Conceived and designed the experiments: WM MF LZH. Performed the experiments: WM IH TPJH GKS MN MF LZH. Analyzed the data: WM IH TPJH LZH. Contributed reagents/materials/analysis tools: WM IH GKS MN MF. Wrote the paper: WM MF LZH.
WM received a research grant from Pari Pharma GmbH, Gräfelfing, Germany for conducting a study on nasal drug delivery devices. This does not alter the authors' adherence to all the PLoS ONE policies on sharing data and materials. All other authors declare that no competing interests exist. There is no product of Pari Pharma where WM is involved in and where the authors' share intellectual property rights or being involved in a patent. The topic of the grant is related to nasal drug delivery and is not in any case related to the topic of the submitted manuscript.
Endotoxin (Lipopolysaccharide, LPS) is a potent inducer of inflammation and there is various LPS contamination in the environment, being a trigger of lung diseases and exacerbation. The objective of this study was to assess the time course of inflammation and the sensitivities of the airways and alveoli to targeted LPS inhalation in order to understand the role of LPS challenge in airway disease.
In healthy volunteers without any bronchial hyperresponsiveness we targeted sequentially 1, 5 and 20 µg LPS to the airways and 5 µg LPS to the alveoli using controlled aerosol bolus inhalation. Inflammatory parameters were assessed during a 72 h time period. LPS deposited in the airways induced dose dependent systemic responses with increases of blood neutrophils (peaking at 6 h), Interleukin-6 (peaking at 6 h), body temperature (peaking at 12 h), and CRP (peaking at 24 h). 5 µg LPS targeted to the alveoli caused significantly stronger effects compared to 5 µg airway LPS deposition. Local responses were studied by measuring lung function (FEV1) and reactive oxygen production, assessed by hydrogen peroxide (H2O2) in fractionated exhaled breath condensate (EBC). FEV1 showed a dose dependent decline, with lowest values at 12 h post LPS challenge. There was a significant 2-fold H2O2 induction in airway-EBC at 2 h post LPS inhalation. Alveolar LPS targeting resulted in the induction of very low levels of EBC-H2O2.
Targeting LPS to the alveoli leads to stronger systemic responses compared to airway LPS targeting. Targeted LPS inhalation may provide a novel model of airway inflammation for studying the role of LPS contamination of air pollution in lung diseases, exacerbation and anti-inflammatory drugs.
Endotoxin (Lipopolysaccharide, LPS) is a constituent of the outer membrane of Gram-negative bacteria and an important microbial trigger that stimulates innate immunity
The human body is confronted with LPS during infection with gram-negative bacteria. There is however also LPS contamination of particulate matter (PM10) in air pollution
Inhalation of LPS can be used to determine the competence of the innate immune system regarding gram-negative bacteria
In addition, we have to consider that the airways and the alveolar space may have different sensitivities to endotoxin challenge. Inhaled particles, including bacteria, viruses and aerosolized drugs have different deposition probabilities in airways and alveoli, depending on particle size and inhalation parameters (flow rate, tidal volume) such that for example larger particles (>4 µm diameter) at a flow rate of 500 mL/s primarily deposit in the airways, while smaller particles at lower flow rates penetrate to the alveoli. The alveolar space is covered with a surfactant monolayer and there is only a 2 µm thick barrier between air and blood. Aerosols reaching this area may therefore more directly interact with pneumocytes and alveolar macrophages, or can penetrate and reach the systemic circulation, as was shown for inhaled nanoparticles
In contrast, the airway is covered with epithelium including the cilia, which form a more than 10 µm thick cellular barrier between the airway lumen and the circulation. In addition, the airways are covered with a mucus layer of 6 µm thickness, although there are some areas without mucus
Based on these differences we hypothesized that there will be different responses when LPS is targeted to the alveoli compared to the airways. We show herein that targeting LPS to the airways does lead to lower inflammatory responses compared to LPS deposition in the alveoli. This may have significant impact on responses to inhaled endotoxin, disease progression and severity.
