Studies on the interaction between poly-phosphane gold(I) complexes and dihydrofolate reductase: An interplay with nicotinamide adenine dinucleotide cofactor

Article abstract of DOI:10.3390/ijms20071802

A class of gold(I) phosphane complexes have been identified as inhibitors of dihydrofolate reductase (DHFR) from E. coli, an enzyme that catalyzes the reduction of dihydrofolate (DHF) to tetrahydrofolate (THF), using NADPH as a coenzyme. In this work, to

Full text of DOI:10.3390/ijms20071802

International Journal of
Molecular Sciences
Article
Studies on the Interaction between Poly-Phosphane
Gold(I) Complexes and Dihydrofolate Reductase:
An Interplay with Nicotinamide Adenine
Dinucleotide Cofactor
Stefania Pucciarelli 1 , Silvia Vincenzetti 1 , Massimo Ricciutelli 2, Oumarou Camille Simon 2,
Anna Teresa Ramadori 2, Lorenzo Luciani 2 and Rossana Galassi 2,
*
1
School of Biosciences and Veterinary Medicine, University of Camerino, Via Gentile III da Varano,
62032 Camerino, Italy; stefania.pucciarelli@unicam.it (S.P.); silvia.vincenzetti@unicam.it (S.V.)
School of Science and Technology, University of Camerino, Via Sant’Agostino, 1, 62032 Camerino, Italy;
massimo.ricciutelli@unicam.it (M.R.); camilleoumarou@yahoo.fr (O.C.S.);
2
annateresa.ramadori@unicam.it (A.T.R.); lorenzo.luciani@unicam.it (L.L.)
*
Correspondence: rossana.galassi@unicam.it
ꢀꢁꢂꢀꢃꢄꢅꢆꢇ
ꢀꢁꢂꢃꢄꢅꢆ
Received: 18 March 2019; Accepted: 9 April 2019; Published: 11 April 2019
Abstract: A class of gold(I) phosphane complexes have been identified as inhibitors of dihydrofolate
reductase (DHFR) from E. coli, an enzyme that catalyzes the reduction of dihydrofolate (DHF) to
tetrahydrofolate (THF), using NADPH as a coenzyme. In this work, to comprehend the nature of the
interaction at the basis of these inhibitory effects, the binding properties of bis- and tris-phosphane
gold(I) chloride compounds in regards to DHFR have been studied by emission spectroscopy and
spectrophotometric assays. The lack of cysteine and seleno-cysteine residues in the enzyme active
site, the most favorable sites of attack of Au(I) moieties, makes this work noteworthy. The interaction
with the gold compounds results into the quenching of the DHFR tryptophan’s emissions and in
an enhancement of their intrinsic emission intensities. Moreover, a modulating action of NADPH
is highlighted by means of an increase of the gold compound affinity toward the enzyme; in fact,
the dissociation constants calculated for the interactions between DHFR and each gold compound
in the presence of saturating NADPH were lower than the ones observed for the apo-enzyme.
The fluorimetric data afforded to Kd values ranged from 2.22
±
0.25 µM for (PPh3)2AuCl in the
presence of NADPH to 21.4 3.85
±
µ
M for 4L3AuTf in the absence of NADPH. By elucidating the
energetic aspects of the binding events, we have attempted to dissect the role played by the gold
phosphane/protein interactions in the inhibitory activity, resulting in an exothermic enthalpy change
and a positive entropic contribution (
∆
H◦ =
−
5.04
±
0.08 kcal/mol and
∆
S◦ = 7.34
±
0.005 cal/mol K).
·
Keywords: protein–metal adducts; enzyme inhibition; DHFR; gold(I) phosphane compounds;
thermodynamic data; enzymatic assays
1. Introduction
Dihydrofolate reductase (DHFR) is a ubiquitous enzyme responsible for maintaining a pool of
tetrahydrofolate (labelled as THF or H4F) in cells. THF is an active form of the vitamin folic acid
and the precursor for purines, pyrimidines, and several amino acids necessary for synthesis in cells
and for their growth [1]. In particular, DHFR catalyzes the reduction of 7,8-dihydrofolate (labelled
as DHF or H2F) to THF by stereospecific hydride transfer from the NADPH cofactor (Nicotinamide
Adenine Dinucleotide Phosphate) to the C6 atom of the pterin ring, with concomitant protonation at
Int. J. Mol. Sci. 2019, 20, 1802; doi:10.3390/ijms20071802
www.mdpi.com/journal/ijms

Int. J. Mol. Sci. 2019, 20, 1802
2 of 17
N5 (Scheme 1) [2]. This catalytic role makes DHFR a therapeutic target by chemical and genetic means,
having the scope to hamper the proliferation and the growth of cells [3,4].
Scheme 1. Schematic representation of the reduction reaction catalyzed by dihydrofolate reductase (DHFR).
Classic antimicrobial DHFR inhibitors [5] are TMP (trimetoprin) and its analogs [4,6]; their action
is mainly based on the competition with the pteridine moiety of DHF. Methotrexate is the DHFR
inhibitor used most often in a clinical setting as an anticancer drug, and as an anti-inflammatory and
immunosuppressive agent. Methotrexate (MTX) acts through a 103 stronger bond to DHFR than to DHF
itself [
serious side effects observed, resulting in a complete loss of efficacy [
new entities possessing the skill to inhibit DHFR becomes a challenging issue, as well as the study of the
mechanism developing the drug resistance [11 12]. Whilst this latter aspect has been approached, the
7
,
8]. The antifolate treatments may afford eventual resistance, which is one of the most important
9,
10]. Then, the identification of
,
research of innovative antifolate compounds has been neglected. The design of antifolate compounds
was mainly led by considering structural DHF analogies [11], and few alternative molecules have been
tested as DHFR inhibitors [12]. In recent research, gold(I) azolate/phosphane compounds (where the
phosphane was PPh3, and the azolate were some 3,5-di-substituted pyrazoles and 4,5-di-substituted
imidazoles) were found to be cytotoxic in vitro against many panels of cancer cells, and two of them
were found to be strongly active also in vivo on the treatment of BBLC sick syngeneic mice. They
were found to possess mechanisms of action that were strongly dependent on their structure. In
this regard, some years ago, our group was involved in a pioneer study where, surprisingly, these
phosphane gold(I) compounds [13,14] and their relative bis-phosphane gold(I) compounds were found
to inhibit DHFR from E. coli in the micromolar range of concentration [15]. This latter work was done
with the double goal to contribute to the shared hypothesis that bioactive anticancer or antimicrobial
gold(I) phosphane compounds interact with different enzymes, depending on (even mild) structural
modifications [16,17], and with the aim to find new inhibitors of DHFR. Previously [15], mono- or
bis- phosphane gold(I) compounds were tested as DHFR inhibitor; nevertheless, a structure activity
relationship (SAR) was difficult to be discussed for mono- and bis- phosphane gold(I) compounds
and some anomalies on the Ki’s values solicited a further study in depth. In order to accomplish
this deepening, some gold(I) phosphane compounds were reconsidered and additional phosphane
gold(I) compounds were synthesized to evaluate the nature of the DHFR/NADPH/gold compounds
interactions. Exploiting the emissive properties of the tryptophan groups in DHFR, spectrofluorimetric
and UV-visible spectroscopy assays [18
–20] were performed to obtain kinetic and thermodynamic data
for the DHFR/gold phosphane compounds adducts. For this purpose, PPh3 (L) and analogs containing

