14-3-3γがTMCC3に結合し、その局在を制御することで、小胞体ネットワークを形成する

概要

Kobe University Repository : Kernel
PDF issue: 2024-05-02

The 14-3-3γ isoform binds to and regulates the
localization of endoplasmic reticulum (ER)
membrane protein TMCC3 for the reticular
network of the ER

SAIHAS, SUHDA
(Degree)
博士（医学）

(Date of Degree)
2023-09-25

(Resource Type)
doctoral thesis

(Report Number)
甲第8719号

(URL)
https://hdl.handle.net/20.500.14094/0100485903
※ 当コンテンツは神戸大学の学術成果です。無断複製・不正使用等を禁じます。著作権法で認められている範囲内で、適切にご利用ください。

（課程博士関係）

学位論文の内容要旨

The 14-3-3γ isoform binds to and regulates the localization of
endoplasmic reticulum (ER) membrane protein TMCC3 for the
reticular network of the ER

14-3-3γ が TMCC3 に結合し、その局在を制御することで、
小胞体ネットワークを形成 する

神戸大学大学院医学研究科医科学専攻
膜動態学
（指導教員：匂坂 敏朗 教授）

SAIHAS SUHDA

Introduction
The endoplasmic reticulum (ER) is continuous membrane system composed
of ER sheets and ER tubules. The ER tubules are interconnected by three-way
junctions formed by conserved ER membrane protein, atlastins and lunapark.
Atlastin, dynamin-like GTPases, drive homotypic membrane fusion between the
two ER tubules, leading to the generation of three-way junctions. Lunapark is then
recruited to the nascent three-way junctions for stabilization. Furthermore, we have
demonstrated that TMCC3 (transmembrane and coiled-coil domain family 3) is ER
membrane protein that specifically localizes to the three-way junctions and binds
to atlastins.
Reticular network of ER tubules undergoes constant remodeling through
formation and loss of three-way junctions. We had previously demonstrated that
TMCC3 acts upstream of atlastins. TMCC3 localizes to three-way junctions
independently of atlastins, and promotes the formation of three-way junctions by
enhancing atlastins activity. However, it still remains unknown how the TMCC3
activity is regulated for formation of the reticular ER network.
The earlier study identified 14-3-3 proteins as TMCC3-binding proteins. It
is well known that 14-3-3 proteins interact with target proteins in phosphorylationdependent manner, thereby regulating various cellular processes. However, it still
remains unknown whether binding of 14-3-3 proteins to TMCC3 is involved in
regulation of the ER morphology. In this study, we characterize the interaction
between 14-3-3γ and TMCC3 in the context of regulation of the ER morphology
and demonstrate that 14-3-3 binding negatively regulates TMCC3 localization to
the three-way junctions for the formation of reticular ER network.
.
Results
1. Overexpression of 14-3-3γ reduces localization of TMCC3 to three-way
junctions
We set out to examine the effect of 14-3-3 proteins on localization of
TMCC3 to three-way junctions. HA-tagged 14-3-3γ (HA-14-3-3γ) or mCherry
was then transfected into U2OS cells stably expressing GFP-TMCC3 (U2OS-

1

GFP-TMCC3 cells), followed by immunostaining. The result showed that 60%
of cells overexpressing HA-14-3-3γ were lost GFP-TMCC3 puncta indicative
of three-way junctions whereas almost all of the cells overexpressing mCherry
retained the puncta, indicating that overexpression of HA-14-3-3γ reduces
localization of TMCC3 to three-way junctions.
To confirm the result, we examined whether overexpression of HA14-3-3γ affected the localization of endogenous TMCC3. HA-14-3-3γ or
control vector was transfected into U2OS cells, followed by immunostaining
with anti-TMCC3 pAb and anti-PDI mAb as an ER marker. Similar to the
previous result, peripheral puncta of endogenous TMCC3 were obviously
reduced in the cells overexpressing HA-14-3-3γ compared to the cells
transfected with control vector.
We next examined whether overexpression of 14-3-3γ affected the
protein level of TMCC3. HA-14-3-3γ or the control vector was transfected into
U2OS cells. The total cell lysates were subjected to immunoblotting with antiTMCC3 pAb and anti-GAPDH. The result showed that overexpression of HA14-3-3γ did not change the protein level of endogenous TMCC3.
2. Overexpression of 14-3-3γ decreases the number of three-way junctions
We reasoned that if 14-3-3γ negatively regulated TMCC3 localization,
overexpression of 14-3-3γ would affect the ER morphology. HA-14-3-3γ or
mCherry was transfected into U2OS cells, followed by immunostaining with
anti-PDI mAb as an ER marker. The number of three-way junctions as detected
by PDI staining was counted within 10 × 10 μm2 of the peripheral ER. In
agreement with the reasoning, the cells overexpressing HA-14-3-3γ
significantly decreased the number of three-way junctions relative to the cells
overexpressing mCherry.
To confirm the result, we also carried out the similar experiment
using ER-GFP, GFP fused to the ER signal sequence of calreticulin and KDEL
(ER retention signal) as an ER marker. HA-14-3-3γ or control vector was
transfected into U2OS cells, followed by transfection with ER-GFP. The cells

