Sie sind auf Seite 1von 9
"Asian Pacific Journal of Alrgy and immunology (1806) 12: 129-137 Antitumor Mechanisms of Eubacterium lentum and Its Components Mochammad Hatta The intestinal microflora have several effects on the host and many studies have been performed on the relationship between tumors and intestinal microflora.!3 However, no definite correlations have been identified between them, but it is possible that the cooperation of some species of normal flora may affect the promotion of inhibition of tumors in the host. We have studied the amitumor activity of bacteria in the intestinal microflora of humans and animals, and re- ported amtitumor activity of several bacterial species;+ our report was the first on the antitumor activity of one of them, Eubacteriuim lentur, which is a well-known anaerobic, gram positive, short rod form bac- terium and natural component of human intestinal flora. Most studies on antitumor activity of various bacteria have reported that their actions were mediated by the activa- tion of immunological effector cells and production of cytokines in the host after administration! We reported in a previous paper that £, entum (TYH-11) ap- parently had adjuvant effects and hhad a strong antitumor effect on 11 strains of experimental tumor cell SUMMARY In the present study, some antitumor mechanisms of Eubacterium Jentum (TYH- 11) and bacterial components having antitumor effects were investigated. Eertum induced maximum NK cell activity in C3HHe mice on day 1 after injection (80.6% against 33.99% of control at E:T ratio 50:1) and the activity was kept at a level of 48.6% on day 7. Tumoricial peritoneal macrophages w ere Induced 9 days after Elenium injection into BALB{c mice (58.2% against 10.1% control at E:T ratio 10:1). Tumoricidal macrophage activity persisted atthe same level for atleast 11 days. Cyto- toxle T lymphocyte (CTL) activity was induced only in tumor bearing mice treated with Erentum, 4 weeks alter tumor inoculation. Antitumor acthity was observed in the cell wall (CW) and membrane fractions (CM) of Elentum. CW induced NK cel activity; the ‘activity was transient while the kinetics of NK activly by CM showed 2 peeks, on day 1 and day 7. Tumoricidal macrophages were induced by CW and the activity level was the same as that induced by whole body, while thet Induced by CM was ata lower level ‘Nolther CW nor CM induced CTL in tumor bearing mice. lines, but showed no direct cytotoxi- used for experiments were main- city, and suggested that E. lenfum tained as follows: Ehrlich ascites might be a biological response tumor cells were maintained in ICR modifier (BRM).!2 Therefore, in mice, L1210 lymphoid leukemia and this paper, mechanisms of antitumor 815 mastocytoma cells in DBA/2 activity of E. lentum were studied mice and EL~4 lymphoma cells in and cellular components of the ba. C57BL/6 mice by successive ascites cillus having an antitumor effect were investigated, MATEaIALA AND merzoog Sot otr‘tusctaetuatae Arima nd tumor ces ra Yar Truro Six to 8 week old ICR, CS7BL/ Correspondence : Mochammad Hatta, De- 6, BALB/c, DBA/2, and C3H/He partment of Medical Microbiology. Faculty Crake, lp) acon ie Bs TESS a 130 HATTA passage. YAC-I lymphoma cells, P81S mastocytoma cells and EL~4 lymphoma cells were also cultured in vitro in RPMI 1640 medium (Nissui Pharmaceutical Co., Tokyo, Japan) supplemented with 10% feral calf serum (RPMI-CFS) (Flow Laboratories, North Ryde, Aus- tralia), 100 U penicillin/ml, 100 ug, streptomycin/ml, 0.25 ug ampho- teticin B/ml, 1 mM sodium pyru- vate, 0.1 mM non-essential amino acids, and 2 mM glutamine. Agents E.lentum (TYH-I1) was cule tured in GAM broth (Nissui) under anaerobic conditions for 18 hours at 37°C, killed with 0.3% formalin and washed with sterilized saline. ‘OK-432 was kindly provided by Chugai Pharmaceutical Co., Ltd. (Tokyo, Japan). Preparation of effector cells NK cells were separated from spleens of C3H/He mice according to