Parkinson-Associated SNCA Enhancer Variants Revealed by Open Chromatin in Mouse Dopamine Neurons

Abstract

The progressive loss of midbrain (MB) dopaminergic (DA) neurons defines the motor features of Parkinson disease (PD), and modulation of risk by common variants in PD has been well established through genome-wide association studies (GWASs). We acquired open chromatin signatures of purified embryonic mouse MB DA neurons because we anticipated that a fraction of PD-associated genetic variation might mediate the variants' effects within this neuronal population. Correlation with >2,300 putative enhancers assayed in mice revealed enrichment for MB cis-regulatory elements (CREs), and these data were reinforced by transgenic analyses of six additional sequences in zebrafish and mice. One CRE, within intron 4 of the familial PD gene SNCA, directed reporter expression in catecholaminergic neurons from transgenic mice and zebrafish. Sequencing of this CRE in 986 individuals with PD and 992 controls revealed two common variants associated with elevated PD risk. To assess potential mechanisms of action, we screened >16,000 proteins for DNA binding capacity and identified a subset whose binding is impacted by these enhancer variants. Additional genotyping across the SNCA locus identified a single PD-associated haplotype, containing the minor alleles of both of the aforementioned PD-risk variants. Our work posits a model for how common variation at SNCA might modulate PD risk and highlights the value of cell-context-dependent guided searches for functional non-coding variation.

Keywords: ATAC-seq; Parkinson disease; alpha-synuclein (SNCA); chromatin accessibility; dopaminergic neurons; enhancer; regulatory variation.

Figure 1

Preliminary Validation of ATAC-seq Catalogs Generated from Ex Vivo DA Neurons (A) The midbrain (MB) and forebrain (FB) of E15.5 brains from Tg(Th-EGFP)DJ76Gsat mice are microdissected, dissociated, and isolated by FACS. (B) Read pileup and called peaks for the MB and FB libraries at the Th locus. (C and D) Chromatin accessibility, genome-wide, is correlated between replicates. (E) The sequences underlying MB and FB peaks display a high degree of evolutionary sequence constraint as measured by phastCons scores. (F and G) For both the MB and FB, gene ontology terms of the expressed genes nearest to each peak reflect the neuronal origin and function of these catalogs.

Figure 2

Validation of the Putative CRE Catalogs In Vivo (A) Of the elements annotated in VISTA as having enhancer activity, 62% and 56% of these are represented in the MB and FB catalogs, respectively. (B) An abundance of open-chromatin regions in the MB and FB catalogs overlap confirmed neuronal enhancers (≥70%). (C) Neuronal enhancers were stratified by the anatomical domains in which they are active; those that are reported active in the MB and FB are enriched in our MB and FB catalogs, respectively. (D–H) Testing five prioritized putative CREs in vivo identifies five neuronal enhancers. (D) A putative CRE in intron 1 of KCNQ3 directs expression in the midbrain, hindbrain, and neural tube of E11.5 lacZ reporter mice. It fails to direct expression in a transgenic zebrafish assay at either 3 or 5 days post fertilization (dpf); reporter expression is present in ≤25% of mosaics. (E, F, and G) Putative CREs downstream of FOXG1, upstream of NR4A2, and in an intron of CRHR1 fail to direct expression in transgenic mice; however, they direct robust neuronal appropriate expression in transgenic zebrafish reporter assays (scored for expression in MB, FB, amacrine cells [ACs], hindbrain [HB], and spinal cord [SC]). (H) A putative CRE downstream of FOXA2 directs neuronal expression in both transgenic mice and zebrafish assays. n mosaic zebrafish scored: ≥141 for 3 dpf, ≥119 for 5 dpf. All constructs have since been deposited in the VISTA database under the supplied hs numbers.