2

were then immunostained. Similar to PDI staining result, overexpression of
HA-14-3-3γ decreased the number of three-way junctions detected by ER-GFP.

3. TMCC3 binds to 14-3-3γ through the N-terminus
We sought to determine which region of TMCC3 was responsible for
binding to 14-3-3γ. We started by confirming that 14-3-3γ binds to TMCC3.
U2OS-GFP-TMCC3 cells were extracted with 1% TritonX-100 and subjected
to immunoprecipitation with anti-GFP pAb, followed by immunoblotting with
anti-14-3-3 mAb. Endogenous 14-3-3 proteins were coimmunoprecipitated
with GFP-TMCC3, confirming that TMCC3 bound to 14-3-3 proteins.
We next transfected fragments of TMCC3 with the N-terminal HA
tags

into

HEK293

cells

along

with

FLAG-14-3-3γ,

followed

by

immunoprecipitation with anti-FLAG mAb. The HA-tagged N-terminal
fragment encompassing the large cytoplasmic region (HA-TMCC3-N) was
coimmunoprecipitated with FLAG-14-3-3γ, indicating that N-terminal
cytoplasmic region had the binding site for 14-3-3γ. Similar to HA-TMCC-N,
the HA-tagged mutant deleting almost C-terminal half from TMCC3-N (HATMCC3-D1) was coimmunoprecipitated with FLAG-14-3-3γ. By contrast, the
HA-tagged mutant deleting the first 89 amino acids (aa) from TMCC3-D1
(TMCC3-D2) was not coimmunoprecipitated with FLAG-14-3-3γ. These
results indicate that 89 aa of N-terminal is binding site for 14-3-3γ.

4. Phosphorylated serine 15 is required for potent binding to 14-3-3γ
We examined whether TMCC3 had the 14-3-3 binding motif in the 89
aa of N-terminal. Full-length aa sequence of mouse TMCC3 was subjected to
14-3-3-Pred, the web server to predict 14-3-3 binding motifs. 14-3-3-Pred
showed there were three possible binding motifs in the 89 aa of N-terminal with
amino acid positioned at 12-17 has the highest score and meet completely to
14-3-3 binding motif. We then generated the point mutant of TMCC3-D1
substituting alanine for serine 15 (TMCC3-D1-S15A). HA-TMCC3-D1 or HAtagged TMCC3-D1-S15A (HA-TMCC3-D1-S15A) was transfected into

3

HEK293 cells along with FLAG-14-3-3γ, followed by immunoprecipitation
with

anti-FLAG

mAb.

