CD44: target for antiangiogenesis therapy

In this issue of Blood, Tremmel and colleagues provide evidence that endothelial cells express a specific splice variant of the CD44 protein family, CD44v6, that constitutively associates with VEGFR-2 to regulate receptor activation and downstream receptor signaling. Significantly, the targeting of CD44v6 inhibited VEGF-dependent in vitro endothelial cell function as well as in vivo tumor growth and angiogenesis.¹ 

CD44 represents a family of class I transmembrane glycoproteins that are expressed by a wide variety of cells.^2-4  Multiple CD44 isoforms arise from extensive alternative exon splicing that most often involves the tandem insertion of sequences coded by variant exons v1 to v10 into the membrane proximal region of the extracellular domain. Posttranslational modifications, including N- and O-linked glycosylation, glycosaminoglycan additions, and sulfation, dependent on the cell type and growth conditions, lead to further molecular diversity. CD44 has long been recognized as one of the primary cell-surface receptors for hyaluronan (HA), while its cytoplasmic domain engages the cytoskeleton through interactions with linker molecules such as ankyrin and ERM (ezrin, radixin, and moesin) proteins.⁵ 

The CD44 proteins participate in the regulation of a number of diverse cellular processes, including the regulation of cell proliferation, differentiation, survival, and motility.^2-4,6  They have therefore been implicated in developmental events (eg, neuronal axon guidance, fetal lymphogenesis, and limb-bud development), physiologic processes (eg, hematopoiesis, leukocyte recruitment, and lymphocyte homing), and pathologic conditions (eg, tumor growth and spread).

With respect to the growth and spread of tumors, CD44 proteins may be involved in at least 3 ways.^2-4  First, CD44 may stimulate tumor cell proliferation, motility, and/or invasiveness. This may include mechanisms in which the molecule functions to recruit and then enable the activity of surface-associated matrix metalloproteinaises to promote tumor invasiveness or growth factor receptor activation.^7,8  Second, variants of CD44 may function as coreceptors for the activation of growth-promoting tumor receptor tyrosine kinases.⁹  Third, specific CD44 variants may function as tumor cell–surface ligands that interact with endothelial selectins to mediate the initial capture and arrest of circulating tumor cells at a secondary organ.¹⁰  Last, by regulating the function of endothelial cells (ECs)¹¹  and/or the recruitment or activity of inflammatory cells,⁶  CD44 might promote tumor angiogenesis.

A role for CD44 in angiogenesis was first suspected given the involvement of low-molecular-weight HA species in blood vessel formation,¹²  a suspicion that was further re-inforced by the finding that anti-CD44 antibodies block in vitro endothelial cell functions such as cell proliferation and tube formation.¹³  However, direct confirmation of CD44's participation during in vivo angiogenesis has only more recently come from investigations of CD44-null mice.¹⁴  Specifically, these studies demonstrated that the absence of CD44 inhibits angiogenesis, including the vascularization of tumors, an effect that appears to be due principally to the loss of endothelial CD44. In the context of these previous studies, Tremmel et al not only provide further confirmation of the involvement of CD44 in tumor angiogenesis, but also present some additional mechanistic insights into the activity of endothelial CD44 in the formation of new vessels.¹ 

In their paper, Tremmel et al demonstrate that human ECs express CD44 isoforms bearing sequences coded by variant exon 6 (designated as CD44v6). Coimmunoprecipitation studies revealed a constitutive association between vascular endothelial growth factor-2 (VEGFR-2) and CD44v6 (unlike the inducible interaction between CD44v6 and c-Met, the receptor for hepatocyte growth factor [HGF]), which did not require heparan sulfate modification of CD44. Importantly, VEGFR-2 and c-Met activation (but not PDGF-dependent activation) were suppressed by a CD44v6 peptide and a soluble CD44v6 ectodomain. In addition, downstream signaling after VEGFR-2 activation was abrogated by the coexpression of a CD44 cytoplasmic domain construct, an effect that was lost by mutation of the ERM binding sequence. Subsequent studies demonstrated that the targeting of CD44v6 inhibited VEGF-dependent in vitro endothelial cell migration, sprouting, and tube formation. Further, in SCID mice, CD44v6 peptide also suppressed the assembly of human vessels in Matrigel/fibrin implants containing human ECs, as well as the growth and vascularization of human pancreatic tumors. Together, these data implicate CD44v6 variant isoforms as coreceptors in the VEGF-dependent signaling involved in endothelial cell functions required for (tumor) angiogenesis. Issues beyond the scope of this report but worthy of future investigation include the role of HA/CD44 binding in these processes; the molecular basis of the interaction between CD44v6 and VEGFR-2; and the precise role that CD44v6 plays in regulating VEGFR-2 activation and subsequent downstream signaling.

The fact that the targeting of CD44v6 may inhibit signaling by 2 angiogenic growth factors (VEGF and HGF) makes it an attractive, potential approach for cancer therapy. Furthermore, given the broad cellular expression of CD44 and its activity in important physiologic processes, the targeting of a specific CD44 isoform with a potentially more limited cell expression is likely to be better tolerated as a human therapeutic agent. However, the current studies do not exclude off-target effects, such as the inhibition of cytotoxic tumor-infiltrating T lymphocytes,¹⁵  or exclude the development of a vasculopathy resulting from the chronic suppression of VEGF-dependent signaling.¹⁶  It is hoped that continued studies by the authors of this paper and others will be able to move our understanding of the role of CD44 in the biology of tumors to the point where human trials can be performed to determine the effectiveness and safety of targeted anti-CD44 cancer therapy.