562 • JID 2007:195 (15 February) • Traina-Dorge et al.
M A J O R A R T I C L E
Human T Cell Leukemia Virus Type 1 Up-Regulation
after Simian Immunodeficiency Virus–1 Coinfection
in the Nonhuman Primate
Vicki L. Traina-Dorge,1Louis N. Martin,1Rebecca Lorino,1Elsa L. Winsor,2and Mark A. Beilke2,a
1Division of Microbiology, Tulane National Primate Research Center, Covington, and
Section of Infectious Diseases, New Orleans, Louisiana
2Tulane University Health Sciences Center,
The effects that human T cell leukemia virus (HTLV) type 1 and simian immunodeficiency virus (SIV) coin-
fection have on HTLV-1 dynamics and disease progression were tested in a nonhuman primate model. Seven
rhesus macaques were experimentally inoculated with HTLV-1, and a persistent infection was established.
Coinfection with SIV/smB670 resulted in increased HTLV-1 p19 antigens in peripheral blood mononuclear
cells and HTLV-1 proviral loads. Circulating CD2+and CD8+T lymphocytes increased over preinoculation
levels, along with a progressive decrease in CD4+T cells, typical for terminal SIV disease. Finally documented
was the striking emergence of up to 19% of HTLV-associated “flower cell” lymphocytes in the circulation, as
seen in patients with adult T cell leukemia/lymphoma. CD8+CD25+T cell subpopulation increases were pos-
itively correlated with flower cell appearance and suggested their possible role in this process. We conclude
that SIV may have the potential to up-regulate HTLV-1 and disease expression.
Human T cell leukemia virus (HTLV) type 1 is an im-
portant pathogen worldwide known to cause a life-long
chronic infectionthat mayleadtoadultTcellleukemia/
lymphoma (ATLL) and a neurodegenerative disease
known as tropical spastic paraparesis/HTLV-1–associ-
ated myelopathy (TSP/HAM) [1–5]. In certain geo-
graphic regions, HTLV-1 coinfection in HIV-1–positive
Received 22 June 2006; accepted 4 October 2006; electronically published 8
Potential conflicts of interest: none reported.
Presented in part: 11th Conference on Human Retrovirology: HTLV and Related
Viruses, San Francisco, 9–12 June 2003 (abstract P139).
Financial support: National Institutes of Health (NIH), National Center forResearch
and Resource (NCRR; grant 2M01RR005096; TulaneNationalPrimateResearchCenter
base grant 2P51RR000164 to V.T.-D. and M.A.B.; and grants 1G20RR016930,
1G20RR018397, 1G20RR019628, 1G20RR03466, 1G20RR012112, and 1G20RR015169
to V.T.-D.); Tulane/Louisiana State UniversityGeneralClinicalResearchCenter(Clinical
Diseases (contract N01AI65310 to L.N.M.).
aPresent affiliation: Division of InfectiousDiseases,MedicalCollegeofWisconsin,
Reprints or correspondence: Dr. Vicki Traina-Dorge, Tulane National Primate Re-
search Center, Div. of Microbiology, 18703 Three Rivers Rd., Covington, LA 70433
The Journal of Infectious Diseases
? 2007 by the Infectious Diseases Society of America. All rights reserved.
individuals is high (5%–10%) [6, 7], and recent data
suggest that HIV/HTLV-1 coinfection is associatedwith
an increase in HTLV-1 expression and disease [8–10].
The effects of highly active antiretroviral therapy
(HAART) in control of HTLV-1 replication is unclear
, and concern exists for the late emergence of
HTLV-1 disease manifestations given that dually in-
fected persons live longer in the era of HAART because
their HIV infection is controlled. This is evidenced by
high observed rates of TSP/HAM in patient population
studies of HIV/HTLV coinfection .
Simian T cell leukemia virus (STLV)–1, the simian
counterpart of HTLV-1, naturally infects Old World
monkeys and shares virologic, immunological, molec-
ular, and pathological features with HTLV-1 [13–15].
that STLV-1 is the simian ancestor of HTLV-1 that re-
sulted from cross-species transmission from multiple
urally infected monkeys, STLV-1 pathogenesisissimilar
to HTLV-1 in humans, causing ATLL-like pathological
features in a minority of individuals after a long period
of latency [21–24].
at Tulane University on August 14, 2015
HTLV-1/SIV Coinfection in Macaques • JID 2007:195 (15 February) • 563
HTLV-1 infection was demonstrated in several monkey spe-
cies inoculated with MT-2 cells, Ra-1 cells, or autologous
HTLV-1-infected cell lines [25, 26]. Rhesus macaques and
squirrel monkeys have been successfully infected with HTLV-
1 and used as NHP models for studying HTLV-1 pathogenesis,
drug testing, and vaccine trials [27–32].
The current study employed the rhesus model to test the
hypothesis that HIV/HTLV-1 coinfection is associated with in-
creased HTLV-1 expression and disease. Rhesus macaques were
infected with a primary clinical isolate of HTLV-1 from an
individual with associated clinical diseases, TSP/HAM and
polymyositis . These macaques were then either simulta-
neously or serially coinfected with a pathogenic simian im-
munodeficiency virus (SIV) [33, 34]. Macaques were followed
longitudinally for HTLV-1 by culture, polymerase chain reac-
tion (PCR), serologic, and flow cytometry testing to assess vi-
rological and immunological responses and for hematological
parameters to assess ATLL cell emergence in the circulation.
