| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,2
* Laboratory of Cell Biology,
Laboratory of Molecular Biology, National Cancer Institute,
Laboratory of Molecular Microbiology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland 20892, USA
3Correspondence: Laboratory of Cell Biology, Building 37, Room 1A09, National Cancer Institute, National Institutes of Health, 37 Convent Drive MSC 4255, Bethesda, MD 20892-4255, USA. E-mail: mgottesman{at}nih.gov
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: ABC transporters • acquired immunodeficiency syndrome • AIDS • P-glycoprotein • virus entry
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. It comprises 12 transmembrane segments and 2
nucleotide binding domains, and belongs to the ATP binding cassette
(ABC) family of transporters. Some members of the family of ABC
transporters are primarily involved in peptide transport. These include
the Salmonella typhimurium oligopeptide permease
(2
3
4
5)
, the yeast STE6 protein involved in the secretion
of the a factor mating pheromone peptide (6)
,
and the mammalian MHC-I antigen-presenting TAP1/2 complex involved in
import of peptides into the lumen of the endoplasmic reticulum
(7)
. P-gp is also capable of transporting various
hydrophobic peptides, such as the toxic peptide ALLN (8)
and various biologically active hydrophobic peptide derivatives,
including ionophores, chemoattractants, immunosuppressants (9
, 10)
, and synthetic peptides (11)
.
Entry of the human immunodeficiency virus type 1 (HIV-1) into host
cells is mediated through the viral envelope protein. Initially
synthesized as precursor gp160, the HIV-1 envelope protein is
proteolytically processed intracellularly into surface gp120 and
noncovalently associated gp41 transmembrane proteins. Binding of
virion-associated gp120 to CD4 and a coreceptor (e.g., CXCR4 for
T-tropic viruses and CCR5 for M-tropic viruses) induces gp41 to undergo
conformational changes and results in the fusion of the virus with the
target cell (see refs 12
). Since entry of HIV-1 into cells
involves interaction of the hydrophobic amino-terminal fusion domain of
gp41 with the cellular plasma membrane and since P-gp is capable of
extruding hydrophobic peptides, including possibly the hydrophobic
protein domains essential for viral fusion, we investigated whether the
overexpression of P-gp interfered with HIV-1 infectivity. Here we show
that P-gp expression in T cells indeed inhibited HIV-1 infectivity both
during fusion with the plasma membrane and at a later step in the virus
life cycle.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 14)
. 12D7 cells expressing
MDR1 alone were initially selected at 5 ng/ml vincristine
(vinc) and maintained at 10 ng/ml vinc. 12D7 cells expressing NEO were
selected in 0.9 mg/ml G418. Those expressing the bicistronic constructs
were initially selected in 25 ng/ml methotrexate (MTX) and maintained
in 40 ng/ml MTX in media where fetal bovine serum (FBS) was substituted
with dialyzed FBS. Once a stable population of cells had been obtained,
expression of the transgene (e.g., MDR1) remained constant
for approximately a month even in the absence of continued drug
selection. Hence, the cells were removed from drug selection 1–2 wk
before HIV-1 infection experiments, and P-gp expression levels were
examined prior to the infection experiments. P-gp expression did not
influence growth rates of the transfected cells.
Kinetics of HIV-1 production in 12D7 cells expressing the various
genes
Parallel cultures of 2 x 106 12D7
cells separately expressing NEO, MDR1,
MDR1wt-DHFR, or MDR1mt-DHFR were infected with
HIV-1NL4–3 at a multiplicity of infection (MOI)
of 0.00375. Media containing no cells served as the negative control
while the parental 12D7 cells served as a positive control. Virus
production was assayed by measuring reverse transcriptase (RT) activity
(15)
in the supernatants at time intervals as indicated.
