Journal of Neuroscience Research 77:532–539 (2004)
Effects of Apolipoprotein E on the Human
Immunodeﬁciency Virus Protein Tat in
Neuronal Cultures and Synaptosomes
Chava B. Pocernich,1,2 Rukhsana Sultana,1,2 Eugene Hone,3 Jadwiga Turchan,4
Ralph N. Martins,3 Vittorio Calabrese,5 Avindra Nath,4 and D. Allan Butterﬁeld1,2,6*
Department of Chemistry, University of Kentucky, Lexington, Kentucky
Center of Membrane Sciences, University of Kentucky, Lexington, Kentucky
Sir James McCusker Alzheimer’s Disease Research Unit, School of Psychiatry and Clinical Neurosciences,
University of Western Australia, Hollywood Private Hospital, Perth, Western Australia, Australia
Department of Neurology, Johns Hopkins University, Baltimore, Maryland
Department of Chemistry, Division of Biochemistry, University of Catania, Catania, Italy
Sanders-Brown Center in Aging, University of Kentucky, Lexington, Kentucky
Human immunodeﬁciency virus type 1 (HIV-1)-associated lipoproteins (Mahley and Rall, 2000). ApoE is well known
dementia is observed in 20 –30% of patients with acquired for its role in lipid and cholesterol homeostasis but has also
immunodeﬁciency syndrome (AIDS). The 4 allele of the been demonstrated to have immunomodulatory properties
apolipoprotein E (APOE) gene currently is thought to play a in vitro and may regulate smooth muscle and endothelial
role as a risk factor for the development of HIV dementia. cell growth and differentiation (Laskowitz et al., 1998).
The HIV protein Tat is neurotoxic and binds to the same The brain is a major site of synthesis for apoE, which is
receptor as apoE, the low-density lipoprotein receptor- produced primarily by astrocytes. In humans, three pre-
related protein (LRP). In this study, we investigated the role dominant isoforms of the apoE protein exists, apoE2,
apoE plays in Tat toxicity. Synaptosomes from wild-type apoE3, and apoE4, which are thought to have varying
mice treated with Tat had increased reactive oxygen spe-
degrees of antioxidant properties (Miyata and Smith, 1996;
cies (ROS), increased lipid and protein oxidation, and de-
creased mitochondrial membrane potential. Synaptosomes
Lauderback et al., 2002). Inheritance of the 4 allele of the
from APOE-knockout mice also had increased ROS, in- APOE gene has been implicated as a major genetic risk
creased protein oxidation, and decreased mitochon- factor for late-onset Alzheimer’s disease (AD; Strittmatter
drial membrane potential, but to a signiﬁcantly lesser de- et al., 1993; Martins et al., 1995).
gree. Treatment of synaptosomes with heparinase and Tat Human immunodeﬁciency virus type 1 (HIV-1)-
increased Tat-induced oxidative stress, consistent with the associated dementia is observed in 20 –30% of patients
notion of Tat requiring interaction with neuronal mem- with acquired immunodeﬁciency syndrome (AIDS;
branes to induce oxidative damage. Human lipidated Kolson and Gonzalez-Scarano, 2000). HIV infection in
apoE3 greatly protected neurons from Tat-induced toxicity, the brain occurs mainly in macrophages, microglia, and
whereas human lipidated apoE4 showed no protection. We multinucleated giant cells (Lee et al., 1993; Bagasra et
demonstrated that human apoE3 has antioxidant proper- al., 1996). There are conﬂicting reports concerning the
ties against Tat-induced toxicity. Taken together, the data inheritance of APOE4 and the development of HIV
suggest that murine apoE and human apoE4 act similarly dementia. Corder et al. (1998) states that twice as many
and do not protect the cell from Tat-induced toxicity. This HIV patients that carry the apoE4 protein were de-
would allow excess Tat to remain outside the cell and mented or had peripheral neuropathy compared with
interact with synaptosomal membranes, leading to oxida- apoE4-negative HIV patients. In contrast, Dunlop and
tive stress and neurotoxicity, which could contribute to collegues (1997) reported no correlation between HIV
dementia associated with HIV. We show that the antioxi- dementia or encephalitis and APOE genotypes.
dant properties of apoE3 greatly outweigh the competition
for clearance in deterring Tat-induced oxidative stress.
© 2004 Wiley-Liss, Inc. *Correspondence to: Prof. D. Allan Butterﬁeld, Department of Chemistry
and Center of Membrane Sciences, 125 Chemistry-Physics Building, Uni-
Key words: LRP receptor; oxidative stress; ApoE3; versity of Kentucky, Lexington, KY 40506. E-mail: email@example.com
ApoE4; HIV dementia; Tat
Received 27 February 2004; Revised 14 April 2004; Accepted 16 April
Apolipoprotein E (apoE) is a small secreted protein Published online 8 June 2004 in Wiley InterScience (www.
of 34 kDa and a component of several different types of interscience.wiley.com). DOI: 10.1002/jnr.20182
© 2004 Wiley-Liss, Inc.
Effects of ApoE on HIV Protein Tat 533
The HIV-1 protein Tat transactivates viral and cel- buffer. One hundred microliters of sample (200 g of synapto-
lular gene expression; is actively secreted into the extra- somes) incubated with DCFH-DA were placed in triplicate in a
cellular environment, mainly from astrocytes, microglia, 96-well plate, and the resulting ﬂuorescence was measured
and macrophages; and is taken up by neighboring, unin- ( ex 485 nm, em 530 nm). Intracellular esterases convert
fected cells, such as neurons (Sabatier et al., 1991; Ensoli et DCFH-DA to the anionic DCFH, which is unable to diffuse
al., 1993; Chang et al., 1997; Kruman et al., 1998). Tat out of the synaptosomes. Reaction with ROS converts DCFH
protein is neurotoxic in vitro (Kruman et al., 1998; New to dichloroﬂuoroscein (DCF), which is highly ﬂuorescent.
