Fluid-phase pinocytosis of native low density lipoprotein promotes murine M-CSF differentiated macrophage foam cell formation

Abstract

During atherosclerosis, low-density lipoprotein (LDL)-derived cholesterol accumulates in macrophages to form foam cells. Macrophage uptake of LDL promotes foam cell formation but the mechanism mediating this process is not clear. The present study investigates the mechanism of LDL uptake for macrophage colony-stimulating factor (M-CSF)-differentiated murine bone marrow-derived macrophages. LDL receptor-null (LDLR-/-) macrophages incubated with LDL showed non-saturable accumulation of cholesterol that did not down-regulate for the 24 h examined. Incubation of LDLR-/- macrophages with increasing concentrations of (125)I-LDL showed non-saturable macrophage LDL uptake. A 20-fold excess of unlabeled LDL had no effect on (125)I-LDL uptake by wild-type macrophages and genetic deletion of the macrophage scavenger receptors CD36 and SRA did not affect (125)I-LDL uptake, showing that LDL uptake occurred by fluid-phase pinocytosis independently of receptors. Cholesterol accumulation was inhibited approximately 50% in wild-type and LDLR-/- mice treated with LY294002 or wortmannin, inhibitors of all classes of phosphoinositide 3-kinases (PI3K). Time-lapse, phase-contrast microscopy showed that macropinocytosis, an important fluid-phase uptake pathway in macrophages, was blocked almost completely by PI3K inhibition with wortmannin. Pharmacological inhibition of the class I PI3K isoforms alpha, beta, gamma or delta did not affect macrophage LDL-derived cholesterol accumulation or macropinocytosis. Furthermore, macrophages from mice expressing kinase-dead class I PI3K beta, gamma or delta isoforms showed no decrease in cholesterol accumulation or macropinocytosis when compared with wild-type macrophages. Thus, non-class I PI3K isoforms mediated macropinocytosis in these macrophages. Further characterization of the components necessary for LDL uptake, cholesterol accumulation, and macropinocytosis identified dynamin, microtubules, actin, and vacuolar type H(+)-ATPase as contributing to uptake. However, Pak1, Rac1, and Src-family kinases, which mediate fluid-phase pinocytosis in certain other cell types, were unnecessary. In conclusion, our findings provide evidence that targeting those components mediating macrophage macropinocytosis with inhibitors may be an effective strategy to limit macrophage accumulation of LDL-derived cholesterol in arteries.

Figure 1. Macropinosome formation is M-CSF dependent.

A and B. Wild-type macrophages were differentiated with M-CSF for 7 days and visualized by phase-contrast microscopy following the treatments described below. Macrophages differentiated with M-CSF were pretreated 30 min with DMSO drug vehicle (A), or 5 µM of cFMS (i.e., M-CSF receptor) tyrosine kinase inhibitor, GW2580 (B). Pretreatment was carried out without either serum or M-CSF. Withdrawal of M-CSF caused disappearance of the macrophage vacuoles. Subsequently, these macrophage cultures were treated 30 min with fresh serum-free medium containing M-CSF (50 ng/ml) without (A) or with GW2580 (B). Macrophages treated with M-CSF without GW2580 showed numerous vacuoles shown to be macropinosomes in Video S1. In contrast, there was complete inhibition of macropinosome formation when macrophage cultures were treated with GW2580 (also see Video S2). Scale bar in B = 75 µm and also applies to A. (C) Wild-type macrophages were incubated 24 h with 1 mg/ml LDL without or with 5 µM GW2580, and then cholesterol accumulation was assessed. Macrophages incubated without LDL had 111±3 nmol cholesterol/mg protein. ** = p<0.01.

Figure 2. Macrophage uptake of ¹²⁵I-LDL is non-saturable.

A. LDLR−/− macrophages were incubated with 1 mg/ml of LDL for 0–24 h and cholesterol accumulation was then assessed. B. LDLR−/− macrophages were incubated 24 h with increasing concentrations of ¹²⁵I-LDL, and then ¹²⁵I-LDL uptake was assessed. Uptake values represent the sum of cell-associated and degraded ¹²⁵I-LDL. The range of cell-associated and degraded ¹²⁵I-LDL was 7–13% and 87–93%, respectively.

