as a PDF

Transcrição

as a PDF

Deficiency of Glutathione Peroxidase-1 Accelerates the Progression of Atherosclerosis in
Apolipoprotein E-Deficient Mice
Michael Torzewski, Viola Ochsenhirt, Andrei L. Kleschyov, Matthias Oelze, Andreas Daiber,
Huige Li, Heidi Rossmann, Sotirios Tsimikas, Kurt Reifenberg, Fei Cheng, Hans-Anton Lehr,
Stefan Blankenberg, Ulrich Förstermann, Thomas Münzel and Karl J. Lackner
Arterioscler Thromb Vasc Biol. 2007;27:850-857; originally published online January 25, 2007;
doi: 10.1161/01.ATV.0000258809.47285.07
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272
Greenville Avenue, Dallas, TX 75231
Copyright © 2007 American Heart Association, Inc. All rights reserved.
Print ISSN: 1079-5642. Online ISSN: 1524-4636
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://atvb.ahajournals.org/content/27/4/850
Data Supplement (unedited) at:
http://atvb.ahajournals.org/content/suppl/2007/01/25/01.ATV.0000258809.47285.07.DC1.html
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the
Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for
which permission is being requested is located, click Request Permissions in the middle column of the Web
page under Services. Further information about this process is available in the Permissions and Rights
Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
at:
http://atvb.ahajournals.org//subscriptions/
Downloaded from http://atvb.ahajournals.org/ by guest on February 23, 2013
Deficiency of Glutathione Peroxidase-1 Accelerates the
Progression of Atherosclerosis in Apolipoprotein
E-Deficient Mice
Michael Torzewski, Viola Ochsenhirt, Andrei L. Kleschyov, Matthias Oelze, Andreas Daiber, Huige Li,
Heidi Rossmann, Sotirios Tsimikas, Kurt Reifenberg, Fei Cheng, Hans-Anton Lehr,
Stefan Blankenberg, Ulrich Förstermann, Thomas Münzel, Karl J. Lackner
Background—We have recently demonstrated that activity of red blood cell glutathione peroxidase-1 is inversely
associated with the risk of cardiovascular events in patients with coronary artery disease. The present study analyzed
the effect of glutathione peroxidase-1 deficiency on atherogenesis in the apolipoprotein E-deficient mouse.
Methods and Results—Female apolipoprotein E-deficient mice with and without glutathione peroxidase-1 deficiency were
placed on a Western-type diet for another 6, 12, or 24 weeks. After 24 weeks on Western-type diet, double-knockout
mice (GPx-1⫺/⫺ApoE⫺/⫺) developed significantly more atherosclerosis than control apolipoprotein E-deficient mice.
Moreover, glutathione peroxidase-1 deficiency led to modified atherosclerotic lesions with increased cellularity.
Functional experiments revealed that glutathione peroxidase-1 deficiency leads to increased reactive oxygen species
concentration in the aortic wall as well as increased overall oxidative stress. Peritoneal macrophages from
double-knockout mice showed increased in vitro proliferation in response to macrophage– colony-stimulating factor.
Also, we found lower levels of bioactive nitric oxide as well as increased tyrosine nitration as a marker of peroxynitrite
production.
Conclusions—Deficiency of an antioxidative enzyme accelerates and modifies atherosclerotic lesion progression in
apolipoprotein E-deficient mice. (Arterioscler Thromb Vasc Biol. 2007;27:850-857.)
Key Words: antioxidants 䡲 atherosclerosis 䡲 nitric oxide
O
xidative stress is defined as an imbalance between the
production and degradation of reactive oxygen species
(ROS). Enzymatic inactivation of ROS is achieved mainly by
superoxide dismutases, catalase and the glutathione peroxidases.1 Indeed, glutathione and the glutathione peroxidases
constitute the principal antioxidant defense system in mammalian cells.2– 4 Glutathione peroxidase-1 (GPx-1), the ubiquitous intracellular form and key antioxidant enzyme within
many cells, including the endothelium, consumes reduced
glutathione to convert hydrogen peroxide to water and lipid
peroxides to their respective alcohols.5 It also acts as a
peroxynitrite reductase.6 Because of its major role in the
prevention of oxidative stress, GPx-1 may be an important
antiatherogenic enzyme.7 In fact, we have recently shown in
patients with coronary artery disease that a low activity of red
blood cell GPx-1 is associated with an increased risk of
cardiovascular events independently from traditional risk
factors or atherosclerosis.8 A mouse model of GPx-1 defi-
ciency is available. These animals appear healthy and are
fertile.9 However, a recent in vitro study showed increased
cell-mediated oxidation of low-density lipoprotein (LDL) in
this model.10 Furthermore, GPx-1 deficiency causes endothelial dysfunction in mice11 that is aggravated by hyperhomocysteinemia.12 GPx-1 deficiency is accompanied by increased
periadventitial inflammation, neointima formation, and collagen deposition surrounding the coronary arteries.13 GPx-1
activity is decreased or absent in carotid atherosclerotic
plaques, and the absence of GPx-1 activity in atherosclerotic
lesions has been linked to the development of more severe
lesions in humans.14 Recently, t’Hoen et al15 reported that the
expression of several antioxidant enzymes including GPx-1
was increased in the aortic arch of apolipoprotein E-deficient
(ApoE⫺/⫺) mice in the period preceding lesion formation, and
that their expression decreased at the time when lesion
formation became apparent, whereas Biswas et al suggested
that decreased glutathione synthesis as well as reduced
Original received April 26, 2006; final version accepted December 21, 2006.
From the Institute of Clinical Chemistry and Laboratory Medicine (M.T., V.O., F.C., H.R., K.J.L.), Department of Medicine II (A.L.K., M.O., A.D.,
S.B., T.M.), Department of Pharmacology (H.L., U.F.), Central Laboratory Animal Facility (K.R.), Johannes Gutenberg University, Mainz, Germany;
Division of Cardiology (S.T), University of California, San Diego, Calif; Institute Universitaire de Pathologie (H-A.L.), Centre Hospitalier Universitaire
Vaudois (CHUV), Lausanne, Switzerland.
M.T. and V.O. contributed equally to this work.
Correspondence to Michael Torzewski, MD, or to Karl J. Lackner, MD, Institute of Clinical Chemistry and Laboratory Medicine, Johannes Gutenberg
University, 55101 Mainz, Germany. E-mail [email protected] or [email protected]
© 2007 American Heart Association, Inc.
Arterioscler Thromb Vasc Biol. is available at http://www.atvbaha.org
DOI: 10.1161/01.ATV.0000258809.47285.07
850
Downloaded from http://atvb.ahajournals.org/
by guest on February 23, 2013
Torzewski et al
transcription and activity of GPx-1 precede lipid peroxidation
and detectable atheroma in these mice.16 Despite this plethora
of both experimental and clinical data linking GPx-1 deficiency to cardiovascular pathology, no direct evidence has yet
established a role of GPx-1 during atherogenesis. Recently,
de Haan et al17 reported that in C57BL/J6 mice GPx-1
deficiency did not accelerate atherosclerotic lesion development after 12 and 20 weeks on a high-fat diet. Since we
obtained the same result in GPx-1– deficient (GPx-1⫺/⫺)
C57BL/J6 mice even after 1 year on a high-fat diet supplemented with cholate (data not shown), we extended our
analysis to the commonly used ApoE⫺/⫺ mouse model by
crossbreeding GPx-1⫺/⫺ mice with this strain.
Methods
Mice
GPx-1⫺/⫺ApoE⫺/⫺ mice (generously provided by Ye-Shi Ho, Department of Biochemistry, Wayne State University, Detroit, Mich) were
bred by generating F2 hybrids from the ApoE⫺/⫺ and GPx-1⫺/⫺
parental strains. The GPx-1⫺/⫺ApoE⫺/⫺ strain could then be propagated successfully by incrossing. Genotype determination was performed as described (please see http://atvb.ahajournals.org).13
Induction of Atherosclerosis
At 7 to 8 weeks of age, female ApoE⫺/⫺ as well as GPx-1⫺/⫺ApoE⫺/⫺
mice were placed on a Western-type diet (WTD) for another 6 weeks
(5 GPx-1⫺/⫺ApoE⫺/⫺ and 5 control ApoE⫺/⫺ mice), 12 weeks (16
GPx-1⫺/⫺ApoE⫺/⫺ and 16 control ApoE⫺/⫺ mice), or 24 weeks (12
GPx-1⫺/⫺ApoE⫺/⫺ and 15 control ApoE⫺/⫺ mice). The WTD contained 21% (wt/wt) fat and 0.15% (wt/wt) cholesterol (ssniff;
Spezialdiäten GmbH, Soest, Germany). Mice were kept in accordance with standard animal care requirements, housed 4 to 5 per
cage, and maintained on a 12-hour light– dark cycle. Water and food
were given ad libitum.
Lipoprotein Analysis and Blood
Pressure Measurements
For detailed description of quantitative cholesterol and triglyceride
analyses, fast protein liquid chromatography gel filtration of pooled
plasma samples as well as blood pressure measurements with a
