MAJOR ARTICLE

Higher Human T Lymphotropic Virus (HTLV)

Provirus Load Is Associated with HTLV-I
versus HTLV-II, with HTLV-II Subtype A
versus B, and with Male Sex and a History
of Blood Transfusion
Edward L. Murphy,1,2 Tzong-Hae Lee,2 Daniel Chafets,2 Catharie C. Nass,3,a Baoguang Wang,4 Katharine Loughlin,4
and Donna Smith,4 for the HTLV Outcomes Study Investigatorsb

Downloaded from http://jid.oxfordjournals.org/ by guest on June 26, 2015

1
Departments of Laboratory Medicine and Epidemiology/Biostatistics, University of California, San Francisco, and 2Blood Systems Research
Institute, San Francisco; 3American Red Cross Blood Services, Chesapeake and Potomac Region, Baltimore, and 4Westat, Rockville, Maryland

Background. High human T lymphotropic virus (HTLV)–I provirus load (VL) has been associated with an
increased risk of HTLV-associated myelopathy, but little is known about variation in HTLV-I or -II VLs by demo-
graphic characteristics and risk behaviors.
Methods. We measured HTLV-I and HTLV-II VLs in a large cohort of 127 HTLV-I–seropositive and 328
HTLV-II–seropositive former blood donors, by use of real-time polymerase chain reaction using tax primers.
Multivariable linear regression was used to control for confounding by relevant covariates.
Results. The mean VLs were 3.28 log10 copies/106 peripheral blood mononuclear cells (PBMCs) (range, 0.5–
5.3 log10 copies/106 PBMCs) for HTLV-I and 2.60 log10 copies/106 PBMCs (range, 0.05–5.95 log10 copies/106 PBMCs)
for HTLV-II (P ! .0001). HTLV-II VLs were higher in those subjects with subtype A infection (mean, 2.82 log10
copies/106 PBMCs) than in those with subtype B infection (mean, 2.29 log10 copies/106 PBMCs) (P p .005). Higher
HTLV-I VL was associated with previous receipt of a blood transfusion (P p .04 ), and lower HTLV-II VL was
associated with female sex (P p .007 ). These associations persisted in virus-specific multivariate linear regression
models controlling for potential confounding variables.
Conclusions. VL was significantly higher in HTLV-I than in HTLV-II infection and was higher in HTLV-II
subtype A than in HTLV-II subtype B infection. Chronic HTLV VLs may be related to the infectious dose acquired
at the time of infection, with higher VLs following acquisition by blood transfusion and lower VLs following
sexual acquisition.

Human T lymphotropic virus (HTLV) types I and II
are human type C retroviruses derived most likely from
simian-to-human transmission thousands of years ago
[1, 2]. The 2 HTLV types share ∼60% nucleotide ho-
Received 24 December 2003; accepted 18 February 2004; electronically published
6 July 2004.
Financial support: National Heart, Lung, and Blood Institute Retrovirus Epi-
demiology Donor Study (research contracts N01-HB-97077 [superseded by N01-HB-
47114], N01-HB-97078, N01-HB-97079, N01-HB-97080, N01-HB-97081, N01-HB-
97082, and R01-HL-62235 [through the HTLV Outcomes Study]).
a
Present affiliation: C.C.N. is retired from the American Red Cross Blood Services,
Chesapeake and Potomac Region.
b
HOST investigators are listed after the text.
Reprints or correspondence: Dr. Edward L. Murphy, c/o Blood Centers of the Pacific,
UCSF, 270 Masonic Ave., San Francisco, CA 94118 (murphy@itsa.ucsf.edu).
The Journal of Infectious Diseases 2004; 190:504–10
 2004 by the Infectious Diseases Society of America. All rights reserved.
0022-1899/2004/19003-0012$15.00
mology but differ in their epidemiology and disease
associations. HTLV-I is prevalent in Africa, the Carib-
bean, Brazil, and southern Japan and causes adult T
cell leukemia (ATL) and HTLV-associated myelopathy
(HAM) [3]. HTLV-II is prevalent among Amerindians,
African pygmies, and injection drug users (IDUs) in the
Americas and Europe and has been linked to HAM, as
well as to a higher occurrence of pneumonia and bron-
chitis [4, 5]. HTLV-II subtype A is more prevalent among
North American and European IDUs and their sex part-
ners, whereas subtype B is more prevalent among Native
Americans [4].
HTLV-I and -II viral RNA is low to undetectable in
human serum, and viral propagation is thought to oc-
cur via the clonal expansion of lymphocytes with in-
tegrated proviral DNA [6] or the transfer of viral RNA

504 • JID 2004:190 (1 August) • Murphy et al.

