Exoplasmic cysteine Cys384 of the HDL receptor SR-BI is critical for its sensitivity to a small-molecule inhibitor and normal lipid transport activity

Abstract

The HDL receptor, scavenger receptor, class B, type I (SR-BI), is a homooligomeric cell surface glycoprotein that controls HDL structure and metabolism by mediating the cellular selective uptake of lipids, mainly cholesteryl esters, from HDL. The mechanism underlying SR-BI-mediated lipid transfer, which differs from classic receptor-mediated endocytosis, involves a two-step process (binding followed by lipid transport) that is poorly understood. Our previous structure/activity analysis of the small-molecule inhibitor blocker of lipid transport 1 (BLT-1), which potently (IC(50) ∼ 50 nM) blocks SR-BI-mediated lipid transport, established that the sulfur in BLT-1's thiosemicarbazone moiety was essential for activity. Here we show that BLT-1 is an irreversible inhibitor of SR-BI, raising the possibility that cysteine(s) in SR-BI interact with BLT-1. Mass spectrometric analysis of purified SR-BI showed two of its six exoplasmic cysteines have free thiol groups (Cys251 and Cys384). Converting Cys384 (but not Cys251) to serine resulted in complete BLT-1 insensitivity, establishing that the unique molecular target of BLT-1 inhibition of cellular SR-BI dependent lipid transport is SR-BI itself. The C384S substitution reduced the receptor's intrinsic lipid uptake activity by approximately 60% without dramatically altering its surface expression, homooligomerization, or HDL binding. Thus, a small-molecule screening approach identified a key residue in SR-BI involved in lipid transport, providing a powerful springboard into the analyses of the structure and mechanism of SR-BI, and highlighting the power of this approach for such analyses.

Fig. 1.

Structure of BLT-1 (A), determination of free thiols in murine SR-BI by mass spectrometry (B and C), and apparent irreversibility of SR-BI inhibition by BLT-1 (D). (A) Structure of BLT-1. The sulfur in the thiosemicarbazone (R₂CH₃N₃S) moiety of BLT-1 is essential for BLT-1’s effects on SR-BI. All activity is lost when the sulfur (shaded) is replaced with oxygen (semicarbazone). (B) Model of murine SR-BI illustrating the approximate locations of the cysteines. Cysteines 251 and 384 are fully reduced (free thiols) whereas there is little or no free thiol in cysteines 280, 321, 323, and 334 suggesting their participation in two intramolecular disulfide bonds. (C) Determination of the free thiols in murine SR-BI by mass spectrometry. Purified SR-BI-t1 was subjected to differential alkylation with NEM before and after reduction, deglycosylated, and proteolytically digested (trypsin and Glu C), and peptides were subjected to LC-MS/MS analysis as described in Experimental Procedures. Cytoplasmic Cys470 is at least partially fatty acylated (52), which presumably accounts for its low level of free thiol. (D) Cells expressing wild-type murine SR-BI (ldlA[SR-BI]) were preincubated with (septuplet determinations) or without (sextuplet determinations) 1 μM BLT-1 for 1 h, and then incubated without BLT-1 for 0 or 4 h to permit dissociation of any reversibly bound BLT-1. The cells were then incubated for 2 h with DiI-HDL (20 μg of protein per mL) without BLT-1. The cells were then washed and SR-BI specific DiI uptake (relative fluorescence units per well) was determined as described in Experimental Procedures. Nonspecific uptake was measured in the presence of a 50-fold excess of unlabeled HDL (duplicate determinations). All incubations were at 37 °C. Similar results were observed in two additional independent experiments.

Fig. 2.

Influence of BLT-1 on the HDL binding (A) and lipid uptake (B and C) activities of Cys-to-Ser mutants of SR-BI. ¹²⁵I-HDL binding (A) and either [³H]CE uptake from [³H]CE-HDL (B) or DiI uptake from DiI-HDL (C) activities of wild-type SR-BI and the indicated Cys-to-Ser mutants in stably transfected ldlA cells. Experiments were performed at 37 °C with a subsaturating concentration of HDL (10 μg of protein per mL) in the absence (black bars) or presence (white bars) of the indicated concentrations of BLT-1 as described in Experimental Procedures. All values represent receptor-specific activities calculated as the differences between activity in the absence (quadruplicate determinations) and presence (duplicate determinations) of a 40-fold excess of unlabeled HDL. The values (not corrected for cell surface receptor expression) were normalized so that 100% activity represents receptor-specific activity in the absence of BLT-1. These values in A (ng bound/mg cell protein) were as follows: SR-BI, 22 ± 2; C251S, 23 ± 2; C384S, 64 ± 1; and C251/384S, 27 ± 1. The 100% of control values in B (nanogram HDL protein equivalent per milligram cell protein) were as follows: SR-BI, 2616 ± 131; C251S, 1899 ± 196; C384S, 1721 ± 98; and C251/384S, 949 ± 42. The 100% of control values in C (relative fluorescence units) were SR-BI, 85 ± 6 and C384S, 49 ± 4. Statistical analyses comparing without or with BLT-1 were performed using either one-way ANOVA with Tukey posttesting (^∗P < 0.0001) or the unpaired two-tailed t test at 95% confidence intervals (⁺P < 0.0025 or ^∧P < 0.001).

Fig. 3.

¹²⁵I-HDL binding (A) and [³H]CE uptake from [³H]CE-HDL (B) activities of Cys-to-Ser mutants of SR-BI. Receptor-specific ¹²⁵I-HDL binding (A) and [³H]CE uptake from [³H]CE-HDL (B) mediated by wild-type SR-BI and the indicated Cys-to-Ser mutants in transiently transfected COS cells were measured at 37 °C with a subsaturating concentration of HDL (10 μg of protein per mL) as described in Experimental Procedures. Receptor-specific values were calculated as the differences between the total binding and uptake values (triplicate determinations) and the nonspecific values measured in the presence of a 40-fold excess of unlabeled HDL (single determination). All values were normalized to correct for differences in receptor surface expression relative to wild-type SR-BI based on flow cytometry (see text). Statistical analyses comparing wild-type SR-BI and the individual mutants were performed using either one-way ANOVA with Tukey posttesting (^∗P < 0.0001) or the unpaired two-tailed t test at 95% confidence intervals (⁺P = 0.0336 or ^∗P < 0.0001).

Fig. 4.

Cell surface cross-linking of SR-BI and C384S in stable transfectants. Stably transfected ldlA[SR-BI] and ldlA[C384S] cells were treated with the indicated concentrations of the water-soluble, membrane impermeable chemical cross-linker BS³ for 60 min at 4 °C. Monomers (ca. 82 kDa) and oligomers (> 150 kDa) were separated by SDS-PAGE and visualized by immunoblotting with an anti-C-terminus SR-BI polyclonal antibody (see Experimental Procedures). Analysis of the same filter with an antibody to ε-COP (Fig. S6) showed that this intracellular protein that is part of a multiprotein complex was not cross-linked by BS³ to itself or other proteins. This result suggests that BS³-mediated cross-linking was restricted to exoplasmic domains of surface proteins.