Oral Abstracts

OP-03 High Mortality in HIV and HTLV-1 Coinfected Patients with Pulmonary Tuberculosis in Guinea-Bissau, West Africa, Despite Higher CD4 Counts Compared to HIV-Positive HTLV-1 Negative Patients

Background: Guinea-Bissau, West Africa, is highly endemic for HIV-1, HIV-2, HTLV-1 and tuberculosis (TB). In a recent study there we have reported that HTLV-1 infection alone was not enough to increase the risk of acquiring TB, whereas HTLV-1 increased the risk of TB among HIV-infected individuals (JAIDS 2008;48:607-10). We have now extended the study to clinical aspects including mortality.

Methods: A prospective study on hospitalized patients with pulmonary tuberculosis was performed in Guinea-Bissau 1994-1997 including HIV, CD4 counts and clinical outcome. We determined the prevalence of HTLV-1 in all patients and compared the levels of CD4 counts at the time of inclusion between HIV-negative and HIV-positive patients, with or without HTLV-1. Mortality was established while patients were on treatment for tuberculosis.

Results: 280 patients were included in the study, and the prevalence of HIV-1, HIV-2 and HIV-1 + HIV-2 was 9.3%, 21.4% and 11.4%, respectively. In total, 32 patients were positive for HTLV-1 (11.4%), significantly more among HIV-positive subjects (20.0%) than among HIV-negative patients (3.7%). No cases of HTLV-2 were found. Median CD4% was significantly higher in HIV-positive subjects coinfected with HTLV-1 (22, IQR 13-37) compared to HIV-positive HTLV-1 negative patients (13, IQR 6.5-26) adjusted for age and gender in a logistic regression analysis (p = 0.01; 95% CI 1.9-13.6). 233 individuals were included in the analysis of mortality, and among HIV-negative subjects the mortality was 18.0/100 person-years (p.y.). In HIV-2-positive HTLV-1-negative subjects the mortality was 39.5/100 p.y and in HIV-2 and HTLV-1 coinfected patients it was 112.6/100 p.y. (adjusted mortality rate ratio 4.7, 95% CI 1.5-14.4; p < 0.01). When all HIV-positive patients were analyzed together corresponding mortality rates were 53.5/100 p.y. and 104.9/100 p.y., respectively, but this difference did not reach significance (adjusted mortality rate ratio 1.7, 95% CI 0.9-3.6; p = 0.13).

Conclusions: HIV and HTLV-1 coinfected patients hospitalized for pulmonary tuberculosis had increased mortality despite significantly higher CD4% compared to HIV singly infected subjects. This may imply that HTLV-1 adversely affects the immune function independently of CD4 counts, rendering coinfected subjects more vulnerable for tuberculosis. HTLV-1 may be an important inducer of HIV-2 associated mortality in tuberculosis patients. HIV-infected patients with concomitant HTLV-1 infection cannot be judged by the same CD4 standards as HIV singly infected individuals.

OP-04 Identification of a Novel, Highly Divergent HTLV-3 in a Primate Hunter in Cameroon

The recent discovery of HTLV-3 and HTLV-4 in primate hunters in Cameroon suggests that additional surveillance studies are needed to determine the geographical distribution and prevalence of these novel viruses in persons exposed to nonhuman primates (NHP). Although a simian counterpart for HTLV-4 has yet to be identified, all three HTLV-3 strains reported to date fall within the STLV-3 subtype b lineage from West-Central Africa rather than within subtypes a and c, or the divergent STLV-3d lineage recently found in two monkey species from Cameroon. Blood specimens for serologic and PCR testing were collected from 402 persons reporting NHP exposure in eight villages in the three forested areas of Cameroon. Plasma samples from 29 persons (7.2%) were HTLV EIA and Western blot (WB) reactive with the following serologic profiles: HTLV-1 (n = 3), HTLV-2 (n = 2), HTLV-positive but untypeable (n = 6), indeterminate (n = 18). Using established generic PTLV (primate T-lymphotropic virus) PCR assays, tax sequences were detected in peripheral blood mononuclear cell DNA of two HTLV-1-seropositive persons (Cam1806 and Cam1902) and one person classified as HTLV-positive but untypeable (Cam2013) with moderate WB reactivity to only gag (p24, p32) and recombinant envelope (GD21) proteins. Phylogenetic analysis of Cam1806 and Cam1902 tax sequences confirmed infection with HTLV-1. However, tax sequences from Cam2103 clustered with strong bootstrap support within the PTLV-3 clade with two STLV-3d sequences from a Cercopithecus mona (Cmo8699AB) and C. nictitans (Cni7867AB) captured in forests near the hunter's village. LTR sequences obtained from Cam2013 shared 98% identity to the two STLV-3d sequences and all clustered together with high bootstrap support in a distinct lineage in the PTLV-3 phylogroup. These results confirm a fourth HTLV-3 infection with a new and highly divergent strain we designate HTLV-3(Cam2013) subtype D. This infection may have originated from a cross-species transmission of an STLV-3d strain and demonstrates that broad HTLV-3 diversity is due to multiple introductions of divergent STLV-3. The wide distribution and inferred ancient origin of STLV-3 in Africa, the proclivity of STLV to cross species barriers, and the identification of multiple HTLV-3 subtypes, all imply that HTLV-3 may have a greater geographic dispersal and prevalence than currently appreciated. Further studies are necessary to determine the public health implications of HTLV-3 infection.

OP-05 Serological and Epidemiological Characteristics of HTLV-1 and HTLV-2 Infections in Gabon, Central Africa

HTLV-1 infection is considered as highly endemic in central Africa. Recently, based on epidemiological studies in Cameroon a new HTLV-3 and HTLV-4 were discovered. To examine the distribution of various strains circulating in Gabon, a central African country, and describe a new cartography of infections by HTLVs, 4381 samples were collected from adults (mean age 47 years-old) living in the nine regions of the country including more than 220 villages. The prevalence of HTLVs, was determined by two ELISA tests (BioRad and Vironostica), and the presence of HTLV-1 and HTLV-2 in the positive samples was confirmed by western blotting. We found a global HTLV-1 seroprevalence of 7.3% (320/4381) and 0,1% (5/4381) were HTLV-2 positives. Twelve percent (527/4381) of the samples had an indeterminate western blot profile. The seroprevalence differed significantly by region; with the lowest prevalence in the coast region (2,3%) and the highest prevalence in the rainforest region (12,5%). Furthermore, the prevalence increased significantly by age, being 4,0% in individuals under 25 years-old and reaching 9,8% in individuals over 55 years-old. After adjustment for sex, the prevalence differed also significantly; women were significantly more infected (9.1%) than men (5.4%). The epidemiological studies underlined some risk factors associated with HTLV-1 infection such as hospitalization, surgery, transfusion, abortion, abdominal pain, and arthritis. A follow-up study of this HTLV-1 infected population as well as the distribution of various HTLVs strains circulating in the country are now in progress.

OP-06 Follow-up of HTLV-1 Infected Pregnant Women and Analyses of Vertical and Sexual Viral Transmission in Their Familiar Groups

Vertical transmission of HTLV is an important source of spread of viral infection in endemic areas. Particular attention has been directed to counseling women of childbearing age, specially considering that ATL may be a consequence of childhood infection. The Interdisciplinary HTLV Research Group (GIPH) leads an open prevalent cohort study of HTLV-1/2 infected individuals since 1997, which comprises a regular follow-up of participants, including epidemiological and clinical evaluations as part of the ongoing research protocols.

Objective: To investigate HTLV transmission among familiar groups of seropositive pregnant women followed up since 1997 by GIPH cohort, who were counseled to be submitted to cesarean delivery and avoid breastfeeding; to verify how this orientation was effective to diminish viral transmission rate.

Results: Twenty two HTLV-1 infected female (16 deferred blood donors and six related to deferred blood donors) with age varying from 17 to 41 years, had 24 children during their follow-up in GIPH cohort, of which 66.7% (16/24) were tested by EIA, WB and PCR for HTLV-1/2. Just one child was HTLV-1 seropositive, and according to her mother, she was delivered by cesarean section and was not breastfed, suggesting intra-uterine viral trasmission. Of the total of children, four (16.7%) were breastfed for different periods, despite medical advise. Three were not tested for HTLV, and the other was seronegative. Familiar groups (parents, sexual partners, brothers and sisters, and children) of these women were investigated, totalizing 103 individuals tested for HTLV, of which 53 (51.46%) were seropositive and 50 (48.54%) seronegative, showing high rate of viral trasmission inside familiar groups. Twenty two (68.75%) individuals had probably been infected by vertical and 10 (31.25%) by sexual transmission. Four events of HTLV vertical transmission without infection of her sexual partner, and three events of sexual transmission without vertical transmission were observed. On the other hand, in three familiar groups, both sexual and vertical viral transmission occurred among the members of the one family. The rate of viral transmission from mothers to children who had been born during GIPH cohort study was 6.25% (1/16), and was of 25% (3/12) in their other children who had been born before their follow-up in the cohort, i.e., four times less than before intervention.

Conclusion: The data show the importance of HTLV transmission among individuals in familiar groups, especially vertical transmission. Counseling HTLV seropositive pregnant women to avoid breastfeeding is an important precaution to diminish the mother to child viral transmission, but the evaluation of the mother's reasons to not taking the medical advice into consideration is necessary to make it more effective.

Finantial support: Fundação Hemominas, FAPEMIG.

OP-07 HBZ Binding to HLA Class 1 Determines the Outcome of HTLV-1

Background: HTLV-I-infected individuals show considerable heterogeneity in their set point proviral load and clinical outcome. There is evidence that HTLV-I-specific CD8 + T cells lower proviral load and reduce the risk of the associated disease HAM/TSP. However, what constitutes a protective CD8 + T cell response is poorly understood.

Hypothesis: We hypothesised that a HTLV-I-infected individual's proviral load and HAM/TSP risk was determined by the epitope binding properties of their HLA class I alleles.

Results: We tested this hypothesis in 432 HTLV-I-infected individuals and showed that:

Class I alleles previously associated with reduced proviral load and HAM/TSP prevalence were predicted to bind epitopes from the viral protein HBZ significantly more strongly than an allele associated with increased proviral load and HAM/TSP prevalence.

Conclusions: We demonstrated that protection is not limited to a small subset of HLA class I alleles previously associated with disease status and proviral load (HLA-A*02 and HLA-Cw*08), but is more generally associated with HLA class I alleles that bind strongly to HBZ. Furthermore, across all alleles and all proteins, the protection from HAM/TSP conferred by HLA binding was significantly associated with the reduction in proviral load associated with HLA binding. This strongly suggests that CD8+T cells play a significant role in the control of HTLV-I infection with HBZ-specific CD8 + T cells being the most effective at reducing proviral load and risk of HAM/TSP.

OP-09 Invariant NKT Cells have Anti-Human T-Lymphotropic Virus Type-1 (HTLV-1) Activity and the Frequency is Significantly Decreased in Patients with HTLV-1 Associated Disorders

Invariant NKT (iNKT) cells constitute a unique T-cell population that regulates the immune response to microbes, cancers and autoimmunity. We assessed the characteristics of iNKT cells in individuals infected with human T-lymphotropic virus type 1 (HTLV-1). While most HTLV-1 infected individuals remain lifelong asymptomatic carriers (AC), a small proportion, usually with higher equilibrium proviral loads, develops two different diseases, HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP) or adult T-cell leukemia (ATL). We demonstrate that the frequency of iNKT-, NK-, and dendritic cells is decreased in peripheral blood of HAM/TSP- and ATL patients. We also observed an inverse correlation between the frequency of iNKT cells and the HTLV-1 proviral load in peripheral blood of HTLV-1 infected individuals. Notably, in vitro stimulation of peripheral blood cells with ?-galactosylceramide increased the number of iNKT cells and subsequently decreased the number of HTLV-1 infected T-cells in samples from ACs, but not in samples from patients with HAM/TSP or ATL. Our results suggest that iNKT cells contribute to the immune defense against HTLV-1, and that iNKT cell depletion is important in the pathogenesis of HAM/TSP and ATL. Therefore, iNKT cell-based immunotherapy may represent an effective strategy to prevent the development of these HTLV-1 associated disorders.

