Download PDF
Synlett 2018; 29(04): 375-382
DOI: 10.1055/s-0037-1609093
synpacts
© Georg Thieme Verlag Stuttgart · New York

Recent Progress in Methylation of (Hetero)Arenes by Cross-Coupling or C–H Activation

Lu Hu
a   Department of Pharmacology and Pharmaceutical Sciences, School of Medicine, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Tsinghua University, Beijing 100084, P. R. of China   Email: liaoxuebin@mail.tsinghua.edu.cn
,
Yahu A. Liu  *
b   Discovery Chemistry, Genomics Institute of the Novartis Research Foundation, San Diego, CA 92121, USA   Email: yliu2@gnf.org
,
Xuebin Liao  *
a   Department of Pharmacology and Pharmaceutical Sciences, School of Medicine, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Tsinghua University, Beijing 100084, P. R. of China   Email: liaoxuebin@mail.tsinghua.edu.cn
› Author Affiliations

This work was supported by the Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Tsinghua–Peking Centre for Life Sciences and by the 1000 Talents Recruitment Program.
Further Information

Publication History

Received: 24 October 2017

Accepted after revision: 20 December 2017

Publication Date:
05 February 2018 (online)

 
 PDF Download Permissions and Reprints All articles of this category


Abstract

Owing to the ‘magic methyl effect’ on a compound’s physical and biological properties, methylation is a strategy frequently used by medicinal chemists in structure–activity relationship studies or in lead optimization. This article highlights the most recent reported methods for the direct methylation of (hetero)arenes, which mainly involve either C–H functionalization or cross-coupling of methylating reagents with (hetero)aryl halides. Methylation of C–H bonds of (hetero)-arenes, which is atom economical, has been explored by several research groups in recent years. Given the unmatchable availability of (hetero)aryl halides, we believe that Ni-catalyzed methylation using iodomethane or deuterated iodomethane as the methyl source is one of the most convenient methods.


#

Key words

methylation - methyl effect - iodomethane - C–H activation - aryl halides - cross-coupling - nickel catalysis
Zoom Image
Lu Hu was born in Wuhan, P. R. of China, in 1991. He received his B.S. in Chinese medicine from the Huazhong University of Science and Technology in 2014. He then joined Professor Liao’s group at the School of Pharmaceutical Sciences, Tsinghua University (P. R. of China) as a graduate student. His main research interests include the development of new synthetic reactions by using late-transition-metal catalysis, and natural-product synthesis.
Yahu A. Liu received his Ph.D. degree in organic chemistry in 2000 from Case Western Reserve University under the guidance of the late Professor Lawrence M. Sayre. Afterward, he worked in medicinal-chemistry teams at Pharmacia/Pfizer, Vertex, and ChemBridge. In 2007, he joined the Genomics Institute of the Novartis Research Foundation, where he is currently an investigator and chemistry outsourcing lead. He has co-authored 68 publications, and has served on the editorial boards of two biochemistry journals. His main research interests include drug discovery, natural-product synthesis, synthesis methodology, and supramolecular chemistry. He served as president of the Sino-American Biotechnology and Pharmaceutical Professional Association from 2015 to 2017.
Xuebin Liao obtained his B.S. degree from Peking University in 1995, and then he did his Ph.D. studies on natural-product synthesis with Professor James M. Cook at the University of Wisconsin at Milwaukee from 1998 to 2004. From 2005 to 2008, he was a postdoctoral associate of Professor John F. Hartwig at Yale University, and later at the University of Illinois, Urbana-Champaign. In 2009, he held a position of research investigator at the Genomics Institute of the Novartis Research Foundation, where his research centered on discovering new drug candidates for pharmaceutical optimization. In September of 2012, he joined the Department of Pharmacology and Pharmaceutical Science of the School of Medicine of Tsinghua University, where his current researches are focused on three major areas: (1) bioactive natural-product-oriented synthesis and the development of new synthetic methods, (2) immuno-oncology projects related to T-cell exhaustion and innate immunity, and (3) structure-based drug discovery.

