Search

Cell & tissue research 387(2), 249-260 (2022). doi:10.1007/s00441-021-03553-1, Published by Springer, Berlin ; Heidelberg

Chronic P. aeruginosa infections of the respiratory tract pose a major health care problem. The bacterial flagellum represents a tightly regulated virulence factor involved in biofilm formation as well as immune recognition of P. aeruginosa. Aim of this study was to delineate the role of flagellin using precision-cut lung slices (PCLS) as an organotypic, immunocompetent model of the human lung. PCLS were prepared from tumour-free lung tissue and infected with wildtype (wt) or flagellin mutant (∆fliC) PA14 strains. Persistent infection of the tissue over 24h was achieved by treatment with subinhibitory concentrations of tobramycin. Bacterial load was determined and lung tissue viability as well as specific cytokine responses were measured to assess the immune response. While the initial attachment of P. aeruginosa to the lung tissue was enhanced by the presence of the flagellum, both wt as well as ∆fliC strains were able to colonize the tissue over 24h. Analysis of cytokine profiles revealed a strong upregulation of pro-inflammatory mediators such as IL-8 (6-fold), IL-6 (6-fold), TNF-α (20-fold), GM-CSF (25-fold), IL-1β (52-fold), and MIP-3α (19-fold) after 24h of infection with wt PA14 compared to non-infected control PCLS. Importantly, infection with ∆fliC PA14 induced 2- to 4-fold lower levels of the respective molecules than the wt strain. Taken together, we show here that wt as well as flagellar mutant P. aeruginosa strains can infect human lung tissue ex vivo whereby flagellin acts as a potent immune stimulus, which could have implications for both immune evasion and biofilm formation of P. aeruginosa. Future studies will look at the transformation of acute towards persistent infection.

Cystic fibrosis patients are highly susceptible to Pseudomonas aeruginosa which rapidly adapts to the lung microenvironment and causes chronic inflammation strongly contributing to disease pathogenesis. In the presented study, fresh human lung tissue was infected ex vivo with P. aeruginosa to investigate bacterial adaption to lung tissue and antibiotic treatment as well as the host immune response. Human lung tissue slices were infected with 105 PAO1, treated with 10, 20 or 40µg/ml tobramycin and incubated for 24h or 96h. Tissue viability was assessed by Calcein AM staining. Bacterial load was determined by dilution plating and enumeration of CFU. Bacterial colonization was imaged by confocal microscopy. Cytokine release was determined using ELISA.P. aeruginosa rapidly colonized the lung tissue and induced a strong and persistent host immune response as indicated by the release of IL-8, IL-6, IL-1ß and TNF-a. Tobramycin treatment reduced CFU and resulted in maintained tissue viability, but without full eradication of P. aeruginosa, thereby enabling the transition to persisting infection up to 96h. Interestingly, this adaption of P. aeruginosa was associated with a decreased sensitivity to repeated antibiotic treatment. In summary, P. aeruginosa infection and respective induction of the host immune response can be reflected in human lung tissue ex vivo. Furthermore, the data indicate that early stages of the transition of acute to persistent infection can be mimicked, thereby reflecting bacterial adaption to antibiotic treatment. This paves the way to future investigations on host-pathogen interactions as well as testing of novel anti-infective or immunomodulatory therapeutics.

