Masao Matsuoka, Institute for Virus Research, Kyoto University, Kyoto 606-8507, Japan, Retrovirology 2005, 2:27 doi:10.1186/1742-4690-2-27
© 2005 Matsuoka; licensee BioMed Central Ltd., This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

How does Human T-cell Leukemia Virus Type 1 transmit and replicate in vivo?

How does Human T-cell Leukemia Virus Type 1 replicate and increase its copy number?

Somatic alterations in Adult T-cell Leukemia cells

The host immune system, especially the cellular response, against Human T-cell Leukemia Virus Type 1 exerts critical control over virus replication and the proliferation of infected cells [84]. CTLs against the virus have been extensively studied, and Tax protein was found to be the dominant antigen recognized by CTLs in vivo [63]. HTLV-I-specific CD8-positive CTLs are abundant and chronically activated. The paradox is that the frequency of Tax-specific CTLs is much higher in HAM/TSP patients than in carriers. Since the provirus load is higher in HAM/TSP patients, this finding suggests that the CTLs in HAM/TSP cannot control the number of infected cells. One explanation for this is that the CTLs in HAM/TSP patients show less efficient cytolytic activity toward infected cells, whereas CTLs in carriers can suppress the proliferation of infected cells [85]. Hence, the gene expression profiles of circulating CD4+ and CD8+ lymphocytes were compared between carriers with high and low provirus loads. The results revealed that CD8+ lymphocytes from individuals with a low Human T-cell Leukemia Virus Type 1 provirus load show higher expressions of genes associated with cytolytic activities or antigen recognition than those from carriers with a high provirus load [86]. Thus, CD8+ T-lymphocytes in individuals with a low provirus load successfully control the number of HTLV-I-infected cells due to their higher CTL activities. Thus, the major determinant of the provirus load is thought to be the CTL response to HTLV-I.

As mentioned above, the provirus load is considered to be controlled by host factors. Considering that the cellular immune responses are critically implicated in the control of Human T-cell Leukemia Virus Type 1 infection, human leukocyte antigen (HLA) should be a candidate for such a host genetic factor. From analyses of HAM/TSP patients and asymptomatic carriers, HLA-A02, and Cw08 are independently associated with a lower provirus load and a lower risk of HAM/TSP. In addition, polymorphisms of other genes (TNF-α, SDF-1, HLA-B54, HLA-DRB-10101 and IL-15) are also associated with the provirus load, although their associations are not as significant compared with HLA-A02, and Cw08 [87,88]. Regarding the onset of Adult T-cell Leukemia, only a polymorphism of TNF-α gene was reported to show an association [89]. However, familial clustering of ATL cases is a well-known phenomenon, strongly suggesting that genetic factors are implicated in the onset of ATL [90-92].

Spontaneous remission is more frequently observed in patients with Adult T-cell Leukemia than those with other hematological malignancies [90,93]. Usually, this phenomenon is associated with infectious diseases, suggesting that immune activation of the host enhances the immune response against ATL cells. If the immune response against Human T-cell Leukemia Virus Type 1 is implicated in spontaneous remission, this suggests the possibility of immunotherapy for ATL patients by the induction of an immune response to HTLV-I [94], for example via antigen-stimulated dendritic cells.

Immunodeficiency in Adult T-cell Leukemia patients is pronounced, and results in frequent opportunistic infections by various pathogens, including Pneumocystis carinii, cytomegalovirus, fungus, Strongyloides and bacteria, due to the inevitable impairment of the T-cell functions [95]. To a lesser extent, impaired cell-mediated immunity has also been demonstrated in Human T-cell Leukemia Virus Type 1 carriers [96]. Such immunodeficiency in the carrier state may be associated with the leukemogenesis of Adult T-cell Leukemia by allowing the proliferation of Human T-cell Leukemia Virus Type 1-infected cells. A prospective study of HTLV-I-infected individuals found that carriers who later develop Adult T-cell Leukemia have a higher anti-HTLV-I antibody and a low anti-Tax antibody level for up to 10 years preceding their diagnosis. This finding indicates that Human T-cell Leukemia Virus Type 1 carriers with a higher anti-HTLV-I titer, which is roughly correlated with the Human T-cell Leukemia Virus Type 1 provirus load, and a lower anti-Tax reactivity may be at the greatest risk of developing ATL [97]. The anti-Human T-cell Leukemia Virus Type 1 antibody and soluble IL-2 receptor (sIL-2R) levels are correlated with the Human T-cell Leukemia Virus Type 1 provirus load [53], and a high antibody titer and high sIL-2R level are risk factors for developing Adult T-cell Leukemia among carriers [98]. Taken together, these findings suggest that a higher proliferation of HTLV-I-infected cells and a low immune response against Tax may be associated with the onset of Adult T-cell Leukemia. Given these findings, potentiation of CTLs against Tax via a vaccine strategy may be useful for preventing the onset of ATL [99].

