Collection of PM₁₀ samples was co-ordinated by Casella Stanger, London, England. Particles were collected onto TEOM filters in Marylebone Road, London, a site which had particularly high levels of traffic and therefore high levels of primary, combustion-derived nanoparticles. Ultrafine carbon black (UfCB) was obtained from Degussa (Printex 90), the average particle size was 14 nm. The characteristics and details of UfCB particles have been published previously 30.

A single PM₁₀ filter was placed into a bijou bottle and 0.5 ml phosphate buffered saline (PBS) added. The bottle was vortexed for 4 minutes to remove the particles from the filter and the resulting suspension transferred to a clean bijou bottle. The mass of particles was assessed by densitometry. As standards, a series of dilutions of UfCB particles were made, ranging from 15.625 μg/ml to 1 mg/ml in saline, sonicated for 5 minutes, and 75 μl of each concentration was added into triplicate groups of wells in a 96-well plate. Seventy-five microlitres of PM₁₀ sample were added into a separate triplicate group of wells. The samples and standards were then read on a plate reader at 340 nm and the mass of particles calculated from a linear regression of the UfCB standards.

The mouse macrophage cell line J774.A1 (a kind gift from Dr W Muller GSF, Gauting, Germany) was routinely cultured in RPMI medium (Sigma) containing 5% foetal calf serum (FCS) and Penicillin/Streptomycin. Cells were cultured until confluency was reached and then scraped from the surface of the flasks using a cell scraper. The cells were counted and adjusted to 5 × 10⁵/ml in RPMI plus 5% FCS. Sterile 10 mm glass cover slips were placed in each well of a 24-well plate and 1 ml of cell suspension added to each well. Cells were incubated at 37°C for 24 hours prior to particle treatment.

Human peripheral blood mononuclear cells were prepared according to the protocol of Dransfield et al, 31. In brief, two separate volumes of 40 ml of blood were withdrawn from healthy consenting volunteers and transferred to 50 ml sterile Falcon tubes containing 4 ml of 3.8% sodium citrate solution. Tubes were gently inverted and centrifuged at 250 g for 20 minutes, the plasma removed from each tube and pooled without disturbing the cell pellet. Dextran (Pharmacia), prepared as a 6% solution in saline was warmed to 37°C, before adding to the cell pellet (2.5 ml/10 ml cell pellet) and the volume made up to 50 ml with sterile saline. Tubes were gently mixed and the cells allowed to sediment at room temperature for 30 minutes. In order to prepare autologous serum, calcium chloride solution (220 μl 1 M/10 ml), was gently mixed with the plasma and incubated in a glass tube at 37°C until the clot retracted. Percoll (Pharmacia) gradients were made from a stock solution of 90% (18 ml Percoll + 2 ml 10x PBS, (Life Technologies, Paisley) without calcium or magnesium) to give final concentrations of 81%, 70% and 55% using 1x PBS. The separating gradient was prepared by layering 2.5 ml of 70% percoll over 2.5 ml 81% percoll. The leukocyte-rich fraction from the dextran sedimentation was transferred to sterile falcon tubes, 0.9% saline added to give a final volume of 50 ml and the tubes centrifuged at 250 g for 6 minutes. The pellet was resuspended in 55% percoll and 2.5 ml layered over the previously prepared separating gradients. Tubes were centrifuged at 290 g for 20 minutes and the mononuclear cells collected from the 55/70 layer. Cells were washed twice with PBS, counted, and resuspended in RPMI medium at a concentration of 5 × 10⁶ cells/ml and 1 ml added to each well of a 24 well plate. For calcium imaging, cells were also set up in 6-well plates containing a 26 mm diameter sterile glass coverslip. The cells were incubated for 1 hour at 37°C, the medium removed and replaced with RPMI plus 10% autologous serum and incubated for 48 hours at 37°C. After the second incubation, the medium was replaced and the cells incubated for a further 72 hours prior to treatment.

