Skip to main content
NCBI home page
Toxicology Research logoLink to Toxicology Research
. 2017 May 23;6(4):448–459. doi: 10.1039/c7tx00109f

Regulating temperature and relative humidity in air–liquid interface in vitro systems eliminates cytotoxicity resulting from control air exposures

Jose Zavala a,✉, Rebecca Greenan b, Q Todd Krantz a, David M DeMarini a, Mark Higuchi a, M Ian Gilmour a, Paul A White b,✉
PMCID: PMC6062410  PMID: 30090513

graphic file with name c7tx00109f-ga.jpgModifications to a VITROCELL exposure system were required to mitigate cytotoxicity caused by the absence of temperature and humidity control.

Abstract

VITROCELL® systems permit cell exposures at the air–liquid interface (ALI); however, there are inconsistent methodologies in the literature for their operation. Some studies find that exposure to air (vehicle control) induced cytotoxicity relative to incubator controls; others do not mention if any cytotoxicity was encountered. We sought to test whether temperature and relative humidity (temp/RH) influence cytotoxicity with an unmodified (conditions A & B) and modified (condition C) VITROCELL® 6 CF with temp/RH controls to permit conditioning of the sampled air-flow. We exposed BEAS-2B cells for 1 h to air and measured viability (WST-1 cell proliferation assay) and lactate dehydrogenase (LDH) release 6 h post-exposure. Relative to controls, cells exposed to air at (A) 22 °C and 18% RH had a 47.9% ± 3.2% (p < 0.0001) reduction in cell viability and 10.7% ± 2.0% (p < 0.0001) increase in LDH release (B) 22 °C and 55% RH had a 40.3% ± 5.8% (p < 0.0001) reduction in cell viability and 2.6% ± 2.0% (p = 0.2056) increase in LDH release, or (C) 37 °C and >75% RH showed no changes in cell viability and no increase in LDH release. Furthermore, cells exposed to air at 37 °C and >75% RH 24 h post-exposure showed no changes in viability or LDH release relative to incubator controls. Thus, reductions in cell viability were induced under conditions used typically in the literature (conditions A & B). However, our modifications (condition C) overcome this shortcoming by preventing cell desiccation; regulating temp/RH is essential for conducting adequate ALI exposures.

Introduction

The use of in vitro methods for characterizing the relative toxicity of air–pollutant mixtures has become a necessity due to the high cost and technical complexity of in vivo methods. Moreover, several organizations (e.g., the US National Academy of Sciences, EURL-ECVAM, etc.) have called for reduction of animal use in toxicology studies.1,2 Typical in vitro exposure studies of air–pollutant mixtures involve, for example, the addition of particulate matter (PM) or PM extracts to cell culture medium, or the bubbling of gases into the culture medium. However, these methods can alter the pollutant's physical and chemical characteristics,3,4 which can lead to misrepresentation of the pollutant's toxicity.

Exposure of cultured mammalian cells at the air–liquid interface (ALI) is a more realistic air-to-cell exposure condition that can be reasonably regarded as an effective in vitro surrogate for inhalation. In this method, cells are cultured on a porous membrane that allows a direct air–cell interaction on the apical surface, while culture medium on the basolateral side provides the necessary nutrients for viability.5–7 Various ALI exposure systems have been developed,7–15 and in recent years in vitro exposures at ALI conditions have been widely accepted as the “gold standard” for in vitro studies of air–pollutant mixtures.

The VITROCELL® 6 CF is one of the few ALI exposure systems that is commercially available. This device relies on diffusion and sedimentation forces to deposit particles onto cells, whereas mixing of air-flow and diffusion of gas molecules permits the interaction of gaseous pollutants with the cells. The 6 CF model, along with similar models available from the same manufacturer, has been used primarily for studying the toxicological properties of cigarette smoke (CS) and its components.16–23 Exposures to CS and its components are typically conducted for ≤30 min. However, in recent years, VITROCELL® systems have also been adopted to expose cells to other test materials, such as gaseous pollutants,24–34 nanomaterials,35–41 and combustion emissions.42–45 These types of studies have employed exposures of up to 4 h.

Although the use of ALI exposure systems is increasing, there are no standardized guidelines outlining how they should be employed (i.e. flow rate, exposure time, pollutant concentrations). In addition, most published studies have focused on toxicological results generated with minimal descriptions of experimental setup, operating conditions, challenges encountered, and required system modifications, if any. With no existing guidelines regarding the critical information that should be reported, the available literature using VITROCELL® systems contains contradictory methodologies on the operating conditions required for successful cell exposures.

Table 1 summarizes studies that have used VITROCELL® exposure systems. Interestingly, some authors do not mention if any significant challenges were encountered and/or if any induction of cytotoxicity by exposure to control air (vehicle control) were observed. In contrast, several report significant reductions of cell viability after exposure to control air alone. However, even those that report cell viability reductions and other issues do so in a brief sentence or two, providing minimal details. Typically, data resulting from exposures to an air pollutant are normalized to cells exposed to air, and presenting data this way could mask any cytotoxicity caused by the exposure condition itself.

Table 1. Peer-reviewed studies using VITROCELL® systems for toxicological assessments of airborne pollutants.

