Proteins / PRAMEF4

DarkO60810 · Homo sapiens (Human) · 478 aa

PRAMEF4

PRAME family member 4

AI prediction, not experimentally verified. The function, GO terms and reasoning are generated by a model (BioReason-Pro). Treat them as a hypothesis to test, not as a fact.

Predicted 3D structure

AlphaFold model.

In plain language

PRAMEF4 has barely been studied. The model's best guess: A soluble adaptor that uses an N-terminal PRAME-like module and a C-terminal leucine-rich repeat scaffold to recruit specific substrates to a Cul2-RING ubiquitin ligase. By assembling and positioning targets for efficient polyubiquitylation, it promotes their proteasome-dependent turnover and thereby contributes to pathways that govern cell-cycle control and cellular differentiation.

Predicted function

A soluble adaptor that uses an N-terminal PRAME-like module and a C-terminal leucine-rich repeat scaffold to recruit specific substrates to a Cul2-RING ubiquitin ligase.

Confidence: the model gives no confidence score, and this protein has no known function to compare with. Treat it as a hypothesis.

Sequence · 478 aa · coloured by model confidence
1MKMSIWTPPR11LLELAGRSLL21RDQALAMSTL31EELPTELFPP41LFMEAFSRRR51CEALKLMVQS61WPFRRLPLRP71LIKMPCLEAF81QAVLDGLDAL91LNLGVRPRRW101KLQVLDLQDV111CENFWMVWSE121AMAHGCFLNA131KRNKKPVEDC141PRMKGRQPLT151VFVELWLKNR161TLDEYLTCLL171LWVKQRKDLL181HLCCKKLKIL191GMPFRNIRSI201LKMVNLDCIQ211EVEVNCKWVL221PILTQFTPYL231GHMRNLQKLI241LSHMDVSRYV251SPEQKKEIVT261QFTTQFLKLR271CLQKLYMNSV281SFLEGHLDQL291LSCLKTSLKF301LTITNCVLLE311SDLKHLSQCP321SISQLKTLDL331SGIRLTNYSL341VPLQILLEKV351AATLEYLDLD361DCGIIDSQVN371AILPALSRCF381ELNTFSFCGN391PICMATLENL401LSHTIILKNL411CVELYPAPRE421SYGADGTLCW431SRFAQIRAEL441MNRVRDLRHP451KRILFCTDYC461PDCGNRSFYD471LEADQYCC

Protein domains

Known building blocks found by InterPro, placed along the 478 amino acids of the chain. Hover or tap a block for details.

150100150200250300350400450478

Predicted function

Model summary

A soluble adaptor that uses an N-terminal PRAME-like module and a C-terminal leucine-rich repeat scaffold to recruit specific substrates to a Cul2-RING ubiquitin ligase. By assembling and positioning targets for efficient polyubiquitylation, it promotes their proteasome-dependent turnover and thereby contributes to pathways that govern cell-cycle control and cellular differentiation. Its nucleocytoplasmic scaffold architecture enables high-specificity substrate capture and handoff within cullin-based ubiquitination assemblies.

Predicted Gene Ontology terms

Molecular function

What it does chemically

  • GO:1990756 ubiquitin ligase-substrate adaptor activity

Biological process

Which larger process it takes part in

  • GO:0043161 proteasome-mediated ubiquitin-dependent protein catabolic process

Cellular component

Where in the cell it sits

Model reasoning

How the model got to its answer

  1. 01

    I begin with the InterPro architecture.

    The N-terminus carries IPR060378 (PRAME-like, N-terminal domain, residues 8–115), followed by an extended leucine-rich repeat scaffold defined by IPR032675 (Leucine-rich repeat domain superfamily, residues 155–413) and IPR060377 (PRAME-like, C-terminal LRR domain, residues 149–478). Two family-level envelopes, IPR026271 (PRAME family, residues 3–476) and IPR050694 (LRRC14/PRAME family, residues 3–464), span essentially the full length, indicating a conserved PRAME/LRRC14 adaptor framework. The ordered layout—an N-terminal PRAME-like module feeding into a long LRR array—creates a bipartite recognition platform: the N-terminal PRAME-like domain often mediates initial docking to partner proteins, while the LRR solenoid provides a curved surface for high-specificity substrate capture and assembly of larger complexes.

  2. 02

    This architecture causes adaptor activity rather than catalysis.

    LRR arrays are classical specificity modules in ubiquitin ligase pathways, and the PRAME/LRRC14 family is known for assembling substrate-adaptor functions. Such a scaffold naturally confers GO:1990756 ubiquitin ligase-substrate adaptor activity: the N-terminal PRAME-like domain nucleates association with cullin-RING machinery, and the C-terminal LRR domain presents a multivalent interface that selects targets for ubiquitination.

  3. 03

    From this molecular function, the biological role follows.

    By recruiting substrates to cullin-RING assemblies, the protein drives GO:0043161 proteasome-mediated ubiquitin-dependent protein catabolic process. The LRR-guided capture of targets and their handoff to E3 complexes channels them into polyubiquitylation and subsequent proteasomal turnover, a pathway central to cell-cycle control and differentiation.

  4. 04

    The cellular context is dictated by the complex it assembles with.

    The presence of a PRAME/LRRC14 scaffold and the inferred role as a substrate adaptor point to operation within a cullin-based E3 environment, specifically the GO:0031462 Cul2-RING ubiquitin ligase complex. Such complexes function in both cytoplasm and nucleus; the absence of transmembrane features and the soluble LRR scaffold support a nucleocytoplasmic adaptor that shuttles between assembly sites of CRL2.

  5. 05

    Mechanistically, I hypothesize that the N-terminal PRAME-like domain engages core CRL2 components, while the LRR array recognizes specific substrate cohorts whose abundance must be tightly tuned during differentiation and cell-cycle transitions.

    Likely partners include the Cul2 scaffold and RING catalytic subunits, with auxiliary adaptors stabilizing the assembly. This arrangement positions the protein as a selective bridge that captures substrates and delivers them to the Cul2-RING catalytic center for efficient polyubiquitylation and proteasomal routing.

Technical details

InterPro entries

  • IPR032675 Leucine-rich repeat domain superfamily homologous superfamily · 155–413
  • IPR060378 PRAME-like, N-terminal domain domain · 8–115
  • IPR060377 PRAME-like, C-terminal LRR domain domain · 149–478
  • IPR026271 PRAME family · 3–476
  • IPR050694 LRRC14/PRAME family · 3–464

Sources

Method

BioReason-Pro with the ESM3 protein language model (esm3-sm-open-v1). How the atlas is made →

API

/api/v1/proteins/O60810 · API docs