- Email: [email protected]
Effects of natural human interleukin-6 on thrombopoiesis and tumor progression in tumor-bearing mice
CANCER LETTERS
ELSEVIER SCIENCE IRELAND
Cancer Letters 79 (1994) 83-89
Effects of natural human interleukin-6 on thrombopoiesis and tumor progression in tumor-bearing mice Ken-ichi
Yoshizawa*, Tasuku Okamoto, Keiko Moriya, Yu-ichiro Satoh, Masanobu Naruto, Nobutaka
Akiko Ida
Izawa,
Basic Research Laboratories, Toray Industries Inc.. I I I I Tebiro, Kamakura, Kanagawa 248, Japan
(Received 25 January 1994; revision received 20 February 1994: accepted 24 February 1994)
Abstract The growth of inoculated colon 26 adenocarcinoma (colon 26) in mice gradually increased the platelet count owing to murine IL-6 secreted from the tumor, while Lewis lung carcinoma (LLC) decreased the platelet count in the hosts, depending on the tumor growth. Natural human IL-6 injections (hIL-6), 280 &kg/day, stimulated the platelet production in both types of carcinoma-bearing mice. When the administration of mitomycin C or cisplatin decreased the platelet number as a side reaction with a concomitant of suppressing the growth of colon 26 and LLC. respectively, hIL-6 could also increase the platelet count without the augmentation of tumor growth. However, loss of carcass weight was observed in colon 26-bearing mice treated with hIL-6, suggesting the development of cachexia is associated with hIL-6 administration. Despite the possibility of inducing cachexia in some types of tumors, our results suggest that IL-6 could be a useful means of restoring the decreased platelet number in cancer patients after intensive chemotherapy. Key words: Interleukin-6;
Thrombocytopenia;
Anti-neoplastic
1. Introduction Bone marrow suppression is a major problem in the treatment of progressive tumors due to intensive chemotherapy or radiation exposure in the case of bone marrow transplantation. G-CSF is efficiently used for the treatment of decreased neutrophil production, but a useful modality for thrombocytopenia has not been established. In ad-
* Corresponding author.
drug; Carcass
weight
dition, malignant tumors frequently alter the hematopoietic balance, depending on the tumor growth, in tumor-bearing animals [7,21] and cancer patients [6,12,15]. In chronic lymphocytic leukemia, thrombocytopenia is observed in association with the presence of auto antibodies against circulating platelets and excessive platelet pooling in an enlarged spleen. Hodgkin’s disease, breast cancer and lung cancer are tumors with which high platelet counts are frequently associated. Interleukin-6 (IL-6) is a multifunctional cyto-
0304-3835/94/$07.00 0 1994 Elsevier Science Ireland Ltd. All rights reserved. SSDI 0304-3835(94)03313-g
84
kine that has various biological activities, including the maturation of megakaryocytes in vitro [1 l] and the production of peripheral platelets in vivo [2, lo]. The thrombopoietic effect of IL-6 was studied in the treatment of platelet depression following hematopoietic damage induced by 5fluorouracil(5-FU) administration [5] or radiation exposure [4,13] in animals. Though IL-6 may be clinically valuable for the treatment of thrombocytopenia in cancer patients, there remain other problems to be solved. For example, Strassmann and coworkers have shown that IL-6 is an important mediator of cachexia [ 17,181 which induces anorexia, asthenia and anemia, and causes a reduced response to anti-neoplastic drugs [20]. This implies that IL-6 treatment may induce side reactions in cancer patients who are in a poor condition due to chemotherapy. However, there have been few studies concerning the effects of IL-6 on thrombopoiesis in tumor-bearing animals. We describe the thrombopoietic effect of natural human interleukind (hIL-6) in mice in which the hematopoietic balance is disordered due to tumor burden and chemotherapy, and clarify the side reactions, such as the induction of cachexia or the augmentation of malignant tumors. 2. Materials and methods 2.1. Mice C57BL/6 and BALB/c male mice were obtained from Charles River Japan (Kanagawa, Japan) and used at the age of 8 weeks. One group consisted of 6 mice. 2.2. Natural human IL-6 hIL-6 was prepared from human diploid fibroblast, DIP-2, supernatant fluid [9]. One milligram of hIL-6 exhibited 3.1 x lo6 units of WHO international standard activity. Endotoxin contamination was less than 0.05 EU/mg protein, as measured with Endospecye (Seikagaku Kogyo, Tokyo, Japan). 2.3. Tumor cells
Lewis lung carcinoma (LLC) was cultured in vitro in RPM11640 containing 10% fetal calf serum (FCS). After suspending the cells with a
K. Yoshi:wu
rt ul. / Cuncer Lear. 79 (1994) 83-89
pipette, 2 x 10’ cells were intradermally inoculated into the inguinal flank of C57BL/6 mice. Colon 26 adenocarcinoma (colon 26) was maintained in carrier mice. Blocks of the tumor were minced into a single cell suspension using a stainless steel mesh, and 1 x lo6 cells were intradermally inoculated and transferred into BALB/c mice. These tumor cell lines were kindly supplied by Dr T. Kataoka (Cancer Chemotherapy Center, Japanese Foundation for Cancer Research, Tokyo) 2.4. Experimental protocol Mice were inoculated with tumor cells on day 0 and were treated with chemotherapy on day 7. Mitomycin C (MMC) (Kyowa Hakko Kogyo, Tokyo, Japan) was administered to BALBic mice as a single i.p. injection, 2 mg/kg. Cisplatin (CDDP) (Bristol Myers, Tokyo, Japan) was given to C57BL/6 mice as a single i.p. injection, 10 mg/kg hIL-6 was consecutively given as S.C. injections. 280 &kg/day, once a day from day 8. Mice treated with sterile saline served as controls. 2.5. Assays Under anesthesia with ether, mice were bled by cutting the armpit artery and dissected at the times given below in Results. Platelet counts were performed with a Sysmex hematology analyser (Toa Medical Electronics, Hyogo, Japan). Tumor, lung, liver, kidney and spleen tissues were weighed on an analytical balance. Carcass weight was calculated as the difference between the whole body and tumor weight. 2.6. Murine IL-6 assay
Murine IL-6 (mIL-6) was assayed by the modified ELISA method of Starnes et al. [ 161. Briefly, the tracer antibody was changed to goat anti-mIL6 polyclonal antibody (R & D Systems, Minneapolis, MN) conjugated with N-hydroxysuccinimide-Lc-biotin (Pierce, Rockford, IL) in the molar ratio of 1 to 40. Streptoavidinhorseradish peroxidase conjugate was used to detect the biotinylated antibody. The threshold sensitivity of the ELISA was 100 pg/ml. Human IL-6 did not interfere with this assay at all.
