Artificially Expanding DNA’s Genetic Code in Designer Microbes

  • DNA isn’t What it Used To Be!
  • Scripps Team Led by Floyd Romesberg Demonstrates Six-Base DNA Replication in Living Bacteria
  • Are We On Our Way to Semi-synthetic Life-forms? With Vast Potential—But at What Risks? 

Doing the Impossible 

Seemingly everyone these days knows about DNA and how seemingly simple pairings of A-T and G-C encode all life forms. Now, a team led by Floyd Romesberg, a biological chemist at the Scripps Research Institute in San Diego, California, has created a synthetic base pair, which you can simply think of as X-Y, to produce artificial DNA that replicates with six bases! Click here to read more about Romesberg’s findings that were reported last month in Nature.

Prof. Floyd E. Romesberg, Department of Chemistry, The Scripps Research Institute, La Jolla, California (taken from via Bing Images).

Prof. Floyd E. Romesberg, Department of Chemistry, The Scripps Research Institute, La Jolla, California (taken from via Bing Images).

Expanding the genetic code of DNA has been pursued for decades. Romesberg’s achievement, however, has set the scientific world—and news media—a buzz because his version of expanded DNA (eDNA) actually replicates, setting the stage for future biosynthetic engineering aimed at microbes having correspondingly expanded RNA (eRNA) and—here’s the punchline—greatly expanded protein (e-protein) diversity.

Whereas natural proteins are comprised of 20 amino acids encoded by four-base RNA/DNA, Romesberg’s e-proteins could be comprised of 172 (!) amino acids—both natural and mostly artificial—as the result of more triplet codons derived from six-base eRNA/eDNA.

As listed in the graphic below taken from The Wall Street Journal, this stunning achievement in DNA manipulation research is a very big deal in view of the large number of potential applications it enables, covering virtually the entire spectrum of biotechnology and health.


The remainder of this post provides a bit of technical detail and, perhaps more interestingly, some reported opinions that are decidedly positive or—not surprisingly—strongly negative because of concerns for “unintended consequences” of the sort that society has experienced in the past.

Look Folks, no Watson-Crick H-bonds!

Structure and paired orientation of the unnatural X-Y base pair compared to natural C-G base pairing (taken from Romesberg and coworkers Nature 2014).

Structure and paired orientation of the unnatural X-Y base pair compared to natural C-G base pairing (taken from Romesberg and coworkers Nature 2014).

What immediately struck me as being quite unexpected—if not amazing—about the unnatural X-Y base pair was the absence of Watson-Crick H-bonding commonly associated with specific A-T and G-C base pairing. Instead, these X and Y bases—with actual abbreviations d5SICS and dNaM—have relatively simple bicyclic aromatic rings and minimalistic substituents. My assumption is that these hydrophobic moieties—which aren’t even actual bases (!)—have very specific complementary geometry and “pi-stacking” interactions with hydrophobic moieties in flanking A-T and/or G-C base pairs.

Readers interested in technical details underlying this feat—of which there are many—will need to read the entire report. However, some of the “tricks” used by the Romesberg team are worth mentioning here, admittedly in over-simplified terms for the sake of brevity.

  • Their earlier results indicated that passive diffusion of unnatural nucleosides into microbes was possible but subsequent conversion to unnatural nucleotide triphosphates that are required for DNA synthesis was inefficient.
  • That problem was cleverly finessed by “borrowing” a suitable nucleotide triphosphate transporter (NTT) from a certain eukaryotic marine phytoplankton. They also had to include a couple of chemicals in growth media for this NTT to adequately function.
  • Conventional molecular biology was used to prepare a circular double-stranded plasmid having X-Y at a specific locus for transfection into E. coli bacteria to determine if bacterial DNA polymerase would replicate X and Y when “fed” the triphosphate forms of X and Y (dXTP and dYTP).

After 15 hours of growth, which represented 24 doublings or 2 X 107 amplification of the initial X-Y plasmid, analysis confirmed what had been hoped—namely, the unnatural X-Y base pair was retained during replication. Voilà!