et al., 1998; Chauhan et al., 2003) and in vivo (Jones et al.,
1998; Rappaport et al., 1999). Increased protein oxidation Assessment of Protein and Lipid Oxidation
is observed in the cerebrospinal ﬂuid (CSF; Turchan et al., The level of protein oxidation was assayed as protein
2003) and increased peroxynitrite, 4-lipid, and protein carbonyl levels, an index of protein oxidation (Butterﬁeld and
oxidation in the brain of HIV-demented patients (Boven Stadtman, 1997). An Oxidized Protein Detection Kit (Oxyblot;
et al., 1999; Turchan, 2003). Tat has been reported to bind Oncor, Gaithersburg, MD; catalog No. S7150-Kit) was used, as
to the low-density lipoprotein receptor-related protein previously described (Butterﬁeld et al., 1999; Lauderback et al.,
(LRP), efﬁciently internalized and transported into neu- 2001, 2002). Samples were incubated for 20 min with 12%
ronal nuclei in a biologically active form (Liu et al., 2000). sodium dodecyl sulfate (SDS) and 2,4-dinitrophenylhydrazine
ApoE4 is also a competitive LRP physiological ligand. (DNPH) in 10% triﬂuoroacetic acid, vortexed every 5 min, and
Because Tat is associated with oxidative stress, and the then neutralized with Oxyblot Neutralization solution.
APOE4 allele may be a risk factor for HIV dementia, the HNE and protein carbonyl samples were diluted to ap-
current study investigated the effect apoE and its alleles propriate concentrations, and 600 ng of protein were loaded
have on Tat-induced oxidative stress in synaptosomes and directly onto nitrocellulose paper by the slot blotting technique.
neuronal cultures. Membranes were blocked with 3% bovine serum albumin [BSA;
in phosphate-buffered saline (PBS) with 0.01% sodium azide
MATERIALS AND METHODS and 0.2% Tween-20] for 30 min at room temperature. The
Materials nitrocellulose membrane was incubated with primary rabbit
anti-DNPH protein antibody from Oncor Oxyblot (1:150
Tat (1–72 aa) was prepared as previously described and working dilution) or anti-HNE antibody (1:4,000) for 90 min
was 98% pure (Conant et al., 1996). Six-week-old wild-type and then to a secondary antibody (anti-rabbit IgG coupled to
(WT) C57BL/6J inbred mice (stock 000664) and age-matched alkaline phosphatase) diluted in the blocking solution 1:15,000,
apoE-knockout (KO) B6.129P2-Apoe tm1Unc (stock for 1 hr at room temperature. Membranes were washed after
002052) mice (no apoE is present) were purchased from Jackson every step in washing buffer (PBS with 0.01% sodium azide and
Laboratories (Bar Harbor, ME). Antibody against the lipid 0.2% Tween-20). The nitrocellulose paper was then developed
peroxidation product 4-hydroxy-2-trans-nonenal (HNE), by Sigmafast tablet. Blots were analyzed by using the computer-
HNE11-S, was purchased from Alpha Diagnostics. Heparinase 1 assisted imaging software Scion Imaging (Lauderback et al.,
from Flavobacterium heparinium was purchased from Sigma (St. 2001).
Mitochondrial Membrane Potential
Synaptosomal Preparation Mitochondrial membrane potential was determined with
Cortical synaptosomes were prepared from apoE-KO and the JC-1 dye. Synaptosomes were incubated for the last 30 min
WT mice by ultracentrifugation on discontinuous sucrose gra- of treatment with 10 M JC-1 dye in the dark at 37°C, then
dients as described previously (Lauderback et al., 2001). Protein washed in Locke’s buffer and centrifuged for 3 min at 3,000g.
concentration was determined by the Pierce BCA method, and Synaptosomes (1 mg) were resuspended in Locke’s buffer
synaptosomes were adjusted to 1 mg/ml in Locke’s buffer (500 l) and a 100- l aliquot was placed in triplicate in a
(154 mM NaCl, 5.6 mM KCl, 2.3 mM CaCl2, 1.0 mM MgCl2, 96-well plate, and the resulting ﬂuorescence was measured
3.6 mM NaHCO3, 5 mM glucose, 5 mM HEPES). Samples ( ex 525 nm, em 590 nm).
were treated with 500 nM Tat (1–72) and Tat with 1 U/ml
heparinase 1 for 3 hr at room temperature, with continuous Lipoprotein Vesicles
shaking. The main advantage to synthesizing lipoproteins is that
they do not contain any of the other proteins that are found on
Reactive Oxygen Species Measurement lipoproteins in vivo, yet closely mimic the behavior of their
The measurement of reactive oxygen species (ROS) for- animal-derived counterparts (Bowler et al., 1991; Mamo et al.,
mation in synaptosomes treated with Tat (1–72) was performed 1991; Redgrave et al., 1993). Lipoprotein vesicles were prepared
with dichlorodihydroﬂuoroscin-diacetate (DCFH-DA) as pre- as previously described (Martins et al., 1998). Brieﬂy, triolen
viously described (Kanski et al., 2001). DCFH-DA was dis- (4.5 mg), free cholesterol (0.9 mg), cholesterol oleate (0.5 mg),
solved in ethanol to a concentration of 10 mM. Aliquots of and egg-yolk phosphatydyl choline (2.5 mg) were combined
synaptosomes (1 mg protein level) were incubated with 10 M and purged with nitrogen gas and then desicated overnight.
