In this issue of Blood, Ouyang et al describe both an important tumor immune evasion strategy and a means by which it can be overcome.1 They show that EBV proteins LMP1 and 2A induce expression of galectin-1 (Gal1) by the B lymphoblasts of EBV+ posttransplantation lymphoproliferative disease (PTLD), and that they could block the apoptosis this sugar-binding molecule would otherwise induce in effector cytotoxic T lymphocytes using a Gal1 directed monoclonal antibody.
Both tumor cells and virus-infected cells have devised multiple strategies to evade the immune system. These include the failure to present tumor or viral antigens appropriately to the immune system, secretion of factors, such as TGFβ, that diminish T-cell survival and function, or the secretion of chemokines that attract regulatory or inhibitory T-cell subsets rather than antitumor effectors.2 An additional inhibitory mechanism mediated by a family of carbohydrate-binding proteins known as galectins is attracting increasing interest for their immunosuppressive activities in the tumor microenvironment.3 Gal1 is an endogenous glycan-binding protein that is expressed by a number of malignancies and at sites of inflammation. Gal1 has broad effects on both the innate and adaptive immune system through its interaction with specific cell-surface glycans on receptors such as CD45, CD43, and CD7 expressed by immune system cells.3 Gal1 induces tolerogenic dendritic cells, regulates the suppressive function of regulatory cells, and induces apoptosis of several T-cell subtypes including antigen-specific T cells. Several groups have shown that Reed-Sternberg cells in classic Hodgkin lymphoma overexpress Gal1, skewing the immune response toward a TH2-type cytokine profile with consequent expansion of regulatory T cells and inhibition of EBV-specific T-cell immune responses.4,5
Because Gal1 has a role in viral infections6 and EBV is detected in Reed-Sternberg cells in a significant percentage of patients with Hodgkin lymphoma, Ouyang et al evaluated whether Gal1 was also expressed in EBV-driven PTLD. They found that Gal1 is expressed in 76% of primary PTLD samples as well as in EBV-transformed B lymphoblastoid cell lines (LCLs) and that expression is driven by the viral LMP1 and 2 genes through AP-1 and P13K/AKT signaling. Taken together these observations suggest that Gal1 expression induced by EBV-encoded proteins may be a means by which the virus can evade an EBV-specific immune response.
The investigators also showed that a newly developed gal1-directed neutralizing monoclonal antibody selectively inhibited galectin-induced apoptosis of EBV-specific CD8 T cells, thereby limiting the development of an immunosuppressive Th2/Treg-skewed tumor microenvironment. Further evidence to support the value of such a neutralizing Gal1 antibody comes from studies in a melanoma model, in which targeted inhibition of Gal1 expression in tumor cells potentiated antitumor effector T cells.7
An alternative approach to bypassing Gal1-mediated immunosuppression is suggested by Khanna and colleagues, who showed that inhibition of LMP-specific T cells could be overcome by ex vivo stimulation with an effective antigen presenting cell—in this case mononuclear cells incubated with an adenoviral vector encoding multiple LMP-derived epitopes.8 Indeed, LMP2-specific cytotoxic T lymphocytes (CTLs) made in a similar manner have had clinical activity against EBV+ Hodgkin lymphoma,9 although it is unclear whether infused cells retain their activity long term or if instead they ultimately become susceptible to Gal1 or other inhibitory molecules such as TGFβ.
How clinically valuable will it be to overcome Gal1-mediated inhibition? In Hodgkin lymphoma, evidence is strong that the presence of an immunosuppressive environment inhibits specific CTL responses. In this disease, a neutralizing Gal1 monoclonal antibody either as monotherapy or in combination with other immunotherapies targeting EBV9 may therefore be of benefit. In PTLD, by contrast, the issue is less the presence of a tumor-associated immunosuppressive environment and more the deficiency in an EBV-specific immune response that can be restored by adoptive transfer of EBV-specific T cells.10 Hence the likely benefits of blocking Gal1 expression may be less apparent.
Conflict-of-interest disclosure: The author declares no competing financial interests. ■