In order to exclude bronchial hyperresponsiveness, candidate participants were subjected to increasing doses of inhaled methacholine. Volunteers were recruited via newspaper adverts. Among 15 healthy non-smoking volunteers tested seven volunteers (5 male, 2 female, age 49+/−17 years, mean +/− standard deviation) did not show any degree of hyperresponsiveness and were selected for participating in the LPS study (
Mean+/−SD | |
Age, years | 49+/−17 |
Number (male/female) | 7 (5/2) |
|
|
FEV1, %pred | 111+/−12 |
FVC, %pred | 114+/−14 |
FEV1%FVC | 77+/−7 |
RTOT, kPa*s/L | 0.17+/−0.04 |
SRTOT, kPa*s | 0.70+/−0.14 |
Among fifteen non-smoking subjects enrolled in the study, seven did not show any kind of hyperresponsiveness and were therefore included in the LPS challenge study.
The study protocol is shown in
In addition body temperature (BT) and lung function (LF) was assessed. Exhaled breath condensate (EBC) was collected and analyzed for H2O2 concentration and acidity (pH). After targeting of 1, 5 or 20 µg LPS to the airways or 5 µg LPS to the alveoli, inflammatory parameters in blood and BT, LF and EBC were assessed according the time scale.
The volunteers sequentially inhaled 1, 5 and 20 µg LPS deposited to the airways with at least 4 weeks between the exposures. The analysis of body temperature, blood neutrophils, CRP and H2O2 demonstrated that 5 µg was effective in inducing responses in all individuals (see under Results). This dose of LPS was then deposited to the alveoli. Targeted delivery of aerosolized LPS to the airways or to the lung periphery (alveoli) was done between 09:00 and 11:00 by aerosol bolus inhalation using the AKITA® device (Activaero GmbH, Gemünden, Germany). Previous studies have shown that the AKITA® device shows little inter- and intra-subject variation of aerosol deposition in the lung
DSB was used as threshold volume for airway (AW) and alveolar (AL) condensate sampling separation. In addition the first 50 mL of the exhaled air were discarded. The grey areas show the size and the penetration of the shallow and the deep LPS aerosol bolus for targeting the airways or the alveolar space.
Exhaled breath condensate was collected using the EcoScreen-II (Filt GmbH, Berlin, Germany). The EcoScreen-II allows the non-invasive collection of volatile and non-gaseous contents in exhaled air in two separate condensation chambers
Because H2O2 is not stable over longer periods of time, immediate analysis of the collected condensate for hydrogen peroxide (H2O2) and pH was performed using the EcoCheck device (Filt GmbH, Berlin, Germany). The EcoCheck is a biosensor device for measuring H2O2 concentrations by enzymatic peroxidase reduction. The lower detection limit was 50 nmol/l
The concentration of IL-6 in the plasma samples was quantified using a customized Milliplex MAP Human Cytokine/Chemokine Panel (# HCYTOMAG-60K, Millipore, Schwalbach, Germany). The assay was performed according to the manufacturer's instructions. Standards and samples were analyzed in duplicates on a Luminex 200 device (BioRad, München, Germany) using the BioPlex Manager Software (Version 5, BioRad).
Data are expressed as mean +/− standard deviation (SD). Although the data sample is small (n = 7) the parameters did not show significant difference from normal distribution (according to the Kolmogorov-Smirnov-test). Differences among study groups and between airway and alveolar study parameters were assessed by the two-sided t-test (Winstat for Microsoft Excel, Version 2008.1,
In order to exclude bronchial hyperresponsiveness, candidate participants were subjected to increasing doses of inhaled methacholine. Among the 15 healthy volunteers tested there was none responding with a more than 20% decrease of FEV1 to the highest dose of methacholine (0. 77 mg of methacholine). Looking at resistance we found eight volunteers, who responded to methacholine with an increase by more than factor 2 above baseline (0.17+/−0.04 kPa*sec/L), suggesting a low level of hyperresponsiveness. Seven volunteers, who did not show any degree of hyperresponsiveness, were included in the LPS challenge study. The anthropometric and lung function data of these seven study subjects are listed in
Body temperature after LPS inhalation peaked between 6 and 12 h and returned to baseline in all subjects at 24 h. For both 1 µg and 5 µg airway LPS we noted only a mild increase to below 37°C. After 20 µg of airway LPS deposition a strong increase to an average of 38.2+/−0.9°C was seen at 12 h (p<0.01,
A) Systemic response parameter ‘body temperature’ during 72 h after targeting different doses of LPS either to the airways (closed symbols) or to the alveoli (open symbols). B) Increase of 12 h - body temperature with increasing LPS dose targeted to the airways in comparison to 5 µg LPS targeted to the alveoli (open symbol). Data represent mean +/− SD (n = 7; *: p<0.05, **: p<0.01 compared to baseline; ++: p<0.01 for 5 µg alveolar compared to 5 µg airway LPS).
