Module 0 · The Largest Membrane

ER Architecture & Discovery

The ER was invisible until 1945. Keith Porter, using the newly developed transmission electron microscope at the Rockefeller Institute, saw for the first time the “canalicular system” pervading the cytoplasm. Over the following decade, George Palade at the same institution worked out its architecture, its ribosome decoration (rough ER), its continuity with the nuclear envelope, and its role in secretion — earning the 1974 Nobel Prize in Physiology or Medicine.

1. Sheets, Tubules, and Junctions

The ER has two dominant morphological sub-domains:

Rough ER cisternae: flat membrane stacks studded with ribosomes on the cytoplasmic face. Site of membrane- and secretory-protein synthesis. Enriched in translocon components (Sec61), signal peptidase, TRAPP, and OST glycosyltransferase.
Smooth ER tubules: ~30–50 nm diameter, branched, ribosome-poor. Enriched in reticulons (RTN1/2/3/4) and DP1/REEP/Yop1, which form transmembrane hairpins in the outer leaflet and stabilise high curvature. Sites of lipid synthesis, calcium handling (SERCA and IP₃R enriched), and contact-site formation.

The two domains interconvert — local upregulation of reticulons drives sheet-to-tubule conversion; silencing drives tubule-to-sheet. Cell type-specific preferences: secretory cells (pancreatic acinar, plasma cells) have predominantly rough cisternae; steroidogenic cells (Leydig, adrenal cortex) and hepatocytes have abundant smooth tubules for their biosynthetic work.

2. The Three-Way Junction & Atlastin

ER tubules connect at characteristic three-way junctions, producing the polygonal network visible in live-cell imaging. Homotypic fusion of tubules is catalysed by atlastin, an ER-membrane-anchored dynamin-related GTPase. Atlastin dimerises between opposing membranes, pulls them together, and on GTP hydrolysis drives lipid mixing.

Loss-of-function of atlastin-1 causes hereditary spastic paraplegia type SPG3A (Zhao 2001): in long motor-neurone axons, ER architecture is corrupted, and the distal axon degenerates. Similar phenotypes occur with REEP1 (SPG31) and spastin (SPG4) — the entire ER-shaping machinery is represented in this gene set. It is the clearest example in biology of an organelle shape defect producing a defined human disease.

Simulation: Tubule Bending Cost & Surface Area

Python

script.py52 lines

import numpy as np, matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

# Bending cost of an ER tubule vs radius
# E = (kappa * L) / r (per unit length, for a cylinder: E = pi*kappa*L/r)
# Compare with cisterna (flat): low bending but large edge cost

kappa_kT = 20
r_nm = np.linspace(10, 200, 500)

# Cylinder bending energy per unit length: pi*kappa/r
E_tube = np.pi * kappa_kT / r_nm   # kT per nm

# Pressure needed to keep tubule from collapsing (schematic)
# Cortical tension pulls; reticulon insertion stabilises
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5.5), facecolor='#0a0a1a')
for ax in (ax1, ax2):
    ax.set_facecolor('#111827'); ax.tick_params(colors='#cbd5e1')
    for s in ax.spines.values(): s.set_color('#334155')
    ax.grid(True, color='#334155', alpha=0.3)

ax1.plot(r_nm, E_tube, color='#34d399', lw=2.6, label='kappa = 20 kT')
ax1.plot(r_nm, np.pi * 40 / r_nm, color='#60a5fa', lw=2.3, ls='--', label='kappa = 40 kT (reticulon-stabilized)')
ax1.axvspan(30, 50, color='#fbbf24', alpha=0.15, label='Physiological ER tubule (30-50 nm)')
ax1.set_xlabel('Tubule radius (nm)', color='#cbd5e1')
ax1.set_ylabel('Bending energy (kT per nm length)', color='#cbd5e1')
ax1.set_title('ER tubule: bending cost vs radius',
              color='#6ee7b7', fontweight='bold')
ax1.legend(facecolor='#1e293b', edgecolor='#334155', labelcolor='#cbd5e1')

# Panel 2: S/V ratio
# For a cell 10 um diameter, ~1000 um^3 volume
# ER surface ~30-70x plasma membrane surface
cell_d_um = np.linspace(5, 30, 50)
V_cell = (4/3) * np.pi * (cell_d_um/2)**3
S_pm = 4 * np.pi * (cell_d_um/2)**2
S_er_min = 30 * S_pm
S_er_max = 70 * S_pm
ax2.fill_between(cell_d_um, S_er_min, S_er_max, color='#34d399', alpha=0.3, label='ER surface')
ax2.plot(cell_d_um, S_pm, color='#fbbf24', lw=2.6, label='Plasma membrane')
ax2.set_xlabel('Cell diameter (um)', color='#cbd5e1')
ax2.set_ylabel('Surface area (um^2)', color='#cbd5e1')
ax2.set_yscale('log')
ax2.set_title('ER surface area exceeds PM by 30-70x',
              color='#6ee7b7', fontweight='bold')
ax2.legend(facecolor='#1e293b', edgecolor='#334155', labelcolor='#cbd5e1')

plt.tight_layout()
plt.savefig('output.png', dpi=120, bbox_inches='tight', facecolor='#0a0a1a')
print('Typical ER tubule: 30-50 nm radius, sheared by reticulons/DP1')
print('Total ER area ~30-70x plasma membrane; 50% of internal membrane in the cell')

Click Run to execute the Python code

Code will be executed with Python 3 on the server

3. Continuity with the Nuclear Envelope

The outer nuclear membrane is continuous with the ER; the inner nuclear membrane is topologically separate but shares the perinuclear luminal space. This topology has two consequences: (i) proteins can traffic between ER and inner nuclear membrane by lateral diffusion through the nuclear pore complex channel; (ii) during mitosis in open-mitosis organisms, nuclear envelope breakdown returns the NE to the bulk ER, and its reassembly at telophase is ER-membrane-templated. The ER is thus the “parent” membrane system from which the nuclear envelope is instantiated each cell cycle.

4. ER-Cortical Tethering & Organelle Positioning

The ER is not free-floating. It is tethered to microtubules via STIM1-microtubule tips (whose dynamic ends drag ER tubules outward; Waterman-Storer 1998), to the cortical actin cytoskeleton, and to essentially every other membrane-bound organelle via contact sites (Module 5). These contacts position the ER spatially within the cell and coordinate its activity with mitochondria, endosomes, lysosomes, Golgi, and peroxisomes.

The result is a single, continuous, architecturally complex organelle that encompasses perhaps a third of the cell’s internal membrane by surface area and touches every other compartment.

5. Visualising the ER

Modern live-cell imaging uses GFP-tagged KDEL-receptor or Sec61β-GFP as general ER markers; mCherry-calreticulin for luminal visualisation. Super-resolution (STED, SIM) imaging (Nixon-Abell 2016; Guo 2018) has revised the classical EM picture: what looked like sheets in EM are often dense matrices of 50 nm tubules imaged at diffraction-limited resolution. The living ER is more tubular, more dynamic, and more branched than a generation of textbooks suggested. The underlying topology — sheets, tubules, junctions — remains, but proportions have shifted.

← Overview Module 1: Cotranslational Translocation →