Group Title: Retrovirology 2007, 4:4
Title: Overlapping enhancer/promoter and transcriptional termination signals in the lentiviral long terminal repeat
Full Citation
Permanent Link:
 Material Information
Title: Overlapping enhancer/promoter and transcriptional termination signals in the lentiviral long terminal repeat
Series Title: Retrovirology 2007, 4:4
Physical Description: Archival
Creator: Yang Q
Lucas A
Son S
Chang LJ
Publication Date: 39104
 Record Information
Bibliographic ID: UF00100251
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: Open Access:


This item has the following downloads:

overlapping ( PDF )

Full Text


BioMed Central

Short report

Overlapping enhancer/promoter and transcriptional termination
signals in the lentiviral long terminal repeat
Qing Yang, Aurore Lucas, Sodany Son and Lung-Ji Chang*

Address: Department of Molecular Genetics and Microbiology, Powell Gene Therapy Center and McKnight Brain Institute, University of Florida,
College of Medicine, Gainesville, Florida 32606, USA
Email: Qing Yang; Aurore Lucas; Sodany Son; Lung-
Ji Chang*
* Corresponding author

Published: 22 January 2007
Retrovirology 2007, 4:4 doi: 10.1 186/1742-4690-4-4
This article is available from:

Received: 2 October 2006
Accepted: 22 January 2007

2007 Yang et al; licensee BioMed Central Ltd.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Oncoretrovirus, but not lentivirus, displays a high transcriptional readthrough activity in the 3' long
terminal repeat (LTR) (Zaiss et al. J. Virol. 76, 7209-7219, 2002). However, the U3-deleted, self-
inactivating (SIN) lentiviral LTR also exhibits high transcriptional readthrough activity. Since the
canonical "core" polyadenylation signal (AAUAAA) of the lentivirus is located in the R-U5 region,
the above finding suggests that additional RNA termination signals must be present in the U3
region. Insertion of alternative termination signals including panhuman T cell leukemia virus type I
polyadenylation signal, a 3' end small intron, and a tertiary tRNA motif into the lentiviral SIN LTR
did not restore the transcriptional termination function. Functional dissection of the U3 region
revealed that 70-80% of the termination signals reside in the transcriptional control region within
124 nt overlapping NFKB, Sp I and TATA binding sites. Serial deletion analysis of the transcriptional
control region indicates that the lentiviral enhancer/promoter elements are essential to the RNA
termination function. These results characterize the mechanism of lentiviral transcriptional
readthrough, which addresses important fundamental and practical issue of RNA readthrough
influencing lentiviral gene function and vector safety.

Lentiviral vectors (LVs) establish long-term transgene
expression in both dividing and non-dividing cells. Exten-
sive deletion of all of the viral genes and most of the LTR
elements are essential to the safety of this vector system [ 1-
3]. The self-inactivating vector (SIN) with minimal
sequence in the viral LTR has been an important safety
improvement in the LV system [4]. However, the LV SIN
LTR displays very high transcriptional readthrough (TR)
activity [5], which potentially increases the risk of activat-
ing downstream cellular oncogenes. Here we examined
activities of potential transcriptional termination ele-

ments in the SIN LVs and functionally dissected U3
sequence to identify key transcriptional termination sig-

Insertion of alternative transcriptional termination
elements in the LV SIN LTR
The RNA readthrough activity was determined using the
sensitive Cre-loxP TE26 reporter cell line as previously
described [5]. We transfected the readthrough reporter
construct, EF-LTR-IRES-nlCre (rtCre), into TE26 cells
which contain a loxP-nlacZ reporter gene whose expres-
sion closely correlates with the readthrough nlCre activity

Page 1 of 8
(page number not for citation purposes)

(Fig. 1). We confirmed that the WT LTR exhibits low tran-
scriptional readthrough activity, whereas the SIN LTR
exhibits high readthrough activity (Fig. 1, bottom).