Int. J. Mol. Sci. 2019, 20, 1802
3 of 17
the COOH in ortho (2L)/para (4L) position, or the COOCH3 (4MeL) in para position of the phosphane
ligand, were used to synthesize mononuclear bis- and tris-phosphane gold(I) compounds.
2. Results and Discussion
2.1. Gold(I) Compounds Preparation
For this work, the synthesis of a selection of gold(I) poly-phosphane complexes was planned to
understand some aspects of the structural/activity/relationship of these gold compounds. Specifically,
(a) the change of the inhibitory activity as a function of the number of phosphane ligands around
the metal center, (b) the effect of a carboxylic group in the phosphane ligand on the mechanism of
interaction with the enzyme, (c) the putative role of the chloride ion in the activity of the gold complex,
and (d) the binding, if indeed it occurs, between the OH of the carboxylic group and the DHFR in the
interaction event. The synthesis of bis- and tris-phosphane gold(I) complexes with PPh3 has already
4
2
been discussed in literature and it represents the inspiration for the preparation of L or L gold
complexes [21]. The syntheses proceed according to the substitution of the labile (Me2S) by direct
addition of the phosphane ligands to [AuCl(Me2S)] in a mole ratio 4:1, respectively; the excess of the
phosphane ligand guarantees the shift of the equilibrium reaction toward the product, which is then
isolated as a solid and characterized by elemental analysis, IR spectroscopy in the solid state, and 1H, 31
P
NMR, UV-visible, spectrofluorimetric spectroscopies, and ESI Mass Spectrometry in solution (Electro
Spray Ionization Mass Spectrometry). The presence of a deactivating group, such as the carboxylic
in para or ortho position of one phenyl group, confers less σ-donor skill to the phosphane ligands,
labelled as 4L or 2L, yielding to quite stable gold complexes with higher coordination numbers around
the gold center. The synthesis of the compound 4L3AuTfl (Tfl = triflate) was performed according
4
to the direct reaction of the ligand, L, with the [AuTfl(Me2S)] in the mole ratio 4:1; this method
4
4
was preferred to the direct methatesis between L3AuCl and AgTfl. The analysis of the L3AuTfl
complex highlights a likely role of the chloride ion in the stability of the gold complexes in solution
which was already discussed for the PPh3 analogs [21]: if we compare the 31P NMR chemical shift of
4L3AuCl and of 4L3AuTfl, we observe a shift to lower frequencies (from 42.1 to 32.6 ppm in CH3OD,
respectively) and the appearance of a signal at 22.6 ppm for the latter (with an intensity of about 1:10
relative to the main product at 32.6 ppm). These data indicate the absence of the chloride markedly
changes the electronic environment of the phosphorous center, and affords a likely dissociation to the
monophosphane gold(I) compound; in fact, the 31P NMR chemical shift of the tris- and bis-phosphane
gold(I) complexes fall quite close to each other, which are 8–10 ppm higher frequencies with respect
to the chemical shift of the LAuCl derivative [15]. Finally, the formation of the methyl ester on the
ligand 4L and of the corresponding gold(I) complex, 4MeL3AuCl, had the scope to suppress the skill of
the 4-benzoic-diphenylphoshane acid ligand to act as a hydrogen bonding donor in the interaction
with DHFR.
2.2. Inhibition Studies
The phosphane gold(I) compounds were used to prepare buffered stock solutions at pH = 7.3 at
30 ◦C for the inhibition tests of DHFR from E coli. In Table 1, the list of the gold(I) compounds used for
the tests and the corresponding inhibition constants are reported.

Int. J. Mol. Sci. 2019, 20, 1802
4 of 17
Table 1. List of the compounds tested and corresponding Ki values.
Compound
Labels
4L3AuCl
4L3AuTfl
4MeL3AuCl
2L3AuCl
4L2AuCl
2L2AuCl
L2AuCl
4LAuCl
2LAuCl
LAuCl
L
Ki (µM)
4.6 ± 0.5
6.33 ± 0.7
12.72 ± 2.1
21.26 ± 3.8
27.84 ± 12.12 *
2.066 ± 0.8 *
0.53 ± 0.03
2.25 ± 0.62 *
1.10 ± 0.3 *
1.214 ± 0.20 *
6.16 ± 0.77 *
No inhibition *
No inhibition *
[(4-COOH Ph)Ph2P]3AuCl
[(4-COOH Ph)Ph2P]3AuTfl
[(4-MeO-Ph)Ph2P]3AuCl
[(2-COOH Ph)Ph2P]3AuCl
(4COOHPh-Ph2P)2AuCl
(2COOHPh-Ph2P)2AuCl
(PPh3)2AuCl
(4COOHPh-Ph2P)AuCl
(2COOHPh-Ph2P)AuCl
PPh3AuCl
PPh3
4L
2L
4COOHPh-Ph2P
2COOHPh-Ph2P
* For these compounds, the Ki values were obtained in a previous work [15].
The inhibition action of the gold phosphane compounds was tested in regards to the reduction of
H2F to H4F reaction, catalyzed by DHFR [ ]. The inhibition tests were led by analyzing the reciprocal
2
reaction rate upon the addition of Hepes/methanol solutions of gold compounds in different set of
tests in the saturated concentration of NADPH. The tests were repeated with different amounts of
dihydrofolate (DHF) (1.5 ÷ 10 µM). Typical sets of tests are reported in Figure 1.
(B)
(A)
Figure 1. Dixon plot with the secondary plot in the inset for compound L2AuCl (A B).
) and 4L3AuCl (
The data were processed according to a linear regression (Equation (1)) obtaining the values
reported in Table 1. The standard deviations were calculated for a minimum of three determinations.
!
1
ν
Km
VmaxKi[S]
1
Km
[S]
=
∗ [I] +
1 +
(1)
M,
Vmax
The analysis of the data shows that the most active inhibitor is the L2AuCl with a Ki = 0.53
µ
followed by the LAuCl and other derivatives as shown in Table 1. Modification on the phosphane
ligand with a reactive group, such as the carboxylic moiety, does not improve the inhibition power,
which is further worsened by its conversion to the methyl ester derivative, [(4-MeO-Ph)Ph2P]3AuCl.
Notably, for the tris-phosphane gold compounds, the substitution of the chloride with the triflate ion
slightly worsens the inhibition property. The change of the position of COOH from para to ortho led to

