Vanderbilt Institute of Chemical Biology



Discovery at the VICB







Discovery of a New Functional Group 


By: Carol A. Rouzer, VICB Communications
Published:  April 11, 2019


The HaloAminoNitroAlkane (HANA) functional group plays a key role as an intermediate in umpolung amide synthesis and related reactions.


Organic chemists, who study the myriad of carbon-containing compounds in the universe, see the world through the lens of the functional group (Figure 1). Functional groups are structurally distinct clusters of atoms that exhibit characteristic chemical behavior. By identifying the functional groups in any given molecule, the organic chemist can design a scheme to synthesize that molecule and predict its behavior in interactions with other compounds. Indeed, it is the understanding of functional groups that enables us to make sense of the infinite number of possible organic molecules, including the complex macromolecular proteins, nucleic acids, and carbohydrates found in biological systems. Considering their importance to the field of organic chemistry, it is hard to imagine that there could be any functional groups that have not yet been discovered. Now, however, Vanderbilt Institute of Chemical Biology member Jeff Johnston and his lab report the isolation and characterization of the HaloAminoNitroAlkane (HANA), a new functional group that serves as an intermediate in an unusual synthetic pathway [M.S. Crocker, et al. (2019) Chem, published online March 28,].



FIGURE 1. Examples of organic functional groups.



The Johnston group's studies actually began with their interest in a very common and well-known functional group, the amide (Figure 1). Amides are present in a wide range of natural and synthetic organic molecules. The peptide bonds that join the amino acid subunits of proteins are amides. Thus, organic chemists have long been interested in efficient methods for amide synthesis, a process that traditionally involves the chemical reaction of a carboxylic acid (or an electron-deficient derivative of a carboxylic acid) and an amine (Figure 2A). In this reaction, the amine group serves as a nucleophile, donating a pair of electrons to the carbon of the carboxylic acid, which as an electron acceptor, serves as an electrophile. Although used successfully for years, this approach has disadvantages, including the requirement for harsh conditions that can actually damage the reactants or products, leading to a reduction in yield. To address this problem, the Johnston group introduced a scheme for umpolung amide synthesis (UmAS), which involves the reaction of an α-bromo nitroalkane with an amine (Figure 2B). In this reaction, N-iodosuccinimide (NIS) first reacts with the amine to yield an N-iodoamine (or a similarly electrophilic amine). A base (frequently potassium carbonate) is also present in the reaction mixture to abstract a proton from the α-bromo nitroalkane to form a nitronate. The nitronate then acts as a nucleophile to donate electrons to the nitrogen of the N-iodoamine, forming a putative tetrahedral intermediate (TI) (shown in red in the figure). Under aerobic conditions, subsequent reaction with oxygen leads to loss of BrONO2 and formation of the desired amide. The reaction earns its name "umpolung" due to the fact that the polarity of electron transfer is reversed, with the nitrogen of the amine now serving as electrophile rather than nucleophile, as occurs in traditional amide synthesis. The reaction proceeds with high yield under very mild conditions, offering distinct advantages for the synthetic chemist.



FIGURE 2. (A) General scheme for traditional amide synthesis using a carboxylic acid and an amine. In this case, the reaction is acid-catalyzed to increase the electrophilicity of the acid. (B) General scheme for UmAS. A base (B:) abstracts a proton from the α-bromo nitroalkane, which then rearranges to a nucleophilic nitronate. N-iodosuccinimide reacts with the amine to generate an electrophilic N-iodoamine. Electron donation from the nitronate to the nitrogen of the N-iodoamine yields the putative tetrahedral intermediate (red), which then reacts with oxygen (under aerobic conditions) to form the amide.



Having established the usefulness of UmAS, the researchers next wanted to understand the reaction mechanism - the step-by-step process of bond making and breaking that leads from reactants to products. Their initial studies supported the hypothesis that UmAS proceeds through formation of the TI as shown in Figure 2B, but despite repeated attempts, they were unable to clearly demonstrate its existence, a failure that called the proposed mechanism into question. The TI is of particular interest because, if it exists, it would be the first example of a new HANA functional group (Figure 3). Notable in HANA is the presence of an electron withdrawing group (nitro), an electron donating group (amine), and a leaving group (halogen) all attached to the same carbon. Such a concentration of complementary reactive groups is likely to result in considerable chemical instability, making the TI too short-lived to be observed. To address this conundrum, the investigators began to search for ways to modify the UmAS reaction in order to generate a more chemically stable HANA-containing TI.



FIGURE 3. The HANA functional group.



