Characterization of a streptococcal antitumor glycoprotein (SAGP)

Life Sciences, Vol. 62, No. 12. pp. 1043-1053, 1998 Copyright8 1998 Elswicr Science Inc. Printedin the USA. All tights rcscrvcd 0024-3205198$19.00 + .OO ELSEVIER

PIISOO24-3205(97)01142-9

MINIREVIEW CHARACTERIZATION OF A STREPTOCOCCAL ANTITUMOR GLYCOPROTEIN (SAGP)

Junk0 Yoshida*, Shozo Takamura and Matomo Nishio Department of Pharmacology, Kanazawa Medical University, Uchinada, Ishikawa 920-0293, Japan (Received in final form December 3 1,1997) Summary An acidic antitumor glycoprotein (SAGP) was purified from a crude extract of Streptococcus pyogenes, Su strain. Intraperitoneal injection with SAGP (20 mg protein&/day for 4 consecutive days) prolonged the life span of mice inoculated i.p. with

Ehrlich ascite carcinoma cells and methylcholanthrene-induced fibrosarcoma cells (Meth A) up to 244% and 169% of that of the control mice, respectively. These in vivo antitumor effects were reduced in immunosuppressed mice. The effector spleen cells from the Meth A-inoculated and SAGP-injected mice showed a considerable cytostatic activity on Meth A cells in vitro, and immunosuppression studies suggested that carrageenan-sensitive and/or asialo-GM1 positive spleen cells are responsible for the in vivo antitumor effect of SAGP. SAGP inhibited the cell growth of cultured cell lines including transformed hamster embryonic lung cells, murine leukemia L1210, Meth A and human promyelocytic leukemia HL60 cells. The ICs,s for the cell growth of these cells were all below 0.1 ug protein/ml. SAGP inhibited the incorporation of nucleic acid precursors into Meth A cells. It seems that sulthydryl groups of the SAGP molecule are essential for the expression of the antitumor action of SAGP. The cell growth-inhibitory activity of SAGP was diminished in Meth A cells preincubated with pertussis toxin (IAP), whereas it was augmented in the cells preincubated with cholera toxin (CTX), suggesting the involvement of toxin-sensitive GTP (G)-proteins in the SAGP-action. IAP and CTX-catalyzed ADP ribosylation assays confirmed that SAGP augmented the activity of LAP-sensitive G-protein. In addition, this augmentation was detected neither in Meth A cells incubated with heat-inactivated SAGP nor in SAGP-insensitive L929 cells. SAGP induced apoptosis in Meth A and HL60 cells as Key Words: antitumor, Streptococcus pyogenes, immunomodulation, cell growth inhibition, apoptosis, GTP binding proteins

*Address correspondence to: Junko Yoshida Department of Pharmacology, Kanazawa Medical University, Uchinada, Ishikawa 920-0293, Japan, Telephone: (076) 286-2211 (Ex. 3724), FAX: (076) 286-8191, E-mail: [email protected]

Antitumor Effectsof StreptococcalGlycoprotein

1044

Vol. 62, No. 12, 1998

assessed by DNA fragmentation.A singledose injection of SAGP (100 mg protein/kg, i.v., s.c., or i.p.) into mice produced no toxic signs except occasional pain responses observed for one week after the injection. Thus, SAGP is a low toxic substancethat shows in vivo

