Content uploaded by Mahmoud Mahmoudi
Author content
All content in this area was uploaded by Mahmoud Mahmoudi on Jan 12, 2015
Content may be subject to copyright.
Differences in viral and host genetic risk factors
for development of human T-cell lymphotropic
virus type 1 (HTLV-1)-associated myelopathy/
tropical spastic paraparesis between Iranian and
Japanese HTLV-1-infected individuals
Amir H. Sabouri,
1
Mineki Saito,
1
Koichiro Usuku,
2
Sepideh Naghibzadeh Bajestan,
1
Mahmoud Mahmoudi,
3
Mohsen Forughipour,
4
Zahra Sabouri,
3
Zahra Abbaspour,
3
Mohammad E. Goharjoo,
4
Esmaeil Khayami,
5
Ali Hasani,
5
Shuji Izumo,
6
Kimiyoshi Arimura,
1
Reza Farid
3
and Mitsuhiro Osame
1
Correspondence
Mineki Saito
mineki@m3.kufm.kagoshima-u.
ac.jp
1,2
Department of Neurology and Geriatrics
1
and Department of Medical Information Science
2
,
Kagoshima University Graduate School of Medical and Dental Sciences, 8-35-1
Sakuragaoka, Kagoshima 890-8520, Japan
3,4
Department of Immunology and Immunology Research Center
3
and Department of
Neurology
4
, Faculty of Medicine, Mashhad University of Medical Sciences, Mashhad, Iran
5
Khorasan Blood Transfusion Center, Mashhad, Iran
6
Department of Molecular Pathology, Center for Chronic Viral Diseases, Kagoshima University,
8-35-1 Sakuragaoka, Kagoshima 890-8520, Japan
Received 9 August 2004
Accepted 2 December 2004
Human T-cell lymphotropic virus type 1 (HTLV-1)-associated myelopathy/tropical spastic
paraparesis (HAM/TSP) is a neurological disease observed only in 1–2 % of infected individuals.
HTLV-1 provirus load, certain HLA alleles and HTLV-1 tax subgroups are reported to be
associated with different levels of risk for HAM/TSP in Kagoshima, Japan. Here, it was
determined whether these risk factors were also valid for HTLV-1-infected individuals in Mashhad
in northeastern Iran, another region of endemic HTLV-1 infection. In Iranian HTLV-1-infected
individuals (n=132, 58 HAM/TSP patients and 74 seropositive asymptomatic carriers),
although HLA-DRB1*0101 was associated with disease susceptibility in the absence of
HLA-A*02 (P=0?038; odds ratio=2?71) as observed in Kagoshima, HLA-A*02 and HLA-Cw*08
had no effect on either the risk of developing HAM/TSP or HTLV-1 provirus load. All Iranian
subjects possessed tax subgroup A sequences, and the protective effects of HLA-A*02 were
observed only in Kagoshima subjects with tax subgroup B but not in those with tax subgroup A.
Both the prevalence of HTLV-1 subgroups and the host genetic background may explain the
different risks levels for HAM/TSP development in these two populations.
INTRODUCTION
Human T-cell lymphotropic virus type 1 (HTLV-1) (Poiesz
et al., 1980; Yoshida et al., 1982) is a causative agent
of adult T-cell leukaemia (Hinuma et al., 1981; Yoshida
et al., 1984) and the chronic neurodegenerative disorder
HTLV-1-associated myelopathy/tropical spastic paraparesis
(HAM/TSP) (Gessain et al., 1985; Osame et al., 1986). Only
a minority of HTLV-1-infected individuals develop
HAM/TSP, and most infected individuals remain healthy
throughout their lives. A previous seroepidemiological
survey in Kyushu Island, in southwestern Japan, where
Kagoshima prefecture is located, estimated the incidence of
HAM/TSP among HTLV-1-infected persons at 3?1610
25
cases per year; assuming a lifespan of 75 years, the lifetime
incidence is therefore approximately 0?25 % (Kaplan
et al., 1990). In HAM/TSP patients from Kagoshima, the
median provirus load in peripheral blood mononuclear
cells (PBMCs) is more than ten times higher than HTLV-
1-seropositive asymptomatic carriers (HCs) and high pro-
virus load is also associated with an increased risk of
progression to disease (Nagai et al., 1998). HTLV-1 provirus
load has been correlated with progression of motor dis-
ability (Takenouchi et al., 2003) and the risk of sexual
transmission of HTLV-1 (Kaplan et al., 1996). Thus,
HTLV-1 provirus load is an important correlate of virus
transmission as well as disease progression. A previous study
0008-0509 G2005 SGM Printed in Great Britain 773
Journal of General Virology (2005), 86, 773–781 DOI 10.1099/vir.0.80509-0
indicated that the provirus load in PBMCs from HCs in
genetic relatives of patients with HAM/TSP in Kagoshima
was significantly higher than that of non-HAM/TSP-related
HCs, suggesting the importance of genetic background for
developing HAM/TSP (Nagai et al., 1998). In the Kagoshima
population, an association between HLA-DRB1*0101,
HLA-B*5401, HLA-A*02 and HLA-Cw*08 and the outcome
of HTLV-1 infection has been reported, where HLA-A*02
and HLA-Cw*08 genes were each independently associated
with a lower HTLV-1 provirus load and with protection
from HAM/TSP, whereas HLA-DRB1*0101 and HLA-
B*5401 were associated with an increased susceptibility to
HAM/TSP (Jeffery et al., 1999, 2000). The association of
HLA-DRB1*0101 with disease susceptibility was only
evident in the absence of the protective effect of HLA-
A*02 (Jeffery et al., 1999). These results are consistent
with the hypothesis that a strong class I-restricted T-cell
response is beneficial (Bangham, 2000). In another study,
an association between HTLV-1 tax gene sequence variation
and the risk of HAM/TSP was reported (Furukawa et al.,
2000). The tax subgroup A was more frequently observed
in HAM/TSP patients than in HCs and this effect was
independent of HLA-A*02. These reports suggested that
both host genetic factors and HTLV-1 subgroup indepen-
dently play a part in determining the risk of developing
HAM/TSP.