EBV-associated lymphomas frequently develop in individuals with an immunodeficient state associated with transplantation or AIDS. This has also been reported in an Adult T-cell Leukemia patient [100]. Does such an immunodeficient state influence the onset of ATL? Among 24 patients with post-transplantation lymphoproliferative disorders (PT-LPDs) after renal transplantation in Japan, 5 cases of ATL have been reported. Considering that most PT-LPDs are of B-cell origin in Western countries, this frequency of ATL in Japan is quite high. Although the high Human T-cell Leukemia Virus Type 1 seroprevalence is due to blood transfusion during hemodialysis, the immunodeficient state during renal transplantation apparently promotes the onset of ATL [101]. In addition, when experimental allogeneic transplantation was performed to 12 rhesus monkeys and immunosuppressive agents (cyclosporine, prednisolone or lymphocyte-specific monoclonal antibodies) were administered to prevent rejection, 4 of the 7 monkeys that died during the experiment showed PT-LPDs. Importantly, the STLV provirus was detected in all PT-LPD samples [102]. These observations emphasize that transplantation into HTLV-I-infected individuals or from HTLV-I positive donors require special attention.

Although the mechanism of immunodeficiency remains unknown, some previous reports have provided important clues. One mechanism for immunodeficiency is that Human T-cell Leukemia Virus Type 1 infects CD8-positive T-lymphocytes, which may impair their functions [48]. Indeed, the immune response against Tax via HTLV-I-infected CD8-positive T-cells renders these cells susceptible to fratricide mediated by autologous HTLV-I-specific CD8-positive T-lymphocytes [103]. Fratricide among virus-specific CTLs could impair the immune control of HTLV-I. Another mechanism for immunodeficiency is based on the observation that the number of naive T-cells decreases in individuals infected with HTLV-I via decreased thymopoiesis [48]. In addition, CD4+ and CD25+ T-lymphocytes are classified as immunoregulatory T-cells that control the host immune system. Regulatory T-cells suppress the immune reaction via the expression of immunoregulatory molecules on their surfaces. The FOXP3 gene has been identified as a master gene that controls gene expressions specific to regulatory T-cells. FOXP3 gene transcription can be detected in some Adult T-cell Leukemia cases (10/17; 59%) [104]. Such ATL cells are thought to suppress the immune response via expression of immunoregulatory molecules on their surfaces, and production of immunosuppressive cytokines.

7. Pathogenesis of Human T-cell Leukemia Virus Type 1 infection

ATL cells are derived from activated helper T-lymphocytes, which play central roles in the immune system by elaborating cytokines and expressing immunoregulatory molecules. Adult T-cell Leukemia cells are known to retain such features, and this cytokine production or surface molecule expression may modify the pathogenesis.

Adult T-cell Leukemia is well known to infiltrate various organs and tissues, such as the skin, lungs, liver, gastrointestinal tract, central nervous system and bone [95]. This infiltrative tendency of leukemic cells is possibly attributable to the expressions of various surface molecules, such as chemokine receptors and adhesion molecules. Skin-homing memory T-cells uniformly express CCR4, and its ligands are thymus and activation-regulated chemokine (TARC) and macrophage-derived chemokine (MDC). CCR4 is expressed on most ATL cells. In addition, TARC and MDC are expressed in skin lesions in ATL patients. Thus, CCR4 expression should be implicated in the skin infiltration [105]. On the other hand, CCR7 expression is associated with lymph node involvement [106]. OX40 is a member of the tumor necrosis factor family, and was reported to be expressed on ATL cells [107]. It was also identified as a gene associated with the adhesion of ATL cells to endothelial cells by a functional cloning system using a monoclonal antibody that inhibited the attachment of ATL cells [108]. Thus, OX40 is also implicated in the cell adhesion and infiltration of ATL cells.

Hypercalcemia is frequently complicated in patients with acute Adult T-cell Leukemia (more than 70% during the whole clinical course) [109]. In hypercalcemic patients, the number of osteoclasts increases in the bone (Figure 3). RANK ligand, which is expressed on osteoblasts, and M-CSF act synergistically on hematopoietic precursor cells, and induce the differentiation into osteoclasts [110]. ATL cells from hypercalcemic ATL patients express RANK ligand, and induced the differentiation of hematopoietic stem cells into osteoclasts when ATL cells were co-cultured with hematopoietic stem cells [111]. In addition, the serum level of parathyroid hormone-related peptide (PTH-rP) is also elevated in most of hypercalcemic ATL patients. PTH-rP indirectly increases the number of osteoclasts, as well as activating them [112,113], which is also implicated in mechanisms of hypercalcemia.

8. Treatment of Adult T-cell Leukemia - the remaining mission and challenges

Regardless of intensive chemotherapies, the prognosis of Adult T-cell Leukemia patients has not so improved. The median survival time of acute or lymphoma-type ATL was reported to be 13 months with the most intensive chemotherapy [114]. Such a poor prognosis might be due to: 1) the resistance of ATL cells to anti-cancer drugs; and 2) the immunodeficient state and complicated opportunistic infections as described above. Regarding the resistance to anti-cancer drugs, one mechanism is the activated NF-κB pathway in ATL cells [115], which increases the transcription of anti-apoptotic genes such as bcl-xL and survivin. A proteasome inhibitor, bortezomib, is currently used for the treatment of multiple myeloma. One of its mechanisms is suppression of the NF-κB pathway by inhibiting the proteasomal degradation of IκB protein. Several groups have shown that bortezomib is effective against ATL cells both in vitro and in vivo [116-119]. Since the sensitivity to bortezomib is well correlated with the extent of NF-κB activation, the major mechanism of the anti-ATL effect is speculated to be inhibition of NF-κB. In addition, an NF-κB inhibitor has also been demonstrated to be effective against ATL cells [120].