via direct contact between infected and uninfected cells [7].
How the level of proviral DNA in lymphocytes is regulated is
unknown but may involve either the activity of the viral transac-
tivator protein Tax, which differs between HTLV-I and some
HTLV-II subtypes [8], or the degree of lymphocyte turnover
in the infected host [6]. High levels of HTLV-I proviral DNA,
HTLV-I Tax mRNA, and anti-HTLV antibody have been as-
sociated with the occurrence of HAM [9, 10]. However, the
pathogenic mechanism is unresolved and may be due to either
direct viral neurotoxicity [11], an autoimmune response di-
rected at viral proteins homologous to neuronal proteins [12],
or “bystander” immunologic damage to neural tissue [13]. Data
are less clear with regard to the association between HTLV-I
provirus load (VL) and ATL, with one study showing signifi-
cantly higher prediagnostic VLs in patients with ATL than in
asymptomatic carriers [14] and another showing an association
between high HTLV-I VLs and the presence of abnormal lym-
phocytes thought to be indicative of preleukemia, although this
was not proven prospectively [15].
HTLV-II VL has received comparatively less study than HTLV-
I VL. One study found no difference in HTLV-II VLs between
IDUs and endemic African Pygmies and no correlation between
HTLV-II VLs and CD8+ lymphocyte counts or stage of AIDS,
in the IDUs who were coinfected with HIV [16]. Other studies
have found high HTLV-II VLs in patients with HIV/HTLV-II
coinfection [17, 18] and, notably, in coinfected patients with
sensory neuropathy [19]. HTLV-II VLs also increased after the
initiation of antiretroviral therapy for HIV in coinfected pa-
tients [20, 21].
HTLV-II VL has not been correlated with age or sex, and
neither HTLV-I VL nor HTLV-II VL has been studied in relation
to the route of infection. We therefore measured HTLV-I and
HTLV-II VLs in a well-characterized cohort of former blood
donors and correlated these levels with data on demographic
characteristics, risk factors for infection, and HTLV-II subtype.
SUBJECTS AND METHODS
Study design and population. The present study was a cross-
sectional analysis of blood samples obtained at the baseline visit
of the HTLV Outcomes Study cohort, formerly known as the
Retrovirus Epidemiology Donor Study HTLV cohort. The study
protocol was approved by the Committee on Human Research
at the University of California, San Francisco, and by institu-
tional review boards at other participating institutions. En-
rollment criteria and methods have been reported in detail
elsewhere [22–24]. In brief, blood donors found to be HTLV
seropositive during predonation screening at 5 centers across
the United States in 1988–1992 had confirmation of seropos-
itivity performed by use of Western blot, followed by HTLV
typing by use of type-specific antibody or polymerase chain
reaction (PCR) assays [25]. All subjects were seronegative for
HIV at baseline. Demographic, risk factor, and health history
information was obtained by interview with trained study co-
ordinators. HTLV-II subtypes were determined previously by
use of restriction fragment–length polymorphism analysis [26].
Laboratory methods. Peripheral blood mononuclear cell
(PBMC) samples were stored at ⫺70C until used, and PBMCs
were digested in a PCR solution with proteinase K. Quantitation
of proviral DNA of HTLV-I and HTLV-II was performed by
use of real-time PCR. We used a single set of primers for both
HTLV-I and -II, from highly conserved sequences of the viral
tax regions, designated as HTLV-F5 (5-CGG ATA CCC IGT
CTA CGT GTT T-3) and HTLV-R4 (5-CTG AGC IGA IAA
CGC GTC CA-3). To quantitate the cellular input for each
reaction, HLA-DQ-a copy number was measured separately.
The primers used for HLA-DQ-a were GH26 (5-GTG CTC
CAG GTG TAA ACT TGT ACC AG-3) and GH27 (5-CAC
GGA TCC GGT AGC AGC GGT AGA GTT G-3). For each
Downloaded from http://jid.oxfordjournals.org/ by guest on June 26, 2015
sample, 25 mL of DNA lysate was added to the PCR mixture
(5 mL for the HLA-DQ-a reactions). Real-time PCR was per-
formed by use of the GeneAmp 5700 machine (Applied Bio-
systems), with the following cycle conditions: 10 min at 95C,
followed by 45 cycles of 30 s at 95C, 30 s at 64C, and 45 s
at 72C. Reactions were performed in triplicate [27].
Fluorescense intensity of Syber green incorporated into the
amplified product was measured at every PCR cycle, and the
threshold cycle number of each sample was recorded. For each
run, a standard curve was generated from 1:10 serial dilutions
of MT2 (HTLV-I; obtained from American Type Culture Col-
lection) or MoT (HTLV-II) cell lines, with a range of 10⫺1–104
copies/reaction. Mean HTLV copy numbers for these cell lines
were 2.1 HTLV-I copies/cell for MT2, as determined elsewhere
[28], and 11.7 HTLV-II copies/cell for MoT, determined ex-
perimentally using HTLV-II plasmids of known concentration
(provided by J. Kropp, Gladstone Institute of Virology and Im-
munology, San Francisco, CA, and E. Wattel, Centre Leon Ber-
nard, Lyon, France). The number of copies in the test sample
was then calculated by interpolation of the experimentally de-
termined threshold cycle number onto the control standard
regression curve. To determine the VL of each sample, the
number of copies of HTLV-I or HTLV-II was divided by the
HLA-DQ-a copy number. The lower limit of detection for the
assay was 1 copy/105 cells. Reproducibility of the assay was
measured by testing 40 specimens in duplicate and comparing
the results, by use of linear regression with R 2 p 0.97.
Statistical analysis. VLs, expressed as HTLV-I or -II copies
per 106 PBMCs, were log10 transformed for all analyses, to
approximate the normal distribution. Undetectable VLs were
assigned the value of 1 copy/106 cells, which is below the lower
limit of detection of the VL assay, before log10 transformation.
Means, SEs, and ranges were calculated for each HTLV type
and in subsets defined by demographic and risk factor variables.
HTLV-I and -II Provirus Load • JID 2004:190 (1 August) • 505
 Table 1. Demographic and risk factor profile of the HTLV-II group. A greater proportion of subjects with HTLV-
study population from the baseline visit of the HTLV I than subjects with HTLV-II had !4 lifetime sex partners,
Outcomes Study.
whereas a greater proportion of subjects with HTLV-II admitted
Variable HTLV-I HTLV-II
to having had 120 partners. A history of IDU was reported by
25% of the subjects with HTLV-II (48 men and 36 women),
All subjects 127 (100) 328 (100)
Age, years
although only ∼1% said they currently injected drugs. More
!20 2 (2) 2 (1) than half of the subjects with HTLV-II (37 men and 145 wom-
20–29 9 (7) 19 (6) en) reported (yes or likely) sexual intercourse with an IDU. One-
30–39 25 (20) 144 (44) third of the HTLV-I group and one-fifth of the HTLV-II group
40–49 55 (43) 110 (34) had a history of blood transfusion; in comparison, !10% of
50–59 16 (13) 35 (11)
HTLV-seronegative blood donors enrolled in the cohort report-
⭓60 20 (16) 18 (5)
Sex ed such a history.
Female 93 (73) 243 (74) VLs for both viral types ranged from 0.05 to 5.95 log10 copies/
6
Male 34 (27) 85 (26) 10 PBMCs. HTLV-I VLs (mean [SE], 3.28 [0.12] log10 copies/
Race/ethnicitya 106 PBMCs) were significantly higher than HTLV-II VLs (mean
White 46 (36) 112 (34)
[SE], 2.60 [0.09] log10 copies/106 PBMCs; P ! .0001; figure 1A
Black 54 (43) 107 (33)
Hispanic 6 (5) 86 (26)
and 1B). Among subjects with HTLV-II, HTLV-II VLs were
significantly higher in the 190 subjects with subtype A (mean