HTLV-1 is also endemic in the Caribbean Basin (Blattner
et al., 1982), Africa (Biggar et al., 1984), South America
(Zamora et al., 1990; Cartier et al., 1993; Zaninovic et al.,
1994) and the Melanesian islands (Yanagihara et al., 1990).
The city of Mashhad in northeastern Iran has also been
reported as an endemic centre for HTLV-1 (Safai et al.,
1996). In a recent study, the prevalence of HTLV-I infec-
tion was reported to be 0?77 % among blood-bank donors
of Mashhad (Abbaszadegan et al., 2003), but the prevalence
and incidence of HAM/TSP are unknown in this popula-
tion. Since there has been no report to compare the genetic
risk factors for HAM/TSP among different ethnic popu-
lations, it was interesting to study whether genetic risk
factors found in Kagoshima, Japan, were also valid for
HAM/TSP development in the Mashhadi Iranian popula-
tion. We therefore analysed the HTLV-1 provirus load,
HTLV-1 tax subgroup and the allele frequencies of HLA-
A*02, HLA-B*5401, HLA-Cw*08 and HLA-DRB1*0101 in
Iranian HTLV-1-infected individuals using the same
methods and techniques that were used in the Kagoshima
studies (Nagai et al., 1998; Jeffery et al., 1999, 2000). The
effect of host genetic factors and HTLV-1 tax subgroups
on the risk of HAM/TSP development in different ethnic
groups is discussed.
METHODS
Study populations. Peripheral blood samples were studied from
58 Iranian patients with HAM/TSP and 74 HCs from blood donors
of the Blood Transfusion Center in the city of Mashhad and
Neyshabour, both located in HTLV-1-endemic northeastern Iran.
The study population from Kagoshima consisted of 222 patients
with HAM/TSP and 184 HCs, all of whom were enrolled in the
previous Kagoshima studies (Nagai et al., 1998; Jeffery et al., 1999,
2000; Furukawa et al., 2000). The diagnosis of HAM/TSP was
made according to the World Health Organization diagnostic
criteria (Osame, 1990). Informed consent was obtained from all
patients. This research was approved by the institutional review
boards of the authors’ institutions.
DNA preparation. All Japanese and Iranian blood samples were
taken by vacuum tube pre-filled with the anticoagulant EDTA.
Genomic DNA extraction procedures were different for each popula-
tion. In the case of Kagoshima samples, fresh PBMCs were isolated
by Histopaque-1077 (Sigma) density-gradient centrifugation and
genomic DNA was extracted using a QIAamp Blood kit (Qiagen).
For Iranian samples, for economical and technical reasons, fresh
blood specimens were frozen immediately after collection and frozen
whole-blood samples were transported to Kagoshima University on
dry ice. Genomic DNA of nucleated blood cells was isolated from
whole blood in Kagoshima University using the PureGene DNA
Purification kit (Gentra Systems).
Provirus load measurement. To assay the HTLV-1 provirus load,
we carried out a quantitative PCR using ABI Prism 7700 (PE
Applied Biosystems) with 100 ng genomic DNA (equivalent to
approx. 10
4
cells) from PBMCs (for Kagoshima samples) or
nucleated blood cells (for Iranian samples) as reported previously
(Nagai et al., 1998). Using b-actin as an internal control, the
amount of HTLV-1 provirus DNA was calculated using the follow-
ing formula: copy number of HTLV-1 tax per 10
4
PBMCs (for
Japanese samples) or nucleated blood cells (for Iranian samples)
=[(copy number of tax)/(copy number of b-actin/2)]610
4
. All
samples were tested in triplicate. The lower limit of detection was
one copy of HTLV-1 tax per 10
4
PBMCs. In this study, we used the
previously analysed provirus load data of Kagoshima samples from
our database (Nagai et al., 1998). All Iranian samples and some
randomly selected Kagoshima samples were analysed using the same
kit (AmpliTaq Gold and TaqMan probe; PE Applied Biosystems)
and machine (ABI Prism 7700) at the same time. The same standard
DNA for tax and b-actin was used throughout the study and there
was no discrepancy between old and new data (not shown).
Sequencing of the HTLV-1 tax gene. Randomly selected Iranian
samples from 10 HAM/TSP patients and 10 HCs were sequenced
over almost the entire HTLV-1 tax gene (nt 7295–8356, nucleotide
numbers correspond to those of the prototypic strain, ATK-1; Seiki
et al., 1983). PCR was done on extracted DNA to amplify provirus
DNA, and nucleotide sequences were determined by direct sequenc-
ing in both directions. We amplified 100 ng DNA in 35 cycles of
PCR, using an expanded high-fidelity PCR system (Boehringer
Mannheim) and 1 mM primers (PXO1
+
,59-TCGAAACAGCCCT-
GCAGATA-39, nt 7257–7276, and PXO2
+
,59-TGAGCTTATG-
ATTTGTCTTCA-39, nt 8447–8467). Each PCR cycle consisted of
denaturation at 94 uC for 60 s, annealing at 58 uC for 75 s, exten-
sion at 72 uC for 90 s and a final extension at 72 uC for 10 min.
Amplified DNA products were purified using a purification kit
(QIAquick; Qiagen) and 0?1mg PCR product was sequenced with a
dye terminator DNA sequencing kit (Applied Biosystems) with 3?2
pmol each primer [PXI1
+
,59-ATACAAAGTTAACCATGCTT-39,nt
7274–7293; PXI2
+
,59-GGCCATGCGCAAATACTCCC-39, nt 7618–
7637; PXI3
+
,59-TTCCGTTCCACTCAACCCTC-39, nt 8001–8020;
PXI1
2
,59-GGGTTCCATGTATCCATTTC-39, nt 7644–7663, PXI2
2
,
59-GTCCAAATAAGGCCTGGAGT-39, nt 8024–8043; and PXI3
2
,
59-AGACGTCAGAGCCTTAGTCT-39, nt 8374–8393]in an auto-
matic DNA sequencer (model 377; Applied Biosystems).