K. Yoshizawa et al. /Cancer
85
Lett. 79 (1994) 83-89
2.7. Statistical analysis
changes of platelet number in peripheral blood of the hosts even in the absence of hIL-6 and/or antineoplastic drugs. In case of colon 26, platelet counts gradually increased from day 14 accompanying the tumor growth, and reached 97 x 104/j.d (day 0) to 130 x 104/pl (day 22) (Fig. 2a and b). Additional treatment with hIL-6 stimulated platelet production further. In contrast, in LLCbearing mice, platelet counts gradually decreased depending on the tumor growth (Fig. 2c and d). In this case, hIL-6 given to the mice substantially suppressed the platelet depression as shown in Fig. 2. In the case of colon 26 we presumed a participation of endogenous murine IL-6 production for the platelet increase, since it has been reported that colon 26 is a cachectic inducing tumor line and IL6 is involved in cachexia [17]. Thus, we analysed the serum levels of endogenous mIL-6 in colon 26bearing mice. Actually, mIL-6 were elevated depending on the tumor growth (Fig. 3). Interestingly, hIL-6 administration suppressed the production of mIL-6. In LLC-bearing mice, mIL-6 was not detected in sera during the experiment.
Student’s f-test was used to determine statistical differences. 3. Results 3.1. Effect of hIL-6 in mice treated with or without chemotherapeutic agents Prior to the experiment using tumor-bearing mice, the optimal dose of hIL-6 to increase platelet number was determined in normal mice. Subcutaneous injections of hIL-6 at various doses for 7 days exhibited a clear dose-dependent platelet production, and the median effective dose (ED,,) was calculated to be 28 kg/kg/day. Thus, we set up to administer 10 EDSo of hIL-6 (280 &kg/day) to obtain sufficient therapeutic effect thereafter. Similarly, suitable doses of anti-neoplastic drugs, which caused thrombocytopenia without decreasing the body weight, were also determined in non-tumor-bearing mice. At a dose of either 2 mg/kg MMC or 10 mg/kg CDDP, although transient thrombopoiesis was observed on day 4, moderate thrombocytopenia occurred thereafter (Fig. 1). In this condition, additional treatment with 10 EDse of hIL-6 could restore platelet number at nadir and shorten the duration of thrombocytopenia.
3.3. Effect of hIL-6 on thrombopoiesis following chemotherapy in tumor-bearing mice Using chemotherapy models, we assessed a clinical possibility of hIL-6 for the treatment of thrombocytopenia. Fig. 4a indicates the changes of platelet number in colon 26-bearing mice. The nadir occurred on day 18 and a rebound throm-
3.2. Effect of hIL-6 alone in tumor-bearing mice The tumor inoculation itself induced the
180’ a
60’ 0
-
-
Saline
1
10
20
60’
0
Saline
I 10
20
Fig. 1.Effect of hlL-6 on platelet production following chemotherapy. Normal C57BLi6 mice were administered anti-neoplastic drugs on day 0. From day 1, mice were consecutively given hlL-6 (280 &kg/day, s.c.) once a day. (a) MMC (2 mg/kg, i.p.): (b) CDDP (IO mg/kg, i.p.). The data represent mean f SE. *P < 0.05 compared with the control group.
K. Yoshizawjaet al. / Cancer Letr. 79 (1994) 83-89
86 ”
2:_;av/jJq
*
-O-
Saline hlL-6
Cl
0
80.
2
--
Saline hlL-6 I
40
I-
O
20
10
30
Day Fig. 2. Effects of hlL-6 on platelet production and tumor growth. Mice were inoculated with tumor cells on day 0 and consecutively administered hlL-6 (280 &kg/day, s.c.) once a day from day 8. (a) and (b) colon 26-bearing mice; (c) and (d) LLC-bearing mice. The data represent mean f SE. *P c 0.05 compared with the control group.
bocytosis was observed on day 22. During the hIL6 treatment, platelets did not decrease below the normal range. However, the thrombopoietic effect of hIL-6 in tumor bearing models was weaker than in normal mice. With respect to an anti-tumor effect, MMC itself significantly suppressed the growth of colon 26 on days 18 and 27, and hIL-6 exhibited no obvious modification for the anti-
2.0 -
-
” E
-
P
-
G A
-m--
Saline hlL-6 MMC+Saline MMC+hlL-6
I
l.O-
E ‘C z 00 0
20
10
30
Day Fig. 3. Murine IL-6 levels in colon 26-bearing mice. Mice were inoculated with colon 26 cells on day 0 and administered MMC (2 mg/kg, i.p.1 on day 7. From day 8, mice were consecutively given hIL-6 (280 &kg/day. s.c.) once a day. The data represent mean f SE. *P < 0.05 compared with the control group.
tumor activity (Figs. 2b and 4b). In LLC-bearing mice, CDDP caused severe platelet depression on day 18 (Fig. 4c). In the additional hIL-6-treated mice, platelet counts at the nadir significantly increased from 66 x lo4 to 86 x 104/pl. The antitumor effect of CDDP alone was observed on days 14 and 22 (Figs. 2d and 4d). but hIL-6 did not show obvious modification of the tumor growth. Although only on day 22 did hIL-6 exhibit a significant anti-tumor activity in combination with CDDP, the efficacy was marginal. 3.4. Carcass weight changes We evaluated whether interleukin-6 might be associated with the development of cachexia as has been proposed [17]. If this is true, exogenous administration of hIL-6 could induce or strengthen the occurrence of cachexia, especially in the mice bearing colon 26. Table 1 shows the carcass weight changes in tumor-bearing mice. The consecutive injections of hIL-6 tended to decrease the carcass weight in colon 26-bearing mice, while in nontumor-bearing BALB/c mice the effects of hIL-6 and/or MMC on the weight loss were not observed in a separate experiment (The weight gain was 1.6 - 3.1 g on day 27). On the other hand, a large tumor mass (about 4 g on day 18) caused the loss of carcass weight in all LLC-bearing mice. This model did not fulfill the criteria that cancer
K. Yoshizawa
et al. /Cancer
Lett.
