| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Atherosclerosis and Lipoproteins |
From the Center for Cardiovascular Research (B.M.P, J.R., T.S.L., A.J.M.), Division of Endocrinology, Metabolism, and Lipid Research (Z.C., J.G.S., T.C., C.F.S.), Department of Cell Biology and Physiology (B.M.P., A.J.M.); Washington University School of Medicine, St. Louis, Mo.
Correspondence to Dr. Anthony J. Muslin, Center for Cardiovascular Research, Washington University School of Medicine, 660 South Euclid Avenue, Box 8086, St. Louis, MO 63110. E-mail amuslin{at}imgate.wustl.edu
| Abstract |
|---|
|
|
|---|
Methods and Results— Grb2+/– apoE–/– and apoE–/– mice exhibited similar body weight and serum lipid profiles. However, Grb2+/– apoE–/– mice on a Western diet had reduced lesion formation compared with apoE–/– mice by aortic sinus and en face assays. Transplantation of apoE–/– mice with Grb2+/– apoE–/– or apoE–/– bone marrow indicated that Grb2 haploinsufficiency in blood-borne cells confers resistance to Western diet–induced atherosclerosis. Cell culture experiments with bone marrow–derived macrophages showed that Grb2 is required for oxidized low density lipoprotein (oxLDL)-induced MAPK activation and foam cell formation.
Conclusions— Grb2 is required for atherosclerotic lesion formation and uptake of oxidized LDL by macrophages.
We tested the role of Grb2 in atherosclerosis using Grb2+/– apoE–/– mice. Grb2+/– apoE–/– mice were resistant to atherosclerosis, and Grb2 haploinsufficiency in blood-borne cells conferred resistance to lesion formation. Also, Grb2 was required for foam cell formation. These results implicate a fundamental role for Grb2 in atherosclerotic lesion formation.
Key Words: atherosclerosis Grb2 MAPK oxidized LDL macrophage
| Introduction |
|---|
|
|
|---|
Previous work in vitro demonstrated a requirement for various MAPK signaling cascades in macrophage proliferation and lipid uptake. In particular, Senokuchi demonstrated that oxLDL-induced granulocyte macrophage colony stimulating factor (GM-CSF) production in cultured macrophages is dependent on the activation of extracellular regulated kinase (ERK) MAPK, and that GM-CSF–induced macrophage proliferation is dependent on p38 MAPK activity.5 In addition, Zhao showed that p38 activity is essential for oxLDL-induced foam cell formation in vitro.6 In another study, Senokuchi showed that statins reduce oxLDL-induced macrophage proliferation in culture by inactivating the small GTPases ras and rho, and also by inactivating p38 MAPK.7 Also, recent work demonstrated that c-jun kinase 2 (JNK2) activity in monocytes and macrophages is important for atherosclerotic lesion formation in mice.8 JNK2–/– apoE–/– mice exhibited reduced atherosclerotic lesion formation when compared with apoE–/– mice.8 Cultured murine macrophages lacking JNK2 exhibited reduced acetylated LDL uptake.8
Growth factor receptor-bound protein 2 (Grb2) is a ubiquitously expressed intracellular scaffolding protein that mediates the activation of the small GTPase ras and MAPK cascades by various growth factor, cytokine, and ligand receptors. Mice that are null for the Grb2 gene do not survive embryonic development and die at embryonic day 5 because of deficient endoderm differentiation.9 Grb2 haploinsufficient mice survive embryonic development and adults are fertile. Signal transduction in T lymphocytes from Grb2+/– mice is abnormal in response to activating stimuli.10 Indeed, p38 MAPK and JNK activation is almost completely blocked, but ERK activation is intact, in activated Grb2+/– T cells. Despite these defects in intracellular T cell signaling, the biological function of circulating T cells is apparently normal.10
We previously investigated the ability of Grb2+/– mice to respond to pressure overload by transverse aortic constriction (TAC).11 Cardiac p38 MAPK and JNK activation was inhibited in Grb2+/– mice after TAC, and the animals were resistant to the development of cardiac hypertrophy. We also evaluated the ability of Grb2+/– mice to develop neointima in response to carotid injury.12 Grb2+/– mice were resistant to neointima formation in response to mechanical injury of the carotid artery. After injury, carotid p38, ERK, and JNK activity was reduced in Grb2+/– mice compared with wild-type mice. Also, Grb2+/– aortic smooth muscle cells were resistant to platelet-derived growth factor–induced proliferation.12
Because of the abnormal signaling properties observed in many cell types in Grb2+/– mice, we investigated the ability of these mice to develop atherosclerosis in the apoE–/– model system.13
| Methods |
|---|
|
|
|---|
High-Fat Western Diet
Beginning at 8 weeks of age, Grb2+/– apoE–/– and apoE–/– littermates were administered a high-fat Western diet, ad libitum, for 2 months as previously described.14–16 The Western diet contained 0.15% cholesterol and provided 42% of calories as fat (Harlan Teklad).
