Induction of micronucleated erythrocytes by a recombinant human erythropoietin (rhEPO) was examined using both in vitro and in vivo test systems. A small, significant and dose-related increase in the frequency of micronucleated polychromatic erythrocytes (MNPCE) in bone marrow of mice administered i.p. with 12,500-50,000 IU/kg rhEPO was induced at 48 h sampling time. A clear positive dose--response relationship and significant increase in the frequency of micronucleated reticulocytes (MNRET) in peripheral blood of mice administered i.p. with 400-50,000 IU/kg rhEPO was noted at 48, 72 and 96 h sampling times. Conversely, in bacterial reverse mutation tests, no noticeable increase of auxotrophic revertants was observed in Salmonella typhimurium, TA100, TA98, TA1535, TA1537, or Escherichia coli, WP2 uvrA-, by treatment with 188-6000 IU/plate of rhEPO, with or without S9 mix. Furthermore, rhEPO at 750-6000 IU/ml did not induce chromosomal aberrations in vitro in CHL cells or human peripheral blood lymphocytes in a direct method nor in a metabolic activation method. Moreover, chromosomal aberrations were not detected in bone marrow cells of CD-1 male mice, even at high rhEPO concentrations (100,000 IU/kg) in vivo. Consequently, it was concluded that errors in the process of enucleation or differentiation of the erythrocytes should be equally considered as possible mechanisms for the increased frequencies of MNPCE and MNRET alongside induction of DNA damage or errors in the process of DNA repair.

Comparative studies on the induction of micronucleated erythrocytes by three recombinant human erythropoietins (rhEPOs) such as epoetin alpha, epoetin beta, and SNB-5001, and by urinary human EPO (uhEPO) were carried out in vivo. Small but significant increases in the frequency of micronucleated polychromatic erythrocytes (MNPCEs) in bone marrow of mice following a single intraperitoneal (i.p.) administration of each of the three rhEPOs and uhEPO (25,000 IU/kg) were observed at the 48 h sampling time. Clear, significant increases in the frequency of micronucleated reticulocytes (MNRETs) in the peripheral blood of mice similarly treated with the four hEPOs at 50,000 IU/kg were noted at the 48, 72 and 96 h sampling times. When mice were dosed with 50,000 IU/kg of each of the hEPOs once each day for 6 days and micronucleus preparations were made 24 h after the last administration, micronucleus induction in the bone marrow was not clearly demonstrated. As before, MNRETs in these mice were significantly induced in all cases. Consequently, it was concluded that induction of MNPCEs and MNRETs was a common characteristic of both native and genetically recombinant EPO.

The nude (athymic) mouse was used as a novel test system for the evaluation of genotoxicity of genetic recombinant human erythropoietin (rhEPO). A fibroblast cell line derived from the kidneys of baby hamsters, BHK-21, or a subclone of BHK-21 transfected with an expression vector containing the human EPO gene, named BXE cells, were implanted in nude mice. The concentration of EPO in the plasma of mice bearing BXE increased in relation to the increase in the weight of the tumor formed from growth of BXE cells. Increased values of hematocrit (Ht), ratio of reticulocytes to erythrocytes (RET ratio) and the number of red blood cells in mice bearing BXE indicated that excessive hematopoiesis was occurring in the host. However, the concentrations of EPO in the plasma of the mice bearing BHK-21 did not increase in relation to the cell mass and consequently the Ht values and RET ratios in these mice were not affected. Marked increases in the frequencies of micronucleated polychromatic erythrocytes (MNPCE) and micronucleated RET (MNRET) were noted in the mice bearing BXE, although no chromosomal aberrations were found in the spleen and marrow cells of the same mice. The increased levels of RET, MNRET and MNPCE seemed to result from acceleration of erythroblastic maturation and proliferation by rhEPO. It is, therefore, concluded that errors in the processes of enucleation or differentiation of erythrocytes should be equally considered as possible mechanisms alongside errors in genetic repair processes for the increased frequencies of MNPCE and MNRET.

A protein-bound polysaccharide(SN-C)obtained from the culture filtrate of Cordyceps ophioglossoides,which has been reported to have antitumor activity,was analyzed for sugar and amino acid components.Gas chromatography revealed that the neutral sugar component of SN-C was mainly composed of glucose.By using an amino acid auto analyzer,it was shown that SN-C was composed of galactosamine as the sole aminosugai.An intraperitoneal administration of SN-C into mice inoculated intraperitoneally with syngeneic murine tumors such as MM46 mammary carcinoma or L5178Y leukemia suppressed the tumor growth and resulted in a significant prolongation of the life span.When used in vitro,SN-C significantly inhibited the proliferation of various murine tumor cells including P388.Futhermore,SN-C inhibited the incorporation of glucose into Meth-A tumor cells with a consequent decrease of deoxyribonucleic acid(DNA) synthesis.SN-C may be a new type of antitumor polysaccharide since this material seems to posess both direct antitumor effect and stimulating activity on the host-mediated effect.

No abstract