MATERIALS AND METHODS
6 males and 1 female) were used. Macaques were 1–2 years old
and confirmed seronegative and PCR negative for STLV-1 and
SIV at the study initiation. Three Indian and 3 Chinese rhesus
macaques from earlier studies served as SIV–monoinfected con-
macaques were also utilized as controls for baseline values.
All animal housing, care, and research were performed in
Animals (National Research Council) and the guidelines at the
Tulane National Primate Research Center, were fully accredited
by the Association for Assessment and Accreditation of Lab-
oratory Animal Care, and were in accordance with Animal
Welfare Act guidelines. Protocols were approved by the Insti-
tutional Animal Care and Use Committee. Physical examswere
performed either biweekly or monthly, and macaqueswerepro-
vided full supportive care. Moribund macaques werehumanely
killed, and necropsies were performed.
A CD4+HTLV-1–producing cell line, denoted
“HTLVKT,” was used for inoculum preparation. HTLVKTwas
established from cultured peripheral blood mononuclear cells
(PBMCs) of a patient with TSP/HAM, polymyositis, uveitis,
and lymphoid interstitial pneumonitis . The inoculum was
shown to be free of HIV and Mycoplasma by culture and PCR.
Macaques were inoculated intravenously(iv)onceortwicewith
cultured withirradiated HTLV-1KTcells, as described5?10
elsewhere  (figure 1). After 1–2 years of HTLV-1 infection,
subjects P574, P671, M744, P442, and P111 were iv inoculated
with 50–100 TCID50of freshly thawed, cell-free, cryopreserved
Seven Chinese rhesus macaques (Macaca mulatta;
phytohemagluttin-stimulated autologus PBMCs co-
stock of SIVsmB670. Macaque N365 was simultaneously iv
inoculated with both PBMC/HTLV-1KTcoculture and cell-free
SIVsmB670. An additional control, rhesus N401, was mock
inoculated with autologous PBMCs and later inoculated with
the HTLV-1KTPBMC coculture, as described above.
Clinical testing, viral detection, and serologic analysis.
Routine monthly clinical testing included complete blood
counts, differential white blood cell (WBC) counts, chemistry
levels, and parasite stool examination. Additional blood was
collected for virologic, serologic, and flow cytometric evalua-
tions. PBMCs were isolated from blood by density Ficoll-Hy-
paque separation, for both culture and cryopreservation for
storage and later nucleic acid extraction and PCR.
HTLV-1 p19 antigen detection in PBMC culture, serologic
analysis, genomic DNA isolation, and end-point PCR of cir-
culating PBMC genomic DNA were performed as described
elsewhere . Culture supernatants were harvested at days 7,
14, and 21 for HTLV-1 viral p19 antigen detection, by use of
a commercial kit (Zeptometrix). A positive culture was defined
as antigen detection at any of the times points tested. Serologic
assays were performed using a commercial ELISA kit (Abbott
Laboratories). Serial 2-fold dilutions of serum (1:20–1:1280)
were assayed in accordance with the manufacturer’s instruc-
tions. A spectrophotometric reading of ?2.5 times the optical
density reading of a serum sample from an uninfected control
macaque was obtained as a positive titer.
Plasma samples for SIVRNAviralantigenanalysiswerecryo-
preserved at ?70?C until analysis. SIV antigenemia was deter-
mined by SIV p27 core antigen ELISA (Beckman Coulter) in
accordance with the manufacturer’s instructions.
Virus load quantification.
was performed with the Prism 7700 sequence detection system
(Applied Biosystems ABI) for HTLV-1tax and RNase P am-
plification. Primers were as follows: HTLV tax, forward primer
7930F,5?-GCCCTA ATAATTCTA CCCGAAGACT-3?;reverse
primer 8040R, 5?-GGT TGA GTG GAA CGG AAG GA-3?; and
probe 7991P, 5?-Reporter dye-FAM-CCG TCA CGC TAA CAG
CCT GGC A-3?-TAMRA-Quencher. Primers were selected ac-
cording to parameters defined by the Primer Express software
(version 2.0; ABI) and were based on strain J02029 . Plas-
mid pMT2 containing the full HTLV-1 tax/rex coding region
in pUC13 was used as HTLV-1 control DNA to generate the
standard curve . Fifty-microliter reactions contained 1?
101– copies of viral control DNA or 500 ng of genomic
PBMC DNA and the following: 1? TaqMan buffer A; 200
mmol/L each dATP, dCTP, and dGTP; 400 mmol/L dUTP; 3.5
mmol/L MgCl2; 300 mmol/L HTLV primers; 200 nmol/LHTLV-
1 probe; 0.625 U of AmpliTaq Gold; and 0.25 U of uracil N-
glycosylase. Reactions were amplified in duplicate in 96-well
optical-grade PCR plates. The reaction proceeded as follows: 2
Real-time PCR quantification
at Tulane University on August 14, 2015
564 • JID 2007:195 (15 February) • Traina-Dorge et al.