Transcriptional termination and RNA polyadenylation are
regulated by different molecular mechanisms [6,7]. To
reduce the readthrough rate of the LV SIN LTR, three dif-
ferent RNA termination signals were tested: the polyade-
nylation signal of HTLV-1 (HTLVpA), which is located in
U3 rather than R-U5 of the HTLV-I LTR; the small intron
near the 3' end of the human growth hormone gene (hGH
intron), which may promote the RNA polymerase II ter-
mination function [8-12]; and a mutated tRNA motif
(mu-tRNA), which forms a cloverleaf structure that may
serve as a boundary element to block transcription (Fig.
2A) [11]. The nlCRE reporter constructs containing these

individual elements (mu-tRNA, inserted in either forward
or reverse orientation) were transfected into TE26 cells
and the transcriptional readthrough activity was deter-
mined by nlacZ enzyme assay. When compared to WT and
SIN LTRs, neither of these alternative termination signals
restored transcriptional termination activity (Fig. 2B).
Instead, all of these chimeric LTRs consistently exhibited
increased readthrough activity (p < 0.05).

Functional dissection of the lentiviral U3
Several genetic elements in U3 and R regions of the
human immunodeficiency virus type I (HIV-1) LTR have
been shown to play a role in transcriptional termination
[10][13-15]. To identify the U3 elements that are critical
to transcriptional termination, we systemically restored
WT U3 elements back into the SIN LTR in the nlCre


M EFa U3 R U s re rtCre Construct
serial -

TE26 1
TEr26 CEFla neo ............. .>
Reporter Cells FFi

.............. e M nlacZ .............

r .

.S :.. '.

* L




Figure I
The lentiviral LTR transcriptional readthrough assay. The expression level of nlCre, which serves as an index of tran-
scriptional readthrough of the LV LTR, is directly correlated with nlacZ expression in TE26 reporter cells as previously estab-
lished [5]. The SIN LTR, which contains only the attL sequence of the U3, exhibits a high readthrough rate, while the WT LTR
exhibits a low readthrough rate.

Page 2 of 8
(page number not for citation purposes)


Retrovirology 2007, 4:4

EFUa HILVpA R U5 IRES gjlre SV4O polyA

bGH intron

y ~~
v *

800 -

1 2 3
WT mu-tRNA

Forward Reverse

4 5


Figure 2
Insertion of alternative termination signals fail to reduce the transcriptional readthrough activity of LV SIN
LTR. A. Schematic diagram of lentiviral SIN LTRs containing various termination signal elements. HTLVpA (a 130 bp fragment
from MT4 cell genomic DNA) and hGH intron (a 193 bp fragment from pXO-hGH) were PCR amplified and cloned into pdl-
EF-3'LTR-IRES-nlCre. The mutated tRNA (mu-tRNA, 236 nt) sequence was generated by annealing several synthetic oligos and
cloned into LV SIN vector. All amplified fragments were verified by sequencing of the subclones and the final reporter clones.
B. Quantitative analysis of termination signal insertion on SIN LTR readthrough. One-way ANOVA analysis reveals a significant
difference between WT and the other groups (p < 0.05). No significant difference is detected between the different chimeric
termination signal SIN LTR constructs. The modest difference seen between the chimeric termination signal SIN LTRs and the
SIN LTR is significant (p < 0.05, marked by asterisk).

reporter construct by sequential PCR (Fig. 3A) and tested
their transcriptional readthrough activity (Fig. 3B &3C).
The upstream signaling element (USE), located 77-94 nt
5' to the HIV-1 polyadenylation site (AAUAAA), has been
shown to bind to the cleavage polyadenylation specificity
factor (CPSF) and directly stabilizes the polymerase II
polyadenylation complex formation [16]. USE has been
reported to enhance HIV-1 3' RNA processing by approxi-
mately 70% [17,18]. We restored the full-length USE ele-
ment and generated USE-nlCre (Fig. 3C bottom), and
transfected TE26 cells with different amounts of USE-

nlCre DNA and counted the blue-nucleated cells 48 h
later. The results indicate that restoring USE reduced TR
frequency by about 20% (Fig. 3C, USE-nlCre vs. SIN-
nlCre, p < 0.05).

