Apolipoprotein E-Mediated Immune Regulation in Sepsis1
Omar M. Kattan, F. Behzad Kasravi, Erica L. Elford, Michael T. Schell, and Hobart W. Harris2
Lipids and lipoproteins have emerged as key constituents of the immune response to microbial infection. We, therefore,
sought to understand the complex interaction between lipoprotein metabolism and sepsis. Apolipoprotein E (apoE), a com-
ponent of plasma lipoproteins, has been suggested to bind and traffic Ags for NKT cell activation. However, apoE’s role in
sepsis has not been demonstrated. In this study, we examined the effect of exogenous apoE in a rat model of septic peritonitis,
induced by cecal ligation and puncture. We demonstrate that 48 h after serial injections of apoE, septic mortality increased
in a dose-dependent manner. While sepsis resulted in increased splenic and decreased hepatic and circulating NKT cell
populations, serial injections of apoE for 24 h after cecal ligation and puncture increased the frequency, cell number, and
BrdU uptake in splenic and hepatic NKT cell populations, while concomitantly depleting these populations in the circulation.
These changes were correlated with elevated alanine amino transferase levels, an indicator of liver injury. Interestingly,
while sepsis increased hepatic T cell apoptosis and necrosis, apoE reversed these changes. apoE also promoted increases in
predominantly Th1 cytokine levels in sera and a decrease in IL-4, the main NKT cell-derived Th2 cytokine. Consequently,
apoE treatment is associated with increased sepsis-induced mortality, and increased NKT cell frequency and proliferation
in the liver and spleen, with concomitant decreases in these NKT cell parameters in the peripheral circulation. apoE
treatment also promoted a Th1 cytokine response, increased the degree of liver injury, and decreased apoptosis in hepatic
The Journal of Immunology, 2008, 181: 1399–1408.
immune response to bacterial infections, the immune system be-
comes hyperactive and the distribution of circulating lipoproteins
changes significantly, a phenomenon clinically termed the “li-
pemia of sepsis” (1). Although this phenomenon was originally
thought to represent the fueling of the host response to infection
(2), increasing evidence suggests a more direct role for lipids and
lipoproteins as agents in the immune system (3).
Apolipoprotein E (apoE)3is a multifunctional component of
plasma lipoproteins that is found on very low density lipopro-
tein, low density lipoprotein (LDL), high density lipoprotein,
and chylomicron remnant lipoprotein complexes. First recog-
nized as a major determinant in lipoprotein metabolism and
cardiovascular disease, apoE, a ligand for the LDLR, has
emerged as an important molecular mediator in immunoregu-
lation (4). However, the evidence regarding apoE’s role in the
immune response is contradictory. Lipoproteins containing
apoE have been shown to inhibit or stimulate Ag-induced T
lymphocyte activation and proliferation (5). Additionally, mice
omponents of lipid metabolism have been co-opted to
participate in microbial immunity. During the onset of
sepsis, a fatal disease characterized by an acute systemic
deficient in apoE appear to have an impaired response to bac-
terial infection (6, 7). Studies have also shown that apoE can
bind LPS, attenuate the host inflammatory response, and, thus,
protect against LPS-induced mortality (8). Distinctively, apoE
has also recently been implicated in the activation of NKT cells
by acting as a molecular chaperone for bacterial Ags, delivering
them to APCs via LDLR to activate NKT cells by CD1d, a
nonclassical class-I-like Ag-presenting molecule (9). NKT cells
are characterized by the NK marker (NK1.1) and a semi-invari-
ant TCR expressing the V?14/J?281 gene segments. Upon
stimulation, NKT cells can promptly secrete large amounts of
Th1 and Th2 cytokines, IFN-? and IL-4, respectively (10, 11).
NKT cells appear to act as a functional bridge between innate
and acquired immune systems, highlighting their importance as
a target for infectious diseases (12).
In light of this contradictory evidence, we sought to characterize
apoE’s specific immunomodulatory role in a clinically relevant rat
model of bacterial sepsis-cecal ligation and puncture (CLP). We
found that apoE increased the response to septic shock through
increased NKT cell frequency, number, and proliferation in the
peripheral organs, and downstream responses, thereby contributing
to increased mortality. Our results provide evidence for a role of
apoE, a LDLR ligand, in increasing the immune response to
polymicrobial sepsis, particularly by NKT cells.
Materials and Methods
Male Sprague-Dawley rats (Charles Rivers) weighing 250–300 g were
maintained under standard conditions. All procedures were performed in
full accordance with the policies of the Institutional Animal Care and Use
Committee at University of California, San Francisco.
Recombinant apoE3 was produced in bacteria by using a vector ex-
pressing a thioredoxin fusion protein, as previously described (13). To
Department of Surgery, University of California School of Medicine, San Francisco,
Received for publication May 15, 2007. Accepted for publication May 8, 2008.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1This work was supported by Grant R01 GM057463 from the National Institutes of
Health (to H.W.H.).
2Address correspondence and reprint requests to Dr. Hobart Harris, University of
California Surgical Research Laboratory, Box 1302, San Francisco, CA 94143.
E-mail address: firstname.lastname@example.org
3Abbreviations used in this paper: apoE, apolipoprotein E; LDL, low density lipopro-
tein; CLP, cecal ligation and puncture; EJV, external jugular vein; 7-AAD, 7-ami-
noactinomycin D; ALT, alanine amino transferase.
Copyright © 2008 by The American Association of Immunologists, Inc. 0022-1767/08/$2.00
The Journal of Immunology
exclude toxicity, apoE was cultured for presence of bacterial contam-
ination and a Limulus assay (Cape Cod) was performed to detect en-
dotoxin contamination. Maximum potential endotoxin contamination is
5 pg/mg apoE.
Induction of sepsis by CLP
After general anesthesia was induced, a catheter was inserted in the
rat’s external jugular vein (EJV), as described by Thrivikraman et al.
(14). In brief, after a 1.0-cm incision was made at the base of the neck,
a catheter was placed into the EJV, tunneled behind the shoulder, and
out through the skin on the rat’s back. All incisions were closed using
4–0 silk sutures. After 24 h, rats either underwent CLP or a sham
operation. CLP was performed as described previously (15, 16). In
brief, rats were anesthetized and a 1.5-cm midline incision was made.
The cecum was exposed and 40% was ligated with a 3–0 silk suture.
The cecum was punctured through and through using an 18-gauge nee-
dle, and a small amount of fecal matter was expressed from each punc-
ture hole. The cecum was returned to the abdominal cavity and the
abdominal wall was closed in two layers using a 3–0 silk suture. Rats
that underwent sham operations were treated identically except the ce-
cum was not ligated or punctured.