OP-11 Elevated Cyclic Adenosine Monophosphate (Camp) Levels in CD4+T Lymphocytes Transformed with HTLV-1

HTLV-1, the cause of adult T cell leukemia/lymphoma (ATLL), transforms CD4+T cells to permanent growth which is mainly mediated by the viral transactivator Tax. As HTLV-transformed cells are phenotypically similar to long-lived T cell populations such as regulatory T cells (Treg) and high cAMP levels are a key component of Treg mediated suppression, we further defined their phenotype and asked whether elevated cAMP levels are also a property of HTLV-transformed cells. Most HTLV-transformed cell lines used in our study exhibited a long-lived phenotype, which is characterized by the expression of characteristic differentiation markers (CD4+CD25+4-1BB+GITR+CD127− CCR7+). Moreover, elevated intracellular cAMP levels were a consistent feature of all HTLV-1-transformed T cell lines, whether patient-derived or in vitro-transformed. The content of intracellular cAMP was significantly elevated in HTLV-transformed cell lines in comparison to both uninfected primary CD4+lymphocytes from healthy donors and to leukemia-derived cell lines. In cells with repressible Tax, cAMP was significantly upregulated in the presence of Tax, but Tax was not sufficient to induce cAMP in various model systems. When analysing cellular determinants of cAMP upregulation, PDE3B, a cAMP-cleaving phosphodiesterase (PDE) known to be downregulated in murine regulatory T cells, was the only PDE found to be dramatically downregulated in two different HTLV-transformed cell lines in comparison to CD4+lymphocytes from healthy donors as detected by microarrays. qPCR analyses revealed a significant downregulation of PDE3B in a variety of HTLV-transformed T cell lines also in comparison to acute lymphoblastic leukemia cell lines. cAMP levels and PDE3B mRNA expression inversely correlated in HTLV-transformed cells, but PDE3B regulation was independent of Tax. FOXP3 binding to a conserved region of the PDE3B gene was observed in chromatin immunoprecipitation (ChIP) analyses in ATLL-derived cells, which could provide an explanation for weak PDE3B expression. In summary, intracellular cAMP levels were found to be elevated in various HTLV-transformed T cell lines. Apparently, the viral Tax protein could account for elevated cAMP levels. Cellular properties, like downregulation of the cAMP dismantling phosphodiesterase PDE3B, which is a determinant of long-lived T cells populations, could support the virus-mediated properties. These studies will give a hint to which extent HTLV-transformed cells share properties with immunosuppressive fractions of T lymphocytes.

OP-12 High Expression of CD244 and SAP Regulated CD8+T Cell Responses of Patients with HTLV-I Associated Neurologic Disease

High frequencies HTLV-I-specific CD8+T cells have been demonstrated in peripheral blood and cerebrospinal fluid of patients with HTLV-I-associated myelopathy/tropical spastic paraparesis (HAM/TSP). Mechanisms whereby the CD8+CTL have been suggested to play a role in HAM/TSP pathogenesis include recognition and lysis of virus infected cells in the CNS and production of proinflammatory cytokines that may contribute to central nervous system inflammation. Little is known about the differences in CD8+T cell activation between HTLV-I infected asymptomatic carriers (ACs) and patients with HAM/TSP. Effector functions of CD8+T cells are known to be regulated by various cellular receptors and their downstream molecules. The signaling lymphocyte activation molecule (SLAM) family of receptors and their associated adaptors play a pivotal role in the control of both innate and adaptive immunity, and also have been shown to be involved in various inflammatory diseases, such as rheumatoid arthritis and inflammatory bowl disease. Since it has been demonstrated that activated HTLV-I specific CTL may play an important role in the pathogenesis of HAM/TSP and that CD244, one of SLAM family receptor, is a crucial regulator of CTL function, we characterized the expression of CD244 on CD8+T cells of HTLV-I-infected patients and examined their functions in patients with HAM/TSP. Here we demonstrate that the expression of CD244 was significantly higher on CD8+T cells in HTLV-I-infected patients, both ACs and patients with HAM/TSP, than those on healthy normal donors (NDs). Blockade of CD244 inhibited degranulation and IFN-gamma production in CD8 + T cells of patients with HAM/TSP, suggesting that CD244 is associated with effector functions of CD8 + T cells in patients with HAM/TSP. Moreover, the SLAM-related adaptor protein (SAP) was specifically overexpressed in patients with HAM/TSP compared to ACs and NDs. SAP expression in Tax-specific CTLs was correlated with the frequency of these cells in HTLV-I-infected patients. SAP knockdown by siRNA also inhibited IFN-gamma production in CD8+T cells of patients with HAM/TSP. Thus, CD244/SAP pathway is involved in the active regulation of CD8+T cells of patients with HAM/TSP, and therefore may play an important role in promoting inflammatory neurological disease.

OP-13 High Frequencies of Functionally Competent Circulating Tax-Specific CD8+T Cells in Human T Lymphotropic Virus Type 2 (HTLV-2) Infection

Human T lymphotropic virus type 2 (HTLV-2) is characterized by a clinically asymptomatic persistent infection in the vast majority of infected individuals. In this study, we have characterized for the first time ex vivo specific cytotoxic T lymphocyte (CTL) responses against the HTLV-2 Tax protein. The CTL responses were found to be solely directed to a single HLA-A*0201-restricted Tax2 epitope, comprising residues 11 to 19 (LLYGYPVYV). Virus-specific CTLs could be detected in the majority of evaluated subjects, with frequencies as high as 24% of circulating CD8 + T cells. The frequency of specific CTLs had a statistically significant positive correlation with proviral load (PVL) levels. The majority of virus-specific CD8 + T-cells exhibited an effector memory/ terminally differentiated phenotype, produced cytotoxicity mediators including perforin and granzyme B, and lysed in vitro target cells pulsed with Tax2(11-19) synthetic peptide. Our findings suggest that a strong, effective CTL response may control HTLV-2 viral burden and this may be a significant factor in maintaining persistent infection and in the prevention of disease in infected individuals.

OP-15 The Regulation of Innate Immune Signaling by HTLV-1 Tax

Activation of host innate immune responses following virus infection is largely mediated by viral encoded RNA and DNA, the mechanisms of which remain to be fully determined. For example, the DExD/H box RNA helicases RIG-I/MDA5 are known to complex with intracellular viral RNA and to require the CARD containing molecule, IPS-1/MAVS/VISA/Cardif, to activate NF-(B and IRF3 and trigger type I interferon (IFN) production. These processes require FADD/RIP1 and TRAF3-κ. A TLR-dependent pathway also exists to recognize viral RNA species, mainly in dendritic cells and macrophages, with TLR3 and TLR 7/8 facilitating activation of the IFN pathway in response to dsRNA and ssRNA, respectively. In contrast, little is known regarding the activation of the IFN pathway in response to intracellular viral DNA, although an endoplasmic reticulum adaptor referred to as STING appears to regulate this important signaling process. It is becoming increasingly clear that a number viruses have devised a variety of ways to avoid host defense mechanisms by inhibiting innate signaling pathways. For example, hepatitis C virus (HCV) reportedly degrades IPS-1, while measles virus V protein inhibits MDA5. Here we show that HTLV-1 Tax inhibits innate signaling processes by similarly blocking the induction of the type I IFN pathway. We will provide data on the mechanisms of retroviral-mediated activation of innate immune pathways, as well as discuss the mechanisms of how Tax blocks the TLR-dependent and independent signaling processes.

OP-16 The Roles of HTLV-1 bZIP Factor Gene in Oncogenesis

Among viral proteins encoded by human T-cell leukemia virus type 1 (HTLV-1), Tax has pleiotropic actions that induce proliferation and inhibit apoptosis of T-cells. Therefore, Tax is considered to play a critical role in oncogenesis. However, Tax expression is frequently lost in fresh adult T-cell leukemia (ATL) cells by genetic changes of tax gene, deletion of 5’ long terminal repeat (LTR), and DNA methylation of LTR. The HTLV-1 bZIP factor (HBZ) gene is expressed as an antisense transcript of HTLV-1 from 3’LTR. Importantly, the HBZ gene is expressed in all ATL cases and their sequences remain intact in fresh ATL cases. We showed that spliced HBZ gene could induce proliferation of T-cells as from of RNA, and HBZ protein has different functions.

To clarify the significance of HBZ gene in oncogenesis, we generated transgenic mice that expressed the spliced HBZ gene under control of CD4 specific promoter/enhancer. In these transgenic mice, the number of CD4 + T-cells increased compared with non-transgenic littermates. BrdU incorporation was increased only in CD4+ T-lymphocytes in vivo, but not in CD8+T-lymphocytes and B-lymphocytes. These findings indicate that the HBZ gene expression promotes proliferation of T-lymphocytes in vivo. In HBZ transgenic mice, CD4 positive T-lymphocytes infiltrated into skin and lung, indicating that HBZ expression confers infiltrative phenotypes to expressing T-cells. In addition, 14 HBZ transgenic mice developed T-cell lymphomas after one year among 36 transgenic mice, while only 2 non-transgenic littermates had T-cell lymphomas. In these lymphomas, we showed monoclonality of lymphoma cells. Immunohistochemical analyses revealed that these lymphoma cells were CD3+, CD4+, and B220−, indicanting that these were CD4+ T-cell lymphomas, In some transgenic mice, lymphoma cells infiltrated into lung, liver and bone marrow.

These results show that the HBZ gene is oncogenic as well as the tax gene. The HBZ gene is expressed in all ATL cases while the tax gene is expressed in about 40% of cases, indicating that HBZ gene is essential to maintain leukemic state of ATL. We will discuss the molecular mechanism how the HBZ gene transforms CD4+ T-cells in vivo.

OP-17 Functional Evaluation of Monocyte-Derived Dendritic Cells from Patients with Chronic Type of Adult T Cell Leukemia

Adult T cell leukemia (ATL) is an aggressive, poor-prognostic CD4+ T cell malignancy caused by Human T-cell leukemia virus type 1 (HTLV-1). Cytotoxic T lymphocytes (CTLs), especially those for Tax, are considered to be essential to control the outgrowth of HTLV-1-infected cells as seen in many other viral infections. However, It has been reported that ATL patients are profoundly immunocompromised and some asymptomatic carriers also show a weak HTLV-1-specific CTL response, indicating that impairment of HTLV-1-specific CTL responses might be one of the risk factors for disease progression. Therefore, vaccines to restore optimal HTLV-1-specific CTL responses would be required to control the expansion of HTLV-1-infected or ATL cells and to prevent disease progression. For this purpose, dendritic cells (DCs) are the most attractive candidates since they are effective in stimulating antigen-specific naïve and memory T cells. In a number of clinical studies, antigen-loaded DC-based vaccines have been tested in multiple malignancies, such as melanoma and breast cancer, and have been found to induce tumor-specific T cell responses. In ATL, monocyte-derived dendritic cells (MDDCs) generated from patients with acute type of ATL have been reported to be less functional than those from healthy individuals, but the function of MDDCs from patients with chronic type of ATL (cATL) remains obscure. In this study, we examined whether MDDCs could be generated from cATL patients and evaluated their functions such as IL-12 production, antigen uptake, and antigen presentation. Our results showed that it was possible to generate MDDCs from cATL patients and that these MDDCs expressed MHC-I, MHC-II, and co-stimulatory molecules (CD40, CD80, and CD86) and produced abundant of IL-12. Furthermore, immature MDDCs from cATL patients showed the ability to take up antigen and mature MDDCs were able to activate allogeneic CD4+ T lymphocytes, suggesting MDDCs from cATL patients may be a promising tool for restoration of HTLV-1-specific CTL responses.

OP-18 Aberrant Expression of the Transcription Factor Twist in Adult T-Cell Leukemia

Adult T-cell leukemia (ATL) is a mature CD4⁺ T-cell malignancy etiologically associated with human T-cell leukemia virus type 1 (HTLV-1). Despite evidence that Twist, a highly conserved basic helix-loop-helix transcription factor, is a novel oncogene, there are no reports describing Twist expression in ATL. To clarify involvement of Twist expression in ATL, we used RT-PCR and Western blot analyses, and examined Twist expression in a panel of T-cell lines and PBMCs from patients with ATL or healthy volunteers. HTLV-1-infected T-cell lines and primary ATL cells expressed high levels of Twist compared with uninfected T-cell lines and normal PBMCs at both mRNA and protein levels. Immunohistochemical staining of Twist in ATL lymph nodes and skin lesions showed that primary ATL cells were indeed stained strongly positive for Twist. In vitro infection of normal PBMCs with HTLV-1 induced Twist mRNA expression. Induction of the viral protein Tax by Cd²⁺ in JPX-9, a subline of Jurkat human T-cell line carrying Tax under the control of metallothionein promoter, led to upregulation of Twist. A luciferase reporter gene under the control of the Twist promoter was expressed by cotransfection of expression vectors for Tax and HBZ. We demonstrated that Tax-induced Twist expression required the NF-kappaB and CREB signaling pathways. NF-kappaB inhibitors suppressed induction of Twist by Tax in JPX-9 cells and Twist expression in HTLV-1-infected T-cell lines. Mutation analysis of the 5’ flanking sequence of the Twist gene promoter revealed that one of the major elements responsible for the induction by Tax is an NF-kappaB binding site located at −100 to −91 relative to the transcriptional start site. Twist also augmented Tax-activated HTLV-1 transcription. Short interfering RNA against Twist inhibited cell growth of an HTLV-1-infected T-cell line, but not an uninfected T-cell line. Downregulation of Twist expression in an HTLV-1-infected T-cell line inhibited the expression of YB-1, AKT1, AKT2, and Tax, but not HBZ. These results suggest that induction of Twist by a human viral protein Tax directly contributes to the leukemogenesis of ATL.

