Easily Available, Amphiphilic Diiron Cyclopentadienyl Complexes Exhibit in Vitro Anticancer Activity in 2D and 3D Human Cancer Cells through Redox Modulation Triggered by CO Release

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

Easily Available, Amphiphilic Diiron Cyclopentadienyl

Complexes Exhibit in Vitro Anticancer Activity in 2D and 3D

Human Cancer Cells through Redox Modulation Triggered

by CO Release

Lorenzo Biancalana,^[a]Michele De Franco,^[b]Gianluca Ciancaleoni,^[a]Stefano Zacchini,^[c]

Guido Pampaloni,^[a]Valentina Gandin,*^[b]and Fabio Marchetti*^[a]

Abstract: A straightforward two-step procedure via single CO

removal allows the conversion of commercial [Fe₂Cp₂(CO)₄]

into a range of amphiphilic and robust ionic complexes based

on a hybrid aminocarbyne/iminium ligand, [Fe₂Cp₂(CO)₃{CN-

(R)(R’)}]X (R, R’=alkyl or aryl; X=CF₃SO₃or BF₄), on up to

multigram scales. Their physicochemical properties can be

modulated by an appropriate choice of N-substituents and

counteranion. Tested against a panel of human cancer cell

lines, the complexes were shown to possess promising

antiproliferative activity and to circumvent multidrug resist-

ance. Interestingly, most derivatives also retained a significant

cytotoxic activity against human cancer 3D cell cultures.

Among them, the complex with R=4-C₆H₄OMe and R’=Me

emerged as the best performer of the series, being on

average about six times more active against cancer cells than

a noncancerous cell line, and displayed IC₅₀values compara-

ble to those of cisplatin in 3D cell cultures. Mechanistic

studies revealed the ability of the complexes to release

carbon monoxide and to act as oxidative stress inducers in

cancer cells.

Introduction

introduction of few platinum complexes in clinical treatments

against various types of cancer.^[2]Then, the severe side effects

caused by limited selectivity, platinum toxicity and resistance

issues^[3]have fuelled an intense research for alternative drugs

based on other transition metals.^[4]In principle, iron is a

convenient choice, since it is a bioavailable element and

nontoxic in many forms;^[5]moreover, its earth abundancy has

made several cost-effective compounds available to the

researchers for the exploration of the reactivity and the

consequent development of new structural motifs. A variety of

mono-iron compounds has been investigated for the anticancer

potential,^[6,7]and substituted ferrocenes (ferrocifens) have been

regarded as promising drug candidates.^[8]The feasible Fe^+IIto

Fe^+IIIoxidation in the physiological environment and the

conjugation of suitable organic moieties to the robust sandwich

scaffold are responsible for a considerable activity, which is

basically related to the production of damaging metabolites

within cancer cells.^[9]

Transition metal complexes have unique characteristics which

constitute an indispensable potential in medicinal chemistry,

such as the availability of adjacent oxidation states of the metal

enabling redox reactions, the reactivity of the coordinated

organic fragments addressed by the metal centre, and a broad

range of possible geometries, stereochemical configurations

and kinetic behaviours.^[1]In the last century, the serendipitous

discovery of the cytotoxic properties of cisplatin preceded the

[a] Dr. L. Biancalana, Prof. G. Ciancaleoni, Prof. G. Pampaloni,

Prof. F. Marchetti

Department of Chemistry and Industrial Chemistry

University of Pisa

Via G. Moruzzi 13, I-56124 Pisa (Italy)

E-mail: fabio.marchetti1974@unipi.it

Homepage: https://people.unipi.it/fabio_marchetti1974/

[b] Dr. M. De Franco, Prof. V. Gandin

Following the findings on ferrocifens, some half-sandwich

compounds, both neutral-charged (Figure 1, structure I)^[10]and

cationic (structures II–III),^[10c,11]were recently assessed for their

cytotoxicity against various cancer cell lines, which might be

enhanced by the synergic effect provided by the release of the

carbon monoxide co-ligands.^[12]A critical aspect is that, in

general, ferrocifens and other organo-iron compounds possess

a poor solubility in physiological environments,^[13]resulting in

serious drawbacks in the view of clinical applications.^[14]In this

overall scenario, the medicinal potential of diiron bis-cyclo-

pentadienyl compounds was unexplored until 2019. The

commercial dimer [Fe₂Cp₂(CO)₄] (Cp=η⁵-C₅H₅) represents an

inexpensive and convenient entry into the chemistry of diiron

Department of Pharmaceutical and Pharmacological Sciences

University of Padova

Via F. Marzolo 5, I-35131 Padova (Italy)

E-mail: valentina.gandin@unipd.it

Homepage: https://www.dsfarm.unipd.it/valentina-gandin

[c] Prof. S. Zacchini

Department of Industrial Chemistry “Toso Montanari”

University of Bologna

Viale Risorgimento 4, I-40136 Bologna (Italy)

Supporting information for this article is available on the WWW under

https://doi.org/10.1002/chem.202101048

VCH GmbH. This is an open access article under the terms of the Creative

Commons Attribution License, which permits use, distribution and re-

production in any medium, provided the original work is properly cited.

Chem. Eur. J. 2021, 27, 1–18

1

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

slight molar excess of [Fe₂Cp₂(CO)₄] in acetonitrile.^[20]The final

reaction mixtures were dried, and the crude residues were

dissolved in dichloromethane or acetonitrile and treated with a

range of alkylating agents under optimized conditions, to afford

the corresponding products [2–5]⁺(Table 1). The products were

purified by alumina chromatography (see the Experimental

Sections for details), and finally isolated in good to quantitative

yields in a 300 mg to 10 g scale. Deliquescent [2B]CF₃SO₃was

converted into the less moisture-sensitive tetrafluoroborate salt

by repeated treatment with NaBF₄in water followed by CH₂Cl₂

extraction. A minor amount of [2C]Cl was isolated from the

chromatography of [2C]CF₃SO₃, as a result of triflate/chloride

exchange during the chromatographic purification, due to

chloride impurities in commercial alumina; then, [2C]Cl was

reliably obtained from [2C]CF₃SO₃using a cation exchange

resin. In order to ensure the solid-state stability of the

compounds, the light-sensitive anions of [3F]I and [5H]Br were

exchanged by metathesis with AgCF₃SO₃in MeCN. In summary,

Table 1 shows a general approach to obtain easily accessible

ionic diiron complexes with variable substituents on the

bridging hydrocarbyl ligand and counter anions, based on the

choice of the isocyanide and the alkylating agent, thus offering

wide opportunity for tuning the physicochemical properties

(Scheme S1 in the Supporting Information). The cations [2A–

C]⁺, [4G]⁺and [5H]⁺are unprecedented in the literature, while

the others were previously isolated as [2D–F]CF₃SO₃,^[21,22][3F]I^[23]

and [3G]CF₃SO₃;^[20]however, the salts [3G]I and [3F]CF₃SO₃are

novel and an optimized synthesis for [3F]I is supplied here. The

incorporation of the indolyl moiety in [2A]CF₃SO₃is of some

relevance, because many indole derivatives have been reported

as potent anticancer agents, and some of them have been

approved by FDA for clinical treatment.^[24]On the other hand,

the introduction of the highly polar diethylphosphonato group

in [2B]BF₄is functional to enhance the water solubility.^[25]

Figure 1. Structures of cyclopentadienyl iron complexes that show anti-

cancer activity (plus year of publication). I) L=PPh₂(alkyl), Y=C(O)Me

(2020);^[10a]L=CO, Y=Cl, Br, I, NCS, SCN (2016);^[10b]L=PPh₃or related ring-

substituted molecule, Y=I (2017).^[10c]II) R=Ph, 4-C₆H₄CO₂H or 4-C₆H₄F,

R’=4-C₆H₄NH₂, X=PF₆(2017);^[10c]R=Ph, R’=aryl or vinyl, X=CF₃SO₃

(2020).^[11c]III) L=imidazole or N-substituted imidazole, X=CF₃SO₃(2013);^[11d]

L=carbohydrate-substituted nitrile, X=PF₆(2015);^[11a]L=N-heteroaromatic

nitrile, X=PF₆or CF₃SO₃(2014).^[11b]IV) R=Me, CH₂Ph or 2,6-C₆H₃Me₂(Xyl),

R’=alkyl, aryl, pyridyl, or thiophenyl, Y=H, Ph or S/Se group (2019,

2020).^[17a,b]

compounds, and a plethora of organometallic motifs has been

constructed on a bridging coordination site, exploiting the

cooperativity of the two metal centres.^[15]Some of us reported a

preliminary evaluation of the anticancer activity of complexes

based on the {Fe₂Cp₂(CO)_x} core (x=2 or 3),^[16]and especially

those ones with a bridging vinyliminium ligand (Figure 1,

structure IV) displayed interesting preliminary cytotoxicity

profiles.^[17]

Here, we show that a general and straightforward two-step

synthetic procedure (reproducible in multigram scales) allows

to access

a family of compounds showing a promising

antiproliferative activity against 2D and 3D human cancer cell

systems. Remarkably, 3D cell culture studies constitute an

excellent base for more advanced in vivo cancer research,

surpassing the traditional assays on monolayer cultures;^[18]

indeed, 2D models provide limited predictivity for drug testing,

while 3D models promise to bridge the gap between traditional

2D cell cultures and in vivo animal models.^[19]

To the best of our knowledge, the present 3D investigation

is one of the first ones ever reported on iron compounds. The

proposed diiron complexes combine within a cationic structure

two Cp rings, three carbonyls and a variable bridging amino-

carbyne ligand, the latter tuning the water solubility, the

lipophilicity and the activity. Several experiments aimed to

clarify the mechanism of action of the complexes will be

discussed.

New compounds were fully characterized by elemental

analysis, IR and multinuclear NMR spectroscopy. In addition, the

Table 1. Two-step preparation of cationic diiron bis-cyclopentadienyl

complexes with bridging aminocarbyne ligands. Experimental conditions

for the second step: [2A–F]⁺, CH₂Cl₂, room temperature; [3F]⁺, MeCN, RT;

+

[3G]⁺and [5H] , MeCN, reflux; [4G] , MeCN, 0 C. *Not isolated.

°

R

R’X

1H-indol-6-yl

CH₂P(O)(OEt)₂

Cy=C₆H₁₁

4-C₆H₄OMe

Xyl=2,6-C₆H₃Me₂

Me

1

CH₃SO₃CF₃

CH₂=CHCH₂I

CH₃I

[2A]⁺

[2B]⁺

[2C]⁺

[2D]⁺

[2E]⁺

[2F]⁺

[3F]⁺

[3G]⁺

[4G]⁺

[5H]⁺

1B

1C

1D

1E

1F

1G

1H

Results and Discussion

Synthesis and characterization of diiron complexes and

structural studies

2-naphthyl

CH₂Ph

The known complexes 1A–H were prepared by the thermal

substitution reaction of the corresponding isocyanides with a

[Et₃O]BF₄

PhCH₂Br

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

2

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

structures of [2A]CF₃SO₃, [2C]CF₃SO₃, [2C]Cl and [2D]CF₃SO₃

were confirmed by single crystal X-ray diffraction. Relevant

bonding parameters are comparatively compiled in Table 2, and

a view of the cations are shown in Figure 2. The structure of the

cations consists of a typical cis-{Fe₂Cp₂(CO)₂} core, connected to

a further CO and an aminocarbyne group occupying the two

bridging sites. Along this series of complexes, the C(4)-N(1)

distance does not significantly vary, indicating a substantial

double-bond character which accounts for the complementary

description of the bridging ligand as iminium (see below).

The bonding description of {μ-CNR(R’)} ligands in dimetal

complexes is not clearly defined in the literature, therefore we

performed a DFT study to shed light on this point. The

representative structure of [2C]⁺was calculated and compared

to that of the parent compound cis-1C with a bridging

cyclohexyl-isocyanide ligand (CNCy). DFT-optimized structures

and relevant bonding parameters are reported in Table S1 and

Figure S1. In 1C, the (μ-C)À N Mayer bond order (b.o.)^[26]is 2.0

and the bond distance is 1.233 Å, in agreement with the X-ray

data available for related systems [(μ-C)À N=1.26(3) Å for CNiPr

instead of CNCy].^[27]The b.o. of (μ-C)À N lowers to 1.58 upon

methylation of the isocyanide moiety leading to [2C]⁺. The

nature of the nitrogen substituents has a negligible influence,

and for instance the corresponding b.o. is 1.54 in the case of

[2A]⁺, the related bond length being consistent with the

crystallographic characterization of [2A]CF₃SO₃[1.300 vs.

1.289(7) Å; Table 2]. These theoretical results strongly support

the description of the bridging N-containing ligand in the

cationic complexes [2–5]⁺as an almost perfect aminocarbyne-

iminium hybrid structure (Figure 3).

Then, in order to clarify the possible effect of the net ionic

charge of the complex, we also synthesized and characterized

the novel neutral derivative [Fe₂Cp₂Cl(CO)(μ-CO){μ-CNMe(Cy)}],

1

6 (see Experimental Section). Based on a comparison of the H

+

°

Table 2. Selected bond lengths [Å] and angles [ ] for [2A] , [2C] , and [2D] .

[2A]⁺

[2C]^{+ (a)}

[2C]^{+ (b)}

[2D]⁺

Fe(1)-Fe(2)

Fe(1)-C(1)

Fe(2)-C(2)

Fe(1)-C(3)

Fe(2)-C(3)

Fe(1)-C(4)

Fe(2)-C(4)

C(1)-O(1)

C(2)-O(2)

2.5136(11)

1.760(6)

1.782(6)

1.954(6)

1.944(6)

1.874(6)

1.870(6)

1.137(7)

1.131(7)

1.155(7)

1.289(7)

80.3(2)

2.509(3)

2.5152(15)

1.776(7)

1.754(8)

1.915(8)

1.962(7)

1.883(8)

1.888(8)

1.139(9)

1.147(10)

1.172(9)

1.274(10)

80.9(3)

2.5161(10)

1.774(6)

1.762(6)

1.935(5)

1.934(5)

1.870(5)

1.866(5)

1.139(6)

1.148(7)

1.161(6)

1.287(6)

81.1(2)

1.765(16)

1.760(16)

1.924(16)

1.965(16)

1.881(14)

1.873(15)

1.119(17)

1.150(18)

1.291(18)

80.3(6)

C(3)-O(3)

C(4)-N(1)

Fe(1)-C(3)-Fe(2)

Fe(1)-C(4)-Fe(2)

Fe(1)-C(1)-O(1)

Fe(1)-C(2)-O(2)

C(4)-N(1)-C(5)

C(4)-N(1)-C(6)

C(5)-N(1)-C(6)

84.3(2)

83.9(6)

83.7(3)

84.7(2)

177.9(6)

177.8(6)

122.2(5)

124.1(5)

113.7(4)

178.9(14)

179.6(15)

122.6(12)

121.0(12)

116.3(11)

178.5(7)

179.5(8)

122.5(7)

121.4(6)

116.0(6)

175.6(5)

176.3(5)

123.4(4)

120.9(4)

115.5(4)

(a) As [2C]CF₃SO₃. (b) As [2C]Cl.

Figure 2. Molecular structures of [2A]⁺, [2C]⁺and [2D]⁺(triflate salts). Displacement ellipsoids are at the 30% probability level. H-atoms have been omitted

for clarity.

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

3

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

resonance at 17 ppm for the phosphonato group, and the ¹³C

signal for the carbyne appears as a doublet (³J_CPca. 6 Hz).

