9.1 The nucleus #
9.1.1 A gate, not a translocon #
The nuclear envelope has holes, the nuclear pore complexes, that are selective rather than sealed. Proteins below about 40 kDa diffuse through slowly. Larger proteins, ribosomal subunits, and messenger ribonucleoproteins are actively carried. The cargo does not have to be unfolded. The pore opens around it. This is topologically different from Sec61 or from the mitochondrial translocons, which thread unfolded chains through a narrow channel.
9.1.2 Signals #
A classical nuclear localisation signal (NLS) is a short run of basic residues, such as PKKKRKV in SV40 large T antigen, or a bipartite signal with two basic clusters. A nuclear export signal (NES) is a short leucine-rich motif recognised by the exportin CRM1. A protein that shuttles has both, and its steady-state location depends on which signal is currently exposed. Phosphorylation often hides or reveals a signal, which is how many transcription factors wait in the cytosol until a pathway calls them in.
9.1.3 Ran and direction #
Direction is set by a gradient of the small GTPase Ran.
- In the nucleus, the guanine-nucleotide exchange factor RCC1 is bound to chromatin, so nuclear Ran is mostly Ran-GTP.
- In the cytosol, Ran-GAP hydrolyses Ran-GTP, so cytosolic Ran is mostly Ran-GDP.
Import uses importin ฮฒ, often with the adaptor importin ฮฑ that binds the classical NLS. The importinโcargo complex crosses the pore by interacting with phenylalanine-glycine repeats of the nucleoporins. In the nucleus, Ran-GTP binds importin ฮฒ and the cargo is released. Importin returns to the cytosol with Ran-GTP, and hydrolysis there frees the importin for another load.
Export is the reverse arrangement. CRM1 binds NES cargo only when Ran-GTP is also bound, so the trimeric complex assembles in the nucleus and falls apart in the cytosol when the GTP is hydrolysed. The same gradient drives both directions because the two receptors respond to Ran-GTP in opposite ways.
Messenger RNA export is an exception to remember: the principal mRNA exporter (NXF1/NXT1 in animals) does not use Ran. The cargo is handed to it only after splicing and processing are complete, so export is coupled to quality control of the RNA.
9.2 Mitochondria #
9.2.1 Post-translational and unfolded #
Mitochondria make only 13 proteins in humans. The other several hundred are encoded in the nucleus, made on cytosolic ribosomes, and imported after translation. Cytosolic Hsp70 chaperones keep the precursor unfolded. Import uses energy at two steps: the electrical potential across the inner membrane pulls the positively charged signal in, and ATP-driven chaperones in the matrix (mtHsp70) complete translocation and help the protein fold. There is no SRP and no co-translational coupling for the standard matrix protein.
9.2.2 Signals and translocons #
A matrix-targeting sequence is an N-terminal amphipathic helix with positive charges on one face. It binds receptors of the TOM complex in the outer membrane and is handed to the TIM23 complex in the inner membrane. The two complexes can line up at contact sites so the protein crosses both membranes at once. The matrix processing peptidase then cleaves the signal, and the protein folds, often with the help of matrix chaperonins.
Other destinations use other machines, and the names are the level of detail expected here.
| Destination | Route |
|---|---|
| Matrix | TOM then TIM23; signal cleaved |
| Inner membrane, carrier proteins such as the ATP/ADP carrier | TOM, chaperones of the intermembrane space (small TIM proteins), then TIM22; internal signals, not a cleavable N-terminal helix |
| Outer membrane, ฮฒ-barrel proteins such as porin | TOM, then the SAM/TOB complex, which inserts the barrel from the intermembrane-space side |
| Proteins encoded by mitochondrial DNA | Inserted from the matrix side by Oxa1, a homologue of a bacterial membrane-insertion protein |
The Oxa1 route is a fossil of the bacterial ancestor. The organelle still inserts its own hydrophobic subunits the way the bacterium did, from the inside.
