EPYC1/LCI5 (Essential Pyrenoid Component 1, originally described as Low-CO2-inducible protein 5; locus Cre10.g436550) is a nuclear-encoded, chloroplast-targeted, intrinsically disordered repeat protein of about 32 kDa that forms the pyrenoid matrix of Chlamydomonas reinhardtii together with Rubisco. Its sequence consists of near-identical ~60-residue repeats, each containing a short Rubisco-binding region (five in total) that folds into an alpha-helix on binding and sits across the two surface-exposed alpha-helices of a Rubisco small subunit, held by salt bridges and a hydrophobic interface; each Rubisco holoenzyme offers eight such sites. By binding several Rubisco holoenzymes at once through multiple low-affinity sites, EPYC1 cross-links Rubisco into a codependent multivalent network that demixes into a liquid-like condensate. Purified EPYC1 and Rubisco are necessary and sufficient to form dynamic liquid droplets in vitro in which Rubisco remains catalytically active, and expression of EPYC1 with a Rubisco carrying the Chlamydomonas small subunit condenses Rubisco into a proto-pyrenoid in Arabidopsis chloroplasts. EPYC1 is about as abundant as the Rubisco small subunit in the pyrenoid, is strongly induced at low CO2, and is required for normal pyrenoid size, number and matrix density, for packaging most cellular Rubisco into the pyrenoid, and for an efficient CO2-concentrating mechanism and photoautotrophic growth at low CO2. EPYC1 is multiply phosphorylated in its repeats, especially under CO2 limitation; the pyrenoid-localised kinase KEY1 is its main kinase, and phosphorylation weakens EPYC1-Rubisco binding, allowing the condensate to dissolve and change size and number during the cell cycle. An earlier fractionation study placed the protein on the stromal side of thylakoid membranes, but EPYC1 has no transmembrane segment and live imaging places it in the pyrenoid matrix, which is crossed by thylakoid-derived tubules; a dilute pool in the stroma becomes visible when the pyrenoid partly disperses during division.
Definition: The aggregation, arrangement and bonding together of a set of components to form a pyrenoid, the Rubisco-containing, phase-separated organelle of the chloroplast stroma of most algae and some hornworts.
Justification: EPYC1 is the multivalent linker whose binding to Rubisco is necessary and sufficient to form the pyrenoid matrix. GO has a pyrenoid cellular component term but no process term for its assembly, so the role can only be captured with the generic membraneless organelle assembly term.
Parent term: membraneless organelle assembly
Supporting Evidence:
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0140693 molecular condensate scaffold activity | IDA PMID:30498228 The phase separation underlying the pyrenoid-based microalga... | NEW | Summary: EPYC1/LCI5 is a multivalent, intrinsically disordered Rubisco linker. Its five Rubisco-binding regions each bind the two alpha-helices of a Rubisco small subunit, and by bridging many holoenzymes it drives Rubisco to demix into the liquid pyrenoid matrix. Purified EPYC1 and Rubisco are necessary and sufficient for this phase separation in vitro, and interface mutations abolish binding, phase separation and pyrenoid formation in vivo. Reason: No GOA annotations exist for Q94ET8. GO:0140693 is defined as "binding and bringing together two or more macromolecules in contact, permitting those molecules to organize as a molecular condensate", which describes what EPYC1 does to Rubisco. The evidence is direct (reconstitution with purified proteins, PMID:30498228), mechanistic (cryo-EM of the binding interface plus mutations, PMID:33230314) and heterologous (condensate formation in Arabidopsis, PMID:33298923). Wunder et al. reject a "scaffold hypothesis" in the sense of a self-assembled EPYC1 lattice that Rubisco then decorates. That does not conflict with this GO term, which only requires bringing macromolecules together so that they condense. This term replaces the uninformative GO:0005515 protein binding and the unsupported GO:0043169 cation binding that were proposed in an earlier draft. Supporting Evidence: PMID:30498228 Here we use biochemical reconstitution to demonstrate that Rubisco and EPYC1 are the two components necessary and sufficient to bring about a liquid-liquid phase separation (LLPS) that recapitulates the liquid-like behavior reported for the microalgal pyrenoid PMID:33230314 Together, our data demonstrate that EPYC1βs Rubisco-binding regions bind to the Rubisco small subunit Ξ±-helices via salt-bridge interactions and a hydrophobic interface, enabling the condensation of Rubisco into the phase separated matrix. PMID:33230314 Interface mutations disrupt binding, phase separation and pyrenoid formation. PMID:33298923 Expression of EPYC1 in the Arabidopsis line S2Cr results in condensate formation. |
