RDX

RDX (radixin) is an actin-cytoskeleton-linking ERM-family protein whose integrity is essential for sensory hair cell function: loss-of-function mutations disrupting either its FERM domain or its C-terminal actin-binding motif segregate with autosomal recessive nonsyndromic sensorineural hearing loss (DFNB24), and radixin localizes along the length of hair cell stereocilia in the organ of Corti and vestibular system, indicating a direct structural role in stereocilia maintenance (PMID:17226784, PMID:31250571). Studies of the Drosophila ortholog (Roadkill/HIB) reveal an independent function as a CUL3 E3 ubiquitin ligase substrate-specifying adaptor: in complex with CAL1 it ubiquitylates the centromeric histone CENP-A in a manner that stabilizes rather than degrades it, so that RDX loss triggers CENP-A/CAL1 degradation and chromosome segregation defects (PMID:24636256), and within Hedgehog signaling it acts in a Su(fu)-dependent dual mode, either blocking nuclear entry of Ci-155 or promoting its nuclear degradation to tune pathway output (PMID:23216924). RDX has also been placed in a parvovirus-induced PKCη/Rdx kinase complex that phosphorylates PDK1 at Ser138 to activate it independently of PI3-kinase (PMID:25742010), and its protein abundance is post-transcriptionally controlled by direct miRNA targeting (miR-196b-5p in ovarian granulosa cells, miR-409-3p in tongue squamous carcinoma), where reducing RDX alters proliferation, apoptosis, migration, and invasion (PMID:35071469, PMID:29928443).

1. 2007 High

   Established that RDX is a human deafness gene and tied its function to stereocilia, answering whether radixin has a non-redundant physiological role in the inner ear.

   Evidence Sanger sequencing of consanguineous families identifying FERM-domain and actin-binding-motif mutations, with confocal localization in mouse inner ear

   PMID:17226784

   Open questions at the time

   • Mechanism by which radixin loss leads to stereocilia degeneration not resolved at the molecular level
   • No direct biochemical assay of the disease mutations' effect on membrane-actin cross-linking
2. 2012 Medium

   Resolved how the Drosophila ortholog can produce opposite effects on Hedgehog signaling, showing Su(fu) levels switch Rdx between blocking Ci-155 nuclear entry and promoting its nuclear degradation.

   Evidence Genetic epistasis and Ci-155 immunofluorescence localization in Drosophila wing discs with rdx/su(fu) loss- and gain-of-function

   PMID:23216924

   Open questions at the time

   • Conducted in Drosophila; relevance to mammalian RDX not addressed
   • Biochemical basis of the Su(fu)-dependent switch not defined
3. 2014 High

   Defined a CUL3-adaptor role for the ortholog in which ubiquitylation stabilizes rather than degrades its target, explaining how RDX maintains centromeric CENP-A.

   Evidence Reciprocal Co-IP, in vivo ubiquitylation assays, and RNAi/genetic loss-of-function with chromosome-segregation phenotyping in Drosophila

   PMID:24636256

   Open questions at the time

   • Demonstrated in Drosophila; whether human RDX functions as a CUL3 adaptor is untested
   • Molecular basis for non-degradative ubiquitylation not defined
4. 2015 Medium

   Implicated an Rdx-containing kinase complex in direct activation of PDK1, identifying a PI3K-independent route to PDK1 activity.

   Evidence In vitro kinase assay with phospho-specific detection plus cell-based rescue using a constitutively active PDK1 phosphomimetic

   PMID:25742010

   Open questions at the time

   • Single lab; reciprocal validation and structural basis lacking
   • Direct kinase role attributed to PKCη/Rdx complex rather than RDX alone
5. 2018 Medium

   Showed RDX abundance is directly controlled by miR-409-3p and that RDX acts downstream to drive tumor cell proliferation, migration, and invasion in tongue squamous carcinoma.

   Evidence Dual-luciferase target validation, si-RDX rescue/epistasis, migration/invasion assays, and nude-mouse xenograft

   PMID:29928443

   Open questions at the time

   • Molecular pathway linking RDX to invasive behavior not defined
   • Single tumor type and cell line
6. 2019 Medium

   Extended the DFNB24 mutational spectrum with new frameshift/nonsense variants, confirming the FERM-domain N-terminus is required for auditory function.

   Evidence NGS deafness-gene panel with Sanger segregation validation in a Chinese family

   PMID:31250571

   Open questions at the time

   • Pathogenicity inferred from domain disruption rather than functional assay
   • No direct measurement of protein loss in patient tissue
7. 2021 Medium

   Identified miR-196b-5p as a direct regulator of RDX in ovarian granulosa cells, linking RDX levels to proliferation, apoptosis, ROS, and steroidogenesis.

   Evidence Dual-luciferase target validation with RT-qPCR/Western and proliferation, apoptosis, and ROS readouts

   PMID:35071469

   Open questions at the time

   • Mechanism connecting RDX to estrogen synthesis and glucose uptake not established
   • Single cell system

• Whether the CUL3-adaptor, Hedgehog-modulating, and PDK1-activating activities defined in Drosophila or in kinase-complex contexts are exercised by mammalian RDX, and how these reconcile with its stereocilia structural role, remains unresolved.
• No direct test of CUL3-adaptor or Hedgehog roles for human RDX
• No unifying biochemical link between the cytoskeletal and ubiquitylation/signaling functions