Modulation of atherosclerosis in mice
by Toll-like receptor 2 (original) (raw)
Total body deficiency of TLR2. To determine the role of TLR2 in atherosclerosis, Tlr2–/– mice were crossed with atherosclerosis-prone _Ldlr–/–_mice to generate double-knockout Ldlr–/–Tlr2–/– mice. Twelve-week-old male mice were fed a high-fat diet (HFD) for 10 and 14 weeks. Relative to Ldlr–/– mice (n = 27), Ldlr–/–Tlr2–/– mice (n = 17) had reduced total plasma cholesterol levels and increased body weight throughout 10 or 14 weeks of an HFD (Supplemental Figure 1, A and B; supplemental material available online with this article; doi:10.1172/JCI25482DS1). Averaged over the HFD feeding period, the reduction in total plasma cholesterol in Ldlr–/–Tlr2–/– mice was 16% less relative to Ldlr–/– mice (1242 ± 50 vs. 1485 ± 34 mg/dl, P < 0.05).
Analysis of aortic lesion area and aortic valve lesion volume revealed a significant decrease in disease in Ldlr–/–Tlr2–/– mice, with an almost 50% decrease in aortic atherosclerosis (at both time points) and a 55% and 30% decrease in aortic valve lesion volume after 10 and 14 weeks, respectively (Figure 1, A and B).
Measurement of atherosclerosis in male Ldlr–/– and Ldlr–/–Tlr2–/– mice after 10 or 14 weeks of consuming an HFD. (A) Aortic lesion area was calculated as fraction of total aorta covered by lesion. (B) Aortic sinus lesion volume was calculated from an integration of the cross-sectional areas of the lesion in the proximal 500 μm of the sinus. Aortic lesion area and aortic valve lesion volume were significantly decreased in Ldlr–/–Tlr2–/– mice.
BM-derived cell expression of TLR2. Due to the pivotal role that macrophages serve in atherogenesis, we hypothesized that the absence or presence of macrophage TLR2 expression would also have an impact on atherosclerosis. Utilizing BM transplantation (BMT), Ldlr–/– mice were generated that lacked TLR2 expression in their BM-derived cells. Similarly, Ldlr–/–Tlr2–/– mice were generated that expressed TLR2 only in their BM cells.
In the first BMT, 20-week-old female Ldlr–/– mice were reconstituted with BM from donor mice lacking TLR2 (Tlr2–/–) (n = 17) or their respective wild-type controls (Tlr2+/+) (n = 21). After 4 weeks of recovery, animals were fed the HFD for 16 weeks, after which they were sacrificed and their aortae and hearts harvested. Ldlr–/– mice receiving Tlr2–/– BM had similar body weight gain as those receiving Tlr2+/+ marrow throughout the study (Supplemental Figure 2A). Both groups also had a similar increase in plasma cholesterol upon an HFD, with plasma cholesterol increasing at each time point thereafter. Nevertheless, there were differences in plasma cholesterol between these groups at the 4- and 16-week time points (Supplemental Figure 2B). In Ldlr–/–Tlr2–/– mice receiving Tlr2+/+ (n = 14) or Tlr2–/– (n = 16) BM, we also observed no changes in weight gain between the 2 groups with consistent growth observed throughout the 16-week feeding period (Supplemental Figure 2C). Plasma cholesterol levels in these 2 groups were similar throughout the study (Supplemental Figure 2D).
Surprisingly, quantitation of atherosclerosis revealed no impact of BM-derived cellular expression of TLR2 on aortic sinus lesion volume in Ldlr–/– recipients (Figure 2; Ldlr–/– recipient, Tlr2+/+ donor vs. Ldlr–/– recipient, Tlr2–/– donor) or in Ldlr–/–Tlr2–/– BM recipients (Figure 2; Ldlr–/–Tlr2–/– recipient, Tlr2+/+ donor vs. Ldlr–/–Tlr2–/– recipient, Tlr2–/– donor). This BMT study was also performed in male mice, and similar results were obtained. Importantly, comparison of groups with only changes in non-BM expression of TLR2 duplicated our finding in the male double-mutant mice that non-BM cell expression of TLR2 modulated atherosclerosis. For example, significant decreases were observed in atherosclerosis in mice that lacked TLR2 expression in non-BM cells but expressed TLR2 in BM-derived cells (Figure 2; Ldlr–/– recipient, Tlr2+/+ donor vs. Ldlr–/–Tlr2–/– recipient, Tlr2+/+ donor) or lacked TLR2 in BM-derived cells (Figure 2; Ldlr–/– recipient, Tlr2–/– donor vs. Ldlr–/–Tlr2–/– recipient, Tlr2–/– donor). The data for total aortic lesion area showed the same trends as the data for aortic sinus lesion volume discussed above.
Measurement of heart aortic sinus atherosclerosis volume after 16 weeks of HFD feeding in female Ldlr–/– and Ldlr–/–Tlr2–/– mice that underwent BM reconstitution with BM from Tlr2+/+ or Tlr2–/– donors. Regardless of BM cell genotype, a loss of TLR2 on non-BM–derived cells significantly reduced disease.