References Air pollutant Cell type Exposure duration (minutes) Flow rate per well (mL min–1) Insert size (mm) Nozzle height (mm) RH of air sample (%) Temp of air sample (°C) Any issues with cell viability reported?
Ritter et al. 2001 (ref. 33) a NO2, O3 Lk004, HFBE-21 60–120 8.3 12 3 90 25 No mention if issues were encountered or not
Knebel et al. 2002 (ref. 43) a Diesel exhaust HFBE-21 60 8.3 12 NR NR NR No mention if issues were encountered or not
Pariselli et al. 2006 (ref. 31) a Toluene, benzene A549, HaCaT 60 2 24 NR NR NR 10% & 15% reduction in viability after air exposure for A549 and HaCaT cells, respectively, compared to incubator controls. Flow rate initially tested at 8 mL min–1, but produced too high cytotoxicity. A bubbler with water was introduced to improve RH, and the flow rate was reduced to 2 mL min–1 to minimized cell desiccation
Seagrave et al. 2007 (ref. 44) Diesel exhaust EpiAirway™ 180 8.3 12 NR NR NR No mention if issues were encountered or not
Anderson et al. 2010 (ref. 34) Dicarbonyls A549 120–240 3 24 5 50 Room temp Nozzle height was raised to 5 mm and flow reduced to minimize cell desiccation. There was a 14% reduction in viability after air exposure compared to controls
Koehler et al. 2010 (ref. 27) NO2 Primary nasal epithelium 30 5 12 NR NR NR No mention if issues were encountered or not
Persoz et al. 2010 (ref. 32) Formaldehyde A549 30 5 12 NR NR 22–25 Exposing cells for 30 min to air yielded no reduction in viability. However, a 60 min exposure yielded 30% reduction compared to controls; therefore all exposures were conducted for 30 min only
Switalla et al. 2010 (ref. 26) NO2, O3 Precision-cut lung slices 60–180 10 12 NR NR NR No mention if issues were encountered or not
Koehler et al. 2011 (ref. 30) NO2 Primary nasal epithelium 60–180 5 12 NR NR NR No mention if issues were encountered or not
Persoz et al. 2011 (ref. 25) Formaldehyde A549 30 5 12 NR NR NR Same as Persoz et al. 2010
Tang et al. 2012 (ref. 45) Laser printer emissions A549 60 5 24 NR NR NR Exposures longer than 60 min yielded high cytotoxicity
Xie et al. 2012 (ref. 41) Zinc oxide nanoparticles C10 10–20 10 24 NR NR NR A bubbler with water was introduced to improve RH. The entire system was housed in a heated incubation chamber at 32 °C to condition the flow. Cytotoxicity increased with increasing exposure time; therefore, only 20 min exposures were conducted
Persoz et al. 2012 (ref. 24) Formaldehyde A549, BEAS-2B 30 5 12 NR NR NR Same as Persoz et al. 2010
Kim et al. 2013 (ref. 38) Copper nanoparticles A549 240 5 24 2 NR NR Sampled air-flow passed through a dryer and condenser to remove RH from nebulizer used to generate nanoparticles. No viability issues were encountered
Frohlich et al. 2013 (ref. 36) Polystyrene nanoparticles, carbon nanotubes A549 60 1 & 5 24 NR NR 21–25 No mention if issues were encountered or not
Anderson et al. 2013 (ref. 28) Limonene, O3 A549, MucilAir™ 60–240 3 (A549) 2 (MucilAir) 24 5 50 NR Same as Anderson et al. 2010
Elihn et al. 2013 (ref. 37) Copper nanoparticles A549 240 20 24 NR NR NR No mention if issues were encountered or not
Klein et al. 2013 (ref. 39) SiO2-rhodamine nanoparticles 3D co-culture 30 5 24 NR NR NR No mention if issues were encountered or not
Bardet et al. 2014 (ref. 29) Formaldehyde hAECN (Epithelix) 60 2 12 NR NR NR Same as Persoz et al. 2010
Panas et al. 2014 (ref. 40) Silica nanoparticles A549 300–420 100 24 2 80–90 37.5 The sampled aerosol was humidified and warmed with a home-made humidifier. Due to long exposure times, 100 μL of HBSS was added apically to cells to minimize desiccation
Mülhopt et al. 2016 (ref. 48) Combustion aerosols A549 360 100 24 NR 85 37 Humidification is needed to prevent cell dehydration. Exposing cells to humidified air yields no loss of viability. Exposing cells to dry air leads to ∼75% loss of viability
Kooter et al. 2016 (ref. 35) cerium oxide particles A549, BEAS-2B, MucilAir™ 60 5 (A549 BEAS-2B) 24 NR 50 NR Air exposure induced significant increases in LDH and IL-8 release for BEAS-2B (+32% LDH and +73% IL8) and A549 (+133% LDH and +25% IL8) cells. No changes were found with MucilAir cells
1.5 (MucilAir) 6.5
Open in a new tab

aStudy used the Cultex system; the predecessor system to the VITROCELL® systems.

Our earlier work using an unmodified VITROCELL® 6 CF system, i.e., no regulation of temperature or relative humidity (RH) of the sampled air-flow, to expose A549 cells showed that a 1 h exposure to air resulted in a 40% reduction in cell viability, measured 24 h post-exposure, compared to incubator controls maintained at ALI.46 We deemed exposures under such conditions to be unsuitable for meaningful toxicological assessments of complex combustion-derived aerosols due to inadequate control of temperature and RH of the sampled air-flow.

The purpose of the present study was to investigate the adverse effects of air as a vehicle control in another commonly used human lung epithelial cell line (BEAS-2B) and to investigate if supplemental humidification and heating of the sampled air-flow would improve cell viability relative to that of incubator controls. We compare our results using our modified VITROCELL® system to those reported by others using unmodified VITROCELL® systems, and we make recommendations regarding the use of such systems for toxicological profiling of airborne contaminants.

Experimental

Exposure system

We used two identical glass modules of a VITROCELL® 6 CF system (VITROCELL Systems GmbH, Waldkirch, Germany) for our experiments. Each module has a glass base that accommodates 3 membrane inserts, and a heated water jacket surrounding the wells maintains the culture medium at 37 °C. We pushed a total flow of 1.0 L min–1 through the glass distribution manifolds of the system using positive pressure and, for all exposures, pulled at a flow rate of 5 mL per min per well using a vacuum. Typical exposure flow rates ranging from 2–10 mL per min per well have been used previously with similar systems. Based on the existing literature (Table 1), users have determined that the higher the nozzle height (distance between cells and nozzle), the less cytotoxic effects are observed. The highest reported nozzle height used is 5 mm, and significant cytotoxicity was still observed. In this study we selected the nozzle height of 1 mm to examine if temperature and RH have more influence on cytotoxicity than does nozzle height.

