Marion Wernli

Transmission of Human Herpesvirus 8 Infection from Renal-Transplant Donors to Recipients

BACKGROUND Human herpesvirus 8 (HHV-8) has been detected in all forms of Kaposis sarcoma, including transplantation-associated Kaposis sarcoma. To investigate the possibility of transmission of HHV-8 through allografts, we measured the seroprevalence of HHV-8 before and after renal transplantation. METHODS Using an enzyme-linked immunosorbent assay with the recombinant HHV-8 protein orf 65.2, we analyzed serum samples from 220 renal-transplant recipients for the presence of antibodies to HHV-8 on the day of transplantation and one year later. Positive results were confirmed by an indirect immunofluorescence assay that detects antibodies to latent antigen and by Western blotting. Follow-up lasted at least four years. RESULTS The seroprevalence of HHV-8 in graft recipients increased from 6.4 percent on the day of transplantation to 17.7 percent one year after transplantation. Seroconversion occurred within the first year after transplantation in 25 patients, and Kaposis sarcoma developed in 2 of them within 26 months after transplantation. Sequential serum samples were obtained from 10 of the patients with seroconversion, and in 8 of these patients, IgM antibodies to HHV-8 appeared within three months after transplantation. In the case of six patients who seroconverted, serum samples from the donors were available, and five (83 percent) tested positive for HHV-8. In a control group of eight patients who were seronegative at the time of transplantation and who received allografts from HHV-8-negative donors, none seroconverted within the year after transplantation. CONCLUSIONS HHV-8 is transmitted through renal allografts and is a risk factor for transplantation-associated Kaposis sarcoma.

Regression of basal cell carcinoma by intralesional interferon-alpha treatment is mediated by CD95 (Apo-1/Fas)-CD95 ligand-induced suicide.

Basal cell carcinoma (BCC) is the most common skin cancer in humans, and although metastasis rarely occurs, the tumor cells are nevertheless able to invade and destroy the surrounding tissue. Intralesional injection of IFN-alpha has been found to be highly effective in inducing BCC regression by an unknown mechanism. We show that in untreated patients, BCC cells express CD95 ligand, but not the receptor, which may allow tumor expansion by averting the attack of activated CD95 receptor-positive lymphoid effector cells. The CD95 ligand of BCC cells is functional as CD95-positive cells incubated on BCC cryosections become apoptotic and are lysed. In IFN-alpha-treated patients BCC cells express not only CD95 ligand but also CD95 receptor, whereas the peritumoral infiltrate that mainly consists of CD4+ T cells predominantly contains CD95 receptor and only few CD95 ligand-positive cells. Thus, in treated patients BCC most likely regresses by committing suicide through apoptosis induction via CD95 receptor-CD95 ligand interaction.

Enhanced gene silencing by the application of multiple specific small interfering RNAs

Small interfering RNA duplexes (siRNA) induce gene silencing in various eukaryotic cells, although usually in an incomplete manner. Using chemically synthesized siRNAs targeting the HIV‐1 co‐receptor CXCR4 or the apoptosis‐inducing Fas‐ligand (FasL), co‐transfection of cells with two or more siRNA duplexes targeting different sites on the same mRNA resulted in an enhanced gene silencing compared with each single siRNA. This was shown in the down‐regulation of protein and mRNA expression, and functionally in the inhibition of CXCR4‐mediated HIV infection and of FasL‐mediated cell apoptosis. Transfection efficiency determined for the FasL‐specific siRNAs was dose‐dependent and varied among the siRNAs tested, but was not the main reason for the enhanced gene silencing.

Green fluorescent protein as a novel tool to measure apoptosis and necrosis.

BACKGROUND Several apoptosis-detecting methods are currently available. Many of them are work intensive and require the additional use of antibodies, dyes, specific substrates, or enzymatic reactions. A simple, fast, and reliable method was developed to test for apoptosis or necrosis using mouse and human cell lines (e.g., Jurkat, A20.2J, and PB3c cells) stably transfected with a vector coding for green fluorescent protein (GFP) as indicator cells. METHODS Apoptosis in GFP-transfected cell lines was induced either by soluble Fas-Ligand (sFasL), recombinant human TRAIL (rhTRAIL), or interleukin-3 (IL-3) deprivation. Necrosis was induced by polyclonal anti-A20 and complement treatment of GFP-transfected A20. Cells were analyzed by flow cytometry for GFP fluorescence. Propidium iodide and Annexin V staining were used to confirm the results obtained with the GFP-method. RESULTS Live GFP-transfected cells show a strong fluorescence intensity, which is significantly diminished upon induction of apoptosis, whereas necrotic GFP-transfected cells almost completely lose their GFP-associated fluorescence. Apoptosis but not necrosis of GFP-transfected cells was blocked by the use of a caspase inhibitor. The results are highly comparable to conventional apoptosis-detecting methods. CONCLUSIONS The advantage of our GFP-based assay compared with other methods is the analysis of apoptosis or necrosis without the necessity for additional staining or washing steps, making it an ideal tool for screening apoptotic or necrotic stimuli.

