Inflammatory Effects of Respirable Quartz Collected in Workplaces versus Standard DQ12 Quartz: Particle Surface Correlates

Anna Clouter*,1, David Brown*, Doris Höhr{dagger}, Paul Borm{dagger} and Ken Donaldson*

* School of Life Sciences, Napier University, 10 Colinton Road, Edinburgh EH10 5DT, Scotland; and {dagger} Department of Fibre and Particle Toxicology, Medical Institute for Environmental Hygiene, PO Box 103571, D-40028 Düsseldorf, Germany

Received January 22, 2001; accepted May 22, 2001


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
In 1997, the IARC (International Agency for Research on Cancer) reevaluated its quartz classification from a class 2 carcinogen, to that of a class 1, stating sufficient evidence for carcinogenicity in both humans and experimental animals. However, tumor development did not occur across all occupational settings. It is probable that this is due to the considerable differences in toxicity between workplace quartz in comparison to quartz used in experimental studies. We therefore hypothesized that workplace quartz samples differ in toxicity from standard experimental quartz samples at equal mass. To test this hypothesis we compared 2 workplace quartz samples (RH1 and OM) with standard experimental quartz (DQ12) in several assays commonly used in particle toxicology. The sizes of the quartz samples were as closely matched as possible. The endpoints of this study were inflammation in the rat lung following intratracheal instillation (1000 µg or 250 µg for 3 or 14 days), release of soluble iron, cytotoxicity to cells in culture, and surface reactivity as assessed by hemolysis and ESR. The workplace samples did not cause inflammation at any dose or time point. DQ12 quartz caused marked inflammatory responses, as measured by an increased number of neutrophils in the lungs of instilled animals for both time points and doses. Protein in the bronchoalveolar lavage also increased in animals exposed to DQ12 but not the workplace samples. In vitro, DQ12 had the greatest hemolytic activity but only RH1 released substantial amounts of soluble iron. The increased inflammogenicity of DQ12 was not wholly explained by a greater surface area, by diameter, or by releasable iron. The hemolytic activity of DQ12, while not being informative in terms of understanding the mechanism of carcinogenicity, was the best in vitro predictor for in vivo activity. Therefore the surface reactivity of DQ12 appears to drive its inflammogenicity.

Key Words: quartz; DQ12; inflammation; silica; occupational exposure.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The likely events leading to silicosis and lung cancer following exposure to quartz (crystalline silica), based on animal and in vitro studies, have been partially elucidated (Donaldson and Borm, 1998Go) with silica causing stimulation of release of cytokines from macrophages and epithelial cells (Barret et al., 1999), leading to inflammation. Oxidative stress from radicals associated with the quartz itself (Fubini et al., 1995Go; Vallyathan et al., 1991Go) and from the inflammatory leukocytes recruited to the lungs (Castranova et al., 1996) very likely contribute to oxidative stress observed in quartz-exposed lung (Janssen et al., 1992Go). The activation of the oxidative stress-responsive transcription factor NF-{kappa}B by quartz has been reported (Chen et al., 1995Go, 1998Go; Schins et al.2000Go). Results from in vivo/ex vivo studies indicate that activated oxygen species from the quartz-induced inflammatory cells are capable of causing mutations in the epithelial cells of the lung (Borm and Driscoll, 1996Go; Driscoll et al., 1995Go). This is very likely a key event in leading to development of epithelial cancer in quartz-exposed lung.

However, there is abundant evidence from experimental studies that the harmfulness of quartz can be modulated (reviewed in Donaldson and Borm, 1998). Several studies have shown that quartz has less biological activity when it is ground with coal mine dust (Le Bouffant et al., 1982Go; Martin et al., 1972Go; Ross et al., 1962Go). As a model to study the protective effects of aluminum silicate clays on the quartz surface, we (Brown et al., 1990Go) and others (Begin et al.1987Go) have utilized aluminum lactate. It was demonstrated that brief incubation of a highly inflammatory quartz sample in a solution of this compound dramatically ameliorates the pro-inflammatory effects. Iron can also reduce the toxicity of quartz as demonstrated by the ability of iron salts to protect against quartz-mediated damage to erythrocyte membranes (Nolan et al., 1981Go). Cullen et al. (1997) showed that metallic iron diminished the ability of quartz to cause inflammation in the lungs of rats, following instillation. However, the state of the iron is likely to be a most important factor in determining the activity of the surface. While metallic iron can have the above inhibitory effects, ferrous or ferric iron contamination could lead to Fenton chemistry-mediated generation of hydroxyl radicals as has been suggested for some quartz samples (Castranova, 1996Go). Various agents such as lipid and proteinaceous surfactant materials, the polymer polyvinyl-pyridine-n-oxide (PVPNO), and organosilane, have been shown to coat the surface of quartz and decrease its toxicity (Antonini and Reasor, 1994Go; Castranova et al., 1996aGo; Nolan et al., 1981Go; Vallyathan et al., 1991Go; Wallace et al., 1985Go).