To accomplish their goal, the researchers first replaced the α-bromo nitroalkane reactant with an α-fluoro nitroalkane. As carbon-to-fluorine bonds are generally stronger than carbon-to-bromine bonds, they hypothesized that this change would lead to a more stable TI. Next, they began searching for compounds that contain an electrophilic nitrogen to serve as alternatives to the N-iodoamine. They found success through the use of dialkyl azodicarboxylates, which, when reacted with an α-fluoro nitroalkane, should yield an imide product through formation of a tetrahedral intermediate (TI׳) very similar in structure to that of the TI proposed in the UmAS reaction mechanism (Figure 4A). Indeed, trial reactions using a series of aryl-substituted α-fluoro nitroalkanes and diethyl azodicarboxylate (DEAD) yielded the expected TI׳ products (identified by NMR spectroscopy) when carried out at room temperature. N-iodosuccinimide was not required, and 4-dimethylaminopyridine (DMAP) was used as the base. The reaction of the α-fluoro nitroalkane bearing a chlorophenyl substituent (X = Cl in Figure 4B) with diisopropyl azodicarboxylate (DIAD) provided a TI׳ of sufficient stability that the investigators were able to crystallize it and verify its structure via X-ray crystallography. Thus, the existence of the HANA functional group was confirmed for the first time.


Figure 4. (A) General scheme for the reaction of an α-fluoro nitroalkane with a dialkyl azodicarboxylate to form an imide via a tetrahedral intermediate. (B) Examples of reactions carried out in this study. The tetrahedral intermediates in all cases are shown in red.



In further studies, the researchers found that each TI׳ obtained from the incubation of DEAD with one of their aryl-substituted α-fluoro nitroalkanes reacted at high temperature to yield the expected imide, confirming that these compounds could serve as an intermediate in a reaction highly similar to UmAS. As in the case of UmAS, water facilitated this reaction, but unlike UmAS, exclusion of oxygen had no effect. Results of kinetic studies of the TI׳-to-imide conversion indicated that the reaction was strongly influenced by the presence of electron-withdrawing substituents on the phenyl ring, and it likely involved the generation of a positive charge at a critical point during the mechanism.


The data generated to this point indicated distinct similarities, but some differences between the reaction of aryl-substituted α-fluoro nitroalkanes with dialkyl azodicarboxylates and UmAS. A key question facing the investigators was the source of the oxygen that replaces the fluoro and nitro groups in TI׳ during formation of the final imide product. In the case of UmAS, prior studies indicated that the amide oxygen comes from the nitro group under anaerobic conditions and atmospheric oxygen under aerobic conditions. Yet, the finding that elimination of oxygen had no effect on imide formation suggested some potential differences in mechanism. To address this question, the researchers first hypothesized that the oxygen comes primarily from the nitro group via nitro-nitrite isomerization (Figure 5). To test this hypothesis, they synthesized an aryl-substituted α-fluoro nitroalkane uniformly labeled with 18O in the nitro group. Following reaction with DEAD, they discovered that 43% of the imide product contained the 18O label,  suggesting that some, but not all of the oxygen is derived from the nitro substituent. When they included 18O-labeled water in the same reaction mixture, the percent of incorporated isotope increased to 69%, supporting the hypothesis that some oxygen could also be supplied by water, but leaving open the possibility that there are additional sources. These results led the researchers to consider several possible fates for TI׳ in the reaction (Figure 5). Nitro-nitrite isomerization can occur through homolytic cleavage of the carbon-to-nitrogen bond, resulting in formation of free radicals which then react with each other to yield the nitrite. Prior data suggested that homolytic cleavage of the TI occurs in UmAS. However, heterolytic cleavage of the carbon-to-nitrogen bond is also possible, yielding a carbocation and nitrite ion. In this case, these two species can react with each other to yield the nitrite, the carbocation can react with water, or the molecule can cyclize through a reaction with one of the carbamate oxygens. These three options lead to retention of oxygen from the nitro group, incorporation of oxygen from water, or acquisition of oxygen through a third source, respectively. This proposed mechanism is consistent with the labeling studies, as well as with the results of the kinetic data that supported the generation of a positive charge during the reaction. It also explains why water, but not molecular oxygen facilitates imide formation in this case.




Figure 5. Possible fates of TI׳. Homolytic cleavage of the C-N bond generates free radicals that can then react with each other to produce the corresponding nitrite. Alternatively nitro-nitrite isomerization can also occur through heterolytic cleavage of the C-N bond yielding a carbocation and nitrite ion that react with each other. The carbocation can also react with water to incorporate oxygen or undergo cyclization to produce an intermediate that can subsequently decompose to yield the imide product.



