Detection of ferric ions in a gram-positive bacterial cell: Staphylococcus aureus

Article abstract of DOI:10.1080/00958972.2020.1868042

A rhodamine-based chemosensor was synthesized and found to selectively bind ferric ions over other metal ions (Na⁺, K⁺, Ca²⁺, Mg²⁺, Fe²⁺, Zn²⁺, Cd²⁺, Co²⁺, Hg²⁺ Cr³⁺, Al³⁺) in an organic-aqueous mixture (CH₃CN-MES). Upon addition of ferric ions, the spirolactam ring opens, producing a visual color change and a fluorescence intensity increase, i.e. a “turn on” optical response at 577 nm is observed. The chemosensor coordinates to ferric ions in 1:1 stoichiometry with a calculated K _a = 3.5 × 10⁴ mol?dm^?3 by fluorescence spectroscopy and a LoD of 27 ppb. The chemosensor was reversible upon addition of the Fe³⁺ chelator desferrioxamine. One- and two-dimensional NMR experiments with Al³⁺ ions aided in understanding of the coordination environment of the ferric ion with the chemosensor, which were confirmed by molecular modeling calculations. X-ray quality crystals of the chemosensor were obtained, and the solid-state structure is reported. Confocal microscopy was used to detect free ferric ions in Staphylococcus aureus.

Full text of DOI:10.1080/00958972.2020.1868042

Journal of Coordination Chemistry
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/gcoo20
Detection of ferric ions in a gram-positive bacterial
cell: Staphylococcus aureus
Erendra Manandhar, Ashley D. G. Johnson, William M. Watson, Shelby D.
Dickerson, Gyan S. Sahukhal, Mohamed O. Elasri, Frank R. Fronczek, Peter J.
Cragg & Karl J. Wallace
To cite this article: Erendra Manandhar, Ashley D. G. Johnson, William M. Watson, Shelby D.
Dickerson, Gyan S. Sahukhal, Mohamed O. Elasri, Frank R. Fronczek, Peter J. Cragg & Karl J.
Wallace (2021) Detection of ferric ions in a gram-positive bacterial cell: Staphylococcusꢀaureus,
Journal of Coordination Chemistry, 74:1-3, 380-401, DOI: 10.1080/00958972.2020.1868042
To link to this article: https://doi.org/10.1080/00958972.2020.1868042
View supplementary material
Published online: 30 Dec 2020.
Submit your article to this journal
Article views: 49
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=gcoo20

JOURNAL OF COORDINATION CHEMISTRY
2021, VOL. 74, NOS. 1–3, 380–401
https://doi.org/10.1080/00958972.2020.1868042
Detection of ferric ions in a gram-positive bacterial cell:
Staphylococcus aureus
Erendra Manandhara, Ashley D. G. Johnsona, William M. Watsona, Shelby D.
Dickersona, Gyan S. Sahukhalb, Mohamed O. Elasrib, Frank R. Fronczekc,
Peter J. Craggd
and Karl J. Wallacea
aDepartment of Chemistry and Biochemistry, University of Southern Mississippi, Hattiesburg, MS,
b
USA; Department of Biological Sciences, University of Southern Mississippi, Hattiesburg, MS, USA;
d
cDepartment of Chemistry, Louisiana State University, Baton Rouge, LA, USA; School of Pharmacy
and Biomolecular Sciences, University of Brighton, Brighton, UK
ABSTRACT
ARTICLE HISTORY
Received 1 October 2020
Accepted 7 December 2020
A rhodamine-based chemosensor was synthesized and found to
selectively bind ferric ions over other metal ions (Naþ, Kþ, Ca2þ
,
Mg2þ, Fe2þ, Zn2þ, Cd2þ, Co2þ, Hg2þ Cr3þ, Al3þ) in an organic-aque-
ous mixture (CH3CN-MES). Upon addition of ferric ions, the spirolac-
tam ring opens, producing a visual color change and a fluorescence
intensity increase, i.e. a “turn on” optical response at 577 nm is
observed. The chemosensor coordinates to ferric ions in 1:1 stoichi-
ometry with a calculated Ka ¼ 3.5ꢀ 104 molꢁdmꢂ3 by fluorescence
spectroscopy and a LoD of 27 ppb. The chemosensor was reversible
upon addition of the Fe3þ chelator desferrioxamine. One- and two-
dimensional NMR experiments with Al3þ ions aided in
understanding of the coordination environment of the ferric ion
with the chemosensor, which were confirmed by molecular model-
ing calculations. X-ray quality crystals of the chemosensor were
obtained, and the solid-state structure is reported. Confocal micros-
copy was used to detect free ferric ions in Staphylococcus aureus.
KEYWORDS
Ferric ions; sensing; gram-
positive bacteria;
coordination; molecular
probes
CONTACT Karl J. Wallace
karl.wallace@usm.edu
Department of Chemistry and Biochemistry, University of
Southern Mississippi, Hattiesburg, MS 39406, USA
Dedication: In honor of Jerry Atwood’s 25 years’ service to the Journal of Coordination Chemistry and his 51 years’
service to the Chemistry World
Supplemental data for this article is available online at https://doi.org/10.1080/00958972.2020.1868042.
ß 2020 Informa UK Limited, trading as Taylor & Francis Group

JOURNAL OF COORDINATION CHEMISTRY
381
1. Introduction
The synthesis of chemosensors to selectively detect biologically and environmentally
essential ions is an area of extensive interest [1–3]. Iron is one of the most abundant
metal ions in the human body, with 3–6 g present in the average adult living in the
United States [4]. It is found predominantly in the þ2 and þ3 oxidation states and
plays significant biological roles in the cell [5] where it is present at approximately
50–100 lM (3 to 6 ppm) [3, 6]. The ferrous ion is utilized in oxygen metabolism, elec-
tron transfer, and DNA and RNA synthesis [7]. The ferric ion is less abundant but is
nevertheless a critical metal ion, typically found in metallobiomolecules [8]. Also, a
fraction of iron is “loosely” bound to organic anions (phosphates, citrates, carbonates,
and carboxylates), polyfunctional ligands (polypeptides and siderophores), and surface
components of membranes (phospholipid head groups) [9–11]. This can lead to labile
iron pools (LIP’s) of free iron [7]. Moreover, iron can readily undergo redox reactions
with molecular oxygen forming both Fe2þ and Fe3þ in these LIPs [10, 12]. The labile
iron is a source for metabolic reactions that occur within the cell and is a site for gen-
eration of highly reactive oxygen species such as hydroxyl radicals via Fenton chemis-
try [13]. These highly reactive radicals can interact with many biologically important
compounds such as sugars, lipids, proteins, and nucleic acids resulting in peroxidative
tissue damage. It is also suggested the cellular toxicity caused by Fe3þ is potentially linked
to conditions such as Alzheimer’s [14], Huntington’s [15], and Parkinson’s diseases [16, 17].
Conversely, deficiency of Fe3þ leads to anemia, kidney and liver damages, diabetes, and
heart disease [18, 19]. Small-molecule recognition of iron in eukaryotic cells has been
explored by Kumar [20, 21], Bhalla [22, 23], Kim [24, 25], Bernhardt [5, 26], Bruckner [27],
Raymond, [28–30], Hider [31], and Critchon [29, 30], but, despite the role that iron plays in
prokaryotic cells, i.e. bacterial cells, surprisingly few examples exist [32, 33]. The growth of
many bacteria such as Neisseria meningitidis depends on the availability of iron [34]. Iron
influences cell composition, metabolism, enzyme activity, and host cell interactions, includ-
ing pathogenicity. The main functions of iron in the bacteria cell are catalytic [32, 35–37].
Iron often acts as a co-factor for different proteins whereby it can influence other compo-
nents in the bacterial cell [38, 39]. For example, iron deficiency in Mycobacterium smegmatis
decreases DNA and RNA levels [32]. Iron-promoted biofilm formation of Pseudomonas aeru-
ginosa [40] is a significant problem for those with cystic fibrosis while the bacteria Neisseria
meningitidis causes pyogenic meningitis and meningococcal septicemia in humans [34].
Staphylococcus aureus is responsible for both pneumonia and bacteremia and is the lead-
ing causes of skin and soft tissue infections such as abscesses, furuncles, and cellulitis [41].
Numerous small-molecule chemosensors can detect ferric ions in eukaryotic cells
[42–44] but monitoring prokaryotic cells is less well explored, in particular Gram-posi-
tive bacteria. Furthermore, chemosensors specifically targeting ferric ions often lack
selectivity. Many chemosensors also respond to competitive metal species such as
Al3þ, Cu2þ, and Cr3þ ions [45–47]. Literature reports often omit Al3þ ions from their
studies, presumably due to the similarities in size and properties to the Fe3þ ion [42,
48]. Recently, our group has written a review article on chemodosimeters and chemo-
reactants for sensing ferric ions, highlighting the area over the last several years [49],
showcasing the need for chemosensors that selectively detect Fe3þ over Al3þ, Cu2þ
and Cr3þ ions.

