Dysregulation of receptor induced apoptosis during human leishmaniasis : A possible mechanism of skin ulceration

Sammanfattning: Leishmaniasis is endemic in 88 countries and 12 million people per year have been estimated to be at risk of infection. The causative agent, the protozoan Leishmania, is spread by sand-flies and infects macrophages in the mammalian host. Leishmaniasis in humans form a spectrum of clinical presentations. In cutaneous leishmaniasis (CL) caused by L. major, the infection is localised in the skin and manifests as one or several ulcers that typically spontaneously heal within one year after infection, often leaving marked scarring. In visceral leishmaniasis (VL) caused by L. donovani, infected macrophages are found in liver, spleen, bone marrow and lymph-nodes and the infection leads to hepatosplenomegaly, wasting, fever and if not treated, to death. Systemic T-cell deficiency occurs early during VL and leads to uncontrolled parasite replication. In general, solid immunity upon healing occurs after resolved VL and CL. In this thesis, the hypothesis that alterations of death receptor-mediated apoptosis have an impact on the pathogenesis of human leishmaniasis has been explored. During VL, dysregulation of the Fas/FasL pathway was investigated, both at one site of infection, the spleen, and on circulating lymphocytes from the peripheral blood. In the case of CL the hypothesis was that increased death receptor-mediated apoptosis in the microenvironment surrounding infected macrophages may induce bystandard apoptosis of keratinocytes, leading to skin ulceration. Dysiregulation of the Fas/FasL pathway occured during human VL and CL. Levels of soluble Fas (sFas) and soluble Fas ligand (sFasL) were elevated in plasma of patients with active VL and individuals co-infected with VL-HIV-1 compared to healthy controls, and the levels of sFas and sFasL were normalized 6 months after successful treatment. During active VL, the expression of membrane bound Fas, and to a lower extent Fast., was up-regulated on spleen cells, where parasites multiply. In contrast, expression of Fas and Fast- were not altered on peripheral blood mononuclear cells (PBMC) during VL. Furthermore, in vitro infection of macrophages with L. donovani results in up-regulation of Fas expression on the surface of infected cells and increases the levels of sFasL in supernatants from infected cultures. During active CL caused by L. major, a disease mainly localised to the skin, the Fas and Fast- levels were not altered in serum or on PBMCs analysed ex vivo. However, when CL PBMCs were restimulated with L. major, Fas was up-regulated on effector T-cells and high levels of sFasL were detected in the supernatants as compared to control PBMCs. Keratinocyte apoptosis is altered during CL. Dysregulation of the Fas/FasL pathway in the microenvironment surrounding L. major infected macrophages under the skin was visualised in biopsies collected from CL patients. A substantial number of apoptotic keratinocytes were observed in the epidermis of morphologically active and healing CL skin samples. Fas expression was increased on the epidermis in active CL, whereas FasL expressing macrophages and T-cells were found in the subepidermal infiltrate during active disease. Supernatants from re-stimulated CL-PBMC cultures containing high levels of sFasL induced apoptosis in human keratinocyte cell line (HaCaT), and apoptosis could be inhibited in 213 supernatants by blocking Fas. A commercial apoptosis-specific microarray was used to assess alterations in keratinocyte RNAexpression during exposure to supernatants from L. major infected PBMCs. Fas and TRAIL mRNA and protein expression were significantly up-regulated compared to untreated keratinocytes. Supernatant induced apoptosis of keratinocytes was partly inhibited through blocking Fas or FasL, and more efficiently through inhibition of TRAIL by neutralising antibodies or soluble TRAIL-R. Furthermore, TRAIL expressing keratinocytes were detected in skin biopsies from CL cases. Blocking the Fas/FasL pathway in vivo may reduce ulceration during murine CL. In order to obtain the proof of the concept that Fas/FasL signalling is involved in keratinocyte-apoptosis leading to ulceration in the skin during CL, the Fas/FasL pathway was blocked in a murine model of CL by intraperitoneal treatment with Fast- neutralising antibodies (MFL-4). Skin inflammation, skin ulceration and ulcer size were followed weekly and compared to infected, untreated mice. Our results suggests that blocking Fas/FasL signalling during murine CL lead to less apoptotic keratinocytes and diminished ulceration. During treatment, the number of IFNgammaproducing CD8+CD3+ cells was increased at the site of infection when Fast- was neutralised which is suggestive of efficient parasite eradication. However, there was no reduction of parasite load at the site of infection or in draining lymph nodes and parasite replication was high upon discontinuation of anti-FasL treatment. Conclusion The Fas/FasL pathway was shown to be dysregulated both in human VL and CL. A possible mechanism of ulcer formation during CL was proposed by apoptotic death of keratinocytes through enhanced Fast- and TRAIL signalling. This data was further strengthened in a treatment experiment in a murine model of CL, where blocking the Fas/FasL system reduced ulcer formation during L. major infection.