160 - -
79 (1994)
87
83-89
-
Saline hlL-6
Saline hlL-6
120 > 80
8040
1
10,
b
E $__+_&I,
I-’
0
20
10
0
30
10
30
20 Day
Day
Fig. 4. Effects of hIL-6 on platelet production and tumor growth following chemotherapy. Mice were inoculated with tumor cells on day 0 and administered anti-neoplastic drugs on day 7. From day 8, mice were consecutively given hIL-6 (280 &kg/day, KC.) once a day. (a) and (b) colon 26-bearing mice that received MMC (2 mg/kg, i.p.); (c) and (d) LLC-bearing mice that received CDDP (IO mg/kg, i.p.). The data represent mean f SE. *P < 0.05 compared with the control group.
ing platelets can still be observed in tumor-bearing as well as in non-tumor-bearing animals. It is clear that our prolonged treatment with hIL-6 could provide a prominent reduction of the thrombocytopenia at nadir caused by anti-neoplastic drugs in both normal and tumor bearing models. However, other groups have reported that IL-6 injections for 6 or 7 days failed to increase platelet
cachexia appears with a small tumor burden [20]. Consequently, hIL-6 treatment did not augment the loss of carcass weight compared with controls. 4. Discussion The aim of the present study is to examine whether the therapeutic benefit of IL-6 in increas-
Table I Carcass weight changes Tumor
Colon 26
Treatment
Drug
Saline hIL-6 Saline hlL-6
LLC
in tumor-bearing
Carcass
weight changes
(8)”
Day 0
Day 7
Day I4
0
0.1 f 0.2s
0
0.5 f 0.6
MMC MMC
Saline hIL-6 Saline hIL-6
mice
CDDP CDDP
“Carcass weight was calculated as the difference between bMean A SE. ‘P < 0.05 compared with the respective control.
1.9 1.5 2.3 0.9
f f f zt
Day 18 0.7 I.0 0.7 0.8
0.7 l 0.8 0.5 f 0.7 -0.8 f 1.0 -0.7 l 0.9
0.7 f 0.7 1.3 + 0.6 1.0 l 0.7 1.6 + 0.5
Day 22
Day 27
2.1 f 0.9 -1.0 l 0.7c 2.3 zt 0.7 0.4 f 0.8
0.8 -1.1 1.4 0.5
-0.6
f 0.5
-2.5
f 0.8
-0.3 -0.9 -0.4
f 0.8 f 0.4 + 0.8
-1.5 -1.4 -1.5
f 0.9 f I.0 l 0.7
the whole body and tumor
weight.
f f f f
I.1 0.8 0.8 0.7
88
counts at nadir after 5-FU treatment [5,19] or caused slight recovery following radiation-induced myelosuppression [ 131in non-tumor bearing mice. The discrepancy between our findings and others might be based solely on the times of IL-6 injection. Another potential reason may be the different mechanisms of the chemotherapeutic agents used, especially acting on the differentiation process of megakaryocyte lineage. In fact, 5-FU administration has induced thrombocytopenia alone [5,19], but MMC and CDDP exhibited transient thrombopoiesis prior to the induction of thrombocytopenia. Although the mechanism of the transient thrombopoiesis is still vague, similar phenomena by MMC or CDDP have been reported in other papers [1,3). The late increase of platelet number on day 22 in colon 26-bearing mice treated with MMC might be a rebound phenomenon. was also Similar rebound thrombocytosis observed in 5-FU-treated normal mice. Sublethal dose of 5-FU induces the de novo hematopoiesis in a host accompanying the elevation of some hematopoietic growth factors. It has been reported that a significant quantity of IL-6 and stem cell factor (SCF) was detected in sera at platelet nadir after 5-FU treatment [8,19]. The rebound thrombocytosis caused by MMC administration may be a reflection of the synergistic effects of these hostderived cytokines and colon 26-secreted mIL-6. A recent study has shown that IL-6 is a major candidate for inducing cachexia in C-26 IVX (derived from colon 26)-bearing mice - known to be a cachectic model - since a monoclonal antibody to mIL-6 suppresses the development of cachexia [ 171. Our studies clearly support this view. Actually, consecutive injections of hIL-6 decreased the carcass weight, compared with controls, in colon 26-bearing mice. However, endogenous mIL-6 levels had an inverse correlation with the degree of the carcass weight loss. In our separate pharmacokinetic experiments, when normal mice received a single S.C.injection of hIL-6 at 300 &kg, the area under the curve of serum concentrations vs. time (AUC) was calculated as 400 ng. h/ml. If murine IL-6 levels of controls on day 22 (0.536 rig/ml) continue for 24 h, AUC is calculated as 13 ng. h/ml. In hIL-6-treated groups, total AUC of endogenous mIL-6 and exogenous hIL-6 are clearly much larger than that of the endogenous mIL-6 in con-
K. Yoshizawa er al. /Cancer
Lert. 79 (1994) 83-89
trols. The data explains well that hIL-6 administration might also play a role in inducing the loss of carcass weight except for the endogenous serum mIL-6 levels. A question arises as to why hIL-6 injections suppressed the production of endogenous mIL-6 (Fig. 3). Strassmann et al. reported that murine macrophage interacting with C-26 IVX produced IL-I, which stimulated the production of IL-6 by the tumor [18]. In contrast, Schindler et al. demonstrated that IL-6 suppressed the secretion of IL-I from human blood mononuclear cells [ 141. These observations suggest that a kind of feedback regulation might be present between these cytokines. Thus it is reasonable to consider that exogenously administered hIL-6 could inhibit the secretion of IL-I resulting in the regression of the endogenous murine IL-6 in mice. There remains a major obstacle regarding IL-6 as a pivotal cancer cachectic substance from our studies alone, because hIL-6 decreased carcass weight in colon 26-bearing mice, but not in LLC-bearing mice or normal BALB/c mice in spite of the administration of a sufficient dose. Other factors produced in the tumor stimulated by IL-6 might also be involved in the induction of the carcass weight loss. These lindings suggest that even in the same treatment, the degree of the physiological changes by IL-6 is quite different depending on tumor species. As cachexia is caused by the interaction between the host and the tumor in a complex manner, IL-6 may not always induce physiological changes on clinical treatment. However, IL-6 administration to cancer patients for the treatment of thrombocytopenia should be carried out, with careful monitoring of the serum IL-6 level and cachectic markers besides the platelet count. 5. Acknowledgments
We wish to thank Dr Tateshi Kataoka (Cancer Chemotherapy Center, Japanese Foundation for Cancer Research, Tokyo) and Dr Masaaki Iigo (Chemotherapy Division, National Cancer Center Research Institute, Tokyo) for the helpful discussions, and Dr J. Allan Waintz (DNAX Research Institute, Palo Alto) for providing the 20F3 antimIL-6 antibody.