and P442 (P111 data not shown). Each virus inoculum is shown with the virus name boxed and positioned above the inoculation date, which is shown
in months after initial human T cell leukemia virus (HTLV) type 1 inoculation. Inoculation with HTLV-1 is shown for M744, P574, P671, and P442. These
same macaques were later inoculated with the second virus, simian immunodeficiency virus (SIV). One macaque (N365) was simultaneously inoculated
with HTLV-1 and SIV. N401, originally the uninfected control, received a mock infection initially. After 20 months, N401 was inoculated with HTLV-1 and
used as the single HTLV-1–infected control for the coinfection experiments. Antibody response to HTLV-1 was tested by EIA, and resultant titers are shown
longitudinally by columns. The presence of HTLV-1 p19 in peripheral blood mononuclear cell culture is shown as + or ? below each of the monthly culture
Macaque culture and serologic results. Results are summarized for the experimentally inoculated macaques, N401, P574, N365, P671, M744,
min at 50?C, 10 min at 95?C, and 40 cycles of 15 s at 95?C
and 60 s at 60?C. Viral copy numbers were interpolated from
the plasmid control regression curve. Finally, RNase P, a single-
copy gene (2 copies/diploid cell) was used as an endogenous
DNA reference to accurately quantify cell equivalents and nor-
malize sample variability. RNase P primers, probe, and stan-
dards were commercially obtained (ABI). Final HTLV-1 pro-
viral copy number was calculated for each sample as follows:
HTLV-1 copies/[RNase P copies/
limit of the assay was 1 copy/
detectable virus were scored as 10 copies/
Flow cytometry and quantification of T lymphocyte
evaluated for lymphocyte subsets by fluorochrome-conjugated
cells). The lower
cells. Samples with un-
at Tulane University on August 14, 2015
HTLV-1/SIV Coinfection in Macaques • JID 2007:195 (15 February) • 565
reaction was used to quantify virus loads, HTLV-1 provirus copies found in the peripheral blood mononuclear cells of dually HTLV/SIV-infected macaques
(see Materials and Methods). Copies of HTLV-1 provirus for each macaque on each of the months after SIV inoculation are shown as copies/
cells. The lower limit of the assay was 1 copy/cells. Samples with undetectable virus loads were scored as 10 copies/
Human T cell leukemia virus (HTLV) type 1 provirus loads after simian immunodeficiency virus (SIV) inoculation. Real-time polymerase chain
monoclonal antibody staining and flow cytometry, as described
elsewhere . CD2+T cells and CD20+B cells were identified
with anti-T11 and anti-B1 (Beckman Coulter). CD4+T cells,
CD8+T cells, and CD25+T cells were identified with anti-CD4
(clone L200), anti-CD8 (clone RPA T8), and anti-CD25 (clone
2A3) (BD Biosciences). Erythrocytes were lysed, and leukocytes
were fixed with ImmunoPrep reagents using the TQ-Prep Sys-
tem (Beckman Coulter). Data acquisition and analysis were
performed on a FACSCalibur device using Cell Quest software
(BD Biosciences). Absolute numbers of cells were determined
by calculating the percentage of the subset multiplied by total
Enumeration of “flower cells.”
blood smears were prepared, 100 WBCs were counted, lineage
was assessed, and smears were analyzed for the presence of the
abnormally lobulated flower cell lymphocytes of HTLV infec-
tion . Mean numbers of flower cells intheHTLV/SIVgroup
compared with those in the controls were analyzed using an
independent t test withunequalvariancesassumedto3months.
Multiple comparisons required P values to be adjusted using
the Bonferroni correction factor.
To control for spontaneous occurrence of flower cells in
uninfected or single virus–infected NHP and to establish base-
lines, blood smear slides from previously tested SIV and STLV-
1–negative or singly STLV-1–positive colony macaques were
evaluated. As a control for accuracy, an independent investi-
gator reading and quantification of flower cells was performed
on a blinded sample set including smears from both virus-
positive and -negative macaques. Readings had a high corre-
lation (96%) between investigators.
Rhesus macaques were inoculated withHTLV-1KT–infectedcells
and SIVsmB670 to study the virological, immunological, and
clinical effects of dual retroviral infection.Fivemacaques(P671,
P574, P442, P111, and N401) were initially singly inoculated
with HTLV-1 (figure 1). Two macaques from a previous study
were also utilized: M744, inoculated with HTLV-1 only; and
N365, inoculated with HTLV-1 simultaneously with SIV .
Five persistently HTLV-1– infected macaques (M744, P574,
P671, P111, and P442) were later inoculated with 50–100
TCID50of SIVsmB670 and monitored along with N365, the
other dually HTLV-1/SIV– infected macaque. SIV infectionwas
confirmed in all coinfected macaques by positive SIV p27 an-
tigen in the circulation (data not shown). Macaque N401 re-
mained as an singly HTLV-1–infected control.
Progression to AIDS was observed at 3–13 months in 5 of
6 the serially coinfected macaques and at 28 months in the
simultaneously coinfected macaque, N365. These macaques
died of colitis, cachexia, dyspnea, lymphoid hyperplasia, and
opportunistic infections, typical for terminal immunodefi-
at Tulane University on August 14, 2015
566 • JID 2007:195 (15 February) • Traina-Dorge et al.
lymphocytes (B), and CD4+T lymphocytes (C), for the dually human T cell leukemia virus (HTLV) type 1/SIV–infected macaques and the 1 HTLV-1–infected
control macaque at each monthly time point after SIV inoculation. Resultant cell nos. are shown as absolute nos. of cells/mm3.
