Accepted Manuscript

Carbohydrate– N-Heterocyclic Carbene Metal Complexes: Synthesis, Catalysisand Biological Studies

Weihao Zhao, Vito Ferro, Murray V. Baker

PII: S0010-8545(17)30022-XDOI: http://dx.doi.org/10.1016/j.ccr.2017.03.005Reference: CCR 112412

To appear in: Coordination Chemistry Reviews

Received Date: 18 January 2017Revised Date: 6 March 2017Accepted Date: 6 March 2017

Please cite this article as: W. Zhao, V. Ferro, M.V. Baker, Carbohydrate– N-Heterocyclic Carbene Metal Complexes:Synthesis, Catalysis and Biological Studies, Coordination Chemistry Reviews (2017), doi: http://dx.doi.org/10.1016/j.ccr.2017.03.005

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customerswe are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, andreview of the resulting proof before it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

Carbohydrate–N-Heterocyclic Carbene Metal Complexes: Synthesis,

Catalysis and Biological Studies

Weihao Zhao,[a] Vito Ferro*[b] and Murray V. Baker*[a]

[a] Mr W. Zhao, Prof. M.V. Baker

School of Molecular Sciences,

The University of Western Australia,

Crawley, Western Australia 6009, Australia

E-mail: murray.baker@uwa.edu.au

[b] Prof. V. Ferro

School of Chemistry and Molecular Biosciences

The University of Queensland

Brisbane, Queensland 4072, Australia

E-mail: v.ferro@uq.edu.au

2

Table of Contents.

Abstract 3 Introduction 4 2 Synthesis 5 3 Carbohydrate N-heterocyclic carbene (NHC) complexes in catalysis 8

3.1 Carbohydrate NHC metal complexes in stereoselective catalysis 9

3.2 Carbohydrate NHC complexes as water soluble catalysts 17

4 Carbohydrate NHC complexes in medicinal chemistry 21

4.1 NHC analogues of Auranofin 22

4.2 Incorporating a carbohydrate onto imidazolidene ligands in metal

complexes 29

5 Structural complexity in carbohydrate NHC transition metal complexes 30

6 Summary and Outlook 37

7 Acknowledgements 38

8 References 38

3

Abstract

N-Heterocyclic carbenes (NHCs) are strong donor ligands that form stable complexes with

many metals, making them popular subjects for research in organometallic chemistry. While

work with NHCs initially was predominantly in the area of catalysis, applications of NHCs in

medicinal chemistry have blossomed in recent years. Chemists are investigating NHC

complexes with increasingly complex NHC components, adding novel functional groups to

the NHC ring. Among possible functional groups, carbohydrates are naturally abundant and

well known for their biocompatibility, chirality, water solubility and non-toxicity. It is not

surprising that carbohydrate groups have been incorporated into NHC ligands to impose

chirality, increase water solubility, and promote biocompatibility. Such novel NHCs and their

metal complexes show much promise in catalysis and medicinal chemistry.

In this review, we focus on metal complexes that contain both an NHC ligand and a

carbohydrate ligand, or have novel NHC ligands bearing pendant carbohydrate groups. We

discuss synthesis, structural aspects, catalysis and medicinal chemistry. Roles of the NHC

ligand and carbohydrate moiety related to these properties are highlighted.

4

1. Introduction

The last decade has witnessed dramatic developments in the field of metal complexes of N-

heterocyclic carbenes (NHCs), especially those involving 5-membered NHC ligands. They

have been investigated as catalysts for olefin metathesis, polymerisation, cyclisation reactions,

cross-coupling reactions, etc. [1,2]. After the initial burst of research in the catalysis area, use

of NHC metal complexes in other areas gained attention and is still developing, especially in

biological and medicinal chemistry. For example, various NHC metal complexes have been

shown to possess anti-infective activities (e.g., many Ag NHC complexes [3]), anti-biotic

activities [4] and anti-cancer activities (e.g. some complexes of Ag, Pd, Au, Pt, Ru and other

metals) [5-7].

Carbohydrates are naturally abundant, chiral, biocompatible molecules that serve multiple

functions, e.g., energy storage, cell recognition, adhesion and signalling, etc., in biological

systems. Thus, carbohydrates have been widely studied across many fields [8-11]. Methods

for the synthesis of a variety of carbohydrate-functionalised alkyl halides, amines, azides and

alkynes are well established and many of these compounds are readily available. Herein we

review metal complexes that incorporate NHC and carbohydrate components. We consider

two broad classes of complexes—those in which the carbohydrate is a substituent attached to

an NHC ligand, and those in which the carbohydrate is tethered to the metal more directly.

Imidazolium salts, imidazolinium salts, and triazoles are generally air-stable and easy to work

with, and their chemistry is well understood. Consequently, modification of the structure of

their corresponding NHCs is easily achieved by simple synthetic procedures, conveniently

enabling studies of structure-activity relationships of NHC complexes. So far, in the fields of

both catalysis and biological chemistry, studies of NHC metal complexes have mostly

5

focused on common NHCs having simple alkyl or aryl substituents on the ring nitrogen(s).

NHCs incorporating more elaborate substituents (carbohydrates, amino acids, nucleotides, etc)

offer unique opportunities, especially in biological fields. These more elaborate NHCs are

being increasingly studied, but much work remains to be done. The purpose of this review is

to summarize work to date concerning NHC complexes in which the NHC ligand carries a

carbohydrate-based substituent.

Many research groups have begun to put effort into studying these new types of NHCs, to

understand their properties and to determine what they can provide to modern chemistry. To

date, the carbohydrate groups investigated have typically been monosaccharides based on D-

glucopyranose, attached via the anomeric carbon (C1), C2, or C6 (Fig. 1). Metal complexes

that contain both a simple NHC ligand and a separate carbohydrate ligand are also included

in our discussion.

Fig. 1. Common modes of attachment of D-glucopyranose moieties to an NHC.

2. Synthesis

While a great variety of NHC complexes are known, including examples based on N,N-

heterocycles, N,S-heterocycles, "normal" and "abnormal" NHCs, to date NHC-carbohydrate

complexes appear only to have been reported for the 5-membered nitrogen-containing NHCs

6

imidazolidinylidene, imidazolylidene and 1,2,3-triazolylidene (Scheme 1). These NHCs are

easily derived from their precursor imidazolinium salts, imidazolium salts and 1,2,3-triazoles

respectively, by deprotonation (at C2 of imidazolinium or imidazolium ions and at C5 of

1,2,3-triazoles) using an organic or inorganic base. Common synthetic pathways [1] to these

precursors are summarized in Scheme 1.

Scheme 1.

When NHCs are synthesized from a pre-formed heterocycle, the substituents on N are

typically limited to primary and secondary alkyl groups (including benzylic and allylic

groups) [12,13]. If aryl or t-butyl substituents on N are required, the NHC can be constructed

from the parent amine(s) via appropriate ring formation pathways [14-16]. For NHCs bearing

aryl substituents on N, the NHC can also be obtained via routes involving N-arylation of a

OO

Cl OEt

1) R-NH2

2) R'-NH2 OO

NH HN

LiAlH4

R R'

NH HNR R'

NH4XN N R'R X

OO

2 R-NH2

NN RRH2CO

HXN N R'R X

NaBH4

H2CO

R-NH2

NH4X

H3PO4

R'-X

N NRN NH

R-X

R N N NR'

catalyst

Et3N

HC(OEt)3

N N

NR R'R-X or R-NH2

Convenient precursorsto incorporate carbohydrate moieties

N N R'R

N N R'R

N N

NR R'

imidazolidinylidene

imidazolylidene

1,2,3-triazolylidene

1

2

3

45

45

1 3

2

1

2 3

45

7

pre-formed heterocycle, for example reactions catalysed by Cu or Pd [17-19]. The Cu(I)-

catalyzed Huisgen 1,3-dipolar cycloaddition (“click”) reaction provides easy access to a great

variety of triazoles [20,21].

