Genetic Background Selectively Influences Innominate Artery Atherosclerosis
Immune System Deficiency as a Probe
Objective— We sought to examine whether there is a site-specific effect on atherosclerosis of the absence of mature T and B cells caused by a recombination activating-gene deficiency in LDL receptor-deficient mice and whether this effect is influence by the extent of backcrossing to C57BL/6 mice.
Methods and Results— Male mice were fed atherogenic diets for 3 months. In strain 1 mice, in which ≈93% of the genes were from C57BL/6 mice, the absence of mature T and B cells led to a significant reduction in atherosclerosis in both the aortic sinus and the innominate artery. In strain 2 mice, in which ≈99+% of the genes were from C57BL/6 mice, immune system deficiency led to a site-specific effect on atherosclerosis, with a reduction in atherosclerosis in the aortic sinus but not in the innominate artery, similar to previous results obtained with apolipoprotein E−/− mice. All of the immune system-incompetent mice had lower plasma total and VLDL cholesterol levels regardless of strain or diet, indicating that differences in lipid levels were unlikely to be responsible for these site-specific effects of immune system deficiency.
Conclusions— These results suggest that immune system deficiency has a site-specific effect on atherosclerosis that is sensitive to the genetic background of the mice.
Atherosclerosis is a chronic, inflammatory response to hyperlipidemia.1 As with other inflammation, components of both innate and adaptive immunity influence the progression of this inflammation. The cells that have received the most attention are the innate responders—monocytes/macrophages—and the T and B cells of the adaptive immunity system. These cells participate either directly through their primary immune functions or indirectly, when mediated by the cytokines that they produce. The involvement of the adaptive immune system has been implicated by 2 kinds of studies in mice. First, the use of immune system-deficient models involving the absence of mature T and B cells (ie, recombination activating-gene deficiency [RAG−/−]), coupled with either apolipoprotein E deficiency (apoE−/−) or LDL receptor deficiency (LDLR−/−), results in a reduction of aortic sinus atherosclerosis,2–5⇓⇓⇓ implying that the net effect of the immune system through the course of atherogenesis is to promote vascular lesions. Consistent with this conclusion are the results from studies involving deficiency or overexpression of some cytokines secreted by T cells.6–9⇓⇓⇓ In addition, recent data have strongly suggested that B cells are protective against the development of atherosclerosis.10,11⇓ Second, and related to the latter finding, is the observation that immunization of LDLR−/− mice with oxidized LDL and even with native LDL reduces atherosclerosis.12
The effect of the immune system on atherosclerosis can be complex. In a previous study, we observed a site-specific effect of immune system deficiency on atherosclerosis. In that study, male apoE−/− RAG2−/− mice had markedly reduced aortic sinus atherosclerosis but not brachiocephalic (innominate) artery atherosclerosis compared with immune system-competent apoE−/− mice.2 In the current study with the LDLR−/− mouse model, we asked whether this site-specific effect was observed only when apoE was deficient. In the course of these investigations, it became evident that the response of innominate artery atherosclerosis to immune system deficiency was quite sensitive to the relative purity of the genetic background. We have studied 2 strains, 1 that is ≈93% C57BL/6 (6 generations backcrossed to C57BL/6 mice) and 1 that is ≈99% C57BL/6 (10 generations backcrossed), with qualitatively different responses in the innominate artery but not in the aortic sinus. Because both strains are immune system deficient, our results are best explained by the presence of 1 or more genes that have a bias toward modifying the response of innominate artery atherosclerosis to immune system deficiency.
Mice and Diets
LDLR−/− mice backcrossed with C57BL/6 mice for 6 generations (Jackson Laboratories, Bar Harbor, Me) and RAG2−/− mice backcrossed with C57BL/6 mice for 4 generations (obtained from F.W. Alt)13 were crossed to generate LDLR−/− RAG2−/− mice that were deficient for both genes. On the basis of the extent of backcrossing, we estimated that ≈93% of the genes were from C57BL/6 mice.