Int. J. Mol. Sci. 2019, 20, 1802
5 of 17
an increase of the Ki values. Remarkably, by comparing the gold complexes Ki values with those of the
free ligands, the central role of the gold atom on the activation of the inhibition properties is visible.
2.3. Fluorimetric Studies
Before approaching the fluorimetric studies for the determination of the binding constants, the
emissive properties of the phosphane gold(I) complexes in Hepes/methanol solutions were assessed.
The fluorimetric data, both in the absence and in the presence of DHFR, are reported in Table 2.
The emission spectra of compounds 4L3AuCl, 4L3AuTfl, 4L2AuCl, 2L3AuCl, 2L2AuCl, and L2AuCl
were recorded in 20 mM Hepes/CH3OH solutions with concentrations of the metal complexes ranging
from 4 to 50 µM. The emission’s spectra of the tested gold(I) phosphane complexes are shown in
Figure 2. The most intense emissions show maxima at about 490 and 520 nm. These values are close
to those observed for the (PPh3)3AuCl and can be attributed to the presence of different geometries
around the gold center in solution [21]. The compound (PPh3)2AuCl, the emissive properties of which
have been largely assessed in literature [21,22], is a good reference for the evaluation of the effect
of the COOH group on the interaction with the enzyme. Most of the complexes showed intrinsic
emissions in the yellow–green visible range by excitation at 280 nm (which corresponds to the typical
absorbance of tryptophan’s in the enzyme), with a strong dependency of the intensity upon the increase
of the concentration of the gold(I) compound. The emissions of these solutions are not affected by
the presence of oxygen; moreover, according to the large Stokes’s shifts and the studies on halftimes
reported in literature for similar (PPh3)3AuCl compounds, a likely phosphorescence decay may be
attributed to these emissions. The intensity resulted in greater strength for the tris-coordinate geometry
rather than for the bis-coordinate compounds [21
,23]. In these PPh3-related compounds, the nature of
the emission was explained by a charge transfer from the metal to the ligand antibonding orbitals
π
*
←
d. Therefore, the introduction of the one COOH group slightly affects the emissive properties, and
for the ester ligand derivative, 4MeL3AuCl, the most intense emission was observed [24].
Table 2. List of fluorescence data for the gold(I) compounds once treated with DHFR (DiHydroFolate
Reductase) and DHFR/NADPH (Nicotinamide Adenine Dinucleotide Phosphate), with the corresponding
Kd values.
Intrinsic Emission λ
Kd ± s.d. * (25 ◦C) (µM);
em
Kd ± s.d. *, T = 25 ◦C (µM); n
= Hill Coefficient *
Compound
(λ = 280 nm)
Stokes Shift
In Presence of
exc
Hepes/methanol
Saturating NADPH
16,483 and 15,095
cm−1
3.84 ± 0.29; n = 1.73; 3.32 ±
4L3AuCl
4L3AuTfl
4MeL3AuCl
4L2AuCl
520 and 485 nm
5.05 ± 2.7
7.36 ± 0.84
4.55 ± 1.04
2.35 ± 0.61
1.28 (10 ◦C); 2.0 ± 0.36 (−4 ◦C)
21.4 ± 3.85; 7.36 ± 0.84 (10 ◦C);
16,106 cm−1
510 nm weak
8.08 ± 1.00 (−4 ◦C)
16,371 and 15,714
cm−1
13.4 ± 2.24; 7.63 ± 0.93 (10 ◦C);
21.2 ± 1.83 (−4 ◦C)
517 and 500 nm
4.42 ± 0.56; 3.52 ± 0.15 (10 ◦C);
No intrinsic emission
No intrinsic emission
n. d.
3.06 ± 0.87 (−4 ◦C)
10.47 ± 1.074; n = 1.23 ± 0.082;
7.12 ± 0.22 (10 ◦C); n = 1.41 ±
0.048; 5.11 ± 4.2 (−4 ◦C)
2L3AuCl
n. d.
n. d.
7.32 ± 2.46
2L2AuCl
No intrinsic emission
521 and 490 nm
10.33 ± 0.65
4.73 ± 0.68
2.22 ± 0.25
16,520 and 15,306
cm−1
5.99 ± 0.40; 2.47 ± 0.29 (10 ◦C);
(PPh3)2AuCl
3.83 ± 1.85 (−4 ◦C)
4COOHPh2P
Benzoic acid
N.A.
N.A.
n. d.
n. d.
59.6 ± 5.7
≈4000
N.A.
N.A.
* when not specified n = 1; s.d. means standard deviation.

Int. J. Mol. Sci. 2019, 20, 1802
6 of 17
Figure 2. Emission spectra recorded in Hepes/methanol 22.86
µ
M solutions of polyphosphane gold(I)
4
2
chloride compounds (
λ
= 280 nm). Compounds L2AuCl, 2L2AuCl, and L3AuCl show weak
exc
intrinsic emissions and their spectra falls in the baseline intensity.
2.4. Enzyme Binding Studies by Fluorimetric Assays
The Hepes/methanol solutions of the gold(I) compounds were used for the fluorimetric assays
(see materials and methods section) regarding DHFR from E. coli. The emission’s maximum of DHFR
was detected at 344 nm upon excitation at 280 nm; by adding additional amounts of the gold complex
in Hepes/methanol solutions (ranging among 4
÷
50 µM), the quenching of this emission intensity was
observed. In Figure 3, spectra showing the effect of 4L3AuCl on DHFR emission are reported. It is
possible to observe the quenching of the Tryptophan’s emission and the corresponding increase of the
intrinsic phosphorescence of the gold compound.
Figure 3. Emission spectra recorded for DHFR upon adding increasing concentration of 4L3AuCl (
λ
exc
= 280 nm). In the inset legend, the color corresponds to the curve and the respective concentrations.
The spectra show the quenching of the DHFR’s tryptophan emission upon the addition of the gold
compound and the concomitant increase of the emission’s intensity, attributed to the gold compound.

Int. J. Mol. Sci. 2019, 20, 1802
7 of 17
The hyperchromic effect is visible in Figure 4, where the overlaid spectra obtained for the 4L3AuCl
in the presence and in the absence of DHFR are reported.
Figure 4. Comparison of the emission curves recorded for the intrinsic emission of 4L3AuCl in
Hepes/methanol solution in the absence and in the presence of DHFR (4L3AuCl 22.86
µM in both
spectra, λ = 280 nm).
exc
The analysis of the emission’s quenching data was processed according to methods discussed
4
in the materials and methods section. Upon treatment of DHFR with L3AuCl, the Kd value of
3.84 ± 0.29 µM was obtained. This value was compared with that obtained for 4L2AuCl (Figure 5).
The emission spectrum shows a very weak emission at 511 nm, while the quenching effect is still
observable. The Kd value for the 4L2AuCl with DHFR adduct is 4.42
analysis are reported in Figure 6.
±
0.56 µM. The plots for the Kd
Figure 5. DHFR’s tryptophan emission quenching upon stepwise additions of 4L2AuCl (
In the inset legend, the color corresponds to the curve and the respective concentrations.
λexc = 280 nm).

Int. J. Mol. Sci. 2019, 20, 1802
8 of 17
Figure 6. Fluorescence intensity values against added concentration of 4L3AuCl (left) and 4L2AuCl
(right) at 25 ◦C.
To investigate the nature of the binding between the quencher-Au, DHFR, the free
4-benzoic-diphenyl-phosphane acid ligand, i.e., 4-COOHPh2P, and the benzoic acid were assayed
in the same conditions. The Kd values are reported in Table 2 and the spectra are reported in the
supplementary material section (Figure S3). These compounds, which are a structural part of the
4L3AuCl, show looser or negligible static interactions than those observed with gold compounds.
These results might be explained in term of accessibility to the tryptophan of the polar benzoic acid on
comparison with the lipophilic phosphane ligand, 4L [25].
With the same aim, the binding of (PPh3)2AuCl (labeled as L2AuCl) in regard to DHFR was
evaluated in terms of the binding interaction (Figure 7).
Figure 7. Quenching of DHFR’s tryptophan emissions according to the stepwise addition of (PPh3)2AuCl
(λexc = 280 nm). In the inset legend, the color corresponds to the curve and the respective concentrations.
The substitution of the chloride with triflate or the esterification of the carboxylic group results in
an increase of the Kd values, indicating less affinity of the gold complex for the enzyme than what was
observed for the 4L3AuCl compound (see Table 2).
After these preliminary emissive property characterizations, the fluorimetric studies were led in
the presence of the enzyme DHFR and the cofactor NADPH [26]. DHFR possesses five amino-acid
tryptophan residues and four tyrosine residues: both aromatic residues show intense emissive
properties, and the enzyme’s tryptophan, if excited at 280 nm, exhibits an emission at 340–350 nm [18
,27].
Additionally, NADPH is fluorescent, with an absorption maximum at about 340 nm and an emission