From Scheme 1, it is apparent that carbohydrates functionalised with halide or amino groups

(highlighted in dashed boxes) are potential building blocks for NHCs that have pendant

carbohydrate groups. For a comprehensive monograph of carbohydrate synthesis see Stick

and Williams [22].

Methods for synthesis of NHC-metal complexes have been reviewed [23,24]. Briefly, the

most common methods for synthesizing NHC-metal complexes fall into the following three

classes.

1. The free carbene method [25]. In this method, the azolium salt is deprotonated by

a strong base, and the resulting NHC is then allowed to react with a metal source

to form the metal-NHC complex. This method is limited to azolium salts that do

not contain base-sensitive functional groups and requires rigorously dry and

oxygen-free conditions, and so is not widely used.

2. The in situ deprotonation method. Here, the azolium salt is allowed to react with a

mild base in the presence of a metal source. During this procedure the NHC is

generated in situ and quickly binds to the metal centre, forming the target complex.

This method can be more tolerant of functionality than the free carbene method

and does not require rigorously dry conditions. The base can either be derived

8

from the metal source (e.g., acetate from Pd(OAc)2 [26], O2- from HgO [27]) or it

can be added separately.

3. The Ag-NHC transfer method [28]. Here, a labile Ag-NHC complex is prepared

first, and then used to transfer the NHC ligand to a second metal. Metals that can

be used as sources of NHCs in NHC-transfer reactions include Hg and W, but Ag

is by far the most commonly used. Ag-NHC complexes are easily prepared from

azolium salts and Ag2O, and when used in NHC-transfer reactions with halide

complexes of other metals, a driving force is precipitation of the silver halide. The

Ag-NHC transfer method is simple, fast, proceeds under very mild conditions, and

is tolerant of a wide range of functionality in the azolium salt/NHC, and the final

product is often easy to separate and purify, thus is very popular amongst NHC

chemists.

Applications of these methods to the synthesis of carbohydrate-containing NHC-metal

complexes are outlined in subsequent sections in this review.

3. Carbohydrate N-heterocyclic carbene complexes in catalysis

N-Heterocyclic carbenes have attracted interest as nucleophilic catalysts in, for example,

polymerizations [29], but the main interest of NHCs in catalysis focuses on their use as

ligands in catalytic metal complexes [30]. As ligands, NHCs have similar properties to

phosphines, such as the ability to stabilise metals in various oxidation states [31], but NHC

complexes have some significant advantages over phosphine complexes. For example, NHCs

tend to bind metals more strongly than phosphines, and resist dissociation from the metal

centre, so NHC-metal complexes tend to be more robust than analogous phosphine

9

complexes [31,32]. Moreover, whereas phosphines are typically air-sensitive and can be

difficult to synthesize and purify, especially when functionalized, NHCs are readily

accessible via azolium salts, which are usually crystalline. Consequently, compared with

phosphine complexes, NHC complexes offer greater opportunities for exploration of

structure-activity relationships. Examples of reactions that have been catalyzed by NHC

metal complexes include, but are not limited to, ruthenium-catalyzed olefin metathesis,

platinum-catalyzed olefin hydrosilylation, palladium catalyzed Mizoroki-Heck reactions,

Suzuki-Miyaura reactions and umpolung reactions [33].

With these advantages in mind, chemists sought to exploit additional functionality in NHCs

as catalysts. In early attempts to employ NHC metal complexes in asymmetric catalysis,

chiral functional groups were incorporated as N-substituents or into the backbone of NHCs.

The challenge in this area is the choice of a cheap yet effective chiral scaffold to attach to the

NHC [34]. Certain carbohydrates have potential as suitable chiral scaffolds as they are cheap,

offer structural diversity, and are readily available compounds with well-defined chirality.

Studies of NHCs containing carbohydrates in order to incorporate a chiral motif into the

ligand have already emerged in the field of catalysis [35].

3.1 Carbohydrate NHC metal complexes in stereoselective catalysis

In 2007, Nishioka and coworkers synthesized an imidazolium salt 1 (Scheme 2) having an

acetylated β-D-glucopyranose moiety as an N-substituent, via an SN2 reaction of α-D-

glucopyranosyl bromide with 1-methylimidazole. Salt 1 was converted into an Ag-NHC

complex 2 and a subsequent transmetallation reaction gave the 1,2,3,4,5-

pentamethylcyclopentadienyl (Cp*) NHC-iridium complex 3 (Scheme 2) [36]. An ESI-MS

study of 3 [36] in methanol solution found no evidence for cleavage of the bond linking the

10

carbohydrate moiety and the NHC nitrogen atom, suggesting that the carbohydrate-NHC has

similar stability to other NHCs [36]. X-ray diffraction studies showed that the Ir-NHC

binding in 3 was similar to that seen in the closely-related complexes 4 [37] and 5 [38],

which are catalysts for C-H activation and Oppenauer oxidation of alcohol. In the solid state,

3 adopts a conformation in which the Cp*IrCl2 moiety is oriented towards the α-face of the

carbohydrate moiety, and NOESY NMR studies showed that this conformation is retained in

solution [36]. The observation that the conformation seen in the solid state persists in solution

suggests that the steric asymmetry in the carbohydrate moiety might influence the

stereochemical course of reactions at the metal centre. Complex 3 promoted H-D exchange

reactions [36], but its use for asymmetric catalysis has not been reported.

Scheme 2.

Also in 2007, Glorius and coworkers synthesized the imidazolium salt 6 bearing a pivalate-

protected D-galactopyranose as one of the N-substituents (Scheme 3), as well as imidazolium

salts 7 and 8, bearing acetate-protected and unprotected D-glucopyranosyl substituents,

respectively (Scheme 3). When AgNO3 was used instead of AgOTf to promote the

11

displacement of bromide from the α-D-glucopyranosyl bromide starting material, the

imidazolium salt 7 was obtained as a 3:7 mixture of α and β anomers [39]. Compound 7

served as a precursor to carbohydrate-bearing Ag- and Pd-NHC complexes 9 and 10 (Scheme

3); attempts to synthesize 10 directly from 7 and a palladium source were unsuccessful [39].

Scheme 3.

While both groups demonstrated that carbohydrate-functionalized NHC complexes could be

synthesized and were robust enough to serve as catalysts [36,39], their use as catalysts in

stereoselective synthesis has not yet been reported. However, Glorius and coworkers [40-42]

explored the use of 1,3-bis(2,4,6-trimethylphenyl)imidazol-2-ylidene (IMes) and various

carbohydrate-NHCs as catalysts for "conjugated umpolung" annulation [43-45] reactions

O

PivO

PivOPivO

OPiv

Br

NN Mes

AgOTf, CH3CNNN Mes

O

PivO

PivOPivO

OPiv

6

OAcO

AcOOAc

OAc

N N Mes OTf

7 (β-anomer), 31%

OHO

HOOH

OH

N N Mes OTf

8 (β-anomer), 49%

KCN

OAcO

AcOAcO

OAc

Br

NN Mes

CH3CN

AgOTf

AgOTf, DBUOAcO

AcOOAc

OAc

N N

Ag

Mes

N NMesOAcO

OAc

OAc

AcO

CH2Cl2, rt

17h

OAcO

AcOOAc

OAc

N N

PdCl Cl

Mes

N NMesOAcO

OAc

OAc

AcO

40%

10

9

THF, rt

OTf

CH3OH

OTf

PdCl2(PhCN)2

12

(Scheme 4). The carbohydrate-NHCs 6-8 (structures shown in Scheme 3) were generated in

situ by treatment of the corresponding imidazolium salts with base, but could not be isolated.

