Fig 2: Yeast models expressing GINS3 D24G or D24N variants present defects in cell growth and protein stability.(A) Structural alignments of human GINS3 (PDB entry 2Q9Q) and yeast Psf3 (PDB entry 5U8S) proteins. A view down helix 1 (left) and a side view of helix 1 (right) are shown. Human GINS3 is in cyan, with residue D24 in ice blue; yeast Psf3 is in red, with residue D8 in orange. (B) Cell population doubling time, as measured by monitoring OD630 of 8 cultures of each genotype during exponential growth. Results are representative of 3 independent experiments, and statistical analyses were performed using 2-tailed t tests. (C) Cell cycle profiles of asynchronously growing (AS) cultures were assessed by flow cytometry. The experiment was performed twice; a representative replicate is shown. (D) Cells were synchronized in G1 by a factor arrest and were then released into S phase. Samples were fixed every 10 minutes after release and cell cycle progression assessed by flow cytometry. The experiment was performed in duplicate; a representative replicate is shown. (E) Serial 5-fold dilutions of yeast were grown on solid media in the presence or absence of nicotinamide (NAM) at 30°C for 72 hours. The experiment was performed in duplicate; a representative replicate is shown. (F) Exponentially growing yeast cultures were treated with 100 µg/mL cycloheximide (CHX). Samples were removed for protein extraction at the indicated times after CHX addition. Psf3 protein levels were assessed by immunoblotting. A representative blot is shown, and the results from 3 biological replicates are summarized in the graph. For details on experimental procedures, see Methods. P values were adjusted for multiple comparisons where appropriate (see Methods). *P < 0.05, **P < 0.01

Fig 3: DNA replication dynamics are altered in U2OS T-REx Flp-In GINS3-KO cells expressing MGS-associated variants.(A and B) Western blots showing levels of GINS3cr-FLAG construct and endogenous GINS3 expressed before (A) and after (B) disruption of the endogenous GINS3 gene by CRISPR-Cas9. For each cell line, a minimum of 3 independent protein extractions and 3 independent Western blots were performed; a representative example is shown. (C) Sample flow cytometry plots showing gating strategy for EdU+ cells (left) and bar graph of median EdU signal intensity in EdU+ cells (right). Three cell lines of each genotype were pulsed with EdU for 30 minutes and analyzed. Bars represent mean ± SD, and statistical analyses were performed using 2-tailed t tests. (D) Schematic summary of thymidine analog treatments for the DNA fiber experiment with sample fibers shown below. (E–G) Scatter plots showing median fork speed (E), interorigin distance (F), and fork symmetry (G). Red bars indicate median, and statistical analyses were performed using 2-tailed t tests. The experiment was performed twice; 1 representative replicate is shown. For details on experimental procedures, see Methods. P values were adjusted for multiple comparisons where appropriate (see Methods). *P < 0.05, **P < 0.01, ****P < 0.0001.

Fig 4: Patients with hypomorphic GINS3 variants present with MGS-like phenotypes.(A) Proband from Family 1 at 4 and 6 years of age. Note the frontal bossing, full lips, and posteriorly rotated ears. Lateral x-ray of the knee shows patella hypoplasia at age 4. (B) Patients from Family 2 at 16 year of age (P3) and at 24 years of age (P4), showing frontal bossing, full lips, prominent nose with long nasal root, and small ears posteriorly rotated ears. (C) Patient P5 at 4 years of age showing frontal bossing, full lips, micrognathia, small ears, and clinodactyly. (D) Patient P7 at 10 years of age showing frontal bossing, full lips, and prominent nose with long nasal root.

Fig 5: Patients with hypomorphic GINS3 variants present with growth deficiencies.(A) Pedigrees of 5 families showing allele segregation. Probands are indicated by an arrow. (B) Growth parameters of 7 individuals with biallelic mutations in GINS3.