Modifying the VITROCELL® 6 CF

As provided by the manufacturer, the system lacks the ability to condition the sampled air-flow to achieve the temperature (37 °C) and RH (>70%) compatible with survival and maintenance of cultured mammalian cells. Temperature and humidity controls were added to provide capability to increase the temperature and RH of the sampled air-flow to incubator-like conditions. We constructed a temperature-regulated Plexiglas enclosure to contain two exposure modules, a humidification system, and sample lines; two small doors provided access to the interior of the enclosure. Heating pads underneath each exposure module, a heater at the rear of the box, and a fan permitted effective temperature regulation. We also placed a custom-designed in-line diffusion humidifier upstream of each exposure module, and we used digital sensors to measure temperature and RH of the sampled air-flow during each exposure.

Air as a vehicle control

Clean, filtered air has been used as a vehicle control in all the studies summarized in Table 1, as well as in studies using other ALI exposure systems. Typically, cells need to be exposed to air either simultaneously or before an exposure to an air pollutant in order to demonstrate that the vehicle control, operating conditions, and/or the device itself does not induce any adverse effects on the cells.

To assess if any adverse effects are induced by exposure to control air, we replicated similar operating and exposure conditions to those commonly found in the published literature. Here we tested three different experimental conditions of the sampled air-flow by varying its temperature and RH to explore the effects on the cells (Table 2). Conditions A and B, using an unmodified system, have been the most common conditions used in previously published studies where low and medium RH levels of the sampled air-flow were used at room temperature. In condition A, we targeted a low RH level by pushing the air directly from our clean air source at room temperature into the glass distribution manifold (Fig. 1A). In condition B, we targeted a medium RH level by bubbling the air through a three-neck flask containing 80 mL of MilliQ water at room temperature before the air entered the glass distribution manifold (Fig. 1B). For condition C, we used our modified system to target the experimental conditions of the sampled air-flow that we deemed optimal for cell viability: RH levels >75% at a temperature of 37 °C (Fig. 1C). Effects of each condition was measured 6 h post-exposure for conditions A–C and at 24 h post-exposure for condition C only.

Table 2. Experimental conditions tested using air as a vehicle control.

  Water bath temp (°C) Sample air temp (°C) Sample air relative humidity (%)
Condition A 37 22 18
Condition B 37 22 55
Condition C 37 37 >75
Open in a new tab

Fig. 1. Experimental setup of conditions A, B, and C. (A) For condition A, the VITROCELL® 6 CF module was mounted on a cart as specified by the manufacturer with no added temp/RH of the air-flow. (B) For condition B, we used a 3-neck flask filled with 80 mL of MilliQ water as a bubbler to increase the RH of the air-flow. (C) For condition C, we modified the VITROCELL® 6 CF system by adding a Plexiglass enclosure, a heating pad, a heater, a fan, a diffusion humidifier, and temp/RH sensors.

Fig. 1

Open in a new tab

Lung cells

We obtained virus-transformed normal human bronchial cells (BEAS-2B) from the American Type Culture Collection (ATCC® CRL-9609™, Manassas, VA) at passage 38, and they were used within 10 passages of receipt. We cultivated cells in Bronchial Epithelial Basal Medium (BEBM) supplemented with the BEGM SingleQuot Kit (Lonza, Walkersville, MD). As recommended by ATCC, we added 0.1% penicillin–streptomycin (Life Technologies, Grand Island, NY) in place of the gentamycin–amphotericin B that was included with the BEGM SingleQuot Kit.

Cell exposures

For all exposures, we used 24 mm Transwell inserts (Corning Life Sciences, Tewksbury, MA) collagen-coated with 10 μg cm–2 of PureCol® (Bovine collagen solution, type I, Advanced BioMatrix, San Diego, CA). Cells were seeded 24 h prior to exposure at a density of 4 × 105 cells per insert with 2.6 mL of medium on the basolateral side and 1 mL on the apical side. We removed the apical medium 30 min prior to exposure and then washed the cells twice with Dulbecco's Phosphate-Buffered Saline (DPBS, Life Technologies). After replacing basolateral medium with fresh medium, we transported the cells from the cell culture laboratory to the exposure facility in an adjoining building.

Each well in the exposure modules contained 9 mL of medium supplemented with 20 mM HEPES buffer [4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, Life Technologies] to maintain pH in the absence of 5% CO2, as reported previously.40,46 We selected a volume of 9 mL in each well because this is enough to surround the Transwell membrane on the outside wall while minimizing the hydrostatic pressure pushing up on the bottom surface of the membrane. After a 1 h exposure, we placed the Transwell inserts into 6-well plates with fresh medium on the basolateral side only. The plates were then incubated for an additional 6 or 24 h. We conducted 4 replicate runs with 3 technical replicates for both exposed and incubator control cells at each air exposure condition tested at the 6 h time point and 6 replicate runs at the 24 h time point.

Markers of toxicity

Cell viability

We used the water-soluble tetrazolium-1 (WST-1) cell-proliferation assay (Roche Applied Science, Indianapolis, IN) to measure cell viability at 6 or 24 h post-exposure. We added 600 μL of a 2% WST-1 reagent solution in medium directly to the apical side of the Transwells and incubated for 30 min at 37 °C. We then transferred 150 μL of the WST-1 samples in triplicate to a 96-well plate for colorimetric analysis (SpectraMax i3, Molecular Devices, Sunnyvale, CA). Absorbance was measured at 435 nm with reference at 620 nm.