While the NHCs served as catalysts and promoted conjugated umpolung annulation in good

yield, the stereoselectivities of the reactions were modest.

Scheme 4.

Keitz and Grubbs synthesized Ru complexes 11 and 12 from similar imidazolium-based

precursors to those used by Glorius and coworkers [46]. These complexes (Scheme 5) were

evaluated as catalysts for ring-opening metathesis polymerization, ring-closing metathesis,

cross-metathesis, and asymmetric ring-opening cross-metathesis reactions at 40 °C. The

results (examples are included in Scheme 6) were compared with those obtained for "non-

carbohydrate" ruthenium-based catalysts 13-15 at 30 °C.

H

O

+ R

O10 mol% catalyst

O

O

Ph Ph

CF3

O

O

Ph CF3

Ph

+

60 mol% DBU

THF, room T

major minor

catalyst

IMes

6

7

8

yield (%)

84

81

79

83

diastereomeric excess (%)

33

46

68

52

13

Scheme 5.

The catalytic activities of 11-15 were evaluated for ring-opening metathesis polymerization

(ROMP) of cyclooctadiene (Scheme 6). At 100:1 monomer to catalyst ratio, 11 and 12

showed similar activity, and although initiation was slow, both catalysts promoted >95%

conversion within 2 h at 30 °C. This activity was lower than that of 14 but similar to that of

second generation Grubbs catalyst 15 and far superior to that of first generation Grubbs

catalyst 13.

Curiously, for ring-closing metathesis (RCM), initiation of catalysis of was more rapid and

the conversion attained was higher using the D-galactose-derived complex 12 compared with

the D-glucose-derived catalyst 11. The activities of the carbohydrate-based catalysts 11 and

12 were on a par with those of the non-carbohydrate catalysts 13-15. The conversions

attained in a cross-metathesis reaction were also similar using either the carbohydrate-based

catalysts 11 and 12 or the non-carbohydrate catalysts 13-15. Interestingly, cross-metathesis

reactions conducted using the carbohydrate-based catalysts had poorer E/Z stereoselectivity,

whereas the non-carbohydrate catalysts 13-15 afforded a significant preference for the

product of E stereochemistry. Complexes 11 and 12 were evaluated in a number of

OAcOAcO

AcO

OAc

N N

RuCl

MesCl

Ph

O

AcOAcO

OAc

N N

RuCl

MesCl

Ph

AcO

12

PCy3 PCy3

11

PCy3

RuCl

Cl

Ph

13

PCy3

14

RuCl

Cl

Ph

15

NNMes Mes

PCy3

RuCl

Cl

Ph

NNMes Mes

PCy3

14

asymmetric ring-opening cross-metathesis reactions, but again showed poor E/Z

stereoselectivity and poor to modest enantioselectivity.

Scheme 6.

More recently, in 2014, Bower and Galan reported a synthetic route from 2-amino-2-

deoxysugars to imidazolium and imidazolinium salts with monosaccharide units as N-

substituents (Scheme 7). In contrast to earlier studies in which the carbohydrate units were

connected to the imidazol(in)ium ring at the anomeric position, in this study carbohydrates

were attached to the ring via the C2 position. The nitrogen atoms on the 2-amino-2-deoxy-

sugars were incorporated into the imidazol(in)ium ring [47]. The imidazol(in)ium salts were

converted into Rh-NHC complexes 16a-e and 17a-c (Scheme 7) that were catalysts for

reduction of ketones by diphenylsilane (Scheme 8).

15

Scheme 7.

Scheme 8.

16

These catalysts promoted reduction of unsymmetrical ketones in yields up to ca. 96% and

with ee's up to 62%. There was no significant difference between results obtained using

complexes incorporating saturated (17) or unsaturated (16) NHCs. It is notable that catalysts

containing the simplest NHCs (those derived from methyl 2-amino-2-deoxy-3,4,6-tri-O-

methyl-β-D-glucopyranose) gave the best results, with bulkier alkoxy groups leading to lower

yields and lower enantioselectivity. Amongst the series of catalysts 17a-c, the catalyst in

which the carbohydrate group is connected via an equatorial bond to the NHC unit (17a),

showed superior activity (both in terms of enantioselectivity and yield obtained). This result

is promising, showing that the carbohydrate NHC is able to induce chirality at the active site

of the Rh complex. However, the achievement of only modest enantioselectivity, and the fact

that larger substituents on the chiral (carbohydrate) component actually led to reduced

enantioselectivity suggests that the asymmetry associated with the carbohydrate group may

be too far from the catalytic metal centre to impart strong chiral induction.

In summary, some synthetic pathways to metal complexes of NHC ligands incorporating

carbohydrate units have been established. Synthesis, characterization, and (in some cases)

catalytic activity of these complexes have been compared with their non-carbohydrate

analogues. The catalytic studies so far are based on a limited range of reactions and

uncomplicated substrates. The overall catalytic activities of complexes containing

carbohydrate-NHC ligands are comparable to those reported for non-carbohydrate catalysts,

but in experiments to achieve chiral induction, the enantioselectivity achieved was not

impressive. These frustrating results perhaps discouraged further studies of the use of

carbohydrate-NHC complexes in catalytic reactions involving more complicated substrates.

The low enantioselectivity might be due to the carbohydrate motif being too far away from

the active site of the catalyst (metal centre). This problem might be ameliorated by

17

incorporating carbohydrate groups into chelating ligands that can more efficiently impose

chirality on the catalytic centre. For example, it might be beneficial to investigate complexes

incorporating more complicated carbohydrate structures (e.g., oligosaccharides), or

carbohydrate derivatives with different degrees of conformational flexibility, despite the

increased synthetic complexity.

3.2 Carbohydrate NHC complexes as water soluble catalysts

Incorporation of carbohydrate groups into NHC ligands is an efficient way of increasing

hydrophilicity, potentially leading to water-soluble catalysts. Towards this aim, several

studies have been conducted to use NHC carbohydrate complexes as catalysts of reactions in

water [35].

Lin and coworkers have reported the synthesis of a series of Ag and Pd complexes

incorporating D-glucopyranoside-NHC ligands (Scheme 9). The carbohydrate moieties were

connected to the N atoms of the NHC via the C6 position [48]. An interesting feature of the

synthesis was the demonstration that after the metal-NHC complexes were formed, it was

possible to deprotect the carbohydrate moieties (i.e., cleave the benzoate esters) under

strongly basic conditions by treatment with NaOMe, without disruption of the NHC-Pd

bonding.

18

Scheme 9.

Scheme 10.