Small differences between the spectroscopic data of [2C]

CF₃SO₃and [2C]Cl in chlorinated solvents are ascribable to the

presence of ion pairs (or higher aggregates);^[31]otherwise, the

two compounds display superimposable ¹H NMR spectra in

protic polar solvents (D₂O and CD₃OD), where ion pairs are

practically absent.^[32]

Figure 3. Resonance structures for the bridging hydrocarbyl ligand in the

diiron complexes [2–5]⁺and 6. Structures I and II: aminocarbyne; structure

III: iminium.

Solubility, partition coefficient, stability in aqueous media

and interaction with a model protein

NMR and IR spectra with data available in the literature for

similar compounds,^[28]6 exists in CDCl₃solution as a nearly

equimolar mixture of two isomers, maintaining the cis-geometry

of the Cp ligands (as ascertained by ¹H NOE experiments,

Figure S32) and differing in the orientation of the N-substituents

with respect to the chloride (E/Z isomers). This feature suggests

that the (μ-C)À N bond in the neutral 6 still holds a substantial

double bond character. The X-ray structure of the E isomer was

determined and is shown in Figure S2, although the low quality

of the crystals prevents a detailed discussion of bonding

parameters.

In agreement with the NMR evidence, DFT outcomes

indicate that 6^Z(Z isomer) and 6^E(E isomer) are practically

isoenergetic, the latter being only 0.2 kcalmol^{À 1}more stable

than the former (Figure S1). In both E/Z isomers, the bridging

{CN(Me)(Cy)} group resembles that in [2C]⁺, with only a slight

decrease of the (μ-C)À N bond order (from 1.58 to 1.52). It is

interesting to note that the bond length of the bridging

carbonyl in 6 is very close to that in 1C (1.187 vs. 1.188 Å),

highlighting that the isocyanide/CO and aminocarbyne/Cl

combinations supply very similar electron densities for back-

donation to the diiron framework.^[29]

A detailed study on the behaviour of a selection of the diiron

complexes in aqueous media was undertaken. Experimental

procedures are provided in the Experimental Section and in the

Supporting Information, with data collected in Tables 3, S3 and

S4. The D₂O solubility of the compounds was assessed by H

NMR, ranging from sub-millimolar to 0.15 M in the case of [2C]

1

Cl. The octanol/water partition coefficients (logP_ow

) were

measured by a UV/Vis method and fall in between 0 and À 1,

reflecting an amphiphilic or slightly hydrophilic character. An

exception is represented by [2B]BF₄, wherein the marked

hydrophilicity imparted by the diethylphosphonato does not

appear sufficiently balanced (logP_ow<À 1.5). Conversely, those

complexes with aromatic groups (i.e., 1H-indol-6-yl, 2-naphthyl

and bis-benzyl in [2A]⁺, [3G]⁺, [4G]⁺and [5H]⁺) are less soluble

in water and display increased affinity for n-octanol (logP_ow�0).

The diiron compounds are inert in aqueous solution at room

°

temperature, and undergo a slow degradation at 37 C. Under

such conditions, a variable amount (50 to 86%; Table S3) of the

starting material was detected by ¹H NMR in D₂O or D₂O/CD₃OD

mixtures after 72 h.^[33]During this time, the precipitation of a

brown solid was observed, which in one case (from a [2C]

CF₃SO₃solution) was separated and identified as γ-Fe₂O₃by

Raman analysis. In addition, headspace GC analyses performed

on 5% MeOH aqueous solutions of the compounds maintained

Salient spectroscopic features of all complexes are summar-

ized in Table S2, and NMR and IR spectra are supplied in

Figures S3–S32. In the IR spectra (CH₂Cl₂solution), the vibrations

due to the terminal carbonyl ligands occur in the ranges 2018–

2028 (anti-symmetric stretching) and 1985–1996 cm^{À 1}(symmet-

ric stretching), while the band related to the bridging carbonyl

is found at 1833–1853 cm^{À 1}. Moreover, the absorption attrib-

uted to the carbyne-N bond falls in between 1530 to 1602 cm^{À 1},

the higher values being associated to the N-substituents with

higher electron-donor ability. The ¹H and ¹³C NMR spectra

(recorded in CDCl₃, CD₃CN or [D₆]DMSO) display one set of

resonances, ascribable to a single isomeric form featuring the

Cp ligands in mutual cis position, as found in the X-ray

°

at 37 C allowed the identification and quantitation of the

carbon monoxide released over a 48 h period (Table S4). These

facts indicate that the thermal degradation process in aqueous

°

Table 3. Solubility in D₂O (21 C) and octanol/water partition coefficient

(logP_ow) for diiron compounds.

Compound

Solubility [molL^{À 1}

]

LogP_ow

[2A]CF₃SO₃

[2B]BF₄

[2C]CF₃SO₃

[2C]Cl

[2D]CF₃SO₃

[2E]CF₃SO₃

[2F]CF₃SO₃

[3F]CF₃SO₃

[3G]CF₃SO₃

[4G]BF₄

ca. 3·10^{À 4}

M

0.0�0.1

1

5.2·10^{À 2}

6.2·10^{À 3}

1.5·10^{À 1}

4.1·10^{À 3}

1.4·10^{À 3}

3.2·10^{À 3}

3.4·10^{À 3}

M

� À 1.5

structures and confirmed by H NOE experiments in the case of

[4G]⁺(Figure S27). Two resonances for the Cp and terminal CO

ligands are distinguishable in the unsymmetrical complexes,

due to hampered rotation around the partially double CÀ N

bond (see above); accordingly, such resonances coincide in the

complexes [2F]⁺and [5H]⁺, displaying two identical N groups.

In the ¹³C NMR spectra, the bridging {CN} unit is detected at low

fields (i.e., around 320 ppm), as it is typical for bridging

aminocarbynes.^[20,30]Compound [2B]⁺features a ³¹P NMR

À 0.46�0.02

À 1.1�0.08

À 0.25�0.04

À 0.27�0.04

À 0.9�0.1

À 1.01�0.02

0.29�0.03

0.1�0.1

[b]

not soluble ^[a]

[b]

not soluble ^[a]

[5H]CF₃SO₃

6·10^{À 4}

M

0.2�0.02

[a] Solubility below the lowest value of quantitation (ca. 3×10^{À 4}M). [b]

Data taken from ref. 16a.

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

4

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

solution leads to the progressive disassembly of the organo-

metallic scaffold.^[16]The presumed, controlled release of the iron

centres (converted to iron oxide, see above) and carbon

monoxide are two aspects that deserve attention with reference

to the biological activity: note that the decay of ferrocene drug

candidates into Fe^IIions was previously associated to their

cytotoxicity,^[34]while the activity of various iron carbonyl

complexes was correlated to CO dissociation in the physiolog-

ical media.^[35]

chromatograms after protein peak reconstruction, are reported

in Table 4. Compounds lacking N-methyl groups (i.e., [4G]⁺and

[5H]⁺) did not react with lysozyme. Methylation is a common

post-translational modification of protein/enzymes (meth-

yltransferase enzymes) but it has also been observed as a result

of the interaction of proteins with various electrophiles.^[37]

Cytotoxic activity in 2D cell cultures

Next, the stability of the diiron compounds in a cell culture

medium (DMEM) solution at 37 C, over a shorter time range

The diiron complexes [2A]⁺, [2C–E]⁺, [3F–G]⁺, [5H]⁺(as CF₃SO₃

salts) and [2B]⁺, [4G]⁺(as BF₄salts) were tested for their

cytotoxic potential by means of the MTT assay, as reported in

the Experimental Section. The most convenient counter anion

in terms of preparation and handling/storage was chosen in

each case; in particular, [2C]CF₃SO₃was included in the study,

while [2C]Cl was excluded, being highly moisture sensitive and

thus affecting the accuracy of the weighting operation. The in-

house human cancer cell line panel contains examples of

ovarian (2008), colon (HCT-15), pancreatic (PSN-1), and breast

(MCF-7) cancers as well as of melanoma (A375). Cisplatin was

used as a reference and tested under the same experimental

conditions. The cytotoxicity parameters, expressed in terms of

IC₅₀obtained after 72 h of exposure, are listed in Table 5. In

addition, the cytotoxicity was also evaluated for shorter time of

exposure (24 h) towards PSN-1 cancer cells (Table S5). Data

analysis reveals that the cytotoxic activity showed by tested

complexes is time- and dose-dependent. In particular, the

highly hydrophilic complexes [2B]BF₄and [3F]CF₃SO₃did not

impact cell viability (average IC₅₀values were over 200 or

100 μM for all tested cell lines), whereas the other compounds

elicited IC₅₀values in the micromolar range. Despite all

cytotoxic complexes showed an average in vitro antitumor

potential lower than that of cisplatin, [2C]CF₃SO₃and [2D]

CF₃SO₃proved to be the most effective derivatives, with

average IC₅₀values of 17.4 and 16.6 μM (72 h time trials), and

showed a similar pattern of response over the five tested cell

lines. In particular, [2D]CF₃SO₃exhibits a comparable activity

with respect to cisplatin against HCT-15, PSN-1 and MCF-7 cells.

As one of the main drawbacks of chemotherapeutics is the

toxic effect toward noncancerous cells, we measured the

°

(24 h), was assessed by ¹H NMR. Notably, a considerable fraction

(70–88%; Table S3) of the starting material was detected at the

end of the experiment, with a limited formation of an orange-

brown precipitate analogous to that described above.

To gain insights into the reactivity of the diiron compounds

with possible biomolecular targets, we selected lysozyme, a

small model protein often used to test interactions with metal

compounds.^[36]Following 24 h incubation at 37 C, diiron

°

compounds/lysozyme solutions (2:1 molar ratio) in ammonium

acetate were filtered and analysed by ESI-MS. In all cases, the

original organometallic cation was identified as the major Fe-

containing compound in solution (Figures S33-S40). Concerning

the protein scaffold, the only identified adduct from the

interaction with [2–3]⁺was the methylated derivative, which

was also checked by HPLC separation (Figure S41). The

expected diiron counterparts, that is, 1A–H, could not be

detected possibly due to their water insolubility. The [lysozyme

+Me⁺] to [lysozyme] ratios, calculated on extracted ion

Table 4. Methylation ratio (%) upon lysozyme interaction with the diiron

°

compounds (1:2 molar ratio, 24 h, 37 C), as determined by ESI-MS

measurements.

Compound % Lysozyme meth-

ylation (alkylation) ^[a]

Compound % Lysozyme meth-

ylation (alkylation) ^[a]

[2A]CF₃SO₃45%

[2F]CF₃SO₃35%

[3F]CF₃SO₃47%

[4G]BF₄

[5H]CF₃SO₃0%^[b]

Blank exp

0%^[b]

[2B]BF₄

49%

[2C]CF₃SO₃50%

[2D]CF₃SO₃31%

[2E]CF₃SO₃59%

0%^[b]

[a] Calculated by the relative peak integrals for flow injection MS analysis.

[b] No other protein MS peak beside that of lysozyme was observed.

Table 5. In vitro antitumor activity of diiron complexes in 2D cell cultures. Cells (3–8×10³mL^{À 1}) were treated for 72 h with increasing concentrations of the

tested compounds. The cytotoxicity was assessed by the MTT test. IC₅₀values were calculated by a four-parameter logistic model 4-PL (P<0.05). In brackets

RF=IC₅₀(resistant cells)/IC₅₀(wild-type cells) for MCF-7 cells.SI=selectivity index (see main text).

Compound

IC₅₀(μM)�SD

SI

2008

HCT-15

A375

PSN-1

MCF-7

MCF-7 ADR

HEK293

[2A]CF₃SO₃

[2B]BF₄

55.5�4.8

>100

28.4�5.4

21.3�3.1

38.8�4.5

>100

36.6�3.4

38.8�6.9

37.4�5.2

2.2�1.4

À

69.5�6.7

>100

9.2�1.8

15.6�3.5

37.3�3.5

>100

24.6�2.9

22.9�5.8

32.5�4.8

16.5�2.2

À

50.8�6.1

>100

16.8�2.1

15.5�2.6

34.6�4.0

>100

25.2�3.3

26.2�5.6

30.8�5.1

2.1�0.3

À

54.3�6.2

>100

15.3�2.7

16.2�2.9

41.7�6.3

>100

26.3�2.8

25.6�4.5

33.7�4.2

12.1�2.9

À

63.2�5.5

>100

17.3�3.3

14.3�2.6

44.7�5.5

>100

20.7�3.0

21.4�3.6

36.2�6.1

8.8�0.2

0.2�0.03

84.5�8.5 (1.3)

>100

23.3�4.1 (1.3)

22.6�4.2 (1.6)

66.9�5.2 (1.5)

>100

135.5�6.4

>100

62.5�4.9

94.3�6.1

96.6�2.9

>100

90.3�5.3

89.5�4.5

98.3�4.2

18.1�3.6

À

2.3

À

[2C]CF₃SO₃

[2D]CF₃SO₃

[2E]CF₃SO₃

[3F]CF₃SO₃

[3G]CF₃SO₃

[4G]BF₄

[5H]CF₃SO₃

cisplatin

doxorubicin

3.6

5.6

2.4

À

30.5�6.7 (1.5)

26.9�5.1 (1.3)

44.7�6.1 (1.2)

À

3.3

2.8

2.2

À

7.8�2.2 (39)

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

5

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

cytotoxicity of tested complexes against a noncancerous cell

line (HEK293) and calculated the selectivity index (SI, defined as

the ratio between the IC₅₀value in noncancerous cells and the

corresponding average IC₅₀value related to cancer cells;

Table 5). The diiron complexes are generally less cytotoxic

against HEK293 noncancerous cells, and again [2D]CF₃SO₃

exhibited the best profile (SI value of 5.6 to be compared with

2.2 for cisplatin). Overall, it appears that slight structural

variations on the iminium group may have a notable impact on

the activity of the cations, although not obvious: for instance,

[2D]CF₃SO₃and [2E]CF₃SO₃differ from each other in the nature

and the position of peripheral arene substituents and display

almost identical logP_owvalues, notwithstanding the former is

significantly more cytotoxic and selective than the latter.

The antiproliferative activity was also investigated in a

multidrug-resistant cancer cell model overexpressing P-gp

(MCF-7 ADR cells). All complexes exhibited activity levels very

similar on both sensitive (MCF-7) and multi-drug resistant (MCF-

7ADR) cell lines; interestingly, the resistance factor (RF) values

(Table 5) range from 24 to 35 times lower than that of

doxorubicin, a drug belonging to the MDR spectrum, thus

indicating the ability of the complexes to overcome MDR

phenomena and not to act as P-gp substrates.

Cellular uptake in cancerous and noncancerous cells

In order to correlate the cytotoxic potential to the ability of the

newly developed complexes to enter cells, we performed

uptake experiments. To this purpose, we selected the most

promising compound [2D]CF₃SO₃(logP_ow=À 0.25) together

with the less active/selective [2E]CF₃SO₃(logP_ow=À 0.27) and

[3G]CF₃SO₃(logP_ow= +0.29), to be assessed for their accumu-

lation in human pancreatic PSN-1 cancer cells and in untrans-

formed HEK293 cells. Thus, cells were treated for 24 h with

equimolar concentrations (50 μM) of the tested compounds.

The cellular iron levels were quantified by means of GF-AAS

analysis, and the results, expressed as μg Fe per 10⁶cells, are

shown in Figure 4A. Complex [3G]CF₃SO₃was ineffective in

entering human cells, while both [2D]CF₃SO₃and [2E]CF₃SO₃

were able to induce cellular iron accumulation, and no

significant differences in the ability to enter cells were

evidenced for the two complexes. Considering the logP_ow

values, the obtained outcomes highlight that the hydrophilic

complexes [2D]CF₃SO₃and [2E]CF₃SO₃are significantly more

effective in entering cells compared to the more lipophilic one

[3G]CF₃SO₃. In addition, both [2D]CF₃SO₃and [2E]CF₃SO₃led to

major iron accumulation in untransformed cells with respect to

pancreatic cancer cells. Intriguingly, these findings attest that

their selectivity towards cancer cells with respect to untrans-

formed ones cannot be attributable to their different ability to

cross cellular plasmalemma and to enter human cells, and the

reason of their preferential activity against cancer cells might

be dependent on their ability to interfere with a cancer specific

target.

