The uptake mechanism and intracellular fate of Paraoxonase-1 in endothelial cells

https://doi.org/10.1016/j.freeradbiomed.2020.03.032Get rights and content

Highlights

  • PON1 and HDL share binding sites on the endothelial membrane lipid-rafts/caveolae.

  • PON1 internalized into the EC cytoplasm in a dynamin-dependent mechanism.

  • PON1 is directed by endosomes toward lysosomal degradation.

  • Under endothelial dysfunction, the PON1 internalization into ECs is increased.

Abstract

Paraoxonase-1 (PON1) is a high-density lipoprotein (HDL)-associated lactonase that plays a significant role in the anti-atherosclerotic activity of HDL. However, several studies have shown that PON1 localizes in cells, where it operates independently of HDL. Previously, we showed that PON1 localizes in endothelial cells (ECs), and impairs vasodilation mediated by the endothelium-derived hyperpolarizing factor (EDHF) 5,6-δ-DHTL. However, the internalization pathway of PON1 into ECs, and the intracellular fate of PON1 are unknown. Therefore, the present study aimed to elucidate the uptake mechanism, intracellular trafficking and the function of PON1 in ECs. ‏We conducted a series of inhibition experiments of fluorescently labeled recombinant PON1 (rePON1) in ECs, followed by FACS analyses. We found that rePON1 binds the EC membrane via specific binding sites located in lipid-rafts/caveolae microdomains that are shared with HDL, and internalized through dynamin-dependent endocytosis. Qualitative assessments of the intracellular trafficking of rePON1, using confocal z-stack images, showed colocalization of the labeled rePON1 with early and late endosome/lysosome markers. Accordingly, a “pulse-chase” incubation of rePON1, followed by lactonase activity measurement in EC lysate, revealed that rePON1 retains its lactonase activity after binding to the cells. However, this activity decreases over time. Finally, induction of endothelial dysfunction with high glucose, angiotensin II, or palmitic acid increased rePON1 uptake by ECs. In conclusion, these results indicate that free PON1 interacts with ECs via binding sites located in lipid-rafts/caveolae, where it is enzymatically active and regulates endothelial functions. However, once internalized, PON1 is degraded. Additionally, alteration in endothelial function affects PON1 uptake by ECs.

Introduction

Paraoxonase-1 (PON1) is a high-density lipoprotein (HDL)-associated lactonase that has been thoroughly studied for its protective functions against cardiovascular disease and atherosclerosis [1,2]. PON1 is synthesized primarily in the liver and secreted into the bloodstream, where it is carried by HDL [3]. PON1 has been attributed major roles in the circulatory system. These include the protection of low-density lipoprotein (LDL), HDL, and macrophages from oxidative stress; enhancing the reverse cholesterol transport process; and maintaining proper endothelial function [[4], [5], [6]].

Although PON1 is found predominantly in the circulation system, carried by HDL, a number of studies indicate that PON1 is not fixed to HDL, but can function outside the lipoprotein environment. Accordingly, PON1 appears to be localized in multiple tissues that do not express the enzyme [4,[7], [8], [9]]. Deakin et al. showed that free PON1, and also HDL-associated PON1, can pass between the external medium and the cell membranes of endothelial cells (ECs) and Chinese‏ ‏hamster ovary (CHO) cells [8]. In the latter, reduced SR-BI expression was correlated with reduced transfer of PON1 to the cell membrane. PON1 has been shown to interact with macrophage cells, and to be specifically internalized into them through HDL binding sites [4,9,10]. The presence of PON1 in these cells was associated with an increase in cholesterol efflux capacity and a positive effect of the redox status of the cells [4,8,9,11].

