Cytotoxicity of heated chrysotile.

Cytotoxicity and hemolysis were studied in chrysotile and quartz. The biological activity of the surface seemed to be different between chrysotile and quartz. Quartz lost its cytotoxicity on heating over about 500 degrees C. However chrysotile showed remarkable toxicity and induced hemolysis on heating between 650 and 800 degrees C, compared with the original unheated specimens. The mice injected intraperitoneally with minerals heated in this temperature range generally died within 48 hr after injection, while those injected with untreated chrysotile or chrysotile heated in the other heat ranges did not. The products in this range were highly disorded materials. It was assumed that the change of biological effects resulting from heat treatment may be related to the disordered state of chrysotile in the process of transformation into forsterite. The relationship between chemical character and cytotoxicity of the heated chrysotile specimens was also studied.

1,25(OH)2D3 is the active metabolite of vitamin D3 and its action is mediated by an intracellular receptor (VDR) which binds to DNA (Mangelsdorf et al., 1984;Reichel et al., 1989). The presence of the VDR in cells of the immunohaemopoetic system first indicated 1,25(OH)2D3 may have a role in regulating their activity. 1,25(OH)2D3 has been shown in vitro to have an antiproliferative effect and promote differentiation in monoblastic and promyelocytic cell lines, to inhibit differentiation in K562 leukaemia cells, to inhibit IL-1 production, to suppress CD4+T cell proliferation and IL2 production and inhibit immunoglobulin production by B cells (Jordan et al., 1990;Rebel et al., 1992;Tsoukas et al., 1989;Rigby et al., 1987;Moore et al., 1991;Iho et al., 1986;Tsoukas et al., 1984;Bhalla et al., 1989). In a number of cell lines the degree of response to 1,25(OH)2D3 appears to be dependent on the level of ligand binding (Chen et al., 1986) however the precise relationship is unknown. Furthermore 1,25(OH)2D3 has been shown to modulate expression of its own receptor (Lee et al., 1989;Strom et al., 1989). In actively proliferating lymphocytes VDR expression appears to be high and correlates with proliferative status except in B-cells in which expression is low (Kizaki et al., 1991). B cell lymphomas have been shown to be sensitive to vitamin D3. In a clinical trial, 1 lsg of l1,25OHD3 (a-calcidol) daily (metabolised to 1,25(OH)2D3 in the liver) produced an overall response rate of 30% in advanced follicular B cell non-Hodgkin's lymphoma (NHL) (Raina et al., 1991). The clinical importance of this finding is that it indicates vitamin D3 may have a role as a maintenance therapy in the setting of minimal residual disease in follicular NHL. Vitamin D3 analogues which have the antiproliferative effect of 1,25(OH)2D3 but lack its effect on calcium metabolism have been evaluated in a number of cell lines (Norman et al., 1990;Zhou et al., 1989;Binderup & Bramm, 1988;Colston et al., 1992). These could be of clinical value as they may enable higher dosages without calcium toxicity. The basis of the antilymphoma effect is uncertain. A problem in studying the biology of low grade lymphoma is the lack of suitable in vitro and in vivo models. Low grade lymphoma cells do not proliferate in vitro without adulteration for example by immortalisation with the Epstein Barr virus or culture with CD4+ cells (Umetsu et al., 1990) and such interventions clearly will change the nature of the cells studied. An in vitro model which may represent some aspects of follicular lymphoma cell biology are those cell lines which have a t(14;18) chromosomal translocation since this abnormality is found in at least 85% of cases of follicular NHL in association with rearrangement of the bcl2 gene (Yunis et al., 1987;Weiss et al., 1987).
The aim of this study was to investigate the antiproliferative effect and the induction of differentiation of vitamin D3 on t(14;18) lymphoma cell lines and relate this to expression of the VDR in lymphoma tumour biopsy material. 1,25(OH)2D3 and an analogue, MC903 (calcipotriol) which has equivalent VDR binding affinity but 100 fold less effect on Ca2+ metabolism were assayed.

Cells
The cell lines studied for their response to 1,25(OH)2D3 and MC903 were as follows: SU-DHL4 and SU-DUL5 (both derived from high grade B cell lymphoms): each carries a t(14;18) with an associated rearrangement of the bc12 gene; in SU-DHL4 rearrangement through the major breakpoint region (Cleary et al., 1986a) and SU-DUL5 rearrangement through the minor cluster region (Cleary et al., 1986b). U937 is a monoblastic cell line for which both 1,25(OH)2D3 and MC903 have been shown to inhibit proliferation and promote differentiation along the monocytic/macrophage pathway. SU-DHL4 cells were a gift from Dr A. Epstein (UCLA, Los Angeles, CA), SU-DUL5 cells were a gift from Dr M. Cleary (Stanford, CA). Cells were cultured in medium supplemented with 10% foetal bovine serum in humidified atmosphere with 5% C02. a stock concentration of 10-3 M and stored at -20°C and protected from light. Dilutions of the stock solutions were made in ethanol and then medium. The maximum concentration of ethanol in culture (0.1%) did not influence cell growth.
Modulation of cell proliferation and analysis of differentiation Cells in log phase growth were seeded at 2 x I0O cell ml-' and either agent (or vehicle = control) were added at the required concentration. Cells were counted daily using a Coulter counter and assayed for viability with Trypan blue. Each experiment lasted 5 days. On days 0 and 4 cells were examined using a panel of monoclonal antibodies to assay for differentiation. The experiment was performed in duplicate and repeated once. J,25(OH)2D3 receptor binding assay Cells were harvested by centrifugation and washed twice in ice cold phosphate buffered saline. Cells were then homogenised in KTMED (KC12 22.4gV1', Tris Cl 1.21 g1V', sodium molybate 2.06 g 1'-l, EDTA 0.336 g 1-, Dithiothreitol 0.62 g 1-' lI07 cells ml-' were sonicated and then centrifuged l00,000g for 1 h. One hundred and ninety jil of cytosol is then added to 10 ftl [3H]-1,25(OH)2D3 (2.6 x 10-8 M) either with or without excess radioinert 1,25(OH)2D3. After 4 h incubation receptor-bound [3]-1,25(OH)2D3 was separated from free [3H]-1,25(OH)2D3 with hydroxylapatite (Colston et al., 1980). Cytosol protein concentration was determined by the method of Bradford (Bradford, 1976). The experiment was performed in duplicate and repeated once.
Tumour samples and assay for VDR expression Tumour biopsy specimens were snap frozen and 5 ytm cryostat sections were mounted onto poly-l-lysine coated slides and fixed in 4% formaldehyde and methanol and washed in phosphate buffered saline. (Fixed sections were stored at -20°C in glycerol/sucrose storage medium before staining.) Sections were analysed for the presence of the VDR using a Vectastain ABC anti-rat alkaline phosphatase kit with the monoclonal antibody 9A7a (provided by Dr J. Wesley Pike). The alkaline phosphatase substrate contained a levamisole block. Sections were processed in duplicate. The MCF-7 cell line which has a high level of VDR expression served as the positive control.