the method of Kumagai ef al. Single cells of mice given 107 cells of formalin killed E.lentum, 300 us cell wall fraction (CW) or 1,000 uz membrane fraction (CM) were suspended in RPMI 1640 culture medium containing 10% of fresh autologous serum and $-10% of KAC-2 (Silica suspension as mono- cyte removal agent, Ohtsuka assay, Tokushima, Japan) and incubated for 45 minutes at 37°C, Then, 6 ml of cell suspension was overlaid onto 3 ml of Ficoll Paque (Pharma- cia, New Jersey, USA) and centri- fuged at 3,000%¢ for 30 minutes to separate lymphocytes from KAC-2 phagocyting macrophages, and removed the adherent cells by a nylon column. Macrophages were obtained as, follows. Peritoneal exudate cells (PEC) of BALB/c mice administered 3 times with 107 cells of formalin killed E.lentum were collected by ‘washing the peritoneal cavity with 7 ml Hank's balanced salt solution (HBSS, Nissui). After centrifuga tion at 3000%g for 7 minutes, PEC were resuspended in RPMI-FCS and were cultured in plastic dish for 2 hours at 37°C in a $% CO at- mosphere. Adherent cells were col- lected with cell scraper (Greiner, Nurtingen, Germany) and used as macrophages. Cytotoxic T lymphocytes (CTL) Mice were injected iv with 107 cells of killed E.lentum everyday for 7 days. CTL were obtained from spleen of ICR mice inoculated with 106 cells of Ehrlich ascites tumor subcutaneously (sc). Besides, CS7BL/6 mice (H-2) were im- ‘munized with 4x 106 spleen cells of BALB/c mice (H-24) intravenously (iv), and injected with $x 108 cells of L1210 (H-24) intraperitoneally (ip) on day 21. A week later, CTL in. spleen of CS7BL/6 mice was collected." Plastic adherent cells were removed from single spleen cells and lymphocyte fractions were separated by centrifugation on Ficoll-Paque. Effector cells from 5 mice in all experimental groups were pooled. Cytotoxicity assay Cytotoxicity of NK cells was measured using YAC-1 lymphoma cell line as a target. Approximately 107 YAC-1 cells in 1 ml RPMI-FCS were incubated with 3.7 MBq of NaySICrOq (Daiichi Radioisotope, Tokyo, Japan) for 1 hour at 37°C in 5% COz atmosphere. The cells were washed with REMI-FCS, and resuspended in 1 ml of RPMI-FCS supplemented with 10% HEPES buffer (Wako Pure Chemical Indus- tries Lid., Osaka, Japan). One hundred microltres of effector cell suspensions at different concentra- tions were placed in round-bottomed wells of microplates (Corning, Iwaki Glass, Tokyo, Japan) in triplicate, followed by addition of 100 ul S1Cr- labeled YAC-1 cells. Plates were incubated for 4 hours at 37°C in a 5% CO atmosphere; maximum SiCr-release was measured using target cell lysate in HCL Culture medium was collected by Super- natant Collection System (Skatron, Lier, Norway) to determine S!Cr- release, which was determined by the following formula : Experimental ~ Spontaneous release release Maximal Spontaneous release release Activated macrophage activity was determined by 5ICr-release from labeled target cells (EL-4 or P815), after effector cells were ccultured with target cells for 20-22 hours at 37°C in a $% CO hum ified incubator. CTL activity in vitro was measured by the same method as NK activity. Ehrlich ascites tumor cells and L1210 cells were used as specific target cells. In all experiments, spontaneous release was less than 20%. CTL activity in vivo against H-24 was determined by the number of living L1210 cells in the peritoneal cavity of CS7BL/6 mice on day 28. Mice were sac ficed under anesthesia and L1210 cells were collected from the pe oneal cavity with 10 ml RPMI-1640. The number of living 1.1210 cells judged by the trypan blue method was counted by blood cell coumter.!4 Fractionation of tumor activity measurement of fractions, Fractionation was performed by the method of Azuma er al. E.lentum was broken up by MINI~ LAB (Rannie, Denmark) under 900 bar. By centrifugation at 20,000 for 1 hour, the crude cell wall frac- tion and the membrane fraction (Fraction 2, Fr2) were separated. ‘The