computerized tail-cuff system (please see http://atvb.ahajournals.
org) were performed.
GPx-1 Activity
GPx-1 activity was determined in washed red blood cells obtained
immediately after sampling from whole blood anticoagulated with
EDTA. Hemolyzed cells were stored frozen for up to 1 week.
Enzyme activity was measured as described, with minor modifications (Ransel; Randox, Crumlin, UK).18
Tissue Preparation and Quantitative Morphometry
of Atherosclerotic Lesion Development
Mice were euthanized by exposure to carbon dioxide. Peritoneal
cavities were opened and the cadavers fixed in 4% buffered
formaldehyde. Hearts and aortas were resected en bloc down to the
iliac bifurcation and carefully cleaned of perivascular adipose tissue
under a dissection microscope (Leica MZ6; Leica, Bensheim, Germany). The aortic arch and the rest of the aorta from the arch to the
iliac bifurcation were separated. Longitudinal sections of the aortic
arch were stained with trichrome and computer-assisted (Image Pro
Discovery; Media Cybernetics, Silver Spring, Md) measurement of
plaque size was performed as described previously19 (supplemental
Figure I, please see http://atvb.ahajournals.org).
GPx-1 and Atherogenesis
851
Immunohistochemical and Histochemical Analyses
Immunostaining of murine tissues with the murine MAbs was
performed using the Vector M.O.M. immunodetection kit (Vector
Laboratories, Burlingham, Calif; please see http://atvb.ahajournals.
org). The following antibodies and dilutions were used: murine
monoclonal IgM EO6 (1:500) binding the phosphorylcholine headgroup of oxidized but not native phospholipids, rabbit anti-murine JE
(MCP-1, 1:100), murine monoclonal IgG NA59 (1:100) binding
4-hydroxynonenal lysine-epitopes, rabbit anti-nitrotyrosine (1:500;
Upstate, Dundee, UK), and murine anti-smooth muscle ␣-actin (1A4,
1:100; Sigma, St Louis, Mo). Collagen content was analyzed by
picrosirius red and polarized light microscopic imaging. Percentpositive area for immunohistochemical or picrosirius red staining of
the inner aortic arch intima (lesser curvature) was quantified by
Photoshop-based image analysis as described.20,21 Briefly, pixels
with similar chromogen characteristics were selected with the
“magic wand” tool and the “select similar” command, and the ratio
of the positively stained area to the total lesion area studied was
calculated with the “histogram” command in Photoshop. For details
of the procedure for determination of MCP-1 immunostaining
intensity please see http://atvb.ahajournals.org. All quantitative morphometric and immunohistochemical data were collected independently by 2 experienced operators blinded to the mice genotypes.
Determination of ApoB IgG and IgM Immune
Complexes (IgG and IgM IC/apoB) and
Autoantibodies to Oxidized LDL
The IC/apoB in mouse plasma were assessed after 12 and 24 weeks
on the WTD with a “sandwich” chemiluminescence immunoassay as
previously described.22,23
The levels of IgG and IgM autoantibodies binding to malondialdehyde (MDA)–LDL and copper– oxidized LDL (low-density lipoprotein) were determined by chemiluminescence-based enzymelinked immunosorbent assay as described.24 MDA-LDL and copper–
oxidized low-density lipoprotein was generated as previously
described.25,26 Mice sera were diluted 1:200.
Oxidative Fluorescent Microtopography and
Detection of ROS Formation Using
Diogenes-Enhanced Chemiluminescence
Superoxide was detected in situ using dihydroethidium fluorescence
as described recently.27 For superoxide and hydrogen peroxide
detection, please see http://atvb.ahajournals.org.
Isolation of Peritoneal Macrophages
Mouse peritoneal macrophages were prepared from GPx-1⫺/⫺
ApoE⫺/⫺, and ApoE⫺/⫺ mice were prepared by intraperitoneal injection of 1 mL 3% thioglycollate (Merck, Darmstadt, Germany). After
4 days, cells were harvested by intraperitoneal lavage with 7 mL
DMEM and centrifuged for 5 minutes at 1250 rpm. The pellet was
resuspended in DMEM with 10% fetal calf serum and plated in
bacterial dishes. After incubation for 4 hours, nonadherent cells were
removed.
Proliferation Assay
The thioglycollate-elicited macrophages were incubated for 4 days
with macrophage– colony stimulating factor (10 ng/mL; PeproTech,
London, UK). Cells were detached with Accutase (PAA Laboratories, Pasching, Austria) and plated in a 96-well plate (2.5⫻104
cells/well). Cells were incubated again with macrophage– colony
stimulating factor and BrdU for another 16 hours, and the proliferation assay (Roche, Mannheim, Germany) was performed according
to the manufacturer’s instruction. Briefly, after fixation and permeabilization of the cells, the incorporated BrdU was detected by an
anti-BrdU—POD antibody followed by incubation with Luminol.
The amount of bound antibody was quantified by determination of
relative light units with a chemiluminescence plate reader (Fluoroscan; Thermo Labsystems, Waltham, Mass).
852
Arterioscler Thromb Vasc Biol.
April 2007
B
% positively stained plaque area
A
Figure 1. Atherosclerosis lesion progression. A, After consuming the
WTD for 12 or 24 weeks, both GPx1⫺/⫺ApoE⫺/⫺ and ApoE⫺/⫺ mice aortae
from the arch down to the iliac bifurcation were isolated and stained for
lipid deposition with Sudan IV. Sudan
stained atherosclerotic lesions en
face were then quantified using
Photoshop-based image analysis.
Data are presented as boxplots with
median, interquartile range, minimum,
and maximum. B, Representative
specimens of GPx-1⫺/⫺ApoE⫺/⫺ and
ApoE⫺/⫺ mice aortae used to calculate the percentages of Sudan
stained areas shown in the boxplots
in (A).
ApoE-/GPx-1-/- ApoE-/-
ApoE
DKO
40,0
40
p<0.01
30,0
30
n.s.
20,0
20
10,0
10
00,0
12 weeks
24 week
s
ApoE-/-
TUNEL Assay
GPx-1-/-ApoE-/-
Thioglycollate-elicited peritoneal macrophages were plated in 16well chamber slides (1⫻105 cells/well) in DMEM supplemented
with 10% fetal calf serum. After 24 hours, nonadherent cells were
removed and the resident macrophages were incubated for another 7
days without additional stimuli. Staining of apoptotic cells was
performed with the in situ cell death staining kit (Roche, Mannheim,
Germany) according to the manufacturer’s instructions. Percentage
of apoptotic cells was evaluated by counting the stained cells/1⫻105
cells seeded.
significant differences between the 2 strains after 6, 12, and
24 weeks, respectively (supplemental Figure IIA). Furthermore, fast protein liquid chromatography analysis of lipoproteins showed that disruption of GPx-1 gene function did not
affect lipoprotein distribution in mice consuming the atherogenic diet (supplemental Figure IIB).
Also systolic blood pressure was similar after 6 weeks on
a WTD (118⫾4.4 mm Hg in GPx-1⫺/⫺ApoE⫺/⫺ mice [n⫽4]
versus 114.6⫾2.7 mm Hg in ApoE⫺/⫺ mice [n⫽3]).
Reporter Cell Assay for Determination of Aortic
Nitric Oxide Production
Atherosclerosis Lesion Progression
Aortic nitric oxide production was bioassayed by determination of
the cGMP content in RFL-6 rat lung fibroblasts as reporter cells
(please see supplemental materials).
Statistical Analyses
Data were analyzed with SPSS 12.0 for Windows (SPSS Inc). Most
of the outcome parameters used in this study did not follow a normal
distribution as judged by Shapiro-Wilk tests, so statistical analyses
were performed with Mann-Whitney U tests. Except for plasma lipid
contents and blood pressure measurements (mean⫾SD), data in the
text are presented as median and interquartile range. Data in the
Figures are presented as boxplots with median, interquartile range,
minimum, and maximum. Differences were considered significant
when P⬍0.05.
Results
ⴚ/ⴚ
Generation of GPx-1
ApoEⴚ/ⴚ Mice
To study the role of GPx-1 in atherogenesis, we generated
double-knockout (GPx-1⫺/⫺ApoE⫺/⫺) mice by crossing GPx1⫺/⫺ with ApoE⫺/⫺ mice, both on the C57/BL6 background
(please see supplemental materials).
GPx-1 Deficiency Does Not Affect Serum
Lipoproteins or Systolic Blood Pressure
The WTD significantly increased plasma total cholesterol but
not triglyceride levels in both GPx-1⫺/⫺ApoE⫺/⫺ mice and
control ApoE⫺/⫺ mice compared with baseline values with no
Compared with control ApoE⫺/⫺ mice (n⫽15), doubleknockout mice (GPx1⫺/⫺ ApoE⫺/⫺, n⫽12) developed significantly more atherosclerosis after 24 weeks on the WTD as
indicated by plaque area en face of the aorta from the arch
down to the iliac bifurcation (Figure 1A and 1B). This
difference was mainly caused by significantly more atherosclerosis in the distal part of the aorta from the diaphragm
down to the iliac bifurcation (P⬍0.01, data not shown). A
similar, albeit statistically not significant, trend could already