Int. J. Mol. Sci. 2019, 20, 1802
9 of 17
maximum at about 450 nm [18]. In the presence of NADPH, the enzyme’s emission intensity is then
quenched by 65%, indicating that tryptophan’s environment in the enzyme is significantly altered
by the binding of cofactor. Moreover, because NADPH has a maximum of absorption at 340 nm,
significant energy transfer from the excited tryptophan of the enzyme to bound NADPH occurs; as a
consequence, a new emission at 420–450 nm is then observed [28]. Figure 8 reports the progressive
quenching of the DHFR and NADPH emissions by stepwise adding of solution of 4L3AuCl. Moreover,
a slight shift of the typical emission at 510 nm (λ = 280 nm) is observed.
exc
Figure 8. Emission spectra recorded upon the addition of Hepes/methanol solution of 4L3AuCl into
the DHFR solution saturated with NADPH (
λ
= 280 nm). In the inset legend, the color corresponds
exc
to the curve and the respective concentration.
To give an insight into the emission spectra shown in Figure 8, the fluorescence spectra were also
recorded for the solution of NADPH upon the addition of the gold complexes in absence of the DHFR
(data reported in Table 2, rightmost column).
In Figure 9, the emissions recorded upon the stepwise addition of solution of 4L3AuCl to NADPH
◦
solution at 25 C are reported (
λ
exc
= 280 nm). The spectra show a modification of the emissions
attributed to NADPH, with a likely quenching either of the nicotinamide or of the adenine moieties,
while the hyperchromic effect of the emission due to the gold complex is not observed. This modification
of the emission maxima might be diagnostic of an interaction of the 4L3AuCl to NADPH, which is also
strongly interactive with the DHFR. An interaction between NADPH and 4L3AuCl is also revealed by
the emission spectrum of the mixture of the two molecules in comparison with the individual spectra,
reported in Figure 10: the cumulative spectrum of NADPH and 4L3AuCl is not the resulting sum of
the emission intensities of the two compounds analyzed separately.

Int. J. Mol. Sci. 2019, 20, 1802
10 of 17
Figure 9. Emission spectra of 50 µ
M solution of NADPH upon adding an increased amount of 4L3AuCl
(λ = 280 nm).
exc
Figure 10. Comparison of the emission spectra recorded for NADPH 50
µM (black line), and the mixture of them (red line).
µ
M (blue line), 4L3AuCl 22.86
As can be seen from Table 2, most of the Kd values have been obtained at different temperatures;
◦
25, 10, and
−
4 C. In general, the trend of the Kd values results in a lowering upon cooling, even
though some discrepancies are observed mainly at
is sometime critical. In Table 3, the fluorimetric data processed according to the vant’Hoff equation and
the standard free energy equation
G◦ = G◦ S◦ are reported. In all the cases, negative values of
−
4 ◦C, where the solubility of the gold compounds
∆
∆
−
T∆
∆
G◦ remark the spontaneous event of the binding, with a prevalent negative enthalpy contribution,
apart from the compound 4L2AuCl, for which the entropic energy prevails. Notably, this compound
shows an anomaly in the trend of the Ki compared to other analogs.

Int. J. Mol. Sci. 2019, 20, 1802
11 of 17
Table 3. Thermodynamic data for tris and bisphosphane gold complexes.
Compounds
4L3AuCl
4L3AuTfl
4MeL3AuCl
2L3AuCl
4L2AuCl
L2AuCl
∆H◦ (kcal/mol)
−5.04 ± 0.08
−5.32 ± 3.2
−4.35 ± n. d. *
−3.94 ± 0.1
∆S◦ (cal/mol K)
7.34 ± 0.005
3.92 ± 8.8
7.69 ± n. d. *
9.58 ± 0.08
T∆S◦ (kcal/mol)
2.19 ± 0.1
1.17 ± 0.9
2.29 ± n. d. *
2.86 ± 0.1
∆G◦ (kcal/mol)
−7.23 ± 0.12
−4,15 ± 3.3
−6.64 ± n. d. *
−6.8 ± 0.14
−2.02 ± 0.2
−4.87 ± n. d.*
17.76 ± 0.1
8.42 ± n. d. *
5.29 ± 0.2
2.5 ± n. d. *
−7.31 ± 0.28
−7.37 ± n. d. *
* n. d. = not determined. Data are available only at two temperatures (10 and 25 ◦C) because of loss of solubility.
2.5. Spectrometric Studies by ESI-MS
Some additional studies have been led by using the ESI Mass Spectrometry [29]. In this regard,
the DHFR enzyme has already been the object of several studies mainly addressed to the structural
characterization [30] or to the binding of the metal to the enzyme’s fragment [31]. At first, buffered solutions
of 4L2AuCl or 4L3AuCl and H2F in the presence or in the absence of NADPH were analyzed by ESI-MS to
rule out the possibility of having the reduction of DHF to THF activated by gold compounds, the catalytic
actions of which, in regards to many reactions, are already known [32,33]. From the analysis of the peaks
and the comparison with standards of H2F and H4F, we could exclude this hypothesis. However, additional
spectra were recorded on buffered solutions containing NADPH and 4L2AuCl or 4L3AuCl to analyze the
interaction between the cofactor and the gold complex. In the former case, in the range 250–800 m/z, peaks
due to [4L]− (304.8 m/z), [4L2Au]2− (403 m/z) and [4LAuCH3COO]− (560.8 m/z) were detected; other peaks
were attributed to NADPH, such as [NADPH]2− at 371 m/z. The most characteristic peaks were found in
the range 600–1000 m/z. In particular, in addition to the peak of [NADPH]− at 743.9 m/z, a quite intense
peak at 765.8 m/z might be attributed to the doubly charged ion [2NADPH + 4L2Au
peak, a reaction between the carboxylic group and the amine function of the adenine moiety with the loss of
water might be speculated. The spectrum is reported in Figure 11.
−
H2O]2−. For this
Figure 11. ESI-MS spectrum in the range 700–830 m/z for negative ions recorded for NADPH in ammonium
acetate and 4L2AuCl upon incubation of 5 min. Magnification of the peak at 765.8 m/z for [2NADPH +
4L2Au − H2O]2− (top, right image) and its simulation with IsoPro3.0 (bottom, right image).
3. Materials and Methods
3.1. Syntheses
The preparation of the gold compounds was performed according to methods already published [14,15]
and was based on the substitution of the labile ligand from the gold precursor [AuCl(Me2S)], by mean
of a more coordinative ligand, such as the PPh3 or its modified analogs, 4L and 2L. Mole ratio with an
excess of the phosphane ligand was pursued with the aim to enhance the reaction yield for tris-phosphane
compounds. The solids obtained by crystallization were characterized by elemental analysis, 1H and 31
P
NMR, IR, and ESI Mass spectrometry. The stability of the stock solutions upon dilution was checked by