While

HA-TMCC3-D1

was

strongly

coimmunoprecipitated with FLAG-14-3-3γ, coimmunoprecipitation of HATMCC3-D1-S15A was significantly decreased relative to HA-TMCC3-D1.
These results indicate that serine 15 was required for potent binding to 14-3-3γ.
Since TMCC3-D1-S15A still showed a weak binding to 14-3-3γ, we
further substituted serine 25 and serine 46 to alanines in TMCC3-D1-S15A
(TMCC3-D1-S15/25/46A), and performed immunoprecipitation analysis. A
small amount of HA-tagged TMCC3-D1-S15/25/46A (HA-TMCC3-D1S15/25/46A), which was comparable to that of HA-TMCC3-D1-S15A, was
coimmunoprecipitated with FLAG-14-3-3γ. These results indicate that the other
two deduced binding motifs predicted with low score are not involved in
binding to 14-3-3γ.
We next examined whether serine 15 was phosphorylated. HATMCC3-D1, HA-TMCC3-D1-S15A, HA-TMCC3-D1-S15/25/46A, or control
vector was transfected into HEK293 cells, followed by immunoprecipitation
with anti-HA mAb. The samples were subjected to immunoblotting with
antibody recognizing phosphorylated serine in the consensus 14-3-3 binding
motif (anti-phospho-14-3-3 binding motif pAb). The anti-phospho-14-3-3
binding motif pAb detected HA-TMCC3-D1, but not recognized HA-TMCC3D1-S15A and HA-TMCC3-D1-S15/25/46A, indicating that serine 15 is
phosphorylated. Similar results were obtained in U2OS cells.
5. TMCC3 mutant subtituting alanine for serine 15 is prone to localize at
three-way junction against overexpression of 14-3-3γ
We generated the full-length TMCC3 mutant by substituting alanine
for serine 15 (TMCC3-S15A), and established U2OS cell lines stably
expressing GFP tagged TMCC3-S15A (U2OS-GFP-TMCC3-S15A). HA-14-33γ or a control vector was then transfected into the U2OS-GFP-TMCC3 cells
or

U2OS-GFP-TMCC3-S15A

cells,

followed

by

immunostaining.

Overexpression of 14-3-3γ significantly increased the number of U2OS-GFPTMCC3 cells and U2OS-GFP-TMCC3-S15A cells losing their puncta relative

4

to the control vector. However, the number of the cells losing GFP-TMCC3S15A puncta was significantly lower than that of the cells losing GFP-TMCC3
puncta, indicating that TMCC3-S15A was more resistant to localize at threeway junctions against 14-3-3γ overexpression. Collectively, these result
indicate that binding to 14-3-3γ through phosphoserine 15 negatively regulates
localization of TMCC3 to three-way junctions.
6. The negative regulation of TMCC3 by 14-3-3γ is involved in formation of
the reticular ER network
We next examined whether negative regulation of TMCC3 by 14-33γ is involved in formation of reticlar ER network. The siRNA targeting
TMCC3 was transfected into U2OS cells followed with the transfection of HATMCC3, HA-TMCC3-S15A, or control vector. As negative control, the
negative control siRNA was transfected followed by transfection with control
vector. The samples were subjected for immunostaining with anti-CLIMP-63
pAb as an ER sheet marker and anti-α-tubulin mAb. The ratio of CLIMP-63
staining area to total area determined by α-tubulin staining significantly
increased in the TMCC3 knockdown-cells transfected with control vector
relative to negative control cells, indicating that, consistent with our previous
finding, TMCC3 knockdown caused ER sheet expansion. On the other hand,
the TMCC3-knockdown cells expressing HA-TMCC3 showed the ratio of
CLIMP-63 staining area to total area were comparable to the negative control
cells, indicating that expression of HA-TMCC3 rescued the phenotype of the
TMCC3 knockdown. Importantly, the TMCC3-knockdown cells expressing
HA-TMCC3-S15A showed the ratio of CLIMP-63 staining area to total cell
area significantly higher than the TMCC3-knockdown cells expressing HATMCC3, but lower than the TMCC3-knockdown cells transfected with the
control vector. These results indicate that expression of HA-TMCC3-S15A
partially rescued the phenotype of the TMCC3 knockdown, suggesting that the
negative regulation of TMCC3 by 14-3-3γ was involved in formation of the
reticular ER network.

5

Conclusion
In this study, we demonstrate that 14-3-3γ binding negatively regulates
TMCC3 localization for the reticular ER network. Binding of 14-3-3γ to TMCC3
is depend on phosphorylation of serine 15 in the 14-3-3 binding motif.

