Here, we present a protocol for direct, early stage guanidinylation that enables rapid total synthesis of aminoguanidine-containing small organic molecules. An advanced synthetic intermediate used in the synthesis of a blood coagulation factor XIa inhibitor was prepared using this protocol.
The guanidine functional group, displayed most prominently in the amino acid arginine, one of the fundamental building blocks of life, is an important structural element found in many complex natural products and pharmaceuticals. Owing to the continual discovery of new guanidine-containing natural products and designed small molecules, rapid and efficient guanidinylation methods are of keen interest to synthetic and medicinal organic chemists. Because the nucleophilicity and basicity of guanidines can affect subsequent chemical transformations, traditional, indirect guanidinylation is typically pursued. Indirect methods commonly employ multiple protection steps involving a latent amine precursor, such as an azide, phthalimide, or carbamate. By circumventing these circuitous methods and employing a direct guanidinylation reaction early in the synthetic sequence, it was possible to forge the linear terminal guanidine containing backbone of clavatadine A to realize a short and streamlined synthesis of this potent factor XIa inhibitor. In practice, guanidine hydrochloride is elaborated with a carefully constructed protecting array that is optimized to survive the synthetic steps to come. In the preparation of clavatadine A, direct guanidinylation of a commercially available diamine eliminated two unnecessary steps from its synthesis. Coupled with the wide variety of known guanidine protecting groups, direct guanidinylation evinces a succinct and efficient practicality inherent to methods that find a home in a synthetic chemist's toolbox.
The objective of this video is to show how using a direct and early guanidinylation method to make a terminal guanidine structure is more practical, rapid, and efficient than traditional guanidinylation methods in organic synthesis. The guanidine functional group, found on the amino acid arginine, is a key structural element in many complex natural products and pharmaceuticals. The discovery and design of new guanidine containing natural products and small molecules establish the need for a more efficient guanidinylation method. The commonly used circuitous approach features the introduction of a latent guanidine precursor that is unmasked at a late stage in the synthesis. In contrast, a straightforward tactic installs a protected guanidine onto a primary amine early in a synthetic route.
The reactive nature of guanidines generally precludes them from routine use without an appropriate protecting group strategy. Traditionally, methods to add a guanidine functional group involved an indirect approach that involved multiple protection steps followed by adding the guanidine at the end of the synthesis. Two recent syntheses illustrate the drawbacks inherent to indirect guanidinylation1,2. The direct method reported herein involves reacting a protected guanidine reagent with a primary amine early on in the synthesis of a given molecule and then deprotecting it at the end of the synthesis. This strategy was deployed successfully in recent total synthesis of biologically active marine alkaloids clavatadine A and phidianidine A and B3,4.
While this direct guanidinylation method does have its advantages over traditional methods of guanidinylation it still has its drawbacks. The chemical conditions that the protected guanidine can survive will depend on the protecting group employed. Despite these potential drawbacks, the direct guanidinylation method is an enabling strategy to add terminal guanidines to primary amines for use in the synthesis of complex organic molecules.
Caution: Please consult and heed Safety Data Sheets (SDS) for each chemical prior to use. A few of the chemicals used in this synthesis are corrosive, toxic, carcinogenic, or otherwise harmful. Consequently, take every precaution to avoid inhalation, ingestion, or skin contact with these chemicals. Please wear appropriate Personal Protective Equipment (PPE) correctly. Proper PPE includes wrap-around safety goggles, nitrile gloves or more chemically resistant gloves, a lab coat, long pants that cover the tops of the shoes, closed-toe shoes. Use a working fume hood with the sash at the lowest possible height, in tandem with additional relevant engineering controls, to minimize the risk of accidental exposure. Portions of the procedure involve standard air- and moisture-free technique, such as the use of a Schlenk line, amber chemical storage bottles with crown caps and elastomer discs, syringe and cannula transfer of liquids and solutions, compressed gasses, and the distillation of flammable liquids under inert atmosphere.5
1. Direct Guanidinylation
2. Synthesis of Carbamate 9 from di-Boc Agmatine Isocyanate (8) and 2,4-DibromoHGAL (3)3
3. Synthesis and Isolation of Clavatadine A (10)3
Direct guanidinylation of a commercially available α,ω-diamine, followed by reaction with triphosgene, afforded the reactive isocyanate 8 as the linear portion of clavatadine A (Figure 1b). Yields of this two-step reaction sequence are invariably high, 95% or greater. Guanidinylation reagent 6 was prepared exactly as described by Goodman.11,24
When isocyanate 8 was combined with dibrominated phenol 3 (whose synthesis is shown in Figure 1a) in the presence of a catalytic amount of the organic base N,N-diisopropylethylamine, carbamate formation provided compound 9 (Figure 1c) in moderate yield. An aliquot of the reaction mixture was taken after 15 minutes and worked up. IR analysis of this mixture showed that the isocyanate had been fully consumed. With this data, it is unclear why the reaction yield is not higher. Reisolation of dibromophenol 3 after chromatography suggests that perhaps some of the isocyanate decomposed under the reaction conditions, or the product may have partially hydrolyzed during workup or chromatography. Finally, hydrolysis of the lactone under acidic conditions was accompanied by deprotection of the guanidine protecting groups leading to the final molecule, clavatadine A (10) (Figure 1d). Exposure of any benzofuranone-containing molecules to methanol at any stage of the synthesis invariably led to irreversible methanolysis of the lactone; therefore, contact with small alcohols is to be avoided.