They also investigated resistance of X-Y base pairs to E. coli DNA excision-repair processes after reaching stationary phase, and found X-Y to be quite stable: 45% and 15% retention (toward replacement by A-T) after days 3 and 6, respectively.

Here’s what Romesberg and coworkers opined in their concluding remarks:

“In the future, this organism, or a variant with the [unnatural base pairs] incorporated at other episomal or chromosomal loci, should provide a synthetic biology platform to orthogonally re-engineer cells, with applications ranging from site-specific labeling of nucleic acids in living cells to the construction of orthogonal transcription networks and eventually the production and evolution of proteins with multiple, different unnatural amino acids.”

At the risk of belittling this major milestone, there is clearly much more to do in order to extend—pun intended—X and Y into unnatural eRNA and, eventually, unnatural e-proteins prophetically imagined in the above graphic. Obviously there are many challenges ahead, but as the saying goes “every journey begins with a single step,” and in my opinion Romesberg’s team has taken a huge leap forward.

eDNA opens a world of possibilities in terms of unique products for companies synthesizing nucleotides. TriLink would certainly like to add all sorts of future dXTP and dYTP (and ribo versions) to their already extensive offering of modified nucleotides. Hopefully, that won’t be too far in the future. Time will tell.

Laudatory Views and Some Expression of Concern

My quickie survey of quotations in various stories reporting this work by Romesberg’s team, which incidentally included scientists at enzyme-purveyor New England Biolabs, indicated mostly high praise.

For example, a May 7th NY Times story by Andrew Pollack quotes Eric T. Kool, a professor of chemistry at Stanford who is also doing research in the area, as saying “it took some clever problem-solving to get where they got,” adding “it is clear that the day is coming that we’ll have stably replicating unnatural genetic structures.”

A May 9th editorial by Robert F. Service in venerable Science magazine quotes Ross Thyer, a molecular biologist at the University of Texas, Austin, as saying that “this is an amazing enabling technology,” and that the feat opens the way to a universe of new proteins—a vastly more diverse menu of proteins with a wide variety of new chemical functions, such as medicines better able to survive in the body and protein-based materials that assemble themselves.

Despite the resoundingly positive feedback, some concern about eDNA has been voiced. Jim Thomas of the ETC Group, a Canadian advocacy organization, said in an email, “The arrival of this unprecedented ‘alien’ life form could in time have far-reaching ethical, legal and regulatory implications. While synthetic biologists invent new ways to monkey with the fundamentals of life, governments haven’t even been able to cobble together the basics of oversight, assessment or regulation for this surging field.”

As for scary possibilities such as creating an unnatural dangerous organism, the editorial in Science says that creating synthetic “superbacteria” might sound ominous, but Kool thinks the risks are low. “These organisms cannot survive outside the laboratory,” Kool is quoted saying. “Personally, I think it’s a less dangerous way to modify DNA than existing genetic engineering,’ Kool is again quoted as saying.

I’m in the camp of carrying on with this line of research as long as the unnatural base pairs have to be chemically synthesized and “fed” to host organisms in a legitimate lab or manufacturing facility, thus having virtually zero possibility of unintended growth and function.

What do you think?

As usual, your comments are welcomed.

Ripples from the 2013 TIDES Conference

Nucleic acids are getting small, SMARTT, long and weird

Like the “kid in the candy store,” it was a challenge for me to decide which of the many scheduled talks on oligonucleotide therapeutics and nucleic acids diagnostics at the 15th Anniversary TIDES 2013 conference would provide “tasty” blog content. Even more challenging was how to get that content since I couldn’t be at this event in Boston on May 12-15! Continuing the metaphor, “candy” I eventually selected scientific diversity to hopefully satisfy the different ”tastes” of readers.

Jerry Zon’s standup & co-blogger Rick Hogrefe

Jerry Zon’s standup & co-blogger Rick Hogrefe

The solution for how to be there, yet not be there, had me stumped. Fortunately my cleaver colleagues at TriLink got me there as a life-size cardboard standup for a “Bit of Boston” contest (check out all the pictures of me and TIDES attendees in my May 12 post), while Rick Hogrefe (President/CEO) offered to be my eyes and ears, so to speak. Following is Rick’s report to which I’ve added some visuals, links, and comments (in italics).

Getting Small:  Spherical Nucleic Acids Nanoparticle Delivery

Prof. Chad A. Mirkin, Northwestern University and AuraSense Founder

JZ Comments: Prof. Mirkin’s outstanding scientific accomplishments, pioneering nanotechnology research, and list of numerous awards – including the 2013 Linus Pauling Medal Award, recipients of which have frequently (~50%) gone on to receive a Nobel Prize – may be found at the Mirkin Research Group website.


Prof. Chad A. Mirkin – the most cited chemist in the world (past decade, Thomson Reuters) and the most cited nanomedicine researcher in the world.

The normal course of drug discovery usually starts with a drug concept in search of a discovery vehicle.  Rarely does a discovery vehicle create a new drug concept with its own set of remarkable properties.  Such may have been the luck of researchers at the International Institute for Nanotechnology directed by Prof. Mirkin – a man of many talents and head of a research group larger than many biopharmaceutical companies – who has led a nearly two-decade effort delving into the creation and use of nanoparticles combined with oligonucleotides.  His earlier collaborative work, with many others at Northwestern, has led to commercial successes in the diagnostic world and nanofabrication (see Nanosphere and NanoInk for details). However, recent discoveries regarding therapeutic applications of oligonucleotide-functionalized nanoparticles has led to what may be the development of an entirely new class of therapeutic nucleic acids called spherical nucleic acids (SNA).


As depicted above, SNA are comprised of an inorganic (e.g., gold, silver, silica, etc.) inner core and a nucleic acid outer shell. These compact, 13 nm particles exhibit unusual hybridization properties that are largely based on the cooperativity of densely packed strands and high local concentrations.  Transitions from duplex to single strands that normally occur over a 20°C range take place in a 2-8°C range on SNA.

This property was exploited as a diagnostic tool by preparing fluorescent probes Mirkin called “flares” (aka nano-flares) that are hybridized to the oligonucleotides on the gold SNA and thereby quenched.  Because of the very sharp melting transition, they are able to detect with greater accuracy a wider range of sequences compared to existing microarray technologies.  When the SNA oligonucleotides hybridize to the correct target, the tagged probes are released and light up, like a flare, due to unquenching of the fluorophore. The SNA wiki site provides many interesting details about SNA structure, function, applications and societal benefits of this novel class of materials, including gene regulation, molecular diagnostics, intracellular probes, and materials synthesis.


Depiction of flare-based detection (Mirkin et al) wherein multiplex detection of actin and survivin mRNA expression in cells was demonstrated using survivin-Cy5 and actin-Cy3 flares in conjunction changing the relative levels of survivin and actin mRNA in the cells by treatment with actin or survivin targeted siRNA.

Mirkin also reported that SNA have a range of other favorable properties useful for a therapeutic.  SNA-bound oligonucleotides are nuclease resistant and very non-immunostimulatory in comparison to other delivery vehicles.  Perhaps the most remarkable property is the ability for SNA to permeate almost any cell and distribute to most tissues including the brain.  Approximately 1% of the material administered to an animal crosses the blood brain barrier.  They demonstrated the ability of an SNA to reduce the level of Bcl2L12 expression in a mouse glioblastoma model using systemic delivery.  Read more about this line of research on anti-glioma therapeutics by the Mirkin group using novel siRNA-conjugated gold nanoparticles.

The mechanism of delivery was explored at length, with the most likely mode being uptake through caveolae-mediated endocytosis.  Class A scavenger membrane receptors, which are known to bind various polyanionic ligands that include polynucleotides, were found to likely be associated with this active transport.  Down-regulation of class A scavenger receptors resulted in cessation of uptake.  Also revealing was the fact that the nanoparticles required a nucleic acid shell in order to be taken up.  Controls using other conjugates were not taken up by cells.  In my conversation with Dr. Sergei Gryaznov – recently appointed as CTO of AuraSense Therapeutics, a subsidiary of AuraSense that commercializes SNA – he speculated that perhaps a high density of anionic ligands is all that is required in order to achieve uptake.