DCFH-DA in 1 ml of Locke’s buffer for 30 min at 37°C. Excess Lipids were mixed with 8.5 ml of 2.2% glycerol in distilled water
dye was removed by washing three times with Locke’s buffer, and sonicated for 1 hr at 55–56°C with nitrogen purging over
and then synaptosomes were resuspended in 0.5 ml of Locke’s the top. The volume was readjusted to 8 ml by using 2.2%
534 Pocernich et al.
Fig. 1. Tat signiﬁcantly increases ROS formation in both wild-type
(WT) and apoE-KO mouse synaptosomes as assessed by DCF ﬂuor- Fig. 2. Mouse synaptosomal protein oxidation is elevated by Tat. Tat
scence. In comparison with corresponding controls, ROS is more signiﬁcantly increased protein carbonyl levels in both WT and
elevated in WT synaptosomes treated with Tat compared with apoE-KO mouse synaptosomes. In comparison with corresponding
apoE-KO synaptosomes treated with Tat (P .05). Data presented are controls, protein oxidation is signiﬁcantly more elevated in WT syn-
means and SEM (n 6; *P .05 vs. corresponding control, Student’s aptosomes treated with Tat compared with apoE-KO synaptosomes
t-test). See text. treated with Tat (P .05). An increase in protein carbonyl levels was
seen in synaptosomes from WT and apoE-KO mice treated with
1 U/ml heparinase and 500 nM Tat compared with synaptosomes
treated with 500 nM Tat. Data presented are means and SEM (n 6;
glycerol, and 1.14 g dibasic sodium phosphate was added. Lipids *P .05 vs. respective control; #P .05, Student’s t-test). See text.
were layered on top of NaCl gradients (1.065, 1.04, 1.02,
1.006 g/ml) and spun at 35,000 rpm for 1 hr at 20°C with a
Beckman SW41 rotor in a Beckman L8-70M ultracentrifuge. ROS generation leads to increased protein and lipid
Lipids were collected, adjusted to 2 ml with distilled water, and oxidation (Butterﬁeld and Stadtman, 1997). Protein oxi-
stored at room temperature with nitrogen purged above. Lipid dation was measured by the levels of protein carbonyls.
vesicle size was analyzed by electron microscope. ApoE was HNE is a marker of lipid peroxidation and is formed by
added to the lipid solution and sonicated three times for 1 sec free radical oxidation of unsaturated lipids (Esterbauer et
each. al., 1991). Consistent with the ROS ﬁndings, both wild-
type and APOE-KO Tat-treated synaptosomes had signif-
RESULTS icant increases (P .05) in protein carbonyls (Fig. 2). An
In this study, the role of apoE in Tat-induced oxi- increase in HNE levels was seen only in the wild-type
dative damage to synaptosomes and neurons in culture was synaptosomes treated with Tat, not the APOE-KO syn-
investigated. Synaptosomes isolated from WT mice con- aptosomes treated with Tat (Fig. 3). Thus, these results
tain apoE as shown by Lauderback et al. (2001) and Keller demonstrate that Tat directly increased protein and lipid
et al. (2000), and WT and apoE-KO mice have similar oxidation in vitro. Tat-treated wild-type synaptosomes
levels of the LRP receptor (data not shown), which is had a signiﬁcant percentage increase over controls in pro-
consistent with previous reports (Games et al., 1995). To tein carbonyl levels (75%) and HNE levels (50%) com-
measure ROS formation by Tat and the inﬂuence of apoE pared with APOE-KO Tat-treated synaptosomes (34%
in Tat-induced formation of ROS, synaptosomes were and 2%, respectively).
incubated with nonﬂuorescent DCFH-DA, which, upon Mitochondrial membrane potential as measured by
oxidation, produces the highly ﬂuorescent DCF. After JC-1 ﬂuorescence was decreased in both wild-type and
treatment with 500 nM Tat for 3 hr at room temperature apoE-KO synaptosomes treated with Tat. Wild-type syn-
(Fig. 1), there is a signiﬁcant increase in ROS in both aptosomes had a signiﬁcantly larger decrease in mitochon-
wild-type and apoE-KO synaptosomes (P .05), as as- drial membrane potential (P .05) compared with
sessed by increased ﬂuorescence. Tat directly induced apoE-KO synaptosomes (Fig. 4).
ROS formation in vitro in synaptosomes. However, wild- Both Tat and apoE require the presence of the
type synaptosomes containing apoE and treated with Tat protein heparan sulfate proteoglycan (HSPG) in order to
exhibited a signiﬁcant 44% increase (P .05) in ROS be taken up by the LRP receptor (Liu et al., 2000), and
compared with the control, whereas apoE-KO synapto- Tat binds tightly to heparan sulfate (Ziegler and Seelig,
somes treated with Tat only increased DCF ﬂuorescence 2004). We added the enzyme heparinase to cleave HSPG,
by 29% (P .05). The 44% increase in wild-type vs. the thus inhibiting the uptake of Tat and apoE and exposing
29% increase of ROS in apoE-KO is signiﬁcantly different Tat to the outer membrane for a longer period. Protein
(P .01). carbonyl and HNE levels signiﬁcantly increased in both
Effects of ApoE on HIV Protein Tat 535
Fig. 3. Mouse synaptosomal lipid peroxidation is induced by Tat. A
signiﬁcant increase in the lipid peroxidation product HNE was found in
WT synaptosomes treated with Tat but not in apoE-KO synaptosomes Fig. 5. Lipidated human apoE3 partially protects hippocampal neurons
treated with Tat. An increase in HNE levels was observed in synapto- from Tat-induced protein oxidation and lipid peroxidation compared
somes from WT and apoE-KO mice treated with 1 U/ml heparinase with lipidated human apoE4. Primary hippocampal neurons were
and 500 nM Tat compared with synaptosomes treated with 500 nM treated with 10 g/ml lipidated speciﬁc human apoE isoform and
Tat. Data presented are means and SEM (n 6; *P .05 vs. 500 nM Tat for 24 hr. Data are mean and SEM (n 3; *P .05 vs.
corresponding control, #P .05 vs. WT-Tat, Student’s t-test). See E4/Tat and Tat, Student’s t-test). See text.
neurons. Lipidated human apoE3 displayed antioxidant
properties by partially protecting neurons from Tat-
induced protein and lipid oxidation compared with the
effects of Tat and human lipidated apoE4 (Fig. 5).