K. Yoshkawa et al. /Cancer
Lett. 79 (1994) 83-89
89
6. References 1
2
3
4
5
6
I
8
9
10
11
Akamatsu, K., Endo, K., Matsumoto, T., Morikawa, K., Koizumi, M., Koizumi, K. and Mitsui, H. (1991) Toxicological and tumoricidal evaluations of a new platinum complex, (-)-(R)-2-aminomethylpyrrolidine (I, 1-cyclebutanedicarboxylato) platinum(n) monohydrate, in rats. Jpn. J. Cancer Res., 82, 724-731. Asano, S., Okano, A., Ozawa, K., Nakahata. T., Ishibashi, T., Koike, K., Kimura, H., Tanioka, Y., Shibuya, A., Hirano. T., Kishimoto, T., Takaku, F. and Akiyama, Y. (1990) In vivo effects of recombinant human interleukin-6 in primates: stimulated production of platelets. Blood, 75, 1602-1605. Ashizawa. T.. Okabe, M., Gomi. K. and Hirata, T. (1993) Reduced bone marrow toxicity of KW-2149, a mitomycin C derivative, in mice. Anti-Cancer Drugs, 4, 181-188. Burstein, S.A., Downs, T., Friese, P.. Lynam, S., Anderson, S., Henthorn, J., Epstein, R.B. and Savage, K. (1992) Thrombocytopoiesis in normal and sublethally irradiated dogs: response to human interleukin-6. Blood, 80, 420-428. Carrington, P.A., Hill, R.J., Levin, J. and Verotta, D. (1992) Effects of interleukin 3 and interleukin 6 on platelet recovery in mice treated with S-fluorouracil. Exp. Hematol., 20. 462-469. Hardisty, R.M. (1990) Platelet disorders. In: Postgraduate Haematology, pp. 598-626. Editors: A.V. Hoffbrand and SM. Lewis, Heinemann Medical Books, Oxford. Hardy, C.L. and Balducci, L. (1986) Early hematopoietic events during tumor growth in mice. J. Natl. Cancer Inst.. 76, 535-540. Hunt, P., Zsebo, K.M.. Hokom, M.M., Hornkohl, A., Birkett, N.C., Castillo, J.C. d. and Martin, F. (1992) Evidence that stem cell factor is involved in the rebound thrombocytosis that follows 5-fluorouracil treatment. Blood, 80, 904-91 I. Ida, N., Sakurai, S., Hosaka, T., Hosoi, K., Kunitomo. T., Matsuura, Y. and Kohase, M. (1990) An enzymelinked immunosorbent assay for the measurement of human interleukin-6. J. Immunol. Methods. 133, 279-284. Ishibashi, T.. Kimura, H.. Shikama. Y., Uchida, T., Kariyone. S., Hirano, T., Kishimoto, T.. Takatsuki, F. and Akiyama, Y. (1989) Interleukin-6 is a potent thrombopoietic factor in vivo in mice. Blood. 74, 1241-1244. Kimura. H., Ishibashi, T., Uchida. T., Maruyama. Y., Friese. P. and Burstein, S.A. (1990) Interleukin 6 is a dif-
I2
I3
I4
I5
16
17
18
19
20
21
ferentiation factor for human megakaryocytes in vitro. Eur. J. Immunol., 20, 1927-1931. Levin, J. and Conley, C.L. (1964) Thrombocytosis associated with malignant disease. Arch. Intern. Med., 114, 497-500. Patchen, M.L., MacVittie, T.J., Williams, J.L., Schwartz, G.N. and Souza, L.M. (1991) Administration of interleukin-6 stimulates multilineage hematopoiesis and accelerates recovery from radiation-induced hematopoietic depression. Blood, 77, 472-480. Schindler, R., Mancilla, J.. Endres. S., Ghorbani, R., Clark, S.C. and Dinarello, C.A. (1990) Correlations and interactions in the production of interleukind (IL-6) IL1, and tumor necrosis factor (TNF) in human blood mononuclear cells: IL-6 suppresses IL-I and TNF. Blood, 75. 40-47. Silvis, S.E., Turkbas, N. and Doscherholmen, A. (1970) Thrombocytosis in paiients with lung cancer. J. Am. Med. Asoc., 211, 1852-1853. Starnes, H.F., Pearce, M.K., Tewari, A., Yim, J.H.. Zou, J.-C. and Abrams, J.S. (1990) Anti-IL-6 monoclonal antibodies protect against lethal Escherichia co/i infection and lethal tumor necrosis factor-o challenge in mice. J. Immunol., 145, 4185-4191. Strassmann, G.. Fong, M., Kenney, J.S. and Jacob, C.O. (1992) Evidence for the involvement of interleukin 6 in experimental cancer cachexia. J. Clin. Invest.. 89, 1681-1684. Strassmann, G., Jacob, C.O., Evance. R.. Beall, D. and Fong, M. (1992) Mechanisms of experimental cancer cachexia: interaction between mononuclear phagccytes and colon-26 carcinoma and its relevance to IL-6mediated cancer cachexia. J. Immunol., 148. 3674-3678. Takatsuki, F., Okano, A., Suzuka, C., Miyasaka, Y., Hirano. T., Kishimoto, T., Ejima, D. and Akiyama, Y. (1990) lnterleukin 6 perfusion stimulates reconstitution of the immune and hematopoietic systems after 5fluorouracil treatment. Cancer Res., 50, 2885-2890. Tanaka, Y.. Eda, H.. Tanaka, T.. Udagawa, T.. Ishikawa, T., Horii. I., Ishitsuka, H., Kataoka, T. and Taguchi, T. (1990) Experimental cancer cachexia induced by transplantable colon 26 adenocarcinoma in mice. Cancer Res.. 50, 2290-2295. Young, M.R.. Newby. M. and Wepsic, H.T. (1987) Hematopoiesis and suppressor bone marrow cells in mice bearing large metastatic Lewis lung carcinoma tumors. Cancer Res., 47. 100-105.