In workplaces where quartz is part of a mixed dust with other minerals, quartz is likely to have its surface affected by interactions with these various other compounds such as clays and other geologically derived materials during geological processes as well as in the industrial crushing and milling processes that could release respirable quartz particles. In the present study we therefore set out to ascertain whether the toxicity and surface of workplace quartzes differed in reactivity from standard experimental respirable quartz in a number of different assays routinely utilized to study the mechanism of particle toxicity. We obtained respirable workplace quartz samples and examined them for their ability to cause lung inflammation in comparison to the standard pathogenic quartz sample DQ12. We also utilized measures of the reactivity of the quartz surfaces to relate to the in vivo biological activity.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Quartz samples.
The standard experimental quartz used was DQ12. This is a special grade of material, produced in 1972 by Robock (1973). This standard quartz sample was from the tertiary Dorentrup kaolinitic sand deposit in Westphalia, Germany. The sample was produced from a ~60 µm sand fraction, which was then centrifuged in a laboratory separator to produce a sample that was predominantly < 5 µm in diameter.

The workplace samples were collected from cyclones used in milling circuits before discharge to a bag filter. These materials were collected from 2 distinctly different production sources. The OM sample represents sand that is produced from medium hard sandstone, crushed, processed, and then milled in a rotary ball mill to produce silica flour. The samples collected represent fines, which if not collected would enter the working atmosphere. The RH1 sample was collected from a similar site location although the feedstock to the plant was a soft, loosely consolidated sand where no crushing was required to produce the sand product and the only grain breakage was due to the milling circuit.

Scanning electron microscope images of all samples are shown in Figure 1Go. These reveal the samples to be broadly similar in appearance with a suggestion that some DQ12 particles may be aggregates of smaller particles.



View larger version (68K):
[in this window]
[in a new window]
 
FIG. 1. SEM images of the particle types used in the study: (A) DQ12, (B) RH1, and (C) OM. White bar represents 10µm.

 
Chemicals.
RPMI 1640 medium (without L-glutamine), penicillin-streptomycin (1000 µg/ml), L-glutamine (200 mM), fetal bovine serum (FBS), serum substitute Ultraser G (10 x), Dulbecco's phosphate-buffered saline (PBS) with calcium and magnesium salts, and a Hemacolour rapid staining set were obtained from Gibco BRL (Paisley, UK). MTT (3-(4,5-dimethylthiazol-2-y1)-2,5-diphenyl tetrazolium bromide) and DPX were obtained from Sigma Chemicals Company (Poole, Dorset, UK). Sterile water was purchased from Baxter (UK). The spin trap 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) and hydrogen peroxide (H2O2) were obtained from Aldrich Chemicals Co. (Milwaukee, WI).

Animals.
Male Sprague-Dawley rats weighing between 250 and 350 g were obtained from an in-house colony bred at Napier University (Edinburgh, Scotland). Animals were maintained on a 12 h light/dark cycle with food and water ad libitum.

Preparation of the dust samples.
The quartz samples, suspended in methanol, were filtrated on a carbon-coated Nuclepore filter (pore size 0.2 µm) and counter coated with carbon. The filter pieces placed on a copper grid were dissolved in a chloroform atmosphere and the dust particles were estimated electron-microscopically (TEM, Philips CM12) at a magnification of 2050. The particle diameters were calculated with a digital imaging system (Soft Imaging System, Münster, Germany) using micrographs composed of maximally 30 single images by a multiple image alignment. Five hundred (DQ12) to 4513 particles (RH1, OM) per sample were used for the statistical analyses.

Characterization of quartz samples: Size and surface area.
The sizes of the quartz samples were as closely matched as possible. Consequently we anticipated that the surface area would also be comparable between the samples. Surface area by nitrogen adsorption (BET) was measured by Morgan Materials Ltd. or obtained from the literature. Chemical composition was assessed by XRF analysis.

It is demonstrated that the DQ 12 sample was the coarsest of the 3 samples tested using TEM methods but the finest by standard Coulter methods (Fig. 2Go). One explanation for this is that there is slight clumping of the DQ12 quartz particles during preparation for EM.



View larger version (16K):
[in this window]
[in a new window]
 
FIG. 2. Size distribution of quartz samples. Upper panel, data prepared by E. M.; lower panel, by Coulter counter.

 
Soluble iron.
Releasable iron (Fe3+) was measured by a spectrophotometric method that detects the color change in desferrioxamine when it combines with Fe3+.

In vitro toxicity.
The cytotoxicity of the quartz samples was assessed using A549 epithelial cells and primary rat alveolar macrophages.

A549 cell culture conditions.
The A549 cell line was obtained from the European Collection of Animal Cell Cultures (Salisbury, England). A549 epithelial cells are an adherent cell line (RPMI 1640 medium supplemented with 1% penicillin-streptomycin, 1% L-glutamine, and 10% FBS), therefore nonadherent cells were disregarded. Cells were detached from the culture dish by the addition of trypsin (5%). The cell suspensions were centrifuged at 900 x g for 10 min at 4°C and the pellets resuspended in 3 ml of culture medium. Cell counts and viability were determined using a haemocytometer and trypan blue dye exclusion (cell viability > 90%). Cells were then resuspended in culture medium according to the concentration required. Cultures were kept in a humidified incubator at 37°C with 95% air and 5% CO2. Culture conditions were maintained as above unless otherwise stated.