Downloaded from http://jid.oxfordjournals.org/ by guest on June 26, 2015

Asian 15 (12) 6 (2)
Other/unspecified 4 (3) 14 (4) [SE], 2.82 [0.11] log10 copies/106 PBMCs), compared with the
Lifetime no. of sex partnersa 138 subjects with subtype B (mean [SE], 2.29 [0.15] log10 cop-
!4 42 (33) 73 (22) ies/106 PBMCs; P p .005; figure 2D). All analyses by demo-
5–8 31 (24) 81 (25)
graphic and risk factor variables were performed separately for
9–20 38 (30) 89 (27)
⭓21 14 (11) 78 (24) HTLV-I and for HTLV-II.
History of IDUa HTLV-I VLs were lower in white (mean [SE], 3.03 [0.22]
No 125 (98) 243 (74) log10 copies/106 PBMCs) than in nonwhite (mean [SE], 3.48
Yes 1 (1) 84 (26) [0.14] log10 copies/106 PBMCs) subjects, although this differ-
History of sex with an IDUa
ence was not statistically significant (P p .08 ; figure 2A). HTLV-
No/unlikely 104 (82) 136 (42)
Yes/likely 10 (8) 182 (55)
I VLs were higher in those with a history of blood transfusion
History of blood transfusiona (mean [SE], 3.62 [0.14] log10 copies/106 PBMCs) than in those
No 85 (67) 251 (77) without a history of blood transfusion (mean [SE], 3.1 [0.16]
Yes 42 (33) 69 (21) log10 copies/106 PBMCs; P p .04; figure 2B). HTLV-I VLs did
NOTE. Data are no. (%) of subjects. IDU, injection drug not differ significantly by age, sex, number of sex partners,
user.
a
Nos. may not add to totals, because of missing data (!10%
per variable).

