ANGPTL3

ANGPTL3 (angiopoietin-like protein 3) is a hepatocyte-specific secreted glycoprotein encoded by the ANGPTL3 gene on human chromosome 1p31.3 (7 exons, 460 amino acids, ~70 kDa glycosylated monomer) that acts as an endocrine master regulator of intravascular lipoprotein catabolism. Together with its paralogs ANGPTL4 (fasting-induced, adipose- and liver-expressed) and ANGPTL8 (feeding-induced, liver- and adipose-expressed), ANGPTL3 forms the ANGPTL3–4–8 triad, which partitions circulating triglyceride-rich lipoproteins (TRLs—chylomicrons and very-low-density lipoproteins, VLDL) between oxidative skeletal/cardiac muscle during fasting and white adipose tissue storage after feeding. ANGPTL3 inhibits two vascular endothelial phospholipases/lipases: lipoprotein lipase (LPL), which hydrolyzes core triglycerides inside chylomicrons and VLDL, and endothelial lipase (EL, encoded by LIPG), which hydrolyzes surface phospholipids on high-density lipoproteins (HDL) and VLDL remnants.

In 2010, whole-exome sequencing of a family with familial combined hypolipidemia (FHBL2) discovered that humans homozygous or compound heterozygous for nonsense loss-of-function (LoF) mutations in ANGPTL3 (S17X and E129X) exhibit >60% reductions in plasma triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), and HDL-C without hepatic steatosis or adverse systemic phenotypes. Subsequent population-scale sequencing established that heterozygous ANGPTL3 LoF carriers (~1 in 300 individuals) experience a 34% to 41% reduction in coronary artery disease (CAD) risk. Remarkably, inhibiting ANGPTL3 lowers plasma LDL-C by ~47% even in patients with homozygous familial hypercholesterolemia (HoFH) who completely lack functional LDLR activity—making ANGPTL3 a cornerstone therapeutic target for drug discovery alongside PCSK9.

Coiled-Coil Inhibitory Domain and Proprotein Convertase Cleavage

Full-length human ANGPTL3 (460 residues) comprises: (1) a 16-residue N-terminal signal peptide (residues 1–16); (2) an N-terminal coiled-coil domain (CCD, residues 17–207) containing the conserved specific epitope 1 (SE1, residues 32–55, sequence LAXGLLXLG) required for direct physical binding and unfolding of LPL and EL; (3) a basic linker region (residues 208–241) containing a canonical proprotein convertase cleavage motif (221RAPR↓TT226); and (4) a C-terminal fibrinogen-like domain (FLD, residues 242–460) that binds endothelial α_Vβ₃ integrins but does not participate in lipase inhibition.

Inside the Golgi apparatus and on the hepatocyte surface, proprotein convertases (furin and PACE4/PCSK6) cleave ANGPTL3 at Arg221–Arg224↓Thr225, liberating the N-terminal coiled-coil fragment (~30 kDa), which exhibits significantly higher inhibitory potency against LPL and EL than uncleaved full-length ANGPTL3. Transcription of ANGPTL3 in hepatocytes is driven primarily by liver X receptors (LXRα/LXRβ) and hepatocyte nuclear factor 1α (HNF1α), rendering basal ANGPTL3 secretion relatively constitutive across fasting and fed states unless modulated by ANGPTL8.

Allosteric Activation by ANGPTL8 and Catalytic Unfolding of LPL

On its own, homotrimeric ANGPTL3 is a comparatively weak inhibitor of LPL (IC₅₀ ≈ 30–70 nM) compared with ANGPTL4 (IC₅₀ ≈ 1–3 nM). However, in the postprandial (fed) state, insulin strongly induces expression of ANGPTL8 (betatrophin/lipasin, a 198-residue atypical member lacking a C-terminal fibrinogen domain) in liver and adipose tissue. ANGPTL8 cannot fold or secrete efficiently on its own; instead, it co-assembles inside the hepatocyte endoplasmic reticulum with ANGPTL3 in a 2:1 or 1:1 heteromeric coiled-coil complex (ANGPTL3/8).