The methyl group is one of the most commonly occurring carbon fragments in small-molecule drugs. A survey of Njarðarson and co-workers’ Top 200 Drugs of 2012 showed that more than 67% of small-molecule drugs contain at least one methyl group bound to a carbon atom, with many of them having one or more methyl groups attached to (hetero)arenes (Figure [1]).[1] [2] A methyl group can modulate both biological and physical properties of a molecule. Although the so-called ‘magic methyl effect’ usually refers to a methyl group’s effect on binding potency,[3] and consequently on the IC50 value of a drug candidate, the effects of a methyl group on a molecule’s physical and biological properties extend beyond potency. It has been reported that adding one or more methyl groups can change a drug candidate’s half-life,[4] solubility,[5] or selectivity against off-targets,[6] and can even convert an agonist into an antagonist.[7] Thus, methylation is among the most widely used strategies for modifying bioactive compounds in medicinal-chemistry research.

Zoom Image
Figure 1 Small-molecule drugs containing at least one carbon-bound methyl group on a (hetero)arene

In drug discovery, occasionally what is old becomes new again. Deuterium substitution in drug-design has faded in and out of fashion over the years,[8] and now drug firms are again banking on its potential to improve existing drugs. In 2017, the US Food and Drug Administration (FDA) approved the deuterated drug SD-809 from Teva Pharmaceuticals, and Vertex Pharmaceuticals bought the cystic-fibrosis Phase II drug candidate CTP-656 from Concert Pharmaceuticals (Figure [2]).[9] Both SD-809 and CTP-656 have much longer half-lives than the corresponding nondeuterated versions. Another example of a deuterated drug candidate is AVP-786 (Phase II) from Avanir Pharmaceuticals, which is a deuterium-substituted version of dextromethorphan (Figure [2]). [8]

Zoom Image
Figure 2 Examples of drugs and drug candidates with deuterated methyl groups

Despite the importance of the methyl group, and the latest retro fashion in drug discovery for deuterium substitution with deuterated methyl groups, the introduction of a methyl group, let alone a deuterated one, is not an easy task in most cases. In many instances, methylation of an advanced intermediate has to be realized by de novo synthesis.[10] It can be difficult to embark on a multistep synthesis to introduce one or more methyl groups in lead optimization, given the tight timeframe for medicinal-chemistry endeavors.

Consequently, reactions that permit such transformations to be achieved in a single step on an advanced drug lead are attracting increasing attention from researchers, but nevertheless remain challenging. The most common method for introducing a methyl group onto (hetero)aromatic rings involves a lithium–halide exchange[11] of an aryl halide, followed by trapping of a methyl electrophile. Recently, Knochel developed a Mg–halogen or Zn–halogen exchange, which is followed by methylation of the metalated arene.[12] Quenching of a metalated (hetero)arene formed by deprotonation of an acidic C–H bond is an another common strategy for methylation. Because several existing reviews focus on the metalation of aryl groups,[13] neither methylation strategy will be covered in this review. In recent studies, the direct methylation of (hetero)aromatic rings has relied mainly on two types of reaction, C–H activation and cross-coupling of carbon–halide bonds, although there were a few other methods that have not attracted the same level of attention as these two methods[10] [14] In this article, we highlight recent progress in the development of methods for transition-metal-catalyzed methylation of (hetero) arenes, including the methylation of C–X and C–H bonds. We also discuss the methyl sources used in the various methods.

Cross-coupling of organic halides with organometallic reagents is one of the most commonly used method for C–C bond formation.[15] Cross-coupling of two electrophiles is a considerable challenge because of the complications arising from competitive homocouplings, hydride functionalization, and catalyst deactivation.[16] Although Weix and co-workers have described an excellent method for reductive coupling of aryl and alkyl halides with high selectivity toward the cross-coupling product over the homocoupling product, they did not demonstrate a coupling reaction with a methyl halide.[17] This might be attributed either to the difficulty of generating free methyl radicals or to their subsequent involvement in C–C bond formation through nickel-mediated reductive elimination, as implied by the proposed free-radical chain mechanism.[18]

