Antitumor Activity of Actinonin in Vitro and in Vivo

Vol. 4, 171-176, January 1998 Clinical Cancer Research 171

Yang Xu, Lawrence T. Lai, Janice L. Gabrilove, 8), mye!oid and monocytic cells, and most myeloblastic leuke- and David A. Scheinberg’ mias as well as on cells and tissues outside the hematopoietic system including fibroblasts, intestinal epithelium, renal tubular Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, New York 10021 epithelium, and synaptic membranes of the central nervous system (I ). APN occurs as a homodimer, and the molecule is a 150-kDa, transmembrane glycoprotein with an intracellular ABSTRACT amino terminus (1). F23, an antihuman CD13/APN mAb, is able Actinonin, an antibiotic and CD13/aminopeptidase N to completely block the active site of the enzyme (9). (APN) inhibitor, has been shown to be cytotoxic to tumor Bestatin, a CD 1 3/APN inhibitor, was examined in preclin- cell lines in vitro. We investigated the antiproliferative ef- ical and clinical studies; bestatin could inhibit lymph node fects of actinonin on human and murine leukemia and lym- metastasis of P388 leukemia in mice (10) and was used in phoma cells. Actinonin inhibited growth of NB4 and HL6O clinical trials in malignant skin tumors (1 1), in head and neck human cell lines and AKR mouse leukemia cells in vitro with cancer (12), in esophageal cancer (13), and in gynecologic an IC50 of about 2-S g/ml. The inhibitory effect on CD13- tumors (14). High doses of bestatin resulted in the significant positive cells was not blocked by pretreatment with the inhibition of preexisting experimental and spontaneous metas- anti-CD13/APN monoclonal antibody F23, which binds with tasis in mice (15). Results with bestatin prompted us to examine high affinity to the active site of CD13/APN and blocks its the actions of actinonin as an anticancer agent. Actinonin, a enzymatic activity. Moreover, F23 alone was not inhibitory naturally occurring antibiotic derivative of L-prolinol and a to CD13-positive cells. Furthermore, a similar inhibitory potent CD13/APN inhibitor, is obtained from the culture fib- IC50 of actinonin was seen in the CD13-negative cell lines trates of a Streptomyces species classified as Streptomyces Cut- RA.JI and DAUDI human lymphoma. These data suggest ter C/2 NCIB 8845 (16). Actinonin has been shown to be that the inhibitory effect of actinonin is not mediated by generally active against Gram-positive bacteria. The action of inhibition of CD13/APN. Cell cycle analysis showed that the antibiotic involves disruption of RNA synthesis in bacteria. actinonin induces a G1 arrest in HL6O and NB4 cells; apop- In vivo, it has no apparent toxicity to mice in doses up to 400 tosis was observed in 20-35% of the cells as measured by mg/kg body weight (16). Actinonin is eight times more potent intracellular flow cytometry. To assess whether these effects than bestatin (9). could be seen in vivo, the effect of actinonin on the syngeneic We report here the antiproliferative effects of actinonin on AKR mouse leukemia model was evaluated. Actinonin human leukemia and bymphoma cells in vitro and the treatment showed dose-dependent antitumor effects on AKR leukemia of syngeneic AKR leukemia in an AKR mouse model using in vivo, resulting in a survival advantage. In conclusion, actinonin. apoptosis, growth inhibition, and therapeutic effects in vivo are induced by actinonin and are not likely to be mediated by CD13/APN. MATERIALS AND METHODS Materials. Actinonin, amastatin, and bestatin were pur- INTRODUCTION chased from Sigma (St. Louis, MO). mAbs 4B4, OKT9, Leu- CD 1 3/APN2 (EC 3.4. 1 1.2) is a ubiquitous cell surface zinc 1 la, Leu-15, Leu-Ml, MY4, control IgG1, and IgG2a were aminopeptidase involved in down-regulation of regulatory pep- purchased from Coulter (Hialeah, FL) or Becton Dickinson (San tide signals (1). Recently, APN has been shown to be the major Jose, CA). Fluoresceinated, affinity-purified goat antiserum to receptor for the enteropathogenic coronavirus TGEV (2) and for mouse immunoglobulins (GAM-FITC) was purchased from human coronavirus 229E (3) and to be involved in tumor cell Kirkegaard & Perry (Gaithersburg, MD). mAbs F23, TA99, invasion (4, 5). Human APN is identical to the myeboid differ- JD12, and M195 were produced in Memorial Sloan-Kettering entiation antigen CD13 (6, 7) found on HL6O leukemia cells (7, Cancer Center. Recombinant human IL-3 was obtained from Amgen (Thousand Oaks, CA). Stock vials of human IL-3 were stored at 4#{176}C.For each experiment, all factors were diluted in serum-containing medium on the day of use. Received 8/14/97; revised 10/6/97; accepted 10/13/97. Cell Separation Techniques. Bone marrow cells were The costs of publication of this article were defrayed in part by the obtained from healthy volunteers after informed consent. The payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to mononuclear cells were isolated by centrifugation on Ficoll- indicate this fact. Hypaque gradients (1.077 g/mb; Pharmacia Fine Chemicals, I To whom requests for reprints should be addressed, at 1275 York Piscataway, NJ), washed twice in PBS and suspended in Avenue, New York, NY 10021. Phone: (212) 639-5010; Fax: (212) Iscove’s modified Dulbecco’s medium containing 10% FCS 717-3068. (Hycbone, Logan, UT) supplemented with penicillin (100 units/ 2 The abbreviations used are: APN. aminopeptidase N; mAb, monoclonal antibody; IL, interleukin; CFU-GM, colony forming unit-granu- ml; Life Technologies, Inc., Grand Island, NY), streptomycin bocyte macrophage. (100 p.g!ml; Life Technologies, Inc.), and 3 mg/ml glutamine

