7.7. Solid-Phases Synthesis of Nucleic Acids
Although there are enzyme-based techniques for making additional copies of pre-existing sequences of DNA/RNA, generating a specific sequence from scratch needs a different approach. Moderately sized nucleic acid chains of a few hundred to a few thousand subunits are often required for research or pharmaceutical purposes. Synthesizing these requires dozens or hundreds of chemical reactions performed in sequence. Because they are frequently needed and tedious to prepare an alternative system for nucleic acid synthesis was developed. This method can be done manually but is very often automated instead. The approach is similar to solid-phase synthesis of peptide polymers (see Section 4.6.4) but has several notable differences.
As with solid-phase peptide synthesis, the main goal of the alternative approach is to make the steps much faster.
As with solid-phase peptide synthesis, the speed of setting up the reactions is improved by making all of the reagents standardized and commercially available. Protected versions of all of the nucleotides, where the protecting groups have already been carefully chosen and added, are synthesized on industrial scale and purchased by chemists. The speed of setup is often also improved by automation.
In principle an activator compound can be used to combine two monomers. In peptide synthesis this is dicyclohexylcarbodiimide (DCC). In practice, most methods to synthesize nucleic acid polymers have the ‘activator’ built-in to the nucleotide monomers. During the reaction a functional group on the monomer changes and becomes a powerful electrophile, which allows one monomer to attach to the next.
As with solid-phase peptide synthesis, the speed of working up (purifying) the product after each step is improved by physically attaching the nucleic acid chain to an object. There are a large variety of items made for this purpose. The most commonly used are beads made from synthetic polymers or glass that have useful functional groups on the surface.
As with solid-phase peptide synthesis, technically some methods and machines are proprietary and use different protecting groups and/or solid supports. The specific details may vary slightly but the overall approach is always the same.
The monomers are purchased with protecting groups already attached to the nucleotides. Each of the common nucleotides is available (Figure 7.24). In all cases there are 2 to 3 modifications: a protecting group for the nucleobase (if needed), a protecting group for the alcohol at position 5 of the saccharide, and a protecting/activating group for the “phosphate” attached to the alcohol at position 3 of the saccharide.
Figure 7.24 – Generalized Structures of Protected Monomers for DNA Solid-Phase Synthesis.
With respect to the nucleobases, only the amine functional groups require protection. The nitrogen atoms in the heterocyclic rings and the carbonyls are functionally non-reactive under the conditions of the polymerization.
In reality, synthesizing the protected monomers is an intensive process that requires both preparation and planning. The order that the protecting groups are added to the compounds, the specific conditions used to add them, and the purification methods all require consideration and vary depending on which monomer is being synthesized. These details are omitted from this text for brevity.
For synthesis of RNA polymers an additional protecting group is needed for the alcohol at position 2 of the saccharide (Figure 7.25). There are several possible protecting groups for this position. The only important consideration is that the group must be orthogonal to all of the other protecting groups. This text will focus on solid-phase synthesis of deoxyribonucleic acid polymers. The specific details of the additional protecting group needed for RNA synthesis is not required but understand that the approach is the same.
Figure 7.25 – Generalized Structure of Protected Monomers for RNA Solid-Phase Synthesis.
7.7.1.1. The Benzoyl (Bz) Protecting Group
The benzoyl (Bz) protecting group is a slightly uncommon choice for protecting amines (Scheme 7.11). This converts the functional group from an amine (moderate nucleophile, moderate base) to an amide (very poor nucleophile, poor base). While it is not necessary to do so, many sources use the abbreviation in place of drawing the protecting group in schemes/figures.
Scheme 7.11 – Generalized Reaction Equations for Adding and Removing the Benzoyl (Bz) Protecting Group to an Amine Using Adenine as an Example.
The benzoyl (Bz) protecting group is occasionally confused with the benzyl (Bn) protecting group (Figure 7.26). Although the names sound similar, the two groups are very different: benzoyl protecting groups form carbonyl-containing functional groups (e.g. an amide) and benzyl protecting groups do not. Benzyl protecting groups are not commonly used for amines.
Figure 7.26 – Comparison of Benzoyl (Bz) and Benzyl (Bn) Protecting Groups.
Bz protecting groups can be added several different ways. By far the most common approach is to add it using benzoyl chloride (sometimes written as BzCl; Scheme 7.12). As with other syntheses of amides from acid halides (see Section 4.6.1.2) a non-nucleophilic base is added to react with the acid side-product. The base is almost always a tertiary amine, typically with more than one equivalent. The mechanism for adding this protecting group is identical to the mechanism for formation of an amide using an acid halide.
Scheme 7.12 – Simplified Reaction Mechanism for Addition of the Benzoyl (Bz) Protecting Group to Adenine Using Benzoyl chloride (BzCl) and Triethylamine.
Bz protecting groups are stable to both moderately acidic and basic conditions. They can be removed several different ways. A semi-common approach is to remove it using ammonia in water (sometimes called “aqueous ammonia”; Scheme 7.13). Aqueous ammonia has a special reactivity: the ammonia is much more basic and nucleophilic than it would otherwise be. The mechanism for removing this protecting group using aqueous ammonia is a standard addition-elimination with a proton transfer.