OP-19 Distinct Genome Profiles of Acute And Lymphoma Type ATLL: An Indication for Two Subtypes in Acute Type ATLL

Adult T-cell leukemia/lymphoma is clinically classified into four subtypes. These include acute, lymphoma, chronic and smoldering subtypes. Acute and lymphoma types are aggressive and fatal with poor prognosis. By means of array CGH with 2304 BAC clones, we previously reported that genomic alteration pattern (genome profile) of these two subtypes are different (Blood 107:4500-4507, 2006). Genomic aberrations are more frequently found in lymphoma type than acute type, and the patterns are also distinct between these two subtypes, suggesting that genetic pathways involved in tumor development for acute and lymphoma types are different. In an attempt to understand the roles of these genetic alterations in ATLL, we examined genome profiles of peripheral T-cell lymphoma-unspecified (PTCL-U), all of which were negative for HTLV-1 (Clin Cancer Res. 15:30-38, 2009). Out of 51 PTCL-U cases, 29 cases showed genomic alterations. Interestingly, the genome profile of these 29 cases is similar to that of lymphoma type ATLL. Histopathological analysis of the PTCL-U with genomic alterations revealed that these had pleomorphic nuclei with frequent CCR4 positive phenotype while PTCL-U without genomic alteration had frequent inflammatory background. Histology of the former is similar to that of lymphoma type ATLL while the latter is similar to angioimmunoblastic T-cell lymphoma. These results suggested that lymphoma type ATLL has common genetic pathways for lymphoma development with some of PTCL-U having genomic alterations. These findings prompted us to analyze acute type ATLL because it is suggested to contain advanced stage patients of lymphoma type ATLL which can be different from acute type with less lymph node involvement (acute leukemic type). Preliminary analysis with array CGH using a paired combination of peripheral blood and lymphoma samples demonstrated that malignant cells in peripheral blood cells had identical genomic alteration patterns with lymph node samples in the same patients, indicating that some of acute type ATLL might be regarded as lymphoma type ATLL. We are currently studying expression analysis in addition to array CGH to further clarify if some of acute type ATLL can be recognized as lymphoma type ATLL. These molecular analyses will lead to new insights into understandings of clinicopathological features and lymphomagenesis of ATLL.

OP-23 SOD1 Ex Vivo Levels Discriminate Between Asymptomatic Infection, HAM/TSP and ATL: A Potential Therapeutic Target in HTLV-1-Associated Pathologies

Background: High dose Vitamin C, IFN-alpha and IFN-beta are currently used in HAM/TSP therapy. Their mechanism of action in vivo remains obscure, but protection against reactive oxygen species (ROS) has been demonstrated for all three drugs in vitro. Since we recently demonstrated increased superoxide dismutase1 (SOD1) expression upon IFN-beta treatment (Khouri et al., J Immunol 2009), we investigated the potential role of SOD1 as a biomarker and possible therapeutic target in HAM/TSP and ATL.

Results: We have quantified (ELISA) SOD1 ex vivo plasma levels of uninfected (3.0 ± 0.4 ng/ml, n = 20) and HTLV-1-infected healthy controls (AS, 24.5 ± 2.4 ng/ml, n = 19), HAM/TSP (3.6 ± 0.7 ng/ml, n = 14) and ATL patients (48.4 ± 8.3 ng/ml, n = 24). SOD1 levels were significantly (p < 0.0001) decreased in HAM/TSP, as compared to AS and ATL. At a cut-off of 8.0 ng/ml, ROC curve analysis allowed discrimination with 100% sensitivity between HAM/TSP and AS (AUC 1.0, p < 0.0001) and 90.9% sensitivity between HAM/TSP and ATL (AUC 0.98, p < 0.0001), but not between AS and ATL (AUC 0.60, p = 0.22). In vitro, SOD1 inhibition by the small molecule compound D1 increases superoxide production (measured by flow cytometry using sensitive dihydroethidine and MitoSox probes) and dose-dependently induces apoptosis (quantified by acitve caspase-3 and Hoechst 33432 staining) in HTLV-1-infected MT-4 cells. D1-induced apoptosis (83% subdiploid cells) was almost completely reverted by the antioxidant NAC (17% subdiploid cells), demonstrating a ROS-dependent mechanism of action. On the other hand, NAC treatment had only a partial effect upon IFN-alpha+AZT-induced apoptosis in MT-4 cells (32% decrease in DNA degradation and 23% decrease in caspase-3 activation). Interestingly, SOD1-overexpressing HTLV-1-infected cell lines (MT-2 and C8166) are resistant to IFN-alpha+AZT in vitro (10% subdiploid cells), but became sensitive to IFN-alpha+AZT-induced apoptosis upon pharmacological inhibition of SOD1 (42.4% subdiploid cells).

Conclusions: 1. SOD1 is a candidate biomarker in HAM/TSP, its decreased ex vivo levels might explain the therapeutic success of antioxidants, such as Vitamin C in HAM/TSP. 2. Increased SOD1 might represent a novel therapeutic target in ATL, since in vitro SOD1 inhibition induces ROS-dependent apoptosis in HTLV-1-infected cell lines.

OP-26 Ocular Manifestations in HTLV-1 Patients and Its Correlation with Proviral Load

Purpose: To demonstrate ocular manifestations in HTLV-1 asymptomatic carriers (AC) and HTLV-1 associated myelopathy/Tropical spastic paraparesis (HAM/TSP) patients and its correlation with proviral load (PL).

Design: Cohort study.

Methods: From January 2007 to December 2007, 94 HTLV-1-infected individuals were submitted to neurological and ophthalmological assessment at the Clinical Research Laboratory on Neuroinfection and Ophthalmology Department (FIOCRUZ, Rio de Janeiro). Patients were classified as AC, HAM/TSP, other neurologic abnormalities (ONA), and other systemic symptoms. HTLV-1 PL in peripheral blood leucocytes was measured by real-time PCR assay (SmartCycle; Cepheid) using the TaqMan system (Applied Biosystems).

Results: 45 had HAM/TSP, 38 were AC, 3 had ONA, 4 had isolated neurogenic bladder dysfunction, 2 had dermatitis, and 2 had fibromyalgia and pulmonary fibrosis. Some ocular changes associated with HTLV-1 infection were diagnosed in 24 out of 94 patients: 18 patients had corneal abnormalities, 4 had uveitis, and 2 patients had other changes (Adie pupil in one and optic neuritis in other). Considering only patients with ocular changes, 19 had HAM/TSP, 3 were AC, 1 had isolated neurogenic bladder dysfunction (INBD), and 1 had fibromyalgia with pulmonary fibrosis. Median HTLV-1 PL was 13.25 copies/100 leukocytes in patients with some ocular changes (ranging from 4.34 to 39.92), significantly higher than in patients without ophthalmic abnormalities (median 4.76 copies/100 leukocytes, ranging from 0.00 to 39.97; p < 0, 0001 – Mann-Whitney test). Median PL among AC with ocular changes was 14.83, significantly higher than among AC without ocular changes (3.22 copies/100 leukocytes; p < 0.0001). Also, HAM/TSP patients with ocular changes had PL higher than HAM/TSP without ocular changes (12.87 vs 5.45 copies/100 leukocytes; p < 0.005).

Conclusion: Corneal abnormalities were the most frequent ocular changes in patients with HTLV-1 infection in this Brazilian cohort study. As in neurological diseases, patients with ocular changes due to HTLV-1 infection had higher PL than patients without ocular changes.

OP-28 Time to Disability in Patients with Rapidly Progressive HTLV-1-Associated Myelopathy/Tropical Spastic Paraparesis (HAM/TSP)

Background: Description of rapidly progressive HAM/TSP bares new insights among the HTLV-1 pathogenesis. We aim to the further understanding of this infrequent HAM/TSP presentation to shared light over the possibility of having a therapeutic window in the early stages of this disease.

Methods: We reviewed clinical records of rapidly progressive HAM/TSP patients from the HTLV-1 cohort at the Institute of Tropical Medicine Alexander von Humboldt in Lima, Peru. Rapidly progressive HAM/TSP was defined as the inability to walk unaided within two years after onset of lower limb symptoms. We determined the time (months), from symptoms' onset to (1) reach a disability requiring bilateral aid or (2) an Expanded Disability Status Score (EDSS) greater or equal to 6.5 (outcome). The Kaplan-Meier method was used to estimate the mean time to reach the outcome. We used the log-rank test to explore the effect of co-variables on the risk of developing the outcome. Continuous variables were tested with the Mann-Whitney U-test. We measured the HTLV-1 proviral load (PVL) with an in-house SYBR-Green-based real-time quantitative PCR.

Results: Twenty-eight patients with rapidly progressive HAM/TSP were identified in 2001–2009; Fifteen of 28 (54%) were women; median age at onset was 46 years (range 9–67). Fifteen of 28 patients (54%) needed bilateral aid (EDSS greater or equal to 6.5) within 846 person-months of follow up (incidence rate = 0,02); the median time from onset to outcome was 16 months (range 4–138). The mean time to reach a 25% risk of developing the outcome was 14 months, and for a 50% risk 27 months. The median maximum EDSS reached was 6.5 (range 4–10). Lower limb weakness was the most frequent symptom at disease onset in 12/29 (41%). Twelve of 28 patients (43%) had received empirical treatment with antispasmodics, antiretrovirals (AZT or 3TC), prednisone, or a combination of these. Univariate analyses showed a statistically significant association with PVL and empirical treatment. The median PVL (HTLV-1 copy number/10⁴ PBMCs) was higher in patients who developed the outcome (3398 vs. 1489; p = 0.02). We found no significant associations with sex, ethnic origin, transfusion history, first symptom, or age at disease onset. The statistical significance for empirical treatment and PVL was lost after cox regression multivariable analysis.

Conclusions: Rapidly progressive HAM/TSP patients have a 25% risk of requiring bilateral aid by the 14^th month of disease onset and a 50% risk by the 27^th month. Lack of empirical treatment and higher PVL were associated with a faster progression towards severe disability. However, this association was lost after adjustment for co-variables.

OP-29 HTLV-1 Induces TLR-Dependent Immune Response by Plasmacytoid Dendritic Cells

Background: Recent research demonstrated that cell free HTLV-1 virions infect plasmacytoid Dendritic Cells (pDC) and consequently transfer virus to T cells (Jones et al. 2008). Furthermore, HTLV-1 initial binding molecule (BDCA-4) (Ghez et al. 2006) is a specific marker mainly expressed by pDC (Tordjman et al. 2002). However, the major role of innate immune pDC is to secrete massive levels of anti-viral IFN-? cytokine after virus exposure (Colonna et al. 2004) that remains unknown concerning HTLV-1.

Methodology/Principal Findings: We performed a cell free HTLV-1 purification method and stimulated freshly isolated pDC from blood donors. We show that purified HTLV-1 virions generated pDC innate immune response by producing massive levels of IFN-?, which were inhibited by neutralizing HTLV- 1 antibodies (HTLV-1-patients' serum or anti-envelope). HTLV-1 induced costimulatory molecules and rapid expression of apoptotic ligand TRAIL, and downregulated chemokine receptors. Furthermore HTLV-1-stimulated pDC induced apoptosis of CD4 + T cells expressing DR5, this killer activity characterized Interferon-producing Killer pDC (IKpDC). Thus, we investigated the mechanism by which HTLV-1 turned pDC into IKpDC. We observed that an endosomal acidification inhibitor (chloroquine) and a TLR7 specific blocker (A151) drastically inhibited pDC response to HTLV-1. Three-dimensional (3D) microscopy analysis revealed that resting pDC were “dormant” IKpDC with high levels of intracellular TRAIL that could be rapidly mobilized at the surface in response to TLR7 activation. Inhibition of viral degradation in endosomes by chloroquine maintained viral integrity allowing virus detection in pDC by 3D microscopy. In chloroquine treated pDC, intact particles were unable to stimulate the TLR7 pathway and consequently preventing TRAIL relocalization and IKpDC generation.