(Life Technologies, Inc.). These cells were used as target cell A populations for the CFU-GM progenitor cell assay. U 1. Cl CFU-GM. Low-density (I X l0 cells/mi) or CD34 e V (5 X l0 ce!!s/ml) bone marrow cells were cultured in 35-mm U tissue culture dishes (Coming, Corning, NY) in McCoy’s modified assay medium containing 0.3% agar (DIFCO Laboratories, Detroit, MI) and 10% FCS (17). Cultures were stimulated by the U

I- addition of 100 ng/ml IL-3. U 0. I- Animals. Five-week-old female AKR mice were pur- U V chased from Jackson Laboratory (Bar Harbor, ME). All bedding Cl material was sterilized before use; the cages were covered with = an air filter and maintained in isolation cabinets. Animal han- dbing and experiments were performed in a sterile atmosphere using a laminar flow hood following institutional care guide- Actinonin (.tg/ml) lines. Cell Lines and Culture Conditions. AKR leukemia cells were obtained from the spleen of old AKR mice who spontaneously developed leukemia at 10 months. HL6O (acute myeloid leukemia, CD13 positive), NB4 (acute promyelocytic

leukemia, CD13 positive, from Dr. M. Lanotte of the Louis U I. Pasteur Institute, Paris, France), and RAJI and DAUDI (B Cl VU lineage Burkitt’s lymphomas, CD13 negative) were maintained U in culture using RPMI 1640 supplemented with 10% Serum Plus (JRH Biosciences, Lenexa, KS) and 10% heat-inactivated FCS (Intergen, Purchase, NY) at 37#{176}Cina humidified atmosphere of .0 5% CO2. Cell viability was always higher than 90%, and cells Cl were free of mycoplasma contamination. Transplantation of AKR Leukemia Cells into AKR Mice and Therapy. A 0.l-ml aliquot containing 2 X 106 AKR leukemia cells from suspension culture was transplanted .001 .01 .1 1 10 100 Actinonin (.tg/ml) s.c. into AKR mice. Tumors grew s.c., and the cutaneous tumor size was measured as a cross product to derive surface area. To Fig. 1 Cytotoxicity and inhibition of protein synthesis in cell lines by protect animals, mice were considered dead and sacrificed when actinonin. A, inhibition of protein synthesis in cells by actinonin. Ce!! lines (5 x l0 cells/mb) were incubated for 5 days at 37#{176}Cinthe the tumor surface area reached more than 400 mm2. The test presence of actinonin. Levels of protein synthesis were determined by animals were treated i.p. with actinonin in a final volume of 0.1 5-h incorporation of tritiated beucine into trichloroacetic acid-precipita- ml. Control mice were treated with 0.1 ml of saline. ble protein. B, cell viability determined by trypan blue exclusion. Cell Flow Cytometry Assays. Cells were washed and resus- lines (5 x l0 cells/mi) were incubated for 5 days at 37#{176}Cinthe pended in 2% rabbit serum (Pci Freeze, Rogers, AK) to reduce presence of actinonin. Trypan blue was added, and live and dead cells were enumerated. nonspecific binding. Cells (5 X 10) in a final volume of 0.1 ml were incubated for 1 h on ice in the presence of primary antibody. Cells were washed twice, incubated for 30 ruin on ice with a secondary FITC-labeled antibody (goat antimouse im- munoglobubin; Kirkegaard & Perry), washed twice, and fixed PBS and fixed in 2% paraformaldehyde for 10 mm, and then with 0.5% paraformaldehyde. FITC fluorescence intensity was exposed to 0.1% Triton X-lOO for 10 mm. After washing with measured on an EPICS Profile H flow cytometer (Coulter). PBS and blocking with 1% human AB serum, cells were incu- Inhibition of Tritiated Thymidine or Leucine Incorpo- bated with FITC-conjugated anti-BCL-2 mAb (DAKO A/S. ration. An aliquot containing 200 p.! of cells was washed and Carpinteria, CA) on ice for 30 mm and then analyzed on an incubated at 37#{176}Cin96-well plates in the presence or absence of EPICS Profile II flow cytometer (Coulter). Ten thousand events actinonin. After an incubation time of 3-7 days, 50 p.1 of 10 were counted for each sample. Mean peak fluorescence intensity p.Ci!ml tritiated thymidine or leucine (DuPont New England for an isotype-matched control antibody was set at 1. Nuclear, Wilmington, DE) were added to each well and allowed Cell Cycle and Apoptosis Analysis by Flow Cytometry. to incorporate for 5- 6 h. Trichloroacetic acid was added at a Cells were collected and fixed in 1.5% paraformaldehyde and final concentration of 10% to precipitate protein for [3H]!eucine PBS for 15 mm. After washing with PBS, the cells were resus- incorporation experiments. Cells were harvested using a semi- pended in 70% ice-cold ethanol and kept at -20#{176}Cfor up to 5 automatic harvester (Skatron, Sterling, VA) and read in a scm- days. Analysis for cell cycle distribution and apoptosis was tillation counter (LS 6000 IC; Beckman, Fullerton, CA). performed according to the instructions in the APOPTAG kit Flow Cytometric Analysis of CD11b and BCL-2 Pro- (Oncor, Gaithersburg, MD). Stained cells were analyzed on a teins. Cells were incubated for 1 h on ice with phycoerythrin- FACScan flow cytometer (Becton Dickinson); evidence of ap- conjugated anti-CD! lb mAb (Becton Dickinson), washed with optosis and percentage of cells in each phase of the cell cycle

Table 1 Effect of actinonin on cell viability determined by trypan 0 GI blue exclusion in vitro as D G2M IC() (p.g/ml) (mean ± SD)”

HL6O cells 2 ± 0.5

NB4 cells 5 ± 1.0 V DAUDI cells 2 ± 0.5 Cl RAil cells 4 ± 1.0 0 V AKR cells 3 ± 0.6 V I. “ The !C5() is the concentration of actinonin required to kill 50% of cells.

were analyzed by Ce11FIT and PC-LYSIS software (Becton 24 48

Dickinson). Times (Hours)

