rs2294918 — PNPLA3 PNPLA3 K434E (E434K)

Second PNPLA3 missense variant that reduces hepatic PNPLA3 mRNA and protein expression by ~50%, independently associated with NAFLD and elevated liver enzymes, and modifies the risk conferred by the co-located I148M variant (rs738409).

See your personal result for PNPLA3

Upload your DNA data to find out which genotype you carry and what it means for you.

Upload your DNA data

Works with 23andMe, AncestryDNA, and other DNA test exports. Results in under 60 seconds.

PNPLA3 K434E — The Volume Control for Liver Damage Risk

Most people who know about PNPLA3 genetics know about rs738409 (I148M) — the strongest common genetic risk factor for fatty liver disease. But PNPLA3 harbours a second functionally important variant, rs2294918, that acts not by changing the protein's enzyme activity but by changing how much of the protein the liver makes.

The rs2294918 A allele encodes lysine (K) at protein position 434 instead of the more common glutamic acid (E). The 434K allele causes the liver to produce approximately 50% less PNPLA3 messenger RNA and protein — effectively turning the gene's volume down by half. This expression reduction has two opposing clinical consequences: it independently predisposes carriers to steatosis and elevated liver enzymes, yet it substantially dampens the risk that the co-located I148M variant (rs738409) would otherwise confer.

The Mechanism

Donati et al.11 Donati et al.
Donati B et al. The rs2294918 E434K variant modulates patatin-like phospholipase domain-containing 3 expression and liver damage. Hepatology, 2016 established that rs2294918 does not alter PNPLA3 enzymatic activity itself — the 434K protein hydrolyses triglycerides at the same rate as 434E. Instead, the A allele reduces hepatic PNPLA3 mRNA and protein levels by approximately 50%. This matters because the hepatic damage associated with I148M (rs738409 G allele) is now understood to operate through a dominant-negative mechanism: the dysfunctional 148M protein accumulates on lipid droplets and inhibits other lipases that would otherwise clear hepatic triglycerides. When the total amount of PNPLA3 protein is halved by 434K, there is simply less dysfunctional 148M protein available to exert this dominant-negative lipase inhibition — reducing the net liver damage.

The expression effect is also clinically relevant in isolation. Even without I148M, reduced PNPLA3 expression appears to shift hepatic lipid handling in ways that predispose to steatosis, as evidenced by the overrepresentation of the 434K allele in NAFLD patients independent of the I148M background.

The Evidence

Donati et al. 201622 Donati et al. 2016 studied 142 early-onset NAFLD patients and 100 controls, then validated in 1,447 subjects. The 434K (A allele) was overrepresented in NAFLD patients (adjusted P=0.01) and was independently associated with serum ALT (P=0.044). Crucially, haplotype analysis revealed the pivotal modifier role: the 148M-434E haplotype (I148M + G at rs2294918) was strongly associated with steatohepatitis and fibrosis (P<0.0001), while the 148M-434K haplotype (I148M + A at rs2294918) showed no association with histological liver damage (P>0.9). The interaction was statistically significant (P=0.006).

The hepatocellular carcinoma dimension emerges from the G-G haplotype: Arreola Cruz et al. 202533 Arreola Cruz et al. 2025 studied 173 biopsy-confirmed HCC cases and 346 controls in Mexico, finding that the G-G combination of rs738409 and rs2294918 (I148M risk allele + 434E, i.e., high expression of the dysfunctional protein) was associated with an OR of 2.2 (95% CI 1.7–2.9) for HCC development. A 2017 Han Chinese case-control study44 2017 Han Chinese case-control study
Gao et al. Scand J Gastroenterol, 2017 of 2,410 subjects found the rs2294918 AG heterozygous genotype associated with HBV-related HCC (OR 1.872, 95% CI 1.256–2.792, p=0.002).

The therapeutic implications are significant: the Schwartz 2020 review noted that rs2294918 provides natural proof-of-concept that reducing PNPLA3 expression is a viable hepatoprotective strategy — an observation that has informed antisense oligonucleotide development targeting PNPLA3.

