CHIMERIC PROTEIN AND CANCER
Chromosome translocations are particularly common in cancers arising in bone marrow cells – leukaemias and lymphomas. Specific translocations are associated with particular types of leukaemia and the introduction of chimeric genes into mice or zebrafish has shown that they are sufficient to trigger the development of the disease, even though additional mutations subsequently accumulate. The best known chromosomal translocation was identified in 1960 by Nowell and Hungerford in a case of myeloid leukaemia (CML). Janet Rowley subsequently showed that it involved a fragment from the tip of chromosome 9 displacing a large part of chromosome 22, producing a shortened, and hybrid chromosome – the Philadelphia chromosome (Fig.1).
1. The Philadelphia chromosome.
The effect of this fusion is that the chromosome 22 gene BCR is placed directly in front of the ABL1 gene that is carried on the fragment of chromosome 9. Because the fusion is ‘in frame’ the transcription and translation machinery now produce a chimeric protein (BCR-ABL1) that is the driving force for leukaemia development. The main reason for this is that ABL1 encodes a tyrosine kinase enzyme whose activity is increased in the chimeric form. Tyrosine kinases can act as powerful proliferation signals and ABL1 appears to be particularly potent in this respect when expressed in bone marrow cells. It is notable that, in a similar manner to the SRC and RAS oncogenes, there is a mouse virus that has picked up a mutated version of the ABL1 gene. This virus causes leukaemia in infected animals and the tyrosine kinase activity of the viral ABL1 protein is similar to that of chimeric BCR-ABL1 in humans. The Philadelphia chromosome is present in over 90% of CML cases and also occurs in 10% to 20% of patients with acute lymphoblastic leukaemias (ALL).
Acute promyelocytic leukaemia (APL), a subtype of acute myelogenous leukaemia (AML), is characterised by the capacity of all-trans retinoic acid (ATRA) to induce remission. ATRA is the acid form of vitamin A and is used to treat acne (it’s also called tretinoin). AML stems from chromosomal translocations involving the retinoic acid receptor alpha (RARA) gene and, most commonly, the promyelocytic leukaemia (PML) gene (Fig. 2).
- Structures of (the retinoic acid receptor family) RARA, RARB and RARG, and PML and PML-RARA.
The PML protein is a transcription factor that functions as a tumour suppressor, regulating cell proliferation, survival and senescence. In part, its tumour suppressor activity derives from interaction (of isoform IV: there are seven known isoforms) with the tumour suppressors p53 and RB1. The fusion protein PML-RARA essentially suppresses this suppressor function, so that PML-RARA acts as the transforming protein of APL. PML-RARA proteins form DNA binding complexes with members of the RXR family of retinoic acid receptors. Thus PML-RARA is a hormone-dependent transcription factor that acts as a dominant negative oncoprotein to inhibit myeloid differentiation and promote proliferation. ATRA is a RAR agonist that blocks the dominant negative effect of PML-RARA, permitting undifferentiated leukaemic blast cells to differentiate into neutrophils. The mechanism of ATRA action is a (rare) example of a therapeutic strategy in which tumour cells are induced to differentiate and hence cease to proliferate: they have exited the cell cycle and entered the G0 quiescent phase or ‘post-mitotic’ state. In combination with other drugs, ATRA gives remission rates of over 90%.
In the 1970s arsenic trioxide (As2O3) was also shown to give high remission rates for APL and both retinoic acid and arsenic are used for the clinical treatment of this disease (Fig. 3).
- Acute promyelocytic leukaemia (APL): induction of differentiation in bone marrow blast cells.
Although they interact with different regions of PML-RARA, they both cause the protein to undergo attachment of ubiquitin that initiates its degradation by the proteasome.
When leukaemias either fail to respond to chemotherapy or patients relapse after a period of remission, the cause appears to lie in a very small population of ‘leukaemia initiating cells’ (LIC). Leukaemia initiating cells resemble normal hematopoietic stem cells in that they are generally quiescent. However, they are pluripotent and retain the capacity to replicate and provide perhaps the strongest evidence for cancer ‘stem cell hypothesis’. Because they are generally not proliferating they tend to be resistant to elimination by drugs. Thus, for example, although imatinib is effective against chronic myeloid leukaemia (CML), LICs are likely to survive chemotherapy to cause a relapse. PML is expressed at high LICs from a mouse model of CML and also in blast cells from CML patients. It transpires that PML plays a critical role in maintaining LICs in a quiescent state. When PML is inhibited by As2O3, LICs re-initiate proliferation and thus become susceptible to anti-leukemic therapy.
The leukaemias have been particularly revealing in terms of cancer development because individual cells carrying ‘driver’ translocations, e.g. BCR-ABL1, can be identified. Fluorescence in situ hybridisation (FISH) uses labelled probes to detect specific translocations, permitting identification of the leukaemic cells in a blood sample. DNA from these then be sequenced to determine the complete mutational pattern in individual tumour cells. Perhaps the most informative evidence relating to stem cells and cancer, and indeed to the general way in which tumours develop, has come from childhood ALL.