ELSEVIER SCIENCE IRELAND
Cancer Letters 79 (1994) 83-89
Effects of natural human interleukin-6 on thrombopoiesis and tumor progression in tumor-bearing mice Ken-ichi
Yoshizawa*, Tasuku Okamoto, Keiko Moriya, Yu-ichiro Satoh, Masanobu Naruto, Nobutaka
Akiko Ida
Izawa,
Basic Research Laboratories, Toray Industries Inc.. I I I I Tebiro, Kamakura, Kanagawa 248, Japan
(Received 25 January 1994; revision received 20 February 1994: accepted 24 February 1994)
Abstract The growth of inoculated colon 26 adenocarcinoma (colon 26) in mice gradually increased the platelet count owing to murine IL-6 secreted from the tumor, while Lewis lung carcinoma (LLC) decreased the platelet count in the hosts, depending on the tumor growth. Natural human IL-6 injections (hIL-6), 280 &kg/day, stimulated the platelet production in both types of carcinoma-bearing mice. When the administration of mitomycin C or cisplatin decreased the platelet number as a side reaction with a concomitant of suppressing the growth of colon 26 and LLC. respectively, hIL-6 could also increase the platelet count without the augmentation of tumor growth. However, loss of carcass weight was observed in colon 26-bearing mice treated with hIL-6, suggesting the development of cachexia is associated with hIL-6 administration. Despite the possibility of inducing cachexia in some types of tumors, our results suggest that IL-6 could be a useful means of restoring the decreased platelet number in cancer patients after intensive chemotherapy. Key words: Interleukin-6;
Thrombocytopenia;
Anti-neoplastic
1. Introduction Bone marrow suppression is a major problem in the treatment of progressive tumors due to intensive chemotherapy or radiation exposure in the case of bone marrow transplantation. G-CSF is efficiently used for the treatment of decreased neutrophil production, but a useful modality for thrombocytopenia has not been established. In ad-
* Corresponding author.
drug; Carcass
weight
dition, malignant tumors frequently alter the hematopoietic balance, depending on the tumor growth, in tumor-bearing animals [7,21] and cancer patients [6,12,15]. In chronic lymphocytic leukemia, thrombocytopenia is observed in association with the presence of auto antibodies against circulating platelets and excessive platelet pooling in an enlarged spleen. Hodgkin’s disease, breast cancer and lung cancer are tumors with which high platelet counts are frequently associated. Interleukin-6 (IL-6) is a multifunctional cyto-
0304-3835/94/$07.00 0 1994 Elsevier Science Ireland Ltd. All rights reserved. SSDI 0304-3835(94)03313-g
84
kine that has various biological activities, including the maturation of megakaryocytes in vitro [1 l] and the production of peripheral platelets in vivo [2, lo]. The thrombopoietic effect of IL-6 was studied in the treatment of platelet depression following hematopoietic damage induced by 5fluorouracil(5-FU) administration [5] or radiation exposure [4,13] in animals. Though IL-6 may be clinically valuable for the treatment of thrombocytopenia in cancer patients, there remain other problems to be solved. For example, Strassmann and coworkers have shown that IL-6 is an important mediator of cachexia [ 17,181 which induces anorexia, asthenia and anemia, and causes a reduced response to anti-neoplastic drugs [20]. This implies that IL-6 treatment may induce side reactions in cancer patients who are in a poor condition due to chemotherapy. However, there have been few studies concerning the effects of IL-6 on thrombopoiesis in tumor-bearing animals. We describe the thrombopoietic effect of natural human interleukind (hIL-6) in mice in which the hematopoietic balance is disordered due to tumor burden and chemotherapy, and clarify the side reactions, such as the induction of cachexia or the augmentation of malignant tumors. 2. Materials and methods 2.1. Mice C57BL/6 and BALB/c male mice were obtained from Charles River Japan (Kanagawa, Japan) and used at the age of 8 weeks. One group consisted of 6 mice. 2.2. Natural human IL-6 hIL-6 was prepared from human diploid fibroblast, DIP-2, supernatant fluid [9]. One milligram of hIL-6 exhibited 3.1 x lo6 units of WHO international standard activity. Endotoxin contamination was less than 0.05 EU/mg protein, as measured with Endospecye (Seikagaku Kogyo, Tokyo, Japan). 2.3. Tumor cells
Lewis lung carcinoma (LLC) was cultured in vitro in RPM11640 containing 10% fetal calf serum (FCS). After suspending the cells with a
K. Yoshi:wu
rt ul. / Cuncer Lear. 79 (1994) 83-89
pipette, 2 x 10’ cells were intradermally inoculated into the inguinal flank of C57BL/6 mice. Colon 26 adenocarcinoma (colon 26) was maintained in carrier mice. Blocks of the tumor were minced into a single cell suspension using a stainless steel mesh, and 1 x lo6 cells were intradermally inoculated and transferred into BALB/c mice. These tumor cell lines were kindly supplied by Dr T. Kataoka (Cancer Chemotherapy Center, Japanese Foundation for Cancer Research, Tokyo) 2.4. Experimental protocol Mice were inoculated with tumor cells on day 0 and were treated with chemotherapy on day 7. Mitomycin C (MMC) (Kyowa Hakko Kogyo, Tokyo, Japan) was administered to BALBic mice as a single i.p. injection, 2 mg/kg. Cisplatin (CDDP) (Bristol Myers, Tokyo, Japan) was given to C57BL/6 mice as a single i.p. injection, 10 mg/kg hIL-6 was consecutively given as S.C. injections. 280 &kg/day, once a day from day 8. Mice treated with sterile saline served as controls. 2.5. Assays Under anesthesia with ether, mice were bled by cutting the armpit artery and dissected at the times given below in Results. Platelet counts were performed with a Sysmex hematology analyser (Toa Medical Electronics, Hyogo, Japan). Tumor, lung, liver, kidney and spleen tissues were weighed on an analytical balance. Carcass weight was calculated as the difference between the whole body and tumor weight. 2.6. Murine IL-6 assay
Murine IL-6 (mIL-6) was assayed by the modified ELISA method of Starnes et al. [ 161. Briefly, the tracer antibody was changed to goat anti-mIL6 polyclonal antibody (R & D Systems, Minneapolis, MN) conjugated with N-hydroxysuccinimide-Lc-biotin (Pierce, Rockford, IL) in the molar ratio of 1 to 40. Streptoavidinhorseradish peroxidase conjugate was used to detect the biotinylated antibody. The threshold sensitivity of the ELISA was 100 pg/ml. Human IL-6 did not interfere with this assay at all.