Fig. 2 Effects of actinonin on cell cycle. Aliquots of HL6O cells were RESULTS collected at different times after actinonin treatment (10 p.glml). Cell Actinonin Inhibited Growth of Leukemia Cells in Vitro. cycle distribution was analyzed by flow cytometry as described in Actinonin was tested for its ability to kill CD13-positive and “Materials and Methods.” CD13-negative cells. Activity and cytotoxicity were determined by inhibition of incorporation of [3H]leucine into protein and by trypan blue exclusion. Dose-response curves were generated by Table 2 Effect of actinonin (10 p.g/ml) on apoptosis in cells at 96 h testing the inhibitory effects of actinonin on the protein synthe- HL6O NB4 RAJI sis of NB4 and HL6O cells (CD 13 positive) or RAJI cells (CD 13 Control 2.9% 9.1% 4.3% negative) in culture (Fig. 1A). In these in vitro studies, 2-5 Actinonin 37.9% 29.4% 13.6% p.g/ml actinonin were found to be required to inhibit protein synthesis by 50% in CDI3-positive and -negative cells. The cytotoxicity of actinonin was initially determined by trypan blue exclusion (Fig. lB and Table I). The IC() (the Effects of Actinonin on AKR Leukemia Cells in Vivo. concentration of actinonin required to kill 50% of cells) ranged The significant antiproliferative effects of actinonin in vitro and from 2-5 p.g/ml, comparable to the concentration of actinonin the previously published antiprobiferative effects in vivo of required to inhibit 50% of protein synthesis. The similar dose- bestatin, another APN inhibitor (1), prompted an analysis of the response curves for cytotoxicity and protein synthesis inhibition effects of actinonin in vivo against leukemia or lymphoma cells. on cells that expressed or did not express CD13/APN suggested To avoid the problems associated with a xenograft model, we that the mechanism of cytotoxicity did not necessarily involve chose to use a syngeneic leukemia/bymphoma model in AKR inhibition of CD13/APN by actinonin. Treatment of NB4 and mice to better approximate actual human use. AKR cells are HL6O cells for 4 days with 100 p.g/mI mAb F23, which blocks inhibited in vitro by actinonin with an IC0 of about 3 p.g/ml substrate binding to CD13/APN and its activity (18), had no (Table 1 ). In these experiments, AKR mice were injected s.c. effect on cell viability (91% alive) in comparison to no treat- with 2,000,000 AKR leukemia cells (day 0) and treated i.p. with ment (90% alive) or an isotype-matched control antibody, TA99 injections of actinonin beginning at day 3. (91% alive; data not shown). In addition, cytotoxicity of acti- The effect of actinonin on tumor growth and survival rates nonin was not abrogated by pretreatment of cells with mAb F23, of AKR mice after transplantation of AKR leukemia cells was which blocks actinonin binding to its active site (19). These data investigated. After transplantation, mice were treated with a together showed that the inhibition of cell growth by actinonin total of five injections of actinonin (one injection daily for S is not through the inactivation of cell surface enzyme. days). As compared to controls, actinonin increased the mean Cell Cycle G1 Arrest and Apoptosi.s in Leukemia Cells survival time of mice by nearly twofold by reducing the rate of during Actinonin Treatment. Cell cycle distribution of leu- tumor growth. Reduction in tumor growth rates and probonga- kemia cells showed small changes early after treatment with 10 tion of survival were actinonin dose-related (Fig. 3A). Toxicity p.g/ml actinonin. After 24 h of exposure to actinonin, the num- due to actinonin was not observed. Actinonin doses up to 8000 ber of HL6O cells arrested in G1 phase increased, and the p.g/mouse (400 mg/kg) were tolerated without apparent toxicity percentage of cells in S phase decreased (Fig. 2). A similar (16). effect was seen in NB4 (G1 phase increased by 24%) and RAil In a second trial, mice were treated with 100 p.g of acti- (G1 phase increased by 37%) after 24 h of treatment with 10 nonin daily for 3 days beginning at day 3 after transplantation p.g/ml actinonin (data not shown). This G1 arrest is in accord- and then treated by an additional three injections of actinonin ance with the growth inhibition observed after 2 days of expo- (every other day). On day 17, the control mice showed tumors sure to actinonin (data not shown). Ninety-six h after exposure with a mean surface area of 287 ± 95 mm2. In contrast, no to actinonin at 10 p.g/ml, 20-35% of HL6O and NB4 cells tumors were found in mice in the actinonin-treated group (Fig. showed apoptosis, whereas only 10% of RAil cells had apop- 3B). These results indicate that actinonin has significant antitu- tosis after treatment (Table 2). mor effects on AKR leukemia in vivo.

