Summary of discussion sessions: workshop on ingested asbestos

The anti-tumour activity induced by glucans (lentinan, yeast cell walls, pseudonigeran, dextran, DEAE-dextran and dextran sulphate) and fructosans (levan and carboxymethyl-levan) was compared with the activity of C. parvum. The following effects on tumour systems in CBA mice were assayed: (a) adjuvant activity on the immune response against tumour-specific transplantation antigens (TSTA) with a methylcholanthrene-induced fibrosarcoma; (b) cytostatic activity of peritoneal macrophages against radiation-induced leukaemia cells; and (c) inhibition of tumour nodule formation in the lungs following i.v. injection of fibrosarcoma cells. All the polysaccharides induced cytostatic macrophages, but the dextrans and levans did so only after i.p. and not i.v. injection. Only lentinan, yeast cell walls and pseudonigeran were active in the lung-nodule inhibition test; and only lentinan and dextran sulphate showed slight adjuvant activity for TSTA. It is concluded that the anti-tumour activity induced by these polysaccharides is predominantly non-specific macrophage-mediated and much weaker than that found with C. parvum. COR YNEBACTERIUMPARVUM is a potent stimulant of the mononuclear phagocytic system (MPS) and causes increases in spleen and liver weight (Halpern et al., 1964). It is also an immunological adjuvant (reviewed by Howard, Scott and Christie, 1973), and an inducer of anti-tumour activity (reviewed by Scott, 1974a). Various glucose polymers (glucans) share one or more of the biological activities of C. parvum (CP). Zymosan (yeast cell walls) stimulates the MPS (Benacerraf and Sebestyen, 1957), and increases resistance to tumour growth (Manowski, Yama-shita and Diller, 1957; Bradner, Clarke and Stock, 1958). The active component of zymosan as regards MPS stimulation is a glucan (Riggi and Di Luzio, 1961). Lentinan, another glucan of fungal origin, is an anti-tumour agent (Chihara et al. Dextran, DEAE-dextran and dextran sulphate also have anti-tumour effects (Ebbesen, 1974). Two mechanisms of anti-tumour resistance caused by CP have been distinguished: (1) The first follows systemic injection of CP and is not abolished by immunosuppressive procedures such as T-cell depletion (Woodruff, Dunbar and Ghaffar, 1973; Scott, 1974b) or irradiation (Bomford and Olivotto, 1974). It is therefore non-specific, independent of the host immune response to tumour-specific transplantation antigens (TSTA) and probably mediated by cytostatic macrophages (Olivotto and Bomford, 1974; Bomford and Christie, 1975). (2) The s.c. injection of CP mixed with irradiated tumour cells generates highly-specific resistance to tumour challenge in immunologically

dextran sulphate showed slight adjuvant activity for TSTA.
It is concluded that the anti-tumour activity induced by these polysaccharides is predominantly non-specific macrophage-mediated and much weaker than that found with C. parvum. COR YNEBACTERIUMPARVUM is a potent stimulant of the mononuclear phagocytic system (MPS) and causes increases in spleen and liver weight (Halpern et al., 1964). It is also an immunological adjuvant (reviewed by Howard, Scott and Christie, 1973), and an inducer of anti-tumour activity (reviewed by Scott, 1974a).
Two mechanisms of anti-tumour resistance caused by CP have been distinguished: (1) The first follows systemic injection of CP and is not abolished by immunosuppressive procedures such as T-cell depletion (Woodruff, Dunbar and Ghaffar, 1973;Scott, 1974b) or irradiation . It is therefore non-specific, independent of the host immune response to tumour-specific transplantation antigens (TSTA) and probably mediated by cytostatic macrophages Bomford and Christie, 1975). (2) The s.c. injection of CP mixed with irradiated tumour cells generates highly-specific resistance to tumour challenge in immunologically intact mice only (Scott, 1975;Bomford, 1975) which is a promotion of specific immunity to TSTA.
The objective of the present work was to analyse further the mechanism of the anti-tumour activities induced by glucans and fructosans, using the tests of nonspecific and specific activity devised for CP.

C. parvum
A killed suspension of CP (Coparvax) was provided by Wellcome Reagents Ltd, Beckenham, Kent.

Yeast cell walls
Commercial yeast cells (Saccharomyces cerevisiae) were washed several times with distilled water, fixed with a 4% v/v solution of formaldehyde in water, and washed with water, methanol, acetone, ether and benzene, and dried. The material was not chemically characterized. Glucans Lentinan.-Lentinan is a /(1-3) glucan of molecular weight about 106, obtained from the mushroom Lentinus edodes (Berk.) Sing. (Chihara et al., 1970 Pseudonigeran.-This a( 1-3) glucan was extracted from Aspergillus niger by the method of Johnston (1965). It was insoluble in water, yielded oligosaccharides of the nigerose series after acid hydrolysis, and contained less than 0 2% nitrogen.