Differences in log10 VLs between groups were tested by use of

2-sample t tests. Associations between log10 VLs and each con-
tinuous variable were tested by use of separate linear regression
models. Finally, multivariate linear regression models were con-
structed to assess independent associations between variables
found to be significantly associated in the univariate analyses.
All analyses were performed using SAS statistical software (ver-
sion 8.2; SAS Institute).

RESULTS

VLs and epidemiologic data were available for 127 subjects with
HTLV-I and 328 subjects with HTLV-II (∼85 % of the HTLV
cohort). Demographic and risk factor characteristics of the sub-
jects are presented in table 1. Most subjects were middle-aged
and women, and there was a substantial proportion of black
persons in both HTLV groups and of Hispanic persons in the
Figure 1. Difference in human T lymphotropic virus (HTLV) provirus load
(VL) between HTLV-I– and HTLV-II–infected subjects (P ! .0001), from the
baseline visit of the HTLV Outcomes Study (HOST). VL is presented as
median (central line), 25th and 75th percentiles (box boundaries), and 10th
and 90th percentiles (error bars). PBMCs, peripheral blood mononuclear
cells.

506 • JID 2004:190 (1 August) • Murphy et al.

 Downloaded from http://jid.oxfordjournals.org/ by guest on June 26, 2015
Figure 2. Human T lymphotropic virus (HTLV) provirus load (VL) presented as median (central line), 25th and 75th percentiles (box boundaries), and
10th and 90th percentiles (error bars). A, HTLV-II VL by HTLV-II subtype A vs. B (P p .005). B, HTLV-II VL by sex (P p .007). C, HTLV-I VL by nonwhite
vs. white race/ethnicity (P p .08). D, HTLV-I VL by history of blood transfusion (P p .04). PBMCs, peripheral blood mononuclear cells.

history of IDU, or history of sex with an IDU. In the multi-
variate linear regression analysis, HTLV-I VL retained its as-
sociation with a history of blood transfusion (b p 0.68; P p
.02) but had only a borderline association with white race (b
p ⫺0.51; P p .06).
HTLV-II VLs were significantly lower among women (mean
[SE], 2.45 [0.10] log10 copies/106 PBMCs) than among men
(mean [SE], 3.02 [0.18] log10 copies/106 PBMCs; P p .007; figure
2C). There was no association between HTLV-II VL and age,
race, number of sex partners, history of IDU, or history of sex
with an IDU. In the multivariate linear regression analysis, HTLV-
II VL remained independently and inversely associated with fe-
male sex (b p ⫺0.68; P p .01) and positively associated with
HTLV-II subtype A (b p 0.54; P p .008). In an analysis re-
stricted to women only, there was an inverse association be-
tween HTLV-II VL and increasing number of sex partners (b
p ⫺0.22; P p .03), which did not persist after adjusting for
other covariates (b p ⫺0.18; P p .12), although the association
with HTLV-II subtype A remained (b p 0.65; P p .004).
DISCUSSION
The present study of HTLV VLs at the baseline visit of a large
cohort of generally asymptomatic former blood donors found
that HTLV-I VLs were significantly higher than HTLV-II VLs
and that HTLV-II VLs were significantly higher in subtype A
than in subtype B infections. Analyses by demographic and
behavioral characteristics revealed higher HTLV-I VLs in those
subjects with a history of blood transfusion and borderline
lower HTLV-I VLs in white subjects. HTLV-II VLs were sig-
nificantly lower in women than in men. There was a borderline
association of lower VLs with increasing numbers of sex part-
ners, but this did not persist after adjusting for covariates.
HTLV-I VLs have been reported in numerous studies, rang-
ing from !1 copy/10,000 lymphocytes to 11 copy/10 lympho-
cytes [6, 9, 10, 14]. The results of the present study are con-
sistent with those of previous reports, with a mean HTLV-I VL
of ∼1000 copies/106 PBMCs and a range of 1–100,000 copies/
106 PBMCs. By use of an end-point dilution PCR assay, a
previous study revealed VLs of 20–200,000 copies/106 PBMCs
in 49 subjects with HTLV-II, with and without HIV coinfection
[17]. A competitive PCR assay found 50–162,390 copies/106
PBMCs in Italian IDUs [16], and those with the highest VLs
were found to have clonal expansion of HTLV-II–infected lym-
phocytes [29]. These VL results are also similar to those of the
present study.
In the only other comparative study of HTLV-I and HTLV-