Pulmonary fibrosis comprises a multitude of rapidly progressing interstitial lung diseases assumed to develop from failure in the process of wound healing. Although TGFβ was reported to play a central role in the pathogenesis, it seems evident that also pro-inflammatory mediators contribute to the initiation and progression of lung fibrosis. In this study, the effect of TGFβ and TNFα on the induction of profibrotic mediators was investigated in human ex vivo lung tissue slices (Precision-Cut Lung Slices; PCLS). PCLS were prepared from tumor-free lung tissue sections and were cultured in the presence of TGFβ, TNFα or combination (TGFβ/TNFα) for up to 48h. Stimulation with TGFβ, TNFα or TGFβ/TNFα lead to specific up- or downregulation of distinct mediators in PCLS as compared to control samples, highlighting the necessity of combining key cytokines for the efficient induction of a pro-fibrotic profile. Upon others, TNFα as well as TGFβ/TNFα stimulation of PCLS lead to specific upregulation of IL-1β, while single stimulation with TGFβ showed no effect. IP10 and IFNγ levels remain unchanged in the presence of TNFα, while TGFβ/TNFα, or TGFβ alone lead to down-modulation of the expression. Stimulation of PCLS with either TNFα or TGFβ showed 2- to 3-fold up-regulation of PAI1. The combination of TNFα and TGFβ lead to further amplification of PAI1 mRNA levels to up to 7-fold. Importantly, ex vivo treatment with Pirfenidone efficiently reverts the TGFβ and TNFα-mediated modulation. We describe here the induction of a specific pro-fibrotic biomarker pattern in PCLS by TGFβ and TNFα stimulation, enabling the investigation of fibrotic signaling pathways with high translational relevance.

Formation of metastases marks the main cause of death in cancer and is characterized by complex processes, which can only partly be reflected in most in-vitro models due to a lacking human microenvironment. Cisplatin, bevacizumab and vemurafenib were used to modulate different aspects of tumorigenesis in co-cultures of patient-derived, disseminated melanoma cells or MDA-MB-231 cells and fresh human lung tissue as well as fresh human tumor tissue slices. Neoangiogenetic biomarker VEGF was elevated 3.5fold in co-cultures and 5.4fold in lung tumor slices after 48h. Treatment with bevacizumab [200 mg/mL] suppressed VEGF-release 24fold in co-cultures and 25fold in lung tumor tissue slices after 48h, also leading to impaired endothelial cell migration of the supernatants (81% in co-cultures, 83% in tumor slices). Cisplatin treatment [50mM] led to a decline of viability in co-cultures by up to 37.5% and in tumor slices up to 48.7% after 72h. Co-cultures and tumor slices were less sensitive in regards to their respective anti-cancer drug efficacy than the 2D culture. To address patient-specific genetic disparities, co-cultures with disseminated melanoma cells were performed with both patient-derived cells carrying BRAF mutation V600E and non-mutated cells. Treatment with vemurafenib reduced V600E cancer cell number (IC50=6µM) after 48 hours while non-mutated cells showed no significant cancer cell decrease. Here we show that tumor cells in solid tumors and seeded into healthy lung tissue are sensitive to tumor treatments ex vivo. This work aims to translate cancer drug efficacy data from animals towards humans to optimize clinical trial outcome.

Cigarette smoke (Cs) inhalation is one of the main reasons to develop chronic obstructive pulmonary disease (COPD), characterised by degradation of alveoli, emphysema development, inflammation and mucus hypersecretion. Mechanisms that underlie various components of COPD can be modelled in vitro, specifically using whole fresh cigarette smoke. We assessed the effects of Cs and Cs condensate (Csc) on fresh human lung tissue. Human Precision-Cut Lung Slices (PCLS) were exposed to Csc or whole Cs in an Air-Liquid Interface (ALI) using the in vitro exposure device P.R.I.T.® ExpoCube®. Tissue viability, release of cytokines and extracellular matrix (ECM) proteins were analysed after single and repeated exposure and cultivation of up to 96h. Inhibitors were applied to suppress inflammatory responses of tissue to Cs. Csc and Cs induced concentration-dependent loss of tissue viability in fresh lung tissue. Smoke exposure induced an increased release of pro-inflammatory cytokines IL-1a (~ two fold), IL-1v (~ three fold) and changes in ECM proteins, e.g. MMP-9 and pro-Collagen1a1 (~ two fold), as well as Receptor for Advanced Glycation Endproducts (RAGE) from human PCLS. Dexamethasone and Roflumilast inhibited Cs-induced increase of pro-inflammatory cytokines. Csc and Cs induced tissue injury, early biomarkers of inflammation and changes in ECM proteins in vital ex vivo human lung tissue. The exposure of the complex mixture of whole cigarette smoke closely reflects the in vivo situation in human lung tissue. Next we will assess the response of human PCLS exposed to Cs and infected with rhinovirus to study exacerbations.

