Angiopoietin-like 3 monomers are abundant in human plasma but are unable to inhibit endothelial lipase

Figures

• 

  Figure 1. Identification of cleavage-resistant and glycosylation-resistant ANGPTL3 mutations. (A) Western blot of conditioned media from cells expressing wild-type ANGPTL3 (WT A3) or ANGPTL3 S220A-R221A. (B) Conditioned media from cells expressing WT or the indicated mutant ANGPTL3 were treated with deglycosylation enzymes. The treated and untreated media were analyzed by Western blotting using an antibody against ANGPTL3. (C and D) The phospholipase activity of EL after incubation with the indicated concentrations of mutant and WT ANGPTL3 for 30 minutes at 37°C was measured using a fluorescence-based phospholipase assay. Data points represent the mean ± SD of 3 independent experiments performed in duplicate. For each graph, the activity was normalized to the control treated with EL and 0 μg/mL ANGPTL3. The experiments shown in panels C and D were run simultaneously and thus used the same WT control.

• 

  Figure 2. Representative levels of EL inhibition by different ANGPTL3 mutants. Increasing concentrations of wild-type ANGPTL3 (WT) and ANGPTL3 I68A-N69A (A), E160A-V161A (B), or S104A-T105A (C) were incubated with EL for 30 minutes at 37°C. The phospholipase activity of EL was measured using a fluorescence-based phospholipase assay. Points and lines represent the data from a single replicate. For each graph, the activity was normalized to the control treated with EL and 0 μg/mL ANGPTL3.

• 

  Figure 3. Identification of regions and residues critical for EL inhibition. Phospholipase activity of EL after incubation with the indicated concentrations of wild-type ANGPTL3 (WT) or ANGPTL3 with mutations in the first 4 amino acids of the SE1 motif (A), in the region downstream of the SE1 motif (B), or in aliphatic hydrophobic residues in the coiled-coil region (C). EL and ANGPTL3 were incubated for 30 minutes at 37°C and then phospholipase activity was measured using a fluorescence-based assay. Data points represent the mean ± SD of 3 independent experiments performed in duplicate. For each graph, the activity was normalized to the control treated with EL and 0 μg/mL ANGPTL3. All experiments in panel B were performed simultaneously and thus used the same WT control. All experiments in panel C were performed simultaneously and thus used the same WT control.

• 

  Figure 4. Wild-type ANGPTL3 forms a homotrimer. (A) Blue NativePAGE and immunoblotting of recombinant wild-type (WT A3) or cleavage-resistant (CR) ANGPTL3 conditioned media detected with an antibody against ANGPTL3. (B) Blue NativePAGE and immunoblotting of endogenous secreted ANGPTL3 from Huh7 cells or recombinant ANGPTL3 (recomb A3) conditioned media detected with an antibody against ANGPTL3. (C and D) Mass photometry of WT (C) or cleavage-resistant (D) ANGPTL3 after purification using affinity chromatography. Peaks represent the estimated molecular weight of subpopulations within the sample. (E) AlphaFold 3 prediction of the mouse ANGPTL3 homotrimer. Colors indicate predicted local distance difference test (pLDDT) values. Dark blue = very high (pLDDT > 90), light blue = confident (90 > pLDDT > 70), yellow = low (70 > pLDDT > 50), and orange = very low (pLDDT < 50).

• 

  Figure 5. Mutations to aliphatic hydrophobic residues prevent the formation of stable homotrimers. (A) AlphaFold 3 prediction of the mouse ANGPTL3 homotrimer showing the indicated residue as spheres at the interface of the 3 ANGPTL3 chains. (B) Blue NativePAGE and immunoblotting of wild-type (WT A3) or mutant ANGPTL3 conditioned media detected with an antibody against ANGPTL3. (C) Mass photometry of ANGPTL3 mutant L124A-E125A after purification using affinity chromatography. Peaks represent the estimated molecular weight of subpopulations within the sample.