Lymphocyte subset populations after simian immunodeficiency virus (SIV) inoculation. Results are shown for CD2+T lymphocytes (A), CD8+T
ciency disease. Macaque P442 did not show clinical signs of
AIDS. For thismacaque,however,anelectednecropsy5months
after SIV coinfection showed severe lymphoid hyperplasia in
both lymph nodes and spleen.
After HTLV-1 inoculation, longitudinal samples for all 7 ma-
caques confirmed HTLV-1 infection by antigen detection in
PBMC culture, serologic analysis, and end-point PCR. Figure
1 depicts these culture and serologic results for 6 of the 7
macaques (excluding P111). HTLV p19 antigens were detected
in PBMC cultures for 6 of the inoculated macaques. HTLV-
p19 antigen in cultures were positive before or during peak
antibody production with differences in peak responses cor-
at Tulane University on August 14, 2015
HTLV-1/SIV Coinfection in Macaques • JID 2007:195 (15 February) • 567
each monthly time point after simian immunodeficiency virus (SIV) inoculation. Flow cytometry results are shown as absolute no. of cells/mm3and are
shown on the left Y-axis. Flower cells are shown as nos./100 white blood cells from the differential analysis of the peripheral blood smear and are shown
on the right Y-axis.
Comparison of CD25+CD4+and CD25+CD8+T lymphocytes with presence of flower cells for the macaques P574, P671, P111, and P442, at
experimentally inoculated with human T cell leukemia virus type 1. Two
multilobulated flower cells are shown in the larger field, and a singleflower
cell is shown enlarged in the inset.
Flower cells in a peripheral blood smear from a macaque
related with viral clearance. Strong peak antibody responses (1:
1280) were shown in M744 and P442; moderate responses (1:
320–1:640) in N365 and P671; and low responses (1:40–1:20)
in P574 and N401. P111 with intermittent positive culture and
PCR results never mounted a detectable antibody response
(data not shown). Macaque P442, with high antibodyresponse,
was culture negative for 12 years. N401 antibody decrease was
associated with intermittent positive cultures. P574, P671,
M744, and P442 cultures decreased to undetectable levels, de-
spite persistence of infection as documented by PCR (data not
shown). After SIV inoculation, HTLV-1 antibody levels were
reduced further and were correlated with positive HTLV p19
detection in cultures in all coinfected macaques.
After SIV coinfection, HTLV loads were monitored and
quantified by real-time PCR amplification of the HTLV-1 tax
gene. At baseline, HTLV-1 tax was either undetectable (M744,
P574, P442, and N365) or low (P671 and P111), with only 37.8
and 36.3 provirus copies/
SIVsmB670 inoculation, distinct increases were documented in
4 macaques by 2 weeks, progressing to peak levels over 1 to
several months. Peak levels of HTLV provirus were as follows:
P111, 83 copies/ PBMCs (1 month); P574, 367 copies/
PBMCs (1.5 months); P671, 156 copies/
(2 months); and M744, 68 copies/
N365, simultaneously HTLV-1/SIV coinfected, was more de-
layed, with provirus emerging at 4 months after infection and
PBMCs (figure 2). After
PBMCs (2 months).
Macaque P442 remained completely negative by PCR for
HTLV-1KTsequences in PBMCs throughout the entire time of
T cell subset population changes after SIV infection were
evaluated by flow cytometry. ThreeIndianand3Chineserhesus
from earlier studies were used as historical controls for singly
at Tulane University on August 14, 2015
568 • JID 2007:195 (15 February) • Traina-Dorge et al.
virus (HTLV) type 1–positive control macaque, over time after simian im-
munodeficiency virus (SIV) infection. Because of each macaque’s individual
progression to terminal disease and death, the no. of the group changed:
6 at baseline through 2 months; 5 at 2.5–5 months; 4 at 6 and 7 months;
and 3 at 8 and 10 months. Flower cells are indicated as percent of the
100 white blood cells identified in the differential analysis of theperipheral
blood smear. Values are means for the experimental HTLV/SIV group and
actual values for the control. Actual flower cell percentage for 4 of the
experimental macaques (P574, P671, P111, and P442) are shown in
Mean levels of abnormal flower cells in the circulation of the
SIV-infected macaques. These SIV-positive macaques showed
CD2+T lymphocytes ranging from 1000 to 3100 cells/mm3,
CD8+T lymphocytes ranging from 500 to 1900 cells/mm3, and
CD4+T lymphocytes decreasing to only a few hundred cells
per cubic millimeter within 1 year (L.N.M., unpublished data)
. No sharp increases in either CD2+or CD8+T cell pop-
ulations were noted after SIV monoinfection.
Baseline CD2+T lymphocytes in the study macaques ranged
from 1230 to 3262 cells/mm3(mean, 2433 cells/mm3) (figure
3A). After SIV infection, at 0.5, 1, and 2.5 months, values in
coinfected macaques (M744, P671, P442, and N365) increased
200%–290% over baseline (3479–6790 cells/mm3) and above
the singly HTLV-1–infected control (522–2159 cells/mm3).