Fructosans
Levan.-Levan, a /(2-6) and /3(2-1) linked polymer of fructose, was prepared from Corynebacterium levaniformis and characterized as previously described (Moreno, Courtenay and Howard, 1976). It was soluble in water, with an average mol. wt of about 2 x 107. Carboxymethyl-levan (CM-levan) was prepared by direct coupling with chloroacetate (Inman, 1975 Tests for anti-tumour activity Macrophage cytostasis.-At various times after either i.v. or i.p. injection of 0-2 ml of CP or polysaccharides in saline, peritoneal cells were harvested and monolayers of macrophages tested for inhibition of RI leukaemia cell DNA synthesis as previously described . Peritoneal cell suspensions were adjusted to 2 x 106 cells/ml, and 2 ml were placed in 30-mm Sterilin plastic Petri dishes in Dulbecco's modification of Eagle's medium with 10% foetal calf serum. After 2 h incubation at 37°C in a CO2 incubator, the non-adherent cells were removed by vigorous and repeated pipetting and washing, and 2 ml fresh medium was added. The number of cells remaining attached to one dish from each group was counted, using a grid eyepiece with an inverted microscope. It was usually about 2 x 105 cells. Dishes with macrophages from treated mice were discarded if the total of macrophages was not within the range + 15% of the total of normal macrophages. 105 syngeneic RI leukaemia cells in 0-2 ml Dulbecco's medium were added to the macrophage cultures, which were pulsed (for 1 h) with [3H]TdR 16 h later, as previously described .
Cultures containing macrophages alone incorporated negligible amounts of [3H]TdR.
Lung nodule inhibition.-Mice treated with CP or polysaccharides as above were injected i.v. with 2 x 104 T3 fibrosarcoma cells, and lung nodules counted 14 days later .
Potentiation of specific immunity.
-5 x 105 irradiated (10,OOOR from a 137Cs source) M4 methylcholanthrene-induced CBA fibrosarcoma cells (Bomford, 1975) in 0-05 ml saline, alone or admixed with CP or polysaccharides, were injected s.c. into a hind footpad. Seven days later, 105 living M4 cells were injected into the contralateral footpad. Tumour growth was monitored by measuring footpad thickness with a dial gauge caliper 42 (Schnelltaster, H.C. Kroplin GmbH, Hessen, West Germany). Experiments were terminated 30 days after injection of the living cells. Mice whose footpad thickness had reached > 3 mm were considered to have developed tumours, since at this size progression was inevitable.

Statistics
The significance of differences in mean footpad thickness between groups of mice, and of mean incorporation of [3H]TdR in the cytostasis test was assessed by Studeint's t test, and of differences in lung nodule numbers by the Mann-Whitney U test. In each case significance corresponded to P < 0.05.

RESULTS
Splenomegaly, cytostatic macrophages and lung nodule inhibition CP was injected 5 to 14 days before measurement of spleen weight, testing the cytostatic activity of macrophages, or i.v. injection of tumour cells for lung nodule inhibition. These times were chosen because both tumour cell cytostasis and lung nodule inhibition are maximal 5 days after the i.v. injection of 350 ,tg of CP Bomford and Olivotto, 1974). Splenomegaly is maximal about 14 days after i.v. injection of a wide range of doses of CP (Adlam and Scott, 1973). Yeast cell walls, lentinan and pseudonigeran also induced the appearance of cytostatic macrophages at Day 5, and at Day 14 also, with the exception of yeast cell walls (Fig. 1). In the case of the dextrans and levans, cytostatic macro-  phages were only obtained from mice injected i.p. rather than i.v. (Table II). At Day 5, both DEAE-dextran and dextran sulphate were more active than neutral dextran. In all these experiments, the polysaccharides were consistently less active than CP. The number of lung nodules developing after i.v. injection of 350 ,ug of CP or 1-3 glucans is shown in Fig. 2. At Day 5 lentinan and yeast cell walls caused significant inhibition, and at Day 14 only pseudonigeran did so.
A temporal correlation between cytostasis by peritoneal macrophages and lung nodule inhibition is apparent when the data from Figs. 1 and 2 are presented together as percentage inhibition of leukaemia cell DNA synthesis or of lung nodules (Fig. 3).
No lung nodule inhibition was observed after i.v. or i.p. injection of 350 ,g of the dextrans (data not shown).