The present study assessed the effects that Tat have
on synaptosomes from WT and APOE-KO mice and the
effects of lipidated human apoE3 and apoE4 alleles on
Tat-induced oxidative stress. Increased ROS formation,
elevated protein and lipid oxidation, and decreased mito-
chondrial membrane potential in Tat-incubated synapto-
somes from WT mice support data suggesting that Tat
induces oxidative stress and is neurotoxic (Chauhan et al.,
2003; Turchan et al., 2003). WT synaptosomes treated
with Tat displayed a greater increase in ROS, protein
Fig. 4. Synaptosomes from WT mice are more vulnerable to Tat- carbonyls, and HNE levels and a greater decrease in mi-
induced alterations in mitochondrial membrane potential than synap- tochondrial membrane potential compared with
tosomes from apoE-KO mice. Synaptosomes from WT or apoE-KO APOE-KO synaptosomes treated with Tat. Not all pro-
mice were treated with Tat (500 nM) and analyzed for alterations in teins and peptides produce ROS in synaptosomes. For
mitochondrial membrane potential by using the ﬂuorescent indicator example, whereas Alzheimer’s amyloid -peptide (1– 42)
JC-1. Both WT and apoE-KO synaptosomes showed a signiﬁcant induces ROS, substitution of the methionine residue of
decrease from corresponding controls, WT mitochondrial membrane this peptide by norleucine, which is not neurotoxic, does
potential being more affected by Tat (P .05). Data are mean and not produce ROS (Varadarajan et al., 2000). We recently
SEM (n 3; *P .05 vs. corresponding control, Student’s t-test). See
have shown that Tat, incubated with primary cortical rat
neurons, also increases protein and lipid oxidation (C.B.
Pocernich and D.A. Butterﬁeld, unpublished observa-
tions). These data suggest two things: ﬁrst, that competi-
WT and apoE-KO mice treated with 1 U/ml heparinase tion for the LRP receptor between apoE and Tat increases
and 500 nM Tat compared with synaptosomes treated just the length of time for which Tat is in contact with the
with Tat (Figs. 2, 3). membrane, thereby increasing Tat-induced oxidative
The effects of the different human apoE alleles on stress and neurotoxicity, and, second, that murine apoE
Tat-induced toxicity were investigated by the addition of confers no antioxidant properties against Tat-induced ox-
lipidated human apoE and Tat to primary cortical rat idative stress. Alternatively, it is possible that murine apoE,
536 Pocernich et al.
which acts like human apoE4, actually potentiates Tat- toxicity, rat cortical neurons were incubated with lipidated
induced oxidative injury to neurons. human apoE3 and apoE4 isoforms and Tat. Lipidated
ApoE4 and Tat are competitive ligands for the LRP human apoE was used, because lipidation is the natural
(Liu et al., 2000). Neuronal uptake of Tat through the state of this lipoprotein in plasma. Human lipidated
LRP receptor-mediated pathway reportedly inhibits neu- apoE3 greatly attenuated Tat-induced protein and lipid
ronal clearance of other LRP ligands, such as apoE4 (Liu oxidation, whereas human lipidated apoE4 seemed to have
et al., 2000). Both apoE and Tat require initial binding to no such properties against Tat-induced oxidative effects
the protein HSPG before uptake by the LRP receptor (Liu (Fig. 5). These data conﬁrm that murine apoE acts simi-
et al., 2000). Recently, it has been demonstrated that larly to human apoE4, offering no protection against Tat-
heparan sulfate can tightly and rapidly bind Tat (Ziegler induced toxicity.
and Seelig, 2004). The enzyme heparinase cleaves HSPG, Tat-induced neurotoxicity is thought to be mediated
not allowing apoE or Tat to bind to HSPG and thereby through excitotoxic mechanisms. Tat speciﬁcally binds to
inhibiting uptake of both by the LRP receptor. Hepari- rat brain synaptosomal membranes with moderate afﬁnity
nase is not known to cleave Tat (Liu et al., 2000). To test (Kd 2 M; Sabatier et al., 1991). Tat is capable of depo-
the hypothesis that Tat-induced oxidative stress increases larizing rat CA1 hippocampal neurons and human cortical
with increased contact with the membrane, we incubated neurons (Magnuson et al., 1995), increasing intracellular
synaptosomes with heparinase. Tat- and heparinase- Ca2 (Nath et al., 1996), and inducing neuronal death
treated synaptosomes from WT and apoE-KO mice both (Tardieu et al., 1992; Hayman et al., 1993; Magnuson et
displayed increased protein and lipid peroxidation com- al., 1995). In neurons, Tat leads to the activation of
pared with that induced by Tat alone (Figs. 2, 3). These phosphatidylinositol 3-kinase (Milani et al., 1996), to in-
data suggest that Tat-induced toxicity involves contact creased levels of inositol triphosphate (IP3), to calcium
with the membrane and that murine apoE does not act as release from IP3-sensitive endoplasmic reticulum (ER)
an antioxidant. It is not known whether Tat binding to the internal stores (Haughey et al., 1999), and to increased
LRP receptor results in toxicity. Our data show that activity of the protein kinase C isoforms , , and
uptake of Tat by the LRP receptor results in decreased (Borgatti et al., 1998). Tat-induced neurotoxicity is pre-
oxidative stress, which could be interpreted as promoting vented by antagonists of phopholipase C and IP3-sensitive
neuronal survival, but, upon internalization, Tat enters the ER calcium release (Haughey et al., 1999). Tat toxicity is
nucleus of the cell, where it stimulates the transcription of also related to glutamate receptor activation; antagonists of
dormant viral genes. This scenario is particularly impor- N-methyl-D-aspartate (NMDA) and non-NMDA recep-
tant, insofar as, in the nervous system, the LRP receptor is tors partially protect neurons from the toxic effects of Tat
abundantly expressed on neurons (Herz and Strickland, (Hayman et al., 1993; Magnuson et al., 1995; Nath et al.,
2001), and internalized Tat may accelerate HIV dementia. 1996; Haughey et al., 2001). Tat introduced intracellularly
Lipid oxidation in apoE-KO synaptosomes treated through patch recording pipettes does not alter neuronal
with Tat was not signiﬁcantly increased, although a small membrane potentials (Cheng et al., 1998). The above-
yet signiﬁcant increase is observed after the addition of Tat described ﬁndings suggest that Tat is capable of directly
and heparinase, albeit to a signiﬁcantly lesser extent com- exciting neurons and causing excitotoxicity by interacting
pared with synaptosomes treated in the same way from with the cell membrane and causing detrimental down-
WT mice. The increased ROS formation, which is stream effects.