antitumor activity by modulating immuneresponsesof the host, and also exhibits in vitro cell-growth inhibition through IAP-sensitiveG-protein. Since the seventeenth century, it has been recognized that malignant tumors occasionally disappear after intercurrent bacterial infection. Busch (1) was the first to describe a clinical effect, in a sarcoma patient treated with an exudate of eryspelas. Fehleisen (2) established streptococcus as the causative agent of erysipelas, and also observed a favorable influence of living streptococci on cancer patients. William B Coley and his colleagues observed that an attack of erysipelas on a patient with an inoperable sarcoma induced a complete disappearance of the tumor. On the basis of this observation, they prepared heat-treated mixtures of streptococci and Serrutiu marcescence, known as “mixed Coley’s toxins” and treated the tumor patients with them (3-5). Tumor therapy using Coley’s toxins fell into virtual disuse with the introduction of radiotherapy and chemotherapy in the 1930s. However, laboratory studies on microbial products including bacteria, yeast and fungi as anti-tumor agents have continued, and interest has now focused on three bacterial groups, gram negative bacteria, mycobacteria (BCG) and corynebacteria (Corynebucterium parvum) (6). Motivated by the reproducible phenomenon “hemonhagic necrosis” induced by the injection of Gram-negative bacteria into tumor-bearing mice, Shear and Turner (7) identified an active substance lipopolysaccharide (LPS) from the bacteria. Afterwards Coley et al. discovered tumor necrosis factor (TNF) in serum of animals injected with BCG and LPS (6, 8). Pleiotropic properties of TNF have been widely reported (9- 11). Meanwhile Okamoto et al. observed that haemolytic streptococci exhibited anti-tumor activity in mice, and prepared a streptococcal antitumor preparation named OK-432 from a non-virulent Su strain of Streptococccus pyogenes (12). Yamamoto et al. (13) reported that the injection of lypoteichoic acid (LTA) from Streptococcus pyogenes strain Sv induced TNF in the serum of mice previously primed with Corynebucterium parvum. Usami et al. showed that LTA induced TNF inhibited the growth of tumor cell lines of both mouse and human origin (14), and further, LTA suppressed the growth of both solid- and ascite-type Meth A fibrosarcoma, suggesting that LTA is an antitumor component of OK-432 (15). On the other hand, Higuchi and Shoin (16) isolated a streptococcal cytotoxic protein (SCP) from a cell-free extract of the Su strain of Streptococcus pyogenes, and showed its physicochemical and antitumor properties. Here, we summarize the physicochemical and biological properties of an acidic antitumor glycoprotein purified by us (17) from an extract of Streptococcus pyogenes Su strain. This new bacterial substance inhibits the growth of tumor cells and has immunomodulating properties.

Purification

of SAGP from the Crude Extract (CE) of Streptococcuspvof?enes. SU Strain

We employed a non-virulent streptococcal strain, Su, previously used to prepare streptococcal preparation OK-432 (12). The cocci grown in medium were mechanically disrupted, extracted with magnesium-acetate buffer and centrifuged at 105,000 g for 2 hr. The crude extract so obtained (CE) was heated (45”C, 30 min), precipitated with streptomycin to remove the nucleic acid fraction, fractionated with ammonium sulfate (55-70%), and subjected to chromatography on octyl-Sepharose CL-4B, DE-52, and Sephadex G-200 gel. The antitumor activity of each chromatographic fraction was evaluated by the cell growth-inhibitory activity on transformed hamster embryonic lung cells (THEL) or murine leukemia L I2 10 cells in vitro. Final active fractions from Sephadex G-200 gel filtration were combined, concentrated, and dialyzed against distilled water, and the resulting precipitate was removed. To the supematant was added l/10 the volume of 10 X phosphate buffered saline (PBS) and the mixture stored at -30” C until use (17).

Vol. 62, No. 12, 1998

Antitumor Effects of Streptococcal Glycoprotein

1045

In 1987, Kanaoka et al. (18) cloned and expressed the structural gene for this substance in Escherichia coli.

SAGP was shown to be homogeneous on polyacrylamide gel containing sodium dodecyl sulfate with a molecular weight of about 48 kDa (19). The molecular weight of the native state of this substance was calculated as 150-160 kDa by gel filtration suggesting that the substance consisted of identical subunits. The protein was also stained with the periodic acid-Schiff s reagent specific for sugar. Gas chromatographic analysis revealed that one of the carbohydrates is d-allose of which the content was 1.5% (w/w). The isoelectric point of this protein was 4.3. Analysis of the amino acid composition showed the protein to be abundant in acidic amino acids, such as glutamic and aspartic acids. Then, we tentatively named this substance streptococcal acid glycoprotein (SAGP) (17).