| GO:1990732 pyrenoid | IDA PMID:27166422 A repeat protein links Rubisco to form the eukaryotic carbon... | NEW | Summary: EPYC1 was found by proteomics of isolated pyrenoids, where it is about as abundant as the Rubisco small subunit. Venus-tagged EPYC1 colocalizes with Rubisco in the pyrenoid matrix, and the protein is itself a structural constituent of the matrix condensate. Reason: Direct localization by pyrenoid fractionation and live-cell fluorescence (PMID:27166422), confirmed by the spatial interactome (PMID:28938113), and inherent to its role in forming the matrix. This is the location where the protein acts. It replaces the earlier proposal of GO:0009535 chloroplast thylakoid membrane, which rested on a 2006 fractionation result (PMID:16572472). EPYC1 has no transmembrane segment, and the thylakoid association is best explained by thylakoid tubules running through the pyrenoid matrix. Supporting Evidence: PMID:27166422 Venus-tagged EPYC1 showed clear colocalization with mCherry-tagged RBCS in the pyrenoid PMID:27166422 Strikingly, a fourth protein, previously identified as a low-CO 2 βinduced nuclear-encoded protein (LCI5; Cre10.g436550) ( 24 ), was found in the low-CO 2 pyrenoid fraction with comparable abundance to Rubisco PMID:28938113 essential pyrenoid component 1 (EPYC1), a Rubisco linker protein |
| GO:0140694 membraneless organelle assembly | IMP PMID:27166422 A repeat protein links Rubisco to form the eukaryotic carbon... | NEW | Summary: EPYC1 builds the pyrenoid matrix, a membraneless organelle, by cross-linking Rubisco. In epyc1 mutants pyrenoids are smaller, more numerous and less dense, and most Rubisco stays outside the pyrenoid. The canonical matrix is absent, though a tubule- and starch-associated pyrenoid-like structure remains. Reason: Participation test (CLAUDE.md): EPYC1 is not only necessary for the process; it does part of the work. It is one of the two macromolecules whose multivalent bonds make up the condensate (the scaffold case), as shown by reconstitution with purified proteins and by interface mutations. Comparator check: proteins with molecular condensate scaffold activity (GO:0140693) are commonly also annotated to GO:0140694, for example FUS and Wnk in QuickGO. No more specific "pyrenoid assembly" term exists in GO; this is recorded as a knowledge gap. The earlier draft proposed GO:0015976 carbon utilization, GO:0015979 photosynthesis and GO:0071244 cellular response to carbon dioxide. These are dropped: the first two are too broad, and EPYC1 does not carry out carboxylation (Rubisco is equally active inside and outside droplets, PMID:30498228). The third rested only on low-CO2 induction of expression, which shows regulation, not participation. Supporting Evidence: PMID:27166422 We conclude that EPYC1 is required for Rubisco localization to the pyrenoid not only at low CO 2 , but also at high CO 2 . PMID:33177094 First, in a mutant lacking EPYC1, a pyrenoid-like structure still assembles around the tubules, containing some Rubisco enclosed by a starch sheath, although the canonical matrix is absent (13). PMID:33230314 Cryo-electron tomography supports a model in which EPYC1 and Rubisco form a codependent multivalent network of specific low-affinity bonds, giving the matrix liquid-like properties. |
| GO:0009570 chloroplast stroma | IDA PMID:28938114 The Eukaryotic CO(2)-Concentrating Organelle Is Liquid-like ... | NEW | Summary: Most EPYC1 is in the pyrenoid. A dilute-phase pool in the chloroplast stroma becomes visible when part of the matrix disperses into the stroma during pyrenoid division. Reason: Live imaging of EPYC1-Venus shows EPYC1 in the stroma during division. This is a secondary, non-core location for the dilute phase of the condensate. It is retained because it is directly observed and because it is where KEY1 phosphorylation-driven dissolution sends the protein. The pyrenoid (GO:1990732) is the primary functional location. Supporting Evidence: PMID:28938114 A portion of the RbcS1-Venus and EPYC1-Venus signals rapidly dispersed from the pyrenoid matrix into the stroma for ~20 minutes near the end of chloroplast and pyrenoid division |
Loading supporting contentβ¦
Download this section (compressed HTML)Q: Should GO add a "pyrenoid assembly" process term (is_a membraneless organelle assembly) so that the role of EPYC1 and other pyrenoid matrix and tubule proteins can be captured more specifically?
