Comparative toxicities of different forms of asbestos on rat pleural mesothelial cells.

The effects of UICC crocidolite and chrysotile A, either oxalic acid-leached or unleached, on the viability, morphology and growth characteristics of rat pleural mesothelial cells (PMC) were examined; DQ12 quartz particles were also used. When asbestos fibers were added for 48 hr at the beginning of exponential growth, 20 or 50 micrograms/mL of chrysotile fibers were cytotoxic and no growth occurred; with 5 or 10 micrograms/mL a latent period was observed, and the mean population doubling time was increased. Chrysotile ingestion was associated with morphological changes (spreading, intense vacuolation); moreover, a large proportion of the cells was binucleated (more than 30% with 10 micrograms/mL). The oxalic acid-leached chrysotile inhibited growth at a concentration of 50 micrograms/mL; with 5 or 10 micrograms/mL, no spreading occurred, but a shrinkage of some cells was observed. A few large vacuoles were seen in the cytoplasm of the cells; there were fewer binucleated cells. Addition of 5 or 10 micrograms/mL of crocidolite, either leached or unleached, did not significantly change the growth rate, in spite of the presence of a large number of fibers inside the cells which persisted when the cells reached confluency. With 20 or 50 micrograms/mL, the mean population doubling time was increased in a dose-dependent manner. A slight vacuolation of the cells occurred. The sample of quartz did not modify the parameters studied in this report. The results confirm the different in vitro reactivities of the two kinds of unleached asbestos fibers. Leaching of chrysotile fibers decreased their reactivity; alternatively, leaching of crocidolite increased the effects on PMC. ImagesFIGURE 1. (a)FIGURE 1. (b)FIGURE 2. (a)FIGURE 2. (b)FIGURE 3.FIGURE 4.FIGURE 5.


Introduction
When crocidolite or chrysotile fibers are injected into the pleural cavity of rats, they both induce mesothelioma (1). However, when tested with in vitro systems, these fibers give different responses (2). In these in vitro experiments, the cells used did not come from the pleura. The aim of the work reported here was to study the in vitro effects of asbestos fibers on the morphology and grQwth characteristics of rat pleural mesothelial cells in short-term experiments.
Since acid-leached chrysotile fibers have been found to be nontumorigenic (3,4), leached chrysotile and leached crocidolite were also tested. In order to *  compare the results with the activity of noncarcinogenic particles, a sample of quartz was also used.

Asbestos Fibers and Control Particles
The asbestos fibers were obtained from the UICC. Chrysotile A and crocidolite were used. Chrysotile and crocidolite were leached with 0.1 N oxalic acid (5). DQ12 quartz from Doirentrup was chosen as a control. The particles were dispersed by sonication in the culture medium (50 kHz, 20 W, The size parameters of the samples were determined by electron microscopy as described elsewhere (4). The leached chrysotile had a slightly higher mean diameter than the unleached samples (0.091 and 0.082 pm) and a lower mean length (1.8 and 3.1 ,m). Rat Pleural Mesothelial Cells (PMC) PMC were cultured as described elsewhere (6). The cells were used between the 6th and 17th passage. The PMC came from six different rats.

Morphological Studies
PMC morphology was observed by use of phasecontrast microscopy and electron microscopy as described by Jaurand et al. (7).

Growth Analysis
The growth curve analysis was performed as described elsewhere (7). Briefly, PMC were plated at 5 x 105 cells per flask, and 24 hr after plating the particles were added to the culture medium for 48 hr. The difference between the number of treated and untreated cells was assessed by the F test. The experiments were carried out in triplicate or quadruplicate for each particle type. The percentage of mitosis or of binucleated cells, following either 48 hr or 72 hr of incubation with the particles, was determined. The difference between the percentage observed in treated and untreated cells was assessed by the E test.

Morphological Studies
The results describing the effects of chrysotile and crocidolite fibers have been reported elsewhere (7). Chrysolite-treated PMC had numerous vacuoles; there was an increase in the cell volume, and a large proportion of binucleated cells appeared (Table 1). Crocidolite fibers did not induce such morphological changes except at the highest concentration (50 pg/mL), but these effects were less than with 5,g/ mL) of chrysotile fibers.
PMC treated with acid-leached chrysotile also showed a vacuolation of the cytoplasm but the vac-uoles were less numerous and generally larger than with unleached chrysotile (Fig. 1). Shrinkage of the cells was sometimes observed. After 72-hr contact with acid-leached fibers, some binucleated cells were observed, but their percentage was less than with the untreated chrysotile ( Table 1). The difference between the number of binucleated cells in treated and untreated cultures was significant with 10 l g/mL of leached chrysotile fibers. With the unleached fibers, it was significant with 5 ,ug/mL; moreover, it was twice the percentage observed in the leached chrysotile-treated culture. Thus, the effect of leached fibers was less than the effect of the unleached.
Very few vacuoles were seen in the cultures treated with leached crocidolite (Fig. 2). Electron microscopy revealed the presence of intracellular fibers (Fig. 3). The number of binucleated cells was less than in the cultures treated with either leached or unleached chrysotile (Table 1). However, the leached sample of crocidolite had more effect on PMC than the unleached sample, since 10 ,ug/mL induced a significant increase in the proportion of binucleated cells as compared to the control cultures. Treatment with 50 ,ug/mL of crocidolite fibers was necessary to obtain a significant increase of this parameter.
The particles of quartz did not modify the morphology of the culture (Fig. 4), in spite of the presence of intraphagosomal particles (Fig. 5).
When the proportion of binucleated PMC was examined, a differential effect of leaching was observed as the proportion of binucleated cells decreased with leaching of chrysotile and increased with leaching of crocidolite. Table 2 shows that there was no effect, due to the treatment with the particles, on the percentage of mitosis determined 48 hr after their addition. However, during this time, the cells were in contact with the particles. This result means that in the five groups of treated PMC there was no inhibition of passage to the M phase and that the toxicity may