CD8+T lymphocytes increased 190%–330% in 5 coinfected
macaques (M744, P671, P111, P442, and N365) (figure 3B).
Peak values (2307–4460 cells/mm3) were observed at 0.5, 1, and
2.5 months after infection and were clearly increased above
CD8+T lymphocyte values in the control (302–903 cells/mm3).
However, a different pattern was observed with CD4+T lym-
phocytes. Baseline values ranged from 907 to 1298 cells/mm3
(figure 3C). After SIV infection, P574, P671, P442, and P111
showed a progressive and steady decrease in CD4+T lympho-
cytes to 108, 286, 119, and 318 cells/mm3, respectively, as ex-
pected with AIDS progression. Two macaques had early and
transient increases (1.3- and 1.6-fold for M744 [1777 and 2016
cells/mm3] at 0.5 and 2.5 months and 2.6-fold for N365 [2403
cells/mm3] at 1 month), which then decreased to 1173 and
803 cells/mm3, respectively. These data collectively suggest that
dual HTLV/SIV coinfections cause a distinct but transient
CD2+CD8+T cell lymphocytosis.
The CD25+T cell surface marker (interleukin-2 receptor a
chain) has long been known to be activated and present on
HTLV-1–infected and transformed ATLL cells [40, 41]. To eval-
uate increases after SIV coinfection, CD25+T lymphocytes in
4 of the macaques (P574, P671, P111, and P442) were moni-
tored by flow cytometry. CD25+cells at baseline for 4 macaques
ranged from 110 to 164 cells/mm3and then progressively de-
creased to very low levels (18–64 cells/mm3) within 8 months
after SIV coinfection (data not shown). One macaque, P574,
showed slight increases above baseline (1 and 2 months) and
then decreased. Further analysis of the CD4+CD25+and
CD8+CD25+subpopulations showed dramatic differences (fig-
ure 4). The larger CD4+/CD25+subpopulation with baseline
values (86–104 cells/ mm3) paralleled total CD25+cells and
showed progressive decreases tolevels !30cells/mm3.Macaques
P574 and P111 showed intermittent increases at 2–3 months
after infection but decreased similarly to levels !30 cells/mm3
by 8 months. The smaller CD8+CD25+subpopulation showed
distinct increases from 0.5 to 2.0 months after SIV infection.
Of particular importance was the abnormal appearance of
multilobulated lymphocytes, resembling the flower cells (figure
5) of ATLL. Flower cells are a hallmark and characteristic find-
ing in HTLV-1 and STLV-1 infections that have progressed to
disease [13, 39, 41]. Readings of archived blood slides from
uninfected or single STLV-1– or SIV-positive macaques estab-
lished baseline values.Thosebaselineresultsshowedflowercells
present at low levels in the absence of STLV-1: 0%–2% flower
cells (average, 0.8%) in SIV/STLV-uninfected macaques (n p
); 0%–3% flower cells (average, 0.8%) in SIV-positive/STLV-
negative macaques (); and 0%–6% flower cells (average,
n p 23
2.2%) in SIV/STLV-positive macaques (
cells observed in the study HTLV-1–positive control, N401,
ranged from 0% to 3% and averaged 1.3%. The coinfected
macaques—M744, P574, P671, P111, and P442—showed dra-
matic increases in flower cells to 7%, 8%, 10%, 6%, and 12%
during the first 3 months (figure 4) (M744 not shown). N365,
the macaque concurrently inoculated with HTLV-1 and SIV,
showed varying levels of flower cells (0%–6%) during the first
2 years after infection, which increased to 19% in a period of
5 months before progression to AIDS and death the following
month (data not shown). Figure 6 shows the calculated
means of flower cells for the coinfected macaques compared
with the control for 10 months after SIV inoculation. t test
analysis of flower cells in HTLV-1/SIV–coinfected macaques
() through 3 months after infection were shown to ben p 6
significantly higher (P p .001
HTLV-1–positive control macaque (
tistical significance, compared with the Bonferroni-adjusted
critical value of.P p .007
). The flower
n p 16
) than those observed in the
) and indicate sta-n p 1
at Tulane University on August 14, 2015
HTLV-1/SIV Coinfection in Macaques • JID 2007:195 (15 February) • 569
This study showed that SIV coinfection of HTLV-1–infected
rhesus monkeys up-regulated HTLV-1 expression and emer-
gence of flower cells in the peripheral circulation. Although
the small numbers of macaques in the study precludes pow-
erful statistical analysis, clear changes were documented. Af-
ter SIV coinfection, specific increases in HTLV-1 p19, HTLV
proviral loads, circulating CD2+CD8+T cells, and abnormal
flower cells were clearly evident. The increased occurrence of
?3% of flower cells in the circulation of coinfected macaques
was significantly higher than in the control, yet the macaques
maintained normal WBC counts. These findings are consis-
tent with the pattern seen in chronic and smoldering forms
of adult T cell leukemia/lymphoma [39, 42]. Furthermore,
peripheral blood smears of healthy carriers of HTLV-1 rarely
if ever show 11% flower cells . The emergence of flower
cells in all 6 coinfected macaques during infection also sug-
gests the possibility that HTLV and SIV may increase the risk
for development of leukemia and lymphoma in dually infect-
It has been reported that HIV appears to increase the risk
for HTLV-1–associated disease, including hematologic ma-
lignancies and TSP/HAM [8, 10, 44]. The dynamics of HIV
and HTLV coinfections remain unclear, but Beilke et al.
 showed that coinfections are associated with increased
HTLV-1 mRNA in PBMCs. HIV/HTLV coinfections pose
diagnostic and prognostic problems for physicians who do
not routinely test their HIV-infected patients for HTLV-1
or HTLV-1I antibodies.