Compared with Pd(OAc)2, compounds 21a-c showed better catalytic activity toward the

Suzuki-Miyaura coupling reaction in water (Scheme 10), especially the coupling of o-

bromobenzoic acid with phenylboronic acid at 0.05 mol% catalyst loading, for which yields

of 88%-95% were obtained. Interestingly, the catalytic efficiency increases with the length of

19

the alkyl substituent on the imidazolylidene ligand. Compound 21a, possessing a methyl

substituent on the imidazolylidene ligand, was not stable and showed consistently lower

catalytic activity than 21b and 21c.

Li and coworkers reported the results of catalyst recycling in the Suzuki-Miyaura cross-

coupling reaction catalyzed by Pd(OAc)2/DABCO/PEG-400 [49]. They found that 21a and

21b could be recycled in the same way. Although catalytic activity started to decline when

the catalyst was used for the third time due to decomposition, catalyst stability was enhanced

by the presence of tetrabutylammonium bromide (TBAB), in which case no significant loss

of activity was noted when catalysts were reused up to three times [48]. An interesting aspect

of this study is the demonstration of the coexistence of two different conformations of the

palladium complexes 20 in solution as revealed by two sets of 1H NMR signals (see Section

5).

In 2014 Nishioka and coworkers synthesized an interesting imidazolium salt 22 (Scheme 11)

that was composed of two NHCs and a triazole moiety incorporating a tetraacetyl-D-

glucopyranosyl moiety as a carbohydrate substituent. This ligand was synthesized using

"click" chemistry, and served as a precursor of Pd complex 23 [50].

20

Scheme 11.

Scheme 12.

Compound 23 served as a catalyst for a Suzuki-Miyaura cross-coupling reaction between

phenylboronic acid and 4'-bromo- or 4'-chloroacetophenone in water (Scheme 12). Under the

reaction conditions employed, the presence of K2CO3 resulted in deprotection of the

glucopyranosyl substituent (i.e., base-catalysed hydrolysis of the acetate groups), which then

enhanced the water solubility of the catalytic species. In the reaction between 4'-

bromoacetophenone and phenylboronic acid, the catalyst promoted formation of the coupling

product (4-phenylacetophenone) in 98% at a catalyst loading of 10-2 mol%, and 76% at a

catalyst loading of 10-3 mol%. The catalyst was less effective in promoting Suzuki-Miyaura

cross-coupling of 4'-chloroacetophenone and phenylboronic acid, with catalyst loadings of

0.5 mol% being required to achieve yields of 56% after extended time periods.

21

While these studies by the groups of Lin and Nishioka demonstrated the potential of

carbohydrate-NHC complexes as catalysts for reactions in water, the spectrum of catalytic

activity has scarcely been explored. It will be beneficial to test these and related complexes in

more complicated catalytic systems involving more highly functionalised substrates in future

studies, and benchmark activities against those of existing catalysts not incorporating

carbohydrates [33].

4. Carbohydrate NHC complexes in medicinal chemistry.

Mostly studied for their catalytic activity, NHC metal complexes have also exhibited

biological activities, both in vitro and in vivo. Complexes of form [Au(NHC)2]+ have been

reported to exhibit anti-cancer activities and were shown to cause mitochondrial membrane

permeabilisation and induce apoptosis in cancer cells [51,52]. The anti-fungal and anti-

bacterial activity of (NHC)AuCl complexes have been reported by Günal et al. [53], and their

compounds had minimum inhibitory concentration (MIC) values lower than 12.5 µg/mL

towards S. aureus and E. facecalis bacteria and the fungi C. albicans and C. tropicalis.

Biological activity (anti-cancer, anti-bacterial, anti-fungal, etc) has also been reported for

NHC complexes of other metals, e.g., Ni [54], Ru [55,56], Ag [57,58], Cu [59], Pd [60,61].

Unlike NHCs, carbohydrates originally attracted the attention of organometallic chemists due

to their roles in a variety of biological signalling and recognition processes. Their widespread

biological importance made carbohydrates interesting starting materials for the synthesis of

biologically active metal complexes. For the purposes of discovering novel classes of metal

drugs, many carbohydrate ligands have been designed, synthesized and attached to different

metal centres. There are many reviews [8-11] on this topic, including a detailed book chapter

covering the research up to 2013 [62]. Despite a number of studies of NHC-metal complexes

22

and carbohydrate-metal complexes in the bio-related research fields mentioned above, the

medicinal properties of metal complexes containing both NHC ligands and carbohydrates are

rarely discussed. Here we review studies of such compounds.

4.1 N-Heterocyclic carbene analogues of Auranofin

Auranofin (Fig. 2) is a well-known anti-arthritic Au(I) drug containing a carbohydrate ligand.

It has been marketed under the brand name Ridaura® for three decades [63]. Auranofin has

been known to show anti-cancer activity [64] and it is currently the subject of at least three

clinical trials concerning a range of cancers [65-67]. Auranofin has attracted a great deal of

interest because it displays a number of interesting biological activities including activity

against some difficult to treat cancers, such as osteosarcoma [68,69].

To date, the anti-cancer activity of Au-based drugs is largely believed to be due to the ability

of Au to bind to the selenocysteine residue at the active site of thioredoxin reductase (TrxR).

TrxR, found in both mitochondria and in the cytosol, is a key player in processes that protect

cells from effects of oxidative damage. The biological pathway involving TrxR is

summarised in Fig. 2 [70,71].

Fig. 2. Structure and proposed mode of action of Auranofin.

OAcO

AcOOAc

OAc

S Au P

NADP+

NADPH

+H+ TrxRS

Se

TrxRSH

SeTrx-S2

Trx-(SH)2Protein-S2

Protein-(SH)2

Auranofin

Se Au PEt3TrxR

SH

oxidative stress

X Xreactiveoxygen species

23

Critical to the functioning of TrxR are cysteine and selenocysteine residues at the active site.

In the reduced state, these residues exist in the thiol (–SH) and selenolate (–Se-) forms, and

when TrxR reduces its target protein (thioredoxin, Trx), these residues form a S–Se bond

(analogous to the disulfide bond). Thioredoxin goes on to reduce disulfide bonds in other

proteins that were formed by reactions with oxidative species in cells. After binding to

Auranofin, TrxR loses its ability to repair proteins that have suffered oxidative damage, thus

making cells susceptible to oxidative damage, and eventually the cells undergo apoptosis

(controlled cell death) [70].

The carbohydrate thiolate ligand of Auranofin is labile and can be displaced by other

biological thiols (cysteine, albumin, etc.) [72-74]. It has been proposed that such exchange

reactions with serum albumin are important in distributing Au via the bloodstream [73].

Baker and coworkers were the first to prepare Auranofin analogues 24a-e (Scheme 13),

which retained the tetraacetylthioglucopyranose moiety but incorporated an NHC ligand in

place of the PEt3 group [75]. The purpose of this study was to obtain complexes that might

mimic Auranofin due to the similarity between NHCs and phosphines as metal-binding

ligands, while exploiting the relative ease of synthesis of imidazolium salts to access

complexes of a range of NHC ligands having different steric and lipophilic characteristics.

Interestingly, whereas cationic [Au(NHC)2]+ complexes are believed to have anti-

mitochondrial activity and cause swelling of isolated mitochondria [76], preliminary

investigations of NHC analogues of Auranofin did not reveal any mitochondrial-swelling

behaviour for these neutral complexes [77].

24

Scheme 13.

Scheme 14.