9.3 Chloroplasts #
Chloroplast proteins encoded in the nucleus carry a cleavable N-terminal transit peptide, different in sequence from the mitochondrial signal, though both are positively charged and both are removed after arrival. Import is post-translational. The outer-membrane machine is TOC and the inner-membrane machine is TIC. Stromal chaperones and ATP drive the process. The thylakoid has a further sorting problem, solved by three pathways that recapitulate bacterial export:
Chloroplast
- a Sec pathway for unfolded proteins;
- a Tat pathway that carries folded proteins and is recognised by a twin-arginine motif;
- an SRP-like pathway for some membrane proteins.
A thylakoid protein can therefore have two signals in series: a transit peptide that gets it into the stroma, and a second peptide, exposed by cleavage of the first, that sends it into the thylakoid. Dual targeting to both mitochondria and chloroplasts occurs for some proteins; the signals are related enough that ambiguity has to be resolved by the receptors, and the cell does resolve it.
Chloroplast Ii (Cropped)
9.4 Peroxisomes #
9.4.1 Folded import #
Peroxisomes import proteins from the cytosol after the proteins have folded, and even as oligomers. This is the opposite of mitochondrial import. The receptors cycle.
- PTS1, the common signal, is a C-terminal tripeptide, typically SKL (serine-lysine-leucine). The receptor Pex5 binds it in the cytosol, docks at the peroxisomal membrane, and escorts the cargo in.
- PTS2 is an N-terminal signal recognised by Pex7, which cooperates with Pex5 in animals.
- Membrane proteins use a different set of Pex factors, including Pex19 as a chaperone and Pex3 as a docking site.
ATP is spent in recycling the receptor, not in pushing an unfolded chain. Peroxisomes can grow by importing material and dividing, and they can also arise de novo by maturation of pre-peroxisomal vesicles that come from the ER. Both routes operate.
9.4.2 Zellweger syndrome as a biogenesis disease #
Zellweger syndrome and related disorders are caused by mutations in PEX genes. Peroxisomal enzymes are made, but the organelle is empty or absent, because import fails. Very-long-chain fatty acids accumulate and plasmalogens are deficient. Compare this with a single-enzyme peroxisomal defect, which damages one pathway and leaves the organelle intact. The distinction, biogenesis versus one enzyme, is the same distinction as I-cell disease versus TayโSachs disease in the lysosomal system.
9.5 A comparison worth memorising #
| Destination | Signal | Folding during import | Energy | Machine |
|---|---|---|---|---|
| ER lumen | N-terminal hydrophobic signal, usually cleaved | Unfolded; co-translational | GTP on SRP and receptor; translation itself pushes | SRP, Sec61 |
| Nucleus | NLS or NES | Folded | GTP on Ran | Importins or exportins, nuclear pore |
| Mitochondrial matrix | N-terminal amphipathic basic helix, cleaved | Unfolded | Membrane potential plus ATP | TOM, TIM23 |
| Chloroplast stroma | N-terminal transit peptide, cleaved | Unfolded | ATP | TOC, TIC |
| Peroxisome | C-terminal SKL, or N-terminal PTS2 | Folded | ATP for receptor recycling | Pex5, Pex7 |
9.6 Key terms, summary, and self-check #
Key terms. NLS; NES; importin; Ran; RCC1; TOM; TIM23; TIM22; transit peptide; TOC; TIC; Tat pathway; PTS1; Pex5; Zellweger syndrome.
Summary. Nuclear pores gate folded cargo, and a Ran-GTP gradient makes import and export directional. Mitochondria and chloroplasts import unfolded precursors through paired translocons and cleave the signal. Thylakoids add a second sorting step. Peroxisomes import folded proteins using cycling receptors, and failure of those receptors empties the organelle.
Self-check.
1. Why can a ribosomal subunit pass the nuclear pore but not the mitochondrial TOM channel?
2. What would happen to nuclear import if Ran-GAP were mistakenly located inside the nucleus?
3. A matrix protein and an ATP/ADP carrier both end in the mitochondrion. Which one uses TIM23, and which uses TIM22?
4. Why does a thylakoid protein often have two signal sequences in series?
5. PTS1 is at the C-terminus and is not cleaved in the way a mitochondrial signal is. Why does that suit a receptor that must bind a folded protein?
