A few weeks ago I was invited by @OphirKlein to give a Zoom talk about @PerlaraPBC’s n-of-1 drug repurposing trials for congenital disorders of glycosylation ( #PMM2-CDG and #NGLY-CDDG) to his @UCSF clinical genetics colleagues.

As previously described ( https://www.g3journal.org/content/9/2/413 ) we created yeast patient avatars expressing #PMM2 disease-causing mutations.

Because yeast can be grown as haploids or diploids, it’s possible to model any PMM2-CDG genotype involving conserved amino acid residues.

The yeast ortholog of the human PMM2 gene is called SEC53. When we express PMM2 in a yeast strain lacking SEC53, cell growth is restored. The severity of a mutation, eg F119L/F126L, is the same whether its expressed as SEC53 or PMM2. The genetics and the biochemistry match up.

Yeast cell growth is a deceptively simple but statistically robust phenotype.

Keeping phenotype constant but varying genotype (dimerization mutant misfolding mutant or mutation in a haploid diploid), we identified several possible repurposable drugs or GRAS compounds.

The 3 most compelling repurposing candidates were:

the phenylpropanoid α-cyano-4-hydroxycinnamic acid, found in cinnamon
the flavonoid 2,2-bisepigallocatechin digallate, found in black tea.
suramin, a century-old drug approved for African sleeping sickness.

As we described in https://dmm.biologists.org/content/12/11/dmm040584 the first-ever worm model of #PMM2-CDG expressing the F119L dimerization-defective variant has reduced phosphomannomutase activity, ~30% of wildtype levels.

F119L pmm-2 homozygote worms are more sensitive to growth inhibition and developmental delay caused by the proteasome inhibitor bortezomib.

Using an unbiased, whole-animal, high-throughput, image-based phenotypic approach, we screened for compounds that rescue worm growth and development.

Out of 20 hits from the worm screen, we see examples of phenylpropanoids (specifically catecholamine mimics) and flavonoids, structural classes that overlap with the yeast PMM2 screen.

Using OI (Organic Intelligence), the common thread connecting the yeast and worm screening results is aldose reductase inhibition.

α-cyano-4-hydroxycinnamic acid and epalrestat, the only aldose reductase inhibitor approved for use in people, share a pharmacophore.

We tested both aldose reductase inhibitors side by side in worms and patient fibroblasts and asked whether the boost PMM2 enzymatic activity.

The answer was: yes!

We then tested epalrestat on three PMM2-CDG patient fibroblast lines.

Epalrestat increased phosphomannomutase activity in all three lines, ranging from 40-400% boosts.

Epalrestat does not appear to increase PMM2 protein abundance, suggesting with a post-translational mechanism of action.

Aldose reductase (AR) is the first enzyme in the polyol pathway, and converts glucose to sorbitol and in the process consumes NADPH.

We believe that epalrestat shunts excess glucose toward generation of glucose-1,6-bisphosphate, an endogenous activator of PMM2.

Working arm in arm with @CarmichaelHolly and her family, we’re now approaching the 4-month mark of epalrestat treatment in our n-of-1 trial.

So far, so good! No serious adverse events and positive signals of efficacy.

A similar story unfolded for #NGLY1 and we’re working with @mattmight @bertrandmight @ngly1org to advance aripiprazole through the clinic.

We published these results ( https://dmm.biologists.org/content/12/11/dmm040576) last year.

#NGLY1 knockout worms and heterozygous flies are hypersensitive to the proteasome inhibitor bortezomib setting the stage for multi-species phenotypic screens similar to what we did for #PMM2-CDG.

Aripiprazole was the only compound that rescues all #NGLY1 models tested: worms, flies and human cells.

Stay tuned for epalrestat and aripiprazole n-of-1 trial updates in the months ahead.

How many other known drugs can be repurposed like this? For starters, there are 146 genes other genes besides #PMM2 and #NGLY1 that cause a CDG when mutated.