Restriction fragment length polymorphism (RFLP) analysis
of the HTLV-1 tax gene. To determine the HTLV-1 tax gene sub-
group (tax A or B) in Iranian samples, we carried out a PCR-RFLP
774 Journal of General Virology 86
A. H. Sabouri and others
analysis as previously described (Furukawa et al., 2000). For RFLP
analysis, 4 ml PCR product was digested with 5 U AccII (Takara) in
10 ml total volume at 37 uC for 1 h followed by electrophoresis on
2 % Nusieve agarose gel. The previously analysed tax subgroup data
of Kagoshima samples (Furukawa et al., 2000) were extracted from
our database. Positive and negative controls of known Japanese
samples of tax gene subgroups A and B, which were confirmed by
direct sequence analysis, were included in all experiments.
HLA typing. PCR sequence-specific primer reactions were per-
formed to detect HLA-A*02, HLA-B*5401, HLA-Cw*08 and HLA-
DRB1*0101 as previously described (Bunce et al., 1995; Olerup &
Zetterquist, 1992). We used previously analysed HLA data of
Kagoshima samples from our database (Jeffery et al., 1999, 2000).
Statistical analysis. Statistical analysis was performed using the
SPSS for Windows release 7.0, run on an IBM-compatible computer
(Analytical Software, version 7). The x
2
test, the Mann–Whitney U
test and the odds ratio (OR) were used for statistical analysis. Values
of P<0?05 were considered statistically significant.
RESULTS
Differences in HTLV-1 provirus load between
HAM/TSP patients and asymptomatic carriers is
significantly lower in Iranian HTLV-1-infected
individuals than in Japanese
We used the previously analysed provirus load data of
Kagoshima samples from our database (Nagai et al., 1998);
all Iranian samples were newly analysed. The median
age of HAM/TSP patients in both Kagoshima (57?3 years,
range 15–80 years, 70?4 % female) and Iran (49?7 years,
range 24–80 years, 72?1 % female) was greater than that of
HCs in Kagoshima (39?4 years, range 16–64 years, 52?7%
female) and Iran (41?4 years, range 22–73 years, 38?3%
female), respectively. There was no significant difference in
age between the control groups (HCs) of the two popula-
tions. All HCs in each group originated from unrelated
blood donors. Since we extracted Japanese genomic DNA
samples from PBMCs but Iranian samples from whole
blood, direct comparison of HTLV-1 provirus load between
the two populations was inappropriate. Since the main
target of HTLV-1 infection is human T cells, whole blood-
derived DNA contains more uninfected nucleated cells
than PBMCs, and therefore the provirus load data in
Iranians was likely to be underestimated if we used b-actin
as an internal control. Thus, we compared the HTLV-1
provirus load between HAM/TSP patients and asympto-
matic carriers within each population. As shown in Fig. 1,
although the HTLV-1 provirus load of Iranian HAM/TSP
patients was significantly higher than that of Iranian HCs
(P=0?009, Mann–Whitney U test), as reported in Japanese
patients (Nagai et al., 1998), the differences in median
provirus load between Iranian HAM/TSP patients and
HCs (twofold greater in the HAM/TSP patients than in
the HCs) was much smaller than that of Japanese subjects
(13-fold). Interestingly, although provirus load data were
probably underestimated in Iranian samples compared
with Japanese samples, the HTLV-1 provirus load in
Iranian HCs was still significantly higher than that of
Japanese HCs (P=0?004, Mann–Whitney U test).
HLA-A*02 and HLA-Cw*08 are not associated
with a lower risk of HAM/TSP and a lower
provirus load in Iranian HTLV-1-infected
individuals
To examine whether the previously reported associations
between class I and class II HLA alleles and HAM/TSP
prevalence in Kagoshima was also valid for HAM/TSP
development in the Iranian population, we genotyped
HLA-DRB1*0101 and HLA-A*02, HLA-B*5401 and HLA-
Cw*08 by PCR-based DNA typing in 132 Iranian HTLV-
1-infected individuals (58 HAM/TSP and 74 HCs). All
Japanese HLA data had been previously analysed and were
extracted from our database (Jeffery et al., 1999, 2000). As
shown in Table 1, the genotype frequency of HLA-A*02
and HLA-Cw*08 in Kagoshima subjects was significantly
lower among the cases of HAM/TSP compared with HCs
(P=0?0006 and 0?0196, respectively). In contrast, the
genotype frequency of HLA-A*02 and HLA-Cw*08 was
not significantly different between Iranian HAM/TSP and
HCs (P=0?346 and 0?940, respectively). Also, whereas
HLA-A*02 and HLA-Cw*08 were associated with a lower
median provirus load in Kagoshima subjects (P=0?0003
for A*02 and P=0?009 for HLA-Cw*08; Mann–Whitney
U test), this effect was not observed in Iranian subjects
Fig. 1. HTLV-1 provirus load of Japanese and Iranian HTLV-1-
infected individuals. Mean HTLV-1 copy numbers per 10
4
PBMCs for Japanese samples and per 10
4
nucleated cells for
Iranian samples determined by quantitative PCR are shown.
The HTLV-1 provirus load of Iranian HAM/TSP patients was
significantly higher than that of Iranian HCs (P=0?009, Mann–
Whitney U test). The difference in median provirus load
between Iranian HAM/TSP patients and HCs was much smaller
than that of Japanese (Kagoshima) subjects, since HTLV-I pro-
virus load in Iranian HCs is significantly higher than in Japanese
HCs (P=0?004). Error bars indicate SEM.
http://vir.sgmjournals.org 775
Risk factors for Iranian HAM/TSP
(P=0?071 for A*02 and P=0?75 for HLA-Cw*08; Mann–
Whitney U test; Table 2), indicating that a protective effect
of HLA-A*02 and HLA-Cw*08 was not observed in Iranian
HTLV-1-infected individuals. As expected, HLA-B*5401,
which is known to be almost exclusively found in East
Asian populations, was not found in the Iranian subjects
analysed.