Figure 3

Identification of Transcription Factors (TFs) Important to DA Neurons (A) The kmer predicted to have the greatest regulatory potential underlying MB ATAC-seq peaks corresponds to the RFX family of TFs. (B, C, and D) RNA-seq quantification in these same cells indicates this enrichment is likely due to RFX3 or RFX7 activity. Examining the ATAC-seq signal over predicted binding sites reveals a robust TF footprint (C) and a general enrichment of reads overlapping RFX sites genome-wide (D). (E–H) Similarly, a kmer corresponding to FOXA1 and/or FOXA2 has similar evidence for the activity of one or both of these TFs. (I–L) The third-ranked motif most likely corresponds to ASCL1, and although it fails to leave a robust TF footprint (K), there is clear enrichment of ATAC-seq signal overlapping genome-wide predicted ASCL1 binding sites (L). (M–P) NR4A2, canonically associated with DA neuron biology, is identified as a highly expressed TF that probably contributes to the regulatory potential of the putative CREs; however, it fails to leave a TF footprint in the cut-site patterns around predicted motif sites (O) and is only mildly enriched for ATAC-seq reads over its predicted binding sites (P).

Figure 4

A MB-Specific Enhancer Directs Expression in Catecholaminergic Populations of Neurons Known to Parkinson Disease Biology (A) An IGV track indicating the location of the MB-specific region of open chromatin located in intron 4 of Snca. (B) Snca is differentially expressed between the MB and FB DA neurons. The red bar is the mean expression of the four replicates (black dots). (C) At 72 hpf, stable transgenic zebrafish reporter assays indicate this putative CRE is capable of directing reporter expression in key catecholaminergic neuronal populations, including the locus coeruleus (LC), the catecholaminergic tract (CT) of the hindbrain, the diencephalic cluster (DC), and the subpallium (SP), into which the DC projects. (D–H) Further studies in lacZ reporter assays in embryonic (E) and post-natal (P) mice indicate dynamic enhancer usage across developmental time. (D) This enhancer directs expression throughout the MB, FB, dorsal root ganglia (DRG), sympathetic chain (SC), and cranial nerves (CN) of E12.5 mice. (E) By E15.5, reporter expression is observed in the amygdala and/or piriform cortex (AM/PC), sympathetic chain, MB, and hypothalamus (Hyp). (F) Patterns of reporter expression at P7 reflect those seen at E15.5. (G) Reporter activity is observed at P30 in the amygdala; hypothalamus and thalamus (Thal); brain stem (BS); substantia nigra (SN); ventral tegmental area (VTA); and the periaqueductal gray area (PAG). (H) In aged mice (P574), reporter expression is detected robustly in the brain stem and faintly in the amygdala.

Figure 5

Identification of Proteins Whose Binding Is Impacted by the Implicated PD-risk SNPs (A and B) MA plots for both dbSNP: rs2737024 and dbSNP: rs2583959 indicate the magnitude of the effect of the minor and major allele on binding. Cut-off for differential binding: log₂(major/minor) ≥ 1.5 or ≤ −1.5. (A) NOVA1 and APOBEC3C (green circles) bind at dbSNP: rs2737024 with greater affinity for the major allele, but PEG10 and SNRPA (red circles) have a greater affinity for the minor allele. (B) CHMP5 (red circle) has a greater affinity for the minor allele of dbSNP: rs2583959. (C) Representative images of the protein binding for each of the differentially bound proteins. (D) Expression analysis in the MB and FB DA neurons for each of the differentially bound proteins indicates Nova1, Peg10, Snrpa, and Chmp5 to be highly expressed in these populations, yet none of the Apobec family member genes are expressed (RPKMs ≤1, data not shown). The red bar is the mean expression of the four replicates (black dots).

Figure 6

A Schematic of the Chromatin Interactions, LD Structure, Variation, and Open Chromatin at the SNCA Locus Publicly available DNase hypersensitivity site (DHS) linkage analysis suggests that the promoter of SNCA possibly interacts with the identified MB-specific enhancer, the lead GWAS variant (dbSNP: rs356182), and a previously functionally validated variant (dbSNP: rs356168). ChIA-PET data suggest that the MB-specific enhancer might interact with variant dbSNP: rs356168. Open-chromatin data from DA neurons do not overlap with any variants at this locus or haplotype other than at the MB-specific enhancer. LD analysis at this locus indicates that despite the low LD structure between the lead GWAS variant (dbSNP: rs356182) and the enhancer-associated variants (dbSNP: rs2737024 and dbSNP: rs2583959), the variants are in the same haplotype. Therefore, the GWAS signal might, at least in part, be flagging the identified enhancer-associated variants.