known to lead to protein and lipid oxidation (Butterﬁeld The 4 allele of APOE is a risk factor for AD and also
and Stadtman, 1997), was measured in the synaptosomal possibly for HIV dementia. As noted above, some re-
cytosol and not in the lipid bilayer. The extended time in searchers report that twice as many HIV patients that carry
which Tat interacts with the membrane after heparinase the apoE4 protein were demented or had peripheral neu-
treatment may allow for greater interaction with the lipid ropathy compared with apoE4-negative HIV patients
bilayer and increased lipid oxidation, although our data (Corder et al., 1998). Other researchers reported no cor-
may suggest that Tat has to interact with apoE to induce relation of HIV dementia or encephalitis with APOE
oxidation in the lipid bilayer. Many proteins in the lipid genotypes (Dunlop et al., 1997). Recently, it has been
bilayer protrude partially into the extracellular space and demonstrated that there are increased HNE levels, a
are accessible to other proteins and ROS that are pro- marker of lipid peroxidation, in brain and CSF of HIV
duced; consequently, such exposed proteins can be easily patients with dementia vs. HIV patients without dementia,
oxidized. This might partially explain why Tat-treated and this ﬁnding correlates with the 4 allele of APOE (A.
WT and KO synaptosomes had a greater increase in pro- Nath, unpublished observations). The three alleles for
tein oxidation compared with lipid peroxidation. HNE human APOE have differential antioxidant capabilities,
binds to proteins by Michael addition, thereby adding a E2 E3 E4 (Lauderback et al., 2002; Miyata and
carbonyl. Thus, HNE contributes to protein carbonyl Smith, 1996), and the reverse order displays increased
levels in oxidative stress. injury from stroke, head injury, and amyloid peptide-
ApoE is thought to have antioxidant capabilities in induced toxicity in brain (Chen et al., 1997; Sheng et al.,
brain (Lauderback et al., 2002). To test the antioxidant 1998; Lauderback et al., 2002). Our ﬁndings suggest that
capabilities of human apoE alleles against Tat-induced human apoE3 does have antioxidant properties against
Effects of ApoE on HIV Protein Tat 537
Tat-induced toxicity, supporting the previous studies oxidative stress and neurotoxicity (Butterﬁeld et al., 2001,
listed above. The APOE alleles differ from each other by 2002). ApoE binds A in an allele-speciﬁc manner (LaDu
two amino acids. ApoE2 has cysteine residues at positions et al., 1994; Yang et al., 1997) and clears this toxic peptide
112 and 158, E3 has a cysteine at 112 and an arginine at from the extracellular space through the LRP receptor in
158, and E4 has an arginine at both positions. The simple an allele-speciﬁc manner (Beffert et al., 1999; Yang et al.,
amino acid differences in alleles could account for the 1999; Kang et al., 2000). The removal of A by an
difference in antioxidant capabilities that apoE is thought LRP-mediated pathway is thought to be a mechanism to
to posses (Lauderback et al., 2002). Only one apoE iso- prevent protein and lipid oxidation induced by A inter-
form has been identiﬁed in mice and is most analogous to action with membranes (Lauderback et al., 2001). Simi-
the human apoE4 protein (Rajavashisth et al., 1985), larly, we have shown that Tat interaction with the cell
although functionally there are clear species differences membrane allows for oxidative stress to occur, possibly
(Fagan et al., 2002). Murine apoE is similar to human through interactions with other receptors, such as the
apoE4 in that it possesses an arginine at position 112 NMDA receptor, and clearance of Tat from the mem-
(Rajavashisth et al., 1985). The similar levels of increase in brane prevents continued oxidative stress. Consistent with
protein carbonyls and HNE suggest that murine apoE is previous studies, human apoE3 acts as an antioxidant,
analogous to human apoE4 in not having antioxidant whereas human apoE4 and mouse apoE, which is similar
capabilities toward Tat-induced toxicity. However, this to human apoE4, do not provide protection against Tat-
notion should be tempered, in that murine apoE does not induced toxicity. Further investigations are currently un-
posses an arginine at 61, unlike human apoE4. derway in our laboratory to gain insights into the molec-
There is isoform-speciﬁc binding of apoE to the ular pathways by which Tat induces oxidative stress and
LRP/apha 2 macroglobulin and low-density lipoprotein neurotoxicity and the relationship of apoE alleles to HIV
(LDL) receptors, which might help to explain the isoform- dementia.
speciﬁc protection against Tat-induced oxidative stress.
Afﬁnity to the apoE receptors decreases according to apoE ACKNOWLEDGMENTS
genotype, E4 E3 E2, with E2 having a much greater This work was supported in part by NIH grants
decrease in afﬁnity than E3 (Bohnet et al., 1996; Mamotte MH-64409, AG-10836, and AG-05119 to D.A.B. and
et al., 1999). ApoE3 and -E2 have a decreased afﬁnity to RO1 NS-39253 and P20 RR-15592 to A.N. We thank
the receptor, which facilitates continued competition for Dr. Mark Kindy for useful discussions. R.N.M. and E.H.
the receptor, thereby increasing Tat interaction with the are supported by the McCusker Foundation for Alzhei-
membrane, likely resulting in increased oxidative stress. mer’s Disease Research.