Intraperitoneal injection of SAGP (20 mg protein/kg/day X 4 days) into ICR mice inoculated i.p. with Ehrlich ascite carcinoma cells prolonged the life span of the mice as compared to that of the control mice (17). SAGP also prolonged the life span of syngeneic BALB/c mice inoculated i.p. with methylcholanthrene-induced fibrosarcoma (Meth A) cells (20) (TABLE 1). Kanaoka et al. prepared SAGP by somewhat different procedures from ours, examined its in vivo and in vitro antitumor effects on some tumor cells and compared the effects with those of OK-432 (2 1). They reported that i.v. injection of SAGP (600 &mouse/day X 4 days) into mice inoculated i.m. with sarcoma 180 inhibited the tumor growth, suggesting a systemic effect of SAGP.

C ell-

r

. .

SAGP inhibits the growth of some cell lines in culture, including THEL (17,22), L12 10 cells (22), Meth A cells (19,23) and human promyelocytic leukemia HL60 cells (unpublished data). Kanaoka ef al. reported that SAGP inhibited the growth of human uterus cancer (HeLa), human acute lymphoblastoid leukemia (CCRF-HSB-2) and three mouse tumor cell lines (Meth A, Ehrlich carcinoma and Sarcoma 180) at doses below 0.1 pg/ml (21). We used mainly Meth A cell line to explore the mechanisms by which SAGP exhibits the cell growth-inhibitory activity, since SAGP shows antitumor effects on Meth A cells both in vitro and in vivo. NCTC clone 929 cell line (L929) was used in some in vitro experiments as it is insensitive to SAGP. The growth of L929 cells is not affected at higher doses of SAGP (up to 100 ug protein/ml) (19) (TABLE 2).

Immunomodulating Action of SAGP As shown in TABLE 1, in viva antitumor effects of SAGP were diminished in X ray-irradiated ICR mice and carrageenan (an antimacrophage agent)-injected BALB/c mice, respectively. These results suggested host-mediated action is involved in the antitumor effects of SAGP. To explore whether effector cells are responsible for the host mediated action of SAGP, the cytostatic activity of effector cells including spleen cells, and plastic adherent and non-adherent peritoneal exudate cells against Meth A cells was assessed by an in vitro 3H-thymidine incorporation assay as previously reported by

EAC*’

EAC

Meth Ae2

1

2

3 27 15 15

20 20

SAGP Carra.*4-control

Carra.- SAGP

94

169 94

100

16

Control

134

25.5

20

X-SAGP

100 202.6 89.5

19 38.5 17

20

Control SAGP X*3-control

116

21

20

CE

244

100

T/C(%)*6

44

18

MST*’

20

(mg protein&day)

Dose

SAGP

Control

Treatment

o/10

2110 o/10

Oil0

O/10

o/10 4/10 o/9

l/8

218

O/l6

(day 60)

Survivors

20

20

17

References

From days 1 to 4, PBS (Control) or SAGP were injected i.p.

On day 0, female BALB/c mice were inoculated i.p. with Meth A cells (2 X 105/mouse).

Carra. : Mice were injected i.p. with carrageenan (2 mg/mouse) on day -2.

*6 T/C(%): T; survival days in the treated group. C; survival days in the control group.

+5 MST: Median survival time

*4

*3 X: Mice received a dose of 5 Gy X ray-irradiation over the whole body on day -6.

*2

From days 1 to 4, PBS (Control), SAGP or CE (crude extract) were injected i.p.

*1 On day 0, female ICR mice were inoculated i.p. with Ehrlich ascite carcinoma cells (EAC)( 1 X 106/mouse).

Tumor

No. of exp.

Effect of SAGP on Survival of Mice Inoculated with Ehrlich Ascite Carcinoma Cells or Meth A Cells

TABLE 1

> z. z 2 ; B G: % e 8 8 8 g 0 q 8 ‘0 a R 5’

Vol. 62, No. 12, 1998

Antitumor

Effectsof StreptococcalGlycoprotein

1047

TABLE 2 Growth Inhibitory Effect of SAGP on Cultured Cell Lines Origin

Cell line

hamster embryonic

lung

Gil (pg protein/ml)* 0.062

references 17,22

THEL

transformed

L1210

murine leukemia

0.08

22

Meth A

murine methylcholanthrene-induced fibrosarcoma A

0.073

19, 23

HL60

human promyelocytic

0.054

L929

murine connective

leukemia

tissue

> 100

unpublished

data

I9

*Tumor cells (5 X lo4 or 1 X 10’ cells/ml) were incubated in the absence and presence of SAGP (0.03 to 3.0 pg protein/ml) at 37°C for 72 hr in humidified 5% CO, and 95% air. The number of cells was determined by trypan blue exclusion. us (24). The spleen cells from Meth A-inoculated and SAGP-injected mice were found to have