HLA-DRB1*0101 increases the odds of HAM/
TSP development in both Japanese and Iranian
HLA-A*02-negative, but not in HLA-A*02-
positive, HTLV-1-infected individuals
In contrast to HLA-A*02, HLA-DRB1*0101 was asso-
ciated with susceptibility to HAM/TSP in both Japanese
(P=0?049) and Iranian (P=0?035) populations (Table 3).
This effect was observed only in the HLA-A*02-negative
subjects but not in the HLA-A*02-positive subjects in
both populations (Table 3). Although possession of HLA-
DRB1*0101 was associated with a significantly lower pro-
virus load in the Japanese HAM/TSP patients (Table 4,
P=0?024) but not in HCs, HLA-DRB1*0101 was not
associated with a difference in the provirus load in the
Iranian HTLV-1-infected HAM/TSP patients and HCs
(Table 4).
All Iranian HTLV-1 isolates have 10 nt
substitutions in the tax region including all the
tax subgroup A substitutions
Based on the LTR gene sequence, HTLV-1 can be classified
into three types: Melanesian, Central African and cosmo-
politan types, while cosmopolitan types can be further
classified into subtypes A, B and C (Miura et al., 1994).
There are two distinct subtypes in Japan; the most fre-
quently observed (nearly 80 %) Japanese subtype belongs
to cosmopolitan subtype B, while a minor subtype (20 %),
which seems to cluster in the southern islands of Kyushu
and the Ryukyu Islands, belongs to cosmopolitan subtype
A. A previous report suggested that, although Mashhadi
HTLV-1 isolates belonged to cosmopolitan subtype A,
this strain formed a tight cluster that was distinct from
the other isolates of cosmopolitan subtype A from Japan,
India, the Caribbean Basin and South America (Yamashita
Table 1. HLA-A*02 and HLA-Cw*08 are not associated with a lower risk of HAM/TSP in Iranian HTLV-1-infected individuals
Population (no. HAM/HCs) HLA allele HAM/TSP HCs x
2
*PORD95 % CI
Iranian (58/74) HLA-A*02
+
21 (36?2 %) 20 (27?0%) 0?887 0?346 1?53 0?73–3?22
HLA-A*02
2
37 (63?8 %) 54 (73?0%)
Japanese (222/184)dHLA-A*02
+
67 (30?2 %) 87 (47?3%) 11?784 0?0006 0?48 0?32–0?72
HLA-A*02
2
155 (69?8 %) 97 (52?7%)
Iranian (58/74) HLA-Cw*08
+
9 (15?5 %) 10 (13?5%) 0?006 0?940 1?18 0?44–3?11
HLA-Cw*08
2
49 (84?5 %) 64 (86?5%)
Japanese (222/184)dHLA-Cw*08
+
24 (10?8 %) 36 (19?6%) 5?45 0?0196 0?50 0?29–0?87
HLA-Cw*08
2
198 (89?2 %) 148 (80?4%)
*With Yates correction.
DOR used the approximation of Woolf (1955).
dJapanese data were extracted from a database from previous analyses (Jeffery et al., 1999, 2000).
Table 2. HLA-A*02 and HLA-Cw*08 are not associated with a lower provirus load in Iranian HTLV-1-infected individuals
Population HLA allele Provirus load (mean±SE)* Provirus load (median)* No. subjects PD
Iranian HLA-A*02
+
262?1±34?5 190?0410?071
HLA-A*02
2
209?6±24?9 120?091
JapanesedHLA-A*02
+
366?8±43?4 118?5 156 0?0003
HLA-A*02
2
525?6±41?5 266?0 250
Iranian HLA-Cw*08
+
198?2±42?8 131?0190?75
HLA-Cw*08
2
233?6±22?9 147?0 113
JapanesedHLA-Cw*08
+
300?7±56?4 120?0600?009
HLA-Cw*08
2
492?0±34?5 234?0 346
*Provirus load is the HTLV-1 tax copy number per 10
4
PBMCs for Japanese samples and per 10
4
nucleated cells for Iranian samples by
quantitative PCR.
DTwo-tailed Mann–Whitney U test.
dJapanese data were extracted from a database of previous analyses (Nagai et al., 1998; Jeffery et al., 1999, 2000).
776 Journal of General Virology 86
A. H. Sabouri and others
et al., 1995). A previous report indicated that the tax
subgroup A was more frequently observed in HAM/TSP
patients in the Kagoshima cohort and that this effect was
independent of HLA-A*02 (Furukawa et al., 2000). The
higher HAM/TSP risk tax subgroup A corresponds to the
cosmopolitan subtype A, and the lower HAM/TSP risk tax
subgroup B corresponds to the cosmopolitan subtype B
according to the LTR sequence (Furukawa et al., 2000).
We sequenced almost the entire tax region of HTLV-1
provirus (nt 7295–8356) from 20 different Iranian subjects
(10 HAM/TSP and 10 HCs) by direct sequencing in both
directions. As shown in Table 5, all Iranian HTLV-1
sequences (EMBL/GenBank/DDBJ accession no. AB181224)
differed at 10 nt compared with the Japanese prototypic
Table 3. HLA-DRB1*0101 increases the odds of HAM/TSP development in Japanese and Iranian HLA-A*02-negative, but not
in HLA-A*02-positive, HTLV-1-infected individuals
Population Subjects HAM/TSP (n) HCs (n)x
2
*PORD95 % CI
DRB1
+
DRB1
”
DRB1
+
DRB1
”
Iranian All 18 40 12 62 3?30?035 2?33 1?01–5?34
A*02
2
13 24 9 45 3?10?038 2?71 1?01–7?24
A*02
+
5 16 3 17 0?10?376 1?77 0?36–8?65
JapanesedAll 34 161 20 163 2?80?049 1?72 0?95–3?12
A*02
2
27 107 10 83 2?90?044 2?09 0?96–4?57
A*02
+
75410800?005 0?47 1?03 0?37–2?89
*Reported as one-tailed with Yates correction.