The transcriptional readthrough activity of USE-nlCRE
was still 8-fold higher than that of the WT LTR. To identify
additional termination signals, we generated a series of
U3-restored rtCre reporter constructs (Fig. 3B). Different
lengths of U3, 349 nt, 228 nt, and 124 nt, designated as
U3A, U3B, and U3C, respectively, were inserted 5' to the

Page 3 of 8
(page number not for citation purposes)

*, p < 0.05, one way ANOVA



Retrovirology 2007, 4:4

EFtc 3'LTR IRES nlCre

mutagenesis primer
prnmer name primer sequegce

HIV-1 3' LTR U3 Deletion Series

attL NRE USF Spl |
n r r- i r- mrm

I 3 I R I Us I
I mA 453kt I R I us |

I U38ae r I R I | I
I U3C124 I A I us I
SInl e Ig NI a I us I

100 200 3o0 4o00 So 600
RNA Readthrough Rate

C HIV-1 3' LTR USE I Transcription Control Region Deletion Series

I Uc 1*nt I A I s I
-dikB)lu3D Sol.TAl R I Us I
[di kBJSpl) IU3TA R US
U3C -1TATA R I I us I
USE-nlCre uSEl R I t
SIN- nCre S sNI R U

~I 1*

0 100 200 300 400 500 600
RNA Readthrough Rate

Figure 3
Serial deletions in U3 and in the transcriptional control region and functional analysis of transcriptional
readthrough. A. The U3 deletion mutagenesis strategy and oligonucleotide primer list. The PCR strategy used for the gener-
ation of nlCre reporter constructs is illustrated. The first PCR product, generated by EF I a and a mutagenesis primer, was used
as a mega-primer in a second PCR with the IRES primer to generate the DNA fragment for nlCre plasmid cloning. All mutagen-
esis constructs were verified by DNA sequencing. The amplified products were cloned into pdl-EF-3'LTR-IRES-nlCre (Fig. I).
B. and C. The 3' LTR deletion series. Genetic structure of the LTR U3 is illustrated at the top diagram. The deletion clones
were generated in pBluescript subclones verified by squencing before being swapped into the final reporter construct. The
reporter constructs were verified again by sequencing. TE26 cells were transfected with 0.1 or 0.2 itg of different plasmid DNA
and 48 h later, fixed and stained with 5-bromo-4-chloro-3-indolyl-3-D-galactopyranoside, and the RNA readthrough rate was
determined. The bar graphs to the right summarize the results of RNA readthrough analysis. The results shown are represent-
ative of more than three independent duplicate or triplicate transfections.

Page 4 of 8
(page number not for citation purposes)

Retrovirology 2007, 4:4

R region in the SIN nlCre reporter construct as illustrated
in Fig. 3B. U3C includes the two NFKB, three Spl binding
sites, and the TATA Box. U3B extends further upstream to
the NF-AT and USF binding sites. U3A retains most of U3
sequence including the 5' NRE. These constructs were
transfected into TE26 cells and the TR activity was deter-
mined. The results reveal that the length of the U3 inser-
tion is inversely correlated with the readthrough activity,
suggesting that the termination signals spread across the
entire U3. Nevertheless, the U3C construct exhibited a
low RNA readthrough rate, 20 to 30% of the SIN-nlCre
readthrough rate, close to that of the WT LTR (U3C 124 nt
vs. SIN and WT 445 nt). Therefore, the majority activity of
the termination signals appear to fall within the 124 nt of

Analysis of key termination signals contributing to LV RNA
U3C contains NFKB and Spl binding sites and TATA box.
To dissect the role of these transcription factor binding
sites, we generated additional deletions in U3C, which
included deletion of NFKB binding sites (U3D), deletion
of NFKB and Spl binding sites (U3E), and deletion of
TATA box alone (U3C-10 TATA), as shown in Fig. 3C.
Results of transcriptional readthrough assays show that
deletion of NFKB binding sites (U3D) significantly
increases the RNA readthrough rate (p < 0.05), suggesting
a critical role in transcriptional readthrough. Further dele-
tion of Spl binding sites (U3E) follows the trend with
more readthrough. Interestingly, the deletion of TATA box
alone (U3C-10 TATA) with intact NFKB and Spl binding
sites also results in high readthrough, indicating that the
basic promoter element (or function) is critical to the
transcriptional termination function. The deletion of all
enhancer/promoter elements plus USE (SIN LTR) results
in the highest RNA readthrough rate. Therefore, the HIV-
1 enhancer/promoter elements are critical transcriptional
termination elements.

