Endosomal lipid flippases and their related diseases

Abstract

 

In eukaryotic cell membranes, phospholipids are asymmetrically distributed between the two leaflets of the lipid bilayer. For example, the extracellular leaflet of the plasma membrane (PM) is enriched with phosphatidylcholine and sphingomyelin, while the cytosolic leaflet of the PM is enriched with phosphatidylserine (PS) and phosphatidylethanolamine. The asymmetric distribution of PS in the PM is crucial for cell life, since PS in the extracellular leaflet of the PM is recognized as an “eat-me” signal by phagocytes. Inside the cells, a high PS concentration in the cytosolic leaflet of the PM is essential to facilitate various cellular events through the recruitment of signaling molecules such as protein kinase C and Akt.

The asymmetric distribution of phospholipids is believed to be generated in part by phospholipid translocases, or “flippases.” The proteins responsible for flippase activity are type IV P-type ATPase (P₄-ATPases). P-type ATPases are multispan transmembrane pumps that use ATP hydrolysis as an energy source. P-type ATPases undergo autophosphorylation of a conserved aspartate residue during the catalytic cycle, hence the designation of “P”-type. P₄-ATPases are unique in that they are phospholipid transporters whereas other types of P-type ATPases are ion transporters.

The human genome contains 14 P₄-ATPases, and mutations in some P₄-ATPases cause inherited genetic diseases. For example, mutations in ATP8B1 are associated with intrahepatic cholestasis and also cause hearing loss. Mutations in ATP8A2 are associated with a severe neurological disorder characterized by cerebellar ataxia, mental retardation, and dysequilibrium syndrome (CAMRQ).¹ Despite the accumulating evidence highlighting the physiological importance of P₄-ATPases, how dysfunction of P₄-ATPases causes diseases is poorly understood.

In a recent study, we revealed the cellular function of the P₄-ATPase, ATP8A1.² ATP8A1 localizes at recycling endosomes (REs), an organelle that functions in recycling transport of internalized molecules back to the PM, thus defining the amount of proteins at the PM. PS is most concentrated in REs among intracellular organelles and we roughly estimated that 70 and 30% of PS are localized in the cytosolic and the luminal leaflets of RE membranes, respectively.² ATP8A1 generates the asymmetric transbilayer distribution of PS at REs. The knockdown of ATP8A1 halted recycling traffic from REs to the PM. At the mechanistic level, we found that EHD1, a dynamin-like membrane fission protein, lost its RE localization upon ATP8A1 knockdown and EHD1 knockdown also blocked recycling traffic. EHD1 bound PS in vitro and lost its membrane localization in cells that are defective in PS synthesis. Thus, we propose that PS flipping by ATP8A1 recruits EHD1 to RE membranes, thereby regulating the recycling traffic from REs to the PM (Fig. 1).

Figure 1.

Model of flippase-related diseases. Under normal conditions, flippases (e.g., ATP8A1 and ATP8A2) translocate PS to the cytosolic leaflet of RE membranes. PS recruits EHD1 to REs, and then EHD1 participates in the fission of RE membranes to generate transport vesicles that contain cell surface receptors. In flippase-dysfunctional situations, PS levels in the cytosolic leaflet of REs would be low. This impairs the PS/EHD1/membrane traffic axis, leading to a lower abundance of cell surface receptors that are critical for responses to extracellular ligands.

ATP8A2 is a tissue-specific ATP8A1 paralogue. We found that a CAMRQ-causative mutation of ATP8A2 (I376M) lost its ATPase and flippase activity toward PS. ATP8A2 is not endogenously expressed in COS-1 cells. Interestingly, the phenotype that was caused by the loss of ATP8A1 in COS-1 cells, was restored by the exogenous expression of wild-type ATP8A2, but not I376M mutant ATP8A2. Moreover, cortical neurons prepared from ATP8A2 knockout mice showed lower abundance of transferrin receptors at the PM. Together, these results indicate that ATP8A2 functions in the recycling traffic in neurons, and that CAMRQ may result from the defect in recycling of important neurological receptor proteins from REs to the PM. One possible candidate protein is very low-density lipoprotein receptor (VLDLR). VLDLR is a receptor for reelin, an extracellular protein that guides neuronal migration in the cerebral cortex and cerebellum. VLDLR circulates between the PM and endosomes (possibly REs) by recycling traffic.³ Significantly, mutations in VLDLR gene are also linked to CAMRQ.^4,5 Therefore, impaired recycling traffic of VLDLR to the PM in neurons with dysfunctional ATP8A2 (I376M) may cause lower expression of VLDLR at the PM, leading to reduced reelin signaling, abnormal neuronal development, and neurological disorder.

dATP8B, a P₄-ATPase in Drosophila melanogaster was recently reported to cause an impaired response to cVA pheromone (a sex-specific social cue) and mislocalization of the pheromone receptor in cVA-sensing neurons.⁶ The impaired response to the pheromone in dATP8B mutant was rescued by expressing bovine ATP8A2. Therefore, from insects to mammals, phospholipid flippases may define the localization of neuronal receptors to the PM.

Lastly, our findings may explain the phenotype of ATP8A1 knockout mice.⁷ ATP8A1 knockout mice are vital but show deficiencies in hippocampus-dependent learning. Hippocampus-dependent learning involves modification of synaptic strength, and one cellular mechanism for tuning synaptic strength is long-term potentiation (LTP). During LTP, REs supply glutamate receptors to the post-synaptic membrane.⁸ Therefore, we speculate that impaired glutamate receptor traffic from REs to the post-synaptic membranes during LTP may underlie the deficiency in learning in ATP8A1 knockout mice. In agreement with this hypothesis, the dominant-negative form of EHD1 inhibits glutamate receptor traffic during LTP.⁸

Many P₄-ATPases are expressed in the Golgi/endosomes and the PM. We expect that they contribute redundantly to the phospholipid asymmetry and membrane traffic through organelles. Simultaneous ablations of P₄-ATPases may dissect their roles and will give more insight into flippase-mediated cellular processes and -related diseases.