Influence of Cyclophosphamide on L-Asparaginase-Induced Allergy in Animal Model

BACKGROUND: L-Asparaginase (L-ASP) is one of the essential drugs for acute lymphoblastic leukemia and lymphoma. Since L-ASP proteins used clinically are derived from bacteria, they often cause allergies, including anaphylaxis. However, the characteristics of L-ASP-induced hypersensitivity are not well documented so far. Also, there is neither suitable treatment for hypersensitivity to L-ASP nor adequate experimental model of L-ASP allergy. In addition, some anti-cancer drugs, used concomitantly with L-ASP, are known to affect lymphocyte activities, although the role of such drugs on L-ASP hypersensitivity is not known. Previously, we have established mouse model of L-ASP allergy and found that a neutralizing antibody (Ab) of IgE is effective to inhibit L-ASP-induced allergic reaction.

Aims: In order to establish in vitro model of L-ASP allergy, we used RBL-2H3 mast cells and evoked degranulation of the cells sensitized with L-ASP-allergic mouse serum using L-ASP as an antigen. Furthermore, we examined the effect of cyclophosphamide (CY), which reportedly impairs the functions of regulatory T (Treg) cells, on L-ASP allergy both in vivo and in vitro.

Methods: Male BALB/c mice (7 week old) were sensitized with L-ASP (100 units) with Al(OH)₃ gel (1 mg) on days 1 and 15. The right ears of the mice were locally sensitized on days 18, 21, 24 by i.d. injection of L-ASP (10 units). Antigen challenge was carried out on day 27 by i.d. injection (10 units). CY (75, 150, 300 mg/kg) was i.p. administrated at day -1 and day 13. Anti-IgE treatment (BD Bioscience, clone R35-92; 100 μg, i.p.) was carried out on the day before L-ASP challenge. The serum was collected on day 27. Total IgE level in the serum was measured by ELISA kit (Yamasa). RBL-2H3 cells were sensitized by the serum and stimulated by L-ASP to determine β-hexosaminidase (β-Hex) release in vitro.

Results: L-ASP-sensitization induced allergic ear edema and an increase in serum IgE level in mice, both of which are augmented by an administration of CY at 150 mg/kg. L-ASP-induced allergic ear edema was inhibited by a pretreatment with anti-IgE Ab. L-ASP-induced CY-enhanced ear edema was also inhibited by anti-IgE Ab. On the other hand, at 300 mg/kg of CY, L-ASP-induced ear edema was much lower than 150 mg/kg group, though IgE concentration in the serum was as high as CY 150 mg/kg group. When RBL-2H3 cells were sensitized by anti-L-ASP serum, L-ASP challenge induced β-Hex release. Anti-L-ASP serum of CY 150 mg/kg-treated mice induced higher β-Hex release than normal anti-L-ASP sera in vitro, though that of CY 300 mg/kg-treated mice did not induce β-Hex release. After removing IgG from the serum of CY 300 mg/kg-treated mice using protein G Mag Sepharose (GE Healthcare), β-Hex release became higher than normally sensitized mice. Anti-IgE Ab treatment both ex vivo and in vitro inhibited L-ASP-induced β-Hex release from the RBL-2H3 cells sensitized by anti-L-ASP serum of CY-treated mice.

Conclusions: In this study, we showed that L-ASP sensitization induced IgE production in vivo and this serum could bring about β-Hex release from RBL-2H3 cells in vitro; both allergic reactions were inhibited by an anti-IgE antibody which binds to Fc region of IgE molecule. From these results, we propose that non-anaphylactogenic neutralizing antibody of IgE, such as omalizumab, can be a candidate drug for the treatment of L-ASP hypersensitivity. Also in this study, CY induced biphasic effects on L-ASP allergy in mice, enhancing it at low dosage and attenuating it at high dosage. Since CY is reportedly impair the functions of Treg cells, CY-induced suppression of Treg may enhance Th2 responses so as to augment L-ASP hypersensitivity. Further investigations for the detection and avoidance of L-ASP hypersensitivity are under the way.