HIV/HTLV-1 coinfections are often associatedwithsustained
periods of normal or even elevated CD4+T cell counts 
and early neurologic complications in the face of normal CD4+
T cell counts . CD4+/CD25+T cells, specifically the T reg-
ulatory (Treg) cell subpopulation, are suggested as the major
reservoir for HTLV-1, express virus, down-regulate the normal
suppressor function, allow expansion of HTLV-1 tax–specific
CD8+T cells, and contribute to viral pathogenesis [ 46, 47].
In addition, CD8+CD25+T cells have been shown to be a viral
reservoir in vivo and involved in pathogenesis of HTLV-1–
mediated disorders [46–48].
This study clearly supports clinical observations in humans
that HIV/HTLV-1 coinfection may up-regulate cell prolifera-
tion, HTLV-1 expression, and increase disease potential. After
SIV coinfection, CD4+CD25+cells progressivelydecreased,with
only 2 of 4 macaques showing intermittent and transient in-
creases over the course of terminal decrease. SIV-specific CD4+
Tregcell depletion is a possible cause for this sharp decrease in
CD4+CD25+T cells . Elimination of the Tregcell suppressor
function by SIV and resultant immune hyperactivation 
along with HTLV-1–induced spontaneous lymphoproliferation
may also explain the subsequent increases in the CD8+CD25+
T cell subpopulation and the increase in HTLV replication that
were documented in the macaques [46, 47]. The emergence of
large numbers of flower cells in macaques, resembling those
seen in the circulation of patients with ATLL, is particularly
striking. The flower cells directly correlated with CD8+/CD25+
T cells and suggested the possible involvement of this subpop-
ulation in disease progression. Additional studies are ongoing
to assess this correlation.
Fultz et al.  obtained similar results in pig-tailed ma-
caques inoculated with SIV-PBj and STLV-I. During 2 years of
in SIV burdens, onset of disease, or survival were detected.
However, in the first coinfected macaque that died of AIDS (1
year after infection), 150% of CD4+and CD8+T lymphocytes
expressed CD25, and lymph nodes showed a monomorphic
population of lymphoblastoid cells. The authors suggested a
preleukemic condition due to STLV/SIV coinfection and spec-
ulated that SIV may have potentiated STLV-related disease.
The exact mechanism of up-regulated HTLV-1 expression
after SIV coinfection cannot be determined throughthecurrent
studies. However, the possibility exists that flower cells would
continue to increase along with T lymphocyte population de-
crease after SIV infection. This would suggest that loss of im-
mune control, as opposed to immune stimulation, would be a
possible mechanism to explain up-regulation of HTLV-1. This
is further evidenced by the late increase in HTLV-1 p19 viral
antigen detection in PBMC cultures after a decrease in HTLV-
1 antibody titers.
Taken together, our results suggest that the SIV/HTLV-1
model of coinfection may be useful for further studies of
HTLV-1 leukemogenesis. The model may also prove very
useful in testing the efficacy of antiretroviral compounds
and antineoplastic agents for the treatment of HTLV-1 dis-
We thank the veterinarians and clinical laboratory and staff at theTulane
National Primate Research Center. We also thank Patricia Kissinger for her
statistical analysis of the flower cell data and Maureen Shuh, Elizabeth
Didier, and Cristian Apetrei for their critical reviews of the manuscript.
1. Hinuma Y, Nagata K, HanoaokaM, etal. AdultT-cellleukemia:antigen
in an ATL cell line and detection of antibodies to the antigen in human
sera. Proc Natl Acad Sci USA 1981;78:6476–80.
2. Poiesz BJ, Ruscetti FW, Gazdar AF, Bunn PA, Minna JK, Gallo RC.
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 USA 1980;77:7415–9.
3. Yamaguchi K, Watanabe T. Human T lymphotropic virus type-I and
adult T-cell leukemia in Japan. Int J Hematol 2002;(Suppl 2):240–5.
4. Gessain A, Caudie C, Gout O, et al. Intrathecal synthesis of antibodies
to human T lymphotropic virus type I and the presence of IgG oli-
at Tulane University on August 14, 2015
570 • JID 2007:195 (15 February) • Traina-Dorge et al.
goclonal bands in the cerebrospinal fluid of patients with endemic
tropical spastic paraparesis. J Infect Dis 1988;157:1226–34.
5. Jacobson S. Immunopathogenesis of human T cell lymphotropic virus
type I–associated neurologic disease. J Infect Dis 2002;186(Suppl 2):
6. Witkor SZ, Cannon RO, Atkinson WL, et al. Infection with human T
lymphotropic virus types I and II in sexually transmitted diseaseclinics
in Baltimore and New Orleans. J Infect Dis 1992;165:920–4.