6

神戸大学大学院医学（
系）
研究科（博士課程）

論 文審 査 の 結 果 の 要 旨
甲第

3308 号

名

受 付番 号

氏

SAIHASSUHDA

risoform binds to and regulates the locali
za
t
ion of

The 14-3-3

endop
lasmi
cr
e
t
iculum(ER)membranepr
o
t
ei
nTMCC3f
ort
her
e
t
ic
u
l
a
r

論文題目

ne
t
workoft
heER

Ti
t
l
eo
f
Di
ssert
at
i
on I14-3-3r
が TMCC3に結合し、その局在を制御することで、小胞体ネット
ワークを形成する

主

査

Ch
i
ef
Exam
i
ner

審査委員
Exam
i
ner

副

査

Vi
c
eexami
ner

副

査

Vi
c
e・
exa
mi
ner

犀碍

釘
犀

え

頁芦

＼灯

g

慮

（要旨は 1
, 000字∼ 2, 000字程度）

（目的）
小胞体は酵母から哺乳動物まで全ての真核生物に保存された最も大きい細胞内小器官である。小胞体は
シート構造とチュープ構造が連続した特徴的な形状を持つ。チ ューブ構造は、 t
h
r
e
e
-w
a
yj
unc
t
ion によっ
hree-wa
yj
unc
t
ionに局在 し
、 その形成
て連結 されている。近年、 我々は小胞体膜タンパク質 TMCC3が t

を促進することを報告した。 しかし、 この TMCC3 の活性がどのように調節さ れるかは不明である。
TMCC3結合タンパク質として報告されている 1
433は、リン酸化依存的に標的タンパク質に結合し、

様々な細胞機能を調節して い ることが知られている。そこで、本研究では、 TMCC3の 1
4
3
-3結合リン
酸化アミノ酸を同定し、その結合が TMCC3のt
l
1
r
ee-wa
yj
unc
t
ionへの局在化
、 小胞体チ ューブネッ トワ
ーク形成に 関与するかを解析した。

（方法）
• 1
4
3
3による TMCC3の t
hree-wa
yj
unc
t
ionへの局在化への効果
GFP-TMCC3を恒常的に発現 してい る U20S細胞(U20S-GFP-TMCC3 細胞）
に HA-1
4
3針 あるいはコン

トロールとして mCher
r
yを トランス フェクションした。Th
r
ee-wa
yj
unc
t
ionに局在する GFPTMCC3の点
状構造の数をカウントした。 同様に、野生型 U20S 細胞 に HA-14-3—祈 あるいは mCherry を トラン スフ

ェクションし、抗 TMCC3抗体と抗 PDl抗 体で染色した。 Three-wa
yj
unc
t
ionに局在する内在性 TMCC3
の点状構造の数をカウン トした
。
• 1
43
3によ る t
h
r
e
e
-w
a
yj
un
c
t
ion形成への効果
野生型 U20S 細胞 に HA- 14-3—舒 ある い はコントロー ル と し て mCherry をトランスフェクションし、抗

PDI抗体で染色した。細胞辺縁の面積 ]
Ox10μm
2において、 t
hree-wa
yj
unc
t
ionの数をカウントした。
• TMCC
3の 1
4
3
3結合領域
HEK293細胞に、 HA-TMCC3-N (
TMCC3の N 末細胞質領域全体）、 HA
TMCC3Dl (
HAT MCC3-N の

c末半分を欠損した欠損体）あるいは

HA-TMCC3D2 (HATM
CC3-D1の N 末 1
89 アミノ酸配列 を欠

損した欠損体）および FLAG-14-3 —舒 をそれぞれトランスフェクションした。 TritonX-100 抽出液に抗

FLAG抗体を加えて免疫沈降した。
• TMCC3の 1
4
3
3結合リン酸化アミノ酸
HA-TMCC3-D1-S15A (
HA-TMCC3-D1 の 1
5 番目のセリンをアラニンに置換 し た変異体）あるいは
HA-TMCC3-D1
-S1
5/25/46A (
HA-TMCC3-Dlの 1
5番目 、25番目 、46番目のセリンをアラニンに置換し

た変異体）および FLAG-14-3-打 をそれぞれトランスフェクションした。 Tri
t
onX-1
00抽 出液に抗 FLAG
抗体を加えて免疫沈降した。
• TMCC3S1
5Aによる t
h
r
ee-wa
yj
unc
t
i
onへの局在化への効果
GFP
-TM
CC3S1
5A(GFPTMCC3の 1
5番目のセリンをアラニンに置換した変異体）を恒常的に発現して