A 59-year-old patient presented in the outpatient clinic because of an increasing mass in his right buttock. He has a history of a testicular tumor 28 years ago, with increased tumor markers (Human Choriongonadotropin (β-HCG) value 4.9 and Alpha-1 Fetoprotein (AFP) 218 ng/ml).

Increasing evidences point to a functional unit from mast cell and sensory nerve as a key component of pathophysiological responses. Here we propose a subpopulation of neurokinin-1 receptor (NK1R) expressing mast cells in the peripheral tissue of humans to be involved in airway hyper responsivness (AHR).Human precision cut lung slices PCLS were generated using resection material from lung tumor patients and passivly sensitized with plasma from HDM-allergic patients. Tissue was treated with 10 µM histamine 1 receptor (H1R) or 100µM NK1R antagonists or a monoclonal antibody against IgE (mab IgE) during sensitization. Subsequently, bronchoconstriction was provoked with the sensory nerve agonist capsaicin. Bronchoconstriction was measured by video microscopy. The reduction of the airway area in percentage of initial airway area was analysed. Mast cell tryptase, IgE and NK1R immunostaining was performed in PCLS in order to visualize mast cell subpopulations. Immunostaining of sensory nerve fibers and mast cells showed anatomical proximity in the airways. A subpopulation of NK1 receptor positive mast cells were identified. After passive sensitization, capsaicin‑induced bronchoconstriction showed reduction of the airway area of 40%. Capsaicin‑induced bronchoconstriction was completely inhibited by pre-incubation with NK1R and H1R antagonists or mab IgE. Capsaicin-induced significant increase in bronchoconstriction in passive sensitized human PCLS. Capsaicin induced bronchoconstriction under asthmatic conditions is mediated by a process involving a subpopulation of NK1R positive mast cells.

Pulmonary fibrosis is a severe and fast progressing lung disease causing irreversible dysfunction of the organ. Current treatments are accompanied by severe side effects and do not accomplish final cure of the disease. With the use of fresh lung tissue, the aim was to identify specific biomarkers of pulmonary fibrosis ex vivo in order to improve the investigation of novel pharmacological approaches. Vital tumor-free lung tissue was used to prepare human ex vivo tissue slices. Slices were cultured in the presence of TGF-β and TNF-α to induce a pro-fibrotic profile. Rat tissue slices were prepared from lungs of bleomycin-treated rats and were cultivated in medium only. Human tissue slices stimulated ex vivo with TGF-β and TNF-α displayed an up-regulation of important pro-fibrotic genes as compared to control samples. Amongst others, mRNA levels of matrix metalloproteinase 13 were 2- to 6-fold and the extracellular enzyme lysil oxidase was 2- to 3-fold elevated. Other genes, e.g caveolin-1 were downregulated in the presence of TGF-β and TNF-α. Tissue slices prepared from bleomycin treated rats revealed a comparable pattern of up-regulated pro-fibrotic genes. Here, e.g. mRNA levels of collagen1a1 or fibronectin 1 were significantly up-regulated as compared to control samples. Moreover, this pattern shows stable expression in culture for 2 to 5 days. Comparison of the mRNA profiles from both species, by using INGENUITY pathway analysis software, revealed identical canonical pathways and upstream regulators. This characteristic pattern of pro-fibrotic biomarkers, described in ex vivo lung tissue slices from two different species, enables the investigation of pulmonary fibrosis with high translational relevance.