Regardless of which applications of SNA meet with success, it’s clear that Prof. Mirkin and his team have done a remarkable job at both developing and commercializing a very interesting class of nanoparticles that we will definitely keep an eye on.

Getting SMARTT:  Block-Polymer Delivery

Dr. Mary Prieve, PhaseRx


Mary Prieve, PhD, is Director of Biology at PhaseRx.

PhaseRx, located in Seattle, entered the nucleic acid delivery fray in 2008.  Their lead technology, called the SMARTT Polymer, is a vinyl polymer delivery system that consists of three domains.  The first is the targeting portion of the molecule.  The second is the payload carrier, to which the nucleic acid is conjugated in the examples shown.  The third is designed to facilitate release from the endosome, which is the mode of uptake of the particle.


SMARTT Polymer Technology (courtesy of PhaseRx)

PhaseRx proof-of-concept studies have focused on hepatocellular cancer (HCC).  Two genes, MET, a cell surface receptor, and CTNNB1 (β-catenin), a cell adhesion protein, are often found to be overexpressed in many HCC patients. siRNA sequences were found that specifically down regulated  those genes.  PhaseRx conducted a very eloquent animal study wherein they introduced the MET and CTNNB1 genes into mice using plasmids, demonstrating that co-expression of these genes can induce HCC, and that they can subsequently turn off the effect using their formulated siRNAs.  The result was a significant reduction in liver tumor size and alpha-fetoprotein (AFP) levels. They then modified the MET gene such that the gene still functioned to induce HCC, but did not match the siRNA sequence developed for wild type MET.  They obtained the result they had hoped for:   the modified MET gene was completely uneffected by the complex.

The audience had several interesting questions, one being whether this vinyl polymer is biodegradable.  Dr. Prieve said those studies are currently ongoing.  Another question was about the payload and the number of siRNAs conjugated to each particle, to which the answer was one per particle. According to PhaseRx’s website, it is “deploying an integrated delivery system based on novel synthetic polymers that delivers RNAi drugs into the cytoplasm by mediating their escape from intracellular vesicles called endosomes, where these drugs are often trapped and sequestered. The synthetic polymers exploit natural cellular processes and pH changes to effect endosome escape, delivering RNAi drugs into the cytoplasm where they can reach and inhibit the desired target of interest. The PhaseRx system offers significant advantages: effective intracellular RNAi delivery, broad applicability, and consistency with pharmaceutical development standards.”

Getting Long:  Targeting Long Non-Coding RNA

Dr. Jim Barsoum, CSO, RaNA Therapeutics


Jim Barsoum, PhD, has over 25 years of experience in biotech at Biogen Synta, and then Theracrine.

RaNA Therapeutics is focusing on the latest in the RNA craze, long non-coding RNA (lncRNA).  The presentation started with a much needed beginners course on lncRNA and what are some of its functions (Wikipedia has some good resources as well if you’re interested). Naturally, new findings will most likely indicate that much of these proposed functions are completely wrong, and that lncRNA operates in some other fashion, but that RaNA’s core technology still works (read on…).  Call that the cynicism of someone who follows the slow unveiling of the roles of microRNAs in the cell.  RNA biology is perhaps at one of its most exciting times.  The range of tools available today for the RNA scientist from commercial sources is remarkable, and “RNA-ologists” are exploiting these tools to make great strides forward.

One new fact that I took home is that lncRNA is often 5’ capped and 3’ polyadenylated, just like mRNA, however they are very poorly expressed, if at all.  The reason for this was not given.  One possibility that came to my mind was that perhaps lncRNAs are degraded by the same mechanism as mRNA.  Another possibility is that they require circularization to function using the same proteins that bind 5’ cap to the 3’ poly-A tail that is required by mRNA in order to be translated. Another interesting fact is that of all the RNA that is transcribed, only ~1% is mRNA for the production of protein.  There is a lot of unknown science in the other ~99%.

