Four subtypes of disease-causing missense mutations underlie pathogenic protein interactions in neurodegenerative VPS13A disease
Xing Lin, Yuta Ryoden, Chigure Suzuki, Hiroyuki Ishikawa, Takaharu Sakuragi, Yasuo Uchiyama, Shigekazu Nagata
Xing Lin, Yuta Ryoden, Chigure Suzuki, Hiroyuki Ishikawa, Takaharu Sakuragi, Yasuo Uchiyama, Shigekazu Nagata
View: Text | PDF
Research Article Genetics Neuroscience

Four subtypes of disease-causing missense mutations underlie pathogenic protein interactions in neurodegenerative VPS13A disease

Abstract

VPS13A is an intracellular lipid transfer protein comprising more than 3,000 amino acids. Mutations in human VPS13A cause VPS13A disease, a neurodegenerative disorder that affects movement and cognition. VPS13A forms a complex with the membrane protein XK to mediate ATP-induced phospholipid scrambling in the plasma membrane. Here, we established a mouse cell system expressing full-length mouse VPS13A and examined its interaction with XK. Mutational analysis revealed that VPS13A binds to XK through a C-terminal β-strand that interacts with a β-hairpin in the central region of XK, an interaction essential for scramblase activity. The XK paralog XKR2, which contains a similar β-hairpin structure, also associates with VPS13A and supports phospholipid scrambling. We analyzed 10 mouse VPS13A variants corresponding to human patient mutations and classified them into 4 groups: (a) L67P, I90K, and W2453R, which showed reduced expression; (b) A1091P and M3080R, which were normally expressed but lacked scramblase activity; (c) S1446P, Q2689H, Y2713C, and R3084H, which modestly impaired expression or activity; and (d) I2763R, which altered cell size and disrupted ER independently of XK. These findings define the VPS13A–XK interaction interface, clarify the functional impact of disease-causing mutations, and reveal an unexpected gain-of-function mutation of a VPS13A variant.

Authors

Xing Lin, Yuta Ryoden, Chigure Suzuki, Hiroyuki Ishikawa, Takaharu Sakuragi, Yasuo Uchiyama, Shigekazu Nagata

×

Figure 1

Requirement of the C-terminal domain of mouse VPS13A for ATP-induced PtdSer exposure.

Options: View larger image (or click on image) Download as PowerPoint
Requirement of the C-terminal domain of mouse VPS13A for ATP-induced Ptd...
(A) AlphaFold 3-predicted mouse VPS13A–XK structure. The structure of the VPS13A-XK complex was predicted using AlphaFold 3 (27) with interface predicted template modeling score (ipTM) = 0.72 and predicted template modeling score (pTM) = 0.54, and viewed at 2 angles. VPS13A is shown in dark cyan and XK in royal blue. The N-terminal chorein domain of VPS13A is shown in orange. The β-hairpin of XK is shown in red, and the 2 β-strands in the PH domain of VPS13A are shown in yellow. The red dotted lines indicate the predicted membrane bilayers. The region where β-hairpin of XK interacts with β-strands of the PH domain of VPS13A is enclosed by a square line. (B–D) Effect of VPS13A C-terminal deletion mutants on their interaction with XK and ATP-induced PtdSer exposure. WT, R3127Δ, and E3136Δ VPS13A were stably expressed in Vps13a–/–DKO-P2X7 cells. (B) Whole-cell lysates were separated by SDS-PAGE and analyzed by Western blotting using anti-VPS13A Ab. Red arrowhead indicates VPS13A; * indicates a nonspecific band. (C) Membrane fractions from DKO-P2X7, Vps13a–/–DKO-P2X7, and their transformants expressing none, WT, R3127Δ, or E3136Δ mouse VPS13A were separated by BN-PAGE and analyzed by Western blotting with anti-VPS13A (upper) or anti-XK (lower) Abs. Red arrowheads indicate the VPS13A–XK complex; the black arrowhead indicates XK. The band intensity for XK in WT-Vps13a–/–DKO-P2X7 is slightly lower than that observed in DKO-P2X7 with unknown reason. (D) DKO-P2X7, Vps13a–/–DKO-P2X7, and their transformants expressing none, WT, R3127Δ, or E3136Δ mouse VPS13A were treated with ATP. PtdSer exposure was assessed by flow cytometry using Cy5–annexin V and expressed as MFI. Data are presented as mean ± SD (bars) of at least 3 independent experiments. The statistical significance was determined using 1-way ANOVA with Dunnett’s multiple comparison test. ****P < 0.0001.