NFKB enhances the SIN LTR transcriptional termination
The above U3 dissection study suggests that NFKB may
play a critical role in the LV RNA 3' transcriptional termi-
nation function. To evaluate the relevance of the NFKB
trans-acting factor, we silenced the expression of NFKB in
TE26 using lentiviral siRNA targeting the p50 of the NFKB
dimer and compared the readthrough efficiency of SIN
LTR, WT LTR and U3C LTR in both control and NFKB-
silenced TE26 cells. TE26 cells were transduced with lenti-
viral siRNA vector (LV/KB-siRNA with a puromycin-resist-
ant gene) and after several passages under puromycin
selection, the p50 NFKB RNA was quantified by real-time
RT-PCR. The RNA analysis showed that the siRNA sup-
pressed more than 95% of the NFKB RNA expression (Fig.
4A). This was confirmed by Western analysis using anti-

NFKB antibody illustrating more than 95% inhibition of
expression of both p105 and p50 of the NFkB family of
proteins (Fig. 4B). To test the effect of NFKB on transcrip-
tional readthrough, TE26 and TE26-siNFKB cells were
transfected with different amount of SIN LTR, U3C LTR or
WT LTR rtCre plasmids and the readthrough activities
were determined. Repeated experiments demonstrate that
in the NFKB-knockdown TE26 cells, the readthrough rate
of U3C LTR and WT LTR, both of which contain two NFKB
binding sites, is consistently up-regulated, but not for the
SIN LTR which does not contain any NFKB binding site.
Statistical analysis also supports that the readthrough rate
in the NFKB-silenced TE26 cells is significantly increased
for both the WT LTR (p < 0.05) and the U3C LTR (p < 0.1,
paired t-test) constructs (Fig. 4C).

The WT U3 contains multiple transcriptional factor bind-
ing sites. Our results suggest that the binding of transcrip-
tional factors may contribute to the enhanced polymerase
II termination function. To test if a foreign transcriptional
factor binding site could block RNA readthrough, we
inserted a synthetic seven-consecutive Tet-responsive ele-
ment (TRE) in the SIN rtCre reporter plasmid and tested
its readthrough activity in the presence or absence of a
TRE-binding reverse tetracycline trans-activator/silencer
(rtTATS). If the binding of rtTATS in the TRE region
enhances transcriptional termination function, a reduced
readthrough activity would be expected. The binding of
rtTATS was induced by doxycycline. The result showed
that insertion of TRE reduced the readthrough rate by
around 20%, regardless of the presence of rtTATS (Fig. 5).
In addition, induction of rtTATS binding to TRE did not
influence the readthrough rate. Therefore, the binding of
a non-native transcriptional factor in the SIN LTR did not
significantly improve the RNA pol II termination activity.

Here we have illustrated that heterologous termination
signal elements such as HTLVpA, a small 3' intron and a
tertiary RNA motif (tRNA) do not restore LV RNA termi-
nation function. In addition to USE, two additional ele-
ments in U3, the transcriptional control region and the
NFAT/USF binding region, contribute significantly to len-
tiviral LTR transcriptional termination (Fig. 3B). Restora-
tion of transcriptional control region alone reduces
readthrough by 70-80%, while insertion of NFAT/USF,
an additional 100 nt upstream of the transcriptional con-
trol region, reduces RNA readthrough to a level even lower
than that of the wild type LTR.

Further dissection of transcriptional control region indi-
cates that the key enhancer/promoter elements overlap
with the transcriptional termination elements, or that
they have an overlapping function. The NFKB binding
sites appear to have a dual role during HIV-1 transcrip-
tion. It is possible that interaction of the NFkB elements

Page 5 of 8
(page number not for citation purposes)

