What is PMM2?

PMM2 is the name of a gene that encodes an enzyme called phosphomannomutase 2. PMM2 converts mannose 6-phosphate (M6P) to mannose 1-phosphate in the one of the first steps in N-linked protein glycosylation. PMM2 has a close sibling gene called PMM1, which stands for phosphomannomutase 1. However, PMM2 and PMM1 and not functionally interchangeable – otherwise a congenital disorder of glycosylation wouldn’t arise when PMM2 function is lost because PMM1 would step up to fill the breach.

PMM2 is the next enzyme in the queue after MPI, or mannose 6-phosphate isomerase, which feeds mannose 6-phosphate to PMM2. PMM2 is a cytoplasmic enzyme meaning it’s not bound to any membrane though it likely doesn’t stray too far from the endoplasmic reticulum (ER) membrane. PMM2 forms a homodimer, meaning two PMM2 proteins snuggle up in an enzymatic embrace.

Orthologs of PMM2 are conserved in mice, zebrafish, flies, worms, budding yeast and fission yeast.

What does PMM2 do?

PMM2 is essential for life in all animals, including budding yeast cells. Starting with the simplest species first, let’s summarize the published literature on animal and cellular models of PMM2-CDG. The budding yeast version of PMM2 is called SEC53. The SEC prefix is short for secretory, and almost all SEC mutants like SEC53 have defects in protein secretion and as a result are inviable. SEC53 was first isolated a secretory-defective mutant by Ferro-Novick et al., 1984, and SEC53 protein was confirmed to have phosphomannomutase enzymatic activity by Kepes & Schekman, 1988.

The first worm PMM2-CDG model was recently described in a biorXiv preprint by Iyer et al. The F119L mutation was knocked into worms and the resulting PMM2-CDG mutant worm has diminished phosphomannomutase enzymatic activity using an assay originally developed for human cells. The first fly model of PMM2-CDG was described by Parkinson et al., 2016. Knocking out PMM2 in flies results in lethality and glycosylation defects, as expected. 

Two groups have modeled PMM2-CDG in zebrafish. Cline et al, 2012 described a PMM2 morpholino knockdown mutant as opposed to a PMM2 genetic knockout mutant. Mukaigasa et al., 2018 described a zebrafish line that was discovered in a genetic screen for mutants that constantly express a stress response master regulator called NRF2. Turns out this zebrafish line has a novel loss-of-function mutation in PMM2. The interpretation is that the expression of NRF2 is turned up in response to the ER stress caused by defective PMM2 and the ensuing global hypoglycosylation of proteins.

Unfortunately there are no viable PMM2-CDG mouse models because they all suffer from early lethality. The first PMM2 knockout mouse was attempted by Thiel et al., 2006. This was followed by a PMM2-CDG knock-in mouse which was genetically engineered to express the two most common PMM2-CDG disease-causing mutations, R141H and F119L (Chan et al., 2016). However, the PMM2-CDG R141H F119L mouse still has early lethality – just later than a PMM2 null mutant.

PMM2-CDG patient fibroblasts, immortalized mammalian cell lines and even induced pluripotent stem cells/iPSCs (Thiesler et al., 2016) have also been used over the years to study PMM2 function and the consequences of loss of PMM2 on the glycosylation status of proteins. PMM2-CDG patient fibroblasts appear to have an Achilles heel in that the levels of detectable phosphomannomutase enzymatic activity are higher than what one would expect – 35%-70% of normal PMM2 enzymatic activity (Grünewald et al., 2001). On the surface, PMM2-CDG disease models are not limiting, especially in yeast where PMM2 pathogenic variants can be studied in parallel (Lao et al., 2019).

Clinical presentation of PMM2-CDG

PMM2-CDG, formerly known as CDG-1a, is by far the most common type of CDG. PMM2-CDG kids have variable multi-system clinical presentation leading to failure to thrive, hypotonia, global developmental delay, ataxia, dysmorphia, skeletal abnormalities, and coagulation defects. There’s up to 20% mortality in the first 5 years due to organ failure and/or severe infections. As PMM2-CDG children age and become adults they show intellectual disability, stroke-like episodes, and retinitis pigmentosa. PMM2-CDG adults are usually wheelchair bound with peripheral neuropathy and stable intellectual disability.

The first published clinical report of PMM2-CDG is Matthijs et al., 1997. Very quickly it was appreciated that there are no living PMM2-CDG patients homozygous for a completely nonfunctional mutation, meaning at least some residual PMM2 enzymatic activity is required for life (Matthijs et al., 1998; Kjaergaard et al., 1998).