Practical Actions

For carriers of the A allele (434K), the primary concern is the independent NAFLD risk associated with reduced PNPLA3 expression. Monitoring liver enzymes and keeping hepatic fat burden low through saturated fat restriction are the most directly supported interventions. The A allele's HCC association in the context of HBV and hepatic inflammation warrants vigilance regarding alcohol, which further amplifies liver inflammation and cancer risk.

Carriers of two G alleles (434E/434E) who also carry the I148M risk allele at rs738409 are in the highest-risk group for liver damage and HCC — the G-G haplotype study confirms this is the combination driving the worst outcomes.

Interactions

The dominant interaction at this locus is with rs738409 (PNPLA3 I148M). The risk haplotype for HCC and steatohepatitis is rs738409-G (I148M) combined with rs2294918-G (434E) — i.e., high expression of the dysfunctional I148M protein. By contrast, the combination of rs738409-G (I148M) with rs2294918-A (434K) substantially attenuates liver damage because the 434K allele halves total PNPLA3 expression, reducing the dominant-negative lipase inhibition of I148M. This interaction is a strong candidate for a compound action: carriers of the I148M risk allele who also carry the 434K allele need different guidance than carriers of I148M with 434E.

GCKR rs780094 and MBOAT7 rs641738 are pathway partners that additively influence hepatic fat accumulation through de novo lipogenesis and phospholipid remodeling respectively — cohort studies have examined cumulative risk allele burden across these loci including rs2294918.

Nutrient Interactions

saturated fat altered_metabolism

Genotype Interpretations

What each possible genotype means for this variant.

GG High PNPLA3 Expression Normal

Common genotype — normal PNPLA3 expression, standard liver fat baseline

You carry two copies of the G allele (434E), the most common variant at this position, found in approximately 41% of people globally and about 36% of Europeans. Your liver produces PNPLA3 protein at the typical level. In isolation, this genotype does not increase liver fat risk through this variant. However, if you also carry the I148M risk allele (rs738409 G allele), your risk profile is determined primarily by that variant — GG here means more of the dysfunctional I148M protein, which is the combination most strongly linked to steatohepatitis and hepatocellular carcinoma in multiple cohort studies.

AG Partial Expression Reduction Intermediate

One copy of the expression-reducing A allele — modestly lower PNPLA3 levels

The AG heterozygous state means roughly half your hepatic PNPLA3 alleles carry the expression-reducing 434K variant. The net effect is a partial reduction in PNPLA3 protein levels — enough to alter lipid droplet dynamics in hepatocytes and produce the modest, independent NAFLD and ALT associations documented in the Donati 2016 study.

The HCC signal observed in the Gao 2017 Chinese cohort for AG carriers (OR 1.872, p=0.002) warrants attention, particularly for individuals who carry hepatitis B virus, have established liver disease, or have significant alcohol exposure — the contexts where PNPLA3 variants most potently influence HCC risk.

AA Reduced PNPLA3 Expression High Risk

Two A alleles — ~50% lower hepatic PNPLA3 expression with elevated NAFLD and liver enzyme risk

The mechanism for independent NAFLD susceptibility in AA carriers is distinct from I148M. While I148M creates a dysfunctional protein that accumulates on lipid droplets and inhibits other lipases (a dominant-negative effect), the AA genotype reduces total PNPLA3 expression. Reduced PNPLA3 appears to impair normal hepatic lipid droplet remodelling capacity, as PNPLA3 at normal levels contributes to triglyceride mobilisation from hepatocytes. The net effect is steatosis risk through reduced clearance capacity rather than through lipase inhibition.

Clinically, this distinction matters because the therapeutic implications differ from I148M. Dietary interventions that reduce hepatic fat substrate (saturated fat, refined carbohydrates driving de novo lipogenesis) are the most evidence-based approach, alongside regular liver function monitoring to detect steatosis early.