A Table 3 Response of bow-density bone marrow cells to actinonin, S amastatin, and bestatin Data represent the mean ± SD for four plates scored on day 7 and day 14. For these studies, 1 X i05 low-density bone marrow cells were 4 L ---, cultured for 7 and 14 days, respectively, with IL-3 (100 ng/ml). V V E With IL-3: no. of colonies 3 - Control 1 - Actinonn 20 g No Day 7 Day 14 . . Actinonin 50 g Drugs (doses) IL-3 (% of inhibition) (% of inhibition) V 2 Actinonin 100 g Control (saline) 0 177 ± 28 100 ± 4 Actinonin (10 p.g/ml) 0 99 ± I 8 (44) 44 ± 8 (56) 1 Actinonin (5 p.g/ml) 0 123 ± 10 (3! ) 49 ± 7 (51) Actinonin (0.5 p.g/ml) 0 148 ± 6 (16) 74 ± 7 (26) Amastatin (5 p.g/ml) 0 157 ± I (11) 82 ± 11 (18) Bestatin (10 p.g/ml) 0 63 ± 27 (64) 30 ± 7 (70) 0 10 20 30 40 Days

B 14 (data not shown). These data further support the contention 400 -..-fr-- Control that the cytotoxic effects of actinonin are not mediated through .-.o--. Actinonin CD13/APN. E E 300

Cl V DISCUSSION I- Cl Actinonin, a naturally occurring derivative of L-prolino!, is V 200 used as a potent inhibitor of CD13/APN. We report here: (a) the antiproliferative effects of actinonin on human leukemias and

I- 0 lymphoma cells in vitro; (b) the induction of growth arrest and E 100 = apoptosis in target cells; (c) the antitumor effects of actinonin on AKR leukemias in AKR mice; and (d) that these effects do not seem to be mediated through inhibition of CD13/APN. 0 5 10 1 5 20 Actinonin had significant antiproliferative effects on hu- Days man leukemia cells of various derivations. The cytotoxicity of