Isolation of rat alveolar macrophages.
Macrophages were isolated using a modified version of the method described by Harper et al. (1994). Briefly, male Sprague-Dawley rats weighing approximately 250 g were sacrificed with an overdose of pentobarbitone. The chest cavity was opened and the trachea was cannulated using a luer port cannula (Portex, Hythe, Kent 200/300/050). A 10 ml syringe, containing Ca/Mg free PBS (8 ml) was attached. The lungs were filled with the solution and massaged to encourage the Ca/Mg free PBS into all alveoli. The fluid was withdrawn and stored on ice. This procedure was repeated for a minimum of 8 washes until the final volume of lavage was approximately 50 ml. The lavage fluid was filtered through a sterile 70 µm filter (Falcon, Becton Dickinson) and the resulting solution centrifuged at 900 x g for 10 min. The supernatant was removed and the pellet resuspended in 2 ml of Ca/Mg free PBS. The cells were counted and further diluted in RPMI 1640 medium containing 1% glutamine (200 mM), 4% Ultroser G, 0.5% Amphotericin B (250 µg/ml), and 2% penicillin/streptomycin (100 units/ml).

Quartz treatments.
A549 cells were resuspended in culture medium at a concentration of 5 x 106/ml. Cell suspensions (200 µl; 1 x 105 cells) were then dispensed (6 control wells and triplicate wells for each particle type) into 96-well plates (Helena BioSciences). Cells were incubated overnight (16 h) for adherence. Culture medium was replaced with 200 µl of quartz/culture medium suspensions (25, 50, 100, and 200 µg/ml) and plates were further incubated for 1, 20, and 48 h. Macrophages were plated in a 96-well plate at a density of 100,000 cells per well (500,000/ml). The medium was removed and replaced with medium containing test samples after a period of 2 h. The cells were incubated with the quartz samples for 1, 20, or 48 h at 37°C, 5% CO2.

Assessment of cytotoxicity.
Cytotoxicity of particles was assessed using an MTT assay based on Mosmann (1983). Briefly, MTT (20 µl, 5mg/ml of PBS) was added to each well and incubated for 1 h. Supernatants were removed and replaced with 200 µl of DMSO. The 96-well plates were then centrifuged at 900 x g for 15 min at 18°C to remove any particles present in the supernatant. Supernatants (100 µl) were aliquoted onto new 96-well plates and the absorbance read at 540 nm in a MRX Microplate Reader.

Particle surface reactivity: Hemolysis.
Red blood cells were incubated with the quartz samples at 30 mg/ml and the percentage hemolysis calculated in comparison to that obtained with the detergent Triton X100 that causes 100% lysis.

ESR measurements: Hydroxyl radical generation.
Prior to use, the spin trap DMPO was dissolved in deionized water and active coal was added (30 mg/ml). Suspension was shaken continuously, incubated for 20 min at 350C, and subsequently centrifuged (2000 x g, 10 min.). This procedure was repeated once and the clear supernatant was filtered (0.45 µm). The final DMPO concentration was measured using spectrophotometry (234 nm) and aliquots were stored at –20°C.

Generation of hydroxyl radicals in solution was measured in unground dust, by adding 250 µl H2O2 (0.5 M in PBS) and 500 µl DMPO (0.05 M in miliQ water) to a 20 mg dust sample. The suspension was incubated for 10 min at 37°C and shaken repeatedly during incubation. The suspension was filtered rapidly through a 0.2 µm filter (Polycarbonate, Schleicher & Schuell Gmbh, Dassel, Germany) and 50 µl of the clear filtrate transferred to the capillary sample tube in the cavity of the ESR spectrometer. ESR spectra were recorded at room temperature using the following instrumental conditions: magnetic field, 3360 G; sweep width, 100 G; modulation amplitude, 1.975 G; receiver gain, 105; microwave frequency, 9.9 GHz; power, 50 mW; time constant, 0.00; scan time, 30 s; number of scans, 3. As a positive control a mixture of 20 mg coal fly-ash (EVA91) was used. As a negative control we used a mixture of 250 µl H2O2 (0.5 M in PBS) and 500 µl DMPO solution (0.05 M in miliQ) without dust sample added. The solutions were incubated, filtered, and analyzed by ESR as described above. ESR spectra were recorded using an MS100 Magnetech ESR-spectrometer (Berlin, Germany).

Quantification was performed by cumulation of 3 different spectra each averaging 3 different runs. All 4 peaks were quantified by measuring ist amplitudes and the 4 peaks of the DMPO-OH spectrum were added. Units are arbitrary (AU) and results were expressed as a ratio of the DQ-12 signal at the same day. Absolute variation of DQ-12 samples over different days was less then 20%. For information the positive control (EVA-91) has a signal intensity which is about 15-fold that of DQ-12 due to high iron release.

Inflammatory activity.
Male Sprague-Dawley rats weighing approximately 250 g were utilized. Test animals were dosed with different quartz samples (250 or 1000 µg) by a single intratracheal instillation, 4 rats per group. Control animals were dosed in a similar manner with the diluent, 0.15M saline (0.5 ml). Rats were sacrificed at 3 and 14 days postexposure. At each time point and for each dose the animals were sacrificed with an overdose of pentobarbital anesthetic and the free cell population obtained by lavage in situ. The samples of bronchoalveolar lavage fluid (BAL) were centrifuged and cytospins prepared of the free cell population. A differential cell count was performed to determine the recruitment of inflammatory cells to the lung.

Lactate dehydrogenase activity (LDH) and protein levels in BAL.
The cell-free supernatant was tested for protein and LDH both are measurable effects of an inflammatory response in the BAL. Protein was measured using Sigma Kit no. TPRO 562 and LDH was measured using a Sigma Kit no. 500-c modified for use on a 96-well plate.