LDLR−/− mice and RAG1−/− mice14 backcrossed for 10 generations to C57BL/6 mice (both from Jackson Laboratories, Bar Harbor, Me) were crossed to generate LDLR−/− RAG1−/− (≈99% of genes were from C57BL/6 mice). All of the doubly deficient mice had fasting plasma cholesterol levels (while being fed a chow diet) >140 mg/dL and lacked immunoglobulins, as determined by Western blotting. The mice were housed in a specific pathogen-free environment in a temperature-controlled room with a 12-hour light/dark cycle. Male mice (12 to 17 per group) were maintained on a standard chow diet (Purina Mills) until placed on a milk fat-based diet, consisting of 0.25% cholesterol and 18% milk fat (Harlan Teklad TD97222), or a Western-type diet, consisting of 0.15% cholesterol and 21% fat (Harlan Teklad TD88137), at 8 weeks of age. The mice were bled monthly by retro-orbital sinus puncture for determination of 4-hour fasting plasma cholesterol and triglyceride levels (Roche Molecular Biochemicals).2 Lipoproteins were separated by fast protein liquid chromatography on tandem Superose 6 columns, and the amount of cholesterol in the even-numbered fractions was determined and expressed as micrograms cholesterol per milliliter of plasma. The area under each lipoprotein peak was quantified by computer digitizer (SigmaScan) and expressed as percent total cholesterol. Lipoprotein cholesterol levels were determined from the total plasma cholesterol and the percentage of total cholesterol in each of the lipoprotein peaks. All procedures performed on the mice were in accordance with National Institutes of Health and institutional guidelines regarding the care and use of laboratory animals.
Analysis of Atherosclerotic Lesions
Atherosclerosis was examined after the mice had been on their assigned diet for 3 months. The mice were anesthetized by inhalation of methoxyflurane or intraperitoneal injection of 80 mg/kg ketamine and 10 mg/kg xylazine, exsanguinated, and perfused with paraformaldehyde, and aortic tissue was obtained as described previously.2 The frozen tissue was serially sectioned into 10-μm sections from the innominate artery through the aortic sinus. Every tenth section was stained with oil red O and Harris hematoxylin and counterstained with fast green. Lesions in the innominate artery were quantitated from 4 oil red O-stained sections, each separated by 100 μm, and located between 150 and 450 μm distal to the branch point of the innominate artery from the aortic arch. Aortic sinus lesions were evaluated from 3 sections, each separated by 100 μm, beginning at the site of appearance of the coronary artery. Atherosclerosis was quantitated from digitally captured images and OpenLab Software, version 1.7.6. The amount of lesion in the innominate artery was expressed as a percentage of total lumen area,15 although the conclusions were not different when total lesion areas were compared.
Results are expressed as mean±SEM. Statistical analysis was performed with StatView 5.0.1 software. The data were analyzed by 1-way ANOVA. Nonparametric analysis of atherosclerotic lesions was analyzed by the Mann-Whitney test. The significance level was set at P<0.05.
Our preliminary goal was to assess whether our previously reported site-selective effect of immune system deficiency on atherosclerosis in the apoE−/− background was also to be found in the presence of apoE, as exemplified by the LDLR−/− model. Our initial set of experiments was performed with animals that had been backcrossed into the C57BL/6 background through 6 generations (strain 1). The results obtained with these mice were in contrast to those we obtained in the apoE−/− model.2 To determine whether the differences in immune system deficiency in the 2 atherosclerotic models could be explained by background modulating genes, we developed a second strain backcrossed to C57BL/6 mice through 10 generations (strain 2). It is the difference between these 2 strains that is the basis of this article. Strain 1 was the result of a cross of the LDLR−/− mouse with the RAG2−/− mouse, whereas the cross to produce strain 2 used the RAG1−/− mouse, largely because at the time the cross was made, the RAG1−/− mouse was available in the C57BL/6 background. There is no reason to believe that RAG2−/− and RAG1−/− mice are phenotypically different, even at a refined cellular and biochemical level.16 RAG1 and RAG2 are contiguous genes expressed primarily in lymphoid tissues, and their products function as a heterodimer that is absolutely required for the recombination of immunoglobulin and T-cell receptor genes. RAG1 or RAG2 gene knockouts produce animals lacking mature B and T cells,13,14⇓ and no differences in immunological phenotype of these mice have been noted. The LDLR−/− RAG−/− mice are healthy and viable when housed in a specific pathogen-free facility. There was no apparent difference in fertility, infection rate, or mortality in either immune system-deficient strain compared with the immune system-competent LDLR−/− mice. Although there was no difference in body weight of the strain 1 mice (28.7±1.1 g and 30.9±1.4 g for LDLR−/− RAG2−/− and LDLR−/− mice, respectively) after 3 months on the milk fat diet, the body weights of strain 2 LDLR−/− RAG1−/− mice fed the milk fat diet were slightly lower than those of the LDLR−/− mice (30.0±0.9 g vs 34.0±1.2 g, respectively; P=0.034) and significantly lower in the LDLR−/− RAG1−/− mice fed the Western-type diet (29.6±1.2 g vs 37.3±1.4 gm, respectively; P<0.0001). The reason for the difference in weight, especially in mice fed the Western-type diet, is unclear. The mice did not appear ill, and there was no systematic increase in the acute-phase protein serum amyloid A in the plasma of the immune system-incompetent mice.