PM₁₀ particles were diluted to give a final concentrations ranging from 5 μg/ml to 40 μg/ml in RPMI medium without serum and the suspension was sonicated for 5 minutes to disperse the particles. Cells which had been set up as described above, were washed twice with sterile PBS and 250 μl of particle suspension added to appropriate wells. UfCB particles were quantified as described for the PM₁₀ and set up in parallel with PM₁₀ particles at similar mass concentrations with J774 cells to investigate TNF-α release. One well received medium only (-ve control) and one received 250 μl of 1 μg/ml LPS (+ve control). The calcium antagonists were added concomitantly with the particles to give final concentrations of verapamil (100 μM), BAPTA-AM (50 μM), W-7 (250 μM), trolox (25 μM), and nacystelyn (5 mM). The cells were then incubated at 37°C for 4 hours and the supernatants removed and stored at -80°C until required. The cells cultured on 10 mm cover slips were fixed by the addition of 3% formaldehyde.

J774.A1 cells were cultured and removed from flasks as described above. Cells were pooled into a single tube, adjusted to 4.5 × 10⁶ cells/ml in RPMI plus 10% FCS and incubated at 37°C until required for the assay. One millilitre of cell suspension was transferred to an Eppendorf tube, centrifuged at 145 g for 2 minutes, the medium removed, the cell pellet resuspended in 1 ml PBS and again centrifuged at 145 g for 2 minutes. The PBS was removed and cells resuspended in serum-free RPMI medium containing 23 mM Hepes buffer. Cells were loaded with 1 μg/μl Fura 2-AM (Sigma) in DMSO, 2 μl/ml cell suspension, the tube wrapped in foil and incubated in a shaking water bath for 20 minutes at 34°C. After incubation, the tube was centrifuged at 145 g for 2 minutes at 4°C, the medium removed and replaced with 1.5 ml fresh RPMI without serum. The Fura 2-AM-loaded cells were transferred to a quartz cuvette with stirrer and placed immediately into a fluorimeter with heated block and basal fluorescence measurements obtained over a 100 second period. The fluorimeter was set up with to give excitation wavelengths of 340 nm and 380 nm, emission 510 nm and excitation and emission slit widths set at 5 nm. During the experiments, the cuvette temperature was kept constant at 37°C. After 100 seconds, 10 μl appropriate treatment in RPMI medium was added to the cuvette and the experiment allowed to run for a further 1700 seconds. Treatments consisted of PM₁₀ to give a final concentration of 10 μg/ml with and without the calcium antagonists at the concentrations described above. Twenty microlitres of 5% Triton solution were added to the cuvette to lyse the cells to give the maximum fluorescence (Rmax) and the experiment continued for 500 seconds. To give the minimum fluorescence value (Rmin), 15 μl of 0.5 M EGTA in 3 M Tris buffer were added to the cuvette. The experiment was terminated after a further 500 seconds. The ratio of the fluorescence measurements at excitation wavelengths of 340 and 380 nm were converted to calcium concentration values using the method of Grynkiewicz et al, 32.

The supernatants previously prepared were assayed for TNF-α protein content using a commercially available human TNF-α kit (Biosource) or mouse TNF-α kit (R&D Systems) according to the manufacturer's instructions. Briefly, each well of a 96-well plate was coated overnight with capture antibody, before washing with PBS containing 0.05% tween, and then adding test supernatant to the appropriate wells in triplicate groups. After incubation for 2 hours at room temperature, the wells were washed, a detection antibody added and incubated for a further hour at room temperature. The wells were then washed with PBS/tween before addition of Horseradish peroxidase (HRP)-conjugated streptavidin and incubated for 45 minutes at room temperature. Finally, the colour was developed by adding peroxidase substrate to each well, before reading the absorbance at 450 nm using a Dynatec plate reader.