Figure 4. Intracellular detection of iron and carbon monoxide. A) Cellular

uptake in PSN-1 and HEK293 cells; cells were incubated with 50 μM of

complexes for 24 h, and the cellular iron content was detected by GFAAS

analysis. B) Intracellular CO levels were determined by using a fluorescent

probe (BioTracker Carbon Monoxide Probe 1 Live Cell Dye®, Merck, Germany)

after 30 min of treatment of PSN-1 or HEK293 cells with 20 μM [2C]CF₃SO₃

and [2D]CF₃SO₃, or 20 μM CORM-401 as positive control. **P<0.01,

*P<0.05.

Cytotoxic activity in 3D cell cultures

The remarkable cell-killing effect observed in 2D cultured cells

prompted us to evaluate the in vitro antitumor activity of the

complexes on 3D cell cultures. Differently from 2D monolayer

culture, 3D spheroid cell culture systems comprise cancer cells

in various cell growth stages. Consequently, the multicellular

cancer spheroid model is recognized to better reflect the

tumour mass in vivo regarding drug permeation, cell interac-

tions, gene expression, hypoxia and nutrient gradients with

respect to monolayer cell cultures, making 3D models more

predictive than conventional 2D monolayer cultures in screen-

ing antitumor drugs.^[38]Table 6 summarizes the IC₅₀values

obtained after treatment of 3D cell spheroids of human ovarian

(2008) and pancreatic (PSN-1) cancer cells with the diiron

complexes and cisplatin. In accordance with 2D chemosensitiv-

ity assays, compounds [2B]CF₃SO₃and [3F]CF₃SO₃were ineffec-

tive against cancer cells spheroids, whereas [2D]CF₃SO₃was the

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

6

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

environments was previously demonstrated.^[16a,39]The relatively

fast release of carbon monoxide within the cell might be

associated to the presence of a consistent amount and variety

of biological substrates, which can quicken the formation of CO

substitution diiron products. As a matter of fact, it is well

documented that [2E–F]CF₃SO₃are susceptible to replacement

of one terminal CO ligand by suitable C-, P-, S- and N-donors to

give adducts of general formula [Fe₂Cp₂(CO)(L)(μ-CO){μ-CNR

Table 6. Activity of diiron complexes in 3D cell cultures.

IC₅₀�SD (μM)

Compound

2008

PSN-1

[2A]CF₃SO₃

[2B]BF₄

>100

78.2�9.1

53.7�6.1

82.3�4.8

>100

[2C]CF₃SO₃

[2D]CF₃SO₃

[2E]CF₃SO₃

[3F]CF₃SO₃

[3G]CF₃SO₃

[4G]BF₄

66.4�8.6

42.8�4.2

65.1�6.8

>100

88.5�3.2

58.9�4.8

73.2�6.2

57.6�3.6

(Me)}]^{0/+ [15b,16a,20]}

Although the nature of L in the cell context is

.

>100

77.4�8.4

86.2�6.4

61.7�12.1

far from being clarified, it is reasonable to speculate that CO

release is readily operative via such substitution pathway.

However, an acceleration of the route leading to the extensive

decomposition of the diiron scaffold (elimination of three COs

per complex), as observed in aqueous media, might also

contribute.

[5H]CF₃SO₃

cisplatin

Spheroids (2.5×10³cells/well) were treated for 72 h with increasing

concentrations of tested compounds. The growth-inhibitory effect was

evaluated by means of the acid phosphatase (APH) test. IC₅₀values were

calculated from the dose-survival curves using a four-parameter logistic

model (p<0.05). SD=standard deviation.

Iron is a redox active metal that can generate ROS in cells

via the Fenton reaction, thus inducing cellular damage and

ultimately leading to cell death.^[40]Coherently, several iron

complexes have been shown to exert antitumor activity

through redox stress induction (see the Introduction).^[8]Based

on this premises, we evaluated the effects induced by the diiron

complexes on the mitochondria activity and the cellular redox

environment. More precisely, we investigated the ROS produc-

tion and the alteration of the mitochondrial membrane

potential. Treatment of PSN-1 cells with [2C]CF₃SO₃and [2D]

CF₃SO₃determined a substantial increase in cellular ROS

production, in a time-dependent manner (Figure 5A). Remark-

ably, after 2 h treatment with [2D]CF₃SO₃, the hydrogen

peroxide content was similar to that induced by antimycin, a

classical inhibitor of the mitochondrial respiratory chain at the

level of complex III. Consistently, a significant dose-dependent

increase of cells with depolarized mitochondria was observed

after 24 h-treatment of PSN-1 cells with [2C]CF₃SO₃and [2D]

CF₃SO₃(Figure 5B). The time-dependent ROS basal production

is probably associated to more than one phenomenon. First of

all, the progressive disruption of the diiron structure leads to

the extrusion of the Fe^Icentres, which are expected to

subsequently convert into Fe^III(in fact, slow precipitation of iron

(III) oxides has been recognized from water solution, see above).

Otherwise, based on previous voltammetric measurements on

[2E–F]CF₃SO₃,^[41]oxidation of iron enclosed in the complex is

out of the range of biologically relevant potentials. In the cases

of [2–3]⁺, some contribution to ROS might be provided by the

most effective complex among the series, thus indicating its

ability to better penetrate the spheroid and reach the hypoxic

core. It is noteworthy that [2D]CF₃SO₃exhibited a level of

activity either comparable (PSN-1 cells) or even superior (2008

cells) to that of cisplatin.

An important aspect of the collected data is that only a

limited decrease of activity is observed with the cytotoxic

compounds, the IC₅₀ratio related to 3D and 2D experiments

ranging from 2 to 5 for both cell lines; in particular, this ratio is

approximately 2 for the cytotoxic diiron complexes against the

2008 cell line, while the corresponding value for cisplatin is 26.

Note that a substantial drop of antiproliferative activity has

been often recognized for potential anticancer metal drugs in

3D cell cultures compared to the corresponding 2D ones,

highlighting the need for a combination of both models for

screening of compounds.^[1c]

Intracellular CO release, ROS production and inhibition of

thioredoxin reductase

Additional studies were conducted in order to elucidate the

mode of action underlying the cell killing effect of the newly

synthesized diiron complexes in cancer cells. In particular, based

on the evidence of their tendency to slowly release CO in

aqueous media (see section on solubility, partition coefficient,

stability in aqueous media), we evaluated the ability of the

most performant complexes [2C]CF₃SO₃and [2D]CF₃SO₃to

release carbon monoxide inside PSN-1 pancreatic cancer cells

and HEK293 noncancer cells. A fluorometric test was conducted

over a short time interval (30 min) by means of a CO sensing

dye (Figure 4B). Interestingly, both complexes were effective in

significantly release CO in PSN-1 cells whereas lower levels of

CO were detected in HEK293 cells. These results attest the

power of the novel diiron complexes to release CO in cellular

milieu and, more importantly, could at least in part explain the

tumour-selective cytotoxic effect exerted by [2C]CF₃SO₃and

[2D]CF₃SO₃. Note that the ability of a range of related diiron

cyclopentadienyl complexes to release CO ligands in aqueous

+

partial generation of the neutral products 1A–H upon CH₃

elimination (see above); in principle, the absence of a net

positive charge in 1A–H might favour the oxidation of these

complexes or their possible derivatives in the cells.^[40]

It has recently emerged that the redox stress induced by

some iron complexes might also be attributed to their ability to

inhibit the selenoenzyme thioredoxin reductase (TrxR). In

particular, Rigobello and co-workers reported the capacity of a

ferrocenyl diphenol and of a Tamoxifen-like ferrocifen to target

and hamper TrxR activity.^[42]On this basis, we thought of

interest to test the ability of the most promising diiron

complexes [2C]CF₃SO₃and [2D]CF₃SO₃to target TrxR both in

cell-free experiments and in cells. The two complexes showed a

similar pattern of response, being completely ineffective in

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

7

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

Figure 5. Effects induced by compounds [2C]CF₃SO₃and [2D]CF₃SO₃at the mitochondrial level. A) ROS production in PSN-1 cells. Cells were pre-incubated in

°

phosphate-buffered saline (PBS)/10 mM glucose medium for 20 min at 37 C in the presence of 10 mM CM-H₂DCFDA and then treated with increasing

concentrations of [2C]CF₃SO₃and [2D]CF₃SO₃or antimycin (3 μM). The fluorescence of DCF was measured at λ_ex=485 nm and λ_em=527 nm. B) Effects of [2C]

CF₃SO₃and [2D]CF₃SO₃on mitochondrial membrane potential. PSN-1 cells were treated for 24 h with increasing concentrations of [2C]CF₃SO₃and [2D]CF₃SO₃

or antimycin (3 μM) and stained with TMRM (10 nM). Fluorescence was estimated at λ_ex=490 nm and λ_em=590 nm. Data are the means of five independent

experiments. Error bars indicate S.D. *P<0.05.

hampering TrxR in cell-free experiment (data not shown) but

elicited a substantial inhibition of the selenoenzyme in human

pancreatic PSN-1 cancer cells (Figure 6). In particular, PSN-1 cells

treated with 25 μM of [2C]CF₃SO₃and [2D]CF₃SO₃for 24 h

showed a residual TrxR cellular activity of 57 and 39%,

respectively. These data well correlate with the observed

methylate specific substrates, as we have assessed for lysozyme.

Actually, it was previously described that several alkylating

agents are able to selectively modify the redox active Cys or Sec

of TrxR, thus causing irreversible inhibition.^[43]

formation of carbon monoxide in cells, and suggest that the Conclusions

inhibition of TrxR is ascribable to the interaction with diiron

complexes and organic fragments generated from [2–5]⁺by

dissociation of CO ligand(s). On the other hand, TrxR inhibition

might also be consistent with the capacity of the complexes to

We have described a synthetic strategy to obtain a family of

diiron carbonyl complexes containing a bridging aminocar-

byne/iminium hybrid ligand from the readily available

[Fe₂Cp₂(CO)₄]. They have ideal properties for

a potential

anticancer drug, such as the presence of a nontoxic metal

element, appreciable water solubility and amphiphilicity, and

stability in aqueous media.^[44]Moreover, the synthesis works up

to multigram scales and features a considerable structural

variability. In general, the complexes showed moderate cyto-

toxic activity against human cancer cells and the ability to

overcome resistance. Compound [2D]CF₃SO₃, with a 4-methoxy

phenyl substituent on the bridging hydrocarbyl ligand, was

revealed to be the most promising and exhibited a noteworthy

selectivity toward cancer cell lines compared to noncancerous

cells. According to cellular-uptake experiments, such selectivity

is not correlated to a different ability to enter cancerous or

noncancerous cells. Remarkably, the cytotoxicity profile of the

complexes does not substantially decrease in 3D cell cultures,

in contrast to what has often been observed in the literature

upon comparison of 2D/3D studies on various transition metal

compounds. Interestingly, the activity exhibited by [2D]CF₃SO₃

in the ovarian cancer cell 3D model was even superior to that

of cisplatin. Experiments on selected complexes outlined their

power to behave as carbon monoxide-releasing molecules

Figure 6. Effects of compounds [2C]CF₃SO₃, [2D]CF₃SO₃and Auranofin as

positive control on TrxR in human pancreatic cells. PSN-1 cells were

incubated for 24 h with the IC₅₀of tested compounds. Afterwards, cells were

washed twice with PBS and lysed. TrxR activity was tested by measuring the

NADPH-dependent reduction of DTNB. *P<0.05.

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

8

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

°

were analysed by isothermal runs (110 C, 4 min) using He as carrier

gas. Raman analysis was conducted with a Renishaw Invia micro-

Raman instrument equipped with a Nd:YAG laser working at

532 nm and 0.1 mW, integration time 10 s. HPLC-MS analyses were

performed with a API3000 instrument (SCIEX) equipped with ESI(+)

source and a quadrupole detector. HPLC separation was performed

using an Agilent 1110 series instrument (Agilent Technologies

Deutschland GmbH, Germany) equipped with a G1312A binary

pump, a G1329A autosampler, a peltier column oven Series 200

(PerkinElmer) and a BioZen C4 intact column (3.6 μm, 150×2.1 mm,

pore size 200 Å; Phenomenex, USA).

(CORMs) inside cells, to unbalance cellular redox homeostasis

and to alkylate biological targets. The preferential CO release in

cancer cells with respect to untransformed cells could, at least

in part, be consistent with the ability to target TrxR. Actually, it

is well known that the Trx system is overexpressed in cancer

cells, and has been recognized as a tumour-specific target for

the development of selective anticancer agents. In summary,

we suggest that a simple diiron platform might offer an arsenal

of tools not available for widely investigated ferrocenes, paving

the way for the development of iron-based drugs with

optimized features. Further biochemical and biophysical studies

will be performed to clarify in more detail the mechanism of

action of this new class of organometallic anticancer candi-

dates.

Synthesis and characterization of compounds

[Fe₂Cp₂(CO)₃(CNR)], [1A–H].^[19]General procedure. In a 250 mL

round-bottom flask equipped with a reflux condenser under N₂, the

selected isocyanide, or its solution in MeCN (10 mL, for solid

compounds), was added dropwise to a suspension of [Fe₂Cp₂(CO)₄]

(1.6 equiv) in anhydrous MeCN (80–100 mL). Alkyl isocyanides: the

Experimental Section

General experimental details: [Fe₂Cp₂(CO)₄] (99%) was purchased

from Strem Chemicals, other reactants and solvents were obtained

from Alfa Aesar, Merck, Apollo Scientific or TCI Chemicals and were

°

mixture was heated at reflux (T�100 C) for 8 h, then stirred at

room temperature for additional 14 h. Aryl isocyanides: the mixture

was stirred at room temperature for 14 h then heated at reflux (T�

°

100 C) for 1–3 h. Next, conversion was checked by IR and the dark

°

of the highest purity available. Isocyanides were stored at 4 C or

red-brown suspension was dried under vacuum (40 C). The

°

À 20 C and used as received; contaminated labware was treated

resulting solid, consisting of

a

mixture of [Fe₂Cp₂(CO)₄] and

[Fe₂Cp₂(CO)₃(CNR)], was directly used in the following alkylation

procedure without any purification.

°

with bleach. Methyl triflate was stored under N₂at 4 C; allyl iodide

and triethyloxonium tetrafluoroborate (1 M solution in CH₂Cl₂) were

[19]

°

stored under N₂at À 20 C. Compounds [1A–H]

and [2D–F]

[Fe₂Cp₂(CO)₂(μ-CO){μ-CNMe(1H-indol-6-yl)}]CF₃SO₃, [2A]CF₃SO₃

[20,21]

CF₃SO₃

were prepared according to published procedures.