いる U20S細胞(U20S-GFPTMC
C3-S1 5A 細胞） あるいは U20S-GFP-TMCC3 細胞 に 、 HA-14-3 —舒あるい
はコントロールベクターをトランスフェクションした。 Threewa
yj
un
ct
i
onに局在する GFP
-TMCC3S1
5A
あるいは GFP
TMCC3の点状構造の数をカウン トした
。
• T MCC3S1
5Aによ る小胞体形態形成への効果
U20S細胞 に TMCC3s
i
RNAを トランスフェクションして TMCC3ノックダウン細胞を調製 した
。 TMCC3

ノックダウン細胞に HA-TMCC3ある いは HA-TMCC3-S1
5Aをトランスフェクションし、抗 CLIMP
63
抗体と抗 a
.
t
ub
u
li
n抗体で染色した。 CLIMP-6
3染色領域（シート構造領域） ／
a
.
t
u
b
u
li
n染色領域（細胞質
全 体） を計測した
。

（結果）
1
. 1
4
3
3による TMCC3の t
hree-wa
yj
unc
t
i
onへの局在化への効果
U20S
GFP-TMCC3 細胞 に HA-143ー
祈 を発現させたと ころ、GFPTMCC3の点状構造の数が減少

した。同様に、野生型 U20S細胞に HA-14-3-3yを発現させたところ、内在性の TMCC3の点状構造
の数が減少した。
2
. 1
433による t
hree-wa
yj
unc
t
ion形成への効果
野生型 U20S 細胞に HA- 1 4-3—行を 発現させたところ 、three-way j
unc
t
i
onの数が減少した。

3
. TMCC3の 1
4
3
3結合領域
1433
yは、 HA-TMCC3N あるいは HA-TMCC3-D1 と結合する が
、
免疫沈降実験により、 FLAGHATMCC3D2と結合しなかった。
4
. TMCC3の 1
4
3
3結合リン酸化アミノ酸
HATMCC3-DI
S1
5Aと FL
AG14
3
3
yの結合は、野生型 HATMCC3-DI との結合に 比べて、減少し

た
。 HA-TMCC3-D1-SI
5/25/46A と FLAG-l4
3
3
yの結合は、 HA-TMCC3-Dl
-S
lSAとの結合と同程度
であ った。
5
. TMCC3-S15Aによる t
hr
e
e
w
a
yj
unc
t
ionへの局在化への効果
U20S-GFP
TMCC3S15A 細 胞 に HA-14-3-3
y を 発 現 さ せ た と こ ろ 、 コ ン トロ ー ル の
U20SGFP
TMCC3 細胞 と比べて、 GFP-TMCC3 の点状構造が減少した細胞の数が少なかった。
1
4
3
3タンパク質の存在下でも、 GFP-TMCC3に比べて、 GFP
TMCC3SI
SAがt
h
r
e
e
wa
yj
unc
t
ionへ

局在 した。
6
. TMCC3-SI
SAによる小胞体形態形成への効果
TMCC3ノ ックダウン細胞ではチューブ構造が減少し、CLIMP-63染色領域 （シー ト構造領域）が増

加する 。TMCC3ノックダウン細胞に HA-TMCC3を発現すると CLIMP-63染色領域が減少し 、チュ
ーブ構造が回復した。 一方、HA-TMCC3-S15A を発現すると、 CLIMP-63 染色領域が HAT MCC3
程 には減少せず、部分的にチューブ構造が回復した。

（結論）
1
4
3
3が TMCC3の 1
5番目セリンにリン酸化依存的に結合することにより 、
TMCC3の t
h
r
ee
wa
yj
unc
t
i
on

への局在化を抑制し、小胞体チュ ー ブネットワ ー ク形成を調節していることを明らかにした。

本研究は、 TMCC3の 143
3結合リン酸化アミノ酸の同定とそれによる小胞体チュ ーブネッ トワーク形
成への影密を解析し たものである 。そ の結果、 1
4
3
3が TMCC3の t
hr
ee
-w
a
yj
unc
t
i
onへの局在化を抑制、
t
h
r
e
e
-wa
yj
unc
t
ionの形成ひいては小胞体チュープネットワーク形成を制御していることを証明した点に
おいて、価値ある業績であると認める 。 よって、本研究者は博士（医学）の学位を得る資格があると認
める。