DOR used the approximation of Woolf (1955).
dJapanese data were extracted from a database of previous analyses (Jeffery et al., 1999).
Table 4. HLA-DRB1*0101 associated with lower HTLV-1 provirus load in Japanese but not in Iranian HAM/TSP patients
The DRB1-positive Japanese HAM/TSP patients developed HAM/TSP with a significantly lower provirus load than DRB1-negative HAM/
TSP patients, but this effect was not observed in Iranian HAM/TSP patients.
Population HLA
allele
HAM/TSP HCs
Median provirus load* No. subjects PDMedian provirus load* No. subjects PD
Iranian DRB1
2
193?0400?31 115?0620?34
DRB1
+
357?0 18 104?012
JapanesedDRB1
2
602?6 161 0?024 34?7 163 0?33
DRB1
+
331?134 49?020
*Provirus load is HTLV-1 tax copy number per 10
4
PBMCs for Japanese samples and per 10
4
nucleated cells for Iranian samples by quantitative
PCR.
DPlevel reported using two-tailed Mann–Whitney U test.
dJapanese data were extracted from the database of previous analyses (Nagai et al., 1998; Jeffery et al., 1999).
Table 5. Nucleotide variations specific to Iranian HTLV-1
Amino acid changes in tax A resulting from the nucleotide substitution are shown. Nucleotide numbers correspond to those of the proto-
typic strain, ATK-1 (Seiki et al., 1983). N, No change.
Subgroup Nucleotide variation (nucleotide position and amino acid change)
7622 7811 7855 7897 7959 7991 8208 8313 8314 8344
MRVIRVN NARVNRHSRNGREN N
ATK-1 (=tax B) A A T C C A G G C A
tax ATTAC
Iranian tax GGCTTC AAGC
http://vir.sgmjournals.org 777
Risk factors for Iranian HAM/TSP
ATK-1 strain (Seiki et al., 1983). Among these, nt 7897,
7959, 8208 and 8344 were exactly the same as those in
tax subgroup A. In addition to these four residues, the
Iranian tax sequences had 6 nt differences, which encoded
four additional amino acid differences from Japanese tax
subgroup A. We further performed PCR-RFLP analysis to
determine the HTLV-1 tax subgroup (tax A or B) of all of
the remaining Iranian samples and found that all Iranian
HTLV-1 isolates had tax subgroup A substitutions.
HLA-A*02 is associated with a lower risk of
HAM/TSP and a lower provirus load only in
HTLV-1-infected individuals with tax subgroup B
in Kagoshima subjects
As the majority of HTLV-1 isolates observed in the
Kagoshima population were tax subgroup B, we examined
further whether the effect of HLA-A*02 on the risk of
HAM/TSP and HTLV-1 provirus load was observed only in
HTLV-1 tax subgroup B-infected individuals in Kagoshima
subjects. Japanese tax subgroup data were extracted from
our existing database (Furukawa et al., 2000). As shown in
Table 6, the effects of HLA-A*02 on the risk of HAM/TSP
and provirus load were not observed in HTLV-1 tax sub-
group A-infected subjects in Kagoshima. We next sought
a possible interaction between HLA-A*02 and HTLV-1
provirus load among HTLV-1 tax subgroup A-infected
subjects in Kagoshima (Table 7). HLA-A*02 was asso-
ciated with a lower provirus load only in the tax subgroup
B subjects in Kagoshima, but not in the tax subgroup A
subjects in either Japan or Iran.
DISCUSSION
Currently, several different approaches including family-
based linkage and population-based case–control studies
have been used to identify genetic susceptibility to
numerous infectious pathogens such as malaria, myco-
bacteria, hepatitis viruses and human immunodeficiency
virus (Hill, 1998). The candidate gene approach (case–
control studies) can only utilize known genes and will not
identify unknown genes, but genome-wide linkage studies
have less power than candidate gene studies to pick up
genes that have only a small or moderate effect on disease
risk; therefore the two approaches are complementary.
Although our Kagoshima cohort of HAM/TSP is the
world’s largest, only 300 HAM/TSP patients were available
for analysis. Also, extensive studies in one ethnic popula-
tion may not disclose the marker-disease distance or
exclude a possible spurious association due to admixture.
Studies in different ethnic populations may thus provide
useful information about marker-disease distance, as well
as confirming the reliability of results from our previous
association studies. In this study, we compared the risk
factors for developing HAM/TSP in two ethnic groups
living in quite different environments, namely, Kagoshima
in southwest Japan and Mashhad in northeast Iran. It is
Table 6. HLA-A*02 is associated with a lower risk of HAM/TSP development only in tax subgroup B subjects in Kagoshima
Japanese data were extracted from a database of previous analyses (Jeffery et al., 1999; Furukawa et al., 2000).
tax subgroup HLA allele HAM/TSP HCs x
2
*PORD95 % CI
tax A HLA-A*02
+
16 (57 %) 6 (60 %) 0 ?047 0?829 0?89 0?20–3?87
HLA-A*02
2
12 (43 %) 4 (40 %)
tax B HLA-A*02
+
51 (26 %) 81 (47 %) 15?5<0?0001 0?41 0?26–0?63
HLA-A*02
2
143 (74 %) 93 (53 %)
*With Yates correction.
DOR used the approximation of Woolf (1955).
Table 7. HLA-A*02 is associated with a lower provirus load only in tax subgroup B subjects in Kagoshima
Japanese data were extracted from a database of previous analyses (Nagai et al., 1998; Jeffery et al., 1999; Furukawa et al., 2000).
tax subgroup HLA allele Provirus load (mean±SE)* Provirus load (median)* No. subjects PD
tax A HLA-A*02
+
635?0±169?3 389?0220?98
HLA-A*02
2
586?4±164?9 356?516
tax B HLA-A*02
+
328?5±41?699?0 132 0?0001
HLA-A*02
2
520?0±42?7 266?0 236
*Provirus load is the HTLV-1 tax copy number per 10
4
PBMCs by quantitative PCR.