Alkylation reactions were carried out under dry N₂using standard

Schlenk techniques and solvents distilled over appropriate drying

agents (MeCN from CaH₂, CH₂Cl₂from P₂O₅). The synthesis and

chromatographic purification of 6 was carried out under N₂using

deaerated solvents. Anion-exchange reactions and all other oper-

ations were conducted under air with common laboratory glass-

ware. Chromatography separations were carried out on alumina

columns (Merck; neutral, 4% w/w water content except where

otherwise noted). Ion-exchange chromatography was performed

with an Amberlyst® 15 hydrogen form resin (Superlco/Merck), pre-

activated with a methanolic solution of NaOH. Once isolated,

compounds [2B]X (X=CF₃SO₃, BF₄), [2C]Cl, [5H]CF₃SO₃(hygroscopic)

and 6 were stored under N₂; all other Fe compounds being air- and

moisture-stable in the solid state. NMR spectra were recorded at

In a 150 mL Schlenk tube under N₂, CF₃SO₃Me (0.11 mL, 1.0 mmol)

was added dropwise to a dark red suspension of 1A (0.975 mmol)

in anhydrous CH₂Cl₂(50 mL) under stirring. Therefore, the mixture

was stirred at room temperature for 6 h and conversion was

checked by IR (CH₂Cl₂). Next the suspension (red solution+solid)

was moved on top of an alumina column (h 7, d 4.3 cm). Impurities

were eluted with neat CH₂Cl₂and THF, then a raspberry red band

containing the title product was eluted with neat MeCN. Volatiles

were removed under vacuum and residue was triturated in Et₂O.

The suspension was stirred at room temperature for a few hours

and then filtered. The resulting red solid was washed with toluene,

°

25 C on a Bruker Avance II DRX400 instrument equipped with a

BBFO broadband probe. Chemical shifts (expressed in parts per

million) are referenced to the residual solvent peaks^[45](¹H, ¹³C) or to

external standards⁴⁶(¹⁴N to CH₃NO₂; ¹⁹F to CFCl₃, ³⁵Cl to 1 M NaCl in

D₂O, ³¹P to 85% H₃PO₄). In [D₆]DMSO/D₂O mixtures, ¹H chemical

shifts are referenced to the [D₅]DMSO signal as in pure [D₆]DMSO

(δ/ppm=2.50); in D₂O/CD₃OD mixtures, ¹H chemical shifts are

referenced to the CD₃OD residual peak as CH₃OH in pure D₂O (δ/

1

ppm=3.34). H and ¹³C spectra were assigned with the assistance

°

Et₂O and dried under vacuum (40 C). Yield: 358 mg, 71%.

1

of H{³¹P}, ¹³C DEPT 135, H,¹H COSY, H-¹³C gs-HSQC and H-¹³C gs-

Compound [2A]CF₃SO₃is soluble in MeCN, THF, MeOH, CH₂Cl₂, less

soluble in CH₂Cl₂, CHCl₃, insoluble in Et₂O, toluene and water. X-ray

quality crystals of [2A]CF₃SO₃were obtained from an acetone

HMBC experiments.^[47]CDCl₃stored in the dark over Na₂CO₃was

used for NMR analysis. IR spectra of solid samples (650–4000 cm^{À 1}

)

were recorded on a Perkin Elmer Spectrum One FTIR spectrometer,

equipped with a UATR sampling accessory; IR spectra of solutions

°

solution layered with Et₂O and settled aside at À 20 C. Anal. calcd.

for C₂₅H₂₁F₃Fe₂N₂O₆S: C 46.47, H 3.28, N 4.34; found: C 46.09, H 3.28,

N 4.34. IR (solid state): ν˜/cm^{À 1}=3330w-br (ν_NH), 3114w, 2004 s

(ν_CO), 1985 s (ν_CO), 1833 s (ν_μ-CO), 1583w-sh, 1555 m, 1540 m (ν_CN),

1475w, 1458w, 1433w, 1420w, 1397 m, 1361w, 1345w, 1323w,

1290–1276 m, 1244 s (ν_SO3), 1223 s (ν_SO3), 1154 s (ν_SO3), 1107 m-sh,

1067w, 1028 s, 1016 s-sh, 1002 m-sh, 942w, 931w, 893w, 875w,

860 m, 845 m, 817w, 797 m, 772 s, 756w-sh, 743 m-sh, 728 m. IR

(CH₂Cl₂): ν˜/cm^{À 1}=2022 s (ν_CO), 1992 m-sh (ν_CO), 1836 s (ν_μ-CO),

1554 m, 1542 m (ν_CN). IR (MeCN): ν˜/cm^{À 1}=2023 s (ν_CO), 1991 m-sh

(ν_CO), 1836 s (ν_μ-CO), 1554 m, 1542 m-sh (ν_CN). ¹H NMR (CD₃CN): δ/

were recorded using

a CaF₂liquid transmission cell (2300–

1500 cm^{À 1}) on a Perkin Elmer Spectrum 100 FTIR spectrometer. UV/

Vis spectra (250–800 nm) were recorded on a Ultraspec 2100 Pro

spectrophotometer using PMMA cuvettes (1 cm path length). IR

and UV/Vis spectra were processed with Spectragryph software.^[48]

Carbon, hydrogen and nitrogen analyses were performed on a

Vario MICRO cube instrument (Elementar). GC analysis was

performed on a Clarus 500 instrument (PerkinElmer) equipped with

a 5 Å MS packed column (Supelco) and a TCD detector. Samples

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

9

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

3

ppm=9.78 (s-br, 1H, NH), 7.85 (s-br, 1H, C9-H), 7.70 (d, ³J_HH=8.6 Hz,

room temperature. ¹³C{¹H} NMR (CDCl₃): δ/ppm=323.7 (d, J_CP

=

3

1H, C4-H), 7.48 (t, J_HH=2.7 Hz, 1H, C6-H), 7.37 (d-br, J_HH=7.2 Hz,

1H, C3-H), 6.67À 6.62 (m, 1H, C7); 5.39, 4.65 (s, 10H, Cp), 4.53 (s, 3H,

C1). ¹³C{¹H} NMR (CD₃CN): δ/ppm=324.9 (CN), 255.9 (μ-CO); 209.8,

209.2 (CO); 144.6 (C2), 136.3 (C5), 129.1 (C8), 128.8 (C6), 118.8 (C3),

117.3 (C9), 113.8 (C4), 103.5 (C7); 91.2, 91.0 (Cp); 58.5 (C1). ¹⁹F{¹H}

6.4 Hz, CN), 254.7 (μ-CO); 207.4, 207.1 (CO); 90.3, 90.1 (Cp); 63.3 (C3),

62.5 (d, J_CP=146 Hz, C2), 53.2 (C1), 16.4 (t, J_CP=5.6 Hz, C4). ¹⁹F{¹H}

NMR (CDCl₃): δ/ppm=-151.9 (¹⁰BF₄), À 152.0 (¹¹BF₄). ³¹P{¹H} NMR

(CDCl₃): δ/ppm=17.3.

1

3

1

NMR (CD₃CN): δ/ppm=À 79.3. H NMR (CDCl₃): δ/ppm=9.57 (s-br,

3

[Fe₂Cp₂(CO)₂(μ-CO){μ-CNMe(C₆H₁₁)}]X, [2C]X (X=CF₃SO₃, Cl)

1H, NH), 8.0–7.0 (br, C3-H+C9-H), ( 7.65 (d, J_HH=8.4 Hz, 1H, C4-H),

7.42–7.39 (m, 1H, C6-H), 6.60 (s-br, 1H, C7-H); 5.41, 4.63 (s, 10H, Cp);

4.56 (s, 3H, C1-H).

[Fe₂Cp₂(CO)₂(μ-CO){μ-CNMe(CH₂PO₃Et₂)}]X, [2B]X (X=BF₄,

CF₃SO₃)

[2C]CF₃SO₃. In a 500 mL round bottom Schlenk flask under N₂,

CF₃SO₃Me (2.8 mL, 26 mmol) was added dropwise to a dark red

suspension of 1C (22.5 mmol) in anhydrous CH₂Cl₂(100 mL) under

stirring. Therefore, the mixture was stirred at room temperature for

4 h and the conversion was checked by IR (CH₂Cl₂). Next, the dark

red solution was moved on top of an alumina column (h 13, d

6 cm). Impurities were eluted with neat CH₂Cl₂and THF, then a red

band containing [2C]CF₃SO₃was eluted with MeCN/MeOH (20:1, v/

v). A red band, containing a minor fraction of [2C]Cl, was collected

using MeOH as eluent. Volatiles were removed under vacuum from

the MeCN/MeOH solution and residue was triturated in Et₂O. The

suspension was stirred at room temperature for a few h then

filtered. The resulting red solid was washed with toluene, Et₂O and

[2B]CF₃SO₃. In a 150 mL Schlenk tube under N₂, CF₃SO₃Me (0.20 mL,

1.8 mmol) was added dropwise to a dark red solution of 1B

(1.38 mmol) in anhydrous CH₂Cl₂(40 mL) under stirring. Therefore,

the mixture was stirred at room temperature for 7 h and conversion

was checked by IR (CH₂Cl₂). Next, the solution was moved on top of

an alumina column (h 8, d 5.5 cm). Impurities were eluted with neat

CH₂Cl₂, THF and MeCN, then a red band containing the title product

was eluted with MeCN/MeOH (10:1, v/v). Volatiles were removed

°

under vacuum (40 C), affording a red, highly hygroscopic solid.

Yield: quantitative. IR (CH₂Cl₂): ν˜/cm^{À 1}=2028 s (ν_CO), 1997w-sh

°

dried under vacuum (40 C). Yield: 11.44 g, 85%. Compound [2C]

(ν_CO), 1836 m (ν_μ-CO). ¹H NMR (CDCl₃): δ/ppm=5.32 (s, 10H, Cp), 5.23

CF₃SO₃is soluble in MeCN, MeOH, CH₂Cl₂, CHCl₃, less soluble in

water, poorly soluble in toluene, insoluble in Et₂O, hexane. X-ray-

quality crystals of [2C]CF₃SO₃were obtained from a CH₂Cl₂solution

(dd, ²J_HH=14.9 Hz, ²J_HP=5.3 Hz, 1H, C2-H), 4.93 (t, ²J_HH=²J_HP

=

15.5 Hz, 1H, C2-H’), 4.38 (s, 3H, C1-H), 4.23–4.10 (m, 4H, C3-H), 1.30

(q, ⁴J_HP=³J_HH=7.1 Hz, 6H, C4-H). ¹⁹F{¹H} NMR (CDCl₃): δ/ppm=

À 78.2. ³¹P{¹H} NMR (CDCl₃): δ/ppm=17.3.

°

layered with heptane and settled aside at À 20 C. Anal. calcd. for

C₂₂H₂₄F₃Fe₂NO₆S: C 44.10, H 4.04, N, 2.34; found: C 44.47, H 4.04, N,

2.34. IR (solid state): ν˜/cm^{À 1}=3102w, 2937w, 2862w, 2185w,

2146w, 2006 s (ν_CO), 1982 s-sh (ν_CO), 1822 s (ν_μ-CO), 1565 m (ν_CN),

1544 m-sh, 1452w, 1435w, 1421w, 1402w, 1366w, 1352w, 1321w,

1272 s-sh, 1257 s (ν_SO3), 1223 s-sh (ν_SO3), 1150s (ν_SO3), 1056 m, 1029 s,

990 m, 864 m, 854 m, 796 s, 746 s. IR (CH₂Cl₂): ν˜/cm^{À 1}=2020s (ν_CO),

[2B]BF₄. In a 500 mL round bottom flask, [2B]CF₃SO₃was suspended

in water (100 mL) with vigorous stirring until completely dissolved

then treated with NaBF₄(500 mg). The red solution was extracted

with CH₂Cl₂(3×20 mL) and the combined organic fractions were

dried under vacuum. The residue was suspended in water and the

ion exchange procedure with NaBF₄was repeated (×3); ¹⁹F{¹H} NMR

analysis of_Àthe final CH₂Cl₂solution indicated the complete removal

of CF₃SO₃anions. Therefore, volatiles were removed under

vacuum; the residue was dissolved in CH₂Cl₂and filtered over celite.

A red foamy hygroscopic solid, obtained upon volatiles removal

without heating, was dried under vacuum and stored under N₂.

Yield: 751 mg, 90% (with respect to 1B). Compound [2B]BF₄is

soluble in MeOH, MeCN, CH₂Cl₂, CHCl₃, water, insoluble in Et₂O and

hexane. Anal. calcd. for C₂₀H₂₅BF₄Fe₂NO₆P: C 39.71, H 4.17, N 2.31;

found: C 39.20, H 4.25, N 2.26. IR (solid state): ν˜/cm^{À 1}=3117w,

2987w, 2938w, 2913w, 2875w, 2012 s (ν_CO), 1986 s-sh (ν_CO), 1823 s

(ν_μ-CO), 1568 m (ν_CN), 1479w, 1446w, 1435w, 1421w, 1399 m, 1370w-

sh, 1297–1286w, 1254 m, 1225 m-sh, 1180w, 1163w, 1075 m-sh,

1033 s-sh (ν_BF4), 1008 s, 973 s, 951 s-sh, 904 m, 864 m-sh, 851 m,

837 m-sh, 780s, 745 s, 710 m-sh. IR (CH₂Cl₂): ν˜/cm^{À 1}=2028 s (ν_CO),

1996w (ν_CO), 1836 m (ν_μ-CO), 1574w (ν_CN). IR (MeCN): ν˜/cm^{À 1}=2026 s

1988w-sh (ν_CO), 1835 m (ν_μ-CO), 1567w (ν_CN). IR (MeCN): ν˜/cm^{À 1}

=

2021 s (ν_CO), 1988w-sh (ν_CO), 1835 m (ν_μ-CO), 1568w (ν_CN). ¹H NMR

3

(CDCl₃): δ/ppm=5.36, 5.27 (s, 10H, Cp); 4.68 (t, J_HH=11.7 Hz, 1H,

C2-H), 4.06 (s, 3H, C1-H), 2.81 (d, J=11.8 Hz, 1H, C4-H), 2.15 (d, J=

14.6 Hz, 1H, C6-H), 2.03–1.94 (m, 2H, C3-H+C4-H’), 1.86–1.72 (m,

3H, C7-H+C5-H), 1.53–1.35 (m, 3H, C3-H’+C6-H’), 1.34–1.20 (m, 1H,

1

C7-H’). No changes were observed in the H spectrum after 14 h at

room temperature. ¹³C{¹H} NMR (CDCl₃): δ/ppm=316.4 (CN), 255.3

(μ-CO); 208.5, 207.6 (CO); 90.3, 90.1 (Cp); 79.8 (C2), 46.8 (C1), 31.2

(C5), 30.7 (C4), 26.1 (C3), 26.0 (C6), 25.1 (C7). ¹⁹F{¹H} NMR (CDCl₃): δ/

ppm=À 78.2. ¹H NMR (CD₃OD): δ/ppm=5.40, 5.36 (s, 10H, Cp), 4.07

(s, 3H, C1-H), 2.41 (d, J=11.1 Hz, 1H, C4-H), 2.17–2.07, 2.03–1.91,

1.85–1.50, 1.42–1.32 (m, C₆H₁₁).

