Human T-cell leukemia virus type 1 (HTLV-1) is the etiological agent of adult T-cell leukemia/lymphoma (ATL), tropical spastic paraparesis/HTLV-1-associated myelopathy (TSP/HAM), and other inflammatory disorders, including infective dermatitis and uveitis (30
). HTLV-1 is not a ubiquitous virus, but its geographical distribution is mainly restricted to some areas of high endemicity, including sub-Saharan Africa, Southern Japan, Central and South America, and some regions of the Middle East and Melanesia (30
). In some of these countries, as well as in some areas of low endemicity (e.g., the United States, Canada, and a few European countries), public health measures aimed at reducing HTLV-1 transmission and dissemination have been taken in the last few decades. This includes the screening of blood donors for the presence of HTLV-1 (30
Sub-Saharan Africa is considered to be one of the largest areas of endemicity for HTLV-1 infection, with an estimated 2 to 4 million HTLV-1-infected individuals. However, most of the early sero-epidemiological studies performed in Africa applied Western blot criteria for HTLV-1 seropositivity that were subsequently shown not to be stringent (3
). This has led to some overestimation of the HTLV-1 seroprevalence in several African areas. Thus, even up to now, there are only a few studies on HTLV-1 infection in the African continent in which the authors used stringent serological and/or molecular criteria for the diagnosis of HTLV-1 infection (18
). Moreover, most of the epidemiological studies have been performed in hospitalized patients or in individuals at risk for acquiring HTLV-1 (prostitutes, human immunodeficiency virus [HIV]-infected patients, etc.) or in rural populations (2
). Therefore, very few data have been reported thus far for large populations of blood donors from west or central Africa (1
). Furthermore, no data are available on the molecular subtyping of the HTLV present in blood donors from western Africa.
The objectives of the present study were thus to perform a seroepidemiological and a molecular study of HTLV-1 in a large population of blood donors from Senegal, a country of western Africa where ATL, TSP/HAM, and infective dermatitis cases have been reported (19
The National Blood Transfusion Center (CNTS) of Senegal currently selects nonpaid, voluntary blood donors on the basis of a health check questionnaire and examination performed by a physician (33
). All donated samples are then tested for detection of HBsAg and of antibodies directed against HIV type 1 (HIV-1), HIV-2, and hepatitis C virus, as well as of syphilitic serology.
Our prospective HTLV study took place from May to October 2002. A total of 4,900 blood donors (mean age, 29.6 years; range, 18 to 65) from the CNTS were included in this serological survey. There were 3,585 men (mean age, 31.8 years; range, 18 to 65 years) and 1,315 women (mean age, 26 years; range, 18 to 52 years). All of the blood donors lived in the Dakar area, and most of them (58%) originated from Dakar. Other blood donors included in the study originated from the northern (17%), southern (13%), and eastern (12%) regions of the country. The majority (53.1%) of them were students from schools or universities. Fourteen percent were unemployed, and the others (33%) had very diverse professions. A total of 61% were born in a town, and 39% were of rural origin. This survey was approved by the Ethic Committee of Senegal and was performed after obtaining the informed consent of the blood donors.
All plasma samples were screened with an enzyme-linked immunosorbent assay (ELISA; HTLV-I+II; Abbot/Murex, United Kingdom). The positive and/or borderline samples were further tested by a confirmatory Western blot assay (WB; HTLV-I/II Blot 2.4; Diagnostic Biotechnology, Singapore). A sample was considered HTLV-1 positive if it reacted to the two Gag proteins (p19 and p24) and both env
-encoded glycoproteins: the HTLV-1-specific recombinant gp46-I peptide (MTA-1) and the specific HTLV-1/HTLV-2 recombinant GD21 protein. It was considered HTLV-2 positive if it reacted to the Gag protein p24 and both env
-encoded glycoproteins: the HTLV-2-specific recombinant gp46-II peptide (K55) and the GD21 protein. Plasma samples were considered negative when they exhibited no bands and indeterminate when they were partially reactive. Moreover, determination of the HTLV-1 and HTLV-2 antibody titers was performed by twofold limited dilution using an immunofluorescence assay (IFA) with either HTLV-1 (MT2)- or HTLV-2 (C19)-producing cell lines, as previously described (21
Among the 4,900 tested samples, only 14 were repeatedly considered to be ELISA positive. Furthermore, only 8 of these 14 samples were considered HTLV seropositive (0.16%) according to stringent WB criteria, 7 of them being HTLV-1 and 1 being HTLV-2 (Table ) . The six other samples exhibited either no reactivity (four cases) or an indeterminate WB pattern (p19 alone and GD21 alone). The HTLV-1 and HTLV-2 seroprevalence increased with age, ranging from 0.1% in blood donors that were <30 years old to 0.8% in blood donors that were >50 years old (Fig. ).