crude cell wall fraction was resuspended in 0,07 M phosphate buffer pH 7.8 containing 10% trypsin (Wako) and chymotrypsin (Merck, Darmstadt, Germany), and digested at room temperature for 24 hours. After centifugation at 20,000x¢ for 1 hour, precipitate ANTITUMOR EFFECTS OF ELENTUM was digested again with pronase (Kaken Pharmaceutical Co, Tokyo, Japan), dissolved in 0.01 M tis HCl-butfer pH 7.2 for 24 hours The sediment was collected as the cell wall fraction (Fr1). In this process, 4 subfractions were also obiained (Frl),2,3,4)- Each frac- tion of E.lenruma Was examined for antitumor activity against Ehrlich ascites tumor calls. ICR mice were inoculated se or ip with 106 or 105 cells of Ehrlich ascites tumor, res- pectively. Fractions were injected intratumoral (it), iv oF ip everyday for 7 days for ascites and solid form, respectively as indicated in. the tables. Tumor weight was calculated by using the following formula : ‘Tumor weight (mg) = [major axis x (minor axis)2) + 2. RESULTS NK activity C3H/He mice were injected iv with 107 cells of E.lentum killed by formalin, and the time course of NK activity in the spleen was measured, NK mice was 33.9% (E:T ratio $0:1) and 24.6% (E:T ratio 25:1). Those in E.lentum given were 90.6% and 66.5% on day 1, respectively (p< 0.01), and decreased till day 7, but NK activity represented still 48.6% (E:T ratio 50:1), significantly higher than in controls (p-<0.05) (Fig. 1) Tumoricidal macrophages induced by Elenum BALB/c mice were ip given 107 cells of formalin killed E,lenttant 3 times every other day for 5 days. As shown in Table 1, tumoricidal macrophages were induced 9 days after the first injection of E.tentum. The activity was $6.28 at E:T ratio 10:1 and $5.4% at E:T ratio S:1 on day 9 as compared to 10.1% and 4.2% in controls (p<0.01), respec tively. On day 11, similar activity was observed (in Exp I and 2) and the same result was also represented by OK-432 (Exp 2) 131 ExT ratlo S01 “+ BT ratio 2541 co 0 1 3 s 7 Days after treatment Fig. 1 NK call activity after iv injection of 107 cells killed EJentum. NK. ‘activity in pieen cell of GSHHe mie wasdetermined against 2 x104 calls well of 51Cr-labeled YAC-1 Iymehoma in a 4 ‘hour assay. Each point and bar indicate the mean standard error of 3 exper ments. *, **: Statistically significant ferences from the control (gr0up (9 <0.05 and p-<0.01, respectively). Table 1. Kinetics of tumoricidal macrophage induced by E.entum in BALBIC mice.® coop Davart % Oytotatcty iat treatment on z expt? Control tasatad 11.1220 Elontum 1 95208 94s19 3 168216 129215 5 149223 144218 7 188217, 113225 " sig248" | 48525.2° ep 2° onto to1si2 42209 Elenum ° 56.2568" 55.4204" " 728278" 593325" oK-432 n 675253" 46.7228" Ayice were injocted ip with 107 cls of killed Elentum oF 2 KE of OK-432 9 times every other day for 5 days. cytotoxicity was measured ina 22 hr 81 Cr-releaseassay against 104 cals of labeled PB18 mastocytoma. oytotoxicity was messured against labeled ELA lymphoma. fe the mean standard error of 3 experiments. 132 tered group was 62.34% in compari son with -3.3% in controls at week 4. AU E:T ratio 50:1, the activity ‘was 29.0% in test group and -2.2% in control. CTL activities in the test group were 66.6% and 28.5% at E:T ratio of 100:1 and 0:1, res- pectively, while those in the control group were -2.4% and -1.1%%, res- pectively, at week 5. As shown in Table 2, on day 28, CTL activity against L1210 was significantly higher in immunized mice given E.lentum on day 0, 1, 2 and 21 than in immu- nized mice without E.lentum ad- minstration (p< 0.01). Antitumor activity of cell fractions Assessment of each E.lemum fraction by weight was approxi- mately 20% in the crude cell wall fraction and 45% in the membrane fraction (Fr2). The cell wall fraction (Fri), obtained from the crude cell wall fraction, was approximately 12% of E.lentum whole body and those of subfractions (Frly,2.3,4) were 18%, 16%, 1.5% and 1.5%, respectively (data not shown). Table 2. Cytotoxicity against L1210 lymphoid leukemia in C57BUE mice.