DNA-protein interactions (DPIs) are critical to all living organisms, particularly in the regulation of gene expression, replication, packing, recombination, and repair, as well as RNA transport and translation.
Since microscopically observing interactions between proteins and DNA in the late nineteenth century, scientists have been intrigued in the mechanisms by which proteins associate with and control both DNA and RNA.
DNA-binding proteins are common and ubiquitous, comprising on average 10% of the proteome/genome of higher plants and animals—about 2000 for the average organism.
Interactions may be specific, that is governed by nucleotide sequences and mediated by hydrogen bonding, ionic interactions, and van der Waal’s forces. Control of transcription is a one example. Nucleotide sequence is irrelevant for nonspecific interactions, which occur through attraction between the protein’s functional groups and the sugar-phosphate backbone of DNA.
Analytical methods
Most DPIs are only partly understood, but not for want of trying. Their study often begins with identifying and characterizing the protein component. Analytical methods include microscopy and classical biochemical assays like chromatin immunoprecipitation analysis (ChIP), Systematic Evolution of Ligands by EXponential enrichment (SELEX), Electrophoretic mobility shift assays, DNA footprinting, and protein-binding microarrays.
ChIP causes proteins to bind covalently to their DNA targets, after which they are unlinked and characterized separately. SELEX exposes target proteins to a random library of oligonucleotides. Those genes that bind are separated and amplified by PCR. DNase footprinting locates protein-DNA binding sites followed by limited DNA digestion. Protein-binding DNA microarrays identify gene sequences that associate with labeled target proteins, followed by fluorescence detection.
Microscopic techniques include optical, fluorescence, electron, and atomic force microscopy (AFM), the latter two providing the highest spatial resolution. Where the latter three resolve dynamic interactions, electron microscopy is limited to static observations.
AFM is arguably the most versatile microscopic method because It offers sub-nanometer resolution, images samples in liquids, and probes intermolecular forces between single molecules.
Sandor Kasas, Ph.D., scientific collaborator at the Swiss Federal Polytechnic University (Lausanne) and an investigator into the microscopy of DPIs, notes that high-speed AFM explores the structures of proteins and associated genes with very high resolution. “With high-speed AFMs, all the dynamics of the interaction can be followed in real time while measuring the interaction forces between the protein and DNA.”
High-speed AFM explores the structures of proteins and associated genes with very high resolution.
This last application involves attaching the protein to the AFM tip and measuring its interaction with free-floating DNA.
Depending on what investigators are looking for the protein sample may or may not be in solution, but in any case it must be immobilized onto a substrate to be visible to the AFM.
For quick and dirty experiments one can image samples in air. In this case attachment is relatively easy. If one is looking for interaction dynamics, however, imaging must occur in liquids, where attachment is more complicated.
“The DNA must be attached to the surface strongly enough to be visible to the AFM, but bound loosely enough to interact with the protein. This compromise is sometimes difficult to achieve,” Kasas adds.
A low speed AFM costs about $100,000 to $200,000, which means groups or departments will often pool resources to acquire one. Dedicated AFM facilities are often happy to collaborate with biologists.
High-speed AFMs are another matter
“These are still rare and only few groups possess them, only about five in France and perhaps a dozen in the United States,” Kasas says. Yet for biologists working on the cutting edge of DPIs they are essential, and for their research transformative.
Biocompare’s Microscope Search Tool
Find, compare and review microscopes
from different suppliers Search
"High-speed AFM is now a reality, delivering video-rate imaging of biological processes like myosin traversing actin filaments, which was unthinkable only few years ago,” Kasas adds.
Constant improvements
As with any burgeoning research field, experimental necessity causes revision and improvements in standard methods.
Protocol for using Promega's HaloCHIP™ to characterize DNA-protein interactions. Image courtesy of Promega.
For example Promega’s HaloCHIP™ System, which the company promotes as an alternative to ChIP, covalently captures intracellular DNA-protein complexes without using antibodies. Cells express proteins as fusion products of the company’s histidine-tagging HaloTag® reagent. These are subsequently crosslinked to DNA with formaldehyde and captured on HaloLink™ Resin. Washing removes nonspecific proteins and DNA. Heating causes the fusion protein to dissociate from the DNA, which is then purified and analyzed. Halo CHIP provides answers in less than 48 hours, according to Promega, with fewer steps and improved signal-to-noise, while minimizing experimental error.
For the rapid characterization of the protein component Promega offers a gel shift or electrophoretic mobility shift assay, a method that is widely used to study sequence-specific DNA-binding proteins, particularly transcription factors. Gel shift is based on the slower migration of DNA-protein complexes through a non-denaturing polyacrylamide gel, compared with free or double-stranded DNA.
Determination of DNA sequences following the HaloCHIP assay can be achieved in several ways, depending on prior knowledge of DNA binding sites. Typical methods include PCR, qPCR, microarrays, or sequencing. Proper controls are required to understand and quantitate enrichment.
”There is a significant increased interest in genome-wide binding analysis of transcription factor and/or epigenetic proteins and characterizing this throughout for many different cell types and disease states,” Danette Daniels, senior research scientist at Promega, explains. “These efforts come primarily from academic researchers, but also in pharma and biotech. Additionally she notes that characterizing and identifying protein-RNA complexes, and understanding the RNA sequence specificity required for those interactions, “is an exciting new direction that is learning a lot from protein-DNA analysis.”
Physical characterization
Biochemical analysis is just one aspect of DPI characterization. Physical methods like X-ray crystallography also provides valuable insights, namely a three-dimensional atomic view of DPIs. The resulting crystal structure may reveal the location of the binding site complex, allowing for identification of essential amino acid residues necessary for complex formation. This can help with downstream pharmaceutical development, in particular, structure-based drug design.
Crystallization of proteins for structural studies has been around for decades. As it is often difficult to predict optimal (and practical) conditions for crystallization, favorable conditions are identified by screening against variations of chemical formulations and concentrations. Yet when conventional techniques are applied to DPIs the resulting crystals often consist of either protein or DNA alone, leading to false positives and wasted time. It appears that the crystallization process itself may cause dissociation of these complexes.
Faced with this problem Thomas Hollis, professor of biochemistry at the Wake Forest School of Medicine, researched the Biological Macromolecule Crystallization Database to find examples of crystallization of protein-DNA complexes. It soon became clear that conditions favoring formation of these crystals were unavailable from most commercial protein-DNA complex screens. Many of the successful protocols employed divalent metal ions. “So we thought it might be useful to create our own screen that had all of the conditions in some array form that promote protein-DNA complex crystallization,” Hollis wrote in his blog.
Applying an experimental discovery technique known as incomplete factorial design in combination with known crystallization conditions led to the Protein-Nucleic Acid Complex Crystal Screen, a product offered by reagents seller Kerafast.
The Protein-Nucleic Acid Complex Crystal Screen consists of a 48-condition screen designed for protein-nucleic acid complex crystallization. Each condition uses a different combination of precipitant, buffer and salt, and 1.5 mL of each mixture is provided. A table maintained by Kerafast lists the 48 conditions in its screen. Investigators plate those conditions into a 96-well plate for crystallization experiments to identify which condition(s) cause crystallization.
“Armed with that information, the researcher can dig deeper into what works, further optimizing the various condition characteristics to determine the ideal condition for X-ray crystallography,” explains Travis Riedel, VP of product development at Kerafast.
The directed, 48-condition screen was designed specifically for protein-nucleic acid complex crystallization by analyzing actual successes. The screen is especially useful when protein is scarce because it offers an efficient way to identify the ideal conditions for crystallization, avoiding sample-wasting trial and error.
“It also decreases the number of false hits consisting of protein or nucleic acid alone,” Riedel says.
But screening is just the first step in the process. After identifying crystal-forming conditions they must be further optimized to produce crystals that diffract strongly. This is a limitation to all crystal screens.
Image: Structural overview of the AmrZ protein – amrZ1 DNA complex (PDB ID: 3QOQ), as determined by the Dr. Thomas Hollis laboratory at Wake Forest University. Image courtesy of Kerafast.