Int. J. Mol. Sci. 2019, 20, 1802
12 of 17
recording UV-visible spectra (see Supplementary Material, Figures S1 and S2). The reactants for the ester
reaction and 4L or 2L ligands were bought from Aldrich and used without any purification. Methanol or
methylene chloride solvents were degassed and dried over molecular sieves.
3.1.1. Synthesis of 4-(diphenylphosphanyl)benzoic acid methyl ester, 4Me
L
The title compound was prepared according to the procedure described by Yoakim.29 To a 15 mL
solution of 4-(diphenylphosphino)benzoic acid (398 mg; 1.30 mmol) in CH2Cl2 anhydrous, methanol
(0.05 mL; 1.4 mmol) and N,N-dimethylaminopyridine (37.0 mg; 0.30 mmol) were added. A solution
of N,N-dicyclohexylcarbodiimide (300 mg; 1.41 mmol) in CH2Cl2 was added over 15 min at 0 ◦C. The
resulting pale yellow suspension was stirred for 16 h at room temperature. The suspension was filtered
and dried at reduced pressure. The solid was dissolved with EtOAc (30 mL), washed successively with
aqueous 10% HCl (15 mL), saturated NaHCO3 (15 mL), and brine (15 mL). The organic fraction was dried
over MgSO4, filtered, and evaporated to dryness. The solid was dissolved in EtOAc and purified with flash
chromatography (5% EtOAc/hexane). The product was crystallized as a white solid, with cyclohexane.
Yield: 60%.
1H NMR (CDCl3, δ): 8.05 (d, 2H); 7.35–7.40 (m, 12 H); 3.95 (s, 3H).
31P NMR (CDCl3, δ): −5.02 (s).
MIR (cm−1): 3067 (w), 2999 (w), 2952 (w), 2924 (w), 2852 (w), 2102 (w), 1943 (w), 1822 (w), 1719 (vs),
1597 (m, sh), 1584 (sh), 1561 (w), 1476 (m), 1433 (s), 1393 (m), 1278 (vs), 1181 (m), 1115 (m), 1088 (s),
1017 (s), 999 (sh), 965 (m), 857 (m), 750 (m), 743 (vs).
3.1.2. Synthesis of [tris(4-benzoate-diphenylphosphanyl)-gold(I)chloride], 4L3AuCl
The solution of 4-(diphenylphosphine)benzoic acid (208.3 mg; 0.4 mmol) in CHCl3; was added
dropwise to the solution of (dimethylsulfide)gold(I)chloride (30 mg; 0.1 mmol) in CHCl3; (3 mL) at 0
◦C. The mixture was stirred at r. t. overnight. A pale yellow precipitate was obtained. Upon filtration,
the solid was recovered and dissolved with MeOH and filtered again over a paper filter. The product
was crystallized, as pale yellow solid, from a mixture of CHCl3; and cyclohexane.
Yield: 86%. M. p. 139–140 ◦C.
Analyzing the product:
1H-NMR (CD3OD, δ): 7.81 (d, 6H); 7.50 (t, 6H), 7.26 (m, 30H).
31P-NMR (CD3OD, δ): 42.09 (s).
MIR (cm−1): 3484 (s, br), 3054 (m, br), 1702.9 (vs), 1597.9 (s), 1560.2 (s), 1481 (m), 1434 (s), 1393 (s),
1310.9 (m), 1275 (m, sh), 1225 (vs), 1183 (s), 1122.6 (m), 1091.8 (s), 1025.9 (m), 1016.8 (s), 998.9 (m), 855
(m–w), 791.9 (m), 764 (s, sh), 742.8 (s), 689.9 (vs).
FIR (cm−1): 534 (w), 513.5 (m), 367.8 (w), 295.68 (w), 176.95 (m), 139 (m), 125.7 (m), 106 (w), 58.43 (m),
39.8 (m).
ESI (
) (CH3OH, m/z): 305.2 (100), 536.9 (40). ESI (+) (CH3OH, m/z): 808.9 (100) [(4-COOH-Ph2P)2Au] +.
−
Elemental analysis for C57H51AuClO9P3, calculated%: C 56.80, H 4.26. Found C 56.86, H 4.13.
3.1.3. Synthesis of [tris(4-benzoate-diphenylphosphanyl)-gold(I)triflate], 4L3AuTfl
A solution of silver trifuoromethanesulfonate (87 mg, 0.34 mmol) was dissolved in 1 mL of dry THF
and then added to a solution of (dimethylsulfide)gold(I)chloride (100 mg; 0.34 mmol) in 12 mL of THF.
A precipitate of AgCl was readily formed. The mixture was rapidly filtered over celite, under nitrogen
flux. The clean filtrate was transferred in another flask containing the 4-(diphenylphosphino)benzoic
acid in THF (2 mL). The mixture was stirred at r. t. for 3h. After evaporation to dryness, a white pale
yellow compound was obtained. Then, it was washed with hexane and diethyl ether and crystallized
from chloroform and hexane.
Yield: 85%. M. p. 153–154 ◦C.
1H-NMR (Acetone-d6, δ) 7.81 (d, J = 7.6 Hz, 6H); 7.48 (m, 6H); 7.28 (m, 30H).
31P-NMR (Acetone-d6, δ, r. t.) 32.61 (s), 22.59 (s).