Human rhinovirus (HRV) is a main cause of the common cold in humans, and for asthmatics a major risk factor for exacerbations. HRV targets the airway mucosa with host epithelial cells and new virions are released without inducing a direct cytopathic effect (CPE). Infection of epithelial cells by HRV induces an innate immune response with secretion of cytokines to mediate the anti-viral host response. The aim of the study was to establish a model of HRV1B infection in human PCLS including prevention of induced immune responses by intervention with 3C protease inhibitor Rupintrivir. Human PCLS containing airways were prepared and inoculated with HRV1B, UV-inactivated HRV, medium or HRV in presence of Rupintrivir for 2h at 33°C. PCLS were incubated for 1d and 3d. Viral load was titrated on HeLa cells and the CPE was analysed. Tissue viability was determined. Pro-inflammatory and anti-viral cytokine regulation was determined by multiplex. After infection with HRV1B no release of virions to supernatant was detected, however replicating virus was confirmed by staining of dsRNA and analysis by confocal LSM in PCLS. HRV infected human PCLS showed no significant loss of vitality and low CPE. HRV infection induced expression of cytokines such as IFN-γ, RANTES, IL-6 and TNF-α. The tissue response was reduced after intervention with Rupintrivir. In particular IFN-γ was increased 150 fold in supernatant of HRV infected PCLS and reduced by Rupintrivir for about 3 fold after 1d. In conclusion, infection of ex vivo human lung tissue with HRV1B induced anti-viral immune responses. This response to HRV was inhibited by Rupintrivir. Next we will assess the response of human PCLS with asthmatic phenotype to HRV infection.

Idiopathic pulmonary fibrosis is a severe lung disease, causing irreversible dysfunction of the organ. However, the exact pathomechanism of IPF is not yet understood and identification of specific biomarkers is urgently needed. Yet, a general shift from in vivo to ex vivo models (and from animals to humans) seems to be necessary in order to support the development of translational models. With the use of PCLS, the aim was to establish a novel experimental method to identify biomarkers of pulmonary fibrosis. PCLS were prepared from lungs of bleomycin-treated rats, NaCl-treated controls or tumor-free lung tissue from cancer patients. Human PCLS were cultured in the presence of TGF-β and TNF-α to induce a pro-fibrotic profile, rat PCLS were cultivated in medium only. Cytokine responses were measured by ELISA and genetic profiles were analyzed by RTqPCR. Analysis of the mRNA profile revealed up-regulation of important pro-fibrotic genes in lungs from bleomycin-treated rats as compared to the controls. Amongst others, extracellular matrix (ECM) genes including different forms of collagen as well as ECM remodeling enzymes (e.g. metalloproteinases) were elevated in bleomycin treated rat lungs as compared to control animals. A similar pattern of up-regulated genes was found in human PCLS stimulated with TGF-β and TNF-α. Furthermore, the pro-inflammatory cytokine IL-1β, involved in the initial wound repair mechanism and early fibrotic response, was elevated on mRNA as well as protein level in TGF-β- and TNF-α-stimulated human PCLS. Overall, this novel ex vivo method shows great potential to investigate important signaling pathways and mechanisms of early onset fibrosis for the identification of pro-fibrotic biomarkers.

Interleukin-13 is a key cytokine of asthma and elevated in asthmatics resulting in airway hyperresponsiveness (AHR). AHR is a hallmark of allergic asthma defined as exaggerated bronchoconstriction in response to contractile stimuli. Research for the development of new drugs is mainly based on appropriate in vivo and in vitro models. There is a need for translational models with improved predictivity for human. For comparison of different species we used precision-cut lung slices (PCLS) and assessed IL-13 induced hyperreactivity in PCLS of mice, rats, and humans. PCLS were prepared from Balb/c mice, Brown Norway rats, and humans. IL-13 receptor was stained in the airways of mouse, rat and human PCLS by immunohistochemistry. Airways of all species were pre-incubated with 100 ng/mL IL-13. Subsequently, bronchoconstriction was induced by addition of methacholine (MCh) and visualized by videomicroscopy. IL-13 receptor was present in epithelial cells and smooth muscle cells in PCLS of all species. Methacholine-induced bronchoconstriction in mouse exhibited an EC50 of 80 nM and decreased by pre-incubation with IL-13 to 50 nM, in rat from 220 nM to 170 nM and human from 180 nM to 47 nM MCh. In general, pre-incubation of PCLS in the presence of IL-13 resulted in all species in stronger bronchoconstriction at maximum methacholine concentration. Maximal constriction of initial airway area resulted in mouse in Cmax 61% by control and decreased in IL-13 pre-incubated tissue to 80%, in rat by 49% to 69% and human by 85% to 94% compared to untreated tissue. This study shows that IL-13 receptor is similar distributed in epithelial cells and smooth muscle cells of all three species. IL-13 induced airway hyperreactivity in all tested species with different methacholine sensitivity. In future studies, PCLS will be used for pre-clinical studies to valuate the antagonist efficacy.

