Published online before print October 25, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Laboratórios de
* Imunorregulação e Microbiologia (LIMI) and
Saúde Pública (LASP) CPqGM, FIOCRUZ, Salvador-BA, Brazil;
Hospital Universitário Professor Edgard Santos (HUPES), Salvador-BA, Brazil;
|| Instituto de Investigação em Imunologia (iii–Millenium Institute), SP, Brazil; and
Clinical and Epidemiological Virology, Rega Institute, Leuven, Belgium
1 Correspondence: FIOCRUZ, LIMI-CPqGM, Rua Waldemar Falcão 121, 40296-710 Salvador-BA, Brazil. E-mail: johan{at}bahia.fiocruz.br
ABSTRACT
Fas (TNFRSF6/Apo-1/CD95) is a type I transmembrane receptor, which mediates apoptosis. Fas gene mutations, aberrant transcripts, and abundant expression of Fas have been reported in adult T cell leukemia (ATL). To further elucidate the role of Fas in ATL pathogenesis, we investigated whether the –670 FAS promoter A/G polymorphism (STAT1-binding site) might contribute to susceptibility and clinical outcome in ATL. Thirty-one patients with ATL, 33 healthy, human T lymphotropic virus type 1-infected individuals, and 70 healthy, uninfected controls were genotyped for the FAS –670 polymorphism by PCR-restriction fragment-length polymorphism. The AA genotype was significantly over-represented in ATL patients in comparison with healthy controls (P=0.006), as well as asymptomatics (P=0.037), corresponding to an odds ratio (OR) of 3.79 [95% confidence intervals (CI; 1.28–11.41)] and 4.58 [95% CI (1.13–20.03)], respectively. The AA group also comprised significantly more aggressive (acute and lymphoma) clinical subtypes [P=0.012; OR=8.40; 95% CI (1.60–44.12)]. In addition, we observed a statistically significant association between GG genotype and survival (log rank test, P=0.032). Finally, IFN-
-induced but not basal FAS mRNA levels were increased significantly (P=0.049) in PBMCs from AA versus GG individuals, demonstrating the IFN-dependent functionality of the –670 polymorphism. In conclusion, our results demonstrate that a functional Fas promoter polymorphism is significantly associated to susceptibility, clinical manifestation, and survival in ATL.
Key Words: TNFRSF6 • ATL • STAT1 • HTLV-1 • apoptosis
Adult T cell leukemia (ATL) is an aggressive neoplasm of activated T lymphocytes caused by human T lymphotropic virus type 1 (HTLV-1) infection and usually occurs in 2–4% of asymptomatic carriers after a 20- to 40-year latency period. As a majority of infected individuals remains asymptomatic, host genetic factors are presumed to play a pivotal role in ATL development. Regarding genetic susceptibility to ATL, few data are available. A significant association to the –857T TNF polymorphism has been described [1
], whereas P53 codon 72 polymorphism was associated to ATL progression but not to susceptibility [2
]. Fas (also known as TNFRSF6, CD95, or APO-1) is a type I transmembrane receptor belonging to the TNFR superfamily, which plays a key role in apoptotic signaling in a variety of cell types. Deregulation of the FAS signaling pathway has been shown to participate in immune escape and tumorogenesis and has been associated with differentiation, invasiveness, and metastasis of cancer cells. Mutations in FAS lead to lymphadenopathy and an increase in the incidence of B cell lymphoma [3
]. Furthermore, functional polymorphisms in the FAS gene are associated with susceptibility to cancer, such as acute myeloid leukemia [4
] and cervical carcinoma [5
]. FAS gene mutations, aberrant transcripts, and deregulation of Fas expression have been reported in ATL cells [6
, 7
]. To further elucidate the role of Fas in ATL pathogenesis, we investigated whether a single nucleotide polymorphism at position –670 in the FAS gene promoter [8
] that occurs in a STAT1-binding motif [IFN--activated sequence (GAS)] might contribute to susceptibility and clinical outcome in ATL.