RAG Deficiency Lowers Plasma Lipid and VLDL Levels in Both Strains
Male mice were maintained on a low-fat chow diet until 8 weeks of age. On the chow diet, the primary lipoprotein in the plasma is LDL (data not shown). In strain 1 (≈93% C57BL/6), plasma cholesterol levels on the chow diet were higher in LDLR−/− mice than in the LDLR−/− RAG2−/− mice (Table 1). However, plasma triglyceride levels were not statistically different. The mice were then switched to a diet containing 0.25% cholesterol and 18% milk fat. After 1 month on the high-fat diet, plasma lipid levels rose in both immune system-competent and -incompetent mice, and these levels were maintained for 3 months. The plasma lipid levels at 3 months are shown in Table 1. Even on the milk fat diet, plasma cholesterol levels were lower in the LDLR−/− RAG2−/− mice. In this case, however, plasma triglyceride levels were almost 3-fold lower in the RAG-deficient mice. This decrease in triglyceride levels is reflected in significantly lower VLDL levels (Table 1 and Figure 1). LDL and HDL levels were not significantly different.
Similar to strain 1, superposition of RAG deficiency in LDLR−/− mice fully backcrossed to C57BL/6 mice (strain 2) reduced plasma cholesterol levels in chow-fed male mice. However, in these ≈99% C57BL/6 mice, triglyceride levels were also reduced in chow-fed animals. Plasma cholesterol levels in immune system-competent strain 2 LDLR−/− mice while on the chow diet also had higher plasma cholesterol and triglyceride levels than strain 1 mice, but there was no difference in plasma lipid levels between the immune system-incompetent strain 1 and strain 2 LDLR−/− RAG−/− mice while on the chow diet. After 3 months on the high-fat diets, plasma cholesterol and triglyceride levels were lower in the LDLR−/− RAG−/− mice, although the difference was significantly different only in the mice fed the Western-type diet. Again, this trend was reflected in the lower VLDL levels. HDL levels were different in the strain 2 mice fed the Western-type diet. There was no significant difference in plasma cholesterol levels among all of the LDLR−/− mice, regardless of extent of backcrossing or type of high-fat/high-cholesterol diet. The same was true for the LDLR−/− RAG−/− mice.
Genetic Background Modifies the Site-Specific Effect of Immune System Deficiency
Atherosclerosis was quantified in the aortic sinus and innominate artery after 3 months on the high-fat diets (Fig. 2). In strain 1 mice (≈93% C57BL/6 background), immune deficiency resulted in a 65% to 70% decrease in atherosclerosis in both the aortic sinus and innominate artery. In contrast, immune deficiency in the strain 2 mice (≈99% C57BL/6) had a site-specific effect on atherosclerosis. In these mice, aortic sinus atherosclerosis was significantly lower in the immune system-deficient mice, but there was no significant difference in the extent of atherosclerosis in the innominate artery. The increase in size of innominate artery atherosclerosis in the milk fat-fed strain 2 mice did not reach statistical significance (P=0.39). Thus, despite similar differences in plasma cholesterol levels between immune system-competent and -incompetent mice within each group, immune system deficiency resulted in a site-specific effect on atherosclerosis in the strain 2 mice, with aortic sinus atherosclerosis being lower but innominate artery atherosclerosis unchanged in the immune system-incompetent mice. In strain 1 mice, immune system deficiency reduced atherosclerosis in both the aortic sinus and innominate artery. Although there is significant variability in the amount of atherosclerosis in each group of mice, this variability was observed between littermates in all of the strains and diet groups, not just between different litters. Therefore, the variability could not have been caused by seasonal differences or differences in lots of the diet.
The 2 major findings of this study are that 2 atherosclerosis-sensitive sites respond differently to global immune system deficiency owing to a lack of mature T and B cells and that the genetic background of the mouse influences this response. This suggests that there is an interaction between 1 or more genes that differentially influence the development of atherosclerosis at the innominate artery and aortic sinus and the immune system. Thus, immune system deficiency can be used as a probe to examine genes that selectively influence innominate artery atherosclerosis.