Specific anti-tumour immunity
Mice were injected s.c. in the footpad with 5 x 105 irradiated M4 tumour celis alone, or admixed with 1, 10, or 100 jug of CP or polysaccharides. Seven days later they were challenged with 105 living M4 cells in the contralateral footpad. Table III shows the proportion of mice developing tumours, and the size of the tumours expressed as average footpad diameter 30 days after challenge. No protection was conferred by irradiated M4 alone. The addition of 1 or 10 ug (but not 100 ,ug) CP resulted in complete protection. The relative inefficacy of larger doses of CP in combination with irradiated cells has been reported before (Scott, 1975;Bomford, 1975  10 ,ug 7 *0±0 8 100 ,g 3-8±0-6* * Jn(licates P < 0 * 05 relative to irradiated M4 only.

DISCUSSION
The activities of the polysaccharides studied are summarized in Table IV. All induced cytostatic macrophages. Lentinan, yeast cell walls and pseudonigeran were also active in the lung-nodule inhibition test, which is considered to be mediated by a non-specific mechanism . Although there is a temporal correlation between the presence of cytostatic macrophages in the peritoneal cavity and lung nodule inhibition after CP injection (Bomford and Olivotto, 1973), evidence is still awaited of a common effector basis for these phenomena. The parallel between the presence of cytostatic macrophages and lung nodule inhibition also held for lentinan, yeast walls and pseudonigeran in the present study, which further strengthens the case for suggesting a causal relationship between them.
Only lentinan and dextran sulphate amongst the polysaccharides displayed even marginal adjuvant activity for TSTA. On the basis of the tests used in this study, therefore, it is concluded that the anti-tumour action induced by the polysaccharides is predominantly non-specific.
We consider to what extent this conclusion is compatible with existing knowledge of the anti-tumour effects of these materials in other systems, and of their adjuvant properties. Previous anti-tumour studies on lentinan (Chihara et al., 1969(Chihara et al., , 1970, zymosan (Manowski et al., 1957;Bradner et al., 1958), dextrans (Ebbesen, 1974) and levan (Leibovici et al., 1975) were aLl performed using the i.p. route of injection which, from the results of this study, might be expected to have induced cytostatic macrophages within the peritoneal cavity. The studies on zymosan and levan showed retardation of growth of transplantable tumours inoculated s.c. or i.p., but did not analyse its mechanism any further.
However, the anti-tumour effect of repeated i.p. administration of lentinan against the s.c. growth of Sarcoma 180 in mice (Chihara et al., 1969(Chihara et al., , 1970 was not found in neonatally thymectomized mice . Although this might suggest that in this svstem lentinan stimulates specific immunity to TSTA, the data on the adjuvanticity of lentinan do not support this contention. Multiple i.p. injections of lentinan given after i.v. injection of sheep red blood cells (SRBC) stimulated the humoral response in normal, but not in T-celldeprived mice (Dresser and Phillips, 1974), but it seems unlikely that i.p. lentinan would modify the humoral response to an s.c. tumour growth. Lentinan also failed to stimulate T-cell cytotoxicity in an allogeneic system of the DBA/2 P815 mastocytomas in C57BL/6 mice (Dennert and Tucker, 1973). An alternative explanation for the T-cell dependence of the anti-tumour effect of lentinan is that the simultaneous influence of cytostatic macrophages and a normal immune response is required, either alone being inadequate.
The effects of multiple i.p. injections of dextran, DEAE-dextran and dextran sulphate have been tested in two systems, the spontaneous leukaemia of AKR mice, and Rauscher leukaemia-virus-induced leukaemias in BALB/c mice (Ebbesen, 1974). If macrophages were effective in these systems, one might have predicted from the present results that all three dextrans should have shown some activity. However, in the AKR system dextran had no effect, DEAE-dextran improved survival, and dextran sulphate accelerated tumour development. In the BALB/c system, dextran and DEAE-dextran prolonged life when treatment started at the time of palpable spleen enlargement, whereas only dextran and dextran sulphate protected when injections started from the time of infection. Clearly factors other than cytostatic macrophages must be involved in these systems. It was suggested that the differential effects of the dextrans on the spread of virus or on the humoral response to it may play a role (Ebbesen, 1974).
Dextran sulphate (but not dextran) injected i.v. before i.v. immunization with SRBC potentiates the humoral response (Diamantstein et al., 1971;Bradfield et al., 1974). The failure of dextran sulphate to inhibitlung nodule formation in our studies is not surprising, as the mechanism does not involve an immune response to tumour cells (see above). Dextran sulphate potentiates killer T cells in the allogeneic P815 mastocytoma system (Vachek and Kolsch, 1975), which might explain why it showed a modest adjuvant effect for TSTA in the present study. CP (McBride et al., 1975), lentinan (Okuda et al., 1972), dextran sulphate and levan (Pryjma, Humphrey and Klaus, 1974) all share the property of activating complement by the alternate pathway. Subsequently a plausible common mechanism for the induction of cytostatic macrophage by CP and polysaccharides has been provided by the finding of Schorlemmer, Davies and Allison (1976) that activated complement components 46 induce lysosomal enzyme release from macrophages.