There are too many n-of-1 or n-of-few PMM2-CDG clinical reports (many of them behind a subscription journal paywall) to list here so I’ll cite a few recent comprehensive reviews. Chang et al., 2018 describe the over 130 CDGs that have been diagnosed today. Despite the large number of CDG types, PMM2-CDG represents the vast majority of patients. There are an estimated 800 to 1,200 PMM2-CDG patients diagnosed and alive in the world today.

The large number of PMM2-CDG patients means a universe of many disease-causing mutations. R141H is the most common mutation seen in PMM2-CDG patients around the world, with a frequency between 20-30%. R141H is a so called catalytic dead mutation. Mutant PMM2 R141H protein folds correctly, is stable, forms dimers but has zero enzymatic activity (Yuste-Checa et al., 2015). This is why PMM2-CDG patients homozygous for R141H are never observed alive. The next most common mutation is F119L (Andreotti et al., 2013). Phenylalanine at position 119 is located in the PMM2 dimer interface. Mutation to leucine at that position destabilizes dimers or prevents them for forming stably in the first place. The most common PMM2-CDG genotype is R141H F119L. 

Then there’s a long tail of missense mutations that are either private to a family or ethnic group. PMM2 does not usually tolerate premature stop, small insertions or deletions or truncating mutations. Andreotti et al., 2015 described a remarkable finding. Even though PMM2 functions as dimer, there only needs to be one active site per PMM2 pairing. Furthermore, a dimerization-defective mutant PMM2 like F119L is unable to dimerize with itself but dimerizes just fine with a catalytic-dead mutant PMM2 like R141H. In other words, if a mutant R141H PMM2 monomer pairs up with a mutant F119L PMM2 monomer, the dimer is just as active as a R141H PMM2 monomer pairing up with a normal PMM2 monomer.  

Why are there so many PMM2 mutations? Citro et al., 2018 raise the fascinating possibility that being a carrier of PMM2-CDG mutations might have an evolutionary fitness advantage, in much the same way the being a carrier of hemophilia offers protect from malarial infection. What metabolic or other advantages does a 50% reduction in PMM2 enzymatic activity confer? That’s an excellent question for future research.

Next steps toward the clinic and beyond

CDG researcher Dr. Hudson Freeze has been in the field for decades. He recommended three principal therapeutic avenues: mannose 1-phosphate supplementation (substrate replacement), small molecule PMM2 enzyme activators or chaperones, and PMM2 enzyme replacement (Freeze, 2009). 

Some of these approaches are being put to the test in high-throughput drug screens and even in the clinic. The biotech company Glycomine is pursuing a mannose-1-phosphate replacement strategy and is currently in Phase 1. Proof of concept for a PMM2 pharmacological chaperone screen was provided by Yuste-Checa et al., 2017. Lao et al., 2019 and Iyer et al have proven the value of yeast and worm PMM2-CDG models for drug repurposing and drug discovery.

The PMM2-CDG is fortunate to have many disease models at its disposal and multiple reasonable therapeutic theses. So now it’s time to screen screen screen.


Andreotti G, Monti MC, Citro V, Cubellis MV. (2015). Heterodimerization of Two Pathological Mutants Enhances the Activity of Human Phosphomannomutase2. PLoS One. 10: e0139882.

Andreotti G, Pedone E, Giordano A, Cubellis MV. (2013). Biochemical phenotype of a common disease-causing mutation and a possible therapeutic approach for the phosphomannomutase 2-associated disorder of glycosylation. Mol Genet Genomic Med. 1: 32-44.

Chan B, Clasquin M, Smolen GA, Histen G, Powe J, Chen Y, Lin Z, Lu C, Liu Y, Cang Y, Yan Z, Xia Y, Thompson R, Singleton C, Dorsch M, Silverman L, Su SM, Freeze HH, Jin S. (2016). A mouse model of a human congenital disorder of glycosylation caused by loss of PMM2. Human Molecular Genetics. 25: 2182-2193.

Chang IJ, He M, Lam CT. (2018). Congenital disorders of glycosylation. Annals of Translational Medicine. 6: 477.

Citro V, Cimmaruta C, Monticelli M, Riccio G, Hay Mele B, Cubellis MV, Andreotti G. (2018). The Analysis of Variants in the General Population Reveals That PMM2 Is Extremely Tolerant to Missense Mutations and That Diagnosis of PMM2-CDG Can Benefit from the Identification of Modifiers. International Journal of Molecular Science. 19.

Cline A, Gao N, Flanagan-Steet H, Sharma V, Rosa S, Sonon R, Azadi P, Sadler KC, Freeze HH, Lehrman MA, Steet R. (2012). A zebrafish model of PMM2-CDG reveals altered neurogenesis and a substrate-accumulation mechanism for N-linked glycosylation deficiency. Molecular Biology of the Cell. 23: 4175-4187.