For descriptions of other methods used in this study please see the online data supplement, available at http://atvb.ahajournals.org.
| Results |
|---|
|
|
|---|
|
Analysis of Atherosclerosis in Grb2+/– ApoE–/– and ApoE–/– Mice
After 2 months of high-fat, Western diet, aortic sinus and en face assays were performed to quantify lesion formation in the aortic sinus and aortae, respectively, of apoE–/– and Grb2+/– apoE–/– mice. Via aortic sinus assays, prominent atherosclerotic lesion formation was detected for apoE–/– mice (Figure 2A). In contrast, atherosclerotic lesions were markedly reduced in size and extent in Grb2+/– apoE–/– mice after 2 months of high-fat feeding. Quantification of lesion area demonstrated that apoE–/– mice had markedly increased lesion area within the aortic sinus, compared with Grb2+/– apoE–/– mice. Specifically, the mean lesion area was 0.263±0.097 mm2 in apoE–/– mice, but was only 0.151±0.075 mm2 in Grb2+/– apoE–/– mice (P=0.003, Figure 2B). Immunohistochemistry indicated similar cellularity of apoE–/– and Grb2+/– apoE–/– aortic sinus lesions. Specifically, oil red O–positive lesions from apoE–/– and Grb2+/– apoE–/– mice demonstrated comparable staining for macrophages and CD4+ T cells (supplemental Figure I).
|
In en face assays, multiple atherosclerotic lesions were detected in the aortic arches and thoracic aortae of apoE–/– mice. In contrast, few atherosclerotic lesions were observed in the aortae of Grb2+/– apoE–/– mice. Quantification of aortic lesion area revealed that the mean decimal fraction of total aortic area covered by atherosclerotic lesion was significantly reduced in Grb2+/– apoE–/– mice when compared with apoE–/– mice. For the aortic arch, the mean decimal fraction of total area covered by lesion for apoE–/– mice was 0.088±0.048 (8.8±4.8%), but was only 0.035±0.028 (3.5±2.8%) for Grb2+/– apoE–/– mice (P=0.003, Figure 2C). A similar reduction in lesion area was observed in the thoracic aortae of Grb2+/– apoE–/– mice (P=0.002, Figure 2C).
Male and female mice were used in the aortic sinus and en face assays. However, the sexes in each group showed comparable lesion formation. Specifically, at all anatomic sites studied, lesion area was comparable and did not statistically differ between male apoE–/– and female apoE–/– mice (supplemental Figures II and III). Likewise, at all anatomical sites studied, lesion area was comparable and did not statistically differ between male Grb2+/– apoE–/– and female Grb2+/– apoE–/– mice (supplemental Figures II and III).
Bone Marrow Transplantation Experiments
Grb2+/– apoE–/– mice exhibited markedly reduced atherosclerotic lesion area, compared with apoE–/– mice. Reduced lesion formation in Grb2+/– apoE–/– mice may be caused by altered physiological properties of Grb2+/– apoE–/– macrophages or other blood-borne cells. To test this possibility, bone marrow transplantation experiments were performed.
In these studies, lethally-irradiated 6-week-old apoE–/– mice were transplanted with bone marrow from Grb2+/– apoE–/– or apoE–/– donor mice. After 4 weeks of recovery, transplanted mice were placed on a high-fat Western diet. Purification of blood-leukocyte genomic DNA and subsequent polymerase chain reaction (PCR) amplification of the Grb2 null allele indicated highly efficient engraftment of Grb2+/– apoE–/– bone marrow (supplemental Figure IV). In addition, analysis of body weights and total sera lipid concentrations demonstrated that apoE–/– mice transplanted with apoE–/– or Grb2+/– apoE–/– bone marrow exhibited nearly identical body weights and lipid abnormalities at baseline and after Western diet (supplemental Figure V). FPLC analysis demonstrated similar distribution of cholesterol among VLDL, LDL, and HDL for both groups of animals (data not shown).
