusion, these results indicate that supplementation with LPS analogs forms a promising strategy that can be used to improve aP vaccines.

In vitro effects of aldehydes present in tobacco smoke on gene expression in human lung alveolar epithelial cells

Tobacco smoke consists of thousands of harmful components. A major class of chemicals found in tobacco smoke is formed by aldehydes, in particular formaldehyde, acetaldehyde and acrolein. The present study investigates the gene expression changes in human lung alveolar epithelial cells upon exposure to formaldehyde, acrolein and acetaldehyde at sub-cytotoxic levels. We exposed A549 cells in vitro to aldehydes and non-aldehyde chemicals (nicotine, hydroquinone and 2,5-dimethylfuran) present in tobacco smoke and used microarrays to obtain a global view of the transcriptomic responses. We compared responses of the individual aldehydes with that of the non-aldehydes. We also studied the response of the aldehydes when present in a mixture at relative concentrations as present in cigarette smoke. Formaldehyde gave the strongest response; a total of 66 genes were more than 1.5-fold differentially expressed mostly involved in apoptosis and DNA damage related processes, followed by acetaldehyde (57 genes), hydroquinone (55 genes) and nicotine (8 genes). For acrolein and the mixture only one gene was upregulated involved in oxidative stress. No gene expression effect was found for exposure to 2,5-dimethylfuran. Overall, aldehyde responses are primarily indicative for genotoxicity and oxidative stress. These two toxicity mechanisms are linked to respiratory diseases such as cancer and COPD, respectively. The present findings could be important in providing further understanding of the role of aldehydes emitted from cigarette smoke in the onset of pulmonary diseases.

Induction of skin sensitization is augmented in Nrf2-deficient mice.

Several in vitro DNA microarray studies have shown the importance of nuclear factor (erythroid-derived 2)-like 2 (Nrf2) in skin sensitization. Nevertheless, the exact in vivo role of the Nrf2-Keap1 pathway during the induction of skin sensitization remains unknown. To study the function of Nrf2, a local lymph node assay was performed in wild-type and Nrf2-deficient mice using 2,4-dinitrochlorobenzene. The Nrf2-deficient mice show a more pronounced response, indicating that Nrf2 is involved in dampening the induction of skin sensitization.

A Dose-Response Modeling Approach Shows That Effects From Mixture Exposure to the Skin Sensitizers Isoeugenol and Cinnamal Are in Line With Dose Addition and Not With Synergism

Currently, hazard characterization of skin sensitizers is based on data obtained from studies examining single chemicals. Many consumer products, however, contain mixtures of sensitizers that might interact in such a way that the response induced by a substance is higher than predicted in the hazard assessment. To assess interaction of skin sensitizers in a mixture, a dose-response modeling approach is applied. With this approach, it is possible to assess whether or not responses from mixtures of sensitizers can be predicted from the dose-response information obtained from individual chemicals using dose addition. We selected the skin sensitizers isoeugenol and cinnamal, frequently occurring together in consumer products, to be examined in an adjusted local lymph node assay (LLNA). Cell number and cytokine production (IL-10 and IFN-γ) of the auricular lymph nodes were measured as hallmarks of the skin sensitization response. We found that dose addition for these 2 skin sensitizers closely predicted the effects from mixtures of both chemicals across the broad dose range tested. Hence, isoeugenol and cinnamal show no synergistic effects in the LLNA. Therefore, hazard assessment and risk assessment of these substances can be performed without taking into account mixture exposure.

Response of MUTZ-3 dendritic cells to the different components of the Haemophilus influenzae type B conjugate vaccine: towards an in vitro assay for vaccine immunogenicity.

Potency testing is mandatory for vaccine registration and batch release. Due to various limitations to in vivo potency testing, there is need for relevant in vitro alternatives. These alternative tests should preferably comprise cells from the target (human) species. The whole suite of immune responses to vaccination that occur in vivo in humans cannot be tested in vitro using a single cell type. Even so, dendritic cells (DC) form an important candidate cell type since they are pivotal in inducing and orchestrating immune responses. Cell lines are preferred over ex vivo cells for reasons of safety, accessibility, and reproducibility. In this first feasibility study we used the human cell line MUTZ-3, because it most closely resembles ex vivo human DC, and compared its response to monocyte-derived DC (moDC). Haemophilus influenzae type B (HiB) vaccine was chosen because its components exert different effects in vivo: while the HiB antigen, polyribosyl ribitol phosphate (PRP) fails to induce sufficient protection in children below 2 years of age, conjugation of this polysaccharide antigen to outer membrane protein (OMP) of Neisseria meningitides, results in sufficient protection. Effects of PRP, OMP, conjugated PRP-OMP, and adjuvanted vaccine (PedVax HiB), on cytokine production and surface marker expression were established. PRP induced no effects on cytokine production and the effect on surface marker expression was limited to a minor decrease in CD209 (DC-SIGN). In both MUTZ-3 and moDC, OMP induced the strongest response both in cytokine production and surface marker expression. Compared to OMP alone conjugated PRP-OMP generally induced a weaker response in cytokine production and surface marker expression. The effects of PedVax HiB were comparable to conjugated PRP-OMP. While moDC showed a larger dynamic range than MUTZ-3 DC, these cells also showed considerable variability between donors, with MUTZ-3 DC showing a consistent response between the replicate assays. In our view, this makes MUTZ-3 DC the cells of choice. In conclusion, our results demonstrate that the MUTZ-3 DC assay allows discrimination between compounds with different immunogenicity. The potential of this cell line as (part of) an in vitro immunogenicity assay should be further explored.

