Cell Development Deficiency and Gene Expression Dysregulation of Trisomy 21 Retina Revealed by Single-Nucleus RNA Sequencing

Prevalence and diversity of retinal disease in adults with Down syndrome

While epidemiologic data exists for some ophthalmic diseases in people with Down Syndrome (DS), like strabismus and amblyopia, no studies explore the prevalence of retinal disease in people with DS on a large scale. This study utilized a literature review and exploratory epidemiology analysis to examine patterns of retinal disease in people with DS. To evaluate previous studies on physiology and/or anatomy in retinal models representing DS or in the retinas of people with DS, all relevant terms related to Down Syndrome, retina, and retinal diseases were searched in PubMed and Scopus. Data from the health platform TriNetX was then utilized to determine the prevalence and prevalence odds ratio (POR) of retinal disorders, including diabetic retinopathy and age-related macular degeneration (AMD), within the U.S. adult population with DS. The final literature review included 28 of 535 screened studies and found that a DS diagnosis was associated with atypical retinal vascularization, retinal thickening, and abnormal neuronal development. Of 55,198,979 individuals included in the population study, 97,795 (0.18%) had a recorded DS diagnosis. Compared to the population without DS, the population with DS had significantly increased PORs for any retinal diagnosis (3.78, 95% CI 3.63-3.93), for 16 of 18 recorded individual retinal diagnoses, and for 4 of 5 major diagnostic categories, including diabetic retinopathy (2.56, 95% CI 2.33-2.82) and macular degeneration (4.01, 95% CI 3.42-4.71). The conclusion is that retinal anomalies common to people with DS likely contribute to higher rates of recorded retinal disease. However, future studies should evaluate this relationship.

Rhodopsin-positive cell production by intravitreal injection of small molecule compounds in mouse models of retinal degeneration

We aimed to verify whether the intravitreal injection of small molecule compounds alone can create photoreceptor cells in mouse models of retinal degeneration. Primary cultured mouse Müller cells were stimulated in vitro with combinations of candidate compounds and the rhodopsin expression was measured on day 7 using polymerase chain reaction and immunostaining. We used 6-week-old N-methyl-N-nitrosourea-treated and 4-week-old rd10 mice as representative in vivo models of retinal degeneration. The optimal combination of compounds selected via in vitro screening was injected into the vitreous and the changes in rhodopsin expression were investigated on day 7 using polymerase chain reaction and immunostaining. The origin of rhodopsin-positive cells was also analyzed via lineage tracing and the recovery of retinal function was assessed using electroretinography. The in vitro mRNA expression of rhodopsin in Müller cells increased 30-fold, and 25% of the Müller cells expressed rhodopsin protein 7 days after stimulation with a combination of 4 compounds: transforming growth factor-β inhibitor, bone morphogenetic protein inhibitor, glycogen synthase kinase 3 inhibitor, and γ-secretase inhibitor. The in vivo rhodopsin mRNA expression and the number of rhodopsin-positive cells in the outer retina were significantly increased on day 7 after the intravitreal injection of these 4 compounds in both N-methyl-N-nitrosourea-treated and rd10 mice. Lineage tracing in td-Tomato mice treated with N-methyl-N-nitrosourea suggested that the rhodopsin-positive cells originated from endogenous Müller cells, accompanied with the recovery of the rhodopsin-derived scotopic function. It was suggested that rhodopsin-positive cells generated by compound stimulation contributes to the recovery of retinal function impaired by degeneration.

Detection of Embryonic Trisomy 21 in the First Trimester Using Maternal Plasma Cell-Free RNA

Prenatal trisomy 21 (T21) screening commonly involves testing a maternal blood sample for fetal DNA aneuploidy. It is reliable but poses a cost barrier to universal screening. We hypothesized maternal plasma RNA screening might provide similar reliability but at a lower cost. Discovery experiments used plasma cell-free RNA from 20 women 11−13 weeks tested by RNA and miRNA microarrays followed by qRT-PCR. Thirty-six mRNAs and 18 small RNAs of the discovery cDNA were identified by qPCR as potential markers of embryonic T21. The second objective was validation of the RNA predictors in 998 independent pregnancies at 11−13 weeks including 50 T21. Initial analyses identified 9−15 differentially expressed RNA with modest predictive power (AUC < 0.70). The 54 RNAs were then subjected to machine learning. Eleven algorithms were trained on one partition and tested on an independent partition. The three best algorithms were identified by Kappa score and the effects of training/testing partition size and dataset class imbalance on prediction were evaluated. Six to ten RNAs predicted T21 with AUCs up to 1.00. The findings suggest that maternal plasma collected at 11−13 weeks, tested by qRT-PCR, and classified by machine learning, may accurately predict T21 for a lower cost than plasma DNA, thus opening the door to universal screening.