HTLV-I and -II Provirus Load • JID 2004:190 (1 August) • 507

II VLs that we were able to find, Kaplan et al. reported that
HTLV-I VLs were significantly higher than HTLV-II VLs [27].
Their study included a subset of individuals from the present
study but used a different assay to measure VLs. The present
report is the only published in vivo study of HTLV-II VLs
according to HTLV-II subtype. Differences in VLs between
HTLV-I and HTLV-II, and between HTLV-II subtype A and
subtype B, may be related to differences in tax gene structure
and protein expression among the HTLV types and subtypes.
HTLV-II Tax has been shown to have a lower transforming
activity than HTLV-I Tax [30], which may account for the lack
of leukemogenesis by HTLV-II. Compared with HTLV-II sub-
type B, HTLV-II subtype A has less ability to inhibit p53 func-
tion in T lymphocytes [8, 31] and to induce CREB- and NF-
kB–mediated transactivation [32]. These in vitro data would
suggest lower VLs in HTLV-II subtype A infection, which is
the opposite of our finding. One explanation may be that tax
genes from wild-type HTLV-II subtype A isolates may have
different biologic activity than the HTLV-II prototypes com-
monly used for in vitro experiments [31–33]. It is also possible
that differences in VLs between HTLV types and subtypes are
due to differences in modes of transmission or to host factors,
and not to biological characteristics of each virus.
The association of higher HTLV-I VLs with a history of blood
transfusion is the most suggestive evidence we found that the
route of infection affects HTLV VLs. Fully 34% of our HTLV-
I cohort had received a blood transfusion, compared with 10%
of our HTLV-seronegative comparison group, which is consis-
tent with our previous report of blood transfusion as a risk
factor for infection (adjusted odds ratio, 4.5; 95% confidence
interval, 2.6–7.8) [34]. These same subjects with HTLV-I and
a history of blood transfusion have now been shown to have
higher VLs than subjects with HTLV-I without this risk factor.
Blood transfusion is associated with a higher dose of HTLV-I–
infected lymphocytes, so it is biologically plausible that VLs are
higher. Since VLs have been shown to remain relatively constant
over time [35], one may infer that a higher infecting dose from
blood transfusion translates to a higher persistent VL, such as
those detected in our subjects. Alternatively, altered immune
function from comorbid illness or severe trauma at the time
of blood transfusion could have affected the VL set point. Our
data are consistent with those of a previous report of median
HTLV-I VLs of 100–200 copies/105 PBMCs in the first year
after blood transfusion–transmitted infection in a Jamaican
study [14]. However, in the same study, asymptomatic carriers
had higher HTLV-I VLs (median, ∼1000 copies/105 PBMCs)
than did asymptomatic carriers in the present study, perhaps
because of a higher proportion of mother-to-child infection.
HTLV-II VLs were significantly lower in women than in men.
Two previous studies of HTLV-I VLs found no differences by
sex among asymptomatic carriers [9, 36], but we were unable
508 • JID 2004:190 (1 August) • Murphy et al.
to find other published reports of differences in HTLV-II VLs
by sex. Other reports found that women with HAM have either
higher or lower HTLV-I VLs than do men with HAM [9, 10].
Women are clearly overrepresented among both patients with
HTLV-I HAM and patients with HTLV-II HAM [37]. Previous
epidemiological data suggest that sexual acquisition of HTLV-
I is associated with development of HAM, whereas mother-to-
child acquisition is associated with ATL [38, 39].
We therefore hypothesize that women are more likely to
acquire HTLV-II sexually, with a smaller infectious dose, and,
hence, to have lower VLs than men, most of whom acquire
the virus by parenteral (IDU or blood transfusion) or mother-
to-child routes. Against this hypothesis are data from the pres-
ent study, which has showed no association between HTLV-II
VLs and the lifetime number of sex partners or history of sex
with an IDU, for both sexes. However, in an analysis limited
to women, we did find a nonsignificant trend toward lower
Downloaded from http://jid.oxfordjournals.org/ by guest on June 26, 2015
HTLV-II VLs with increasing number of lifetime sex partners.
Retrospective interview data may not be the most accurate in
