Article pubs.acs.org/joc
Cite This: J. Org. Chem. 2018, 83, 9774−9786
Modular Strategy To Expand the Chemical Diversity of DNA and Sequence-Controlled Polymers Donatien de Rochambeau,† Yuanye Sun,† Maciej Barlog,‡ Hassan S. Bazzi,‡ and Hanadi F. Sleiman*,† †
Department of Chemistry, McGill University, 801 Sherbrooke St. West, Montréal, Québec H3A 0B8, Canada Department of Chemistry, Texas A&M University at Qatar, P.O. Box 23874, Doha, Qatar
‡
Downloaded via UNIV OF SOUTH DAKOTA on September 8, 2018 at 06:58:16 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
S Supporting Information *
ABSTRACT: Sequence-defined polymers with customizable sequences, monodispersity, substantial length, and large chemical diversity are of great interest to mimic the efficiency and selectivity of biopolymers. We report an efficient, facile, and scalable synthetic route to introduce many chemical functionalities, such as amino acids and sugars in nucleic acids and sequence-controlled oligophosphodiesters. Through achiral tertiary amine molecules that are perfectly compatible with automated DNA synthesis, readily available amines or azides can be turned into phosphoramidites in two steps only. Individual attachment yields on nucleic acids and artificial oligophosphodiesters using automated solid-phase synthesis (SPS) were >90% in almost all cases. Using this method, multiple water-soluble sequence-defined oligomers bearing a range of functional groups in precise sequences could be synthesized and purified in high yields. The method described herein significantly expands the library of available functionalities for nucleic acids and sequence-controlled polymers.
■
INTRODUCTION The precise molecular recognition and folding properties of nucleic acids and proteins derived from their primary sequence give them central roles in biological systems. Driven by the goal of mimicking the structures and functions of these biopolymers,1 interest in sequence-controlled polymers (SCP) has grown tremendously in the past few years.2 Digital storage molecules,3,4 foldamers−artificial oligomers with highly ordered secondary structures,5−8 and single-chain nanoparticles that fold into protein-like structures9−15 are some of the applications envisaged by these studies. For these purposes, a strategy allowing the synthesis of monodisperse oligomers with a large number of chemical monomer variants is of great interest. Novel methods, such as iterative exponential growth16 and other elegant approaches17−20 have been introduced to make sequence-controlled polymers in a scalable manner. However, solid-phase synthesis remains the method of choice for the most precise sequence control.21−25 Automated solid-phase phosphoramidite chemistry, in particular, has shown exceptional coupling yields for the synthesis of DNA and RNA.26,27 Decades of optimization have allowed it to attain the highest degrees of polymerization (DP) for a solid-phase synthesis. Up to 200 monomer-long oligonucleotides can be made in good yields and with simple purification methods.28 Importantly, the cost of phosphoramidite synthesis has been significantly and steadily declining, making it a practical as well as powerful strategy.29 © 2018 American Chemical Society
We have developed an efficient strategy to make sequencedefined polymers based on automated phosphoramidite solidphase synthesis.30−34 This method allowed the formation of polyphosphodiesters35 that are stable and highly soluble in water due to their anionic nature. Other groups subsequently showed the synthesis of very long artificial sequence-controlled polyphosphodiesters (DP of 100)36 and reported a method for post-synthesis dual functionalization.37 Phosphoramidite chemistry has also been used to make “oligopyrenotides” with novel supramolecular assembly properties.38,39 Thus, monodispersity, cost-efficiency, length of the oligomers, and perfect sequence-control have been achieved using automated solid-phase synthesis with phosphoramidite chemistry. The next goal for this method is to expand the library of available compatible phosphoramidites in a straightforward and scalable manner. Solid-phase synthesis on a DNA synthesizer requires the phosphoramidite monomers to stay unaltered to a great number of chemical conditions (repetitive treatment with oxidant and mild acid and final deprotection in aqueous base). For the purpose of versatile and multiple functionalization of synthetic oligophosphodiesters, even more restrictive conditions apply, making the task arduous. (i) Due to the need for numerous monomers, their synthesis must be fast, costeffective, and scalable. This notably prevents the use of Received: May 22, 2018 Published: August 20, 2018 9774
DOI: 10.1021/acs.joc.8b01184 J. Org. Chem. 2018, 83, 9774−9786
Article
The Journal of Organic Chemistry
Figure 1. Synthesis of sequence-controlled oligophosphodiesters made of monomers based on the novel tertiary amine backbone. The three moieties shown on the oligophosphodiester (phenylalanine, β-D-glucose, and alkyne) were used as examples. CEP = cyanoethylphosphoramidite, DMT = dimethoxytrityl.
Scheme 1. Different Backbones Exploreda
a
Molecules with a prime (e.g., 1′ and 2′) designate the DMT derivatives before conversion to the phosphoramidite.
nucleoside derived phosphoramidites.40 (ii) Very high coupling yields, leading to high DPs, are essential. As an example, a 22mer would be obtained in 95%),31 whereas molecule 1′ consistently led to unsatisfactory coupling yields followed by spontaneous degradation of the DNA strand in water at 4 °C after a few days. We hypothesized that the nucleophilic nitrogen atom in the case of molecule 1′ can attack an electrophilic phosphorus atom 9775
DOI: 10.1021/acs.joc.8b01184 J. Org. Chem. 2018, 83, 9774−9786
Article
The Journal of Organic Chemistry forming a favorable five-membered ring (Figure S1). On the other hand, the fluorine atoms in 2′ exert an inductive effect that reduces the nitrogen lone pair nucleophilicity, thus possibly explaining the good yields obtained with this molecule. As further evidence for this mechanism, a phosphoramidite bearing an anthraquinone moiety directly attached to the diethanolamine nitrogen atom (thus also reducing its nucleophilicity) was reported elsewhere and led to good yields.52 This observation led to further monomer optimization. Molecules 3, 4, and 5 were used to check backbone suitability. For 3 and 4, the previous degradation mechanism is less likely to happen because they do not have a nucleophilic nitrogen, while 5 has a longer spacer separating the nucleophilic nitrogen from the phosphorus, thus avoiding the five-membered ring intermediate in this degradation mechanism. 3 was made following a straightforward synthesis: Myristoyl chloride was added onto diethanolamine leading to an amide in quantitative yields, which was transformed to phosphoramidite 3 following standard procedures (Scheme S2). 4 was made from the precursor for 1′ using methyliodide, followed by DMT protection and phosphoramidation using standard procedures (Scheme S3). Finally, bromohexanol was substituted onto hexadecylamine, leading to compound 5′ further transformed to 5 (Scheme S4). These monomers were coupled to the 5′ end of a DNA strand as a first step. Molecule 3 did not lead to good yields (