The current study confirms the site-selective effect of immune system deficiency on atherosclerosis previously reported in the apoE-deficient mouse model of atherosclerosis.2 The fact that similar data were obtained with both apoE−/− and LDLR−/− mice suggests that a precise lipoprotein profile and a lack of apoE expression in the vessel wall are not requirements for this site selectivity. Site-specific influences on atherogenesis have also been observed with probucol treatment in apoE−/− mice,17 with von Willebrand factor deficiency in apoE−/− and LDLR−/− mice,18 and in female LDLR−/− mice subjected to interleukin-4 deficiency,19 but none of these studies examined the innominate artery, a high-probability site for the development of atherosclerosis.
One of the significant observations reported here is that immune system deficiency influences innominate artery atherosclerosis differently, depending on the genetic background (strain 1 vs strain 2). It is well known that mouse strains differ profoundly in their susceptibility to atherosclerosis. The strains used here were all mixtures of the 129 strain and the C57BL/6 strain, and because strain 1 and strain 2 mice were backcrossed to different extents, these mice will have different numbers of residual 129 genes. Because these strains were independently derived, it is unclear whether the residual 129 genes in strain 2 would necessarily be included among the 129 genes in strain 1. A likely but not exclusive model to account for our results is that 1 or more of the129 genes interact with the immune system-deficient setting (eg, spectrum of cytokine profiles) to reduce innominate artery atherosclerosis. There is no significant effect of genetic background on aortic sinus atherosclerosis, suggesting that some of the genes that modulate the response of the innominate artery to immune deficiency are less important for the response of aortic sinus atherosclerosis. We have noted that immune system deficiency reduces atherosclerosis in the aortic sinus to a greater extent in strain 1 mice than strain 2 mice. Thus, we cannot exclude the possibility that a gene(s) in strain 1 mice that influences innominate artery atherosclerosis response to immune system deficiency is having at least a partial effect on the aortic sinus.
The genes that influence innominate artery atherosclerosis are not those that directly influence B-cell and T-cell function, because both strains 1 and 2 were similarly deficient in these functions. Rather, they are modifier genes that are manifested in the immune system-deficient setting. In this connection, immune system deficiency can be considered a biologic probe that uncovers the operation of the gene(s) that selectively influences innominate artery atherosclerosis. In studies that explored genetic-susceptibility loci in atherosclerosis of apoE−/− and LDLR−/− mice, aortic sinus lesion area has been used as the screening readout.20–25⇓⇓⇓⇓⇓ In LDLR−/− mice, loci on chromosomes 4 and 6 have been found that result in variations in aortic sinus lesion area.21 Candidate genes within these loci include phospholipase A2 and members of the tumor necrosis factor receptor superfamily. In the apoE-deficient background, loci were identified on chromosomes 10, 14, and 19 that account for variations in aortic sinus lesion area.22 The strongest candidate on the chromosome 10 interval is the interferon-γ receptor, whereas others include connective tissue growth factor gene, estrogen receptor-α, and tumor necrosis factor-induced protein 3. There is no overlap in the loci that modulate aortic sinus atherosclerosis in apoE-deficient and LDLR-deficient, hyperlipidemic mice. The genes that account for variations within the innominate artery might be different from those suggested in the screening for genetic variations that influence aortic sinus atherosclerosis.
Some sites in the vascular tree are more likely to develop atherosclerosis than others.26 These are determined in large part by the regional hemodynamics at each of these sites, with the high-probability sites being in areas of disturbed flow.26,27⇓ Endothelial cells are the major transducers of regional hemodynamic stress, and this is manifested in significant differences in gene expression in microenvironments of laminar or turbulent shear stress.28 It is very likely that disturbed flow, including reverse flow and hence, hemodynamic stresses, differs between the aortic sinus and the innominate artery, both being high-probability sites. From this, one might expect that the pattern of gene expression would differ between these sites, although this has yet to be categorically demonstrated. Our ongoing work will address this issue. Differences in endothelial cell gene expression could be manifested in the alteration of cell adhesion and transendothelial migration of cells and lipoproteins into the evolving lesion.1,29–31⇓⇓⇓ This or endothelial cell interaction with underlying intimal cells (macrophages, smooth muscle cells, and T cells) could well result in manifest changes in the rate of lesion evolution at these various sites and might account for our reported observations.