Structural hydrogen–deuterium exchange mass spectrometry (HDX-MS) and cryo-EM models show that ANGPTL8 binding unmasks a cryptic high-affinity LPL-binding motif, increasing the potency of the ANGPTL3/8 complex against capillary GPIHBP1-bound LPL by 50- to 100-fold (IC₅₀ ≈ 0.2–0.6 nM). Meanwhile, in white adipose tissue, ANGPTL8 simultaneously complexes with local ANGPTL4 to form an ANGPTL4/8 heterodimer whose LPL-inhibitory activity is reduced ~75% relative to ANGPTL4 alone. This dual switch directs circulating postprandial triglycerides away from oxidative heart/skeletal muscle (inhibited by circulating endocrine ANGPTL3/8) and into white adipose tissue (disinhibited by local ANGPTL4/8) for energy storage. Mechanistically, both ANGPTL3/8 and ANGPTL4 catalyze the irreversible unfolding of the LPL α/β-hydrolase catalytic domain from an active homodimer/monomer into inactive unfolded monomers, promoting PCSK3/furin cleavage and clearance of LPL from capillary GPIHBP1 tethers.

A pivotal physiological puzzle emerged when ANGPTL3 inhibition was tested in Ldlr^−/− mice and human patients with homozygous familial hypercholesterolemia (HoFH) carrying two receptor-negative LDLR null alleles: unlike statins and PCSK9 inhibitors, which depend on upregulating functional hepatocyte LDLR and therefore show little or no efficacy in receptor-null HoFH, the anti-ANGPTL3 monoclonal antibody evinacumab reduced plasma LDL-C by 47.1% (ELIPSE HoFH Phase 3 trial) regardless of whether residual LDLR activity was present.

Daniel Rader, Gusarova, and colleagues resolved this mechanism by demonstrating that LDLR-independent LDL-C lowering requires endothelial lipase (EL). When ANGPTL3 is inhibited, derepressed EL hydrolyzes the surface phosphatidylcholine shell of nascent hepatic VLDL particles as they circulate through capillary beds. This rapid phospholipid depletion remodels VLDL into lipid-depleted VLDL remnants BEFORE they can be converted by hepatic lipase into mature LDL particles. These EL-remodeled VLDL remnants are then cleared directly by hepatocytes via multiple redundant non-LDLR remnant receptors—including LRP1 (LDL receptor-related protein 1), SDC1 (syndecan-1 heparan sulfate proteoglycans), and SR-B1—cutting off the intravascular production rate of LDL at its VLDL precursor source.

Convergent Evolutionary Relaxation in Diving and Hibernating Mammals

Across comparative vertebrate genomics, ANGPTL3 exhibits striking signatures of relaxed purifying selection (HyPhy RELAX selection intensity parameter k < 0.08, with local d_N/d_S ω rising toward 0.75–0.92 across the coiled-coil domain) or pseudogenic inactivation in lineages subjected to prolonged fasting, hibernation, or extreme breath-hold diving (such as cetaceans, pinnipeds, and hibernating bears/bats). In marine mammals that consume lipid-dense prey (~40%–50% milk and blubber fat) while preventing intravascular hypertriglyceridemic sludge during diving ischemia, dampening ANGPTL3 activity ensures constitutive high-flux LPL clearance of circulating TRLs without causing hepatic steatosis—mirroring the benign phenotype of human FHBL2 knockouts.

Clinical Therapeutic Modalities (Monoclonal Antibodies, GalNAc-siRNA, and Base Editing)

Building on human knockout validation, four therapeutic platforms target ANGPTL3 (and differs functionally from earlier antisense oligonucleotides like vupanorsen, which was discontinued due to intracellular hepatocyte ASO off-target transaminase elevations not seen with extracellular mAbs or siRNAs):