Zoom Image
Scheme 1 Previous methods for the methylation of aromatic carbon–halide bonds
Zoom Image
Scheme 2 Liao and co-workers’ nickel-catalyzed methylation of aryl halides. Reaction conditions (unless otherwise stated): 11 or 13 (0.5 mmol, 1.0 equiv), MeI (3.5 equiv), NiCl2(dppp) (10 mol%), Zn (3.0 equiv), NaI (1.5 equiv), THF (5 mL), 24 h. Yields are those of the isolated products. a12f was isolated as a mixture; the ratio was determined by 1H NMR. b 11p = 4-bromo-4′-iodobiphenyl. c MeI (7.0 equiv), Zn (6.0 equiv), NaI (3.0 equiv). d Yield determined by 1H NMR with 1,3,5-trimethoxybenzene as internal standard.

The first examples of metal-catalyzed methylation of arenes through cross-coupling of carbon halides were reported by Hatanaka and Hiyama in 1988, who used organosilicon reagent as their methyl source (Scheme [1]).[19] In 1996, Schumann and Blum and their co-workers synthesized a series of air-stable methylaluminum complexes that they successfully used for methylation of aryl halides in the presence of nickel or palladium catalysts (Scheme [1]).[20] The analogous gallium complexes were also reported to be suitable methyl sources in this kind of transformation.[20] By using commercially available 1,4-diazabicyclo[2.2.2]octane (DABCO) and AlMe3, Woodward’s group synthesized an air-stable methylaluminum complex, DABAL-Me3, which allowed the methylation reaction to be carried out without using an inert atmosphere (Scheme [1]).[21] In 2007, Love and co-workers reported a Pt-catalyzed methylation of C–F bonds in difluoroarylimines by using Me2Zn as the methyl source (Scheme [1]).[22] Shortly afterwards, the use of methyl Grignard reagents as methyl sources in Ni-catalyzed methylation of C–O or C–Br bonds was reported by Shi’s group, and in Fe-catalyzed methylations was described by Cook’s group (Scheme [1]).[23] All of these reported works relied on the use of air-sensitive organometallic reagents (Mg, Al, Zn, or Si) as methyl sources.

This limitation becomes more problematic when the installation of a CD3 group is required, because CD3-containing organometallic reagents are scarce and either not commercially available or prohibitively expensive [e.g., D3CB(OH)2]. Methyl iodide (MeI) and its deuterated analogue (D3CI) are probably the least expensive and most readily available sources of CH3and CD3 groups, respectively. MeI is naturally emitted by rice plantations, algae, and kelp,[24] and it was approved for use as a pre-plant biocide by the US Environmental Protection Agency in 2007.[25] We therefore began our work on metal-catalyzed methylation of aryl halides by using MeI and D3CI as sources of methyl and deuteromethyl groups, respectively.[26] The investigation of the reactions commenced with the use of a nickel catalyst in the presence of zinc. No methylation product was obtained when nitrogen-containing ligands were used, which led us to suspect that such ligands might undergo N-methylation, thereby preventing C-methylation. By switching to the phosphorus ligand 1,3-bis(diphenylphosphino)propane (dppp) and adding NaI to generate more reactive organozinc species, the methylation of 4-butyl-1-iodobenzene (11a) was achieved at room temperature in 74% yield (Scheme [2]).[26] With the optimized conditions in hand, we explored the scope of the reaction with respect to various aryl iodides and bromides. A variety of aryl iodides 11 or bromides 13 underwent methylation with MeI to give the desired products 12 and 14, respectively (Scheme [2]). Electron-withdrawing, electron-donating, and electron-neutral substituents were all well tolerated in this reaction system. Various functional groups, including amide, ester, ketone, formyl, and sulfonamide groups, were compatible with the reaction conditions, providing the corresponding methylation products in good to excellent yields. Gratifyingly, even a vinyl group was tolerated under these reaction conditions.[26]

Notably, by using our method, SD-560 was prepared in gram quantities in 69% overall yield by a straightforward two-step synthesis (Scheme [2]), whereas a previous method afforded this compound in less than 5% overall yield in seven steps.[27] Very recently, Falb et al. reported a new method for the synthesis of SD-560, which involves a palladium-catalyzed process.[28]