* HATTA tamer £2 1901 Week Effect of Elentum on CTL induction in ICA mice inoculated se with 108 Enrlicn ascites tumor cola. Mice wore injected iv with 107 calls of killed Elentum for 7 days alter tumor inoculation. CTL ‘activity in spleen calls was measured against 2 <1 04 celldwell of 519r- labeled Enrich ascites tumor cols in a 4 hour assay. Each point and bar indicate the mean = standard error of 3 experiments **: Statistically significant difference from the normal group at p<0.01 Group Injection ot No, of alive L1210 % Cytotoxicity? Efentum® calls (x107)¢ 00:1 aa ontot pia=10 sonra] r8s15] nit a02s3-) 2382255) © teee4g Immun + sasoad] 50,9258, azarae] Mice were immunized with 4 108 spleen cells of BALB/c mice and injected ip with 5 x108 cals of L1210 at 3 weeks later. Price were injected W with 107 calls of killed Elertum on day 0, 1, 2, and 24 lin peritoneal cavity of C57BU8 mice on day 28 Incicates the mean standard error Number of living ce of 15 mice, ‘oytotoxic test was performed on day 28 and cytotoricity was determined by 4~Pr 5*Cr-release assay against 5 109 colls of labeled L1210. Figures indicate the mean = standard error of 3 experiments. “p<0.05, **p<0.01 ANTITUMOR EFFECTS OF £ LENTUM 133 Table 3. Elects of components in E lentum on Ehlch ascites tumor in ICR mic. Tumor : ist? N2-oteunivor Tumor sue on TUM Materia 3° grow ote aay 21 (9) No tested — day 2r ig SON contol 47.7492 ono 5108 Eien (8009) ND wo t1s03" 216 m ( 500p9) ND zoe 1.4203" 27.8 Fe (1.80049) 51.7 24.5 oin0 29206" 56.9 Fi-t ( T00u9)57-728.8 ovo 40205 784 Fi-2 ( 6504ig) 93.424.0 ono 5209 1278 Fi-3 | 60u9) 54.6240 ano 55207 1078 Fi-¢ —( 60Hg) 48.626.0 an 50208 98.0 Recovery 56.8265 ano 212058" 412 Fractions were injected inratumoally every day for 7 days starting on day 0. Pean survival time (days) indicates the mean-estandard error. © Figure in parenthesis indicates the number of tumor-bearing animal Tumor weight indicates the mean standard err. ND Not determined p<0.01, Table 4. Effects of various doses of cell wall fration.* Tumor No. of suvivors! Tumor see on Materiah, mst? : growth No.tested — day2t (gt Fo" eet Controt sr7ats oto 60812 Eleotum —( 6009) No 8 0.420 80 Celtwall ( 300,ua) No 2noiye 1.8204" 38.0 (5009) No sion) 0920.2" 18.0 (800 ya) No v9 1.0203" 20.0 (1,000 ND 29 08205" 16.0 ep? onto! 92.1236 0120 4.0208 Cella ( 50yq) 48.82147 ono 21207, 528 {100 us) No wor) 4280.2" 30.0 { 200 3) No gio) 1.4409" 278 {500 49) No 402) 1.2404" 300 2 Fractions ware injected intratumorally every cay tor 7 days starting on day 0 Pxtean survival time (days) indicates the mean + stancard err. Figure in parenthesis indicates the number of turer- bearing animal ‘rumor weight indicates the mean + standard eror. NOMNot determine <0.01 134 HATTA Table 5. Effects of various doses of membrane fraction # Tumor growth Te) No, of survivors! Tumor size on No. tested day21 (gy? msi Bt Control ono Membrane (1,800 9) 31104) @3.600p0) -3110(2) (6.4009) 411012) Bp? Control 21292 ono 11.9207 Membrane ( 180yq) 96.622.7 on 10.2209 (360u9) 54.6225 ono 10.8214 ( 72049) NO. 0 40212" (1,000) ND 310 2920.7" Fractions were injected intratumorally every day for 7 days starting on day 0. Pmiean survival time (days) indicates the mean + standard error. Figure in parenthesis indicates the number of tumor-bearing animal Tumor weight indicates the mean = standard eror ND Not determined. + pco.o1 Antitumor activity of each activity was recognized within a pared with 33.3% and 22.5% in fraction against Ehrlich ascites range of 100 to 1,000 ug. When control mice, respectively (p< 0.01), tumor is shown in Table 3. Tested tumor-bearing mice were injected it On and after day 5, NK act dose of each fraction corresponded with various doses of membrane decreased to control level (Fi to the 500 yg of cell wall fraction fraction for 7 days, as shown in while that in C3H/He mice injected (Frl) which was about 5 times the Table 5, an antitumor effect was iv with 1,000 ug of membrane weight of cell of 107 E.lenium cells. observed at doses more than 720 ug. fraction represented the maximum All the mice in untreated group died A maximum effect was observed at (85.2% and 68.79% at E:T ratio 50:1 from on day 34 to 66. Antitumor the dose of 1,000 to 1,800 yg. Simi and 25:1) on day 1 (p< 0.01). Al- activity was retained in cell wall lar results of cell wall and membrane though NK activity decreased to fraction (Fr1) and membrane frac- fraction were also obtained by intra- control level on day 5, it was elevated tion (Fr2), however, the activity venous and intraperitoneal injection again on day 7 to 49.0% (E:T ratio ‘was mild compared to the activity (data not shown). 501) and 31.3% (E:T ratio 25:1), of 107 cells of E-lentum (800 u). values which were significantly Induction of NK activity tumoricidal higher than those in controls (p< Dose dependency of cell wall and macrophage and CTL activity by 0.01), membrane fraction effective component of E.lentum BALB/c mice were given ip ‘As shown in Table 4, when Asshown in Fig. 3, NK activity 100 g of cell wall or 720 ug of mice were administrated with various in spleens of C3H/He mice given iv membrane fraction for 5 days, and doses of cell wall fraction for 7 days, 300g of cell wall fraction, elevated on day 9, macrophage mediated antitumor activities were shown at to 89.7% (E:T ratio 50:1) and 87.6% cytotoxicity in PEC was determined doses more than 100g and a similar (E:T ratio 25:1) on day 1 as com- against EL4 in vitro. As shown in ANTITUMOR EFFECTS OF ELENTUM 135 Days after treatment Fig. 3 Kinetics of NK call activation folowing Ww injection of 200 49 cell wall or 1,000 14g membrane fraction. NK activity in spleen cells of C3 HiHe mice was determined against 2x10 eatywell ot 51Gr- labeled YAC-1 in 4 hour assay. Each point and bar indicate the ‘mean = standard errr of 3 experiments. **: Statistically significant tference trom the control group at p<0.01 eT Table 6. Effects of fractions on tumorieidal macrophage induction in peritoneal exudate cells of BALB/c mice.® b =, ‘% Cytotoricty! y0:1 sit Control 70.1226° 8521.5 Esertum (800g) 3.929.4°" 49.421.8" Collwall (100g) 85.922.8°" 38.522.0" Coll membrane (720g) _24.922.0° ——20.321.5" ® Mice were injected ip with Elentum or each traction every other cay for days. P cytotoxic test was performed at 9 days after the first injection and cytotoxicity was assessed in 20 hr assay against 104 cells of, beled ELA. Figures indicate the mean + standard error of 3 experiments. “p<0.05, *"p<0.01 Table 6, macrophages from mice phages from mice given the mem- treated with E.lentum and the cell brane fraction represented only wall fraction displayed the same twice the control level (p< 0.05). level of cytotoxicity (p< 0.01). On CTL activity was not induced in the other hand, tumoricidal macro- spleens of ICR mice inoculated with Ehrlich ascites tumor cells by admi- nistration of either cell wall frac- tion or membrane fraction until week $ (data not shown) DISCUSSION E.lentum does not have a direct cyrotoxic action against Ehrlich ascites tumor cells and produced a marked increase in plaque-forming cells and humoral antibody against SRBC. Furthermore, E.lentum ‘enhanced DTH against SRBC signi- ficantly to the control level.!2 Thus, Wwe suggested that E.lentum might be BRM and its antitumor effect was exerted via a host-mediated action. In this study, we demons- trated that NK cell activity was ‘enhanced remarkably one day after injection of E.lentum and de- creased later, but was higher than the control level even on day 7. In cancer patients receiving OK-432, NK cell activity was detectable on day 1, reached a peak on day 3, and returned to the pretreatment level by day 7 and 8.167 Oshimi er ai.!8 have reported that NK cell activity in PEC of mice given OK-432 peaked on day 3. From these results, NK cell activity