The ethical committee of Hospital Universitário Professor Edgard Santos (Salvador-BA, Brazil) approved this study; informed consent was obtained from all participants. Diagnosis of ATL, HTLV-1 infection, and clinical subtype classification was made according to published criteria [9
], combining clinical and anatomo-pathological data, ELISA (Murex, Paris, France), Western blot (INNO-LIA, Innogenetics, Belgium), flow cytometry, and inverse PCR to detect monoclonal integration [10
, 11
]. Genotyping for the Fas –670 A/G polymorphism by PCR-restriction fragment length polymorphism (RFLP) was performed as described [8
]. Briefly, a 331-bp fragment of the FAS promoter was amplified using specific primers (5'-CTACCTAAGAGCTATCTACCGTTC-3' and 5'-GGCTGTCCATGTTGTGGCTGC-3') and digested with MvaI. We compared the allelic distribution of this polymorphism in 31 patients with ATL, 33 healthy HTLV-1-infected individuals (seropositives), and 70 healthy uninfected controls (HUC; seronegatives), genotyped by PCR-RFLP. With this sample size, we had 80% power to detect significant differences in genotype frequencies at odds ratios (OR) of at least 3.75. Allele frequencies and genotype distributions of FAS –670 polymorphism for each group are shown in Table 1
. The functional AA genotype (corresponding to the consensus STAT1-binding motif) was the most frequent in patients with ATL (38.7%), whereas 35.5% displayed the GG genotype, and 25.8% were heterozygous GA. ATL genotype distribution was significantly different from HUC (
2, P=0.019) and HIC (2, P=0.014). The AA genotype was significantly over-represented in ATL patients in comparison with healthy controls (2, P=0.006), as well as asymptomatics (Fisher’s exact test, P=0.037), corresponding to an OR of 3.79 [95% confidence intervals (CI; 1.28–11.41)] and 4.58 [95% CI (1.13–20.03)], respectively. In addition, neither allelic distribution (P=0.44) nor genotype frequencies (P=0.76) were significantly different between asymptomatic HTLV-1-infected subjects and healthy controls. Thus, Fas promoter polymorphism does not seem to be associated with risk of infection but more specifically, with the process of leukemogenesis. Next, we examined if the –670 polymorphism was associated to clinical manifestation and outcome in ATL, which is usually classified into four subtypes: acute, lymphoma, chronic, and smoldering, according to published criteria [9
]. Chronic and smoldering ATL have a less-aggressive, clinical course, whereas acute and lymphoma subtypes display a decreased survival time. When analyzing clinical subtypes in ATL patients according to Fas promoter genotypes, the AA group comprised significantly more aggressive clinical subtypes (acute and lymphoma) in comparison with GA + GG groups [P=0.012; OR=8.40; 95% CI (1.60–44.12)]. We also observed a statistically significant linear tendency (2 for trend, P=0.006) for the presence of none, one, or two functional A alleles (GG, AG, and AA, respectively) and manifestation of aggressive clinical subtypes. The AA group comprised 57.1% of acute patients. Moreover, a lymphoma subtype was observed only in patients with at least one functional A allele [homozygous AA (71.4% of lymphoma forms) and heterozygous GA (28.6% of lymphoma forms)], as none of the lymphoma subtype patients presented a GG genotype. In addition, 63.6% of GG and the 50% of GA patients manifested a smoldering type of ATL, whereas in the AA group, only 25.0% of cases presented a smoldering subtype. Therefore, the AA genotype of the FAS –670 polymorphism might not only predispose to ATL development but also to a more aggressive clinical presentation. We also investigated if the –670 polymorphism had any effect on survival in ATL, as determined from onset of disease until death or during at least 1 year of follow-up for patients who were still alive. We observed a statistically significant association between GG genotype and survival (log rank test, P=0.032), as shown in Figure 1
. Thus, the presence of at least one A allele was associated significantly with increased mortality during the first year of follow-up from ATL onset, whereas no significant difference in survival was observed between AA and GA genotypes.