In 2011, Nolan and coworkers screened a wide variety of Au(I) complexes, including neutral

Au(I) complexes {(NHC)AuX (X = halide or a carboxylate-bound amino acid)} and cationic

Au(I) complexes {[Au(NHC)2]+, [Au(NHC)(PR3)]

+} against human breast cancer cell line

MDA MB231 and human prostate cancer cell line LNCaP [78]. Amongst the complexes

tested, the NHC Auranofin analogue 25 and the cationic Au(I) complex 26 (Scheme 14),

which both incorporated the same sterically-encumbered NHC ligand, showed potential anti-

cancer activities (IC50 < 10µM). The authors also reported that generally cationic Au(I)

complexes (IC50 < 1µM) were more effective than neutral analogues bearing the same NHC

ligand (2µM < IC50 <10µM) against both cell lines. It was suggested that this phenomenon

25

was perhaps because lipophilic cations are more likely to accumulate in mitochondria, thus

affecting the overall activity of the drug.

Scheme 15.

Later, Tacke and coworkers tested Auranofin analogues 27-29 (Scheme 15) and their

(Cl,Br)Au(NHC) precursors against Caki-1 (renal) and MCF7 (breast) cancer cells in

vitro [79]. Compounds 27 and 28 exhibited low IC50 values (48 h) against both cell lines,

while 29 was active against the MCF7 cell line but not the Caki-1 cell line. This study

concluded that, generally, (Ac4GlcS)Au(NHC) complexes are more effective than their

(Cl,Br)Au(NHC) precursors. While partitioning coefficients were not reported, the four

aromatic substituents on the NHC ligand would clearly cause complexes 27-29 and their

(Cl,Br)Au(NHC) precursors to be highly lipophilic and poorly soluble in biological media. It

was suggested that the carbohydrate moiety in 27-29 might have increased the solubility of

these complexes, thereby accounting for their improved activity compared with the

(Cl,Br)Au(NHC) complexes.

Recently, Casini’s group synthesized a new Auranofin analogue (32a and 32b, Scheme 16)

containing an NHC ligand incorporating a coumarin derivative as a fluorescent

appendage [80]. These fluorescent NHC complexes, along with their precursor imidazolium

salt (30) and (NHC)AuCl complex (31) and cisplatin, were tested against the human cancer

26

cell lines A2780 (ovarian), MCF-7 (breast), and A549 (lung), with (non-cancerous) human

embryonic kidney HEK-293T cells as a control. The precursors 30 and 31 showed much

lower activity compared with cisplatin, while the glucosethiolate complex 32b showed only

weak activity, and only against MCF-7 cells. The tetraacetylglucosethiolate complex 32a,

however, exhibited cytotoxicity against both A2780 and MCF-7 cells and was similar in

efficacy to cisplatin against MCF-7 cells.

Scheme 16.

The selectivity of these complexes to affect cancer cells rather than the non-cancerous HEK-

293T cells was poor, although, interestingly, the imidazolium salt 30 appeared to be more

toxic to the cancer cell lines than to the HEK-293T cells. Due to the lack of activity of the

glucosethiolate complex 32b, the authors concluded that the presence of the glucopyranosyl

moiety did not enhance the uptake of the compound via the GLUT-1 transporter.

27

In studies [80] using isolated and purified enzymes, complexes 32a and 32b were shown to

inhibit both cytosolic and mitochondrial thioredoxin reductase (TrxR), the latter more

strongly, but were less effective inhibitors than Auranofin. However, 32a and 32b, as well as

Auranofin, showed little or no inhibition of glutathione reductase and glutathione peroxidase.

Interestingly, the inhibitory profile demonstrated by 32a and 32b against isolated TrxR and

GR is similar to that seen previously for cationic [Au(NHC)2]+ complexes [51,52]. When

A2780 cells were incubated with 32a, fluorescence microscopy studies [80] showed that the

fluorescent tag was taken up by cells and was accumulated in the nuclei. Unfortunately, the

limited fluorescence of the complex required concentrations of 20 µM for efficient

visualization, and at these concentrations the complex may have rapidly led to cell death.

More detailed studies are needed to determine whether the complex 32a remains intact

throughout the experiment, or whether the NHC-coumarin unit and/or the

glucopyranosylthiolate moiety become displaced from gold by other ligands in the biological

milieu.

In 2015, Casini's group published a new study of series of homobimetallic complexes 33-35

that each contain two Au(I) centres linked by an NHC-phosphine, but which contain 0, 1, or 2

carbohydrate moieties (Scheme 17) [81]. These complexes showed inhibition of isolated

TrxR and showed some anticancer activity in vitro (Scheme 17). It is notable that addition of

the sugar component did not increase anti-cancer activity; for the lung cancer cell line A549,

the addition of the carbohydrate moieties actually decreased the cytotoxicity.

28

Scheme 17

29

The same group also synthesized Au(I) complexes 36 and 37, which served as precursors of

several heterobimetallic complexes, including 38. Heterometallic Au(I)-Ru(II) complex 38

showed activity only against the A2780 ovarian cancer cell line. Based on these data, it is

difficult to draw conclusions about the role and significance of the carbohydrate moiety in the

biological activity of these compounds.

4.2 Incorporating a Carbohydrate onto Imidazolidene Ligands in Metal Complexes

In 2010, Skander et al. reported the synthesis of a series of NHC platinum complexes

containing mutually trans amine and NHC ligands bound to a PtI2 core [82]. Amongst this

series was 39 (Scheme 18), which incorporated a tetraacetylglucopyranosyl substituent on the

NHC ligand. Complex 39 was synthesized via a transmetallation route involving an Ag-NHC

complex as an intermediate, generated in situ. Many of the trans-(NHC)(amine)PtI2

complexes exhibited reasonable cytotoxic activity against both cisplatin-sensitive and

cisplatin-resistant cell lines. It was suggested the activity against cisplatin-resistant cell lines

might be due to the trans geometry the complexes, which would result in formation of Pt-

DNA adducts different to those resulting from cisplatin, and which would be less susceptible

to the DNA repair mechanisms present in cells that had already acquired resistance to

cisplatin [83]. Sadly, no biological studies of 39 were reported, so whether or not the

presence of the carbohydrate moiety affects the cytotoxicity of trans-(NHC)(amine)PtI2

complexes is unknown.

30

Scheme 18.

5. Structural complexity in carbohydrate NHC transition metal complexes

As mentioned earlier, Lin’s group found that the NMR spectra of solutions containing Pd

complexes 20a-c (Scheme 8) each showed two distinct sets of signals, a result that was

attributed to the existence of two distinct conformational isomers in solution [48]. The

authors considered the possibility of isomers in which the two NHC groups are mutually cis

or trans about the Pd centre, but this possibility would seem unlikely given that the 13C NMR

signals due to the Pd-C carbons in each form had very similar chemical shifts. Instead, the

authors concluded [48] that the distinct NMR spectra corresponded to trans-syn and trans-

anti isomers (conformations) (Fig. 3). Evidently, steric interactions between the NHCs

(carrying a bulky carbohydrate substituent) and the two bromido ligands attached to the Pd

centre are substantial, so that rotation of the NHC units about the Pd-C axis is sufficiently

slow that the two conformations do not interconvert on the NMR timescale.

31

Fig. 3. Isomerism in Pd-NHC complexes 20a-c

Interestingly, Lin and coworkers were able to isolate trans-syn 20a (R = Me) and trans-anti

20b (R = Et) and characterised both by X-ray diffraction studies. In the crystal structures of

20a and 20b, the NHC planes are (very) roughly parallel to one another, and (very) roughly

orthogonal to the Pd coordination plane. This structural motif is characteristic of many trans-

M(X)2(NHC)2 systems. Glorius and coworkers reported a similar motif for trans-

PdCl2(NHC)2 complex 10. The crystal structure determined for 10 contained two

independent Pd complexes, but both were trans-anti forms [40-42].