382
E. MANANDHAR ET AL.
2. Experimental
2.1. General techniques
One-dimensional (1H, 13C) and two-dimensional (HMBC, HSQC, and ROESY) NMR spec-
tra were recorded on a Bruker Ultrashield plus 400 MHz spectrometer in deuterated
solvents or on a Bruker Advance 600 MHz spectrometer. Chemical shifts are reported
in parts per million (ppm) downfield from tetramethylsilane (0 ppm) as the internal
standard, and coupling constants (J) are recorded in Hertz (Hz). The multiplicities in
1
the H NMR spectra are reported as (br) broad, (s) singlet, (d) doublet, (dd) doublet of
doublets, (ddd) doublet of doublet of doublets, (t) triplet, (sp) septet, and (m) multi-
plet. All spectra are recorded at ambient temperature. UV-Vis experiments were per-
formed on a Beckman DU-70 UV-Vis spectrometer. Low-resolution mass spectra were
measured with Finnigan TSQ70 and VG Analytical ZAB2-E instruments. IR spectra were
recorded on a Nicolet Nexus 470 FT-IR paired with a Smart Orbit ATR attachment.
Characteristic functional groups are reported in wavenumbers (cmꢂ1). Fluorescence
experiments were carried out on a QuantaMasterTM 400 intensity-based spectrofluor-
ometer from PTI technologies in the steady-state; slit-widths 0.5 mm; kex ¼ 550 nm
kem ¼ 555 to 800 nm. Elemental analysis was carried out at Atlantic Microlab Inc.
2.2. General procedure UV-Vis and fluorescence experiments
Stock solutions of all compounds were prepared in CH3CN, from which the desired con-
centration required for the optical studies were then made. A 20 or 50 mmolꢁdmꢂ3 concen-
tration solution was used for UV-Vis spectroscopy studies, and a 5 mmolꢁdmꢂ3 solution
concentration was used for the fluorescence studies. The absorption and emission data are
shown in Supporting information Figure S33. The solution was excited at k ¼ 550 nm and
scanned from 555 to 800 nm with slit widths set to 0.5 mm, for the fluorescence titration a
500 mmolꢁdmꢂ3 of ferric perchlorate or triflate was prepared in CH3CN. Aliquots of ferric
perchlorate were added to 3. Fluorescence spectra were recorded after each addition.
Dilution factors were taken into consideration during binding studies. The binding con-
stants were determined from fluorescence titrations using HypSpec 2006 [50].
2.3. Computational methods
Calculations were undertaken using the Spartan ’18 Parallel Suite running on a Mac
Pro with 3.5 GHz 6-Core Intel Xenon E5 processors and two threads per core.
2.4. X-ray crystallography
Data collection: COLLECT [51]; cell refinement: HKL SCALEPACK [52]; data reduction:
HKL DENZO and SCALEPACK [52]; program(s) used to solve structure: SIR97 [53]; pro-
gram(s) used to refine structure: SHELX2014/7 [54].
Crystal data for 3, M ¼ C35H37N5O3ꢁ0.21H2O; F.W ¼ 579.39, colorless needle fragment,
0.34 ꢀ 0.22 ꢀ 0.20 mm3, monoclinic, space group P21/n (No. 14), a ¼ 9.7604(4),
b ¼ 11.3709(6), c ¼ 26.8708(11) Å, b ¼ 97.613(2)ꢃ, V ¼ 2955.9(2) Å3, Z ¼ 4, Dc ¼

JOURNAL OF COORDINATION CHEMISTRY
383
1.302 gꢁcmꢂ3
,
F000 ¼ 1232, Nonius KappaCCD, MoKa radiation, k ¼ 0.71073 Å,
T ¼ 90.0(5) K, 2hmax ¼ 55.7ꢃ, 24901 reflections collected, 7029 unique (Rint ¼ 0.0514).
Final GooF ¼ 1.021, R1 ¼ 0.0495, wR2 ¼ 0.1015, with R indices based on 4498 reflec-
tions with I I > 2(I) (refinement on F2), 406 parameters, 0 restraints. Lp and absorption
corrections were applied, m ¼ 0.085 mmꢂ1
.
1
2.5. H NMR titrations
A solution of 3 (10.4 mmolꢁdmꢂ3) was prepared by dissolving 3 mg in 0.5 mL CD3CN. A
10-fold more concentrated solution of Al(ClO4)3 was prepared in CD3CN, and 5 mL (5 mL
1
¼ 0.1 equivalents of metal salt) aliquots were added to the compound, and the H
1
NMR spectrum recorded after each addition. Binding studies were undertaken by H
NMR titrations using HypNMR 2016 [55]. The fully labeled numbering system
(Supporting information Figure S5) and full 1D and 2D spectra of 3 and the Al3þ com-
plex (Supporting information Figures S5–S27) are shown in the ESI.
2.6. Protocol for bacterial cell imaging
The stock solution of 3 in DMSO (10 mM) was prepared. The working solution was
made by diluting the stock solution 10 times (200 mL 3 in 1800 mL sterile 1X PBS or
sterile distilled water). The final concentration of organic solvent utilized in cell growth
was 10%.
Bacterial cells were grown in a tryptic soy agar plate from ꢂ80 ꢃC freezer stock culture.
A loopful of bacteria was inoculated into 5 mL of tryptic soy broth (TSB) and incubated
overnight at 37ꢃC with shaking at 200 rpm. The overnight bacterial cell culture was diluted
1:10 in 4mL of pre-warmed TSB and further incubated at 37ꢃC with shaking at 200 rpm
for 3 h. After incubation, 400 lL of FeCl3 (1 mM) was added to the culture yielding a final
iron concentration of 100 lM. Pre-warmed TSB (400 lL) was added to the culture as a
negative control. Both test and control cells were then incubated for 30 min. One mL of
cells was aliquoted into 1.5 mL microfuge tube and centrifuged at 10,000ꢀ g for 5 min to
harvest the cells. Cells were washed twice with sterile distilled water. The supernatants
were removed completely and discarded in each of the washing steps. The cells were
resuspended in 100 lL of 1mmolꢁdmꢂ3 of 3 and 20 lL of bacteria spotted on a clean glass
slide, covered with the lifter slide, and the edges sealed. Confocal microscopy images were
taken in 100 X/0.63 oil immersions. Phase-contrast photos were taken to visualize the cells,
and the same field was laser scanned in cy3/rhodamine settings with emission at 550 nm
and excitation at 580 nm. The same settings were used to scan both control and experi-
mental (iron supplemented condition) for taking images, and the experiments were per-
formed in triplicate to verify the result.
2.7. Synthesis of 2-amino-30,60-bis(diethylamino-2,3-dihydrospiro[isoindole-1, 90-
xanthene]-3-one (1) [56]
Rhodamine B (1.2 g, 2.5 mmol) was dissolved in ethanol (30 mL). Hydrazine hydrate
(3.0 mL, 96 mmol) was then added dropwise with vigorous stirring at room temperature