Figures 2-8 include NMR or IR spectra that confirm the structure of each compound whose preparation is described herein. Comparison of the NMR spectrum of each synthesized compound with the NMR spectrum of its synthetic precursor reveals structural changes that confirm the identity of the prepared molecule. Each NMR spectrum is festooned with arrows that show likely or confirmed assignments for each spectral resonance with each group of unique hydrogen atoms in a prepared molecule. Additional supporting data that further confirms the structural assignments within synthesized molecules has been published elsewhere.3
Figure 1. Step-wise synthesis of clavatadine A (10) by a direct, early stage guanidinylation approach. Schemes a-d illustrate the sequence of chemical reactants and reaction conditions for the preparation of clavatadine A (10) by a direct, early stage guanidinylation approach. (a) Synthesis of the aromatic portion, 2,4-dibromohomogentisic acid lactone (3). (b) Synthesis of the linear portion, isocyanate 8, by direct guanidinylation. (c) Carbamate-forming reaction uniting the aromatic and linear subunits. (d) Acidic hydrolysis and Boc-deprotection of carbamate 9 leading to the hydrochloride salt of clavatadine A (10). Please click here to view a larger version of this figure.
Figure 2. Proton NMR spectrum confirming the preparation of linear compound di-Boc-agmatine (7). Numerical values pertaining to chemical shifts and relative integration of visible proton signals are labeled above and below the spectrum, respectively; peak assignments originate from the structure shown; and known impurities are listed above each relevant peak3. Please click here to view a larger version of this figure.
Figure 3. Proton NMR spectrum confirming the preparation of linear compound di-Boc-agmatine isocyanate (8). Numerical values pertaining to chemical shifts and relative integration of visible proton signals are labeled above and below the spectrum, respectively; peak assignments originate from the structure shown; and known impurities are listed above each relevant peak3. Please click here to view a larger version of this figure.
Figure 4. Infrared (IR) spectrum confirming the preparation of linear compound di-Boc-agmatine isocyanate (8). Numerical values pertaining to wave numbers of bond absorptions are labeled below the spectrum, but above the abscissa. The characteristic isocyanate stretch of compound 8 can be found at 2,265 cm-1. Please click here to view a larger version of this figure.
Figure 5. Proton NMR spectrum confirming the preparation of carbamate 9, in CDCl3. Numerical values pertaining to chemical shifts and relative integration of visible proton signals are labeled above and below the spectrum, respectively; peak assignments originate from the structure shown; and known impurities are listed above each relevant peak3. Please click here to view a larger version of this figure.
Figure 6. Proton NMR spectrum confirming the preparation of carbamate 9, in DMSO-d6. Numerical values pertaining to chemical shifts and relative integration of visible proton signals are labeled above and below the spectrum, respectively; peak assignments originate from the structure shown; and known impurities are listed above each relevant peak. Please click here to view a larger version of this figure.
Figure 7. Proton NMR spectrum confirming the preparation of clavatadine A (10), in DMSO-d6. Numerical values pertaining to chemical shifts and relative integration of visible proton signals are labeled above and below the spectrum, respectively; peak assignments originate from the structure shown; and known impurities are listed above each relevant peak3. Please click here to view a larger version of this figure.