be observed as early as after 12 weeks on the WTD (Figure
1A), but not yet after 6 weeks on the WTD (% plaque area,
median and interquartile range: GPx-1⫺/⫺ApoE⫺/⫺, 0.7/0.65;
ApoE⫺/⫺, 1.1/1.0). Taken together, these data indicate that
GPx-1 deficiency accelerates atherosclerotic lesion development in ApoE⫺/⫺ mice.
Phenotypic Analysis of Atherosclerotic Lesions
Quantification of the maximal area and thickness (median/
interquartile range) of the inner aortic arch intima (lesser
curvature) revealed no significant differences in control mice
compared with double-knockout mice after 6 weeks on the
WTD (area, 446.372/575.210 ␮ m 2 versus 184.386/
185.062 ␮m2; thickness, 298/380 ␮m versus 203/167 ␮m.
The relatively high medians and interquartile ranges are
attributable to extreme outliers and high interindividual vari-
Torzewski et al
12 weeks WTD
*
*
ApoE
DKO
100
100,0
853
A
24 weeks WTD
A
ApoE-/-
100
100,0
GPx-1-/- ApoE-/80,0
80
p<0.01
80,0
80
ApoE-/60,0
60
60,0
60
40,0
40
40,0
40
20,0
20
20,0
20
0
0,0
0
0,0
n.s.
ApoE-/-
11
*
*
macrophages
ApoE-/-GPx-1-/-
ApoE-/-GPx-1-/-
ApoE-/-
ApoE-/-
ApoE-/-GPx-1-/-
ApoE-/-GPx-1-/-
B
ApoE
DKO
n.s.
ApoE-/-
100,0
100
GPx-1-/- ApoE-/-
80,0
80
80,0
80
12 weeks WTD
100,0
100
p<0.01
1
60,0
60
60,0
60
40,0
40
40,0
40
20,0
20
20,0
20
0
0,0
0
0,0
smooth muscle cells
C
C
24 weeks WTD
B
ApoE
DKO
100,0
100
n.s.
80,0
80
p<0.01
*
ApoE-/-
100,0
100
GPx-1-/- ApoE-/-
80,0
80
60,0
60
60,0
60
40,0
40
40,0
40
20,0
20
20,0
20
0,0
0
0,0
ApoE-/-
0
collagen
Figure 2. Phenotypic analysis of atherosclerotic lesions after 12
(left panels) and 24 weeks (right panels) on the WTD. Atherosclerotic lesions of the inner aortic arch intima (lesser curvature)
of GPx-1⫺/⫺ApoE⫺/⫺ and ApoE⫺/⫺ mice were quantified for macrophages (A), smooth muscle cells (B), and collagen (C).
Percent-positive area for macrophages, smooth muscle cells,
and collagen was quantified by Photoshop-based image analysis (see also Figure 3). Data are presented as boxplots with
median, interquartile range, minimum, and maximum.
ations at this early time point). In contrast, consistently, albeit
not significantly, less atherosclerosis was observed in control
mice compared with double-knockout mice after 12 weeks
(area, 115.962/114.594 ␮m2 versus 128.633/87.096 ␮m2;
thickness, 215/151 ␮m versus 237/173 ␮m) or 24 weeks on
the WTD (area, 382.551/182.114 ␮m2 versus 416.530/
71.370 ␮m2; thickness, 328/201 ␮m versus 334/100 ␮m).
Furthermore, the histomorphological aspect of lesions in
double-knockout mice differed markedly from lesions in
control mice in terms of plaque composition. At 12 weeks on
*
ApoE-/-GPx-1-/-
*
ApoE-/-
*
ApoE-/-GPx-1-/-
Figure 3. Representative examples of atherosclerotic lesion
composition after 12 (left panels) and 24 weeks (right panels) on
the WTD. Atherosclerotic lesions of the inner aortic arch intima
(lesser curvature) of GPx-1⫺/⫺ApoE⫺/⫺ and ApoE⫺/⫺ mice were
stained with trichrome (A) (for quantification of macrophages),
mouse anti-smooth muscle ␣-actin (1A4) (B) (for quantification
of smooth muscle cells), and picrosirius red with subsequent
polarization (C) (for quantification of collagen). Percent-positive
area for macrophages (*), smooth muscle cells (brown), and collagen (yellow, green, orange, and red polarized color) were
quantified by Photoshop-based image analysis (see also Figure
4). The lumen is to the upper left corner. The demarcation
between intima and media is indicated by an arrowhead. C,
Note that the adventitial tissue (*) also polarizes after picrosirius
red staining (internal positive control). Magnification ⫻64.
the WTD, atherosclerotic lesions in ApoE⫺/⫺ mice contained
significantly less macrophages (area covered by macrophages; Figures 2A and 3A, left panels) and significantly
more collagen (area covered by collagen; Figures 2C and 3C,
left panels) than lesions in double-knockout mice. Further-
854
*
1250
1250
E
April 2007
ApoE -/ApoE-/-GPx-1-/-
Figure 4. In situ detection of superoxide in the
aorta from GPx-1⫺/⫺ApoE⫺/⫺ and ApoE⫺/⫺ mice
after 6 weeks on the WTD. Left, Fluorescent photomicrographs of microscopic sections of aortic
rings. Vessels were labeled with dihydroethidium
dye, which produces red fluorescence when oxidized to ethidine by superoxide. The green
autofluorescence corresponds to the basal laminae.
E indicates endothelium; M, media; A, adventitia.
Right, Data presented as boxplots with median,
interquartile range, minimum, and maximum.
1000
1000
IOD [arbitrary units]
M
A
ApoE-/-
E
750750
500500
250250
M
00
A
GPx-1-/-ApoE-/-
superoxide formation
more, a slight but yet not significant increase in the amount of
smooth muscle cells could be observed (area covered by
smooth muscle cells; Figures 2B and 3B, left panels). At 24
weeks on the WTD, the difference in the amount of smooth
muscle cells in double-knockout mice was highly significant
(Figures 2B and 3B, right panels). Macrophages and collagen
content were not significantly different at this time point
(Figures 2A, 2C, 3A, 3C, right panels). Collectively, these
data indicate that GPx-1 deficiency leads to increased cellularity with a higher relative number of macrophages in early
and a higher relative number of smooth muscle cells in
advanced atherosclerotic lesions when compared with control
ApoE⫺/⫺ mice.
Superoxide Formation in the Aortic Wall and
Detection of ROS Formation in Aortic Vessel
Segments, Heart Mitochondria, and
Membrane Fractions
We assessed superoxide formation in cryosectioned aortic
rings after 6 weeks on a WTD by means of dihydroethidiumderived fluorescence. Staining of aortic sections with dihydroethidium revealed a marked (50%) increase in vascular
superoxide in GPx-1⫺/⫺ApoE⫺/⫺ mice compared with control
ApoE⫺/⫺ mice. This increase was observed primarily in the
endothelium, but less pronounced in the intima and adventitia
(Figure 4). Furthermore, detection of ROS formation in aortic
vessel segments, heart mitochondria, and membrane fractions
point toward increased oxidative stress in GPx-1⫺/⫺ApoE⫺/⫺
mice because of activation of mitochondria and NADPH
oxidases or because of impaired antioxidative defense mechanism in the GPx-1⫺/⫺ animals (supplemental Figure III).
We did not find evidence for an increased lipoprotein
oxidation in GPx-1⫺/⫺ApoE⫺/⫺ mice as determined by both
immunohistochemical quantification of oxidized LDL and
measurement of ApoB IgG and IgM immune complexes and
autoantibodies to oxidized LDL (supplemental Figure IV).
Proliferation, Apoptosis, and Recruitment
of Monocytes
Next, we investigated whether proliferative activity and/or
less apoptosis of monocyte-derived macrophages might
account for the increased cellularity of early atherosclerotic lesions in GPx-1⫺/⫺ApoE⫺/⫺ mice. The proliferation
rate of mouse peritoneal macrophages was investigated
with a BrdU-based chemiluminescence assay. After stimulation with macrophage– colony-stimulating factor macrophages from GPx-1⫺/⫺ApoE⫺/⫺ mice showed significantly more BrdU incorporation than macrophages from
ApoE⫺/⫺ control mice (Figure 5A). Apoptosis of mouse
peritoneal macrophages was investigated with a terminal
deoxynucleotidyl transferase-mediated dUTP nick endlabeling assay. After cultivation for 7 days without additional stimuli, percentage of apoptotic cells was not
significantly different in control mice compared with
double-knockout mice (Figure 5B).
These results indicate that increased proliferative activity
rather than diminished apoptotic rates of macrophages contribute to the increased cellularity of early atherosclerotic
lesions in GPx-1⫺/⫺ApoE⫺/⫺ mice. Furthermore, we did not
find evidence for increased expression of MCP-1 in lesions of
B
A
ApoE
DKO
25.000
25.000
p<0,001
20.000
20.000
48
15.000
15.000
10.000
10.000
5.000
5.000
00
100
100
positively stained cells/
1 x 105 cells seeded
30.000
30.000
relative light units [RLUs]/
2,5 x 104 cells
Influence of GPx-1 Deficiency on
Lipoprotein Oxidation
8080
6060
4040
n.s.
2020
00
proliferation
apoptosis
Figure 5. Proliferative activity and apoptosis. A,
After differentiation of thioglycollate-elicited
mouse peritoneal monocytes for 4 days in macrophage– colony stimulating factor (10 ng/mL),
cells were incubated with BrdU for another 16
hours and the proliferation rate was investigated
with a BrdU-based chemiluminescence assay
(n⫽8). B, After cultivation of thioglycollate-elicited
mouse peritoneal monocytes for 7 days without
additional stimuli, staining of apoptotic cells was
performed with an in situ cell death staining kit.
Percentage of apoptotic cells was evaluated by