Occupational asthma can be induced by a number of chemicals at the workplace. Risk assessment of potential sensitizers is mostly performed in animal experiments. With increasing public demand for alternative methods, human precision-cut lung slices (PCLS) have been developed as an ex vivo model. Human PCLS were exposed to increasing concentrations of 20 industrial chemicals including 4 respiratory allergens, 11 contact allergens, and 5 non-sensitizing irritants. Local respiratory irritation was characterized and expressed as 75% (EC 25 ) and 50% (EC 50 ) cell viability with respect to controls. Dose–response curves of all chemicals except for phenol were generated. Local respiratory inflammation was quantified by measuring the production of cytokines and chemokines. TNF-α and IL-1α were increased significantly in human PCLS after exposure to the respiratory sensitizers trimellitic anhydride (TMA) and ammonium hexachloroplatinate (HClPt) at subtoxic concentrations, while contact sensitizers and non-sensitizing irritants failed to induce the release of these cytokines to the same extent. Interestingly, significant increases in T H 1/T H 2 cytokines could be detected only after exposure to HClPt at a subtoxic concentration. In conclusion, allergen-induced cytokines were observed but not considered as biomarkers for the differentiation between respiratory and contact sensitizers. Our preliminary results show an ex vivo model which might be used for prediction of chemical-induced toxicity, but is due to its complex three-dimensional structure not applicable for a simple screening of functional and behavior changes of certain cell populations such as dendritic cells and T-cells in response to allergens.

Increasing incidence and substantial morbidity and mortality of respiratory diseases requires the development of new human-specific anti-inflammatory and disease-modifying therapeutics. Therefore, new predictive animal models that closely reflect human lung pathology are needed. In the current study, a tiered acute lipopolysaccharide (LPS)-induced inflammation model was established in marmoset monkeys (Callithrix jacchus) to reflect crucial features of inflammatory lung diseases. Firstly, in an ex vivo approach marmoset and, for the purposes of comparison, human precision-cut lung slices (PCLS) were stimulated with LPS in the presence or absence of the phosphodiesterase-4 (PDE4) inhibitor roflumilast. Pro-inflammatory cytokines including tumor necrosis factor-alpha (TNF-alpha) and macrophage inflammatory protein-1 beta (MIP-1 beta) were measured. The corticosteroid dexamethasone was used as treatment control. Secondly, in an in vivo approach marmosets were pre-treated with roflumilast or dexamethasone and unilaterally challenged with LPS. Ipsilateral bronchoalveolar lavage (BAL) was conducted 18 hours after LPS challenge. BAL fluid was processed and analyzed for neutrophils, TNF-alpha, and MIP-1 beta. TNF-alpha release in marmoset PCLS correlated significantly with human PCLS. Roflumilast treatment significantly reduced TNF-alpha secretion ex vivo in both species, with comparable half maximal inhibitory concentration (IC50). LPS instillation into marmoset lungs caused a profound inflammation as shown by neutrophilic influx and increased TNF-alpha and MIP-1 beta levels in BAL fluid. This inflammatory response was significantly suppressed by roflumilast and dexamethasone. The close similarity of marmoset and human lungs regarding LPS-induced inflammation and the significant anti-inflammatory effect of approved pharmaceuticals assess the suitability of marmoset monkeys to serve as a promising model for studying anti-inflammatory drugs.