Cytotoxicity

We measured the amount of lactate dehydrogenase (LDH) release into the basolateral medium as an indicator of cytotoxicity at 6 or 24 h post-exposure. We lysed cells from an unexposed insert with 2.6 mL of 1× Lysis Buffer (same volume used for basolateral medium) included in the assay kit (Pierce LDH Cytotoxicity Assay kit, Thermo Fisher Scientific, Rockford, IL) as a measure of maximal (100%) LDH content. The Lysis Buffer is provided in a 10× solution; therefore, it was diluted to 1× in culture medium. The 1× Lysis Buffer was added apically and the insert was incubated at 37 °C for 30 minutes. Using medium as a baseline (0%) measurement, we analyzed 3, 50 μl aliquots from each insert using an LDH cytotoxicity kit that we modified and adapted for use on a Konelab Arena 30 clinical chemistry analyzer (Thermo Clinical Labsystems, Espoo, Finland).

Statistical analysis

We present data as the mean ± standard error of the mean (SEM). We analyzed the data using the two-tailed unpaired t-test, with the threshold for statistical significance set at 0.05.

Results

Air-exposed cells vs. incubator controls

We assessed the effects of air as a vehicle control on the cells 6 h post-exposure under three different experimental conditions for the sampled air-flow tested (Table 2).

Viability 6 h post-exposure

Cells were exposed under condition A where the air-flow was controlled at 22 °C and 18% RH. Results show that cell viability was reduced significantly by 47.9% ± 3.2% (p < 0.0001) under this condition when compared to incubator controls (Fig. 2A). The air-flow in condition B was controlled at 22 °C and 55% RH during exposures and a significant reduction in viability of 40.3% ± 5.8% (p < 0.0001) was observed when compared to incubator controls (Fig. 2B). For condition C, the modified system was used, and the sampled air-flow was conditioned and controlled upstream of the exposure modules to 37 °C and >75% RH. No significant reductions in viability (4.8% ± 3.2%; p = 0.14) were observed under this condition (Fig. 2C).

Fig. 2. Viability of BEAS-2B cells exposed to vehicle air at ALI (n = 4) measured 6 h post-exposure. (A) Cells exposed to air experienced a significant reduction in viability (p < 0.001) compared to incubator controls. (B) Cells exposed to air experienced a significant reduction in viability (p < 0.001) compared to incubator controls. (C) No change in viability was observed (p = 0.14) between condition C and incubator controls. Bars represent average percent viability ± SEM (two-tailed unpaired t-test compared with incubator controls; * denotes significant difference).

Fig. 2

Open in a new tab

LDH release 6 h post-exposure

Similarly, the release of LDH protein into the basolateral medium, which is an indicator of cytotoxicity, was increased significantly (10.7% ± 2.0%; p < 0.0001) in the air-exposed cells under condition A when compared to incubator controls (Fig. 3A). Under condition B, a small increase in LDH release was observed (2.6% ± 2.0%; p = 0.21), but it was not statistically significant when compared to incubator controls (Fig. 3B). For condition C, a small decrease in LDH release was observed (0.7% ± 0.3%; p = 0.02) when compared to incubator controls; however, levels for both controls and exposed samples were under 1.5% (Fig. 3C).

Fig. 3. LDH release by BEAS-2B cells exposed to vehicle air at ALI (n = 3) measured 6 h post-exposure. (A) Cells exposed to air experienced a significant increase in LDH release (p < 0.001) compared to incubator controls. (B) Cells exposed to air experienced a non-significant increase in LDH release (p = 0.21) compared to incubator controls. (C) Cells exposed to air experienced a small decrease in LDH release (p = 0.02) compared to incubator controls. Bars represent average percent LDH release ± SEM (two-tailed unpaired t-test compared with incubator controls; * denotes significant difference).

Fig. 3

Open in a new tab

24. h post-exposure

Because we demonstrated that exposures under condition C were optimal, we also evaluated the effects of air on the cells 24 h post-exposure for this condition. We found no reduction in viability (4.8% ± 3.0%; p = 0.11) in cells exposed to air relative to incubator controls (Fig. 4). We found no changes in LDH release into the basolateral medium (0.4% ± 0.7%; p = 0.61) by the air-exposed cells compared to the incubator control cells (Fig. 5).

Fig. 4. Viability of BEAS-2B cells exposed to vehicle air by ALI (n = 6) measured 24 h post-exposure for condition C. No reduction in viability was observed (p = 0.1098) compared to incubator controls. Bars represent average percent viability ± SEM (two-tailed unpaired t-test compared with incubator controls).

Fig. 4

Open in a new tab
Fig. 5. LDH release by BEAS-2B cells exposed to vehicle air at ALI (n = 6) measured 24 h post-exposure for condition C. Cells exposed to air experienced no change in LDH release (p = 0.61) compared to incubator controls. Bars represent average percent LDH release ± SEM (two-tailed unpaired t-test compared with incubator controls).

Fig. 5

Open in a new tab

Modifications to system

During exposures under conditions A and B, excessive condensation was observed at the outlet of each well because the air-flow picks up moisture from the cell culture medium that is maintained at 37 °C as it exits the well. However, because the outlet of the well was at room temperature, the moisture condensed on the surface walls of the outlet and tubing, which are in a vertical position. Excessive condensation could lead to water droplets dripping down and potentially blocking the air flow at the outlet. Enclosing all components in the heated enclosure as described above for condition C prevented any condensation in the sample lines from occurring. These enhancements allowed cell cultures to be maintained and exposed in an ideal environment, similar to that of a cell culture incubator. Collectively, these biological measurements summarized above indicated that our enhancements to the exposure system effectively eliminated the cytotoxic effects of vehicle air that was observed previously.

Discussion

Unmodified VITROCELL® 6 CF

VITROCELL® systems have been used to conduct ALI exposures to a variety of airborne pollutants and materials. However, peer-reviewed studies using such systems provide a wide range of operating conditions and parameters that in many cases produce cytotoxic effects, making it difficult to understand what the critical conditions are for an exposure environment suitable for maintaining viable mammalian cells exposed to air. We observed previously significant reductions in viability after 1 h exposures to air; thus in our experience, the unmodified VITROCELL® 6 CF system was not suitable for conducting our desired exposures because the exposure condition itself compromised the validity of the responses elicited by the air–pollutant mixtures. Specifically, exposure to air alone was cytotoxic relative to incubator controls.