Statistical analysis.
Statistical significance was calculated by means of two-way ANOVA with comparisons of the means using Tukey's multiple comparison test to demonstrate differences between the quartz samples. For simplicity, significance is not shown on the graphs but significant changes between each of the samples and from the samples and the controls are noted in the results.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Quartz Samples Analyses
Surface area.
As shown in Table 1Go, the surface areas of the samples as measured by BET method were broadly similar, although DQ12 did have the highest surface area.


View this table:
[in this window]
[in a new window]
 
TABLE 1 Surface Area of Quartz Samples as Measured by Nitrogen Adsorption
 
Size distribution and chemical analysis.
Figure 2Go shows the size distribution of the 3 samples and reveals that DQ12 quartz particles are generally larger than the other particles.

As shown in Table 2Go, the chemical analysis of the 3 samples did highlight some differences. Iron was present at less than 0.5% in any sample but to a 10 times greater extent in RH1 than in the other 2 quartz samples. Ca and Al also showed quite marked differences although again, generally at less than 1%. All of the samples had approximately 98% Si of which 85% was quartz in OM, 89% in DQ12, and 95% in RH1. Mineralogically the 3 samples are very different, which is not evident directly from the chemistry of each sample. DQ12 is composed of 87% crystalline silica ({alpha}-quartz) but with the balance quoted as amorphous SiO2 and with small contaminations of kaolinite (Robock, 1973Go). The OM percentage of crystalline silica is combined with a high level of aluminosilicate minerals in the form of microcline feldspar and some illite. No amorphous silica is reported in either the OM or RH samples. The RH sample has the highest level of crystalline silica but with a higher presence of contaminant oxides (particularly iron).


View this table:
[in this window]
[in a new window]
 
TABLE 2 Chemical Analysis of the Three Quartz Samples
 
Iron release.
Figure 3Go shows the release of iron from the different quartzes at acid and neutral pH. There was substantial release from RH1 but no difference between DQ12 and OM in the small amounts of iron they released.



View larger version (29K):
[in this window]
[in a new window]
 
FIG. 3. Release of iron from the quartz samples.

 
In Vitro Toxicity
A549 epithelial cells and rat alveolar macrophages were treated with test samples (25, 50, 100, and 200 µg/ml) for 1, 20, and 48 h before the MTT assay. Performing the assay on wells containing dusts but no cells controlled for any absorbance caused by the quartz alone.

Epithelial cells.
Data not shown. The only effect seen was a slight but significant decrease (p < 0.05) in MTT activity after 20 h with DQ12 compared to OM at the highest dose.

Rat alveolar macrophages.
There was a clear dose-response relationship for all quartz samples tested at all time points in terms of MTT activity in macrophages (Figs. 4 and 5GoGo). After 20 hours, MTT activity in DQ12 treated cells was significantly lower (p < 0.01) than RH1- and OM-treated cells (all doses). After 48 h exposure, MTT activity was significantly lower (p < 0.01) in DQ12 than in RH1 and OM treated cells at all doses.



View larger version (21K):
[in this window]
[in a new window]
 
FIG. 4. MTT activity in rat alveolar macrophages following a 20 h exposure to quartz. Data is expressed relative to the control figures (100%) and is the mean (± SD) of 4 repeat measurements made in a single experiment.

 


View larger version (23K):
[in this window]
[in a new window]
 
FIG. 5. MTT activity in rat alveolar macrophages following a 48 h exposure to quartz. Data is expressed relative to the control figures (100%) and is the mean (± SD) of 4 repeat measurements made in a single experiment.

 
Surface Reactivity
As shown in Table 3Go, the reactivity of the quartz samples measured by hemolysis of red blood cells was negligible for the workplace samples. Only the experimental quartz, DQ12, disrupted membranes of the red blood cells. However, Figure 6Go shows the generation of hydroxyl radicals as measured by ESR. It can be seen that both workplace samples have a greater ability to generate hydroxyl radicals than the experimental quartz.


View this table:
[in this window]
[in a new window]
 
TABLE 3 Surface Reactivity of Quartz Samples as Measured by Hemolysis of Red Blood Cells
 


View larger version (34K):
[in this window]
[in a new window]
 
FIG. 6. Generation of hydroxyl radicals in a cell free system.

 
Inflammatory Activity
The number of total cells, comprising neutrophils and macrophages, in the alveolar region demonstrates the inflammatory activity of the particles. The data is expressed as total number of cells (Table 4Go) and number of PMN (Fig. 7Go). The data is internally consistent in showing significant inflammation over the controls (p < 0.01) with DQ12 alone. This is manifest as significantly increased total cell numbers at 1000 µg for both days, increased proportions, and total numbers of PMN at all time points and doses and increased macrophage numbers with the high dose and significantly decreased proportions of macrophages where the proportions of PMNs were significantly raised where significance is equivalent to p < 0.01 for all data.


View this table:
[in this window]
[in a new window]
 
TABLE 4 Total Number of Cells in Bronchoalveolar Lavage after Quartz Treatment
 


View larger version (14K):
[in this window]
[in a new window]
 
FIG. 7. PMNs in bronchoalveolar lavage from animals exposed to quartz samples by intratracheal instillation for either (A) 3 or (B) 14 days. Data is expressed as the mean of 4 animals (± SD) for 2 doses (open bar: 1000 µg/rat, shaded bar: 250 µg/rat).