384
E. MANANDHAR ET AL.
before being refluxed for 16 h. During this time, the solution turned from a dark purple to
light orange. The solution was allowed to cool and the solvent was removed under
reduced pressure. The oil was then dissolved in CH2Cl2 (150 mL), washed with water
(200 mL) and brine (200 mL), and dried over Na2SO4. The filtrate was evaporated to dry-
ness, dissolved in CH2Cl2 (30 mL), and stored at room temperature for 1–2days. During this
time, a white solid precipitated, the crystals were filtered and washed with CH3CN (10 mL)
1
to yield 2 (855 mg, 1.87 mmol, 75% yield). H NMR (400 MHz, CDCl3, 300 K): d 7.93–7.95 (m,
1H), 7.43–7.45 (m, 2H), 7.09 (m, 1H), 6.46 (d, J¼ 8.81 Hz, 2H), 6.42 (d, J¼ 2.49 8.80 Hz, 2H),
6.27–6.30 (dd, J¼ 8.84 and 2.52 Hz, 2H), 3.62 (br s, 2H), 3.34 (q, J¼ 6.78 Hz, 8H), 1.17 (t,
J¼ 6.85 Hz, 12H) ppm. 13 C NMR (100 MHz, CDCl3, 300 K): d 166.1, 153.9, 151.6, 148.9,
132.5, 130.0, 128.1, 128.0, 123.9, 123.0, 108.1, 104.6, 98.0, 65.9, 44.4, 12.7 ppm. IR (ATR solid);
3220–3350 mNH2 (vw), 2961mC-H (w), 1761 mC¼O (s) cmꢂ1
.
2.8. Synthesis of 2-formyl-6-hydroxymethylpyridine (2) [57]
2,6-Bis(hydroxymethyl)pyridine (1.0 g, 7.2 mmol) was dissolved in hot CHCl3 (20 mL). The
solution was cooled, and activated MnO2 (1.3 g, 14 mmol) was added slowly with stirring.
The mixture was refluxed for 3 h. After this time, additional MnO2 (1.0 g, 11.6 mmol) was
added and allowed to reflux for an additional 3 h. Then the reaction mixture was filtered
hot through Celite. Upon cooling, the black inorganic salts were washed with CH3OH
(3ꢀ 25 mL). The filtrate was concentrated under reduced pressure and the resulting oily
residue was subjected to gradient column chromatography. Initially, the eluent started at
ethyl acetate:hexane (1:9), with the pure product eluting with ethyl acetate:hexane (4:6) as
1
a yellowish oil which turned into a solid after cooling (502 mg, 3.67 mmol g, 51%). H NMR
(400 MHz, CDCl3, 300 K): d 10.09 (s, 1H), 7.87–7.89 (m, 2H), 7.53 (t, J¼ 4.39 Hz, 1H), 4.88 (s,
2H); IR (ATR solid); 3125 mO-H (br), 2715 mCHO (w), 1710 mC¼O (s) cmꢂ1
.
2.9. Synthesis of 2-(hydroxymethylpyridine)methleneamine-spiro[isoindo-1,90
xanthene]-3-one (3)
Compounds 1 (0.228 g, 0.5 mmol) and 2 (0.68 g, 0.5 mmol) were dissolved in CH3OH
(20 mL) and refluxed for 5–6 h. The solution was allowed to cool and stand at room
temperature for 24 h, during which time crystals formed. These were filtered and
washed with cold CH3OH (3 ꢀ 10 mL) to yield the desired product as a light yellow
1
solid; yield (215 mg, 0.37 mmol, 75%, m.p 245 ꢃC). H NMR (400 MHz, CD3CN, 300 K): d
8.38 (s, 1H), 7.92 (d, J ¼ 7.46 Hz, 1H), 7.70 (t, J ¼ 7.71 Hz, 12H), 7.51–7.64 (m, 2H), 7.32
(d, J ¼ 7.52 Hz, 1H), 7.05 (d, J ¼ 7.39 Hz, 1H), 6.50 (d, J ¼ 8.87 Hz, 1H), 6.45 (d,
J ¼ 2.51 Hz, 1H), 6.31–6.34 (dd, J ¼ 8.87 and 2.55 Hz, 2H), 4.55 (d, J ¼ 5.76 Hz, 1H), 3.48
(t, J ¼ 5.78 Hz, 12H), 3.34 (q, J ¼ 7.04 Hz, 8H), 1.10 (t, J ¼ 7.01 Hz, 12H); 13C NMR
(101 MHz, CD3CN, 300 K): d 164.5, 160.8, 152.7, 152.6, 152.1, 148.9, 145.5, 137.0, 133.8,
128.4, 127.9, 127.3, 123.1, 122.9, 120.4, 117.4, 117.4, 107.9, 105.0, 97.3, 65.1, 63.9,43.6,
11.5 ppm. IR (ATR solid); 3435 mO-H (br), 2965, 2927, 2892 mC-H (w), 1703 mC¼O (vs)
cmꢂ1. LRMS-ESI; [M þ H]þ ¼ 576.4 (100%), [M þ Na]þ ¼ 598.5 (60%), [2M þ H]þ
¼
1150.8 (20%), [2M þ Na]þ ¼ 1173.1 (37%). Anal. Calcd for C35H37N5O3: H 6.48%; N
12.18%; C 73.02%. Found for C35H37N5O3: H 6.45%; N 12.07%; C 72.55%.

JOURNAL OF COORDINATION CHEMISTRY
385
Scheme 1. Synthesis of 3.
3. Results and discussion
Rhodamine-based chemosensors have attracted attention since the work by Czarnik et al.,
who developed a molecular probe for Cu2þ ions in which the Cu2þ induced ring-opening
of the lactam ring on the rhodamine B derivatives. This ring-opening produced a strong
fluorescence signal upon hydrolysis of the chemodosimeter-copper2þ adduct [58].
Xanthene derivatives have excellent photophysical properties; the neutral spirolac-
tam ring is non-fluorescent and colorless but upon ring-opening produces a strong
fluorescent signal (kem ꢄ 575 nm) and a distinct pink color (kab ꢄ 560 nm) [59]. As a
result of this ring opening and closing phenomenon, a number of rhodamine deriva-
tives have been utilized to detect biologically relevant ions and biomolecules [59–61].
Consequently, the xanthene family of dyes is ideal molecular compounds from which
to prepare an “off–on” chemosensor to detect ferric ions [62, 63].
Over the years, our group has designed and synthesized molecular probes for the moni-
toring of toxic analytes, including Fe3þ and Zn2þ [64–69], anions [70], and neutral species
[71–73]. Our previously published systems, and those of many other groups, work well in
organic solvents, which hinder their use in many biological or environmental applications.
Preparing water-soluble chemosensors is an ongoing endeavor for our group. Here we
describe the design and synthesis of a rhodamine B derived molecular probe that binds
Fe3þ in Staphylococcus aureus and the chemosensor’s ability to monitor ferric ions in a
mixed organic-aqueous water system, using MES as a buffer at pH 5.5.
3.1. Synthesis and pH studies
The commercially available rhodamine B was reacted with excess hydrazine hydyrate
to form 2-amino-30,60-bis(diethylamino)-2,3-dihydrospiro[isoindole-1,90-xanthene]-3-one
(1) in reasonable yield (75%). Compound 2 was prepared by reacting 2,6-bis(hydroxy-
methyl)pyridine with MnO2, a commonly used mild oxidizing agent, often used to con-
vert primary alcohols to aldehydes in the presence of other functional groups (ESI
Supporting information Scheme S1) [74, 75]. Finally, 1 and 2 were coupled via a con-
densation reaction to form 3 as shown in Scheme 1.
It is well known that pH is crucial when using rhodamine-based molecular probes
[44, 76]. Under acidic conditions, the spirolactam ring opens and pink color is
observed along with a strong fluorescent signal observed at kem ¼ 577 nm in CH3CN.
The correct pH window must be chosen so that the initial optical state is in the lactam
“off” form and, upon addition of metal ions, we can be confident that the fluorescence
signal is due to coordination of ferric ions. This eliminates false positives from changes
in pH that are encountered under physiological conditions. The optimum window for

386
E. MANANDHAR ET AL.
Figure 1. (A) The molecular structure of 3 showing displacement ellipsoids at the 50% probability
level; a single H2O molecule has been omitted for clarity. (B) Crystal packing showing how two
molecules are related in the unit cell.
chemosensor 3 in CH3CN is between pH 4 and 10 in the absence of metal ions (see
ESI, Supporting information Figure S1), suggesting that the molecular probe is in the
spirolactam form in this range.
3.2. X-ray crystallography and modeling calculations
Crystals of 3 were grown from a saturated methanolic solution over several days and
crystallized in the monoclinic P21/n space group (Figure 1(A)). There are two independ-
ent molecules in the asymmetric unit cell forming a dimer which is stabilized by C-Hꢁꢁꢁp
(3.588 Å) and a single hydrogen bonding interaction between the hydroxyl group to the car-
bonyl oxygen (O30-H3O0ꢁꢁꢁO20 (DꢁꢁꢁA¼ 2.7345(17)) via an inversion center, generating the

JOURNAL OF COORDINATION CHEMISTRY
387
Figure 2. DFT fully optimized structures of (A) [Fe(3)(CH3CN)3]3þ and (B) [Fe(3)(H2O)3]3þ with
hydrogens omitted for clarity.
two other molecules (z¼ 4) (see ESI, Supporting information Figure S2). The structure shows
that the substituted phenyl ring, to which the metal ion is expected to ligate, is orthogonal
to the xanthene ring (Figure 1(A)). This geometrical arrangement is typical for rhodamine
molecules [77, 78]. However, a particularly interesting structural feature is the significant
curvature of the xanthene ring systems (Figure 1(B)) as xanthenes are typically planar
[79–82]. Numerous attempts were made to grow crystals from 3 and metal salts to no avail.
To gain insight into possible ligand and coordination environments, a computational
approach was adopted. An initial molecular geometry was generated using molecular
mechanics energy minimization methods (Merck Molecular Force Field). The resulting atomic
coordinates were used as the input to a semi-empirical calculation, where formal bonds
were introduced between the solvent molecules and the metal center. Following geometry
optimization, formal bonds were removed, and further refinement was undertaken by dens-
ꢅ
ity functional methods (B3LYP/6-31G ) to give a final geometry. Both the corresponding
Fe3þ and Al3þ (see ESI, Supporting information Figure S3) coordination complexes were cal-
culated. As solution studies were carried out in a mixture of CH3CN (Figure 2(A)) and water
(Figure 2(B)), both solvents were incorporated into the coordination sphere of the metal ion
to mimic the solution work. The ferric ion is coordinated in a tridentate fashion, occupying
the equatorial positions, forming two five-membered chelating ring systems. Solvent mole-
cules occupy the remaining sites. The tris-aqua compound seems to be stabilized by a water
molecule bound in the axial position, which participates in an intramolecular hydrogen
bonding interaction with the carbonyl group. The same coordination environment is seen in
the Al3þ complex (see ESI, Supporting information Figure S3B).
3.3. IR Spectroscopy
In order to verify the binding mode of Fe3þ suggested by the molecular modeling cal-
culations, an IR study was conducted. The infrared spectrum of free chemosensor 3
was recorded as a solid using an ATR-FTIR and compared with that of the metal com-
plex. A 1:1 ratio of 3 and Fe(ClO4)3 was refluxed in CH3CN overnight, evaporated to
dryness, and the resulting solid allowed to air dry over several days. The pink/red solid