Based on the studies described in Table 1, there is clear evidence that the air being used as a vehicle control is in fact inducing significant adverse effects under the conditions tested relative to incubator controls. Several studies, however, make no mention of any challenges encountered using these exposure systems, even in the absence of conditioning the sampled air-flow to increase its temperature and/or RH. It is important to note that not mentioning any challenges with the system like we have experienced does not necessarily mean that they were not encountered. To mitigate challenges associated with air exposures, researchers have increased the distance between the inlet nozzle and the cell surface and/or reduced the flow rate through each well. Additionally, some studies attempted to introduce moisture into the sampled air-flow, whereas others just restricted exposure times to those that showed no or minimal reductions in viability.

We postulated that temperature and humidity played a key role in maintaining optimal cell viability; therefore, we selected the nozzle height of 1 mm to demonstrate that temperature and RH influence cytotoxicity more than does the nozzle height itself or exposure time to a certain degree. Thus, temperature and RH need to be regulated in an effective manner. In an effort to assess if any adverse effects were induced by exposure to vehicle air, we replicated similar operating and exposure conditions (i.e., conditions A and B) to those commonly found in the published literature. Exposing BEAS-2B cells under these conditions resulted in large reductions of viability (i.e., WST-1 assay) and induction of cytotoxicity (i.e., LDH assay) that are unacceptable for a vehicle control. The magnitude of these measurements may vary depending on the cell model used; therefore, it is critical to demonstrate that the operating conditions of any ALI exposure system do not adversely affect the cells being tested. Failure to do so can result in adverse effects to the control cells, compromising the results and making them unsuitable for meaningful toxicological assessments.

Modified VITROCELL® 6 CF

After making substantial modifications to the system, we observed no adverse effects in cells exposed to air for 1 h (condition C). The required modifications involved conditioning the air-flow upstream of the exposure modules by increasing the temperature and RH to incubator-like conditions. Air-flow conditioning minimized adverse effects due to the exposure condition itself, providing suitable experimental controls and, moreover, maximizing the dynamic range of the endpoints examined. Similar air-flow conditioning is provided by other in vitro ALI exposure systems, such as the Gillings Sampler,7 the EAVES,14 and the NACIVT system.47 The inventors of these in-house devices, which include in-line humidifiers and temperature control, indicate that conditioning of the sampled air-flow is an essential system component necessary to establish a suitable exposure environment that permits realistic ALI toxicity assessments of air–pollutant mixtures. More recently, an automated ALI exposure system integrated with three VITROCELL® 6/6 CF Stainless modules (similar size and dimensions as the glass modules) was developed where internal controls are exposed to air at 37 °C and 85% RH because not doing so will lead to cell dehydration.48 In that system, A549 cells exposed to 4 h of humidified air showed no loss of viability when compared to controls, whereas cells exposed to 4 h of dry air showed a ∼75% reduction in viability when compared to controls. These observations agree with our results. In humans, inhaled air is moistened and warmed as it enters the nasal passages and continues through the trachea and distal respiratory airways;49 thus, conditioning the sampled air-flow should be a critical component of ALI exposure systems in order to mimic in vivo conditions.

RH is highly dependent on temperature. When working at 37 °C and with RH > 75%, small fluctuations in temperature can lead to major fluctuations in RH, as observed in a psychrometric chart. As mentioned above and in Table 1, some users have targeted RH values of 80–90%. Although working under high RH conditions (>85%) might be seen as optimal, it can become unfavorable because a small decrease in temperature can lead to condensation in the sample lines or any air-flow path. As described in the results above, we observed condensation in the outlet tubing of the system under conditions A and B. This occurred because the air-flow picked up moisture from the cell culture medium, which is maintained at 37 °C, as it exits the well, whereas the outlet tubing was at room temperature (22 °C). Calculating the dew point, which is the temperature at which water in the air condenses, is an easy way to understand how critical temperature control is. For example, if the air-flow is at 37 °C and 90% RH, the dew point will be 35 °C; a 2 °C decrease on any surface will cause condensation. If the air-flow is at 37 °C and 75% RH, the dew point will be 32 °C; a 5 °C decrease on any surface will cause condensation. Thus, working under RH levels of 75–80% is more practical.

Our modifications do not alter the principle of operation of the system, which relies on diffusion and sedimentation for particle deposition. At the typical flows used with these systems (2–10 ml per min per well), aerosol physics dictates that diffusion will be the dominant deposition force for particles smaller than 100 nm, whereas sedimentation will be the dominant force for particles larger than 1 μm. Using similar exposure modules, studies have reported deposition efficiencies of less than 2% for particles ranging from 50 to 500 nm,6,50 whereas efficiencies for a 1 μm particle are <4%.51 In addition, it has been shown that the particle deposition efficiency increases as the nozzle height decreases from 4 to 0.5 mm.51 With our modification, the system can be operated at a nozzle height of 1 mm, thus increasing the expected deposition efficiency for PM exposures. Additionally, longer exposure periods (multiple hours) could be employed to ensure sufficient particle dosing without inducing cytotoxicity due to the length of the exposure time. More testing is needed to determine the maximum exposure duration without inducing cytotoxic effects.

Although some studies briefly mentioned difficulties in maintaining cell viability after air exposures (Table 1) with a VITROCELL® system, only one study included a description of the adverse effects of air without humidification. This is surprising because these systems have been used in a wide variety of applications with various cell types and air pollutants. Only one other study52 using a CULTEX® RFS system (a similar commercial system) provided a detailed description of adverse cellular effects (i.e., morphological changes and reduction in viability) following exposures to air for 30 min or less with no added heating or humidification. In that study, changes to the exposure protocol only (i.e., cell handling, type of membrane insert, procedure for removing inserts from system, etc.) were employed to improve cell integrity and viability.