was augmented more quickly and durably by adminis- tration of E.lentum than by that of OK-432. The mechanism of NK cell activation by E.lentum has not been clarified. It has been reported that IFN, IL-2 and other Iympho- kines induced by bacteria might activate NK cells n vivo.9-Ms!6 Thus, it seems that E.lentum may also induce several cytokines and acti- vate NK cells by that mechanism, Elenium induced tumoricidal ma- crophages in PEC of BALB/c mice on day 9 and its activity remained at the same level at least until day 11 Furukawa er al.!? reported that @ mycolic acid derivative containing high proportion of unsaturated fauy acids rendered macrophages cytotoxic for over 14 days after injection. Furthermore, Keller et al.2 reported that tumoricidal 136 HATTA activity induced by lymphokines was only short lived, while that induced by bacteria persisted and enhanced the secretion of IL-6 and PGE2. Dazord eral’ reported that activity of tumoricidal macro- phages induced by killed Brucella reached maximum on day 5 and remained for more than 20 days. The mechanism of induction of tumoricidal macrophage by E.lentum is not yet known. Paulanock and Lambert®! demonstrated that macro- phages might be activated in a 2-step mode by IFN gamma and LPS. However, it seems that macrophages may be activated by gram positive bacteria such as E.lentum and OK- 432 having no LPS with the co- operation of lymphokine, ie IFN ‘gamma induced by the bacteria, and by the bacterium itself. In tumor- bearing mice given E.lentum, CTL against the tumor were induced 4 weeks after tumor inoculation, while CTL were not induced in untreated tumor-bearing mice given Ellentum was remarkably slower than that in untreated mice 21 days after tumor inoculation. There was a clear correlation between the induction time of CTL and growth or cure of the tumor in E.lentum- treated mice. The mechanism of CTL induction by E.lentum is also not known. Roberson and ElgerS have reported that C.parvum might induce CTL via the interleukin cas- cade, Since E.lentum induces GM~ CSF (unpublished), it seems that CTL induction by E.lentum may ‘occur through the same mechanism as that by C,parvumn. In this paper, we indicated that cell wall and membrane frac- tions of E.lentum had active anti- tumor components. Cell wall frac- tiotis of Lactobacillus,2? Bifido- bacterium,23 Nocardia,’ BCG ete. have been reported to act as BRM. NK cell activation by the cell wall fraction reached a maximum con day 1 like that by Z-lentum whole cell, but the activity remained for a short time. On the other hand, NK cell activation by the membrane fraction reached a maximum on day 1 and decreased to the control level fon day $, but increased again on day 7. These results suggest a reason why NK activation by E.lenzum whole cells persisted for a relatively Jong time. In contrast to the results in tumor-bearing mice given E. lentum whole cell, CTL induction was not observed in mice given cell wall and membrane fractions. These results indicate the reason why E. lentum whole cell has a superior antitumor effect than each of its components. Further studies should be carried out on the mechanism of effector cell activation by E.lentum. ACKNOWLEDGEMENTS: We than Dr Atsushi Muraguchi for his helpful suggestion References 1. Moore WEC, Holdeman LV, Discus sion of current bacteriological inves: ions of the relationships between testinal lora, diet and colon cancer. Cancer Res 1975; 35 + 418-20, 2. Hill Ms, Drasar BS, Aries V, Crowther JS, Hawksworth ©, Willam REO. Bacceria and etiology of cancer of large bowel, Lance 1971: |: 95-100 3. Finegold SM, Flora DJ, Attebery HR, Satter VL. Fecal bacteriology of colo: ic polyp patients and control patents. ‘Cancer Res 1975; 35 = 3407-17. 4. Sakamoto K, Konishi K. Antitumor fect of normal intestinal neroflora on Ehrlich ascites tumor. Jpn J. Cancer Res (Gann) 1988; 9 10946. 