• 

  Figure 6. Aliphatic hydrophobic residues are essential for EL inhibition and ANGPTL3 homotrimerization. (A) The phospholipase activity of EL after incubation with the indicated concentrations of mutant and wild-type (WT) ANGPTL3 for 30 minutes at 37°C was measured using a fluorescence-based phospholipase assay. Data points represent the mean ± SD of 3 independent experiments performed in duplicate. For each graph, the activity was normalized to the control treated with EL and 0 μg/mL ANGPTL3. The experiments shown in the 2 left-hand graphs were performed simultaneously and thus use the same WT control. (B) Blue NativePAGE followed by immunoblotting of WT or mutant ANGPTL3 conditioned media with an antibody against ANGPTL3.

• 

  Figure 7. Subpopulations of ANGPTL3 in normal human serum. (A) Pooled human serum was fractionated by SEC, and total ANGPTL3, trimeric ANGPTL3, or ANGPTL3/8 complexes in the individual fractions were measured using dedicated immunoassays. (B) ANGPTL3/8 in pooled human serum was depleted using an anti-ANGPTL8 antibody (IBA363). The depleted serum was then fractionated by SEC and total ANGPTL3, ANGPTL3 trimers, and ANGPTL3/8 complex levels for each fraction were measured using dedicated immunoassays. (C and D) Western blot analysis of ANGPTL3 in different serum SEC fractions. Neighboring SEC peak fractions from the 3 major peaks in B (peak 1: 23 and 24, peak 2: 30 and 31, peak 3: 58 and 59) were pooled. After immunoprecipitating ANGPTL3, Western blotting was performed using antibodies against the N-terminus (C) or C-terminus (D) of ANGPTL3.

• 

  Figure 8. Large amounts of ANGPTL3 monomers are present in human serum. (A) Neighboring SEC peak fractions from the first 2 major peaks were pooled and immunoprecipitated for ANGPTL3. ANGPTL3 was then subjected to blue NativePAGE and immunoblotting with an ANGPTL3 antibody. Recombinant human ANGPTL3 (rhANGPTL3) conditioned media served as a control. Red coloring shows the position of non-specific fluorescence from the Coomassie G-250 dye. (B–E) SEC fractionation of pooled human serum that was run without recombinant protein (B) or was spiked with recombinant wild-type (WT) human ANGPTL3 (C) or human ANGPTL3 L124A (D). Human ANGPTL3 L124A was also run without serum (E). Total ANGPTL3 and trimeric ANGPTL3 in the individual fractions were measured using dedicated immunoassays.

• 

  Figure 9. Levels of monomeric and trimeric ANGPTL3 in human serum under different feeding states. (A) Total ANGPTL3, trimeric ANGPTL3, and ANGPTL3/8 complex levels as measured by immunoassays in healthy human plasma donors (n = 10) after an overnight fast and 1 and 2 hours after a mixed meal. Bars represent mean ± SD. (B and C) Concentrations of ANGPTL3/8 complexes (B) and trimeric ANGPTL3 (C) graphed by individual. P values were calculated by 1-way repeated-measures ANOVA followed by Tukey’s multiple-comparison test.

• 

  Figure 10. Structure-function models of ANGPTL3. (A) Predicted structure of ANGPTL3 showing the lipase interaction domain (magenta), the single glycosylation site in the N-terminal domain (orange), the aliphatic residues necessary for trimer formation, and the cleavage site (purple). The location of mutants that did not express are shown in red and 1 of the 3 ANGPTL3 chains is shown in green. (B) Alignment of the N-terminal domains of mouse and human ANGPTL3 using ClustalOmega (62). The lipase interaction domain, glycosylation site, critical trimer residues, and cleavage recognition site are colored as in A. The classical SE1 motif is also shown. (C and D) AlphaFold 3 prediction of the ANGPTL3 homotrimer (gray with the lipase interaction domain in magenta) interacting with EL (slate blue). (E and F) AlphaFold 3 predictions of an ANGPTL3 monomer interacting with EL.