ER(소포체) 구조, 기능, 기능부전 --＞ 개선 영양소(특히 타우린)

[문형철]

https://pmc.ncbi.nlm.nih.gov/articles/PMC3653419/

세포 내 소포체(ER) 스트레스와

이를 해결하는 비접힘 단백질 반응(Unfolded Protein Response, UPR)이

어떻게 다양한 질병의 발생과 진행에 관여하는지 체계적으로 정리한 리뷰.

핵심 내용

• ER은 단백질 합성·접힘의 주요 장소인데, 스트레스가 쌓이면 UPR이 작동.
• UPR의 3대 주요 경로:
  • IRE1 → XBP1 스플라이싱 (ER 접힘 능력 증가)
  • PERK → eIF2α 인산화 (단백질 합성 일시적 억제)
  • ATF6 → 전사인자 활성화 (ER 샤페론 등 유전자 발현 증가)
• UPR은 초기에는 보호적이지만, 스트레스가 지속되면 세포 사멸을 유발 (CHOP, JNK 경로 등).

질병과의 연관 (특히 대사 질환)

• 제2형 당뇨병: 비만 시 간·지방·근육에서 ER 스트레스가 증가 → JNK 활성화로 IRS-1 인산화 억제 → 인슐린 저항성 유발. β세포에서는 PERK 경로 이상으로 인슐린 분비 장애와 세포 사멸 발생.
• 기타: 신경퇴행성 질환(알츠하이머, 파킨슨), 암, 바이러스 감염, 염증성 질환 등에서 ER 스트레스가 중요한 병인으로 작용.

결론 ER 스트레스와 UPR은 세포 항상성 유지에 필수적이지만, 만성적으로 활성화되면 여러 질병의 공통 병리 기전

https://www.nature.com/articles/s12276-021-00560-8

ER(소포체) 스트레스와

산화환원(redox) 신호(특히 ROS, 산화 스트레스) 사이의 복잡한 상호작용을 정리하고,

이 interplay가 세포 운명(생존 vs 사멸)을 어떻게 결정하며, 그 결과가 어떤 질병으로 이어지는지 설명.

핵심 내용

• ER 스트레스가 발생하면 UPR(Unfolded Protein Response)이 활성화. (IRE1α, PERK, ATF6 경로)
• ER 내에서 단백질 접힘 과정(ERO1-PDI 경로) 자체가 ROS(H₂O₂)를 생성하여 redox 환경을 변화시킵니다.
• 산화 스트레스는 ER 칼슘 채널(IP₃R, RyR)을 산화시켜 ER에서 칼슘 방출을 유발 → 미토콘드리아로 칼슘 과부하 → 미토콘드리아 ROS 증가 → 악순환.
• 만성 ER 스트레스 시 UPR 실패(ER stress response failure)가 발생:
  • XBP1s의 핵 이동 장애
  • IRE1α의 S-니트로실화(SNO) 등으로 XBP1 스플라이싱 감소
  • 결과: 적응 실패 → 세포 사멸(apoptosis) 유발

ER 스트레스 시 활성화되는 3가지 주요 센서와 그 downstream 신호:

1. IRE1 경로
   • XBP1 mRNA를 스플라이싱 → sXBP1 (강력한 전사인자)
   • sXBP1 → 샤페론, ERAD, 자가포식, 지질 합성 증가 (보호적)
   • 과도 활성화 시 → RIDD (mRNA 분해) + JNK 활성화 → 세포 사멸 신호
2. PERK 경로
   • eIF2α 인산화 → 전반적인 단백질 합성 감소 (세포 부담 줄임)
   • 선택적으로 ATF4 번역 증가 → CHOP, 산화 스트레스 대응, 자가포식 유전자 발현
3. ATF6 경로
   • Golgi로 이동 후 절단(S1P/S2P) → 활성형 ATF6 (p50)
   • 핵으로 들어가 샤페론과 XBP1 유전자 발현 촉진

최종 결과 (핵심):

• 초기/적당한 스트레스: sXBP1, ATF4, ATF6 → ER 기능 회복, 단백질 접힘 능력 증가, 세포 생존
• 만성/강한 스트레스: CHOP 증가, JNK 활성화, Caspase 활성화 → 세포 사멸

• ER 안에서 단백질이 제대로 접히지 않으면 (Unfolded/misfolded proteins) UPR이 활성화됩니다. (IRE1, PERK, ATF6)
• ER 단백질 접힘 과정 자체가 ROS(활성산소)를 대량 생산합니다.
  • ERO1 + PDI → 단백질에 disulfide bond (SS-bond) 형성하면서 ROS 발생
  • NOX (NADPH oxidase)도 ER에서 ROS를 추가로 생성
• ER에서 발생한 ROS + Ca²⁺ 유출 (IP₃R 통해) → 미토콘드리아로 Ca²⁺ 과부하
• 미토콘드리아에서도 ROS가 폭발적으로 증가 → Oxidative stress
• 결과: ER dysfunction + Mitochondrial dysfunction → 세포 사멸(apoptosis) 신호 (Bax, Cytochrome c, Caspase-9 → Caspase-3)

핵심 메시지: ER 스트레스와 산화 스트레스는 서로를 증폭시키는 악순환

좌측: 정상적인 ER stress response (성공적 적응)

• IRE1이 활성화되어 unspliced XBP1 (uXBP1)을 spliced XBP1 (sXBP1)로 변환
• sXBP1이 핵으로 들어가 ER chaperones (단백질 접힘 도우미)와 ERAD components (잘못 접힌 단백질 제거 시스템)를 대량 발현
• 결과: ER 기능이 회복되고 세포 생존

우측: ER stress response failure (실패)

• sXBP1 생성이 제대로 안 되거나 (X 표시), sXBP1의 기능이 저하됨
• 대신 TRAF2 → ASK1 → JNK 경로가 강하게 활성화
• sXBP1의 기능 저하 원인으로 Sdf2L1-p85α/β 결합, SNO (S-니트로실화), SO3H (황산화) 등이 의심됨
• 결과: ER 기능 회복 실패 → 세포 사멸(apoptosis) 신호가 강해지고, 대사 장애(당뇨, 비만, 노화 등)가 발생

대사 질환과의 연결

• 비만·제2형 당뇨병에서 ER 스트레스와 redox 스트레스가 인슐린 신호를 방해합니다.
• XBP1s와 PI3K p85 서브유닛의 상호작용 장애 → 인슐린 저항성 악화.
• 이는 이전에 보신 ER Ca²⁺ dysregulation → UPR → 인슐린 저항성 기전과 잘 맞물립니다.

결론 및 의의

ER 스트레스와 redox 신호의 상호작용은

초기에는 세포를 보호하지만,

지속되면 세포 사멸과 다양한 질환(대사질환, 신경퇴행성 질환 등)을 촉진

https://pmc.ncbi.nlm.nih.gov/articles/PMC5117888/

내용이 가장 큰 세포 소기관인 소포체(Endoplasmic Reticulum, ER)가

다른 소기관들과 형성하는 막 접촉 부위(Membrane Contact Sites, MCS)의

구조와 기능을 체계적으로 정리한 리뷰.

ER은

다른 소기관과 실제로 융합하지 않으면서도 매우 가까이(보통 3~15 nm) 붙어 있으며,

이 접촉 부위를 통해 지질 전달, 칼슘 신호, 소기관 움직임 조절, 분열(fission) 등을 조절한다는 점을 강조.

1. ER의 구조와 MCS의 특징
   • ER은 핵막, sheet(평평한 판, 리보솜 많음), tubule(가느다란 관, 리보솜 적음)로 이루어져 있습니다.
   • Peripheral ER (세포 주변부 tubule 네트워크)이 특히 많은 MCS를 형성합니다.
   • MCS에서는 리보솜이 배제되고, 특이한 tether 단백질들이 막을 가까이 붙여둡니다.
   • 접촉 면적은 미토콘드리아 표면의 2~5%, 엔도솜 표면의 3~5% 정도로 작지만, 여러 곳에서 동시에 일어나며 기능적으로 중요합니다.
2. 주요 MCS 대상 소기관과 기능
   • 미토콘드리아 (ER–mitochondria MCS): 칼슘 전달(Ca²⁺), 지질 합성(인산화티딘 → 포스파티딜세린), 미토콘드리아 분열(fission) 부위 표시, 움직임 조절.
   • 엔도솜/라이소좀 (ER–endosome MCS): 엔도솜 성숙과 위치 조절, 콜레스테롤 전달, 칼슘 신호 교환, 엔도솜 분열 조절. 특히 late endosome일수록 ER과의 접촉이 증가합니다.
   • 기타: Golgi (지질 전달), peroxisome, lipid droplet (지질 저장 및 합성), plasma membrane 등도 언급되지만, 미토콘드리아와 엔도솜이 가장 중점적으로 다뤄집니다.
3. MCS가 하는 중요한 일
   • 비-소포성 지질 전달 (Non-vesicular lipid transfer): LTP (lipid transfer proteins, 예: ORP, OSBP 등)를 통해 지질을 직접 전달.
   • 칼슘 신호 조절: ER에서 방출된 Ca²⁺가 미토콘드리아나 엔도솜으로 선택적으로 들어감.
   • 소기관 역학(dynamics): 엔도솜과 미토콘드리아가 움직일 때 ER tubule에 붙어서 함께 이동하거나, 분열 부위를 지정.
   • 소기관 생합성(biogenesis): 일부 소기관(퍼옥시좀, lipid droplet 등)의 형성이 ER에서 시작됨.

ER은 단순한 “단백질 공장”이 아니라,

다른 소기관들과 연결된 거대한 네트워크로서

세포 내 신호 전달과 항상성을 유지하는 핵심 허브입니다.

MCS는

막 융합 없이도 빠르고 정밀한 물질 교환과 신호 조절을 가능하게 하며,

향후 MCS 단백질의 정확한 작용 기전 연구가 질환(신경퇴행성 질환, 대사 질환 등) 이해에 중요할 것으로 전망

이 논문은

단순한 개요가 아니라

ER의 다중 도메인 구조와 신호-구조 연계(signal-to-shape coupling)에 초점을 맞춘 고급 리뷰.

ER을

정적 organelle이 아닌,

동적이고 적응적인 네트워크로 재정의하며,

세포 주기, 발달, 스트레스 반응 동안의 remodeling 메커니즘을 강조합니다.

1. ER의 기본 구조와 도메인 특이성 (Structural Domains)

ER은

세포에서 가장 큰 막 기관으로,

nuclear envelope와 peripheral ER로 나뉩니다.

Peripheral ER은 다시 두 가지 주요 형태로 존재합니다:

• Sheets (평면 시트, rough ER): 리보솜이 많이 붙어 있어 단백질 합성·접힘·수송에 특화. 단백질 분비가 활발한 세포(예: 분비세포)에서 우세.
• Tubules (관형, smooth ER): 곡률이 높고, 지질 합성, 스테로이드 합성, 칼슘 저장·방출에 관여. 칼슘 신호나 지질 대사가 중요한 세포에서 많음.

이 두 형태 사이의 비율과 연결(three-way junctions)은

세포 유형, 세포 주기 단계, 발달 단계에 따라

동적으로 변합니다.

구조를 결정짓는 핵심 단백질:

• Reticulons (RTN)과 DP1/Yop1/REEP family: 막 내 hairpin topology로 고곡률을 유도 → tubule 형성 촉진. 과발현 시 ER을 tubule-dominant로 바꿈.
• Atlastins (ATL, GTPases): homotypic fusion (동종 막 융합) 담당. ER 네트워크 형성에 필수. GTP 가수분해로 tubule 간 연결.
• Lunapark: three-way junction 안정화.
• Rab GTPases (Rab10, Rab18 등): subdomain 특이적 fusion/fission 조절. sheet/tubule 균형 유지.

이 단백질들의 topology와 oligomerization이

ER의 물리적 curvature와 membrane tension을 결정한다는 점이 핵심입니다.