Based on our previous and current experiences with the VITROCELL® 6 CF system, we conclude that it is necessary to condition the air-flow by increasing its temperature and RH such that it constitutes a suitable environment for BEAS-2B cells. As indicated in Table 1, most published studies that reported reduced cell viability after air exposure used A549 or BEAS-2B cells. The studies that did not report problems with cell viability after air exposure included those using fibroblast cells, human bronchial epithelial cells (HFBE-21), or primary multicellular constructs such as EpiAirway™ and MucilAir™. EpiAirway™ constructs have been shown to be more resistant to toxicological effects because they contain ciliated and mucus-secreting goblet cells and basal cells that likely reduce contact between the target cells and the air pollutant.53 The same toxicological resistance can be assumed for the MucilAir™ constructs.

Several studies listed in Table 1 in which A549 cells were exposed for 1–4 h without added humidification or increased temperature did not mention any reduction in cell viability after air exposure. Similarly, a study in which A549 cells were exposed for 1 h to air at 22 °C and <20% RH using the analogous ExpoCube system reported an average reduction in viability of only 3% immediately after exposure.54 These studies contradict our observations as well as those reported by others (Table 1), and it is difficult to surmise the reason(s) underlying the lack of significant cytotoxicity in the absence of temperature and RH adjustment of the sampled air-flow. A possible explanation could be that the total volume of medium used in the basolateral side of the Transwell inserts was excessive, causing the height of the medium on the outside of the insert to be much higher than that of the Transwell membrane, subjecting the membrane to high hydrostatic pressure. If high hydrostatic pressure exists, the medium would push up on the porous membrane, causing some of the medium to penetrate into the apical side and provide a thin liquid layer protecting the cells; thus there would no longer be an ALI exposure. Another explanation could be that some of the cytotoxicity/viability assays used were more sensitive than others – as we also observed in this study.

Although our viability measurements were robust, the LDH release measurements may be underestimating the actual cytotoxicity. A combination of two possible scenarios could explain the differences in sensitivity. For the LDH assay, any LDH released into the basolateral medium during the exposure was not captured in our measurements because the Transwell inserts were transferred into fresh medium after the exposure was completed. A second possibility is that a 6 h post-exposure time point might not have been long enough for any LDH that is released apically to diffuse down through the membrane and into the basolateral medium.

Conclusions

With the recent increase in ALI exposure studies being conducted in the absence of standardized exposure procedures and parameters, it is critical for the scientific community to establish some general guidelines regarding ALI exposure protocols. These can be separated into two categories: (1) biological conditions and (2) engineering conditions. Currently, most published studies adequately describe the biological conditions: cell type, passage number(s), type of culture medium and volumes, seeding density, growth time, type of membrane insert, exposure times, post-exposure recovery times, and biological endpoints/assays. In contrast, some of the critical engineering conditions are not reported, and these are essential for others to be able to reproduce the work. The engineering conditions that should be reported when using ALI exposure systems include total flow rate, flow per well (if applicable), temperature and RH of the sampled air-flow, nozzle height (if applicable), electric field strength (if applicable), thermal gradient (if applicable), how the air pollutant was generated (i.e., ozone generator, diesel generation, Dynacalibrator® for use of permeation tubes, etc.), concentration of the air pollutant at the generation source, concentration entering the exposure system (chemical compounds or PM can be altered with increased RH or loss to surface walls of tubing), air–pollutant properties (i.e., particle size, particle count), and how the air pollutant was monitored.

Further studies are needed to evaluate other important parameters described in the suggested guidelines above with the VITROCELL® systems. Flow rate, for example, is a critical factor that varies from study to study. Changes to the flow rate alter the aerosol dynamics as an aerosol moves through the exposure chamber, and ultimately the deposition rate, thus limiting the assessment of similar studies that used different flow rates. The format of the membrane insert (i.e. 24-well vs. 6-well) is another variable that should be considered. Typically, the exposure systems with smaller inserts are operated under lower flow rates, whereas the larger inserts use higher flow rates; again, the aerosol dynamics, for example, may differ. These factors could compromise comparisons between experiments that used different sized inserts.

ALI in vitro exposures for characterizing the relative toxicity of air–pollutant mixtures are becoming widely accepted as an effective in vitro surrogate for inhalation studies. As our results and an overview of the literature indicate, ALI systems can have serious limitations, and the lack of guidelines for describing the systems and reporting the results prevent a careful evaluation of the state-of-the-art. We encourage the inclusion of both the biological and engineering conditions described above in future ALI studies so that more informed conclusions can be made regarding the effectiveness of various ALI exposure systems for reliable toxicological profiling of complex aerosols and air pollutant mixtures.

Conflict of interest

There are no conflicts of interest to declare.

Acknowledgments

This study was funded in part by an Oak Ridge Institute for Science and Education (ORISE) postdoctoral fellowship, the Government of Canada Clean Air Regulatory Agenda (CARA) research fund and Program of Energy Research and Development (PERD), and the intramural research program at the U.S. EPA. We would like to thank Judy Richards for all LDH measurements and Nancy Hanley for technical support. This manuscript was reviewed by Health Canada and the National Health and Environmental Effects Research Laboratory of the U. S. Environmental Protection Agency and approved for publication. Approval does not signify that the contents reflect the views of any agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.