Analysis of atherosclerotic lesion formation by aortic sinus assays, after 12 weeks of Western diet, revealed that apoE–/– mice transplanted with Grb2+/– apoE–/– bone marrow exhibited significantly less atherosclerotic lesion area compared with apoE–/– mice transplanted with apoE–/– marrow (Figure 3A). Specifically, the mean lesion area was 0.484±0.071 mm2 in mice transplanted with apoE–/– bone marrow, but was only 0.316±0.102 mm2 in mice transplanted with Grb2+/– apoE–/– bone marrow (P=0.004, Figure 3B).
|
Similarly, as shown by en face assays conducted after 8 weeks of Western diet, apoE–/– mice transplanted with Grb2+/– apoE–/– bone marrow were resistant to the development of aortic lesions, compared with apoE–/– mice transplanted with apoE–/– marrow (Figure 3C). Indeed, in the aortic arch, the mean decimal fraction of total area covered by lesion for apoE–/– mice transplanted with apoE–/– marrow was 0.036±0.006 (3.6±0.6%), but only 0.010±0.002 (1.0±0.2%) for apoE–/– mice transplanted with Grb2+/– apoE–/– marrow (P=0.004, Figure 3C). With Grb2+/– apoE–/– donor bone marrow, a similar reduction in percent lesion area was observed in the thoracic and abdominal aortae (P=0.034 and 0.007, respectively; Figure 3C). For the bone marrow transplantation studies, both male and female recipient mice were used. However, the sexes in each group showed comparable lesion formation (supplemental Figures VI and VII).
Macrophage Function of Grb2+/– ApoE–/– Mice
Bone marrow transplantation experiments suggested that reduced lesion formation in Grb2+/– apoE–/– mice might be a consequence of the altered physiological properties of macrophages. To evaluate this possibility, bone marrow–derived macrophages (BMMs) were cultured from Grb2+/– apoE–/– and apoE–/– mice.
Production of BMMs requires the in vitro differentiation of macrophages from bone marrow precursor cells. To confirm the purity of this cell culture system and to ascertain the maturity of apoE–/– and Grb2+/– apoE–/– BMMs, fluorescence activated cell sorting (FACS) was performed with a monoclonal antibody directed against the F4/80 cell surface marker. F4/80 is expressed by mature murine macrophages, but not by other hematopoietic or unrelated cells.17 F4/80 expression progressively increases during macrophage differentiation.17 ApoE–/– and Grb2+/– apoE–/– BMMs displayed similar staining profiles for F4/80 by FACS. In both groups,
90% of cells were positive for F4/80. In addition, the peak and distribution of fluorescence signal was nearly identical for apoE–/– and Grb2+/– apoE–/– BMMs (data not shown).
As predicted, Western blot analysis of BMMs demonstrated that Grb2+/– apoE–/– macrophages had approximately 50% of Grb2 protein compared with apoE–/– macrophages (P=0.008, supplemental Figure VIII). Cultured BMMs were treated with radio-labeled oxLDL, acetylated LDL (acLDL), or LDL and uptake was measured. Grb2+/– apoE–/– BMMs exhibited markedly reduced oxLDL uptake when compared with apoE–/– BMMs (P=0.020, Figure 4). In addition, Grb2+/– apoE–/– BMMs exhibited markedly reduced degradation of oxLDL (P=0.010, Figure 4). However, Grb2+/– apoE–/– and apoE–/– BMMs exhibited similar cell surface binding of oxLDL (Figure 4).
|
Cultured macrophages take up oxLDL, at least in part, via CD36-mediated endocytosis.18 Western blot analysis with total BMM lysates indicated that apoE–/– and Grb2+/– apoE–/– BMMs express comparable amounts of CD36 (data not shown). In addition, FACS analysis indicated that apoE–/– and Grb2+/– apoE–/– BMMs expressed similar amounts of CD36 at the cell surface (supplemental Figure IX). Therefore, Grb2 haploinsufficiency limits uptake of oxLDL, but does not alter cell surface expression of an important oxLDL receptor, CD36, or cell surface binding of oxLDL.