Scheme 7.13 – Simplified Reaction Mechanism for Removal of the Benzoyl (Bz) Protecting Group from Adenine Using Aqueous Ammonia (NH3).
7.7.1.2. The Dimethoxytrityl (DMT) Protecting Group
The dimethoxytrityl (DMT) protecting group is an uncommon choice for protecting alcohols (Scheme 7.14). This converts the functional group from an alcohol (moderate nucleophile, moderate base, moderate acid) to an ether (poor nucleophile, poor base, non-acid). While it is not necessary to do so, many sources use the abbreviation in place of drawing the protecting group in schemes/figures. In the case of DMT because the structure is very large this is often necessary for speed/ease of drawing.
Scheme 7.14 – Generalized Reaction Equations for Adding and Removing the Dimethoxytrityl (DMT) Protecting Group to an Alcohol Using Deoxyribofuranose as an Example.
The dimethoxytrityl (DMT) protecting group is a special modification of the trityl (Tr) protecting group (Figure 7.27). Trityl protecting groups are much more commonly used in other chemical reactions. They are simple but slow to remove. The ethers make removal of the protecting group faster. This increases the overall speed of polymer synthesis. It also makes the conditions required for removal slightly less harsh (see below).
Figure 7.27 – Comparison of Dimethoxytrityl (DMT) and Trityl (Tr) Protecting Groups.
DMT protecting groups can be added several different ways. By far the most common approach is to add it using dimethoxytrityl chloride (sometimes written as DMTCl or DMT-Cl; Scheme 7.15). A non-nucleophilic base is added to react with the acid side-product. The base is often a tertiary amine or pyridine. The mechanism for adding this protecting group is a standard SN1 nucleophilic substitution.
Scheme 7.15 – Simplified Reaction Mechanism for Addition of the Dimethoxytrityl (DMT) Protecting Group to Ribofuranose Using Dimethoxytrityl chloride (DMT-Cl) and Triethylamine.
DMT protecting groups are stable to moderately basic conditions. The most common approach to remove it uses a moderately strong acid (Scheme 7.16). It is possible to use strong acids (e.g. hydrochloric acid (HCl) or trifluoroacetic acid (CF3CO2H, sometimes abbreviated as TFA)). However, these strong acids may cause small amounts of side reactions to occur. When the reaction is performed many times, such as in polymer synthesis, even a small number of side reactions can exponentially lower the overall yield over time. Instead, use of related trichloroacetic acid (CCl3CO2H, sometimes abbreviated as TCA) is common. While this acid is still relatively strong (pKa = 0.6) it is significantly less strong (viz. harsh) than TFA. As a result, undesired side reactions are minimized. The mechanism for removing this protecting group is similar to an E1 elimination.
Scheme 7.16 – Simplified Reaction Mechanism for Removal of the Dimethoxytrityl (DMT) Protecting Group from Ribofuranose Using Trichloroacetic Acid.
7.7.1.3. The β-Cyanoethyl Protecting Group (and Why a Phosphoramidite is Used)
The final group that requires protection is the “phosphate” attached to the oxygen at position 3 of the saccharide. However, this group also contains a built-in activator for combining monomers to build the polymer. As a result, it is more complex. This section includes a brief discussion of the functional group used as an activator and the protecting group placed on it.
In place of an actual phosphate functional group the phosphorous is incorporated into the monomers as part of an unusual functional group, sometimes referred to as a “phosphoramidite” or “phosphoramidite ester” (Figure 7.28).
Figure 7.28 – Comparison of Select Phosphorous-Containing Functional Groups.
This functional group is similar to a phosphite rather than a phosphate. Although unusual, this functional group allows the phosphorous atom to become electrophilic under certain conditions (see Section 7.7.2). This is what allows combining monomers to form the nucleic acid polymer. The specific alkyl groups on the nitrogen are preferred only because it is easily synthesized and the side product after coupling is unreactive; there is no special reactivity or reason for using isopropyl groups beyond convenience.
To avoid chemoselectivity problems when attaching a new monomer (see Section 7.7.2) the oxygen of the phosphoramidite requires a protecting group (Figure 7.29).
Figure 7.29 – Comparison of Unprotected and Protected Phosphoramidites of Deoxyribofuranose.
The beta-cyanoethyl (β-cyanoethyl) protecting group is a very rare choice for protecting OH’s in functional groups (Scheme 7.17). Use of this protecting group is functionally exclusive to nucleic acid polymer synthesis methods. This converts the functional group from an “OH” (moderate nucleophile, moderate base, moderate acid) to an “OR” (poor nucleophile, poor base, poor acid). Because this protecting group is so rare there is no abbreviation for this protecting group.
Scheme 7.17 – Generalized Depiction of a Nucleoside Phosphoramidite and Reaction Equation for Removing the Beta-Cyanoethyl (β-Cyanoethyl) Protecting Group.
