B R I E F R E P O R T
HTLV-2 APH-2 Expression Is
Correlated With Proviral Load but
APH-2 Does Not Promote
Edward L. Murphy,4and Renaud Mahieux1,2,3
1Equipe Oncogene `se Re ´trovirale, INSERM-U758 Virologie Humaine,
2Ecole Normale Supe ´rieure de Lyon,3IFR 128 Biosciences Lyon-Gerland, Lyon,
France;4University of California, San Francisco and Blood Systems Research
Institute; and5Laboratory of Virus Control, Institute for Virus Research, Kyoto
We recently discovered the antisense protein of human T-cell
leukemia virus (HTLV) type 2 (APH-2), whose messenger
RNA is encoded by the antisense strand of the HTLV-2 ge-
nome. Wequantified proviral load, level of tax,and APH-2 in
a series of blood samples obtained from a cohort of HTLV-2
carriers. We determined whether APH-2 promotes cell pro-
liferation. APH-2 was detectable in most samples tested and
was correlated with proviral load. APH-2 levels were not
correlated with lymphocyte count in vivo, consistent with
the inability of APH-2 to promote cell proliferation in vitro.
APH-2 does not promote cell proliferation and does not
Human T-cell leukemia virus type 1 (HTLV-1) and type 2
(HTLV-2) infect T lymphocytes in vivo . Their 5# long
terminal repeat (LTR) serves as a viral promoter to encode,
among others, the Tax transactivator protein, whereas another
promoter located in the 3#-LTR is used to encode the antisense
HTLV-1 bZip factor (HBZ) protein or the antisense protein of
HTLV-2 (APH-2), respectively [2, 3].
HTLV-2 infection is associated with lymphocytosis , rarely
with cases of myelopathies that may resemble HTLV-1 Associated
Myelopathy/Tropical Spastic Paraparesis (HAM/TSP), and with
rare cases of lymphoproliferative disease in patients coinfected
with human immunodeficiency virus. The discovery of HBZ
opened new avenues of research. Surprisingly, HBZ mRNA and
HBZ protein may have distinct functions. Indeed, HBZ promotes
proliferation of T-cell lines, whereas HBZ downregulates Tax-
mediated viral transcription . In line with these observations,
tax mRNA is not detected in most leukemic Adult T cell Leuke-
mia/Lymphoma (ATLL) cases, whereas HBZ is always expressed
[5, 6]. Furthermore, HBZ expression is correlated with proviral
load (PVL) and HAM/TSP severity . Even if a recent report
demonstrated that HBZ promotes cell transformation when ex-
pressed in transgenic animals , the model of ATLL de-
velopment suggests that Tax stimulates the initiation of
transformation, whereas HBZ is involved in the maintenance of
the transformed stage and in cell proliferation .
We have discovered APH-2, which represents the HTLV-2
counterpart of HBZ and which represses Tax-2–mediated
transcription , suggesting that APH-2 might function like
HBZ . Consistent with this, lymphocytosis is commonly
observed in HTLV-2 carriers .
We quantified APH-2 and tax expression in addition to
HTLV-2 PVL in a series of samples obtained from HTLV-2
asymptomatic carriers. APH-2 was expressed in most samples
tested. APH-2 and tax levels are correlated with PVL. However,
neither APH-2 nor tax is correlated with lymphocyte count.
Finally, in constrast to HBZ, APH-2 is not capable of promoting
HTLV-2 carrier blood samples were obtained from human
subjects enrolled in the HTLV Outcomes Study cohort, which
followed for .15 years the 151 HTLV-1 subjects, 387 HTLV-2
subjects, and 799 uninfected controls . From the total of 387
HTLV-2 subjects, we selected 3 groups of 20 based on PVL
measurements at the baseline visit of the cohort: 20subjectswith
the lowest (1026to 1024.7copies/cell), 20 subjects surrounding
the median (1024.5to 1023.19copies/cell) and 20 subjects with
the highest PVLs (1022.8to 1021.6copies/cell), as described in a
previously published longitudinal study . Adequate remain-
ing specimens and complete current measurements were avail-
able from 51 of these 60 subjects. Subjects included 36 women
and 15 men (age range, 37–85 years); 18 were Hispanic, 12 were
black, and 18 were white (data not available for 3 subjects). In
total, 33 samples belonged to HTLV-2 subtype A and 5 be-
longed to HTLV-2 subtype B, while data were unavailable for
the remaining samples. None of these 51 samples originated
Received 23 May 2011; accepted 23 August 2011; electronically published 7 November
Presented in part: XVth International Meeting on Human Retrovirology, Leuven, Belgium,
4–8 June 2011. Abstract 184.