K. Yoshizawa et al. /Cancer
85
Lett. 79 (1994) 83-89
2.7. Statistical analysis
changes of platelet number in peripheral blood of the hosts even in the absence of hIL-6 and/or antineoplastic drugs. In case of colon 26, platelet counts gradually increased from day 14 accompanying the tumor growth, and reached 97 x 104/j.d (day 0) to 130 x 104/pl (day 22) (Fig. 2a and b). Additional treatment with hIL-6 stimulated platelet production further. In contrast, in LLCbearing mice, platelet counts gradually decreased depending on the tumor growth (Fig. 2c and d). In this case, hIL-6 given to the mice substantially suppressed the platelet depression as shown in Fig. 2. In the case of colon 26 we presumed a participation of endogenous murine IL-6 production for the platelet increase, since it has been reported that colon 26 is a cachectic inducing tumor line and IL6 is involved in cachexia [17]. Thus, we analysed the serum levels of endogenous mIL-6 in colon 26bearing mice. Actually, mIL-6 were elevated depending on the tumor growth (Fig. 3). Interestingly, hIL-6 administration suppressed the production of mIL-6. In LLC-bearing mice, mIL-6 was not detected in sera during the experiment.
Student’s f-test was used to determine statistical differences. 3. Results 3.1. Effect of hIL-6 in mice treated with or without chemotherapeutic agents Prior to the experiment using tumor-bearing mice, the optimal dose of hIL-6 to increase platelet number was determined in normal mice. Subcutaneous injections of hIL-6 at various doses for 7 days exhibited a clear dose-dependent platelet production, and the median effective dose (ED,,) was calculated to be 28 kg/kg/day. Thus, we set up to administer 10 EDSo of hIL-6 (280 &kg/day) to obtain sufficient therapeutic effect thereafter. Similarly, suitable doses of anti-neoplastic drugs, which caused thrombocytopenia without decreasing the body weight, were also determined in non-tumor-bearing mice. At a dose of either 2 mg/kg MMC or 10 mg/kg CDDP, although transient thrombopoiesis was observed on day 4, moderate thrombocytopenia occurred thereafter (Fig. 1). In this condition, additional treatment with 10 EDse of hIL-6 could restore platelet number at nadir and shorten the duration of thrombocytopenia.
3.3. Effect of hIL-6 on thrombopoiesis following chemotherapy in tumor-bearing mice Using chemotherapy models, we assessed a clinical possibility of hIL-6 for the treatment of thrombocytopenia. Fig. 4a indicates the changes of platelet number in colon 26-bearing mice. The nadir occurred on day 18 and a rebound throm-
3.2. Effect of hIL-6 alone in tumor-bearing mice The tumor inoculation itself induced the
180’ a
60’ 0
-
-
Saline
1
10
20
60’
0
Saline
I 10
20
Fig. 1.Effect of hlL-6 on platelet production following chemotherapy. Normal C57BLi6 mice were administered anti-neoplastic drugs on day 0. From day 1, mice were consecutively given hlL-6 (280 &kg/day, s.c.) once a day. (a) MMC (2 mg/kg, i.p.): (b) CDDP (IO mg/kg, i.p.). The data represent mean f SE. *P < 0.05 compared with the control group.
K. Yoshizawjaet al. / Cancer Letr. 79 (1994) 83-89
86 ”
2:_;av/jJq
*
-O-
Saline hlL-6
Cl
0
80.
2
--
Saline hlL-6 I
40
I-
O
20
10
30
Day Fig. 2. Effects of hlL-6 on platelet production and tumor growth. Mice were inoculated with tumor cells on day 0 and consecutively administered hlL-6 (280 &kg/day, s.c.) once a day from day 8. (a) and (b) colon 26-bearing mice; (c) and (d) LLC-bearing mice. The data represent mean f SE. *P c 0.05 compared with the control group.
bocytosis was observed on day 22. During the hIL6 treatment, platelets did not decrease below the normal range. However, the thrombopoietic effect of hIL-6 in tumor bearing models was weaker than in normal mice. With respect to an anti-tumor effect, MMC itself significantly suppressed the growth of colon 26 on days 18 and 27, and hIL-6 exhibited no obvious modification for the anti-
2.0 -
-
” E
-
P
-
G A
-m--
Saline hlL-6 MMC+Saline MMC+hlL-6
I
l.O-
E ‘C z 00 0
20
10
30
Day Fig. 3. Murine IL-6 levels in colon 26-bearing mice. Mice were inoculated with colon 26 cells on day 0 and administered MMC (2 mg/kg, i.p.1 on day 7. From day 8, mice were consecutively given hIL-6 (280 &kg/day. s.c.) once a day. The data represent mean f SE. *P < 0.05 compared with the control group.
tumor activity (Figs. 2b and 4b). In LLC-bearing mice, CDDP caused severe platelet depression on day 18 (Fig. 4c). In the additional hIL-6-treated mice, platelet counts at the nadir significantly increased from 66 x lo4 to 86 x 104/pl. The antitumor effect of CDDP alone was observed on days 14 and 22 (Figs. 2d and 4d). but hIL-6 did not show obvious modification of the tumor growth. Although only on day 22 did hIL-6 exhibit a significant anti-tumor activity in combination with CDDP, the efficacy was marginal. 3.4. Carcass weight changes We evaluated whether interleukin-6 might be associated with the development of cachexia as has been proposed [17]. If this is true, exogenous administration of hIL-6 could induce or strengthen the occurrence of cachexia, especially in the mice bearing colon 26. Table 1 shows the carcass weight changes in tumor-bearing mice. The consecutive injections of hIL-6 tended to decrease the carcass weight in colon 26-bearing mice, while in nontumor-bearing BALB/c mice the effects of hIL-6 and/or MMC on the weight loss were not observed in a separate experiment (The weight gain was 1.6 - 3.1 g on day 27). On the other hand, a large tumor mass (about 4 g on day 18) caused the loss of carcass weight in all LLC-bearing mice. This model did not fulfill the criteria that cancer
K. Yoshizawa
et al. /Cancer
Lett.