Beginning immediately at the time of CLP, recombinant apoE3 (1.6
mg/kg or 114 ?g/kg), or an equivalent volume of saline, was injected
into the rats by using the EJV catheter every 4 h for a total of 48 h.
Mortality was measured at day 7 in four groups: sham/saline, sham/
apoE, CLP/saline, and CLP/apoE. For in vivo characterization, rats
were given 114 ?g/kg of apoE or an equivalent volume of saline exactly
as above for 24 h, at which point liver and spleen were harvested for
FACS analysis and blood samples were obtained to measure cytokine
and transaminase levels.
Preparation of lymphocytes
Intrahepatic lymphocytes were isolated as described by Dao et al. (17). In
brief, rats were perfused via the portal vein with digestion medium [RPMI
1640 medium containing 0.2 mg/ml collagenase and 0.02 mg/ml Dnase I
(Roche) and 5% FCS]. The liver was minced and passed through a 70 ?m
nylon cell strainer (BD Falcon) and washed. The suspension was differ-
entially centrifuged over a Percoll gradient [40% Percoll (EMB Bio-
sciences), 150 ? g, 15 min, 4°C]. Cells were washed in RPMI 1640 (300 ?
g, 5 min, 4°C) and resuspended in a red cell lysing buffer (0.15M NH4Cl,
1.0 mM KHCO3, 0.1 mM Na2EDTA) for 3 min. Cells were washed with
PBS (300 ? g, 5 min, 25°C) and resuspended in FACS staining buffer
(0.1% BSA, 0.02% Na2N3in PBS) for FACS analysis. Cell viability was
determined by trypan blue exclusion.
The spleen was minced with two sterile glass slides, passed through a 40
?m nylon cell strainer (BD Falcon) and washed. Cells were resuspended in
a red cell lysing buffer for 3 min. Splenocytes were washed with PBS and
resuspended in FACS staining buffer for FACS analysis. Cell viability was
determined by trypan blue exclusion.
Whole blood was incubated with red cell lysing buffer in a 1:14 ml ratio
for 5 min on a rotator. Cells were washed (2000 rpm, 5 min) and resus-
pended in FACS staining buffer for FACS analysis. Cell viability was
determined by trypan blue exclusion.
Flow cytometry analysis
Abs used included CD45-PECy5, CD161a-PE (NKR-P1A), ??TCR-FITC,
CD8a-PE, anti-CD23, Annexin V-APC, and 7-aminoactinomycin D (7-
AAD) (BD Pharmingen). For T cell counts, 1.0 ? 106cells from the liver
and spleen were preincubated with anti-CD23 to block FC receptors and
decrease nonspecific binding. Samples were then incubated with Abs for 30
min at 4°C in the dark, washed (1200 rpm, 5 min), and resuspended in
FACS staining buffer. To determine positive double-staining, single-stain
controls for each surface marker and IgG controls for each color were used.
For apoptotic staining, 1.0 ? 105cells from the liver were incubated with
Annexin V and/or 7-AAD and incubated for 15 min at 25°C. Three-color
analysis was performed on a FACSCalibur (BD Biosciences) with a
100,000–200,000 event count. Data was analyzed using FlowJo software
BrdU administration and detection
Rats received an i.p. injection of BrdU (50 mg/kg) 3 h after CLP. To detect
BrdU incorporation in lymphocytes harvested 24 h after CLP, we used the
Allophycocyanin BrdU Flow kit (BD Pharmingen). After surface stain-
ing, cells were fixed and permeabilized with Cytofix/Cytoperm buffer
for 15–30 min on ice and then treated with DNase to expose incorpo-
rated BrdU. Subsequently, cells were stained with allophycocyanin-
conjugated anti-BrdU Ab.
Multiplex analysis of serum cytokine concentration
Concentrations of IFN ?, IL-2, IL-4, IL-10, IL-1?, and TNF-? were si-
multaneously quantified in serum samples using an ELISA-based bead
multiplex assay (Linco Research) according to the manufacturer’s instruc-
tions. Samples were analyzed with a Luminex100 plate reader (Linco Re-
search) to determine cytokine concentration. Concentrations for each cy-
tokine in the multiplex assay were calculated from calibration curves using
individual recombinant proteins as standards, according to the manufac-
turer’s instructions. Serum samples were diluted (1/5). StatLIA software
(Brendan Scientific) with a 5-parameter logistic curve-fitting method was
used for data reduction. All specimens were tested in duplicate wells to
assess interassay variability. The sensitivity of the cytokine assay was ?5
pg/ml for all cytokines measured.
A total of 50 ?l of rat serum was used to obtain alanine amino transferase
(ALT) values by using ALT L3K reagent (Diagnostic Chemicals) run by a
standardized, automated analyzer (Cobas Mira Plus; Roche) at room
Flow cytometry and ALT values were compared using two-tailed t tests. Sur-
vival analysis was analyzed by the log-rank test. Multiplex serum cytokine
levels were compared using Kruskal-Wallis and non-parametric Mann-Whit-
ney U tests. A value of p ? 0.05 was regarded as statistically significant.
apoE increases mortality in sepsis
We first investigated whether apoE injected serially at the induc-
tion of and for the duration of septic insult affected the host defense
against endotoxin-induced mortality. Mortality was determined in
four groups of rats treated serially every 4 h for 48 h and followed
for 7 days: sham/apoE-1.6 mg/kg, CLP/saline, CLP/apoE-114 ?g/
kg, and CLP/apoE-1.6 mg/kg. We found that apoE (1.6 mg/kg) had
apoE on survival of wild-type rats subjected to CLP.
After CLP or a sham operation, apoE (114 ?g/kg or 1.6
mg/kg) or vehicle (saline) was immediately injected via
the EJV catheter. Rats then received the same dose con-
centration every 4 h for 48 h and survival was monitored
for 7 days. Differences between survival curves were
determined by log-rank tests. Results are expressed as
percentage of survival. Data represent three independent
experiments. p ? 0.05 [CLP/saline (n ? 20) vs CLP/
apoE 114 ?g/kg (n ? 28) and sham/apoE (n ? 7) vs
CLP/saline]; p ? 0.01 [CLP/apoE 114 ?g/kg vs CLP/
apoE 1.6 mg/kg (n ? 8)]; p ? 0.001 (sham/apoE vs
CLP/apoE 114 ?g/kg, sham/apoE vs CLP/apoE 1.6 mg/
kg, and CLP/saline vs CLP/apoE 1.6 mg/kg).