Fluorescence-activated cell sorting (FACS) analyses of 12D7 or
12D7-MDR1 cells triply stained with antibodies (Ab) against
P-gp (MRK16), CD4, and CXCR4 (12G5)
Approximately 2 x 105 cells were
incubated on ice for 40 min with 4 µl of either 12G5 Ab or its
isotype control. After three washes in Hank’s medium containing 1%
bovine serum albumin (HB), the cells were incubated with 1 µl of FITC
anti-mouse IgG2a (PharMingen, San Diego, Calif.)
for another 40 min. After another three washes in HB, these cells were
incubated with either biotinylated MRK16-Fab or its isotype control at
4°C for 40 min. The cells were washed three more times in HB and
incubated with 4 µl streptavidin-RPE (Life Technologies, Inc.-Gibco,
Gaithersburg, Md.) and 5 µl CD4-APC (Becton Dickinson, Franklin
Lakes, N.J.) or its isotype controls. After a final three washes in HB,
the cells were resuspended in phosphate-buffered saline (PBS)
containing 3.7% formaldehyde for FACS analyses using a FACSort flow
cytometer equipped with CellQuest software (Becton Dickinson, San Jose,
Calif.). Single antibody-stained and isotype control-stained cells were
used as controls.
Cell fusion between MDR1-expressing and HIV-1
envelope-expressing cells
A recombinant vaccinia virus (vv) -based transient expression
system as described previously (16
, 17)
was used to study
cell fusion. Briefly, effector HeLa cells were coinfected with 10
pfu/cell of VCB21 (vv expressing ß-galactosidase under the control of
the T7 promoter) and either VCB60 (HIV-1 wild-type envelope expressing
vv) or VCB16 (vv expressing mutant HIV-1 envelope, which was deleted in
the gp120/gp41 cleavage site rendering the protein nonfusogenic).
Target HeLa cells were coinfected with VCB3 (CD4-expressing vv),
VTF7–3 (vv expressing T7 RNA polymerase), and either of the following
vv: WR (wild-type control), MDR1, D555N (amino-terminal ATP
binding site mutant of MDR1), V185 (substrate mutant of
MDR1 that changes substrate specificity of MDR1)
(18)
, or CFTR (19)
. Infection was initiated
in media containing 2.5% FBS at 32°C with gentle shaking every 15
min. After 1.5 h, the cells were trypsinized, washed and
resuspended in media containing 10% FBS, and incubated at 32°C
overnight. The next day, 1 x 105 effector
and target cells were allowed to fuse at 37°C for 2.5 h. Cells
were then lysed in 0.5% Nonidet P-40, and aliquots of the samples were
mixed with an equal volume of 2x ß-gal substrate (chlorphenol
red-ß-D-galactopyranoside) (CPRG). Rates of substrate hydrolysis
(OD/min) were monitored at room temperature by measuring absorbance at
570 nm using a SpectraMax 250 microplate reader (Molecular Devices,
Sunnyvale, Calif.). CPRG hydrolysis from mixing of cells containing
mutant HIV-1 envelope lacking the gp120/gp41 cleavage site that
represents background ß-gal activity was subtracted.
Transfection of pNL4–3 DNA into MDR1-expressing cells
pHaMDR1 plasmid was transfected into KB-3–1, a
subclone of HeLa, and selected at increasing concentrations of
vincristine (V) (expressed as ng/ml) to obtain escalated expression of
P-gp. HeLa cells were cotransfected with pCMV-ß-gal,
HIV-1pNL4–3 DNA, and the various constructs
shown in Fig. 1A
. Western blotting using anti-P-gp monoclonal Ab, C219, an
anti-p24 Ab (from NIAID AIDS Reagent and Reference Program, Bethesda,
Md.), or an HIV-1 patient’s serum was performed as described
(20)
to evaluate the expression of P-gp or HIV-1 proteins,
respectively. The MDR1-reversing agents quinidine (20 µM)
and PSC 833 (1 µM) were added to either parental KB-3–1 cells or
MDR1-expressing KB-V1 before
HIV-1pNL4–3 DNA was introduced into the cells by
transfection. HIV-1pNL4–3 DNA was then
introduced into these cells by calcium phosphate precipitation
(21)
. Two days later, RT activity (15
, 22)
was determined in the media, which had been centrifuged to remove
floating cells/cell debris. Quantitation of RT activity was performed
using the STORM 860 phosphorImaging system. Transfection differences
were normalized by cotransfection of pCMV-ß-gal plasmid and
determination of ß-gal activity as described earlier.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 14)
and controls (Fig. 1A
) into
12D7, a human CEM T cell line (23)
, followed by infection
with HIV-1NL4–3. Figure 1B
shows that
whereas the parental 12D7 and control neomycin (NEO) -expressing 12D7
cells supported virus replication as monitored by production of
supernatant RT activity, MDR1-expressing 12D7 cells did not.