JZ Comments: To put Rick’s further remarks in context, I’ve included the mechanism from RaNA’s website for long noncoding RNA.  The description for the mechanism says, “RaNA’s proprietary technology upregulates the expression of desirable genes that can prevent or treat disease. RaNA’s approach operates epigenetically and reverses the endogenous repression of gene expression. RaNA’s therapeutic oligonucleotide can be administered as a subcutaneous injection in saline, and is then taken up by cells in most tissues of the body, crossing the endosome membrane to enter the cytoplasm and nucleus. When transcription of the unique lncRNA target reveals the PRC2 binding domain the RaNA antagonist therapy binds the lncRNA, which blocks PRC2 recruitment and allows for transcription to proceed resulting in mRNA upregulation.”


RaNA is exploiting one function of lncRNA which is the control of transcription in a cell.  The lncRNA silences genes by recruiting transcriptional repressor proteins, such as polycomb repressor 2 (PRC2) by binding to the target gene in a standard hybridization mode with part of lncRNA and to the protein with another part of the molecule in more of an aptamer fashion.  They found they can upregulate expression by blocking the region of the lncRNA that binds to the protein using a short DNA oligonucleotide comprised of 50% LNA.  Their initial target is the binding site for PRC2 on the lncRNA that appears to control transcription of the erythropoietin (EPO) gene.  They demonstrated they were able to upregulate the production of EPO. Preliminary data showing a 25-fold upregulation of Erythrid Krüoppel-like Factor (EKLF; aka KLF-1) in mice was presented as well.

We expect that the story on lncRNA is just beginning and has the potential to far surpass siRNA because of what may soon be found to be a vast number of different functions in the cell.

RaNA Therapeutics was co-founded by Art Krieg, MD, who has more than 20 years of experience in immune stimulatory CpG oligonucleotide R&D in academia, and co-founded Coley Pharmaceutical Group (acquired by Pfizer) based on “CpG-ology,” as well as co-founded the first antisense journal, Oligonucleotides, and the Oligonucleotide Therapeutics Society. According to the RaNA Therapeutics website, “the capitalized “R”, “N”, and “A” in RaNA spell out RNA, the type of nucleic acid underlying our approach.  The lowercase “a” represents activation.  Together the letters stand for RNA activation, the very basis of our drug development approach.”

Getting Weird:  Unnatural Base Pairs

Dr. Ichiro Hirao, RIKEN and President/CEO, TAGCyx Biotechnologies


As is usual for TIDES, there was very little “hard core” chemistry at this meeting.  However, of the few chemistry talks, one of the more interesting was that given by Dr. Hirao of RIKEN.  His research has been focused on the development of unnatural base pairs.  He described the base pair that appears to be the winning combination, which consists of a purine analog (Ds) and a pyrimidine analog (Px) that can bind to each other with high affinity but not to the natural bases.


Taken from Hirao and coworkers NAR 2012.

The authors demonstrated that these unnatural bases can also get incorporated by various polymerases with high fidelity.  Misincorporation of the wrong base against these unnatural bases was only 0.005%, which is only slightly higher than the misincorporation rate of 0.002% for standard bases.  The utility of the Ds base as a novel base for aptamer selection was described.  Several random libraries that contained from one to three substitutions with the Ds base at specific locations were prepared.  The oligos did not contain any Px bases, thus forming unusual structures that were not found in the natural sequence.  This opened up the structural space for aptamer selection greatly.  Using an ingenious method that substituted the Px base pair with a less specific base, Pa allows misincorporation of either dA or T at that site.

According to, RIKEN was founded in 1917 and has ~3000 scientists on seven campuses across Japan, the main one in Wako, just outside Tokyo. RIKEN is an Independent Administrative Institution whose formal name in Japanese is Rikagaku Kenkyūjo and in English is The Institute of Physical and Chemical Research. RIKEN conducts research in many areas of science ranging from basic research to practical applications. It is almost entirely funded by the Japanese government, and its annual budget is approximately $760 million.

JZ Comments: I wish to thank Rick Hogrefe for being a guest blogger and providing  content!  As always, I welcome reader comments!