Next, we analyzed the uptake and degradation of acLDL and LDL by apoE–/– and Grb2+/– apoE–/– BMMs. In contrast to oxLDL, the uptake and degradation of both acLDL and LDL was nearly identical in apoE–/– and Grb2+/– apoE–/– BMMs (Figure 4 and data not shown).
Lipid accumulation in BMMs treated with oxLDL was also determined by microscopic analysis of BMMs stained with oil red O. After incubation with oxLDL for 24 hours, BMMs from Grb2+/– apoE–/– mice showed much less lipid accumulation than BMMs from apoE–/– mice (Figure 5). Specifically, oil red O staining was substantially reduced for Grb2+/– apoE–/– BMMs compared with apoE–/– BMMs.
|
The signal transduction properties of cultured BMMs were also examined. Cultured BMMs from Grb2+/– apoE–/– and apoE–/–mice were treated with unlabeled oxLDL for 24 hours and BMM lysates were generated. Analysis of lysates by anti–phospho-JNK and anti–phospho-p38 MAPK immunoblotting revealed that Grb2+/– apoE–/– BMMs had reduced activation of JNK and p38 MAPK in response to oxLDL (Figure 6).
|
In addition to defective oxLDL uptake, impaired lesion formation in Grb2+/– apoE–/– mice could also be a result of reduced monocyte/macrophage proliferation or increased macrophage apoptosis.19,20 BMM proliferation was assessed in response to M-CSF stimulation. In response to 24 hours of stimulation with 5 or 50 ng/mL M-CSF, apoE–/– and Grb2+/– apoE–/– BMM cell numbers increased to a similar extent (data not shown). To measure apoptosis in BMMs, staurosporine (STS)-induced apoptosis was examined. STS and the oxysterol components of oxLDL induce apoptosis of macrophages via the mitochondrial death pathway.21,22 STS induced similar apoptosis of apoE–/– and Grb2+/– apoE–/– BMMs after 3 and 6 hours of treatment (data not shown).
| Discussion |
|---|
|
|
|---|
In this work, the role of Grb2 in the pathogenesis of murine atherosclerosis was examined. Grb2 links both receptor and nonreceptor tyrosine kinases, and integrins to ras activation and subsequent activation of MAPK pathways; including JNK MAPK, p38 MAPK, and ERK MAPK. Mice that are haploinsufficient for Grb2, in general, display reduced JNK and p38 MAPK signaling in various tissues.10–12 Moreover, Grb2–/– embryonic fibroblasts exhibit markedly reduced ERK activation.9 In this study, Grb2 haploinsufficient mice in the apoE null model were shown to be resistant to high-fat feeding–induced atherosclerosis.
Reduced atherosclerotic lesion formation in Grb2+/– apoE–/– mice could be a result of defective signaling in smooth muscle cells, macrophages, endothelial cells, lymphocytes, or other cell types. In previous work, we demonstrated that Grb2 is required for growth factor–stimulated vascular smooth muscle cell (VSMC) proliferation in culture.12 Furthermore, we showed that Grb2+/– mice are resistant to carotid injury-induced neointima formation.12 Smooth muscle cells are found in atherosclerotic plaques, and smooth muscle cell proliferation is thought to play an important role in the growth of lesions.23 Furthermore, smooth muscle cell apoptosis may play a critical role in the development of plaque instability that can lead to arterial thrombosis.24 If defective smooth muscle cell proliferation was responsible for the phenotype observed in Grb2+/– apoE–/– mice, then transplantation of bone marrow from Grb2+/– apoE–/– animals into apoE–/– would not be expected to confer resistance to atherosclerosis. However, our results suggest that macrophage-specific Grb2, but not smooth muscle cell–specific Grb2, is critical for atherosclerotic lesion formation.
JNK2 knockout mice have altered macrophage function with reduced ability to form foam cells,8 and given the important role of Grb2 in JNK activation, the function of Grb2+/– apoE–/– macrophages was examined. Cultured Grb2+/– apoE–/– macrophages exhibited normal oxLDL binding but reduced oxLDL uptake and degradation when compared with apoE–/– macrophages. However, Grb2+/– apoE–/– macrophages displayed normal acLDL and LDL uptake and degradation.