Introduction: A number of low-molecular weight (LMW) chemicals at workplaces are involved in the development of occupational asthma. There are no currently accepted and validated test methods to identify chemicals with the potential to cause respiratory sensitization. Risk assessment is normally performed in animal experiments, however, in the context of REACH and its principle of 3Rs there is an increasing public demand for alternative methods. Human precision cut lung slices (PCLS) is an ex vivo approach in which all relevant cell types of lung tissue are present in their natural position. The aim of the study was to analyse chemical-induced inflammation and irritation by assessing a variety of immunotoxic endpoints in human PCLS. Methods: Human PCLS were prepared from tumor-free tissue from resected lung lobes of cancer patients. PCLS were incubated with 20 substances in DMEM under standard submerged cell culture conditions. PCLS without test substances were incubated as controls. After 24 hours incubation, induced toxicity was assessed by vitality staining and determination of enzymatic activity using WST-1 assay. EC25 with respect to vehicle controls were calculated for all chemicals. Cytokine contents were detected with MSD technology and ELISA. Results: Individual EC25 values correlated significantly with data published for in vitro approaches with THP-1 and NCTC cell lines (r2 = 0.87 and 0.83, respectively) (Mitjans et al. 386-95;Corsini et al. 789-96). Furthermore EC25 of human PCLS correlated with LD50 data published for in vivo rat inhalation toxicity with r of 0.53 (p value of 0.08). Subtoxic concentrations were used to investigate cytokine pattern for the differentiation between allergen and non-allergen. Respiratory allergens trimellitic anhydride and ammonium hexachloroplatinate increased the expression of the cytokines TNF? and IL-1? significantly while contact allergens (cinnamaldehyde, 2-bromo-2-glutaronitrile) and controls (phenol, SLS) failed to induce these cytokines to the same extent. Conclusion: The toxicity of chemicals in human PCLS resembles the in vitro situation very closely and also correlates with LD50 values from in vivo studies. The pro-inflammatory effects could be shown for some respiratory allergens. However, some respiratory allergens like glutaraldehyde did not induce cytokine due to the chemical properties. Altogether, PCLS could be regarded as an ex vivo immunotoxicity model, displaying chemical-induced inflammation and irritation on all relevant cell types of lung tissue.

Introduction: There is growing burden of chronic obstructive pulmonary diseases (COPD). Currently available treatments have, however, only minimal impact on disease progression. Thus, new pre-clinical in vitro / ex vivo animal models which closely reflect anatomy and organ physiology of the human respiratory tract are needed to enhance therapeutic approaches. Here we display inflammatory aspects of COPD in a lipopolysaccharide (LPS) induced acute inflammation in vitro model in marmoset monkeys (Callithrix jacchus) and humans. Methods: Peripheral blood and precision cut lung slices (PCLS) of marmoset monkeys and healthy human subjects were incubated with LPS for 24 hrs. The inflammatory response was modulated with roflumilast and dexamethasone. Supernatants were analysed for TNF-a by ELISA. The results for marmosets were correlated with the human data. Results: Peripheral blood and PCLS showed a dose-dependent a response to LPS. The TNF-a response in marmoset peripheral blood correlated significantly with the TNF-a response in marmoset PCLS (r²=0.96, p=0.0007). This was also true for human. The correlation of the ex vivo cytokine response between both species was also high in PCLS (r²=0.89). Anti-inflammatory treatment with roflumilast and dexamethasone suppressed LPS-induced TNF-a in lung tissue. Roflumilast reduced TNF-a to e.g. 61 % in marmoset and 66 % in human. The were calculated to be about 1.6 nM and 1.3 nM for marmoset and humans, respectively. Dexamethasone was also efficient. Here, the marked reduction of cytokine response was also used to calculate the IC50 which was about 0.4 nM (marmoset) and 8.4 pM (human). Discussion: There is a high analogy in the lung specific cytokine response of humans and marmosets as seen by the great similarities concerning LPS-induced inflammation in both species. Moreover, for marmoset monkeys the peripheral blood cytokine response is very similar to the lung specific cytokine response. Dexamethasone and roflumilast were effective in suppressing inflammatory responses in peripheral blood and PCLS of both species. Conclusion: In vitro models of marmoset monkeys are a promising preclinical model to study human inflammatory diseases.