Conclusion/Significance: We establish here that pDC respond to cell free HTLV-1 by producing high levels of IFN-? and by mobilizing TRAIL on cell surface after TLR7 triggering. This is the first demonstration of pDC innate immune response to HTLV-1 retrovirus. Furthermore, HTLV-1 virions transform pDC into functional killer cells (IKpDC). This characterization of the new IKpDC subset opens a new field in HTLV-1 investigation.

Supported by the ANRS GRANT, Paris, France.

OP-31 The Interferon/Arsenic Combination Eradicates Leukaemia-Initiating Cells in Tax-Driven Murine Adult T Cell Leukemia

Adult T cell leukemia (ATL) is a fatal disease that complicates chronic HTLV-I infection. High response rates were achieved with the combination of zidovudine and interferon-a (IFN). However, a majority of patients eventually relapse, which underlines the need for new therapeutic approaches. The viral oncoprotein Tax plays a critical role in initiating the leukemic process, because Tax transgenics develop a disease with striking ATL features. Hence, Tax represents an ideal target for targeted therapy of ATL. In HTLV-I-infected human ATL cell-lines ex vivo, we previously showed that arsenic trioxide shuts off the constitutive NF-kB activation and potentiates the apoptotic effects of IFN at least in part through proteasomal degradation of TAX. To assess any in vivo efficacy of IFN/arsenic, we established an ATL transplantation model where ATL spleen cells from TAX transgenic mice are inoculated to SCID mice. Recipients developed murine ATL with massive hyperleucocytosis, splenomegaly, hypercalcemia, multiple organ invasion, and constitutive NF-kB activation, as the original transgenics. We then investigated the effect of the Arsenic/IFN combination on the survival of these ATL SCID mice. We demonstrate that the combination of Arsenic and IFN, synergistically cures murine ATL through leukemia initiating cell (LIC) exhaustion, rather than immediate growth inhibition or apoptosis. Indeed, the disease initially thrives under therapy and only vanishes 3 weeks later. A brief Arsenic/IFN treatment that does not change ATL cell abundance or viability, results in a dramatic loss of the ability of ATL cells to transplant in secondary recipients. Strikingly, the addition of the proteasome inhibitor to As/IFN protects LIC and surprisingly results in a dramatic increase in circulating malignant cells. Our promising results open new perspectives for ATL cure through specific targeting of leukaemia stem cells that are oncogene addicted, potentially leading to ATL cure and not only long term disease control.

OP-33 Inhibition of the SDF-1F ¿-CXCR4 Axis by the CXCR4 Antagonist AMD3100 Suppresses the Migration of Human ATLL Cells and Murine Lymphoblastoid Cells from HTLV-I Tax Transgenic Mice

Adult T cell Leukemia/lymphoma (ATLL) is a T-cell malignancy caused by human T-lymphotropic virus type I (HTLV-I), and presents as an aggressive leukemia with characteristic widespread leukemic cell infiltration into visceral organs and skin. The molecular mechanisms associated with leukemic cell infiltration are poorly understood. We have recently established a mouse model of ATLL by generating HTLV-I tax transgenic mice using the lck proximal promoter to restrict transgene expression to developing thymocytes. The mice developed leukemia and diffuse large cell lymphomas histologically identical to those in ATLL with extensive peri-vascular infiltration of many organs. Transfer of splenic lymphomatous cells from transgenic to SCID mice rapidly reproduces a leukemia and lymphoma which is histologically identical to human disease. We have characterized the expression of chemokines and their receptors associated with cellular invasion in this model. It could be shown that lymphomatous cells exhibit specific chemotactic activity in response to SDF-1ƒ ¿. We also observed that CXCR4, the specific receptor of SDF-1 ƒ¿ is expressed on the surface of the lymphomatous cells. Chemotaxis of the cells in the presence of SDF-1 ƒ¿ was associated with rapid phosphorylation of intracellular ERK1/2. AMD3100, a specific CXCR4 antagonist, was found to markedly inhibit both SDF-1 ƒ¿-induced migration and phosphorylation of ERK1/2. Investigation of primary human ATLL cells revealed identical findings. SDF-1ƒ¿- immunostaining demonstrated positive signals in epithelial cells of regenerative hepatic ducts surrounded by leukemic cell infiltration. These results suggest that expression of SDF-1 ƒ¿ might be involved in leukemic cell invasion and that CXCR4 antagonists could be important in combination therapy of ATLL.

OP-34 HTLV-II Establishes a Persistent Infection in Rhesus Macaques and Does Not Exacerbate SIV Infection

The human T cell leukemia virus-II (HTLV-II) is a T cell tropic retrovirus that establishes a chronic infection in most individuals with low levels of virus replication. HTLV-II has not been clearly defined as the etiologic agent of any human disease. Indeed, co-infection with HTLV-II and HIV may have a potential benefit on HIV disease progression by reducing viral replication and modulating chemokine secretion. This study establishes an animal model of HTLV-II infection and evaluates the effects of HTLV-II preinfection on viral and T cell dynamics during the acute and chronic phases of SIV infection. Our findings indicate that HTLV-II replicates in the gastrointestinal tract of rhesus macaques and stimulates humoral and cellular immune responses. Macaques dually infected with HTLV-II and SIV demonstrate similar levels of SIV viral replication, similar rates of mucosal and peripheral CD4+ T cell loss, and increased T cell proliferation. Additionally, neither the magnitude nor the functional capacity of the SIV- specific T cell mediated immune response was significantly different in HTLV-II/SIV co-infected animals when compared to SIV singly infected animals. Thus we demonstrate no exacerbation or protection from disease in HTLV-II/SIV co-infected macaques. Furthermore, we believe the tropism and compartmentalization of HTLV-II replication warrants its evaluation as a potential candidate for an HIV vaccine vector. Candidate HIV vaccines that demonstrate low-level persistent antigenic stimulation at mucosal sites such as the gastrointestinal tract, may protect immunized hosts from HIV infection.

OP-35 Bovine Leukemia Virus (BLV) Could be Subdivided into up to Seven Groups Based on Phylogenetic and Similarity Analyses of Sequences from the ENV Gene

We sequenced the env gene from 28 BLV field strains from Argentina, and combined these sequences with 45 env sequences representing all the genetic groupings identified previously. These data were analyzed by phylogenetic analysis and similarity comparisons. Phylogenetic analyses showed the existence of five groups of sequences or genotypes. Four genotypes were supported by high bootstraps (89–100) and posterior probabilities (p = 1.0), and the fifth one was supported by a high posterior probability (p = 1.0) and moderate bootstrap (80–96) values. All these genotypes have one or more counterparts among the groupings identified previously, and thus many groups that were originally named differently correspond actually to the same genotype. Two sequences did not cluster with any other sequence, and were highly divergent when compared to the five groups mentioned above. This indicates that two extra genotypes could exist. The similarity comparisons were highly congruent with the phylogenetic analyses. The mean number of pair wise nucleotide substitutions between sequences from the same genotype was 11.1 (min. 6.5; max. 19), whereas it was 39.6 (min. 36.8; max. 47.9) between sequences from different genotypes. Likewise, the mean number of amino acid substitutions in pair wise comparisons at the intra-genotype level was 3.3 (min. 0.4; max. 6.2), whereas it was 9.6 (min. 8.3; max. 13.5) at the inter-genotype level. Our analyses confirmed the existence of geographic clusters (p = 6.838E-15, Chi2 test; p = 4.583E-12, Fisher's exact test).

OP-36 A Combination of Reverse Transcriptase and Histone Deacetylase Inhibitors Promotes an STLV-1 Specific Cellular Response and Depletes STLV-1 Latent Infection in Vivo

Since a nice correlation exists between the clinical status of the carrier and the HTLV-1 proviral load (PVL) in the peripheral mononuclear cells, we hypothesized that diminishing the PVL in HTLV-1 asymptomatic carriers would prevent later occurrence of a disease. In vivo, most infected cells contain transcriptionnally silent proviruses and the apparent viral shut-off seems to play an important role in the viral escape from the host immune response. A transient transcriptional activation of the viral reservoir could therefore allow a reduction of the PVL in vivo by prompting a cellular response towards the infected cells. We conducted a study combining or not valproate (VPA), an inhibitor of histone deacetylases (HDACi), and AZT, an inhibitor of the reverse transcriptase, in a series of 20 STLV-1 naturally infected asymptomatic P. papio whose proviral load was equivalent to that of asymptomatic HTLV-1 individuals. Treatment with VPA or AZT alone had no effect on the PVL. On the contrary, the combination of drugs induced a strong and significant decrease of the PVL. In the VPA/AZT treated group, the PVL decline was associated with an increased number of lymphocytes, which was not linked to any obvious alteration in CD4+ T-cell subpopulations. For this reason, CD8+ naïve, central memory and effector cell populations were analyzed in the VPA/AZT group of animals and were compared to those of the other groups. The number of naïve and central memory CD8+ T-cells was constant over time in all groups of animals. However, animals treated only with VPA had a global decrease in the number of effector CD8+ T-cells, while this number increased for the VPA/AZT treated animals. We wondered if the accumulation of effector memory CD8 T-cells represented indeed an increase of cytotoxic lymphocytes (CTL) directed against STLV-1. The rate of CD8+ cell–mediated lysis of STLV-1-infected cells was therefore determined. We observed a nice correlation between the variation on the lysis rate and the variation on PVL. Thus the PVL decrease is likely to be due to the establishment of an effective anti-STLV-1 immune response that would lyse cells that, in response to VPA, express the virus. Our study indicates that inducing viral expression with HDACi and preventing de novo infection with AZT is a potential therapy for decreasing PVL in asymptomatic carriers at risk for developing a disease.

OP-37 Virus-Infected IFN-Gamma Producing CD4+CD25+CCR4+Foxp3^low T-Cells Are Abnormally Increased and Proinflammatory in HAM/TSP

The human T-lymphotropic virus type 1 (HTLV-1) is a human retrovirus that is associated with both a chronic neuroinflammatory disease termed HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP) and adult T-cell leukemia (ATL). The pathogenesis of HAM/TSP is known as HTLV-1 infected T-cells triggered hyper immune response, which leads to neuroinflammation. However, it has never been demonstrated which T-cell subset in HTLV-1 infected T-cells play major role in the accelerated immune response. Here, we demonstrate that CD4⁺CD25⁺CCR4⁺ T-cells are the predominant viral reservoir in HAM/TSP patients. Notably, we demonstrate that in healthy individuals, CD4⁺CD25⁺CCR4⁺Foxp3^low T-cells produce multiple proinflammatory cytokines such as IL-2, IL-17, IFN-gamma while in HAM/TSP patients, IFN-gamma production is increased and IL-17 production is decreased. Importantly, the frequency of IFN-gamma producing CD4⁺CD25⁺CCR4⁺Foxp3^low T-cells is dramatically increased in HAM/TSP patients, and is correlated with disease activity and severity. Thus, we demonstrate that virus infected abnormally increased IFN-gamma producing CD4⁺CD25⁺CCR4⁺Foxp3^low T-cells are important for the pathogenesis of this retrovirus associated inflammatory disorder of the central nervous system.

OP-38 The Plasma Level of CXCL10 and Soluble IL-2 Receptor Are Useful Biomarkers and Decreased in HAM/TSP Patients with the Long-Term Corticosteroid Therapy

HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP) is a chronic inflammatory disease of the central nervous system. Advances in our understanding of the pathogenesis of HAM/TSP have revealed that patients with HAM/TSP have high frequencies of HTLV-1–infected T-cells, high cellular and humoral immune responses, and increased proinflammatory cytokines and chemokines production. However, among these characteristics, the identification and validation of relevant HAM/TSP biomarkers have been insufficient.

In this study, we measured the concentrations of interferon-ã, IL-17, soluble IL-2 receptor (sIL-2R), CXCR3 ligands (CXCL10, MIG, I-TAC), CCR5 ligands (RANTES, MIP-1á, MIP-1â) in plasma from 24 asymptomatic carriers (ACs) and 21 HAM/TSP patients. The HTLV-1 provirus DNA load in peripheral blood mononuclear cells (PBMCs) from ACs and HAM/TSP patients was analyzed using real time PCR method. We also measured the concentration of neopterin in cerebrospinal fluid (CSF) of HAM/TSP patients. Then, the level of these markers was initially evaluated for the sensitivity and specificity to discriminate HAM/TSP from ACs. Next, in HAM/TSP patients, we analyzed the correlation between the level of these markers and disease severity. Based on these analyses, we identified CXCL10 and sIL-2R in plasma and neopterin in CSF as relevant biomarkers of HAM/TSP disease activity.