Simple in vitro models can predict pulmonary toxicity of silver nanoparticles

Abstract To study the effects of nanomaterials after inhalation, a large number of in vitro lung models have been reported in literature. Although the in vitro models contribute to the reduction of animal studies, insufficient data exists to determine the predictive value of these in vitro models for the in vivo situation. The aim of this study was to determine the correlation between in vitro and in vivo data by comparing the dose metrics of silver nanoparticles in an in vitro lung model of increasing complexity to our previously published in vivo inhalation study. In vivo, the previously published study showed that the alveolar dose expressed as particle surface area is the most suitable dose metric to describe the toxicity of silver nanoparticles after inhalation. The results of the present study show that particle surface area is a suitable dose metric to describe the effects of silver nanoparticles when using a simple monolayer of lung epithelial cells. The dose metric shifted from particle surface area to particle mass when adding an increasing number of macrophages. In addition, a co-culture of endothelial cells, epithelial cells and macrophages on a Transwell® insert correlated less well to the in vivo results compared to the epithelial monolayer. We conclude that for studying the acute pulmonary toxicity of nanoparticles simple in vitro models using an epithelial monolayer better predict the in vivo response compared to complex co-culture models.

The crystal structure of titanium dioxide nanoparticles influences immune activity in vitro and in vivo

BackgroundThe use of engineered nanoparticles (NP) is widespread and still increasing. There is a great need to assess their safety. Newly engineered NP enter the market in a large variety; therefore safety evaluation should preferably be in a high-throughput fashion. In vitro screening is suitable for this purpose. TiO2NP exist in a large variety (crystal structure, coating and size), but information on their relative toxicities is scarce. TiO2NP may be inhaled by workers in e.g. paint production and application. In mice, inhalation of TiO2NP increases allergic reactions. Dendritic cells (DC) form an important part of the lung immune system, and are essential in adjuvant activity. The present study aimed to establish the effect of a variety of TiO2NP on DC maturation in vitro. Two NP of different crystal structure but similar in size, uncoated and from the same supplier, were evaluated for their adjuvant activity in vivo.MethodsImmature DC were differentiated in vitro from human peripheral blood monocytes. Exposure effects of a series of fourteen TiO2NP on cell viability, CD83 and CD86 expression, and IL-12p40 and TNF-α production were measured. BALB/c mice were intranasally sensitized with ovalbumin (OVA) alone, OVA plus anatase TiO2NP, OVA plus rutile TiO2NP, and OVA plus Carbon Black (CB; positive control). The mice were intranasally challenged with OVA. OVA-specific IgE and IgG1 in serum, cellular inflammation in bronchoalveolar lavage fluid (BALF) and IL-4 and IL-5 production in draining bronchial lymph nodes were evaluated.ResultsAll NP dispersions contained NP aggregates. The anatase NP and anatase/rutile mixture NP induced a higher CD83 and CD86 expression and a higher IL-12p40 production in vitro than the rutile NP (including coated rutile NP and a rutile NP of a 10-fold larger primary diameter). OVA-specific serum IgE and IgG1 were increased by anatase NP, rutile NP, and CB, in the order rutile<anatase<CB. The three particles similarly increased IL-4 and IL-5 production by bronchial LN cells and eosinophils and lymphocytes in the BALF. Neutrophils were induced by rutile NP and CB but not by anatase NP.ConclusionsOur data show that measuring CD83 and CD86 expression and IL-12p40 and TNF-α production in DC in vitro may provide an efficient way to screen NP for potential adjuvant activity; future studies should establish whether this also holds for other NP. Based on antigen-specific IgE and IgG1, anatase NP have higher adjuvant activity than rutile NP, confirming our in vitro data. Other parameters of the allergic response showed a similar response for the two NP crystal structures. From the viewpoint of safe(r) by design products, rutile NP may be preferred over anatase NP, especially when inhalation exposure can be expected during production or application of the product.

A quantitative approach to assess the potency of skin sensitizers in the elicitation phase.

The concept that thresholds exist for the induction of allergic contact dermatitis by chemicals with skin sensitizing properties has been used for a quantitative risk assessment approach. In this approach the potency of skin sensitizers as determined in the Local Lymph Node Assay is used to calculate the threshold for induction of sensitization. These are then used to estimate safe exposure levels for consumers. Whether these exposure levels will protect subjects that are already sensitized is unknown. The elicitation of allergic contact dermatitis supposedly occurs above a certain threshold as well and this threshold is most likely lower than that for the induction. It is unclear if induction thresholds can be extrapolated to elicitation thresholds. The aim of this study was to assess the potency of sensitizers with different sensitizing potencies in the elicitation phase in a mouse model for elicitation. Mice were sensitized by topical application on days 0 and 7 using equipotent concentrations of oxazolone, 2,4-dinitrochlorobenzene (DNCB) and eugenol to ensure that the sensitization strength would not influence the elicitation potency. Mice were challenged on day 21 by topical application on the ears in a dose-dependent manner and dose-response data were used to calculate the elicitation potency. Unexpectedly, sensitizers with different sensitizing potencies induced not the same dose-response curves in sensitized mice. The most potent sensitizer in the elicitation phase was oxazolone, followed by DNCB and eugenol. Similar to the induction phase, under equipotent sensitization conditions strong sensitizers such as oxazolone and DNCB elicit allergic reactions at lower concentrations than weak sensitizers such as eugenol. Our results indicate that elicitation thresholds cannot be readily deduced from sensitization thresholds.