However, the antioxidant properties of apoE2 and apoE3,
resulting from the cysteine residues or possible continued REFERENCES
interaction of apoE2 and E3 with Tat, appear to outweigh Bagasra O, Bachman SE, Jew L, Tawadros R, Cater J, Boden G, Ryan I,
competition for the receptor, thereby reducing Tat- Pomerantz RJ. 1996. Increased human immunodeﬁciency virus type 1
induced oxidative stress. replication in human peripheral blood mononuclear cells induced by
ethanol: potential immunopathogenic mechanisms. J Infect Dis 173:550 –
In the human brain, apoE is produced mainly by 558.
astrocytes. Neuronal apoE is most likely internalized from Beffert U, Aumont N, Dea D, Lussier-Cacan S, Davignon J, Poirier J.
the extracellular space, insofar as in vitro studies have 1999. Apolipoprotein E isoform-speciﬁc reduction of extracellular amy-
demonstrated that apoE is internalized by neurites (Poirier loid in neuronal cultures. Brain Res Mol Brain Res 68:181–185.
et al., 1993). Astrocytes, microglia, and macrophages are Bohnet K, Pillot T, Visvikis S, Sabolovic N, Siest G. 1996. Apolipoprotein
infected with HIV (Nath, 1999), and Tat can be detected (apo) E genotype and apoE concentration determine binding of normal
in HIV-infected astrocytes in vivo (Hudson et al., 2000). very low density lipoproteins to HepG2 cell surface receptors. J Lipid Res
Tat is actively secreted into the extracellular environment 37:1316 –1324.
mainly from astrocytes, microglial, and macrophages and is Borgatti P, Zauli G, Cantley LC, Capitani S. 1998. Extracellular HIV-1 Tat
protein induces a rapid and selective activation of protein kinase C
taken up by neighboring uninfected cells, such as neurons (PKC)-alpha, and -epsilon and -zeta isoforms in PC12 cells. Biochem
(Sabatier et al., 1991; Ensoli et al., 1993; Chang et al., Biophys Res Commun 242:332–337.
1997; Kruman et al., 1998). The interaction of apoE and Boven LA, Gomes L, Hery C, Gray F, Verhoef J, Portegies P, Tardieu M,
Tat with the same receptor places them in close proximity Nottet HS. 1999. Increased peroxynitrite activity in AIDS dementia
to each other, potentially allowing interaction between complex: implications for the neuropathogenesis of HIV-1 infection.
these two proteins to take place. We recently showed that J Immunol 162:4319 – 4327.
Tat released from astrocytes caused mitochondrial dys- Bowler A, Redgrave TG, Mamo JC. 1991. Chylomicron-remnant clear-
function, trimming of neurites, and cell death in neurons ance in homozygote and heterozygote Watanabe-heritable-
but was not toxic to astrocytes (Chauhan et al., 2003). The hyperlipidaemic rabbits is defective. Lack of evidence for an independent
chylomicron-remnant receptor. Biochem J 276:381–386.
decreased mitochondrial membrane potential produced by Butterﬁeld DA, Stadtman ER. 1997. Protein oxidation processes in aging
Tat conceivably could play a role in apoptotic mechanisms brain. Adv Cell Aging Gerontol 2:161–191.
of neuronal death. Butterﬁeld DA, Yatin SM, Varadarajan S, Koppal T. 1999. Amyloid
The relationship of apoE to A may provide insight -peptide-induced oxidative pathways of cell death. Meth Enzymol
into how mouse apoE and Tat might interact. A causes 309:746 –768.
538 Pocernich et al.
Butterﬁeld DA, Drake J, Pocernich C, Castegna A. 2001. Evidence of tricular injection of human immunodeﬁciency virus type 1 (HIV-1) tat
oxidative damage in Alzheimer’s disease brain: central role for amyloid protein causes inﬂammation, gliosis, apoptosis, and ventricular enlarge-
beta-peptide. Trends Mol Med 7:548 –554. ment. J Neuropathol Exp Neurol 57:563–570.
Butterﬁeld DA, Castegna A, Lauderback CM, Drake J. 2002. Evidence that Kang DE, Pietrzik CU, Baum L, Chevallier N, Merriam DE, Kounnas MZ,
amyloid beta-peptide-induced lipid peroxidation and its sequelae in Alz- Wagner SL, Troncoso JC, Kawas CH, Katzman R, Koo EH.
heimer’s disease brain contribute to neuronal death. Neurobiol Aging 2000. Modulation of amyloid beta-protein clearance and Alzheimer’s
23:655– 664. disease susceptibility by the LDL receptor-related protein pathway. J Clin
Chang HC, Samaniego F, Nair BC, Buonaguro L, Ensoli B. 1997. HIV-1 Invest 106:1159 –1166.
Tat protein exits from cells via a leaderless secretory pathway and binds to Kanski J, Lauderback C, Butterﬁeld DA. 2001. 5-Aminosalicylic acid
extracellular matrix-associated heparan sulfate proteoglycans through its protection against oxidative damage to synaptosomal membranes by
basic region. AIDS 11:1421–1431. alkoxyl radicals in vitro. Neurochem Res 26:23–29.
Chauhan A, Turchan J, Pocernich C, Bruce-Keller A, Roth S, Butterﬁeld Keller JN, Lauderback CM, Butterﬁeld DA, Kindy MS, Markesbery WR.
DA, Major EO, Nath A. 2003. Intracellular human immunodeﬁciency 2000. Potential role for apolipoprotein E in maintaining synaptic ho-
virus Tat expression in astrocytes promotes astrocyte survival but induces meostasis. J Neurochem 74:1579 –1586.
potent neurotoxicity at distant sites via axonal transport. J Biol Chem Kolson DL, Gonzalez-Scarano F. 2000. HIV and HIV dementia. J Clin
278:13512–13519. Invest 106:11–13.