Since extrusion of drugs and peptides by P-gp requires active
transport, we evaluated whether functional MDR1 is required
for this effect. Bicistronic constructs expressing either wild-type
(wt) MDR1 or MDR1 mutated (D555N) at the ATP
utilization site (mt) to inactivate P-gp pump function and DHFR were
stably introduced into 12D7 cells by selection with methotrexate
(24
, 25)
. When these selected cells were infected with the
HIV-1NL4–3, neither wt nor mt
MDR1-expressing cells supported HIV-1 production (Fig. 1C
). These results suggest that the physical presence of
P-gp protein, albeit not functioning as a molecular pump, might be
sufficient to block HIV-1 infectivity. Nonetheless, there is a slight
possibility that the effects observed with the MDR1 mutant
in the bicistronic format may be due to the DHFR gene.
The above observation that overexpression of MDR1 inhibits
HIV-1 entry was further confirmed using a single cycle infectivity
assay. HeLa cells, stably expressing CD4
(CD4+-Hela), were transfected with
pHaMDR1 and selected with vincristine.
CD4+-HeLa cells that stably express
MDR1 were then infected with
HIV-1NL4–3, and virus-specific p24 protein
synthesis in the infected cells was examined by Western analyses. HIV-1
p24 expression was reduced approximately fivefold in MAGI cells
expressing MDR1 (data not shown). This result is in
agreement with the MDR1-induced inhibition of HIV-1 progeny
particle production shown in Figs. 1B
, C
.
Since an active P-gp pump does not seem to be required for anti-HIV-1
effects, we asked whether overexpression of MDR1 on the cell
surface alters the amount and/or distribution of the HIV-1 receptor CD4
and/or the coreceptor CXCR4. No major differences were observed between
parental and MDR1-expressing 12D7 cells either by
three-color FACS analyses (Fig. 2
) or fluorescence microscopy (data not shown) although three-color FACS
did show a very small population of cells that expressed
MDR1 but exhibited reduced CD4 and CXCR4 surface expression
(Fig. 2
. 12D7-MDR1 lower leftmost and rightmost). It is unlikely that
this subpopulation of MDR1-expressing, low
CD4/CXCR4-expressing cells was responsible for the significant
attenuation of HIV-1 infectivity in these cells.
|
MDR1 interferes with HIV-1-induced fusion
Fusion between MDR1, CD4-expressing cells and HIV-1
envelope-expressing cells was examined using a previously described,
recombinant vaccinia virus-based transient expression system (16
, 17)
. Effector HeLa cells were coinfected with vaccinia viruses
expressing HIV-1 envelope and a ß-galactosidase (ß-gal) reporter
gene under the control of a T7 promoter. Target HeLa cells were
coinfected with vaccinia viruses that express CD4 receptors,
MDR1 or its mutants/family members and T7 RNA polymerase.
Fusion between the effector and target cells results in activation of
the ß-gal reporter gene, which is monitored by the hydrolysis of
chlorphenol red-ß-D-galactopyranoside (CPRG) (16)
.