Next, to evaluate whether the long-term corticosteroid therapy improve the disease activity and prognosis of HAM/TSP patients, we compared retrospectively the level of CXCL10 and sIL-2R in plasma, the level of neopterin in CSF and disease severity between HAM/TSP patients with or without long term corticosteroid therapy. Importantly, the levels of these three markers and disease severity were significantly lower in HAM/TSP patients with long-term corticosteroid therapy. These results suggest that the plasma level of CXCL10 and soluble IL-2 receptor are clinically useful biomarkers and long-term corticosteroid therapy modify the prognosis of HAM/TSP by the remission of inflammation in the lesions.

OP-39 Mechanisms of Action of IFN-Alpha in ATL and HAM/TSP Patients: In Vitro Pro-Apototic and Antiproliferative But Not Antiviral Activity Predict in Vivo Outcome

Background: IFN-alpha+AZT is currently the major treatment option in ATL, but therapeutic success is limited to non-lymphoma subtypes. On the other hand, IFN-alpha monotherapy (but not AZT or antiretrovirals) has shown clinical benefit in HAM/TSP. However, the mechanism of action of IFN-alpha is largely unknown and no biomarkers for survival and/or therapeutic monitoring are currently available for clinical use in ATL or HAM/TSP.

Methods: We have quantified pro-apoptotic, antiproliferative, and antiviral activity (p19 secretion) of IFN-alpha in vitro in primary cells (PBMCs) from 30 ATL and 25 HAM/TSP patients. Molecular and clinical data (2–5 years of follow-up) were subjected to univariate analysis, Bayesian network learning and classification modeling. For ATL patients, FAS-670 genetic polymorphism was included in the analysis and survival was dichotomized as short-term (<2 years) or long-term (>2 years). For HAM/TSP, both Kurtzke's and Osama's disability scales were used.

Results: IFN-alpha exerted a significant (p = 0.0076) antiviral effect in all ATL patients, including the lymphoma subtype, demonstrating functional IFN signaling. However, its effect upon lymphoproliferation in vitro was significantly associated with clinical forms (p = 0.0077), being absent in lymphoma patients, suggesting therapeutic failure might be linked to resistance to the antiproliferative but not antiviral effect of IFN-alpha. The significant (p = 0.044) pro-apoptotic effect of IFN-alpha was found to be critically dependent upon FAS-670 genotype, which was confirmed in an independent set of 20 healthy controls. Multivariate Bayesian network analysis confirmed the association between in vitro apoptosis, FAS genotype and survival. In addition, classification modeling allowed the elaboration of a Bayesian algorithm that reliably predicted short- or long-term survival of ATL patients. In HAM/TSP patients, lymphoproliferation was significantly correlated to disease status (EDSS, p = 0.034) but the antiproliferative effect of IFN-alpha was only significant in patients with short disease duration (<3 years, p = 0.017). In contrast to ATL, no significant pro-apoptotic activity was observed in HAM/TSP patients.

Conclusions: 1. Survival of ATL patients can be predicted based upon in vitro data available at diagnosis. 2. Pro-apoptotic and antiproliferative activity probably decide therapeutic success of drugs/drug combinations in both ATL and HAM/TSP.

OP-40 Minocycline Modulates Antigen-Specific CTL Activity through Inactivation of Mononuclear Phagocytes in Patients with HAM/TSP

The activation of mononuclear phagocytes (MPs), including monocytes, macrophages and dendritic cells, contribute to central nervous system inflammation in various neurological diseases. In HTLV-I-associated myelopathy/tropical spastic paraparesis (HAM/TSP), MPs have been shown to be reservoirs of HTLV-I in addition to virus infected CD4+ and CD8+ T cells. HTLV-I-infected or activated MPs induce proinflammatory cytokines and dysregulation of CD8+ T cell function mediated by viral infection with concomitant expression of IL-15. Here we demonstrate increased production of TNF-alpha and IL-1 beta in CD14+ cells of patients with HAM/TSP after short term in vitro culture compared to healthy normal donors. In addition, we have examined the inhibitory effects of minocycline, a known inhibitor of activated MPs, on PBMCs of patients with HAM/TSP. Our results demonstrated that in patients with HAM/TSP, minocycline effectively inhibited IL-1 beta release from PBMCs and TNF-alpha expression in CD14+ monocytes cells but not in CD4+ T cells, after culture for 24 hours. These results showed that minocycline inhibited MP-specific immune responses in HAM/TSP patients. Although treatment with minocycline did not directly inhibit HTLV-I Tax expression in either CD4+ T cells or CD14+ monocytes, interestingly, minocycline significantly inhibited spontaneous degranulation and IFN-gamma expression in CD8+ T cells of HAM/TSP patients after 24 hours culture. Moreover, Tax11–19 specific CD8+ T cells showed inhibition of IFN-gamma expression but not degranulation after treatment with minocycline. These results suggest that minocycline may have differential effects on CD8+ CTL function. While minocycline did not inhibit the lytic function of CD8+ CTL it can modulate cytokine expression. The downmodulation of IFN-gamma expression in CD8+ T cell by treatment with minocycline was also associated with a downregulation of HLA class I expression on CD14+ cells, suggesting that immunological interactions between CD14+ monocytes (targets) and CD8+ T cells (effectors) was reduced. These results demonstrate that minocycline in HAM/TSP can interfere with a number of MP functions including IL-1 beta, TNF-alpha, and down regulation of HLA class I, which can also modulate CD8+T cells function. Thus, activated MP may be important therapeutic target in HAM/TSP.

OP-44 Increased HTLV-1 Proviral Load and T Cell Proportion are Associated with HAM/TSP Evolution, According a New Proposal for Diagnostic Criteria of HAM/TSP

Introduction: A high HTLV proviral load is associated with development of disease, particularly HAM/TSP. However, HAM/TSP is a syndrome that can include individuals with different clinical stages.

Objective: To evaluate the relationship between proviral load and the proportion of CD4⁺ and CD8⁺ T cells in HTLV-1-infected groups classified following neurological findings as possible-Ps, probable-Pb and definite-D.

Methods: 302 HTLV-1-infected individuals followed at The Reference Center of HTLV cohort in Salvador-Bahia were included in this study: 199 healthy carries-HC and 103 TSP/HAM classified as: 22-Ps, 23-Pb, and 58-D, in Salvador/Bahia-Brazil. HTLV-1 proviral load was quantified using the real-time TaqMan PCR method and the CD4 and CD8 T cell subsets by flow cytometry.

Results: The male:female ratio was 2:1. The median age was lower in the HC (41yr) group than in HAM/TSP patients: Ps = 50yr, Pb = 47yr and D = 55yr. There was a statistical difference between the age from Hc vs D (p < 0.001), and Pb vs D patients (p = 0.03). The median CD4+T lymphocytes was higher in the HAM/TSP -PB (720 cell/mm3) and -D (774 cell/mm3), compared with HC (570 cell/mm3) (p = 0.02, p = 0.006, respectively). The number of CD8 T cell was increased only in the HAM/TSP-Pb-patients (529 cell/mm3), compared with the HC (354 cell/mm3)(p = 0.01). The median of HTLV-1 proviral load (copies/10⁶ PBMC) in HC (log-4.0) and in TSP/HAM -Ps (log-4.5) were lower than in Pb (log-5.0) and D (log-5.5) patients (p < 0.0001). In addition, there was a positive correlation between HTLV proviral load and age (R = 0.15 p = 0.036), proportion of T CD4+lymphocytes (R = 0.2, p = 0.003) and the Osame's motor disability graded scale (R = 0.3, p = 0.03).

Conclusion: Proviral load and levels of CD4⁺ and CD8 T are higher in the HTLV-infected groups with neurological symptoms and could be useful as a biological marker of disease progression and to estimate the time point to introduce the therapy.

OP-45 Ciclosporin for the Treatment of Patients with Early or Progressive HAM: 24-Week Data from an Open, Pilot Study

Background: Disease modifying treatments for HTLV-1-associated myelopathy are urgently required. Many therapies have been tried but few have shown any clinical benefit. Responses to interferon-? and prednisolone have been reported but are frequently transient and side-effects are problematic. Although the pathogenesis of HAM is not fully understood there is evidence that cytokines released by activated lymphocytes cause spinal cord by-stander damage. We therefore sought to block T-cell activation with ciclosporin.

Methods: Patients were recruited into an open, pilot 48 week study if they had motor symptoms due to HAM for less than 2 years (early disease) or there was objective evidence of deterioration in patients with disease of longer duration (progressive). Two baseline clinical assessments were made along with MR imaging of the neuraxis and CSF examination. Ciclosporin was prescribed at a dose of 2.5–5mg/kg/day divided into 2 doses with the dose adjusted according to trough plasma ciclosporin concentrations. Clinical, viral and immunological assessments were made every 4 weeks to week 24 and then every 8 weeks to week 72. Treatment was stopped after 48 weeks.

Results: 7 patients were enrolled, 5 females and 5 were Afro-caribbean. Mean age was 51.7 years, median duration of symptoms 4.6 years including 2 patients ‘early’ HAM. 6 patients had spasticity and used a walking aid, the mean time to walk 10 meters (10mTW) was 37 seconds. All patients had nocturia except the two using urinary catheters. 6 patients complained of pain with a median visual analogue pain score of 8.4. Baseline HTLV-I viral load in PBMCs was 15.9 copies/100 PBMCs (%) and 95.5% in CSF.

Data to 24 weeks are presented: 5 patients completed 24 weeks treatment. 2 withdraw 1 due to depression and a urinary tract infection with sepsis requiring IV antibiotic and 1 due to headache and tremor. A 3rd patient with bronchiectasis was admitted for treatment of a lower respiratory tract infection and continued study treatment. In an intention to treat analysis at 6 months, compared with baseline 10 m TW improved in 7/7 patients by on average 18%, spasticity was reduced in 4 and unchanged in 3, pain improved in 5/6.

HTLV-I viral load was reduced by 22% in PBMCs but by 40% in CSF (repeat LP at 12 weeks). Frequency of CD4+CD25+cells decreased by 26%. ? 2 microglobulin concentrations fell in 2/3 patients with raised levels at baseline.

Conclusion: Significant clinical improvement was seen documented in pain, spasticity and timed walk. Two patients discontinued therapy due to side-effects. Reductions in viral load were more marked in CSF than peripheral and markers of T-cell activation where raised at baseline tended to normalise. Further studies of ciclosporin for the treatment of HAM are warranted.

OP-47 What Can Walking Related Outcome Measures Tell Us about Patients with HAM

Background: Patients with Human T-lymphotrophic virus type 1( HTLV1)-associated myelopathy (HAM) have lifelong disability, with restriction of their quality of life and activities of daily living. Treatment options are limited. Clinical trials of new therapy require a robust, clinically useful tool to measure their efficacy with meaningful implication for patient function. Disability scales are insensitive to slow progressive deterioration or improvement.

Aim: To compare 3 walking related outcome measures: the 10 metre timed walk (10mTW); Timed Up and Go (TUAG) & MSWS 12 in patients with HAM.

Methods: A repeated study design. A single observer applied the 3 outcome measures to 8 patients who attended the HTLV outpatient clinic on seven successive occasions over 18 months. Pain was assessed with a visual analogue scale.

Results: The 10mTW correlated well with TUAG (R²-0.99). MSWS 12 scores varied widely & correlated poorly with TUAG(R²-0.24) and 10mTW (R²-0.40). None of the outcome measures showed a clinically significant change (p = < 0.05). Pain scores were consistent over time and pain did not correlate with TUAG.

Discussion: Robust, clinically useful & sensitive measurement tools for patients with HAM need to take into account the varying factors that may influence the patient's functional ability. In this cohort TUAG & 10mTW were stable over 18 months but were sensitive to fluctuations in the patients' condition whereas MSWS12 reflected better patients' perceptions of their disability. TUAG unmasked areas of motor disability (rising from the chair). Surprisingly, pain did not correlate with changes in mobility, however other potential contributing factors include the influence of spasticity on mobility. We conclude that movement kinematics should be utilised in the assessment of these patients.