[2C]Cl. Compound [2C]CF₃SO₃(110 mg, 0.184 mmol) was dissolved

in MeOH (2 mL) then moved on top of an Amberlyst 15 column

(Na⁺form; h 7 cm, d 2.3 cm). NaCF₃SO₃was eluted with neat MeOH

then a red band containing [2C]⁺was eluted with NaCl-saturated

MeOH. Volatiles were removed under vacuum and the residue was

suspended in MeCN. The suspension was_Àfiltered over celite; the

filtrate was checked for absence of CF₃SO₃by ¹⁹F NMR then dried

1

(ν_CO), 1994w (ν_CO), 1838 m (ν_μ-CO), 1569w (ν_CN). H NMR (CDCl₃): δ/

ppm=5.29 (s, 10H, Cp), 5.15 (dd, ²J_HH=16.4 Hz, ²J_HP=4.0 Hz, 1H,

C2-H), 4.93 (t, ²J_HH=²J_HP=15.6 Hz, 1H, C2-H’), 4.35 (s, 3H, C1-H),

4

3

4.24–4.10 (m, 4H, C3-H), 1.30 (dt, J_HP=9.8 Hz, J_HH=7.1 Hz, 6H, C4-

H). No changes were observed in the ¹H spectrum after 14 h at

°

under vacuum (40 C). The resulting red, hygroscopic solid was

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

10

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

stored under N₂. Yield: 83 mg, 93%. Compound [2C]Cl shows the

same solubility pattern as the triflate salt. X-ray quality crystals of

[2C]Cl·2.34H₂O were obtained from a CH₂Cl₂solution layered with

[Fe₂Cp₂(CO)₂(μ-CO){μ-CNMe(CH₂CHCH₂)}]X, [3F]X (X=I,

CF₃SO₃)

°

hexane and settled aside at À 20 C. Anal. calcd. for C₂₁H₂₄ClFe₂NO₃:

C 51.94, H 4.98, N 2.88; found: C 51.99, H 5.02, N 2.83. IR (solid

state): ν˜/cm^{À 1}=3376 m-br* (ν_OH), 3098–3059w, 2933 m, 2858w,

2180–2160w, 1997 s (ν_CO), 1969 s-sh (ν_CO), 1811 s (ν_μ-CO), 1662w,

1631w, 1565 s (ν_CN), 1543 m, 1450 m, 1434w, 1419 m, 1400 m,

1350w, 1319w, 1263w-sh, 1249w, 1173w, 1154w, 1054 m, 1026w,

1015w, 990 m, 924w-sh, 893w-sh, 852 m, 795 s, 746 s, 729 s-sh.

*Due to moisture. IR (CH₂Cl₂): ν˜/cm^{À 1}=2018 s (ν_CO), 1985w-sh (ν_CO),

1833 m (ν_μ-CO), 1569w (ν_CN). IR (MeCN): ν˜/cm^{À 1}=2020s (ν_CO), 1987w-

1

sh (ν_CO), 1834 m (ν_μ-CO), 1568w (ν_CN). H NMR (CDCl₃): δ/ppm=5.51,

[3F]I. The preparation of the title compound was optimized with

respect to the literature (69% yield with 10–30 equiv. of allyl

iodide).^[22]In a 150 mL Schlenk tube under N₂, allyl iodide (0.39 mL,

4.3 mmol) was added dropwise to a dark red solution of 1F

(1.54 mmol) in anhydrous MeCN (30 mL) under vigorous stirring.

The mixture was stirred for 4.5 h under protection from the light

and conversion was checked by IR (MeCN). Next, the dark red

solution was moved on top of an alumina column (Brockmann

activity I; h 6.5, d 4.3 cm). Impurities were eluted with neat Et₂O,

THF and MeCN, then a red band containing the title product was

eluted with MeCN/MeOH (10:1, v/v). Volatiles were removed under

5.39 (s, 10H, Cp); 4.75–4.66 (m, 1H, C2-H), 4.14 (s, 3H, C1-H), 3.16 (d,

²J_HH=13.4 Hz, 1H, C4-H), 2.15 (d, J_HH=14.2 Hz, 1H, C6-H), 2.03–1.93

2

(m, 2H, C3-H+C4-H’), 1.84–1.72 (m, 3H, C7-H+C5-H), 1.51–1.35 (m,

2H, C3-H’+C6-H’), 1.33–1.23 (m, 1H, C7-H’). No changes were

observed in the H spectrum after 24 h at room temperature. ¹³C

1

{¹H} NMR (CDCl₃): δ/ppm=316.2 (CN), 255.8 (μ-CO); 208.7, 207.8

(CO); 90.7, 90.3 (Cp); 79.7 (C2), 47.1 (C1), 31.3 (C5), 31.0 (C4), 26.1

(C3), 26.0 (C6), 25.2 (C7). ³⁵Cl NMR (CDCl₃): δ/ppm=7.61 (Δν_1/2

=

266 Hz).

°

vacuum (40 C) and the residue was triturated in Et₂O. The

suspension was stirred at room temperature then filtered. The

resulting red solid was washed with Et₂O and dried under vacuum

[Fe₂Cp₂(CO)₂(μ-CO){μ-CNMe(4-C₆H₄OMe)}]CF₃SO₃, [2D]CF₃SO₃

°

(40 C). Yield: 646 mg, 78%. Compound [3F]I is soluble in MeOH,

CH₂Cl₂, CHCl₃, poorly soluble in water, insoluble in Et₂O, hexane. IR

(CH₂Cl₂): ν˜/cm^{À 1}=2021 s (ν_CO), 1989w (ν_CO), 1835 m (ν_μ-CO), 1585w

(ν_CN). ¹H NMR (CDCl₃): δ/ppm=6.09–5.97 (m, 1H, CH=C, C3-H), 5.53–

5.46 (m, 3H, C2-H+C4-H); 5.45, 5.53 (s, 10H, Cp); 5.17–5.11 (m, 1H,

C2-H’), 4.25 (s, 3H, C1-H).

[3F]CF₃SO₃. In a 50 mL round bottom flask, a solution of Ag(CF₃SO₃)

(468 mg, 1.79 mmol) in MeCN (5 mL) was added to a dark red

°

suspension of [3F]I (950 mg, 1.78 mmol) in MeCN (6 mL) at 0 C. The

°

mixture stirred at 0 C for 2.5 h while maintaining the system under

The title compound was prepared with minor modifications of the

literature procedure.^[21]In

250 mL Schlenk tube under N₂,

protection from the light. The resulting suspension (dark red

solution+yellow AgI precipitate) was filtered over celite. Volatiles

were removed under vacuum from the filtrate solution and the

residue was triturated in a Et₂O/toluene mixture. The suspension

was stirred at room temperature for a few h, then filtered. The

resulting red solid was washed with toluene, Et₂O and dried under

a

CF₃SO₃Me (1.0 mL, 5.5 mmol) was added dropwise to a dark red

solution of 1B (4.63 mmol) in anhydrous CH₂Cl₂(40 mL) under

stirring. The mixture was stirred at room temperature for 5.5 h then

moved on top of an alumina column (Brockmann activity I; h 6.0, d

4.3 cm). Impurities were eluted with CH₂Cl₂and CH₂Cl₂/THF (1:1, v/

v), then a red band containing [2D]CF₃SO₃was eluted with neat

°

vacuum (40 C). Yield: 928 mg, 94%. Compound [3F]CF₃SO₃is

soluble in MeCN, acetone, MeOH, water, less soluble in CH₂Cl₂

>

°

MeCN. Volatiles were removed under vacuum (40 C) to afford a red

CHCl₃, poorly soluble in toluene, insoluble in Et₂O. Anal. calcd. for

C₁₉H₁₈F₃Fe₂NO₆S: C 40.96, H, 3.26, N, 2.51; found: C 41.20, H, 3.20, N,

2.72. IR (solid state): ν˜/cm^{À 1}=3106w, 2213w, 2178w, 2012 s (ν_CO),

1988 s (ν_CO), 1813 s (ν_μ-CO), 1644w, 1581 m (ν_CN), 1435w, 1416w,

1395w, 1261 s (ν_SO3), 1243 s-sh, 1222 m-sh (ν_SO3), 1169 s (ν_SO3),

1050 m, 1028 s, 989w-sh, 931 m, 888 m, 854 m, 758 s, 666 m. IR

(CH₂Cl₂): ν˜/cm^{À 1}=2022 s (ν_CO), 1990w (ν_CO), 1836 m (ν_μ-CO), 1584 m

(ν_CN). IR (MeCN): ν˜/cm^{À 1}=2023 s (ν_CO), 1990w (ν_CO), 1836 m (ν_μ-CO),

foamy solid. Yield: 2.446 g, 85%. Compound [2D]CF₃SO₃is soluble

in MeCN, MeOH, CH₂Cl₂, CHCl₃, water, insoluble in Et₂O and hexane.

X-ray quality crystals of [2D]CF₃SO₃were obtained from a CHCl₃

°

solution layered with hexane and settled aside at À 20 C. Anal.

calcd. for C₂₃H₂₀F₃Fe₂NO₇S: C 44.33, H 3.24, N 2.25; found: C 43.92, H

3.31, N 2.17. IR (solid state): ν˜/cm^{À 1}=3113w, 3091w, 2946w,

2849w, 2011 s (ν_CO), 1978 s (ν_CO), 1821 s (ν_μ-CO), 1608w, 1563 m,

1539 m (ν_CN), 1506 s, 1468w, 1449w, 1433w, 1420w, 1395 m, 1304w-

sh, 1278 s, 1251 s (ν_SO3), 1224 s-sh (ν_SO3), 1176 m-sh, 1146 s (ν_SO3),

1117 m-sh, 1108 m, 1067w, 1028 s, 890w, 855 s, 825w-sh, 773 s,

756 m-sh, 736w, 702 s. IR (CH₂Cl₂): ν˜/cm^{À 1}=2021 s (ν_CO), 1989w

(ν_CO), 1836 m (ν_μ-CO), 1607w, 1562w, 1540w (ν_CN), 1507 m. IR (MeCN):

ν˜/cm^{À 1}=2024 s (ν_CO), 1991w (ν_CO), 1837 m (ν_μ-CO), 1610w, 1563w,

1539w (ν_CN), 1508w. ¹H NMR (CDCl₃): δ/ppm=8.3–7.6 (br, 2H, C3-H),

1

3

1582 m (ν_CN). H NMR (CDCl₃): δ/ppm=5.98 (dddd, J_HH=17.4, 10.0,

7.5, 5.0 Hz, 1H, C3-H), 5.50–5.45 (m, 2H, C4-H+C4-H’); 5.34, 5.32 (s,

10H, Cp); 5.30–5.23 (m, 1H, C2-H), 5.15–5.08 (m, 1H, C2-H’), 4.17 (s,

1

3H, C1-H). No changes were observed in the H spectrum after 14 h

at room temperature. ¹³C{¹H} NMR (CDCl₃): δ/ppm=318.1 (CN),

255.3 (μ-CO); 208.1, 207.8 (CO), 130.2 (C3), 122.2 (C4), 121.0 (¹J_CF

=

319 Hz, CF₃); 90.20, 90.16 (Cp); 70.6 (C2), 51.3 (C1). ¹⁹F{¹H} NMR

3

7.10 (d, J_HH=7.4 Hz, 2H, C4-H); 5.43, 4.78 (s, 10H, Cp); 4.54 (s, 3H,

(CDCl₃): δ/ppm=À 78.2. ¹H NMR ([D₆]acetone): δ/ppm=6.33–6.19

C1-H), 3.90 (s, 3H, C6-H). ¹⁹F{¹H} NMR (CDCl₃): δ/ppm=À 78.1.

3

(m, 1H, C3-H), 5.62 (d, J_HH=17.2 Hz, 1H, C4-H); 5.58, 5.54 (s, 10H,

Cp); 5.41–5.16 (m, 3H, C2-H+C2-H’+C4-H’), 4.28 (s, 3H, C1-H).

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

11

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

(ν_CO), 1836 s (ν_μ-CO), 1596w, 1532 s (ν_CN), 1506 m-sh, 1458w, 1434w,

1421w, 1381w, 1360w, 1341w, 1275w, 1247w, 1211w, 1179w,

1134w, 1075 s-sh, 1049 s-br, 1033 s (ν_BF4), 1014 s-sh, 963w, 935w,

865w, 850w, 825w, 808 m, 745 s, 707 m, 657 m. IR (CH₂Cl₂): ν˜/

cm^{À 1}=2021 s (ν_CO), 1989w (ν_CO), 1836 m (ν_μ-CO), 1538w (ν_CN). IR

(MeCN): ν˜/cm^{À 1}=2023 s (ν_CO), 1990w (ν_CO), 1837 m (ν_μ-CO), 1537w

[Fe₂Cp₂(CO)₂(μ-CO){μ-CNMe(2-naphthyl)}]I, [3G]I

1

(ν_CN). H NMR ([D₆]acetone): δ/ppm=8.7–8.3 (br, 1H, C4-H/C12-H),

8.30 (d, ³J_HH=8.6 Hz, 1H, C5-H), 8.16–8.09 (m, 2H, C7-H+C10-H),

8.1–7.7 (br, 1H, C4-H/C12-H), 7.75–7.70 (m, 2H, C8-H+C9-H), 5.68 (s,

5H, Cp), 5.19 (q, ³J_HH=7.0 Hz, 2H, C1-H), 4.87 (s, 5H, Cp), 1.51 (t,

³J_HH=7.2 Hz, 3H, C2-H). No changes were observed in the ¹H

spectrum after 14 h at room temperature. ¹³C{¹H} ([D₆]acetone): δ/

ppm=325.2 (CN), 255.2 (μ-CO); 210.1, 209.8 (CO); 146.1 (C3); 134.3,

133.9 (C6+C11); 131.4 (C5); 129.5, 129.0 (C7+C10); 128.8, 128.6

(C8+C9); 126.3, 125.1 (C4+C12); 91.7, 91.2 (Cp); 66.8 (C1), 14.4

(C2). ¹⁹F{¹H} NMR ([D₆]acetone): δ/ppm=À 151.0 (¹⁰BF₄), À 151.1

In

a 25 mL Schlenk tube under N₂, methyl iodide (110 mL,

1.8 mmol) was added to a dark red solution of 1G (0.48 mmol) in

anhydrous MeCN (8 mL). The mixture was stirred for 2.5 h and

conversion was checked by IR (MeCN). Next, the dark red solution

was moved on top of an alumina column (h 10, d 2 cm). Impurities

were eluted with neat CH₂Cl₂and THF, then a red band containing

the title product was eluted with THF/MeOH (4:1, v/v). Next,

volatiles were removed under vacuum and the residue was

dissolved in CH₂Cl₂. The suspension was filtered over celite and the

1

(¹¹BF₄). H NMR (CDCl₃): δ/ppm=8.5–7.8 (br, 4H), 7.70–7.60 (m, 2H),

7.60–7.48 (m-br, 1H), 5.44 (s, 5H), 5.20–5.05 (s-br, 1H), 4.95 (td, J=

14.5, 7.3 Hz, 1H), 4.65 (s, 5H), 1.44 (t, J=7.3 Hz, 3H). ¹³C{¹H} NMR

(CDCl₃): δ/ppm=323.2 (br), 254.9, 209.2 (br), 208.4, ca. 145 (br); 133

(br), 132.9 (br); 130.1 (m), 128.1, 127.9, 126–125 (m-br), 124.0, 120.5,

117.1, 90.6, 90.3, 66 (br), 14.0. ¹⁹F{¹H} NMR (CDCl₃): δ/ppm=À 150.9,

À 151.0. The cis orientation of the Cp ligands in [4G]BF₄was

ascertained by ¹H NOE experiment in CDCl₃. Thus, irradiation of one

Cp resonance (δ_H5.44 or 4.65 ppm) evidenced a NOE interaction

with the other.

°

filtrate was taken to dryness under vacuum (40 C), affording a dark

red solid. Yield: 208 mg, 69%. A considerable decrease in yield and

purity was observed for reactions carried out in refluxing CHCl₃or

in MeCN at lower temperatures. Anal. calcd. for C₂₅H₂₀Fe₂INO₃: C

48.35, H 3.25, N 2.56; found: C 48.20, H 3.15, N 2.49. Compound

[3G]I is soluble in MeOH, CH₂Cl₂, CHCl₃, sparingly soluble in water,

insoluble in hexane. IR (CH₂Cl₂): ν˜/cm^{À 1}=2021 s (ν_CO), 1989 m-sh

1

(ν_CO), 1836 m (ν_μ-CO), 1564w, 1545w, 1538w (ν_CN). H NMR (CDCl₃): δ/

[Fe₂Cp₂(CO)₂(μ-CO){μ-CNBn₂}]X, [5H]X (X=Br, CF₃SO₃)

ppm=8.10 (d+m, 2H), 7.98À 7.95 (m, 1H), 7.94À 7.77 (m, 1H),

7.66À 7.52 (m, 2H), 7.55À 7.47 (m, 1H) (C₁₀H₇); 5.62, 4.82 (s, 10H,

Cp); 4.75 (s, 3H, NCMe).