considerable cytostatic activity against Meth A cells, while no activity of plastic adherent and nonadherent peritoneal exudate cells was found (20). However, the activity of the spleen cells horn Meth A-inoculated and SAGP-injected mice was reduced when the assay was done in the presence of carrageenan or ithen the spleen cells were pretreated with anti-asialo GM1 antibody and complement (Fig. 1). Carrageenan and anti-asialo GM1 antibody have been shown to deplete macrophage functions (25, 26) and natural killer (NK) cell functions by recognizing NK cell antigen (27, 28) respectively. These results suggested that splenic macrophages and NK cells are involved in the cytostatic activity of the spleen cells induced by SAGP. Ehl et al. recently reported that anti-asialo GM1 antibody is less specific in its discrimination between NK cells and cytotoxic T cells (29). So, a role for cytotoxic T cells in the activity can not be ruled out. Thus, SAGP showed an immunomodulating action on splenic cell populations of mice inoculated with tumor cells. Inhibition

of Incorporation

of Nucleic Acid Precursors

into Meth A Cells by SAGP

When SAGP was added to Meth A cell culture on day 0, a growth-inhibitory effect was apparent on day 2 and increased up to day 4. When SAGP was removed from the culture medium on day 2 by washing the cells by centrifugation, the cells once again grew as in control cultures. This indicates that the growth-inhibitory action of SAGP was reversible and time dependent (19). Subsequently, the effect of SAGP on nucleic acid synthesis was examined. SAGP reduced the incorporation of ‘Hthymidme and 3H-uridme into Meth A cells in a concentration and a time-dependent manner. When Meth A cells were incubated in the presence of SAGP (0.1-3.0 ng protein/ml) for 16 hr, then pulsed with 3H-thymidine or 3H-uridine for 3 hr, the incorporation of two precursors into Meth A cells was reduced in a concentration-dependent manner. Assay of the time course on the incorporation of ‘Hthymidine into Meth A cells revealed that 1.0 ug protein/ml SAGP reduced the ‘H-thymidine incorporation during the first 2 hr of incubation. These results indicate that the inhibitory effects of SAGP on nucleic acid synthesis preceded cell-growth inhibition (19).

Antitumor Effectsof StreptococcalGlycoprotein

1048

A

Vol. 62, No. 12, 1998

% Cytostasis (Mean f S. E.)

Control-SPC SAGP-SPC Control-SPC+carra. 1

N.S.

SAGP-SPC+carra.

B

% Cytostasis

0 I

(Mean f S. E.) 50

100

1

I

Control-SPC SAGP-SPC

1

Control-SPC + anti-asialo GM1 + C SAGP-SPC + anti-asialo GM1 + C Control-SPC + C

N.S.

1 pc

0.05

,

SAGP-SPC + C

FIG. 1. Cytostatic activity of the effector spleen cells from Meth A cell-inoculated and SAGP-injected mice. On day 0, female BALBk mice were inoculated i.p. with Meth A cells (1 X lO’/mouse) and injected i.p. with SAGP (20 mg protein/kg/day) from day I to 4. On day 7, the effector spleen cells (SPC) of each mouse were collected and assayed for the cytostatic activity against Meth A cells using in vitro ‘H-thymidine incorporation assay. A. The assay was done in the presence of carrageenan (carra.)( 10 &ml) (n = 5). B. The spleen cells were pretreated with anti-asialo GM1 antibody (anti-asialo GMl) and complement(C) then subjected to ‘H-thymidine incorporation assay at an effector and target cells ratio of 1:100 (n = 4).