Int. J. Mol. Sci. 2019, 20, 1802
13 of 17
MIR (cm−1): 3053 (m, br), 2546 (m, br), 1693 (vs), 1597.7 (m), 1560 (m), 1482 (m), 1435 (vs), 1414 (m, sh),
1395 (m), 1315 (m), 1290 (s), 1221 (vs), 1179 (m), 1157 (s,sh), 1095 (vs), 1027 (s, sh), 1017 (m), 999 (m),
926 (m, br), 854 (m), 793.3 (w), 743 (m), 690 (vs).
ESI (−) (CH3OH, m/z): 305.2 (100). ESI (+) (CH3OH, m/z): 808.9 (100) [(4-COOH-Ph2P)2Au] +.
Elemental analysis for C58H45AuF3O9P3S + 3H2O, calculated%: C 56.20; H 3.69, S 2.06. Found%: C
56.07; H 3.59, S 2.53.
3.1.4. Synthesis of [tris(2-benzoate-diphenylphosphanyl)-gold(I)chloride], 2L3AuCl
The solution of 2-(diphenylphosphine) benzoic acid (208.3 mg; 0.4 mmol) in CHCl3; was added
dropwise to the solution of (dimethylsulfide)gold(I)chloride (30 mg; 0.1 mmol) in CHCl3; (3 mL) at 0
◦C. The mixture was stirred at r.t overnight. A pale yellow precipitate was obtained. The compound
was recovered upon filtration, completely dissolved with MeOH, and evaporated. The product was
crystallized, as pale yellow solid, from a mixture of CHCl3; and cyclohexane.
Yield: 88%.
1H-NMR (CD3OD, δ): 8.51 (d, 3H); 7.73 (t, 6H), 7.45 (dt, 6H), 7.06 (m, 30H).
31P-NMR (CD3OD, δ): 47.09 (s).
IR (cm−1): 3054 (w), 1699 (vs), 1596 (w), 1584 (w), 1565 (w), 1471 (w), 1433 (vs), 1388 (vs), 1265 (w),
1222 (vs), 1179 (w), 1114 (s), 1091 (s), 1027, 999, 799 (s), 742 (vs), 688 (vs), 642 (vs), 522 (s), 506 (s), 446
(m), 432 (m).
ESI (
) (CH3OH, m/z): 305.2 (100), 536.9 (40). ESI (+) (CH3OH, m/z): 808.9 (100) [(2-COOH-Ph2P)2Au]+.
−
Elemental analysis for C57H45AuClO6P3, calculated%: C 61.35, H 4.06. Found C 60.98, H 4.23.
3.1.5. Synthesis of [tris(4-methylbenzoate)-diphenylphosphanyl)-gold(I)chloride], 4MeL3AuCl
The solution of methyl 4-(diphenylphosphino)benzoate (64 mg; 0.2 mmol) in anhydrous CHCl3;
(2 mL) was added dropwise to the solution of (dimethylsulfide)gold(I)chloride (15 mg; 0.05 mmol) in
CHCl3; (3 mL) at 0 ◦C. The mixture was stirred at r.t. overnight, then filtered and dried at reduced
pressure. The pale yellow solid was washed in hexane to remove the excess of dimethylsulfide and
crystallized by CHCl3 solutions.
Yield: 67%.
1H-NMR (CDCl3, δ): 7.97 (d, 6H); 7.69 (d, 6H), 7.29–7.44 (m, 30H), 3.95 (s, 9H).
31P-NMR (CDCl3, δ): 30.57 (s).
MIR (cm−1): 3052 (w), 3008 (w), 2953 (w), 2924 (w), 2852 (w), 2103 (w, br), 1980 (w, br), 1812 (w), 1718
(vs), 1597 (m), 1559 (sh), 1560 (w), 1480 (m), 1434 (s), 1394 (m), 1273 (vs), 1186 (m), 1110 (s), 1089 (s),
1018 (s), 999 (sh), 963 (w), 952 (m), 804 (m), 743 (vs).
ESI (+) (CH3OH, m/z): 517 m/z (30) [(4-COOCH3-Ph2P)Au]+; 838 (100) [(4-COOCH3-Ph2P)2Au]+.
Elemental analysis for C60H51AuClO6P3, calculated%: C 60.39, H 4.31. Found C 60.77, H 4.76.
3.2. Enzyme Inhibition Studies
3.2.1. Preparation of the NADPH Stock
3.2 mg of NADPH was dissolved in 1 mL of Hepes (4-(2-hydroxyethyl)-1-piperazine-
1-ethanesulfonate) 50 mM buffer solution at pH = 7.3. The concentration of NADPH solution
was determined by UV-Vis spectroscopy, using the buffer solution as a blank, and diluting the stock
solution 1: 50 (20 mL of NADPH stock in 1 mL solution). The absorbance was recorded at 340 nm (for
NADPH ε340 = 6220 M−1·cm−1).
3.2.2. Preparation of the H2F Stock
1 mg of H2F stock was dissolved in 600 µL of HEPES buffer 50 mM and 38 µL of β-mercapto-ethanol.
The KOH 0.01 M was used as the blank and to dilute the stock solution to 1:200. The absorbance was
read at 283 nm (for H2F ε283 = 28,000 cm−1).

Int. J. Mol. Sci. 2019, 20, 1802
14 of 17
3.2.3. Preparation of the DHFR Stock
The overexpressed and purified DHFR from Escherichia coli was stored at 4 ◦C and precipitated
in a 90% (NH4)2SO4 solution. The dialysis of DHFR was performed in two steps with a 10 mM
K2HPO4-KH2PO4 buffer, pH =7.3, the first in 3 h and the second overnight. The concentration of
DHFR was determined by recording the absorbance of the stock solution at 280 nm (E. coli DHFR ε280
= 31,100 M−1 cm−1) using the potassium phosphate buffer as the blank.
3.2.4. Preparation of the Gold(I) Compounds Stocks
5 mg of gold(I) compounds were dissolved in 5 mL of methanol, affording approximately 1 µM
solutions. The concentration of the gold(I) compounds were calculated knowing the molar masses.
3.3. Inhibition Studies on DHFR
The activity of the DHFR was followed by recording the absorbance with a Shimadzu UV-2450 (UV-Vis
spectrophotometer). The enzyme assays were made in a quartz cuvette by using similar experimental
conditions: DHFR 3-5 nM, NADPH 60–80
preparation of the following mixtures: 800–971
10 l of DHFR, and increasing quantities of gold(I) compounds stock solutions in order to achieve higher
accuracy in the measurement of the inhibitory effect. After storing the mixture for 5 min at 30 ◦C, 4–33
of H2F was added. The decreasing absorbance at 340 nm, due to the oxidation of NADPH to NADP+, was
detected after adding the H2F, and expressed in U mL−1 using the equation: U mL−1
abs/ t/11.8
µM, and H2F 1.5, 4, and 10
µM. The procedure consisted in the
µL of 50 mM Hepes buffer at pH = 7.3, 20
µL of NADPH,
µ
µL
·
=
∆
∆
×
dilution factor, where 11.8 is the mMolar extinction coefficient when NADPH and H2F are simultaneously
present in solution. The Ki values were calculated with the Dixon plot from the second plot obtained
computing the slope values against the reciprocal of the H2F concentrations.
3.4. Fluorimetric Analysis
The fluorimetric analysis was performed using a 1 mL quartz cuvette and a Perkin–Elmer
Fluorescence Spectrometer LS 55. For all the analysis we used 5–12
increasing concentrations of gold(I) compounds in the 4–60 M range. For some experiments,
saturating concentrations of NADPH (50 M) were added to the initial solutions of DHFR to observe
µM solutions of DHFR and
µ
µ
the affinity variation of the complexes with respect to the DHFR.
To observe the intrinsic emission of the gold(I) complexes, and eventually their hyperchromic
effect, emission spectra were recorded in the same range of concentration used for the quenching
studies (4–60 µM).
All the emission spectra have been interpolated graphically (Origin Lab® 2019) [34] in order to
correct for the second order Rayleigh scattering band, centered at 560 nm.
All the spectra were recorded in the range between 250 and 600 nm using an excitation wavelength
characteristic of the Tryptophan, equal to 280 nm. For most of the experiments the temperature was
kept constant at 25 ◦C; furthermore, to analyze the thermodynamic aspects of the reactions (via the
van’t Hoff equation), quenching spectra of DHFR in presence of the gold(I) complexes were recorded
at −4 and 10 ◦C as well.
The fluorimetric data were analyzed using the binding isotherm (Equation (2)), where [Q] is
the concentration of the added quenching ligand, F is the measured fluorescence, F0 is the starting
fluorescence, Fc is the fluorescence of the fully complexed protein, Kd is the dissociation constant, and
n is the Hill coefficient. Indeed, n is a phenomenological parameter, an interaction coefficient, such
as when n = 1 the equation describes a complex formation with one site of interaction, when n = 2
binding occurs with a dimer, when 1 < n < 2 a cooperative interaction occurs [35].
[Q]n
(Kd + [Q])n
F0 − F
=
(2)
F − Fc