(Fig. 1~2에서 live imaging과 electron microscopy로 시각화)

A, B: HeLa 세포(인간 암세포)에서 관찰한 소포체 네트워크

• A: 핵(nucleus)과 핵막(Nuclear Envelope)을 GFP-Sec61 단백질로 표시
• B: 세포 전체에서 소포체가 복잡하게 퍼져 있는 모습

C, D: Xenopus (아프리카 발톱개구리) 난자 추출물에서 본 소포체

• ER Sheet: 평평하고 판처럼 생긴 부분 (C, D에 표시)
• ER Tubule: 가느다란 관 모양의 부분
• Three-way junction: 세 개의 관이 만나서 Y자 형태로 연결된 지점

E: Rough ER (조면소포체)

• 전자현미경 사진으로, 리보솜이 붙어 있어서 거칠게 보임
• 단백질을 합성하는 곳

F: Smooth ER (활면소포체)

• 전자현미경으로 본 단면
• 리보솜이 없어서 매끄럽게 보임
• 지질 합성, 해독, 칼슘 저장 등의 기능을 함

A: 소포체의 전체 모양

• 소포체는 크게 두 가지 형태로 존재합니다:
  • Sheet (평평한 판 모양) → 주로 Rough ER에 많음
  • Tubule (가느다란 관 모양) → 주로 Smooth ER에 많음
• 리보솜(보라색 동그라미)이 붙어 있는 부분은 단백질 합성이 활발한 Rough ER입니다.
• 오른쪽 큰 그림은 실제 세포 안에서 sheet와 tubule이 연결된 복잡한 3차원 구조를 보여줍니다.

B: REEP 단백질의 구조와 차이 소포체 막의 곡률(curvature)을 조절하는 중요한 단백질인 REEP 계열 단백질을 비교한 것.

• REEP1-4 (왼쪽 두 그림):
  • 막을 관통하는 transmembrane domain이 두 개 있습니다.
  • N말단과 C말단이 세포질 쪽을 향하고 있습니다.
  • 막을 살짝 구부리는 역할을 합니다.
• REEP5/6 (오른쪽 그림):
  • transmembrane domain이 하나만 있습니다.
  • N말단은 세포질 쪽, C말단은 ER 내부(루멘) 쪽을 향합니다.
  • REEP1-4와는 다르게 막을 더 강하게 구부리거나 tubule 형성에 관여할 가능성이 높습니다.

2. ER Dynamics와 신호 응답 (Response to Cellular Signaling)

논문의 가장 중요한 기여는

ER이 외부/내부 신호에 어떻게 구조를 재조정하는가입니다.

이는 단순한 형태 변화가 아니라,

기능적 적응입니다.

• 세포 주기 (특히 Mitosis):
  • Interphase → Mitosis 진입 시 cyclin B-CDK1 kinase가 ER-shaping protein을 인산화. 예: Climp63 (sheet stabilizer) 인산화 → microtubule과의 interaction 차단 → ER이 tubule에서 sheet-dominant로 shift.
  • STIM1 (calcium sensor) 인산화 → store-operated calcium entry (SOCE) 조절 변화.
  • 결과: mitotic ER은 주로 sheets 형태로 재구성되어 spindle과 상호작용. (Xenopus egg extract 시스템에서 in vitro 재구성 실험으로 증명)
• 난모세포 성숙과 수정 (Oocyte maturation & Fertilization):
  • ER이 cluster나 sheet wrapping 형태로 재배치.
  • 칼슘 wave (IP3R, RyR-mediated)와 microtubule dynamics가 coupling.
  • XendoU (calcium-dependent RNase): RNA 분해를 통해 fusion 촉진 → ER 구조 변화.
• 스트레스 반응 (Unfolded Protein Response, UPR):
  • ER 내 misfolded protein 축적 → UPR 활성화 (IRE1, PERK, ATF6 pathway).
  • ER membrane 확장 (주로 sheet 증가)으로 folding capacity 증대.
  • 칼슘 homeostasis 붕괴 시 STIM1/Orai1-mediated SOCE 활성화.

이 모든 과정에서

phosphorylation,

GTPase activity,

calcium signaling이 구조 변화의 molecular switch 역할을 합니다.

논문은

live-cell imaging, RNAi knockdown, phospho-proteomics 등을 통해

이러한 메커니즘을 종합합니다.

https://www.nature.com/articles/s12276-023-01067-0

이 논문은
세포 내 포스포이노시티드(PIPs)와 칼슘(Ca²⁺) 신호의 상호작용을 리뷰한 논문.

특히
Ca²⁺ 과부하가 PIP 신호를 부정적으로 억제하는 새로운 메커니즘을 중점적으로 설명.

핵심 발견

• 높은 Ca²⁺ 농도가 PIPs(PI(4,5)P₂, PI(3,4,5)P₃ 등)와 직접 결합하여 Ca²⁺-PIP 복합체를 형성.
• 이 복합체는 PH 도메인이나 C2 도메인을 가진 중요한 신호 단백질(AKT, IRS1, PI3K 등)의 막 결합을 방해.
• 결과적으로 인슐린 신호(PI3K-AKT 경로)가 억제되어 인슐린 저항성이 발생.
• 비만이나 대사 스트레스 상태에서 ER(소포체)의 Ca²⁺ 고갈과 미토콘드리아 Ca²⁺ 과부하가 악순환을 일으키며, 이는 염증, ROS 증가, 세포 기능 저하로 이어집니다.

Ca²⁺와 PIP의 결합이
세포 신호의 중요한 음성 조절자(negative regulator)라는 점을 새롭게 강조.

이는
비만, 당뇨, 대사 질환의 병인 이해와
새로운 치료 타겟(예: Ca²⁺ 조절제)을 제시

인슐린(Insulin)이

세포막의 인슐린 수용체(IR)에 결합하면 다음과 같은 신호가 시작.

1. IR → IRS → PI3K 활성화
   • 인슐린 수용체가 인산화 → IRS(IRS-1)가 인산화
   • IRS가 PI3K(p85 + p110)를 활성화
   • PI3K가 PI(4,5)P₂ → PI(3,4,5)P₃ (PIP3)로 변환
2. PIP3 → AKT 활성화 (주요 신호 허브)
   • PIP3가 막에 모이면 PH 도메인을 가진 PDK1과 AKT가 막으로 이동
   • PDK1이 AKT를 인산화 → mTORC2가 추가로 인산화 → AKT 완전 활성화
3. 활성화된 AKT가 하는 일 (주요 downstream 효과):
   • GLUT4 이동: AS160을 인산화 → GLUT4가 세포막으로 이동 → 포도당 흡수 증가 (혈당 낮춤)
   • GSK3β 억제 → Glycogen synthesis (글리코겐 합성) 증가
   • FOXO 억제 → Gluconeogenesis (포도당신생합성) 감소
   • BAD 인산화 → 세포 사멸(Apoptosis) 억제
   • SREBP 활성화 → 지방 합성 증가
   • mTORC1 활성화 (TSC2 억제 통해) → 단백질 합성 증가 (4EBP1, p70S6K)

참고) Rab GTPase는 세포 내 소포(vesicle) 수송과 막 트래픽을 조절하는 중요한 작은 G단백질(small GTPase) 가족.

• Rab = Ras-related in brain (뇌에서 처음 발견된 Ras 유사 단백질)
• 소형 GTPase로, GTP를 결합하면 활성 상태(GTP-bound), GTP를 GDP로 가수분해하면 비활성 상태(GDP-bound)가 됩니다.
• 세포 내 다양한 막(소포체, 골지체, 엔도솜, 리소솜, 세포막 등)에 특이적으로 위치하여, 소포의 형성, 이동, 도킹(docking), 융합(fusion)을 정확하게 제어합니다.
• 인간에는 약 60여 종 이상의 다양한 Rab 단백질(Rab1, Rab4, Rab5, Rab7, Rab8, Rab10, Rab11, Rab13, Rab14 등)이 있으며, 각각 특정 막 트래픽 단계를 담당

음성 조절자 (Negative regulators)

• PTEN, SHIP2, INPP4B: PIP3를 분해하여 신호를 끔
• PP2A, PHLPP: AKT를 탈인산화하여 비활성화

한 줄 요약:

인슐린 → IR → PI3K → PIP3 → AKT 활성화

→ 포도당 흡수↑, 글리코겐 합성↑, 지방 합성↑, 단백질 합성↑, 포도당신생합성↓을 통해

혈당을 낮추고 세포 성장을 촉진하는 대표적인 성장·대사 신호 경로

• 세포 외부 → 세포질 Ca²⁺ 유입: PMCA (펌프, Ca²⁺ 배출), ORAI (SOCE), TRP, P2RX 등 채널을 통해 Ca²⁺가 들어옴.
• ER에서의 Ca²⁺ 조절:
  • SERCA: Ca²⁺를 ER 안으로 펌핑 (저장)
  • IP₃R: IP₃에 의해 ER에서 Ca²⁺ 방출
• 미토콘드리아: MCU를 통해 Ca²⁺ 유입 → ATP 생산 증가, 하지만 과다하면 문제 발생.
• 주요 downstream:
  • GPCR → PLC → IP₃/DAG → Ca²⁺ 방출 + PKC
  • Ca²⁺ → CaMK, CnA (칼시뉴린) → NFAT, CREB, FOXO, CRTC2 등 전사인자 활성화
• 결과: 염증(Inflamation), 포도당신생합성(Gluconeogenesis) 조절 등.

• 인슐린 신호: Insulin → IR → IRS → PI3K → AKT 경로 (정상적으로 작동)
• Ca²⁺ 과부하(overload) 시 문제점:
  • ER Ca²⁺ 고갈 → ER stress & UPR 활성화
  • 미토콘드리아 Ca²⁺ 과다 → ROS 증가, ATP 변화
  • SERCA, IP₃R, MCU 등의 조절 변화 (화살표 ↑↓ 표시)
  • Ca²⁺ → CaMK → FOXO, CRTC2 등 활성화 → 자가포식(Autophagy) 억제, Lysosome 경로(TFEB) 변화
• 핵심 포인트: 높은 Ca²⁺가 인슐린 신호를 방해하고, ER stress, 염증, 대사 이상을 유발하는 악순환을 강조.

3. 방법론적 접근 (Graduate-level Insight)

이 논문은

실험 논문이 아닌 리뷰이지만,

인용된 연구들의 방법이 고급입니다:

• Live-cell imaging + electron microscopy (ultrastructure).
• Xenopus egg extract를 이용한 in vitro reconstitution (cell-free system으로 ER dynamics 재현).
• RNAi, overexpression‎, proteomics/phospho-proteomics.
• Functional assay (calcium imaging, protein secretion assay 등).

이 접근은 ER의 multi-scale regulation (molecular → organelle → cellular level)을 밝히는 데 필수적입니다.

4. 주요 발견과 함의

• ER 구조는 세포 기능에 최적화되어 있다 (protein-secreting cell → sheet dominant; lipid/calcium cell → tubule dominant).
• 신호(phosphorylation, calcium, GTP)와 구조 변화 간의 coordinated response가 ER의 다기능성을 가능하게 함.
• 질병 연계: ER shaping protein 돌연변이 → hereditary spastic paraplegia (HSP), Alzheimer’s 등. UPR dysregulation → 여러 대사·신경 질환.
• 미래 방향: signal-to-shape translation의 정확한 메커니즘, meiosis/fertilization에서의 세부 분자 경로, 개발 과정에서의 ER remodeling.

Cell Mol Life Sci

. 2015 Oct 3;73(1):79–94. doi: 10.1007/s00018-015-2052-6

The endoplasmic reticulum: structure, function and response to cellular signaling

Dianne S Schwarz 1,2,3, Michael D Blower 1,2,✉

• Author information
• Article notes
• Copyright and License information

PMCID: PMC4700099  PMID: 26433683

Abstract

The endoplasmic reticulum (ER) is a large, dynamic structure that serves many roles in the cell including calcium storage, protein synthesis and lipid metabolism. The diverse functions of the ER are performed by distinct domains; consisting of tubules, sheets and the nuclear envelope. Several proteins that contribute to the overall architecture and dynamics of the ER have been identified, but many questions remain as to how the ER changes shape in response to cellular cues, cell type, cell cycle state and during development of the organism. Here we discuss what is known about the dynamics of the ER, what questions remain, and how coordinated responses add to the layers of regulation in this dynamic organelle.

소포체(Endoplasmic Reticulum, ER)는

세포 내에서 매우 크고 동적인 구조로,

칼슘 저장, 단백질 합성, 지질 대사 등 다양한 역할을 수행한다.

골지체(Golgi apparatus, 골지체 또는 골지 복합체)는
세포 내 우체국 또는 포장 공장으로 불리는
중요한 소기관입니다.

골지체의 기본 구조 (그림 기준) 제공하신 그림에서 보이는 것처럼:

• ER(소포체)에서 노란색 관을 통해 단백질이나 지질이 들어옵니다.
• 골지체는 납작한 주머니(시스테르나, cisternae)가 여러 개 쌓여 있는 파란색 구조로 표현됩니다.
• 한쪽 면(cis face, 들어오는 면)은 ER 쪽에 가깝고, 반대쪽 면(trans face, 나가는 면)에서 소포(vesicle)가 나가서 다른 곳으로 이동합니다.

