Chemicals

Piperazine: Sigma Aldrich P45907-100G (99% purity). Chloramine-T: Sigma Aldrich 857,319-100G (assay: 98–103%; impurities: ≤ 1.5% insoluble in ethanol). 2,4-Toluene-diisocyanate (TDI): Santa Cruz Biotechnology SC-251262 (≥ 80% purity).

Test atmosphere generation, measurement, and cell exposure

For each chemical, three independent exposure experiments were performed. Piperazine and TDI were both dosed in the gas phase as these chemicals have sufficient vapor pressure at room temperature to reach the target concentration. Vapor pressure of Chloramine-T at room temperature is very low, so it was dosed as an aerosol of a solution in water.

Piperazine

The highest piperazine dose to test was informed by Mori et al. (2012), where mice were skin sensitized and intratracheally challenged with, among other sensitizers, piperazine and TDI. For skin sensitization, piperazine was used at a dose 5 times higher than TDI, whereas for intratracheal challenge the piperazine dose was 1% that of TDI. Thus, we decided for piperazine to use as highest test dose the same as that for TDI, that is 100 mg/m3.

The intended dose of piperazine was different between experiment 1 (90, 3, and 0.09 mg/m3 for the high, middle, and low dose, respectively) and experiments 2 and 3 (90, 18, and 3 mg/m3 for the high, middle, and low dose, respectively). Based on the outcome of experiment 1, we were able to test a narrower dose range in experiments 2 and 3.

Test atmosphere generation

A gas washing bottle was filled with the solid (flakes) test material. To generate the test atmosphere the gas washing bottle was flushed with a mass flow controlled stream of compressed dry air. To reach the target concentration for the high dose (100 mg/m3) exposure group the stream of piperazine vapor in air was mixed with a (mass flow controlled) stream of compressed dry air. The gas washing bottle was held at a constant temperature by means of a water bath. For the first exposure, 20 °C was sufficient to reach the target concentration, for the following exposures 31 °C was needed, probably due to reduction of the surface of the piperazine crystals. The exhaust of the gas washing bottle and the T-piece in which the stream of piperazine vapor was mixed with compressed dry air were held at an elevated temperature (40–45 °C) by means of a water bath, to prevent condensation of piperazine. The concentrations of the middle dose (18 mg/m3) and low dose (3 mg/m3) exposure groups were generated by extracting a small mass flow controlled stream of the high and middle dose concentration, respectively, and mixing this with a stream of mass flow controlled compressed dry air. To prevent gradual declining of the generated concentration the gas washing bottle with the test material was shaken daily to redistribute the test material. Part of the stream of the high dose exposure test atmosphere was sampled by a Total Carbon Analyzer (TCA, Ratfisch RS55-T; Ratfisch, Poing, Germany). The experimental set-up is illustrated in Fig. 1.

Fig. 1

Experimental set-up of piperazine exposure and measurement

Test atmosphere measurement

The piperazine concentration was determined by weighing the gas washing bottle. As the weight decrease during the exposure period (2 h daily) was too small to be reliably measured, it was chosen to generate the test atmosphere continuously. The bottle was weighed before the start of the test atmosphere generation and daily after exposure (~ 24 h of generation time). The high dose was calculated as the weight loss divided by the total volume of air (through the gas washing bottle and the first dilution). The concentrations of the middle and low dose exposures were calculated using the calculated high and middle dose, respectively, and the dilution factor derived from the measured sample and dilution flows.

Cell exposure

From the stream of test atmosphere (high, middle or low dose, or clean air), 1 l/min was extracted to a manifold connected to a VITROCELL® Modular Exposure System (VITROCELL® Systems GmbH, Germany). The inserts were placed in the VITROCELL® module in insert holders (Institute’s design) which prevented humidification of test atmosphere by evaporation of basolateral medium. The VITROCELL® module was held at 37 °C by means of a water bath. The test atmosphere was extracted from the manifold with a mass flow controlled flow of 5 ml/min for each insert. The inserts were exposed to clean air (air control) or piperazine in clean air (exposure) for 2 h/day, for 3 consecutive days. Before the first exposure and after each of the three (daily) exposures, the basolateral medium was aliquoted. Part was stored at 4 °C and used for LDH measurements on the same day, the remainder was frozen at −80 °C for cytokine measurement. The inserts were placed in fresh, pre-warmed (37 °C) basolateral medium.

chloramine-T

The highest dose of chloramine-T to test was based on a 4-week inhalation exposure study, with the highest dose 4 mg/m3 chloramine-T (Shim et al. 2013). This corresponds to an inhaled volume of (4 weeks × 5 days × 6 h/day × 0.24 l/min Respiratory Minute Volume (rat)) 1728 l, or an inhaled dose of 7 mg. With an estimated deposition efficiency of 20% and a lung surface of 5000 cm2 this amounts to 280 ng/cm2. We thus chose to start with 300, 30, and 3 ng/cm2 as target deposited high, middle, and low dose, respectively.