388
E. MANANDHAR ET AL.
was used to take an IR spectrum of the complex. Chemosensor 3 has distinctive func-
tional groups that are excellent IR handles to help our understanding of the coordin-
ation environment, as we were unable to grow crystals of the coordination complex.
The carbonyl group on the phenyl moiety has a characteristic IR stretching band at
1703 cmꢂ1 and a hydroxyl group at 3435 cmꢂ1, typical for hydrogen bonding interac-
tions (see ESI, Supporting information Figure S4A), supporting the hydrogen bonding
interaction seen in the crystal structure. The stretching band assigned to the carbonyl
group at 1703 cmꢂ1 is shifted to lower wavenumber and seen at 1645 cmꢂ1 in the
[Fe(3)]3þ complex. Upon coordination of the iron, the carbonyl group on the phenyl ring
loses its double bond character as the functional group now participates in coordination
with Fe3þ. The appearance of a sharp stretching band at 619 cmꢂ1 is indicative of a Fe-OH
bond (see ESI, Supporting information Figure S4B) [83] that also participates in hydrogen
bonding (see ESI, Supporting information Figure S4C). This is supported by change in the
C¼ O stretching frequency. Moreover, the analogous stretch in the ionic iron perchlorate
hydrate spectrum is weak and broad. Additionally, the appearance of several broad bands
between 3000 and 3500 cmꢂ1 is also indicative of the presence of coordinated water mole-
cules (see ESI, Supporting information Figure S4A).
Significant infrared spectral changes occur when perchlorate coordinates to a metal
ion through different modes [84]. Determination of the binding mode due to chang-
–
ing symmetry of the ClO4 ion upon binding, from Td (ionic) to C3v (unidentate) and
C2v (bidentate or bridging), can help identify the binding mode of the anion [85–87].
The IR spectrum of ionic Fe(ClO4)3 shows a broad symmetrical band at 1043 cmꢂ1, typ-
ical of the Td symmetry of the ion (see ESI, Supporting information Figures S4C). The
anion in the complex shows distinctive degeneracies, presumably due to the counter-
ion hydrogen bonding to the water molecules coordinated, which subsequently
changes the Td to C3v. This is in agreement with our previously published work [66].
1
3.4. H NMR studies
To obtain an understanding of the binding environment, we carried out 1D and 2D
NMR experiments, although, due to the paramagnetic nature of Fe3þ ions, we were
unable to perform NMR experiments directly with ferric salts. However, the Al3þ ion is
diamagnetic, with an ionic radius of 0.54 Å, which is similar to Fe3þ ions (0.55 Å), add-
itionally, Al3þ ions are known to have common coordination numbers of four and six,
making Al3þ ion an excellent substitute for ferric ions in the structural elucidation
studies. The 1D and 2D NMR spectra of chemosensor 3 and the [Al(3)]3þ, prepared in
situ by adding one equivalent of Al(ClO4)3 to 3, are shown in the Supporting
Information (Supporting information Figures S5–S27). Even though Al3þ ions do not
display any optical response at the mM concentration (vide infra), it is expected that at
significantly higher concentration (mM range) the equilibrium can be shifted towards
[Al(3)]3þ product. This is advantageous as this concentration is now within the NMR
concentration limits to aid in our understanding of the coordination environment.
The proton and carbon chemical environments close to the coordination environ-
ment had the greatest chemical shift change as shown in Figure 3. The imine proton
C(80)H has the most significant downfield chemical shift (D1.2 ppm), consistent with

JOURNAL OF COORDINATION CHEMISTRY
389
Figure 3. Partial 1H NMR of 3 (A) alone and upon addition of 1.2 equivalents of Al(ClO4)3
in CD3CN.
Figure 4. 1H NMR titration between 3 and Al(ClO4)3 in CD3CN, log K11 > 7.
the nitrogen coordinating to the metal center thereby making the proton more acidic,
and a considerable upfield carbon shift (D 6.2 ppm) is observed. A distinct triplet is
seen at 3.48 ppm, which can be assigned to the hydroxyl group, which disappears
upon metal coordination, supporting the binding of the oxygen to Al3þ. As a number
1
of the protons produced significant chemical shifts we carried out a H NMR titration
of 3 by adding aliquots of Al(ClO4)3 in CD3CN (Supporting information Table S1).
1
There were significant downfield chemical shifts in H NMR signals upon addition of
Al3þ for the protons of imine C(80)H, the methylene group C(140)H, and protons on the
pyridine ring. The changes in chemical shifts for C(80)H, C(140)H, and C(110)H after add-
ition of one equivalent of Al(ClO4)3 were 1.3, 0.4, and 0.7 ppm, respectively (Figure 4
and ESI, Supporting information Figures S28 and S29). Downfield shifts of pyridine pro-
tons can be explained by the inductive effect when the Al3þ ion is coordinated to the
pyridine nitrogen. The protons on the xanthene moiety also shift significantly in the

390
E. MANANDHAR ET AL.
Figure 5. UV-Vis optical screening of metal salts with 3 (50 lmolꢁdmꢂ3 in CH3CN upon addition of
–
ꢅ
three equivalents of metal ions, as their ClO4 salts ( Cl- salt).
downfield direction upon addition of Al3þ ions. This is due to the conjugation and
decrease in electron density of the xanthene ring once spirolactam ring of 3 opens
because of binding to Al3þ. No further chemical shifts were observed after the add-
ition of one equivalent of Al(ClO4)3 (Figure 4). The binding affinity between 3 and Al3þ
could not be calculated by nonlinear regression due to the high concentration used
for the NMR experiments (0.01 molꢁdmꢂ3) and only an approximate K11 value could be
obtained (log K > 7) using HypNMR (see ESI, Supporting information Figure S30) [55].
3.5. Optical studies
3.5.1. UV-Vis spectroscopy
The spectroscopic response of 3 was initially investigated by screening a range of bio-
logically relevant metal ions (Naþ, Kþ, Ca2þ, Mg2þ, Zn2þ, Cd2þ, Co2þ, Fe2þ Cu2þ, Hg2þ
,
Cr3þ, Al3þ, and Fe3þ) by adding 150 mmolꢁdmꢂ3 of metal salts into a solution of
CH3CN. The initial UV-Vis spectrum of 3 does not show the typical absorption band at
557 nm for rhodamine compounds, therefore the molecule is in the "off" state. Upon
addition of 150 lmolꢁdmꢂ3 ferric ions to 3, the solution turned from colorless to pink.
The absorption band at 557 nm can be attributed to the ring opened form because of
coordination of Fe3þ. No other metal ions (Naþ, Kþ, Ca2þ, Mg2þ, Cr3þ, Al3þ, Fe2þ
Zn2þ, Cd2þ, Co2þ, Hg2þ) gave a color change (Figure 5).
,
The UV-Vis screening studies suggest that 3 could serve as a "naked-eye" chemo-
sensor towards ferric ions in CH3CN and CH3CN aqueous mixture (MES buffer at pH
5.5). A UV-Vis titration was carried out with Fe(triflate)3. The counter-ion also plays a
role in the speciation; both the triflate and perchlorate salts were used and did not
affect the calculated binding constants [88]. The binding affinity was calculated using
the nonlinear regression program HypSpec [50, 89].To obtain a good fit between the

JOURNAL OF COORDINATION CHEMISTRY
391
Figure 6. (A) UV-Vis titration between 3 upon addition of Fe(CF3SO3)3 in a 10 lmolꢁdmꢂ3 in
CH3CN, (B) binding isotherm, (C) HypSpec 2014 [50] data and (D) HySS 2009 [89] simulated binding
isotherms, in excellent agreement with the experimental data.
observed experimental data and the model, different speciations had to be included.
An example of the UV-vis titration data is shown in Figure 6. Ferric triflate was added
to probe 3 in CH3CN. The UV-Vis spectra and the experimental binding isotherm are
shown in Figure 6(A,B), respectively. As the rhodamine dye ring opens, a hyperchromic
shift is seen at 560 nm. Moreover, a band at 375 nm suggests that the ferric ion has
coordinated to the molecular probe. The 1:1 binding stoichiometry between 3 and
Fe3þ ions was calculated to have an association constant of log K11 ¼ 6.9. Figure
6(C,D) show the least-squares fitting from the experimental data in Figure 6(B) and the
simulated binding isotherms (Figure 6(D)), which is in excellent agreement with the fit.
Monitoring and probing ferric ions in an aqueous environment is incredibly chal-
lenging. The concentration of ferric ions is very low under physiological conditions,
[Fe(H2O)6]3þ 10ꢂ18 M, [90], therefore organic ligands with a high affinity for ferric ions
(Kaff > 1030) need to be used to bind iron in the þ3 oxidation state (siderophores)
[91–93]. Another factor affecting the concentrations of ferric and ferrous ions in water
is the speciation of the metal, and this factor is often overlooked in the sensing com-
munity [94].
The hydrolysis of ferric ions in aqueous solutions is a convoluted time-dependent
system. It can be defined as hydrolysis-polymerization-precipitation. A simple mechan-
ism describes the process in several steps: (a) primary hydrolysis giving rise to low-

392
E. MANANDHAR ET AL.
Figure 7. UV-Vis titration (spectra and binding isotherms) between 3 upon addition of Fe(CF3SO3)3
in a 10 lmolꢁdmꢂ3 solution in (A) CH3CN:MES (95:5) and (B) CH3CN:MES (90:10).
molecular-weight complexes (monomer and dimer), i.e. Fe(OH)2þ
Fe2(OH)24þ; (b) formation and aging of polynuclear polymers, i.e. Fen(OH)m(H2O)x
,
Fe(OH)2
,
þ
(3n-m)þ
or FenOm(OH)x(3n-2m-x)þ; (c) precipitation of ferric oxides and hydroxides, i.e. Fe(OH)3,
FeOOH, and Fe2O3. The whole process, from hydrolysis to precipitation, can take sev-
eral years [95–99], however, initial hydrolysis is very rapid [95]. Different concentrations
of ferric species over a given pH range have been reported in predominance diagrams at
25 ꢃC [100]. Dimeric iron species have also been seen in solid-state structures. For
4þ
example, the dimeric [Fe2(OH)]2 ion has been suggested as the first condensation prod-
uct of [Fe(OH2)6]3þ, but with little evidence (outside kinetics), reported as far back as
Cotton and Wilkinson 1st edition [101]. In fact, hard evidence is in favor of a m-oxo
bridged species, as first shown by Junk et al. isolating [(H2O)5Fe–O–Fe(H2O)5]4þ in the
solid state as crown ether adducts characterized by X-ray diffraction [102]. In situ studies
by EXAFS on ferric nitrate solutions identified the m-oxo bridged dimer in solutions [103].
Despite substantial research on iron hydrolysis reactions, a significant amount of
conjecture remains with respect to the stability of iron hydrolytic species [104]. The
predominance diagrams suggest that at deficient iron concentrations, it is possible to
inhibit the formation of iron hydroxide precipitates in solution. One drawback of this,
however, is that extremely sensitive analytical techniques would be required to meas-
ure the concentration of iron complexes in solution precisely. To investigate the effect
of molecular probe 3 in an aqueous environment, we carried out UV-vis titrations

JOURNAL OF COORDINATION CHEMISTRY
393
Figure 8. A comparison of the binding isotherms obtained between 3 and Fe(CF3SO3)3 in a
10 lmolꢁdmꢂ3 solution in CH3CN, CH3CN:MES (95:5) and CH3CN:MES (90:10) (MES pH ¼
5.5, 10 mM).
Table 1. Binding constant (b) determined by UV-Vis titrations for the interaction between 3 and
ferric triflate, in different solvent systems.
Probe 3
aK11 (M–1
)
K12 (M–2
)
K21 (M–2
)
b
CH3CN
CH3CN:MES (95:5)
CH3CN:MES (90:10)
7.76 ꢀ 106
1.20 ꢀ 107
2.07 ꢀ 107
6.17 ꢀ 104
–
4.79 ꢀ 1011
1.82 ꢀ 1012
6.54 ꢀ 1010
–
–
1.51 ꢀ 105
3.16 ꢀ 103
aK values were obtained from HypSpec 2016. Different binding species had to be taken into account when fitting
the nonlinear regression data. Errors in K11 are <5%. K11 ¼ [Fe(3)]3þ; K12 ¼ [Fe(3)2]3þ; K21 ¼ [Fe2(3)]3þ
.
between 3 and ferric triflate in a CH3CN and MES buffer (pH ¼ 5.5; 10 mM) (Figure 7).
Initially, the UV-Vis titration behaves similar to the 100% CH3CN study. However, as
the MES buffer concentration increases the shape of the binding isotherm changes,
and the binding isotherm appears to continuously increase, without reaching satur-
ation, until five equivalents. The UV-Vis spectra, shown in Figures 7(A,B), show absorb-
ance intensity increases. This increase is indicative of other ferric species in solution,
most likely ferric hydroxide solid particulate that rapidly forms in the organic-aqueous
media in the experiment, which means that the ferric hydrolysis is occurring quicker
than binding between 3 and ferric ions. This is reasonable, as the binding environ-
ment in 3 does not resemble that of a siderophore motif and merely binds tridentate.
As the water content is increased, the shape of the binding isotherm changes; in a
noncompetitive solvent, like CH3CN, a typical 1:1 binding isotherm is obtained. As the
competitive nature of the medium increases from 5% to 10% (MES buffer pH ¼ 5.5),
the binding isotherms differ significantly (Figure 8). K11 is approximately the same (K11
ꢄ log 7), Table 1 (ESI, Supporting information Figures S34 and S35). However, the
model used to calculate the binding constants required different species to be part of
the non-linear fit algorithm. In the organic solvent, a 2:1 (2 x probe 3:Fe3þ) was
needed, but in both of the aqueous systems, the speciation was 1:2 (probe 3: 2 x
Fe3þ). This agrees with our understanding of the aqueous chemistry of ferric ions. As

394
E. MANANDHAR ET AL.
Figure 9. Fluorescence screening of metal salts with 3 (50 mM in CH3CN upon addition of three
–
ꢅ
equivalent of metal ions, as their ClO4 salts ( Cl- salt).
Figure 10. Fluorescence titration between 3 upon addition of Fe(ClO4)3 in a 5 lmolꢁdmꢂ3 solution
in CH3CN:H2O (95:5).
more of the ferric salt (>5 equivalents), the excess ferric ion undergoes hydrolysis,
which will form hydroxide species that precipitate out of solution or form bridging
complexes with the iron metal that has already coordinated to the molecular probe
[Fe(3)]3þ, in agreement with the species required by the non-linear fit model.