1. Evinacumab (Evkeeza): A fully human IgG4 monoclonal antibody (Regeneron, FDA-approved 2021) that binds the N-terminal coiled-coil domain of ANGPTL3 (both free ANGPTL3 and the ANGPTL3/8 complex), administered 15 mg/kg IV every 4 weeks for adults and pediatric patients (≥5 years) with HoFH.
2. Zodasiran (ARO-ANG3) and Solbinsiran (LY3561774): Hepatocyte-targeted GalNAc-siRNA therapeutics that degrade ANGPTL3 mRNA via RISC, achieving dose-dependent −74% to −84% reductions in circulating ANGPTL3, −54% to −63% reductions in triglycerides, −36% to −44% reductions in non-HDL-C, and a relative −24% reduction in hepatic fat fraction (ARCHES-2 trial) with quarterly subcutaneous dosing and zero hepatic steatosis.
3. VERVE-201 (CRISPR Adenine Base Editing): A GalNAc-LNP formulation delivering an adenine base editor to permanently inactivate hepatic ANGPTL3 via a single intravenous infusion for refractory hypercholesterolemia and HoFH.

Why can ANGPTL3 inhibitors lower LDL cholesterol in HoFH patients who have zero functional LDL receptors, whereas PCSK9 inhibitors cannot?

PCSK9 inhibitors work by preventing lysosomal degradation of the hepatocyte LDLR; if a patient has two receptor-null LDLR mutations (HoFH), there is no functional LDLR to rescue. By contrast, inhibiting ANGPTL3 derepresses endothelial lipase (EL), which strips surface phospholipids from circulating VLDL particles. These phospholipid-depleted VLDL remnants are rapidly cleared from the bloodstream by non-LDLR hepatocyte remnant receptors (LRP1 and syndecan-1 heparan sulfate proteoglycans) before they can mature into LDL, reducing LDL-C production by ~47% independent of LDLR.

Why does ANGPTL3 inhibition lower HDL-C alongside triglycerides and LDL-C without increasing cardiovascular risk?

Because ANGPTL3 normally inhibits endothelial lipase (EL)—the primary phospholipase that catabolizes HDL phosphatidylcholine—loss or pharmacological inhibition of ANGPTL3 increases EL activity and lowers plasma HDL-C by ~30% to 60%. However, human genetic studies show that this HDL-C reduction reflects accelerated HDL remodeling and reverse cholesterol flux rather than impaired atheroprotection; because atherogenic apoB-containing particles (VLDL remnants and LDL) are simultaneously reduced by 30% to 65%, ANGPTL3 loss-of-function carriers enjoy a ~34%–41% lower lifetime risk of coronary artery disease.

How does ANGPTL8 reverse the tissue specificity of ANGPTL3 and ANGPTL4 between fasting and feeding?

During fasting, adipose tissue secretes high levels of ANGPTL4 homodimers, which inhibit local adipose LPL and divert circulating triglycerides to heart and skeletal muscle for oxidative phosphorylation. After a meal, insulin triggers ANGPTL8 synthesis: in the liver, ANGPTL8 binds ANGPTL3 to form the endocrine ANGPTL3/8 complex (~50–100× more potent at inhibiting muscle LPL), while in adipose tissue ANGPTL8 binds ANGPTL4 to form an ANGPTL4/8 complex (~75% less potent than ANGPTL4 alone), shunting postprandial lipids into adipose storage.

1. Musunuru, K., Pirruccello, J. P., Do, R., et al. (2010). "Exome sequencing, ANGPTL3 mutations, and familial combined hypolipidemia." New England Journal of Medicine, 363(23), 2220–2227. https://doi.org/10.1056/NEJMoa1002926
2. Dewey, F. E., Gusarova, V., Dunbar, R. L., et al. (2017). "Genetic and pharmacologic inactivation of ANGPTL3 and cardiovascular disease." New England Journal of Medicine, 377(3), 211–221. https://doi.org/10.1056/NEJMoa1612790
3. Raal, F. J., Rosenson, R. S., Reeskamp, L. F., et al. (2020). "Evinacumab for homozygous familial hypercholesterolemia." New England Journal of Medicine, 383(8), 711–720. https://doi.org/10.1056/NEJMoa2004215
4. Adam, R. C., Mintah, I. J., Alexa-Braun, C. A., et al. (2020). "Angiopoietin-like protein 3 governs LDL-cholesterol levels through endothelial lipase-dependent VLDL clearance." Journal of Lipid Research, 61(9), 1271–1286. https://doi.org/10.1194/jlr.RA120000888