Our DNA skills are lopsided: We are much better at dissecting DNA structure and organization than we are at synthesizing DNA. But our DNA skills are starting to become more balanced, now that we are benefitting from exponential decreases in the cost of oligonucleotide synthesis, dramatic improvements in technology, and the incorporation of novel paradigms.
These developments have already culminated in the generation of synthetic viral, bacterial, and eukaryotic chromosomes. Going forward, we will continue to refine our skills in DNA synthesis, supporting the use of synthetic DNA in the study of biological processes.
“We are always thinking about how to stay at the cutting edge of DNA synthesis and assembly technologies,” says Daniel G. Gibson, Ph.D., vice president of DNA Technologies at Synthetic Genomics, Inc. (SGI) and associate professor in the synthetic biology group at the J. Craig Venter Institute (JCVI). In 2004, when Dr. Gibson and colleagues resolved to synthesize bacterial genomes, generating even 5,000 base pairs of synthetic DNA was challenging. “Our goal at the time,” recalls Dr. Gibson, “was to build something that was two orders of magnitude larger, and that required developing brand-new methods.”
“We began [by] thinking of how to stitch together pieces of DNA to build large pieces, such as genetic pathways, bacterial chromosomes, and beyond,” relates Dr. Gibson. One of the early efforts undertaken by Dr. Gibson’s group made use of cellular extracts from Deinococcus radiodurans, a bacterium that can tolerate very high doses of ionizing radiation. It sustains damage but remains viable because it can rejoin fragmented chromosomal DNA.
“Our idea,” emphasizes Dr. Gibson, “was to somehow leverage what the cell was doing to stitch together our pieces of DNA.”
Synthesizing Whole Chromosomes
This stitching approach seemed promising at first, but it had multiple shortcomings. “We needed something faster,” explains Dr. Gibson. “Our goal was to synthesize a whole chromosome within a year, and we still did not have a method that worked.”
Dr. Gibson’s group started exploring the possibility of using purified enzymes to link DNA fragments. “Ultimately, it became pretty simple to do this, provided the DNA fragments were designed to overlap each other, allowing the ends to come together specifically,” explains Dr. Gibson.
This process requires three enzymes: an exonuclease, which degrades the ends of the DNA fragments; a DNA polymerase, which fills gaps and anneals and repairs damaged DNA; and a DNA ligase, which covalently joins the molecules. This in vitro recombination method has become known as Gibson Assembly.
“We used this method to synthesize whole chromosomes, but we also used natural homologous recombination machinery in yeast,” elaborates Dr. Gibson. This machinery, he adds, is remarkable because it enables yeast to take up and assemble dozens of overlapping large DNA fragments into large pieces. These key discoveries set the stage for the creation of synthetic cells. When scientists at the JCVI announced the creation of the first synthetic cell in 2010, they credited their success to the combined use of whole-genome synthesis and transplantation technologies.
Gene-synthesis technologies provided crucial assistance when the JCVI, SGI, and Novartis collaborated to accelerate the generation of synthetic flu vaccines. Conventional influenza vaccine preparation currently requires at least a month after a viral sample has been obtained, and shortening this time would significantly improve epidemic and pandemic preparedness.
“We teamed up and showed that we can synthesize influenza viruses for vaccine production using DNA synthesis in less than a week,” asserts Dr. Gibson. As part of this collaboration, Dr. Gibson and colleagues carried out the viral nucleic acid synthesis. “What we needed was to completely change the way we synthesize genes,” explains Dr. Gibson.
The chemical synthesis approach used by Dr. Gibson and colleagues combined enzymatic cell-free gene synthesis with enzymatic error correction. The chemical synthesis of DNA is error-prone, with an error being introduced approximately once every 500–1000 base pairs of DNA, and this precludes the synthesis of large genomic regions. To ensure that the DNA of the influenza virus genome fragments is copied both accurately and efficiently, Dr. Gibson and colleagues used an approach that can increase the overlap between the nucleotides, incorporate an enzymatic correction step, and concomitantly assemble a larger number of oligonucleotides.
“We ended up developing a process using endonucleases and exonucleases, where the enzymes could identify and cleave mutations, so that only the error-free DNA molecules were left,” states Dr. Gibson.
This process not only generated high-quality influenza virus genes, it also reduced the time between receiving the DNA sequences and generating highly accurate influenza genes to about 16 hours. “These genes,” maintains Dr. Gibson, “were demonstrated to rescue flu virus by Novartis, which used them in vaccine production.”
The technologies that catalyzed these rapid, robust, and accurate DNA-synthesis methods became the driving force behind the digital-to-biological converter (DBC), a biomanufacturing unit developed by Dr. Gibson and his colleagues at SGI. “As a proof of concept, we demonstrated that we could build a DBC that could accept digitized DNA information,” states Dr. Gibson. This information may originate anywhere in the world, be conveyed via email, and arrive at the DBC. The digital DNA sequence information can then be converted into biological material such as DNA, RNA, and proteins. The DBC is not commercially available yet, but the BioXp 3200 system, a benchtop DNA printer, is already being marketed through the SGI subsidiary SGI-DNA. Whereas the DBC starts with DNA sequence information, the BioXp 3200 instrument starts with oligonucleotide pools that are obtained from a vendor and then pooled and sent to customers.
More New Tools
“We are focusing on developing new tools for detecting and analyzing different types of DNA and RNA sequences,” says Ken Halvorsen, Ph.D., principal investigator at The RNA Institute, University at Albany, State University of New York. A major effort in Dr. Halvorsen’s group is focusing on studying DNA nanoswitches, which are molecular structures that comprise a linear DNA duplex and an inducible loop.
Upon exposure to external target molecules, DNA nanoswitches undergo reversible topological changes that can provide information about molecular association and dissociation reactions. The nanoswitch starts out as a linear piece of DNA. When the nanoswitch interacts with a target oligonucleotide, a bulge or loop structure is generated.
“That is the main outcome of the switch function,” explains Dr. Halvorsen. “There are several ways to determine whether a molecule that a researcher is looking for is present.”
One strategy is to perform agarose gel electrophoresis, which can separate a structure with a loop from a structure without a loop based on the differential migration patterns of the two. Another strategy is to detect the presence of a loop, at the single-molecule level, and then measure its length. The second strategy, notes Dr. Halvorsen, is one of the main DNA-related nanotechnologies that has been used in his laboratory.
In collaboration with Alan Chen, Ph.D., and Mehmet Yigit, Ph.D., both chemists at the University of Albany, Dr. Halvorsen and colleagues modeled the molecular dynamics of an experimentally observed phenomenon: the preferential affinity of graphene oxide to single-stranded DNA over double-stranded DNA. Graphene oxide, the water-soluble form of graphene, can be used for DNA and RNA detection and has shown promise for applications including environmental monitoring and biosensor development. “We wanted to shed light on some of the basic principles of the thermodynamics of the interaction between graphene surfaces and oligonucleotides,” says Dr. Halvorsen.
The molecular models that were developed by Dr. Halvorsen and colleagues may be used to characterize the adsorption of oligonucleotides to graphene surfaces. By revealing details of the complex interaction between graphene surfaces and nucleic acids, these models provide a framework for the rational design of graphene-based biosensors.
“Rather than synthesize individual sequences one at a time via a solid support, as is done in conventional DNA synthesis, we have a product called OligoMix®, which is generated on a microfluidic chip,” says Chris Hebel, Ph.D., vice president of business development at LC Sciences.
OligoMix consists of pooled oligonucleotids, which are generated with a technology that has also been used to synthesize microarrays in gene-expression applications. “We discovered that these oligonucleotides, if they are cleaved off the array into the mix, have quite a few applications in addition to the applications of the oligonucleotide arrays,” asserts Dr. Hebel. OligoMix contains thousands of single-stranded oligonucleotides per tube. The oligonucleotides can be up to 150-mers, and they can be prepared with a variety of 5’-, 3’-, or internal modifications.
The approach used to prepare the OligoMix is based on three different technologies. The first technology is the microfluidic chip itself. (“We are not just spotting reagents on the chip, but the chip contains microchannels that allows us to control the flow of fluid across the array,” notes Dr. Hebel.) The second technology involves a photogenerated mask. (It provides the ability to gate and control which chambers on the chip will be exposed to light.) The third technology involves a photogenerated acid.
In conventional oligonucleotide synthesis, an acid-labile protecting group is generally placed on each nucleotide, added to the growing nucleotide chain, and removed with acid; in synthesis on solid support, the column is flooded with acid to deprotect the nucleotide. “On arrays, we only want to remove that acid-labile group on certain features of the array at a time,” advises Dr. Hebel. After the array is flooded with an acid precursor, exposure of the acid precursor to light generates the acid that deprotects the previous nucleotide only on the desired feature. “That allows us to selectively deprotect and synthesize all the different sequences at one time,” explains Dr. Hebel.
While microarrays have become somewhat less popular for certain applications in recent years, the technology allows oligonucleotides to be synthesized and cleaved. “Once we cleave the oligonucleotides, they may be used for quite a few different applications by our customers, so even though the arrays may be somewhat obsolete for gene expression, they are still very relevant for other applications,” insists Dr. Hebel.
The OligoMix can also be used for sequence capture. “In many cases, investigators want to perform targeted sequencing, where instead of sequencing the entire genome, they need to selectively enrich for certain genomic regions,” says Dr. Hebel. Several different target-capture technologies have been developed over the last few years. “The common thing between all these technologies is the need for lots of primers to do the capture,” notes Dr. Hebel. Sometimes, thousands or tens of thousands of oligonucleotides are needed to capture the regions that are of interest.