The experiments described above were also used to generate cells for total RNA extraction. After removal of the supernatant, 400 μl Tri reagent (Sigma) was added to each well. The lysed cells were then scraped from the surface of the plate using a cell scraper and transferred to Eppendorf tubes. Two hundred microlitres of chloroform were added to each Eppendorf, vortexed for 15 seconds and allowed to stand at room temperature for 15 minutes. The resulting mixture was centrifuged at 12000 g for 15 minutes at 4°C. The colourless upper phase was transferred to a fresh Eppendorf, before adding 450 μl isopropanol. The mixed samples were allowed to stand for a further 10 minutes at room temperature. Again the tubes were centrifuged at 12000 g for 10 minutes at 4°C, the supernatant removed and the RNA pellet washed in 1 ml of 75% ethanol. The resulting samples were then vortexed briefly, centrifuged at 7500 g for 5 minutes at 4°C and the RNA pellet air-dried for 10 minutes. The RNA was then suspended in 50 μl diethylpyrocarbonate (DEPC)-treated water and stored at -70°C until required for quantification and Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR).

The RT-PCR procedure was carried out using the Promega Access Kit. Briefly, a master mix of the kit reagents was prepared according to the manufacturers instructions. Ten microlitres of RNA at 0.03 μg/ml was added to 40 μl of the master mix, containing 10 μl of the appropriate human primers Glyceraldehyde Phosphate Dehydrogenase (GAPDH) or IL-1-α (MWG AG Biotech, Ebersberg, Germany). Tubes were placed in a thermal cycler which was programmed for the following temperatures and times. Following an initial 45 minute incubation at 48°, samples were cycled as follows: 94°C for 2 minutes, 95°C for 30 seconds, 60°C for 1 minute, 68°C for 2 minutes. This cycle was repeated 25 times for GAPDH and 30 times for IL-1 alpha. To conclude, the sample was incubated at 68°C for 7 minutes and then cooled to 4°C. The resulting RT-PCR products were separated by electrophoresis using a 2% agarose gel containing 1 μg/ml ethidium bromide and viewed under UV light. The RT-PCR bands were quantified by densitometry using Syngene software and the IL-1α band intensity was expressed as a percentage of the corresponding GAPDH band. These results were then expressed as a percentage of the untreated control.

Human mononuclear cells were isolated from blood as previously described followed by adhesion onto 26 mm glass coverslips contained in 6-well plates. Cells were seeded in RPMI medium containing 0.1% BSA and penicillin/streptomycin at a density of 5 × 10⁵ cells/ ml and incubated at 37°C, 5% CO₂ for 1 hour before washing with 1 ml PBS. Prior to particle treatment and digital enhanced video microscopy (Roper scientific), cells were loaded with the calcium-sensitive dye, Fura 2-AM (2 μg/ml in RPMI) (Sigma) for 30 minutes at 37°C. The coverslips were then washed with PBS, assembled into the microscope holder and 400 μl RPMI medium without phenol red (Sigma) added. The fluorescence ratio was observed (excitation 340 and 380 nm, emission 510 nm) at a magnification of 63× (Zeiss Axiovert microscope). Images were captured every 2 seconds by a Coolsnap fx Photometrics (Roper Scientific) camera controlled by Metafluor software. After 100 seconds particle treatment was added to the cells (100 μl of a 250 μg/ml stock solution of particles to give a final concentration of 50 μg/ml) contained in phenol red-free RPMI medium.

The cells cultured on cover slips and fixed with formaldehyde were washed three times with PBS and permeablised with 0.1%Triton for 4 minutes. Cells were then washed three times with PBS and the F-actin stained using 33 ng/ml Phalloidin-FITC (Sigma) in PBS for 30 minutes at room temperature. Cells were washed three times with PBS before staining with propidium iodide (10 μg/ml in PBS) for 5 seconds. Cells were further washed three times with PBS before being mounted on glass slides using Citifluor mounting medium. The cells were then examined using an Axiofluor fluorescence microscope. Images were captured and quantified using Metamorph software (Universal Imaging Corporation). Seven fields of view were captured from each treatment and the images deconvolved using the image software. The staining intensity of each cell was then measured using the analysis software.

Data from all of the experiments were analysed using analysis of variance with Tukey or Fishers multiple comparison test. Significance was set at p < 0.05.