Correspondence: Renaud Mahieux, PhD, Equipe Oncogene `se Re ´trovirale, INSERM-U758
Virologie Humaine, 69364 Lyon Cedex 07, France (firstname.lastname@example.org).
The Journal of Infectious Diseases
? The Author 2011. Published by Oxford University Press on behalf of the Infectious
Diseases Society of America. All rights reserved. For Permissions, please e-mail:
d JID 2012:205 (1 January)
d BRIEF REPORT
from patients presenting a clinical manifestation. The 336
samples that were not used had also a detectable PVL. Frozen
peripheral blood mononuclear cells (53106PBMCs) were
obtained from visits 7 or 8, approximately 15 years after the
baseline visit. All subjects gave written informed consent for
future research use of their biologic specimens, and the pro-
tocol was approved by the Committee on Human Research of
the University of California, San Francisco, consistent with the
Declaration of Helsinki.
High-molecular-weight DNA as well as total RNA was ex-
tracted from the same cryopreserved PBMC samples, thus al-
lowing us to perform an accurate measurement of HTLV-2 PVL
and of both APH-2 and tax transcripts. Because these PBMCs
were never placed in culture, the level of tax and APH-2 tran-
C19 cells, uninfected Jurkat cells, or Kit225 cells were grown in
Roswell Park Memorial Institute (RPMI) 1640 medium, sup-
plemented with 10% fetal bovine serum (FBS) and antibiotics,
and maintained at 37?C in 5% CO2.
High-molecular-weight DNA was extracted, and HTLV-2
PVL was measured in duplicate using a previously described
real-time polymerase chain reaction(PCR)technique .RNA
was extracted from the same ex vivo samples using TRIzol re-
agent(Invitrogen)and keptat280?C until usein real-time PCR
experiments. Complementary DNA (cDNA) was obtained from
500 ng of RNA using High Capacity RNA-to-cDNA Master
Mix (Applied Biosystems). RNA samples were incubated for
5 minutes at 25?C, 30 minutes at 42?C, and 5 minutes at 85?C in
a GeneAmp PCR System 9700 thermocycler (Applied Bio-
systems). The cDNA samples were diluted (1/10), and 2.5 lL of
Master (Roche) containing ROX (Carboxy-X-rhodamine) as an
internal reference. The cDNA samples were amplified with
APH-2 primers (5#-CCCCAAGACTATTTTAGGAGATTGC-
3# (sense) and 5#-CCGATCCCGACCCCAGAG-3#) (antisense)
or tax primers (5#-GACAGAGCCTCCTATATGG-3# (sense)
and 5#-GGTATTGGAGAGGAGAGC-3#) (antisense) . The
18S primers (5#-GTGGAGCGATTTGTCTGGTT-3# and 5#-
CGCTGAGCCAGTCAGTGTAG-3#) were used for normali-
zation. Samples were incubated for 10 minutes at 95?C; then
40 cycles were performed using a StepOne Plus thermocycler
(Applied Biosystems) (10 seconds at 95?C and 30 seconds at
60?C), and the melt curve was performed between 60?C and
95?C. APH-2 and Tax-2 primers’ efficacy was determined on
a series of C19 cDNA dilutions (0.5 pg to 500 ng). As a control
for DNA contamination, quantitative PCR was directly per-
formed without performing the reverse transcriptase (RT)
step on RNA obtained from 2 randomly chosen HTLV-2
carriers. RNA extracted from uninfected Jurkat cells was also
included in each run. As an interassay reproducibility control,
a known concentration of HTLV-2 C19 cDNA samples was
used in duplicate for each experiment. The data were then
analyzed with StepOne software (version 2.1; Applied
Total RNA was extracted from stable transfectants as de-
scribedabove.One microgram ofDNaseI-treatedtotalRNAwas
used to synthesize cDNA using reverse-transcriptase SuperScript
III and random primers (Invitrogen). To quantify the level of
APH-2 and b-actin transcripts, real-time PCR (see above) was
performed using APH-2 primers (see above) and human b-actin
primers: 5#-AGGCCAACCGCGAGAAGATG-3# (sense) and
5#-CCAGAGGCGTACAGGGATAG-3# (antisense). Relative ex-
pression level of APH-2 was calculated by the delta cycle
threshold (DCt) method.