Fig. 3 Effects of actinonin on AKR leukemia cells in vivo. A, survival actinonin was directly determined by trypan blue analysis and curves of AKR mice after transplantation of AKR leukemia cells. AKR [3H]leucine incorporation. The IC50 was about 2-5 p.g/ml. leukemia cells (2 x 106) were transplanted into AKR mice as described However, the IC50 for other CD13/APN inhibitors, amastatin in “Materials and Methods.” Three days after the transplantation, the and bestatin, was above 100 p.g/ml (data not shown). Actinonin mice were treated with actinonin i.p. daily for 5 days. B. treatment of not only inhibited growth of CD13-positive cells (NB4 or HL6O AKR leukemia cells in vivo by actinonin. AKR mice were transplanted s.c. with 2 X 106 AKR leukemia cells as described in “Materials and cells) but also of CD13-negative cells (RAJI or DAUDI cells; Methods.” After the third day, mice were treated i.p. with 100 p.g of Table 1), suggesting that the effect is not mediated by CD 13/ actinonin/mouse daily for 3 days and then treated with additional injec- APN. Experiments with mAb F23 also suggested that the effect tions of 100 i.g of actinonin (one injection of 100 p.g of actinonin every is not mediated by CD13/APN, because cell viability was not other day for three times). At the times indicated on the X axis, tumor changed by treatment with mAb F23 in comparison to no surface area (2) was measured. treatment or treatment with an isotype-matched control antibody. In addition, the inhibitory effect of actinonin on CD 13- positive cells was not blocked by pretreatment with the anti- In contrast, actinonin did not have any significant growth- CD13/APN mAb F23 (19). Because actinonin binds to F23 inhibiting properties for s.c. implanted RAJI lymphoma in nude epitopes and is blocked by prior incubation with mAb F23 (18) mice (data not shown) over the same dose range as was effective and mAb F23 is not inhibitory of cell growth, we conclude here in the AKR model. that actinonin-induced cell death in human leukemia cells is not Effects of Actinonin on Norma! Human Bone Marrow likely to be associated with binding and inhibition of cell surface Cell Growth ex Vivo. The effects of actinonin, amastatin, and CD13/APN. bestatin were evaluated on human bone marrow CFU-GM (Ta- The effect of actinonin in vitro was mediated at least partly ble 3). Actinonin decreased colony formation 16-56% in a through G3 arrest and apoptosis. After 10 p.g/ml actinonin dose-dependent manner. Bestatin (10 p.g/ml), at doses designed treatment for 96 h, 20-35% of NB4 and HL6O cells showed to approximate its dose level in clinical trials (20), also de- evidence of apoptosis. The 1,10-phenanthroline, which inhibits creased normal human bone marrow CFU-GM by 64-70%. APN activity by chelating the zinc ion (3, 21), also induces This suggested that actinonin may have some myelosuppressive apoptosis in HL6O cells (22). Actinonin binds to zinc domains of activities at high doses. F23 (antiactive site of CD! 3) and M 195 CD13/APN, as determined by competition assays (18). This (anti-CD33), both of which bind to bone marrow progenitors, may suggest that the zinc-binding motif of important intraceb- had no effect on human bone marrow CFU-GM on days 7 and lular enzymes may account for apoptosis, but cell surface CD 13/