We performed three experiments. The intended dose of chloramine-T was different between experiment 1 (300, 30, and 3 ng/cm2 for the high, middle, and low dose) and experiments 2 and 3 (3000, 300, and 30 ng/cm2 for the high, middle, and low dose). Based on the outcome of experiment 1, we decided to test a higher dose range in experiments 2 and 3.

In contrast to cell exposures to piperazine and TDI, exposure to chloramine-T was performed by deposition, using a “radial in vitro aerosol exposure system” (RIVAES; Melzi et al. 2024). The RIVAES is inspired by the design of the VITROCELL® Cloud (VITROCELL®, Waldkirch, Germany). The RIVAES was equipped with a Quartz Crystal Microbalance (QCM; VITROCELL®) and an Aeroneb® Lab nebulizer (4–6 μm pore size, Aerogen, Ireland). Details on the principle and setup of the VITROCELL® Cloud system, QCM, and nebulization process can be found in Ding et al. (2020).

The Cloud system was heated to 37 °C before use. QCM data acquisition was started and the QCM stability was measured for 1 min. For blank/sham correction, 200 μl ultrapure water spiked with 1% isotonic (i.e. 9 mg/mL) NaCl (90 μg/ml NaCl final concentration) was nebulized to check proper operation of the VITROCELL® Cloud12 system; QCM values should be < 250 μg/cm2. Chloramine-T was used in 3 concentrations: 23 (high dose), 2.3 (middle dose), and 0.23 (low dose) µg/ml. For each of these concentrations, chloramine-T was spiked with 1% isotonic NaCl. Two hundred μl of each test solution (3 concentrations of salt-spiked chloramine-T, or isotonic NaCl solution (air control)) was used for nebulization immediately after preparation. The inserts were exposed to chloramine-T or isotonic NaCl (air control). After app. 10 min the QCM signal dropped, followed by a stable horizontal line. Then, the cover of the chamber was lifted and the chamber and cover were cleaned using a cloth. Next, the cover was placed back to prevent ventilation and temperature effects that could influence the QCM signal. Data acquisition was stopped after an additional 3 min to assure a stable QCM signal. For data evaluation, the mean of the values recorded in the last 30 s was calculated. For cleaning, the reservoir of the nebulizer was rinsed with water prior to each usage. To check for proper cleaning, deposition of water with 1% isotonic NaCl was additionally measured; a QCM signal < 250 μg/cm2 indicated clean conditions. Medium was stored at 4 °C and used for LDH measurements on the same day, the remainder was frozen at − 80 °C for cytokine measurement. The inserts were placed in fresh, pre-warmed (37 °C) basolateral medium.

TDI

The highest dose of TDI to test was chosen to be saturated vapor (100 mg/m3 at 20 °C). We chose 100, 33, and 10 mg/m3 as target deposited high, middle, and low dose, respectively.

Test atmosphere generation

The test atmosphere was generated by dosing 0.55 µl/min TDI in a heated glass evaporator (Dimroth-type cooler) and mixing it with a mass flow controlled stream of 6.4 l/min compressed dry air. The TDI was dosed using a syringe pump (540,200; TSE Systems, Bad Homburg, Germany) and a glass Syringe (Hamilton gas tight model #1001). The evaporator was heated to 55 °C by means of a water bath (Fig. 2). In order to calculate the amount of TDI used, the syringe was weighed before the start and at the end of the exposure.

Fig. 2

Experimental set-up of TDI exposure and measurement

The resulting test atmosphere was then brought to a similar Dimroth-type cooler held at a temperature of 20 °C by means of a water bath. There, the excess TDI condensed, resulting in a saturated vapor of TDI (at 20 °C).

Part of the stream of test atmosphere was sampled by a Total Carbon Analyzer (TCA, Ratfisch RS55-T). Saturation of the test atmosphere was checked during preliminary experiments by raising the TDI flow and observing the change in response from the TCA. Based on these experiments the test atmosphere was found to be saturated at a TDI flow slightly below 0.55 µl/min.

From the stream of test atmosphere (clean air or clean air with TDI) 1 l/min was extracted to a manifold connected to a VITROCELL® Modular Exposure System. The inserts were placed in the VITROCELL® module in insert holders (Institute’s design) which prevented humidification of test atmosphere by evaporation of basolateral medium. The VITROCELL® module was held at 37 °C by means of a water bath. The test atmosphere was extracted from the manifold with a mass flow controlled flow of 5 ml/min for each insert.

Test atmosphere measurement

During and after generation of the test atmosphere the response of the analyzer was recorded. In this way, the steady-state concentration (during the exposure) as well as the decline in the concentration was measured. Recording of the response of the analyzer was ended when the response of the analyzer was close to zero and stable.

The concentration during exposure was calculated by using the nominal concentration of TDI during the total recording period divided by the average response of the analyzer during the same period. The nominal concentration of TDI is calculated by dividing the used mass of TDI (mg) by the total volume of air in which the TDI was evaporated. In these calculations the following assumptions are made: (1) all TDI dosed will evaporate (generation efficiency of 100%), and (2) the response of the analyzer is linear and unbiased.

