CD40/TRAF6 switch in neointimal hyperplasia

An article by Donner and colleagues in this issue of Blood describes a specific role for the CD40L/CD40/TRAF6 signaling axis in neointima formation upon carotid injury. In this model, several key parameters of neointimal remodeling are demonstrated to be dependent on CD40 signaling through the TRAF6 binding site, but not the TRAF2,3,5 binding site.

CD40 receptor is a member of the TNF receptor family with a prominent role in the regulation of adaptive immune response. The receptor's stimulation by specific CD40 ligand (CD40L) results in activation of macrophages, B cells, and dendritic cells. Vascular cells also express CD40, and stimulation by CD40L mediates inflammatory responses in the vascular wall. The 2 major signaling mediators downstream of CD40 are TRAF2 and TRAF6. Whereas TRAF2 function seems to be confined to the signaling of the TNF receptor family, TRAF6 seems to function more broadly as a mediator of IL-1 and Toll-like receptor signaling.¹  TRAF2 and TRAF6 have distinct binding sites in the CD40 intracellular domain, structurally different from each other.

CD40/CD40L signaling appears to underlie a number of cardiovascular pathologies, and elevated levels of soluble CD40L have been shown to predict cardiovascular events. Previously published analyses of animal models have demonstrated opposing effects of CD40L knock-out (KO) on arteriosclerosis²  and acute collar-induced arterial injury.³ 

The study by Donner et al shows that, upon carotid ligation, neointima formation is substantially reduced in CD40 KO, which is attributed to bone marrow–dependent functions. Indeed, the study documents defective infiltration of CD40-null leukocytes to endothelium, which might result in reduced infiltration of inflammatory cells into the vessel wall. Analysis of the neointima of CD40^−/− animals showed dramatic reduction in CD45-positive cells. Similar reduction is observed in animals expressing CD40 receptor with impaired TRAF6 binding site (CD40-T6), but not TRAF2 binding site (CD40-T2). These analyses conclusively demonstrate that CD40/TRAF6 regulates infiltration of pro-inflammatory cells into the vessel wall and, consequently, neointima formation. Intriguingly, the effect of CD40L KO was minimal compared with CD40 KO, suggesting the possibility of an alternative CD40 ligand. CD40L interaction with αIIb/β3 integrins in thrombosis⁴  also complicates analysis while suggesting that the phenotypes of CD40L and CD40-null mice are not completely identical.

The intriguing novelty of the study by Donner and colleagues lies in its demonstration of a functional dichotomy downstream of CD40. In CD40L and CD40 KO animals, as well as in the case of CD40-T6 mutants, carotid ligation leads to reduced leukocytes infiltration, vessel restructuring, and metalloprotease activity. However, CD40 mutation with an impaired TRAF2,3,5 binding site (CD40-T2) has no such effect.

Either TRAF6 or the TRAF2,3,5 binding site is sufficient for CD40L-triggered NF-κB and JNK stimulation. Nevertheless, knockout of TRAF6 expression rendered cells CD40L-unresponsive. The explanation for this phenomenon is the formation of TRAF2/TRAF6 heterodimers compensating for the mutation in the TRAF6 binding site. However, IL-6 induction has been reported as fully dependent on the TRAF6 binding site (reviewed in Hostager¹ ), similar to the effect on neointima formation, while CD80 up-regulation is induced when either of the 2 binding sites is present. The molecular mechanism of CD40/TRAF6 signaling awaits further exploration. Importantly, careful analysis of the CD40/TRAF6 axis might result in the development of novel therapeutic strategies designed to block only selected CD40 functions without affecting the others.