Effect of immediate administration of
1400apoE INCREASES SEPTIC MORTALITY
no effect on the mortality of sham-operated rats (Fig. 1). The mor-
tality of CLP/saline rats increased significantly when compared
with sham rats, and mortality after CLP was directly related to
apoE concentration. Seven days after CLP, rats that received 114
?g/kg apoE had a mortality rate of 82.14%, whereas all of the rats
that received 1.6 mg/kg died. These findings indicate that apoE,
when administered at the time of and for the duration of the septic
insult, increases mortality in a dose-dependent manner in an in vivo
model of sepsis.
apoE increases splenic and hepatic NKT cell frequencies in
sepsis while depleting circulating NKT cells
In patients with sepsis, the subpopulation of lymphocytes changes
considerably (18). NKT cells, a subset of T lymphocytes, have a
phocytes were isolated from spleens, livers, and whole blood of septic or sham rats, treated with apoE or vehicle. A, Numbers in flow cytometry panels
indicate the percentage of gated lymphocytes. ??TCR: ?? TCR; NK1.1: NK cell-associated marker (CD161). B–D, Rats that underwent sham surgery with
apoE or vehicle treatment had no significant differences in NKT cell frequencies in the spleen (B), liver (C), and circulation (D). Splenic lymphocytes from
septic rats had 18% more NKT cell frequency than did lymphocytes from sham/saline rats (p ? 0.01). NKT cell frequencies were reduced by 40% in the
liver and by 42% in the circulation of septic rats as compared with sham/saline rats (p ? 0.001; p ? 0.01, respectively). Septic rats treated with apoE (114
?g/kg) immediately at the time of CLP and every 4 h for 24 h had increased (p ? 0.01) NKT cell frequencies in their spleen and liver as compared with
septic rats treated with vehicle, (17 and 300% more, respectively), whereas this decreased (p ? 0.01) 47% further in the circulation. Percentages for
individual cell types are shown. Data represent five to nine independent experiments and are reported as a fraction of sham/saline. ??, p ? 0.01; ???, p ?
0.001. Error bars indicate SDs.
In vivo stimulation of CD1d-restricted NKT cell frequencies with apoE treatment during poly-microbial Gram-negative infection. Lym-
1401 The Journal of Immunology
relevant but poorly characterized role in the immunopathogenesis
of septic shock induced by CLP (19). Therefore, we next
investigated whether, in our model, apoE changed NKT cell
frequency in the liver, spleen, and circulation, thereby contributing
to immune activation and mortality. Lymphocytes from liver and
spleen were harvested 24 h after CLP and treatment with saline
Surface marker staining with the ??TCR and NK1.1 Abs
revealed that while there was no statistically significant differ-
ence in NKT cell frequencies in the liver, spleen, and circula-
tion between the sham control groups, CLP-induced sepsis re-
sulted in an increase in the splenic and a decrease in the hepatic
and circulating NKT cells (Fig. 2B, C, and D). Septic rats re-
ceiving 114 ?g/kg of apoE had significantly elevated splenic
and hepatic NKT cell frequencies, whereas those in the circu-
lation were further depleted compared with septic rats that did
not receive apoE (Fig. 2B, C, and D). This finding suggests that
when apoE is added to an in vivo model of sepsis, NKT cell
frequency increases in the liver and spleen, whereas it is de-
pleted in the circulation.
apoE increases NKT cell proliferation and trafficking after CLP
Since the percentage of NKT cells was changing after CLP and
treatment with apoE, we sought to determine the role of prolifer-
ation in these changes by BrdU labeling in NKT cells. Rats re-
ceived one i.p. injection of BrdU 3 h after induction of CLP, and
lymphocytes harvested at 24 h after CLP were measured for BrdU
Flow cytometric analysis showed no significant changes be-
tween sham control groups in all three cell populations, as well as
mg/kg) i.p. 3 h after CLP and lymphocytes were harvested 21 h later and were measured for BrdU incorporation in NKT cells. A, Percentages for BrdU?
cells shown. B–D, Hepatic lymphocytes from septic rats had 60% more BrdU?NKT cells compared with sham rats treated with saline. Splenic and hepatic
lymphocytes from septic rats treated with apoE had 54 and 31% more BrdU?NKT cells than septic rats that did not receive apoE, respectively. Data
represent four independent experiments and are reported as a fraction of sham/saline. ?, p ? 0.05, ??, p ? 0.001. Error bars indicate SD.
apoE-mediated increases in NKT cell proliferation in the spleen, liver, and circulation of septic rats. Rats were injected with BrdU (50
1402apoE INCREASES SEPTIC MORTALITY
no significant change in BrdU?NKT cells after CLP in the spleen
and circulation (Fig. 3B, C, and D). However, CLP-induced sepsis
did result in a significant increase in hepatic NKT cell proliferation
(Fig. 3C). Septic rats that received apoE had significant increases
in splenic and hepatic BrdU?NKT cells compared with those that
did not receive apoE (Fig. 3, B and C). These results suggest that
apoE causes increases in splenic and hepatic NKT cell prolifera-
tion in sepsis, whereas no significant changes occur in the
Despite clear differences in NKT and BrdU?NKT cell percent-
ages, we sought to determine whether changes in an NKT cell
population after apoE treatment were due to trafficking or local
expansion by determining the absolute lymphocyte and NKT cell
numbers. After CLP was induced, the total number of NKT and
BrdU?NKT cells decreased in the liver by 62 and 40%, respec-
tively, and in the circulation by 71 and 78%, respectively, when
compared with sham controls (Table I). However, in septic rats
that received apoE, the total number of NKT and BrdU?NKT
cells increased significantly in the spleen (by 35 and 102%, re-
spectively), and in the liver (by 275 and 437%, respectively), when
compared with septic rats that received saline, but decreased in the
circulation (by 78 and 64%) (Table I). Additionally, in septic rats
treated with apoE, the total number of lymphocytes did not change
in the spleen, increased by 35% in the liver, and decreased by 62%
in the circulation when compared with septic rats that received
saline. This suggests that apoE causes the absolute number of NKT
and BrdU?NKT cells in the circulation to decrease while con-
comitantly increasing them in the spleen and liver during CLP-
apoE selectively mediates stimulation in NKT cells
To investigate whether apoE selectively acts upon NKT cell pop-
ulations, rather than all T lymphocytes, we next measured the lev-
els of CD8?T cells in the spleen and liver of septic rats treated
with apoE for 24 h. Flow cytometry analysis showed that after CLP,
CD8?cytotoxic T cell populations increased by 12% in the spleen,
21% in the liver, and decreased by 48% in the circulation when com-
pared with sham controls. After apoE treatment for 24 h, septic rats
had 28, 101, and 68% more CD8?T cells in the spleen, liver, and
circulation, respectively, than septic rats that received saline; how-
ever, these differences were not statistically significant (Fig. 4B, C,
and D). Therefore, these results suggest that apoE selectively acts
upon NKT cells, rather than producing a generalized stimulatory ef-
fect on all T lymphocytes within the spleen, liver, and circulation.