 
Protein Levels and Lactate Dehydrogenase in BAL
Protein levels in BAL are shown in Figure 8Go. After 3 days exposure, there was no significant difference between the animals treated with 250 µg but at the 1000 µg dose there was a significant increase (p < 0.01) in BAL protein levels following DQ12 treatments in comparison to the control and all other quartz-treated animals. After 14 days, there was significantly more protein in the BAL of DQ12 treated animals (p < 0.01) than RH1 (250 µg), and, as at 1000 µg after 3 days, DQ12 caused a significant increase in protein concentration compared to controls and all other quartz-treated animals (p < 0.01).



View larger version (17K):
[in this window]
[in a new window]
 
FIG. 8. Protein levels in bronchoalveolar lavage from animals exposed to quartz samples by intratracheal instillation for either (A) 3 or (B) 14 days. Data is expressed as the mean of 4 animals (± SD) for 2 dose regimes (open bar: 1000 µg/rat, shaded bar: 250 µg/rat).

 
There were no biologically significant differences at either the 250 or 1000 µg dose either 3 or 14 days postexposure in bronchoalveolar lavage LDH levels (data not shown).


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
We have previously discussed the variation in the quartz hazard (Donaldson and Borm, 1998Go) and its relevance for the IARC decision that quartz is a carcinogen (IARC, 1997Go). The likely physicochemical basis for this variation has been further discussed by Fubini (1998). Based on these arguments, we concluded that the geological or industrial history of quartz may greatly modulate its potential to cause harm to the lungs, especially if aluminum-based minerals are present with the quartz. As a first step towards addressing this issue we hypothesized that workplace quartz samples might differ in their toxicity from standard experimental quartz samples at equal mass. To test this hypothesis we compared 2 workplace quartz samples (RH1 and OM) with standard experimental quartz (DQ12) in several assays of toxicity. In addition we measured key parameters of the quartz samples—surface area, size distribution, iron release, and surface reactivity by hemolysis and ESR, to determine whether these correlated with toxicity in any of the assays.

We found that the 2 workplace quartz samples had virtually no ability to cause inflammation in the rat lung following instillation at 2 doses (250 and 1000 µg) and 2 time points (3 and 14 days postexposure). By comparison, the standard DQ12 sample, was highly inflammogenic at all doses and time points. We were surprised that there was no evidence of cell damage as measured by LDH but this might have been detected at earlier time points than those used here. Toxicity to macrophages in vitro supported this pattern of difference in potential to cause lung injury between the workplace quartzes and the DQ12 quartz. Together with these indices of toxicity, we quantified a number of potential factors relating to the physicochemistry of the 3 particle samples that might account for the observed differences in biological activity. These include: surface area, size distribution, iron release, chemical analysis, hemolytic activity towards red blood cells, and hydroxyl radical activity as measured by ESR.

The surface area of the samples was measured by BET analysis and compared with the known values for DQ12. BET measures the adsorption of nitrogen molecules onto the particle surface. This method resolves minute molecular irregularities in the particle surface. DQ12 was shown to have the greatest surface area in comparison to the workplace quartz samples but the samples were broadly similar. The size distribution data revealed that the DQ12 particles were larger than the 2 workplace samples. However, close inspection of the images suggests that the DQ12 particles are likely formed of aggregates of smaller particles to a greater extent than the workplace samples. Nitrogen molecules would readily penetrate between the single particles comprising the aggregates and bind to them, which might explain why the surface areas by BET are similar for all the samples despite the apparent greater size of the DQ12 particles.

Other pathogenic particles such as asbestos (Lund and Aust, 1992Go), PM10 (Gilmour et al., 1996Go), and residual oil fly ash (Dreher et al., 1997Go) are considered to generate free radicals via released transition metals, principally iron, and thereby cause lung inflammation. Measurements showed that the RH1 workplace quartz contained considerably more releasable iron than all other samples tested. This is in agreement with the information in Table 3Go, which shows the RH1 sample to have approximately 10-fold more iron than the other samples. However, this sample did not show any special toxicity compared to the other samples and so there does not appear to be a clear role for releasable iron in the toxicity of these quartzes. This contrasts with several studies indicating that the toxicity of quartz involves iron (Kim et al., 2000Go; Shi et al., 1994Go). Our finding of the absence of an obvious role for freely mobilized iron in causing inflammation by quartz is in contradiction to previous findings (Kim et al., 2000Go; Shi et al., 1994Go). Again this highlights the differences between quartz samples in their mode of action and the perils of generalizing on the mechanism of the quartz hazard (Donaldson and Borm, 1998Go).

The in vitro cytotoxicity data showed very little effect with epithelial cells although there was a tendency for the DQ12 to show modest toxicity. Macrophages tended to be more sensitive to all quartz samples and while there were less differences between quartzes at the earliest time points, there was a strong suggestion, partially backed up in the statistical analysis, that the toxicity could be ranked OM = RH < DQ12. In vitro cytotoxicity may be very coarse measure of quartz potential for in vivo toxicity. More subtle effects, at lower doses, such as cytokine expression (Yuen et al., 1996Go) or transcriptional activation (Chen et al., 1998Go) may be more predictive. These endpoints are related to oxidative stress (Barrett et al., 1999Go; Chen et al., 1998Go) and reflect important aspects of the in vivo response to quartz and are therefore highly relevant endpoints that should be utilized in future studies.

