CD40, a member of the tumour necrosis factor receptor (TNFR) superfamily, has the capacity to cause extensive apoptosis in carcinoma cells, while sparing normal epithelial cells. Yet, apoptosis is only achieved by membrane-presented CD40 ligand (mCD40L), as soluble receptor agonists are but weakly pro-apoptotic. Here, for the first time we have identified the precise signalling cascade underpinning mCD40L-mediated death as involving sequential TRAF3 stabilisation, ASK1 phosphorylation, MKK4 (but not MKK7) activation and JNK/AP-1 induction, leading to a Bak- and Bax-dependent mitochondrial apoptosis pathway. TRAF3 is central in the activation of the NADPH oxidase (Nox)-2 component p40phox and the elevation of reactive oxygen species (ROS) is essential in apoptosis. Strikingly, CD40 activation resulted in down-regulation of Thioredoxin (Trx)-1 to permit ASK1 activation and apoptosis. Although soluble receptor agonist alone could not induce death, combinatorial treatment incorporating soluble CD40 agonist and pharmacological inhibition of Trx-1 was functionally equivalent to the signal triggered by mCD40L. Finally, we demonstrate using normal, ‘para-malignant' and tumour-derived cells that progression to malignant transformation is associated with increase in oxidative stress in epithelial cells, which coincides with increased susceptibility to CD40 killing, while in normal cells CD40 signalling is cytoprotective. Our studies have revealed the molecular nature of the tumour specificity of CD40 signalling and explained the differences in pro-apoptotic potential between soluble and membrane-bound CD40 agonists. Equally importantly, by exploiting a unique epithelial culture system that allowed us to monitor alterations in the redox-state of epithelial cells at different stages of malignant transformation, our study reveals how pro-apoptotic signals can elevate ROS past a previously hypothesised ‘lethal pro-apoptotic threshold' to induce death; an observation that is both of fundamental importance and carries implications for cancer therapy.

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.

No abstract

Chemotherapy-induced alopecia (CIA) represents the most distressing side-effect for cancer patients. Scalp cooling is currently the only treatment to combat CIA, yet little is known about its cytoprotective effects in human hair follicles (HF). We have previously established in vitro human keratinocyte models to study the effects of taxanes and anthracyclines routinely-used clinically and reported that cooling markedly-reduced or even completely-prevented cytotoxicity in a temperature dependent manner. Using these models (including HFderived primary keratinocytes), we now demonstrate that cooling markedly attenuates cellular uptake of the anthracyclines doxorubicin and epirubicin to reduce or prevent drug-mediated human keratinocyte cytotoxicity. We show marked reduction in drug uptake and nuclear localization qualitatively by fluorescence microscopy. We have also devised a flow cytometry-based methodology that permitted semi-quantitative analysis of differences in drug uptake, which demonstrated that cooling can reduce drug uptake by up to~8-fold in comparison to normal/physiological temperature, an effect that was temperature-dependent. Our results provide evidence that attenuation of cellular drug uptake represents at least one of the mechanisms underpinning the ability of cooling to rescue human keratinocytes from chemotherapy drug-cytotoxicity, thus supporting the clinical efficacy of scalp cooling.

The capacity to induce tumour-cell specific apoptosis represents the most unique feature of the TNF receptor (TNFR) family member CD40. Recent studies on the signalling events triggered by its membrane-presented ligand CD40L (mCD40L) in normal and malignant epithelial cells have started to unravel an exquisite context and cell type specificity for the functional effects of CD40. Here, we demonstrate that, in comparison to other carcinomas, mCD40L triggered strikingly more rapid apoptosis in colorectal carcinoma (CRC) cells, underpinned by its ability to entrain two concurrently operating signalling axes. CD40 ligation initially activates TNFR-associated factor 3 (TRAF3) and subsequently NADPH oxidase (NOX)/Apoptosis signal-regulating kinase 1 (ASK1)-signalling and induction of reactive oxygen species (ROS) to mediate p38/JNK- and ROS-dependent cell death. At that point, p38/JNK signalling directly activates the mitochondrial pathway, and triggers rapid induction of intracellular TNF-related apoptosis-inducing ligand (TRAIL) that signals from internal compartments to initiate extrinsic caspase-10-asscociated apoptosis, leading to truncated Bid (tBid)-activated mitochondrial signalling. p38 and JNK are essential both for direct mitochondrial apoptosis induction and the TRAIL/caspase-10/tBid pathway, but their involvement follows functional hierarchy and temporally controlled interplay, as p38 function is required for JNK phosphorylation. By engaging both intrinsic and extrinsic pathways to activate apoptosis via two signals simultaneously, CD40 can accelerate CRC cell death. Our findings further unravel the multi-faceted properties of the CD40/mCD40L dyad, highlighted by the novel TNFR crosstalk that accelerates tumour cell-specific death, and may have implications for the use of CD40 as a therapeutic target.