The burden of trisomy 21 disrupts the proteostasis network in Down syndrome

Abstract

Down syndrome (DS) is a genetic disorder caused by trisomy of chromosome 21. Abnormalities in chromosome number have the potential to lead to disruption of the proteostasis network (PN) and accumulation of misfolded proteins. DS individuals suffer from several comorbidities, and we hypothesized that disruption of proteostasis could contribute to the observed pathology and decreased cell viability in DS. Our results confirm the presence of a disrupted PN in DS, as several of its elements, including the unfolded protein response, chaperone system, and proteasomal degradation exhibited significant alterations compared to euploid controls in both cell and mouse models. Additionally, when cell models were treated with compounds that promote disrupted proteostasis, we observed diminished levels of cell viability in DS compared to controls. Collectively our findings provide a cellular-level characterization of PN dysfunction in DS and an improved understanding of the potential pathogenic mechanisms contributing to disrupted cellular physiology in DS. Lastly, this study highlights the future potential of designing therapeutic strategies that mitigate protein quality control dysfunction.

Fig 1. A simplified schematic illustrating the three major functions of the proteostasis network (Blue), and the key modulators (Red), UPR and HSF-1.

Fig 2

LCLs from DS patients enhanced basal levels of ROS production (A) and expression of ER stress genes (B). Unstimulated DS and CTL LCLs were evaluated for basal superoxide and hydrogen peroxide production using flow cytometry. These unstimulated cells were also assessed for ER stress gene expression using qRT-PCR. N = 3; * P<0.05; ** P<0.01; *** P<0.001; **** P<0.0001.

Fig 3. Grp78 and the PERK-eIF2α pathway are not basally up-regulated in DS LCLs or fibroblasts.

Western blot analyses in both LCLs and fibroblasts from DS and euploid controls did not show a difference in either basal Grp78 abundance (A) or abundance and phosphorylation status of eIF2α (B).

Fig 4. LCLs and fibroblasts from DS patients and mouse models of DS display increased abundance of XBP1s and XBP1s was found localized to the nucleus.

Unstimulated LCLs and fibroblasts from DS individuals and age-matched controls were examined for XBP1s abundance using Western blot (A-C). Both DP16 and DP17 (D) mice display significant increase in XBP1s abundance in the brain (E). Western blots were conducted in triplicate and the images are representative of these replications. WT, wild type; TG, transgenic (DP16 or DP17); n = 3, * P<0.05.

Fig 5. Phosphorylation of IRE1α does not differ greatly between cells from DS and euploid controls.

Three pairs of DS and CTL LCL, as well as one pair of DS and CTL fibroblasts, were analyzed for basal levels of IRE1α phosphorylation (A). These data show that LCLs did not display a clear DS-mediated phenotype, while the fibroblasts showed a DS-mediated increase in phosphorylation (B). Inhibition of the IRE1α endonuclease domain with 4μ8c did not significantly decrease XBP1s abundance in DS fibroblasts (C). Western blots were conducted in triplicate and the images are representative of these replications.

Fig 6

LCLs and fibroblasts from DS patients show increased ATF6 cleavage (A) and increased nuclear localization (B) compared to euploid controls. Both DS LCLs and fibroblasts displayed increased ATF6 cleavage, as evidenced by the presence of a band located at approximately 50 kD (arrow). Significantly increased ATF6 cleavage and nuclear localization was observed in the DS cells compared to euploid controls (C). Western blots were conducted in triplicate and the images are representative of these replications.

Fig 7. Impaired heat shock response in DS fibroblasts.

Basal levels of most heat shock proteins (Hsp) investigated were not significantly different controls; however, Hsp27 was significantly decreased in DS cells (A). Stimulation of these cells for 2h with 40°C heat stress did not result in a significant increase in Hsp90 or 70 (B). These results indicate an abnormal HSF-1 response in DS fibroblasts. Graphs represent results of Western blotting experiments that were conducted in triplicate. Data is represented as the mean ± SEM (*P<0.05, ****P<0.0001).

Fig 8. DS cells display an increased abundance of ubiquitinated proteins and impaired proteasomal function.

Untreated fibroblasts from DS and CTL were stained for ubiquitinated protein and visualized using fluorescent microscopy (A). The amount of basal ubiquitin staining was found to be significantly increased in DS fibroblasts compared to CTL (B), indicating increased load of misfolded proteins in DS cells. Western blot analyses after treatment with the proteasomal inhibitor, MG132 (5μM), show a more rapid increase in polyubiquitinated proteins (C) in DS compared to CTL. Further investigation of proteasomal activity demonstrated that in DS fibroblasts there are significant decreases in both chymotrypsin-like and trypsin-like activity of the proteasome (D). N = 3, * P<0.05, ** P<0.01; *** P<0.001.

Fig 9. Cell viability in DS cells is altered to a greater extent due to exposure to ER stressors compared to CTL.

Cells were treated for 24h with Tm (A) or MB (B) and cell viability was assessed using a WST-1 assay. N = 3–4, * P<0.05; ** P<0.01; *** P<0.001.