assigning the actual mode of transmission. In support of the
hypothesis are data showing lower mean HTLV-I VLs in 11
cases of sexually acquired HTLV-I, compared with the 11 trans-
mitting persons [40].
That infectious dose may predict the chronic set point for
VLs is consistent with the observation that expansion of HTLV
within an infected human is predominantly due to clonal ex-
pansion of lymphocytes with integrated HTLV provirus [6].
Thus, the number of initially infected lymphocytes would in-
fluence the number of lymphocyte clones, although the size of
each clone may depend on factors that cause multiplication of
lymphocytes. Recent data from the Wattel laboratory support
this hypothesis, in that humans with presumed breast-milk
acquisition of HTLV-II as infants had increased clonality and
VLs, compared with those with risk factors suggesting infection
acquired later in life [41]. An alternative explanation is that
infection during infancy or childhood produces higher VLs
because of a weaker initial immune response.
The borderline association we found between lower HTLV-
I VLs and white race may be explained either by genetic back-
ground or by route of infection associated with a particular
racial group. Bangham et al. have previously reported that
HTLV-I VLs may be correlated with specific HLA subtypes [42]
and that these subtypes are also associated with an increased
risk of developing HAM. Since HLA phenotypes differ by racial
origin, it is conceivable that the race/ethnicity association we
observed is due to genetic background. On the other hand,
other host characteristics, such as route of HTLV infection,
lymphocyte activation, and concomitant infections, may also
differ by race. It has been recognized that the incidence of
HTLV-associated diseases, such as ATL, may differ between
endemic Japanese and Jamaican populations, perhaps because
of differences in biologic response to infection [23]. Further-
more, comparative studies across race and ethnicity populations
using similar methodologies to measure VL and controlling for
route of infection and HLA type should help resolve this issue.
Strengths of the present study include the use of a well-
defined, generally asymptomatic group of human subjects and
a sensitive, reproducible, and high-throughput assay to measure
VLs. Potential weaknesses include the relatively large variability
of VLs and the fact that the differences in VLs, although statis-
tically significant, were of relatively small magnitude, compared
with the interpersonal range of observed values. Determining
causality for the associations we observed is also problematic in
this cross-sectional study, since different characteristics of the sub-
jects (i.e., race/ethnicity and mode of acquisition) may be con-
founded. That associations persisted after multivariable analysis
is reassuring but not definitive.
In conclusion, we have demonstrated associations between
higher HTLV-I VLs and history of blood transfusion and, per-
haps, nonwhite race and between lower HTLV-II VLs and fe-
male sex. Discovering the biological determinants of VL is im-
portant because the incidence of HTLV-associated diseases
seems to be related to VL. Although reasons for the race/eth-
nicity association remain speculative, we hypothesize that as-
sociations with sex and history of blood transfusion result from
a biological relationship between infectious dose of HTLV-I and
HTLV-II and the chronic VL measured in an individual. Further
studies in humans with well-defined time and mode of infection
or experimental studies in animals are required to investigate
this hypothesis.
HTLV OUTCOMES STUDY (HOST) MEMBERS
The HOST is presently the responsibility of the following persons.
Study headquarters. University of California, San Fran-
cisco (San Francisco, CA): E. L. Murphy (Principal Investigator)
and J. Engstrom.
Blood centers. American Red Cross Blood Services Greater
Chesapeake and Potomac Region (Baltimore, MD): C. C. Nass,
C. Conry-Cantilena, and J. Gibble; American Red Cross Blood
Services Southeastern Michigan Region (Detroit, MI): B. New-
man; American Red Cross Blood Services Southern California
Region (Los Angeles, CA): G. Garratty, S. Hutching, and A.
Ziman; Blood Centers of the Pacific (San Francisco, CA): M.
P. Busch; and Oklahoma Blood Institute (Oklahoma City, OK):
J. W. Smith and E. Moore.