7. Khabbaz RF, Onorato IM, Cannon RO, et al. Seroprevalence of HTLV-
I and HTLV-II among intravenous drug users and persons in clinics
for sexually transmitted diseases. N Engl J Med 1992;326:375–80.
8. Harrison LH, Vaz B, Taveira DM, et al. Myelopathy among Brazilians
co-infected with HTLV-I and HIV. Neurology 1997;48:13–8.
9. Beilke MA, Japa S, Vinson DG. HTLV-I and HTLV-II virus expression
increase with HIV-1 co-infection. J Acquir Immune Defic Syndr Hum
10. Beilke MA, Theall KP, O’Brien M, et al. Clinical outcomes and disease
progression among patients coinfect with HIV and human T lympho-
tropic viruses types 1 and 2. Clin Infect Dis 2004;39:256–63.
11. Murphy EL, Grant RM, Kropp J, Oliveira A, Lee TH, Bush MP. In-
creased human T-lymphotropic virus type II proviral load following
highly active retroviral therapy in HIV-coinfected patients. J Acquir
Immune Defic Syndr 2003;33:655–6.
12. Beilke MA, Japa S, Moeller-Hadi C, Martin-Schild S. Tropical spastic
paraparesis/HTLV-1-associated myelopathy in HIV-1 coinfected pa-
tients. Clin Infect Dis 2005;41:e57–63.
13. Hayami M. Simian T-cell leukemia virus—HTLV-1 related virus in non-
human primates. In: Takatsuki K, Hinum Y, Yoshida M, eds. Gann
monograph on cancer research. Vol 39.Tokyo:JapanScientificSocieties
14. Saksena NK, Herve V, Durand JP, et al. Seroepidemiologic, molecular,
and phylogenetic analyses of simian T-cell leukemia viruses (STLV-I)
from various naturally infected monkey species from central and west-
ern Africa. Virology 1994;198:297–310.
15. Slattery JP, Franchini G, Gessain A. Genomic evolution, patterns of
global dissemination, and interspecies transmission of humanandsim-
ian T-cell leukemia/lymphotropic viruses. Genome Res 1999;9:525–40.
16. Koralnik IJ, Boeri E, Saxinger WC, et al. Phylogenetic associations of
human and simian T-cell leukemia/lymphotropic virus type 1 strains:
evidence for interspecies transmission. J Virol 1994;68:2693–707.
17. Liu HF, Goubou P, Van Brussel M, et al. The three distinct human T-
lymphotropic virus type I subtypes arose from three geographically
distinct simian reservoirs. J Gen Virol 1996;77:359–68.
18. Mahieux R, Ibrahim F, Mauclere P, et al. Molecular epidemiology of
58 new African human T-cell leukemia virus type 1 (HTLV-1) strains:
identification of a new and distinct HTLV-1 molecular subtype in Cen-
tral Africa in Pygmies. J Virol 1997;71:1317–33.
19. Van Dooren S, Salemi M, Vandamme AM. Dating the origin of the
African human T-cell lymphotropic virus type 1 subtypes. Mol Biol
20. Meertens L, Rigoulet J, Mauclere P, et al. Molecular and phylogenetic
analyses of 16 novel simian T cell leukemia virus type 1 from Africa:
close relationship of STLV-1 from Allenopithecus nigroviridis to HTLV-
1 subtype B strains. Virology 2001;287:275–85.
21. Tsujimoto H, Noda Y, Ishikawa K, et al. Development of adult T-cell
clonal integration of simian T-cell leukemia virus type I. Cancer Res
22. Traina-Dorge V, Blanchard J, Martin L, Murphey-Corb M. Immuno-
deficiency and lymphoproliferative disease in an African greenmonkey
dually infected with SIV and STLV-I. AIDS Res Hum Retroviruses
23. Hubbard GB, Mone JP, Allan JS, et al. Spontaneously generated non-
Hodgkin’s lymphoma in twenty-seven simian T-cell leukemia virus
type 1 antibody-positive baboons (Papio species). Lab Anim Sci 1993;
24. Allen JS, Leland M, Broussard S, Mone J, Hubbard G. Simian T-cell
lymphotropic viruses and lymphomas in African nonhuman primates.
Cancer Invest 2001;19:383–95.
25. Yamamoto N., Hayami M, Komuro A,et al. Experimental infection of
cynomolgus monkeys with a human retrovirus, adult T-cell leukemia
virus. Med Microbiol Immunol 1984;173:57–64.
26. Nakamura H, Tanaka Y, Komuro-Tsujimoto A, et al. Experimental
inoculation of monkeys with autologous lymphoid cell lines immor-
talized by and producing human T-cell leukemia virus type-I. Int J
27. Beilke MA, Traina-Dorge VL, England JD, Blanchard J. Polymyositis,
arthritis and uveitis in a macaque experimentally infected with HTLV-
1. Arthritis Rheum 1996;39:610–5.
28. Kazanji M, Moreau JP, Mahieux R, et al. HTLV-I infection in squirrel
monkeys (Saimiri sciureus) using autologous, homologous or heter-
ologous HTLV-I transformed cell lines. Virology 1997;231:258–66.
29. Kazanji M, Tartaglia J, Franchini G, et al. Immunogenicity and pro-
tective efficacy of recombinanthumanT-cellleukemia/lymphomavirus
type 1 NYVAC and naked DNA vaccine candidates in squirrelmonkeys
(Saimiri sciureus). J Virol 2001;75:5939–48.