JOURNAL OF COORDINATION CHEMISTRY
395
3.6. Fluorescence spectroscopy
As fluorescence spectroscopy is more sensitive than absorbance, we investigated the
fluorometric response of 3 towards the same metal salts (Naþ, Kþ, Ca2þ, Mg2þ, Zn2þ
,
Cd2þ, Co2þ, Fe2þ Cu2þ, Hg2þ, Cr3þ, Al3þ and Fe3þ). Ferric ions quench the fluores-
cence intensity of organic fluorophores [105]. To investigate the extent of the quench-
ing in this system, we calculated the quantum yield (/) for 3 with ferric ions using the
comparative method [49]. The quantum yield was calculated to be 0.029 in ethanol
using rhodamine 6 G as a reference (/ ¼ 0.95 in EtOH). To confirm that the reduction
emission intensity is due to the quenching of the ferric ion, we calculated the quan-
tum yield of rhodamine 6 G-Hþ. This study showed that the quantum yield was ten
times higher than that of the iron complex (/ ¼ 0.29), in excellent agreement with
other rhodamine B sensors for ferric ions (see ESI, Supporting information Figures S31
and S32) [106]. Upon addition of 30 equivalents of Fe3þ to a CH3CN of 3
(5 mmolꢁdmꢂ3) and exciting at 550 nm, a strong fluorescence emission band at 577 nm
was observed. There was no evidence of a fluorescence band appearing on the add-
ition of other metal ions (Figure 9), strongly suggesting that 3 is selective for ferric
ions. To further investigate the interaction of 3 with Fe3þ, fluorescence titrations were
carried out in H2O-CH3CN solution. The addition of increasing concentrations of Fe3þ
resulted in a gradual increase of fluorescence intensity at 577 nm (Figure 10). The
binding affinity was calculated by the nonlinear regression curve fitting program
HypSpec using different speciation in the model. The association constant was calcu-
lated to be log K11 ¼ 4.54 [50].
Considering there were subtle fluorescence changes for the other metal ions, we
carried out competition experiments with metal ions that can exhibit cross-sensitivity
and reactivity to eliminate false positives in cell studies. A solution of 3 in CH3CN-MES
buffer (9:1, 1 m mM, pH 5.5) (5 mmolꢁdmꢂ3) contained 150 mmolꢁdmꢂ3 of metal salts
(Al3þ, Cr3þ, Fe2þ, Zn2þ, Cd2þ, Cu2þ, Naþ, Kþ, Co2þ and Hg2þ). Once the solution
reached equilibrium, 166 mmolꢁdmꢂ3 ferric perchlorate was added and an increase in
fluorescence was observed in all cases (see ESI, Supporting information Figure S36).
There was no significant interference with the fluorescence intensity. There was a
noticeable decrease with Cu2þ, Hg2þ, and Cd2þ ions which is not surprising as these
metal ions are known to quench fluorescence intensity [107, 108]. This suggests that 3
can be used as a selective fluorescent chemosensor for recognition of Fe3þ ion in the
presence of other metal ions.
An important facet of molecular sensor design is the concept of reversibility. If a
molecular probe is to be reused for treatment over several iterations in real applica-
tions, it should show reversibility, otherwise, it would be a “one shot” chemodosimeter
[109]. To demonstrate the reversibility of 3 with Fe3þ ions the siderophore DFB (des-
feroxamine B) was used to scavenge the ferric ion, as it is well known that sidero-
phores target ferric ions with very high binding affinities [68, 69, 110]. A solution of 3
and Fe3þ ions (five equivalents of Fe3þ ions to 20 lmolꢁdmꢂ3 solution) in CH3CN (see
ESI, Supporting information Figure S38) was prepared. Fluorescence was quenched
upon addition of DFB as it stripped the Fe3þ ions from the [Fe(3)]3þ complex, return-
ing 3 to its original lactam state, but this was reversed upon addition of two equiva-
lents of 2 mM Fe3þ ions. The observation confirms that 3 is a reversible fluorescent

396
E. MANANDHAR ET AL.
probe for Fe3þ ions and the spectral response of 3 to Fe3þ ions is due to chelation-
induced ring-opening of the rhodamine spirolactam. To determine the limit of detec-
tion (LoD) the method of least squares was used to give a line of regression. The con-
fidence limit of the slope is defined as b tsb, where t is the t-value taken from the
desired confidence and n-2 degrees of freedom. In our experiment, we chose a 95%
confidence level (t-value 2.306, df ¼ 10) (see ESI, Figure S39). It is generally accepted
that the LoD is the analyte concentration giving a signal equal to the blank signal
plus three standard deviations from the blank, i.e. y ¼ yB
3SB. The calculated LoD
value for 3 was 4 ꢀ 10ꢂ7 molꢁdmꢂ3 (ꢄ 27 ppb).
3.7. Confocal microscopy
The solution studies described were carried out in an organic-aqueous mixture (95:5
or 90:10) and this high concentration of organic solvent would certainly be an extreme
environment for cells to thrive. For a long time it was believed that any amount of
organic solvent was severe for microorganisms [111]. However, Inoue and Horikoshi
showed that a strain of Pseudomonas putida, a Gram negative bacterium, could
actively grow in the presence of 50% toluene solution [105], which was a surprising
result at the time. Since this original study, both Gram-negative and Gram-positive
bacteria have been found to tolerate high levels of organic solvents [112–114]. This
work suggests that bacteria have evolved their biochemical mechanisms to endure
these harsh conditions. Even though there is ample evidence that bacteria can tolerate
organic solvents, it is generally accepted that any biological studies that are carried
out in an organic environment should be kept to a minimum with a 5–15% organic
solvent-to-aqueous ratio being acceptable [106]. This ratio is often used to solubilize
molecular probes so they can be utilized in the cell medium. Knowing that bacteria
can thrive in an organic solvent, a 10% organic solvent system was used to show that
ferric ions can be detected in a Gram-positive bacterium.
Chemosensor 3 was also used to detect the free intracellular iron in bacterial cells
(Figure 11). The Gram-positive bacterium Staphylococcus aureus was used as it can
sequester and use intracellular iron in some of its virulence factor pathways. Bacterial
cells were grown for 3 h in TSB, treated with FeCl3 (100 lmolꢁdmꢂ3), and incubated for
a further 30 min. The cells were collected by centrifugation, washed twice with sterile
double distilled water, and treated with 100 lL of 1 mM of probe 3. As a control, bac-
terial cells grown under the same condition as above were also treated with 3 without
additional iron. A phase-contrast image of the bacteria was first taken to visualize cells,
and the same field was laser scanned in cy3/rhodamine settings with emission and
excitation wavelengths of 550 and 580 nm, respectively, to detect fluorescence. Equal
numbers of bacteria were observed in both test and control, but red fluorescence was
only detected for those treated with iron (Figure 9). This suggests that the fluores-
cence produced in the test is due to free intracellular iron detected by the chemosen-
sor and not background fluorescence. The results showed that 3 could be used to
detect free intracellular iron in prokaryotic cells like bacteria.

JOURNAL OF COORDINATION CHEMISTRY
397
Figure 11. Confocal microscopy for detecting intracellular free iron using 3. Phase-contrast images
(A and C) and fluorescence images (B and D) of S. aureus. The upper panel is the iron-free control.
4. Conclusion
Many molecular probes that are designed to bind Fe3þ ions often fail to consider the
potential for interference from other trivalent species, in particular, Cr3þ and Al3þ ions
[78, 81, 115–117]. We believe that this is intentional as these ions are also hard Lewis
acids, which can also have high affinities similar to Fe3þ ions and elicit optical
responses. There are a small number of reports of probes that do take Al3þ into con-
sideration [7, 118] and, ironically, the Al3þ ion has turned out to be helpful in our
studies. We have used 2D NMR to investigate the binding mode of Al3þ to these
probes, as this gave insight into the Fe3þ coordination environment.
Many ferric ion probes reported in the literature are a little misleading, often claim-
ing that their sensors can monitor ferric ions under conditions in which ferric ions
alone cannot exist, a phenomenon we have observed in our own laboratory. This
work is the first example of a rhodamine probe utilized to monitor ferric ions in a pro-
karyotic system, Staphylococcus aureus. Whereby, the phase-contrast images, besides
showing the presence of bacteria, also showed the presence of fairly equal numbers
of cells in both the test and negative control, which suggests that the fluorescence
signal is due to the presence of free intracellular iron detected by 3 and not due to
any background fluorescence. It is a significant result as this is one of a few examples
where iron can be confidently detected in bacteria cells. Rhodamine probes have pre-
viously been used to detect intracellular iron in eukaryotic living cells; however, the

398
E. MANANDHAR ET AL.
published work is questionable as eukaryotic cells have low levels of Fe3þ ions,
required for cell growth, already present. Therefore, often, the "model" systems (i.e. no
iron added) already exhibit an emission signal as it is very difficult to grow eukaryotic
cells that are iron-free.
In summary, we have synthesized and characterized a new fluorescent “turn-on
probe” for Fe3þ ions, which operates in CH3CN-MES buffer. The chemosensor exhibits
high fluorescence enhancement and selectivity for ferric ions over other metal ions.
1
Binding was also studied by H NMR, IR-spectroscopy, and DFT calculations. Confocal
microscopy experiments showed that 3 can monitor free intracellular iron in the
Gram-positive bacterium Staphylococcus aureus.
Funding
KJW would like to thank the NSF (OCE-0963064 and CHE-0840390 for the Acquisition of a cyber
accessible 400 MHz NMR at the University of Southern Mississippi) for financial support. PJC
thanks the US Army Research Office for funding computational resources (Contract W911NF-15-
1-0624). Additional support for the study was made possible by the Mississippi INBRE, an
Institutional Development Award (IDeA) from the National Institute of General Medical Sciences
(grant number P20GM103476).
Accession codes
CCDC 1560969 contain the supplementary crystallographic data for this paper. These data can
be obtained via XXXX, or by contacting the Cambridge Crystallographic Data Centre, 12 Union
Road, Cambridge CB2 IEZ, UK; Fax þ44 1223 336033.
ORCID
Peter J. Cragg
Karl J. Wallace
http://orcid.org/0000-0002-6134-2093
http://orcid.org/0000-0003-4541-7523
References
[1] J. Yin, Y. Hu, J. Yoon. Chem. Soc. Rev., 44, 4619 (2015).
[2] Y.L. Pak, K.M.K. Swamy, J. Yoon. Sensors (Basel), 15, 24374 (2015).
[3] K.P. Carter, A.M. Young, A.E. Palmer. Chem. Rev., 114, 4564 (2014).
[4] C.D. Fryar, Q. Gu, C.L. Ogden. Anthropometric reference data for children and adults:
United States, 2007-2010.U.S. Department of health and human services, Centers for
Disease Control and Prevention, Washington DC, Vol. 252 (2012).
[5] Y. Yu, D.S. Kalinowski, Z. Kovacevic, A.R. Siafakas, P.J. Jansson, C. Stefani, D.B. Lovejoy,
P.C. Sharpe, P.V. Bernhardt, D.R. Richardson. J. Med. Chem., 52, 5271 (2009).
[6] S. Epsztejn, H. Glickstein, V. Picard, I.N. Slotki, W. Breuer, C. Beaumont, V.L. Cabantchik.
Blood, 94, 3593 (1999).
[7] S.K. Sahoo, D. Sharma, R.K. Bera, G. Crisponi, J.F. Callan. Chem. Soc. Rev., 41, 7195 (2012).
[8] P. Aisen, M. Wessling-Resnick, E.A. Leibold. Curr. Opin. Chem. Biol., 3, 200 (1999).
[9] A. Jacobs. Blood, 50, 433 (1977).
[10] O. Kakhlon, Z.I. Cabantchik. Free Radic. Biol. Med., 33, 1037 (2002).
[11] M. Kruszewski. Mutat. Res., 531, 81 (2003).
[12] M. Valko, C.J. Rhodes, J. Moncol, M. Izakovic, M. Mazur. Chem. Biol. Interact., 160, 1
(2006).