Int. J. Mol. Sci. 2019, 20, 1802
15 of 17
3.5. ESI-MS Spectrometry
The instrument used was a HPLC 1100 Agilent Technologies connected to a mass spectrometer
Agilent Technologies (LC/MSD Trap SL) equipped with an electrospray ion source (ESI). The analyses
have been performed in FIA (flow injection analysis) mode. The mobile phase was H2O: MeOH 90:10
at 300
µ
L/min, with the addition of 0.1% formic acid. The ionization source paramet◦ers were the
following: nebulizer pressure 50 psi, gas flow 10 L/min, drying gas temperature 350 C, capillary
voltage 3500 Volts, injection volume 2 µL.
The solutions of 80
in ammonium acetate 10 mM pH 7.3 and by adding 100
µM NADPH and 100
µ
M 4L2AuCl have been prepared by dissolving NADPH
L of a 1 mM stock solution of the gold
µ
compound dissolved in methanol, to a final volume of 1 mL.
4. Conclusions
In this work, a class of polyphosphane gold(I) complexes have been assayed as DHFR from E. coli
inhibitors and a study about the mechanism of their interaction has been led. For this set of phosphane
gold compounds, the strongest inhibition was detected for (PPh3)2AuCl with a value of Ki = 0.53
and the same compound shows a remarkable affinity for DHFR with a thermodynamic constant for
the dissociation of the adduct [DHFR:(PPh3)2AuCl] equal to about 2 M. The presence of the COOH
group in the phosphane ligand affords to compounds 4L2AuCl or 2L2AuCl, and it does not introduce a
better ability to inhibit the reduction of H2F to H4F by the DHFR (Ki = 27 and 2 M, respectively) and
not even to bind the enzyme. Moreover, the conversion of the COOH into the methyl esther function
in the ligand does not introduce big changes in the inhibition (Ki 13 M), nor in the affinity to the
DHFR (Kd 12.7 M). Nevertheless, the presence of gold atom plays a central role in both inhibitory
µM,
µ
µ
≈
µ
≈
µ
and enzyme-binding activities. The affinity of gold compounds for DHFR in general is enhanced by
the presence of NADPH, which yields the interaction to be more specific and behaves as a positive
allosteric modulator with respect to all the tested gold compounds. The NADPH cooperative effect on
gold compounds binding by DHFR is consistent with its role in the enzyme catalytic mechanism and
with the behavior observed for other DHFR inhibitors [36].
Indications about an additional action of NADPH by binding the 4L2AuCl has been also evidenced
by mass spectrometry, even though the investigation was not exhaustive, the formation of an adduct
between NADPH and the gold compound has been observed. In conclusion, the compounds analyzed in
this paper show a reversible inhibitory effect that foresees a reversible bond, modulated by the presence
of the cofactor, affected by the nature of the phosphane ligands and, unquestionably, by the presence
of the gold atom. The reversible bonding is also evidenced by the hyperchromic effect on the intrinsic
emission of the phosphane gold compound and by the shift of the emission maxima, due to both the
stabilization of the excited state and the binding to the enzyme, respectively. The effect of the temperature
on the Kd values (Table 3) characterizes the binding, attributing a modest exothermic contribute in spite
of the entropic one. These results may be discussed by considering previous studies on the interaction
of gold phosphane compounds and target enzymes, such as Thioredoxine reductase. For this latter,
nanomolar IC50 values have been found [13], but for this enzyme, the formation of covalent bonds with
cysteine or selenocysteine residues has been largely hypothesized [17,31]. In this work, from the kinetic
inhibition and thermodynamic data, we hypothesize a reversible bond, with the enzyme DHFR having
both an electrostatic and hydrophobic nature with an interplay mechanism involving the cofactor NADPH,
depending on the type of the gold compound. Based on these results, we can infer that enzymes in which
the catalytic site does not contain either cysteines or selenocysteines might interact with gold complexes
through non-covalent interactions, endowing Ki values in the micromolar order of magnitude.
Supplementary Materials: Supplementary materials can be found at http://www.mdpi.com/1422-0067/20/7/1802/s1.
Author Contributions: R.G. and S.P. equally contributed to this work. S.V. provided partial financial coverage.
O.C.S. and A.T.R. prepared the gold compounds and put the basis for this study. M.R. took care about the ESI
mass spectra acquisitions in the proper operative conditions. L.L. performed and refined the spectroscopic studies
and took part to the elaboration of the data.

Int. J. Mol. Sci. 2019, 20, 1802
16 of 17
Funding: This research received no external funding.
Acknowledgments: R.G. and S.P. are grateful to CIRCMSB (Consorzio Interuniversitario Ricerca sulla Chimica
dei Metalli e le Sostanze Biologiche) and to FAR of the University of Camerino (Fondi di Ateneo per la Ricerca).
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
Cannon, W.R.; Garrison, B.J.; Benkovic, S.J. Electrostatic Characterization of Enzyme Complexes: Evaluation
of the Mechanism of Catalysis of Dihydrofolate Reductase. J. Am. Chem. Soc. 1997, 119, 2386–2395. [CrossRef]
Schnell, J.R.; Dyson, H.J.; Wright, P.E. Structure, Dynamics, and Catalytic Function of Dihydrofolate Reductase.
Annu. Rev. Biophys. Biomol. Struct. 2004, 33, 119–140. [CrossRef] [PubMed]
Schweitzer, B.I.; Dicker, A.P.; Bertino, J.R. Dihydrofolate Reductase as a Therapeutic Target. FASEB J. 1990, 4,
2441–2452. [CrossRef]
Hawser, S.; Lociuro, S.; Islam, K. Dihydrofolate Reductase Inhibitors as Antibacterial Agents. Biochem.
Pharmacol. 2006, 71, 941–948. [CrossRef] [PubMed]
5.
6.
Sköld, O. Resistance to Trimethoprim and Sulfonamides. Vet. Res. 2001, 32, 261–273. [CrossRef] [PubMed]
Huovinen, P.; Sundström, L.; Swedberg, G.; Sköld, O. Trimethoprim and Sulfonamide Resistance. Antimicrob.
Agents Chemother. 1995, 39, 279–289. [CrossRef]
7.
8.
9.
Matthews, D.; Alden, R.; Bolin, J.; Freer, S.; Hamlin, R.; Xuong, N.; Kraut, J.; Poe, M.; Williams, M.;
Hoogsteen, K. Dihydrofolate Reductase: X-Ray Structure of the Binary Complex with Methotrexate. Science
1977, 197, 452–455. [CrossRef] [PubMed]
Rajagopalan, P.T.R.; Zhang, Z.; McCourt, L.; Dwyer, M.; Benkovic, S.J.; Hammes, G.G. Interaction of
Dihydrofolate Reductase with Methotrexate: Ensemble and Single-Molecule Kinetics. Proc. Natl. Acad. Sci.
USA 2002, 99, 13481–13486. [CrossRef]
Cammarata, M.; Thyer, R.; Lombardo, M.; Anderson, A.; Wright, D.; Ellington, A.; Brodbelt, J.S.
Characterization of Trimethoprim Resistant E. Coli Dihydrofolate Reductase Mutants by Mass Spectrometry
and Inhibition by Propargyl-Linked Antifolates. Chem. Sci. 2017, 8, 4062–4072. [CrossRef] [PubMed]
10. Banerjee, D.; Mayer-Kuckuk, P.; Capiaux, G.; Budak-Alpdogan, T.; Gorlick, R.; Bertino, J.R. Novel Aspects of
Resistance to Drugs Targeted to Dihydrofolate Reductase and Thymidylate Synthase. Biochim. et Biophys.
Acta 2002, 1587, 164–173. [CrossRef]
11. Kuyper, L.F.; Roth, B.; Baccanari, D.P.; Ferone, R.; Beddell, C.R.; Champness, J.N.; Stammers, D.K.;
Dann, J.G.; Norrington, F.E.A. Receptor-Based Design of Dihydrofolate Reductase Inhibitors: Comparison of
Crystallographically Determined Enzyme Binding with Enzyme Affinity in a Series of Carboxy-Substituted
Trimethoprim Analogs. J. Med. Chem. 1985, 28, 303–311. [CrossRef]
12. Singh, A.; Deshpande, N.; Pramanik, N.; Jhunjhunwala, S.; Rangarajan, A.; Atreya, H.S. Optimized Peptide Based
Inhibitors Targeting the Dihydrofolate Reductase Pathway in Cancer. Sci. Rep. 2018, 8, 3190. [CrossRef] [PubMed]
13. Galassi, R.; Burini, A.; Ricci, S.; Pellei, M.; Rigobello, M.P.; Citta, A.; Dolmella, A.; Gandin, V.; Marzano, C.
Synthesis and Characterization of Azolate Gold(I) Phosphane Complexes as Thioredoxin Reductase Inhibiting
Antitumor Agents. Dalton Trans. 2012, 41, 5307–5318. [CrossRef] [PubMed]
14. Gambini, V.; Tilio, M.; Maina, E.W.; Andreani, C.; Bartolacci, C.; Wang, J.; Iezzi, M.; Ferraro, S.; Ramadori, A.T.;
Simon, O.C.; et al. In Vitro and in Vivo Studies of Gold(I) Azolate/Phosphane Complexes for the Treatment
of Basal like Breast Cancer. Eur. J. Med. Chem. 2018, 155, 418–427. [CrossRef] [PubMed]
15. Galassi, R.; Oumarou, C.S.; Burini, A.; Dolmella, A.; Micozzi, D.; Vincenzetti, S.; Pucciarelli, S. A
Study on the Inhibition of Dihydrofolate Reductase (DHFR) from Escherichia Coli by Gold(I) Phosphane
Compounds. X-Ray Crystal Structures of (4,5-Dichloro-1H-Imidazolate-1-Yl)-Triphenylphosphane-Gold(I)
and (4,5-Dicyano-1H-Imidazolate-1-Yl)-Triphenylphosphane-Gold(I). Dalton Trans. 2015, 44, 3043–3056.
[CrossRef] [PubMed]
16. Carlos Lima, J.; Rodriguez, L. Phosphine-Gold(I) Compounds as Anticancer Agents: General Description
and Mechanisms of Action. Anti-Cancer Agents Med. Chem. 2011, 11, 921–928. [CrossRef]
17. Gandin, V.; Fernandes, A.P.; Rigobello, M.P.; Dani, B.; Sorrentino, F.; Tisato, F.; Björnstedt, M.; Bindoli, A.;
Sturaro, A.; Rella, R.; et al. Cancer Cell Death Induced by Phosphine Gold(I) Compounds Targeting
Thioredoxin Reductase. Biochem. Pharmacol. 2009, 79, 90. [CrossRef]

