CD40 induces renal cell carcinoma-specific differential regulation of TRAF proteins, ASK1 activation and JNK/p38-mediated, ROS-dependent mitochondrial apoptosis.

ABSTRACT

A unique feature of CD40 among the TNF receptor (TNFR) superfamily is its exquisitely contextual effects, as originally demonstrated in normal and malignant B-lymphocytes. We studied renal cell carcinoma (RCC) in comparison to normal (human renal proximal tubule) cells, as a model to better understand the role of CD40 in epithelial cells. CD40 ligation by membrane-presented CD40 ligand (mCD40L), but not soluble CD40 agonist, induced extensive apoptosis in RCC cells; by contrast, normal cells were totally refractory to mCD40L. These findings underline the importance of CD40 'signal-quality' on cell fate and explain the lack of pro-apoptotic effects in RCC cells previously, while confirming the tumour specificity of CD40 in epithelial cells. mCD40L differentially regulated TRAF expression, causing sustained TRAF2/TRAF3 induction in RCC cells, yet downregulation of TRAF2 and no TRAF3 induction in normal cells, observations strikingly reminiscent of TRAF modulation in B-lymphocytes. mCD40L triggered reactive oxygen species (ROS) production, critical in apoptosis, and NADPH oxidase (Nox)-subunit p40phox phosphorylation, with Nox blockade abrogating apoptosis thus implying Nox-dependent initial ROS release. mCD40L mediated downregulation of Thioredoxin-1 (Trx-1), ASK1 phosphorylation, and JNK and p38 activation. Although both JNK/p38 were essential in apoptosis, p38 activation was JNK-dependent, which is the first report of such temporally defined JNK-p38 interplay during an apoptotic programme. CD40-killing entrained Bak/Bax induction, controlled by JNK/p38, and caspase-9-dependent mitochondrial apoptosis, accompanied by pro-inflammatory cytokine secretion, the repertoire of which also depended on CD40 signal quality. Previous reports suggested that, despite the ability of soluble CD40 agonist to reduce RCC tumour size in vivo via immunocyte activation, RCC could be targeted more effectively by combining CD40-mediated immune activation with direct tumour CD40 signalling. Since mCD40L represents a potent tumour cell-specific killing signal, our work not only offers insights into CD40's biology in normal and malignant epithelial cells, but also provides an avenue for a 'double-hit' approach for inflammatory, tumour cell-specific CD40-based therapy.