References

  1. National Research Council, Toxicity testing in the 21st century: A vision and a strategy, National Academies Press, 2007. [Google Scholar]
  2. Stephens M. L., Goldberg A. M. and Rowan A. N., The first forty years of the alternatives approach: Refining, reducing, and replacing the use of laboratory animals, 2001. [Google Scholar]
  3. Bakand S., Hayes A. J. Pharmacol. Toxicol. Methods. 2010;61:76–85. doi: 10.1016/j.vascn.2010.01.010. [DOI] [PubMed] [Google Scholar]
  4. Lichtveld K. M., Ebersviller S. M., Sexton K. G., Vizuete W., Jaspers I., Jeffries H. E. Environ. Sci. Technol. 2012;46:9062–9070. doi: 10.1021/es301431s. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Dvorak A., Tilley A., Shaykhiev R., Wang R., Crystal R. Am. J. Respir. Cell Mol. Biol. 2011;44:465. doi: 10.1165/rcmb.2009-0453OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Paur H. R., Cassee F. R., Teeguarden J., Fissan H., Diabate S., Aufderheide M., Kreyling W. G., Hänninen O., Kasper G., Riediker M., Rothen-Rutishauser B., Schmid O. J. Aerosol Sci. 2011;42:668–692. [Google Scholar]
  7. Zavala J., Lichtveld K., Ebersviller S., Carson J. L., Walters G. W., Jaspers I., Jeffries H. E., Sexton K. G., Vizuete W. Chem.-Biol. Interact. 2014;220:158–168. doi: 10.1016/j.cbi.2014.06.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Aufderheide M., Mohr U. Exp. Toxicol. Pathol. 2000;52:265–270. doi: 10.1016/S0940-2993(00)80044-5. [DOI] [PubMed] [Google Scholar]
  9. Aufderheide M., Scheffler S., Möhle N., Halter B., Hochrainer D. Anal. Bioanal. Chem. 2011;401:3213–3220. doi: 10.1007/s00216-011-5163-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. de Bruijne K., Ebersviller S., Sexton K. G., Lake S., Leith D., Goodman R., Jetters J., Walters G. W., Doyle-Eisele M., Woodside R., Jeffries H. E., Jaspers I. Inhalation Toxicol. 2009;21:91–101. doi: 10.1080/08958370802166035. [DOI] [PubMed] [Google Scholar]
  11. Lenz A. G., Karg E., Lentner B., Dittrich V., Brandenberger C., Rothen-Rutishauser B., Schulz H., Ferron G. A., Schmid O. Part. Fibre Toxicol. 2009;6:32. doi: 10.1186/1743-8977-6-32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Phillips J., Kluss B., Richter A., Massey E. Altern. Lab. Anim. 2005;33:239–248. doi: 10.1177/026119290503300310. [DOI] [PubMed] [Google Scholar]
  13. Savi M., Kalberer M., Lang D., Ryser M., Fierz M., Gaschen A., RicÃåka J., Geiser M. Environ. Sci. Technol. 2008;42:5667–5674. doi: 10.1021/es703075q. [DOI] [PubMed] [Google Scholar]
  14. Volckens J., Dailey L., Walters G., Devlin R. B. Environ. Sci. Technol. 2009;43:4595–4599. doi: 10.1021/es900698a. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Stevens J., Zahardis J., MacPherson M., Mossman B., Petrucci G. Toxicol. in Vitro. 2008;22:1768–1774. doi: 10.1016/j.tiv.2008.05.013. [DOI] [PubMed] [Google Scholar]
  16. Aufderheide M., Knebel J. W., Ritter D. Exp. Toxicol. Pathol. 2003;55:51–57. doi: 10.1078/0940-2993-00298. [DOI] [PubMed] [Google Scholar]
  17. Aufderheide M., Ritter D., Knebel J. W., Scherer G. Exp. Toxicol. Pathol. 2001;53:141–152. doi: 10.1078/0940-2993-00187. [DOI] [PubMed] [Google Scholar]
  18. Li X., Shang P., Peng B., Nie C., Zhao L., Liu H., Xie J. Food Chem. Toxicol. 2012;50:545–551. doi: 10.1016/j.fct.2011.12.008. [DOI] [PubMed] [Google Scholar]
  19. Ritter D., Knebel J. W., Aufderheide M. Inhalation Toxicol. 2003;15:67–84. doi: 10.1080/08958370304449. [DOI] [PubMed] [Google Scholar]
  20. Schlage W. K., Iskandar A. R., Kostadinova R., Xiang Y., Sewer A., Majeed S., Kuehn D., Frentzel S., Talikka M., Geertz M., Mathis C., Ivanov N., Hoeng J., Peitsch M. C. Toxicol. Mech. Methods. 2014;24:470–487. doi: 10.3109/15376516.2014.943441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Thorne D., Kilford J., Payne R., Haswell L., Dalrymple A., Meredith C., Dillon D. BMC Res. Notes. 2014;7:367. doi: 10.1186/1756-0500-7-367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Weber S., Hebestreit M., Wilms T., Conroy L. L., Rodrigo G. Toxicol. in Vitro. 2013;27:1987–1991. doi: 10.1016/j.tiv.2013.06.016. [DOI] [PubMed] [Google Scholar]
  23. Wolz L., Krause G., Scherer G., Aufderheide M., Mohr U. Food Chem. Toxicol. 2002;40:845–850. doi: 10.1016/s0278-6915(02)00034-0. [DOI] [PubMed] [Google Scholar]
  24. Persoz C., Achard S., Momas I., Seta N. Toxicol. Lett. 2012;211:159–163. doi: 10.1016/j.toxlet.2012.03.799. [DOI] [PubMed] [Google Scholar]
  25. Persoz C., Leleu C., Achard S., Fasseu M., Menotti J., Meneceur P., Momas I., Derouin F., Seta N. Toxicol. Lett. 2011;207:53–59. doi: 10.1016/j.toxlet.2011.07.028. [DOI] [PubMed] [Google Scholar]
  26. Switalla S., Knebel J., Ritter D., Krug N., Braun A., Sewald K. Toxicol. Lett. 2010;196:117–124. doi: 10.1016/j.toxlet.2010.04.004. [DOI] [PubMed] [Google Scholar]
