iiinova — where your digital self keeps evolving

Andrew Smith

@aj-smith.bsky.social

Staff Scientist in the Pagliarini Lab at WashU and HHMI. Views my own.

Andrew Smith @aj-smith.bsky.social · 30/09/2026

Huge thanks to Patrick, Sean, Merima, and @pagliarini-lab.bsky.social, and to NIDDK, NIGMS and HHMI. Questions and critiques very welcome! #mitochondria #phosphoproteomics

000

Andrew Smith @aj-smith.bsky.social · 30/09/2026

The result: Increased phosphorylation of cristae forming complexes in PGAM5 knockdown cells leads to denser cristae, more MIB assembly, fewer ATP synthase dimers, faster growth, and greater spare respiratory capacity. Full preprint: www.biorxiv.org/content/10.6...

100

Andrew Smith @aj-smith.bsky.social · 30/09/2026

Mode #3: cristae architecture Loss of PGAM5 increases MICOS and ATP synthase phosphorylation. MIC19 S58 phosphorylation is required to build normal cristae. ATP5I S68 phosphorylation blocks ATP synthase dimerization and prevents cristae tip formation.

100

Andrew Smith @aj-smith.bsky.social · 30/09/2026

Mode #2: import Dynamic sites are 4x enriched in the first 30 residues after the MTS cleavage site. Phosphomimetic BCKDK (S31/S33) imports slowly and never gets processed, so BCKDH is no longer inhibited, and cells burn through branched chain amino acids faster.

100

Andrew Smith @aj-smith.bsky.social · 30/09/2026

Three different approaches to analyzing this data revealed three modes of phosphoregulation. Mode #1: catalysis Adenylate Kinase 2 (AK2) S151 sits right in the ATP-binding pocket. A phosphomimetic at this site blunts nucleotide binding, increasing the Km.

100

Andrew Smith @aj-smith.bsky.social · 30/09/2026

SILAC phosphoproteomics on isolated mitochondria across all 10 knockdowns: 350 mito phosphosites, ~45% of them dynamic. No single knockdown showed more than a slice; the full picture only emerged across the panel. Dynamic sites are enriched in disordered regions.

100

Andrew Smith @aj-smith.bsky.social · 30/09/2026

~90% of mitochondrial proteins have at least one phosphosite. Only ~1% of these sites have a reported function. Is this regulation or noise? We silenced 10 resident mitochondrial phosphatases, one at a time, to find out. New preprint 🧵👇

110