It is clear that immune system competence is not an absolute requirement for atherogenesis. The phenotype of global immune system deficiency is the result of removal of an immune component network involving, at a minimum, T helper (Th) 1 cells, Th2 cells, B cells, and cytolytic T cells. Each of these cell components of the adaptive immune system produces its particular spectrum of secreted molecules and cytokines. Th1 cells secrete interferon-γ, which is clearly proatherogenic.6,7,32⇓⇓ Th2 cells and their characteristic cytokines interleukins-4 and -10 appear to be predominately antiatherogenic.19,33,34⇓⇓ There is a suggestion that with profound hypercholesterolemia, there is a switch from a predominance of Th1 cells to 1 of Th2 cells.35 Recent evidence suggests that B cells are antiatherogenic.10–12⇓⇓ The net outcome with respect to atherosclerosis would be the result of the balance of the inputs from each of these components of the adaptive immune system. The balance between proatherogenic and antiatherogenic modifiers might differ at different vascular sites.
Two systemic factors that could influence atherogenesis are changed by immune system deficiency. Our immune system-deficient mice had lower plasma cholesterol levels attributable to decreases in VLDL in the high-fat diet-fed mice. Similar observations were made in chow-fed apoE−/− and apoE−/− RAG2−/− mice2 and in LDLR−/− and LDLR−/− RAG1−/− mice by Song et al,5 who also suggested that immune deficiency influences VLDL metabolism. However, in the latter study, VLDL levels were decreased, without a significant difference in total cholesterol levels in mice fed a Western-type diet. The major difference between our study and the study by Song et al is that our immune system-competent LDLR−/− mice responded to the high-fat/high-cholesterol diets with higher plasma cholesterol levels. The cholesterol levels in our high-fat/high-cholesterol-fed, immune system-incompetent LDLR−/− RAG−/− mice were very similar to those reported by Song et al. The strain 1 mice, both immune system competent and incompetent, had lower levels of VLDL than did the strain 2 mice. This is a systemic change and could only be related to the differential response of aortic sinus and innominate artery atherosclerosis in these 2 strains of mice if triglyceride and VLDL were selective risk factors for innominate artery atherosclerosis. Even if this were the case, it would require that the same VLDL particles interact differently with the innominate artery than with the aortic sinus and that VLDL is more atherogenic for the innominate artery, with the maximal atherogenic concentration of VLDL for the innominate artery being >400 mg/dL. This would suggest that the atherogenic potential of VLDL for the aortic sinus and the innominate artery exhibits a substantially different concentration dependence. With this possibility in mind, we examined whether there was a relation between triglyceride levels (a surrogate for VLDL levels) and atherosclerosis at either site when compared for each individual animal. No such correlation was observed.
The second systemic metabolic effect of immune system deficiency is the lower body weight, compared with immune system-competent mice, especially in strain 2. We have no explanation for this, and we suspect that it has no influence on the differential atherosclerosis response. The weight difference is mainly attributable to better weight gain in the immune system-competent strain 2 mice on the high-fat diet. There was no difference in body weights among the immune system-deficient mice, despite the substantial difference in innominate artery atherosclerosis between strain 1 and strain 2 mice.
In conclusion, our data confirm that innominate artery and aortic sinus atheroscleroses respond selectively to immune system deficiency in the LDLR−/− mouse model, just as we had previously shown for the apoE−/− model. We also show that the innominate artery, but not the aortic sinus, responds to immune system deficiency in a fashion sensitive to the genetic background. There are 2 possibilities to explain this genetically driven differential susceptibility of the innominate artery to atherosclerosis in the context of immune system deficiency. On the one hand, 1 or more genes expressed in this location (perhaps hemodynamically conditioned) interact with the immune system deficiency state to modify the atherosclerosis response. Alternatively, the genetic background could influence risk factors, eg, VLDL level or composition, to which the innominate artery responds differently than the aortic sinus in the context of immune system deficiency. We favor the first explanation but cannot exclude the second. In any event, the data highlight the need to extend studies that seek to characterize genetic modifier genes that influence atherosclerosis to sites other than the aortic sinus.
This research was supported by National Institutes of Health (Bethesda, Md) grants HL15062 and HL56827.
- Received April 17, 2003.
- Accepted May 20, 2003.
- ↵Reardon CA, Blachowicz L, White T, Cabana V, Wang Y, Lukens J, Bluestone J, Getz GS. Effect of immune deficiency on lipoproteins and atherosclerosis in male apoprotein E-deficient mice. Arterioscler Thromb Vasc Biol. 2001; 21: 1011–1016.