Figure 3. Fluid-phase pinocytosis mediates LDL uptake.

(A) Wild-type macrophages were incubated 6 h with either 25 µg/ml ¹²⁵I-LDL alone, or 25 µg/ml ¹²⁵I-LDL and 500 µg/ml unlabeled LDL. CD36 KO macrophages (ΔCD36) and SRA KO macrophages (ΔSRA) macrophages were incubated 6 h with 25 µg/ml ¹²⁵I-LDL alone. (B) Wild-type macrophages were incubated 6 h with either 25 µg/ml ¹²⁵I-AcLDL alone, or 25 µg/ml ¹²⁵I-AcLDL and 500 µg/ml unlabeled AcLDL. CD36 KO macrophages (ΔCD36) and SRA KO macrophages (ΔSRA) macrophages were incubated 6 h with 25 µg/ml ¹²⁵I-AcLDL alone. Incubations were performed in serum-free medium containing 50 ng/ml M-CSF. Uptake values represent the sum of cell-associated and degraded ¹²⁵I-labeled lipoprotein. The range of cell-associated and degraded ¹²⁵I-LDL was 17–21% and 79–83%, respectively. The range of cell-associated and degraded ¹²⁵I-AcLDL was 16–18% and 82–84%, respectively. ¹²⁵I-LDL uptake was not competed with excess unlabeled LDL consistent with fluid-phase pinocytosis mediating uptake. Statistical tests compare each treatment group with wild-type macrophages incubated with 25 µg/ml ¹²⁵I-AcLDL. * = p<0.05. ** = p<0.01. *** = p<0.001. There was no statistical difference between macrophage groups incubated with ¹²⁵I-LDL.

Figure 4. Macropinocytosis is inhibited by PI3K and actin polymerization inhibitors.

Wild-type macrophages differentiated with M-CSF were pretreated 30 min with the drugs indicated below but without serum and M-CSF. Then, macrophages were treated 30 min with fresh serum-free medium containing M-CSF (50 ng/ml) in the presence of either DMSO vehicle (A), 100 nM wortmannin (B), 4 µg/ml cytochalasin D (C), or 50 µM LY294002 (D), and then examined by phase-contrast microscopy. Arrows indicate macropinosomes. Scale bar in A = 75 µm and applies to all. (Also see Videos S5, S6, S7, S8).

Figure 5. Wild-type and LDL−/− macrophages incubated with LDL accumulate similar levels of cholesterol.

Wild-type and LDLR−/− macrophages were incubated with 1 mg/ml LDL for 24 h and then total cholesterol accumulation was assessed. The baseline cholesterol levels for wild-type and LDLR−/− macrophages were 107±10 nmol cholesterol/mg protein and 122±9 nmol cholesterol/mg protein, respectively.

Figure 6. LDL-derived cholesterol accumulation occurs independently of class I PI3K isoforms.

Cholesterol accumulation after 24 h-incubation without or with 1 mg/ml LDL was assessed in M-CSF differentiated macrophages cultured from wild-type, PI3Kγ-KI, PI3Kβ-KI, and PI3Kδ-KI mice.

Figure 7. Macrophage macropinosome formation is dependent on dynamin and vacuolar type H(+)-ATPase.

Wild-type M-CSF differentiated macrophages were pretreated 30 min with the indicated drugs but without serum and M-CSF. Subsequently, macrophages cultures were treated 30 min with fresh serum-free medium containing M-CSF (50 ng/ml) and either drug vehicle (A), 80 µM dynasore (B), 500 nM bafilomycin A1 (C), or 100 µM NSC23766 (D). Then, macropinosome formation was assessed by phase-contrast microscopy. Scale bar in A = 75 µm and applies to all. (Also see Videos S9, S10, S11, S12, S13, S14).