Main epidemiological and virological features of HTLV-1- or HTLV-2-infected blood donors from Dakar, Senegala
Age-dependent seroprevalence rates for HTLV-1 and HTLV-2 in 4,900 blood donors from all Senegal and in 637 blood donors originating only from the southern region of Senegal (Casamance). An increase with age is observed in both populations.
The seven HTLV-1-seropositive plasma originated from six men and one woman with a mean age of 34.5 years, while the only HTLV-2-seropositive individual was a woman of 53 years (Table ). None of these individuals had been previously transfused, and they all denied any intravenous drug use. Three of the six children of these donors were tested for HTLV and were found to be ELISA seronegative.
The eight blood donors had given during their life a total of 79 blood gifts (Table ). Of these eight blood donors, six, including the HTLV-2-infected woman, originated from the southern part of the country, the Casamance, an area from which only 13% of the blood donors originated. Thus, in the subpopulation of the 637 blood donors from Casamance, the HTLV-1 and HTLV-2 seroprevalence reached nearly 1% (0.94%) with an increase with age (Fig. ).
High-molecular-weight DNA was extracted from the buffy coats from four of the six HTLV-1-seropositive blood donors, from the HTLV-2-seropositive blood donors, and from the two blood donors with HTLV-indeterminate results by means of the QIAamp DNA Blood Minikit (QIAGEN GmbH, Hilden, Germany). The seven DNA samples were subjected to PCR with primers specific for the human β-globin gene to check that cellular DNA was amplifiable. The DNA samples were then subjected to two series of PCR in order to obtain the complete long terminal repeat (LTR; 755 bp) and a 522-bp region of the HTLV-1 env
gene as previously described (22
). For HTLV-2, only a fragment comprising most of the LTR (622 bp) was amplified by using specific primers as described previously (28
). The PCR products were then cloned and sequenced, and phylogenetic studies were performed as previously described (25
Analysis of both LTR and env
sequences demonstrated that all of the four new HTLV-1 strains were closely related together with a nucleotidic interstrain difference ranging from 0 to 0.9% for the LTR sequences and from 0 to 0.2% for the 522-bp env
genomic fragments. Furthermore, phylogenetic analyses, using several representative strains of the different HTLV-1 molecular subtypes, indicated clearly that the four novel strains belonged to the large cosmopolitan HTLV-1 subtype (Fig. ). They constituted more specifically a highly phylogenetically supported clade (bootstrap of 90% in the neighbor-joining analysis), located between the northern African and the western African HTLV-1 subgroups (Fig. ) and comprising most of the other HTLV-1 strains known from Senegal (20
FIG. 2. Phylogenetic tree generated with the neighbor-joining method, performed by using the PAUP program (v4.0b10), on a 750-bp fragment of the LTR region. The sequence alignment was submitted to the Modeltest program (version 3.6) to select the best model to (more ...)
Regarding the sole HTLV-2 strain, LTR sequence analysis (data not shown) indicated that this strain belonged surprisingly to the HTLV-2 subtype C phylogenetic clade, being closely related to some strains from Brazilian Amerindians (28
In conclusion, we demonstrate here the presence of HTLV-1 infection in blood donors from Senegal. These strains belong to the cosmopolitan subtype. Moreover, our data suggest that the Southern part of the country (a forest area much more humid than the other areas of Senegal) is a probable cluster of HTLV-1 infection, with a level for areas of endemicity reaching 1 to 2% in the adult population. This finding resembles the situation found in Guinea, a neighboring country of the southern region of Senegal, where the prevalence in the general population has been found to range from 0.2 to 1.9% (16
). In Guinea, it is also worthwhile to note that a similar level of HTLV-1 seroprevalence (1%) was found in a series of 1,700 blood donors from the capital city, Conakry (12
). Another north-to-south gradient of HTLV-1 seroprevalence was also observed in Cameroon, with the highest level of viral infection being also clearly in the humid rain forest area (23
Finally, the data presented herein illustrate the current transfusional risk of HTLV-1 in Senegal. This concerns not only the dissemination of such viral infection by transfusion but also the occurrence of TSP/HAM. Indeed, several posttransfusion cases of this chronic neurological disorder have been reported (11
), and TSP/HAM cases have been diagnosed in neurological departments in Dakar (27
). Moreover, transfusion is a major risk factor for TSP/HAM development in several areas of endemicity, such as Japan, the West Indies, and South America (8
). Lastly, history of blood transfusion has been demonstrated to be an important risk factor for HTLV-1 infection in Africa, including urban population of Guinea Bissau (26
) and children from Gabon, Central Africa (31
We also demonstrated here the presence of HTLV-2 infection, which has been very rarely reported in western Africa (12
) and whose origin is still a matter of discussion (13