OP-48 Interaction of Cytotoxic T Lymphocytes and Monocytes: Role in HTLV-I Associated Neurologic Disease

The pathogenesis of human T lymphotropic virus type I (HTLV-I)-associated myelopathy/tropical spasticparaparesis (HAM/TSP) is thought to involve the interaction between HTLV-I-specific CD8+ cytotoxic T lymphocyte (CTL), which are known to be elevated in the periphery and cerebrospinal fluid (CSF) of patients with neurologic disease, and HTLV-I-infected CD4 positive T lymphocytes. However, within the central nervous system (CNS), inflammatory CD4 positive T lymphocytes and HTLV-I proviral loads gradually decrease with duration of illness while the symptoms of the disease progress. We wondered what cells are targets of potentially immunopathogenic infiltrating CD8+ CTL in the CNS of the patients with HAM/TSP considering that CD4+ lymphocytes decrease with HAM/TSP disease duration. Recently we have demonstrated that CD14+ monocytes are infected with HTLV-I and that activated monocytes influence CD8+ T cell activity. We hypothesized that the CTLs interaction with not only HTLV-I-infected CD4+ lymphocyte but also infected mononuclear phagocyte (MP) such as monocytes and macrophages may play a role in the pathogenesis of HAM/TSP. We have analyzed monocyte populations and their interactions in cultured peripheral blood mononuclear cells (PBMCs) of patients with HAM/TSP and normal donors (NDs). Flow cytometry analysis revealed a significant decrease of CD14+ cells after culture in the PBMC of patients with HAM/TSP compared to NDs. In addition, we have characterized the target cells of these CD8+ CTL by fluorescence microscopy in PBMC of patients with HAM/TSP and NDs. Although the interactions between CTL and CD4+ or CD14+ cells in the PBMC of NDs were detected as frequently as expected with flow cytometry, CTL interactions with CD14+ cells in PBMCs of HAM/TSP patients were more frequently detected than expected. Strikingly, CTL interactions with CD14+ cell were more frequently observed by microscopy relative to the total number of CD14+ in the periphery than CTL interactions with CD4+ cells (relative to the total number of CD4+ cells). This suggests a greater affinity of CD8+ CTL for CD14+ monocytes compared with CD4+ T cells. Three-dimensional analysis of microscopic images demonstrated that HTLV-I Tax could be transferred from virus-infected CD4+ T cells to CD14+ monocytes suggesting a possible mechanism whereby viral antigen can be spread to uninfected cells that can be potential targets for CTL. Lastly, these in vitro observations were extended to in situ studies in which we can demonstrate a number of CD8+ T cells and macrophages in the area of the HAM/TSP spinal cord where CD4+ cells are rarely seen. These results suggest that the CTL interaction with CD14+ cells might play an important role in the pathogenesis of HAM/TSP.

OP-49 Roc Curve Analyses to Estimate the Threshold of HTLV-1 Proviral Load Level as a Risk Marker to HAM/TSP

HTLV-1 is associated with hematological (ATL) and inflammatory (HAM/TSP) diseases, but almost 95% of HTLV-1 carriers remain asymptomatic lifelong. HTLV-1 proviral load (pvl) is a marker of development of diseases, since levels of pvl are significantly higher in symptomatic than in healthy carriers. However, pvl quantified in peripheral blood shows overlapping levels in asymptomatic carriers and HAM/TSP patients, causing doubt about which threshold the pvl level may be considered as a risk marker to development of HAM/TSP.

Objective: Analyse HTLV-1 pvl of asymptomatic carriers (ASY) and HAM/TSP patients (HAM) to determine a cutoff (CO) pvl level using ROC curve analysis.

Methodology: HTLV-1 proviral load was measured in 46 ASY and 59 HAM individuals. Proviral load was quantified by real time PCR using SYBR Green system to amplify pol (number of proviral copy) and albumin (input of cells) genes. Standard curves were done using MT2 DNA. All samples were tested in duplicate. ROC curve was utilized to estimate if there was a suitable CO pvl level to predict HAM/TSP individuals.

Results: The mean of pvl in ASY and HAM was 108/10000 cells and 508/10000 cells, respectively (p < 0.05). The ROC curve area was 0.803 (CI 95%: 0.718–0.888) and std. error was 0.043. The better CO pvl level was defined as 227/10000 cells, with a sensitivity of 78% (true HAM patient) and 1-specificity of 28.3% (false HAM patient). To verify if this CO might be useful to identify individuals with risk to development of HAM/TSP, we analysed the pvl of HTLV-1 carriers that have been followed up by GIPH cohort. From seven individuals who have showed clinical progression from asymptomatic to neurologic signals and symptoms, but not enought to be classified as HAM/TSP, five (71.4%) showed pvl level higher than CO. In other 27 individuals considered oligosymptomatic (hyperrreflexia patelar or Babinski sign or urinary disturbance) since the beginning of follow-up, just eight (29.6%) showed pvl higher than the CO.

Conclusion: The definition of CO values to pvl that could be indicative of actual risk to HAM/TSP is important in the follow-up of asymptomatic carriers. Cohort studies are necessary to support prognostic indicators, and statistical tools may help to better elucidate criteria to laboratorial quantitative data.

Financial support: FAPEMIG, CNPq and Fundação Hemominas.

OP-50 First Line Antiviral Therapy Significantly Improves Survival of Patients with Acute, Chronic and Smoldering ATL: A Worldwide Meta-Analysis

HTLV-I associated adult T cell leukemia/lymphoma (ATL) is an aggressive T cells malignancy, with poor prognosis due to chemotherapy resistance. Multiple small studies using zidovudine (AZT) and interferon-á (IFN) showed response in ATL patients. However, the global impact of this innovative antiviral treatment strategy on long-term survival of ATL patients in routine practice remains undetermined. We realized a worldwide meta-analysis on the use of AZT/IFN treatment. Charts of 254 ATL patients treated from 1994 to 2008 in the USA (59 patients), the UK (13 patients), Martinique (111 patients) and France metropolitaine (67 patients) were individually reviewed. Collected data included epidemiological, clinical and biological characteristics, ATL subtype, first line treatment, response to first line therapy, and overall survival. In 231 patients with available survival data, first line therapy was recorded in 207 patients only. Median age was 50 years (range 16 to 95). According to Shimoyama classification, there were 116 acute ATL, 18 chronic ATL, 11 smoldering ATL, 100 ATL lymphoma, and 9 patients with an unknown subtype. Hypercalcemia was present in 48% of patients. Five year overall survival rates were 20% for 77 patients who received chemotherapy alone, 46% for 75 patients who received first line antiviral therapy alone (p = 0.012), and 12% for 55 patients who received chemotherapy followed by antiviral therapy. Response rate to first line antiviral therapy was 66%, including 35% complete remission (CR). When overall survival analysis was performed by ATL subtype, patients with acute, chronic, and smoldering ATL significantly benefited from first line antiviral therapy, whereas patients with ATL lymphoma had a better outcome with first line chemotherapy. Achievement of CR with first line antiviral therapy resulted in prolonged survival of more than 10 years in 82% of the acute ATL subgroup. Finally, first line antiviral therapy in chronic and smoldering ATL resulted in 100% overall survival at 10 years. Multivariate analysis using Cox regression showed that first line antiviral therapy significantly improves overall survival of ATL patients. In conclusion, these results confirm that treatment of ATL using AZT and IFN results in a high response and CR rates particularly in acute, chronic and smoldering ATL, but not in lymphoma, resulting in impressive prolonged survival and hence should be considered as gold standard first line therapy. Predictive factors of response are currently under investigations in order to improve prognosis of this still uncurable disease.

OP-51 Phase II Study of the Efficacy and Safety of the Combination of Arsenic Trioxide, Interferon Alpha, and Zidovudine in Newly Diagnosed Chronic ATL

HTLV-I associated adult T-cell leukemia/lymphoma (ATL) carries a very poor prognosis due to its intrinsic resistance to chemotherapy. Antiretroviral agents such as zidovudine (AZT) and interferon alpha (IFN) have demonstrated clinical efficacy, but acute ATL still carries a dismal prognosis. Patients with chronic ATL have a relatively better prognosis, but a poor long-term survival when they are managed with a watchful-waiting policy until disease progression or with chemotherapy. In ATL cell-lines, arsenic trioxide shuts off the constitutive NF-?B activation and potentiates the apoptotic effects of IFN through proteasomal degradation of Tax viral oncoprotein. Clinically, As/IFN therapy has shown some efficacy in refractory aggressive ATL patients. These results prompted us to investigate the efficacy and safety of the combination of AZT, IFN and arsenic in 10 newly diagnosed chronic ATL patients from North East Iran. An impressive 100% response rate was observed, including 7 complete remissions (CR), 2 CR but with more than 5% circulating atypical lymphocytes, and one partial response. HTLV-I proviral load significantly decreased from an average of 1415 copy/μl of blood to 226 copy/μl (p < 0.05). Responses were rapid and no relapse was noted. Side effects were moderate and mostly hematological. In conclusion, addition of arsenic to reduced dose of AZT and IFN is feasible and exhibits an impressive response rate with moderate toxicity. Long-term follow up will clarify whether this impressive response rate will translate in disease cure. Overall, these clinical results strengthen the concept of oncogene targeted cancer therapy.

OP-53 A Critical Analysis of Prognostic Factors in Patients with HTLV-1 Adult T-Cell Leukemia: A Multicenter Clinicopathologic Experience and New Prognostic Score

ATLL has 4 recognized subtypes: acute, lymphomatous, chronic, and smoldering and studies have sought to identify additional prognostic factors. We assessed prognostic models in use for aggressive NHL to develop a novel risk stratification scheme. Data were collected between 8/92 and 5/07. Descriptive statistics were used to assess categorical and continuous variables. OS was defined as time from diagnosis to death. Survival curves for OS were estimated using the Kaplan-Meier method. Univariate associations between individual clinical factors and OS were evaluated using the log-rank test for categorical variables and the Cox model for continuous variables. Recursive partitioning analysis was used to develop a new prognostic model. 89 patients with ATLL were identified; 37 males (41.6%) and 52 females (58.4%) and median age 50 years (range 22 to 82). The acute subtype of ATLL predominated (68.5%), followed by lymphomatous (20.2%), chronic (6.8%) and smoldering (4.5%). Median OS for all sub-types was 24 weeks (range 0.9 to 315). According to the IPI, 8 patients (9.1%) were classified as low risk, 11 patients (12.5%) as low intermediate risk, 13 patients (14.8%) as high intermediate risk, and 56 patients (63.6%) as high risk, 1 patient could not be evaluated due to missing data. Median OS by IPI risk group was 271, 65, 31 and 16 weeks, respectively (p < 0.01). The PIT could be determined in 68 patients; 10 patients (14.7%) had a score of 0-1 (group 1), 19 patients (27.9%) had a score of 2 (group 2), 31 patients (45.6%) had a score of 3 (group 3), and 8 patients (11.8%) had a score of 4 (group 4). Median OS by PIT risk group was 61.1, 28, 24, and 11.3 weeks respectively (p < 0.01). A new risk model was developed using the variables of the IPI, PIT and calcium level as it had independent prognostic value. Recursive partitioning of OS based on these variables gave a tree with 5 nodes, which fell into three risk categories: low risk patients with Stage I-II disease and a performance status <2; the medium risk group composed of two sets of patients: (a) those with Stage III-IV disease with an ECOG performance status <2 or (b) those with an ECOG performance status ¡Ý2 with calcium ¡Ü11 mg/dL and age ¡Ü 60; and the high risk group (also comprising 2 sets of patients): (a) those with a performance status ¡Ý 2 with calcium ¡Ü11 mg/dL and age >60 or (b) those with a performance status ¡Ý 2 and calcium >11 mg/dL. There were 10 patients (11.2%) in the low risk (median survival = 156.6 weeks), 31 (34.8%) in the intermediate risk (median survival = 45.4 weeks), and 48 (53.9%) in the high risk (median survival = 13 weeks) categories (p < 0.01). We propose a new prognostic model based on recursive partitioning analysis that successfully identifies three prognostic categories based on performance status, stage, age and calcium level at diagnosis.

OP-56 A Rapid in Vitro Multi-Parametric Drug Screening Assay for Adult T-Cell Leukaemia/Lymphoma

Adult T-cell leukaemia/lymphoma (ATLL) is an aggressive T-cell malignancy of mainly CD4+CD25+ cells, which develops in a minority of HTLV-1-infected individuals. ATLL is resistant to current chemotherapy and prognosis is poor. A significant advance in the treatment of ATLL was obtained with the combination of zidovudine (AZT) and interferon-alpha (IFN-a). However, most of the patients only achieve a partial response and eventually relapse, whereas patients with the lymphoma subtype do not respond. Considering the current inefficient therapeutic strategies, we developed a rapid and sensitive flow cytometry-based in vitro screening assay to identify new lead compounds with a therapeutic potential.