To further explore the mechanism by which the TI׳ converts to the imide, the investigators employed computational approaches with collaborators at Brock University. They evaluated the energetics of four potential pathways by which the conversion might occur. These were: 1) concerted nitro-nitrite isomerization; 2) heterolytic cleavage of the carbon-to-fluorine bond; 3) homolytic cleavage of the carbon-to-nitrogen bond; and 4) multi-step nitro-nitrite isomerization via carbocation formation (Figure 6). The results indicated that pathway 4 offered the most energetically favored mechanism, providing further support for the hypothesis that the TI׳-to-imide conversion pathway includes carbocation formation. The researchers noted that pathway 4 includes transition state intermediates in which a carbamate N-H plays a role in nitro-nitrite isomerization. This is not possible in the classic UmAS mechanism, as no carbamate groups are present in TI. Thus, there is reason to hypothesize that the conversion of TI to the amide in UmAS and TI׳ to the imide as studied here will happen via somewhat distinct mechanisms.




Figure 6. Possible pathways for the conversion of TI׳ to the imide product. Shown are the four pathways investigated computationally, including: 1) concerted nitro-nitrite isomerization; 2) heterolytic cleavage of the carbon-to-fluorine bond; 3) homolytic cleavage of the carbon-to-nitrogen bond; and 4) multi-step nitro-nitrite isomerization via carbocation formation. Only selected intermediates/transition states are shown for pathways 1 and 4, but in every case, the transition state representing the highest energy barrier is included. Values (in kcal/mol) provided indicate the difference in energy between each species and the TI׳ starting material. Note that pathway 4 is the only one that does not include a transition state with an energy difference >30 kcal/mol.



In a final set of experiments, the researchers noted that the reaction of an α-fluoro nitroalkane and a dialkyl azodicarboxylate as studied here bears some similarity to the previously reported reaction of an α-bromo nitroalkane and an acyl hydrazide to form a 1,3,4-oxadiazole (Figure 7A). The latter reaction, referred to as diverted UmAS, is believed to occur via a tetrahedral intermediate formed by the same general mechanism as seen in UmAS. However, the intermediate in this case cyclizes to yield the oxadiazole rather than producing a diacyl hydrazide. As in the case of UmAS, the TI in diverted UmAS has not been identified, leading the researchers to explore the reaction of an aryl-substituted α-fluoro nitroalkane with ethyl acyl hydrazide (Figure 7B). Although the reaction did not enable observation of the predicted tetrahedral intermediate, the researchers identified a fluoro hydrazone that was likely derived by loss of the nitro group from the intermediate. Upon longer incubation, the fluoro hydrazone was replaced by the expected oxadiazole. These results provide additional indirect support for the presence of a tetrahedral intermediate during diverted UmAS.




Figure 7. (A) General scheme of diverted UmAS in which an α-bromo nitroalkane reacts with an acyl hydrazide to form a 1,3,4-oxadiazole via a tetrahedral intermediate (shown in red). (B) Reaction of an aryl-substituted α-fluoro nitroalkane with ethyl acyl hydrazide. The predicted tetrahedral intermediate is shown in red. After short incubations, the indicated fluoro hydrazone was detected in the reaction mixture. Longer incubation periods yielded the oxadiazole.



Elucidating reaction mechanisms is the holy grail of organic chemistry, the ultimate goal that remains forever elusive. One can never actually prove a mechanism. Rather, through an ongoing process of hypothesis testing, modification, and refinement, a likely reaction pathway can be outlined. The importance of this process cannot be overstated, because a well-supported mechanism enables the chemist to predict how previously untested compounds will behave under a given set of reaction conditions, enabling new applications for known reactions and the design of new ones. Fundamental to this process is the understanding of functional group chemistry. Thus, the discovery of a new functional group, HANA, and the exploration of its chemistry in UmAS and related reactions represents a major contribution to the organic chemist's armamentarium. We look forward to learning more about HANA and the applications of UmAS in the near future.




View Cell Press article: Direct Observation and Analysis of the Halo-Amino-Nitro Alkane Functional Group










The Vanderbilt Institute of Chemical Biology, 896 Preston Building, Nashville, TN 37232-6304, phone 866.303 VICB (8422), fax 615 936 3884
Vanderbilt University is committed to principles of equal opportunity and affirmative action. Copyright © 2014 by Vanderbilt University Medical Center