APN does not seem to be associated with actinonin action. In REFERENCES addition, there is no increase in the percentage of CD! lb in 1. Shipp, M. A., and Look, A. T. Hematopoictic differentiation antigens NB4, HL6O, or RAJI cells (data not shown) after actinonin that are membrane-associated enzymes: cutting is the key! Blood, 82: 1052-1070, 1993. treatment, suggesting that there is no differentiating activity induced by actinonin. 2. Delmas, B., Gelfi, J., L’Haridon, R., Vogel, L. K., Sjostrom, H.. Noren, 0., and Laude, H. Aminopeptidase N is a major receptor for the Actinonin was also cytotoxic to RAJI cells, without an enteropathogcnic coronavirus TGEV. Nature (Lond.), 357: 4 17-420, increase in apoptosis; in preliminary experiments designed to 1992. screen for other mechanisms or pathways involved in the anti- 3. Ycager, C. L., Ashmun, R. A., Williams, R. K., Cardcllichio, C. B., proliferative effects in these cells, BCL-2 expression was stud- Shapiro, L. H., Look, A. T., and Holmes, K. V. Human aminopeptidase ied. BCL-2 expression, observed by intracellular flow cytometry N is a receptor for human coronavirus 229E. Nature (Lond.), 357: 420-422. 1992. in about 20% of HL6O cells and 60% of NB4 cells, showed no 4. Mcnrad, A., Speicher, D., Wacker, J., and Hcrlyn, M. Biochemical significant changes after treatment with actinonin. BCL-2 cx- and functional characterization of aminopeptidase N expressed by hu- pression, however, was decreased by 80% in RAJI cells after man melanoma cells. Cancer Res., 53: 1450-1455, 1993. treatment with actinonin. Other cell surface proteins (CD1 lb. 5. Saiki, I., Fujii, H., Yoneda, I., Abe, F., Nakajima, M., Tsuruo, T., and 13, 15, 29, 33, and HLA-A) significantly decreased in NB4 cells Azuma, I. Role of aminopeptidase N (CD1 3) in tumor-cell invasion and extracellular matrix degradation. Int. J. Cancer, 54: 137-143, 1993. after treatment with actinonin, consistent with the toxic effects 6. Olsen, J., Cowell, G. M., Konigshofcr, E., Danielsen, E. M., Moller, of actinonin. J., Laustsen, L., Hansen, 0. C., Welinder, K. G., Engberg, J., Hunziker, Bestatin, another APN inhibitor, has shown antitumor ther- W., Spiess, M., Sjostrom, H., and Noren, 0. Complete amino acid apeutic effects in several clinical trials (23). In a multi-institu- sequence of human intestinal aminopeptidase N as deduced from cloned tional study, 101 patients with acute nonlymphocytic leukemia cDNA. FEBS Lett., 238: 307-314, 1988. were randomized to receive bestatin or control. The bestatin 7. Look, A. T., Ashmun, R. A., Shapiro, L. H., and Peiper, S. C. Human mycloid plasma membrane glycoprotcin CD13 (gpl5O) is identical to group achieved a statistically significant prolongation of both aminopeptidase N. J. Clin. Invest., 83: 1299-1307, 1989. the remission duration and survival in patients aged 50-65 8. Griffin, J. D., Ritz, J., Nadler, L. M., and Schbossman, S. F. Expres- years (24). It also normalized the CD4:CD8 ratio in peripheral sion of mycboid differentiation antigens on normal and malignant my- blood and maintained the immune homeostasis in cancer pa- eloid cells. J. Clin. Invest., 68: 932-941, 1981. tients (25, 26). This may be beneficial to AIDS patients with 9. Xu, Y., Wcblncr, D., and Schcinberg, D. A. Substance P and brady- kinin are natural inhibitors of CD13/aminopeptidase N. Biochem. Bio- lymphoma. Recently, bestatin also showed direct inhibition of phys. Res. Commun., 208: 664-674, 1995. the growth of human chonocarcinoma in nude mice in a dose- 10. Tsuruo, T., Naganuma, K., lida, H., Yamori, T., Tsukagoshi, S., and dependent manner (27). A recent randomized study in patients Sakurai, Y. Inhibition