DTwo-tailed Mann–Whitney U test.
778 Journal of General Virology 86
A. H. Sabouri and others
almost certain a priori that there will be significant differ-
ences between populations in the genetic contribution to
susceptibility to HAM/TSP, since HLA-B*5401 is prevalent
in Japan and elsewhere in East Asian populations, but is
virtually absent from many other populations. Since HLA-
B*5401 has an important influence on the risk of disease
in Kagoshima (Jeffery et al., 2000), its presence in the
population is certain to influence the risk associated
with other HLA alleles, and the absence of HLA-B*5401
in other populations with endemic HTLV-1 infection will
alter the relative importance of other genes to the risk of
developing HAM/TSP.
We first examined the HTLV-1 provirus load in Iranian
HAM/TSP patients and HCs, since one of the major risk
factors for developing HAM/TSP is the provirus load
(Nagai et al., 1998). The median HTLV-1 provirus load of
Iranian HAM/TSP patients was twofold greater in HAM/
TSP patients than in HCs, whereas that of Japanese HAM/
TSP patients was 13-fold greater than in HCs. Interestingly,
despite differences in the methods of DNA extraction
between the two study groups (whole blood-derived DNA
for Iranian samples vs PBMC-derived DNA for Japanese
samples), the HTLV-1 provirus load in Iranian HCs was
still significantly higher than Japanese HCs (P=0?004,
Mann–Whitney U test). This may be the main cause of the
smaller observed ratio of median provirus load between
HAM/TSP patients and HCs in the Iranian study group. To
investigate the reason for this difference between the two
populations, we further analysed the frequencies of certain
HLA alleles and the HTLV-1 tax subgroup in the Iranian
population.
In the Kagoshima population, possession of either of the
HLA class I genes HLA-A*02 or HLA-Cw*08 was associated
with a statistically significant reduction in both HTLV-1
provirus load and the risk of HAM/TSP (Jeffery et al., 1999,
2000). However, in Mashhadi Iranian subjects, both HLA-
A*02 and HLA-Cw*08 had no effect on either the risk of
HAM/TSP or provirus load. In contrast, HLA-DRB1*0101
was associated with increased susceptibility to HAM/TSP
both in Kagoshima (P=0?049) and Iran (P=0?035). In
HAM/TSP, CD4
+
cells are the predominant cells present
early in the active lesions (Umehara et al., 1993) and are also
the HTLV-1-infected cells in the inflammatory spinal cord
lesions (Moritoyo et al., 1996). Moreover, HLA-DRB1*0101
restricts CD4
+
T-cell immunodominant epitopes of
HTLV-1 env gp21 (Yamano et al., 1997; Kitze et al.,
1998). Therefore, it is possible that HLA-DRB1*0101 is
associated with susceptibility to HAM/TSP via an effect on
CD4
+
T-cell activation and subsequent bystander damage
in the central nervous system (Ijichi et al., 1993; Bangham,
2000). However, since possession of HLA-DRB1*0101 was
associated with a significantly lower provirus load in the
Japanese HAM/TSP patients but not in the Iranian HAM/
TSP patients, the underlying mechanism involving HLA-
DRB1*0101 may not be the same between Iranian and
Japanese HTLV-1-infected individuals. Differences in other
genetic factors, including non-HLA genes, may also be
important for explaining the observed differences between
the populations.
Another possible explanation of the observed differences
in the present study is that certain HLA genotypes are
associated with different effects on different subtypes of
the virus. In human papilloma virus (HPV) infection,
the association of the DRB1*1501–DQB1*0602 haplotype
with HPV-related cervical carcinoma was reported to be
specific for the viral type HPV-16, suggesting that specific
HLA haplotypes may influence the immune response to
specific virus-encoded epitopes and affect the risk of viral
disease (Apple et al., 1994). To test this possibility, we
sequenced almost the entire region of the tax gene
in 20 Mashhad Iranian HTLV-1-infected individuals (10
HAM/TSP and 10 HCs) and compared the sequence with
that of two Japanese strains, tax subgroups A and B.
Although we could not identify any amino acid differences
in the Tax11–19 immunodominant epitope between the
Iranian and Japanese tax subgroups A and B, we found that
Iranian HTLV-1 possessed 10 different nucleotides in the
tax region compared with Japanese tax subgroup B.
Among these, nt 7897, 7959, 8208 and 8344 were identical
to tax subgroup A. Therefore, Iranian tax sequences
have four additional different amino acids compared with
Japanese tax subgroup A and six additional different
amino acids compared with Japanese tax subgroup B.
These findings suggest that both the lack of consistency
of host genetic influences and the smaller difference in
median provirus load between HAM/TSP patients and
HCs in Iran may be due in part to different strains of
HTLV-1. Our present observation that HLA-A*02 was
associated with a lower provirus load only in the tax
subgroup B-infected subjects in Kagoshima, but not in
tax subgroup A-infected subjects, is consistent with this
hypothesis. Further studies to examine functional differ-
ences between Iranian and Japanese HTLV-1 Tax proteins
will provide important information to clarify this point.
The interaction between different genes and/or environ-
mental factors is also likely to contribute to the observed
differences between the two populations. For example, co-
infection with Strongyloides stercoralis (Gabet et al., 2000)
can affect the HTLV-1 provirus load. In Japan, S. stercoralis
infection is endemic in the southwestern islands Amami
and Ryukyu, but is rarely reported on the mainland
including Kagoshima (Arakaki et al., 1992). However,
there are no data on the prevalence of S. stercoralis infec-
tion in Mashhad, Iran, and therefore future epidemiolog-
ical studies are necessary to clarify this possibility.
It seems likely that the same evolutionary selection pres-
sures that induce polymorphisms in ‘infection-resisting
genes’ have contributed to marked allele-frequency differ-
ences at the same loci. When geographical variation in
pathogen polymorphism is superimposed on this host
genetic heterogeneity, considerable variation in detectable
allelic associations is likely to result in the different
http://vir.sgmjournals.org 779
Risk factors for Iranian HAM/TSP
populations. In other words, genetic resistance to infec-
tious diseases that is formed by complex host genetic effects
is complicated further by pathogen diversity and environ-
mental factors. Considering this background of complexity,
the most practical approach to finding reliable results may
be first to identify disease-associated genes in a single large
population, and secondly to analyse subsequently whether
a similar effect is found in other ethnic populations, as we
have shown in this study.
