Skip to content

Romeo S, Kozlitina J, Xing C, et al. Genetic variation in PNPLA3 confers susceptibility to nonalcoholic fatty liver disease. Nat Genet. 2008;40(12):1461-5. PMID 18820647

One-paragraph summary

A genome-wide association scan of 9,229 nonsynonymous sequence variants was run in a multi-ancestry population of Hispanic, African American and European American individuals, with hepatic fat content measured directly rather than by proxy. One allele stood out: PNPLA3 rs738409[G], encoding the I148M substitution, was strongly associated with increased hepatic fat (p = 5.9 × 10⁻¹⁰) and with hepatic inflammation (p = 3.7 × 10⁻⁴). Hepatic fat content was more than twofold higher in rs738409[G] homozygotes than in non-carriers. The allele was most common in Hispanic Americans, the ancestry group with the highest NAFLD prevalence. Resequencing identified a second allele, rs6006460[T] (S453I), associated with lower hepatic fat and more common in African Americans, the group at lowest risk. The paper therefore explained a substantial part of both inter-individual and between-population variation in liver fat with two coding variants in a single gene.

Key findings

  • rs738409[G] (I148M): hepatic fat p = 5.9 × 10⁻¹⁰; hepatic inflammation p = 3.7 × 10⁻⁴.
  • Hepatic fat >2-fold higher in GG homozygotes than non-carriers.
  • Allele frequency tracks population risk: highest in Hispanics, protective variant more common in African Americans.
  • Identified a candidate mechanism for ancestry differences in NAFLD prevalence that had previously been attributed to environment alone.

Limitations

  • Cross-sectional, with hepatic fat as the phenotype — not steatohepatitis, fibrosis or clinical outcome.
  • Association with ancestry-linked prevalence is inference from allele frequency, not a formal decomposition of population variance.
  • The mechanism of I148M was unknown at the time and remains only partly resolved.
  • Modest sample by contemporary GWAS standards; the locus was large enough to be found anyway.

Why it matters

It opened the genetic era of this disease and remains its dominant common variant. Everything downstream follows: the histology-anchored GWAS confirming PNPLA3 for steatosis, fibrosis and NAS at genome-wide significance (Anstee 2020, PMID 32298765); the 17-locus multi-ancestry meta-analysis in which top-decile genetic risk carries 2.5–6-fold increased risk of NAFLD, cirrhosis and HCC (Chen 2023, PMID 37709864); reclassification of indeterminate-FIB-4 patients into a high-risk band by genotype plus diabetes (Chen 2023, PMID 36758837); divergence of fibrosis trajectories by GG genotype from age 44 (Díaz 2025, PMID 40334848); and the multiplicative interaction with female sex mediated by an ER-α binding site in a PNPLA3 enhancer (Cherubini 2023, PMID 37749332). Most consequentially, the variant became a drug target: AZD2693, a liver-targeted antisense oligonucleotide, achieved 89% hepatic mRNA knockdown with placebo-corrected liver-fat reductions of 7.6–12.2% (Armisen 2025, PMID 39798707), and siRNA programmes are in phase 2. Eighteen years on, genotype still appears in no clinical risk-stratification pathway.

Cited by wiki pages

  • genetics.md
  • overview.md