Ferro-Novick S, Hansen W, Schauer I, Schekman R. (1984). Genes required for completion of import of proteins into the endoplasmic reticulum in yeast. Journal of Cell Biology. 98: 44-53.

Freeze HH. (2009). Towards a therapy for phosphomannomutase 2 deficiency, the defect in CDG-Ia patients. Biochimica Biophysica Acta. 1792: 835-840.

Grünewald S, Schollen E, Van Schaftingen E, Jaeken J, Matthijs G. (2001). High residual activity of PMM2 in patients’ fibroblasts: possible pitfall in the diagnosis of CDG-Ia (phosphomannomutase deficiency). American Journal of Human Genetics. 68: 347-354.

Iyer S, Murthy K, Parton Z, Tsang H, Sam FS, DiPrimio N, Lao J, Perlstein EO. Repurposing the aldose reductase inhibitor and diabetic neuropathy drug epalrestat for the congenital disorder of glycosylation PMM2-CDG. biorXiv. 

Kepes F & Schekman R. (1988). The yeast SEC53 gene encodes phosphomannomutase. Journal of Biological Chemistry. 263: 9155-9161.

Kjaergaard S, Skovby F, Schwartz M. (1998). Absence of homozygosity for predominant mutations in PMM2 in Danish patients with carbohydrate-deficient glycoprotein syndrome type 1. European Journal of Human Genetics. 6: 331-336.

Lao JP, DiPrimio N, Prangley M, Sam FS, Mast JD, Perlstein EO. (2019). Yeast Models of Phosphomannomutase 2 Deficiency, a Congenital Disorder of Glycosylation. G3. 9: 413-423.

Matthijs G, Schollen E, Pardon E, Veiga-Da-Cunha M, Jaeken J, Cassiman JJ, Van Schaftingen E. (1997). Mutations in PMM2, a phosphomannomutase gene on chromosome 16p13, in carbohydrate-deficient glycoprotein type I syndrome (Jaeken syndrome). Nature Genetics. 16: 88-92.

Matthijs G, Schollen E, Van Schaftingen E, Cassiman JJ, Jaeken J. (1998). Lack of homozygotes for the most frequent disease allele in carbohydrate-deficient glycoprotein syndrome type 1A. American Journal of Human Genetics. 62: 542-550.

Mukaigasa K, Tsujita T, Nguyen VT, Li L, Yagi H, Fuse Y, Nakajima-Takagi Y, Kato K, Yamamoto M, Kobayashi M. (2018). Nrf2 activation attenuates genetic endoplasmic reticulum stress induced by a mutation in the phosphomannomutase 2 gene in zebrafish. Proc Natl Acad Sci U S A. 115: 2758-2763.

Parkinson WM, Dookwah M, Dear ML, Gatto CL, Aoki K, Tiemeyer M, Broadie K. (2016). Synaptic roles for phosphomannomutase type 2 in a new Drosophila congenital disorder of glycosylation disease model. Disease Models & Mechanisms. 9: 513-527.

Thiel C, Lübke T, Matthijs G, von Figura K, Körner C. (2006). Targeted disruption of the mouse phosphomannomutase 2 gene causes early embryonic lethality. Molecular Cell Biology. 26: 5615-5620.

Thiesler CT, Cajic S, Hoffmann D, Thiel C, van Diepen L, Hennig R, Sgodda M, Weiβmann R, Reichl U, Steinemann D, Diekmann U, Huber NM, Oberbeck A, Cantz T, Kuss AW, Körner C, Schambach A, Rapp E, Buettner FF. (2016). Glycomic Characterization of Induced Pluripotent Stem Cells Derived from a Patient Suffering from Phosphomannomutase 2 Congenital Disorder of Glycosylation (PMM2-CDG). Mol Cell Proteomics. 15(4):1435-52.

Yuste-Checa P, Gámez A, Brasil S, Desviat LR, Ugarte M, Pérez-Cerdá C, Pérez B. (2015). The Effects of PMM2-CDG-Causing Mutations on the Folding, Activity, and Stability of the PMM2 Protein. Human Mutations. 36: 851-860.

Yuste-Checa P, Brasil S, Gámez A, Underhaug J, Desviat LR, Ugarte M, Pérez-Cerdá C, Martinez A, Pérez B. (2017). Pharmacological Chaperoning: A Potential Treatment for PMM2-CDG. Human Mutations. 38: 160-168.

Sign up for the CDG Hub mailing list

Would you like to be contacted by (pick all that apply)