In contrast to Grb2+/– apoE–/– macrophages, JNK2 null macrophages exhibited defective acLDL uptake in previous work, perhaps because of reduced scavenger receptor A (SR-A) activation and function.8,18 Despite reduced JNK activity, acLDL uptake was not decreased in Grb2+/– apoE–/– BMMs. One explanation for this apparent discrepancy is that loss of JNK2 protein in JNK2–/– apoE–/– macrophages results in macrophage developmental abnormalities that alter SR-A function. Another explanation is that JNK2 has a kinase-independent function in the regulation of acLDL uptake, for example, by functioning as a scaffolding protein.
Previous studies with SR-A–/–, CD36–/–, and SR-A–/– CD36–/– macrophages, demonstrated that SR-A is important for a fraction of oxLDL binding and uptake. In contrast, CD36 is required for the majority of oxLDL binding and uptake.18 In the current work, Grb2+/– apoE–/– macrophages displayed normal oxLDL binding and cell surface expression of CD36. However, Grb2+/– apoE–/– macrophages exhibited reduced uptake of oxLDL compared with apoE–/– macrophages. The number of CD36 molecules on the cell surface of Grb2+/– apoE–/– macrophages was unaltered. This data suggests that oxLDL-induced endocytosis of CD36 in Grb2+/– apoE–/– macrophages is abnormal. Little is known about the cellular mechanism(s) governing the endocytosis of CD36 upon oxLDL binding. In one study, oxLDL binding resulted in CD36 endocytosis via a clathrin- and caveolin-1–independent mechanism.25 However, oxLDL binds to additional scavenger receptors on macrophages in vivo such as SR-A, SR-B1, LOX-1, PSOX, and MARCO.26,27 Grb2 is known to bind to dynamin, a GTPase essential for types of clathrin- and caveolin-1–independent endocytosis.28,29 It is possible that Grb2-mediated recruitment of dynamin to oxLDL-bound CD36, or other scavenger receptor(s), facilitates the endocytosis of oxLDL.
It is important to note that the specific contribution of SR-A and CD36 to atherosclerotic lesion formation has become controversial because of the recent work of Freemans group.30 Previous work by Kodamas group demonstrated that apoE–/– mice with targeted deletion of SR-A were resistant to lesion formation when compared with apoE–/– mice.31 Moreover, a study by Lintons group that involved the use of bone marrow transplantation, demonstrated that macrophage- or other blood-borne cell-specific SR-A was important for lesion formation in LDLR–/– mice.32 Furthermore, Silversteins group demonstrated that CD36–/– apoE–/– mice exhibit reduced lesion formation compared with apoE–/– mice.33 Bone marrow transplantation studies by Silverstein demonstrated a requirement for macrophage- or other blood-borne cell-specific CD36 for atherosclerotic lesion formation in apoE–/– mice.34 In contrast to the work of Kodama, Linton, and Silverstein, the recent study by Freeman suggests that apoE–/– mice with deletion of SR-A or CD36 are not resistant to atherosclerosis compared with apoE–/– mice.30 However, in the setting of hypercholesterolemia, there may be scavenger receptors, in addition to SR-A and CD36, that contribute to atherosclerotic lesion formation.26,27,35 Furthermore, in cell culture experiments, macrophages from SR-A–/– CD36–/– double knockout mice exhibit a larger decrease in uptake of modified LDLs than macrophages from SR-A–/– or CD36–/– single knockout mice.18 Therefore, future in vivo atherosclerosis experiments with SR-A–/– CD36–/– apoE–/– triple knockout mice will help to clarify the roles of SR-A and CD36 in atherosclerotic lesion formation.35
In summary, Grb2 is required for atherosclerotic lesion formation in mice and for oxLDL uptake by macrophages. The precise role of CD36 or other scavenger receptors in Grb2-facilitated oxLDL uptake by macrophages remains to be determined.
| Acknowledgments |
|---|
This work was supported by grants from the National Institutes of Health awarded to A.J.M. (HL61567, HL57278), the Burroughs Welcome Fund (A.J.M.), and a grant from the National Institutes of Health awarded to Z.C. (RO1HL73939). J.R. is a recipient of an American Heart Association Heartland Affiliate postdoctoral fellowship.
Disclosures
None.
| Footnotes |
|---|
Original received October 13, 2006; final version accepted February 21, 2007.
| References |
|---|
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2007 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |