The G6PC3 c.210delC variant found in Mexico originated from a founder effect
Among all the mutations causing G6PC3 deficiency, a few are only found in specific ethnic groups, implying founder effects6. For instance, the c.210delC variant has been reported in 13 G6PC3 deficient patients, who are either homozygous or compound heterozygous for this variant 4,32–34. Among them, 12 patients are of Mexican descent, while another patient is included from the North American Severe Chronic Neutropenia International Registry32. These observations led us to interrogate whether a founder effect causes the recurrence of this mutation or if it is a mutational hotspot. To study this, we recruited four patients from central Mexico who are homozygous for this variant and were born to unrelated, non-consanguineous parents (Fig. 1A-B). We collected genomic DNA from these four patients plus six of their heterozygous healthy parents. We performed whole-genome sequencing (WGS) followed by haplotype analysis on the region of chromosome 17 surrounding G6PC3. Since samples were unavailable from 2 of the patients' fathers, these two carrier haplotypes were inferred computationally from the mother and patient haplotypes. Our analysis revealed a shared haplotype segment surrounding the mutation site among all carriers (Fig. 1C). The length of this shared haplotype suggests that the mutation originated 26 generations ago (95% confidence interval, 2.3–50.5) in a common ancestor. This analysis indicates that the G6PC3 c.210delC single-nucleotide deletion is a founder mutation in Mexico.
The G6PC3 c.210delC mutation is of Indigenous American origin
Based on the assumption that each generation interval is between 20 and 25 years, the estimated age of the G6PC3 c.210delC variant is 520–650 years. This allele age raised the question of whether the shared haplotype originated from native American ancestry or was introduced to Mexico by Europeans. To investigate this, we conducted principal component analysis (PCA) with WGS data of 10 carriers, using reference populations from the combined 1000 Genomes Project (1KGP) and Human Genome Diversity Project (HGDP) dataset. Ancestry inference analysis on the whole genome and chromosome 17 showed that the carriers of this mutation cluster closely with the indigenous American populations from HGDP (Fig. 2A-B). Moreover, we used 2,343 deeply sequenced reference samples of individuals of European, African, and Ad Mixed American Ancestry from the 1KGP + HGDP dataset to perform local ancestry estimation across chr17. Our analysis indicates that the chromosomal region containing the G6PC3 c.210delC mutation (chr17: 44,071,175) is of American origin (Fig. 2C). Overall, our analysis shows that the mutation c.210delC originated in the indigenous Mexican population.
The G6PC3 c.210delC variant results in reduced G6PC3 mRNA and complete loss of protein expression
G6PC3 encodes a protein with nine transmembrane domains that localizes to the endoplasmic reticulum35,36. The c.210delC variant observed in our patients is a single-nucleotide deletion in exon 1 of G6PC3 (Fig. 3A). It is predicted to cause a shift in the reading frame after amino acid 70 (in the second transmembrane domain of the protein), thus introducing a premature stop codon 46 amino acids after the mutation site (p.F71Sfs*46) (Fig. 3B). To examine the impact of this variant at mRNA and protein levels, we first transfected human embryonic kidney 293T (HEK293T) cells with plasmids containing C- or N-terminally Histidine-tagged versions of wildtype (WT) and p.F71Sfs*46 G6PC3. RT-qPCR results suggest that the mutation induces a moderate reduction in G6PC3 mRNA expression (Fig. 3C). Western blotting of whole cell lysate with an anti-His-tag antibody yielded a 10kDa band in the cells transfected with N-terminal tagged p.F71Sfs*46 G6PC3, aligning with the anticipated molecular weight of a truncated mutant protein (Fig. 3D). This predicted mutant protein would not retain enzymatic activity as it lacks the active site of G6PC3, which is composed of amino acids R79, H114, and H16736 (Fig. 3B). By analyzing the C-terminal His-tagged mutant G6PC3, we showed that there is no translation re-initiation after the premature termination codon (Fig. 3D). To gain additional insights in a more physiologically relevant context, we utilized Epstein-Barr Virus immortalized B (EBV-B) cells derived from two patients and healthy controls. We observed markedly reduced levels of G6PC3 mRNA, potentially due to nonsense-mediated mRNA decay (Fig. 3E). Furthermore, using membrane protein fractions extracted from EBV-B cells for western blotting and using a polyclonal antibody with epitope spanning the N-terminal region of human G6PC3, we showed that neither the WT nor the mutant size protein was expressed in cells from the patients (Fig. 3F). Altogether, our data demonstrate that the G6PC3 c.210delC (p.F71Sfs*46) variant disrupts mRNA and protein levels, causing a complete loss of expression.
Cells from patients with the G6PC3 c.210delC variant show abolished G6PC3 enzymatic function
G6PC3 is a metabolite-repair enzyme involved in the hydrolysis of 1,5-anhydroglucitol-6-phosphate (1,5-AG6P), which is the phosphorylated form of a food-derived polyol named 1,5-anhydroglucitol (1,5-AG)37. When a functional G6PC3 is absent, 1,5-AG6P can accumulate to a concentration that inhibits hexokinase activity37. Accumulation of 1,5-AG6P thereby impairs glycolysis since hexokinase mediates the rate-limiting first step of the glycolytic pathway38. This mechanism leads to neutropenia since neutrophils rely heavily on glycolysis to fulfill their energetic needs39. To directly assess the metabolic consequence of the absence of G6PC3 caused by the c.210delC variant, we used patient-derived EBV-B cells to determine if they have defective hexokinase activity in mediating glycolysis. EBV-B cells express all four isoforms of hexokinase (I, II, III, and ADP-glucokinase) identified in mammalian cells, thus permitting us to test the functional impact of the G6PC3 deficiency (Fig. S1). EBV-B cells from patients and healthy controls were treated with either 1,5-AG or glycolytic inhibitor 2-Deoxy-D-Glucose (2-DG) for five days. Then, we quantified the glycolytic activities of these cells through measurements of extracellular acidification rate (ECAR). Patient EBV-B cells exhibited similar levels of glycolytic activity to cells derived from healthy controls, whether untreated or treated with 2-DG. However, patient cells showed significantly impaired glycolysis rate and glycolytic capacity following the 1,5-AG treatment, while healthy control cells remained unaffected (Fig. 4A-C). These results illustrate the metabolic disturbance in cells from patients with the G6PC3 c.210delC variant due to the absence of G6PC3, establishing ECAR measurements in EBV-B cells as a means to examine the functional consequences of G6PC3 mutations.
Patients with the G6PC3 c.210delC variant show a clinical profile similar to other G6PC3-deficient patients
As all patients with the G6PC3 c.210delC variant are from the same geographical area, we evaluated whether they exhibit any characteristic in their clinical presentation that may differentiate them from the rest of the reported G6PC3 deficient patients denoting some environmental aspects of the disease. To this end, we collected the clinical information from all published cases of G6PC3 deficiency to compare the frequency of appearance of nine prominent clinical features between patients with and without the G6PC3 c.210delC mutation (n = 14, some published, some unpublished) (Fig. 5). None of these patients showed isolated neutropenia but present with features of syndromic severe congenital neutropenia including extra-hematological abnormalities. Except for hepatosplenomegaly, all other features of G6PC3 deficiency have been observed in these patients. We also noticed that patients with the c.210delC variant display a higher occurrence of thrombocytopenia, endocrine abnormalities, and hearing loss. Although this may imply a specific characteristic of this group of patients, it might also reflect some variability in standard clinical testing. Overall, our analysis shows that patients who are carriers of the G6PC3 c.210delC mutation display all main clinical characteristics described in G6PC3 deficiency, indicating that the mutation may be the main driver of their disease.