JOURNAL OF COORDINATION CHEMISTRY
399
[13] B. Halliwell, J.M.C. Gutteridge. Methods Enzymol., 186, 1 (1990).
ꢀ
ꢀ
ꢀ
ꢀ
[14] S. Rivera-Mancia, I. Perez-Neri, C. Rıos, L. Tristan-Lopez, L. Rivera-Espinosa, S. Montes.
Chem. Biol. Interact., 186, 184 (2010).
[15] K.J. Thompson, S. Shoham, J.R. Connor. Brain Res. Bull., 55, 155 (2001).
[16] H. Kozlowski, A. Janicka-Klos, J. Brasun, E. Gaggelli, D. Valensin, G. Valensin. Coord. Chem.
Rev., 253, 2665 (2009).
[17] P. Zatta, D. Drago, S. Bolognin, S.L. Sensi. Trends Pharmacol. Sci., 30, 346 (2009).
[18] J.D. Haas, T.I.V. Brownlie. J. Nutr., 131, 676S (2001).
[19] C. Brugnara. Clin. Chem., 49, 1573 (2003).
[20] M. Kumar, R. Kumar, V. Bhalla, P.R. Sharma, T. Kaur, Y. Qurishi. Dalton Trans., 41, 408
(2012).
[21] M. Kumar, R. Kumar, V. Bhalla. Tetrahedron Lett, 51, 5559 (2010).
[22] V. Bhalla, N. Sharma, N. Kumar, M. Kumar. Sens. Actuators B, 178, 228 (2013).
[23] V. Bhalla, A. Gupta, M. Kumar. Dalton Trans., 42, 4464 (2013).
[24] M.H. Lee, T. van Giap, S.H. Kim, Y.H. Lee, C. Kang, J.S. Kim. Chem. Commun. (Camb), 46,
1407 (2010).
[25] M.H. Lee, H. Lee, M.J. Chang, H.S. Kim, C. Kang, J.S. Kim. Dyes Pigm., 130, 245 (2016).
[26] D.R. Richardson, P.C. Sharpe, D.B. Lovejoy, D. Senaratne, D.S. Kalinowski, M. Islam, P.V.
Bernhardt. J. Med. Chem., 49, 6510 (2006).
€
[27] N.C. Lim, S.C. Pavlova, C. Bruckner. Inorg. Chem., 48, 1173 (2009).
[28] S.M. Cohen, M. Meyer, K.N. Raymond. J. Am. Chem. Soc., 120, 6277 (1998).
[29] R.J. Ward, F.A. Zucca, J.H. Duyn, R.R. Crichton, L. Zecca. Lancet. Neurol., 13, 1045 (2014).
[30] R.J. Ward, D.T. Dexter, R.R. Crichton. J. Trace Elem. Med. Biol., 31, 267 (2015).
[31] R. Cusnir, C. Imberti, R.C. Hider, P.J. Blower, M.T. Ma. Int. J. Mol. Sci., 18, 1 (2017).
[32] A.J.M. Messenger, R. Barclay. Biochem. Educ., 11, 54 (1983).
[33] T.C. Johnstone, E.M. Nolan. Dalton Trans., 44, 6320 (2015).
[34] D. Perkins-Balding, M. Ratliff-Griffin, I. Stojiljkovic. Microbiol. Mol. Biol. Rev., 68, 154
(2004).
[35] A. Pandey, F. Bringel, J.-M. Meyer. Appl. Microbiol. Biotechnol., 40, 765 (1994).
[36] M. Imbert, R. Blondeau. Curr. Microbiol., 37, 64 (1998).
[37] K.H. Ebrahimi, E. Bill, P.-L. Hagedoorn, W.R. Hagen. Nat. Chem. Biol., 8, 941 (2012).
[38] I.U. Heinemann, M. Jahn, D. Jahn. Arch. Biochem. Biophys., 474, 238 (2008).
[39] A.S. Fleischhacker, P.J. Kiley. Curr. Opin. Chem. Biol., 15, 335 (2011).
[40] E. Banin, M.L. Vasil, E.P. Greenberg. Proc. Natl. Acad. Sci. USA, 102, 11076 (2005).
[41] J. Kluytmans, A. van Belkum, H. Verbrugh. Clin. Microbiol. Rev., 10, 505 (1997).
[42] N.R. Chereddy, S. Thennarasu, A.B. Mandal. Dalton Trans., 41, 11753 (2012).
[43] S.-R. Liu, S.-P. Wu. Sens. Actuators B, 171–172, 1110 (2012).
[44] M. Tian, X. Peng, J. Fan, J. Wang, S. Sun. Dyes Pigm., 95, 112 (2012).
[45] J. Wang, Y. Li, N.G. Patel, G. Zhang, D. Zhou, Y. Pang. Chem. Commun. (Camb), 50, 12258
(2014).
[46] S. Goswami, K. Aich, S. Das, A.K. Das, D. Sarkar, S. Panja, T.K. Mondal, S. Mukhopadhyay.
Chem. Commun. (Camb), 49, 10739 (2013).
[47] S.B. Maity, P.K. Bharadwaj. Inorg. Chem., 52, 1161 (2013).
[48] Y.-S. Wu, C.-Y. Li, Y.-F. Li, J.-L. Tang, D. Liu. Sens. Actuators B, 203, 712 (2014).
[49] A.T.R. Williams, S.A. Winfield, J.N. Miller. Analyst, 108, 1067 (1983).
[50] P.; Gans. Protonic Software: Leeds, UK (2014)
[51] Nonius; Nonius BV, Delft, The Netherlands (2000)
[52] Z. Otwinowski, W. Minor. In Methods in Enzymology, C.W. Carter Jr., R.M. Sweet (Eds.),
Vol. 276, Academic Press, New York (1997).
[53] A. Altomare, M.C. Burla, M. Camalli, G.L. Cascarano, C. Giacovazzo, A. Guagliardi, A.G.G.
Moliterni, G. Polidori, R. Spagna. J. Appl. Crystallogr., 32, 115 (1999).
[54] G.M. Sheldrick. Acta Crystallogr. C Struct. Chem., 71, 3 (2015).
[55] P. Gans. Protonic Software: Leeds, UK (2016)
[56] Z. Jin, D.-X. Xie, Z.-B. Zhang, Y.-J. Gong, W. Tan. Anal. Chem., 84, 4253 (2012).