For solutions containing 10, in contrast to the situation for 20a-c, only a single set of signals

was seen in the 1H and 13C NMR spectra [40-42]. It may be that in 10, the PdCl2 core is

sufficiently small to permit rotation of the NHC ligands about the Pd-C axis that is rapid on

the NMR timescale, so that trans-syn and trans-anti forms interconvert rapidly and present a

single set of averaged signals in the NMR spectra. An alternative possibility [48] is that the

NHC groups inherently provide sufficient steric bulk around the metal centre in 10 to prevent

formation of the trans-syn form, whereas the presence of the methylene linking group

between the carbohydrate and imidazolyl units in 20a and 20b allows the bulky carbohydrate

to adopt positions that permit the formation of both the trans-syn and trans-anti forms.

N

OBzOBzO

BzO

N

OCH3

R

OBzOBzO

BzOOCH3

N

N

R

Pd

Br

Br

OBzO

CH3O N

N

OBzOBzO

BzOOCH3

N

N

R

Pd

Br

BrR

trans-syn trans-anti

isomer ratio

87 : 13

56 : 44

77 : 23

BzOBzO

R = Me

R = Et

R = n-Bu

20a

20b

20c

32

Examination of the PdBr2(NHC)2 analogues of 10, the PdCl2(NHC)2 analogues of 20a-20c,

and variable temperature NMR studies, would shed light on this interesting subject.

A more recent study suggests that the isomerism seen by Lin and coworkers was indeed a

consequence of the bulk of the ancillary ligands (i.e., the PdBr2 core) in 20a-c. Nishioka and

coworkers reported Ni(II) complexes 40 and 41 that were analogues of the Pd(II) complex 10

studied by Glorius’ group. Unlike 10, both 40 and 41 showed two sets of NMR signals in

solution, a result consistent with the presence of trans-syn and trans-anti isomers, and the

presence of the bulky NiBr2 core prevents rapid rotation of the NHC units about the Ni-C

bonds, so that these two forms remain distinct on the NMR timescale. In an X-ray diffraction

study, 40 was found to exist in the trans-syn conformation, while 41 adopted the trans-anti

conformation [84]. It was suggested that in the latter case, interactions between axial

hydrogens of the glucopyranosyl substituent of one NHC and the π-system of the benzyl

substituent of the other NHC unit may have contributed to the preference for the trans-anti

conformation.

Scheme 19.

33

Nishioka and coworkers reported variable temperature NMR studies of Ni(II) complexes 40

and 41 (Scheme 19). The NMR spectra showed complicated temperature dependencies,

which were interpreted in terms of (i) a change in the relative amounts of syn and anti forms

with temperature and (ii) occurrence of more than one exchange process. They suggested that

in addition to the substantial rotation of NHC ligands about the Ni-C axis that results in syn /

anti exchange, for the syn form, a smaller twisting motion would result in interconversion of

diastereomers syn-Ra and syn-Sa (Fig. 4) that differ due to axial chirality about the C-Ni-C

axis, the effects of the latter only becoming manifest at lower temperatures. Additional

processes, such as rotation of the carbohydrate moiety about the carbohydrate-N bond, might

also have an effect on the NMR spectra.

Fig. 4. Equilibrium proposed to account for the temperature-dependence of NMR spectra of

40 and 41.

Nishioka’s group also demonstrated the first example of a diastereoselective synthesis of an

NHC complex induced by anomeric isomerism of a carbohydrate unit in a ligand. They

synthesized the imidazolium salts 42 (Scheme 20) containing a tetraacetylglucopyranosyl

34

moiety in either the α- or β-D-anomeric forms (S and R configurations respectively), and used

this imidazolium salt as a precursor to Rh(III) and Ir(III) complexes 43 and 44 (Scheme

20) [85]. In these complexes the metal centre is pseudo-tetrahedral and therefore a

stereocentre that can be assigned an R or S configuration. The carbohydrate moiety induced

chirality at the metal centre — for imidazolium salts containing the carbohydrate moiety in

the α-D-anomer (anomeric carbon of S configuration) the dominant configuration of the

metal centre in the resulting complex was S, and for imidazolium salts containing the

carbohydrate moiety in the β-D-anomer (anomeric carbon of R configuration) the dominant

configuration of the metal centre in the resulting complex was R. In each experiment, the

major diastereomer was isolated and the configuration of the metal centre was confirmed by

X-ray diffraction studies.

In compound 23 (Scheme 11), the chelating (NHC)2(triazole) ligand is not planar, but forms

a helical arrangement around the Pd centre. The chelating ligand can adopt M- or P-helicity,

and the presence of the chiral D-tetraacetylglucosyl moiety causes the two forms to be

diastereoisomers (Scheme 21) [50]. These two diasterisomers exist in a 1:1.3 ratio in solution

and give rise to distinct sets of signals in the 1H NMR spectrum at room temperature [50]. At

150 °C, the signals merge into a single set of sharp signals, indicating that the M- and P-

isomers interconvert rapidly on the NMR timescale at elevated temperatures.

35

Scheme 20.

Scheme 21.

36

Shi et al. synthesized an imidazolinium salt 45 containing carbohydrate moieties attached to

both nitrogen atoms and converted this precursor into a Rh complex 46 (Scheme 22) [86].

The NMR spectra of 45 indicated that the imidazolinium salt had C2 symmetry, which

indicates that the two carbohydrate moieties had the same stereochemical configuration (a C2

axis co-linear with the imidazolinium C2-H bond renders the two carbohydrate environments

equivalent). In 46, coordination of the NHC to the Rh(Cl)(COD) unit breaks the C2 symmetry,

but the two carbohydrate environments would be equivalent on the NMR timescale if the

NHC ligand were to undergo rapid rotation about the Rh-C axis. The observation that the

carbohydrate environments in 46 are not equivalent and NMR spectra are more complex

suggests that rotation about the Rh-C axis is slow on the NMR timescale.

Scheme 22.

37

6. Summary and Outlook

Combining NHCs and carbohydrates together as ligands in metal complexes is a challenging

yet potentially rewarding endeavour in organometallic chemistry.

In catalysis research, a carbohydrate moiety in the catalyst can provide a chiral motif and

potential water solubility. Thus, it might be a cheap and simple way of achieving both

asymmetric catalysis and catalysis in an aqueous medium. Preliminary results obtained for

some carbohydrate-NHC complexes as catalysts for reactions in aqueous media are

promising, with high yields being obtained in several cases. On the other hand, existing

studies of complexes with carbohydrate-NHCs showed little utility in asymmetric catalysis.

Their enantioselectivity was relatively poor compared with existing chiral catalysts,

presumably due to the carbohydrate motif being too far from the metal centre. Better

enantioselectivity might be achieved by catalysts incorporating more complex carbohydrate

moieties, or designed so that the carbohydrate moieties are directed more towards the metal

centre; investigation of asymmetric catalysis by complexes such as 46 would be interesting in

this regard.