Figure 8. Carbon (13C) NMR spectrum confirming the preparation of clavatadine A (10), in DMSO-d6. Numerical values pertaining to chemical shifts are labeled above the spectrum3. Please click here to view a larger version of this figure.
Initial efforts to prepare clavatadine A enlisted a traditional, indirect approach to guanidinylation from a suitable amine precursor, which in this case was a terminal azide. Central to this effort was the union of the two halves of the molecule to construct the carbamate moiety. Unfortunately, all attempts to realize an azide reduction in anticipation of a planned late-stage guanidinylation were unsuccessful.25,26 These setbacks inspired the pursuit of compound 7, which could be prepared in a single step by direct guanidinylation from commercially available materials. Though this method had been used in prior total syntheses, in this instance, a direct approach circumvented a critical impasse encountered amid myriad attempts to install the protected guanidine functionality into an advanced synthetic intermediate.27-29
The application of this direct aminoguanidinylation approach began with the preparation of the known N,N´-di-Boc-protected guanidine 7 from Goodman's reagent (6) and 1,4-butanediamine (5) in high yield.3,6,30 The terminal amine of protected guanidine 7 was converted into the reactive isocyanate 8 by exposure to triphosgene in a bi-phasic solvent mixture.3 In the penultimate synthetic transformation, the electrophilic isocyanate 8 was treated with 2,4-dibromohomogentisic acid lactone (3) to form the central carbamate linkage in compound 9.3 Finally, tandem lactone hydrolysis and guanidine deprotection occurred under dilute acidic conditions to provide clavatadine A (10) in 93% yield.3 The overall yield for the entire four-step synthesis (longest linear sequence) is 41-43%.3
For each chemical reaction described in the protocol that was not conducted in aqueous media, the use of high-purity, moisture-free solvents was critical. Some of the reactive intermediates formed during these transformations likely react with adventitious water, leading to decomposition. Although Goodman's reagent (6) is commercially available, its substantial cost and relative ease of synthesis made its preparation a reasonable choice. Again, minimizing moisture by distilling each reagent and painstaking temperature control were critical to its successful synthesis, as stated in the published procedure.11
Despite the expediency inherent to this direct approach for the synthesis of biologically relevant aminoguanidine-containing compounds, there are limitations to this method. Using different amine protection groups is possible in this direct guanidinylation method, but the overall success will always depend on the chosen protecting group strategy. Principally, the selection of aminoguanidine protecting groups requires substantial forethought, because the masked guanidine must remain intact throughout every subsequent synthetic step. In addition, the advanced target molecule must be able to endure the conditions and reagents required for guanidine deprotection at the appropriate time. In the synthesis of clavatadine A (10), acid-sensitive Boc groups were used to protect the guanidine, which necessitated the avoidance of reactions that required or created an acidic environment. In this case, the need to employ acidic conditions to hydrolyze the lactone was optimal due to the fact that acid is a convenient means to cleave Boc carbamates.31 Although clavatadine A represented an ideal template to showcase this approach, direct guanidinylation should be amenable to the preparation of many other natural and non-natural organic molecules. To this end, efforts are underway in our laboratory to prepare several non-natural analogues of clavatadine A as part of a drug-discovery program to develop a reversible and selective, natural-product based inhibitor of human blood coagulation factor XIa.32
What makes this direct method potentially better than the traditional, indirect approach is that it can shorten an organic synthesis route by multiple steps, removing the need to protect and deprotect a terminal amine multiple times before installing the desired guanidine functionality. Though traditional, indirect guanidinylation methods are effective, such as that illustrated in Looper's recent total synthesis of saxitoxin, the inclusion of extraneous steps in a synthesis is time-intensive and can potentially lower the overall yield.33 Moreover, the value of direct guanidinylation was highlighted in a recent total synthesis of 1,2,4-oxadiazole-containing natural products phidianidine A and B. This total synthesis was two steps shorter than the synthesis reported one year earlier by Snider and co-workers.4,34
In the future, the direct guanidinylation method needs to be expanded and tested on different aminoguanidine-containing scaffolds, inescapably, toward the exploration of varied guanidine protecting groups. Clavatadine A and phidianidine A and B both used Boc protecting groups to mask the guanidine functionality. The next stage in the refinement of this method would be to try the same reactions with different protecting groups to see if higher yields can be obtained.4 Recent work by Pfeffer,35 Looper,36 and Nagasawa37 suggests that a variety of aminoguanidine protecting groups in addition to Boc, such as Cbz, as well as derivatives of Cbz may be enlisted. Another approach would involve the use of two different protecting groups on the aminoguanidine scaffold. Judiciously chosen masking groups with orthogonal reactivity may enable the aminoguanidine to survive reaction conditions that cleave one protecting group while leaving the other intact.38 In conclusion, the direct guanidinylation method used for the total synthesis of clavatadine A and variations thereof may be used to synthesize newly discovered guanidine-containing natural products and designed pharmaceuticals.39,40
The authors have nothing to disclose.