An MTT assay was performed. Kit225 cells stably transfected
with pME18Sneo/APH-2, pME18Sneo/sHBZ or control empty
vector were maintained in RPMI medium containing 10% FBS
and antibiotics, supplemented with recombinant interleukin
2 (rIL-2; 85 U/mL) and G418 (600 lg/mL). Thirty days after
transfection, stable transfectants were washed 3 times with
RPMI containing 10% FBS and antibiotics and cultured in this
same medium in the absence of rIL-2 and G418. Forty-eight
hours later, cells were counted and resuspended in medium
supplemented with rIL-2 (2.5 U/mL). Twenty thousand cells
were seeded per well in a 96-well plate, and viability was
determined by measuring 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide dye absorbance at the indicated
The Shapiro-Wilk test was used to test the data for normality.
The variables HTLV-II PVL, APH-2, and tax showed skewed
distributions and were log10transformed to approximate nor-
mal distributions. The assay limit of detection (data not shown)
and tax. We used linear regression (PROC CORR) with cal-
culation of coefficients of determination (R2) to explore the
hypothesis of linear correlations between APH-2 and tax
versus other variables of interest. The strength of relationship
between covariates was also assessed using general linear
models (PROC GLM), with similar results. For comparisons
between groups, we used t tests with equal variance (except
tax high vs tax intermediate, which used the Satterthwaite
method). The contingency tables comparing APH-2 versus
tax detection within the 3 PVL groups were analyzed using
Fisher exact test due to low cell counts. All statistical analyses
were performed using SAS software, version 9.1.3 (SAS
PVL values were determined for each sample obtained at visit
7 or 8 and compared to those measured at baseline more than
10 years before. Those results demonstrated that PVL from visit
7 or 8 was strongly correlated with PVL measured at baseline,
confirming, as previously reported , that HTLV-2 PVL does
not increase with time (Figure 1A).
d JID 2012:205 (1 January)
We quantified APH-2 and tax on the same samples and de-
was expressed in 94% of the samples independent of HTLV-2
subtype, and its level was significantly correlated (P 5 .001) with
PVL (Figure 1B). Expression of tax was not detected in the
samples with the lowest PVL but was also correlated with PVL
across all samples (P 5 .0002) (Figure 1C). Consistent with our
previous data showing that APH-2 suppresses Tax-mediated
transactivation on the 5#-LTR , APH-2 was more likely than
tax to be detected in samples with low and intermediate PVL
(Figure 1D and 1E). The level of APH-2 expression differed
significantly between low and high PVL groups (P 5 .0473). The
level of tax expression also differed significantly between the
low and high PVL groups, as well as between the intermediate
and high PVL groups (P , .0001 and P 5 .0074, respectively)
(Figure 1D). Both APH-2 and tax were expressed in samples with
some cells might express both transcripts of the high PVL group
represents heterogeneous populations of cells that express
either tax or APH-2 (Figure 1D and 1E). In addition, APH-2 was
more frequently detected than tax in the low and intermediate
PVL groups (P 5 .046 and P , .0001, respectively) (Figure 1E).
virus type 2 (HTLV-2)–infected individuals. Each dot represents the log PVL measured at baseline as a function of the log PVL measured at visit 7or 8 for
a given HTLV-2 carrier. Coefficients of determination (R2) and P values were derived from linear regression. B, C, Log PVL as a function of (B) log APH-2 or
(C) log tax mRNA expression. Each dot represents 1 sample. The vertical dotted lines represent the limit of detection of the assays. Coefficients of
determination (R2) and P values were derived from linear regression. D, Median log APH (top) or median log tax (bottom) values according to PVL divided
into 3 groups. The t tests were used to derive P values between groups. E, Number of HTLV-2 carriers with detectable or undetectable APH-2 and tax
mRNA, according to PVL divided into 3 groups. Fisher exact tests were used to derive P values for each contingency table.