Increases in T cell populations predominantly correlate with
increases in Th1 cytokines
Because activated NKT cells produce a variety of cytokines, we
sought to test the systemic effect and contribution of apoE on
cytokine secretion. Release of IL-10, IL-2, IL-4, IL-1?, TNF-?,
and IFN-? cytokine levels was measured in sera from septic rats
24 h after they received apoE. Multiplex cytokine analysis dem-
onstrated that septic rats had higher levels of the Th1 cytokines
IL-1? and TNF-? (Fig. 5, A and B), as well as the Th2 cytokines
IL-10 and IL-2 (Fig. 5, D and E), than did sham/saline rats.
Septic rats treated with apoE had higher levels of IL-1? and
IFN-? (Fig. 5, A and C), and IL-10 and IL-2 (Fig. 5, D and E),
present in the sera than did septic rats that did not receive
An unexpected finding was that apoE treatment in sham rats
resulted in a higher level of IFN-?, a prototypical NKT cell
cytokine, than was found in sham/saline conditions (Fig. 5C).
Table I. apoE stimulates changes in absolute lymphocyte, NKT, and BrdU?NKT cell numbers in sepsisa
2.36 ? 0.3 ? 108
5.16 ? 1.0 ? 106
1.59 ? 0.4 ? 105
1.59 ? 0.1 ? 107
4.13 ? 1.5 ? 105
2.97 ? 1.0 ? 104
3.9 ? 0.1 ? 107
5.18 ? 1.0 ? 105
2.26 ? 0.8 ? 104
2.63 ? 0.4 ? 108
5.13 ? 0.2 ? 106
1.63 ? 0.2 ? 105
1.68 ? 0.2 ? 107
4.45 ? 1.8 ? 105
3.02 ? 1.1 ? 104
3.65 ? 0.2 ? 107
4.78 ? 0.9 ? 105
2.21 ? 0.6 ? 104
1.94 ? 0.1 ? 108†
5.24 ? 0.7 ? 106
1.41 ? 0.2 ? 105
9.84 ? 0.1 ? 106†††
1.56 ? 0.5 ? 105††
1.78 ? 0.5 ? 104†
1.94 ? 0.1 ? 107†††
1.46 ? 0.3 ? 105†††
5.02 ? 2.5 ? 103†††
2.08 ? 0.2 ? 108
7.07 ? 1.6 ? 106* 2.85 ? 0.5 ? 105* 1.33 ? 0.1 ? 107**
5.86 ? 1.1 ? 105** 9.55 ? 1.9 ? 104**
7.4 ? 0.3 ? 106**
3.14 ? 0.4 ? 104**
1.83 ? 0.6 ? 103*
aComparison of absolute lymphocyte, NKT, and BrdU?NKT cell numbers in the spleen, liver, and circulation after CLP and apoE treatment. Numbers are averages from five separate experiments. †, p ? 0.05; ††, p ? 0.01; †††, p ? 0.001
for sham/saline v. clp/saline comparison; ?, p ? 0.05; ??, p ? 0.001 for CLP/saline vs CLP/saline vs CLP/apoE comparison.
1403The Journal of Immunology
Moreover, apoE treatment in septic rats decreased the elevated
IL-4 levels of septic rats that received saline by 4-fold, to those
of the sham/saline rats (Fig. 5F). These findings, therefore, sug-
gest that whereas apoE increases Th1 and Th2 cytokines in the
sera of septic rats, the predominant increase is in Th1 cytokines.
apoE increases immune cell activity by decreasing hepatic T
cell apoptosis and necrosis in septic rats
Because apoptosis is known to result in immune dysfunction
through the induction of Th2 responses in sepsis (20–22), we
investigated whether our cytokine findings correlated with de-
creased apoptosis. To test apoE’s effect on apoptosis-induced
mortality in sepsis, T cell apoptosis and necrosis were measured
in the liver. Flow cytometry staining for Annexin V and 7-AAD
demonstrated that 24 h after CLP, septic rats had 167% more
hepatic T cells undergoing apoptosis than did sham controls. apoE
in septic rats induced a 58% reduction in hepatic T cell apoptosis
when compared with septic rats that did not receive apoE (Fig.
6A). Septic rats also had 37% more hepatic T cell necrosis 24 h
after CLP than did sham controls. Similarly to its effect on apo-
ptosis, apoE reduced T cell necrosis levels to those of sham control
rats that did not receive apoE, although only by 29% (Fig. 6B).
These results indicate that apoE prevented liver T cell apoptosis
and necrosis, both of which are characteristic of sepsis, and
thereby contributed to increased survival and activity of liver im-
of rats from each treatment group are shown. B–D, Splenic lymphocytes (B) and hepatic lymphocytes (C) from rats subjected to CLP had more CD8?T
cell populations than did sham/saline controls, whereas CD8?T cells in the circulation were reduced (D). The frequency of CD8?T cells in the spleens,
livers, and circulation of septic rats treated with apoE (114 ?g/kg) for 24 h were increased. Data represent three independent experiments and are reported
as a fraction of sham/saline. Error bars indicate SDs.
apoE-mediated stimulation is specific to NKT cell levels. A, Percentages of CD8?T cell types from the spleens, livers, and whole blood
1404apoE INCREASES SEPTIC MORTALITY
apoE accentuates liver injury in septic rats
Hepatic dysfunction frequently accompanies a variety of bacterial
infections (23). To examine the effect of apoE-induced NKT cell
proliferation on hepatotoxicity, ALT serum levels were measured
24 h after CLP and apoE treatment. We found that serum ALT
levels were increased in septic rats that had undergone CLP and
were further increased in septic rats that received apoE. These
findings indicate that bacterial sepsis induced by CLP resulted in
liver injury, which was further exacerbated by apoE treatment
concentrations 24 h after CLP were determined by multiplex ELISA in sera of septic rats treated with apoE. A, IL-1? concentrations in rats treated with
apoE increased by 2-fold (p ? 0.05) over those in sham/saline rats. After treatment with apoE (114 ?g/kg) at the time of CLP and every 4 h for 24 h, IL-1?
