Studies on in vitro chrysotile-pleural mesothelial cell interaction: morphological aspects and metabolism of benzo-3,4-pyrene.

Cultures of rat pleural mesothelial cells (PMC) were exposed to nonlethal doses of UICC chrysotile A. The morphology was studied by optical and electron microscopy. The consequences of chrysotile ingestion on the rate of pinocytosis of horseradish peroxidase (HPR) metabolism and benzo-3-4-pyrene (BP) were studied. Nonlethal doses of chrysotile (5 micrograms/mL) induced a time-dependent vacuolation of PMC; a dose-dependent inhibition of the vacuolation was observed when PMC were pretreated with DMSO. The origin of the vacuoles is not clear, but some features of autophagy and lysosomal storage were observed. Chrysotile fibers did not modify the rate of pinocytosis of HRP. Similarly, the metabolism of BP was unchanged when BP and chrysotile were both added to the culture medium or when PMC were preincubated with the fibers 24 hr prior to the addition of BP.


Introduction
In vitro studies performed with various cell lines have demonstrated that chrysotile fibers are cytotoxic (1). However, the effect is dose-dependent, and low concentrations do not kill the cells but induce some pathological features. The aim of this work was, first, to study the morphology of rat pleural mesothelial cells, (PMC) treated in culture with nonlethal concentrations of chrysotile fibers and, secondly, to examine the consequences of chrysotile ingestion on the metabolism of the PMC with regard to pinocytosis of exogenous macromolecules (horseradish peroxidase, HRP) or polycyclic hydrocarbons (benzo-3-4-pyrene, BP). *

Material and Methods
Cell Culture PMC were cultured in NCTC 109 + 10% fetal bovine serum as described by Jaurand et al. (2). The cells were used at confluency between the 8th and 20th passage.

Chrysotile Fibers
UICC chrysotile A fibers were used. The fibers were dispersed by sonication in the culture medium (20 kHz, 20 W, 5 min).

Cell Viability
The cell viability was determined by trypan blue exclusion and also by measuring the lactate dehydrogenase (LDH) released in the culture medium as described elsewhere (3).

Phase Contrast Microscopy
The enumeration of vacuoles was carried out by using phase contrast microscopy (x 370). The fields were randomly selected, and 100 cells crossing a line in the eye piece were counted. The treatment of PMC was unknown by the reader in order to avoid subjective results.

Electron Microscopy
Cells were prepared for electron microscopy as described elsewhere (2). The acid phosphatase staining was performed on cells fixed with 2.3% glutaraldehyde in 0.045 M cacodylate buffer and stained by the technique of Miller and Palade (4). The HRP was detected by the technique of Graham and Karnovsky (5).

Spectrophotometry for HRP
The HRP activity was measured by using the assay of Steiman and Cohn (6). No endogenous peroxidase was found in PMC.

Metabolism of BP
14C-BP was added to the cells either untreated or treated with 2 ,g/mL of chrysotile fibers. Both BP and chrysotile were added to the medium or PMC was preincubated with chrysotile for 24 hr prior to BP addition. After 24 hr of incubation with BP, the culture medium was extracted with ethyl acetate. The organosoluble metabolites were separated by HPLC and the proportion of conjugates, in the aqueous phase, was also determined.

Results
In order to incubate the cells with a nonlethal concentration of chrysotile fibers, a dose-effect relationship was established. The results are expressed in Table 1 and show that 5,g/mL did not highly modify the viability of the PMC. The effect of BP was also studied, and no modification of the growth curves was observed with 1 ,pg/mL of BP. Morphological Studies Addition of 5,g/mL of chrysotile fibers to confluent PMC resulted in the appearance of regular  (Fig. 1). Their number was time-dependent and after 20 hr of incubation 80.2 ± 20% of PMC contained vacuoles (mean obtained with four different cell strains) (Fig. 2). The mean number of vacuoles per PMC was 3.0 ± 2.6 and 6.6 ± 3.7 after 8 hr and 20 hr of incubation with the fibers, respectively (Fig. 3).
Previous studies (7) showed that chrysotile fibers were ingested by PMC in culture and that degranulation of lysosomal enzymes occurred into the phagocytic vacuole. Following 2 hr of incubation with chrysotile fibers, electron microscopy revealed both the presence of phagocytic vacuoles containing chrysotile fibers and large regular vacuoles which were generally empty or contained traces of membrane whorls ressembling myelin figures (Fig. 4). Acid phosphatases were present not only in the phagosome ( these structures, suggesting that they were secondary lysosomes. In some cells cytoplasmic inclusions could be seen which had the appearance of lipid inclusions and were either highly ordered crystalloid or lamellated structures. These inclusions were contained by a single membrane.