|
View this table: [in a new window] |
Table 1. Genotype Frequencies of FAS–670 Polymorphism in ATL and Control Groups
|
 View larger version (13K): [in a new window] |
Figure 1. Patients with AA and GA genotypes were grouped (no significant difference in survival was observed between AA and GA genotypes), and survival rates during the first year of follow-up were compared with the GG group. Onset was calculated from beginning of clinical symptoms; no significant differences in a patient’s age at onset were observed between groups. Log rank test was used for survival curves (P=0.032).
|
-induced Fas mRNA levels in PBMCs from 12 healthy donors (four GG, five GA, and three AA genotypes). As shown in Figure 2
, IFN- was unable to induce Fas mRNA levels (0.6±2.0% increase) in GG individuals, but a gradual increase was observed with each functional A allele: 8± percent in GA and 19.2± percent in AA individuals (P=0.049, as compared with GG, t-test). However, no differences in basal FAS mRNA levels were observed between genotypes (not shown), indicating that the –670 polymorphism determines IFN-induced and hence, STAT1-mediated FAS expression rather than interfering with basal promoter activity. We hypothesize that increased Fas expression (functional AA and/or GA genotypes) might contribute to ATL susceptibility and clinical progression by affecting normal host cells involved in tumor surveillance, in agreement with recent observations in cervical cancer, where genetic polymorphisms in the Fas-Fas ligand pathway might confer susceptibility through enhanced activation-induced cell death of tumor-specific T cells [14
]. Leukemogenesis in ATL is expected to be a multifactorial process as a result of several genetic and epigenetic changes described affecting cell cycle and apoptosis regulatory genes, generally requiring several decades [15
]. Furthermore, additional apoptosis-independent functions of Fas have been described [16
], including induction of proliferation in T cells, through which Fas might contribute to ATL pathogenesis.
 View larger version (18K): [in a new window] |
Figure 2. PBMCs from healthy controls (three AA, five GA, and four GG) were cultured for 12 h in the absence or presence of 1000 IU/ml IFN-. Total RNA was extracted, reverse-transcribed, and amplified using specific primers for β-2-microglobulin (CATCCAGCGTACTCCAAAGA and GACAAGTCTGAATGCTCCAC) and FAS (TCCTACCTCTGGTTCTTACG and AGATCTGGATCCTTCCTCTT), followed by PAGE and quantification using ImageJ software. Fas mRNA levels were normalized to β-2-microglobulin levels; IFN- induction of FAS mRNA is represented as percent increase over unstimulated cells. One-tailed t-test of log-transformed values was used to compare the effect of IFN- between genotypes (*, P=0.049 for AA vs. GG).
|
+ AZT, whereas milder forms mostly remain untreated. Confirmation of our results in other ATL endemic areas will be necessary before Fas genotyping might enter into practice in ATL clinical management. In conclusion, our results demonstrate that a functional Fas promoter polymorphism is significantly associated to susceptibility, clinical manifestation, and survival in ATL.
ACKNOWLEDGEMENTS
Financial support was provided by FAPESB, CNPq, and K. U. Leuven "Leerstoel voor wetenschappelijk onderzoek naar infectieziekten in ontwikkelingslanden." A. L. B. and B. G-C. are CNPq senior investigators; L. F., A. A., and G. M. S. received a FAPESB fellowship. We thank the patients and the volunteers who participated in this study.
Received April 2, 2007; revised October 2, 2007; accepted October 2, 2007.
REFERENCES
polymorphism associated with increased susceptibility to development of adult T-cell leukemia/lymphoma in human T-lymphotropic virus type 1 carriers Cancer Res. 61,3770-3774This article has been cited by other articles:
 |
|
 |
|
  Z. Zhang, H. Xue, W. Gong, M. Wang, L. Yuan, S. Han, and Z. Zhang FAS promoter polymorphisms and cancer risk: a meta-analysis based on 34 case-control studies Carcinogenesis, March 1, 2009; 30(3): 487 - 493. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
