BROX

BROX is a farnesylated, single Bro1 domain-containing ESCRT-associated protein that links the ESCRT-III machinery to membrane remodeling events including viral budding and nuclear envelope repair (PMID:18190528, PMID:34818527). C-terminal CAAX-mediated farnesylation directs BROX to perinuclear membranes and is required for efficient engagement of CHMP4 proteins in cells, with a farnesylation-defective C408S mutant becoming diffusely distributed and failing to redistribute CHMP4b (PMID:18190528). Its boomerang-shaped Bro1 domain presents a concave hydrophobic surface that binds both the CHMP4B C-terminal tail, which docks as an amphipathic α-helix, and the CHMP5 C-terminal tail, which docks as a tandem β-hairpin engaging a unique pocket at residue Y348; CHMP5 specifically recruits endogenous BROX to detergent-resistant membranes (PMID:22162750, PMID:21889351, PMID:22484091). BROX lacks the ALIX-specific Phe105 loop and the ESCRT-0 (STAM2)-binding determinant present in HD-PTP, restricting its repertoire of ESCRT contacts (PMID:21889351, PMID:26866605). Functionally, the BROX Bro1 domain binds the HIV-1 Gag nucleocapsid and stimulates virus-like particle production through both CHMP4-dependent and CHMP4-independent routes (PMID:19403673), and BROX is required for the budding step of infectious HCV release (PMID:21264300). In nuclear envelope rupture repair, BROX binds the LINC complex protein Nesprin-2G, promoting its ubiquitination and relaxation of compressive actin-based mechanical stress to accelerate ESCRT-dependent resealing (PMID:34818527).

1. 2007 Medium

   Establishing whether BROX is a genuine prenylation substrate addressed how this uncharacterized protein might be targeted to membranes.

   Evidence in vivo prenylation/biochemical verification of selective farnesylation in a large-scale prediction study

   PMID:17411337

   Open questions at the time

   • Reported briefly within a prediction-focused study
   • Did not define the functional consequence of farnesylation
2. 2008 High

   Demonstrating that CAAX-mediated farnesylation controls BROX perinuclear localization and CHMP4 engagement connected a lipid modification to ESCRT recruitment.

   Evidence farnesyltransferase inhibitor treatment, C408S mutagenesis, co-pulldown, and fluorescence microscopy in HEK293/HeLa

   PMID:18190528

   Open questions at the time

   • Did not resolve the structural basis of CHMP4 binding
   • Cellular pathway requiring this localization not defined
3. 2009 Medium

   Showing the BROX Bro1 domain binds HIV-1 Gag nucleocapsid and stimulates VLP production placed BROX functionally alongside ALIX in viral budding while revealing a CHMP4-independent activity.

   Evidence minimal Gag rescue VLP assay with CHMP4-binding-surface mutagenesis

   PMID:19403673

   Open questions at the time

   • Molecular basis of the CHMP4-independent activity not identified
   • Single lab; physiological relevance to authentic infection untested
4. 2011 High

   Crystal structures of the BROX Bro1 domain defined its boomerang fold and two conserved hydrophobic surfaces, and explained why it is a weaker HIV-1 budding factor than ALIX by the absence of the Phe105 loop.

   Evidence X-ray crystallography (two independent structural studies) with structure-guided functional comparison to ALIX in budding assays

   PMID:21889351 PMID:22162750

   Open questions at the time

   • Did not capture BROX bound to a partner
   • Function of Surface 2 at the narrow end undefined
5. 2011 Medium

   Loss-of-function in cells showed BROX is required for infectious HCV release at the production/budding step, extending its ESCRT-associated role to a second enveloped virus.

   Evidence siRNA knockdown with HCV supernatant infectivity, Core quantification, and intracellular RNA replication measurement

   PMID:21264300

   Open questions at the time

   • Direct interaction with HCV components not shown
   • Single method/single lab
6. 2012 High

   Structures of BROX:CHMP4B and BROX:CHMP5 complexes showed the same concave surface binds two ESCRT-III tails in distinct modes and that CHMP5 recruits BROX to detergent-resistant membranes, defining the molecular logic of its ESCRT contacts.

   Evidence crystal structures of two complexes plus cell fractionation and co-immunoprecipitation

   PMID:22484091

   Open questions at the time

   • Functional consequence of dual CHMP4B/CHMP5 binding for membrane scission not established
   • Whether binding is competitive in cells unresolved
7. 2016 Medium

   Comparative structural analysis identified Arg145 as the determinant excluding ESCRT-0/STAM2 binding, explaining BROX's restricted interactome relative to HD-PTP.

   Evidence crystal structure of HD-PTP Bro1:STAM2 with site-directed mutagenesis and binding assays

   PMID:26866605

   Open questions at the time

   • BROX-relevant conclusion inferred from HD-PTP mutagenesis rather than direct BROX mutagenesis
   • Biological consequence of lacking ESCRT-0 engagement untested
8. 2020 Medium

   A clean knockdown established that BROX is dispensable for HSV-1 replication, bounding its viral requirement to specific enveloped viruses.

   Evidence siRNA knockdown (Western-confirmed) with quantitative single-step HSV-1 growth assay

   PMID:31748394

   Open questions at the time

   • Negative result; does not rule out subtle or context-specific roles
   • Single lab
9. 2021 High

   Defining BROX as a Nesprin-2G-binding factor that promotes its ubiquitination and mechanical stress relaxation established a physiological role in accelerating ESCRT-dependent nuclear envelope rupture repair.

   Evidence loss-of-function NERDI repair kinetics, BROX–Nesprin-2G co-immunoprecipitation, ubiquitination assay, and live-cell imaging

   PMID:34818527

   Open questions at the time

   • The ubiquitin ligase acting on Nesprin-2G not identified
   • Whether farnesylation/CHMP binding are required for the NERDI role not dissected

• How BROX coordinates its farnesylation, dual ESCRT-III contacts, and substrate recognition to drive distinct membrane-remodeling outcomes remains unresolved.
• The CHMP4-independent budding mechanism is unidentified
• No structure of full-length farnesylated BROX on membranes
• Endogenous physiological pathways beyond NERDI not mapped