APH-2 and tax messenger RNA (mRNA) levels vary according to proviral load (PVL). A, PVL is stable over time among human T-cell leukemia
d JID 2012:205 (1 January)
d BRIEF REPORT
Because APH-2 was expressed in most samples tested
(Figure 1B and 1D) and because HBZ promotes T-cell pro-
liferation and HTLV-2 carriers develop lymphocytosis, we then
asked whether APH-2 could also stimulate cell proliferation.
Kit225 CD41T cells were transfected with HBZ, APH-2, or an
empty vector and grown in the presence of neomycin (G418) to
select for transfected cells (Figure 2A). Interleukin 2 (IL-2) was
then removed from the cell culture medium for 48 hours. Cell
proliferation was then assessed in the presence of a low con-
centration of recombinant exogenous IL-2 (2.5 U/mL). As
previously observed, under those conditions, HBZ-positive
cells proliferated, whereas APH-2 cells behaved as control
cells, suggesting that APH-2 is not involved in cell pro-
liferation in vitro (Figure 2B). As a control, APH-2 expres-
sion was assessed and normalized to b-actin (Figure 2C). This
confirmed that suboptimal levels of IL-2 have no effect on
APH-2 expression. These results demonstrate that although
APH-2 is widely expressed in vivo, its function is not to
promote cell proliferation. Consistent with those results, we
did not find any correlation between APH-2 levels and the
number of lymphocytes in vivo (Figure 2D).
The discovery of HBZ and its effects on cell proliferation and
transformation solved a number of questions that have re-
mained unanswered for several years. We previously demon-
strated that HTLV-2 cell lines also express an antisense
transcript that we named APH-2 . As is the case with HBZ,
APH-2 expression is correlated with 5#-LTR downregulation.
Because persons with HTLV-2 develop lymphocytosis, this
prompted us to evaluate APH-2 expression in vivo and APH-2
properties in vitro.
Our results demonstrate that APH-2 is expressed in vivo in
most HTLV-2 carriers and is correlated with PVL, a situation
plasmid. After neomycin selection, cell viability was tested by removing interleukin 2 (IL-2) from the culture medium for 48 h. Recombinant IL-2 (2.5 U/mL)
was later added. Cell viability was then assessed 1, 2, 3, and 4 days later. B, Functional analyses of APH-2 or HBZ genes on proliferation of T cells.
C, Determination of APH-2 expression by quantitative real-time polymerase chain reaction (RT-PCR). Relative expression level of APH2 was calculated by
the DCt method. Values were first normalized to b-actin expression and then compared to the normalized expression of APH-2 in IL-2 (85 U/mL)
conditions, which was considered as 1. B, C, Data represent the mean value of 2 independent experiments. D, APH-2 messenger RNA (mRNA) expression
does not correlate with lymphocyte counts in human T-cell leukemia virus type 2 (HTLV-2) carriers. Each dot represents the log of the absolute lymphocyte
count as a function of the log APH-2 mRNA value obtained for a given HTLV-2 carrier. Coefficients of determination (R2) and P values were derived from
APH-2 expression does not promote lymphocyte proliferation. A, Kit225 lymphocytes were transfected with APH-2, HBZ, or the backbone
d JID 2012:205 (1 January)
that is reminiscent of HBZ [6, 7]. Contrary to previous reports Download full-text
on HTLV-1 , HTLV-2 tax levels were also correlated with
situation, (1) HBZ and tax are generally mutually exclusive
and (2) only HBZ levels are correlated with PVL . Our data
demonstrate that APH-2, in contrast with HBZ, cannot pro-
mote cell proliferation.
It is established that Tax-1 and Tax-2 present a number of
that, even if Tax plays a major role in HTLV-1 pathogenesis,
HBZ is a key player in the late steps of disease development .
The lack of APH-2 effect onlymphocyte proliferation might also
explain the differences in leukemogenesis that are observed be-
tween HTLV-1– and HTLV-2–infected individuals .
with the settings of the quantitative PCR experiments. We thank the dif-
ferent members of the Mahieux laboratory for their helpful suggestions.