Involvement

of Sulfhydryl Groups of SAGP in the Expression of SAGP-activity

activity of SAGP on Meth A cells was reduced when the Meth A cells were incubated with SAGP in the presence of sulfhydryl agents such as cystamine and 5,5’-dithiobis (2nitrobenzoic acid). The activity of SAGP was also diminished by pretreatment with cystamine, and the inactivation of SAGP by cystamine was reversed by dithiothreitol, suggesting that cystamine interacts with SH-groups of the cysteinyl residues on SAGP but not with all SH-groups of the target cells. Furthermore, the survival-prolonging effect of CE on mice inoculated with Meth A cells was reduced by the administration of cystamine. These findings suggest that sulfhydryl groups of the SAGP molecule are involved in the expression of both in vitro and in vivo antitumor activities of SAGP or CE (23). The growth-inhibitory

Induction of Apoptosis by SAGP Meth A or HL 60 cells were incubated in the absence or presence of SAGP for 48 hr. The DNA of cultured cells was extracted by the method of Ishizawa et al. (30) and subjected to agarose gel

Vol. 62, No. 12, 1998

Antitumor

Effects of Streptococcal

Glycoprotein

1049

bp 23130

9416 6557 4361

2322 2027 564

FIG. 2. Effect of SAGP on DNA fragmentation of Meth A cells (A) and HL60 cells (B). Meth A or HL60 cells (1 X lo6 cells40 ml/well) were incubated in the absence or presence of SAGP (0.1 to 1.O pg protein/ml) for 48 hr at 37”C, then subjected to DNA extraction by the method of Ishizawa et al (30). Each DNA was assayed for DNA fragmentation by 2.0% agarose gel electrophoresis. The right lane in (A) and the left lane in (B) indicate DNA size markers (Lambda pharge DNA/Hind III digest).

electrophoresis for DNA fragmentation analysis (3 1). As shown in Fig. 2, SAGP at concentrations of 0.3 and 1.O pg protein/ml induced DNA fragmentation in Meth A and HL 60 cells. Flow cytometric analysis showed that the rate of apoptosis was increased in cells incubated with SAGP compared to

control cells. In addition, Meth A and HL 60 cells incubated with SAGP for 48 hr showed typical morphologic signs of apoptosis such as chromatin condensation and nuclear fragmentation as demonstrated by 4’, 6-diamidine-2’-phenylindole dibydrochloride (DAPI) staining (unpublished data). Involvement of Pertussis Toxin-sensitive G-protein in the Cell-Growth Inhibitory Action of SAGP Studies with several reagents affecting intracellular signal transduction pathways showed that the Meth A cell-growth inhibitory effect of SAGP was diminished by preincubating the cells for 16 hr with l-100 &ml pertussis toxin (IAP), whereas the SAGP- activity was augmented by preincubating the cells for 16 hr with 0.3 - 3.0 &ml cholera toxin (CTX) (19). IAP catalyses ADP ribosylation of a cysteine residue in the a subunit of inhibitory GTP-binding (Gi) protein and thereby blocks interactions between Gi protein and receptors (32,33), while CTX activates stimulatory GTP-binding (Gs) protein by ADP-ribosylating an arginine residue of the a subunit in Gs protein (34, 35). The fmding that the activity of SAGP was diminished by IAP suggested the involvement of Gi protein in the expression of SAGP-activity. To investigate the effect of SAGP on the activity of G protein, IAP or CTX-catalyzed ADP ribosylation assay with [‘“PI NAD was performed. The IAP-catalyzed ADP ribosylation assay showed that the labeling intensity of the Gia 41 kDa membrane protein of Meth A cells was augmented when the cells were incubated with SAGP. The IAP-catalysed ADP