  27. Koehler C., Ginzkey C., Friehs G., Hackenberg S., Froelich K., Scherzed A., Burghartz M., Kessler M., Kleinsasser N. Toxicol. Appl. Pharmacol. 2010;245:219–225. doi: 10.1016/j.taap.2010.03.003. [DOI] [PubMed] [Google Scholar]
  28. Anderson S. E., Khurshid S. S., Meade B. J., Lukomska E., Wells J. Toxicol. in Vitro. 2013;27:721–730. doi: 10.1016/j.tiv.2012.11.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Bardet G., Achard S., Loret T., Desauziers V., Momas I., Seta N. Toxicol. Lett. 2014;229:144–149. doi: 10.1016/j.toxlet.2014.05.023. [DOI] [PubMed] [Google Scholar]
  30. Koehler C., Ginzkey C., Friehs G., Hackenberg S., Froelich K., Scherzed A., Burghartz M., Kessler M., Kleinsasser N. Toxicol. Lett. 2011;207:89–95. doi: 10.1016/j.toxlet.2011.08.004. [DOI] [PubMed] [Google Scholar]
  31. Pariselli F., Sacco M. G. and Rembges D., Dynamic in vitro Exposure of Human Derived Cells to Indoor Priority Pollutants, European Commission, Directorate-General, Joint Research Centre, 2006. [Google Scholar]
  32. Persoz C., Achard S., Leleu C., Momas I., Seta N. Toxicol. Lett. 2010;195:99–105. doi: 10.1016/j.toxlet.2010.03.003. [DOI] [PubMed] [Google Scholar]
  33. Ritter D., Knebel J. W., Aufderheide M. Exp. Toxicol. Pathol. 2001;53:373–386. doi: 10.1078/0940-2993-00204. [DOI] [PubMed] [Google Scholar]
  34. Anderson S. E., Jackson L. G., Franko J., Wells J. Toxicol. Sci. 2010;115:453–461. doi: 10.1093/toxsci/kfq067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Kooter I. M., Gröllers-Mulderij M., Steenhof M., Duistermaat E., van Acker F. A., Staal Y. C., Tromp P. C., Schoen E., Kuper C. F., van Someren E. Appl. In Vitro Toxicol. 2016;2:56–66. [Google Scholar]
  36. Frohlich E., Bonstingl G., Hofler A., Meindl C., Leitinger G., Pieber T. R., Roblegg E. Toxicol. in Vitro. 2013;27:409–417. doi: 10.1016/j.tiv.2012.08.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Elihn K., Cronholm P., Karlsson H. L., Midander K., Odnevall Wallinder I., Moller L. J. Aerosol Med. Pulm. Drug Delivery. 2013;26:84–93. doi: 10.1089/jamp.2012.0972. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Kim J. S., Peters T. M., O'Shaughnessy P. T., Adamcakova-Dodd A., Thorne P. S. Toxicol. in Vitro. 2013;27:164–173. doi: 10.1016/j.tiv.2012.08.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Klein S. G., Serchi T., Hoffmann L., Blomeke B., Gutleb A. C. Part. Fibre Toxicol. 2013;10:31. doi: 10.1186/1743-8977-10-31. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Panas A., Comouth A., Saathoff H., Leisner T., Al-Rawi M., Simon M., Seemann G., Dossel O., Mulhopt S., Paur H. R., Fritsch-Decker S., Weiss C., Diabate S. Beilstein J. Nanotechnol. 2014;5:1590–1602. doi: 10.3762/bjnano.5.171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Xie Y., Williams N. G., Tolic A., Chrisler W. B., Teeguarden J. G., Maddux B. L., Pounds J. G., Laskin A., Orr G. Toxicol. Sci. 2012;125:450–461. doi: 10.1093/toxsci/kfr251. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Diabate S., Mulhopt S., Paur H. R., Krug H. F. ATLA, Altern. Lab. Anim. 2008;36:285–298. doi: 10.1177/026119290803600306. [DOI] [PubMed] [Google Scholar]
  43. Knebel J. W., Ritter D., Aufderheide M. Toxicol. in Vitro. 2002;16:185–192. doi: 10.1016/s0887-2333(01)00110-2. [DOI] [PubMed] [Google Scholar]
  44. Seagrave J., Dunaway S., McDonald J. D., Mauderly J. L., Hayden P., Stidley C. Exp. Lung Res. 2007;33:27–51. doi: 10.1080/01902140601113088. [DOI] [PubMed] [Google Scholar]
  45. Tang T., Gminski R., Konczol M., Modest C., Armbruster B., Mersch-Sundermann V. Environ. Mol. Mutagen. 2012;53:125–135. doi: 10.1002/em.20695. [DOI] [PubMed] [Google Scholar]
  46. Greenan R., MSc, University of Ottawa, 2015. [Google Scholar]
  47. Jeannet N., Fierz M., Kalberer M., Burtscher H., Geiser M. Nanotoxicology. 2015;9:34–42. doi: 10.3109/17435390.2014.886739. [DOI] [PubMed] [Google Scholar]
  48. Mülhopt S., Dilger M., Diabaté S., Schlager C., Krebs T., Zimmermann R., Buters J., Oeder S., Wäscher T., Weiss C. J. Aerosol Sci. 2016;96:38–55. [Google Scholar]
  49. Morris J. B. and Shusterman D., Toxicology of the nose and upper airways, Informa Healthcare USA, New York, 2010. [Google Scholar]
  50. Desantes J., Margot X., Gil A., Fuentes E. J. Aerosol Sci. 2006;37:1750–1769. [Google Scholar]
  51. Fujitani Y., Sugaya Y., Hashiguchi M., Furuyama A., Hirano S., Takami A. J. Aerosol Sci. 2015;81:90–99. [Google Scholar]
  52. Tsoutsoulopoulos A., Möhle N., Aufderheide M., Schmidt A., Thiermann H., Steinritz D. Toxicol. Lett. 2016;244:28–34. doi: 10.1016/j.toxlet.2015.09.003. [DOI] [PubMed] [Google Scholar]
  53. Zavala J., O'Brien B., Lichtveld K., Sexton K., Rusyn I., Jaspers I., Vizuete W. Inhalation Toxicol. 2016;28:251–259. doi: 10.3109/08958378.2016.1157227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Ritter D., Knebel J. Adv. Toxicol. 2014;2014:185201. [Google Scholar]

Articles from Toxicology Research are provided here courtesy of Oxford University Press

RESOURCES