Exogenous agonist activation of TLR2. Studies were performed to evaluate the impact of an exogenous TLR2 agonist on atherosclerosis. Twelve-week-old mice received weekly i.p. injections of Pam3, a synthetic TLR2 agonist that mimics the triacylated amino terminus of bacterial lipoproteins and activates TLR2/TLR1 heterodimers (24). Twenty-two female Ldlr–/– mice were split into 3 groups that received weekly i.p. injections of vehicle, 25 μg of Pam3, or 50 μg of Pam3. An ELISA measurement of the acute phase protein, serum amyloid A (SAA), was performed 24 hours after initiation of the HFD and i.p. injections (Figure 3). The vehicle injection group experienced a 9-fold increase in SAA (16 ± 3 vs. 140 ± 68, P < 0.05). Relative to the increase seen with vehicle injections, we measured a 10-fold and 24-fold increase with 25 and 50 μg of Pam3, respectively (1,380 ± 250 vs. 3,420 ± 420 μg/ml SAA, P < 0.05). This demonstrated a systemic inflammatory reaction in response to i.p. administration of Pam3. Injection of Pam3 in the Ldlr–/–Tlr2–/– mice was similar to the vehicle injections in Ldlr–/– mice, confirming the specificity of the Pam3 preparation for TLR2. Throughout the study, all mice experienced similar rates of weight gain (Supplemental Figure 3A). After 4 weeks of consuming the HFD and weekly injections, total plasma cholesterol was significantly increased in all groups, from 240 to 1,330 (vehicle), 270 to 1,750 (25 μg of Pam3) and 260 to 1,340 mg/dl (50 μg of Pam3) (Supplemental Figure 3B). By the eighth and twelfth weeks, the Pam3 groups had similar plasma cholesterol levels, which were reduced relative to the vehicle group: 1,110 (vehicle), 900 (25 μg of Pam3), and 850 mg/dl (50 μg Pam3) (P < 0.05). Averaging each group’s plasma cholesterol during the HFD feeding revealed a slight cholesterol-lowering effect from Pam3 exposure. Additionally, 7 female Ldlr–/–Tlr2–/– mice were used as negative controls to test for the specificity of Pam3 for TLR2. Ldlr–/–Tlr2–/– mice receiving 50 μg of Pam3 exhibited no differences in body weight or total plasma cholesterol levels relative to vehicle-injected Ldlr–/–Tlr2–/– mice.
ELISA measurement of plasma SAA 24 hours following i.p. injections of vehicle, 25 μg of Pam3, or 50 μg of Pam3 demonstrated a dose-dependent systemic inflammatory response. Double-mutant mice exhibited baseline (time = 0) SAA values similar to those of Ldlr–/– mice. Error bars indicate SEM.
Analysis of aortic atherosclerosis demonstrated a significant increase in lesion areas with Pam3 administration. Relative to the vehicle-injected group, a dose-response effect of Pam3 was observed in aortic atherosclerosis, with significant increases in lesion severity (relative to the vehicle group) of 220% and 480% observed in the 25 and 50 μg Pam3 groups, respectively (Figure 4A). Pam3 administration resulted in a peculiar distribution of lesion development with profuse abdominal atherosclerosis observed in the 50 μg Pam3 group and sporadic abdominal lesions in the 25 μg Pam3 group (Figure 4, C and D). A similar dose-response effect was observed with aortic sinus lesion volume, with an increase in lesion volume (relative to the vehicle group) of 54% and 82% in the 25 μg Pam3 and 50 μg Pam3 groups, respectively (P < 0.05) (Figure 5A). Mice deficient in TLR2 (Ldlr–/–Tlr2–/–) exhibited no changes in aortic or aortic sinus atherosclerosis and confirmed the specificity of Pam3 for TLR2-mediated cell activation (Figures 4 and 5). Additionally, comparison of aortic and aortic sinus atherosclerosis in Ldlr–/– and Ldlr–/–Tlr2–/– animals receiving vehicle injections revealed the atheroprotective effect of complete TLR2 deficiency, as demonstrated previously. Finally, a comparison of the trends observed in male cohorts (Figure 1) and female cohorts (Figures 4 and 5; vehicle controls) confirmed that the atheroprotection seen with TLR2 deficiency was not sex-dependent.
Measurement of aortic atherosclerosis in Pam3-exposed mice. (A) Aortic atherosclerosis expressed as a fraction of total area after 12 weeks of an HFD and weekly i.p. injections of vehicle (veh) or Pam3 in female Ldlr–/– and Ldlr–/–Tlr2–/– mice. (B–D) Representative aortae from Ldlr–/– mice in the following groups: vehicle (B), 25 μg Pam3 (C), and 50 μg Pam3 (D). A dose-response effect of Pam3 administration and lesion development was observed. Profuse abdominal aortic atherosclerosis was observed in mice exposed to 50 μg Pam3. Lesion severity in TLR2-deficient animals was not affected by Pam3. Scale bar: 0.5 cm.