ACKNOWLEDGEMENTS
We thank the staff of the Blood Transfusion Center in Mashhad and
Neyshabour, the personnel of the Bu-Ali Research Institute and the
Faculty of Pharmacology in Mashhad University, and Dr Mahbubeh
Naghibzadeh Bajestan for their cooperation, Professor Charles R. M.
Bangham of Imperial College, London, for critical reading and
comments on the manuscript, and Ms Tomoko Muramoto and Yoko
Nishino of Kagoshima University for their excellent technical assis-
tance. This work was supported by the Grant in Aid for Research on
Brain Science of the Ministry of Health, Labor and Welfare, Japan.
REFERENCES
Abbaszadegan, M. R., Gholamin, M., Tabatabaee, A., Farid, R.,
Houshmand, M. & Abbaszadegan, M. (2003). Prevalence of human
T-lymphotropic virus type 1 among blood donors from Mashhad,
Iran. J Clin Microbiol 41, 2593–2595.
Apple, R. J., Erlich, H. A., Klitz, W., Manos, M. M., Becker, T. M.
& Wheeler, C. M. (1994). HLA DR-DQ associations with cervical
carcinoma show papillomavirus-type specificity. Nat Genet 6,
157–162.
Arakaki, T., Kohakura, M., Asato, R., Ikeshiro, T., Nakamura, S. &
Iwanaga, M. (1992). Epidemiological aspects of Strongyloides
stercoralis infection in Okinawa, Japan. J Trop Med Hyg 95, 210–213.
Bangham, C. R. (2000). The immune response to HTLV-1. Curr
Opin Immunol 12, 397–402.
Biggar, R. J., Saxinger, C., Gardiner, C., Collins, W. E., Levine, P. H.,
Clark, J. W., Nkrumah, F. K. & Blattner, W. A. (1984). Type-I HTLV
antibody in urban and rural Ghana, West Africa. Int J Cancer 34,
215–219.
Blattner, W. A., Kalyanaraman, V. S., Robert-Guroff, M. & 7 other
authors (1982). The human type-C retrovirus, HTLV, in Blacks
from the Caribbean region, and relationship to adult T-cell
leukemia/lymphoma. Int J Cancer 30, 257–264.
Bunce, M., O’Neill, C. M., Barnardo, M. C., Krausa, P., Browning,
M. J., Morris, P. J. & Welsh, K. I. (1995). Phototyping: comprehen-
sive DNA typing for HLA-A, B, C, DRB1, DRB3, DRB4, DRB5 &
DQB1 by PCR with 144 primer mixes utilizing sequence-specific
primers (PCR-SSP). Tissue Antigens 46, 355–367.
Cartier, L., Araya, F., Castillo, J. L. & 8 other authors (1993).
Southernmost carriers of HTLV-I/II in the world. Jpn J Cancer Res
84, 1–3.
Furukawa, Y., Yamashita, M., Usuku, K., Izumo, S., Nakagawa, M.
& Osame, M. (2000). Phylogenetic subgroups of human T cell
lymphotropic virus (HTLV) type I in the tax gene and their
association with different risks for HTLV-1-associated myelopathy/
tropical spastic paraparesis. J Infect Dis 182, 1343–1349.
Gabet, A. S., Mortreux, F., Talarmin, A. & 7 other authors (2000).
High circulating proviral load with oligoclonal expansion of HTLV-1
bearing T cells in HTLV-1 carriers with strongyloidiasis. Oncogene
19, 4954–4960.
Gessain, A., Barin, F., Vernant, J. C., Gout, O., Maurs, L., Calender, A.
& de The, G. (1985). Antibodies to human T-lymphotropic virus
type-I in patients with tropical spastic paraparesis. Lancet ii, 407–410.
Hill, A. V. (1998). The immunogenetics of human infectious diseases.
Annu Rev Immunol 16, 593–617.
Hinuma, Y., Nagata, K., Hanaoka, M., Nakai, M., Matsumoto, T.,
Kinoshita, K. I., Shirakawa, S. & Miyoshi, I. (1981). Adult
T-cell leukemia: antigen in an ATL cell line and detection of
antibodies to the antigen in human sera. Proc Natl Acad Sci U S A
78, 6476–6480.
Ijichi, S., Izumo, S., Eiraku, N. & 8 other authors (1993). An
autoaggressive process against bystander tissues in HTLV-1-infected
individuals: a possible pathomechanism of HAM/TSP. Med
Hypotheses 41, 542–547.
Jeffery, K. J., Usuku, K., Hall, S. E. & 14 other authors (1999). HLA
alleles determine human T-lymphotropic virus-I (HTLV-I) proviral
load and the risk of HTLV-1-associated myelopathy. Proc Natl Acad
SciUSA96, 3848–3853.
Jeffery, K. J., Siddiqui, A. A., Bunce, M. & 8 other authors (2000).
The influence of HLA class I alleles and heterozygosity on the
outcome of human T cell lymphotropic virus type I infection.
J Immunol 165, 7278–7284.
Kaplan, J. E., Osame, M., Kubota, H., Igata, A., Nishitani, H.,
Maeda, Y., Khabbaz, R. F. & Janssen, R. S. (1990). The risk of
development of HTLV-1-associated myelopathy/tropical spastic
paraparesis among persons infected with HTLV-1. J Acquir
Immune Defic Syndr 3, 1096–1101.
Kaplan, J. E., Khabbaz, R. F., Murphy, E. L. & 12 other authors
(1996). Male-to-female transmission of human T-cell lymphotropic
virus types I and II: association with viral load. The Retrovirus
Epidemiology Donor Study Group. J Acquir Immune Defic Syndr
Hum Retrovirol 12, 193–201.