[Fe₂Cp₂(CO)₂(μ-CO){μ-CNEt(2-naphthyl)}]BF₄, [4G]BF₄

[5H]Br. In a 100 mL Schlenk tube under N₂, benzyl bromide (1.0 mL,

8.4 mmol) was added dropwise to a dark red solution of 1H

(1.58 mmol) in anhydrous MeCN (50 mL) under vigorous stirring at

°

60 C. The mixture was stirred at reflux for 1.5 h and conversion was

checked by IR (MeCN). Next, the dark red solution was moved on

top of an alumina column (h 6.5, d 5.5 cm). Impurities were eluted

with neat CH₂Cl₂, THF and MeCN, then a red band containing the

title product was eluted with MeCN/MeOH (10:1, v/v). Volatiles

°

were removed under vacuum (40 C), affording a red solid. Yield:

In a 150 mL Schlenk tube under N₂, [Et₃O]BF₄(1 M solution in

CH₂Cl₂; 0.71 mL, 0.71 mmol), was added dropwise to a dark red

892 mg, 92%. Compound [5H]Br is soluble in CH₂Cl₂, CHCl₃, THF,

iPrOH, poorly soluble in toluene, EtOAc, insoluble in Et₂O. IR

(CH₂Cl₂): ν˜/cm^{À 1}=2020s (ν_CO), 1988w (ν_CO), 1837 m (ν_μ-CO), 1550w

°

solution of 1G (0.724 mmol) in anhydrous MeCN (30 mL) at 0 C

under vigorous stirring. Therefore, the mixture was stirred for 3 h,

while allowing to reach room temperature. Next, conversion was

checked by IR (MeCN) then volatiles were removed under vacuum.

The residue was dissolved in a small volume of THF then moved on

top of an alumina column (h 6.5, d 4.3 cm). Impurities were eluted

with neat Et₂O and THF, then a red band containing the title

product was eluted with MeCN/MeOH (10:1, v/v). Volatiles were

1

3

(ν_CN), 1534w. H NMR (CDCl₃): δ/ppm=7.48 (t, J_HH=7.4 Hz, 4H, C4-

3

H), 7.40 (t, J_HH=7.3 Hz, 2H, C5-H), 7.30–7.26 (m, 4H, C3-H), 5.69 (d,

²J_HH=14.7 Hz, 2H, C1-H), 5.56 (s, 10H, Cp), 5.52 (d, ²J_HH=14.7 Hz, 2H,

C1-H’).

[5H]CF₃SO₃. In a 50 mL round bottom flask, a solution of Ag(CF₃SO₃)

(440 mg, 1.71 mmol) in MeCN (5 mL) was added to a dark red

solution of [5H]I (1045 mg, 1.70 mmol) in MeCN (40 mL) at 0 C. The

mixture was stirred at 0 C for 1.5 h while maintaining the system

under protection from the light. The resulting suspension (dark red

solution+pale yellow AgBr precipitate) was filtered over celite and

the filtrate solution was dried under vacuum. The residue was

treated with hexane (10 mL) then Et₂O (50 mL) and settled aside for

30 min. Therefore, the solution was discarded and the washing

procedure was repeated (x 2). The resulting red solid was dried

°

removed under vacuum (40 C) and the residue was triturated in

°

Et₂O. The suspension was stirred at room temperature for a few h

then filtered. The resulting dark red solid was washed with Et₂O

°

and dried under vacuum (40 C). Yield: 284 mg, 66%. A considerable

decrease in yield and purity was observed for reactions carried out

in CH₂Cl₂or in MeCN at room temperature. Compound [4G]BF₄is

soluble in MeOH, MeCN, CH₂Cl₂, CHCl₃, poorly soluble in Et₂O,

insoluble in water. Anal. calcd. for C₂₆H₂₂BF₄Fe₂NO₃: C 52.49, H 3.72,

N 2.35; found: C 52.11, H 3.67, N 2.30. IR (solid state): ν˜/cm^{À 1}

=

°

under vacuum (40 C) and stored under N₂(slightly hygroscopic).

3116w, 3060w, 2979w, 2937w, 2160w-br, 2007 s (ν_CO), 1937 s-sh

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

12

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

Yield: 1076 mg, 93%. Compound [5H]CF₃SO₃is soluble in MeCN,

acetone, MeOH, CH₂Cl₂, poorly soluble in Et₂O, toluene, water and

insoluble in hexane. Anal. calcd. for C₂₉H₂₄F₃Fe₂NO₆S: C 50.98, H,

3.54, N, 2.05; found: C 50.59, H, 3.58, N, 2.26. IR (solid state): ν˜/

cm^{À 1}=3106w, 3070–3230w, 2166w, 2013 s (ν_CO), 1985 s-sh (ν_CO),

1827 s (ν_μ-CO), 1588w, 1549 m (ν_CN), 1533 m, 1497w, 1454w, 1433w,

1420w, 1352w, 1260s (ν_SO3), 1223 s-sh (ν_SO3), 1153 s (ν_SO3), 1096w,

86.6 (Cp); 86.4, 86.3 (Cp’); 76.6, 75.4 (C2-H); 44.4, 43.6 (C1-H), 32.5,

32.1 (C3-H); 32.3, 30.6, 26.3, 26.2, 26.0, 25.6, 25.6, 25.5 (C3’-H+C4-H

+C5-H).

X-ray crystallography

1079w, 1029 s, 858 m, 767 s, 737 m-sh, 698 s. IR (CH₂Cl₂): ν˜/cm^{À 1}

=

2023 s (ν_CO), 1991 m (ν_CO), 1840 m (ν_μ-CO), 1550w (ν_CN), 1534w. IR

(MeCN): ν˜/cm^{À 1}=2023 s (ν_CO), 1990w (ν_CO), 1839 m (ν_μ-CO), 1552w

(ν_CN), 1538w. ¹H NMR (CDCl₃): δ/ppm=7.50–7.44 (m, 4H, C4-H),

7.44–7.39 (m, 2H, C5-H), 7.20 (d, ³J_HH=7.3 Hz, 4H, C3-H), 5.66 (d,

Crystal data and collection details for [2A]CF₃SO₃, [2C]

Cl·2.34H₂O, [2C]CF₃SO₃, [2D]CF₃SO₃and 6 are reported in

Table 7. Data were recorded on a Bruker APEX II diffractometer

equipped with a PHOTON2 detector using Mo_Kαradiation. The

structures were solved by direct methods and refined by full-

matrix least-squares based on all data using F².^[49]Hydrogen

atoms were fixed at calculated positions and refined using a

riding model, except H(2) in [2A]CF₃SO₃which was located in

the Fourier map and refined isotropically using the 1.2-fold U_iso

of the parent N(2) atom. The H₂O molecules of [2C]Cl·2.34H₂O

are disordered and it was not possible to locate the hydrogen

atoms. All non-hydrogen atoms were refined with anisotropic

displacement parameters. The crystals of 6 display a low quality,

allowing the full determination of the overall connectivity and

geometry of the complex, while the bonding parameters

cannot be discussed in detail.

2

²J_HH=14.6 Hz, 2H, C1-H), 5.50 (d, J_HH=14.7 Hz, 2H, C1-H’), 5.40 (s,

10H, Cp). No changes were observed in the ¹H spectrum after

2 days at room temperature. ¹³C{¹H} NMR (CDCl₃): δ/ppm=324.4

(CN), 253.7 (μ-CO), 208.2 (CO), 132.1 (C2), 129.9 (C4), 129.4 (C5),

127.8 (C3), 121.1 (¹J_CF=313 Hz, CF₃), 90.6 (Cp), 68.9 (C1). ¹⁹F{¹H} NMR

(CDCl₃): δ/ppm=À 78.1 ppm.

[Fe₂Cp₂Cl(CO)(μ-CO){μ-CNMe(C₆H₁₁)}], 6

Computational studies

All geometries were optimized with ORCA 4.0.1.2,^[50]using the

B97 functional in conjunction with a triple-ζ quality basis set

(def2-TZVP). The dispersion corrections were introduced using

the Grimme D3-parametrized correction and the BeckeÀ

Johnson damping to the DFT energy.^[51]Most of the structures

were confirmed to be local energy minima (no imaginary

frequencies), but in some case a small, unavoidable negative

frequency relative to the Cp rotation around the MÀ Cp axis was

observed. The solvent was considered through the continuum-

like polarizable continuum model (C-PCM, dichloromethane).

In a 25 mL Schlenk tube under N₂, a red suspension of [2C]CF₃SO₃

(103 mg, 0.172), Me₃NO·2H₂O (35 mg, 0.32 mmol) and LiCl (23 mg,

0.54 mmol) in acetone (8 mL) was stirred at reflux for 5 h.

Conversion was checked by IR (acetone) then volatiles were

removed under vacuum. The brown residue was suspended in

CH₂Cl₂and moved on top of an alumina column (h 2.5, d 4.3 cm),

under N₂. Impurities were eluted with CH₂Cl₂then a brown band

containing the title product was eluted with THF. The eluate was

dried under vacuum and the residue was suspended hexane. The

suspension was filtered; the resulting brown solid was washed with

°

hexane and dried under vacuum (40 C). Yield: 55 mg, 70%. A

related one-pot reaction with [2C]CF₃SO₃/Me₃NO·2H₂O (1.6 equiv)/

°

NaCl (3.6 equiv) in deaerated MeOH (8 mL) at 40 C overnight gave

Behaviour in aqueous media

6 in 43% yield. A two-step procedure comprising the [2C]CF₃SO₃/

Me₃NO·2H₂O (1 equiv) reaction in MeCN at room temperature

followed by LiCl (3.0 equiv) in refluxing acetone for 3 h gave 6 in

54% yield. Compound 6 is soluble in CH₂Cl₂, CHCl₃, scarcely soluble

in MeOH, insoluble in Et₂O and hydrocarbons. Crystals of 6 for X-ray

analysis were obtained from a CHCl₃solution layered with pentane

Solubility in water. A suspension of the selected Fe compound

(3–5 mg) in a D₂O solution (0.7 mL) containing Me₂SO₂as

internal standard⁵²(3.36·10^{À 3}M) was vigorously stirred at 21 C

°

for 10 h. The resulting saturated solution was filtered over

and settled aside at À 20 C. IR (CH₂Cl₂): ν˜/cm^{À 1}=1977 s (ν_CO),

°

1

1798 s (ν_μ-CO), 1537 m (ν_CN). IR (MeCN): ν˜/cm^{À 1}=1971 s (ν_CO),

1795 m (ν_μ-CO), 1539w (ν_CN). IR (solid state): ν˜/cm^{À 1}=3104w, 3070w,

1939 s (ν_CO), 1782 s (ν_μ-CO), 1558w-sh, 1536 s (ν_CN), 1456w, 1447w,

1419w, 1394 m, 1360w, 1346w, 1317w, 1251w, 1182w, 1153w,

1116w, 1062 m, 1016w, 1001w, 993 m, 857w-sh, 838 m, 814 m,

celite, transferred into an NMR tube and analysed by H NMR

spectroscopy (delay time=3 s; number of scans=20). For the

most soluble compounds (S�0.05 M), a saturated solution was

prepared in a smaller volume of D₂O/Me₂SO₂(0.2–0.3 mL), then

filtered and diluted with pure D₂O. The concentration (solubil-

ity) was calculated by the relative integral with respect to

Me₂SO₂(δ/ppm=3.14 (s, 6H)). Results are compiled in Table 3.

Octanol/water partition coefficients (logP_ow). Partition coef-

800s, 757 s. H/¹³C NMR: mixture of isomers in 3:2 ratio (¹H CDCl₃).

1

Signals attributable to the minor isomer are italicized. ¹H NMR

(CDCl₃): δ/ppm=6.09, 5.01 (tt, ³J_HH=12.0, 3.2 Hz, 1H, C2-H); 4.72,

4.70 (s, 5H, Cp); 4.64, 4.62 (s, 5H, Cp’); 4.53, 4.05 (s, 3H, C1-H); 2.59,

2.37 (d, J=12.0 Hz, 1H, C3-H); 2.28–2.06, 1.98–1.69, 1.42–1.24 (m,

ficients (P_ow; IUPAC: K_Dpartition constant^[53]), defined as P_ow

_org/c_aq, where c_organd c_aqare molar concentrations of the

=

1

9H, C3-H’+C4-H+C5-H). No change in the H NMR spectrum was

c

observed after 14 h at room temperature. ¹³C{¹H} NMR (CDCl₃): δ/

ppm=334.8, 334.5 (CN); 268.1, 267.5 (μ-CO); 212.5, 211.9 (CO); 86.8,

selected compound in the organic and aqueous phase,

respectively, were determined by the shake-flask method and

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

13

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

Table 7. Crystal data and measurement details for [2A]CF₃SO₃, [2C]Cl·2.34H₂O, [2C]CF₃SO₃, [2D]CF₃SO₃and 6.

[2C]Cl·2.34H₂O

[2C]CF₃SO₃

[2A]CF₃SO₃

[2D]CF₃SO₃

6

Formula

FW

T, K

C₂₁H₂₄ClFe₂NO_5.34

523.04

100(2)

C₂₂H₂₄F₃Fe₂NO₆S

599.18

100(2)

C₂₄H₁₉F₃Fe₂N₂O₆S

632.17

100(2)

C₂₃₂H₂₀F₃Fe₂NO₇S

623.16

100(2)

C₂₀H₂₄ClFe₂NO₂

457.55

100(2)

λ, Å

Crystal system

0.71073

triclinic

0.71073

orthorhombic

0.71073

monoclinic

0.71073

monoclinic

0.71073

monoclinic

Space group

Pna2₁

P2₁/c

P2₁/n

�

P1

a, Å

b, Å

c, Å

10.8828(6)

15.1007(8)

15.1537(8)

77.457(3)

87.276(3)

69.367(3)

2273.8(2)

4

21.468(3)

11.9137(18)

8.9585(16)

90

2291.3(6)

4

1.737

9.3327(8)

20.1635(17)

12.8705(10)

90

100.145(4)

90

2384.1(3)

4

1.761

9.9264(9)

14.5214(12)

17.1821(15)

90

102.180(3)

90

2421.0(4)

4

1.710

6.9490(15)

22.695(5)

12.227(2)

90

97.775(6)

90

1910.7(7)

4

1.591

°

α,

°

β,

γ,

°

Cell Volume, Å³

Z

D_c, gcm^{À 3}

1.528

1.426

μ, mm^{À 1}

1.423

1.374

1.353

1.672

F(000)

Crystal size, mm

θ limits,

Reflections collected

Independent reflections

Data/restraints /parameters

Goodness on fit on F²

R₁(I >2σ(I))

1075

1224

1280

1264

944

0.21×0.18×0.15

1.790–26.000

28243

8885 [R_int=0.0496]

8885/0/567

1.246

0.22×0.19×0.12

1.897–25.488

20143

4040 [R_int=0.1419]

4040/193/318

1.278

0.21×0.16×0.13

1.898–25.050

38770

4173 [R_int=0.0569]

4173/1/347

1.243

0.24×0.21×0.12

1.854–25.097

54774

4267 [R_int=0.0875]

4267/0/336

1.181

0.22×0.16×0.13

1.795–23.999

22519

2933 [R_int=0.1406]

2933/264/235

1.238

°

0.0871

0.0939

0.0729

0.0614

0.1692

wR₂(all data)

0.2056

0.901/À 0.810

0.1657

1.090/À 0.698

0.1521

0.905/À 0.752

0.1426

1.346/À 0.936

0.4021

2.986/À 2.257

Largest diff. peak and hole, e Å^{À 3}

UV/Vis measurements.^[16a,54]Deionized water and octan-1-ol

were vigorously stirred for 24 h, to enable saturation of both

phases, then separated by centrifugation. A stock solution of

the selected Fe compound (ca. 2 mg) was prepared by first

adding acetone (50 μL, to help solubilization), followed by

water-saturated octanol (2.5 mL). The solution was diluted with

water-saturated octanol (ca. 1:3 v/v ratio, c_Fe2�10^{À 4}M, so that

1.5�A�2.0 at λ_max) and its UV/Vis spectrum was recorded (

A⁰_org). An aliquot of the solution (V_org=1.2 mL) was transferred

into a test tube and octanol-saturated water (V_org=V_aq=1.2 mL)

was added. The mixture was vigorously stirred for 30 min at

Carbon monoxide release. A stock solution was prepared by

dissolving the selected Fe compound (ca. 10 mg) in MeOH

(0.50 mL) then diluting with H₂O up to 10.0 mL total volume

(5% v/v MeOH). Next, 3.20 mL of the solution (n_Feca.