concentrations increased by 4-fold (p ? 0.05). B, TNF-? serum concentrations increased by at least 2-fold (p ? 0.05) over those in sham controls, but did
not change after apoE treatment. C, IFN-? concentrations decreased by 3-fold (p ? 0.05) over those in sham/apoE rats and were 2-fold higher in septic
rats treated with apoE than in CLP/saline rats (p ? 0.05). D, IL-10 serum concentrations in septic rats were 10-fold higher (p ? 0.01) than those in
sham/saline rats. After apoE treatment, IL-10 concentrations increased by 2-fold (p ? 0.01) over those in rats subjected to CLP only. E, IL-2 concentrations
in septic rats treated with apoE were 5-fold higher than those in CLP/saline rats (p ? 0.01). F, apoE treatment reversed IL-4 concentrations in rats treated
with CLP back to those of sham/saline controls (p ? 0.01). Data represent means ? SDs of duplicate samples in each of at least four independent
experiments. ?, p ? 0.05; ??, p ? 0.01.
Predominant Th1 cytokine production in serum with apoE-mediated CD1d-restricted NKT cell proliferation during septic insult. Cytokine
apoE treatment in rats subjected to CLP. A and B, T cell apoptosis (A) and
necrosis (B) levels were measured via Annexin V and 7-AAD cytometry
analysis 24 h after CLP and apoE treatment. Levels in the liver were higher
in CLP/saline rats than in rats that underwent sham surgery followed by
either saline or apoE treatment. Treatment with apoE (114 ?g/kg) in rats
subjected to CLP serially every 4 h for 24 h reduced levels in the liver to
those of sham controls. Data represent three independent experiments and
are reported as a fraction of sham/saline. Error bars indicate SDs.
Reduction of apoptotic and necrotic T cells in the liver by
concentrations of ALT were measured 24 h after rats were subjected to
CLP and apoE treatment. Twenty-four hours after CLP, serum ALT levels
were higher than in sham/saline rats (69.0 ? 14.7 vs 37.25 ? 6.9; p ?
0.01). Treatment with apoE (114 mg/kg) every 4 h for 24 h after CLP
further increased ALT serum levels (98.5 ? 19.4 vs 69.0 ? 14.7; p ?
0.001) for rats subjected to CLP and saline treatment. ???, p ? 0.001. Error
bars indicate SDs.
apoE increases liver injury in rats subjected to CLP. Serum
1405The Journal of Immunology
Among previous characterizations of its immunomodulatory ef-
fects, apoE has recently been implicated in lipid Ag presentation to
NKT cells, via CD1d (24). However, apoE’s specific immuno-
modulatory role in sepsis is currently unknown. This study pro-
vides evidence that apoE, a LDLR ligand, has a role in increasing
the immune response to polymicrobial sepsis, particularly through
increased NKT cell number, proliferation, trafficking, and down-
stream responses, and thereby contributes to increased mortality.
In our in vivo model of sepsis induced by CLP, apoE treatment
increased sepsis-induced mortality, and increased NKT cell fre-
quency and proliferation in the liver and spleen, with concomitant
decreases in these NKT cell parameters in the peripheral circula-
tion. apoE treatment also promoted a Th1 cytokine response, in-
creased the degree of liver injury, and decreased apoptosis in he-
Although apoE had no toxic effects on mortality in sham rats,
the effect of apoE concentration on mortality during sepsis was
dose-dependent; both concentrations tested resulted in signifi-
cantly higher mortality in septic rats than did saline. The effect of
sepsis on NKT cell frequency and proliferation differed in the
spleen, liver, and circulation. The frequency of NKT cells in-
creased in the spleen but significantly decreased in the liver and
circulation. One possible explanation for this difference is the find-
ing that hepatic apoE production decreases significantly after sep-
sis (25). Additionally, it has been shown that the percentages of
circulating NKT cells are significantly lower in individuals with
liver and infectious diseases (26–28). When compared with sham
controls, septic rats had fewer NKT cells in the liver, but a greater
proportion of these NKT cells were proliferating. apoE given dur-
ing sepsis resulted in a significant increase in NKT cell frequency
in the spleen and liver, and a decrease in the circulation, which
probably contributed to the mortality we observed. Not only were
the frequencies of splenic and hepatic NKT cells increased, a
greater proportion of NKT cells were proliferating. This increase is
consistent with studies demonstrating that mice deficient in apoE
exhibited lower splenic and hepatic NKT cell activation than wild-
type mice (29).
The increased percentage and proliferation of NKT cells in the
spleen and liver, and the concomitant decrease in the frequency of
NKT cells in the circulation after the addition of apoE, indicate
that both trafficking from the circulation to the spleen and liver and
local expansion of NKT cells may be occurring. The absolute num-
ber of lymphocytes, NKT, and BrdU?NKT cells confirm this. The
addition of apoE in sepsis caused the total lymphocyte, NKT, and
proliferating NKT cell populations to increase the most in the liver,
while causing them to decrease in the circulation, suggesting a role
for both pathways in the liver. However, there was no significant
increase in the number of lymphocytes in the spleen, suggesting
that the increase in the absolute number of splenic NKT cells may
be due to local expansion, consistent with the increased number of
proliferating NKT cells observed in the spleen. Furthermore, the
circulation’s small lymphocyte population may not significantly
contribute in absolute cell number to the spleen as much as it can
in the liver, which may also explain the greater hepatic response to
apoE in sepsis. Evidence for preferential homing of NKT cells to
the liver also supports this conclusion (30); however, apoptosis or
down-regulation of T cell receptors cannot be ruled out as a cause
for the depletion of circulating NKT cells (27). Studies on the
mechanics of NKT cell trafficking, including the expression of
chemokine and adhesion molecules, and contributions from the
thymus, lymph nodes, and other organs may further characterize
the dynamic redistribution of NKT cells. The lack of transgenic
models and a functional CD1d tetramer for detecting NKT cells in
rats (31) limited this study. Future studies using murine transgenic
models and appropriate blocking experiments could further char-
acterize apoE’s causative role in activation and proliferation of
NKT cells in sepsis.
To test whether apoE affects activity of other T lymphocytes, we
measured CD8?T cell populations. Despite a trend toward in-
creased proliferation of CD8?T cells in septic rats treated with
apoE as compared with septic rats that did not receive apoE, the
difference was not statistically significant, indicating that apoE
may selectively activate NKT cells rather than producing a gen-
eralized stimulatory effect on all T lymphocytes. As CD8?T cells
are known to be influenced by NKT cell activation (32, 33), this
trend could be attributed to the downstream effects of NKT cells.