Kitze, B., Usuku, K., Yamano, Y., Yashiki, S., Nakamura, M.,
Fujiyoshi, T., Izumo, S., Osame, M. & Sonoda, S. (1998). Human
CD4
+
T lymphocytes recognize a highly conserved epitope of
human T lymphotropic virus type 1 (HTLV-1) env gp21 restricted by
HLA DRB1*0101. Clin Exp Immunol 111, 278–285.
Miura, T., Fukunaga, T., Igarashi, T. & 7 other authors (1994).
Phylogenetic subtypes of human T-lymphotropic virus type I and
their relations to the anthropological background. Proc Natl Acad Sci
USA91, 1124–1127.
Moritoyo, T., Reinhart, T. A., Moritoyo, H., Sato, E., Izumo, S.,
Osame, M. & Haase, A. T. (1996). Human T-lymphotropic virus
type I-associated myelopathy and tax gene expression in CD4
+
T
lymphocytes. Ann Neurol 40, 84–90.
Nagai, M., Usuku, K., Matsumoto, W. & 8 other authors (1998).
Analysis of HTLV-1 proviral load in 202 HAM/TSP patients and 243
asymptomatic HTLV-1 carriers: high proviral load strongly pre-
disposes to HAM/TSP. J Neurovirol 4, 586–593.
Olerup, O. & Zetterquist, H. (1992). HLA-DR typing by PCR
amplification with sequence-specific primers (PCR-SSP) in 2 hours:
an alternative to serological DR typing in clinical practice including
donor–recipient matching in cadaveric transplantation. Tissue
Antigens 39, 225–235.
Osame, M. (1990). Review of WHO Kagoshima meeting and
diagnostic guidelines for HAM/TSP. In Human Retrovirology: HTLV,
pp. 191–197. Edited by W. A. Blattner. New York: Raven Press.
Osame, M., Usuku, K., Izumo, S., Ijichi, N., Amitani, H., Igata, A.,
Matsumoto, M. & Tara, M. (1986). HTLV-1 associated myelopathy, a
new clinical entity. Lancet i, 1031–1032.
780 Journal of General Virology 86
A. H. Sabouri and others
Poiesz, B. J., Ruscetti, F. W., Gazdar, A. F., Bunn, P. A., Minna, J. D.
& Gallo, R. C. (1980). Detection and isolation of type C retrovirus
particles from fresh and cultured lymphocytes of a patient with
cutaneous T-cell lymphoma. Proc Natl Acad Sci U S A 77, 7415–7419.
Safai, B., Huang, J. L., Boeri, E., Farid, R., Raafat, J., Schutzer, P.,
Ahkami, R. & Franchini, G. (1996). Prevalence of HTLV type I
infection in Iran: a serological and genetic study. AIDS Res Hum
Retroviruses 12, 1185–1190.
Seiki, M., Hattori, S., Hirayama, Y. & Yoshida, M. (1983). Human
adult T-cell leukemia virus: complete nucleotide sequence of the
provirus genome integrated in leukemia cell DNA. Proc Natl Acad
SciUSA80, 3618–3622.
Takenouchi, N., Yamano, Y., Usuku, K., Osame, M. & Izumo, S.
(2003). Usefulness of proviral load measurement for monitoring of
disease activity in individual patients with human T-lymphotropic
virus type I-associated myelopathy/tropical spastic paraparesis.
J Neurovirol 9, 29–35.
Umehara, F., Izumo, S., Nakagawa, M., Ronquillo, A. T.,
Takahashi, K., Matsumuro, K., Sato, E. & Osame, M. (1993).
Immunocytochemical analysis of the cellular infiltrate in the spinal
cord lesions in HTLV-1-associated myelopathy. J Neuropathol Exp
Neurol 52, 424–430.
Woolf, B. (1955). On estimating the relationship between blood
group and disease. Ann Hum Genet 19, 251–253.
Yamano, Y., Kitze, B., Yashiki, S. & 7 other authors (1997).
Preferential recognition of synthetic peptides from HTLV-1 gp21
envelope protein by HLA-DRB1 alleles associated with HAM/TSP
(HTLV-1-associated myelopathy/tropical spastic paraparesis).
J Neuroimmunol 76, 50–60.
Yamashita, M., Achiron, A., Miura, T. & 7 other authors (1995).
HTLV-I from Iranian Mashhadi Jews in Israel is phylogenetically
related to that of Japan, India, and South America rather than to that
of Africa and Melanesia. Virus Genes 10, 85–90.
Yanagihara, R., Jenkins, C. L., Alexander, S. S., Mora, C. A. &
Garruto, R. M. (1990). Human T lymphotropic virus type I infec-
tion in Papua New Guinea: high prevalence among the Hagahai
confirmed by western analysis. J Infect Dis 162, 649–654.
Yoshida, M., Miyoshi, I. & Hinuma, Y. (1982). Isolation and
characterization of retrovirus from cell lines of human adult T-cell
leukemia and its implication in the disease. Proc Natl Acad Sci U S A
79, 2031–2035.
Yoshida, M., Seiki, M., Yamaguchi, K. & Takatsuki, K. (1984).
Monoclonal integration of human T-cell leukemia provirus in all
primary tumors of adult T-cell leukemia suggests causative role of
human T-cell leukemia virus in the disease. Proc Natl Acad Sci U S A
81, 2534–2537.
Zamora, T., Zaninovic, V., Kajiwara, M., Komoda, H., Hayami, M. &
Tajima, K. (1990). Antibody to HTLV-1 in indigenous inhabitants of
the Andes and Amazon regions in Colombia. Jpn J Cancer Res 81,
715–719.
Zaninovic, V., Sanzon, F., Lopez, F. & 9 other authors (1994).
Geographic independence of HTLV-I and HTLV-II foci in the Andes
highland, the Atlantic coast, and the Orinoco of Colombia. AIDS
Res Hum Retroviruses 10, 97–101.
http://vir.sgmjournals.org 781
Risk factors for Iranian HAM/TSP
