What is ALG3?
ALG3 is the name of a gene that encodes an enzyme called Dolichyl-P-Man:Man(5)GlcNAc(2)-PP-dolichyl mannosyltransferase, or alpha-1,3- mannosyltransferase for short. ALG3 is a mannosyltransferase that adds a mannose building block to the middle branch of a maturing glycan. Specifically, it transfers mannose from dolichyl-phosphate mannose to the lipid-linked oligosaccharide (LLO) intermediate Man(5)GlcNAc(2)-PP-dolichol.
ALG3 protein resides in the endoplasmic reticulum (ER) with its business end located in the ER lumen. ALG3 is situated smack dab in the middle of the N-linked glycosylation biosynthetic pathway. Downstream of ALG3 is an assembly line of ER-lumen-facing mannosyltransferases: ALG9, ALG12, ALG5, ALG6, ALG8 and ALG10.
Orthologs of ALG3 are conserved in mice, zebrafish, flies, worms, budding yeast and fission yeast.
What does ALG3 do?
The gene prefix ALG stands for ALtered in Glycosylation. ALG genes were first identified as mutants in budding yeast in the 1980s and 1990s. Like other ALG genes that cause a CDG, ALG3 was initially characterized in budding yeast (Huffaker & Robbins, 1983). In fact, a rich yeast literature thrived in the 1980s and 1990s but then yeast models of CDGs began to fade from view after the year 2000.
Studies in the fruit fly showed that while loss of ALG3 causes system defects in hypoglycosylation, at least one specific glycoprotein appears to be uniquely under-glycosylated by the loss of ALG3, leading to aberrant signaling by the fly version of a gene encoding tumor necrosis factor (de Vreede et al., 2018).
Clinical presentation of ALG3-CDG
ALG3-CDG is formerly known as CDG-1d. ALG3-CDG has a primarily neurologic presentation. Unfortunately, most of the clinical reports are behind a paywall.
Korner et al, 1999 described the first ALG3-CDG patient. The expected accumulation of the glycan intermediate Man(5)GlcNAc(2)-PP-dolichol in ALG3-CDG patient fibroblasts was observed. They also showed that expression of the human ALG3 gene in a yeast mutant lacking yeast ALG3 rescued the glycosylation defect of CPY, a protein secreted by yeast that serves as a biomarker analog of serum transferrin.
In 2008, a clinical report described the seventh ALG3-CDG patient (Rimella-Le-Huu et al., 2008). They noted that: “All patients with CDG Id display a slowly progressive encephalopathy with microcephaly, severe psychomotor retardation and epileptic seizures. They also share some typical dysmorphic features but they do not present the multisystem involvement observed in other CDG syndromes or any biological marker abnormalities.”
The most recent clinical report (Himmelreich et al., 2019) presents data on four new ALG3-CDG patients, finding that “Additional clinical symptoms observed in our patients comprise sensorineural hearing loss, right-descending aorta, obstructive cardiomyopathy, macroglossia, and muscular hypertonia.”
Broader implications beyond ALG3-CDG
Like we’ve seen for other CDG genes, ALG3 can be viewed in the context of improving recombinant/synthetic protein production. Using the yeast Yarrowia lipolytica, which has naturally suited to protein production and secretion, De Pourcq et al., 2012 showed that knocking out ALG3 and concomitantly overexpressing ALG6 leads to production of proteins with humanized glycans.
Next steps toward the clinic
The justification for creating ALG3-CDG yeast models is strong. The first ALG3-CDG worm model is begging to be made. As mentioned above, although a ALG3-CDG fly model hasn’t been created yet the fly version of the ALG3 gene has been characterized. There is no ALG3-CDG zebrafish disease model and that should rectified as well. It is likely that an ALG3 knockout mouse is inviable so strategies to create conditional or tissue-restricted knockout is advisable.
Any and all of these animal models can be paired with studies of ALG3-CDG patient fibroblasts and one day ALG3-CDG iPSC models.
A targeted approach of supplying a hydrophobic derivative of mannose 1-phosphate, a limiting substrate for all N-linked glycosylation pathways, has been proposed by Eklund et al., 2005. The biotech company Glycomine is running a trial for a hydrophobic mannose 1-phosphate substrate in the treatment of PMM2-CDG.
De Pourcq K, Tiels P, Van Hecke A, Geysens S, Vervecken W, Callewaert N. (2012). Engineering Yarrowia lipolytica to produce glycoproteins homogeneously modified with the universal Man3GlcNAc2 N-glycan core. PLoS One. 7: e39976.
de Vreede G, Morrison HA, Houser AM, Boileau RM, Andersen D, Colombani J, Bilder D. (2018). A Drosophila Tumor Suppressor Gene Prevents Tonic TNF Signaling through Receptor N-Glycosylation. Developmental Cell. 45: 595-605.
Eklund EA, Merbouh N, Ichikawa M, Nishikawa A, Clima JM, Dorman JA, Norberg T, Freeze HH. (2005). Hydrophobic Man-1-P derivatives correct abnormal glycosylation in Type I congenital disorder of glycosylation fibroblasts. Glycobiology. 15: 1084-1093.
Himmelreich N, Dimitrov B, Geiger V, Zielonka M, Hutter AM, Beedgen L, Hüllen A, Breuer M, Peters V, Thiemann KC, Hoffmann GF, Sinning I, Dupré T, Vuillaumier-Barrot S, Barrey C, Denecke J, Kölfen W, Düker G, Ganschow R, Lentze MJ, Moore S, Seta N, Ziegler A, Thiel C. (2019). Novel variants and clinical symptoms in four new ALG3-CDG patients, review of the literature, and identification of AAGRP-ALG3 as a novel ALG3 variant with alanine and glycine-rich N-terminus. Human Mutations. May 8.
Huffaker TC & Robbins PW. (1983). Yeast mutants deficient in protein glycosylation. Proc Natl Acad Sci U S A. 80: 7466-70.
Körner C, Knauer R, Stephani U, Marquardt T, Lehle L, von Figura K. (1999). Carbohydrate deficient glycoprotein syndrome type IV: deficiency of dolichyl-P-Man:Man(5)GlcNAc(2)-PP-dolichyl mannosyltransferase. EMBO Journal. 18: 6816-6822.
Rimella-Le-Huu A, Henry H, Kern I, Hanquinet S, Roulet-Perez E, Newman CJ, Superti-Furga A, Bonafé L, Ballhausen D. (2008).Congenital disorder of glycosylation type Id (CDG Id): phenotypic, biochemical and molecular characterization of a new patient. Journal of Inherited Metabolic Diseases. 31 Suppl 2: S381-6.