Medical coordinating center. Westat (Rockville, MD): G.B.
Schreiber, D. Ameti, and B. Wang.
Central laboratory. Blood Centers of the Pacific (San Fran-
cisco, CA): M. P. Busch and L. H. Tobler.
Diagnostic review panel. E. L. Murphy, R. Sacher, and J.
Fridey.
References
1. Cann AJ, Chen ISY. Human T-cell leukemia virus types I and II. In:
Fields B, ed. Virology. 2nd ed. New York: Raven Press, 1990:1501–27.
2. Slattery JP, Franchini G, Gessain A. Genomic evolution, patterns of
global dissemination, and interspecies transmission of human and sim-
ian T-cell leukemia/lymphotropic viruses. Genome Res 1999; 9:525–40.
3. Manns A, Hisada M, La Grenade L. Human T-lymphotropic virus type
I infection. Lancet 1999; 353:1951–8.
4. Hall WW, Ishak R, Zhu SW, et al. Human T lymphotropic virus type
II (HTLV-II): epidemiology, molecular properties, and clinical features
of infection. J Acquir Immune Defic Syndr Hum Retrovirol 1996;
13(Suppl 1):S204–14.
5. Murphy EL. The clinical epidemiology of human T-lymphotropic virus
type II (HTLV- II). J Acquir Immune Defic Syndr Hum Retrovirol
1996; 13(Suppl 1):S215–9.
6. Wattel E, Vartanian JP, Pannetier C, Wain-Hobson S. Clonal expansion
of human T-cell leukemia virus type I–infected cells in asymptomatic
and symptomatic carriers without malignancy. J Virol 1995; 69:2863–8.
7. Igakura T, Stinchcombe JC, Goon PK, et al. Spread of HTLV-I between
lymphocytes by virus-induced polarization of the cytoskeleton. Science
2003; 299:1713–6.
8. Mahieux R, Pise-Masison CA, Nicot C, Green P, Hall WW, Brady JN.
Downloaded from http://jid.oxfordjournals.org/ by guest on June 26, 2015
Inactivation of p53 by HTLV type 1 and HTLV type 2 Tax trans-
activators. AIDS Res Hum Retroviruses 2000; 16:1677–81.
9. Nagai M, Usuku K, Matsumoto W, et al. Analysis of HTLV-I proviral
load in 202 HAM/TSP patients and 243 asymptomatic HTLV-I carriers:
high proviral load strongly predisposes to HAM/TSP. J Neurovirol 1998;
4:586–93.
10. Yamano Y, Nagai M, Brennan M, et al. Correlation of human T-cell
lymphotropic virus type 1 (HTLV-1) mRNA with proviral DNA load,
virus-specific CD8+ T cells, and disease severity in HTLV-1–associated
myelopathy (HAM/TSP). Blood 2002; 99:88–94.
11. Osame M. Overview, epidemiology and clinical features of HTLV-I–
associated myelopathy (HAM/TSP). Clin Neurol 1988; 28:1375–8.
12. Levin MC, Lee SM, Kalume F, et al. Autoimmunity due to molecular
mimicry as a cause of neurological disease. Nat Med 2002; 8:509–13.
13. Jacobson S. Immunopathogenesis of human T cell lymphotropic virus
type I–associated neurologic disease. J Infect Dis 2002; 186(Suppl 2):
S187–92.
14. Manns A, Miley WJ, Wilks RJ, et al. Quantitative proviral DNA and
antibody levels in the natural history of HTLV-I infection. J Infect Dis
1999; 180:1487–93.
15. Tachibana N, Okayama A, Ishihara S, et al. High HTLV-I proviral DNA
level associated with abnormal lymphocytes in peripheral blood from
asymptomatic carriers. Int J Cancer 1992; 51:593–5.
16. Cimarelli A, Duclos CA, Gessain A, et al. Quantification of HTLV-II
proviral copies by competitive polymerase chain reaction in peripheral
blood mononuclear cells of Italian injecting drug users, central Afri-
cans, and Amerindians. J Acquir Immune Defic Syndr Hum Retrovirol
1995; 10:198–204.
17. Woods TC, Graber JM, Hershow RC, Khabbaz RF, Kaplan JE, Heneine
W. Investigation of proviral load in individuals infected with human
T-lymphotropic virus type II. AIDS Res Hum Retroviruses 1995; 11:
1235–9.
18. Beilke MA, Japa S, Vinson DG. HTLV-I and HTLV-II virus expression
increase with HIV-1 coinfection. J Acquir Immune Defic Syndr Hum
Retrovirol 1998; 17:391–7.
19. Zehender G, Meroni L, Varchetta S, et al. Human T-lymphotropic virus
type 2 (HTLV-2) provirus in circulating cells of the monocyte/mac-
rophage lineage in patients dually infected with human immunode-
ficiency virus type 1 and HTLV-2 and having predominantly sensory
polyneuropathy. J Virol 1998; 72:7664–8.
20. Murphy EL, Grant RM, Kropp J, Oliveira A, Lee T-H, Busch MP.
Increased HTLV-II proviral load following highly active retroviral ther-
apy in HIV-coinfected patients. J Acquir Immune Defic Syndr 2003; 33:
655–6.
HTLV-I and -II Provirus Load • JID 2004:190 (1 August) • 509
21. Machuca A, Soriano V. In vivo fluctuation of HTLV-I and HTLV-II
proviral load in patients receiving antiretroviral drugs. J Acquir Immune
