Science - USA (2021-12-10)

PI(3,4)P 2 levels at the midbody as well as elon-
gated intercellular bridges (Fig. 4, F and G,
and fig. S14A). Removal of TSG101 together
with PI3K-C2asynergized and almost com-
pletely abolished VPS36 recruitment to the

midbody (fig. S13G). Thus, PI3K-C2aÐderived

PI(3,4)P 2 facilitates VPS36 recruitment and

ESCRT-III assembly. In agreement, PI4KA

inhibition reduced the ESCRT-III subunit

CHMP4B at the midbody (fig. S14B), but supres-

sion of CHMP4B had no effect on VPS36 lo-

calization (Fig. 4E), confirming that VPS36

is recruited before ESCRT-III during abscis-

sion. At the midbody, depletion of either PI3K-

C2aor VPS36 reduced CHMP4B enrichment

Gulluniet al.,Science 374 , eabk0410 (2021) 10 December 2021 6 of 14

1.25

GFP-VPS36

RFP-α-tubulin

DNA VPS36 PI(3,4)P2

midbody

C D

1um 2um

00:15

00:05 00:25

00:10 00:30

00:35

00:00 00:20

E

150 -

100 -

20 -

Head Lens

PI3K-C2α

ALIX

α-Cry

AB

H

CHMP4B norm fluorescence

intensity at midbody (AU)

0

ctrl

PI3K-C 2 α

siRNA

VPS36

siRNA

1.0 2.0 3.0

***

***

α-tubulin CHMP4B Merge

Midbody section

I

0

0.5

1.0

1.5

2.0

2.5

3.0

(^) e
cn
ec
se
ro
ulf
mr
on
(^6)
(^3) S
PV
ody (AU)b
di
m (^) t
a
yti
sn
et
ni

• wt kd c3

***

3.5

GFP-C2α

siRNA C2α ++++

--

** **

DNA PI3K-C2α

lens

DNA C2α Merge

lens lens

DNA ALIX DNA ALIX Merge

lens

lens

VPS36

Vinculin

50 -

100 -

1

0.0625

0.5

8

Log2 Head/Lens

2

0.25

ALIX

PI3K-C2

α

VPS36

4

0.125

**

**

lens

VPS36 norm fluorescence

intensity at midbody (AU)

0

ctrl

PI 3 K-C2α

siRNA

CHMP4B

siRNA

1.0 2.0 3.0

***

α-tubulin VPS36 Merge

Midbody section

F

0

0.75

1.00

1.25

1.50

0.25

0.50

(^) e
cn
ec
se
ro
ulf
mr
on
(^6)
3
S
P
V
)
U
A(
yd
ob
di
m (^) t
a
yti
sn
et
ni
+/+ --/
Fam-I Fam-II
--/ --/ --/
α-tubulin VPS36

---

--/

---

Fam-III

---

α-tubulin PI(3,4)P2
0
0.50
0.75
1.00
1.50
0.25
PI(3,4)P2 norm fluorescence
intensity at midbody (AU)

---

G
+/+ --/
Fam-I Fam-II
--/ --/ --/ --/
Fam-III
Fig. 4. PI3K-C2a/VPS36 axis is an ALIX alternative pathway to recruit
CHMP4B to the midbody in lens.(A) Immunoblot analysis (left) and
quantification plot (right) showing Log2 of head/lens protein level for PI3K-C2a,
ALIX, and VPS36 in adult zebrafish. (B) Immunofluorescence showing (top) ALIX
or (bottom) PI3K-C2aexpression in 72-hpf zebrafish lens. (C) Immuno-
fluorescence staining showing colocalization between endogenous VPS36 and
PI(3,4)P 2 at the midbody ring. (D) Snapshots taken from time-lapse imaging of
GFP-VPS36 during cytokinesis in HeLa cells stably expressing red fluorescent
protein (RFP)–a-tubulin. Arrowheads indicate enrichment of VPS36 at the
abscission site. (E) Quantification of VPS36 levels at midbody in control and
siRNA-treated cells.n≥200 cells, mean ± SD. (FandG) Immunofluorescence
staining and quantification of endogenous (F) VPS36 and (G) PI(3,4)P 2 levels at the
midbody in fibroblasts derived from patients with homozygous deletion of
PI3K-C2a. +/+ genotypes were pulled together.n≥50 cells, mean ± SD.
(H) Quantification of CHMP4B levels at midbody in control and siRNA-treated
HeLa cells.n≥120 cells, mean ± SD. (I) Quantification of VPS36 levels at
midbody in control and siRNA-treated cells upon transfection of siRNA-resistant
WT, KD, and PI(3)P–producing (C3) forms of PI3K-C2a.n≥30 cells, mean ± SD.
If not previously specified, all results are shown as mean or representative picture
of at least three independent experiments ± SEM. **P< 0.01; ***P< 0.001.
RESEARCH | RESEARCH ARTICLE