A similar pattern of responses was seen for blood neutrophils. Here the peak response was at 6 h and the values were back to baseline at 48 h. LPS at 1 µg airway deposition showed an increase from 3400+/−1300/µL to 5400+/−1300/µL at 6 h (p<0.005). The peak response at 6 h increased with increasing LPS dose (p<0.001,
A: absolute blood neutrophils (6 h after LPS challenge, baseline value = 3.4+/−1.3*103/µL), B: CRP (24 h after LPS challenge, baseline value = 1.4+/−0.9 mg/L) and C: IL-6 (baseline value 2.0+/−0.4 pg/mL). Data represent mean +/− SD (n = 7; *: p<0.05, **: p<0.01 compared to baseline; ++: p<0.01 for 5 µg alveolar compared to 5 µg airway LPS).
Also there was a clear increase of CRP from 1.4+/−0.9 mg/L before to 3.4+/−2.3 mg/L after 1 µg of airway LPS challenge at 24 h (p<0.05) and at 72 h after LPS challenge CRP was still significantly above baseline. The 24 h CRP peak value significantly increased with LPS dose (p<0.001,
Serum samples taken from experiments with alveolar and bronchial exposure to 5 µg LPS were tested for IL-6 protein levels. As shown in
When LPS is applied to the airways then a local inflammation may lead to air flow limitation. We therefore monitored FEV1 using a hand held spirometer. For 1 µg airway LPS there was a significant decrease of FEV1 at 12 h (97.5+/−2.5% of the baseline value (p<0.05), and there was a further decrease of 12 h FEV1 with increasing airway LPS dose to 93.4+/−4.6% and 84.8+/−8.8% of the baseline value after 5 µg and 20 µg LPS, respectively (p<0.01 for both doses,
Data represent mean +/− SD (n = 7, *: p<0.05, **: p<0.01 compared to baseline).
LPS can trigger reactive oxygen production by inducing assembly of the NADPH oxidase complex. We therefore have asked whether an increase of H2O2 can be detected in exhaled breath after LPS inhalation. For this we collected EBC samples separated into an airway and an alveolar fraction, where the airway fraction represents about one third and the alveolar fraction two thirds of the collected volume (see
When looking at the EBC-airway fraction after LPS was targeted to the airways then the induced H2O2 peaked at 2 h with values of 526+/−280 81 nmol/L, 442+/−208 nmol/L and 538+/−173 nmol/L for 1 µg, 5 µg and 20 µg, respectively (all p<0.05 compared to baseline values, see
When analyzing induced H2O2 in the same EBC-airway fraction after 5 µg LPS dose targeted to the alveoli then we also saw significant induction at 328+/−135 nmol/L at 2 h (p<0.05 compared to baseline EBC-H2O2). The airway response to 5 µg LPS targeted to the airways was higher in tendency compared to this response when targeted to the alveoli.
In the alveolar EBC fraction (data not shown) we detected much lower 2 h values with 216+/−83, 182+/−177 and 196+/−104 nmol/L after 1 µg, 5 µg and 20 µg bronchial LPS, respectively (all p<0.01 compared to baseline values). After alveolar LPS challenge the alveolar EBC did not show a significant induction of H2O2.