of lymph node metastasis of P388 leukemia by with non-Hodgkin’s bymphoma after autologous bone marrow bestatin in mice. J. Antibiot. (Tokyo), 34: 1206-1209, 1981. transplant who were treated with bestatin showed increases in I 1. Ikeda, S., Miyasato, H., Saito, K., Nakayama, H., and Tajima, K. Phase I study of bestatin: (II) clinical trial of bestatin in malignant skin natural killer activity in pokeweed mitogen and phytohemag- tumors. In: H. Umezawa (ed), Small Molecular Immunomodifiers of glutinin responses of lymphocytes in CD4 T-cel! and B-cell Microbial Origin, pp. 143-158, Elmsford, NY: Pergamon Press, 1981. numbers (26). Therefore, we investigated whether a more potent 12. Inuyama, Y., Noriuchi, M., Kohno, N., and Mashino, S. Clinical APN inhibitor, actinonin, could inhibit the growth of syngeneic experience with bestatin in head and neck cancer. In: H. Umezawa (ed). AKR leukemia cells in a mouse model in vivo. Actinonin Small Molecular Immunomodifiers of Microbial Origin, pp. 179-186. Elmsford, NY: Pergamon Press, 1981. showed dose-dependent antitumor effects on AKR leukemia in 13. Isono, K., Onoda, S., Ishikawa, T., Koide, Y., and Sato, H. Clinical vivo. There was a significant effect at doses of 100 p.g/mouse (5 and experimental study of bestatin for esophageal cancer. In: H. mg/kg). Prolonged treatment seemed to further improve the Umezawa (ed), Small Molecular Immunomodifiers of Microbial Ori- activity against the leukemias. Because actinonin shows no gin, pp. 187-200. Elmsford, NY: Pergamon Press. 1981. apparent toxicity in doses up to 400 mg/kg in mice (about 8 14. Akiya. K., and Okabe, K. Clinical effect of bestatin on malignant mg/mouse; Ref. 16) and no apparent toxicity was seen in the tumors offemabe genital organs. ln: H. Umezawa (ed), Small Molecular Immunomodifiers of Microbial Origin, pp. 201-210. Elmsford, NY: experiments conducted here, the drug may be considered safe in Pergamon Press, 1981. this mouse model at these doses. In contrast, little antitumor 15. Abe, F., Schneider, M., Black, P. L., and Talmadge, J. E. Chemo- activity was seen in a nude mouse model. One explanation for immunotherapy with cyclophosphamide and bestatin in experimental this may be related to the hypothesis that these inhibitors work metastasis in mice. Cancer Immunol. Immunother., 29: 231-236, 1989. via a nonspecific immune augmentation in vivo (26, 28) that 16. Gordon, J. J., Kelly, B. K.. and Miller, G. A. Actinonin an antibiotic substance produced by an actinomycete. Nature (Lond.), 195: 70 1-702. may not be possible in nude mice. Alternatively, there may 1962. simply be differences in the biology of the cells that account for 17. Gabribove, J. L., White, K., Rahman, Z., and Wilson, E. L. Stem cell the lack of effects with RAJI cells. factor and basic fibroblast growth factor arc synergistic in augmenting In conclusion, the data showed that actinonin induces committed mycloid progenitor cell growth. Blood, 83: 907-910, 1994. arrest and apoptosis in human leukemia and lymphoma cells; 18. Xu, Y., Wellner, D., and Schcinberg, D. A. Cryptic and regulatory moreover, actinonin can treat AKR leukemia in AKR mice with epitopes in CD13/aminopeptidasc N. Exp. Hematol., 25: 521-529. 1997. minimal toxicity. The site of action does not seem to be via 19. Xu, Y., Lai, L. T., and Scheinberg, D. A. Actinonin-induced growth inhibition of CD13/APN. Details of the mechanisms of acti- arrest and apoptosis in human leukemia cells is associated with p53 nonin antileukemia and anti-lymphoma activity will require modulation but not with aminopeptidase inhibition. Proc. Am. Assoc. further study. Cancer Res., 38: 1368, 1997.