Figure 3
shows that fusion was greatly reduced in target cells expressing either
wt MDR1, MDR1 with an inactivating amino-terminal
ATP-utilization site (D555N) mutation that renders the transporter
inactive or MDR1 with a single mutation in the substrate
binding site (G185V) (18)
. These results correlated with
the observation that 12D7 cells expressing MDR1,
MDR1-IRES-DHFR, or MDR1mt-IRES-DHFR inhibited
HIV-1 infection (Fig. 1B
, C
). Surface expression,
as monitored by FACS analyses, of P-gp or its mutants, CD4, and CXCR4,
was similar in all these cells (data not shown). Cells infected with
vaccinia viruses expressing CFTR (19)
, another member of
the ABC family of transporters, also exhibited reduced fusion activity
(Fig. 3)
. Since the amount of CFTR on the surface of cells cannot be
compared directly with the amount of MDR1 on the surface of
cells, it is not possible to compare the extent of inhibition of fusion
by these two molecules.
|
MDR1 also interferes with HIV-1 infection at steps
downstream of fusion
We also examined whether overexpression of MDR1
interfered with steps in the virus life cycle downstream of HIV-1
fusion. To investigate this, the fusion step was bypassed by directly
transfecting HIVpNL4–3 cDNA into KB-3–1 cells,
a subclone of HeLa, or KB-3–1 cells that had been stably transfected
with pHaMDR1 and selected with various concentrations of
vincristine to achieve different levels of P-gp expression. Virus
production was determined by measuring reverse transcriptase activity
in the media of these cells. Reduction in HIV-1 virus production
correlates with increasing expression of P-gp (Fig. 4A
). To rule out the possibility that overexpression of MDR1
interfered with the synthesis of HIV-1 proteins, parental KB-3–1 or
its multidrug-resistant derivative, KB-V1 (expressing high levels of
P-gp) were transfected with the HIV-1pNL4–3
molecular clone. Viral proteins inside the transfected cells were
determined using Western blot analyses whereas virus production was
examined by measuring reverse transcriptase activity in the media of
these cells. The major HIV-1 protein bands were detected in KB-3–1 and
KB-V1 cells, although there was a reduction of RT activity in KB-V1
cells (Fig. 5
).
|
|
To determine whether functional MDR1 is required for this
effect, HeLa cells were cotransfected with pNL4–3 cDNA, control
pCMV-ß-gal, and each of the constructs shown in Fig. 1A
,
and progeny virion production was determined. Although high levels of
virus were released after cotransfection of HIV-1 (plus salmon sperm
DNA) or HIV-1 and pHaNEO, cotransfection of HIV-1 with either
pHaMDR1, pHMIDwt, or pHMIDmt greatly reduced virus
production (Fig. 4B
).
MDR1 function requires substrate binding and
substrate-induced ATPase activity that results in substrate
translocation (pumping). The data presented above indicate that pump
function is not needed for the MDR1 effect on HIV-1
production. To determine whether MDR1-reversing agents,
which interfere with substrate binding, can restore HIV-1 infectivity
in MDR1-expressing cells, KB-3–1 (parental) or its
multidrug-resistant derivative KB-V1 were incubated with P-gp
inhibitors, quinidine, and PSC 833 and then transfected with pNL4–3
cDNA as described (22)
. Whereas KB-V1 cells showed reduced
virus production compared to parental KB-3–1 cells, incubation of
KB-V1 cells with either quinidine or PSC 833 increased virus
production, but these agents could not fully restore susceptibility to
HIV-1 (Fig. 4C
). These observations suggest that although a
functional MDR1 pump may not be required for inhibition of
virus production, binding of MDR1-reversing agents to the
hydrophobic substrate binding site(s) of P-gp interferes in part with
the ability of P-gp to block HIV-1 infection.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
that hematopoietic stem cells are resistant to
HIV-1 infection although they express the HIV-1 receptor CD4 and
coreceptor CXCR4, since these cells are known to express P-gp
(27)
. These data are also consistent with another recent
observation indicating that P-gp overexpressing cells are resistant to
other enveloped viruses that enter via the plasma membrane (Y. Raviv,
A. Puri, and R. Blumenthal, unpublished results).
Using the recombinant vaccinia virus-based transient expression system
for assessing fusion, we have some indications that in addition to
MDR1, overexpression of other members of the ABC family of
transporters—e.g., CFTR (Fig. 4)
, STE6, or PDR5 (data not shown)—also
inhibit HIV-1 envelope fusion. This phenotype could be the result of
overexpression of any multimembrane domain-spanning protein or may
reflect the ability of each of these ABC family members to form
hydrophobic binding sites within the plasma membrane. However, though
more extensive studies are needed to evaluate the significance of these
observations, the idea that ABC family members might dominantly inhibit
HIV-1 infection raises the consideration that some multiply
exposed/uninfected individuals might be explained by a gain in
P-gp-like function rather than a loss of coreceptor function.