Chen P, Mayne M, Power C, Nath A. 1997. The Tat protein of HIV-1 Kruman II, Nath A, Mattson MP. 1998. HIV-1 protein Tat induces
induces tumor necrosis factor-alpha production. Implications for HIV-1- apoptosis of hippocampal neurons by a mechanism involving caspase
associated neurological diseases. J Biol Chem 272:22385–22388. activation, calcium overload, and oxidative stress. Exp Neurol 154:276 –
Cheng J, Nath A, Knudsen B, Hochman S, Geiger JD, Ma M, Magnuson 288.
DS. 1998. Neuronal excitatory properties of human immunodeﬁciency LaDu MJ, Falduto MT, Manelli AM, Reardon CA, Getz GS, Frail DE.
virus type 1 Tat protein. Neuroscience 82:97–106. 1994. Isoform-speciﬁc binding of apolipoprotein E to beta-amyloid.
Conant K, Ma M, Nath A, Major EO. 1996. Extracellular human immu- J Biol Chem 269:23403–23406.
nodeﬁciency virus type 1 Tat protein is associated with an increase in both Laskowitz DT, Horsburgh K, Roses AD. 1998. Apolipoprotein E and the
NF-kappa B binding and protein kinase C activity in primary human CNS response to injury. J Cereb Blood Flow Metab 18:465– 471.
astrocytes. J Virol 70:1384 –1389. Lauderback CM, Hackett JM, Keller JN, Varadarajan S, Szweda L, Kindy
Corder EH, Robertson K, Lannfelt L, Bogdanovic N, Eggertsen G, Wilkins M, Markesbery WR, Butterﬁeld DA. 2001. Vulnerability of synapto-
J, Hall C. 1998. HIV-infected subjects with the E4 allele for APOE have somes from apoE knock-out mice to structural and oxidative modiﬁca-
excess dementia and peripheral neuropathy. Nat Med 4:1182–1184. tions induced by A beta(1– 40): implications for Alzheimer’s disease.
Dunlop O, Goplen AK, Liestol K, Myrvang B, Rootwelt H, Christo- Biochemistry 40:2548 –2554.
phersen B, Kvittingen EA, Maehlen J. 1997. HIV dementia and apoli- Lauderback CM, Kanski J, Hackett JM, Maeda N, Kindy MS, Butterﬁeld
poprotein E. Acta Neurol Scand 95:315–318. DA. 2002. Apolipoprotein E modulates Alzheimer’s Abeta(1– 42)-
Ensoli B, Buonaguro L, Barillari G, Fiorelli V, Gendelman R, Morgan RA, induced oxidative damage to synaptosomes in an allele-speciﬁc manner.
Wingﬁeld P, Gallo RC. 1993. Release, uptake, and effects of extracellular Brain Res 924:90 –97.
human immunodeﬁciency virus type 1 Tat protein on cell growth and Lee SC, Hatch WC, Liu W, Brosnan CF, Dickson DW. 1993. Productive
viral transactivation. J Virol 67:277–287. infection of human fetal microglia in vitro by HIV-1. Ann N Y Acad Sci
Esterbauer H, Schaur RJ, Zollner H. 1991. Chemistry and biochemistry of 693:314 –316.
4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radic Liu Y, Jones M, Hingtgen CM, Bu G, Laribee N, Tanzi RE, Moir RD,
Biol Med 11:81–128. Nath A, He JJ. 2000. Uptake of HIV-1 tat protein mediated by low-
Fagan AM, Watson M, Parsadanian M, Bales KR, Paul SM, Holtzman DM. density lipoprotein receptor-related protein disrupts the neuronal meta-
2002. Human and murine ApoE markedly alters A beta metabolism bolic balance of the receptor ligands. Nat Med 6:1380 –1387.
before and after plaque formation in a mouse model of Alzheimer’s Magnuson DS, Knudsen BE, Geiger JD, Brownstone RM, Nath A. 1995.
disease. Neurobiol Dis 9:305–318. Human immunodeﬁciency virus type 1 tat activates non-N-methyl-D-
Games D, Adams D, Alessandrini R, Barbour R, Berthelette P, Blackwell aspartate excitatory amino acid receptors and causes neurotoxicity. Ann
C, Carr T, Clemens J, Donaldson T, Gillespie F, et al. 1995. Alzheimer- Neurol 37:373–380.
type neuropathology in transgenic mice overexpressing V717F beta- Mahley RW, Rall SC Jr. 2000. Apolipoprotein E: far more than a lipid
amyloid precursor protein. Nature 373:523–527. transport protein. Annu Rev Genom Hum Genet 1:507–537.
Haughey NJ, Holden CP, Nath A, Geiger JD. 1999. Involvement of Mamo JC, Bowler A, Elsegood CL, Redgrave TG. 1991. Defective plasma
inositol 1,4,5-trisphosphate-regulated stores of intracellular calcium in clearance of chylomicron-like lipid emulsions in Watanabe heritable
calcium dysregulation and neuron cell death caused by HIV-1 protein tat. hyperlipidemic rabbits. Biochim Biophys Acta 1081:241–245.
J Neurochem 73:1363–1374. Mamotte CD, Sturm M, Foo JI, van Bockxmeer FM, Taylor RR. 1999.
Haughey NJ, Nath A, Mattson MP, Slevin JT, Geiger JD. 2001. HIV-1 Tat Comparison of the LDL-receptor binding of VLDL and LDL from apoE4
through phosphorylation of NMDA receptors potentiates glutamate ex- and apoE3 homozygotes. Am J Physiol 276:E553–E557.
citotoxicity. J Neurochem 78:457– 467. Martins IJ, Vilcheze C, Mortimer BC, Bittman R, Redgrave TG. 1998.