Retrovirology 2007, 4:4


p105 1







Z 100-


p50 p


I M anti-f-tubulin


o TE26 control
l NFKB knockdowr

*: p <0.05, paired t test



Figure 4
Lentiviral siRNA silencing of NFKB reduces RNA readthrough of LV SIN LTR. The NF kB siRNA lentiviral vector
was made by annealing two oligos pre-treated at 95oC for 5 min and gradually cooled down to room temperature. The
annealed products were cloned into pTYF-Puro-hU6r LV-driven by a human U6 promoter. The NFKB silenced TE26 cell line
was generated after puromycin selection and used for RNA TR analysis as described in the article. A. Real-time quantitative
RT-PCR analysis of NFKB mRNA. B. Western analysis of NFKB expression in LV-NFKB siRNA-transduced TE26 cells. C.
Comparison of RNA readthrough rate in TE26 and NFKB-silenced TE26 cells. The significance of difference is analyzed by
paired student t test and shown by asterisks.

directly or indirectly affect binding of the 3' RNA process-
ing factors. In addition to the analysis of "cis-elements",
the readthrough study with the NFKB siRNA supports that
"trans-factors" also play a role. The artificial introduction
of a transcriptional factor binding cassette (seven copies
of the Tet-responsible element, TRE) into SIN LTR resulted
in a modest decrease in RNA readthrough. These addi-
tional findings support that both "cis-elements" and
"trans-elements" (transcriptional factors or DNA binding
proteins) play an important role in lentiviral transcrip-
tional readthrough.

The high RNA readthrough frequency of the lentiviral SIN
vector could compromise the intended safety feature. This
may be overcome by further characterization of the RNA
transcription elongation and termination machinery and
genetically modify the control elements to reduce
readthrough without adverse effects. Recent studies have
established that components of the pol II holoenzyme can
interact with transcription factors, HIV-1 Tat as well as
mRNA processing factors involving capping, splicing, ter-
mination and polyadenylation [19,20][21,22]. Further
analysis of the transcriptional readthrough process of len-

Page 6 of 8
(page number not for citation purposes)

Retrovirology 2007, 4:4

TRE effect on RNA readthrough

EFIa TRE R U5 RES nlCre SV40 poyA
..... 1 Lll~





I *


+ +

- +

Figure 5
Quantitative analysis of TR rate of LV SIN LTR containing TRE binding sites. Tetracycline responsive elements
(TRE) were derived from the Xho I to Xma I fragment of plasmid TREd2eGFP (BD Clontech) and ligated into the LV SIN vec-
tor. The RNA readthrough analysis was performed as previously described in the presence or absence of doxycycline as indi-
cated. One-way ANOVA analysis shows a significant difference between the different TRE constructs and the SIN or WT LTR
construct (p < 0.05).

tiviral LTR will address the fundamental mechanism of
RNA termination and help design for a safer and more
efficient lentiviral vector system.

HIV-1, human immunodeficiency virus type 1; LV, lenti-
viral vector; SIN, self-inactivating; WT, wild type; LTR,
long terminal repeats; hGH, human growth hormone;
HTLV, human T cell leukemia virus; TR, transcriptional
readthrough; USE, upstream signal element; CPSF, cleav-
age polyadenylation specificity factor; TRE, tetracycline
responsive element; ANOVA, analysis of variance.

Competing interests
LJC has declared a financial interest as consultant to a
company and LJC holds patents related to the work that is
described in the present study.

Authors' contributions
All authors participated in the molecular biology studies.
LJC conceived the study. LJC, SS and AL initiated the
design of the study and QY performed the statistical anal-
ysis. LJC and QY participated in the final figure prepara-
tion and drafted the manuscript. All authors read and
approved the final manuscript.

We thank G. Eubanks for discussion and preparation and M. Cotter for crit-
ical reading of the manuscript. This work is supported by a grant from
National Institutes of Health.

I. Federico M: Lentiviruses as gene delivery vectors. Curr Opin Bio-
technot 1999, 10:448-453.
2. Naldini L, Verma IM: Lentiviral vectors. Adv Virus Res 2000,
3. Chang LJ, Gay EE: The molecular genetics of lentiviral vectors
current and future perspectives. Current Gene Therapy 2001,

Page 7 of 8
(page number not for citation purposes)


