Dissolution of man-made vitreous fibers in rat alveolar macrophage culture and Gamble's saline solution: influence of different media and chemical composition of the fibers.

The effect of different chemical compositions of man-made vitreous fibers (MMVF) on their dissolution by alveolar macrophages (AM) in culture and in Gamble's solution was studied. The fibers were exposed to cultured rat AMs, culture medium alone; or Gamble's saline solution for 2, 4, or 8 days. The dissolution of the fibers was studied by measuring the amount of silicon (Si), iron (Fe), and aluminum (Al) in each medium. The AMs in culture dissolved Fe and Al from the fibers but the dissolution of Si was more marked in the cell culture medium without cells and in the Gamble's solution. The dissolution of Si, Fe, and Al was different for different fibers, and increased as a function of time. The Fe and Al content of the fibers correlated negatively with the dissolution of Si by AMs from the MMVF, i.e., when the content of Fe and Al of the fibers increased the dissolution of Si decreased. These results suggest that the chemical composition of MMVFs has a marked effect on their dissolution. AMs seem to affect the dissolution of Fe and Al from the fibers. This suggests that in vitro models with cells in the media rather than only culture media or saline solutions would be preferable in dissolution studies of MMVFs.


Introduction
Man-made vitreous fibers (MMVFs) are amorphous silicates manufactured from glass, rock, or other minerals by drawing, blowing or spinning the material into fibers. They are composed of oxides of Si, Al, Fe, alkalis and alkaline earths and other elements to add specific properties. The chemical composition of MMVF determines their chemical resistance and solubility in different environments (1)(2)(3). Biological activity of the fibers depends not only on their respirability, but also on their chemical durability and persistency (4)(5)(6)(7). If the fiber dissolves rapidly, it is less likely to have long-term deleterious effects in the lungs (8).
The aim of the present study was to investigate the effects of different chemical compositions of MMVF on the dissolution of fibers by rat alveolar macrophage (AM) culture, culture medium alone, and Gamble

Man-made Vitreous Fibers
The fibers A (F), B, C, and G were prepared by conventional spinning techniques on a pilot line by Paroc Oy Ab Parainen, Finland. Neither binding material nor oil was added. Fibers D and E were normal rockwool samples received from Paroc Oy Ab. Sample H was commercially available ceramic fiber and the sample J commercially available glasswool. The very fine fibers were collected in a piece equipment, consisting of a rotating vulcanized steel container, a separation chamber, and a filter system for collecting the airborne fibers.
The majority (56-79%) of the collected fibers were less than 1 pm in diameter.
The fibers for the study were chosen based on their chemical composition (Table 1), particularly their Al and Fe content, because these elements markedly influence the dissolution of the fibers in aqueous solutions (18). The dimensions of the fibers were measured from SEM photographs magnified x2600. A part of the Environmental Health Perspectives    (20). Cell viability was never less than 95% (21). The AMs were resuspended in regular medium 1640 containing 100 IU/ml of penicillin, 100 pg/ml of streptomycin, 10% fetal calf serum (FCS) and 2 mM of L-glutamine and 10 IU/ml of heparin. The cell concentration was adjusted to 1 x 106cells/ml, and the AMs were allowed to adhere to the culture wells for 1.5 hr in an incubator at 37°C in 5% CO2 /95% air mixture, then the medium in the wells was changed to remove the nonadherent cells. Fresh medium (2 ml) was added, and the culture wells were incubated overnight. Following incubation, the medium was removed, and fresh medium containing 200 pl/ml of MMVF was added to each well. Control wells contained only culture medium with mineral fibers. The wells were cultured for 2, 4, or 8 days.
Fresh medium (1 ml) was added to the wells on day 4. For each time point, separate incubations were carried out. After incubation, the medium was collected, and the cells were detached with a 0.25% trypsin solution containing 0.006% EDTA, lysed with 0.1% Triton X-100 solution, sonicated for 40 min, and filtered through a 0.45-pm Millipore filter using vacuum. Si-, Fe-, and Al-concentrations in the filtrates were analyzed with a microcomputer-controlled atomic absorption spectrometer (AAS, Perkin-Elmer 2100) using a graphite furnace (Perkin-Elmer, HGA-700) technique. Three separate determinations were made of the solubility of each fiber sample.