Suggested experts: Jonikas MC, Mackinder LCM
Q: Is there a GO convention for linking a condensate scaffold to the carbon-concentrating function of the organelle it forms (for example how carboxysome linkers such as CcmM and CsoS2 are annotated), given that EPYC1 does not catalyse carbon fixation itself?
Q: What are the roles of the 14-3-3 proteins FTT1/FTT2 and the kinase KIN4-2 found with EPYC1 in the spatial interactome, and is any phosphatase responsible for re-condensation after division?
Suggested experts: Mackinder LCM
Q: Does EPYC1 have a direct, specific interaction with thylakoid tubules or tubule proteins, or did the thylakoid association reported in 2006 arise only from co-fractionation of the pyrenoid matrix?
Experiment: Measure Rubisco binding (SPR/ITC) and phase-separation boundaries for phosphomimetic and phospho-null variants at the mapped KEY1 sites in each binding region, and test the variants in an epyc1 background for pyrenoid number and dissolution during division.
Hypothesis: Phosphorylation of the EPYC1 Rubisco-binding regions by KEY1 directly lowers Rubisco affinity and so controls condensate size.
Type: biophysics and complementation
Experiment: Repeat the thylakoid fractionation with pyrenoid-less or matrix-less strains (for example the RBCS alpha-helix swap line) and compare EPYC1 partitioning; complement with cryo-ET immunolabelling relative to tubule membranes.
Hypothesis: The apparent thylakoid membrane association of LCI5 reported in 2006 reflects pyrenoid matrix co-sedimentation with thylakoid tubules, not a membrane location.
Type: cell fractionation and imaging
Experiment: Replace endogenous EPYC1 with variants carrying 3 to 6 binding regions or altered linker lengths, then measure FRAP, pyrenoid morphology, Ci affinity and growth at low CO2.
Hypothesis: The number and spacing of Rubisco-binding regions sets the material properties of the matrix and the efficiency of the CCM in vivo.
Type: genetic complementation and physiology
What is not known β curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: The kinase acting on EPYC1 is now known (KEY1), but no phosphatase or 14-3-3-dependent mechanism that resets EPYC1 phosphorylation to re-condense the pyrenoid after division has been identified.
OPEN BIOLOGY RESIDUAL_SUBGAP
What is known: EPYC1 is phosphorylated in its Rubisco-binding regions mainly by KEY1, and phosphorylation disrupts Rubisco binding and promotes condensate dissolution.
Provenance (the field's own admissions):
Gap: GO has no process term specific to pyrenoid assembly, so the role of EPYC1 can only be captured as membraneless organelle assembly.
OPEN ONTOLOGY RESIDUAL_SUBGAP
What is known: EPYC1 with Rubisco is necessary and sufficient to form the liquid matrix of the pyrenoid.
What would resolve it: Request a new GO BP term "pyrenoid assembly" (is_a GO:0140694 membraneless organelle assembly).
Provenance (the field's own admissions):
Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)π View Pathway Visualization Interactive pathway diagram with detailed annotations