400
E. MANANDHAR ET AL.
[57] D.J. Hutchinson, L.R. Hanton, S.C. Moratti. Inorg. Chem., 49, 5923 (2010).
[58] V. Dujols, F. Ford, A.W. Czarnik. J. Am. Chem. Soc., 119, 7386 (1997).
[59] M. Beija, C.A.M. Afonso, J.M.G. Martinho. Chem. Soc. Rev., 38, 2410 (2009).
[60] H.N. Kim, M.H. Lee, H.J. Kim, J.S. Kim, J. Yoon. Chem. Soc. Rev., 37, 1465 (2008).
[61] X. Chen, T. Pradhan, F. Wang, J.S. Kim, J. Yoon. Chem. Rev., 112, 1910 (2012).
[62] Y. Ma, W. Luo, P.J. Quinn, Z. Liu, R.C. Hider. J. Med. Chem., 47, 6349 (2004).
[63] M. Kumar, R. Kumar, V. Bhalla. Org. Lett., 13, 366 (2011).
[64] A.B. Davis, F.R. Fronczek, K.J. Wallace. Acta Crystallogr, E72, 1032 (2016).
[65] A.M. Mallet, A.B. Davis, D.R. Davis, J. Panella, K.J. Wallace, M. Bonizzoni. Chem. Commun.
(Camb), 51, 16948 (2015).
[66] E. Manandhar, P.J. Cragg, K.J. Wallace. Supramol. Chem., 26, 141 (2014).
[67] E. Manandhar, J.H. Broome, J. Myrick, W. Lagrone, P.J. Cragg, K.J. Wallace. Chem.
Commun. (Camb), 47, 8796 (2011).
[68] K.M. Orcutt, W.S. Jones, A. McDonald, D. Schrock, K.J. Wallace. Sensors (Basel), 10, 1326
(2010).
[69] K.J. Wallace, M. Gray, Z. Zhong, V.M. Lynch, E.V. Anslyn. Dalton Trans., 2436 (2005).
[70] A.B. Davis, R.E. Lambert, F.R. Fronczek, P.J. Cragg, K.J. Wallace. New J. Chem., 38, 4678
(2014).
[71] I. Walton, M. Davis, L. Yang, Y. Zhang, D. Tillman, W.L. Jarrett, M.T. Huggins, K.J. Wallace.
Magn. Reson. Chem., 49, 205 (2011).
[72] W. Walton, D. Marauo, L. Munro, V.J. Catalano, P.J. Cragg, M.T. Huggins, K.J. Wallace.
Org. Lett., 14, 2686 (2012).
[73] K. Kellett, J.H. Broome, M. Zloh, S.B. Kirton, S. Fergus, U. Gerhard, J.L. Stair, K.J. Wallace.
Chem. Commun. (Camb), 52, 7474 (2016).
[74] J.-E. Backvall. Modern Oxidation Methods, Wiley-VCH, Weinham (2003).
[75] G. Tojo, M. Fernandez. Oxidation of Alochols to Aldehydes and Ketones: A Guide to Current
Common Practice, Springer Science & Business Media, New York (2006). https://www.
springer.com/gp/book/9780387236070
[76] Q.A. Best, R. Xu, M.E. McCarroll, L. Wang, D.J. Dyer. Org. Lett., 12, 3219 (2010).
[77] S. Kenmoku, Y. Urano, H. Kojima, T. Nagano. J. Am. Chem. Soc., 129, 7313 (2007).
[78] W. Huang, C. Song, C. He, G. Lv, X. Hu, X. Zhu, C. Duan. Inorg. Chem., 48, 5061 (2009).
[79] J.D. Chartres, M. Busby, M.J. Riley, J.J. Davis, P.V. Bernhardt. Inorg. Chem., 50, 9178
(2011).
[80] Z. Yang, M. She, B. Yin, J. Cui, Y. Zhang, W. Sun, J. Li, Z. Shi. J. Org. Chem., 77, 1143
(2012).
[81] X. Chen, X. Meng, S. Wang, Y. Cai, Y. Wu, Y. Feng, M. Zhu, Q. Guo. Dalton Trans., 42,
14819 (2013).
[82] A. Sahana, A. Banerjee, S. Lohar, A. Banik, S.K. Mukhopadhyay, D.A. Safin, M.G.
Babashkina, M. Bolte, Y. Garcia, D. Das. Dalton Trans., 42, 13311 (2013).
[83] Y.J. Kim, C.R. Park. Inorg. Chem., 41, 6211 (2002).
[84] M.R. Rosenthal. J. Chem. Educ., 50, 331 (1973).
[85] G. Herzberg. Infrared and Raman Spectra, Litton Educational Publishing, New York
(1944).
[86] A.S. Goldin, R.J. Velten, G.W. Frishkorn. Anal. Chem., 31, 1490 (1959).
[87] R.D. Armstrong, D.F. Porter, H.R. Thirsk. J. Phys. Chem., 72, 2300 (1968).
[88] A.D.G. Johnson. Dissertation, The University of Southern Mississippi (2020).
[89] P. Gans. Protonic Software: Leeds, UK (2009).
[90] K.N. Raymond, C.J. Carrano. Acc. Chem. Res., 12, 183 (1979).
[91] E. Griffiths. Iron and Infection, In Iron and infection, Vol. 69, 2nd Edn, pp. 69–137,
Chichester
UK
(1987).
https://books.google.com/books/about/Iron_and_Infection.
html?id=Bw5rAAAAMAAJ&source=kp_book_description
[92] K.G. Wooldridge, P.H. Williams. FEMS Microbiol. Rev., 12, 325 (1993).
[93] R.C. Hider, X. Kong. Nat. Prod. Rep., 27, 637 (2010).
[94] K.J. Wallace, A.D.G. Johnson, W.S. Jones, E. Manandhar. Supramol. Chem., 30, 353 (2018).

JOURNAL OF COORDINATION CHEMISTRY
401
[95] A.N. Pham, A.L. Rose, A.J. Feitz, T.D. Waite. Geochim. Cosmochim. Acta, 70, 640 (2006).
[96] M. Zhu, C. Frandsen, A.F. Wallace, B. Legg, S. Khalid, H. Zhang, S. Morup, J.F. Banfield,
G.A. Waychunas. Geochim. Cosmochim. Acta, 172, 247 (2016).
[97] P.G. Daniele, C. Rigano, S. Sammartano, V. Zelano. Talanta, 41, 1577 (1994).
[98] P. Hemmes, L.D. Rich, D.L. Cole, E.M. Eyring. J. Phys. Chem., 75, 929 (1971).
[99] R.L. Martin, J.P. Hay, L.R. Pratt. J. Phys. Chem. A, 102, 3565 (1998).
[100] D.G. Kinniburgh, D.M. Cooper. Environ. Sci. Technol., 38, 3641 (2004).
[101] A.F. Cotton, G. Wilkinson. Advanced Inorganic Chemistry: A Comprehensive Text, John
Wiley and Sons, United States (1962).
[102] P.C. Junk, B.J. McCool, B. Moubaraki, K.S. Murray, L. Spiccia, J.D. Cashion, J.W. Steed. J.
Chem. Soc, Dalton Trans., 1024 (2002).
[103] M. Zhu, B.W. Puls, C. Frandsen, J.D. Kubicki, H. Zhang, G.A. Waychunas. Inorg. Chem., 52,
6788 (2013).
[104] R.H. Byrne, W. Yao, Y.-R. Luo, B. Wang. Marine Chem., 97, 34 (2005).
[105] J.R. Lakowicz. Principles of Fluorescence Spectroscopy, 3rd edn, Springer, New York (2006).
[106] A.J. Weerasinghe, C. Schmiesing, S. Varaganti, G. Ramakrishna, E. Sinn. J. Phys. Chem. B.,
114, 9413 (2010).
[107] V. Bhalla, M. Kumar, P.R. Sharma, T. Kaur. Inorg. Chem, 51, 2150 (2012).
[108] J. Wang, L. Long, D. Xie, X. Song. Sens. Actuators B, 177, 27 (2013).
[109] B. Wang, E.V. Anslyn. Chemosensors: Principles, Strategies, and Applications, John Wiley &
Sons, Hoboken, New Jersey (2011).
ꢀ
~
[110] S.C. Andrews, A.K. Robinson, F. Rodrıguez-Quinones. FEMS Microbiol. Rev., 27, 215 (2003).
[111] R. Aono, K. Aibe, A. Inoue, K. Horikoshi. Agric. Biol. Chem., 55, 1935 (1991).
[112] D. K. Hong, S.-H. Jang, C. Lee. J. Mol. Catal. B Enzym., 133, S337 (2017).
[113] Y.N. Sardessai, S. Bhosle. Biotechnol. Prog., 20, 655 (2004).
[114] N. Doukyu, H. Ogino. Biochem. Eng. J., 48, 270 (2010).
[115] F. Ge, H. Ye, H. Zhang, B.-X. Zhao. Dyes Pigm., 99, 661 (2013).
[116] M. Chai, D. Zhang, M. Wang, H. Hong, Y. Ye, Y. Zhao. Sens. Actuators B, 174, 231 (2012).
[117] Z.-Q. Hu, Y.-C. Feng, H.-Q. Huang, L. Ding, X.-M. Wang, C.-S. Lin, M. Li, C.-P. Ma. Sens.
Actuators B, 156, 428 (2011).
[118] Y. Zhou, H. Zhou, J. Zhang, L. Zhang, J. Niu. Spectrochim. Acta A Mol. Biomol. Spectrosc.,
98, 14 (2012).