counting the stained cells/1⫻105 cells (n⫽8).
Data are presented as boxplots with median,
interquartile range, minimum, and maximum.
Torzewski et al
GPx-1⫺/⫺ApoE⫺/⫺ mice, that might account for increased
recruitment of monocytes/macrophages into the lesions (supplemental Figure V).
Vascular Nitric Oxide Production and
Protein Nitration
The bioassay of aortic nitric oxide production indicated that
less bioactive nitric oxide is generated in response to acetylcholine in aortae of double-knockout mice compared with
control mice. Likewise, the observation of extensive protein
nitration in atherosclerotic lesions of double-knockout mice
suggests that peroxynitrite is produced and may be involved
in the different evolution of the lesions in control and
double-knockout mice (supplemental Figure VI).
Discussion
The present study on atherosclerosis development in GPx-1⫺/⫺
ApoE⫺/⫺ mice has provided several key results. Atherosclerotic lesion development is more rapid in GPx-1⫺/⫺ ApoE⫺/⫺
double-knockout mice with larger lesions after 12 (not
significant) and 24 weeks on a WTD compared with ApoE⫺/⫺
mice. The lesions are more cellular with an increase in
macrophage content in early lesions and an increase in
smooth muscle cell content in advanced lesions. As expected,
arterial walls of double-knockout mice contain more ROS
than tissue from controls. Consistent with the previously
reported endothelial dysfunction in GPx-1⫺/⫺ mice,11–13 aortic
tissue from double-knockout mice produced less bioactive
nitric oxide in response to acetylcholine than that of ApoE⫺/⫺
control mice. Somewhat unexpectedly, we could not find
consistent evidence of increased lipoprotein oxidation in
GPx-1⫺/⫺ApoE⫺/⫺ mice by immunohistochemistry as well as
determination of apoB-immune complexes and autoantibodies against oxidized LDL. Taking into account the inverse
association of GPx-1 activity and the risk for cardiovascular
events in humans with coronary artery disease,8 these results
provide further strong evidence that GPx-1 plays an important antiatherogenic role in the arterial wall.
The accelerated development of atherosclerosis in the
present study is in contrast to recent observations in simple
GPx-1⫺/⫺ mice, which have no enhanced atherosclerotic
lesion development.17 One should keep in mind though, that
lesion development in fat fed C57BL/6 mice is very discrete
and confined to the aortic root. Furthermore, this model is
associated with severe nonvascular pathology (eg, hepatic)
caused by the extreme diet modification. Therefore, it has
been generally abandoned for atherosclerosis research after
the availability of hyperlipidemic ApoE⫺/⫺ and LDL-receptor⫺/⫺ mice.
Besides the increased lesion size in double-knockout mice,
another striking difference between the 2 genotypes in atherosclerotic lesion development relates to histomorphological
composition in the aortic arch lesion. At 12 weeks on the
WTD, lesions in the aortic arch were particularly rich in
macrophages. GPx-1⫺/⫺ApoE⫺/⫺ mice showed a lesional macrophage content that was approximately twice that of corresponding lesions in ApoE⫺/⫺ mice (Figure 4A). This increase
in macrophage cellularity was paralleled by a comparative
decrease in extracellular matrix as determined by the collagen
855
content (Figure 4C). At 24 weeks on the WTD, lesions in the
aortic arch were much more advanced with smooth muscle
cells being the predominant cell type. At this time point, the
relative number of smooth muscle cells was significantly
higher in GPx-1⫺/⫺ ApoE⫺/⫺ mice when compared with
ApoE⫺/⫺ mice with no concomitant increase in collagen
content, which is surprising at first glance (Figure 4B).
However, this latter observation is in line with recent data
demonstrating that increased oxidative stress activates matrix
metalloproteinases and decreases fibrillar collagen synthesis
in rat cardiac fibroblasts28 and, vice versa, overexpression of
GPx attenuates matrix metalloproteinase-9 zymographic and
protein levels in mouse myocardium.29
This propensity toward more cellular atherosclerotic lesions of GPx-1⫺/⫺ ApoE⫺/⫺ mice was accompanied by severely impaired disposal of ROS in mitochondria. Whereas
severe oxidative stress is cytotoxic, mild oxidative stress may
stimulate cell proliferation, as shown recently in various cell
types including vascular smooth muscle cells.30 –33 Proliferation of vascular smooth muscle cells is a hallmark of
atherosclerosis development. Furthermore, many data support
the concept that also macrophages in early lesions may
derive, at least in part, from local proliferation, especially
under circumstances of enhanced oxidative stress.34 The
combination of hypertension and hyperlipidemia in LDLreceptor⫺/⫺ mice with hyperglycemia and further oxidative
stress by glucose-oxidized LDL has recently been shown to
induce macrophage proliferation in vascular lesions.35 We
could show that GPx-1 deficiency increases proliferation of
macrophages in vitro in response to macrophage– colonystimulating factor. In contrast, GPx-1 deficiency obviously
does not influence the extent of apoptosis of macrophages in
vitro or MCP-1 protein expression in vivo. Therefore, we
propose that increased cellularity of early atherosclerotic
lesions in GPx-1⫺/⫺ApoE⫺/⫺ mice is mainly caused by increased proliferative activity rather than diminished apoptotic
rates or increased recruitment of monocytes/macrophages
into the lesions.
The data obtained in GPx-1⫺/⫺ ApoE⫺/⫺ mice add further
information to the role of antioxidant defenses in atherosclerosis. Antioxidant enzymes have been the topic of many
investigations in experimental animals and epidemiologic
studies in humans. In mouse models of atherosclerosis, data
are available on the 3 superoxide dismutases (SOD), catalase,
and glutathione peroxidase. It is remarkable that neither
overexpression of copper, zinc–SOD, or of endothelial cell–
SOD protects against atherosclerosis in ApoE⫺/⫺ or LDLreceptor⫺/⫺ mice.36 –38 Accordingly, deletion of endothelial
cell–SOD is not proatherogenic.39 In contrast to these models
of atherosclerosis, neointima formation in balloon injury
models has been reported to be inhibited by overexpression of
endothelial cell–SOD.40 Because homozygous Mn–SOD deficiency leads to severe pathology and death soon after birth,
only mice with heterozygous Mn-SOD deficiency have been
analyzed for atherosclerosis development. In these mice,
there was an increase in branchpoint lesions observed in the
aorta, interpreted as interaction of high shear-stress with the
genotype. However, only 4 animals per group were analyzed
and no data on the overall size of lesions were presented.41 In
856
April 2007
summary, the SODs with the possible exception of Mn-SOD
apparently do not protect hyperlipidemic mice against atherosclerosis. Similarly, also, inactivation of NADPH oxidase
(with a significant reduction in vascular superoxide generation) had no protective effect.42,43 Finally, no association of
red blood cell copper-zinc-SOD activity with cardiovascular
risk was observed by us in the same study that revealed the
inverse association of GPx-1 activity and cardiovascular
risk.8
While the role of SODs in atherogenesis has been surprisingly minor, it has been shown that overexpression of catalase
is protective in a hyperlipidemic mouse model.38 Like glutathione peroxidases, catalase inactivates H2O2. Even though
definitive experimental evidence is missing to date, these data
implicate that H2O2 may be relatively more important in
atherogenesis than superoxide anion. This may be related to
the fact that the latter is very short-lived (also in the absence
of SOD) and contributes to the generation of H2O2. It is
conceivable that as long as H2O2 is effectively converted to
water, minor changes in the production and half-life of
superoxide outside mitochondria may not be critical for
atherogenesis. Aside from these considerations, it is wellestablished that superoxide and H2O2 induce different cellular
responses, eg, by different effector mechanisms in transcriptional regulation.33,44
An intriguing aspect of the increased lesion size relates to
the reduction of antioxidant enzymes including GPx-1 in the
aortic arch of ApoE⫺/⫺ mice before or at least when lesions
develop.15,16 Even though these 2 studies come to slightly
different results, this may be interpreted as indicating that
GPx-1 and other antioxidant enzymes cannot be critical for
atherosclerotic lesion development, because they are no
longer expressed at high levels when lesions developed. It
should be noted, though, that there is definitely remaining