Man-made Vitreous Fibers in Gamble's Saline Solution
The chemical composition of the Gamble's solution (22) was modified according to Klingholz and Steinkopf (9). For determining the dissolution of fibers during continuous flow, a modification of an apparatus designed by Forster (7) was used. A sample of 200 mg of each fiber was placed on the 0.45-pm pore Millipore filter paper and Gamble's saline solution was passed through the sample using a peristaltic pump (Desaga PLG 132100, Heidelberg, Germany) at a flow rate of 3 to 5 ml/hr. The samples were kept at 37°C, and pH was adjusted to 7 with CO2. The dissolution of the fibers was determined after 2, 4, and 8 days, with each fiber being tested twice. Two sample holders were included as controls in each battery of ten. The concentrations of Al, Fe, and Si in the effluents were measured with an atomic absorption spectrometer (AAS, Perkin-Elmer 5000). More Si was dissolved in the culture medium than in the AM culture; the opposite was true for Fe and Al. The dissolution of Si was the greatest for the fiber C with a low Fe and Al content (Table 1), and the smallest for fibers D, E, and H, with high Fe or Al contents. The largest amounts of Fe were dissolved from the fibers B and E, which had relatively high Fe content, and the smallest from the fiber H, which had a low Fe content. The dissolution of Al was the highest from the fiber H with a high Al content, and much smaller from fibers C, G, and J, all with low Al content. Fibers D, E, and H seemed to show a negative correlation between their Fe and Al contents and the dissolution of Si from them.

Man-made Vitreous Fibers in Gamble's Saline Solution
In the Gamble's solution the dissolution of Si was highest from fiber C and lowest from fibers D, E, and H: in general it was higher in Gamble's solution than in AM culture for fibers A, F, and G. For the other fibers it was lower or similar, while the dissolution of Fe was minimal. The dissolution of Al was higher from fibers D, E, and H, which all had high Al content, and low from all the others.

Discussion
The main thrust of these studies was to explore whether AMs can dissolve respirable-sized MMVF, and how the chemical composition of the fibers modifies their dissolution in different media. The results show that AMs play an active role in fiber dissolution but their effect seems to be complex. The dissolution profiles of the fibers exposed to AMs were different from those obtained with either Gamble's solution or culture medium. The lower dissolution of Si by AMs could be explained partly by AMs preventing or slowing down the diffusion of elements by blocking the surface of the fibers. Another reason could be the lower pH in the AM cultures (15)(16)(17), which actually decreased during 2 to 3 days of incubation. In the controls, where the fibers were in culture medium alone, the pH increased, although the change of pH was smaller than in AM culture.
In AM culture, the dissolution of the fibers takes place both intra-and extracellularly, depending on the length of the fibers. Those of less than 20 pm are mainly dissolved intracellularly. There are, of course, many differences between intra-and extra-Environmental Health Perspectives LUOTO ETAL.     cellular milieus. Within the phagolysosomes of the AMs the pH is much lower than normal extracellular pH, which is near to pH 7 (15,17). Comparing rockwool samples (D and E) with the glasswool sample (J), the dissolution in AM culture is higher for glasswool in the extracellular medium, while rockwool dissolves more readily within the alveolar macrophages. Si dissolution from glasswool increases as pH increases. For rockwool the opposite is true, which explains why intracellular dissolution of fibers by AMs is probably more effective for rockwool than for glasswool. Moreover, Si forms stable insoluble silicates at low pH, whereas at higher pH, the formation of free silicon hydroxyl complexes increases (23). The fiber samples, collected from the air after dusting of the raw material, provided the kind of airborne fibers that would exist in a real working environment. These were typically longer than 20 pm, but there were also shorter particles in the samples. Therefore, in the AM experiments, there could be both extraand intracellular dissolution, with AMs blocking the surface of the long fibers in the first case, and dissolving the short fibers within the phagolysosomes in the second case.
The dissolution of Si was the lowest for rockwool samples D and E and the ceramic fiber sample H, in all three types of solutions used (Figure 1), and this may be due to the high aluminium content in these fibers. The influence of different media on the dissolution of Fe and Al from the fibers is more marked for these ions than for Si (Figure 2,3). The dissolution of Fe is minimal in Gamble's saline solution, quite high in cell culture medium, and highest in the AM culture. This may be because there are different mechanisms due to the different chemistries of the media. Even stronger than the pH effect could be that caused by complex formation between Fe and Al, and chelating agents, such as amines and aminocarboxylic acids (23). Gamble's solution does not contain any of those components. In the cell culture medium, the dissolution of Fe and Al may be reduced because of the formation of hydroxyl compounds, which precipitate at the surface of the fibers and thereby inhibit further dissolution. The increased dissolution of Fe and Al by AMs can have an important impact on the biopersistence of the fibers. Fe and Al stabilize and strengthen the glassy silicate network of the fiber (18), and their dissolution weakens the network and so makes further dissolution possible.
Fibers A, B, C, D, and E were studied in one experiment and fibers F, G, H, and J in another. The chemical composition of fibers A and F were identical they were prepared by identical methods (see Material and Methods), but their surface areas differed from each other (Table 1). These fiber samples were intended to be used as standards in comparing the two different experiments; unfortunately, they were not optimal for this purpose.
The results of the present study are in agreement with those of Lundborg et al. (15)(16)(17), which indicate that AMs can also dissolve metal oxide particles in vitro. The differences in dissolutions in AM culture compared with that in the culture medium and in Gamble's solution, indicate the importance of also using in vitro cell culture methods for in vitro assessment of the solubility of MMVF.