How Atherosclerosis
Actually Develops

This section provides detailed scientific context for researchers, clinicians, and health enthusiasts who want to understand the full evidence base.

The Historical Development of the Response-to-Retention Hypothesis

The Response-to-Retention (RtR) hypothesis was formally articulated by Kevin Jon Williams and Ira Tabas in their landmark 1995 paper in Arteriosclerosis, Thrombosis, and Vascular Biology. Building on decades of observations about lipid accumulation in arterial walls, they proposed that the subendothelial retention of ApoB-containing lipoproteins—not passive accumulation at sites of injury—is the key initiating event in atherogenesis.

The hypothesis was definitively validated by the 2002 Nature paper from Borén and colleagues, who created transgenic mice expressing proteoglycan-binding-defective LDL carrying the K3363E mutation in apoB100. Despite equivalent hyperlipidemia to control mice, these animals developed significantly less atherosclerosis, proving that retention via proteoglycan binding is mechanistically essential.

The 2017 European Atherosclerosis Society Consensus Panel, representing over 30 international experts, declared that evidence "unequivocally establishes that LDL causes ASCVD" and that subendothelial retention is "the key initiating process." No substantive scientific opposition to this mechanism exists in mainstream cardiovascular literature.

Molecular Details of Transcytosis

The discovery that LDL entry occurs via active transcytosis rather than passive diffusion resolved a long-standing biophysical paradox: how do 22-28 nm LDL particles cross an endothelial barrier with tight junctions of only 3-6 nm?

SR-B1 (Scavenger Receptor Class B Type 1): Originally characterized as an HDL receptor mediating reverse cholesterol transport in the liver, SR-B1's pro-atherogenic role in arterial endothelium was established by Huang et al. (Nature 2019). The mechanism involves binding of LDL to SR-B1's extracellular domain, recruitment of DOCK4 (Dedicator of Cytokinesis 4) via an 8-amino-acid cytoplasmic domain (residues 487-494), activation of the small GTPase Rac1, and subsequent actin-mediated internalization and transcellular transport.

Endothelial-specific SR-B1 knockout reduces LDL accumulation in the aorta by approximately 50% and proportionally reduces atherosclerosis—despite identical plasma cholesterol levels. Notably, HDL competes with LDL for SR-B1 binding, providing one mechanism for HDL's apparent protective effects.

ALK1 (Activin Receptor-Like Kinase 1): Identified through unbiased genome-wide RNAi screening by Kraehling et al. (Nature Communications 2016), ALK1 is a TGF-β type I receptor that binds LDL with lower affinity than LDLR (Kd ~200 nM vs. 7 nM) but crucially does not saturate at physiological LDL concentrations. This makes ALK1 the dominant transcytosis mediator under hypercholesterolemic conditions. A 2023 study confirmed that monoclonal antibody blockade of ALK1-LDL binding reduces transcytosis by ~50% and doubles the rate of lesion regression in mice.

Simultaneous knockdown of both SR-B1 and ALK1 reduces LDL transcytosis by approximately 70%, confirming that these two receptors account for the majority of LDL entry into the arterial wall.

The Biochemistry of Proteoglycan Binding

The retention step depends on electrostatic interactions between positively charged residues on ApoB-100 and negatively charged glycosaminoglycan (GAG) chains on arterial proteoglycans. Site-directed mutagenesis has identified "Site B" (residues 3359-3369) on ApoB-100 as the critical domain, enriched with arginine and lysine residues that facilitate binding.

The arterial wall actively modulates retention susceptibility. In atherosclerosis-prone regions (bifurcations, branch points) where hemodynamic shear stress is disturbed, vascular smooth muscle cells synthesize proteoglycans with hyperelongated chondroitin sulfate chains and increased sulfation patterns. These modifications can increase binding affinity for atherogenic lipoproteins by 5-10 fold, creating "sticky" zones in the arterial tree.

Importantly, this biochemical entrapment occurs in the absence of significant inflammatory infiltrate, contradicting the "firefighter" premise that LDL responds to pre-existing inflammation.

Quantifying the Relative Contributions

The INTERHEART study, a massive case-control study across 52 countries, provides the most robust population-level data for attributing risk to different factors. The ApoB/ApoA1 ratio showed a population attributable risk (PAR) of 49.2%—the highest of any single factor measured. Other major contributors included smoking (35.7%), psychosocial factors (32.5%), and abdominal obesity (20.1%).

However, PAR statistics can be misleading because they assume independent, additive effects. The mechanistic data clarifies the hierarchy: ApoB accounts for ~50% of the quantifiable risk but represents 100% of the mechanistic necessity. Without ApoB retention, the risk attributable to other factors (smoking, hypertension, inflammation) largely collapses because there is no plaque substrate to destabilize.

The correlation between LDL and inflammation (measured by hs-CRP) is notably weak at r ≈ 0.08, indicating these are largely independent pathways that can be targeted separately. The CANTOS trial proved inflammation's independent causal contribution by reducing cardiovascular events 15% with canakinumab (IL-1β inhibitor) without changing LDL. However, this effect size is smaller than that achieved by aggressive LDL lowering, and CANTOS showed no mortality benefit (possibly offset by increased infections).

The Tsimane Evidence in Detail

The Tsimane of Bolivia represent one of the most studied non-industrialized populations. Kaplan et al. (Lancet 2017) performed CT coronary angiography on 705 Tsimane adults aged 40-94 and found that 85% had a CAC score of zero, with only 3% having CAC >100. For comparison, in the US MESA cohort, only ~14% of adults over 45 have CAC = 0.

The paradox: Tsimane adults have chronically elevated hs-CRP, often exceeding 3 mg/L (the clinical threshold for high cardiovascular risk), due to endemic parasitic infections and recurrent illnesses. Yet they have among the lowest rates of atherosclerosis ever documented in any human population.

Their mean LDL-C is approximately 1.8-2.4 mmol/L—low by Western standards but not extraordinarily so. The key insight is that at these LDL levels, even chronic inflammation cannot generate atherosclerotic plaque. The inflammatory "fire" has no "fuel" to burn.

Similar findings come from the Hadza of Tanzania (mean LDL-C ~1.6 mmol/L) and from neonatal cord blood data (LDL-C 0.8-1.3 mmol/L in healthy full-term infants), suggesting that human physiology functions optimally at lipid levels far below the Western average of 3.0-3.4 mmol/L.

Animal Model Evidence

The claim that atherosclerosis cannot develop without elevated ApoB is strongly supported by animal data. Wild-type C57BL/6 mice have total cholesterol of 1.9-2.8 mmol/L, carried mainly in HDL, and do not develop atherosclerosis even when fed atherogenic diets for over 12 months—they develop only primitive fatty streaks at most.

Atherosclerosis research requires hypercholesterolemic models: ApoE-/- mice (cholesterol 10.3-15.5 mmol/L on chow diet) or LDLR-/- mice (which require high-fat diet to elevate cholesterol). In these models, inflammatory stimuli alone (e.g., inducing lupus-like conditions) cause arteritis but not typical atheroma unless cholesterol is also elevated.

Mullick et al. (J Lipid Research 2011) provided elegant dose-response evidence using ApoB antisense oligonucleotides in LDLR-/- mice. Reducing hepatic ApoB production by 60-90% reduced atherosclerotic lesions by 50-90% in a strictly linear dose-dependent manner. At the highest ASO dose, achieving LDL of 2.0 mmol/L, atherosclerosis was 87% lower than controls and 54% lower than even chow-fed animals.