Screening for Functional Non-coding Genetic Variants Using Electrophoretic Mobility Shift Assay (EMSA) and DNA-affinity Precipitation Assay (DAPA)

Summary

We present a strategic plan and protocol for identifying non-coding genetic variants affecting transcription factor (TF) DNA binding. A detailed experimental protocol is provided for electrophoretic mobility shift assay (EMSA) and DNA affinity precipitation assay (DAPA) analysis of genotype-dependent TF DNA binding.

Abstract

Population and family-based genetic studies typically result in the identification of genetic variants that are statistically associated with a clinical disease or phenotype. For many diseases and traits, most variants are non-coding, and are thus likely to act by impacting subtle, comparatively hard to predict mechanisms controlling gene expression. Here, we describe a general strategic approach to prioritize non-coding variants, and screen them for their function. This approach involves computational prioritization using functional genomic databases followed by experimental analysis of differential binding of transcription factors (TFs) to risk and non-risk alleles. For both electrophoretic mobility shift assay (EMSA) and DNA affinity precipitation assay (DAPA) analysis of genetic variants, a synthetic DNA oligonucleotide (oligo) is used to identify factors in the nuclear lysate of disease or phenotype-relevant cells. For EMSA, the oligonucleotides with or without bound nuclear factors (often TFs) are analyzed by non-denaturing electrophoresis on a tris-borate-EDTA (TBE) polyacrylamide gel. For DAPA, the oligonucleotides are bound to a magnetic column and the nuclear factors that specifically bind the DNA sequence are eluted and analyzed through mass spectrometry or with a reducing sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) followed by Western blot analysis. This general approach can be widely used to study the function of non-coding genetic variants associated with any disease, trait, or phenotype.

Introduction

Sequencing and genotyping based studies, including Genome-Wide Association Studies (GWAS), candidate locus studies, and deep-sequencing studies, have identified many genetic variants that are statistically associated with a disease, trait, or phenotype. Contrary to early predictions, most of these variants (85-93%) are located in non-coding regions and do not change the amino acid sequence of proteins^1,2. Interpreting the function of these non-coding variants and determining the biological mechanisms connecting them to the associated disease, trait, or phenotype has proven challenging^3-6. We have developed a general strategy to identify the molec....

Protocol

1. Preparation of Solutions and Reagents

1. Order custom DNA oligonucleotide probes for use in EMSA and DAPA.
   1. To reduce non-specific protein binding, design short oligos (between 35-45 base pairs (bp) in length)³⁰, and place the variant of interest directly in the center flanked by its 17 bp endogenous genomic sequence. For EMSA oligos, add a 5' fluorophore. For DAPA oligos, add a 5' biotin tag.
   2. Order both the sense strand and its reverse complement strand. Alternatively, order duplex (pre-annealed) oligos. When naming the oligos, base the nomenclature on an established reference genome.
      Note: "Risk" a....

Results

In this section, representative results of what to expect are provided when performing an EMSA or DAPA, and the variability with regards to the quality of lysate is characterized. For example, it has been suggested that freezing and thawing protein samples multiple times may result in denaturation. In order to explore the reproducibility of EMSA analysis in the context of these "freeze-thaw" cycles, two 35 bp oligos differing at one genetic variant were incubated with a single bat.......

Discussion

Although advances in sequencing and genotyping technologies have greatly enhanced our capacity to identify genetic variants associated with disease, our ability to understand the functional mechanisms impacted by these variants is lagging. A major source of the problem is that many disease-associated variants are located in n on-coding regions of the genome, which likely affect harder-to-predict mechanisms controlling gene expression. Here, we present a protocol based on the EMSA and DAPA techniques, valuable molecular t.......

Materials

Name	Company	Catalog Number	Comments
Custom DNA Oligonucleotides	Integrated DNA Technologies		http://www.idtdna.com/site/order/oligoentry
Potassium Chloride	Fisher Scientific	BP366-500	KCl, for CE buffer
HEPES (1M)	Fisher Scientific	15630-080	For CE and NE buffer
EDTA (0.5M), pH 8.0	Life Technologies	R1021	For CE, NE, and annealing buffer
Sodium Chloride	Fisher Scientific	BP358-1	NaCl, for NE buffer
Tris-HCl (1M), pH 8.0	Invitrogen	BP1756-100	For annealing buffer
Phosphate Buffered Saline (1X)	Fisher Scientific	MT21040CM	PBS, for cell wash
DL-Dithiothreitol solution (1M)	Sigma	646563	Reducing agent
PMSF	Thermo Scientific	36978	Protease Inhibitor
Phosphatase Inhibitor Cocktail	Thermo Scientific	78420	Prevents dephosphorylation of TFs
Nonidet P-40 Substitute	IBI Scientific	IB01140	NP-40, for nuclear extraction
BCA Protein Assay Kit	Thermo Scientific	23225	For measuring protein concentration
Odyssey EMSA Buffer Kit	Licor	829-07910	Contains all necessary EMSA buffers
TBE Gels, 6%, 12 Wells	Invitrogen	EC6265BOX	For EMSA
TBE Buffer (10X)	Thermo Scientific	B52	For EMSA
FactorFinder Starting Kit	Miltenyi Biotec	130-092-318	Contains all necessary DAPA buffers
Licor Odyssey CLx	Licor		Recommended scanner for DAPA/EMSA
Antibiotic-Antimycotic	Gibco	15240-062	Contains 10,000 units/mL of penicillin, 10,000 µg/mL of streptomycin, and 25 µg/mL of Fungizone® Antimycotic
Fetal Bovine Serum	Gibco	26140-079	FBS, for culture media
RPMI 1640 Medium	Gibco	22400-071	Contains L-glutamine and 25mM HEPES