Int. J. Mol. Sci. 2019, 20, 1802
17 of 17
18. Degan, P.; Carpano, P.; Cercignani, G.; Montagnoli, G. A Fluorescence Study of Substrate and Inhibitor
Binding to Bovine Liver Dihydrofolate Reductase. Int. J. Biochem. 1989, 21, 291–295. [CrossRef]
19. Fierke, C.A.; Johnson, K.A.; Benkovic, S.J. Construction and Evaluation of the Kinetic Scheme Associated
with Dihydrofolate Reductase from Escherichia Coli. Biochemistry 1987, 26, 4085–4092. [CrossRef]
20. Smith, V.F.; Matthews, C.R. Testing the Role of Chain Connectivity on the Stability and Structure of
Dihydrofolate Reductase from E. Coli: Fragment Complementation and Circular Permutation Reveal Stable,
Alternatively Folded Forms. Protein Sci. 2001, 10, 116–128. [CrossRef] [PubMed]
21. King, C.; Khan, M.N.I.; Staples, R.J.; Fackler, J.P. Luminescent Mononuclear Gold(I) Phosphines. Inorg. Chem.
1992, 31, 3236–3238. [CrossRef]
22. Hoshino, M.; Uekusa, H.; Ishii, S.; Otsuka, T.; Kaizu, Y.; Ozawa, Y.; Toriumi, K. Polymorphic Crystal Approach
to Changing the Emission of [AuCl(PPh3)2], Analyzed by Direct Observation of the Photoexcited Structures
by X-Ray Photocrystallography. Inorg. Chem. 2010, 49, 7257–7265. [CrossRef]
23. Assefa, Z.; McBurnett, B.G.; Staples, R.J.; Fackler, J.P.; Assmann, B.; Angermaier, K.; Schmidbaur, H. Syntheses,
Structures, and Spectroscopic Properties of Gold(I) Complexes of 1,3,5-Triaza-7-Phosphaadamantane (TPA).
Correlation of the Supramolecular Au . . . ..Au Interaction and Photoluminescence for the Species (TPA)AuCl
and [(TPA-HCl)AuCl]. Inorg. Chem. 1995, 34, 75–83. [CrossRef]
24. Ziolo, R.F.; Lipton, S.; Dori, Z. The Photoluminescence of Phosphine Complexes of d10 Metals. J. Chem. Soc.
D 1970. [CrossRef]
25. Lee, B.; Richards, F.M. The Interpretation of Protein Structures: Estimation of Static Accessibility. J. Mol. Biol.
1971, 55. [CrossRef]
26. Talukder, P.; Chen, S.; Roy, B.; Yakovchuk, P.; Spiering, M.M.; Alam, M.P.; Madathil, M.M.; Bhattacharya, C.;
Benkovic, S.J.; Hecht, S.M. Cyanotryptophans as Novel Fluorescent Probes for Studying Protein
Conformational Changes and DNA–Protein Interaction. Biochemistry 2015, 54, 7457–7469. [CrossRef]
27. Ohmae, E.; Sasaki, Y.; Gekko, K. Effects of Five-Tryptophan Mutations on Structure, Stability and Function of
Escherichia Coli Dihydrofolate Reductase. J. Biochem. 2001, 130, 439–447. [CrossRef]
28. Available online: https://www.sciencedirect.com/topics/neuroscience/dihydrofolate-reductase (accessed on
27 November 2018).
29. Boeri Erba, E.; Petosa, C. The Emerging Role of Native Mass Spectrometry in Characterizing the Structure
and Dynamics of Macromolecular Complexes. Protein Sci 2015, 24, 1176–1192. [CrossRef]
30. Yamamoto, T.; Izumi, S.; Gekko, K. Mass Spectrometry on Hydrogen/Deuterium Exchange of Dihydrofolate
Reductase: Effects of Ligand Binding. J. Biochem. 2004, 135, 663–671. [CrossRef]
31. Casini, A.; Messori, L. Molecular Mechanisms and Proposed Targets for Selected Anticancer Gold Compounds.
Curr. Top Med. Chem. 2011, 11, 2647–2660. [CrossRef]
32. Hashmi, A.S.K.; Rudolph, M. Gold Catalysis in Total Synthesis. Chem. Soc. Rev. 2008, 37, 1766–1775. [CrossRef]
33. Gorin, D.J.; Toste, F.D. Relativistic Effects in Homogeneous Gold Catalysis. Nature 2007, 446, 395–403. [CrossRef]
34. OriginLab—Origin and OriginPro—Data Analysis and Graphing Software. Available online: https://www.
originlab.com/ (accessed on 15 March 2019).
35. Weiss, J.N. The Hill Equation Revisited: Uses and Misuses. FASEB J. 1997, 11, 835–841. [CrossRef]
36. Batruch, I.; Javasky, E.; Brown, E.D.; Organ, M.G.; Johnson, P.E. Thermodynamic and NMR Analysis of
Inhibitor Binding to Dihydrofolate Reductase. Bioorg. Med. Chem. 2010, 18, 8485–8492. [CrossRef]
©
2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).