이 구조는
단백질이 한쪽에서 들어와 가공되면서
다른 쪽으로 나가는 조립 라인 역할을 합니다.

골지체는
ER에서 온 물질을 받아서 최종 가공 → 분류 → 포장 → 수송하는 역할.

1. 단백질과 지질의 변형(Modification)
   • 글리코실화(Glycosylation): 당(糖)을 붙여 당단백질(glycoprotein)이나 당지질(glycolipid)로 만듦. (단백질이 안정화되고, 제대로 기능할 수 있게 해줍니다.)
   • 인산화(phosphorylation), 절단(cleavage) 등 다른 화학적 변형도 수행.
   • 식물세포에서는 세포벽을 구성하는 복잡한 다당류(polysaccharides)도 합성합니다.
2. 분류와 표지(Sorting & Tagging)
   • 단백질에 “주소 표지”를 붙여 어디로 갈지 결정합니다. 예: 만노스-6-인산 표지가 붙으면 → 리소좀(lysosome)으로 이동.
   • 세포막으로 갈 단백질, 세포 밖으로 분비될 단백질, 세포 내 다른 소기관으로 갈 물질 등을 구분합니다.
3. 포장과 수송(Packaging & Transport)
   • 가공된 물질을 소포(vesicle)에 싸서 목적지로 보냅니다.
     • 세포 외부로 분비 (분비 단백질: 호르몬, 효소 등)
     • 세포막으로 삽입 (막 단백질)
     • 리소좀으로 이동 (가수분해효소 등)
4. 기타 기능
   • 지질 합성 및 수송에도 관여.
   • 소포체로 잘못 온 단백질을 되돌려 보내는 역할도 합니다 (재활용).

ER의 다양한 기능은

각각 별개의 도메인에서 이루어지며,

이 도메인은 세관(tubules), 시트(sheets), 그리고 핵막(nuclear envelope)으로 구성된다.

ER의 전체적인 구조와 동역학에 기여하는 여러 단백질들이 이미 밝혀졌으나,

세포 신호, 세포 유형, 세포 주기 상태, 그리고 개체 발생 과정에서

ER이 어떻게 형태를 변화시키는지에 대해서는 아직 많은 질문이 남아 있다.

본 리뷰에서는

ER의 동역학에 대해 알려진 내용,

아직 해결되지 않은 문제점들,

그리고 이 동적인 소기관의 여러 층위의 조절에 기여하는 협력적 반응에 대해 논의하고자 한다.

키워드: 간기(Interphase), 유사분열(Mitosis), 비접힘 단백질 반응(Unfolded protein response), 조직화(Organization), 수정(Fertilization), 인산화(Phosphorylation)

Keywords: Interphase, Mitosis, Unfolded protein response, Organization, Fertilization, Phosphorylation

Introduction

The ER is the largest organelle in the cell and is a major site of protein synthesis and transport, protein folding, lipid and steroid synthesis, carbohydrate metabolism and calcium storage [1–7]. The multi-functional nature of this organelle requires a myriad of proteins, unique physical structures and coordination with and response to changes in the intracellular environment. Work from a variety of systems has revealed that the ER is composed of multiple different structural domains, each of which is associated with a specific function or functions. However, it is not yet clear how these functional subdomains are organized and how different functional domains translate into different structures.

서론 

소포체(ER)는

세포 내에서 가장 큰 소기관으로서,

단백질 합성과 수송,

단백질 접힘,

지질 및 스테로이드 합성,

탄수화물 대사,

그리고 칼슘 저장의 주요 장소이다 [1–7].

이 소기관의 다기능성은

수많은 단백질, 독특한 물리적 구조,

그리고 세포 내 환경 변화에 대한 조정 및 반응을 필요로 한다.

다양한 생물 시스템에서 수행된 연구를 통해 ER은

각각 특정 기능과 연관된 여러 구조적 도메인으로 구성되어 있음이 밝혀졌다.

그러나

이러한 기능적 하위 도메인이 어떻게 조직화되는지,

그리고 서로 다른 기능적 도메인이 어떻게 다른 구조로 나타나는지에 대해서는 아직 명확하지 않다.

Protein synthesis and folding

One of the major functions of the ER is to serve as a site for protein synthesis for secreted and integral membrane proteins [8], as well as a subpopulation of cytosolic proteins [1]. Protein synthesis requires localization of ribosomes to the cytosolic face of the ER, and the canonical pathway that regulates protein synthesis involves co-translational docking of the mRNA:ribosome complex on the ER membrane. Translation of secretory or integral membrane proteins initiates in the cytosol, then ribosomes containing these mRNAs are recruited to the ER membrane via a signal sequence within the amino terminus of the nascent polypeptide that is recognized and bound by the signal recognition particle (SRP) [9, 10]. The complex of mRNA:ribosome:nascent polypeptide:SRP is targeted to the ER where it docks on the SRP receptor [11, 12]. Translation continues on the ER and the emerging polypeptide can co-translationally enter the ER through the translocon [2], which is a channel that contains several Sec proteins and spans the lipid bilayer [13].

Also during this time, or in some cases once translation is complete [3], a signal peptidase cleaves the short signal peptide allowing the free protein to enter the ER lumen [14]. If the protein is destined to be an integral membrane protein, determined by the presence of a stretch of hydrophobic residues or stop-transfer membrane anchor sequence, translocation will pause [15]. At this point the protein will be shifted laterally and become anchored within the phospholipid bilayer where it remains [15]. Transmembrane proteins can either contain one hydrophobic stretch of amino acids, and are classified as single pass transmembrane proteins, or contain multiple regions that cross the membrane and are classified as multi-pass transmembrane proteins [3].

If the protein is not destined to be integrated into the membrane, but instead enter the secretory pathway or the lumen of membrane-bound organelles, the protein begins the process of transport. Once translation is complete and the signal peptide has been cleaved the ribosomes are released back into the cytosol [16, 17]. For mRNAs translated by stably-bound ER ribosomes, mRNAs are released and ribosomes may remain bound to the ER and participate in multiple rounds of translation [18, 19]. For cytosolic proteins translated on ER-bound ribosomes it is not clear how these mRNAs are recruited to the ER or what populations of ribosomes are utilized to initiate translation, although a recent study indicates that the ER-resident protein p180 may play a role in the translation-independent recruitment of mRNAs to the ER [20].

Following protein synthesis and translocation into the ER lumen, a protein destined for secretion must undergo proper folding and modifications, with the aid of chaperones and folding enzymes. These modifications include N-linked glycosylation, disulfide bond formation and oligomerization [3]. At this point the fate of the secretory proteins is determined. If the protein functions in the ER, for example as a chaperone, then proper folding will commence. If the protein is destined for secretion, it will be released by the chaperones and packaged for travel through the Golgi on to a final destination (such as the plasma membrane or secreted) or move into peroxisomes [21]. Additionally, the cytosolic regions of the transmembrane protein may interact with cytosolic proteins or chaperones to properly fold these domains.

On the other hand, even with several proteins and complexes dedicated to folding proteins properly, a fraction of proteins do not achieve native and functional form and are either misfolded or aggregated [22]. These proteins can either remain in the ER or enter the ER-associated degradation (ERAD) pathway mediated by the proteasome, assuring that aberrant polypeptides do not inadvertently enter the secretory pathway [23]. Recognition of misfolded proteins, followed by clearing of these aggregates through the ERAD pathway, needs to be tightly controlled so as not to affect cellular function [23]. Interestingly, there are several connections to activation of ER stress response pathways and pathological human conditions. Several neurodegenerative protein misfolding diseases, such as Alzheimer’s disease, activate ER stress response pathways. Additionally, activation of the ER stress response pathway is observed in diabetes, inflammatory bowel disease, and various cancers. How ER stress response pathways play a role in these pathologies is an active area of research and various components of the stress response pathways are being investigated as potential therapeutic targets [24]. In general, the protein synthesis functions of the ER are confined to ER sheets and regulation of ER structure by RNA localization and ER stress will be covered later in this review.

단백질 합성과 접힘 

ER의 주요 기능 중 하나는

분비 단백질과 막 관통 단백질 [8],

그리고 일부 세포질 단백질 [1]의 합성 장소로 작용하는 것이다.

단백질 합성은

리보솜이 ER 막의 세포질 쪽 면에 국소화되는 것을 필요로 하며,

단백질 합성을 조절하는 정형 경로는

mRNA:리보솜 복합체가 ER 막에 공-번역(co-translational) 방식으로 도킹하는 것이다.

분비성 또는 막 관통 단백질의 번역은 세포질에서 시작되며,

신생 폴리펩타이드의 아미노 말단에 있는 신호 서열(signal sequence)이

신호 인식 입자(Signal Recognition Particle, SRP)에 의해 인식되고 결합되어

리보솜이 ER 막으로 모집된다 [9, 10].

mRNA:리보솜:

신생 폴리펩타이드:SRP 복합체는 ER로 이동하여 SRP 수용체에 도킹한다 [11, 12].

번역은 ER 상에서 계속 진행되며,

신생 폴리펩타이드는 트랜스로콘(translocon)을 통해 ER 안으로 공-번역적으로 들어갈 수 있다 [2].

트랜스로콘은

여러 Sec 단백질을 포함하며 지질 이중층을 가로지르는 채널이다 [13].

이 과정 중, 또는 때로는 번역이 완료된 후 [3], 신호 펩티다아제(signal peptidase)가 짧은 신호 펩티드를 절단하여 자유 단백질이 ER 내강(lumen)으로 들어갈 수 있게 한다 [14]. 만약 단백질이 막 관통 단백질로 운명지어진 경우(소수성 아미노산 서열 또는 stop-transfer 막 고정 서열의 존재에 의해 결정됨), 전위(translocation)가 일시 정지한다 [15]. 이 시점에서 단백질은 측면으로 이동하여 인지질 이중층 내에 고정되어 남게 된다 [15]. 막 관통 단백질은 하나의 소수성 아미노산 서열을 포함하는 단일 통과(single-pass) 막 단백질이거나, 막을 여러 번 가로지르는 다중 통과(multi-pass) 막 단백질로 분류될 수 있다 [3].

단백질이 막에 통합되지 않고 대신 분비 경로나 막 결합 소기관의 내강으로 들어갈 운명이라면, 단백질은 수송 과정을 시작한다. 번역이 완료되고 신호 펩티드가 절단된 후 리보솜은 다시 세포질로 방출된다 [16, 17]. ER에 안정적으로 결합된 리보솜에 의해 번역된 mRNA의 경우, mRNA는 방출되지만 리보솜은 ER에 남아 여러 번의 번역 라운드에 참여할 수 있다 [18, 19]. ER 결합 리보솜에서 번역된 세포질 단백질의 경우, 이 mRNA들이 ER로 어떻게 모집되는지 또는 어떤 리보솜 집단이 번역을 시작하는 데 이용되는지는 아직 명확하지 않다. 다만 최근 연구에 따르면 ER 상주 단백질인 p180이 mRNA의 번역-비의존적 ER 모집에 역할을 할 수 있음이 시사되었다 [20].

단백질 합성과 ER 내강으로의 전위 이후, 분비될 운명의 단백질은

샤페론(chaperones)과 접힘 효소의 도움을 받아

적절한 접힘과 수정을 거쳐야 한다.

이러한 수정에는

N-결합 글리코실화(N-linked glycosylation),

이황화 결합(disulfide bond) 형성,

올리고머화(oligomerization) 등이 포함된다 [3].

이 시점에서

분비 단백질의 운명이 결정된다.

단백질이 ER 내에서 기능하는 경우(예: 샤페론으로서),

적절한 접힘이 진행된다.

분비될 운명이라면

샤페론으로부터 방출되어 골지체를 거쳐

최종 목적지(예: 세포막 또는 세포 외 분비)로 이동하거나

과산화물소체(peroxisomes)로 이동하게 된다 [21].

또한

막 관통 단백질의 세포질 영역은

세포질 단백질이나 샤페론과 상호작용하여 이 영역을 제대로 접히도록 한다.

반면,

단백질을 제대로 접히게 하는 여러 단백질과 복합체가 존재함에도 불구하고,

일부 단백질은 본래의 기능적 형태를 이루지 못하고

잘못 접히거나 응집된다 [22].