1050

Antitumor

Effects of Streptococcal

Glycoprotein

Vol. 62, No. 12, 1998

ribosylation of the Gia protein was augmented neither in the SAGP-insensitive L929 cells incubated with SAGP nor in the Meth A cells incubated with heat-inactivated SAGP. On the other hand, there was no modification in the CTX-catalysed ADP ribosylation of Gsa in the membranes from either Meth A or L929 cells incubated with SAGP (19). Thus, IAP and CTX-catalyzed ADP ribosylation assay confirmed that SAGP augmented the activity of IAP-sensitive G protein, suggesting the involvement of IAP-sensitive G protein in the expression of SAGP-action. At present, we do not know why the cell growth inhibitoty effect of SAGP was augmented by preincubating the cells with CTX. It is demonstrated that the py subunits of heterotrimeric G proteins play important roles in regulating receptor-stimulated signal transduction processes (36, 37). As continuous stimulation of Gs protein by CTX promotes the dissociation of its a- and &-subunits (38), py-subunits dissociated by CTX might participate in the signal cascades to augment the SAGP-activity. The above findings on the in vitro SAGP-action led to the assumption that SAGP binds to an unknown receptor molecule coupled with IAP-sensitive G protein, which triggers events leading to the inhibition of nucleic acid synthesis and cell proliferation. Unfortunately, we failed to characterize the receptor molecule for SAGP by binding assay and chemical cross-linking studies with cross linker (DSG: disuccimidyl glutarate or DSS: disuccinimidyl suberate). SAGP was iodinated with Na”‘I by the chloramine-T method or Bolton-Hunter reagent according to the manufacturers’ recommendations. ‘251-labeledSAGP (‘*?-SAGP) retained the growth inhibitory activity on Meth A cells. Binding assays with lZ51-SAGPwere carried out on Meth A or SAGP-insensitive L929 cells. Although the binding of SAGP to Meth A cells had a tendency to increase in a time (up to 5 hr) and temperature (O”C,24°C and 37”(Z)-dependent manner and also the binding to L929 cells was iess than that to Meth A cells, high levels of nonspecific binding were observed even after addition of a large excess of unlabeled competitor SAGP. Binding of ‘251-SAGPto Meth A cells was barely competed up to 40% by a 600 fold excess of unlabeled SAGP (unpublished data). These findings suggest that SAGP binds to a large number of low affinity binding sites or alternatively, SAGP-receptor internalization occurs. Effect of SAGP on Intracellular CAMP Levels To examine whether or not the G proteins involved in the expression of the SAGP-action are functionally related to adenylate cyclase, the intracellular CAMP contents of Meth A cells incubated with SAGP were measured. Meth A cells were incubated in the absence and presence of 0.1 to 3.0 pg protein/ml SAGP for 24 hr, or with 1 pg/ml CTX or 100 ng/ml IAP as positive controls. The CAMP level in the cells exposed to SAGP was reduced slightly with increasing concentrations of SAGP (19). Cho-Chung et al. have shown that CAMP exerts dual controls, either positive or negative, on cell proliferation, depending on the physiological status of the cell (39). It is unlikely that reduced CAMP levels in the cells contribute to the SAGP-induced cell-growth inhibition, because dibutyrylCAMP(30-100 PM) or CAMPphosphodiesterase inhibitor, 1-methyl-3-isobutylxanthine (30-300 PM) could not reverse the SAGP-activity (data not shown) and also the cell-growth inhibitory effect of SAGP was increased in the cells preincubated with CTX, which actually caused an elevation of CAMP level owing to activation of the Gs protein (19). At present, we do not know what effector molecules are associated with Gi protein modulated by SAGP. Furthermore, it is unclear what kinds of cellular proteins in signal transduction pathways are responsible for the SAGP-induced cell-growth inhibition and apoptosis. In this regard, we found in preliminary experiments that a tyrosine kinase inhibitor, herbimycin A (40) (100 nM) augmented DNA fragmentation and the growth inhibition induced by SAGP in Meth A cells, and a tyrosine phosphatase inhibitor, sodium orthovanadate (30 PM) (4 l-43) but not a serine/threonine phosphatase inhibitor, sodium fluoride (1 mM) (43) diminished the Meth A cell-growth inhibition by SAGP

Antitumor

Vol. 62, No. 12, 1998

Effectsof Streptococcal Glycoprokin

1051

TABLE 3 Effect of SAGP on Female ddY Mice by a Single Dose (100 mg protein/kg) Injection Route of administration

No. of l-day survivors/ No. of animals tested

Toxic signs

i.v.

515

None

i.p.

515

A.S.M.* in all mice soon after the injection and stretching responses in 3 mice between 20 and 60 min after the injection

S.C.

313

Squeaking responses during the injection

in two mice

*A. S. M. : Arrest of spontaneous motor activity (unpublished data). These results suggest that inhibition of tyrosine phosphorylation or dephosphorylation of cellular proteins plays a critical role in the cell-growth inhibition by SAGP.

