THUMPD3

THUMPD3 is the catalytic subunit of a tRNA N2-methylguanosine (m2G) methyltransferase that controls global protein synthesis and cell growth (PMID:34669960). It heterodimerizes with TRMT112, which is required to activate its otherwise inactive catalytic function; the complex recognizes the 3'-CCA end of mature cytoplasmic tRNAs and installs m2G at position 6 of all tested G6-containing tRNAs, as well as m2G7 of tRNATrp (PMID:34669960). Loss of THUMPD3 in human cells impairs global protein synthesis and reduces cell growth, establishing m2G6/7 as a translation-supporting modification (PMID:34669960). In cancer, the THUMPD3-TRMT112 complex promotes pancreatic tumor cell growth and autophagy by sustaining m2G modification of tRNALeu(CAG), which is required for efficient TFEB translation (PMID:41530782). THUMPD3 additionally maintains a pro-tumor EDB-containing isoform of Fibronectin-1 and influences alternative splicing of extracellular matrix and cell adhesion transcripts, and its depletion impairs cancer cell proliferation and migration (PMID:39656728). Beyond its tRNA m2G writing activity, transcriptome-wide mapping indicates THUMPD3-dependent m2G sites also exist in mRNAs and other RNA classes [PMID:bio_10.1101_2024.08.15.608094], but the mechanism linking THUMPD3 to splicing has not been resolved in the available corpus.

1. 2021 High

   Established THUMPD3 as the catalytic m2G writer for human cytoplasmic tRNAs and defined how its activity is switched on and targeted, answering what enzyme installs tRNA m2G6/7 and how substrate specificity is achieved.

   Evidence Co-immunoprecipitation, in vitro enzymatic reconstitution with substrate mapping, RNA-mass spectrometry, and THUMPD3-knockout cells

   PMID:34669960

   Open questions at the time

   • Structural basis for TRMT112-dependent activation not resolved
   • Mechanism of 3'-CCA end recognition not defined at atomic level
   • Whether m2G7 of tRNATrp has a distinct functional role unaddressed
2. 2021 High

   Linked loss of THUMPD3-catalyzed tRNA m2G to a cellular phenotype, showing the modification supports global translation and proliferation rather than being dispensable.

   Evidence THUMPD3-knockout cell lines with protein synthesis and cell growth assays

   PMID:34669960

   Open questions at the time

   • Which specific transcripts are most translationally sensitive not identified at this stage
   • Whether the growth defect is fully attributable to translation impairment unresolved
3. 2024 Medium

   Connected THUMPD3 to cancer-relevant alternative splicing, revealing a function beyond bulk translation by showing it maintains a pro-tumor Fibronectin-1 isoform and shapes ECM/adhesion transcript splicing.

   Evidence siRNA/shRNA depletion with transcriptome-wide RNA-seq, exogenous overexpression in fibroblasts, proliferation and migration assays in lung cancer cells

   PMID:39656728

   Open questions at the time

   • Mechanism coupling THUMPD3 to splicing (tRNA-dependent vs independent) not established
   • Direct vs indirect control of FN1 EDB exon inclusion not distinguished
   • No reciprocal validation that splicing changes drive the proliferation/migration phenotype
4. 2024 Medium

   Extended the THUMPD3 m2G target landscape beyond tRNAs by transcriptome-wide mapping, addressing whether m2G occurs in other RNA classes and whether THUMPD3 is the sole writer.

   Evidence Optimized PhOxi-seq chemical sequencing with a bioinformatic pipeline, enzyme-dependent comparison in THUMPD3-depleted cancer cells (preprint)

   PMID:bio_10.1101_2024.08.15.608094

   Open questions at the time

   • Preprint, single lab, not peer-reviewed
   • Functional consequence of mRNA m2G sites not tested
   • Identity of the non-THUMPD3 m2G writers not determined
5. 2026 Medium

   Provided a molecular route from THUMPD3 tRNA modification to a cancer phenotype, showing m2G of tRNALeu(CAG) is required for TFEB translation and thus for autophagy and tumor growth.

   Evidence shRNA knockdown, translation reporter assays, autophagic flux assays, tRNA modification analysis, and in vitro plus in vivo xenograft pancreatic cancer models

   PMID:41530782

   Open questions at the time

   • Whether TFEB is the principal effector or one of several codon-sensitive targets unclear
   • Codon-specific decoding mechanism by which m2G affects tRNALeu(CAG) not dissected
   • Single lab; generality across cancer types not established

• The mechanism by which THUMPD3 influences alternative splicing, and whether this is separable from its tRNA m2G activity, remains unresolved.
• No direct RNA-binding or spliceosome-interaction data linking THUMPD3 to splicing
• Functional role of THUMPD3-dependent mRNA m2G sites untested
• Structural model of the THUMPD3-TRMT112 complex absent