Additional applications are for gene synthesis—when oligonucleotides synthesize gene fragments that are subsequently assembled—and for library construction. One of the current challenges is that the yield of the oligonucleotides in OligoMix is small because they are synthesized on the array. “We generally recommend that customers amplify the OligoMix before they use it, and while that is feasible, it adds another step, and it increases the level of complexity to what end-users are already trying to do,” explains Dr. Hebel. Current efforts at LC Sciences are focusing on increasing the yield by improving the synthesis technology and on developing amplification technologies prior to delivery of the product to the customers.
“The oligonucleotide industry has changed dramatically in the past 10 years,” says Mark Behlke, M.D., Ph.D., CSO at Integrated DNA Technologies (IDT). During the early 2000s, the main use of oligonucleotides was for PCR primers, and at that time they were often needed in large numbers for this application. “Our largest order was over a million oligonucleotides for a single project, all of which were unmodified PCR primers,” recalls Dr. Behlke. The main challenge in that setting was to generate a very simple, uncomplicated product of a very high quality and at a very high throughput.
“While researchers still need PCR primers, those products do not tend to be the major issue, and today we have instead moved toward larger scale, much more complex synthesis as the most important items,” says Dr. Behlke. A crucial effort at IDT is focusing on NGS adaptors, which are indispensable for applications such as sequencing. Generating these more complex products, at extreme purity and larger scale, is currently a major focus, and it requires different manufacturing processes and different quality control measures.
Additionally, there is an increasing demand for complex modifications. “The fraction of our business that requires complex modification has increased. In addition, the modifications we are dealing with today are more complicated than the ones we dealt with a decade ago,” says Dr. Behlke. Generating long, highly modified, high-quality oligonucleotides poses a different set of challenges. “We still could incorporate those modifications a decade ago. At that time, the need for oligonucleotides was much lower. Currently, we need to make them by the thousands, and the production lines have become more complex,” says Dr. Behlke.
Another recent challenge stems from the need for cGMP-quality products. “One of the goals of entering into the genomics era was transitioning research into diagnostics and therapeutics, and that is happening, but the result of that is that people working in clinical diagnostics need quality products that can receive FDA approval or CE Mark approval,” says Dr. Behlke. These developments reflect the success that DNA diagnostic applications have seen in recent years. Manufacturing these products requires certain closed off, restricted access areas that operate under a different regulatory compliance environment.
“At IDT, we have separate cGMP suites that are under ISO 13845 guidance, which enables us to sell products that can be used in FDA-approved kits and are CE Mark-approved in Europe,” says Dr. Behlke. Recent research advances also opened the interest for using oligonucleotides in clinical trials, with the hope for human therapeutic applications. “That is a more specialized area that we have not entered into, and we refer customers to other companies that specialize in pharmaceutical-grade oligonucleotides for human administration,” says Dr. Behlke.
“We are primarily performing quantitative and qualitative analyses for oligonucleotide/nucleic acid and protein therapeutics and biomarkers,” says Laixin Wang, Ph.D., vice president at NovaBioAssays, and cofounder and senior vice president at Chongqing Denali Medpharma. Scientists at NovaBioAssays provide various innovative analytical services for drug discovery and development applications, such as quantification and metabolite identification of small molecule drugs and large biopharmaceutics in biological matrices with and without automated immunoaffinity enrichment.
At NovaBioAssays, a recognized oligonucleotide bioanalysis service provider, one of the main activities is to perform novel liquid chromatographic-based assays to quantify and characterize oligonucleotide therapeutics and their metabolites in biological matrices. In addition to their sensitivity and specificity, these approaches allow multiple oligos to be quantitated in a single assay. NovaBioAssays also provide in vitro stability and protein-binding evaluations.
In a recent study that aimed to improve the assay sensitivity and throughout provided by liquid chromatography–mass spectrometry (LC–MS/MS), Dr. Wang and scientists at NovaBioAssays developed a new solid-phase extraction (SPE) method that produces high-quality oligonucleotide extracts from plasma or tissue homogenate samples. In another study, to examine the performance of the three major platforms (triple quad, Q-Exactive, and Q-Tof) that are currently being used in bioanalytical laboratories, investigators at NovaBioAssays performed a side-by-side comparison to examine the platforms’ strengths and limitations in performing the quantitative and qualitative bioanalysis of oligonucleotides.
“With CRISPR gene-editing techniques emerging as a new modality of gene therapy, the characterization and quantitation of large RNA oligos (100-mer or longer) are posing challenges for analytical and bioanalytical chemists in the development of oligonucleotide/mRNA therapeutics,” says Dr. Wang.