Major facilitator superfamily domain-containing 2a (Mfsd2a)
is
a sodium-dependent lysophosphatidylcholine cotransporter that plays
an important role in maintaining the integrity of the blood–brain
barrier and neurological function. Abnormal degradation of Mfsd2a
often leads to dysfunction of the blood–brain barrier, while
upregulation of Mfsd2a can retrieve neurological damage. It has been
reported that Mfsd2a can be specifically recognized and ubiquitinated
by neural precursor cell-expressed developmentally downregulated gene
4 type 2 (NEDD4-2) ubiquitin ligase and finally degraded through the
proteasome pathway. However, the structural basis for the specific
binding of Mfsd2a to NEDD4-2 is unclear. In this work, we combined
deep learning and molecular dynamics simulations to obtain a Mfsd2a
structure with high quality and a stable Mfsd2a/NEDD4-2-WW3 interaction
model. Moreover, molecular mechanics generalized Born surface area
(MM-GBSA) methods coupled with per-residue energy decomposition studies
were carried out to analyze the key residues that dominate the binding
interaction. Based on these results, we designed three peptides containing
the key residues by truncating the Mfsd2a sequences. One of them was
found to significantly inhibit Mfsd2a ubiquitination, which was further
validated in an oxygen-glucose deprivation (OGD) model in a human
microvascular endothelial cell line. This work provides some new insights
into the understanding of Mfsd2a and NEDD4-2 interaction and might
promote further development of drugs targeting Mfsd2a ubiquitination.