NKT cell activation reportedly plays a critical role in the intrahe-
patic immunity to several infections and liver injury (9, 34). We
found ALT levels to be significantly increased after apoE treat-
ment, which suggests that NKT cell activation contributes to sep-
sis-induced mortality. NK cells, which are key to the innate im-
mune response, have been demonstrated to increase in sepsis (35).
Although NK cells were significantly increased in the spleen and
liver of rats 24 h after CLP, apoE treatment did not result in sig-
nificant changes (data not shown), which suggests that the ob-
served apoE-mediated Th1 response was due primarily to prolif-
eration of NKT cells.
Endotoxic shock is frequently caused by a huge systemic cyto-
kine response to Gram-negative bacteria and their characteristic
cell-wall component, LPS. Clinically, bacterial sepsis is character-
ized by an imbalance between the pro- and anti-inflammatory re-
sponses of the immune system. Previous research suggests that
both proinflammatory cytokines such as TNF-?, IL-1?, and IFN-?,
as well as anti-inflammatory cytokines such as IL-4, and IL-10,
have relevant roles in sepsis (19). We found that sera from septic
rats that received apoE had higher levels of IL-10, IL-1?, TNF-?,
and IFN-? than septic rats that received saline. Interestingly, IL-4,
the main Th2 cytokine produced by NKT cells, was not elevated
by apoE treatment after septic challenge. Instead, it was signifi-
cantly reduced to that of control levels. Although both Th1 and
Th2 cytokines were elevated, the observation that IL-4 was not
suggests that apoE activation of NKT cells may predominantly
elicit a Th1 cytokine response. This possibility is supported by
recent data that suggest NKT cells exclusively produce IFN-? after
stimulation with LPS (36). A role for Th2 cytokines, such as IL-4,
in counteracting LPS-induced shock, has been recently reported
(37) and provides support for our findings. In addition, injection of
a synthetic NKT cell ligand has been shown to protect against a
systemic Shwartzmann reaction by increasing Th2 cytokines, de-
pending on the timing of injection (37). Clearly, timing of NKT
cell activation is a very important component of the host response
Clinical studies have shown sepsis to be a combination of a
hyperinflammatory and immunosuppressive states (38). Lympho-
cytes (B and T cells) are central to the adaptive immune system.
The profound decrease in their numbers documented in sepsis re-
sults in induction of Th2 responses by surviving immune cells
(20–22). As dysregulated apoptotic immune cell death is proposed
to contribute to this loss in lymphocytes, and, thus, increase in the
immunosuppressive phase in sepsis (39), we sought to examine
whether apoE’s effect on inducing a Th1 response correlated with
changes in apoptosis and necrosis. We found that whereas hepatic
T cell apoptosis and necrosis were increased in septic rats 24 h
after CLP, Annexin V and 7-AAD staining showed that apoE treat-
ment equalized levels of hepatic T cell apoptosis and necrosis to
those of sham animals. Although lymphocyte apoptosis is known
1406 apoE INCREASES SEPTIC MORTALITY
to contribute to septic mortality (39), as we have shown, apoE’s
activation of the splenic and hepatic lymphocytes mediated by
NKT cells seems to have excited these cell populations into a
hyperactivated state. The profound decrease in the number of T
and B cells in sepsis impairs both the adaptive and innate immune
response because of the important cross-talk between the systems
(22, 40). Similarly, apoE’s effect on preventing lymphocyte death
increases the immune response, enhancing the proinflammatory
response to polymicrobial infections, and, thus, increased mortal-
ity. Our findings support reports which demonstrate that inhibition
of apoptosis by lipoproteins depends on APOE genotype (41), and
that apoptotic cells and fragments accumulate markedly in a range
of tissues in apoE-deficient mice (42). The predominant Th1 re-
sponse observed in serum from septic rats treated with apoE cor-
relates with our observations of decreased apoptosis and Th2
Among the evidence linking apoE and host immune responses
to infection is the observation that apoE can bind LPS, attenuate
the host inflammatory response, and protect against LPS-induced
mortality (8). In contrast with these earlier findings, our results
argue against a protective role for apoE. However, whereas we
used serial injections of apoE in our CLP model of sepsis, the
earlier study used injection of LPS as a model for sepsis and dem-
onstrated protectability by preincubating LPS with apoE before
injection, a possible reason for the different findings. Contradictory
reports also exist regarding the role of NKT cells in bacterial sep-
sis. Some investigators have shown that ?-GalCer treatment,
which activates NKT cells, can induce septic shock, and propose it
as a model for bacterial sepsis (43), while others have shown that
activation of NKT cells can help reduce septic mortality (37). Con-
firming our finding that NKT cell activation contributes to CLP-
induced mortality is the evidence that anti-CD1d reduces CLP-
induced mortality in mice (19). Moreover, NKT cells have been
shown to amplify the innate immune response to LPS (36). Al-
though NKT cells are known to recognize glycolipid Ags, the iden-
tity of foreign and endogenous Ags remains a major question. The
prototypic NKT cell Ag ?-GalCer is derived from a marine sponge
and is not recognized as a product of mammalian cells or patho-
gens. However, NKT cells have recently been found to react with
a pathogen-derived lipid Ag from Borrelia burgdorferi, which
causes Lyme disease (44, 45). We found that when apoE was given
to sham rats, the level of IFN-?, a prototypical cytokine produced
by NKT cells, was higher than in sham rats treated with saline,
perhaps suggesting that apoE was presenting an endogenous Ag to
NKT cells. Other “indirect” mechanisms that have been suggested
to amplify NKT cell activation are the dendritic cell-derived IL-12,
a response to TLR activation by LPS (12, 46), and other APC-
derived cytokines (36).
In conclusion, our study provides evidence that apoE hyperac-
tivates the immune system, leading to increased morbidity and
mortality in a rat model of sepsis. apoE, one possible vehicle for
increased bacterial Ag presentation, increases CD1d-restricted
NKT cell activity, Th1 cytokine release, and liver injury, and de-
creases lymphocyte apoptosis in polymicrobial infection. These
findings emphasize that lipids do indeed play a significant role in
the immune system and suggest that apoE may regulate the pro-
cessing of foreign lipid Ags during bacterial sepsis. The APOE
gene codes for three main isoforms of the protein: apoE2, 3, and 4
(4), which range in their binding affinities to the LDL receptor
(E4 ? E3 ? ? E2) (47). Two intriguing questions are whether en-
dogenous apoE could regulate host defense during sepsis, and
whether deletion interferes with innate immune responses. Protec-
tion from bacterial sepsis may occur by modulating apoE’s effect
with anti-apoE Abs, or by taking advantage of other apoE isoforms
that may occupy LDL-receptor binding sites, to either block or
delay apoE-mediated internalization of foreign Ags. Our findings
raise new prospects for the role of apoE in regulating infection and
We thank Pamela Derish for excellent editorial assistance and Dr. Karl
Weisgraber for reagents and helpful discussions.
