Tsuyoshi Tanabe ^{1, 2}, Tomoko Kuwabara ^{2, 3}, Masaki Warashina ^{2, 3}, Kenzaburo Tani ¹, Kazunari Taira ^{2, 4} & Shigetaka Asano ¹

E-Mail: taira@chembio.t.u-tokyo.ac.jp

Chronic myelogenous leukaemia (CML) is a haematopoietic malignant disease associated with the expression of a chimaeric BCR-ABL gene [1,2]. We have designed an allosterically controllable ribozyme that specifically cleaves BCR-ABL messenger RNA and induces apoptosis in cultured CML cells [3], and here we test it as a possible treatment of CML in a mouse model. We find that this ribozyme completely inhibits tumour-cell infiltration in these mice. To our knowledge, this is the first application of an artificial, allosterically controllable enzyme in animals, opening up the possibility of using ribozyme technology in the treatment of CML.

The chimaeric BCR-ABL gene generates two different BCR-ABL mRNAs, known as K28 b3a2 and L6 b2a2, both of which yield p210^BCR-ABL on translation, an oncoprotein [4,5] with tyrosine kinase activity [2] that causes malignant transformation of white blood cells in CML.

Although ribozyme-based gene therapy is arousing great interest, its application to disrupt chimaeric mRNA has been restricted by the lack of a specific cleavable site at the junction of the chimaeric gene. To disrupt BCR-ABL mRNA, a ribozyme must exclusively target the junction sequence, otherwise normal ABL mRNA that shares part of the chimaeric sequence will also be cleaved by the ribozyme, with resultant damage to host cells [3].

We have previously generated a novel ribozyme ('maxizyme') that is dimeric [3, 6-8] and specifically cleaves BCR-ABL mRNA, inducing apoptosis in CML cells [3]. The maxizyme has sensor arms that recognize two target sequences that induce it to form a cavity for capturing Mg²⁺ ions essential for catalysis. The activity of the heterodimeric ribozyme can be allosterically controlled through one of the two substrate-binding regions [3]. The maxizyme remains in an inactive conformation in the presence of normal ABL mRNA or in the absence of the junction sequence [3].

To test this technology in vivo, we embedded genes encoding the maxizyme downstream of genes for the human transfer RNA tRNA^Val (refs 3,7,8). We used a retroviral system for the expression of the maxizyme in leukemic cells [9], which involved subcloning two tRNA^Val-driven expression cassettes, corresponding to each component of the heterodimer, in tandem into the retroviral vector. A line of CML cells (BV 173) was transduced either with a control vector in which the maxizyme sequence had been deleted, or with the maxizyme-encoding vector. We then injected 2 x 10⁶ transduced BV 173 cells into the tail veins of individual NOD-SCID mice. (These mice, produced by crossing SCID and non-obesity diabetic mice, are a powerful tool for growing human haemopoietic cells in vivo). Survival of the animals was monitored daily for 100 days after inoculation. Eight weeks after inoculation, eight animals from each group were killed by cervical disloaction and their tissues examined.

The pathological changes found at necroscopy were remarkably consistent in mice that had received control-transduced BV 173 cells (Fig. 1a).

Figure 1. The antitumour effects of the maxizyme in a murine model of chronic myelogenous leukaemia (CML).
a, Infiltration by BV 173 leukaemic cells, as revealed by haematoxylin-and-eosin staining of livers, spleens, kidneys and brains of NOD-SCID mice injected with BV173 cells that had been transduced either with a control vector or with the maxizyme-expressing vector. In most cells stained with haematoxylin and eosin, the nucleus is blue and the cytoplasm is pink, but the extent of staining is different for each tissue. Because infiltrated leukaemic cells almost replace normal cells in control spleen and disrupt its structure, the degree of staining is different from that of maxizyme-containing spleen.
b. Reduction of tumorigenicity of BV 173 cells in vivo. A total of 2 X 10⁶ BV 173 cells transduced with either control vector or maxizyme were selected by incubation with puromycin and injected into the tail vein of NOD-SCID mice (two independent experiments, with 4-8 animals per group in each experiment.). The survival of the animals was monitored daily for more than 20 weeks after inoculation; all control mice died within 13 weeks, whereas maxizyme-treated mice remained disease-free for the entire period of the investigation. Scale bars, 50um.

The differences between the two groups of mice were reflected in their mortality rates (Fig. 1b). All of the mice injected with control BV 173 cells died of diffuse leukaemia, confirmed at necroscopy, 6-13 weeks afterwards (median survival time, 9 weeks), whereas mice injected with maxizyme-transduced BV 173 cells remained healthy (Fig. 1b).

Our results indicate that each subunit of the maxizyme introduced by the retroviral vector is produced at the appropriate concentration to support dimerization in vivo. Also, the maxizyme apparently functions successfully in animals, cleaving BCR-ABL mRNA with exceptional efficiency.

At present, allogeneic transplantation is the only effective therapy for this type of leukaemia [10, 11], with only half of all patients on average being eligible for this treatment because of limited donor availability and age restrictions. Our results raise the possibility that this maxizyme could be useful for purging bone marrow in cases of CML treated by autologous transplantation, when it would presumably reduce the incidence of relapse by decreasing the tumorigenicity of contaminating CML cells in the transplant [12].

We have also constructed five other maxizymes that successfully target other chimaeric genes [13], suggesting that maxizymes could be a new class of powerful gene-inactivating agents that can cleave any type of chimaeric mRNA.

1. Bartram CR, et al, Nature 306: 277-280 (1983).

2. Konopka JB, Watanabe SM, and Witte ON, Cell 37: 1035-1042 (1984).

3. Kuwabara T, et al, Mol. Cell 2: 617-627 (1998).

4. Daly GQ, Van Etten RA, and Baltimore D, Science 247: 824-830 (1990).

5. Gishizky ML, and Witte ON, Science 256: 836-839 (1992).

6. Amontov S, and Taira K, J. Am. Chem. Soc. 118: 1624-1628 (1996).

7. Kuwabara T, et al, Nature Biotechnol. 16: 961-965 (1998).

8. Kuwabara T, Warashina M, Nakayama A, Ohkawa J, and Taira K, Proc. Natl. Acad. Sci. U.S.A., 96: 1886-1891 (1999).

9. Onishi M, Exp. Hematol. 24: 324-329 (1996).

10. Snyder DS, and McGlave PB, Hematol. Oncol. Clin. North Am. 4: 535-557 (1990).

11. McGlave PB, et al, Blood 81: 543-550 (1993).

12. Deisseroth AB, et al, Blood 83: 3068 (1994).

13. Tanabe T, et al, Biomacromolecules 1: 108-117 (2000).

Additional References:

1. "Effect of RNA from Normal Human Marrow on Leukaemic Marrow In-Vivo".

2. "Mated Models of Gene Regulation in Eukaryotes".

3. "Oncogenes as Molecular Targets within Active Chromatin".