이러한 단백질은 ER에 남아 있거나,

프로테아좀에 의해 매개되는 ER 관련 분해(ER-associated degradation, ERAD) 경로로 들어가,

비정상적인 폴리펩타이드가

실수로 분비 경로로 들어가지 않도록 한다 [23].

잘못 접힌 단백질의 인식과 응집체 제거는

세포 기능에 영향을 주지 않도록 엄격하게 조절되어야 한다 [23].

흥미롭게도

ER 스트레스 반응 경로의 활성화와

여러 인간 병리 상태 사이에는 밀접한 연관이 있다.

알츠하이머병과 같은 여러 신경퇴행성 단백질 접힘 질환은

ER 스트레스 반응 경로를 활성화시킨다.

또한

당뇨병, 염증성 장질환, 다양한 암에서도

ER 스트레스 반응 경로의 활성화가 관찰된다.

ER 스트레스 반응 경로가

이러한 병리 현상에서 어떤 역할을 하는지는 활발한 연구 분야이며,

스트레스 반응 경로의 다양한 구성 요소들이 잠재적 치료 표적으로 연구되고 있다 [24].

일반적으로

ER의 단백질 합성 기능은

ER 시트에 국한되어 있으며,

RNA 국소화와 ER 스트레스에 의한 ER 구조 조절은

본 리뷰의 후반부에서 다루어질 것이다.

Lipid biogenesis

While the ER is a major site of protein synthesis, it is also a site of bulk membrane lipid biogenesis [4], which occurs in the endomembrane compartment that includes the ER and Golgi apparatus. Proteins and phospholipids, which are the major lipid component of membranes, are transferred and biochemically modified in the region of the ER that is in close juxtaposition to the Golgi apparatus [25]. This region, known as the ER-Golgi intermediate compartment (ERGIC), is rich in tubules and vesicles [4]. Once lipids are mobilized to the ERGIC they are distributed throughout the cell through organelle contacts or secretory vesicles [26]. The cis-Golgi, which is the closest structure to the ERGIC, leads to the trans-Golgi network where vesicles carrying newly synthesized secretory proteins from the ER form and bud [4]. The trans-Golgi network has traditionally been viewed as the main sorting station in the cell where cytosolic cargo adaptors are recruited to bind, indirectly or directly, and transport proteins or lipids [27].

지질 생합성

ER은

단백질 합성의 주요 장소인 동시에,

대량 막 지질 생합성의 주요 장소이기도 하다 [4].

1. 지질 합성의 중심지 = ER (소포체)

• 대부분의 인지질(phospholipids)과 중성지질(neutral lipids, TAG)이 ER에서 합성.
• ER은 지질 합성 공장 역할을 하며, 여기서 만들어진 지질이 다른 소기관으로 공급되어야 합니다.

2. 지질 이동의 핵심: 막 접촉 부위(MCS)를 통한 직접 전달 세포 내에서 지질은 물에 잘 녹지 않기 때문에 소포(vesicle)를 타지 않고, 접촉 부위에서 직접 전달됩니다. (효율적이고 빠름)

• ER ↔ Mitochondria (ERMES 복합체) 가장 중요한 접촉 부위. ER에서 합성된 포스파티딜세린(PS) 등이 미토콘드리아로 이동 → 미토콘드리아에서 포스파티딜에탄올아민(PE), 카르디올리핀 등으로 바뀜. ERMES (Mmm1, Mdm12, Mdm34, Mdm10)가 tether + lipid transfer tube 역할을 합니다.
• ER ↔ Lipid Droplets (LD) ER에서 만들어진 중성지질(TAG)이 ER 막에 축적되어 지질 방울(Lipid Droplets)로 분리됩니다. Seipin (Fld1/Ldb16) 복합체가 LD 형성과 ER-LD 접촉을 조절합니다.
• ER ↔ Vacuole (액포) NVJ (Nuclear-Vacuole Junction) 등을 통해 지질 이동과 저장이 조절됩니다.
• 기타 접촉 부위
  • Vacuole ↔ Mitochondria (vCLAMP)
  • Lipid Droplets ↔ Mitochondria
  • Autophagosome 등

3. 전체적인 의미 (지질 합성 관점)

• ER이 지질을 “만들고”, MCS를 통해 미토콘드리아(에너지 생산에 필요), 지질 방울(에너지 저장), 액포(저장/분해) 등으로 빠르게 공급

이는 ER과 골지체를 포함하는 엔도막 구획(endomembrane compartment)에서 일어난다. 단백질과 인지질(막의 주요 지질 성분)은 ER과 골지체가 밀접하게 인접한 ER-골지 중간 구획(ER-Golgi intermediate compartment, ERGIC) 영역에서 이동되고 생화학적으로 수정된다 [25]. 이 영역은 세관과 소포가 풍부하다 [4]. ERGIC로 이동된 지질은 소기관 접촉 또는 분비 소포를 통해 세포 전체로 분배된다 [26]. ERGIC에 가장 가까운 cis-골지체는 trans-골지 네트워크로 이어지며, 여기서 ER에서 새로 합성된 분비 단백질을 실은 소포가 형성되고 출아(bud)한다 [4]. trans-골지 네트워크는 전통적으로 세포 내 주요 분류 정거장으로 여겨져 왔으며, 여기서 세포질 화물 어댑터(cargo adaptors)가 모집되어 단백질이나 지질을 간접적 또는 직접적으로 결합하고 수송한다 [27].

https://pmc.ncbi.nlm.nih.gov/articles/PMC12522807/

1. ER stress 억제 기전 (핵심)

• 타우린은 UPR(Unfolded Protein Response) 주요 경로를 억제합니다:
  • PERK, ATF6, IRE1α 경로 ↓
  • GRP78(BiP), XBP1, CHOP 발현 ↓
• 특히 tunicamycin(ER stress 유도제)로 유발된 모델에서 IRE1α, ATF6, XBP1, GRP78, CHOP를 유의하게 감소시켰습니다.
• 결과적으로 CHOP 의존성 세포 사멸을 막고, 간세포 보호 효과를 나타냅니다.

2. 지질 합성·대사 개선 기전

• SREBP-1c 억제: Insig-2a mRNA 발현을 증가시켜 SREBP-1c를 ER에 붙잡아 둡니다. → 지질 합성 유전자(Fas, SCD-1) 발현 ↓ → de novo lipogenesis(새로운 지방 합성) 감소.
• 담즙산 대사 촉진: CYP7A1 발현 ↑ → 콜레스테롤을 담즙산으로 전환·배출 증가 → 간 내 콜레스테롤 ↓.
• 항산화 효과: MDA, GSSG ↓ / GSH, SOD ↑ → 산화 스트레스 감소 (ER stress와 악순환 차단).
• 기타: AMPK 활성화, SIRT1 ↑, Nrf2 경로 활성화, CD36(지방산 uptake) ↓ 등.

3. 전체적인 MASLD 개선 효과

• 고지방 식이 또는 고탄수화물 식이 모델에서 타우린 투여 시:
  • 간 지방 축적(hepatic steatosis) 감소
  • 혈중 TG, TC, LDL 개선
  • 인슐린 저항성 개선
  • 장-간 축(gut-liver axis) 조절 (장벽 강화, 내독소 감소, SCFA 증가)

결론 (논문의 핵심 메시지)

타우린은
항산화 + ER stress 완화 + 지질 대사 조절을 통해
MASLD의 초기 단계(단순 지방간)부터 진행(염증, 섬유화)까지 억제할 수 있는
유망한 영양 보조제

Calcium (Ca2+) metabolism

Finally, while the ER is a major site of synthesis and transport of a variety of biomolecules, it is also a major store of intracellular Ca2+ [28, 29]. The typical cytosolic concentration of Ca2+ is ~100 nM, while the Ca2+ concentration in the lumen of the ER is 100–800 μM, and the extracellular Ca2+ concentration is ~2 mM [6, 30]. The ER contains several calcium channels, ryanodine receptors and inositol 1,4,5-trisphosphate (IP3) receptors (IP3R) that are responsible for releasing Ca2+ from the ER into the cytosol when intracellular levels are low [6]. Ca2+ release occurs when phospholipase C (PLC) is stimulated through G protein-coupled receptor (GPCR) activation [31] and cleaves phosphatidylinositol 4,5 bisphosphate (PIP2) into diacyl-glycerol (DAG) and IP3, which can then bind the IP3R leading to Ca2+ release and transient increase in intracellular Ca2+ levels [6]. Ryanodine receptors (RyRs) act through Ca2+-induced Ca2+ release (CICR), when the receptors bind Ca2+ in response to increased cytoplasmic levels of Ca2+ [32]. In addition, depolarization of t-tubule membranes can lead to conformational changes in voltage-dependent Ca2+ channels, such as dyhydropyridine receptors (DHPRs), which interact and activate RyRs leading to Ca2+ release [33]. Furthermore, Ca2+ can leak from the ER into the cytoplasm only to be pumped back into the ER via sarcoendoplasmic reticular Ca2+ ATPases (SERCAs), or can enter the cell from the extracellular media, adding to the layers of regulation [6]. If ER stores of Ca2+ are rapidly depleted through IP3 receptor (IP3R)-mediated release a mechanism for Ca2+ entry into the cell is activated, known as store-operated Ca2+ entry (SOCE) [6, 34]. After ER luminal Ca2+ depletion, STIM1 proteins cluster in regions of ER abutting the plasma membrane. At these regions, clustered STIM1 traps plasma membrane-diffusing Orai1 subunits [35, 36] and assembles them into active Ca2+ release-activated channels (CRAC) allowing for uptake of extracellular Ca2+ into the ER lumen to restore Ca2+ levels [37–39]. Interestingly, SOCE and activation of CRAC does not depend on, nor sense, changes in levels of Ca2+ in the cytoplasm [6], but senses and responds to changes in luminal Ca2+ concentration.

Calcium is a widespread signaling molecule that can affect diverse processes including localization, function and association of proteins, either with other proteins, organelles or nucleic acids. Release of Ca2+ can result in a wave of Ca2+ that moves through the entire cell [40], a gradient of Ca2+ from the source of release, or a spatially-restricted wave from clustered channels known as a Ca2+ spark [41]. One of the most well-studied Ca2+ release events occurs at fertilization following sperm entry [40, 42], but also occurs during muscle contraction and secretion [6] as well as neuronal processes including neurotransmitter release [43]. We will highlight recent evidence that Ca2+ may also play a role in reshaping the ER in response to cellular signals.

칼슘(Ca²⁺) 대사 

마지막으로,

ER은 다양한 생체 분자의 합성과 수송의 주요 장소인 동시에,

세포 내 칼슘의 주요 저장소이기도 하다 [28, 29].

일반적인

세포질 칼슘 농도는 약 100 nM인 반면,

ER 내강의 칼슘 농도는 100–800 μM이며,

세포 외 칼슘 농도는 약 2 mM이다 [6, 30].

ER에는

여러 칼슘 채널, 라이아노딘 수용체(ryanodine receptors), 그리고

이노시톨 1,4,5-삼인산(IP₃) 수용체(IP₃R)가 존재하여

세포 내 칼슘 농도가 낮아졌을 때

ER에서 세포질로 칼슘을 방출한다 [6].

칼슘 방출은

G 단백질 결합 수용체(GPCR) 활성화를 통해 포스포리파아제 C(PLC)가 자극되어

포스파티딜이노시톨 4,5-비스인산(PIP₂)을

디아실글리세롤(DAG)과 IP₃로 절단할 때 발생하며,

IP₃가 IP₃R에 결합하여 칼슘 방출을 유발하고 세포 내 칼슘 농도를 일시적으로 증가시킨다 [6].

라이아노딘 수용체(RyRs)는 세포질 칼슘 농도가 증가하면 칼슘에 의해 유발되는 칼슘 방출(Ca²⁺-induced Ca²⁺ release, CICR) 기전을 통해 작용한다 [32]. 또한 t-소관 막의 탈분극은 전압 의존성 칼슘 채널(예: 디하이드로피리딘 수용체, DHPRs)의 형태 변화를 유발하여 RyRs를 활성화시켜 칼슘 방출을 일으킨다 [33]. 더 나아가, 칼슘은 ER에서 세포질로 누출된 후 사르코엔도플라스믹 레티큘럼 칼슘 ATPase(SERCAs)에 의해 다시 ER로 펌핑되거나, 세포 외 매질로부터 유입되어 조절의 여러 층위를 더한다 [6]. IP₃ 수용체(IP₃R)에 의한 방출로 ER 칼슘 저장소가 빠르게 고갈되면, 저장소 작동 칼슘 유입(store-operated Ca²⁺ entry, SOCE)이라는 메커니즘이 활성화된다 [6, 34]. ER 내강 칼슘이 고갈된 후, STIM1 단백질은 세포막에 인접한 ER 영역에 클러스터를 형성한다. 이 영역에서 클러스터된 STIM1은 세포막을 확산하는 Orai1 서브유닛을 포획하여 [35, 36] 활성 칼슘 방출 활성화 채널(CRAC)을 조립하고, 세포 외 칼슘을 ER 내강으로 유입시켜 칼슘 수준을 회복시킨다 [37–39]. 흥미롭게도 SOCE와 CRAC 활성화는 세포질 칼슘 수준의 변화에 의존하지 않으며 이를 감지하지도 않지만 [6], 내강 칼슘 농도의 변화를 감지하고 반응한다.