160 - -
79 (1994)
87
83-89
-
Saline hlL-6
Saline hlL-6
120 > 80
8040
1
10,
b
E $__+_&I,
I-’
0
20
10
0
30
10
30
20 Day
Day
Fig. 4. Effects of hIL-6 on platelet production and tumor growth following chemotherapy. Mice were inoculated with tumor cells on day 0 and administered anti-neoplastic drugs on day 7. From day 8, mice were consecutively given hIL-6 (280 &kg/day, KC.) once a day. (a) and (b) colon 26-bearing mice that received MMC (2 mg/kg, i.p.); (c) and (d) LLC-bearing mice that received CDDP (IO mg/kg, i.p.). The data represent mean f SE. *P < 0.05 compared with the control group.
ing platelets can still be observed in tumor-bearing as well as in non-tumor-bearing animals. It is clear that our prolonged treatment with hIL-6 could provide a prominent reduction of the thrombocytopenia at nadir caused by anti-neoplastic drugs in both normal and tumor bearing models. However, other groups have reported that IL-6 injections for 6 or 7 days failed to increase platelet
cachexia appears with a small tumor burden [20]. Consequently, hIL-6 treatment did not augment the loss of carcass weight compared with controls. 4. Discussion The aim of the present study is to examine whether the therapeutic benefit of IL-6 in increas-
Table I Carcass weight changes Tumor
Colon 26
Treatment
Drug
Saline hIL-6 Saline hlL-6
LLC
in tumor-bearing
Carcass
weight changes
(8)”
Day 0
Day 7
Day I4
0
0.1 f 0.2s
0
0.5 f 0.6
MMC MMC
Saline hIL-6 Saline hIL-6
mice
CDDP CDDP
“Carcass weight was calculated as the difference between bMean A SE. ‘P < 0.05 compared with the respective control.
1.9 1.5 2.3 0.9
f f f zt
Day 18 0.7 I.0 0.7 0.8
0.7 l 0.8 0.5 f 0.7 -0.8 f 1.0 -0.7 l 0.9
0.7 f 0.7 1.3 + 0.6 1.0 l 0.7 1.6 + 0.5
Day 22
Day 27
2.1 f 0.9 -1.0 l 0.7c 2.3 zt 0.7 0.4 f 0.8
0.8 -1.1 1.4 0.5
-0.6
f 0.5
-2.5
f 0.8
-0.3 -0.9 -0.4
f 0.8 f 0.4 + 0.8
-1.5 -1.4 -1.5
f 0.9 f I.0 l 0.7
the whole body and tumor
weight.
f f f f
I.1 0.8 0.8 0.7
88
counts at nadir after 5-FU treatment [5,19] or caused slight recovery following radiation-induced myelosuppression [ 131in non-tumor bearing mice. The discrepancy between our findings and others might be based solely on the times of IL-6 injection. Another potential reason may be the different mechanisms of the chemotherapeutic agents used, especially acting on the differentiation process of megakaryocyte lineage. In fact, 5-FU administration has induced thrombocytopenia alone [5,19], but MMC and CDDP exhibited transient thrombopoiesis prior to the induction of thrombocytopenia. Although the mechanism of the transient thrombopoiesis is still vague, similar phenomena by MMC or CDDP have been reported in other papers [1,3). The late increase of platelet number on day 22 in colon 26-bearing mice treated with MMC might be a rebound phenomenon. was also Similar rebound thrombocytosis observed in 5-FU-treated normal mice. Sublethal dose of 5-FU induces the de novo hematopoiesis in a host accompanying the elevation of some hematopoietic growth factors. It has been reported that a significant quantity of IL-6 and stem cell factor (SCF) was detected in sera at platelet nadir after 5-FU treatment [8,19]. The rebound thrombocytosis caused by MMC administration may be a reflection of the synergistic effects of these hostderived cytokines and colon 26-secreted mIL-6. A recent study has shown that IL-6 is a major candidate for inducing cachexia in C-26 IVX (derived from colon 26)-bearing mice - known to be a cachectic model - since a monoclonal antibody to mIL-6 suppresses the development of cachexia [ 171. Our studies clearly support this view. Actually, consecutive injections of hIL-6 decreased the carcass weight, compared with controls, in colon 26-bearing mice. However, endogenous mIL-6 levels had an inverse correlation with the degree of the carcass weight loss. In our separate pharmacokinetic experiments, when normal mice received a single S.C.injection of hIL-6 at 300 &kg, the area under the curve of serum concentrations vs. time (AUC) was calculated as 400 ng. h/ml. If murine IL-6 levels of controls on day 22 (0.536 rig/ml) continue for 24 h, AUC is calculated as 13 ng. h/ml. In hIL-6-treated groups, total AUC of endogenous mIL-6 and exogenous hIL-6 are clearly much larger than that of the endogenous mIL-6 in con-
K. Yoshizawa er al. /Cancer
Lert. 79 (1994) 83-89
trols. The data explains well that hIL-6 administration might also play a role in inducing the loss of carcass weight except for the endogenous serum mIL-6 levels. A question arises as to why hIL-6 injections suppressed the production of endogenous mIL-6 (Fig. 3). Strassmann et al. reported that murine macrophage interacting with C-26 IVX produced IL-I, which stimulated the production of IL-6 by the tumor [18]. In contrast, Schindler et al. demonstrated that IL-6 suppressed the secretion of IL-I from human blood mononuclear cells [ 141. These observations suggest that a kind of feedback regulation might be present between these cytokines. Thus it is reasonable to consider that exogenously administered hIL-6 could inhibit the secretion of IL-I resulting in the regression of the endogenous murine IL-6 in mice. There remains a major obstacle regarding IL-6 as a pivotal cancer cachectic substance from our studies alone, because hIL-6 decreased carcass weight in colon 26-bearing mice, but not in LLC-bearing mice or normal BALB/c mice in spite of the administration of a sufficient dose. Other factors produced in the tumor stimulated by IL-6 might also be involved in the induction of the carcass weight loss. These lindings suggest that even in the same treatment, the degree of the physiological changes by IL-6 is quite different depending on tumor species. As cachexia is caused by the interaction between the host and the tumor in a complex manner, IL-6 may not always induce physiological changes on clinical treatment. However, IL-6 administration to cancer patients for the treatment of thrombocytopenia should be carried out, with careful monitoring of the serum IL-6 level and cachectic markers besides the platelet count. 5. Acknowledgments
We wish to thank Dr Tateshi Kataoka (Cancer Chemotherapy Center, Japanese Foundation for Cancer Research, Tokyo) and Dr Masaaki Iigo (Chemotherapy Division, National Cancer Center Research Institute, Tokyo) for the helpful discussions, and Dr J. Allan Waintz (DNAX Research Institute, Palo Alto) for providing the 20F3 antimIL-6 antibody.