In medicinal chemistry, the introduction of a carbohydrate moiety into NHC complexes

opens up the possibility for modulation of solubility and lipophilicity, and targeted delivery

of compounds to particular cells (by commandeering, for example, glucose transporters). For

such applications, to prevent displacement of the carbohydrate moiety from the metal by

other ligands (albumin, cysteine etc.) in the biological milieu, the carbohydrate moiety should

be an appendage to the strongly-bound NHC ligand, rather than bound to the metal directly as

(for example) a thiolate ligand. With appropriate molecular design, this approach can also

provide longer distances between the carbohydrate group and metal centre and its ancillary

38

ligands, which may be important for carbohydrate presentation for recognition by

transporters or other types of biological activity. The ease of modification of NHCs/azolium

salts via well-established procedures allows for further tuning of lipophilicity and reactivity

of the NHC-carbohydrate metal complex if need be. Results obtained from studies of the

biological activity of carbohydrate-NHC metal complexes that have been reported to date

have been mixed. At this stage, it is too early for firm conclusions to be drawn about the best

ways to incorporate carbohydrates or their derivatives into NHC metal complexes for

optimum catalyst design or to modulate specific biological targets of therapeutic interest.

Given the vast number of possible carbohydrate and carbohydrate-derived structures and the

number of potential applications, further work is required to unlock the full potential of this

exciting new field.

7. Acknowledgements

We thank the University of Western Australia and the University of Queensland for a UWA-

UQ Partnership Research Collaboration Award (to VF and MVB). Weihao Zhao thanks the

Government of the People's Republic of China for a travelling scholarship, and the National

Laboratory of Biochemistry and Pharmacy of Hebei (China) for additional financial support.

8. References

[1] L. Cavallo, C.S.J. Cazin, N-Heterocyclic Carbenes in Transition Metal Catalysis and Organocatalysis, Springer, New York, 2011.

[2] A.W. Salman, S. Budagumpi, R.A. Haque, Coord. Chem. Rev., 256 (2012) 1787-1830.

[3] K.M. Hindi, M.J. Panzner, C.A. Tessier, C.L. Cannon, W.J. Youngs, Chem. Rev., 109 (2009) 3859-3884.

[4] A. Kascatan-Nebioglu, M.J. Panzner, C.A. Tessier, C.L. Cannon, W.J. Youngs, Coord. Chem. Rev., 251 (2007) 884-895.

39

[5] L. Oehninger, R. Rubbiani, I. Ott, Dalton Trans., 42 (2013) 3269-3284.

[6] S.A. Patil, S.A. Patil, R.P. Rangappa, G.R. Balakrishna, M. Tacke, Future Med. Chem., 7 (2015) 1305-1333.

[7] A.R. Kapdi, I.J.S. Fairlamb, Chem. Soc. Rev., 43 (2014) 4751-4777.

[8] B. Gyurcsik, L. Nagy, Coord. Chem. Rev., 203 (2000) 81-149.

[9] T. Storr, C.L. Fisher, Y. Mikata, S. Yano, M.J. Adam, C. Orvig, Dalton Trans., (2005) 654-655.

[10] M. Gottschaldt, U. Schubert, Chem. Eur. J., 15 (2009) 1548-1557.

[11] Y. Mikata, S. Yano, Curr. Top. Med. Chem., 12 (2012) 145-157.

[12] C.T. Yang, H.M. Lee, S.P. Nolan, Org. Lett., 3 (2001) 1511-1514.

[13] P.L. Chiu, H.M. Lee, Organometallics, 24 (2005) 1692-1702.

[14] A.J. Arduengo, III, R. Krafczyk, R. Schmutzler, H.A. Craig, J.R. Goerlich, W.J. Marshall, M. Unverzagt, Tetrahedron, 55 (1999) 14523-14534.

[15] F.E. Hahn, M.C. Jahnke, T. Pape, Organometallics, 25 (2006) 5927-5936.

[16] H.M. Lee, J.Y. Zeng, C.H. Hu, M.T. Lee, Inorg. Chem., 43 (2004) 6822-6829.

[17] H. Chen, D. Wang, X. Wang, W. Huang, Q. Cai, K. Ding, Synthesis, (2010) 1505-1511.

[18] R.A. Altman, E.D. Koval, S.L. Buchwald, J. Org. Chem., 72 (2007) 6190-6199.

[19] A.H. Sandtorv, C. Leitch, S.L. Bedringaas, B.T. Gjertsen, H.-R. Bjørsvik, ChemMedChem, 10 (2015) 1522-1527.

[20] H.C. Kolb, M.G. Finn, K.B. Sharpless, Angew. Chem. Int. Ed., 40 (2001) 2004-2021.

[21] C.W. Tornøe, C. Christensen, M. Meldal, J. Org. Chem., 67 (2002) 3057-3064.

[22] R.V. Stick, S.J. Williams, Carbohydrates: The Essential Molecules of Life, 2nd ed., Elsevier Science, Amsterdam, 2009.

[23] P. Eduardo, Top. Organomet. Chem., 21 (2007) 83-116.

[24] F.E. Hahn, M.C. Jahnke, Angew. Chem. Int. Ed., 47 (2008) 3122-3172.

[25] W.A. Herrmann, L.J. Goossen, M. Spiegler, J. Organomet. Chem., 547 (1997) 357-366.

[26] W.A. Herrmann, C.-P. Reisinger, M. Spiegler, J. Organomet. Chem., 557 (1998) 93-96.

[27] H.-W. Wanzlick, H.-J. Schönherr, Angew. Chem. Int. Ed., 7 (1968) 141-142.

[28] H.M.J. Wang, I.J.B. Lin, Organometallics, 17 (1998) 972-975.

[29] N. Marion, S. Díez-González, S.P. Nolan, Angew. Chem. Int. Ed., 46 (2007) 2988-3000.

40

[30] S. Díez-González, N. Marion, S.P. Nolan, Chem. Rev., 109 (2009) 3612-3676.

[31] M.N. Hopkinson, C. Richter, M. Schedler, F. Glorius, Nature, 510 (2014) 485-496.

[32] D.M. Khramov, V.M. Lynch, Organometallics, 26 (2007) 6042-6049.

[33] H.D. Velazquez, F. Verpoort, Chem. Soc. Rev., 41 (2012) 7032-7060.

[34] A. Monney, M. Albrecht, Coord. Chem. Rev., 257 (2013) 2420-2433.

[35] L.A. Schaper, S.J. Hock, W.A. Herrmann, F.E. Kuhn, Angew. Chem. Int. Ed., 52 (2013) 270-289.

[36] T. Nishioka, T. Shibata, I. Kinoshita, Organometallics, 26 (2007) 1126-1128.

[37] R. Corberán, M. Sanaú, E. Peris, J. Am. Chem. Soc., 128 (2006) 3974-3979.

[38] F. Hanasaka, K. Fujita, R. Yamaguchi, Organometallics, 25 (2006) 4643-4647.

[39] F. Tewes, A. Schlecker, K. Harms, F. Glorius, J. Organomet. Chem., 692 (2007) 4593-4602.

[40] C. Burstein, F. Glorius, Angew. Chem. , 43 (2004) 6205-6208.

[41] C. Burstein, S. Tschan, X. Xie, F. Glorius, Synthesis, 14 (2006) 2418-2439.

[42] W. Schrader, P.P. Handayani, C. Burstein, F. Glorius, Chem. Commun., 21 (2007) 716-718.

[43] S.S. Sohn, E.L. Rosen, J.W. Bode, J. Am. Chem. Soc., 126 (2004) 14370-14371.

[44] M. He, J.W. Bode, Org. Lett., 7 (2005) 3131-3134.