Acute Toxic Effect of SAGP on Mice A single dose of SAGP (100 mg protein/kg) was administered i.v., i.p. or S.C.into female ddY mice. As shown in TABLE 3, except stretching or squeaking pain responses, no toxic signs were observed in mice injected with SAGP for one week after the injection suggesting low toxicity of SAGP in mice.

An acidic glycoprotein (SAGP) purified from a crude extract of Srreptococcus pyogenes (Su strain) showed antitumor effects on mice inoculated with either allogenic or syngeneic tumor cells. It was suggested that augmentation of the cytostatic activity of spleen cells by SAGP is in part responsible for the in vivo antitumor effect of SAGP. SAGP inhibited the growth of tumor cell lines in culture including some human cell lines in a concentration-dependent manner. Nucleic acid synthesis of tumor cells was inhibited by SAGP, preceding the cell-growth inhibition. Sulfhydryl groups of SAGP molecules seem to be essential for the expression of the cell-growth inhibitory activity of SAGP. SAGP induced apoptosis in mouse Meth A and human I-IL60 cells. The Meth A cell growth-inhibitory effect of SAGP was reduced by incubating the cells with IAP, whereas it was augmented by incubating the cells with CTX. IAP and CTX-catalyzed ADP ribosylation assays suggested that the activity of IAP-sensitive G-protein was augmented by SAGP. Although the exact mechanism of the immunomodulating action of SAGP on tumor-bearing host was not elucidated, clarifying the molecular mechanisms of the action of SAGP may help in understanding the tumor cell proliferation mechanism, especially the cross-regulation between bacterial toxin-

1052

AntitumorEffects of StreptococcalGlycoprotein

Vol. 62, No. 12, 1998

sensitive G protein-mediated signaling pathways in tumor cells, and in the search for an effective low toxic substance in cancer therapy.

AcknowledgementS We wish to thank Prof. K. Nishikawa of the second Department of Biochemistry of this University for valuable discussions. We thank Prof. K. Fujikawa-Yamamoto and Dr. Y. Nagao of Division of Basic Science, Research Institute of Medical Science of this University for critical discussions and valuable advice.

References 1. BUSCH, Berl. Klin. Wchnschr. 5 137-138 (1868). 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.

FEHLIESEN, Deutsche med. Wchnschr. 8 553-554 (1882). H.C. NAUTS, W.E. Swift and B.L. Coley, Cancer Res. 6 205-216 (1946). CO. STARNES, Nature 357 1 l-12 (1992). G. ROOK, Nature 357 545 (1992). L.J. OLD, Science 230 630-632 (1985). M.J. SHEAR and F.C. TURNER, J. Natl. Cancer Inst. 4 81-97 (1943). E.A. CARSWELL, L.J. OLD, R.L. KASSEL, S. GREEN, N. FIORE and B. WILLIAMSON, Proc. Natl. Acad. Sci. USA 72 3666-3670 (1975). M. KLUNKE, S. SCHGTZE, P. SCHEURICH, A. MEICHLE, G. HENSEL, B. THOMA, G. KRUPPA and K. PFIZENMAIER, Cell Signal. 2 l-8 (1990). J.M. KYRIAKIS, 0. BANERJEE, E. NIKOLAKAKI, T. DAI, E.A. RUBIE, M.F. AHMAD, J. AVRUCH, and R. WOODGE’IT, Nature 369 156-l 60 (1994). W. FIERS, FEBS Lett. 285 199-212 (1991). H. OKAMOTO, S. SHOIN and S. KOSHIMURA, Bacterial Toxins and Cell Membranes, J. Jeljaszewicz and T. Wadstrljm (Eds), 259-289, Academic Press, London (1978). A. YAMAMOTO, H. USAMI, M. NAGAMUTA, Y. SUGAWARA, S. HAMADA, T. YAMAMOTO, K. KATO, S. KOKEGUCHI and S. KOTANI, Brit. J. Cancer 51 739-742 (1985). H. USAMI, A. YAMAMOTO, Y. SUGAWARA, S. HAMADA, T. YAMAMOTO, K. KATO, S. KOKEGUCHI, H. TAKADA, S. KOTANI, Brit. J. Cancer 56 797-799 (1987). H. USAMI, A. YAMAMOTO, W. YANASHITA, Y. SUGAWARA, S. HAMADA, T. YAMAMOTO, K. KATO, S. KOKEGUCHI, H. OHOKUNI and S. KOTANI, Brit. J. Cancer 57 70-73 (1988). Y. HIGUCHI and S. SHOIN, Biochim. Biophys. Acta. 966 239-247 (1988). J. YOSHIDA, M. YOSHIMURA, S. TAKAMURA and S. KOBAYASHI, Jpn. J. Cancer Res. 76 2 13-223 (1985). M. KANAOKA, C. KAWANAKA, T. NEGORO, Y. FUKITA, K. TAYA and H. AGUI, Agr. Biol. Chem. 51 2641-2648 (1987). J. YOSHIDA, S. TAKAMURA, S. SUZUKI and M. NISHIO, Brit. J. Cancer 73 917-923 (1996). J. YOSHIDA, S. TAKAMURA and S. SUZUKI, Biotherapy 3 331-336 (1991). M. KANAOKA, Y. FUKITA, K. TAYA, C. KAWANAKA, T. NEGORO and H. AGUI, Jpn. J. Cancer Res. (Gann) 78 1409-1414 (1987). J. YOSHIDA, S. TAKAMURA and S. SUZUKI, I. Antibiot. 42 448-453 (1989). J. YOSHIDA, S. TAKAMURA and S. SUZUKI, Anticancer Res. 14 1833-1838 (1994). J. YOSHIDA, S. TAKAMURA, S. SUZUKI and S. KOSHIMURA, J. Kanazawa Med. Univ. 17 330-335 ( 1992). G. CUDKOWICZ, Y.P. YUNG, J. Immunol. 119 483-487 (1977). R. KELLER, J. Natl. Cancer Inst. 59 175 l-1753 (1977). M. KASAI, M. IWAMORI, Y. NAGAI, K. OKUMURA and T. TADA, Eur. J. lmmunol. 10 175-180 (1980). W. W. YOUNG, S. HAKOMORI, J.M. DURDIK and C.S. HENNEY, J. Immunol. 124 199-20 1 (1980).