Defic Syndr 2000; 24:189–93.
22. Murphy EL, Watanabe K, Nass CC, Ownby H, Williams A, Nemo G.
Evidence among blood donors for a 30-year-old epidemic of human
T lymphotropic virus type II infection in the United States. J Infect
Dis 1999; 180:1777–83.
23. Murphy EL, Glynn SA, Fridey J, et al. Increased incidence of infectious
diseases during prospective follow-up of human T-lymphotropic virus
type II– and I–infected blood donors. Retrovirus Epidemiology Donor
Study. Arch Intern Med 1999; 159:1485–91.
24. Murphy EL, Wang B, Sacher RA, et al. Respiratory and urinary tract
infections, arthritis and asthma associated with HTLV-I and HTLV-II
infection. Emerg Infect Dis 2004; 10:109–16.
25. Busch MP, Laycock M, Kleinman SH, et al. Accuracy of supplementary
serologic testing for human T-lymphotropic virus types I and II in US
blood donors. Retrovirus Epidemiology Donor Study. Blood 1994; 83:
1143–8.
26. Liu H, Leung P, Glynn S, Murphy EL. Human T-lymphotropic virus
type II RFLP subtypes a0 and b4/b5 are associated with different de-
mographic and geographic characteristics in the United States. Virology
2001; 279:90–6.
27. Kaplan JE, Khabbaz RF, Murphy EL, et al. 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 1996; 12:193–201.
28. Albrecht B, Collins ND, Newbound GC, Ratner L, Lairmore MD.
Quantification of human T-cell lymphotropic virus type 1 proviral load
by quantitative competitive polymerase chain reaction. J Virol Methods
1998; 75:123–40.
29. Cimarelli A, Duclos CA, Gessain A, Casoli C, Bertazzoni U. Clonal
expansion of human T-cell leukemia virus type II in patients with high
proviral load. Virology 1996; 223:362–4.
30. Endo K, Hirata A, Iwai K, et al. Human T-cell leukemia virus type 2
(HTLV-2) Tax protein transforms a rat fibroblast cell line but less
efficiently than HTLV-1 Tax. J Virol 2002; 76:2648–53.
31. Mahieux R, Pise-Masison CA, Lambert PF, et al. Differences in the
ability of human T-cell lymphotropic virus type 1 (HTLV-1) and HTLV-
2 Tax to inhibit p53 function. J Virol 2000; 74:6866–74.
510 • JID 2004:190 (1 August) • Murphy et al.
32. Lewis MJ, Sheehy N, Salemi M, VanDamme AM, Hall WW. Compar-
ison of CREB- and NF-kB–mediated transactivation by human T lym-
photropic virus type II (HTLV-II) and type I (HTLV-I) Tax proteins.
Virology 2002; 295:182–9.
33. Ross TM, Minella AC, Fang ZY, Pettiford SM, Green PL. Mutational
analysis of human T-cell leukemia virus type 2 Tax. J Virol 1997; 71:
8912–7.
34. Schreiber GB, Murphy EL, Horton JA, et al. Risk factors for human
T-cell lymphotropic virus types I and II (HTLV-I and -II) in blood
donors: the Retrovirus Epidemiology Donor Study. NHLBI Retrovirus
Epidemiology Donor Study. J Acquir Immune Defic Syndr Hum Ret-
rovirol 1997; 14:263–71.
35. Taylor GP, Tosswill JH, Matutes E, et al. Prospective study of HTLV-
I infection in an initially asymptomatic cohort. J Acquir Immune Defic
Syndr 1999; 22:92–100.
36. Etoh K, Yamaguchi K, Tokudome S, et al. Rapid quantification of
HTLV-I provirus load: detection of monoclonal proliferation of HTLV-
I–infected cells among blood donors. Int J Cancer 1999; 81:859–64.
37. Orland JR, Engstrom J, Fridey J, et al. Prevalence and clinical features
of HTLV neurologic disease in the HTLV outcomes study. Neurology
2003; 61:1588–94.
38. Kramer A, Maloney EM, Morgan OS, et al. Risk factors and cofactors
Downloaded from http://jid.oxfordjournals.org/ by guest on June 26, 2015
for human T-cell lymphotropic virus type I (HTLV-I)–associated mye-
lopathy/tropical spastic paraparesis (HAM/TSP) in Jamaica. Am J Ep-
idemiol 1995; 142:1212–20.
39. Wilks R, Hanchard B, Morgan O, et al. Patterns of HTLV-I infection
among family members of patients with adult T-cell leukemia/lym-
phoma and HTLV-I associated myelopathy/tropical spastic paraparesis.
Int J Cancer 1996; 65:272–3.
40. Iga M, Okayama A, Stuver S, et al. Genetic evidence of transmission
of human T cell lymphotropic virus type 1 between spouses. J Infect Dis
2002; 185:691–5.
41. Moules V, Gabet AS, Nass CC, et al. Endemic versus epidemic viral
spreads display distinct patterns of intraindividual HTLV-2 replication
(abstract O60). AIDS Res Hum Retroviruses 2003; 19(Suppl):S24.
42. Jeffery KJ, Usuku K, Hall SE, et al. HLA alleles determine human T-
lymphotropic virus–I (HTLV-I) proviral load and the risk of HTLV-
I–associated myelopathy. Proc Natl Acad Sci USA 1999; 96:3848–53.