칼슘은 단백질의 국소화, 기능, 그리고 다른 단백질, 소기관, 또는 핵산과의 결합에 영향을 미치는 광범위한 신호 분자이다. 칼슘 방출은 세포 전체를 가로지르는 칼슘 파동 [40], 방출원으로부터의 칼슘 농도 구배, 또는 클러스터된 채널로부터의 공간적으로 제한된 파동(칼슘 스파크) [41]을 일으킬 수 있다. 가장 잘 연구된 칼슘 방출 사건 중 하나는 정자 유입 후 수정(fertilization) 시 발생하는 것이며 [40, 42], 근육 수축, 분비 [6], 신경 전달물질 방출 [43] 등에서도 일어난다. 최근 연구에서는 칼슘이 세포 신호에 반응하여 ER의 형태를 재구성하는 데에도 역할을 할 수 있다는 증거가 제시되고 있다.

Regulation of ER shape and function

The ER is a complex organelle, involved in protein and lipid synthesis, calcium regulation and interactions with other organelles. The complexity of the ER is reflected in an equally complex physical architecture. The ER is composed of a continuous membrane system that includes the nuclear envelope (NE) and the peripheral ER, defined by flat sheets and branched tubules (Fig. 1). The shape and distribution of these ER domains is regulated by a variety of integral membrane proteins and interactions with other organelles and the cytoskeleton. These interactions are dynamic in nature and reflect changes within the cell, either through cell cycle or developmental state, cell differentiation, intracellular signals or protein interactions. While it is generally known how the basic shapes of ER sheets and tubules are determined, it is relatively unclear how changes in shape or the ratio of sheets to tubules occur in response to specific cellular signals.

ER 형태와 기능의 조절 

ER은

단백질과 지질 합성, 칼슘 조절, 그리고 다른 소기관과의 상호작용에 관여하는 복잡한 소기관이다.

ER의 복잡성은

그에 상응하는 복잡한 물리적 구조에 반영된다.

ER은

핵막(nuclear envelope, NE)과 말초 ER(peripheral ER)로 이루어진 연속적인 막 시스템으로,

평평한 시트(flat sheets)와 분지된 세관(branched tubules)으로 정의된다(Fig. 1).

이러한

ER 도메인의 형태와 분포는

다양한 막 관통 단백질과 다른 소기관, 세포골격과의 상호작용에 의해 조절된다.

이러한 상호작용은

본질적으로 동적이며,

세포 주기나 발생 상태, 세포 분화, 세포 내 신호, 또는 단백질 상호작용 등 세포 내 변화에 따라 달라진다.

ER 시트와 세관의 기본 형태가 어떻게 결정되는지는 대체로 알려져 있지만,

특정 세포 신호에 반응하여 형태가 변화하거나 시트와 세관의 비율이 어떻게 조절되는지에 대해서는 상대적으로 불분명하다.

Fig. 1.

Open in a new tab

Various ER structural morphologies. a Location of the ER visualized in a HeLa cell transfected with GFP-Sec61β. Inset shows the polygonal network of the peripheral ER magnified ×3 relative to the magnification in a. This view highlights the relationship of the ER to the nuclear envelope (red arrow). b ER morphology from the same HeLa cell depicting an image plane closer to the coverslip. This highlights the complexity of the peripheral ER. c ER network formed in Xenopus egg extracts. Three-way junctions, ER tubules and small ER sheets are highlighted (red arrows). d ER network formed in Xenopus egg extracts highlighting large ER sheets containing ribosomes (red arrow). Scale bar for a–d is 10 μM and is shown in a. e Electron micrograph (EM) of rough ER from guinea pig pancreas. Reprinted with permission from James Jamieson. Scale bar is 0.1 μM. f EM of smooth ER from ocular rabbit muscle. Reprinted with permission from Fig. 4 [164]. Magnification is ×50,000

Here, we will discuss what is known about how the structures of ER are formed, how the dynamics of the ER are regulated, and how these dynamics change in response to cell cycle state and cellular cues. In addition, we provide examples of how the proteins that are involved in contributing to ER shape are influenced by these cellular cues, such as calcium release, and how this is reflected in the dynamics of the ER and ultimately the function of specialized cells that display varying ratios of sheets to tubules.

ER structure

There have been several excellent, recent reviews that cover the topic of general ER structure in detail [7, 44–48], so we will limit our review of the basic ER structure to only those factors that may play a role in changing the shape of ER in response to signaling. The ER consists of the nuclear envelope and the peripheral ER, which includes smooth tubules and rough sheets. While the ER is defined as an interconnected network with a continuous membrane, the different structures that make up the ER perform very diverse and specialized functions within the cell.

The nuclear envelope is made up of two lipid bilayers, the inner nuclear membrane (INM) and outer nuclear membrane (ONM), and shares a common lumen with the peripheral ER. Hundreds of nuclear pores spanning the ONM and INM of the nuclear envelope allow transport of molecules, including RNAs and proteins, at various rates of diffusion or regulated transport depending on the size of the molecule. The nuclear envelope is connected to sheets, or cisternae, that make up part of the peripheral ER. Sheets are flat in nature consisting of two lipid bilayers with an intervening lumen, with curved regions located only at the membrane edges. Peripheral ER Sheets may vary in size, but the luminal spacing is very consistent, usually about 50 nm in mammals and 30 nm in yeast [49] (Fig. 2). Sheets are usually observed in a stacked conformation and are connected via regions of twisted membranes with helical edges [50]. These rough sheets, as defined by the high density of ribosomes on the cytosolic surface [51, 52], are the main site of synthesis, folding and post-translational modifications for secreted or membrane-bound proteins. In turn, far fewer ribosomes are present on the membrane surface of ER tubules [52], which is highly curved and smooth and may not accommodate the binding of large polysomes (Fig. 2). The tubular network is dynamic, continually rearranging and growing, and is defined by three-way junctions that connect individual tubules (Fig. 1). While tubules and sheets possess very different structural features, and hence play a role in different cellular processes, the luminal spacing of both tubules and sheets is similar [49, 52].

Fig. 2.

Open in a new tab

Structure of ER sheets and tubules. a ER sheets and tubules have a diameter of 30–50 nm in eukaryotes. Eukaryotic ribosomes are 25–30 nm and localize to the flat regions of ER sheets, giving the sheets a rough appearance (rough ER). Ribosomes are present in much lower numbers on tubules, giving the tubules a more smooth appearance (smooth ER). b Models of potential hairpin topologies of REEP family proteins that act as wedges to promote bending of the membrane, adapted from [63]

Interestingly, ER tubules and sheets are found in all eukaryotic cells [53], though the ratio of sheets to tubules varies in different cell types and reflects the different functions of these cells. For example, the ER architecture of specialized cells that synthesize vast amounts of secreted proteins, such as pancreatic secretory cells and B cells, is largely made up of sheets (Fig. 1). In turn, cells that are involved in processes including lipid synthesis, calcium signaling and sites of contact for other organelles possess an ER composed of primarily tubules (Fig. 1). Adrenal, liver and muscle cells are all examples of specialized cells with a predominantly tubular network and reflects the function of these cells [54].

An additional configuration of the peripheral ER includes cortical ER, which abuts the plasma membrane and displays an intermediate phenotype between sheets and tubules with membranes that are both highly curved as well as regions that are flat in nature. Calcium signaling occurs at the contact sites between the plasma membrane and the abutting cortical ER and is necessary for muscle contraction [55, 56]. Therefore, the morphology and intracellular location of the ER subdomains contribute to the function of these structures and hence the role of the specialized cell in which they are located.

Improved microscopy techniques have allowed for the characterization of different ER structures, and the ratios of these structures to one another, in specialized cell types. When comparing the roles of these cells in the organism, it is clear that the type and amount of peripheral ER present reflects the function of that particular cell type. It is still unclear how these ratios are generated and what cellular signaling pathways play a role in designating which ER type will be most prominent in a particular cell type.

ER shaping proteins

ER tubules

Peripheral ER structures are just as distinct and diverse as the set of proteins that contribute to their shape. Several proteins have been identified that promote specific ER structures, but perhaps the most well-studied group of proteins include the reticulon family of proteins that localize to tubules and the highly curved edges of ER sheets [51, 57]. These integral membrane proteins contribute to the bending of the membrane by forming a transmembrane hairpin topology that acts as a wedge, displacing lipids in the outer leaflet of the bilayer leading to curvature of the membranes [57]. These proteins tend to form oligomers and are much less mobile than other ER-resident proteins [58]. Overexpression‎ of some reticulon isoforms leads to formation of long ER tubules at the expense of sheets [58]. In turn, depletion of reticulons, and hence the ability to bend membranes, leads to a reduction in the number of ER tubules, leading to an expansion of peripheral sheets [57, 59, 60]. Therefore, the level of reticulons within a cell determines the abundance and fine structure of ER tubules.

Reticulons do not act alone in shaping ER tubules. Members of the DP1/Yop1/REEP5/6 and REEP1-4 family, which are abundant ER-resident proteins that specifically localize to tubules and edges of sheets, also act as tubule-promoting factors. DP1/Yop1, or REEP5/6 [61], proteins share a similar transmembrane hairpin architecture with the reticulons (Fig. 2), leading to the stabilization of the curved membranes of tubules [57, 58, 62]. Interestingly, REEP1-4 proteins have a topology distinct from REEP5/6 suggesting that these proteins may have slightly different functions in shaping the ER than the closely related REEP5/6 proteins [63] (Fig. 2). Additionally, purified reticulons and DP1/Yop1 family proteins were able to induce tubule formation from purified vesicles [62], demonstrating that these proteins play an essential role in ER tubule growth.

Reticulons and DP1/Yop1 promote tubule formation, but additional factors are required to promote the formation of the tubular network and characteristic three-way junctions through homotypic fusion. Atlastins, members of the dynamin-like GTPase family, mediate these homotypic fusion events. Depletion by RNAi or expression‎ of dominant-negative atlastin in cells results in a lack of fusion events leading to an abundance of long, unbranched tubules [61]. When a dominant-negative cytoplasmic fragment from Xenopus, which contains the GTPase domain but lacks the transmembrane domain and cytoplasmic tail [64], are introduced into Xenopus interphase extracts ER network formation was blocked [65]. Comparable point mutations that prevent dimerization of the cytoplasmic fragment of human atlastin [66] were made in the Xenopus cytoplasmic atlastin protein, added into interphase extract and had no effect on ER network formation [65]. Furthermore, antibodies directed against atlastin inhibit ER network formation when introduced into Xenopus egg extracts [61]. In Drosophila, atlastin depletion leads to ER fragmentation and purified atlastin is sufficient to catalyze GTP-dependent fusion of proteoliposomes [64, 66, 67]. Therefore, studies from multiple organisms, extracts and purified components indicate that atlastin is likely required for catalyzing homotypic vesicle fusion between ER membranes, which is important for proper network formation.

Recently, a few new key players have been identified that are involved in ER dynamics. Work using purified ER vesicles derived from Xenopus eggs has demonstrated that GTP is required for homotypic ER vesicle fusion in the absence of cytosolic factors [57, 68]. Previous studies indicated that GTPases are required for ER fusion events [69, 70], and a recent study utilized a proteomics approach to identify Rab10 as a factor required for ER assembly [71]. Knock-down of Rab10, or overexpression‎ of a GDP-locked dominant-negative point mutant, in cultured human cells caused an increase in ER sheets and a decrease in tubules [71]. ER–ER fusion events occurred at regions where Rab10 was enriched. Rab10 was found to co-localize with several lipid-synthesizing enzymes, including phosphoinositol synthase (IS) and choline/ethanolamine phosphotransferase (CEPT1) [71], leading to the possibility that this may represent a previously unidentified ER subdomain or compartment. It is currently not clear what role Rab10 plays in the ER vesicle fusion reaction or how homotypic ER vesicle fusions are coupled to lipid synthesis.