MT-2 and MT-4 were cultured in vitro for 48h under control treatment conditions (no compound; AZT (10μg/ml); IFN-a (1000U/ml); or AZT+IFN-a) and with novel small molecule lead compounds (Q2, Q3). Apoptotic (active-caspase-3, Fas) and proliferation markers (PCNA), DNA content (Hoechst 33342) together with virological (Tax) and immunological markers (CD4, CD25) were quantified by flow cytometry (FACSCanto II) using FACSDiva software. Statistical analysis was performed with GraphPad Prism 5 software.

Flow cytometric analysis showed high levels of CD4, CD25 and Fas for both untreated HTLV-1 infected cell lines, consistent with ex vivo results of ATLL patients (Debatin et al. 1993). Analysis of MT-4 revealed no anti-proliferative nor pro-apoptotic activity for monotherapy with AZT or IFN-a, but a synergistic effect with the combination of AZT+IFN-a (33.4 ± 8.0% increase in active-caspase-3, p < 0.0001; 49.0 ± 10.0% decrease in PCNA, p < 0.0001) was observed, highly similar to the in vivo “responsive” phenotype in ATLL. Moreover, 89.0 ± 1.7% of subdiploid apoptotic cells were observed with AZT+IFN-a treatment as compared to 7.4 ± 1.3% in untreated cells (p < 0.0001). Analysis of MT-2 revealed no effect of AZT, IFN-a nor AZT+IFN-a upon apoptosis and/or proliferation, indicating a resistant “lymphoma-like” phenotype. However, pro-apoptotic activity (subdiploid cells) was reported for the novel candidate compounds Q2 (MT-4: 22.4 ± 5.0%, p = 0.03) and Q3 (MT-2: 26.7 ± 2.4%, p < 0.0001; MT-4: 59.6 ± 8.5%, p < 0.0001) as compared to untreated cells (MT-2: 16.0 ± 1.5%, MT-4: 7.4 ± 1.3%). Cytotoxicity of control and novel compounds was tested in PHA-stimulated PBMCs and uninfected T-cell lines. All compounds revealed either absence of or limited toxicity for uninfected PBMCs and T-cell lines CEM and MOLT-4 (<9% of cell death at the concentrations tested), demonstrating an excellent selectivity.

Conclusions:

We have established a novel six-colour flow cytometric in vitro assay for drug screening in ATLL, which mimics the in vivo synergistic effect of AZT+IFN-a.

OP-57 Multiplex, Real-Time PCR Assay with Molecular Beacon Probes for Simultaneous Detection, Differentiation and Quantification of HTLV Types 1, 2, 3 and 4 in Rural Population of Gabon, Central Africa

A single-tube, multiplex, real-time PCR assay with molecular beacons was established, in which various probes were used for the simultaneous detection, differentiation and quantification of human T-cell leukaemia virus (HTLV) types 1, 2, 3 and 4 and of simian T-cell leukaemia virus (STLV) types 1 and 3. MT-4 (HTLV-1), C19 (HTLV-2) cell lines, a molecular clones of HTLV-3 and the recently discovered HTLV-4 (synthetic constructed gene) were used as standards as well as albumin gene to standardize the copy number. Comparable results for the detection and quantification of HTLV-1 were obtained with our new methods and with other real-time PCR methods described previously. With our new multiplex assay, however, we were able to detect and quantify HTLV-1, -2, -3, and -4 as well as STLV-1 and -3, with an excellent dynamic range of 10(6)–10(0) copies per assay, which the other assays could not do. It will thus be possible to determine a wide range of HTLVs types in both standards and samples, with detection of 1–10 HTLV copies in samples containing at least 100 cells. A large epidemiological study was conducted in our laboratory, with more than 4,000 samples collected from rural central Africa. Therefore, these clinical specimens were first screened using ELISA and positive samples were confirmed with western blotting. HTLVs positive or western blot indeterminate samples were tested with our new multiplex assay. Overall, 328 samples were found to be positive by our multiplex assay including 118 men (mean age 53 years, range from 22 to 75 years old) and 211 women (mean age 51 years, range from 19 to 75 years old). Among them, 99.1% were HTLV-1 positive (n = 325) and only three (0.9%) were HTLV-2 positive. There was no significant correlation between the HTLVs proviral load and sex or age. Nevertheless, high proviral loads were seen in patients over 45 years old. Further clinical and epidemiological studies are needed to evaluate the role of the high proviral load in the associated disease to HTLVs. These studies are now in progress.

OP-58 Development of a Serological Assay for Evidence of XMRV Infection

Xenotropic Murine Leukemia Virus-related Retrovirus (XMRV) is a newly identified retrovirus discovered in prostate cancer tissue using DNA based ViroChip technology. To establish the etiologic role of XMRV infection in prostate cancer and its association with other diseases will require epidemiological studies of large cohorts of cancer patients and control populations. The apparent localization of XMRV to the prostate and cumbersome nature of molecular PCR and RT-PCR assay technologies present a significant challenge to carrying out these studies. Consequently, we are focusing on development of a high throughput serological assay that detects XMRV-specific antibodies.

To identify XMRV-specific serological markers, we experimentally infected three rhesus macaques with XMRV produced from DU145 prostate cancer cells. Serologic reactivity was monitored by Western Blot (WB) using the native viral proteins. The WB results showed that all three rhesus macaques developed antibody responses to envelope and gag proteins, providing direct evidence of XMRV infection in the model close to human. Based on the antibody pattern of XMRV seroconversion in the primate model, we constructed and evaluated several XMRV recombinant proteins derived from the env and gag regions as potential diagnostic agents. Selected recombinant antigens were utilized to develop a double antigen sandwich assay on a high-throughput (200 tests per hour), fully automated ARCHITECT instrumental system. Preliminary evaluation of the prototype XMRV antibody assay performance demonstrated 100% sensitivity by detecting all WB positive serial bleeds (66/66) from the XMRV infected macaques. Specificity of the prototype assay was 99.9% (879/880) on the US blood donor population. Based on these data, the prototype XMRV antibody assay should be an effective test for epidemiological studies of XMRV infection.

OP-59 Biofilm-Like Extracellular Viral Assemblies Mediate HTLV-1 Cell-To-Cell Transmission

Human T cell leukemia virus type 1 (HTLV-1) cell-to-cell transmission in vitro and in vivo requires cell contacts. Although HTLV-1-infected T lymphocytes are prompted to form “virological synapses”, through which HTLV-1 propagates from cell to cell, the detailed mechanism of HTLV-1 cell-to-cell transmission remains poorly understood. We will discuss that HTLV-1-infected T lymphocytes transiently store infectious viral particles in extracellular adhesive assemblies, forming a biofilm-like structure that also contains cellular vesicles and extracellular matrix. These viral assemblies remain adhered to the surface of virus-producing cells. Upon cell contact with non-infected lymphocytes, HTLV-1 assemblies can rapidly adhere and allow fusion of virus to the target cell plasma membrane. Finally, extracellular virus assemblies are important for productive HTLV-1 transmission, since their removal strongly impaired the capacity of HTLV-1-producing cells to infect target cells. Our data unveil a novel mechanism for HTLV-1 transmission through cell contacts involving “viral biofilm-like structures”, that enable efficient virus propagation from cell to cell.

OP-60 A Quantum Dot-Based Antiviral Assay with Potential Utilization in the High Throughput Screening of Novel Antiretroviral Compounds

Quantum dots (QDots) are fluorescent semiconductor nanocrystals with a narrow emission spectrum, high quantum yield, and excellent photostability. Furthermore, QDots can be conjugated with biological molecules while retaining their optical properties. These unique properties of QDots have been utilized to develop a fluorescent binding assay using biotinylated HTLV-1 conjugated with streptavidin-coated QDots that enabled both qualitative and quantitative analyses of viral binding. The specificity and linearity of the assay was demonstrated utilizing T cells, the primary HTLV-1-susceptible cell population. Furthermore, Differential binding of HTLV-1 was analyzed in various cell types of clinical relevance including primary CD4+ and CD8+ T cells, dendritic cells (DCs), monocytes, bone marrow progenitor cells, and epithelial cells. DCs exhibited maximum binding affinity when compared to other primary cells. Finally, the high throughput use of the assay was demonstrated by using an array of blocking antibodies against a putative HTLV-1 receptor on DCs; DC-SIGN (dendritic cell-specific ICAM-3-grabbing non-integrin). The assay is currently being utilized in an automated robotic system to screen a large library of small molecule inhibitors (average molecular weight 350 Dalton) against retroviral envelope protein (HTLV-1 gp46 and HIV-1 gp120) binding to cell surface receptors. Overall, these results demonstrated that this novel high throughput assay can be utilized to study the binding of any biotinylated virus and has implications for identification of viral binding inhibitors as well as host membrane proteins that may serve as receptors for viral entry.

OP-61 A New Double-Antigen Sandwich ELISA Kit for Highly Sensitive and Specific Detection of Antibodies to HTLV-I/II Virus in Serum Samples

A new double-antigen sandwich-based HTLV-I/II ELISA 4.0 kit for the detection of total antibodies (IgG, IgM and IgA) specific to Human T-cell Lymphotropic Virus type I (HTLV-I) and type II (HTLV-II) was developed utilizing a combination of recombinant proteins and a tri-fusion recombinant protein labeled with horseradish peroxidase. These recombinant proteins correspond to the highly antigenic segments of HTLV-I and HTLV-II viruses. Our study showed that the ELISA kit detected all confirmed HTLV-I/II positive patient samples (n = 515), while maintaining an excellent specificity of 99.82% with samples from various patient (n = 305) and healthy control groups (n = 5001). The positive predictive value (PPV) and the negative predictive value (NPV) for the new test reached 98.3% and 100%, respectively. In addition, external test results indicated that the ELISA kit is also capable of detecting antibodies specific to HTLV-3 and STLV-3 virus in serum samples. External evaluation of the ELISA kit on two automated systems confirmed that the kit is compatible with Dade Behring BEP III and Hamilton Microlab FAME.

Presenting author: Wei-Ping Hu. Mailing address: MP Biomedicals Asia Pacific Pte Ltd., 2 Pioneer Place, Singapore 627885. Phone: (65) 67750008, Fax: (65) 68628968. E-mail: weiping.hu@mpbio.com

OP-65 Abstract Withdrawn

Nejmeddine M. 1 Lewis P.N. 2 Majorovits E. 2 Negi V.S. 3 Mukherjee S. 3 Merriman C. 2 Orth K. 3 Tanaka Y. 4 Taylor G. 5 Fuller S. 2 Bangham C. 1

1 Department of Immunology, Wright-Fleming Institute; Imperial College London 2 Department of Structural biology, Welcome Trust Centre for Genetics, University of Oxford 3 Department of Molecular Biology, University of Texas Southwestern Medical Centre, Dallas, Texas, USA 4 Department of Immunology, Graduate School and Faculty of Medicine, University of the Ryukyus, Okinawa, Japan 5 Department of GU Medicine and Communicable Diseases, Wright-Fleming Institute, Imperial College London. OP-66 HTLV-1 Virological Synapse: Ultrastructure and Role of Tax Protein

It's increasingly recognized that human retroviruses spread efficiently between T-cells through an infectious synaptic structure, termed the virological synapse (VS). The concept of virological synapse was first established for human T-lymphotropic virus type 1 (HTLV-1). The VS resemble the immunological synapse (IS) both are triggered by the interaction between intercellular adhesion molecule-1 (ICAM-1) and its ligant LFA-1 (leukocyte-function associated antigen-1). This early engagement stimulate T-cell signalling by launching a cascade of tyrosine phosphorylation events that leads both to the rearrangement of the actin microfilaments network beneath the plasma membrane and to the polarization of the microtubules organising centre (MTOC) toward the target T-cell. MTOC polarization is particularly crucial in cytotoxic T-cells (CTL), it stabilize the IS and induces a polarized secretion allowing direct delivery of cytotoxic granules to the IS formed between the CTL and the target cells. Similar mechanism is thought to be used by HTLV-1 to deliver virions into the VS. The expression of HTLV-1-Tax protein is sufficient to trigger the polarization of the MTOC from within the cell. By testing the effect of a series of mutants Tax proteins, we are investigating the relationship between the known functions of Tax, its intracellular localization, and its ability to trigger the polarization of the MTOC that is observed during the establishment of the VS.