GPx-1 activity at all stages of atherosclerotic lesions in the
aortic arch. This could provide some protection in the
ApoE⫺/⫺ mouse, which is lost in the double-knockouts.
Furthermore, in the early phase of lesion development, when
GPx-1 expression has been reported to be normal in one
study, there is a significant difference in lesion composition
toward a more macrophage-rich atheroma in GPx-1⫺/⫺
ApoE⫺/⫺ mice. Previous work has already demonstrated that
GPx-1 was expressed at higher level in the descending aorta
than in the aortic arch in Apo E⫺/⫺ mice from the age of 12
weeks onwards.15 The substantially higher residual expression of GPx-1 in atherosclerotic lesions of the descending
aorta would indicate that the difference in enzyme activity
between ApoE⫺/⫺ and GPx-1⫺/⫺ ApoE⫺/⫺ is more pronounced
in this part of the aorta. This observation might explain the
larger difference in lesion development observed in the
abdominal aorta after 24 weeks.
Our knowledge concerning the contributions of different
ROS and their reaction products to atherogenesis is still
incomplete. This fact is impressively underscored by the
surprising but notorious lack of any true preventive effects of
different antioxidant treatments in large clinical trials.45,46
However, the results from different genetically modified
mouse models show that modification of antioxidant defense
systems may indeed add important clues to our understanding
of oxidative stress in atherogenesis and may eventually
redirect clinical interest toward the development of effective
preventive interventions in patients at risk for cardiovascular
disease.
Acknowledgments
The authors thank Ye-Shi Ho, Department of Biochemistry, Wayne
State University, Detroit, Mich, for generously providing GPx-1⫺/⫺
mice backcrossed to C57BL/6. The authors are indebted to Antje
Canisius, Adriana Degreif, Elizabeth Miller, and Carolin Orning for
their excellent technical assistance.
Sources of Funding
This work was supported by the Mainzer Forschungsföderungsprogramm (MAIFOR 2004 to M.T.), the Deutsche Forschungsgemeinschaft (DFG LA 499/3-1 and 499/3-2 to K.J.L. and M.T.), SFB 553
(TP1 to H.L. and U.F., TP17 to T.M.), and the Center of Preventive
Medicine of the Medical Faculty (University of Mainz).
Disclosure
None.
References
1. Forsberg L, de Faire U, Morgenstern R. Oxidative stress, human genetic
variation, and disease. Arch Biochem Biophys. 2001;389:84 –93.
2. Raes M, Michiels C, Remacle J. Comparative study of the enzymatic
defense systems against oxygen-derived free radicals: the key role of
glutathione peroxidase. Free Radic Biol Med. 1987;3:3–7.
3. Ursini F, Maiorino M, Brigelius-Flohe R, Aumann KD, Roveri A,
Schomburg D, Flohe L. Diversity of glutathione peroxidases. Methods
Enzymol. 1995;252:38 –53.
4. Arthur JR. The glutathione peroxidases. Cell Mol Life Sci. 2000;57:
1825–1835.
5. Flohe L. Glutathione peroxidase. Basic Life Sci. 1988;49:663– 668.
6. Sies H. Glutathione and its role in cellular functions. Free Radic Biol
Med. 1999;27:916 –921.
7. Glass CK, Witztum JL. Atherosclerosis. the road ahead. Cell. 2001;104:
503–516.
8. Blankenberg S, Rupprecht HJ, Bickel C, Torzewski M, Hafner G, Tiret L,
Smieja M, Cambien F, Meyer J, Lackner KJ. Glutathione peroxidase 1
activity and cardiovascular events in patients with coronary artery
disease. N Engl J Med. 2003;349:1605–1613.
9. Ho YS, Magnenat JL, Bronson RT, Cao J, Gargano M, Sugawara M,
Funk CD. Mice deficient in cellular glutathione peroxidase develop
normally and show no increased sensitivity to hyperoxia. J Biol Chem.
1997;272:16644 –16651.
10. Guo Z, Van Remmen H, Yang H, Chen X, Mele J, Vijg J, Epstein CJ, Ho
YS, Richardson A. Changes in expression of antioxidant enzymes affect
cell-mediated LDL oxidation and oxidized LDL-induced apoptosis in
mouse aortic cells. Arterioscler Thromb Vasc Biol. 2001;21:1131–1138.
11. Forgione MA, Weiss N, Heydrick S, Cap A, Klings ES, Bierl C,
Eberhardt RT, Farber HW, Loscalzo J. Cellular glutathione peroxidase
deficiency and endothelial dysfunction. Am J Physiol Heart Circ Physiol.
2002;282:H1255–H1261.
12. Dayal S, Brown KL, Weydert CJ, Oberley LW, Arning E, Bottiglieri T,
Faraci FM, Lentz SR. Deficiency of glutathione peroxidase-1 sensitizes
hyperhomocysteinemic mice to endothelial dysfunction. Arterioscler
Thromb Vasc Biol. 2002;22:1996 –2002.
13. Forgione MA, Cap A, Liao R, Moldovan NI, Eberhardt RT, Lim CC,
Jones J, Goldschmidt-Clermont PJ, Loscalzo J. Heterozygous cellular
glutathione peroxidase deficiency in the mouse: abnormalities in vascular
and cardiac function and structure. Circulation. 2002;106:1154 –1158.
14. Lapenna D, de Gioia S, Ciofani G, Mezzetti A, Ucchino S, Calafiore AM,
Napolitano AM, Di Ilio C, Cuccurullo F. Glutathione-related antioxidant
defenses in human atherosclerotic plaques. Circulation. 1998;97:
1930 –1934.
15. t’Hoen PA, Van der Lans CA, Van Eck M, Bijsterbosch MK, Van Berkel
TJ, Twisk J. Aorta of ApoE-deficient mice responds to atherogenic
stimuli by a prelesional increase and subsequent decrease in the
expression of antioxidant enzymes. Circ Res. 2003;93:262–269.
Torzewski et al
16. Biswas SK, Newby DE, Rahman I, Megson IL Depressed glutathione
synthesis precedes oxidative stress and atherogenesis in Apo-E(⫺/⫺)
mice. Biochem Biophys Res Commun. 2005;338:1368 –1373; Epub 2005
Oct: 1326.
17. de Haan JB, Witting PK, Stefanovic N, Pete J, Daskalakis M, Kola I,
Stocker R, Smolich JJ. Deficiency in the antioxidant enzyme glutathione
peroxidase-1 (Gpx1) does not increase atherosclerosis in C57BL/J6 mice
fed a high fat diet. J Lipid Res. 2006;28:28.
18. Paglia DE, Valentine WN. Studies on the quantitative and qualitative
characterization of erythrocyte glutathione peroxidase. J Lab Clin Med.
1967;70:158 –169.
19. Mach F, Schonbeck U, Sukhova GK, Atkinson E, Libby P. Reduction of
atherosclerosis in mice by inhibition of CD40 signalling. Nature. 1998;
394:200 –203.
20. Lehr HA, Mankoff DA, Corwin D, Santeusanio G, Gown AM. Application of photoshop-based image analysis to quantification of hormone
receptor expression in breast cancer. J Histochem Cytochem. 1997;45:
1559 –1565.
21. Lehr HA, van der Loos CM, Teeling P, Gown AM. Complete chromogen
separation and analysis in double immunohistochemical stains using
Photoshop-based image analysis. J Histochem Cytochem. 1999;47:
119 –126.
22. Svenungsson E, Jensen-Urstad K, Heimburger M, Silveira A, Hamsten A,
de Faire U, Witztum JL, Frostegard J. Risk factors for cardiovascular
disease in systemic lupus erythematosus. Circulation. 2001;104:
1887–1893.
23. Schneider M, Witztum JL, Young SG, Ludwig EH, Miller ER, Tsimikas
S, Curtiss LK, Marcovina SM, Taylor JM, Lawn RM, Innerarity TL, Pitas
RE High-level lipoprotein [a] expression in transgenic mice: evidence for
oxidized phospholipids in lipoprotein [a] but not in low density
lipoproteins. J Lipid Res. 2005;46:769 –778; Epub 2005 Jan: 2016.
24. Freigang S, Horkko S, Miller E, Witztum JL, Palinski W. Immunization
of LDL receptor-deficient mice with homologous malondialdehydemodified and native LDL reduces progression of atherosclerosis by mechanisms other than induction of high titers of antibodies to oxidative
neoepitopes. Arterioscler Thromb Vasc Biol. 1998;18:1972–1982.
25. Horkko S, Bird DA, Miller E, Itabe H, Leitinger N, Subbanagounder G,
Berliner JA, Friedman P, Dennis EA, Curtiss LK, Palinski W, Witztum
JL. Monoclonal autoantibodies specific for oxidized phospholipids or
oxidized phospholipid-protein adducts inhibit macrophage uptake of
oxidized low-density lipoproteins. J Clin Invest. 1999;103:117–128.
26. Tsimikas S, Palinski W, Witztum JL. Circulating autoantibodies to
oxidized LDL correlate with arterial accumulation and depletion of
oxidized LDL in LDL receptor-deficient mice. Arterioscler Thromb Vasc
Biol. 2001;21:95–100.
27. Munzel T, Afanas’ev IB, Kleschyov AL, Harrison DG. Detection of
superoxide in vascular tissue. Arterioscler Thromb Vasc Biol. 2002;22:
1761–1768.
28. Siwik DA, Pagano PJ, Colucci WS. Oxidative stress regulates collagen
synthesis and matrix metalloproteinase activity in cardiac fibroblasts.
Am J Physiol Cell Physiol. 2001;280:C53–C60.
29. Shiomi T, Tsutsui H, Matsusaka H, Murakami K, Hayashidani S, Ikeuchi
M, Wen J, Kubota T, Utsumi H, Takeshita A Overexpression of glutathione peroxidase prevents left ventricular remodeling and failure after
myocardial infarction in mice. Circulation. 2004;109:544 –549; Epub
2004 Jan: 2026.
857
30. Martindale JL, Holbrook NJ. Cellular response to oxidative stress: signaling for suicide and survival. J Cell Physiol. 2002;192:1–15.