Recent work has also identified a role for Rab18, which is targeted to the ER by Rab3 GTPase activating protein (GAP) complex, in ER dynamics. Depletion of Rab18 leads to a phenotype similar to that observed following Rab10 inhibition [72]. Additionally, when Rab10 is depleted, Rab18 redistributes to peripheral sheets [72]. Therefore, it appears that depletion of either Rab10 or Rab18 prevents the stabilization of ER tubule fusion, reducing the density of tubules resulting in an increase in ER sheets. Depletion of the Caenorhabditis elegans RAB-5, which has been previously implicated in early endosome function [73], phenocopies the peripheral ER defects seen in the RET-1 and YOP-1 (homologs of Rtn4a and DP1) depletions [70]. In addition to the role RAB-5 plays in peripheral ER formation, kinetics of nuclear envelope disassembly is affected in these mutants [70].

In addition to GTPases that may play a direct role in homotypic membrane fusion of vesicles, recent work has demonstrated a role for lipid synthesizing enzymes in controlling the shape and organization of the ER. Inhibition of C-terminal domain (CTD) nuclear envelope phosphatase-1 (CNEP-1), which is enriched on the nuclear envelope and promotes the synthesis of membrane phospholipids, led to the appearance of ectopic sheets that encased the nuclear envelope, interfering with nuclear envelope breakdown [74]. These results reflect the interconnected network of proteins and functions that play a role in shaping the structures of the ER.

The ER is a very dynamic network that is constantly undergoing rearrangements and remodeling [75]. ER tubules are continually fusing and branching resulting in the creation of new three-way junctions. In a competing process, junction sliding and tubule ring closure leads to loss of three-way junctions and the characteristic polygonal structure [76]. Very little is known about the complexes controlling this process, but it was recently discovered that Lunapark (Lnp1) localizes to and stabilizes three-way junctions [77, 78]. Lnp1 binds to reticulons and Yop1, and localization of Lnp1 to junctions is regulated by Sey1p, the yeast homolog of atlastin [78]. Loss of Lnp1 leads to a collapsed and densely reticulated ER network in yeast and human cultured cells [77, 78], though only half of the junctions are bound to Lnp1 [77], which reflects the fluidity of the ER network. If Lnp1 is overexpressed, the protein localizes to the peripheral ER and induces the formation of a large polygonal tubular network [79]. Additionally, formation of this network was inhibited by Lnp1 mutations that blocked N-myristoylation [79], an attachment of myristic acid (a 14-carbon saturated fatty acid), indicating that this modification plays a critical role in Lnp1-induced effects on ER morphology. N-myristoylation is not required for membrane translocation, topology formation, or protein localization to the ER but may play a role in protein–protein or protein-lipid interactions that are required for morphological changes in the ER, though the exact molecular mechanism of action remains to be elucidated [79].

The actual mechanism for Lnp1-mediated stabilization of three-way junctions is unknown, though recent studies and insights from the structure and domains within the protein shed light on how Lnp1 stabilizes junctions [77, 78]. First, Lnp1 contains two transmembrane domains as well as a zinc finger domain, which is located on the cytoplasmic face of the ER membrane [77]. When cysteines were mutated within the zinc finger domain, the polygons became smaller and regions lacking cortical ER were more apparent as the number of cysteines mutated increased [78]. Therefore, mutations in the zinc finger domain may affect protein–protein interactions, complex formation or interfere with the distribution of resident lipids on the cytoplasmic face of the membrane causing deleterious effects on junction stabilization. In addition, the transmembrane domains may be acting as an inverted wedge, adding to the local negative curvature characteristic of three-way junctions [77], and acting opposite to the positive curvature promoted by reticulons. Another possibility is that multiple Lnp1 proteins may also act cooperatively together to stabilize the junction, or Lnp1 may be acting transiently to stabilize or modify lipids or other proteins at junctions [77].

In addition to proteins that regulate membrane structure and dynamics, there is accumulating evidence that changing the nucleic acid content of the ER can also impact ER shape. Early experiments showed that brief treatment of tissue culture cells with the translation inhibitor puromycin, which dissociates mRNA:ribosome complexes, leads to loss of ribosomes from the ER and a loss of ER sheets [51, 80]. This suggests that the presence of mRNA:ribosome complexes may stabilize ER sheets. In support of this hypothesis, our recent work identified an ER-localized ribonuclease, XendoU [81], that changes the RNA content of the ER in response to changes in free Ca2+ concentration [82, 83]. These changes occur at physiologically relevant levels of ~1.5 μM, which mimics release of Ca2+ from intra- and extracellular stores at fertilization [42, 84]. A subpopulation of XendoU localizes to the ER and co-immunoprecipitates with a number of ER-resident proteins [82]. Depletion of XendoU leads to the formation of long, unbranched tubules in Xenopus leavis egg extract, and rescue of this phenotype requires intact catalytic activity of the protein, indicating that the nuclease function is critical to proper ER network formation [82]. Furthermore, antibody addition to purified vesicles leads to a block in network formation, demonstrating that XendoU acts on the surface of ER membranes to regulate ER structure [82]. Interestingly, addition of 5′5′-dibromo BAPTA, a strong calcium chelator, blocked vesicle fusion in this system [68]. Depletion of XendoU also leads to a delay in replication and nuclear envelope closure [82], and BAPTA blocks nuclear envelope formation in Xenopus egg extract reconstitution experiments [85]. Together these results suggested that XendoU acts on membranes to degrade RNAs.

Upon vesicle fusion it was found that RNAs were degraded and released from the surface of membranes, suggesting that XendoU acts to degrade these RNAs, as well as release proteins, to clear patches of membrane to allow for vesicle formation leading to network formation [82]. Interestingly, when purified vesicles were treated with increasing concentrations of RNaseA and subjected to the same assay, an increasingly aberrant network formed with large vesicles that were unable to fuse [82]. Results from in vitro studies indicate that XendoU is activated on membranes in coordination with calcium release to locally degrade RNAs and clear patches of membranes leading to fusion in a controlled manner to fine tune network formation.

Lastly, similar to other proteins that play a role in tubule formation, knock-down of the human homolog EndoU in cultured human cells leads to an expansion of sheets [82]. Additionally, rescue of the expanded sheet phenotype depended on intact catalytic function as observed with recombinant protein in the extract system. Therefore, XendoU is an example of a protein that is activated in response to cellular cues to regulate proper ER formation, and further studies may reveal additional proteins that are regulated in this manner to fine tune organelle structure.

ER sheets

We have considered how tubules are formed and maintained, which leads the discussion to sheets, the other peripheral ER structure. First, we must consider how sheets are formed. Several mechanisms have been proposed, including the idea that integral membrane proteins can span the intraluminal space and form bridges, connecting the lipid bilayers [51, 86, 87]. These proteins may either stabilize the structure or define the distance between the two lipid layers based on the size of the proteins. Additionally, these proteins or protein complexes may form a scaffold that aids in the stabilization of the sheets or bring the two lipid membranes in closer proximity [86]. Several proteins including Climp63, p180 and kinectin have been implicated in the generation, maintenance and stabilization of ER sheets [51].

In addition to highly enriched membrane proteins and core components of the translocon, Climp63, a coiled–coiled protein with a single transmembrane domain, was identified along with kinectin and p180 in a mass spectrometry screen for abundant integral ER membrane proteins [51]. Through various techniques and in various cell types Climp63 was shown to be a highly abundant protein [88–90] that localizes to perinuclear ER and is absent from the nuclear envelope [91, 92]. Very stable oligomers of Climp63 can form, restricting mobility of the protein along the membrane, promoting localization to the rough ER [92]. Overexpression‎ of Climp63 leads to a massive proliferation of ER sheets while reduction in expression‎ surprisingly does not lead to loss of sheets but instead a decrease in the distance between sheets [51]. Moreover, these sheets are spread diffusely throughout the cytoplasm, reminiscent of the phenotype of cells treated with the translation inhibitor puromycin [51]. This is interesting as the core components of the translocon, the protein channel that interacts with ribosomes and is responsible for translocating nascent peptides into the ER or anchoring transmembrane segments of newly synthesized proteins, were found to be enriched on sheets [93]. Therefore, these results suggest that the role of Climp63 in formation of sheets is likely to involve additional factors and acts as a part of an elaborate regulatory network that balances the production of sheets and tubules.

ER microtubule interactions

It is clear that proteins involved in the promotion, maintenance or stabilization of peripheral ER structures function through interactions with additional proteins or structures, and these interactions are key to proper formation of the ER network. Interestingly, several of the proteins discussed above have been shown to interact with microtubules, including Climp63 [91], p180 [94], kinectin [95] and STIM1 (discussed below). One important interaction discussed below is with microtubules. The ER network exhibits several dynamic interactions with microtubules that are important for determining the distribution of the ER within the cell. The two main types of interactions between the ER and microtubules are Tip Attachment Complexes (TACs) and sliding along preformed microtubules by the action of kinesin and dynein motors [96–100]. In cultured cells treated with nocodazole to depolymerize microtubules, the ER retracts from the periphery [101], though the retraction does not occur immediately. Further investigation revealed that sliding events occurred mainly on a small subset of microtubules, modified by acetylation, that are more resistant to nocodazole treatment [76]. Furthermore, ER tubules can form in the absence of microtubules [57, 65, 68], raising many questions and leading several groups to study the interaction between ER and microtubules more in-depth.

In the past 10 years we have learned a great deal about what proteins are responsible for the intrinsic shape of the ER and how these proteins are connected to specific ER subdomains. However, we know very little about how cellular signals communicate with ER shaping proteins to change the shape of the ER in response to cellular signals.

Changes in ER structure during mitosis

During mitosis many cellular structures are dramatically remodeled to facilitate chromosome segregation. One of the most dramatic examples is changes to the microtubule cytoskeleton that occur as a result of increased microtubule dynamics caused by the action of cyclin-dependent kinases. The increase in microtubule dynamics during mitosis is important for the bipolar attachment of chromosomes to the mitotic spindle and accurate segregation to daughter cells during anaphase [102]. In addition to changes to the microtubule cytoskeleton, essentially all organelles change shape and function during mitosis to facilitate accurate organelle inheritance and orderly chromosome segregation. The ER undergoes dramatic shape changes during mitosis and recent studies are beginning to uncover the mechanisms linked to these structural changes.

In organisms with an open mitosis the nuclear envelope breaks down at the onset of mitosis to allow free exchange between the nucleus and cytoplasm. Nuclear envelope breakdown (NEBD) is a carefully orchestrated process that begins during mitotic prophase [103]. During prophase components of the nuclear pore dissociate from the pore, the nuclear lamina depolymerizes, and the membrane-bound proteins of the nuclear envelope retract into the general ER. These events free the chromosomes of nuclear lamina and membranes to facilitate chromosome condensation and segregation. In general, the events of nuclear envelope breakdown are thought to be driven by the phosphorylation of components of the NE during mitosis by various mitotic kinases, especially cyclinB:cdk1, although many molecular details are still unclear.

Concomitant with changes that occur to the nuclear envelope during NEBD the ER also begins to undergo dramatic shape changes. Changes in ER shape during mitosis have been studied in many different organisms by both light and electron microscope and these studies have resulted in a conflicting series of reports about the shape of the ER during mitosis. However, during the last few years a consensus has begun to emerge that the mitotic ER is primarily composed of sheets. Early studies using live cell microscopy in both Drosophila and C. e legans embryos demonstrated that the ER changed from a mixture of sheets and tubules to almost exclusively sheets during mitosis [104, 105]. Additionally, work using thin section transmission EM in HeLa cells also concluded that the majority of the ER was present in sheets throughout mitosis [106]. However, two studies in a variety of mammalian tissue culture cells [80, 107] have used both live cell microscopy and electron microscopy to suggest that the ER is primarily tubular during mitosis, and two additional studies [60, 108] also suggested that the ER remained tubular during mitosis and further suggested that end-on binding of ER tubules to chromatin during mitosis initiates nuclear envelope reassembly at the end of mitosis. One potential difficulty in interpreting the shape of the mitotic ER is that most cells round up during mitosis which can make acquisition of light and electron microscopy images difficult and require laborious reconstruction of the images into a three dimensional model. In addition, the mitotic ER is highly dynamic, which can complicate acquisition of live cell images during mitosis. To address these questions a series of recent studies have used both high-resolution, high-speed live cell microscopy and high-resolution EM to demonstrate that the ER is almost exclusively composed of sheets during mitosis [109, 110]. In addition, these studies demonstrate that the nuclear envelope reforms through the docking of ER sheets onto regions of chromatin that are isolated from spindle microtubules [109]. Finally, to circumvent many or the problems associated with imaging large, three dimensional cells during mitosis a recent study has examined the structure of the ER in vitro using ER reconstituted from Xenopus egg extracts [65]. This study convincingly demonstrated that ER formed in mitotic extracts is primarily composed of sheets while interphase ER is primarily composed of tubules. In addition, the authors demonstrated that active cyclinB:cdk1 was sufficient to convert a tubular ER into a primarily sheet based ER. Taken together all of these studies present conflicting views of the shape of the ER during mitosis, but a consensus is emerging from a wide variety of organisms that the mitotic ER is primarily composed of sheets and that the shape changes in the ER are related to changes in cyclin:cdk activity.