Hayman M, Arbuthnott G, Harkiss G, Brace H, Filippi P, Philippon V, Sterol side chain length and structure affect the clearance of chylomicron-
Thomson D, Vigne R, Wright A. 1993. Neurotoxicity of peptide ana- like lipid emulsions in rats and mice. J Lipid Res 39:302–312.
logues of the transactivating protein tat from Maedi-Visna virus and Martins RN, Clarnette R, Fisher C, Broe GA, Brooks WS, Montgomery
human immunodeﬁciency virus. Neuroscience 53:1– 6. P, Gandy SE. 1995. ApoE genotypes in Australia: roles in early and late
Herz J, Strickland DK. 2001. LRP: a multifunctional scavenger and signal- onset Alzheimer’s disease and Down’s syndrome. Neuroreport 6:1513–
ing receptor. J Clin Invest 108:779 –784. 1516.
Hudson L, Liu J, Nath A, Jones M, Raghavan R, Narayan O, Male D, Milani D, Mazzoni M, Borgatti P, Zauli G, Cantley L, Capitani S. 1996.
Everall I. 2000. Detection of the human immunodeﬁciency virus regu- Extracellular human immunodeﬁciency virus type-1 Tat protein activates
latory protein tat in CNS tissues. J Neurovirol 6:145–155. phosphatidylinositol 3-kinase in PC12 neuronal cells. J Biol Chem 271:
Jones M, Olafson K, Del Bigio MR, Peeling J, Nath A. 1998. Intraven- 22961–22964.
Effects of ApoE on HIV Protein Tat 539
Miyata M, Smith JD. 1996. Apolipoprotein E allele-speciﬁc antioxidant AM, Pearlstein RD, Roses AD, Warner DS. 1998. Apolipoprotein E
activity and effects on cytotoxicity by oxidative insults and beta-amyloid isoform-speciﬁc differences in outcome from focal ischemia in transgenic
peptides. Nat Genet 14:55– 61. mice. J Cereb Blood Flow Metab 18:361–366.
Nath A. 1999. Pathobiology of human immunodeﬁciency virus dementia. Strittmatter WJ, Saunders AM, Schmechel D, Pericak-Vance M, Enghild J,
Semin Neurol 19:113–127. Salvesen GS, Roses AD. 1993. Apolipoprotein E: high-avidity binding to
Nath A, Psooy K, Martin C, Knudsen B, Magnuson DS, Haughey N, beta-amyloid and increased frequency of type 4 allele in late-onset familial
Geiger JD. 1996. Identiﬁcation of a human immunodeﬁciency virus type Alzheimer disease. Proc Natl Acad Sci USA 90:1977–1981.
1 Tat epitope that is neuroexcitatory and neurotoxic. J Virol 70:1475– Tardieu M, Hery C, Peudenier S, Boespﬂug O, Montagnier L. 1992.
1480. Human immunodeﬁciency virus type 1-infected monocytic cells can
New DR, Maggirwar SB, Epstein LG, Dewhurst S, Gelbard HA. 1998. destroy human neural cells after cell-to-cell adhesion. Ann Neurol 32:
HIV-1 Tat induces neuronal death via tumor necrosis factor-alpha and 11–17.
activation of non-N-methyl-D-aspartate receptors by a NFkappaB- in- Turchan J, Pocernich CB, Gairola C, Chauhan A, Schiﬁtto G, Butterﬁeld
dependent mechanism. J Biol Chem 273:17852–17858.
DA, Buch S, Narayan O, Sinai A, Geiger J, Berger JR, Elford H, Nath A.
Poirier J, Baccichet A, Dea D, Gauthier S. 1993. Cholesterol synthesis and
2003. Oxidative stress in HIV demented patients and protection ex vivo
lipoprotein reuptake during synaptic remodelling in hippocampus in adult
with novel antioxidants. Neurology 60:307–314.
rats. Neuroscience 55:81–90.
Varadarajan S, Yatin S, Aksenova M, Butterﬁeld DA. 2000. Review:
Rajavashisth TB, Kaptein JS, Reue KL, Lusis AJ. 1985. Evolution of
apolipoprotein E: mouse sequence and evidence for an 11-nucleotide Alzheimer’s amyloid beta-peptide-associated free radical oxidative stress
ancestral unit. Proc Natl Acad Sci USA 82:8085– 8089. and neurotoxicity. J Struct Biol 130:184 –208.
Rappaport J, Joseph J, Croul S, Alexander G, Del Valle L, Amini S, Khalili Yang DS, Smith JD, Zhou Z, Gandy SE, Martins RN. 1997. Character-
K. 1999. Molecular pathway involved in HIV-1-induced CNS pathol- ization of the binding of amyloid-beta peptide to cell culture- derived
ogy: role of viral regulatory protein, Tat. J Leukoc Biol 65:458 – 465. native apolipoprotein E2, E3, and E4 isoforms and to isoforms from
Redgrave TG, Ly HL, Quintao EC, Ramberg CF, Boston RC. 1993. human plasma. J Neurochem 68:721–725.
Clearance from plasma of triacylglycerol and cholesteryl ester after intra- Yang DS, Small DH, Seydel U, Smith JD, Hallmayer J, Gandy SE, Martins
venous injection of chylomicron-like lipid emulsions in rats and man. RN. 1999. Apolipoprotein E promotes the binding and uptake of beta-
Biochem J 290:843– 847. amyloid into Chinese hamster ovary cells in an isoform-speciﬁc manner.
Sabatier JM, Vives E, Mabrouk K, Benjouad A, Rochat H, Duval A, Hue Neuroscience 90:1217–1226.
B, Bahraoui E. 1991. Evidence for neurotoxic activity of tat from human Ziegler A, Seelig J. 2004. Interaction of the protein transduction domain of
immunodeﬁciency virus type 1. J Virol 65:961–967. HIV-1 TAT with heparan sulfate: binding mechanism and thermody-
Sheng H, Laskowitz DT, Bennett E, Schmechel DE, Bart RD, Saunders namic parameters. Biophys J 86:254 –263.