Prediction of lung innate immune responses is critical for developing new drugs. Well-established immune modulators like lipopolysaccharides (LPS) can elicit a wide range of immunological effects. They are involved in acute lung diseases such as infections or chronic airway diseases such as COPD. LPS has a strong adjuvant activity, but its pyrogenicity has precluded therapeutic use. The bacterial lipopeptide MALP-2 and its synthetic derivative BPPcysMPEG are better tolerated. We have compared the effects of LPS and BPPcysMPEG on the innate immune response in human precision-cut lung slices. Cytokine responses were quantified by ELISA, Luminex, and Meso Scale Discovery technology. The initial response to LPS and BPPcysMPEG was marked by coordinated and significant release of the mediators IL-1β, MIP-1β, and IL-10 in viable PCLS. Stimulation of lung tissue with BPPcysMPEG, however, induced a differential response. While LPS upregulated IFN-γ, BPPcysMPEG did not. This traces back to their signaling pathways via TLR4 and TLR2/6. The calculated exposure doses selected for LPS covered ranges occurring in clinical studies with human beings. Correlation of obtained data with data from human BAL fluid after segmental provocation with endotoxin showed highly comparable effects, resulting in a coefficient of correlation > 0.9. Furthermore, we were interested in modulating the response to LPS. Using dexamethasone as an immunosuppressive drug for anti-inflammatory therapy, we found a significant reduction of GM-CSF, IL-1β, and IFN-γ. The PCLS-model offers the unique opportunity to test the efficacy and toxicity of biological agents intended for use by inhalation in a complex setting in humans.

Question: Human risk assessment of chemicals is currently based on the evaluation of in vivo animal studies. The use of high animal numbers and the suffering of animals in experiments such as repeated dose toxicity studies (RDT) demand for alternative test approaches. One alternative approach currently used in human risk assessment is read-across (RAX). In a RAX approach toxicity data of one or several tested source compound(s) are used to predict the toxicity of ""similar"" non-tested target compound(s). Similar compounds shall share structural- and physico-chemical proper-ties as well as a similar mode of action. This study was designed to evaluate the use of data from ex vivo experiments such as rator human precision-cut lung slices (PCLS). Three RAX-categories were tested, namely vicinal halogenides, naphthalene derivatives and vinyl esters. Each RAX category was selected based on shared structural properties and similar toxicological effects in RDT studies extracted from the FhG RepDose database (www.fraunhofer-repdose.de). Methods: Exposure of rat and human PCLS was performed on three days for three hours daily. The cytotoxicity of chemicals was assessed by LDH and WST-1 assay. Ex vivo IC50 values were calculated by sigmoidal curve fitting. These values were correlated to the LD50 values. Results: At least one substance in each category showed cyto-toxic effects in dose dependent manner. Linear regression analysis of public available in vivo LD50 values to the obtained ex vivo IC50values showed good linear correlation. Vicinal halogenides and naphthalene derivatives were less toxic in human lung sections compared to rat sections, where as the vinyl esters showed the comparable cytotoxicity in both species. Thus only one substance group showed similar cytotoxicity in both tested species, whereas two other groups reveal interspecies and metabolism diversity based on cytotoxicity endpoint. Conclusion: This first evaluation of rat and human and PCLs show that ex vivo chemical testing in human lung sections result in a promising approach for toxicity screening as no human in vivo reference data exist.

High-density lipoprotein (HDL) of 100-400 micrograms/ml did not prevent morphological alterations of human blood platelets treated with serotonin (1-5 microM). Highly concentrated HDL (1,200 micrograms/ml) appeared to activate platelets in vitro. These findings indicate that whole HDL may not inhibit agonist-induced platelet activation.