Because the beta-cyanoethyl protecting group is so rare there is no standard way of adding this protecting group. The group itself is added to the monomers as the phosphoramidite functional group is being formed rather than protecting a pre-existing “OH”. The exact reactions used to form the phosphoramidite are interesting but not relevant for discussion.
Beta-cyanoethyl protecting groups are stable to both moderately acidic and basic conditions. The traditional approach to remove it uses ammonia in water (sometimes called “aqueous ammonia”; Scheme 7.18). Aqueous ammonia has a special reactivity: the ammonia is much more basic and nucleophilic than it would otherwise be. The mechanism for removing this protecting group using aqueous ammonia is a standard E2 elimination followed by a proton transfer. To avoid confusion the beta-cyanoethyl group is shown being removed from a generalized group (the phosphoramidite undergoes several chemical reactions before removal of the protecting group).
Scheme 7.18 – Generalized Reaction Mechanism for Removal of the Beta-Cyanoethyl (β-Cyanoethyl) Protecting Group Using Aqueous Ammonia (NH3).
7.7.2. Steps for Solid-Phase Synthesis of Nucleic Acids
First, the 3’-terminus of the target nucleic acid is covalently attached to the solid support. There are many different ways to do this. Often the support is coated in one (or more) compounds with functional groups that may form bonds with the alcohol at position 3 of the saccharide. One of the common approaches is to form an ester (Scheme 7.19). There are, again, many ways to do this that vary depending on the manufacturer of the system. In some cases DCC is used to form the ester. The mechanism is identical to amide formation (see Scheme 4.23) but with an alcohol instead of an amine. As a result, this step requires a modified monomer without the phosphoramidite but with the other protecting groups. For convenience a single (small) bead with a single polymer chain is shown. In reality the beads are relatively large and their surface is covered in many nucleic acid chains.
Scheme 7.19 – Generalized First Step of Solid Phase Synthesis of a Target Nucleic Acid: Attachment to Solid Support.
As with solid-phase peptide synthesis, attaching the nucleotides this way means that the chain will be added to “backwards” (start with the last nucleotide, then add the second-last, then the third-last, etc.).
Second, the polymer is added to (Scheme 7.20). This phase also has chemical reactions to modify the phosphoramidite group (see below). The bead is washed in between each step with solvent to remove any unreacted materials and side products. First, the alcohol’s protecting group is removed. Then, a new monomer is attached. This step involves a weak acid. There are, again, many ways to do this that vary depending on the manufacturer of the system. In most cases the acid chosen is tetrazole (or similar). This uncommon acid is preferred because its conjugate base is non-nucleophilic and, more importantly, it is easily removed during washing.
Scheme 7.20 – Generalized First Half of the Second Step of Solid Phase Synthesis of a Target Nucleic Acid: Chain Growth.
The mechanism for adding a new monomer involves formation of an electrophilic phosphorous (Scheme 7.21). The nitrogen of the phosphoramidite (base) removes a proton from tetrazole (acid). This greatly increases its leaving group ability. A lone pair of the oxygen forms a π bond with the carbon, ejecting the leaving group (amine). The oxygen of the alcohol (nucleophile) attacks the phosphorous (electrophile). Finally, the amine removes a proton to generate the final phosphite product.
Scheme 7.21 – Generalized Reaction Mechanism for Combining Two Monomers Using an Acid (Tetrazole) and the Phosphoramidite Group.
Unfortunately, the phosphite group is slightly nucleophilic and can cause chemoselectivity issues in subsequent steps. Before the next monomer is added it must be oxidized to a phosphate (Scheme 7.22), which does not have these issues. There are, again, many ways to do this that vary depending on the manufacturer of the system. In most cases the oxidation involves a dihalide, often iodine (I2). The mechanism for this oxidation is similar to the oxidation of alcohols to aldehydes using bromine (Br2) and water (see Oxidation of Alcohols via Elimination) but slightly more complicated/longer. The specific details of this mechanism are not relevant. Then, the deprotection, addition of a new monomer, and oxidation steps are repeated as each nucleotide in the chain is added to the nucleic acid.
Scheme 7.22 – Generalized Second Half and Repetition of the Second Step of Solid Phase Synthesis of a Target Nucleic Acid: Chain Growth.
Finally, all of the protecting groups are removed and the nucleic acid is detached from the solid support (Scheme 7.23).
Scheme 7.23 – Generalized Third Step of Solid Phase Synthesis of a Target Nucleic Acid: Removal of Protecting Groups and Detachment from Solid Support.
The DMT protecting group (on the last nucleotide that was added) is removed using the standard procedure. Then the polymer is detached. There are several different ways to do this. Most systems use ammonia in water (aqueous ammonia). This breaks the ester through an addition-elimination mechanism. As with solid-phase peptide synthesis, the protecting groups (Bz and β-cyanoethyl) are also removed by these conditions. In a single step the covalent bead-nucleic acid attachment is broken and all of the protecting groups of the nucleotides are removed (Scheme 7.24).
Scheme 7.24 – Example of Third Step of Solid Phase Synthesis of a Target Nucleic Acid with Concurrent Removal of Protecting Groups.