The authors have no financial conflict of interest.
1. Gallin, J. I., D. Kaye, and W. M. O’Leary. 1969. Serum lipids in infection.
N. Engl. J. Med. 281: 1081–1086.
2. Wolfe, R. R., J. H. Shaw, and M. J. Durkot. 1985. Effect of sepsis on VLDL
kinetics: responses in basal state and during glucose infusion. Am. J. Physiol. 248:
3. Harris, H. W., J. E. Gosnell, and Z. L. Kumwenda. 2000. The lipemia of sepsis:
triglyceride-rich lipoproteins as agents of innate immunity. J. Endotoxin Res. 6:
4. Mahley, R. W., and S. C. Rall, Jr. 2000. Apolipoprotein E: far more than a lipid
transport protein. Annu Rev. Genomics Hum. Genet. 1: 507–537.
5. Mahley, R. W. 1988. Apolipoprotein E: cholesterol transport protein with ex-
panding role in cell biology. Science 240: 622–630.
6. Roselaar, S. E., and A. Daugherty. 1998. Apolipoprotein E-deficient mice have
impaired innate immune responses to Listeria monocytogenes in vivo.
J. Lipid Res. 39: 1740–1743.
7. de Bont, N., M. G. Netea, P. N. Demacker, B. J. Kullberg, J. W. van der Meer,
and A. F. Stalenhoef. 2000. Apolipoprotein E-deficient mice have an impaired
immune response to Klebsiella pneumoniae. Eur. J. Clin. Invest. 30: 818–822.
8. Van Oosten, M., P. C. Rensen, E. S. Van Amersfoort, M. Van Eck,
A. M. Van Dam, J. J. Breve, T. Vogel, A. Panet, T. J. Van Berkel, and J. Kuiper.
2001. Apolipoprotein E protects against bacterial lipopolysaccharide-induced le-
thality: a new therapeutic approach to treat gram-negative sepsis. J. Biol. Chem.
9. Chida, Y., N. Sudo, J. Sonoda, H. Sogawa, and C. Kubo. 2004. Electric foot
shock stress-induced exacerbation of ?-galactosylceramide-triggered apoptosis in
mouse liver. Hepatology 39: 1131–1140.
10. Matsuda, J. L., O. V. Naidenko, L. Gapin, T. Nakayama, M. Taniguchi,
C. R. Wang, Y. Koezuka, and M. Kronenberg. 2000. Tracking the response of
natural killer T cells to a glycolipid antigen using CD1d tetramers. J. Exp. Med.
11. Yoshimoto, T., and W. E. Paul. 1994. CD4pos, NK1.1pos T cells promptly pro-
duce interleukin 4 in response to in vivo challenge with anti-CD3. J. Exp. Med.
12. Taniguchi, M., K. Seino, and T. Nakayama. 2003. The NKT cell system: bridging
innate and acquired immunity. Nat. Immunol. 4: 1164–1165.
13. Morrow, J. A., K. S. Arnold, and K. H. Weisgraber. 1999. Functional character-
ization of apolipoprotein E isoforms overexpressed in Escherichia coli. Protein
Expr. Purif. 16: 224–230.
14. Thrivikraman, K. V., R. L. Huot, and P. M. Plotsky. 2002. Jugular vein cathe-
terization for repeated blood sampling in the unrestrained conscious rat. Brain
Res. Brain Res. Protoc. 10: 84–94.
15. Remick, D. G., D. E. Newcomb, G. L. Bolgos, and D. R. Call. 2000. Comparison
of the mortality and inflammatory response of two models of sepsis: lipopoly-
saccharide vs. cecal ligation and puncture. Shock 13: 110–116.
16. Read, T. E., C. Grunfeld, Z. L. Kumwenda, M. C. Calhoun, J. P. Kane,
K. R. Feingold, and J. H. Rapp. 1995. Triglyceride-rich lipoproteins prevent
septic death in rats. J. Exp. Med. 182: 267–272.
17. Dao, T., W. Z. Mehal, and I. N. Crispe. 1998. IL-18 augments perforin-dependent
cytotoxicity of liver NK-T cells. J. Immunol. 161: 2217–2222.
18. Holub, M., Z. Kluckova, B. Beneda, J. Hobstova, I. Huzicka, J. Prazak, and
A. Lobovska. 2000. Changes in lymphocyte subpopulations and CD3?/DR?ex-
pression in sepsis. Clin. Microbiol. Infect. 6: 657–660.
19. Rhee, R. J., S. Carlton, J. L. Lomas, C. Lane, L. Brossay, W. G. Cioffi, and
A. Ayala. 2003. Inhibition of CD1d activation suppresses septic mortality: a role
for NK-T cells in septic immune dysfunction. J. Surg. Res. 115: 74–81.
20. Griffith, T. S., X. Yu, J. M. Herndon, D. R. Green, and T. A. Ferguson. 1996.
CD95-induced apoptosis of lymphocytes in an immune privileged site induces
immunological tolerance. Immunity 5: 7–16.
21. Voll, R. E., M. Herrmann, E. A. Roth, C. Stach, J. R. Kalden, and I. Girkontaite.
1997. Immunosuppressive effects of apoptotic cells. Nature 390: 350–351.
22. Hotchkiss, R. S., and D. W. Nicholson. 2006. Apoptosis and caspases regulate
death and inflammation in sepsis. Nat. Rev. Immunol. 6: 813–822.
23. Chand, N., and A. J. Sanyal. 2007. Sepsis-induced cholestasis. Hepatology 45:
24. van den Elzen, P., S. Garg, L. Leon, M. Brigl, E. A. Leadbetter, J. E. Gumperz,
C. C. Dascher, T. Y. Cheng, F. M. Sacks, P. A. Illarionov, et al. 2005. Apoli-
poprotein-mediated pathways of lipid antigen presentation. Nature 437:
25. Tripp, R. J., A. Tabares, H. Wang, and S. Lanza-Jacoby. 1993. Altered hepatic
production of apolipoproteins B and E in the fasted septic rat: factors in the
development of hypertriglyceridemia. J. Surg. Res. 55: 465–472.