K. Yoshkawa et al. /Cancer
Lett. 79 (1994) 83-89
89
6. References 1
2
3
4
5
6
I
8
9
10
11
Akamatsu, K., Endo, K., Matsumoto, T., Morikawa, K., Koizumi, M., Koizumi, K. and Mitsui, H. (1991) Toxicological and tumoricidal evaluations of a new platinum complex, (-)-(R)-2-aminomethylpyrrolidine (I, 1-cyclebutanedicarboxylato) platinum(n) monohydrate, in rats. Jpn. J. Cancer Res., 82, 724-731. Asano, S., Okano, A., Ozawa, K., Nakahata. T., Ishibashi, T., Koike, K., Kimura, H., Tanioka, Y., Shibuya, A., Hirano. T., Kishimoto, T., Takaku, F. and Akiyama, Y. (1990) In vivo effects of recombinant human interleukin-6 in primates: stimulated production of platelets. Blood, 75, 1602-1605. Ashizawa. T.. Okabe, M., Gomi. K. and Hirata, T. (1993) Reduced bone marrow toxicity of KW-2149, a mitomycin C derivative, in mice. Anti-Cancer Drugs, 4, 181-188. Burstein, S.A., Downs, T., Friese, P.. Lynam, S., Anderson, S., Henthorn, J., Epstein, R.B. and Savage, K. (1992) Thrombocytopoiesis in normal and sublethally irradiated dogs: response to human interleukin-6. Blood, 80, 420-428. Carrington, P.A., Hill, R.J., Levin, J. and Verotta, D. (1992) Effects of interleukin 3 and interleukin 6 on platelet recovery in mice treated with S-fluorouracil. Exp. Hematol., 20. 462-469. Hardisty, R.M. (1990) Platelet disorders. In: Postgraduate Haematology, pp. 598-626. Editors: A.V. Hoffbrand and SM. Lewis, Heinemann Medical Books, Oxford. Hardy, C.L. and Balducci, L. (1986) Early hematopoietic events during tumor growth in mice. J. Natl. Cancer Inst.. 76, 535-540. Hunt, P., Zsebo, K.M.. Hokom, M.M., Hornkohl, A., Birkett, N.C., Castillo, J.C. d. and Martin, F. (1992) Evidence that stem cell factor is involved in the rebound thrombocytosis that follows 5-fluorouracil treatment. Blood, 80, 904-91 I. Ida, N., Sakurai, S., Hosaka, T., Hosoi, K., Kunitomo. T., Matsuura, Y. and Kohase, M. (1990) An enzymelinked immunosorbent assay for the measurement of human interleukin-6. J. Immunol. Methods. 133, 279-284. Ishibashi, T.. Kimura, H.. Shikama. Y., Uchida, T., Kariyone. S., Hirano, T., Kishimoto, T.. Takatsuki, F. and Akiyama, Y. (1989) Interleukin-6 is a potent thrombopoietic factor in vivo in mice. Blood. 74, 1241-1244. Kimura. H., Ishibashi, T., Uchida. T., Maruyama. Y., Friese. P. and Burstein, S.A. (1990) Interleukin 6 is a dif-
I2
I3
I4
I5
16
17
18
19
20
21
ferentiation factor for human megakaryocytes in vitro. Eur. J. Immunol., 20, 1927-1931. Levin, J. and Conley, C.L. (1964) Thrombocytosis associated with malignant disease. Arch. Intern. Med., 114, 497-500. Patchen, M.L., MacVittie, T.J., Williams, J.L., Schwartz, G.N. and Souza, L.M. (1991) Administration of interleukin-6 stimulates multilineage hematopoiesis and accelerates recovery from radiation-induced hematopoietic depression. Blood, 77, 472-480. Schindler, R., Mancilla, J.. Endres. S., Ghorbani, R., Clark, S.C. and Dinarello, C.A. (1990) Correlations and interactions in the production of interleukind (IL-6) IL1, and tumor necrosis factor (TNF) in human blood mononuclear cells: IL-6 suppresses IL-I and TNF. Blood, 75. 40-47. Silvis, S.E., Turkbas, N. and Doscherholmen, A. (1970) Thrombocytosis in paiients with lung cancer. J. Am. Med. Asoc., 211, 1852-1853. Starnes, H.F., Pearce, M.K., Tewari, A., Yim, J.H.. Zou, J.-C. and Abrams, J.S. (1990) Anti-IL-6 monoclonal antibodies protect against lethal Escherichia co/i infection and lethal tumor necrosis factor-o challenge in mice. J. Immunol., 145, 4185-4191. Strassmann, G.. Fong, M., Kenney, J.S. and Jacob, C.O. (1992) Evidence for the involvement of interleukin 6 in experimental cancer cachexia. J. Clin. Invest.. 89, 1681-1684. Strassmann, G., Jacob, C.O., Evance. R.. Beall, D. and Fong, M. (1992) Mechanisms of experimental cancer cachexia: interaction between mononuclear phagccytes and colon-26 carcinoma and its relevance to IL-6mediated cancer cachexia. J. Immunol., 148. 3674-3678. Takatsuki, F., Okano, A., Suzuka, C., Miyasaka, Y., Hirano. T., Kishimoto, T., Ejima, D. and Akiyama, Y. (1990) lnterleukin 6 perfusion stimulates reconstitution of the immune and hematopoietic systems after 5fluorouracil treatment. Cancer Res., 50, 2885-2890. Tanaka, Y.. Eda, H.. Tanaka, T.. Udagawa, T.. Ishikawa, T., Horii. I., Ishitsuka, H., Kataoka, T. and Taguchi, T. (1990) Experimental cancer cachexia induced by transplantable colon 26 adenocarcinoma in mice. Cancer Res.. 50, 2290-2295. Young, M.R.. Newby. M. and Wepsic, H.T. (1987) Hematopoiesis and suppressor bone marrow cells in mice bearing large metastatic Lewis lung carcinoma tumors. Cancer Res., 47. 100-105.