Zoom Image
Scheme 3 Syntheses of SD-560

Through mechanistic studies, we proposed a plausible mechanism involving a Ni0/NiII catalytic cycle (Scheme [4]), which is somewhat similar to that of the palladium-catalyzed aqueous Lipshutz–Negishi cross-coupling of alkyl halides with aryl electrophiles,[16d] although no methylation product was formed under Lipshutz’s reaction conditions. Note that previous mechanistic studies by the groups of Fu,[29] Vicic,[30] Weix,[18] and others[31] [32] all suggested that a NiI/NiIII catalytic cycle is involved in such transformations. Nevertheless, we believe that nickel-catalyzed cross-coupling between aryl electrophiles and alkyl electrophiles involves a Ni0/NiII catalytic cycle.[25]

Zoom Image
Scheme 4 Plausible mechanism for Liao’s methylation of aryl halides

Although much progress has been made, the transition-metal-catalyzed methylation of aryl halides requires further optimization. Future work should focus on broadening the substrate scope, for example, to include C–Cl or C–F bonds, and on exploring greener reaction systems that, ideally, would involve combinations of cheap, air-stable, and nontoxic methylating reagents, catalysts, additives, and solvents under mild reaction conditions.

Earlier direct conversions of C–H bonds into C–Me bonds relied on the deprotonation of an acidic C–H bond by a strong base, followed by quenching with an electrophile, which only works for acidic C–H bonds under harsh conditions, and has limited functional-group tolerance.[33] A radical methylation of C–H bonds by using a silver or iron salt and acetic acid to generate methyl radicals has been reported.[34] The first meaningful methylation of C–H bonds through functionalization was reported in 1984 by Tremont and co-workers.[35] Although a stoichiometric amount of palladium acetate had to be used as a catalyst, or a stoichiometric amount of AgOAc had to be used to regenerate Pd(OAc)2, this pioneering work revealed the feasibility of catalytic C–H bond functionalization with weakly coordinating directing groups. Significant progress was made when Yu and co-workers reported a methylation of aryl C–H bonds by using a catalytic amount of a Pd catalyst, rather than a stoichiometric amount as used in previous work; however, toxic Me4Sn had to be used as the methyl source and batchwise addition of the tetramethyltin reagent was necessary (Table [1], entry 1).[36] Subsequently, several groups described Pd-catalyzed methylations of aromatic C–H bonds with methylboronic acid (entries 2 and 3)[37] or with potassium trifluoro(methyl)borate (entry 4).[38] The use of the latter permitted the methylation to be carried out with MnF3 as an oxidant at 40 °C, whereas a silver salt and heating to 100–110 °C were required when the former was used as the methylating reagent. Nakamura’s group reported a nontoxic cobalt-catalyzed methylation carried out at room temperature by using a methyl Grignard reagent as a methyl source (entry 5).[39] The methylation of C–H bonds by using DMSO (entry 6) or acetic acid (entry 7) as a methyl source has also been reported.[40] In 2008, Li and co-workers developed a methylation of aryl C–H bonds with dicumyl peroxide as both the methylating reagent and reaction solvent (entry 8).[41] Although moderate to good yields were obtained by this process, the relatively low chemoselectivity in some cases resulted in the formation of mono- and dimethylated products. In addition, this method has safety concerns since heating of large amounts of the peroxide at high temperatures is potentially dangerous. Because of the ready availability of MeI, researchers have reported its use as a replacement for organoboron or Grignard reagent (entries 9–11).[42]

In 2014, Beller’s group developed the first examples of a catalytic methylation of C–H bonds by using CO2 and hydrogen as the methyl source (Scheme [3]). A ruthenium triphos catalyst, formed in situ in the presence of a Lewis acid, was the key to success, and a formylated(hetero)arene intermediate was believed to be involved in the reaction.[43] The C–H methylation of indoles, pyrroles, and activated benzene derivatives was achieved in moderate to excellent yields.