6·10^{À 3}mmol) and a 7x2 mm stir bar were transferred into a 4-

mL screw top vial (1.72 mL neat gas phase volume). The mixture

was sealed with a screw cap with a PTFE/silicone septum and

°

heated at 37 C for 24 h under stirring. Afterwards, the head-

space was sampled with a gas tight microsyringe (250 mL) and

analysed by GC-TCD. Therefore, the vial was vented, sealed and

heated for further 24 h, and GC-TCD analysis was repeated. The

amount of released CO (ν_CO, mmol) was calculated based on a

calibration curve obtained from analyses of known air/CO

mixtures (1–10% v/v), and the number of equivalents released

over a 24 h period refers to the initial amount of compound

(eq_CO=n_CO/n_Fe). The residual amount of starting diiron complex

after 48 h was calculated by assuming the release of 3

equivalents of CO per mole and the total amount of CO

released. The procedure was repeated three times for each

sample (from the same stock solution); results are given as

mean � standard deviation (Table S4).

°

21 C then centrifuged (RCF 805, 5 min; Hettich EBA 8G). The

UV/Vis spectrum of the organic phase was recorded (A^f_org) and

the partition coefficient was calculated as P_ow=A^f_org/(A_o⁰_rggÀ A_o^f_rg

)

where A⁰_organd A_o^f_rgare the absorbance in the organic phase

before and after partition with the aqueous phase,

respectively.^[53]For [2B]BF₄, [2C]Cl and [3F]CF₃SO₃an inverse

procedure was followed, starting from a solution of the

compound in octanol-saturated water. The partition coefficient

was calculated as P_ow=(A⁰_aqÀ A^f_aq)/A^f_aqwhere A⁰_aqand A_a^f_qare the

absorbance in the aqueous phase before and after partition

with the organic phase, respectively. The wavelength of the

maximum absorption of each compound (320–390 nm range)

was used for UV/Vis quantitation. The procedure was repeated

three times for each sample (from the same stock solution);

results are given as mean � standard deviation (Table 3).

Naphthoquinone was used as a reference compound (logP=

1.8�0.2; literature:^[55]1.71).

Reactivity with model protein.

Lysozyme from chicken egg white (M=14300 Da, 62971 Sigma-

°

Aldrich) was stored at 4 C as received. A stock solution was

prepared by dissolving the powder protein (125 μM) and

Stability in water (D₂O) and cell culture medium. The

procedures are described in the Supporting Information, and

data are collected in Table S3.

NH₄OAc (1.25×10^{À 2}M)^[56]in HPLC water and was immediately

used. Stock solutions of diiron complexes (c_Fe=1×10^{À 3}M)

2

were prepared by dissolving the selected compound in DMSO

(0.25 mL) or MeOH (0.7 mL; used only for [4G]BF₄), then diluting

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

14

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

with HPLC water up to 5.0 mL total volume. Next, aliquots of

the Fe (0.50 mL) and lysozyme (2.0 mL) stock solutions were

mixed and the final solution (Fe₂200 μM, lysozyme 100 μM,

treated 96 well-plate (Greiner Bio-one, Kremsmünster, Austria)

in phenol red free RPMI-1640 medium (Sigma Chemical Co.),

containing 10% FCS and supplemented with 20% methyl

cellulose stock solution.

°

NH₄OAc 10 mM; 1% DMSO) was kept at 37 C for 24 h. The

resulting suspensions were centrifuged (RCF 805, 5 min; Hettich

EBA 8G) to separate the orange-brown precipitate. A blank

solution (lysozyme only) was prepared and treated by the same

procedure. Next, samples were diluted 1:20 (v/v) with a water/

MeCN (1:1, v/v) solution containing 1% formic acid and

analysed by HPLC-MS and flow injection MS. For HPLC analyses,

elution was conducted with a MeCN/water linear gradient

(90:10 to 5:95 v/v over 60 min), containing 0.05% trifluoro-

acetic acid. The following compounds were identified by their

MS pattern: [2A]⁺(t_R=18.0 min), [2B]⁺(t_R=13.8 min), [2C]⁺

(t_R=18.8 min), [2D]⁺(t_R=17.0 min), [2E]⁺(t_R=18.3 min), [2F]⁺

(t_R=9.7 min), [3F]⁺(t_R=13.1 min), [4G]⁺(t_R=23.0 min), [5H]⁺

(t_R=24.3 min), lysozyme-CH₃(t_R=21.0 min), lysozyme (t_R=

21.6 min). In all cases, the starting organometallic cation was

identified as the major Fe-containing compound in solution.

Protein peaks pattern were obtained following peak reconstruc-

tion. Peak integrals for lysozyme and methyl lysozyme were

calculated from the extracted ion chromatograms from the flow

injection MS analysis (Table 4). MS patterns are displayed in

Figures S33–S41.

MTT assay. The growth inhibitory effect towards tumour

cells was evaluated by means of MTT assay as previously

described.^[57]Briefly, 3–8×10³cells/well, dependent upon the

growth characteristics of the cell line, were seeded in 96-well

microplates in growth medium (100 μL). After 24 h, the medium

was removed and replaced with a fresh one containing the

compound to be studied at the appropriate concentration.

Triplicate cultures were established for each treatment. After 24

or 72 h, each well was treated with 10 μL of a 5 mg/mL MTT

saline solution, and following 5 h of incubation, 100 μL of a

sodium dodecyl sulfate (SDS) solution in HCl 0.01 M were

added. After an overnight incubation, cell growth inhibition

was detected by measuring the absorbance of each well at

570 nm using a Bio-Rad 680 microplate reader. Mean absorb-

ance for each drug dose was expressed as a percentage of the

control untreated well absorbance and plotted against drug

concentration. IC₅₀values, the drug concentrations that reduce

the mean absorbance at 570 nm to 50% of those in the

untreated control wells, were calculated by the four-parameter

logistic (4-PL) model. Evaluation was based on means from at

least four independent experiments.

Acid phosphatase (APH) assay. An APH modified assay was

used for determining cell viability in 3D spheroids, as previously

described.^[58]IC₅₀values were calculated with a four-parameter

logistic (4-PL) model. All the values are the means � SD of not

less than four independent experiments.

Mitochondrial membrane potential (ΔΨ). The ΔΨ was

assayed using the Mito-ID® Membrane Potential Kit according

to the manufacturer’s instructions (Enzo Life Sciences, Farm-

ingdale, NY) as previously described.^[59]Briefly, PSN-1 cells (8×

10³per well) were seeded in 96-well plates; after 24 h, cells

were washed with PBS and loaded with Mito-ID detection

Biological studies

Compounds [2B]BF₄, [2C]CF₃SO₃, [2D]CF₃SO₃and [3F]CF₃SO₃

were dissolved in water while the other compounds were

dissolved in the minimum DMSO amount prior to cell culture

testing. A calculated amount of the stock drug DMSO solution

was added to the cell culture media to reach a final maximum

DMSO concentration of 0.5%, which had no effects on cell

viability.

Cisplatin was dissolved in 0.9% sodium chloride solution.

°

MTT

[3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium

reagent for 30 min at 37 C in the dark. Afterwards, cells were

bromide], cisplatin and ImmunoPure p-nitrophenyl phosphate

(APH) were obtained from Sigma-Aldrich.

incubated with increasing concentrations of tested complexes.

Fluorescence intensity was estimated using a VICTOR X3

(PerkinElmer, USA) plate reader at λ_ex=490 and λ_em=590 nm.

Antimycin (3 μM) was used as positive control.

Cell cultures. Human colon (HCT-15), pancreatic (PSN-1) and

breast (MCF-7) carcinoma cell lines along with human melano-

ma cells (A375) were obtained from American Type Culture

Collection (ATCC, Rockville, MD). Embryonic kidney HEK293 cells

were obtained from the European Collection of Cell Cultures

(ECACC, Salisbury, UK). Human ovarian 2008 cancer cells were

kindly provided by Prof. G. Marverti (Dept. of Biomedical

Science of Modena University, Italy). MCF-7 ADR cells were

kindly provided by Prof. N. Colabufo (Dept. Pharmacy, Univer-

sity of Bari, Italy). Cell lines were maintained in the logarithmic

ROS production. The production of ROS was measured in

PSN-1 cells (10⁴per well) grown for 24 h in a 96-well plate in

RPMI medium without phenol red (Merck). Cells were then

washed with PBS and loaded with 10 μM 5-(and-6)-chlorometh-

yl-2’,7’-dichlorodihydrofluorescein diacetate acetyl ester (CM-

H₂DCFDA; Molecular Probes-Invitrogen) for 25 min, in the dark.

Afterwards, cells were washed with PBS and incubated with

increasing concentrations of tested compounds. Fluorescence

increase was estimated by using λ_ex=485 nm and λ_em=527 nm

in a VICTOR X3 (PerkinElmer) plate reader. Antimycin (3 μM,

Merck), a potent inhibitor of Complex III in the electron

transport chain, was used as positive control.

°

phase at 37 C in a 5% CO₂atmosphere using the following

culture media containing 10% fetal calf serum (Euroclone,

Milan, Italy), antibiotics (50 units/mL penicillin and 50 μg/mL

streptomycin) and 2 mM l-glutamine: i) RPMI-1640 medium

(Euroclone, Milan, Italy) for PSN-1, 2008, HCT-15, MCF-7 and

MCF-7 ADR cells; ii) DMEM for A375 and HEK293 cells.

Intracellular CO release. PSN-1 cell (1×10⁴) were seeded in

96-well plates. After 24 h, cells were treated with 20 μM of

°

Spheroid cultures. Spheroid cultures were obtained by

tested complexes or CORM-401 (Merck) for 15 min at 37 C and

followed by 30 min incubation with a fluorescent probe

seeding 2.5×10³cells/well in round bottom non-tissue culture

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

15

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

BioTracker Carbon Monoxide Probe 1 Live Cell Dye® (Merck).

The intracellular fluorescence, indicative of CO accumulation in

cells, was quantified using a VICTOR X3 (PerkinElmer) plate

reader.

Keywords: bioinorganic chemistry · CO release; cytotoxicity ·

diiron complexes · 3D cancer cell models

[1] a) E. Boros, P. J. Dyson, G. Gasser, Chem. 2020 , 6 , 41–60; b) K. L. Haas,

K. J. Franz, Chem. Rev. 2009 , 109 , 4921–4960; c) E. J. Anthony, E. M.

Bolitho, H. E. Bridgewater, O. W. L. Carter, J. M. Donnelly, C. Imberti, E. C.

Lant, F. Lermyte, R. J. Needham, M. Palau, P. J. Sadler, H. Shi, F.-X. Wang,

W.-Y. Zhang, Z. Zhang, Chem. Sci. 2020 , 11 , 12888–12917.

[2] a) M. G. Apps, E. H. Y. Choi, N. J. Wheate, Endocrine-Related Cancer 2015,

22, R219–R233; b) D. Gibson, Dalton Trans. 2009 , 10681–10689.

[3] a) R. Oun, Y. E. Moussa, N. J. Wheate, Dalton Trans. 2018 , 47 , 6645–6653;

b) Z. H. Siddik, Oncogene 2003 , 22 , 7265–7279; c) S. Dasari, P. B.

Tchounwou, Eur. J. Pharmacol. 2014 , 740 , 364–378.

[4] Selected recent references: a) B. S. Murray, P. J. Dyson, Curr. Opinion

Chem. Biol. 2020 , 56 , 28–34; b) P. Štarha, Z. Trávníček, Coord. Chem. Rev.

2019, 395, 130–145; c) I. Bratsos, T. Gianferrara, E. Alessio, C. G.

Hartinger, M. A. Jakupec, B. K. Keppler, Ruthenium and Other Non-

Platinum Anticancer Compounds, in Bioinorganic Medicinal Chemistry, Ed.

E. Alessio, Wiley-VCH, Weinheim, 2011, 151–174; d) P. Zhang, P. J.

Sadler, J. Organomet. Chem. 2017 , 839 , 5–14.

[5] a) W. Kaim, B. Schwederski, A. Klein, Bioinorganic Chemistry: Inorganic

Elements in the Chemistry of Life, 2nd ed., Wiley 2013; b) M. A. Zoroddu,

J. Aaseth, G. Crisponi, S. Medici, M. Peana, V. M. Nurchi, J. Inorg. Biochem.

2019 , 195 , 120–129.

[6] U. Basu, M. Roy, A. R. Chakravarty, Coord. Chem. Rev. 2020 , 417 , 213339.

[7] a) A. A. Simenel, E. A. Morozova, L. V. Snegur, S. I. Zykova, V. V. Kachala,

L. A. Ostrovskaya, N. V. Bluchterova, M. M. Fomina, Appl. Organomet.

Chem. 2009 , 23 , 219–224; b) L. V. Snegur, Y. S. Nekrasov, N. S. Sergeeva,

Z. V. Zhilina, V. V. Gumenyuk, Z. A. Starikova, A. A. Simenel, N. B.

Morozova, I. K. Sviridova, V. Babin, Appl. Organomet. Chem. 2008 , 22 ,

139–147.

[8] a) R. Wang, H. Chen, W. Yan, M. Zheng, T. Zhang, Y. Zhang, Eur. J. Med.

Chem. 2020 , 190 , 112109; b) M. Patra, G. Gasser, Nat. Chem. Rev. 2017, 1,

0066; c) G. Jaouen, A. Vessières, S. Top, Chem. Soc. Rev. 2015 , 44 , 8802–

8817; d) S. Peter, B. A. Aderibigbe, Molecules 2019 , 24 , 3604.

Cellular uptake. PSN-1 and HEK293 cells (3×10⁶) were

seeded in 75 cm²flasks in growth medium (20 mL). After

overnight incubation, the medium was replaced and the cells

were treated with tested compounds for 24 h. Cell monolayers

were washed twice with cold PBS, harvested and counted.

Samples were than subjected to three freezing/thawing cycles

°

at À 80 C, and then vigorously vortexed. The samples were

treated with highly pure nitric acid (Fe:�0.01 μg kg^{À 1}, Trace-

SELECT ® Ultra, Sigma) and transferred into a microwave Teflon

vessel. Subsequently, samples were submitted to standard

procedures using a speed wave MWS-3 Berghof instrument

(Eningen, Germany). After cooling, each mineralized sample was

analysed for iron using a Varian AA Duo graphite furnace

atomic absorption spectrometer (Varian) at the wavelength of

248.3 nm. The calibration curve was obtained using known

concentrations of standard solutions purchased from Sigma

Chemical Co.

Statistical analysis. All values are the means � SD of no less

than three measurements starting from three different cell

cultures. Multiple comparisons were made by ANOVA followed

by the Tukey-Kramer multiple comparison test (*p<0.05, **p<

0.01), using GraphPad software.