31. Matsuoka T, Wada J, Hashimoto I, Zhang Y, Eguchi J, Ogawa N, Shikata
K, Kanwar YS, Makino H. Gene delivery of Tim44 reduces mitochondrial superoxide production and ameliorates neointimal proliferation
of injured carotid artery in diabetic rats. Diabetes. 2005;54:2882–2890.
32. Hu Y, Rosen DG, Zhou Y, Feng L, Yang G, Liu J, Huang P Mitochondrial manganese-superoxide dismutase expression in ovarian cancer:
role in cell proliferation and response to oxidative stress. J Biol Chem.
2005;280:39485–39492; Epub 32005 Sep: 39422.
33. Stocker R, Keaney JF, Jr. Role of oxidative modifications in atherosclerosis. Physiol Rev. 2004;84:1381–1478.
34. Sakai M, Kobori S, Miyazaki A, Horiuchi S. Macrophage proliferation in
atherosclerosis. Curr Opin Lipidol. 2000;11:503–509.
35. Lamharzi N, Renard CB, Kramer F, Pennathur S, Heinecke JW, Chait A,
Bornfeldt KE. Hyperlipidemia in concert with hyperglycemia stimulates
the proliferation of macrophages in atherosclerotic lesions: potential role
of glucose-oxidized LDL. Diabetes. 2004;53:3217–3225.
36. Tribble DL, Gong EL, Leeuwenburgh C, Heinecke JW, Carlson EL,
Verstuyft JG, Epstein CJ. Fatty streak formation in fat-fed mice
expressing human copper-zinc superoxide dismutase. Arterioscler
Thromb Vasc Biol. 1997;17:1734 –1740.
37. Laukkanen MO, Leppanen P, Turunen P, Porkkala-Sarataho E, Salonen
JT, Yla-Herttuala S. Gene transfer of extracellular superoxide dismutase
to atherosclerotic mice. Antioxid Redox Signal. 2001;3:397– 402.
38. Yang H, Roberts LJ, Shi MJ, Zhou LC, Ballard BR, Richardson A, Guo
ZM Retardation of atherosclerosis by overexpression of catalase or both
Cu/Zn-superoxide dismutase and catalase in mice lacking apolipoprotein
E. Circ Res. 2004;95:1075–1081; Epub 2004 Nov: 1074.
39. Sentman ML, Brannstrom T, Westerlund S, Laukkanen MO, YlaHerttuala S, Basu S, Marklund SL. Extracellular superoxide dismutase
deficiency and atherosclerosis in mice. Arterioscler Thromb Vasc Biol.
2001;21:1477–1482.
40. Laukkanen MO, Kivela A, Rissanen T, Rutanen J, Karkkainen MK,
Leppanen O, Brasen JH, Yla-Herttuala S. Adenovirus-mediated extracellular superoxide dismutase gene therapy reduces neointima formation
in balloon-denuded rabbit aorta. Circulation. 2002;106:1999 –2003.
41. Ballinger SW, Patterson C, Knight-Lozano CA, Burow DL, Conklin CA,
Hu Z, Reuf J, Horaist C, Lebovitz R, Hunter GC, McIntyre K, Runge MS.
Mitochondrial integrity and function in atherogenesis. Circulation. 2002;
106:544 –549.
42. Kirk EA, Dinauer MC, Rosen H, Chait A, Heinecke JW, LeBoeuf RC.
Impaired superoxide production due to a deficiency in phagocyte
NADPH oxidase fails to inhibit atherosclerosis in mice. Arterioscler
Thromb Vasc Biol. 2000;20:1529 –1535.
43. Hsich E, Segal BH, Pagano PJ, Rey FE, Paigen B, Deleonardis J, Hoyt
RF, Holland SM, Finkel T. Vascular effects following homozygous disruption of p47(phox) : An essential component of NADPH oxidase.
Circulation. 2000;101:1234 –1236.
44. Liu H, Colavitti R, Rovira II, Finkel T. Redox-dependent transcriptional
regulation. Circ Res. 2005;97:967–974.
45. Brown BG, Cheung MC, Lee AC, Zhao XQ, Chait A. Antioxidant
vitamins and lipid therapy: end of a long romance? Arterioscler Thromb
Vasc Biol. 2002;22:1535–1546.
46. Lonn E, Bosch J, Yusuf S, Sheridan P, Pogue J, Arnold JM, Ross C,
Arnold A, Sleight P, Probstfield J, Dagenais GR. Effects of long-term
vitamin E supplementation on cardiovascular events and cancer: a randomized controlled trial. Jama. 2005;293:1338 –1347.
Data Supplement
Methods
Mice and Genotype Determination
Laboratory mice were maintained at the Central Laboratory Animal Facility of the University of
Mainz under strict SPF conditions. C57BL/6J-ApoEtm1Unc knockout mice (ApoE-/-) backcrossed to
the C57BL/6 strain for 10 generations were purchased from the Jackson Laboratory (Bar
Harbor, ME, strain No 2052). GPx-1-/- mice backcrossed to C57BL/6 were generously provided
by Ye-Shi Ho.1, 2 Both individual knockout mutants were effectively propagated by incrossing.
Heterozygous GPx-1+/- mice were identified by having both a 509-bp and a 293-bp PCR product.
Homozygous GPx-1-/- mice were identified by an exclusive 509-bp PCR product while wild type
mice were identified by an exclusive 293-bp PCR product. The ApoE knockout and wild type
alleles were discriminated as described by the Jackson Laboratory.
Lipoprotein Analysis, Gel Filtration Chromatography of Total Plasma and Blood Pressure
Measurements
Murine sera were diluted 1:3 before quantitative cholesterol and triglyceride analyses.
Quantitative cholesterol determinations were conducted using a colorimetric assay (CHOD-PAP,
Roche Diagnostics, Mannheim, Germany). Triglycerides were determined by quantifying free
glycerine originating from hydrolytic cleavage (GPO-PAP, Roche Diagnostics).
Pooled plasma samples from 6 mice (100 µl) were subjected to fast protein liquid
chromatography (FPLC) gel filtration using a Superose 6 column (Amersham Biosciences,
Piscataway, NJ). Elution was performed at a flow rate of 0.5 ml/min. 60 fractions (0.5 ml) were
collected, and cholesterol concentrations were determined enzymatically (see above).
Systolic blood pressures (BP) were measured with a computerized tail-cuff system using a
ML125 NIBP (Non-Invasive Blood Pressure) Controller in conjunction with a PowerLab system
(ADInstruments, Spechbach, Germany). After 6 weeks on a WTD, mice (n = 3-5) were trained
for three days to be acclimatized to the measurement protocol. Then, BP measurement were
performed at 30°C for 3 consecutive days. 9 measurements were obtained and averaged for
each mouse. This non-invasive method of measuring blood pressure correlates well with intraarterial measurements in normotensive and hypertensive mice.3
Tissue
Preparation
and
Quantitative
Morphometry
of
Atherosclerotic
Lesion
Development
The maximal area and thickness of the inner-aortic-arch intima (lesser curvature) of each
mouse were used to compute averages per group. The rest of the aortae from the arch down to
the iliac bifurcation was opened longitudinally and stained with freshly prepared Sudan IV as
previously described.4 Sudan stained atherosclerotic lesions en face were then quantified using
Photoshop-based image analysis (Version 8.0.1, Adobe Systems Inc., San Jose, CA) as
described.4
Immunohistochemical and Histochemical Analyses
Serial 5 µm-thick longitudinal sections of the paraffin-embedded aortic arch were deparaffinized
in xylene. All slides were treated with 3% H2O2 to block endogenous peroxidase activity. Slides
were incubated consecutively with 5% normal serum to block non-specific binding, primary
antibody for 1 hour, biotin-conjugated secondary anti-mouse or anti-rabbit antibody for 30
minutes and avidin-biotin-peroxidase reagent for 45 minutes at room temperature. The reaction
products were revealed by immersing the slides in diaminobenzidine tetrachloride to give a
brown reaction product. The slides were then counterstained with hematoxylin and mounted.
Negative controls included replacement of the primary antibody by irrelevant isotype-matched
antibodies.
Determination of MCP-1 immunostaining intensity was performed by Photoshop-based image
analysis as described5 with slight modifications. After selecting pixels with similar chromogen
Torzewski et al.
2
characteristics with the “magic wand” tool and the “select similar” command, the image was
transformed to 8-bit grayscale. An optical density plot of the selected area was generated using
the “histogram” tool in the “image” menu. The mean staining intensity (in arbitrary units, AU)
was recorded. Subsequently, the background of the lesion area was selected using the
“inverse” tool in the “select” menu, and immunostaining was quantified again using the
“histogram” tool in the “image” menu. Immunostaining intensity was then calculated as the
difference between MCP-1 immunostaining and background immunostaining and was
designated immunohistochemical index with arbitrary units (AU).
Detection of Reactive Oxygen Species (ROS) Formation Using Diogenes™ Enhanced
Chemiluminescence
The Diogenes™ cellular chemiluminescence enhancement system (National Diagnostics Inc.,
Atlanta, USA) for superoxide and hydrogen peroxide detection was prepared according to the
manufacturer’s instructions. Aortic vessel segments,6 isolated heart mitochondria7
and
membrane fractions8 were prepared as described. For detection of vascular ROS, aortic vessel
segments (0.4 cm length) were placed in Diogenes™ (ready-to-use) and measured at intervals