Effect of DMSO
As BP was solubilized in DMSO (0.1%), the effect of DMSO was observed. DMSO was added to the culture medium 1 hr prior to the addition of chryso-  Figure 2. Numerous vacuoles are seen, generally empty but sometimes with membrane whorls (arrow). The chrysotile fibers were generally in phagosomes (head of arrow) but were sometimes present inside the large regular vacuoles. x 5300.
tile. There was a dose-dependent inhibition of the vacuolation of PMC. In Figure 6, the variation of the mean number of vacuoles per PMC with the DMSO concentration is reported.

HRP Pinocytosis
The spectrophotometric assay revealed that HRP entered the cells when incubated with PMC. The amount of HRP ingested was timeand dose-dependent. The addition of 5 Mig/mL of chrysotile fibers did not change the pinocytic rate (Table 2).
Pinocytic vesicles were also demonstrated by electron microscopy (Fig. 7). This revealed that HRP and chrysotile fibers were generally in separate vacuoles when both materials were incubated with PMC.

Metabolism of BP
BP was metabolized by PMC in culture. PMC were used at the 7th passage. After 24 hr of incubation 60% of the BP was metabolized. The propor- tions of metabolites in the aqueous and organic phases were 60 and 40%, respectively. When PMC were incubated with both 2 j*g/mL of chrysotile fibers and BP, the fraction of BP metabolized was not significantly different; similarly, the proportion of metabolites obtained in the aqueous or organic phases was unchanged. No modification of the proportion of water-soluble or organo-soluble radioactivity was found when PMC were pretreated with crystotile fibers (Table 3).

Discussion
The results reported here clearly demonstrate that nonlethal concentrations of UICC chrysotile A fibers induce morphological abnormalities in rat pleural mesothelial cells in culture. As far as the rate of pinocytosis of HRP and the metabolism of BP are concerned, the metabolism of PMC was unchanged.
This intense vacuolation of mammalian cells has been observed by others when cells were exposed to chloroquine (8,9). The pictures shown by Wibo and Poole (10) of fibroblasts exposed to chloroquine are very similar to those observed with PMC. Some results reported here show that features of autophagy and lysosomal storage can be observed. However, large empty vacuoles are different from autophagic vacuoles; their significance is not clear, but it could be due to the storage of undigestible or undigested material conducive to an osmotic effect. The mechanism is not defined, and further studies must be performed in order to explain this effect. Such studies are of particular interest, since crocidolite fibers, in spite of being ingested, did not induce such an effect (11). The inhibition of vacuolation observed after DMSO treatment is not clear; it could be due to a change in the physical state of the membranes since it is known that DMSO increases the transition temperature of some phospholipids (12). It is not due to an inhibition of phagocytosis, since numerous intracellular chrysotile fibers were identified by electron microscopy.
It is interesting to note that HRP pinocytosis is not increased during chrysotile exposure. This result indicates a distinction between the sites of at-   (13). Other authors have studied the effect of asbestos fibers on the metabolism of polycyclic hydrocarbons. Our results do not show an increase in the metabolism of BP as observed by Mossman et al. (14) using methylcholanthrene (MCA) and crocidolite on tracheal organ cultures. The difference could be due to the fact that MCA was adsorbed on the fibers, to the origin of the cells or to the nature of the fibers.
Daniel et al. (15) found a similar HPLC profile for untreated and chrysotile-treated human fibroblasts except for the appearance of an early eluting polar FIGURE 7. Electron microscopy of a PMC incubated with 0.5 mg/mL of HRP for 2 hr showing the pinocytic vacuoles (arrows). x 20200. material observed when cells were treated with chrysotile 24 hr before the hydrocarbon was added. The stability in those aspects of metabolism described here confirmed that the viability of chrysotile-treated PMC was similar to the control cells.