We utilized hemolysis here not as part of a mechanistic analysis of the effects of quartz in the lung but as a way to measure the direct reactivity of the quartz surface (i.e., the ability of the quartz surface to lyse erythrocyte membranes). At the dose used, DQ12 caused about 35% of the red cell hemolysis that was produced by the detergent, demonstrating the intense reactivity of the DQ12 surfaces. This was considerably more than that of the workplace samples, which fell between 4 and 7%. Hemolysis has been used to study the reactivity of the quartz surface (Razzaboni and Bolsaitis, 1990Go) and a role for oxidative stress was demonstrated, as seen in other studies (Barrett et al., 1999Go). However, when surface reactivity for hydroxyl radical was investigated by means of ESR, the RH1 and OM had considerably greater potential to generate free radicals at their surface than DQ12. Therefore it cannot be simply the generation of excess hydroxyl radicals at the surface of the DQ12 that renders it proinflammogenic.

It is difficult to understand why we see more ESR signal with the workplace quartzes than DQ12. In a companion study to the present one, we utilized DQ12 again and measured its ability to cause inflammation after treatment with aluminum lactate as described previously by Brown and Donaldson (1990). In this case the ESR radical signal was decreased in the aluminum lactate-treated quartz in tandem with its decreased ability to cause inflammation (Duffin et al., in preparation). Surface generation of hydroxyl radical may not be directly related to quartz toxicity and other radicals at the quartz surface could be important (Castranova et al., 1996aGo).

Ability of quartz samples to cause inflammation was related to their ability to cause hemolysis and this supports the contention that some property of the quartz particle that is intrinsic to the surface rather than diffusible, like hydroxyl radical, is responsible for biological activity. A previous study failed to show a simple relationship between ability to cause hemolysis and fibrogenicity of crystalline silica (Hemenway et al., 1990Go, 1993Go) although the sample used was cristobalite, a polymorph of crystalline silica whose mode of action is less understood than alpha quartz.

It has also been suggested that quartz has its effects forming a complex with host iron and that this is incomplete. Ghio et al. (1994) reported that ionizable Fe3+ complexed to the surface of silica increased more than 3-fold after instillation, while concentrations of iron in bronchoalveolar lavage fluid and lung tissue also increased. The workplace dusts may be much less able to complex iron than DQ12 and this could explain the difference in inflammogenicity. However, we did not measure iron complexed to particle surface after instillation in this study.

We conclude that DQ12 has more ability to cause prolonged inflammation than 2 workplace quartz samples of similar crystalline silica purity. Furthermore, this greater inflammogenicity of DQ12 is not explained by either a greater surface area (as measured by BET), chemical composition, or releasable iron. (See Table 5Go for a summary of the relationship between the explanatory variables and inflammation.) There is a contradiction between the large particle size of the DQ12 and its surface area as measured by BET. However the exceedingly small resolution of surface area detected by BET and penetration of the nitrogen between aggregates of small DQ12 particles may provide an explanation. The hemolytic activity of the DQ12 was markedly greater than for the other quartz samples. While not being informative in terms of understanding the mechanism of the DQ12 toxicity, hemolytic activity nevertheless demonstrates the much greater reactivity of the DQ12 surface. The ability of the 3 different samples to generate hydroxyl radicals with a spin trap using ESR was not related to the ability to cause inflammation. Our findings are the same as demonstrated by Vallyathan (1994). Some other physicochemical correlate of the quartz surface is responsible for the inflammogenicity and hemolytic activity of DQ12 and this forms the basis of future work in our laboratories.


View this table:
[in this window]
[in a new window]
 
TABLE 5 Summary of the Relationship of the Various Explanatory Variables to the Inflammation Caused by the Three Quartzes
 
The caveat inserted in the IARC monograph to the effect that not all quartz is carcinogenic (IARC, 1997Go) can be better understood in light of our data. Samples with high quartz purity collected from a workplace had virtually no ability to cause inflammation compared to the standard experimental quartz DQ12. Given the role of inflammation in silicosis and cancer (Donaldson and Borm, 1998Go; IARC, 1997Go), we may conclude that workers exposed to quartz who had reactivity similar to the workplace quartzes used here would be at much less risk of disease than workers with similar exposure to particles with reactivity equal to that of DQ12.

The implications for risk assessment of large variations in the quartz hazard need to be addressed. Furthermore experimental studies with quartz have historically utilized high reactivity standard quartzes such as DQ12 in Europe and Min-U-Sil in the USA (IARC, 1997Go). It is important to note that the surface of DQ12 must be very different from that of the workplace silicas and this is reflected in the higher levels of inflammation seen with this sample. In studies on the pathological effects of crystalline silica, the particle surface is never fully mineralogically or physicochemically characterized, and from the present work this appears to be of great importance. If the data shown here has general application, then such standard samples as DQ12 and the responses they cause may have limited relevance to groups of people exposed in occupational settings. An understanding of the mechanistic events leading to lung disease from quartz in humans requires that quartzes with reactivities and modes of action similar to those that occur in human exposures are used in such studies. Further work on the characteristics of the quartz surface that mediate inflammation is required and is ongoing in our laboratories.