[45] S.S. Sohn, J.W. Bode, Org. Lett., 7 (2005) 3873-3876.

[46] B.K. Keitz, R.H. Grubbs, Organometallics, 29 (2010) 403-408.

[47] A.S. Henderson, J.F. Bower, M.C. Galan, Org. Biomol. Chem., 12 (2014) 9180-9183.

[48] C. Yang, P. Lin, F. Liu, I. Lin, Organometallics, 29 (2010) 5959-5971.

[49] J.H. Li, W.J. Liu, Y.X. Xie, J. Org. Chem., 70 (2005) 5409-5412.

[50] Y. Imanaka, H. Hashimoto, I. Kinoshita, T. Nishioka, Chem. Lett., 43 (2014) 687-689.

[51] J.L. Hickey, R.A. Ruhayel, P.J. Barnard, M.V. Baker, S.J. Berners-Price, A. Filipovska, J. Am. Chem. Soc., 130 (2008) 12570-12571.

[52] M.M. Jellicoe, S.J. Nichols, A.B. Callus, M.V. Baker, P.J. Barnard, S.J. Berners-Price, J. Whelan, G.C. Yeoh, A. Filipovska, Carcinogenesis, 29 (2008) 1124-1133.

[53] I. Özdemir, N. Temelli, S. Günal, S. Demir, Molecules, 15 (2010) 2203-2210.

[54] C.H. Wang, W.C. Shih, H.C. Chang, J. Med. Chem., 54 (2011) 5245-5249.

41

[55] E.S. Antonarakis, A. Emadi, Cancer Chemother. Pharmacol., 66 (2010) 1-9.

[56] J.M. Rademaker-Labkhai, D. Van den Bongard, D. Pluim, J.H. Beijnen, J.H.M. Schellens, Clin. Cancer Res., 10 (2004) 3717-3727.

[57] T.J. Siciliano, M.C. Deblock, K.M. Hindi, J. Organomet. Chem., 696 (2011) 1066-1071.

[58] S.A. Patil, J. Claffey, A. Deally, Eur. J. Inorg. Chem., 7 (2010) 1020-1031.

[59] C. Marzano, M. Pellei, F. Tisao, C. Santini, Anticancer Agents Med Chem, 9 (2009) 185-211.

[60] R.A. Haque, A.W. Salman, S. Budagumpi, A.A. Abdullah, A.M.S.A. Majid, Metallomics, 5 (2011) 760-769.

[61] M.L. Teyssot, A.S. Jarrousse, A. Chevry, Chem. Eur. J., 15 (2009) 314-318.

[62] Y. Mikata, M. Gottschaldt, Metal Complexes of Carbohydrate-targeted Ligands in Medicinal Inorganic Chemistry, in: T. Storr (Ed.) Ligand Design in Medicinal Inorganic Chemistry, John WIley & Sons, Ltd, Chichester, UK, 2014.

[63] W.F. Kean, W.F. Hart, W.W. Buchanan, Br. J. Rheumatol., 36 (1997) 560-572.

[64] E.R.T. Tiekink, Inflammopharmacology, 16 (2008) 138-142.

[65] ClinicalTrials.gov, Auranofin in Treating Patients With Recurrent Epithelial Ovarian, Primary Peritoneal, or Fallopian Tube Cancer. http://clinicaltrials.gov/ct2/show/NCT01747798, 2016 (accessed 17.01.2017).

[66] ClinicalTrials.gov, Sirolimus and Auranofin in Treating Patients With Advanced or Recurrent Non-Small Cell Lung Cancer or Small Cell Lung Cancer. http://clinicaltrials.gov/show/NCT01737502, 2016 (accessed 17.01.2017).

[67] ClinicalTrials.gov, Phase I and II Study of Auranofin in Chronic Lymphocytic Leukemia (CLL). http://clinicaltrials.gov/ct2/show/NCT01419691, 2016 (accessed 17.01.2017).

[68] E. Topkas, N. Cai, A. Cumming, M. Hazar-Rethinam, O.M. Gannon, M. Burgess, N.A. Saunders, L. Endo-Munoz, Oncotarget, 7 (2016) 831-844.

[69] C.A. Stiller, Cancer Treat. Rev., 33 (2007) 631-645.

[70] S. Talbot, R. Nelson, W.T. Self, Br. J. Pharmacol., 154 (2008) 940-948.

[71] K.E. Hill, G.W. McCollum, M.E. Boeglin, R.F. Burk, Biochem. Biophys. Res. Commun., 234 (1997) 293-295.

[72] F.C. Shaw, III, Chem. Rev., 99 (1999) 2589-2600.

[73] H. Kamel, D.H. Brown, J.M. Ottaway, W. E. Analyst, 102 (1977) 645-657.

[74] C.K. Mirabelli, R.K. Johnson, D.T. Hill, L.F. Faucette, J. Med. Chem., 29 (1986) 218-223.

42

[75] M.V. Baker, S.J. Berners-Price, P.J. Barnard, S.K. Brayshaw, J. Organomet. Chem., 690 (2005) 5625-5633.

[76] P.J. Barnard, M.V. Baker, S.J. Berners-Price, D.A. Day, J. Inorg. Biochem., 98 (2004) 1642-1647.

[77] J.L. Hickey, Gold(I)-carbene complexes as potential antimitochondrial agents, Honours Thesis, The University of Western Australia, 2003.

[78] J. Weaver, S. Gaillard, C. Toye, S. Macpherson, S.P. Nolan, Chem. Eur. J., 17 (2011) 6620-6624.

[79] F. Hackenberg, H. Müller-Bunz, R. Smith, W. Streciwilk, X. Zhu, M. Tacke, Organometallics, 32 (2013) 5551-6660.

[80] B. Bertrand, A. Almeida, E.P.M. Van der Burgt, M. Picquet, A. Citta, A. Folda, M.P. Rigobello, P. Le Gendre, E. Bodio, A. Casini, Eur. J. Inorg. Chem., (2014) 4532-4536.

[81] B. Bertrand, A. Citta, I.L. Franken, M. Picquet, A. Folda, V. Scalcon, M.P. Rigobello, L.P. Gendre, A. Casini, E. Bodio, J. Biol. Inorg. Chem., 20 (2015) 1005-1020.

[82] M. Skander, P. Retailleau, B. Bourrie, L. Schio, P. Mailliet, A. Marinetti, J. Med. Chem., 53 (2010) 2146-2154.

[83] U. Kalinowska-Lis, J. Ochocki, K. Matlawska-Wasowska, Coord. Chem. Rev., 252 (2008) 1328-1345.

[84] T. Shibata, S. Ito, M. Doe, R. Tanaka, H. Hashimoto, I. Kinoshita, S. Yano, T. Nishioka, Dalton Trans., 40 (2011) 6778-6784.

[85] T. Shibata, H. Hashimoto, I. Kinoshita, S. Yano, T. Nishioka, Dalton Trans., 40 (2011) 4826-4829.

[86] J. Shi, N. Lei, Q. Tong, Y. Peng, J. Wei, L. Jia, Eur. J. Inorg. Chem., (2007) 2221-2224.

Highlights

• It is relatively easy to synthesize NHC ligand with carbohydrate moieties attached.

• Carbohydrate groups impart water solubility to NHC metal complexes

• The complexes can serve as water soluble and/or asymmetric catalysts

• NHC analogs of Auranofin exhibit promising anti-cancer properties in vitro.