We thank Dr. John Greaves and Ms. Soroosh Sorooshian, Department of Chemistry, University of California, Irvine Mass Spectrometry Facility, for mass spectrometric analyses. We also thank Mr. Jacob Buchanan for helpful discussions, as well as Miss Stephanie J. Conn, Mrs. Shannon M. Huffman (Vreeland), and Miss Alexandra N. Wexler for early stage work on this project. Partial funding was provided by the Central Washington University (CWU) School of Graduate Studies (C.E.M), the CWU Seed Grant Program, and the CWU Faculty Research Program.
Chloroform-d | Sigma-Aldrich | 612200-100G | 99.8% D, 0.05% v/v tetramethylsilane, Caution: toxic |
Dimethylsulfoxide-d6 | 185965-50G | 99.9% D, 1% v/v tetramethylsilane | |
sodium thiosulfate pentahydrate | Sigma-Aldrich | S8503-2.5KG | |
sodium sulfate, anhydrous | Sigma-Aldrich | 238597-2.5KG | |
silica gel | Fisher Scientific | S825-25 | Merck, Grade 60, 230-400 mesh |
washed sea sand | Sigma-Aldrich | 274739-5KG | |
hexane | Sigma-Aldrich | 178918-20L | Caution: flammable |
ethyl acetate | Sigma-Aldrich | 319902-4L | |
methylene chloride | Sigma-Aldrich | D65100-4L | |
sodium chloride | Sigma-Aldrich | S9888-10KG | |
sodium bicarbonate | Sigma-Aldrich | S6014-2.5KG | |
acetic acid | Sigma-Aldrich | 695092-2.5L | |
hydrochloric acid | Sigma-Aldrich | 258248-2.5L | Caution: Corrosive |
bromine | Sigma-Aldrich | 470864-50G | >99.99% trace metals basis Caution: Corrosive, causes severe burns |
hydrobromic acid | Sigma-Aldrich | 244260-500ML | 48% aqueous, Caution: Corrosive |
2,5-dimethoxyphenylacetic acid | ChemImpex | 26909 | |
chloroform | Sigma-Aldrich | 132950-4L | Caution: Toxic |
tetrahydrofuran | Sigma-Aldrich | 360589-4x4L | Caution: highly flammable |
N,N-diisopropylethylamine | Sigma-Aldrich | D125806-500ML | Caution: Corrosive |
triethylamine | Sigma-Aldrich | T0886-1L | Caution: Corrosive |
3 Angstrom molecular sieves | Sigma-Aldrich | 208574-1KG | |
calcium hydride | Sigma-Aldrich | 213268-100G | Caution: Corrosive, reacts violently with water |
ammonium molybdate | Sigma-Aldrich | 431346-50G | |
phosphomolybdic acid | Sigma-Aldrich | 221856-100G | |
cerium (IV) sulfate | Sigma-Aldrich | 359009-25G | |
1-butanol | Sigma-Aldrich | 537993-1L | |
1,4-butanediamine | Sigma-Aldrich | D13208-100G | Caution: Corrosive / warm in hot water bath to melt prior to use |
triphosgene | VWR | 200015-064 | Caution: Highly Toxic |
methanol | Sigma-Aldrich | 646377-4X4L | |
sodium acetate | Sigma-Aldrich | 241245-100G | |
Dimethylsulfoxide-d6 | Sigma-Aldrich | 570672-50G | Anhydrous, 99.9% D |
sodium hydroxide | Sigma-Aldrich | 221465-500G | Caution: Corrosive |
guanidine hydrochloride | Sigma-Aldrich | G4505-25G | Caution: Toxic, Corrosive |
di-tert-butyl dicarbonate | VWR | 200002-018% | Caution: Toxic / may warm in hot water bath to melt prior to use |
trifluoromethanesulfonic anhydride | Fisher Scientific | 50-206-771 | 98%, anhydrous, Caution: toxic, corrosive, extremely moisture sensitive |