A systemic search for novel congenital disorders of Glycosylation type 1 (CDG-I), with a comprehensive evaluation of the clinical phenotypes.

Glycosylation is one of the most abundant protein modifications found in nature. It results from a meticulously orchestrated process involving numerous proteins for the assembly and modification of oligosaccharide chains, and their attachment onto proteins and lipids. The importance of glycosylation is illustrated by a group of diseases called Congenital Disorders of Glycosylation (CDG). To date, almost 100 distinct disorders have been identified encompassing defects in N- and O-linked protein glycosylation, but also in the synthesis of GPI-anchors and glycolipids. Considering the possibility to screen for most deficiencies in protein N-glycosylation by means of isoelectric focusing of serum transferrin, the project focused on this group of disorders.

The genetic heterogeneity of CDG, but also the phenotypic overlap between the different disorders is remarkable. A clinical ‘hit and run’ diagnosis forms thus rather the exception than the rule. Therefore, patients with a biochemically proven glycosylation deficiency remain often without a molecular diagnosis. In this study, we aimed to circumvent the bottleneck of a gene by gene approach through the implementation of massive parallel sequencing techniques in as well CDG research as diagnostics (Chapter 3).

For the elucidation of novel CDG, whole exome sequencing was performed in 24 individuals with a presumed deficiency in the N-linked glycosylation pathway (Chapter 4 and 5). Once the genetic defect was identified, its pathogenic nature was confirmed using cell biological assays. In this way, a genetic diagnosis could be obtained in nine patients (i.e. 38%), while the most likely candidate gene is still under investigation in seven additional cases (i.e. 30%).

In parallel, a targeted assay for a panel of 79 genes was developed to improve CDG diagnostics (Chapter 6). Over a period of two years, the panel was used for molecular testing in a total number of 86 patients with a presumed deficiency in the N-linked glycosylation pathway. A final molecular diagnosis could be obtained in 38 of them (i.e. 44%). Based on these results, we proposed a tentative novel flowchart wherein a patient considered to have CDG first enters a diagnostic setting for gene panel testing. A close collaboration between the diagnostic and research department would then allow those patients, in whom the culprit gene could not be identified or in whom the pathogenicity of a variant needs to be verified, to subsequently enter a research setting for further biochemical testing.

During this study, mutations in MAN1B1 were identified to cause a novel CDG-II. The biochemical characteristics of the index case allowed for the rapid identification of 18 additional patients (Chapter 7). All cases displayed a similar phenotype characterized by intellectual disability, delayed motor and speech development, hypotonia, macrocephaly and truncal obesity.

During the time span of this PhD project, the intracellular localization of MAN1B1 became the subject of a still ongoing debate. Indeed, besides its role in N-glycan processing, the α(1,2)-mannosidase has been proposed to act as a key factor in ER quality control by targeting terminally misfolded proteins for proteasomal degradation. Since all mediators of ERAD are assumed to reside within the ER, it only seemed natural that MAN1B1 would execute its function within the same organelle. However, today opinions are changing. While some researchers still believe that MAN1B1 resides within the ER, others are convinced that the enzyme localizes to a presumed ERQC compartments or resides within the Golgi apparatus.

In Chapter 9, we could clearly demonstrate that the endogenous MAN1B1 in primary skin fibroblasts is localized within the Golgi apparatus, thereby confirming the initial –but still controversial– results of Sifers and coworkers. Our findings were further supported by the observation that MAN1B1 deficient fibroblasts display an aberrant Golgi morphology in the absence of an ER stress response (i.e. UPR or unfolded protein response) (Chapter 8 and 9).

While former studies mainly focused on the effect of MAN1B1 deficiency on the fate of misfolded cargo, we investigated –with respect to the phenotype– the effect on secretory proteins that attained their native folding state (Chapter 9). In this way, we could demonstrate that MAN1B1 deficiency does not only enable the intracellular accumulation and partial secretion of nonnative proteins, but in addition impairs the anterograde trafficking of the properly folded cargo.

In Chapter 10 of this manuscript, we assumed that the aforementioned accumulation of (mis-)folded proteins within the Golgi apparatus could overwhelm the capacity of the secretory pathway, thereby generating a primary Golgi stress response. Through several pilot experiments we could show that MAN1B1 deficiency generates a mild to moderate transcriptional response that was not only uniformly present among the different patients, but that also differed from the responses observed in other CDG-II cell lines. However, additional investigations are necessary to further address the extent of a possible Golgi stress response in MAN1B1-CDG and to understand how this transcriptional response might impact patient management.