During chemotherapy for Adult T-cell Leukemia, chemotherapeutic agents worsen the immunodeficient state of ATL patients. In this regard, antibody therapy against ATL cells has advantages due to its decreased adverse effects. A humanized monoclonal antibody to CD25 has been clinically administered to patients with ATL [121,122]. In addition, a monoclonal antibody to CD2 is at the preclinical stage [123]. As described above, most ATL cells express CCR4 antigen on their surfaces, and a humanized antibody against CCR4 is being developed as an anti-ATL agent [124].

Advances in the treatment of Adult T-cell Leukemia were brought about by allogeneic bone marrow or stem cell transplantation [125,126]. Absence of graft-versus-host disease (GVHD) was linked with relapse of ATL, suggesting that GVHD or graft-versus-ATL may be implicated in the clinical effects of allogeneic stem cell transplantation [125]. Furthermore, 16 patients with ATL, who were over 50 years of age, were treated with allogeneic stem cell transplantation with reduced conditioning intensity (RIST) from HLA-matched sibling donors [127]. Among 9 patients in whom ATL relapsed after transplantation, 3 achieved a second complete remission after rapid discontinuation of cyclosporine A. This finding strongly suggests the presence of a graft-versus-ATL effect in these patients. In addition, Tax peptide-recognizing cells were detected by a tetramer assay (HLA-A2/Tax 11-19 or HLA-A24/Tax 301-309) in patients after allogeneic stem cell transplantation [128]. In 8 patients, the provirus became undetectable by real-time PCR. Among these, 2 patients who received grafts from HTLV-I-positive donors also became provirus-negative by real-time PCR after RIST. Since the provirus load is relatively constant in HTLV-I-infected individuals [53], this finding indicates an enhanced immune response against Human T-cell Leukemia Virus Type 1 after RIST, which suppresses the provirus load. This may account for the effectiveness of allogeneic stem cell transplantation to ATL. However, Tax expression is frequently lost in ATL cells as described above. Many questions arise, such as whether the tax gene status is correlated with the effect of allogeneic stem cell transplantation, and whether the effectiveness of the anti-HTLV-I immune response is against leukemic cells or non-leukemic HTLV-I-infected cells. Nevertheless, these data suggest that potentiation of the immune response against viral proteins such as Tax may be an attractive way to treat ATL patients [94]. Such strategies may enable preventive treatment of high-risk HTLV-I carriers, such as those with familial ATL history, predisposing genetic factors to ATL, a higher provirus load, etc.

9. Two human retroviruses - Human T-cell Leukemia Virus Type 1 and HIV-1

As described in the first section, Human T-cell Leukemia Virus Type 1 has resided in humans for a long time. On the other hand, HIV-1 has only been recently transmitted to humans, probably from chimpanzees. Due to the comparatively small genomic differences between humans and chimpanzees, this virus can quickly adapt to human cells. These two human retroviruses are opposite in many aspects. HIV-1 vigorously replicates in vivo, and the maximum production of HIV-1 virions in the body can reach 10¹⁰ per day. Since reverse transcriptase is an error-prone enzyme due to its lack of proof-reading activity, it produces about one mistake per replication, resulting in tremendous errors in the proviral sequence during replication. Although most of these variations ruin the virus replication due to nonsense mutations or impairment of viral gene functions, some become capable of replicating under different circumstances such as the presence of anti-HIV drugs and activation of the host immune system. This can account for why HIV-1 acquires resistance against anti-HIV drugs, and escape from CTLs. On the other hand, Human T-cell Leukemia Virus Type 1 increases its copy number in two ways, namely replication of HTLV-I itself and the proliferation of HTLV-I-infected cells in vivo.

Although immune responses (antibodies, CTLs) against viral proteins suggest the presence of active viral replication in vivo, most of increased Human T-cell Leukemia Virus Type 1 provirus load (the number of infected cells) is considered to be due to proliferation of infected cells since CTLs efficiently eliminate virus-expressing cells. Therefore, there is much less variation in the HTLV-I provirus sequence compared with HIV-1 [129]. However, this strategy by which Human T-cell Leukemia Virus Type 1 increases the number of infected cells due to clonal expansion generates unfortunate side effects for both the host and the virus, namely oncogenesis of CD4-positive T-lymphocytes and the development of ATL.

Acknowledgements

I would like to thank my colleagues Jun-ichirou Yasunaga, Kisato Nosaka, Mika Yoshida, Yorifumi Satou, Yuko Taniguchi, Satoshi Takeda, Ken-ichirou Etoh and Sadahiro Tamiya for their excellent studies.