Cell exposure

The inserts were exposed to clean air (air control) or TDI in clean air (exposure) for 2 h/day, for 3 consecutive days. Before the first exposure and after each of the three (daily) exposures, the basolateral medium was aliquoted. Part was stored at 4 °C and used for LDH measurements on the same day, the remainder was frozen at − 80 °C for cytokine measurement. The inserts were placed in fresh, pre-warmed (37 °C) basolateral medium.

Cell line

The Calu-3 cell line has been chosen because these cells form a tight monolayer with a high trans-epithelial electrical resistance (TEER) (He et al. 2021), they can be cultured at the ALI without a decrease in TEER for weeks (Braakhuis et al. 2020), and they are easy to handle. Human bronchial epithelial cells (Calu-3) were purchased from American Type Culture Collection (ATCC, Rockville, USA). Calu-3 cells were cultured in Minimal Essential Medium + GlutaMAX (GIBCO® 42,360–024), supplemented with 10% FCS (GIBCO® A5256801), 100 U/ml penicillin + 100 μg/ml streptomycin (GIBCO® 15,140), and 1% Non-Essential Amino Acids solution (GIBCO® 11,140). The cells were incubated at 37 °C in a 100% humidified atmosphere containing 5% CO2. The medium was renewed every 2–3 days. After thawing the purchased cells, they were passaged 2–25 times before use.

Preparation of the lung model

Calu-3 cells were seeded at a density of 100,000 cells/cm2 on Transwell® inserts (12-well inserts, 1.12 cm2, 0.4 μm pores; Corning CLS 3460). The cells were cultured under submerged conditions for 7 days. After 7 days, the apical medium was removed, and the cells were cultured at the ALI for an additional 7 days.

Measurements of transepithelial electrical resistance (TEER)

Transepithelial electrical resistance (TEER) was measured using the Evom2 Voltohmmeter equipped with a chopstick electrode (World Precision Instruments, Friedberg, Germany). TEER was measured the day before exposure and 24 h after exposure. As TEER can fluctuate depending on the temperature, measurements were performed within a 10 min period after taking the cells from the incubator. The TEER values were obtained by multiplying by the surface area of the insert.

Cytotoxicity

Lactate dehydrogenase (LDH) release was measured in the apical cell culture medium the day before exposure and 24 h after (the last) exposure, and in the basolateral cell culture medium immediately before (the first) exposure and immediately after (each) exposure, using an LDH kit (11,644,793,001, Roche Diagnostics GmbH, Mannheim, Germany). Hundred μl of sample was taken off and incubated with 100 μl of reaction mix for 20 min in the dark. After adding the stop solution (HCl), absorbance was measured using a spectrophotometer at a wavelength of 490 nm and a reference wavelength of 650 nm. All measurements were corrected for the background of the cell culture medium and normalized to the maximum LDH release after lysis of the control cells.

Mitochondrial metabolic activity

Cell mitochondrial activity was evaluated using the WST-1 cell Proliferation Reagent (11,644,807,001, Roche Diagnostics GmbH, Mannheim, Germany). At 24 h after exposure, WST-1 reagent (10 × diluted in medium) was added to the apical side of the inserts for 1 h. Subsequently, 100 μl of supernatant of each insert was transferred to a 96-wells plate in duplicate. Absorbance was measured using a spectrophotometer at a wavelength of 440 nm and a reference wavelength of 620 nm.

Cytokine measurementELISA

The day before exposure and 24 h after (the last) exposure, 500 μl cell culture medium was added to the apical side of the inserts. After 10 min, the supernatants were collected for cytokine analysis. The basolateral cell culture medium was harvested immediately before (the first) exposure and 24 h after (each) exposure, and analyzed for cytokine content. Release of interleukin (IL)−6 and IL-8 were measured in the supernatants using ELISA kits (Invitrogen, Thermo Fisher Scientific). Cytokine analysis was performed according to the manufacturer’s instructions. Absorbance was measured using a spectrophotometer at a wavelength of 450 nm and a reference wavelength of 570 nm.

Bead-based multiplex immunoassay

Based on the ELISA data, a subset of samples was selected for a screening of 24 cyto- and chemokines using a bead-based multiplex immunoassay. The Human Proinflammatory Chemokine Panel 1 and the Human Inflammation Panel 1 were used to assess these cyto- and chemokines (LEGENDplex™; BioLegend, Amsterdam, the Netherlands). Levels of the following 12 cyto- and chemokines were assessed: CCL2 (MCP-1), CCL5 (RANTES), CCL17 (TARC), CXCL8 (IL-8), CXCL9 (MIG), CXCL10 (IP-10), CXCL11 (I-TAC), IFN-α2, IFN-λ1, IFN-γ, IL-1β, and TSLP. Data were acquired using the FACS Canto II (Becton Dickinson) and generated data was interpolated against provided reference using the manufacturer’s cloud-based The LEGENDplex™ Data Analysis Software Suite.

Statistical analysis

The Independent Samples T-Test with Levene’s test for Equality of Variances was performed (SPSS, Chicago, IL).

Dose response curve fitting (benchmark dose modelling (BMD)) for IL-6 production was performed with statistical software package PROAST (Slob 2002) within the software environment ‘R’ (R 2019). BMD results were calculated using the data from all N = 3 experiments.