30. Mortreaux F, Kazanji M, Gabet AS, de Thoisy B, Wattel E. Two-step
nature of human T-cell leukemia virus type 1 replication in experi-
mentally infected squirrel monkeys (Saimiri sciureus). J Virol 2001;75:
31. Heraud JM, Mortreaux F, Merien F, et al. The efficacy of combined
therapy of arsenic trioxide and alpha interferon in human T-cell leu-
kemia virus type-1-infected squirrel monkeys (Saimiri sciureus). An-
tiviral Res 2006;70:132–9.
32. Kazanji M, Heraud JM, Merien F, et al. Chimeric peptide vaccine
composed of B- and T-cell epitopes of human T-cell leukemia virus
type 1 induces humoral and cellular immune responses and reduces
the proviral load in immunized squirrel monkeys (Saimiri sciureus). J
Gen Virol 2006;87:1331–7.
33. Baskin GB, Murphey-Corb M, Watson EA, Martin LN. Necropsy find-
ings in rhesus monkeys experimentally infected with cultured simian
immunodeficiency virus (SIV)/delta. Vet Pathol 1988;25:456–67.
34. Martin LN, Murphey-Corb M, Soike KF, Davison-Fairburn B, Baskin
GB. Effects of initiation of 3?-azido, 3?-deoxythymidine (zidovudine)
treatment at different times after infection of rhesus monkeys with
simian immunodeficiency virus. J Infect Dis 1993;168:825–35.
35. Ling B, Veazey RS, Penedo C, Xu K, Lifson JD, Marx PA. Longitudinal
follow up of SIVmac pathogenesis in rhesus macaques of Chinese
origin: emergenceof Bcell lymphoma.JMedPrimatol2002;31:154–63.
36. Beilke MA, Greenspan DL, Impey A, Thompson J, Didier PJ. Labo-
ratory study of HIV-1 and HTLV-I/IIco-infection.JMedVirol1994;44:
37. Seiki M, Hattori S, Hirayama Y, Yoshida M. Human adult T-cell leu-
kemia virus: complete nucleotide sequence of the provirus genome
integrated in leukemia cell DNA. Proc Natl Acad Sci USA 1983;80:
38. Clarke MF, Gelmann EP, Reitz MS Jr. Homology of HTLV envelope
gene with class I HLA gene. Nature 1983;305:60–2.
39. Catovsky D, Foa R. The lymphoid leukemias. London: Butterworths,
40. Inoue J, Seiki M, Taniguchi T, Tsuru S, Yoshida M. Induction of in-
terleukin 2 receptor gene expression by p40x encoded by human T-
cell leukemia virus type 1. EMBO J 1986;5:2883–8.
41. Franchini G. Molecular mechanisms of human T-cell leukemia/lym-
photropic virus type 1 Infection. Blood 1995;86:3619–39.
42. Yamaguchi K, Kiyokawa T, Futami G, Ishii T, TakatsukiK.Pathogenesis
of adult T-cell leukemia from clinical pathologic features. In: Blattner
WA, ed. Human retrovirology: HTLV. New York: Reven Press; 1990:
43. Sacher RA, Luban NL, Ameti DI, Friend S, Schreiber GB, Murphy EL.
Low prevalence of flower cells in USA blood donors infected with
human T-lymphoctrophic virus types I and II. Br J Haematol 1999;
44. Shibata D, Byrnes RK, Rabinowitz A, et al. Human T-celllymphotropic
at Tulane University on August 14, 2015
HTLV-1/SIV Coinfection in Macaques • JID 2007:195 (15 February) • 571 Download full-text
virus type I (HTLV-I)-associated adult T-cell leukemia-lymphoma in
a patient infected with human immunodeficiency virus type 1 (HIV-
1). Ann Int Med 1989;111:871–5.
45. Nadler JP, Bach MC, Godofsky E. Management of co-infection with
human immunodeficiency virus and human T-cell lymphotropic virus
type I. Clin Infect Dis 1996;23:415.
46. Sakai JA, Nagai M, Brennan MB, Mora CA, Jacobson S. In vitro spon-
taneous lymphoproliferation in patients with human T-cell lympho-
tropic virus type 1- associated neurologic disease: predominant ex-
pansion of CD8+ T cells. Blood 2001;98:1506–11.
T lymphocyte virus type I (HTLV-I)Tax11-19peptide-human-histocom-
patibility leukocyte antigen A*201 complexes on CD4+CD25+ T cells
detected by peptide-specific, majorhistocompatibilitycomplex-restricted
antibodies in patients with HTLV-I-associated neurologic disease. J Exp
48. Nagai M, Brennan M, Sakai JA, Mora CA, Jacobson S. CD8+ T cells
are an in vivo reservoir for human T-cell lymphotropic virus type I.
49. Eggena MP, Barugahare B, Jones N, et al. Depletion of regulatory T
cells in HIV infection is associated with immune activation. J Immunol
50. Fultz PN, McGinn T, Davis IC, Romano JW, Li Y. Co-infection of
macaques with SIV and STLV type I: effects on virus burdens and
disease progression. J Infect Dis 1999;179:600–11.
at Tulane University on August 14, 2015