20. Ota, K., and Uzuka, Y. Clinical trials of bestatin for leukemia and 25. Takeuchi, 1. Antitumor antibiotics discovered and studied at the solid tumors. Biotherapy, 4: 205-214. 1992. Institute of Microbial Chemistry. J. Cancer Res. Clin. Oncol., 121: 21. Vallce, B. L., and Auld, D. S. Zinc coordination, function, and 505-510, 1995. structure of zinc enzymes and other proteins. Biochemistry. 29: 5647- 26. Talmadge. J. E.. Ino, K., Walker, S., Heimann, D., and Bierman, 5659, 1990. P. Immune augmentation by an orally active peptide following trans- 22. Byrnes, R. W. Inhibition of hydroperoxide-induced DNA single- plantation for lymphoma. Proc. Am. Assoc. Cancer Res., 37: 3340, strand breakage by 1,10-phenanthroline in HL-60 cells implications for 1996. iron speciation. Arch. Biochem. Biophys., 332: 70-78, 1996. 27. Inoi, K., Goto, S., Nomura, S., Isobe, K., Nawa, A., Okamoto, 1., 23. Urabe, A., Mutoh. Y., Mizoguchi, H., Takaku, F., and Ogawa, N. and Tomoda, Y. Aminopeptidase inhibitor ubenimex (bestatin) inhibits Ubenimex in the treatment of acute nonlymphocytic leukemia in adults. the growth of human choriocarcinoma in nude mice through its direct Ann. Hematol., 67: 63-66, 1993. cytostatic activity. Anticancer Res.. 15: 2081-2087, 1995. 24. Ota. K., Kurita, S., Yamada, K.. Masaoka, 1., Uzuka, Y., and 28. Talmadge, J. E., Lenz, B. F., Pennington, R., Long. C., Phillips, H., Ogawa, N. Immunotherapy with bestatin for acute nonlymphocytic Schneider, M., and Tribble, H. Immunomodulatory and therapeutic leukemia in adults. Cancer Immunol. Immunother., 23: 5-10, 1986. properties of bestatin in mice. Cancer Res., 46: 4505-4510, 1986.

Y Xu, L T Lai, J L Gabrilove, et al.

Clin Cancer Res 1998;4:171-176.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/4/1/171

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/4/1/171. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.