[9] a) Y. Wang, P. M. Dansette, P. Pigeon, S. Top, M. J. McGlinchey, D.

Mansuy, G. Jaouen, Chem. Sci. 2018 , 9 , 70–78; b) Y. Wang, M.-A. Richard,

S. Top, P. M. Dansette, P. Pigeon, A. Vessières, D. Mansuy, G. Jaouen,

Angew. Chem. Int. Ed. 2016 , 55 , 10431–10434; Angew. Chem. 2016 , 128 ,

10587–10590.

Supporting Information

IR and NMR spectra of products and comparison of selected

spectroscopic data; DTF optimized structures; stability studies in

water and cell culture media; carbon monoxide release studies;

IC₅₀values after 24 h of incubation; MS spectra from lysozyme

interaction experiments. Deposition numbers 2045779 (for [2A ]

CF₃SO₃), 2045778 (for [2C ]CF₃SO₃), 2045777 (for [2C ]Cl), 2045780

(for [2D ]CF₃SO₃) and 2045781 (for 6 ) contain the supplementary

crystallographic data for this paper. These data are provided

free of charge by the joint Cambridge Crystallographic Data

Centre and Fachinformationszentrum Karlsruhe Access Struc-

tures service.

[10] a) D. BenYosef, D. Romano, M. Hadiji, P. J. Dyson, B. Blom, Inorg. Chim.

Acta 2020 , 510 , 119731; b) H. T. Poh, P. C. Ho, W. Y. Fan, RSC Adv. 2016 ,

6 , 18814–18823; c) A. Pilon, P. Gírio, G. Nogueira, F. Avecilla, H. Adams, J.

Lorenzo, M. H. Garcia, A. Valente, J. Organomet. Chem. 2017 , 852 , 34–42.

[11] a) P. R. Florindo, D. M. Pereira, P. M. Borralho, C. M. P. Rodrigues, M. F. M.

Piedade, A. C. Fernandes, J. Med. Chem. 2015 , 58 , 4339–4347; b) A.

Valente, A. M. Santos, L. Côrte-Real, M. P. Robalo, V. Moreno, M. Font-

Bardia, T. Calvet, J. Lorenzo, M. H. Garcia, J. Organomet. Chem. 2014 ,

756 , 52–60; c) A. Pilon, A. R. Brás, L. Côrte-Real, F. Avecilla, P. J. Costa, A.

Preto, M. H. Garcia, A. Valente, Molecules 2020 , 25 , 1592; d) A. C.

Gonçalves, T. S. Morais, M. P. Robalo, F. Marques, F. Avecilla, C. P. Matos,

I. Santos, A. I. Tomaz, M. H. Garcia, J. Inorg. Biochem. 2013 , 129 , 1–8.

[12] X. Jiang, L. Chen, X. Wang, L. Long, Z. Xiao, X. Liu, Chem. Eur. J. 2015 , 21 ,

13065–13072.

[13] A. Vessières, Y. Wang, M. J. McGlinchey, G. Jaouen, Coord. Chem. Rev.

2021 , 430 , 213658.

[14] a) O. Buriez, J. M. Heldt, E. Labb, A. Vessières, G. Jaouen, C. Amatore,

Chem. Eur. J. 2008 , 14 , 8195–8203; b) A. Notaro, G. Gasser, A.

Castonguay, ChemMedChem 2020 , 15 , 345–348.

[15] Selected references: a) R. Mazzoni, M. Salmi, V. Zanotti, Chem. Eur. J.

2012 , 18 , 10174–10194, and references therein; b) F. Marchetti, Eur. J.

Inorg. Chem. 2018 , 3987–4003, and references therein; c) M. E. García, D.

García-Vivó, A. Ramos, M. A. Ruiz, Coord. Chem. Rev. 2017 , 330 , 1–36;

d) Y. Chen, L. Liu, Y. Peng, P. Chen, Y. Luo, J. Qu, J. Am. Chem. Soc. 2011 ,

133 , 1147–1149; e) J. He, C.-L. Deng, Y. Li, Y.-L. Li, Y. Wu, L.-K. Zou, C. Mu,

Q. Luo, B. Xie, J. Wei, Organometallics 2017 , 36 , 1322–1330.

Acknowledgements

We gratefully thank the Universities of Pisa (Fondi di Ateneo

2020 and PRA_2020_39) and Padova (DOR 2019) for financial

support, Dr. Beatrice Campanella (ICCOM – CNR, Pisa, Italy) for

Raman analysis and Dr. Andrea Raffaelli (ICF – CNR, Pisa, Italy)

for MS and HPLC-MS analyses.

[16] a) G. Agonigi, L. Biancalana, M. G. Lupo, M. Montopoli, N. Ferri, S.

Zacchini, F. Binacchi, T. Biver, B. Campanella, G. Pampaloni, V. Zanotti, F.

Marchetti, Organometallics 2020 , 39 , 645–657; b) S. Schoch, L. K.

Batchelor, T. Funaioli, G. Ciancaleoni, S. Zacchini, S. Braccini, F. Chiellini,

T. Biver, G. Pampaloni, P. J. Dyson, F. Marchetti, Organometallics 2020 ,

39 , 361–373.

Conflict of Interest

The authors declare no conflict of interest.

[17] a) D. Rocco, L. K. Batchelor, G. Agonigi, S. Braccini, F. Chiellini, S. Schoch,

T. Biver, T. Funaioli, S. Zacchini, L. Biancalana, M. Ruggeri, G. Pampaloni,

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

16

��

These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202101048

Chemistry—A European Journal

P. J. Dyson, F. Marchetti, Chem. Eur. J. 2019 , 25 , 14801–14816; b) G.

Agonigi, L. K. Batchelor, E. Ferretti, S. Schoch, M. Bortoluzzi, S. Braccini,

F. Chiellini, L. Biancalana, S. Zacchini, G. Pampaloni, B. Sarkar, P. J.

Dyson, F. Marchetti, Molecules 2020 , 25 , 1656.

2019 , 8 , 6; c) H. Carlsson, H. von Stedingk, U. Nilsson, M. Törnqvist,

Chem. Res. Toxicol. 2014 , 27 , 2062–2070.

[38] D. V. LaBarbera, B. G. Reid, B. H. Yoo, Expert Opin. Drug Discovery 2012 ,

7 , 819–830.

[18] R. Maddaly, A. Subramaniyan, H. Balasubramanian, J. Cell. Biochem.

2017 , 118 , 2544–2558.

[19] a) R. Edmondson, J. Jenkins Broglie, A. F. Adcock, L. Yang, Assay Drug

Dev. Technol. 2014 , 12 , 207–218; b) D. Loessner, K. S. Stok, M. P. Lutolf,

D. W. Hutmacher, J. A. Clements, S. C. Rizzi, Biomaterials 2010 , 31 , 8494–

8506.

[39] D. Rocco, N. Busto, C. Pérez-Arnaiz, L. Biancalana, S. Zacchini, G.

Pampaloni, B. Garcia, F. Marchetti, Appl. Organomet. Chem. 2020, 34,

e5923.

[40] K. Jomova, S. Baros, M. Valko, Trans. Met. Chem. 2012 , 37 , 127–134.

[41] R. Mazzoni, A. Gabiccini, C. Cesari, V. Zanotti, I. Gualandi, D. Tonelli,

Organometallics 2015 , 34 , 3228–3235.

[20] L. Biancalana, G. Ciancaleoni, S. Zacchini, G. Pampaloni, F. Marchetti,

[42] V. Scalcon, M. Salmain, A. Folda, S. Top, P. Pigeon, L. H. Z. Shirley, G.

Jaouen, A. Bindoli, A. Vessières, M. P. Rigobello, Metallomics 2017 , 9 ,

949–959.

[43] a) E. S. Arnér, M. Björnstedt, A. Holmgren, J. Biol. Chem. 1995 , 270 , 3479–

3482; b) J. Nordberg, L. Zhong, A. Holmgren, E. S. Arnér, J. Biol. Chem.

1998 , 273 , 10835–10842.

Inorg. Chim. Acta 2020, 517, 120181.

[21] G. Agonigi, M. Bortoluzzi, F. Marchetti, G. Pampaloni, S. Zacchini, V.

Zanotti, Eur. J. Inorg. Chem. 2018 , 960–971, and references therein.

[22] L. Busetto, F. Marchetti, S. Zacchini, V. Zanotti, Organometallics 2006 , 25 ,

4808–4816.

[23] S. Willis, A. R. Manning, F. S. Stephens, J. Chem. Soc. Dalton Trans. 1980 ,

[44] S. Keller, Y. C. Ong, Y. Lin, K. Cariou, G. Gasser, J. Organomet. Chem.

2020 , 906 , 121059.

186–191.

[24] Y. Han, W. Dong, Q. Guo, X. Li, L. Huang, Eur. J. Med. Chem. 2020 , 203 ,

112506.

[25] a) P. Finkbeiner, J. P. Hehn, C. Gnamm, J. Med. Chem. 2020 , 63 , 7081–

7107; b) J. B. Rodriguez, C. Gallo-Rodriguez, ChemMedChem 2019, 14,

190–216.

[45] G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb, A. Nudelman,

B. M. Stoltz, J. E. Bercaw, K. I. Goldberg, Organometallics 2010 , 29 , 2176–

2179.

[46] R. K. Harris, E. D. Becker, S. M. Cabral De Menezes, R. Goodfellow, P.

Granger, Pure Appl. Chem. 2001, 73, 1795–1818.

[26] A. J. Bridgeman, G. Cavigliasso, L. R. Ireland, J. Rothery, J. Chem. Soc.

Dalton Trans. 2001 , 2095–2108.

[27] M. Ennis, R. Kumar, A. R. Manning, J. A. S. Howell, P. Mathur, A. J. Rowan,

F. S. Stephens, J. Chem. Soc. Dalton Trans. 1981 , 1251–1259.

[28] V. G. Albano, L. Busetto, F. Marchetti, M. Monari, S. Zacchini, V. Zanotti,

Z. Naturforsch. 2007, 62b, 427–438.

[47] W. Willker, D. Leibfritz, R. Kerssebaum, W. Bermel, Magn. Reson. Chem.

1993 , 31 , 287–292.

[48] F. Menges, “Spectragryph—optical spectroscopy software”, version

1.2.9, @ 2018, http://www.effemm2.de/spectragryph.

[49] G. M. Sheldrick, Acta Crystallogr. Sect. C 2015 , 71 , 3–8.

[50] F. Neese, Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2017, 8, e1327.

[51] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010 , 132 ,

154104.

[29] G. Bistoni, S. Rampino, N. Scafuri, G. Ciancaleoni, D. Zuccaccia, L.

Belpassi, F. Tarantelli, Chem. Sci. 2016 , 7 , 1174–1184.

[30] a) M. Knorr, I. Jourdain, A. S. Mohamed, A. Khatyr, S. G. Koller, C.

Strohmann, J. Organomet. Chem. 2015 , 780 , 70–85; b) M. A. Alvarez,

M. E. García, D. García-Vivo, M. A. Ruiz, M. F. Vega, Organometallics 2013 ,

32 , 4543–4555.

[31] In both the X-ray structures, the anion strongly interacts with the CÀ H

of the Cp ligands (shortest X.HÀ C distances 2.797 and 2.165 Å for [2C]Cl

and [2C]CF₃SO₃, respectively). Assuming that in low-polar solvents the

ion pair structures are preserved, it appears evident that the anion can

affect the Cp chemical shifts.

[52] T. Rundlöf, M. Mathiasson, S. Bekiroglu, B. Hakkarainen, T. Bowden, T.

Arvidsson, J. Pharm. Biomed. Anal. 2010, 52, 645–651.

[53] N. M. Rice, H. M. N. H. Irving, M. A. Leonard, Pure Appl. Chem. 1993, 65,

2373–2396.

[54] a) OECD Guidelines for Testing of Chemicals, Vol. 107, OECD, Paris, 1995;

b) J. C. Dearden, G. M. Bresnen, Quant. Struct.-Act. Relat. 1988 , 7 , 133–

144.

[55] D. J. Currie, C. E. Lough, R. F. Silver, H. L. Holmes, Can. J. Chem. 1966 , 44 ,

1035–1043.

[32] Related aminocarbyne complexes behave as 1:1 electrolytes in nitro-

methane (μ=3.46 D): I. S. Willis, A. R. Manning, F. S. Stephens, J. Chem.

Soc. Dalton Trans. 1980 , 186–191.

[33] Stability experiments performed in D₂O/[D₆]DMSO mixtures differ in

that formation of [Fe₂Cp₂(CO)(μ-CO){μ-CNMeR}(DMSO)]⁺is observed,

without any precipitate (Scheme S2).

[34] a) A. Leonidova, P. Anstaett, V. Pierroz, C. Mari, B. Spingler, S. Ferrari, G.

Gasser, Inorg. Chem. 2015 , 54 , 9740–9748; b) P. Marzenell, H. Hagen, L.

Sellner, T. Zenz, R. Grinyte, V. Pavlov, S. Daum, A. Mokhir, J. Med. Chem.

2013 , 56 , 6935–6944.

[56] L. Konermann, J. Am. Soc. Mass Spectrom. 2017 , 28 , 1827–1835.

[57] M. Pellei, V. Gandin, L. Marchiò, C. Marzano, L. Bagnarelli, C. Santini,

Molecules 2019 , 24 , 1761.

[58] V. Gandin, C. Ceresa, G. Esposito, S. Indraccolo, M. Porchia, F. Tisato, C.

Santini, M. Pellei, C. Marzano, Sci. Rep. 2017, 7, 13936.

[59] K. Dammak, M. Porchia, M. De Franco, M. Zancato, H. Naïli, V. Gandin, C.

Marzano, Molecules 2020 , 25 , 5484.

[35] X. Jiang, Z. Xiao, W. Zhong, X. Liu, Coord. Chem. Rev. 2021 , 429 , 213634.

[36] L. Messori, A. Merlino, Chem. Commun. 2017 , 53 , 11622–11633.

[37] a) W. Ki Paik, D. C. Paik, S. Kim, Trends Biochem. Sci. 2007 , 32 , 146–152;

b) H. Carlsson, S. M. Rappaport, M. Törnqvist, High-Throughput Screening

Manuscript received: March 23, 2021

Version of record online: ■■■, ■■■■

Chem. Eur. J. 2021, 27, 1–18

www.chemeurj.org

17

��

These are not the final page numbers!

FULL PAPER

A class of cationic diiron complexes

containing a variable bridging

Dr. L. Biancalana, Dr. M. De Franco,

Prof. G. Ciancaleoni, Prof. S. Zacchini,

Prof. G. Pampaloni, Prof. V. Gandin*,

Prof. F. Marchetti*

iminium ligand displays promising

antiproliferative activity towards a

panel of 2D and 3D cancer cell lines,

and a selectivity not correlated to a

different degree of uptake in

cancerous and noncancerous cells.

The mode of action appears to be

mainly related to the interference

with cell redox processes (ROS pro-

duction, inhibition of TrxR), initiated

by intracellular CO release.

1 – 18

Easily Available, Amphiphilic Diiron

Cyclopentadienyl Complexes Exhibit

in Vitro Anticancer Activity in 2D

and 3D Human Cancer Cells through

Redox Modulation Triggered by CO

Release

Article Doi

Easily Available, Amphiphilic Diiron Cyclopentadienyl Complexes Exhibit in Vitro Anticancer Activity in 2D and 3D Human Cancer Cells through Redox Modulation Triggered by CO Release

DOI: 10.1002/chem.202101048

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.202101048

Full text of DOI:10.1002/chem.202101048

Products guided by the article

R&D Labs maybe for 69346-73-2

Relevant to this article

Hot Product