Financial support. R. M. is supported by Ecole Normale Supe ´rieure de
Lyon, E. D. by the Ministe `re de la Recherche, and E. L. M. by the National
Heart Lung and Blood Institute (NHLBI) (grant K24-HL-75036). We ac-
knowledge the financial support from l’Association de Recherche sur le
Cancer, from l’Institut National du Cancer, from INSERM, and from
l’Ecole Normale Supe ´rieure de Lyon to our group. Samples were obtained
from the HTLV outcomes study funded by NHLBI grant 2R01-HL-62235.
Potential conflicts of interest. All authors: No reported conflicts.
All authors have submitted the ICMJE Form for Disclosure of Potential
Conflicts of Interest. Conflicts that the editors consider relevant to the
content of the manuscript have been disclosed.
We acknowledge Bariza Blanquier for her help
1. Journo C, Douceron E, Mahieux R. HTLV gene regulation: because
size matters, transcription is not enough. Future Microbiol 2009;
2. Gaudray G, Gachon F, Basbous J, Biard-Piechaczyk M, Devaux C,
Mesnard JM. The complementary strand of the human T-cell leukemia
virus type 1 RNA genome encodes a bZIP transcription factor that
down-regulates viral transcription. J Virol 2002; 76:12813–22.
3. Halin M, Douceron E, Clerc I, et al. Human T-cell leukemia virus type
2 produces a spliced antisense transcript encoding a protein that lacks
a classic bZIP domain but still inhibits Tax2-mediated transcription.
Blood 2009; 114:2427–38.
4. Bartman MT, Kaidarova Z, Hirschkorn D, et al. Long-term increases in
lymphocytes and platelets in human T-lymphotropic virus type II in-
fection. Blood 2008; 112:3995–4002.
5. Satou Y, Yasunaga J, Yoshida M, Matsuoka M. HTLV-I basic leucine
zipper factor gene mRNA supports proliferation of adult T cell leukemia
cells. Proc Natl Acad Sci U S A 2006; 103:720–5.
6. Saito M, Matsuzaki T, Satou Y, et al. In visvo expression of the HBZ
gene of HTLV-1 correlates with proviral load, inflammatory markers
and disease severity in HTLV-1 associated myelopathy/tropical spastic
paraparesis (HAM/TSP). Retrovirology 2009; 6:19.
7. Usui T, Yanagihara K, Tsukasaki K, et al. Characteristic expression of
HTLV-1 basic zipper factor (HBZ) transcripts in HTLV-1 provirus-
positive cells. Retrovirology 2008; 5:34.
8. Satou Y, Yasunaga J, Zhao T, et al. HTLV-1 bZIP factor induces T-cell
lymphoma and systemic inflammation in vivo. PLoS Pathog 2011;
9. Matsuoka M, Green PL. The HBZ gene, a key player in HTLV-1
pathogenesis. Retrovirology 2009; 6:71.
10. Murphy EL, Glynn SA, Fridey J, et al. Increased prevalence of infectious
diseases and other adverse outcomes in human T lymphotropic virus
types I- and II-infected blood donors. Retrovirus Epidemiology Donor
Study (REDS) Study Group. J Infect Dis 1997; 176:1468–75.
11. Kwaan N, Lee TH, Chafets DM, et al. Long-term variations in human T
lymphotropic virus (HTLV)-I and HTLV-II proviral loads and asso-
ciation with clinical data. J Infect Dis 2006; 194:1557–64.
12. Lee TH, Chafets DM, Busch MP, Murphy EL. Quantitation of HTLV-I
and II proviral load using real-time quantitative PCR with SYBR Green
chemistry. J Clin Virol 2004; 31:275–82.
13. Besson G, Kazanji M. One-step, multiplex, real-time PCR assay with
molecular beacon probes for simultaneous detection, differentiation,
and quantification of human T-cell leukemia virus types 1, 2, and 3.
J Clin Microbiol 2009; 47:1129–35.
14. Roucoux DF, Murphy EL. The epidemiology and disease outcomes of
human T-lymphotropic virus type II. AIDS Rev 2004; 6:144–54.
d JID 2012:205 (1 January)
d BRIEF REPORT