A new face of BCL-2 inhibition in CLL

In this issue of Blood, Zhong and colleagues report on a peptide that is selectively toxic to chronic lymphocytic leukemia (CLL) cells through perturbation of cytoplasmic calcium levels via disruption of the BCL-2:IP3R interaction.¹ 

Cytoplasmic calcium levels are largely controlled by uptake and release from the endoplasmic reticulum (ER). Calcium levels can fluctuate dramatically over time in the form of discrete oscillations of cytoplasmic Ca²⁺ concentrations. These oscillations govern many important cellular processes, including proliferation and programmed cell death. An important protein controlling the pulsatile release of calcium is the inositol 1,4,5-triphosphate receptor (IP3R).² 

It has been shown that the BCL-2 family of proteins, best known for controlling mitochondrial permeabilization during apoptosis, also controls cytoplasmic calcium levels.³  BCL-2 itself is present at the ER, which usually exists in very close conjunction with the mitochondria. BCL-2 can act to dampen high amplitude Ca²⁺ oscillations that can induce apoptosis. The observation of a complex between IP3R and BCL-2 affords the critical link between BCL-2 and physiologic control of calcium levels.^4,5 

In no malignancy is BCL-2 more consistently expressed at high levels than in CLL. In fact, CLL has been shown to be a consistently BCL-2–dependent disease.⁶  Most of the study of BCL-2 function in CLL has focused on the classic BCL-2 function of binding proapoptotic BCL-2 family members by their BH3 domain, sequestering them, and preventing execution of their pro-death function. In these circumstances, where “primed” BCL-2 is antagonized by a BH3 mimetic small molecule, such as ABT-737 (or ABT-263, the clinical derivative), the proapoptotic molecules are released to commit the CLL cell to death. This type of death relies on permeabilization of the mitochondrial outer membrane by BAX or BAK, pro-death BCL-2 family proteins. CLL remains an active area of clinical investigation of BH3 mimetic small molecules.

Zhong et al focus on a different facet of BCL-2, the BH4 domain that is involved in the interaction with IP3R. Using an oligopeptide derived from a site on IP3R found to be involved in binding BCL-2, the authors had previously demonstrated the ability to disrupt the BCL-2:IP3R complex and alter calcium signaling.⁷  This current report is noteworthy in two ways: first, it reports a modification of the peptide that increased cytoplasmic calcium concentrations; and second, it finds that CLL cells are selectively susceptible to death induced by the calcium signaling induced by the peptide. The original peptide was linked to the HIV TAT peptide to foster intracellular penetration via pinocytosis. However, after cell entry, the peptide was subject to proteolytic degradation. Changing 2 aspartic acid residues to alanines removed key protease sites, allowing for greater intracellular accumulation. This is reflected in greater increases in high amplitude calcium oscillations and greater cell death.

When the authors apply this peptide to primary CLL cells in vitro, they find that calcium oscillations are augmented, and that CLL cells die what appears to be an apoptotic death, because caspases are activated and their nuclei have a characteristic morphology. Furthermore, normal lymphocytes are relatively less affected, suggesting that CLL cells are more dependent on the BCL-2:IP3R interaction than are normal lymphocytes, and providing the possibility of a useful therapeutic window.

However, even though caspases are activated, it is not certain that this represents a traditional BAX- and BAK-dependent apoptotic death. This is important; because it is quite possible that much of the resistance to therapy in CLL is based on increased resistance to apoptosis via the BCL-2 family governed pathway. If this peptide mediates an “end-run” around this control of mitochondrial apoptosis, it might work even in cases of previously poor therapeutic response.

In the case of the BH3 mimetic ABT-737, it is well established that the antiapoptotic proteins MCL-1 and BFL-1, which are relatively weakly engaged by ABT-737 and ABT-263, can act as resistance factors to ABT-737 treatment.^8,9  While there is evidence that MCL-1 can play a similar role as BCL-2 in controlling calcium flux via IP3R, it seems likely that the binding site on IP3R is the same as for BCL-2, and hence the peptide strategy might simultaneously inhibit multiple interactions between antiapoptotic proteins and IP3R.¹⁰  Thus, it is possible there will be instances in which a cell that has acquired resistance to a BH3-mimetic targeting BCL-2 can nonetheless be treated by an agent that targets the BCL-2:IP3R complex.

It is clearly too early to speculate on precise clinical applications of such a strategy. Nonetheless, Zhong et al have provided an interesting avenue of investigation for inhibiting a critical survival pathway in many malignancies, one that does not overlap with previous strategies to target BCL-2.