Apoptosis and Pathogenesis of Melanoma and Nonmelanoma Skin Cancer

Skin cancers, i.e., basal cell carcinoma (BCC), squamous cell carcinoma (SCC) and melanoma, belong to the most frequent tumors. Their formation is based on constitutional and/or inherited factors usually combined with environmental factors, mainly UV-irradiation through long term sun exposure. UV-light can randomly induce DNA damage in keratinocytes, but it can also mutate genes essential for control and surveillance in the skin epidermis. Various repair and safety mechanisms exist to maintain the integrity of the skin epidermis. For example, UV-light damaged DNA is repaired and if this is not possible, the DNA damaged cells are eliminated by apoptosis (sunburn cells). This occurs under the control of the p53 suppressor gene. Fas-ligand (FasL), a member of the tumor necrosis superfamily, which is preferentially expressed in the basal layer of the skin epidermis, is a key surveillance molecule involved in the elimination of sunburn cells, but also in the prevention of cell transformation. However, UV light exposure downregulates FasL expression in keratinocytes and melanocytes leading to the loss of its sensor function. This increases the risk that transformed cells are not eliminated anymore. Moreover, important control and surveillance genes can also be directly affected by UV-light. Mutation in the p53 gene is the starting point for the formation of SCC and some forms of BCC. Other BCCs originate through UV light mediated mutations of genes of the hedgehog signaling pathway which are essential for the maintainance of cell growth and differentiation. The transcription factor Gli2 plays a key role within this pathway, indeed, Gli2 is responsible for the marked apoptosis resistance of the BCCs. The formation of malignant melanoma is very complex. Melanocytes form nevi and from the nevi melanoma can develop through mutations in various genes. Once the keratinocytes or melanocytes have been transformed they re-express FasL which may allow the expanding tumor to evade the attack of immune effector cells. FasL which is involved in immune evasion or genes which govern the apoptosis resistance, e.g., Gli2 could therefore be prime targets to prevent tumor formation and growth. Attempts to silence these genes by RNA interference using gene specific short interfering RNAs (siRNAs) or short hairpin RNAs (shRNAs) have been functionally successful not only in tissue cultures and tumor tissues, but also in a mouse model. Thus, siRNAs and/or shRNAs may become a novel and promising approach to treat skin cancers at an early stage.

CMX001 (1-O-Hexadecyloxypropyl-Cidofovir) Inhibits Polyomavirus JC Replication in Human Brain Progenitor-Derived Astrocytes

ABSTRACT Polyomavirus JC (JCV) replication causes progressive multifocal leukoencephalopathy (PML), a frequently fatal brain disease in immunodeficient patients, yet antiviral drugs are lacking. We characterized the lipid conjugate 1-O-hexadecyloxypropyl-cidofovir (CMX001) regarding JCV (Mad-4) replication in human brain progenitor-derived astrocytes (PDA) and the simian virus 40 (SV40) large-T-antigen-expressing COS-7 cells up to 7 days postinfection (dpi). We examined JCV loads by PCR, the infection rate by immunofluorescence, and host cell toxicity by WST-1 and BrdU incorporation assays. Supernatants from CMX001-treated PDA demonstrated a drug concentration-dependent decrease in JCV loads and infectivity. CMX001 had only a modest effect on host cell metabolism but reduced overall BrdU incorporation. In PDA at 7 dpi, the CMX001 50% effective concentration (EC50) was 5.55 nM, the 50% cytotoxic concentration (CC50) was 184.6 nM, and the 50% selectivity index (SI50) was 33.3. The EC90 was 19.7 nM, the CC90 was 5,054 nM, and the SI90 was 256.1. In COS-7 cells, JCV replication was faster and the EC50 and EC90 were 18- and 37-fold higher than those in PDA, i.e., 0.1 μM and 0.74 μM (CC50, 0.67 μM; SI50, 6.7; CC90, 12.2 μM; SI90, 16.5) at 5 dpi. We conclude that CMX001 inhibits JCV replication at concentrations in vitro that can be attained by oral administration without significant side effects in clinical studies.

Suppression of short interfering RNA-mediated gene silencing by the structural proteins of hepatitis C virus.