of 60 s for 20 min and the last value was expressed as counts/min/mg dry weight. Each value
was background corrected. For detection of NADPH oxidase generated ROS in membrane
fractions the suspensions were diluted to a final protein concentration of 0.2 mg/ml in
Diogenes™
(ready-to-use)/PBS
1:4.
Upon
stimulation
with
NADPH
(200
µM)
the
chemiluminescence signal was measured at intervals of 30 s for 5 min and the last value was
expressed as counts/min. For detection of mitochondrial ROS levels the suspensions were
diluted to a final protein concentration of 0.1 mg/ml in Diogenes™ (ready-to-use)/PBS 1:4. Upon
stimulation with succinate (2.5 mM) and myxothiazol (10 µM) the chemiluminescence signal was
measured at intervals of 30 s for 5 min using a Lumat LB9507 (Berthold Technologies, Bad
Wildbad, Germany) and the last value was expressed as counts/min.
Torzewski et al.
3
Determination of ApoB IgG and IgM Immune Complexes (IgG and IgM IC/apoB)
MAb LF3 (or F
(ab)2
fragments of LF3) specific for mouse apoB-1009 was plated on microtiter
plates in order to capture mouse apoB-100, and mouse serum (diluted 1:100) was added. In
parallel plates, biotin-labeled mAb LF5 (5 µg/ml) was used to determine the relative amounts of
mouse apoB-100 bound by LF3. Goat-anti mouse IgG-Fc or IgM (alkaline phosphatase labelled)
were used to detect immune complexes to mouse apoB-100 lipoproteins. The relative number of
IgG or IgM immune complexes bound per apoB-100 particle was then determined by dividing
the bound IgG or IgM relative light units (RLUs,
chemiluminescence was read on a MLX
microtiter plate luminometer (Dynex Technologies, Chantilly, VA)) by the apoB-100 RLU in
parallel wells (IgG-IC/apoB-100, IgM-IC/apoB-100). Each result is the mean of triplicate values
per mouse.
Reporter Cell Assay for Determination of Aortic NO Production
Aortic NO production was bioassayed by determination of the cGMP content in RFL-6 rat lung
fibroblasts as reporter cells. Thoracic aortae of GPx-1-/-ApoE-/- and ApoE-/- mice after 6 weeks
on the WTD were dissected, cleaned of adherent connective tissue, and cut into 2 mm long
rings in ice-cold Hank's balanced salt solution (HBSS, Invitrogen, Carlsbad, CA). Confluent
RFL-6 cells grown in 12-well plates were washed twice with HBSS and then incubated in HBSS
containing 200 U/ml SOD, 10 µM (6R)-5,6,7,8-tetrahydro-L-biopterin (BH4), 1 mM L-arginine, 0.6
mM 3-isobutyl-1-methylxanthine (IBMX), and two aortic rings in each well for 15 min at 37°C.
After preincubation, 300 nM acetylcholine was added and incubation maintained for another 3
min. The reaction was stopped by aspiration of the solution, adding 300 µl of ice-cold 50 mM
sodium acetate, pH 4.0, and rapidly freezing the cells with liquid nitrogen. The cGMP content of
the RFL-6 samples was determined by a radioimmunoassay as described.10
Torzewski et al.
4
Results
-/-
-/-
Generation of GPx-1 ApoE Mice
PCR whole blood analysis confirmed the presence of mutant alleles for both Gpx-1 (509 bp) and
ApoE (245 bp), as well as the absence of their corresponding wild-type alleles (data not shown).
Both GPx-1-/-ApoE-/- and ApoE-/- mice were apparently healthy and fertile, and exhibited
mendelian transmission of the mutant alleles. We noticed no obvious abnormalities in organ
development, gestation time, food/water consumption as well as sleeping and cleaning habit.
Compared with ApoE-/- mice (309 ± 38.6 U/g Hb), GPx-1-/-ApoE-/- mice had no detectable red
blood cell GPx-1 activity.
Detection of Reactive Oxygen Species Formation in Aortic Vessel Segments, Heart
Mitochondria and Membrane Fractions
Basal vascular ROS production (including superoxide, hydrogen peroxide and peroxynitrite) in
the aorta was significantly increased in GPx-1–/–ApoE–/– mice compared to control ApoE-/- mice
(Figure III, left panel). NADPH-stimulated ROS production in membrane fractions was
significantly increased in GPx-1–/–ApoE–/– mice (Figure III, middle panel). Succinate-stimulated
mitochondrial ROS formation showed a significant increase in GPx-1–/–ApoE–/– mice, which was
also present upon blockade of mitochondrial respiration by myxothiazol (Figure III, right panel).
Succinate is a substrate of mitochondrial complex II and myxothiazol an uncoupler of complex
III. These results point towards increased oxidative stress in GPx-1–/–ApoE–/– mice
due to
activation of mitochondria and NADPH oxidases or due to impaired antioxidative defense
mechanism in the GPx-1 deficient animals.
Influence of GPx-1 Deficiency on Lipoprotein Oxidation
OxLDL could be detected both extra- and intracellularly in the atherosclerotic lesions of both
control and double-knockout mice using antibodies against 4-hydroxynonenal (HNE)-lysine
epitopes11 and oxidized phospholipid epitopes (EO6).12 The MAb EO6 resulted in a stronger and
Torzewski et al.
5
more extensive staining than the MAb NA59 but there were no significant differences between
control and double-knockout mice at 12 (data not shown) or 24 weeks (Figure IV) on the WTD.
IgG-IC/apoB-100, IgM-IC/apoB-100 and IgG and IgM OxLDL autoantibodies were not
statistically significant between groups after 12 or 24 weeks on the WTD. However, compared
to ApoE–/– mice, double-knockout mice (GPx-1–/–ApoE–/–) showed a trend towards higher titers of
IgG-IC/apoB-100 and IgM-IC/apoB-100 as well as IgG autoantibody titers against Cu-OxLDL
(data not reaching statistical significance due to high interindividual variations within the groups;
not shown).
Influence of GPx-1 Deficiency on Expression of MCP-1
To investigate whether increased recruitment of monocytes into the lesions might account for
the increased cellularity of early atherosclerotic lesions in GPx-1-/-ApoE-/- mice, we determined
MCP-1 protein expression in lesions of mice fed the WTD for 12 weeks. Since MCP-1 protein
expression was almost exclusively confined to macrophages, we did not determine percentpositive area for immunohistochemical staining but rather immunostaining intensity. We did not
find evidence for an increased staining intensity in macrophages of GPx-1-/-ApoE-/- mice
compared to control mice as quantified by Photoshop-based image analysis (supplemental
Figure V).
Vascular NO Production
To investigate whether endothelial dysfunction might be a mechanism operative during
accelerated atherosclerosis in GPx-1-/-ApoE-/- mice, aortic NO production was bioassayed with
RFL-6 rat lung fibroblasts as reporter cells. RFL-6 fibroblasts showed basal cGMP levels of
approximately 1 pmol/105 cells. In general, incubation of the RFL-6 cells with mouse aortic rings
in the presence of 300 nM acetylcholine markedly increased the cGMP content due to
acetylcholine-stimulated NO production of the aortic rings. However, aortic rings of GPx-1-/ApoE-/- mice were significantly less efficient in inducing cGMP production than aortic rings from
Torzewski et al.
6
ApoE-/- mice (Figure VI A). These results indicate that less bioactive NO is generated in
response to acetylcholine in aortae of double-knockout mice compared to control mice.
Protein Nitration
Nitration on the ortho position of tyrosine has been demonstrated to be a major product of
peroxynitrite attack on proteins.13,
14
Using immunohistochemistry, we detected a significant
increase in the amount of nitrotyrosine at 12 weeks on the WTD in double-knockout mice
compared to control mice (Figure VI B). Immunohistochemical staining was pronounced in and
around macrophage-derived foam cells. The observation of extensive protein nitration suggests
that peroxynitrite is produced and the significant increase in double-knockout mice suggests that
it may be involved in the different evolution of the lesions in control and double-knockout mice.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Ho YS, Magnenat JL, Bronson RT, Cao J, Gargano M, Sugawara M, Funk CD. Mice deficient in
cellular glutathione peroxidase develop normally and show no increased sensitivity to hyperoxia. J
Biol Chem. 1997;272:16644-16651.
Guo Z, Van Remmen H, Yang H, Chen X, Mele J, Vijg J, Epstein CJ, Ho YS, Richardson A.
Changes in expression of antioxidant enzymes affect cell-mediated LDL oxidation and oxidized
LDL-induced apoptosis in mouse aortic cells. Arterioscler Thromb Vasc Biol. 2001;21:1131-1138.
Krege JH, Hodgin JB, Hagaman JR, Smithies O. A noninvasive computerized tail-cuff system for
measuring blood pressure in mice. Hypertension. 1995;25:1111-1115.
Lehr HA, Sagban TA, Ihling C, Zahringer U, Hungerer KD, Blumrich M, Reifenberg K, Bhakdi S.
Immunopathogenesis of atherosclerosis: endotoxin accelerates atherosclerosis in rabbits on
