rs117896735 — INPP5F

Intronic variant in the INPP5F/BAG3 locus associated with increased risk of REM sleep behavior disorder — an early marker of Lewy body neurodegeneration

See your personal result for INPP5F

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.

INPP5F and the Protein-Clearance Gateway to REM Sleep Behavior Disorder

REM sleep behavior disorder (RBD)11 REM sleep behavior disorder (RBD)
During RBD, the normal motor paralysis of REM sleep fails, causing people to act out their dreams — punching, kicking, yelling. It is the strongest single predictor of Parkinson's disease and Lewy body dementia: over 80% of those diagnosed with idiopathic RBD eventually convert to a synucleinopathy within 14 years. is no ordinary sleep disturbance. It sits at the far upstream end of the Parkinson's disease continuum — years before motor symptoms emerge and dopamine neurons begin to die. This variant, rs117896735, sits in an intron of the INPP5F gene and emerged from the first major RBD GWAS as one of only five loci reaching genome-wide significance out of the entire human genome.

The locus spans two genes: INPP5F, which encodes an enzyme called Sac222 Sac2
INPP5F is also known as Sac2 — a member of the SAC phosphatase family that dephosphorylates phosphatidylinositol 4-phosphate on endosomal membranes, and BAG3, a co-chaperone that sits at the intersection of two major protein-disposal systems.

The Mechanism

INPP5F/Sac2 controls the phosphoinositide composition of endosomal membranes. It functions as a PI4P phosphatase33 PI4P phosphatase
Phosphatidylinositol 4-phosphate is a lipid signal that recruits trafficking proteins to endosomes; its timely removal is required for endosomal cargo sorting and recycling, dephosphorylating PI(4)P in concert with OCRL to regulate endocytic recycling.

Immediately neighboring INPP5F, BAG344 BAG3
BCL2-associated athanogene 3 — a co-chaperone that bridges Hsp70 and autophagy receptor p62/SQSTM1 to direct misfolded client proteins into autophagosomes rather than the proteasome orchestrates the switch from proteasomal degradation to selective macroautophagy when the proteasome is overwhelmed by misfolded proteins — precisely the scenario in aging neurons accumulating alpha-synuclein. BAG3 forms a trimeric complex with Hsp70 and p62/SQSTM1, colocalizing with autophagy markers in brain tissue. When BAG3 is overexpressed, alpha-synuclein clearance via macroautophagy increases in an Atg5-dependent manner; when BAG3 is reduced, toxic alpha-synuclein accumulates.

rs117896735 is an intronic variant, so it does not alter either protein's coding sequence.

The Evidence

The landmark genome-wide association study by Krohn et al. 202255 Krohn et al. 2022
Meta-analysis of iRBD + PD-with-probable-RBD cohorts spanning multiple European and North American sites combined 2,843 RBD cases with 139,636 controls, identifying rs117896735 with an odds ratio of 1.80 (95% CI 1.48–2.19, p=4.70×10⁻⁹). The A allele effect was consistent across the isolated RBD cohort (OR 1.88) and the PD-with-probable-RBD cohort (OR 1.57), supporting genuine association rather than a population-specific artefact. The A allele frequency in the GWAS was approximately 2% — enriched above the gnomAD European frequency of ~1.4%.

The four other genome-wide significant RBD loci map to SNCA (alpha-synuclein itself), GBA (lysosomal glucocerebrosidase), TMEM175 (lysosomal K⁺ channel), and SCARB2 (lysosomal membrane protein) — all genes with direct roles in lysosomal protein degradation.

On the BAG3 side, Cao et al. 201766 Cao et al. 2017
Cell-based and transgenic mouse study demonstrating BAG3's role in Atg5-dependent selective macroautophagy of alpha-synuclein directly demonstrated that BAG3 modulates alpha-synuclein quality control in dopaminergic neurons. Ying et al. 202277 Ying et al. 2022
Mouse striatal BAG3 overexpression reduced dopaminergic neuron loss and microglial activation following inflammatory challenge extended this to show BAG3 also suppresses NLRP3 inflammasome activation, adding a neuroinflammatory dimension to the locus's relevance.

Practical Implications

The A allele does not cause RBD; it shifts the probability. RBD is itself a prodromal state for synucleinopathies with ~80% conversion rates over 14 years. The practical value of knowing this genotype lies in motivating early surveillance.

Genotype Interpretations

What each possible genotype means for this variant.

GG Standard Clearance Normal

No increased RBD risk from this variant

You carry two copies of the common G allele at rs117896735. This is the population-majority genotype, present in approximately 96% of people of European descent. You do not carry the A allele that confers elevated RBD risk at this locus.

AG Elevated RBD Risk High Risk

One copy of the RBD risk allele — moderately increased risk of REM sleep behavior disorder

rs117896735 sits in an intron of INPP5F, the gene encoding Sac2/INPP5F, a PI4P phosphatase that regulates endosomal trafficking. Its neighboring gene BAG3 is a co-chaperone that directs misfolded proteins — including alpha-synuclein — toward selective macroautophagy when the proteasome is saturated.

The evidence for OR=1.80 comes from a meta-analysis of 2,843 RBD cases and 139,636 controls. The effect was consistent across isolated RBD and PD-with-probable-RBD cohorts, and the locus sits alongside four other genome-wide significant hits all mapping to the same lysosomal/ autophagic degradation axis (SNCA, GBA, TMEM175, SCARB2).

The A allele does not guarantee any outcome. Absolute lifetime RBD risk even for carriers remains relatively low, but the downstream conversion risk to synucleinopathy makes early awareness actionable.

AA High RBD Risk High Risk

Two copies of the RBD risk allele — substantially elevated risk of REM sleep behavior disorder

The AA genotype represents the most extreme end of the rs117896735 risk spectrum.

At this rarity (below 0.1% of population), formal clinical data on the AA genotype specifically is limited — the GWAS used an additive model across all carriers. Nevertheless, the strong, genome-wide significant per-allele effect and the clear biological plausibility of the locus make proactive monitoring highly appropriate.