The events leading to the formation of the VS, including the polarization of the cytoskeleton and the transfer of the virus particles can be microscopically visualized. We are extending our recent work on the three-dimensional ultrastructure of the VS (Majorovits E; Nejmeddine M. et al.; PLoS ONE. 2008; 3(5):e2251), by carrying out new studies combining light microscopy and electron tomography to examine the transport and the distribution of HTLV-1 proteins to the VS and from cell to cell. We show that the virological synapse that forms spontaneously between lymphocytes of HTLV-1-infected individuals allows direct cell-cell transmission of the virus by triggered, directional release of enveloped HTLV-1 particles into multiple confined intercellular spaces within the VS. These results reconcile the known requirement for HTLV-1-Env protein with the absence of detectable cell-free virions in the serum.

OP-67 Human T-Cell Leukemia Virus Type 1 Antisense Encoded Gene, HBZ, Promotes T-Lymphocyte Proliferation

Human T-cell leukemia virus type 1 (HTLV-1) basic leucine zipper factor (HBZ) is dispensable for HTLV-1-mediated cellular transformation in cell culture, but is required for efficient viral infectivity and persistence in rabbits. In most adult T-cell leukemia (ATL) cells, Tax oncoprotein expression is typically low or undetectable, whereas Hbz gene expression is maintained suggesting that Hbz expression may support infected cell survival and ultimately, leukemogenesis. Data indicates that HBZ protein can interact with CREB and Jun family members modulating transcription factor binding and transactivation of both viral and cellular promoters. Lentiviral vectors that express Hbz-specific short hairpin (sh)RNA effectively decreased both Hbz mRNA and HBZ protein expression in transduced HTLV-1-transformed SLB-1 T-cells. Hbz knockdown correlated with a significant decrease in T-cell proliferation in culture. Both SLB-1 and SLB-1-Hbz knockdown cells engrafted into inoculated NOD/SCIDgamma chain-/- mice to form solid tumors that also infiltrated multiple tissues. However, tumor formation and organ infiltration was significantly decreased in animals challenged with SLB-1-Hbz knockdown cells. We utilized our rabbit model of HTLV-1 infection and real-time RT-PCR to determine the kinetics of viral gene expression in PBMCs from newly infected rabbits. We observed that tax/rex and gag/pol mRNA levels peaked early after inoculation and progressively decreased, which correlated inversely with the proviral load and host antibody response against viral proteins. Interestingly, Hbz mRNA was detectable at one week post-infection and increased and stabilized. Expression of all other HTLV-1 genes in infected rabbit PBMCs were at or below our limit of detection. Taken together our data indicate that Hbz expression enhances the proliferative capacity of HTLV-1 infected T-cells and that in newly infected rabbits, Hbz mRNA expression over time directly correlates with proviral load, which provides the first evidence linking Hbz expression to proviral load and the survival of the virus infected cell in the host.

OP-68 Human T-Cell Leukemia Virus Type 2 Produces a Spliced Antisense Transcript Encoding a Protein That Lacks a BZIP Domain But Inhibits Tax2-Mediated Transcription

While HTLV-1 infection causes leukemia and inflammatory diseases like the neurological disorder HTLV-1 associated myelopathy/tropical spastic paraparesis (HAM/TSP), HTLV-2 has not been demonstrated to be the agent of a hematological malignant disease but has been recently associated with lymphocytosis. The minus strand of the HTLV-1 genome encodes HBZ, a protein which plays a role in of the maintenance of the leukemic cells proliferation. Herein, we demonstrate that the complementary strand of the HTLV-2 genome also encodes a protein that we named APH-2 (Antisense Protein of HTLV-2). APH-2 mRNA is spliced, polyadenylated and initiates in the 3′LTR at different positions. This transcript was detected in different tested HTLV-2-infected cell lines as well as in HTLV-2 cultured primary cells and HTLV-2 carriers. The APH-2 protein is localized in the nucleus but not the nucleolus of transfected cells. APH-2 binds CREB but not the p300 and represses Tax2-mediated transcription. The existence of an antisense strand-encoded protein in HTLV-2 could represent an important player in the development of disorders such as the lymphocytosis that is frequently observed in HTLV-2 patients.

OP-70 Site-Specific Phosphorylation Regulates Human T-Cell Leukemia Virus Type 2 Rex Function In Vivo

The human T-cell leukemia virus type-2 (HTLV-2) Rex is a transacting regulatory protein required for efficient cytoplasmic expression of the unspliced and incompletely spliced viral mRNA transcripts encoding the structural and enzymatic proteins. Previously, it was demonstrated that phosphorylation of Rex-2, predominantly on serine residues, correlated with an altered conformation as observed by a gel mobility shift and the detection of two related protein species (p24Rex and p26Rex). Rex-2 phosphorylation is required for specific binding to its viral mRNA target sequence, inhibition of mRNA splicing, and may be linked to subcellular compartmentalization. Thus, the phosphorylation state of Rex in the infected cell may be a switch that determines whether HTLV exists in a latent or productive state. We have conducted a phosphoryl and functional mapping of mammalian cell-expressed Rex-2 using affinity purification, liquid chromatography tandem mass spectrometry, and site-directed substitutional mutational analysis. We identified two novel phosphorylation sites in p24Rex at Ser-117 and Thr-164. We also have identified six phosphorylation sites in p26Rex at Thr-19, Ser-117, Ser-125, Ser151, Ser-153, and Thr-164. We evaluated the functional significance of these novel phosphorylation events and found that phosphorylation on Thr-164, Ser-151, and Ser-153 is critical for Rex-2 function in vivo, and that phosphorylation of Ser-151 is correlated with nuclear/nucleolar subcellular localization. Overall, this work is the first to completely map the phosphorylation sites in Rex-2 and provides important insight into the phosphorylation continuum that tightly regulates Rex-2 function, HTLV replication, and survival in the infected cell.

IN VIVO”);'>

OP-71 The HTLV-1 P13 II Protein Interacts with Tax, Inhibits Tax Dependent Transcription and Decreases Viral Replication

Human T- cell leukemia virus type 1 (HTLV-1) is the causative agent of adult T-cell leukemia/lymphoma and HTLV-1-associated myelopathy/tropical spastic paresis (HAM/TSP). The viral genome encodes the two essential positive regulators of viral replication Tax and Rex. Tax is activating viral transcription through the HTLV-1 LTR promoter by association with CBP/p300. In addition, Tax interacts with over 100 different other cellular proteins affecting cellular gene expression, cell signaling and cell proliferation. Rex functions at the post-transcriptional level interacting with the Rex responsive element (RXRE) located in the 3’ LTR allowing nuclear export of unspliced and partially spliced viral mRNA. HTLV-1 is thought to propagate by clonal expansion of infected cells rather than de novo viral infection, suggesting that suppression of viral expression in replicating infected T-cells may be the key for their survival in the presence of immune surveillance. Indeed, there are two recognized viral encoded negative regulators of viral expression. The non-shuttling nuclear protein p30II, encoded by ORF-II that inhibits the export of tax/rex mRNA from the nucleus and the HBZ protein, encoded from the anti-sense strand of the viral genome, inhibiting Tax transcriptional activity. Here we present evidence that the HTLV-1 encodes an additional negative regulator of its own expression: the p13II protein. The p13II protein is expressed from ORF-II through a private single spliced mRNA. It alters the function and morphology of the mitochondria, suppresses cell proliferation, represses tumorgenicity in a murine model and sensitizes Jurkat T cells to Ras-mediated apoptosis. We have found that p13II suppresses HTLV-1 by inhibiting Tax mediated transcription from the viral LTR. The p13II protein binds and co-immunoprecipitates with Tax and competes with the association of Tax to CBP/p300, however, no interaction between p13II and CBP/p300 was observed. Tax localizes both to the nucleus and cytoplasm and our preliminary data support interaction between p13II and Tax in the cytoplasm. Interestingly, we find that the presence of Tax greatly stabilizes the expression of p13II while p13II appears to have the opposite effect on Tax resulting in degradation. We find that in the presence of p13II, Tax is degraded through the proteasome. Since Tax is mostly K63 ubiqutinated, we are currently investigating the status of Tax ubiquitination in the presence of p13II. In addition we find that p13II dimerizes through the unique cysteine at position 27. In summary, we provide evidence that p13II suppresses HTLV-1 replication by targeting Tax. This work provides additional evidence that HTLV-I has evolved several functions to maintain Tax expression at a minimum, perhaps to both help T-cells to maintain “a low profile” in an immune competent host and to curtail the effect of Tax on cell cycle progression.

OP-73 The HTLV-I Tax Oncoprotein Interferes with DNA Double

Approximately 500,000 DNA lesions occur each day in a cell due to inherent errors during DNA replication, and to extrinsic stimuli (e.g. ionizing radiation, UV light). To protect the cell from toxic effects of these DNA lesions, the DNA damage response coordinates the recognition and repair of DNA damage to maintain genome integrity. Genome instability, a hallmark of cancer cells, is associated with malfunction of the DNA damage response caused by defective cell cycle checkpoints and/or DNA repair.

HTLV-I transformed cells display extensive genome instability. The ability of the HTLV-I oncoprotein, Tax, to modulate transcription of cellular genes involved in proliferation, interact directly with cell cycle regulators, and interfere with the repair of DNA damage, is thought to mediate its transforming properties. Since double-strand breaks (DSBs) are some of the most deleterious types of DNA damage, we investigated the effects of Tax on the recognition and repair of DSBs.

Double strand break repair is activated when the ATM kinase senses DNA breaks. Recruitment of activated ATM to sites of DNA damage results in phosphorylation of the histone variant H2AX (gH2AX), which subsequently recruits proteins such as MDC1. MDC1 recruitment is essential for the accumulation of activated ATM at damage sites to maintain a DNA damage response. We have shown that Tax-expressing cells appear to recognize DNA damage appropriately, but display premature attenuation of ATM activity following ionizing radiation which induces DSBs. Thus, the cells progress more rapidly toward DNA replication, despite inefficient DNA repair. Premature attenuation of ATM activity appears to be due to defects in the recruitment of MDC1 to chromatin after DNA damage and to disruption of gH2AX-mediated feedback signaling in Tax-expressing cells. As a consequence of this attenuation, the cells fail to maintain the DNA damage response, resulting in premature DNA replication in the presence of genomic lesions. Attempts to replicate through these lesions may result in gradual accumulation of mutations, leading to genome instability and cellular transformation, as observed in ATL cells.

OP-74 Modulation of HTLV-1 Tax Protein Ubiquitination by Cellular Optineurin: Functional Implications for the NF- ?B Pathway Activation

HTLV-1 Tax protein has been shown to play a central role in the process of host cell transformation. Tax alters the expression of many cellular genes by modulating the activity of specific cellular transcription factors, among which NF- ?B plays a critical role. Interaction of Tax with the regulatory component of I ?B kinase, IKK- ?/NEMO has been shown to be necessary for Tax-induced NF- ?B activation. Moreover, several groups have recently demonstrated that Tax association with NEMO is correlated with K63-linked polyubiquitination of Tax. We recently identified optineurin, a cellular protein that shares sequence similarity with NEMO, as a potential interactor of Tax. Optineurin is a Golgi-resident protein, which is mutated in certain forms of glaucoma. Cellular functions associated with optineurin encompass architecture of the Golgi, regulation of the secretory pathway, and inhibition of TNF-a signaling. Using co-immunoprecipitation assays, we demonstrate the interaction between Tax and optineurin in Tax-transfected cells as well as HTLV-infected cell lines. Our immunofluorescent studies also suggest that this interaction occurs in Golgi-associated structures, where Tax recruits NEMO. Furthermore, we demonstrate that similarly to NEMO, Tax interaction with optineurin is dependent upon Tax ubiquitination. Interestingly, overexpression of optineurin leads to a dose-dependant increase of Tax-induced NF- ?B activation. Conversely, silencing of optineurin decreases Tax activation of NF- ?B. We also show that interaction with optineurin stabilizes Tax1 ubiquitinated form. To further understand the mechanism underlying the modulating effect of optineurin on Tax ubiquitination, we investigate the involvement of TAX1BP1, a cellular protein known to interact both with Tax and with the deubiquitination machinery composed of A20 and Itch. We show that Tax, optineurin and TAX1BP1 form a complex. Finally our results suggest that TAX1BP1 acts down-stream of optineurin for the modulation of Tax function. Altogether, these results bring new insights into the mechanisms accounting for the modulation of Tax post-translational status in regards to Tax activity on the NF-? B pathway, and indicate that optineurin could be a major modulator of Tax activity.