Growth Analysis
The results concerning the mean population doubling time 0 are expressed in Table 3. DQ12 quartz had no significant effect on the value of 0. Leached crocidolite and chrysotile, either leached or unleached, were cytotoxic, but the concentration needed to prevent growth depended on the nature of the fibers. Unleached crocidolite had some effect if added at the highest concentration (50 ,ug/mL).  (0) 5 pg/mL; (0) 10 pg/mL; (A) 20 pg/mL; (A) 50 pg/mL.
The growth curves showed that a lag time occurred with the chrysotile-treated cultures but this was not observed with leached-chrysotile or with crocidolite, either leached or unleached (Fig. 6).
It is known that the number of cells N during the exponential growth is N = No 20 (1) where No is the number of PMC at the beginning of growth, t the time and 0 the population doubling time. Then the derivative of Eq. (1) may be written dN/dt = 1/0 (N log 2) (2) When the particles are introduced into the medium, an inhibition can occur; it may be proportional to a power (x) of the particle concentration c, and it causes a decrease in the number of PMC: where k is a constant. Then, the variation in the number of PMC will be dN/dt = N(1/0 log 2 -kcx) (4) If 9' is the mean population doubling time in the presence of the particles, the variation of 9' with c can be written 0' = a9/(acx) (5) In this relationship, when c --0, then 0 '-0; and when cx-a, 0'--om. From the value a it is possible to deduce allx, which is the value of the concentration at which no growth occurs (Clin).
From the values reported in Table 3, the calculation gives the values of cim expressed in Table 4. When the particles are compared by weight, the toxicity will increase in the order Cr < LCr < LCh < Ch, where Cr, Ch are crocidolite and chrysolite, Our results are in good agreement with previous data reported by Neugut et al. (8) and Reiss et al. (9) using chrysotile and amphiboles on various cell lines. These authors found a higher toxicity of chrysotile than of crocidolite. It seems that proliferative epithelioid cell lines react to asbestos treatment by somewhat similar responses, possibly related to the amount of particles ingested. Reiss et al. (9) found that the toxicity of the leached samples of amphiboles increased, whereas that of chrysotile decreased with leaching. A similar result is reported here. It is not known if the batch of leached chrysotile used for the PMC treatment was carcinogenic. However, other leached samples did not induce mesothelioma (3,4), suggesting that this cytotoxicity was not related to the probability of causing mesothelioma.
There were some differences in the size of the fiber samples, as determined by electron microscopic and optical measurements, and the repartition of the number of fibers with size is different from one sample to another one. The results may have to be reconsidered taking this observation into consideration.
Another way to compare the effects of the samples used in this study was to determine the time, . Level at which the number of treated cells was lower than that of the untreated cells in all of the cell strains used to determine the growth rate (see Table 3). For example, with chrysotile fibers, the difference was significant following 24 hr of incubation with a concentration equal to or higher than 20 Ag/mL, or following 48 hr with a concentration equal to or higher than 5 Ag/mL. for a given concentration, where the value of F, which measures the difference between the number of PMC, in treated cultures and the number of PMC in the control culture is such that p < 0.5. For each treatment, the F test was applied every day, and the results are expressed in Figure 7. It can be seen that chrysotile and leached crocidolite were both effective at a concentration of 5,g/mL, but the time necessary to observe a significant effect was longer with leached crocidolite than with chrysotile. This may be due to similar "toxic power" of the samples and to a different sensitivity of the mesothelial cells (expressed by the lag time and possibly related to the intensity of phagocytosis). With the leached samples, no significant effect was observed after 24 hr of contact with the fibers, and with 10 ,ug/mL, the relationship was Cr < LCh < LCr < Ch. When considering the effects on growing cells, the comparison between different varieties of fibers must be made carefully, since an effect may be reversed due to the form of the dose-effect relationship.