Vol. 62, No. 12, 1998

Antitumor Effects of Streptococcal Glycoprotein

1053

29. S. EHL, R. NUESCH, T. TANAKA, M. MYASAKA, H. HENGARTNER and R. ZINKERNAGEL, J. Immunol. Methods 199 149-153 (1996). 30. M. ISHIZAWA, Y. KOBAYASHI, T. MIYAMURA and S. MATSUURA, Nucleic acid Res. 19 5792 (1991). 3 1. C.A. SMITH, G.T. WILLIAMS, R. KINGSTON, E.J. JENKINSON and J.T. OWEN, Nature 337 181-184 (1989). 32. T. KATADA and M. UI, Proc. Natl. Acad. Sci. USA 79 3129-3133 (1982). 33. T. MURAYAMA and M. UI, J. Biol. Chem. 258 33 19-3326 (1983). 34. J.K. NORTHUP, P.C. STERNWEIS, M.D. SMIGEL, L.S. SCHLEIFER, E.M. ROSS and A.G. GILMAN, Proc. Natl. Acad. Sci. USA 77 65 16-6540 (1980). 35. J.R. HEPLER and A.G. GILMAN, Trends. Biochem. Sci. 17 383-387 (1992). 36. E.J. NEER and D.E. CLAPHAM, Nature 333 129-134 (1988). 37. L. BIRNBAUMER, Cell 71 1069-1072 (1992). 38. R.A. KAHN and A.G. GILMAN, J. Biol. Chem. 259 6235-6240 (1984). 39. Y.S. CHO-CHUNG, T. CLAIR, G. TORTORA, H. YOKOZAKI and S. PEPE, Life Sci. 48 1123-l 132 (1991). 40. Y. UEHARA, H. FUKAZAWA, Y. MURAKAMI and S. MIZUNO, Biochem. Biophy. Res. Commun. 163 803-809 (1989). 41. T. HUNTER, Cell 80 225-236 (1995). 42. N. BEGUM, L. RAGOLIA and M. SRINIVASAN, Eur. J. Biochem. 238 214-220 (1996). 43. H.C. R. WANG, J. Cell. Physiol. 168 87-96 (1996).