hypercholesterolemic diet. Circulation. 2001;104:914-920.
Lehr HA, Mankoff DA, Corwin D, Santeusanio G, Gown AM. Application of photoshop-based
image analysis to quantification of hormone receptor expression in breast cancer. J Histochem
Cytochem. 1997;45:1559-1565.
Daiber A, Oelze M, Coldewey M, Bachschmid M, Wenzel P, Sydow K, Wendt M, Kleschyov AL,
Stalleicken D, Ullrich V, Mulsch A, Munzel T. Oxidative stress and mitochondrial aldehyde
dehydrogenase activity: a comparison of pentaerythritol tetranitrate with other organic nitrates. Mol
Pharmacol. 2004;66:1372-1382 Epub 2004 Aug 1326.
Daiber A, Oelze M, August M, Wendt M, Sydow K, Wieboldt H, Kleschyov AL, Munzel T. Detection
of superoxide and peroxynitrite in model systems and mitochondria by the luminol analogue L-012.
Free Radic Res. 2004;38:259-269.
Daiber A, August M, Baldus S, Wendt M, Oelze M, Sydow K, Kleschyov AL, Munzel T.
Measurement of NAD(P)H oxidase-derived superoxide with the luminol analogue L-012. Free
Radic Biol Med. 2004;36:101-111.
Zlot CH, Flynn LM, Veniant MM, Kim E, Raabe M, McCormick SP, Ambroziak P, McEvoy LM,
Young SG. Generation of monoclonal antibodies specific for mouse apolipoprotein B-100 in
apolipoprotein B-48-only mice. J Lipid Res. 1999;40:76-84.
Li H, Forstermann U. Structure-activity relationship of staurosporine analogs in regulating
expression of endothelial nitric-oxide synthase gene. Mol Pharmacol. 2000;57:427-435.
Torzewski et al.
7
11.
Palinski W, Yla-Herttuala S, Rosenfeld ME, Butler SW, Socher SA, Parthasarathy S, Curtiss LK,
Witztum JL. Antisera and monoclonal antibodies specific for epitopes generated during oxidative
modification of low density lipoprotein. Arteriosclerosis. 1990;10:325-335.
Horkko S, Bird DA, Miller E, Itabe H, Leitinger N, Subbanagounder G, Berliner JA, Friedman P,
Dennis EA, Curtiss LK, Palinski W, Witztum JL. Monoclonal autoantibodies specific for oxidized
phospholipids or oxidized phospholipid-protein adducts inhibit macrophage uptake of oxidized lowdensity lipoproteins. J Clin Invest. 1999;103:117-128.
Beckman JS, Ischiropoulos H, Zhu L, van der Woerd M, Smith C, Chen J, Harrison J, Martin JC,
Tsai M. Kinetics of superoxide dismutase- and iron-catalyzed nitration of phenolics by
peroxynitrite. Arch Biochem Biophys. 1992;298:438-445.
Ischiropoulos H, Zhu L, Chen J, Tsai M, Martin JC, Smith CD, Beckman JS. Peroxynitritemediated tyrosine nitration catalyzed by superoxide dismutase. Arch Biochem Biophys.
1992;298:431-437.
12.
13.
14.
Figure Legends
Figure I. Measurement of plaque size in longitudinal sections of the aortic arch stained with
trichrome. A 2-mm segment of the lesser curvature of the aortic arch was defined proximally by
a perpendicular axis dropped from the right side of the innominate artery origin (dashed line)
and the aortic-arch wall area subtended by this 2-mm stretch of intima (green line) was
calculated for each section of all mice by computerized image analysis. The aortic-arch intima
thickness (red line) was determined on this same segment of the lesser curvature. IA,
innominate artery; LCCA, left common carotid artery; LSA, left subclavian artery.
Figure II. GPx-1 deficiency does not affect serum lipids. A, Plasma lipid contents in GPx-1-/-/-
-/-
ApoE and ApoE mice at baseline and after 6, 12, and 24 weeks, respectively, on the WTD
(mean ± SD). B, Representative lipoprotein profiles after 12 weeks on the WTD. Pooled plasma
(n=6) was size-fractionated by FPLC, and the cholesterol content of each fraction was
determined by a colorimetric assay.
Figure III. Determination of vascular, mitochondrial and NADPH oxidase dependent ROS
formation. Left panel: Aortic rings (n=3-6) were placed in undiluted Diogenes™ reagent and
basal ROS formation was measured over 20 min. Middle panel: Heart membrane suspensions
(n=5, 0.2 mg/ml protein) in Diogenes™ reagent/PBS 1:4 were counted for 5 min upon
Torzewski et al.
8
stimulation with NADPH (200 µM). Right panel: Cardiac mitochondrial suspensions (n=4-8, 0.1
mg/ml protein) in Diogenes™ reagent/PBS 1:4 were counted for 5 min upon stimulation with
succinate (2.5 mM) and myxothiazol (10 µM). Data are expressed as counts/min and presented
as boxplots with median, interquartile range, minimum and maximum. * p<0.05 vs. ApoE-/- mice,
#
p<0.05 vs. GPx-1-/-ApoE-/- mice and $ p<0.05 vs. ApoE-/- + myxothiazol group.
Figure IV. Influence of GPx-1 deficiency on lipoprotein oxidation. After consuming the WTD for
-/-
24 weeks, atherosclerotic lesions of the inner aortic arch intima (lesser curvature) of GPx-1
-/-
-/-
ApoE and ApoE mice were stained with MAb anti-oxidized phospholipid epitopes (EO6, 1:500
(A)), and MAb anti-4-hydroxononenal lysine (NA59, 1:100 (B)). Immunostaining of murine
tissues with the murine MAbs was performed using the Vector M.O.M. immunodetection kit.
Percent-positive area for immunohistochemical staining (brown color deposits) was quantified
by Photoshop-based image analysis. The lumen is to the upper left-hand corner. The
demarcation beween intima and media is indicated by an arrowhead. Magnification x64.
Data are presented as boxplots with median, interquartile range, minimum and maximum.
Figure V. Influence of GPx-1 deficiency on MCP-1 protein expression. After consuming the
WTD for 12 weeks, atherosclerotic lesions of the inner aortic arch intima (lesser curvature) of
GPx-1-/-ApoE-/- and
-/-
ApoE
mice
were
stained
with
rabbit
anti-murine
JE
(MCP-1
(macrophage/monocyte chemotactic protein-1), 1:100). Immunostaining intensity was then
quantified by Photoshop-based image analysis and calculated as the difference between MCP-1
immunostaining and background immunostaining (designated immunohistochemical index with
arbitrary units (AU)). The lumen is to the upper left-hand corner. The demarcation beween
intima and media is indicated by an arrowhead. Magnification x64.
Torzewski et al.
9
Figure VI. Vascular NO production and protein nitration. A, After 6 weeks on the WTD,
-/-/acetylcholine stimulated NO production of isolated thoracic aortic rings from GPx-1 ApoE and
ApoE-/- mice was bioassayed with RFL-6 rat lung fibroblasts as reporter cells. The cGMP
content of the RFL-6 samples was determined by radioimmunoassay. B, After consuming the
WTD for 12 weeks, atherosclerotic lesions of the inner aortic arch intima (lesser curvature) of
GPx-1-/-ApoE-/- and ApoE-/- mice were stained with a polyclonal goat anti-nitrotyrosine antibody
using the Vector ABC immunodetection kit. Percent-positive area for immunohistochemical
staining (brown color deposits) was quantified by Photoshop-based image analysis. The lumen
is to the upper left-hand corner. The demarcation beween intima and media is indicated by an
arrowhead. Magnification x64.
Torzewski et al.
10
IA
LCCA
LSA
Figure I
A
Cholesterol, mg/dl
GPx-1-/-ApoE-/ApoE-/-
Triglycerides, mg/dl
GPx-1-/-ApoE-/ApoE-/-
baseline
361 ± 75
383 ± 51
158 ± 36
148 ± 30
6 weeks
1537 ± 261
1632 ± 167
161 ± 35
146 ± 42
12 weeks
1497 ± 186
1535 ± 360
170 ± 69
150 ± 39
24 weeks
1338 ± 383
1546 ± 391
138 ± 68
133 ± 43
B
80
-/ApoE-/GPx-/ApoE
70
Cholesterol (mg/ml)
ApoE
-/- -/- ApoE-/GPx-1
VLDL
IDL
60
50
40
30
LDL
20
HDL
10
0
0
5
10
15
20
25
30
35
40
45
Fraction
Figure II
ApoE
1.000
1.000
DKO
chemiluminescence [counts/min]
20.000
20.000
15.000
15.000
*
ApoE
*
#$
800
800
40.000
40.000
600
600
30.000
30.000
400
400
20.000
20.000
200
200
10.000
10.000
00
00
ApoE-/ApoE-/-GPx-1-/-
ApoE
DKO
50.000
50.000
DKO
10.000
10.000
*
5.000
5.000
00
*
succinate
aortic rings
membrane fractions
myxothiazol
mitochondria
Figure III
ApoE-/-
A
80,0
80
9
n.s.
40,0
40
20,0
20
00,0
EO6
ApoE-/-GPx-1-/-
ApoE-/-
7
60,0
60
ApoE-/-GPx-1-/-
B
ApoE
DKO
100,0
100
ApoE
DKO
100,0
100
80,0
80
n.s.
60,0
60
40,0
40
20,0
20
00,0
NA59
Figure IV
immunohistochemical index [AU]
ApoE-/-
ApoE
DKO
5050
4040
n.s.
3030
2
2020
1
1010
00
MCP-1
ApoE-/-GPx-1-/-
Figure V
cGMP (pmol/105 RFL-6 cells
A
ApoE
DKO
10,0
10
88,0
p<0,05
66,0
44,0
22,0
5
00,0
NO production
ApoE-/-
ApoE-/-GPx-1-/-
B
ApoE
DKO
100,0
100
80,0
80
60,0
60
40,0
40
p<0,05
20,0
20
00,0
nitrotyrosine
Figure VI

as a PDF

Transcrição

Documentos relacionados

EFFECT OF PROBIOTICS IN PREVENTION OF INFECTION: FROM

Expert Reader 1000 Chapter 12

Paresis of peristalsis and ileus lead to death in lactating mice

participation of p2x7 purinergic receptors in the hemorrhagic cystitis

CAFFEINE PREVENTS SCOPOLAMINE-INDUCED

Regulation of body temperature and brown adipose

Antidiabetic Activity of Green Tea - J

Pubertal high fat diet: effects on mammary cancer

Cancer microbiota

Endocrine and Gastrointestinal