Results
Effect of 1,25(OH)2D3 and MC903 on cell proliferation These agents inhibited the proliferation of SU-DHL4 and SU-DUL5 with no effect on cell viability (Figure 1) at 10-7 M. There was no effect on proliferation at lower concentrations. U937 was inhibited at lower concentrations of each as recorded elsewhere (Binderup & Bramm, 1988). At 10-7 M there was an approximately 50% reduction of proliferation of SU-DHL4 and SU-DUL5 on day 4. Effect of 1,25(OH)2D3 and MC903 on induction of differentation (Table I) SU-DHL4 acquired markers of mature B-cell differentiation (Bi, B4 and CD38) whilst SU-DUL5 lost a marker of immaturity (CD 14) when cultured with each agent at I0-7 M.
U937 has been demonstrated to differentiate along the monocyte/macrophage pathway when cultured in the presence of 1,25(OH)2D3 and MC903 (Dodd et Figure 1 Growth curves (means ± s.e.m.) for the SU-DHL4 and SU-DUL5 cell tines after culture in the presence of vehicle (control), 1,25(OH)2D3 and MC903 each at a concentration of I0-7 M. Cell viability was unaffected. There was no effect of either agent at lower concentrations (10-8 to 10-9M, data not shown).  Figure 2). Biopsy samples were analysed for VDR expression from 13 patients with various categories of NHL as defined by the Working Formulation. VDR expression was detectable in 11/13 samples. Only in a case of high grade lymphoblasticlymphoma was there strong staining comparable to the MCF-7 cell line. In the two negative cases, the macrophages stained positively providing an internal positive control. In all cases macrophages stained strongly positive as did cells in the paracortex which by morphology and location were considered to be T cells. node biopsy from a case of follicular mixed small cleaved and large cell lymphoma (working formulation category C). Note weak positive staining and occasional intensely staining cells which in the section shown are macrophages. c, A lymph node biopsy from a case of lymphoblastic lymphoma. The tumour cells are strongly positive.  (Kizaki et al., 1991) but was slightly altered ( ± 20%) by incubation with 1,25(OH)2D3 and not MC903 over 96 h. The reason for the apparent discrepancy between the two drugs in impact on VDR expression, particularly as they had a similar effect on proliferation, is unclear. The time-course of VDR modulation in response to each drug may differ and hence the response to MC903 may have been missed. The observation that 1,25(OH)2D3 produced a reduction in VDR expression in SU-DHL4 and an increase in SU-DUL5 on day 4 may also reflect different time-course functions of VDR modulation in each cell line. However these effects occurred at a relatively high concentration (10-M) of 1,25(OH)2D3 which exceeds the physiological concentration by approximately 103-104 fold. Furthermore expression of VDR in the follicular NHL tumour samples was low. These data suggest th.t nike observed clinical response of advanced follicuiar NHL to 1,25(OH)2D3 may not be due to a c'irect action of the agent on the lymphoma cells. CD4+T helper cells which recognise alloantigens expressed by follicular lymphoma cells induce the lymphoma cells to proliferate (Umetsu et al., 1990) indicating a possible role for CD4+T cells in promoting follicular NHL. 1,25(OH)2D3 inhibits CD4+T cell proliferation over a concentration range I0-'-10-1 M and this effect on CD4+T cells appears to be both direct and also indirect through suppression of ILl production by monocytes (Jordan et al., 1990;Tsoukas et al., 1989;Rigby, 1988;Binderup, 1992). Therefore the antifollicular NHL effect of 1,25(OH)2D3 may be mediated (at least in part) by an inhibitory effect on CD4+T cells. The development of the bcl2 transgenic mouse (McDonnell et al., 1989) provides a novel model of follicular NHL and may enable an in vivo analysis of the interplay between B and T cells under the influence of 1,25(OH)2D3 and its analogues. Certainly the t(14;18) cell lines are limited as a model of follicular NHL since they are derived from high grade B cell lymphomas. Nevertheless, if the antiproliferative and differentiationpromoting effect produced by 1,25(OH)2D3 on the SU-DHL4 and SU-DUL5 cell lines extends to lymphoma of follicle centre cell type, then using 1,25(OH)2D3 analogues it may be possible to achieve serum levels that act directly on the centrocytes, in addition to the postulated indirect mechanism of T cell inhibition, without perturbing calcium metabolism.
This work was supported in part by the Cancer Research Campaign.