Viruses have evolved strategies to overcome the antiviral effects of the host at different levels. Besides specific defence mechanisms, the host responds to viral infection via the interferon pathway and also by RNA interference (RNAi). However, several viruses have been identified that suppress RNAi. We addressed the question of whether hepatitis C virus (HCV) suppresses RNAi, using cell lines constitutively expressing green fluorescent protein (GFP) and inducibly expressing HCV proteins. It was found that short interfering RNA-mediated GFP gene silencing was inhibited when the entire HCV polyprotein was expressed. Further studies showed that HCV structural proteins, and in particular envelope protein 2 (E2), were responsible for this inhibition. Co-precipitation assays demonstrated that E2 bound to Argonaute-2 (Ago-2), a member of the RNA-induced silencing complex, RISC. Thus, HCV E2 that interacts with Ago-2 is able to suppress RNAi.

Intralesional Interferon in Basal Cell Carcinoma: How Does It Work?

Basal cell carcinoma (BCC) is the most common skin cancer among Caucasians, and its incidence is increasing. Intralesional injection of interferon alpha (IFN alpha) has been shown to provide a safe and effective treatment for BCCs. The predominant mechanism for the effect of IFN alpha on BCC has been partially identified. We have shown that in untreated patients, BCC cells constitutively express CD95 ligand (CD95L), but not the receptor. BCC cells make use of the CD95 ligand to escape from a local immune response by averting the attack from activated CD95 receptor-positive CD4+ T cells. The CD95L of BCC cells is functional as CD95+ target cells incubated on BCC cryosections become apoptotic and are lysed. In IFN alpha-treated patients BCC cells express not only CD95L but also CD95 receptor, and regress by committing suicide or fratricide through apoptosis induction via CD95 receptor-CD95L interaction. Peritumoral infiltrating cells, predominantly CD4+ T cells, may support regression of BCC by the secretion of cytokines such as IFN gamma or interleukin-2 which may also be responsible for the up-regulation of CD95 on BCC cells.

Gene silencing of transcription factor Gli2 inhibits basal cell carcinomalike tumor growth in vivo.

Basal cell carcinoma (BCC) belongs worldwide to the most frequent malignancy among Caucasians. The understanding of the molecular mechanisms of BCC formation, which is a prerequisite for the development of efficient new therapies, is still incomplete. The formation of sporadic BCCs in the skin is associated with uncontrolled hedgehog signaling, and the transcription factor Gli2 has been identified as a key mediator or effector of this signaling. There is indication in the literature that preventing Gli2 function may inhibit BCC formation and growth in vivo; however, the mechanism is unclear and difficult to study in humans. Therefore, we used a mouse tumor allograft model to investigate the role of Gli2 in tumor formation. A constitutively Gli2 expressing mouse tumor cell line was stably transfected with Gli2‐specific shRNA to induce Gli2 gene silencing or with control shRNA. Injecting the Gli2 gene silenced cells into nude mice for tumor formation we detected a strongly retarded tumor growth compared with control tumor cells. Investigating the mechanisms, we found that Gli2 gene silencing has led to the disruption of the tumor structure as demonstrated by staining tumor sections with hematoxylin. Two main reasons for the tumor destruction were identified. We found that apoptosis was markedly increased while vascularization was strongly decreased in these tumors. Thus, important functions of the transcription factor Gli2 in this tumor model are the prevention of apoptosis and the promotion of microvascularization.

Tumor necrosis factor-related, apoptosis-inducing ligand supports growth of mouse mastocytoma tumors by killing tumor-infiltrating macrophages.

TRAIL antisense transfected mastocytoma cells (R56VTas) injected into syngeneic DBA/2 mice demonstrate significantly delayed tumor growth compared to mock transfected cells (R56VTMo). TRAIL expression in R56VTas cells was successfully, albeit not completely, downregulated, as shown by Western blots, flow‐cytometric analysis and functionally by loss of cytolytic activity against TRAIL‐R‐bearing target cells. Immunohistochemic and immunoblotting analyses of ex vivo tumors confirmed the lower expression of TRAIL by the antisense transfection compared to the mock transfection. Investigating the mechanism of the delayed tumor growth, it was found that neither T nor NK cells but activated macrophages infiltrated the tumors. The number of infiltrating macrophages was significantly lower in the mock transfected compared to the TRAIL antisense transfected tumor sections, indicating that TRAIL‐expressing tumor cells may lyse macrophages. Indeed, activated macrophages proved to be sensitive to TRAIL‐mediated apoptosis. This indicates that, although macrophages can infiltrate the mastocytoma R56VT, they are in part eliminated by TRAIL‐expressing tumor cells, allowing the tumor to rapidly grow. Hence, downregulation of TRAIL allows more macrophages to survive and to better attack the tumor cells, slowing down tumor growth. In conclusion, TRAIL expressed on R56VT tumor cells can impair an important innate immune defense mechanism against tumors by eliminating effector macrophages.