    ACKNOWLEDGMENTS
 
This research was funded by EUROSIL, the European Silica Producer's Association. K.D. is the British Lung Foundation Transco Fellow in Air Pollution and Respiratory Health.


    NOTES
 
1 To whom correspondence should be addressed. Fax: (+44) 131 455 2291. E-mail: a.clouter{at}napier.ac.uk. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Antonini, J. M., and Reasor, M. J. (1994). Effect of short-term exogenous pulmonary surfactant treatment on acute lung damage associated with the intratracheal instillation of silica. J. Toxicol. Environ. Health 43, 85–101.[ISI][Medline]

Barrett, E. G., Johnston, C., Oberdorster, G., and Finkelstein, J. N. (1999). Antioxidant treatment attenuates cytokine and chemokine levels in murine macrophages following silica exposure. Toxicol. Appl. Pharmacol. 158, 211–220.[ISI][Medline]

Begin, R., Masse, S., Sebastien, P., Martel, M., Geoffroy, M., and Labbe, J. (1987). Late aluminum therapy reduces the cellular activities of simple silicosis in the sheep model. J. Leukoc. Biol. 41, 400–406.[Abstract]

Borm, P. J. A., and Driscoll, K. E. (1996). Particles, inflammation and respiratory tract carcinogenesis. Toxicol. Letters 88, 109–113.[ISI][Medline]

Brown, G. M., Brown, D. M., Slight, J., and Donaldson, K. (1991). Persistent biological reactivity of quartz in the lung: Raised protease burden compared with a non-pathogenic mineral dust and microbial particles. Br. J. Ind. Med. 48, 61–69.[ISI][Medline]

Brown, G. M., and Donaldson, K. (1990). Modulation of quartz toxicity by aluminum. In Silica and Silica-Induced Diseases (V. Castranova, V. Vallyathan, and W. E. Wallace, Eds.), pp. 299–304. CRC Press, Boca Raton, FL.

Brown, G. M., Donaldson, K., Brown, D. M. (1990). Bronchoalveolar leukocyte response in experimental silicosis: Modulation by a soluble aluminum compound. Toxciol. Appl. Pharmacol. 101, 95–105.

Castranova, V. (1996). Suppression of the cytotoxicity and fibrogenicity of silica by PVPNO. In Silica and Silica-Induced Diseases (V. Castranova, V. Vallyathan, and W. E. Wallace, Eds.), pp. 293–298. CRC Press, Boca Raton, FL.

Castranova, V., Antonini, J. M., Reasor, M. J., Wu, L., and VanDyke, K. (1996b). Oxidant release from pulmonary phagocytes. In Silica and Silica-Induced Diseases (V. Castranova, V. Vallyathan, and W. E. Wallace, Eds.), pp. 185–195. CRC Press, Boca Raton, FL.

Castranova, V., Dalal, N. S., and Vallyathan, V. (1996a). Role of surface free radicals in the pathogenicity of silica. In Silica and Silica-Induced Diseases (V. Castranova, V. Vallyathan, and W. E. Wallace, Eds.), pp. 91–105. CRC Press, Boca Raton, FL.

Chen, F., Lu, Y., Demers, L. M., Rojanasakul, Y., Shi, X., Vallyathan, V., and Castranova, V. (1998). Role of hydroxyl radical in silica-induced NF-kappa B activation in macrophages. Ann. Clin. Lab. Sci. 28, 1–13.[Abstract]

Chen, F., Sun, S. C., Kuh, D. C., Gaydos, L. J., and Demers, L. M. (1995), Essential role of NF-kappa B activation in silica-induced inflammatory mediator production in macrophages. Biochem. Biophys. Res. Commun. 214, 985–992.[ISI][Medline]

Cullen, R. T., Vallyathan, V., Hagen, S., and Donaldson, K. (1997). Protection by iron against the toxic effects of quartz. Ann. Occup. Hyg. 41(Suppl. 1), 420–425.

Donaldson, K., and Borm, P. J. (1998) The quartz hazard: A variable entity. Ann. Occup. Hyg. 42, 287–294.[ISI][Medline]

Dreher, K. L., Jaskot, R. H., Lehmann, J. R., Richards, J. H., McGee, J. K., Ghio, A. J., and Costa, D.L. (1997). Soluble transition metals mediate residual oil fly ash induced acute lung injury. J. Toxicol. Environ. Health 50, 285–305.[ISI][Medline]

Driscoll, K. E., Deyo, L. C., Howard, B. W., Poynter, J., and Carter, J. M. (1995). Characterising mutagenesis in the hprt gene of rat alveolar epithelial cells. Exp. Lung Res. 21, 941–956.[ISI][Medline]

Fubini, B. (1998). Surface chemistry and quartz hazard. Ann. Occup. Hyg. 42, 521–30.[ISI][Medline]

Fubini, B., Bolis, V., Cavenago, A., and Volante, M. (1995). Physicochemical properties of crystalline silica dusts and their possible implication in various biological responses. Scand. J. Environ. Health 21(Suppl. 2), 9–14.[ISI]

Ghio, A. J., Jaskot, R. H., and Hatch, G. E. (1994). Lung injury after silica instillation is associated with an accumulation of iron in rats. Am. J. Physiol. 267, L686–L692.[Abstract/Free Full Text]