Also in 2014, Baran’s group presented a practical two-step methylation of C–H bonds in heteroarenes (Scheme [5]).[44] When treated with zinc bis(phenylsulfonylmethanesulfinate) (PSMS), a specially designed reagent, a hetero­arene was transformed into a (phenylsulfonyl)methylated intermediate that could be easily separated from the unreacted starting material. This intermediate was then desulfonylated in open air at room temperature to give the methylated product 26. Although the methylation of a C–H groups by this approach has to be realized in two steps, the straightforwardness of the isolation of the sulfonylmethylated intermediate and the desulfonylation render the method attractive.

Table 1 Previous Methods for Methylation of C–H bonds of (Hetero)arenes

Entry

Methyl source

Conditions

Product

Yield (%)

 1

SnMe4

Pd(OAc)2 (10%), BQ,a Cu(OAc)2, MeCN, 100 °C, 40 h

86

 2

MeB(OH)2

Pd(OAc)2 (10%), BQ, AgOAc, EtC(Me2)OH, 100 °C, 3 h

62

 3

MeB(OH)2

Pd(OAc)2 (10%), BQ, Ag2CO3, t-BuOH, 100 °C, 3 h

75

 4

MeBF3K

Pd(OAc)2 (10%), MnF3, AcOH, THF–H2O, 40 °C, 3 h

83

 5

MeMgCl

Co(acac)2 (10%), DMPU, THF, 23 °C, 12 h

66

 6

DMSO

PdCl2(MeCN)2 (10%), n-Bu4NOAc, ZnO
Bu3N, air, 120 °C, 48 h

64

 7

AcOH

[Rh(CO)2]2 (2.5%), Boc2O, toluene, 140 °C, 24 h

76

 8

dicumyl peroxide

Pd(OAc)2 (2.5%), dicumyl peroxide, 130 °C, 12 h

63

 9

MeI

Pd(OAc)2 (5%), JohnPhos,b Cs2CO3, 110 °C, 18 h

41

10

MeI

Pd(OAc)2(5%), O2, NaOTf, K2CO3,
EtC(Me2)OH, 125 °C, 36 h

90

11

MeI

Pd(OAc)2 (5%), AgOAc, Cu(OTf)2, THF–CH2Cl2, 25 °C, 11 h

86

a 1,4-Benzoquinone

b 2-(t-Bu2P)C6H4Ph.

In 2015, Shang et al. reported a methylation of a C–H bond by using AlMe3 in the presence of a catalytic amount of an inorganic iron(III) salt and a diphosphine, along with 2,3-dichlorobutane as a stoichiometric oxidant (Scheme [5]).[45a] Impressively, the use of the mild aluminum reagent prevents undesired reduction of iron and permits the reaction to proceed with catalyst turnover numbers as high as 6500. Furthermore, pyrophoric AlMe3can be replaced by its air-stable diamine complex. Despite these advantages, the reaction is only applicable to amide substrates bearing a picolinoyl or 8-aminoquinolyl directing group. Recently, the same workers reported a methylation using a new tridentate phosphine ligand 4-Me2NC6H4P(C6H4-2-PPh2)2 (Me2N-TP) for metal catalysis as a replacement for the diphosphine ligand employed in their previous reported work (Scheme [5]).[45b] The change of ligand extended the substrate scope to include ortho C−H bonds of aromatic acids, esters, amides, and ketone, thereby dramatically broadening the range of applications of the reaction. Bao’s group also achieved an efficient Fe-catalyzed methylation of vinyl arenes (Scheme [5]).[46]

Zoom Image
Scheme 5 Recent progress in the methylation of (hetero)aromatic C–H bonds

Direct methylation of C–H bonds through C–H functionalization is atom economical, and can be very green in ideal cases, for example, by using O2 or air as the oxidant and abundant metals (Fe, Ni, etc.) as catalysts at low reaction temperature near room temperature. However, such methylations still have some challenges in their broad application in lead optimization. Generally speaking, C–H functionalization is only applicable to a limited range of C–H bonds in appropriate positions relative to directing groups, which limits the substrate scope. More than stoichiometric amounts of the oxidant are required, although O2can serve as the oxidant in rare cases. We believe that future work will overcome these limitations, but it is less likely that a fully general method will be found, because selectivity is desired where the substrates used only permit access to C–H bonds in certain positions.