Most studies published on human LPS challenge used full breath LPS inhalation and thereby cannot account for differences in defence and immune responses in the different regions of the respiratory tract. A novel human airway inflammation study limiting LPS challenge to the airways used segmental endotoxin challenge in healthy subjects during bronchoscopy
Our study used aerosol bolus inhalation in order to enable controlled LPS targeting either to the airways or to the alveoli. When controlling LPS inhalation by the Akita device then there is minor variation of the delivered LPS dose to either of the target sites
However, targeting by the bolus technique does not exclusively deposit aerosol in either of the anatomical sites although the delivery protocols and particle size were optimized. Although the major fraction of inhaled aerosol is deposited either in the central airways (generations 1–10) after shallow bolus inhalation or in the alveoli (generations 18–23) after deep bolus inhalation, as illustrated in
Nevertheless, as illustrated in
Previous studies suggested that allergy can influence responses to inhaled LPS, with either increased or impaired responses
The participants repeatedly inhaled LPS targeted to the airways in increasing doses and finally a medium dose to the alveoli. Since repeated LPS exposure can induce a non-responsiveness also termed tolerance
When looking at systemic responses to LPS inhalation we see a dose dependent rise in body temperature which peaks at 12 h and has resolved after 24 h. This is in line with a transient induction and release of endogenous pyrogens like IL-1 and IL-6, which are produced locally and then act on the hypothalamus
For neutrophils we saw a much earlier peak response at 6 h and this again was dose dependent with a higher response to 5 µg alveolar as compared to 5 µg airway deposition. The major factor involved in immediate rise of neutrophils is G-CSF (granulocyte-colony-stimulating factor), which can mediate release of neutrophils from bone marrow by interference with CXCL12-CXCR4 interactions that retain these cells in bone marrow
CRP peaked much later at 24 h, which is in line with the clinical experience in infection and inflammation. CRP is synthesized in the liver and it is under control of cytokines like IL-6
When looking at local responses we studied airway constriction and noted a dose dependent decrease of FEV1. This is best explained by the local induction of inflammation with subsequent thickening of the airway wall leading to a reduced width of the airway. There also may be a contribution by smooth muscle cells via a cytokine mediated enhancement of acetylcholine triggered contraction
Of note, in our study there also was a reduction in FEV1 after alveolar LPS deposition. We hypothesize that this response is due to airway deposition of some LPS as it passes through the airways. Since only a minor fraction of the LPS is deposited in the airways during the alveolar targeting and since there is a clear linear dose dependence this would predict a less pronounced effect on FEV1 for 5 µg alveolar compared to 5 µg airway LPS. The decrease of FEV1 is, however, similar for the two deposition sites at the same 5 µg dose. Therefore, we assume that also a systemic component of the inflammatory response contributes to the transient airway obstruction (subjects reported chest tightness).
One major defence mechanism induced by LPS is via the induction of reactive oxygen species. LPS can induce assembly of the NADPH-oxidase complex leading to H2O2 production
In a previous study using this fractionated EBC sampling technique we have shown that basal levels of H2O2 in EBC are higher in the airway compartment compared to the alveoli and this was true for non-smokers, smokers and in COPD patients
With LPS inhalation we noted a pronounced rise in H2O2 production with a peak at 2 h both, in the airway and in the alveolar EBC. Since in-vitro 20 minutes is sufficient to generate an optimum oxidative burst
When comparing airway and alveolar deposition at 5 µg of LPS each it is apparent that the H2O2 production in both the airway and the alveolar EBC fraction is higher with the airway LPS deposition. This higher signal in the airway fraction is conceivable since there is a higher area concentration of LPS at this site (see
Systemic and local inflammation parameters assessed in our study peaked at characteristic time points after LPS challenge, as summarized in
Our data clearly show that targeted LPS delivery by controlled inhalation will lead to a highly reproducible inflammatory response, with predictable peaking times for blood neutrophils, body temperature, FEV1 impairment, CRP and H2O2 production. Also, clinical symptoms consistently have disappeared after 24 h. Taken together we have described herein that LPS targeted to the airways compared to the alveoli generates significantly lower systemic responses, but similar local H2O2 responses. The human inflammation model proposed in our study allows controlled LPS challenge to the airways or to the pulmonary region. Note that our study provides inflammatory responses with respect to deposited LPS dose in the respective lung region while all other studies provide nebulized LPS dose. Since there is great variability in regional particle deposition with respect to particle size, size distribution, inhalation pattern and disease severity, the deposited dose in the target region, the airways, is very variable and partly unknown. In addition, as our study showed higher systemic responses of inhaled LPS in the pulmonary region, most of the systemic responses reported in other studies using tidal breathing may suffer from these not wanted side effects. In addition using shallow bolus LPS inhalation in COPD patients for studying exacerbation one may significantly reduce risks of severe side effects, since the site of deposition and the deposited dose are under control. This may open the opportunity of studying new anti-inflammatory drugs in COPD, such as steroids, β2-agonists, or anti-MCP-1 monoclonal antibody (controlling monocyte recruitment)
Exacerbations are important events in patients with asthma and chronic obstructive pulmonary disease (COPD)
(DOC)
Deposition distribution in the different generations of the human lung.
(TIF)
(TIF)
We thank Activaero GmbH, Gauting, Germany for providing the Akita inhalation device.