Measurement of heart aortic sinus atherosclerosis volume in Pam3-exposed mice. (A) Heart aortic sinus atherosclerosis volume after 12 weeks of an HFD and weekly i.p. injections of Pam3 in female Ldlr–/– and Ldlr–/–Tlr2–/– mice. (B–D) Representative cross-sections of the aortic sinus from Ldlr–/– mice of the following groups: vehicle (B), 25 μg Pam3 (C), and 50 μg Pam3 (D). A dose-response effect of Pam3 administration and lesion development was observed. Lesion severity in TLR2-deficient animals was not affected by Pam3. Scale bar: 0.5 mm.
Utilizing BMT, the contribution of macrophage TLR2 expression was ascertained in the setting of exogenous ligand activation of TLR2 by Pam3. In this BMT study, 40 female Ldlr–/– mice were split equally into 2 groups, with 1 group receiving BM from wild-type Tlr2+/+ mice and the other from Tlr2–/– mice, similar to our previous BMT studies. Mice in each group were further subdivided into groups receiving weekly i.p. injections of 50 μg Pam3 or vehicle.
Among the 4 groups (wild-type vs. Tlr2–/– BM recipients ± 50 μg Pam3), the wild-type Ldlr–/– recipients receiving Pam3 experienced a slight decrease in body weight toward the end of the treatment period (Supplemental Figure 4A) whereas the Tlr2–/– recipients receiving Pam3 demonstrated slight increases in body weight (Supplemental Figure 4C). Animals with BM-derived cell expression of TLR2 demonstrated large increases in atherosclerosis with Pam3 administration (Table 1 and Figures 6, A–E, and 7, A–E). In these groups, aortic atherosclerosis increased by 280% and aortic sinus lesion volume increased by 90% relative to the vehicle controls (Table 1). These profound increases in disease were similar to those observed in the non-BMT agonist injection studies (Table 1), with profuse abdominal atherosclerosis observed with Pam3 administration (Figure 6C). A comparison of the vehicle control groups in this study duplicated the observation of our previous BMT study that BM-derived cell expression of TLR2 did not modulate atherosclerosis in the absence of an exogenous agonist (Figures 6, A, B, and D, and 7, A, B, and D). In the absence of BM-derived cell expression of TLR2, Pam3 administration had only a slight effect on increasing atherosclerosis in the aorta (Table 1 and Figure 6, A, D, and E) and no effect on aortic sinus lesion volume (Table 1 and Figure 7, A, D, and E).
Measurement of aortic atherosclerosis in Pam3-exposed TLR2 chimeric mice. (A) Aortic atherosclerosis expressed as a percentage of total area after 12 weeks of an HFD and weekly i.p. injections of Pam3 in female Ldlr–/– mice that underwent BM reconstitution from Tlr+/+ or Tlr2–/– donor mice. (B–E) Representative aortae from Ldlr–/– recipient mice of the following groups: Tlr2+/+ BM donor, vehicle injection (B); Tlr2+/+ BM donor, 50 μg Pam3 per week (C); Tlr2–/– BM donor, vehicle injection (D); and Tlr2–/– BM donor, 50 μg Pam3 per week (E). Pam3 administration increased lesion severity in animals expressing TLR2 in their BM-derived cells resulting in profuse abdominal aortic atherosclerosis. Slight increases in lesion severity were observed in Tlr2–/– recipient animals receiving Pam3. Scale bar: 0.5 cm.
Measurement of heart aortic sinus atherosclerosis volume in Pam3-exposed TLR2 chimeric mice. (A) Heart aortic sinus atherosclerosis after 12 weeks of an HFD and weekly i.p. injections of Pam3 in female Ldlr–/– mice that underwent BM reconstitution from Tlr+/+ or Tlr2–/– donor mice. (B–E) Representative aortic sinus cross-sections from Ldlr–/– mice of the following groups: Tlr2+/+ BM donor, vehicle injections (B); Tlr2+/+ BM donor, 50 μg Pam3 injections (C); Tlr2–/– BM donor, vehicle injections (D); and Tlr2–/– BM donor, 50 μg Pam3 injections (E). Pam3 administration increased lesion severity only in animals expressing TLR2 in their BM-derived cells. Scale bar: 0.5 mm.
Summary of TLR2 agonist studies
Lipoprotein cholesterol distribution. Because TLR2 deficiency or activation may change the lipoprotein distribution of plasma cholesterol, the distribution of total cholesterol within the major lipoprotein fractions was analyzed by fast protein liquid chromatography (FPLC). FPLC was performed on plasma samples obtained from mice 8 weeks after initiation of the HFD. Equal volumes of samples were pooled from 4–8 representative mice per group. As previously reported (28), the vast majority of total plasma cholesterol was located within the VLDL/LDL lipoprotein fractions. No changes were detected in the cholesterol distribution within the lipoprotein fractions from each group of mice (Figure 8, A and B).
Plasma lipoprotein cholesterol distribution. Pooled plasma samples taken from representative mice after 8 weeks of consumption of the HFD were fractionated by FPLC, and total cholesterol was measured. (A) Complete or BM cell–derived loss of TLR2 did not alter the distribution of cholesterol relative to control. (B) Administration of 50 μg Pam3 also did not have an impact on the distribution of cholesterol relative to control mice given the vehicle injections.