In addition to changes in the gross morphology of the ER during mitosis there are also dramatic changes in the distribution of proteins throughout the ER. During interphase the ER is organized into distinct domains with certain proteins defining different domains. For example, the tubule-shaping reticulon protein Rtn4 is exclusively present in the peripheral ER and excluded from the nuclear envelope [57, 60, 110]. In contrast, some proteins, such as the Lamin B receptor and components of the nuclear pore, are exclusively present in the nuclear envelope and are excluded from the peripheral ER [60, 110], while some proteins, like Sec61β, are present in all ER subdomains. However, during mitosis the NE retracts into the ER and there is nearly complete mixing of the specialized ER-shaping proteins [60, 110]. At the end of mitosis proteins that define the NE and peripheral ER are rapidly resorted such that they reestablish their characteristic interphase organization [60, 110]. In addition, it has been shown that overexpression‎ of Rtn4 or knockdown of three reticulons (Rtn1, Rtn3, Rtn4) can either slow or speed the rate of NE reassembly at the end of mitosis, although the mechanism through which these proteins affect NE formation is currently unknown. These studies highlight the massive reorganization that takes place in the ER during mitosis and suggests that different expression‎ levels of specific ER shaping proteins can control ER reorganization during mitosis. However, we know very little about how various ER shaping proteins are resorted to specific domains at the end of mitosis.

Two very recent studies [111, 112] have begun to provide insight into the specialized processes that regulate nuclear envelope reformation at the end of mitosis. Both of these studies identified a transient localization of the ESCRT-III complex to the surface of chromatin during late anaphase when the nuclear envelope is beginning to reform. ESCRT-III is best known for its role in the formation of multivesicular bodies during endocytosis, but also has well-documented roles in cytokinesis and viral budding from the plasma membrane [113]. Both studies demonstrated that the membrane binding and deformation properties of ESCRT-III are required for nuclear envelope formation. Additionally, interactions with the microtubule severing enzyme spastin and the ubiquitin recognition factor UFD1 are important for nuclear envelope reformation. These results demonstrate that an endosomal complex is important for regulating NE reformation and suggest that ESCRT-III could potentially play a role in additional aspects of ER dynamics.

The redistribution of ER shaping proteins during mitosis suggests that the fundamental activities of some of these proteins are modified during mitosis. For example, the mitotic ER is composed of primarily sheets, yet Rtn4, which promotes tubule formation [57], is distributed throughout the ER [60, 110]. This result suggests that the tubule-promoting activity of Rtn4 may be modified during mitosis to facilitate the tubule-to-sheet transition observed during mitosis. Inspection of large-scale phospho-proteomics studies reveals that a large number of ER-shaping proteins have identified mitosis-specific phosphorylation sites [114–121]. Although none of the phosphorylation sites identified in these large-scale screens has been studied in detail their presence and specificity to mitosis suggests that these are likely to be involved in reshaping the ER during mitosis.

In support of the hypothesis that mitosis-specific phosphorylation of ER-shaping proteins regulates ER remodeling during mitosis two studies have examined this phenomenon in detail. A study of the ER sheet promoting protein Climp63 [51] has demonstrated mitosis-specific phosphorylation on three N-terminal residues [121]. Phosphorylation of Climp63 blocks the interaction of Climp63 with microtubules. Additionally, phosphomimetic mutants blocked the interaction of the ER with microtubules during interphase and resulted in an ER composed primarily of sheets, while nonphosphorylatable mutants tethered the ER to microtubules and resulted in an extremely distorted ER. These results suggest that mitotic phosphorylation of Climp63 likely blocks the interaction of the ER with microtubules and could be an important step in the tubule-to-sheet transition that occurs during mitosis. A second study examined the interaction of the ER with growing microtubule plus ends during mitosis. During interphase the ER-associated protein STIM1 interacts with the microtubule plus end-binding protein EB1 to couple ER reshaping to microtubule polymerization [122]. However, during mitosis the ER is excluded from the mitotic spindle and does not exhibit plus tip growth events. A recent study [123] has demonstrated that STIM1 is specifically phosphorylated during mitosis to control the interaction of the ER with microtubules. Specifically, phosphorylation of STIM1 blocks the interaction with the plus-end tracking protein EB1. Nonphosphorylatable mutants of STIM1, created by mutation of 10 S/T residues that block all mitotic phosphorylation, result in a recruitment of the ER throughout the spindle by restoration of the interaction of STIM1 with EB1, demonstrating that phosphorylation is a major mechanism that regulates the association of the ER with microtubules during mitosis. Interestingly, phosphorylation of STIM1 also blocks activation of SOCE, although this occurs independently of the STIM1:EB1 interaction [118]. Clearly much more work remains before we have a clear understanding of how cell cycle signaling cascades contribute to reshaping of the mitotic ER.

While the above studies demonstrated that phosphorylation of key proteins that link the ER to the microtubule cytoskeleton is important for excluding the ER from the spindle during mitosis a recent study demonstrated the importance of an interaction of the ER with microtubules for clearing the ER from mitotic chromatin. During mitosis the nuclear envelope is absorbed into the ER and is cleared from the surface of the chromatin, however little is known about the mechanisms that regulate ER removal from the chromatin. A recent study used a biochemical approach to identify proteins that bind to both membranes and microtubules to identify new ER proteins REEP3/4 [124]. The authors demonstrate that RNAi against REEP3/4 results in a failure to remove membranes from chromosomes during mitosis, resulting in chromosome segregation defects and internuclear membrane inclusions. Interestingly, the authors further demonstrate that removal of membranes from mitotic chromatin requires the interaction of REEP3/4 with microtubules. However, it is not known if REEP3/4 is subject to phosphoregulation during mitosis or if the microtubule-binding activity or REEP3/4 is required for shaping the ER during interphase. Taken together these three studies demonstrate that interaction of the ER with microtubules is a major mechanism that contributes to shape rearrangement during mitosis and that ER:microtubule interactions are regulated by mitotic phosphorylation. In addition, these studies demonstrate that the ER interacts with microtubules using many different adaptor proteins and that these different adaptor proteins serve different functions during mitosis.

Changes in ER during oocyte maturation and fertilization

One of the greatest changes during development occurs at fertilization. As in mitosis, the transition from oocyte to embryo requires many coordinated cellular changes including release from meiotic arrest, resumption of mitosis, fusion of pronuclei, activation of signaling cascades and changes in protein expression‎ [125–128]. In order for development to proceed normally, the egg must undergo the proper calcium response in order to initiate the developmental program and embryogenesis [129].

While the exact mechanism and conformational changes vary slightly among all organisms studied, the ER architecture in oocytes of all animals changes including Xenopus [130, 131], sea urchin [132], starfish [133] and mouse [134]. Initial studies in starfish oocytes revealed that the ER is comprised of interconnected sheets of membranes, though following germinal vesicle breakdown (GVBD), the ER sheets wrap around yolk platelets resembling a shell [133]. In immature mouse oocytes, large clusters were found deep within the cytoplasm [134]. Following GVBD, the spindle and surrounding ER migrate to the cortex leading to another round of ER reorganization into vegetally localized clusters in the metaphase II egg in addition to a finer reticular network throughout the egg [134, 135]. Interestingly, these steps are dependent on the microtubule network as nocodazole and inhibition of cytoplasmic dynein both prevent the ER reorganization [135]. Formation of the ER clusters is prevented by the depolymerization of microfilaments, but not microtubules [135]. Given the timing of each of these reorganizations, it seems likely that they are related to increases in cyclinB:cdk1 activity that occurs upon oocyte maturation [136]. These observations show an additional time in development where the ER and microtubule network interact to regulate ER structure.

In Xenopus immature oocytes, the network in both the animal (pigmented) half and vegetal (unpigemented) half appears to be uniform and consists of tubules and individual, unstacked sheets [130]. Additionally, the vegetal half contains annulate lamellae, stacks of sheets with membranes containing densely packed nuclear pores [130]. In mature eggs, the ER in the animal half is unchanged, however the annulate lamellae in the vegetal half disappeared. Interestingly, it has been proposed that the annulate lamellae share many properties with the nuclear envelope [137]. In place of the annulate lamellae dense, irregularly shaped ER clusters were present. The appearance of these clusters coincided with germinal vesicle breakdown. These clusters disappeared and reappeared throughout maturation and upon fertilization dispersed and permanently disappeared. The reorganization of the ER is coupled to the cell cycle as the clusters present in mature eggs contain IP3 receptors [130] and release calcium from IP3 channels at fertilization [138, 139].

Along with these changes comes a transient intracellular calcium wave, initiated during sperm entry, released from the ER and extracellular stores [40, 42, 140–142]. There is one major difference in eggs of mice versus eggs of frogs. Frogs, as well as sea urchin [143] and starfish [133, 144] have a single calcium transient at fertilization [145]. Other animals, including mice and humans, have multiple smaller calcium transients following fertilization, and these differences may be reflected in the ER organization in mature eggs [145]. Mice [134] and frogs display ER clusters that are similar in size and location (the side opposite the meiotic spindle) and possess IP3 receptors [130, 146]. However, fertilization in mice occurs on the side with the ER clusters whereas fertilization in frogs occurs in the animal pole where the meiotic spindle is located. Therefore, the clusters may be involved in secondary calcium wave propagation. The organization of the ER network, and the reorganization throughout oogenesis, serves as a functional consequence of calcium signaling and propagation in these organisms [129]. We currently do not know much about the molecular mechanisms that lead to changes in ER shape during meiotic maturation and fertilization, and this should be a major are of research interest.

ER changes in response to ER stress

As seen so far, the ER is an organelle of many different functions that must be tightly regulated to carry out the proper functions. One of the most prominent functions of the ER is protein synthesis. Even with several chaperones and folding enzymes in place, an accumulation of unfolded or misfolded proteins in the lumen of the ER can occur. When the cell undergoes this type of stress there are several things that must occur to retain balance and proper function, including translational inhibition, degradation of unfolded or misfolded proteins, and an increase in the production of chaperones and folding enzymes to restore normal function of the ER and the cell. If the balance is not restored it can lead to cell death or apoptosis [147], therefore achieving normal function is critical to the survival of the cell.

As discussed above, once a peptide destined for secretion has entered the lumen of the cell, there are several modifications that occur, including N-linked glycosylation, disulfide bond formation and oligomerization [3]. N-linked glycosylation can occur co-translationally as the protein is translocated into the ER lumen. The oligosaccharyltransferase (OST) can modify the Asparagine within the Asn-X-Ser/Thr sequence once it has traversed approximately 13 amino acids into the ER lumen [148], which improves the kinetics and thermodynamics of folding for proteins [149, 150]. Misfolding can occur due to the unique environment of the lumen and the high protein concentration of both newly synthesized proteins, proteins ready for secretion and proteins that act as molecular chaperones and folding enzymes. Logistically, due to the high protein concentration and packing in the lumen, the folding enzymes must first identify and find the proper target protein for folding to take place. If proteins are not modified correctly, the lack of glucose residues is recognized by the ER and proteins including UDP-glucose:glycoprotein glucosyltransferase (UGGT) in an attempt to re-glycosylate the protein [151–153]. If the normal folding process is not restored, hydrophobic residues are exposed and bound by Grp78, accumulation of these proteins occurs and the unfolded protein response (UPR) is activated [154, 155]. The first action of the UPR is to increase ER abundance to accommodate the needs of the cell to properly fold the proteins, leading to an expansion of the ER through the generation of sheets [156] and an increase in the ER folding machinery.