5. Roberson AM, Elger KD. Splenoestes {rom tumor-bearing Corynebacterium parsum tread mice maintain inter- leukin-2and -3 activity. Cancer Immunol Immunother 1986; 22: 49.88, 6 Dazord L, le Garree Y, Florentin 1 [Antitumor activity of living or killed Brucella; modification of the non specific etotoxic effector eels. Cancer [Immunol Immunother 1985; 15: 63°. 7. taumiS, Ueda H, Okuhara M, Aoki Hy ‘Yamamura Y. Effect of Nocardia rubre cel wall skeleton on murine interferon production in vitro. Cancer Res 1986; 46 1960, Rossol S, Voth R, Brunner S, et al Corynebacterium parrum (Propion- bacterium acnes) an inducer of tumor necrosis factor-alpha in human per- Pheral blood mononuclear cells and monocytes im vitro. Eur J Immunol 1990; 20 1761-5 Wakasugi H, Kasahara T, Minato N, Hamuro J, Miyata M, Morioka Y. In vero potentiation of human natural Killer cell activity by @ streptococcal preparation, OK-432 : interferon and ietleukin-2 participation in the ‘stimulation with OK-432, J Nat Cancet Ins 1982; 69: 807-12, ). Matsuzaki T, Shimizw Y, Yokokura T. ‘Augmentation of antimetasatc effect fon Lewis lung carcinome (LL) in (CS7BL/6 mice by priming with Lacto- Dacilus case. Med Microbiol Immunol 1990; 179 1618 Zhang SR, Salup RR, Uras PE, ef a ‘Augmentation of NK activity. and/or macrophage-mediated cytotoxicity in the liver by biological response moins including human recombinant intr Jeukin 2. Cancer Immuno! Immunother 1986; 21 : 1925 Morinaga S, Sakamoto K, Konishi K. Antitumor activity and its properties ‘of Eubacerimt lentum. Jpn J Cancer Res (Gann) 1988; 79 : 117-24 Kumagai K, Koh K, Suruhi R, Hinuma 5, Saitoh F. Studies of murine large franular Iymposytes. 1. Identification as elfector cells in NK and K eytotox ‘es, J Immunol 1982; 128: 38.94 Mizoguchi K, Nakashima 1, Isobe K, etal, Genetic and stimulatory cell type requirements for inducing lass 1 major histocompatibilty comple alloantigen specific m vivo cytotoxic T cll imme nity. Eur J tmmunol 1985; 18: 487-94 ‘Azuma I, Ribi EE, Meyer TS, Zbar B Biologically active components trom mycobacterial cell wall 1. Isolation land composition of cell wall skeleton land component P3, J Nat! Cancer Inst 1974; 32 95-01 [No ¥, Tsucitani T, Ohgak K,Inamoto T, Kan N, Tobe T. Effects of orl ad ministration of streptococcal prepara- tion OK-432 (Piibani on natural killer cell activity and lymphoprolierative response in mice with transplanted ANTITUMOR EFFECTS OF € LENTUM cxcal tumors, Exerpta Medica 1986 : Ms. Vinzenz K, Mateike M, Watzck G, Pocteder Il, Neubold N, Mickiche M. Modulation of NK activity in region lymph nodes by preoperative immune- therapy with OK-432 in patents with ‘cancer of the oral eavity. Cancer Detect, Prev (Supp). 1987; 1: 463-75. (shim K, Kano, Takaku F, Okumura K. Augmentation of mouse Killer cell activity by a streptococcal preparation, OK-432. J Natl Cancer Inst 1980; 65: 1265.69, Furukawa M, Ohtsubo Y, Suglmoto N, Katoh Y, Doh Y. Induction of tumor cial activated macrophage by a lipo- some-encapaulated glycolipid, trehalose 2, 3, G-trimycolate from gordona furantiaca, FEMS, Mirobio! Immunol 1990; 2: 8338. Keller R, KestR, Frei K, Lymphokines and bacteria that induce tumorcidal activity, tigger a different secretory ‘response in macrophages. Eur J immu ‘no 1990; 20 : 6958 Paulnock DM, Lambert LE. Identii- cation and characterization of mono- onal antibodies specific for macro- phages at intermediate stages in the tumoricidal activation pathway. J Immunol 1990; 148 : 765.73. 2. 137 Sato K, Sato H, Tomioka H, Yokokura TT. Enhancement of host resistance sgsinst Listeria infection by Lecto actus case: efficacy of cell wall preparation of Lactobactlus cate ‘Microbiol Immunol 198; 32 : 1189-200. Ishihara K. Clinical effects induced by invratumoral administration of anti~ cancerous drugs in skin malignant tumors. Jpn J Cancer chemother 1989; 16: 1739, Richman SP, Hersh EM, Gutterman JU, ef al, Administration of BCG cell ‘wall skeleton into malignant effusion toxic and therapeutic effects. Cancer ‘Treat Rep 1981; 65 : 3837