In line with these findings, we previously showed that PON1 can be internalized into ECs and localized in the perinuclear area [12]. This finding is particularly interesting since, unlike the protective activity of PON1 exhibited in the circulation and macrophages, the presence of PON1 in ECs has been associated with high blood pressure. Notably, PON1 knockout mice exhibited low blood pressure compared to the wild type [13]. This finding was inversely correlated with the level of 5,6-epoxyeicosatrienoic acid (5,6-EET), an arachidonic acid metabolite that mediates vasodilation through a mechanism involving EDHF [14,15]. Follow-up studies in our laboratory showed that the 5,6-EET metabolite is transformed into the 5,6-dihydroxytrienoic lactone (5,6-δ-DHTL) isomer. This isomer also functions as EDHF and is a potential substrate for PON1 [[16], [17], [18]]. In addition, the presence of recombinant PON1 (rePON1) in ECs disrupts the [Ca+2] influx and vascular dilation mediated by 5,6-δ-DHTL [12]. However, such a disruption did not occur in the presence of a PON1 mutant that had limited ability to bind to the cell membrane (PON1-Δ20) [19]. This indicates that PON1 structure and its specific location in ECs affect its ability to regulate endothelial functions and vascular dilation.

Unlike PON1, whose role in ECs is unclear, HDL is known to undergo endocytosis in ECs, coincident with its protective functions. Studies that used different subclasses of HDL have revealed the involvement of multiple endocytic pathways of the HDL particle and receptors in HDL particle endocytosis in ECs. These include SR-B1, ABCA1, ABCG1, and ecto-F1-ATPase [[20], [21], [22]]. Consistent with the variability recognized in the internalization mechanism, the intracellular pathways of HDL particles comprise transcytosis, storage, and cellular function [[23], [24], [25]]. This variability may be dependent on the HDL-subclass used and its proteins, including PON1. Therefore, this study aimed to elucidate the uptake mechanism and intracellular trafficking and function of PON1 in ECs.

Section snippets

Cell culture

Human umbilical vein ECs (HUVECs) were isolated from umbilical cords provided by Baruch Padeh Medical Center in Poriya, Israel, as previously described [12]. The cells were maintained in a full-growth medium (EGM-2, Lonza) at 37 °C in a humidified 5% CO2 incubator. HUVECs were passaged by trypsinization at 90–95% confluence and used between passages 5 and 7.

Recombinant PON1(rePON1)

rePON1 was purchased from The Israel Structural Proteomics Center (ISPC, Weizmann Institute). The G3C9 variant of PON1 was based on

PON1 and HDL compete for internalization into ECs

We previously showed that PON1 is internalized into ECs in a concentration- and time-dependent, and saturable manner [12]. In addition, we showed that the correct 3D structure was required for the internalization of PON1 into the cells. In the current study, we conducted a series of competition experiments to examine the specificity of PON1 binding to the ECs. Accordingly, ECs were incubated with AF488-rePON1, together with increasing concentrations of either unlabeled rePON1 or HDL, in a

Discussion

In this study, we investigated the endocytosis pathway and the functionality of PON1 in ECs, and examined whether endothelial function affects PON1 uptake. Using a combination of quantitative, qualitative, and functional analyses of PON1 in ECs, we showed that (1) PON1 and HDL share specific binding sites located in the lipid-rafts/caveolae microdomains on the endothelial membrane surface; (2) PON1 is internalized into the EC cytoplasm in a dynamin-dependent mechanism; (3) after

Conclusion

We suggest that free-PON1 is localized in lipid-raft/caveolae microdomains throughout the endothelial membrane, where it is enzymatically active and regulates endothelial signals, independent of HDL. However, once internalized, PON1 is degraded and is, therefore, no longer active. Also, under endothelial dysfunction, characterized by a reduction in NO bioavailability, and a transition from NO-dependent to a compensatory EDHF-dependent (e.g. 5,6-DTHL) vasodilation [41], the rate of

Funding sources

This research was supported by The Israel Science Foundation (A.Szuchman-Sapir, Grant No. 1323/15).

Acknowledgment

The authors would like to thank Prof. Doron Goldberg, for valuable scientific editing.

References (41)

Cited by (0)

View full text