Gilmour, P. S., Brown, D. M., Lindsay, T. G., Beswick, P. H., MacNee, W., and Donaldson, K. (1996) Adverse health effects of PM10 particles: Involvement of iron in generation of hydroxyl radical. Occup. Environ. Med. 53, 817–822[Abstract]

Harper, R. A., Stirling, C., Townsend, K. M. S., Kreyling, W. G., and Patrick, G. (1994). Intracellular particle dissolution in macrophages isolated from the lung of the Fischer (F-344) rat. Exp. Lung. Res. 20, 143–156.[ISI][Medline]

Hemenway, D. R., Absher, M. P., Trombley, L., and Vacek, P. M. (1990). Comparative clearance of quartz and cristobalite from the lung. Am. Ind. Hyg. Assoc. J. 51, 363–369.[ISI][Medline]

Hemenway, D. R., Absher, M. P., Fubini, B., and Bolis, V. (1993). What is the relationship between hemolytic potential and fibrogenicity of mineral dusts. Arch. Environ. Health 48, 343–347.[ISI][Medline]

IARC (1997). IARC Monograph on the Evaluation of the Carcinogenic Risk of Chemicals to Humans: Silica, Some Silicates, Coal Dust and Para-Aramid Fibrils, Vol. 68. IARC Press, Geneva.

Janssen, Y. M., Marsh, J. P., Absher, M. P., Hemenway, D., Vacek, P. M., Leslie, K. O., Borm, P. J., and Mossman, B. T. (1992). Expression of antioxidant enzymes in rat lung after inhalation of asbestos or silica. J. Biol. Chem. 267, 10625–10630.[Abstract/Free Full Text]

Kim, K. -A., Kim, E. -K., Chang, H. -S., Kim, J. -H., Lim, Y., and Park, C. -Y. (2000). Effect of desferrioxamine on silica-induced pulmonary reaction. Inhal. Toxicol. 129(Suppl. 3), 117–123.

Le Bouffant, L., Daniel, H., Martin, J. C., and Bruyere, S. (1982). Effect of impurities and associated minerals on quartz toxicity. Ann. Occup. Hyg. 26, 625–634.[ISI][Medline]

Lund, L. G., and Aust, A. E. (1992), Iron mobilization from crocidolite asbestos greatly enhances crocidolite-dependent formation of DNA single-strand breaks in phi X174 RFI DNA. Carcinogenesis 13, 637–642.[Abstract]

Martin, J. C., Daniel-Moussard, H., Le Bouffant, L., and Policard, A. (1972). The role of quartz in the development of coal workers pneumoconiosis. Ann. N.Y. Acad. Sci. 200, 127–141.[ISI][Medline]

Mosman, T. (1983) A rapid colorimetric assay for cellular growth and survival: Application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55–63.[ISI][Medline]

Nolan, R. P., Langer, A. M., Harington, J. S., Oster, G., and Selikoff, I. J. (1981). Quartz hemolysis as related to its surface functionalities. Environ. Res. 26, 503–520.[ISI][Medline]

Razzaboni, B. L., and Bolsaitis, P. (1990). Evidence of an oxidative mechanism for the hemolytic-activity of silica particles. Environ. Health Perspect. 87, 337–341.[ISI][Medline]

Robock, K. (1973). Standard quartz DQ12 < 5µM for experimental pneumoconiosis research projects in the federal republic of Germany. Ann. Occup. Hyg. 16, 63–66.[Medline]

Ross, H. F., King, E. J., Yognathan, M., and Nagelschmidt, G. (1962). Inhalation experiments with coal dust containing 5 percent, 10 percent, 20 percent and 40 percent quartz: Tissue reactions in the lungs of rats. Ann. Occup.Hyg. 5, 149–161.

Schins, R. P. F., McAlinden, A., MacNee, W., Jimenez, L. A., Ross, J. A., Guy, K., Faux, S. P., and Donaldson, K. (2000). Persistent depletion of I kappa B alpha and interleukin-8 expression in human pulmonary epithelial cells exposed to quartz. Toxicol. Appl. Pharmacol. 167, 107–117.[ISI][Medline]

Shi, X. L., Mao, Y. Daniel, L. N., Saffiotti, U. Dalal, N. S., and Vallyathan, V. (1994) Silica radical-induced DNA-damage and lipid-peroxidation. Environ. Health Perspect. 102, 149–154.[ISI][Medline]

Vallyathan, V. (1994). Generation of oxygen radicals by minerals and its correlation to cytotoxicity. Environ. Health Perspect. 102(Suppl. 10), 111–115.

Vallyathan, V., Kang, J. H., VanDyke, K., Dalal, N. S., and Castravova, V. (1991). Response of alveolar macrophages to in vitro exposure to freshly fractured versus aged silica dust: The ability of Prosil 28, an organosilane material, to coat silica and reduce its biological activity. J. Toxicol. Environ. Health 33, 303–315.[ISI][Medline]

Wallace, W. E., Vallyathan, V., Keane, M. J., and Robinson, V. (1985). In vitro biologic toxicity of native and surface-modified silica and kaolin. J. Toxicol. Environ. Health 16, 415–424.[ISI][Medline]

Yuen, I. S., Hartsky, M. A., Snajdr, S. I., and Warheit, D. B. (1996). Time course of chemotactic factor generation and neutrophil recruitment in the lungs of dust-exposed rats. Am. J. Respir. Cell Mol. Biol. 15, 268–274.[Abstract]