The effects of PM₁₀ on J774 murine macrophages was investigated at final concentrations of 5, 10 or 25 μg/ml PM₁₀. Treatment of the cells with these particles produced a dose-dependent increase in cytosolic calcium concentration [Ca²⁺]_c up to a concentration of 10 μg/ml. At 25 μg/ml the [Ca²⁺]_c decreased to a value similar to the 5 μg/ml particle concentration. (Figure 1a). Subsequent treatment with thapsigargin to release the endoplasmic reticulum calcium store produced a further increase in cytosolic Ca²⁺ indicating that the cells remained viable and confirming previous studies 33. There was a statistically significant difference between control and PM₁₀ treatments at 10 μg/ml (p < 0.05). The [Ca²⁺]_c following concomitant treatment of cells with particles and calcium antagonists was reduced (figure 1b). Both the calcium channel blocker verapamil, and the calcium chelator BAPTA-AM significantly reduced (p,0.05) the intracellular calcium compared with PM₁₀ alone. In contrast, the antioxidant Nacystelyn did not significantly reduce the PM₁₀-stimulated [Ca²⁺] increase, with Ca²⁺ concentration remaining significantly greater than the control value (p < 0.05). In our previous studies 3435 we demonstrated that the antagonists used at the same concentrations used here caused no toxic effects to cells and that the drug treatment produced similar results to the untreated controls.

PM₁₀ also induced a significant increase in cytosolic calcium in the primary human monocytes (p < 0.05). The dose of 10 μg/ml final concentration was chosen as the dose at which a significant increase in [Ca²⁺]_c had previously been observed (figure 1a). At time points from 600 to 800 seconds after the addition of particle/antagonist treatment, there was a rapid increase in [Ca²⁺]_c with particles alone compared with the antagonists. In contrast to the antagonist treatment effects reported in figure 1, the antioxidant nacystelyn significantly inhibited the [Ca²⁺]_c changes induced in the human monocytes treated with 10 μg/ml PM₁₀ (figure 2). Both the intracellular calcium chelator BAPTA-AM and the calcium channel blocker verapamil also significantly inhibited the [Ca²⁺]_c rise compared with particles alone (p < 0.05).

A comparison of the gram for gram dose effect of PM₁₀ and UfCB particles on TNF-α release by J774 macrophages is shown in figure 3. The data show that PM₁₀ particles caused significantly more TNF-α release as the same mass of UfCB particles by the mouse macrophage cell line.

The release of TNF-α protein by human monocytes after treatment with varying concentrations of PM₁₀ is shown in figure 4. The dose of particles ranged from 5 μg/ml to 40 μg/ml with concomitant treatment with calcium antagonists. There was a clear dose response of particle treatment from 10 μg/ml to 40 μg/ml. Within this range, the TNF-α concentration was approximately 170 pg/ml to 1000 pg/ml. At higher particle doses, the calcium antagonists reduced TNF-α release only marginally with the most dramatic and significant effect being seen with a particle concentration of 10 μg/ml with verapamil (V) and BAPTA-AM (B) treatments which reduced TNF-α release to 29 pg/ml and 7 pg/ml respectively (p < 0.05). There was no reduction in PM₁₀ induced TNF-α release with W-7 (W), Trolox (T) or Nacystelyn (N) at any particle dose tested.

Treatment of human peripheral blood monocytes with 10 μg/ml PM₁₀ for 4 hours produced a significant increase in IL-1α mRNA content compared with unstimulated cells (p < 0.05) (figure 5). The IL-1α band intensities were expressed as a percentage of the GAPDH band intensities and then normalised to the unstimulated control. PM₁₀ induced a five fold increase in IL-1α mRNA expression compared with the control and on treatment with the calcium antagonists, this was reduced to values similar to the control. There was a significant difference between the PM₁₀ exposed cells and concomitant treatment with all of the calcium antagonists and antioxidants tested (p < 0.05).

The fluorescence intensity of cells stained for F-actin after treatment with PM₁₀ particles and calcium antagonists is shown in figure 6. Particles alone significantly increased the phalloidin-FITC fluorescence and hence the F-actin content of the cells compared with untreated cells. All of the calcium antagonists tested inhibited the PM₁₀ induced increase in F-actin intensity to control levels and this was significantly different from particle only treatment (p < 0.05), although the increase in the fluorescence intensity of the PM₁₀-treated cells was modest (a 5% difference).