The UPR consists of three parallel branches that are activated upon stress and include inositol requiring enzyme 1 (IRE1) by nonconventional splicing, double-stranded RNA-activated protein kinase (PKR)-like ER kinase (PERK) through translational control by phosphorylating eIF2α, and activating transcription factor 6 (ATF6) through regulated proteolysis [155]. Briefly, activation of these pathways lead to production of b-ZIP transcription factors that activate UPR genes [155]. First, ER-resident IRE1, a transmembrane endoribonuclease, mediates the post-transcriptional, non-canonical splicing of XBP1 mRNA that is localized to the ER [157–159] and encodes a transcription factor involved in upregulating additional stress response genes. Additionally, the nuclease activity of IRE1 is involved in degradation of a subset of ER-associated RNAs in a process known as IRE1-dependent decay (RIDD) [160, 161]. The cell has evolved this mechanism to reduce the translational load on the ER by removing mRNAs that otherwise would be translated, and may be one way for the cell to upregulate stress-response genes that are needed in the UPR. Although it is clear that ER-stress leads to large scale changes in the protein and RNA content of the ER, it is not yet clear if this leads to immediate structural reorganization in order to accommodate the new needs of the organelle. In addition, it is not yet clear if activation of stress-responsive signaling pathways leads to the modification of intrinsic structural components of the ER. Interestingly, it has been observed that splicing of XBP1 is activated during meiosis in both Xenopus and budding yeast [162, 163], suggesting that changes in ER structure during meiosis could be linked to the ER stress response. These would both be interesting avenues of future research exploring structural changes in the ER in response to cellular signaling cues.

소포체 스트레스에 대한 ER의 변화

지금까지 살펴본 바와 같이,

소포체(ER)는 다양한 기능을 수행하는 소기관으로,

이 기능들을 제대로 수행하기 위해서는 세밀한 조절이 필요하다.

소포체의 가장 두드러진 기능 중 하나는

단백질 합성이다.

여러 샤페론(chaperone)과

접힘 효소(folding enzyme)가 존재함에도 불구하고,

소포체 내강(lumen)에는 접히지 않거나 잘못 접힌 단백질(unfolded or misfolded proteins)이 축적될 수 있다.

세포가 이러한 스트레스를 받을 때,

균형과 정상 기능을 유지하기 위해 번역 억제(translational inhibition),

잘못 접힌 단백질의 분해,

그리고 샤페론과 접힘 효소의 생산 증가 등이 일어나 소포체와 세포의 정상 기능을 회복해야 한다.

만약 균형이 회복되지 않으면

세포 사멸(apoptosis)로 이어질 수 있으므로 [147],

정상 기능을 회복하는 것은 세포 생존에 매우 중요하다.

위에서 논의한 바와 같이,

분비될 펩티드가 소포체 내강으로 들어오면

N-연결 글리코실화(N-linked glycosylation),

이황화 결합(disulfide bond) 형성,

올리고머화(oligomerization) 등의 여러 변형이 일어난다 [3].

N-연결 글리코실화는

단백질이 소포체 내강으로 이동되는 동안 공번역적으로(co-translationally) 일어날 수 있다.

올리고당 전이효소(oligosaccharyltransferase, OST)는

Asn-X-Ser/Thr 서열 내의 아스파라진(Asn)을,

약 13개의 아미노산이 소포체 내강으로 들어간 후에 변형시킬 수 있으며 [148],

이는 단백질 접힘의 동역학(kinetics)과 열역학(thermodynamics)을 개선한다 [149, 150].

잘못된 접힘(misfolding)은

소포체 내강의 독특한 환경과,

새로 합성된 단백질,

분비 준비가 된 단백질,

그리고 분자 샤페론 및 접힘 효소 등으로 인한 높은 단백질 농도 때문에 발생할 수 있다.

물류적으로,

높은 단백질 농도와 밀집으로 인해 접힘 효소는

먼저 목표 단백질을 정확히 찾아야 접힘이 일어날 수 있다.

단백질이 제대로 변형되지 않으면,

포도당 잔기가 부족한 것을 소포체가 인식하고,

UDP-포도당:당단백질 글루코실전이효소(UGGT) 등이

단백질을 다시 글리코실화하려고 시도한다 [151–153].

정상적인 접힘 과정이 회복되지 않으면

소수성 잔기(hydrophobic residues)가

노출되어 Grp78(BiP)에 결합되고,

이러한 단백질이 축적되면서 미접힘 단백질 반응(Unfolded Protein Response, UPR)이 활성화된다 [154, 155].

UPR의 첫 번째 반응은

단백질을 제대로 접히게 하기 위해 세포의 필요에 맞춰 소포체의 양을 증가시키는 것이다.

이는 시트(sheet) 형태의 소포체를 생성하여

ER을 확장(expansion)시키고,

소포체 접힘 기계장치(folding machinery)를 증가시킨다 [156].

UPR은 스트레스 시 활성화되는 세 가지 병렬 경로로 구성되어 있다:

• 이ositol 요구 효소 1(IRE1)에 의한 비전형적 스플라이싱(nonconventional splicing)
• 이중가닥 RNA 활성화 단백질 키나아제(PKR)-유사 ER 키나아제(PERK)에 의한 eIF2α 인산화로 인한 번역 조절
• 활성화 전사인자 6(ATF6)에 의한 조절된 단백질 분해(regulated proteolysis) [155]

간단히 말하면,

이 경로들의 활성화는 b-ZIP 전사인자(transcription factor)를 생산하여

UPR 유전자를 활성화한다 [155].

먼저,

ER에 상주하는 IRE1(막 관통 엔도리보뉴클레아제)은

ER에 위치한 XBP1 mRNA의 비전형적 스플라이싱을 매개하며 [157–159],

이는 추가적인 스트레스 반응 유전자를 상향조절(upregulating)하는 전사인자를 암호화한다.

또한 IRE1의 뉴클레아제 활성은

IRE1 의존적 분해(RIDD, IRE1-dependent decay) 과정에서

특정 ER 관련 RNA를 분해하는 데 관여한다 [160, 161].

세포는 이 기전을 통해

ER에 대한 번역 부하를 줄이고,

UPR에 필요한 스트레스 반응 유전자를 상향조절한다.

ER 스트레스가

소포체의 단백질과 RNA 함량에 대규모 변화를 일으킨다는 것은 분명하지만,

이러한 변화가 소포체의 새로운 필요를 수용하기 위해

즉각적인 구조적 재구성(structural reorganization)으로 이어지는지는 아직 명확하지 않다.

또한,

스트레스 반응 신호 전달 경로의 활성화가

소포체의 고유한 구조적 구성 요소를 변형시키는지도 아직 분명하지 않다.

흥미롭게도,

XBP1 스플라이싱이 Xenopus와 효모에서 감수분열(meiosis) 동안 활성화되는 것이 관찰되었으며 [162, 163],

이는 감수분열 중 ER 구조 변화가 ER 스트레스 반응과 연결될 수 있음을 시사한다.

이는 세포 신호 전달 자극에 대한

ER 구조 변화 탐구의 흥미로운 미래 연구 방향이 될 것이다.

Closing remarks

The ER is a complex organelle that plays a pivotal role in protein and lipid synthesis, calcium storage and stress response. Changes in structure in response to cell cycle or developmental state render this organelle highly dynamic. Several proteins play a role in the proper formation of the different structures of the peripheral ER including the nuclear envelope, sheets and tubules. Regulation exists at multiple steps in the formation and maintenance of these structures, and the ratios of these structures are very different in cells of different functions. In general, cells involved in synthesizing large amounts of protein have higher ratios of sheets, whereas cells involved in lipid synthesis or signaling with other organelles would have higher ratios of tubules. The generation of these structures relies on a myriad of proteins, involved in either structural aspects of ER morphology by directly affecting the phospholipid bilayer and curvature of membranes or mediating interactions with other organelles or the cytoskeleton. In addition, proteins with other functions, including nucleases and GTPases, also play a role in network formation. Recent work has begun to connect our knowledge of the proteins that provide the fundamental shape of the ER to signaling pathways, but much work remains to be done to understand how developmental, cell cycle, and stress pathways change the fundamental shape of the ER in different circumstances. Recent work on several different human diseases has highlighted a role for several different ER-shaping proteins in diverse diseases such as Alzheimer’s and Hereditary Spastic Paraplegia (HSP) [reviewed in 7]. The strong link of ER-shaping proteins to hereditary human diseases highlights the need for further research into the basic biology of the ER and how this biology changes in response to changes in cellular environment.

References

• 1.Reid DW, Nicchitta CV. Diversity and selectivity in mRNA translation on the endoplasmic reticulum. Nat Rev Mol Cell Biol. 2015;16(4):221–231. doi: 10.1038/nrm3958. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 2.Rapoport TA. Protein translocation across the eukaryotic endoplasmic reticulum and bacterial plasma membranes. Nature. 2007;450(7170):663–669. doi: 10.1038/nature06384. [DOI] [PubMed] [Google Scholar]
• 3.Braakman I, Hebert DN. Protein folding in the endoplasmic reticulum. Cold Spring Harb Perspect Biol. 2013;5(5):a013201. doi: 10.1101/cshperspect.a013201. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 4.Fagone P, Jackowski S. Membrane phospholipid synthesis and endoplasmic reticulum function. J Lipid Res. 2009;50(Suppl):S311–S316. doi: 10.1194/jlr.R800049-JLR200. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 5.Hebert DN, Garman SC, Molinari M. The glycan code of the endoplasmic reticulum: asparagine-linked carbohydrates as protein maturation and quality-control tags. Trends Cell Biol. 2005;15(7):364–370. doi: 10.1016/j.tcb.2005.05.007. [DOI] [PubMed] [Google Scholar]
• 6.Clapham DE. Calcium signaling. Cell. 2007;131(6):1047–1058. doi: 10.1016/j.cell.2007.11.028. [DOI] [PubMed] [Google Scholar]
• 7.Westrate LM, Lee JE, Prinz WA, Voeltz GK. Form follows function: the importance of endoplasmic reticulum shape. Annu Rev Biochem. 2015;84:791–811. doi: 10.1146/annurev-biochem-072711-163501. [DOI] [PubMed] [Google Scholar]
• 8.Jan CH, Williams CC, Weissman JS. Principles of ER cotranslational translocation revealed by proximity-specific ribosome profiling. Science. 2014;346(6210):1257521. doi: 10.1126/science.1257521. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 9.Walter P, Blobel G. Translocation of proteins across the endoplasmic reticulum. II. Signal recognition protein (SRP) mediates the selective binding to microsomal membranes of in-vitro-assembled polysomes synthesizing secretory protein. J Cell Biol. 1981;91(2 Pt 1):551–556. doi: 10.1083/jcb.91.2.551. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 10.Walter P, Ibrahimi I, Blobel G. Translocation of proteins across the endoplasmic reticulum. I. Signal recognition protein (SRP) binds to in-vitro-assembled polysomes synthesizing secretory protein. J Cell Biol. 1981;91(2 Pt 1):545–550. doi: 10.1083/jcb.91.2.545. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 11.Gilmore R, Blobel G, Walter P. Protein translocation across the endoplasmic reticulum. I. Detection in the microsomal membrane of a receptor for the signal recognition particle. J Cell Biol. 1982;95(2 Pt 1):463–469. doi: 10.1083/jcb.95.2.463. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 12.Meyer DI, Krause E, Dobberstein B. Secretory protein translocation across membranes-the role of the “docking protein’. Nature. 1982;297(5868):647–650. doi: 10.1038/297647a0. [DOI] [PubMed] [Google Scholar]
• 13.Deshaies RJ, Sanders SL, Feldheim DA, Schekman R. Assembly of yeast Sec proteins involved in translocation into the endoplasmic reticulum into a membrane-bound multisubunit complex. Nature. 1991;349(6312):806–808. doi: 10.1038/349806a0. [DOI] [PubMed] [Google Scholar]
• 14.Evans EA, Gilmore R, Blobel G. Purification of microsomal signal peptidase as a complex. Proc Natl Acad Sci USA. 1986;83(3):581–585. doi: 10.1073/pnas.83.3.581. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 15.Blobel G. Intracellular protein topogenesis. Proc Natl Acad Sci U S A. 1980;77(3):1496–1500. doi: 10.1073/pnas.77.3.1496. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 16.Seiser RM, Nicchitta CV. The fate of membrane-bound ribosomes following the termination of protein synthesis. J Biol Chem. 2000;275(43):33820–33827. doi: 10.1074/jbc.M004462200. [DOI] [PubMed] [Google Scholar]