Abstract 1640

Poster Board I-666


Natural killer (NK) cells are CD3- CD56+ lymphocytes demonstrating confirmed cytotoxicity against neoplastic and virus infected host cells. Increasing data provide evidence of a direct NK cell effect against extracellular pathogens, such as bacteria, parasites and yeasts, but there is a relative lack of data on their interaction with filamentous fungus and especially with Aspergillus fumigatus. Aspergillus is an omnipresent mold, living in close vicinity with humans, being constantly inhaled in the lungs and thereafter cleared by the innate immune system. Otherwise harmless for healthy people, it is at the origin of invasive Aspergillosis (IA), an extremely devastating disease for immunocompromised subjects. Host's innate immune system controls Aspergillus growth through a complex system of potent effector cells, mediating their antifungal activity mainly by phagocytosis. Our study aims to shed light for the first time on the direct interaction between human NK cells, mediators of extracellular cytotoxicity, and Aspergillus.


NK cells were isolated after magnetic depletion of the peripheral blood of healthy volunteers and they were used after 24h priming with 500 U/ml recombinant interleukin – 2 rhIL-2. To determine gene expression and cytokine release of interferon gamma (IFNg) and Tumor Necrosis Factor- a (TNF-a), NK cells were stimulated for 0, 3, 6 and 12h with different morphologies of Aspergillus: conidia and germlings. To evaluate the lethal impact of NK cells on Aspergillus, plate killing assays were performed at 0, 3 and 6h time points. To illustrate the role of antibody dependent cellular cytotoxicity, ADCC a monoclonal IgG antibody, against germlings, was tested. Transwell permeable membranes, with pores of 0,4 μm, prohibiting the direct contact of cells placed on their opposite sides, but allowing the free circulation of molecules, were used to estimate the effect of cell-fungal contact. To investigate the cytotoxic mechanism involved, NK cells were depleted from perforin and granzymes by treatment with strontium chloride and they had their death ligands, TNF- related apoptosis- inducing ligand (TRAIL) and FasL, neutralised by means of blocking antibodies. The release of cytotoxic granules was estimated by the NK cell surface expression of the marker of degranulation CD107a/b.


Observing the in vitro interaction of NK cells with Aspergillus, fungal germinated morphologies (germlings) showed to be highly immunogenic towards NK cells, compared to conidia, inducing the gene expression and cytokine release of Th1 immune mediators such as IFN-g (p <0,05) and TNF-a.(p <0,1). NK cells demonstrated also a strong lethal impact against germlings (p <0,05). Moreover, the presence of antifungal antibody further potentiated both immunoregulatory and cytotoxic activities. Investigating the means engaged by NK cells to perceive and kill Aspergillus, direct effector–pathogen cell to cell contact was revealed as prerequisite; when this condition was not present there was neither cytokine induction, nor fungal damage (p <0,05). This finding was confirmed by the lack of surface expression of CD107a/b, after NK cell- Aspergillus co-incubation. Investigating the killing pathway we compared the effectiveness of perforin – granzymes depleted NK cells to this of intact cells against germlings and it was found equivalent (p =NS). In a similar way, neutralisation of TRAIL and FasL ligands did not alter the cytotoxic ability of NK cells towards Aspergillus.


Our data show that human NK cells are stimulated in vitro by Aspergillus germlings, which triggers an immunoregulatory Th1 orientated response and causes important fungal killing. NK cells are not aware of conidia, they are not stimulated by them and par consequence they do not kill them. Finally, we showed that NK cells do not mediate their cytotoxic effect via perforin – granzymes pathway, neither through the engagement of TRAIL, FasL death receptors, suggesting that another pathway is involved in NK cell – Aspergillus fumigatus interplay. We suggest that further investigation of these striking findings might offer a potent immunotherapeutic tool against IA.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.