One important implication of these results is that there may be pockets
of cells within humans (e.g., hematopoietic stem cells)
(26)
expressing high levels of P-gp that are resistant to
HIV-1 infection. Paradoxically, high P-gp expression also reduces
uptake of HIV-1 protease inhibitors (22
, 28)
, so although
P-gp-expressing cells may be relatively resistant to HIV-1 infection,
once these cells are infected it may be more difficult to eradicate the
virus. To produce virus-resistant cell populations, a gene therapy
approach could be used to introduce the MDR1 gene into these
cells. The ability of P-gp to confer multidrug resistance provides a
powerful selection both ex vivo and in vivo for
cells transduced with the MDR1 gene (1)
.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Present address: Canji, Inc., 3525 John Hopkins Ct., San Diego, CA 92121, USA.
Received for publication June 17, 1999. Accepted for publication October 25, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Z. Wang, J. Wang, E. Tantoso, B. Wang, A. Y.P. Tai, L. L.P.J. Ooi, S. S. Chong, and C. G.L. Lee Signatures of recent positive selection at the ATP-binding cassette drug transporter superfamily gene loci Hum. Mol. Genet., June 1, 2007; 16(11): 1367 - 1380. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. H. Chang, C. J. Kochansky, and M. Shou The Role of P-glycoprotein in the Bioactivation of Raloxifene Drug Metab. Dispos., December 1, 2006; 34(12): 2073 - 2078. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Wang, S. Ngoi, J. Wang, S. S. Chong, and C. G. L. Lee The Promoter Region of the MDR1 Gene Is Largely Invariant, but Different Single Nucleotide Polymorphism Haplotypes Affect MDR1 Promoter Activity Differently in Different Cell Lines Mol. Pharmacol., July 1, 2006; 70(1): 267 - 276. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Wang, B. Wang, K. Tang, E. J.D. Lee, S. S. Chong, and C. G.L. Lee A functional polymorphism within the MRP1 gene locus identified through its genomic signature of positive selection Hum. Mol. Genet., July 15, 2005; 14(14): 2075 - 2087. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Owen, B. Chandler, P. G. Bray, S. A. Ward, C. A. Hart, D. J. Back, and S. H. Khoo Functional Correlation of P-Glycoprotein Expression and Genotype with Expression of the Human Immunodeficiency Virus Type 1 Coreceptor CXCR4 J. Virol., November 1, 2004; 78(21): 12022 - 12029. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Tang, L. P. Wong, E. J.D. Lee, S. S. Chong, and C. G.L. Lee Genomic evidence for recent positive selection at the human MDR1 gene locus Hum. Mol. Genet., April 15, 2004; 13(8): 783 - 797. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. U. C. Sankatsing, J. H. Beijnen, A. H. Schinkel, J. M. A. Lange, and J. M. Prins P Glycoprotein in Human Immunodeficiency Virus Type 1 Infection and Therapy Antimicrob. Agents Chemother., April 1, 2004; 48(4): 1073 - 1081. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Siegsmund, U. Brinkmann, E. Schaffeler, G. Weirich, M. Schwab, M. Eichelbaum, P. Fritz, O. Burk, J. Decker, P. Alken, et al. Association of the P-Glycoprotein Transporter MDR1C3435T Polymorphism with the Susceptibility to Renal Epithelial Tumors J. Am. Soc. Nephrol., July 1, 2002; 13(7): 1847 - 1854. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
CD4: CD4 antibody; 