1407The Journal of Immunology
26. van der Vliet, H. J., B. M. von Blomberg, M. D. Hazenberg, N. Nishi, S. A. Otto, Download full-text
B. H. van Benthem, M. Prins, F. A. Claessen, A. J. van den Eertwegh,
G. Giaccone, et al. 2002. Selective decrease in circulating V?24?V?11?NKT
cells during HIV type 1 infection. J. Immunol. 168: 1490–1495.
27. Snyder-Cappione, J. E., D. F. Nixon, C. P. Loo, J. M. Chapman,
D. A. Meiklejohn, F. F. Melo, P. R. Costa, J. K. Sandberg, D. S. Rodrigues, and
E. G. Kallas. 2007. Individuals with pulmonary tuberculosis have lower levels of
circulating CD1d-restricted NKT cells. J. Infect. Dis. 195: 1361–1364.
28. Lucas, M., S. Gadola, U. Meier, N. T. Young, G. Harcourt, A. Karadimitris,
N. Coumi, D. Brown, G. Dusheiko, V. Cerundolo, and P. Klenerman. 2003.
Frequency and phenotype of circulating V?24/V?11 double-positive natural
killer T cells during hepatitis C virus infection. J. Virol. 77: 2251–2257.
29. Major, A. S., M. T. Wilson, J. L. McCaleb, Y. Ru Su, A. K. Stanic, S. Joyce,
L. Van Kaer, S. Fazio, and M. F. Linton. 2004. Quantitative and qualitative
differences in proatherogenic NKT cells in apolipoprotein E-deficient mice. Ar-
terioscler. Thromb. Vasc. Biol. 24: 2351–2357.
30. Miyamoto, M., M. Emoto, V. Brinkmann, N. van Rooijen, R. Schmits, E. Kita,
and S. H. Kaufmann. 2000. Cutting edge: contribution of NK cells to the homing
of thymic CD4?NKT cells to the liver. J. Immunol. 165: 1729–1732.
31. Pyz, E., O. Naidenko, S. Miyake, T. Yamamura, I. Berberich, S. Cardell,
M. Kronenberg, and T. Herrmann. 2006. The complementarity determining re-
gion 2 of BV8S2 (V? 8.2) contributes to antigen recognition by rat invariant
NKT cell TCR. J. Immunol. 176: 7447–7455.
32. Schaible, U. E., and S. H. Kaufmann. 2000. CD1 molecules and CD1-dependent
T cells in bacterial infections: a link from innate to acquired immunity? Semin.
Immunol. 12: 527–535.
33. Ueda, N., H. Kuki, D. Kamimura, S. Sawa, K. Seino, T. Tashiro, K. Fushuku,
M. Taniguchi, T. Hirano, and M. Murakami. 2006. CD1d-restricted NKT cell
activation enhanced homeostatic proliferation of CD8?T cells in a manner de-
pendent on IL-4. Int. Immunol. 18: 1397–1404.
34. Ishigami, M., H. Nishimura, Y. Naiki, K. Yoshioka, T. Kawano, Y. Tanaka,
M. Taniguchi, S. Kakumu, and Y. Yoshikai. 1999. The roles of intrahepatic
V?14?NK1.1?T cells for liver injury induced by Salmonella infection in mice.
Hepatology 29: 1799–1808.
35. Giamarellos-Bourboulis, E. J., T. Tsaganos, E. Spyridaki, M. Mouktaroudi,
D. Plachouras, I. Vaki, V. Karagianni, A. Antonopoulou, V. Veloni, and
H. Giamarellou. 2006. Early changes of CD4-positive lymphocytes and NK cells
in patients with severe Gram-negative sepsis. Crit. Care 10: R166.
36. Nagarajan, N. A., and M. Kronenberg. 2007. Invariant NKT cells amplify the
innate immune response to lipopolysaccharide. J. Immunol. 178: 2706–2713.
37. Sireci, G., M. P. La Manna, C. Di Sano, D. Di Liberto, S. A. Porcelli,
M. Kronenberg, F. Dieli, and A. Salerno. 2007. Pivotal advance: ?-galactosyl-
ceramide induces protection against lipopolysaccharide-induced shock. J. Leu-
kocyte Biol. 81: 607–622.
38. Oberholzer, A., C. Oberholzer, and L. L. Moldawer. 2001. Sepsis syndromes:
understanding the role of innate and acquired immunity. Shock 16: 83–96.
39. Wesche, D. E., J. L. Lomas-Neira, M. Perl, C. S. Chung, and A. Ayala. 2005.
Leukocyte apoptosis and its significance in sepsis and shock. J. Leukocyte Biol.
40. Fearon, D. T., and R. M. Locksley. 1996. The instructive role of innate immunity
in the acquired immune response. Science 272: 50–53.
41. DeKroon, R. M., M. Mihovilovic, Z. V. Goodger, J. B. Robinette, P. M. Sullivan,
A. M. Saunders, and W. J. Strittmatter. 2003. ApoE genotype-specific inhibition
of apoptosis. J. Lipid Res. 44: 1566–1573.
42. Grainger, D. J., J. Reckless, and E. McKilligin. 2004. Apolipoprotein E modu-
lates clearance of apoptotic bodies in vitro and in vivo, resulting in a systemic
proinflammatory state in apolipoprotein E-deficient mice. J. Immunol. 173:
43. Ito, H., N. Koide, F. Hassan, S. Islam, G. Tumurkhuu, I. Mori, T. Yoshida,
S. Kakumu, H. Moriwaki, and T. Yokochi. 2006. Lethal endotoxic shock using
?-galactosylceramide sensitization as a new experimental model of septic shock.
Lab. Invest. 86: 254–261.
44. Kinjo, Y., E. Tupin, D. Wu, M. Fujio, R. Garcia-Navarro, M. R. Benhnia,
D. M. Zajonc, G. Ben-Menachem, G. D. Ainge, G. F. Painter, et al. 2006. Natural
killer T cells recognize diacylglycerol antigens from pathogenic bacteria. Nat.
Immunol. 7: 978–986.
45. Godfrey, D. I., and S. P. Berzins. 2006. NKT cells join the war on Lyme disease.
Nat. Immunol. 7: 904–906.
46. Brigl, M., L. Bry, S. C. Kent, J. E. Gumperz, and M. B. Brenner. 2003. Mech-
anism of CD1d-restricted natural killer T cell activation during microbial infec-
tion. Nat. Immunol. 4: 1230–1237.
47. Harris, H. W. 2005. Apolipoprotein E: from Alzheimer’s to sepsis. Crit. Care
Med. 33: 2696–2697.
1408apoE INCREASES SEPTIC MORTALITY