- ↵Dansky HM, Charlton SA, Harper MM, Smith JD. T and B lymphocytes play a minor role in atherosclerotic plaque formation in the apolipoprotein E-deficient mouse. Proc Natl Acad Sci U S A. 1997; 94: 4642–4646.
- ↵Mallat Z, Corbaz A, Scoazec A, Graber P, Alouani S, Esposito B, Humbert Y, Chvatchko Y, Tedgui A. Interleukin-18/interleukin-18 binding protein signaling modulates atherosclerotic lesion development and stability. Circ Res. 2001; 89: e41–e45.
- ↵Whitman SC, Ravisankar P, Daugherty A. Interluekin-18 enhances atherosclerosis in apolipoprotein E−/− mice through release of interferon-γ. Circ Res. 2002; 90: e34–e38.
- ↵Major A, Fazio S, Linton MF. B lymphocyte deficiency increases atherosclerosis in LDL receptor-null mice. Arterioscler Thromb Vasc Biol. 2002; 22: 1892–1898.
- ↵Freigang S, Hörkkö S, Miller E, Witztum JL, Palinski W. Immunization of LDL receptor-deficient mice with homologous malondialdehyde-modified 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.
- ↵Witting PK, Pettersson K, Letters J, Stocker R. Site-specific antiatherogenic effect of probucol in apolipoprotein E-deficient mice. Arterioscler Thromb Vasc Biol. 2000; 20: e26–e33.
- ↵Methia N, Andre P, Denis CV, Economopoulos M, Wagner DD. Localized reduction of atherosclerosis in von Willebrand factor-deficient mice. Blood. 2001; 98: 1424–1428.
- ↵King VL, Szilvassy SJ, Daugherty A. IL-4 deficiency decreases atherosclerosis formation in female LDL receptor deficient mice. Arterioscler Thromb Vasc Biol. 2002; 22: 456–461.
- ↵Mehrabian M, Allayee H, Wong J, Shih W, Wang X-P, Shaposhnik Z, Funk CD, Lusis AJ. Identification of 5-lipoxygenase as a major gene contributing to atherosclerosis susceptibility in mice. Circ Res. 2002; 91: 120–126.
- ↵Welch CL, Bretschger S, Latib N, Bezouevski M, Guo Y, Pleskac N, Liang C-P, Barlow C, Dansky H, Breslow JL, Tall AR. Localization of atherosclerosis susceptibility loci to chromosomes 4 and 6 using the Ldlr knockout mouse model. Proc Natl Acad Sci U S A. 2001; 98: 7946–7951.
- ↵Dansky HM, Shu P, Donavan M, Montagno J, Nagel DL, Smutko JS. Roy N, Whiteing S, Barrios J, McBride TJ, Smith JD, Duyk G, Breslow JL, Moore KJ. A phenotype-sensitizing apoE-deficient genetic background reveals novel atherosclerosis predisposition loci in the mouse. Genetics. 2002; 160: 1599–1608.
- ↵Dansky HM, Charlton SA, Sikes JL, Heath SC, Simantov R, Levin LP, Shu P, Moore KJ, Breslow JL, Smith JD. Genetic background determines the extent of atherosclerosis in apoE-deficient mice. Arterioscler Thromb Vasc Biol. 1999; 19: 1960–1968.
- ↵Mehrabian M, Wong J, Wang X, Jiang Z, Shi W, Fogelman AM, Lusis AJ. Genetic loci in mice that block the development of atherosclerosis despite extreme hyperlipidemia. Circ Res. 2001; 89: 125–130.
- ↵Garcia-Cardena G, Comander J, Anderson KR, Blackman BR, Gimbrone MA. Biomechanical activation of vascular endothelium as a determinant of its functional phenotype. Proc Natl Acad Sci U S A. 2001; 98: 4478–4485.
- ↵Pinderski LJ, Fischbein MP, Subbanagounder G, Fishbein MC, Kubo N, Cheroutre H, Curtiss LK, Berliner JA, Boisvert WA. Overexpression of interleukin-10 by activated T lymphocytes inhibits atherosclerosis in LDL receptor-deficient mice by altering lymphocyte and macrophage phenotypes. Circ Res. 2002; 90: 1064–1071.
- ↵Mallat Z, Besnard S, Duriez M, Deleuze V, Emmanuel F, Bureau MF, Soubrier F, Esposito B, Duez H, Fievet C, Staels B, Duverger N, Scherman D, Tedgui A. Protective role of interleukin-10 in atherosclerosis. Circ Res. 1999; 85: e17–e24.