The transition-metal-catalyzed direct and selective functionalization of C–H bonds is an atom-economical alternative to conventional cross-coupling reactions for C–C bond formation. Although much progress has been made in C–H methylation, a major limitation remains in that the substrate scope is limited by the need for a directing group. Given the unmatchable availability of (hetero)aryl halides, methylation through cross-coupling of aryl halides has great potential. Future work on methylation of either carbon–halogen bonds or C–H bonds should focus on exploring green reaction systems in which inexpensive and nontoxic catalysts, ligands, oxidants (for C–H bonds), additives, and solvents are used, and reactions can be carried out under mild conditions.


#

Acknowledgment

The authors thank Dr. Daniel Raymond for his critical reading and helpful suggestions during the preparation of the manuscript.



   Permissions and Reprints All articles of this category
Zoom Image
Lu Hu was born in Wuhan, P. R. of China, in 1991. He received his B.S. in Chinese medicine from the Huazhong University of Science and Technology in 2014. He then joined Professor Liao’s group at the School of Pharmaceutical Sciences, Tsinghua University (P. R. of China) as a graduate student. His main research interests include the development of new synthetic reactions by using late-transition-metal catalysis, and natural-product synthesis.
Yahu A. Liu received his Ph.D. degree in organic chemistry in 2000 from Case Western Reserve University under the guidance of the late Professor Lawrence M. Sayre. Afterward, he worked in medicinal-chemistry teams at Pharmacia/Pfizer, Vertex, and ChemBridge. In 2007, he joined the Genomics Institute of the Novartis Research Foundation, where he is currently an investigator and chemistry outsourcing lead. He has co-authored 68 publications, and has served on the editorial boards of two biochemistry journals. His main research interests include drug discovery, natural-product synthesis, synthesis methodology, and supramolecular chemistry. He served as president of the Sino-American Biotechnology and Pharmaceutical Professional Association from 2015 to 2017.
Xuebin Liao obtained his B.S. degree from Peking University in 1995, and then he did his Ph.D. studies on natural-product synthesis with Professor James M. Cook at the University of Wisconsin at Milwaukee from 1998 to 2004. From 2005 to 2008, he was a postdoctoral associate of Professor John F. Hartwig at Yale University, and later at the University of Illinois, Urbana-Champaign. In 2009, he held a position of research investigator at the Genomics Institute of the Novartis Research Foundation, where his research centered on discovering new drug candidates for pharmaceutical optimization. In September of 2012, he joined the Department of Pharmacology and Pharmaceutical Science of the School of Medicine of Tsinghua University, where his current researches are focused on three major areas: (1) bioactive natural-product-oriented synthesis and the development of new synthetic methods, (2) immuno-oncology projects related to T-cell exhaustion and innate immunity, and (3) structure-based drug discovery.
Zoom Image
Figure 1 Small-molecule drugs containing at least one carbon-bound methyl group on a (hetero)arene
Zoom Image
Figure 2 Examples of drugs and drug candidates with deuterated methyl groups
Zoom Image
Scheme 1 Previous methods for the methylation of aromatic carbon–halide bonds
Zoom Image
Scheme 2 Liao and co-workers’ nickel-catalyzed methylation of aryl halides. Reaction conditions (unless otherwise stated): 11 or 13 (0.5 mmol, 1.0 equiv), MeI (3.5 equiv), NiCl2(dppp) (10 mol%), Zn (3.0 equiv), NaI (1.5 equiv), THF (5 mL), 24 h. Yields are those of the isolated products. a12f was isolated as a mixture; the ratio was determined by 1H NMR. b 11p = 4-bromo-4′-iodobiphenyl. c MeI (7.0 equiv), Zn (6.0 equiv), NaI (3.0 equiv). d Yield determined by 1H NMR with 1,3,5-trimethoxybenzene as internal standard.
Zoom Image
Scheme 3 Syntheses of SD-560
Zoom Image
Scheme 4 Plausible mechanism for Liao’s methylation of aryl halides
Zoom Image
Scheme 5 Recent progress in the methylation of (hetero)aromatic C–H bonds