Retrovirology 2007, 4:4

4. Iwakuma T, Cui Y, Chang LJ: Self-inactivating lentiviral vectors
with U3 and U5 modifications. Virology 1999, 261:120-132.
5. Zaiss AK, Son S, Chang LJ: RNA 3'-readthrough ofoncoretrovi-
rus and lentivirus: implications in vector safety and efficacy.
journal of Virology 2002, 76:7209-7219.
6. Colgan DF, Manley JL: Mechanism and regulation of mRNA
polyadenylation. Genes Dev 1997, I 1:2755-2766.
7. Zhao J, Hyman L, Moore C: Formation of mRNA 3' ends in
eukaryotes: mechanism, regulation, and interrelationships
with other steps in mRNA synthesis. Microbiol Mol Biol Rev 1999,
8. Ahmed YF, Gilmartin GM, Hanly SM, Nevins JR, Greene WC: The
HTLV-1 Rex response element mediates a novel form of
mRNA polyadenylation. Cell 1991, 64:727-737.
9. Niwa M, Rose SD, Berget SM: In vitro polyadenylation is stimu-
lated by the presence of an upstream intron. Genes Dev 1990,
10. DeZazzo JD, Scott JM, Imperiale MJ: Relative roles of signals
upstream of AAUAAA and promoter proximity in regula-
tion of human immunodeficiency virus type I mRNA 3' end
formation. Mol Cell Biol 1992, 12:5555-5562.
I I. Donze D, Adams CR, Rine J, Kamakaka RT: The boundaries of the
silenced HMR domain in Saccharomyces cerevisiae. Genes
Dev 1999, 13:698-708.
12. Dye MJ, Proudfoot NJ: Terminal exon definition occurs cotran-
scriptionally and promotes termination of RNA polymerase
II. Mol Cell 1999, 3:371-378.
13. Brown PH, Tiley LS, Cullen BR: Efficient polyadenylation within
the human immunodeficiency virus type I long terminal
repeat requires flanking U3-specific sequences. j Virol 1991,
14. Valsamakis A, Zeichner S, Carswell S, Alwine JC: The human
immunodeficiency virus type I polyadenylylation signal: A 3'
long terminal repeat element upstream of the AAUAAA
necessary for efficient polyadenylylation. Proc Natl Acad Sci Uso
1991, 88:2108-2112.
15. Gilmartin GM, Fleming ES, Oetjen J: Activation of HIV-I pre-
mRNA 3' processing in vitro requires both an upstream ele-
ment and TAR. Embo 1992, 11:4419-4428.
16. Gilmartin GM, Fleming ES, Oetjen J, Graveley BR: CPSF recogni-
tion of an HIV-I mRNA 3'-processing enhancer: multiple
sequence contacts involved in poly(A) site definition. Genes
Dev 1995, 9:72-83.
17. Valsamakis A, Schek N, Alwine JC: Elements upstream of the
AAUAAA within the human immunodeficiency virus polya-
denylation signal are required for efficient polyadenylation in
vitro. Mol Cell Biol 1992, 12:3699-3705.
18. Klasens BI, Thiesen M, Virtanen A, Berkhout B: The ability of the
HIV-I AAUAAA signal to bind polyadenylation factors is
controlled by local RNA structure. Nucleic Acids Res 1999,
19. Dye MJ, Proudfoot NJ: Multiple transcript cleavage precedes
polymerase release in termination by RNA polymerase II.
Cell 2001, 105:669-68 1.
20. Neugebauer KM: On the importance of being co-transcrip-
tional. j Cell Sci 2002, I 15:3865-3871.
21. Meinhart A, Cramer P: Recognition of RNA polymerase II car-
boxy-terminal domain by 3'-RNA-processing factors. Nature
2004, 430:223-226.
22. BradyJ, Kashanchi F: Tat gets the "green" light on transcription
initiation. Retrovirology 2005, 2:69. Publish with BioMed Central and every
scientist can read your work free of charge
"BioMed Central will be the most significant development for
disseminating the results of biomedical research in our lifetime."
Sir Paul Nurse, Cancer Research UK
Your research papers will be:
available free of charge to the entire biomedical community
peer reviewed and published immediately upon acceptance
cited in PubMed and archived on PubMed Central
yours you keep the copyright

Submit your manuscript here: BioMedcentral adv.asp -

Page 8 of 8
(page number not for citation purposes)

Retrovirology 2007, 4:4

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs