Glycans as Ligands in Bioinorganic Chemistry. Probing the Interaction of a Trinuclear Platinum Anticancer Complex with Defined Monosaccharide Fragments of Heparan Sulfate

Article abstract of DOI:10.1021/acs.inorgchem.8b03035

We report herein a detailed NMR study of the aquation and subsequent covalent binding of the trinuclear clinical agent [{trans-PtCl(¹⁵NH₃)₂}₂{μ-trans-Pt(¹⁵NH₃)₂(¹⁵NH₂(CH₂)₆¹⁵NH₂)₂}]⁴⁺ (1, 1,0,1/t,t,t or Triplatin) with three d-glucosamine residues containing varied O-sulfate and N-sulfate or N-acetyl substitutions, which represent monosaccharide fragments present within the repeating disaccharide sequences of cell surface heparan sulfate (HS). The monosaccharides GlcNS(6S), GlcNS, and GlcNAc(6S) were synthesized in good yield from a common 4,6-diol α-methyl glucopyranoside intermediate. The reactions of ¹⁵N-1 with sodium sulfate, GlcNS(6S), GlcNS, and GlcNAc(6S) were followed by 2D [¹H,¹⁵N] heteronuclear single quantum coherence (HSQC) NMR spectroscopy using conditions (298 K, pH ≈5.4) similar to those previously used for other anionic systems, allowing for a direct comparison. The equilibrium constants (pK₁) for the aquation of 1 in the presence of GlcNS(6S) and GlcNS were slightly higher compared to that of the aquation in a sulfate solution, while a comparable pK₁ value was observed in the presence of GlcNAc(6S). A comparison of the rate constants for sulfate displacement of the aqua ligand showed preferential binding to 2-N-sulfate compared to 6-O-sulfate but a more rapid liberation. For disulfated GlcNS(6S), equilibrium conditions were achieved rapidly (9 h) and strongly favored the dichloro form, with <2% sulfato species observed. The value of k_L1 was up to 15-fold lower than that for binding to sulfate, whereas the rate constant for the reverse ligation (k_-L1) was comparable. Equilibrium conditions were achieved much more slowly (～100 h) for the reactions of 1 with GlcNS and GlcNAc(6S), attributed to covalent binding also to the N-donor of the sulfamate (GlcNS) group and the O-donor of the N-acetyl [GlcNAc(6S)] group. The rate constants (k_L2) were 20-40-fold lower than that for binding to the 2-N- or 6-O-sulfate, but the binding was less reversible, so that their equilibrium concentrations (5-8%) were comparable to the 2-N- or 6-O-sulfate-bound species. The results emphasize the relevance of glycans in bioinorganic chemistry and underpin a fundamental molecular description of the HS-Pt interactions that alter the profile of platinum agents from cytotoxic to metastatic in a systematic manner.

Full text of DOI:10.1021/acs.inorgchem.8b03035

Article
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
pubs.acs.org/IC
Glycans as Ligands in Bioinorganic Chemistry. Probing the
Interaction of a Trinuclear Platinum Anticancer Complex with
Defined Monosaccharide Fragments of Heparan Sulfate
Anil Kumar Gorle,† Premraj Rajaratnam, Chih-Wei Chang, Mark von Itzstein,*
†
†
,†
,
†
,†,‡
Susan J. Berners-Price,* and Nicholas P. Farrell*
†
Institute for Glycomics, Griffith University, Gold Coast Campus, Southport, Queensland 4222, Australia
‡
Department of Chemistry, Virginia Commonwealth University, Richmond, Virginia 23284, United States
*
S Supporting Information
ABSTRACT: We report herein a detailed NMR study of the
aquation and subsequent covalent binding of the trinuclear
1
5
clinical agent [{trans-PtCl( NH ) } {μ-trans-Pt-
3
2 2
1
5
15
15
4+
(
NH ) ( NH (CH ) NH ) }] (1, 1,0,1/t,t,t or Triplatin)
3 2 2 2 6 2 2
with three D-glucosamine residues containing varied O-sulfate
and N-sulfate or N-acetyl substitutions, which represent
monosaccharide fragments present within the repeating
disaccharide sequences of cell surface heparan sulfate (HS).
The monosaccharides GlcNS(6S), GlcNS, and GlcNAc(6S)
were synthesized in good yield from a common 4,6-diol α-
1
5
methyl glucopyranoside intermediate. The reactions of N-1
with sodium sulfate, GlcNS(6S), GlcNS, and GlcNAc(6S) were
1
15
followed by 2D [ H, N] heteronuclear single quantum
coherence (HSQC) NMR spectroscopy using conditions (298 K, pH ≈5.4) similar to those previously used for other
anionic systems, allowing for a direct comparison. The equilibrium constants (pK ) for the aquation of 1 in the presence of
1
GlcNS(6S) and GlcNS were slightly higher compared to that of the aquation in a sulfate solution, while a comparable pK value
1
was observed in the presence of GlcNAc(6S). A comparison of the rate constants for sulfate displacement of the aqua ligand
showed preferential binding to 2-N-sulfate compared to 6-O-sulfate but a more rapid liberation. For disulfated GlcNS(6S),
equilibrium conditions were achieved rapidly (9 h) and strongly favored the dichloro form, with <2% sulfato species observed.
The value of k was up to 15-fold lower than that for binding to sulfate, whereas the rate constant for the reverse ligation (k−L1)
L1
was comparable. Equilibrium conditions were achieved much more slowly (∼ 100 h) for the reactions of 1 with GlcNS and
GlcNAc(6S), attributed to covalent binding also to the N-donor of the sulfamate (GlcNS) group and the O-donor of the N-
acetyl [GlcNAc(6S)] group. The rate constants (k ) were 20−40-fold lower than that for binding to the 2-N- or 6-O-sulfate,
L2
but the binding was less reversible, so that their equilibrium concentrations (5−8%) were comparable to the 2-N- or 6-O-
sulfate-bound species. The results emphasize the relevance of glycans in bioinorganic chemistry and underpin a fundamental
molecular description of the HS−Pt interactions that alter the profile of platinum agents from cytotoxic to metastatic in a
systematic manner.
INTRODUCTION
the roles of metal ions in biology. Studies of individual amino
acid and purine and pyrimidine nucleic acid bases have
contributed significantly to the understanding of the kinetics
and thermodynamics of metal complex (M) protein and M−
DNA (RNA) interactions of biological significance. Yet, as the
third major class of biomolecules, the bioinorganic chemistry
of glycans in the broadest sense is significantly less developed
■
Sugars are considered to be the most abundant class of organic
molecules and are the third major class of biomolecules after
proteins and nucleic acids. Glycomics is a broad scientific
discipline elucidating the diverse array of structure and
function of glycans in biological systems. Glycosaminogly-
cans (GAGs) play a role in a vast number of biological events
including cellular adhesion and migration, angiogenesis, and
the modulation of immunological responses and also act as
receptors for bacterial and viral pathogen adhesion and
invasion to host cells. GAGs are highly anionic and, as with
DNA or RNA, are associated with endogenous inorganic
cations in nature. The details of metal ion−biomolecule
interactions are of fundamental importance in understanding
1
,2
1
,2
than that of proteins and nucleic acid interactions.
Metalloglycomics, a term first coined by Codd, is the
3
systematic study of the effects of metal ions and coordination
4
compounds on oligosaccharides. The O-donor polyanions of
Received: October 26, 2018
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.8b03035
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
glycan biomolecules may be receptors and possible targets in
their own right for coordination compounds. Examples
glycans represent an extension from simple individual O-
donors to multiple donor sites on a more complex polymeric,
biomolecular template with multiple possible binding sites.
Heparin is considered to have the highest negative charge
density of any known biomolecule with an average of
4
include the interaction of chromium(V) compounds with
sialic acids and its relevance to the toxicity and metabolism of
3
,5,6
chromium species
and the use of ruthenium bipyridines as
7
−10
17−19
fluorescent probes of heparin concentration.
Cobalt(III)
approximately three negative charges per disaccharide.
ammine complexes have also been proposed for the
To date, the molecular mechanism of action of platinum
drugs has focused on their interactions with DNA as the target.
Metabolic interactions with proteins, especially human serum
albumin, and small peptides such as glutathione are considered
deactivating. Detailed pathways for interaction with nucleic
acids and proteins (N- and/or S-donors) are by now well
documented. The vast majority of discussion on the cellular
formation of cisplatin−DNA adducts focuses on the monoaqua
or diaqua species produced upon intracellular aquation of the
Pt−Cl bonds. Interactions with O-donor groups as bases have
not been considered to be as essential to explain cisplatin
activity, in part because their hard nature is less compatible
with the soft acid characteristics of platinum, noting that Pt−
OH bonds can be quite strong. Depending on extracellular and
intracellular concentrations, however these weak anions
carboxylate, carbonate, phosphate, and sulfateshould be
considered as potential ligands in the cellular and biological
milieu. Appleton et al. showed that buffer solutions containing
acetate or phosphate may contain significant concentrations of
acetate/phosphate adducts, which could affect interpretation of
quantification of sulfated glycans including in clinical
1
1−13
studies.
These apparently disparate examples testify to
the broad relevance of metalloglycomics. Glycans offer a wide
variety of binding possibilities from purely carboxylate-based
(
sialic acids and hyaluronan) to the mixed carboxylated and
sulfated GAGs including heparin and heparan sulfate (HS) and
14
the related dermatan and chondroitin sulfates. HS sequences
consist of repeating 1→4-linked disaccharide units of D-
glucosamine and glucuronic or iduronic acid residues that are
variably substituted by sulfate or acetyl groups at the 2-N
position, by sulfate at the 6-O position of D-glucosamine, and
by sulfate at the 2-O position of, predominantly, iduronic acid
(Chart 1). To understand the full potential of metallogly-
Chart 1. Major Repeating Disaccharide Unit (a) and
Variable Repeating Disaccharide Unit (b) of HS
20
the kinetic results. A further case in point is that of
carbonate, where the formation of carbonatoplatinum
intermediates has been proposed as a possible mechanism of
21−25
the cellular accumulation of cisplatin.
Indeed, even the
substitution-inert carboplatin can undergo biotransformation
reactions including ring opening of the dicarboxylate ligand in
2−
22,23
the presence of physiological concentrations of CO3
.
These transformation reactions have been observed in cells,
and carbonatoplatinum species may have a role in the
cytotoxicity, accumulation, and even resistance mechanisms
24,25
of cisplatin and carboplatin.
Carbonate may also activate
the formally inert bidentate malonate ligand in [Pt(dach)-
26
(
malonato)]. Intracellular biotransformation of the malonate
comics and the molecular details of glycan bioinorganic
chemistry, basic studies on the mono- and disaccharide
components are necessary, in much the same way that early
M−DNA chemistry was concentrated on single purine and
pyrimidine chemistry to understand nucleobase/nucleoside/
nucleotide preferences. The detailed mechanistic study of the
role of HS in biology, in general, is hampered by limited access
to defined site-specific oligosaccharides. In this paper, we
present novel synthetic methods for the preparation of site-
specific oligosaccharides and the first examples of the effect of
site-specific glycan substitution on the kinetics of binding to
clinically relevant platinum antitumor agents using 2D
compound in L1210 cells showed the presence of minor
species containing arginine, glutamate, and aspartate along
with those from the expected S-donor ligands methionine,
27
cysteine, and glutathione. Biotransformation of the clinically
relevant oxaliplatin in L1210 cells also showed the presence of
2
7,28
O-bound metabolites including even citrate species.
Finally, interactions of sulfate at physiologically relevant
2
9
concentrations have been observed for cisplatin and
30
dinuclear platinum agents. While many of these identified
species may be transitory and may not affect the final
interaction with DNA, nevertheless a full description of these
interactions is required.
1
15
[
H, N] HSQC NMR spectroscopy, the first example of the
use of this technique in glycan bioinorganic chemistry.
In this paper, we study the aquation and subsequent
15
15
covalent binding of the N-labeled trinuclear clinical agent
We have chosen to study polynuclear platinum complexes
PPCs) as an excellent example (case study) because we have
Triplatin (1, 1,0,1/t,t,t) with a sulfate anion, as well as three
monosaccharide fragments of D-glucosamine residues with
variable sulfate and acetyl substitution at the 2-N and 6-O
positions [GlcNS(6S), 2-N and 6-O sulfated; GlcNS, 2-N-
sulfated; GlcNAc(6S), 2-N acetylated and 6-O sulfated; Chart
2] as representative D-glucosamine fragments of HS. The
conditions used allow a direct comparison of the derived rate
and equilibrium constants with those obtained for the reaction
(
shown that positively charged PPCs interfere with HS function
through strong binding of the highly charged PPCs to the HS
backbone with the potential for the development of
14,16
antimetastatic, rather than cytotoxic, properties.
As a soft
acid, platinum is expected to form coordination bonds easily
with S- and/or N- donors, but chemically weak ligand
interactions with O-donors can be favored because of higher
concentrations in the biological milieu. In one sense, these
31
of 1 in a phosphate solution and also for the reactions of the
32
dinuclear analogue 1,1/t,t (Chart 2) with acetate, phosphate,
B
DOI: 10.1021/acs.inorgchem.8b03035
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Chart 2. Structures of (a) the PPCs 1,0,1/t,t,t (1, Triplatin)
and 1,1/t,t and (b) the Three D-Glucosamine
Scheme 1. Synthesis of D-Glucosamine Monosaccharide HS
Fragments
a
Monosaccharide HS Fragments Used in the Present Study
a
Reagents and conditions: (a) H , Pd(OH) /C, ethanol/phosphate
2
2
buffer (1:1, v/v), 15 h; (b) SO /NMe , NEt , H O, 80 °C, MW, 20
3
3
3
2
min, three cycles; (c) NEt , DMAP, TBDPSCl, DMF, RT, 15 h; (d)
3
BnBr, 60% NaH, 0 °C to RT, 6 h (one-pot, two steps); (e) 1 M
TBAF in THF, 50 °C, 15 h (75% yield over three steps from
compound A to C); (f) SO /NMe , DMF, 75 °C, 1 h; (g) H ,
3
3
2
Pd(OH) /C, ethanol/phosphate buffer (1:1, v/v), 15 h; (h) Ac O,
2
2
NEt , MeOH/H O (1:1, v/v); (i) SO /NMe NEt , H O, 80 °C,
3
2
3
3
3
2
MW, 20 min, three cycles.
nium fluoride (TBAF) to obtain compound C in 75% yield
over three steps. Sulfation at the 6-O position of C was readily
achieved by heating a mixture of C and the sulfur trioxide/
trimethylamine complex in N,N-dimethylformamide (DMF) at
75 °C for 1 h, giving D. Hydrogenolysis of the benzyl ethers
and azide of D afforded the fully deprotected 6-O-sulfated 2-
amino derivative E in quantitative yield. The product E was
divided into two equal portions. One portion was used for
chemoselective N-acetylation using acetic anhydride and
triethylamine to obtain the final 2-N-acetylated 6-O-sulfated
D-glucosamine product F. The other portion was subjected to
N-sulfation conditions to synthesize the 2-N- and 6-O-
disulfated D-glucosamine product G. Upon purification of the
crude mixtures, compounds F and G were obtained in 53% and
30
and sulfate, allowing the influence of the charged {PtN }2+
central linker of 1 to be explored.
4
RESULTS
■
The synthesis of three D-glucosamine monosaccharide HS
fragments with variable substitution at the 2-N and 6-O
positions has been achieved in good yield following the
synthetic route shown in Scheme 1. The common 4,6-diol α-
methyl glucopyranoside (A), which was used as a starting
material to synthesize the final monosaccharide fragments, was
prepared in five steps starting from readily available glucos-
4
0% yield, respectively. The exclusive sulfation positions and
high purity of the monosaccharide units B, E, F, and G were
1
confirmed by high-resolution mass spectrometry and H and
1
3
3
3,34
C NMR spectroscopy.
amine hydrochloride following literature procedures.
NMR Studies. The aquation reactions of 15N-1 (1 mM) in
Hydrogenolysis of diol A using Pd(OH) on active carbon
2
a 15 mM sodium sulfate solution (pH 5.4) and in the presence
under a hydrogen atmosphere provided methyl 2-amino-2-
deoxy-α-D-glucopyranoside as a pale-yellow oil, which was then
treated with a sulfur trioxide/trimethylamine complex and
triethylamine in water/methanol (1:1, v/v). This mixture was
heated under microwave irradiation at 80 °C for 20 min, over
three cycles, affording the crude compound B. Purification of
the product by size-exclusion chromatography and subsequent
passage through a cation-exchange column provided a mixture
of compound B contaminated with inseparable sodium methyl
sulfate. Formation of this undesired byproduct was successfully
avoided by changing the solvent from 1:1 methanol/water to
of the three sulfated glucosamine residues (B, F, and G; 1:15
1
15
mole ratio, pH ∼5.4) at 298 K were followed by 2D [ H, N]
HSQC NMR spectroscopy. The conditions were similar to
those used previously for the reaction of 1 in a phosphate
31
solution (15 mM) and also for the reactions of dinuclear 1,1/
3
2
30
t,t with acetate, phosphate, and sulfate, thus allowing for a
1
15
direct comparison. The [ H, N] NMR method requires that
studies are conducted below physiological pH in these systems
because it depends on a measurable J( N− H) spin−spin
coupling. Representative [ H, N] HSQC NMR spectra
recorded for equilibrium solutions of the four reactions are
shown in Figure 1, and the H and N chemical shifts of the
NH and NH groups of the different species observed (see
1
15
1
15
1
15
1
00% water, resulting in the isolation of pure 2-N-sulfated D-
1
15
glucosamine (product B) in quantitative yield (purity >95%,
determined by H NMR), which is superior to the previously
reported method in the literature.
1
5
15
1
3
2
35
Scheme 2) are listed in Table 1.
Further, compound A was subjected to one-pot 6-O-tert-
butyldiphenylsilyl (TBDPS) protection and 4-O-benzylation,
followed by subsequent desilylation using tetra-n-butylammo-
Aquation of 1 in a Sulfate Solution. In addition to the
1
15
H/ N crosspeaks at δ 3.89/-64.4 (NH ) and 5.06/-47.0
3
(NH ) corresponding to the parent dichloro species (Pt−Cl,
2
C
DOI: 10.1021/acs.inorgchem.8b03035
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1
15
Figure 1. 2D [ H, N] HSQC NMR (600 MHz) spectra of 1,0,1/t,t,t (1) after reaction with (a) 15 mM sodium sulfate at 298 K (pH 5.4) after 43
h, b) GlcNS(6S) at 298 K (pH 5.3) after 70 h, (c) GlcNS at 298 K (pH 5.4) after 174 h, and (d) GlcNAc(6S) at 298 K (pH 5.3) after 160 h. Peaks
1
5
15
have been assigned to the Pt− NH and Pt− NH groups of the species shown in Scheme 2 Models: A (a and b); B (c); C (d). Peak assignments
3
2
1
5
15
2+
are shown in Table 1. Peaks labeled “L” are from the Pt− NH and Pt− NH groups of the central {PtN } linker, and those labeled “o” are
other minor products (see the text).
3
2
4
label I, Figure 1a), peaks assignable to the aquated species
Pt−OH , label II, Figure 1a) appeared at δ 4.11/-62.1 (NH )
for 1, and 2-fold lower anation rate constant (k−H) for the
reformation of the chloro species. A similar 2-fold lowering of
k−H is seen for the reactions in phosphate, but in this case the
aquation rate constant is 1.4 times higher for the trinuclear
(
2
3
and 5.13/-63.7 (NH ). Two additional peaks at δ 4.03/-60.9
2
(
NH ) and 5.14/-66.0 (NH ) became visible as the reaction
3
2
proceeded and are assigned to the sulfato species (Pt-OSO3,
complex, resulting in a significantly lower pK value. For 1,1/t,t
1
label III, Figure 1a). The chemical shifts for the terminal
the pK value in sulfate is almost 0.5 log units lower compared
to that in either phosphate or acetate, but for trinuclear 1, the
1
1
5
15
NH and NH groups are identical to those observed for
3
2
3
0
1
15
the reaction of 1,1/t,t with sulfate. The H/ N chemical
pK values in sulfate and phosphate are comparable. The rate
1
shifts of the linker 15NH and NH groups did not change
15
constants for sulfate displacement of the aqua ligand (kL1) and
aquation of the sulfato ligand (k−L1) are both about 2 times
lower for 1 than dinuclear 1,1/t,t, resulting in similar pK2
values. In stark contrast to the reactions with 1,1/t,t, where the
rate constant for sulfate displacement of the aqua ligand is
about 3 times higher than that of phosphate or acetate, for 1
the kL1 values are similar for sulfate and phosphate. On the
other hand, similar to the reaction with 1,1/t,t the aquation of
the sulfato ligand is significantly faster (close to an order of
magnitude) in comparison to displacement of phosphate.
Reactions of 1 with the D-Glucosamine Monosac-
3
2
over the course of the reaction (Table 1).
Similar to the reaction with 1,1/t,t, the system in sulfate
solution reached equilibrium after 12 h. The majority (64.7%)
of 1 remained in the chloro form (Pt−Cl, I), with the aquated
(
Pt−OH , II) and sulfato species (Pt-OSO , III) accounting
2
3
for 24.0% and 11.3%, respectively. The kinetic profile for the
reaction is shown in Figure 2a and the rate and equilibrium
constants are summarized in Table 2 and compared to the
values obtained for reaction of 1 with phosphate and for 1,1/t,t
with sulfate, phosphate and acetate under the same
3
0−32
1
15
conditions.
a similar distribution at equilibrium (I, 68.4%, II, 20.9, III,
0.7%), inspection of the rate constants in Table 2 reveals
While the reaction of 1,1/t,t with sulfate has
charide Fragments. H/ N chemical shifts for the
1
5
15
Pt− NH and Pt− NH groups of dichloro (I), aqua (II)
3
2
1
and sulfato (III) species observed in all three reactions are
similar to those observed for the aquation of 1 in sulfate
solution (see Table 1). Although the 2-N and 6-O sulfate
environments in GlcNS and GlcNAc(6S) are different,
identical chemical shifts are observed for the 2-N- and 6-O-
sulfate bound species, and thus it cannot be determined for
GlcNS(6S) (which contains sulfate groups at both 2-N- and 6-
O- positions) whether one or both the sulfate groups are
involved in covalent interaction with 1.
interesting differences that may be attributed to a balance of
−
ion pairing (Cl ) and/or formation of sulfate (or phosphate)
clamps with the charged {PtN4}2 central linker. For aquation
+
in the absence of coordinating anions the equilibrium constant
for the aquation (pK ) is lower for 1 than the related dinuclear
1
complex (i.e., further toward the aquated species) due to a
lower anation (k−H) rate constant, attributed to ion-pairing
+
31
between chloride ions and the {PtN4} unit. For the
reactions in sulfate the two pK values are comparable, as a
The time dependence of the concentrations of the species
1
consequence of a slightly lowered forward rate constant (kH)
observed in the reactions of 1 with GlcNS(6S), GlcNS and
D
DOI: 10.1021/acs.inorgchem.8b03035
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 2. Simplified Kinetic Models Used To Study the
Aquation of 1,0,1/t,t,t (1) in the Presence of Sodium Sulfate
and the Three D-Glucosamine Monosaccharide HS
(12 h). There is a marked difference in the equilibrium
concentrations with 75.9% remaining in the chloro form, a
similar amount (22.3%) in the aqua form but only 1.8% as
sulfate-bound species (Figure 2b).
a
Fragments
Equilibrium conditions were achieved much more slowly for
the reactions with GlcNS (93 h, Figure 2c) and GlcNAc(6S)
(
110 h, Figure 2d) compared to the reactions of 1 with either
GlcNS(6S) or sodium sulfate, attributed to the slow formation
of species in addition to the initial 2-N-sulfate or 6-O-sulfate
bound adducts, for which peaks were first visible within 30
min. In the case of GlcNS, two new sets of peaks, first visible
1
15
after about 2 h (labeled IV in Figure 1c), have H/ N shifts (δ
4
.13/-63.0 (NH ) and 5.02/-47.5 (NH )) consistent with Pt−
3
2
N bound species and are attributed to covalent binding of 1
with the N-donor of the sulfamate (−NHSO ) group. For
3
GlcNAc(6S) a new set of peaks at δ 3.97/-61.1 (NH ) and
3
4
.92/-63.8 (NH ) (labeled V in Figure 1d) appeared at a
2
similar time and are attributed to binding to the O-donor of
the N-acetyl group, based on the H/ N shift of the NH
1
15
2
group, which is in the region of Pt−O bound species.
Interestingly, although GlcNS(6S) also contains a sulfamate
group no similar peaks for N-bound species appeared in this
case.
Rate and equilibrium constants derived using the kinetic
models for the respective reactions (illustrated in Scheme 2)
are summarized in Table 3. In the case of GlcNAc(6S) the
model (C), takes into account the formation of additional
minor species which formed slowly over time (peaks labeled
“
o” in Figure 1d). These peaks were first visible after 6.7 h and
a
increased very slowly accounting for 2.2% of the total species at
20 h and 5.3% in the equilibrium solution (110 h). Although
unassigned, the species are not due to binding to the
monosaccharide as we have observed identical peaks in the
reaction of the dinuclear 1,1/t,t with sulfate, where sulfuric acid
was used for pH adjustment. These species are observed to a
lesser extent in the reaction with GlcNS (Figure 1c).
Roman numerals represent the mononuclear species [trans-[Pt-
n+
(
NH ) (NH (CH ) NH )Y] (Pt−Y), where Y represents the
3
2
2
2
6
2
coordinated ligand Cl, H O, SO , etc. Model A was used for sulfate
2
4
and GlcNS(6S), model B for GlcNS, and model C for GlcNAc(6S).
1
15
15
Table 1. H and N Chemical Shifts for the Pt− NH and
3
1
5
Pt− NH Groups of the Species Observed during the
2
1
5
Comparison of the pK1 values shows some interesting
differences for the initial aquation of 1 in the presence of the D-
Reaction of N-1 with Sulfate, GlcNS(6S), GlcNS, and
a
GlcNAc(6S) at 298 K (pH ∼5.4)
glucosamine monosaccharide fragments. The lower pK (in
1
15NH3
15NH2
δ 15
N
comparison to 1,1/t,t) that was observed for 1 in the presence
of sulfate is similarly observed for GlcNAc(6S) due to a similar
lowering of the anation (k−H) rate constant. On the other hand
for GlcNS and GlcNS(6S) the pK1 values are higher
(indicating equilibrium toward the parent chloro form) and
the anation rate constants are now similar to that for 1,1/t,t in
the presence of sulfate. Inspection of the DFT model (Figure
3a) of the interaction of the disulfated GlcNS(6S) with
1
δ 15
1
1
,0,1/t,t,t species
δ H
N
δ H
I (Pt−Cl)
II [Pt−OH(H)]
III(Pt−OSO /OSO N)
IV (Pt−NHSO3)
3.88
4.09
4.03
4.13
3.97
64.4
62.2
60.9
63.0
61.1
5.05
5.07
5.14
5.02
4.92
47.1
63.3
66.0
47.5
63.9
b
c
3
2
V (Pt−ONAc)
a
1
1
5
H
r
e
f
e
r
e
n
c
e
d
t
o
1
,
4
-
d
i
o
x
a
n
e
(
i
n
t
e
r
n
a
l
)
a
n
d
N
r
e
f
e
r
e
n
c
e
d
t
o
2
+
15
[
Pt(NH ) ] illustrates how in this case formation of sulfate
NH Cl (external). Roman numerals represent the mononuclear
3 4
4
n+
clamps, involving both the 2-N- and 6-O sulfate, would limit
species [trans-[Pt(NH ) (NH (CH ) NH )Y] , as shown in the
kinetic models in Scheme 2. Only the shifts for the terminal platinum
am(m)ine groups are listed; the H and N shifts of the linker group
were unchanged over the course of any of the reactions, but the H
3
2
2
2
6
2
the ion pairing between the released chloride and the charged
1
15
2+
{PtN4} central linker. Why there should be a difference for
1
the monosulfated GlcNS and GlcNAc(6S) is less clear based
shifts were slightly shielded for the reaction with GlcNS(6S) [δ 4.15/
2+
on the single sulfate clamps formed with [Pt(NH ) ]
3
4
−
63.7 (NH ) and 4.68/−44.3 (NH )] compared to the reaction with
2+
3
2
(
representing the central {PtN } ) in each case (Figure 3b
4
sulfate and the other two monosaccharides [δ 4.17/−63.7 (NH ) and
3
b
1
15
and c).
Comparison of the rate constants for sulfate displacement of
4
.72/−44.3 (NH )]. H/ N chemical shifts for the aquation
2
c
product are pH-sensitive; for titration curves, see ref 31. Identical
1
15
the aqua ligand show preferential binding to 2-N-sulfate
compared to 6-O sulfate; the value of kL1 is 3-fold higher for
GlcNS compared to GlcNAc(6S), but 0.6 times lower than for
binding to sulfate. In the case of GlcNS(6S) kL1 is lower than
for binding to the 2-N-sulfate (5 fold) and 6-O sulfate (1.5
fold) in GlcNS and GlcNAc(6S), respectively. It is 7-fold lower
than for binding to sulfate. When consideration is given to
H/ N chemical shifts for the sulfate-bound species were observed in
all cases.
GlcNAc(6S) are shown in Figure 2 (b-d). In the case of
GlcNS(6S), equilibrium conditions were achieved slightly
faster (9 h) when compared to the reaction of 1 with sulfate
E
DOI: 10.1021/acs.inorgchem.8b03035
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 2. Plots of relative concentrations of species observed during the reaction of 1,0,1/t,t,t (1) at 298 K with a) sodium sulfate (pH 5.4); b)
1
5
1
GlcNS(6S) (pH 5.3); c) GlcNS (pH 5.4); d) GlcNAc(6S) (pH 5.3), based on intensities of the crosspeaks in the Pt− NH region of the 2D [ H,
3
15
N] HSQC spectra (Figure 1). The curves are computer best fits for the rate constants shown in Tables 2 and 3, which were fit to Model A (a and
b); Model B (c); Model C (d) (Scheme 2). Key I is open squares (□); Cl¯ is “x”; II is open circles (Ο); III is open triangles (△); IV is asterisks
×); V is open diamonds (◇). In (d) the other minor products (‘o’) are not shown for clarity but were included in the kinetic model (see Table 3).
(
a
b
Table 2. Rate and Equilibrium Constants for the Aquation and Ligation of 1,1/t,t and 1,0,1/t,t,t (1) in 15 mM Sulfate,
Acetate and Phosphate
1
5
a
b
1
,0,1/t,t,t ( N-1)
1,1/t,t
SO42
−
PO43−
SO42−
PO43−
OAc−
parameter
kH (10− s−1
5
)
)
3.43 ± 0.01
0.1324 ± 0.0007
0.0131 ± 0.0006
41 ± 2
3.4 ± 0.1
3.85 ± 0.05
0.229 ± 0.04
0.025 ± 0.004
70 ± 10
2.49 ± 0.04
0.40 ± 0.01
0.0086 ± 0.0002
3.9 ± 0.1
1.83 ± 0.03
0.262 ± 0.009
0.0086 ± 0.0001
0.56 ± 0.02
k−H (M s−1
−
1
0.275 ± 0.009
0.0166 ± 0.0002
5.4 ± 0.1
kL1 (M−
1
s
−1)
k−L1 (10− s−1
5
)
pK1
pK2
3.587 ± 0.003
−1.50 ± 0.03
3.73 ± 0.02
−2.51 ± 0.01
b
3.77 ± 0.01
−1.6 ± 0.1
4.21 ± 0.02
−2.34 ± 0.02
4.16 ± 0.02
−3.19 ± 0.02
a
Data for 15N-1 binding with PO43− taken from ref.31 Data taken from ref.
30,32
GlcNS(6S) containing two sulfate binding groups (effective
concentration of 30 mM) the lowering of kL1 is even greater,
nearly 15-fold lower than for binding to sulfate (Table S1).
Once again electrostatic interaction between the disulfated
monosaccharide (2- charge) and the {PtN4}2 linker, through
sulfate clamp formation would limit the covalent binding of the
sulfate groups to the terminal {PtN OH } groups. There are
subtle differences also in the rate constant for aquation of the
sulfato ligand. For GlcNS and GlcNS(6S) the value k−L1 is
similar to that of the reaction of sulfate but for GlcNAc(6S) it
is 3-fold lower. Thus, while binding to 2-N-sulfate is preferred
it is liberated more quickly than binding to the 6-O sulfate.
In the case of GlcNS the rate constant (k ) for
L2
displacement of the aqua ligand by the N-donor of the
sulfamate group is close to 40 fold lower than that for binding
to the 2-N-sulfate, but the rate constant for the reverse reaction
(k−L2) is also significantly lower (nearly 70 fold) so this species
accumulates over time. The distribution of species at
equilibrium is I, 72.4%, II, 15.2%, III, 4.9%, IV, 7.5%.
+
3
2
Similarly, for GlcNAc(6S) the rate constant (k ) for
L2
F
DOI: 10.1021/acs.inorgchem.8b03035
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 3. Rate and Equilibrium Constants for the Aquation and Ligation of 1,0,1/t,t,t (1) with GlcNS(6S), GlcNS and
a
GlcNAc(6S) (at 298 K, pH ∼5.4).
b
Parameter
GlcNS(6S)
GlcNS
GlcNAc(6S)
kH (10− s−1
5
)
)
3.06 ± 0.02
0.217 ± 0.001
0.0018 ± 0.0003
34 ± 6
3.16 ± 0.02
0.284 ± 0.002
0.0086 ± 0.0007
41 ± 4
0.000218 ± 0.000004
0.59 ± 0.02
2.94 ± 0.03
0.132 ± 0.002
0.0027 ± 0.0002
14 ± 1
0.000142 ± 0.000008
0.85 ± 0.07
k−H (M s−1
−1
kL1 (M−
1
s
5
−1)
k−L1 (10 s−1
−
)
)
kL2 (M− s−1
1
)
k−L2 (10−
5
s
−1
pK1
pK2
pK3
3.850 ± 0.003
−0.7 ± 0.1
3.953 ± 0.004
−1.32 ± 0.06
−1.57 ± 0.02
3.652 ± 0.009
−1.28 ± 0.04
−1.22 ± 0.05
a
The rate and equilibrium constants were derived from the kinetic models shown in Scheme 2; for the reaction with GlcNS(6S) rate and
b
equilibrium constants determined by considering the concentration of GlcNS(6S) as 30 mM are given in Supporting Information. The two minor
peaks observed in the reaction with GlcNAc(6S) (labeled ‘o’ in Figure 1d) together accounted for 5.4% of the total species at the equilibrium.
These minor products were summed as other products and taken into account in the kinetic fit analysis (Model C, Scheme 2). The forward and
reverse rate constants for these species are k (M s ), 0.0000019 ± 0.0000001; k (10 s ), 0.71 ± 0.07 and the equilibrium constant (pK )
is 0.57 ± 0.06.
−
1
−1
−5 −1
O
−O
O
2
+
Figure 3. DFT optimized models for the interaction of [Pt(NH ) ] with the sulfated monosaccharide fragments a) GlcNS(6S); b) GlcNS; c)
3
4
GlcNAc(6S).
1
4,16
displacement of the aqua ligand by the O-donor of the N-acetyl
group is nearly 20-fold lower than that for binding to the 6-O
sulfate, while the reverse reaction (k−L2) is 16-fold lower
resulting in the equilibrium distribution I, 64.7%, II, 19.3%, III,
antimetastatic.
A comparison of the reactions with
monosulfated GlcNS and GlcNAc(6S) reveals the higher
reactivity of 1 with the 2-N-sulfate and also the higher
substitution-labile nature of 2-N-sulfato species (kL1 and k−L1
are both 3 times higher for 2-N-sulfate than 6-O-sulfate).
Significantly lower sulfate binding determined in the case of
disulfated GlcNS(6S) suggests that the incorporation of a
second sulfate group at either the 2-N or 6-O position makes it
less favorable toward covalent interaction, which may reflect on
the steric effects of the relatively large but highly flexible
trinuclear complex or a greater preassociation of the two
molecules due to the higher charge on GlcNS(6S) as noted
above. The effects of preassociation on the final products of the
covalent interaction are often quite subtle, and a relevant
example is in the mechanism of the formation of long-range
5
.7%, V, 4.8% (Figure 2).
Although previous studies involving binuclear complexes of
3
6
37
cisplatin with N-glycylglycine and α-pyridone as the
bridging ligands showed covalent interaction with the O-
donor groups (similar to the N-acetyl group of GlcNAc(6S))
the low reactivity observed in the present study demonstrate
the weaker nature of N-acetyl O-donors compared to sulfate O-
II
donors for binding with the Pt center.
DISCUSSION
■
GAGs such as HS are highly sulfated polysaccharides with
variable length,, disaccharide sequences, and sulfation
{
Pt,Pt} 1,4 and 1,6-interstrand DNA−DNA cross-links formed
by Triplatin (1), where minor groove preassociation affects
both the kinetics and mechanism of the cross-link formation as
1
2
patterns. Among the building units of the GAG repeating
sequences, D-glucosamine residues act as key synthons for the
assembly of these complex biomolecules. The results presented
in this study show the strong substitution labile nature of the
species formed by the covalent interaction of 1 with sulfate O-
donors of D-glucosamine residues. This work represents an
example of the relevance of glycans in bioinorganic chemistry,
where, specifically, PPCs may represent a new multifunctional
chemotype for interference with HS−enzyme−protein inter-
actions, expanding the potential utility of platinum-based
anticancer agents from cytotoxic to antiangiogenic and
38
well as the final adduct structure. Similarly, preassociation is
implicated in the intimate mechanism of formation of the
unique 5′−5′and antiparallel 3′−3′ directional isomers of
39,40
{Pt,Pt} 1,4-interstrand cross-links.
The analogy with DNA as the template is relevant also when
we consider extension of the monosaccharide studies here to
the cellular reality of the HS polymer. The nature of the
template (single-stranded vs double-stranded) affects the
41
kinetics of aquation of the dinuclear 1,1/t,t complex, which
G
DOI: 10.1021/acs.inorgchem.8b03035
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
emphasizes further the useful analogy between the helical HS
and single-stranded DNA. Extending from the simple
individual O-donors to multiple donor sites on a more
complex polymeric, biomolecular template with multiple
possible binding sites raises the possibility that the
oligosaccharide polymers are “sinks” for electrostatic and
hydrogen-bonding interactions. A strong preassociation could,
role for 2-NHSO binding in the mediation of PPC cellular
3
accumulation. Further, GAGs are emerging biomarkers for
many tumor types including triple-negative breast and ovarian
46,50,51
cancers.
Platinum drugs enter tumor cells by comple-
mentary pathways including passive diffusion and active
transport, such as through the copper-transport and/or
48
organic-cation-transport system. The unique HSPG-medi-
ated accumulation pathway for PPCs such as Triplatin can
confer a clinical advantage over the current agents in tumors
overexpressing GAGs. Fundamental studies as discussed in this
contribution are essential to our understanding of how to
exploit cellular accumulation mechanisms with the potential
for a new approach to precision medicine for platinum.
−
for example, result in a lower local Cl concentration and allow
the albeit weak O-donor substitution reactions to occur even in
−
the presence of the expected high extracellular Cl
concentrations. These features can be examined as we move
1
to longer site-specific oligosaccharides, such as studies by H
NMR spectroscopy of the interaction of Triplatin with the
pentasaccharide Fondaparinux, the highly sulfated synthetic
GAG-based fragment used clinically as an antithrombotic
CONCLUSIONS
■
1
7
agent. The initial spectral changes are entirely indicative of
the preassociation of Triplatin on the oligosaccharide skeleton,
followed by bond formation through displacement of the Pt−
In summary, at the monosaccharide level, the substituents on
the sugar ring of the biologically important D-glucosamine
affect the kinetics of aquation and substitution of Triplatin
1
6,42
Cl bond.
The spectral changes in the alkanediamine linkers
(
1,0,1/t,t,t; BBR3464). The general concept of preassociation
of Triplatin also mirrored those observed upon covalent Pt−
on a biomolecule, followed by “fixation” in a covalent bond-
forming reaction, is validated in the present studies for GAG
fragments, even with formally weak O-donors. These
interactions will underpin the fundamental molecular descrip-
tion of the HS−Pt(M) complex interaction, leading to
inhibition of the HS function as well as providing a holistic
description of the metabolic fate of metal-based anticancer
complexes in general in the complex cellular milieu.
38
DNA bond formation. The interactions are sufficiently
strong to inhibit cleavage of Fondaparinux by bacterial
14,16
heparinase and mammalian heparanase.
HS is a significant participant in major cellular processes
through binding to a host of structural and signaling proteins
18
such as growth factors. HS−protein binding is mediated
through electrostatic and hydrogen bonding of positively
charged protein residues to the negatively charged carboxylate
and sulfate groups on the HS polymer. The 6-O-sulfate of D-
glucosamine is intimately involved in the recognition of a host
of growth factors and the formation of stable ternary HS−
growth factor−growth factor receptor complexes with well-
ASSOCIATED CONTENT
Supporting Information
■
*
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.8b03035.
43−45
documented structurally characterized examples.
Triplatin
inhibits fibroblast growth factor (FGF-2) binding to
biotinylated HS, and sub-cytotoxic concentrations also inhibit
the FGF-2-induced migration of human umbilical vein
Rate and equilibrium constants for the aquation and
1
5
ligation of N-1 (1 mM) with GlcNS(6S) (30 mM),
scientist models used to fit the kinetic data, and
experimental procedures (PDF)
2
1
epithelial cells. The studies presented here contribute to
our understanding of the detailed molecular mechanism of the
inhibition of growth factor−HS interactions by Triplatin (and
PPCs, in general) and confirm the utility of the metal-
loshielding approach to targeting heparin and HS−protein
AUTHOR INFORMATION
Corresponding Authors
E-mail: m.vonitzstein@griffith.edu.au.
E-mail: s.berners-price@griffith.edu.au.
*E-mail: npfarrell@vcu.edu.
■
*
*
1
9,46
interactions.
The results may also have implications for the intimate
mechanism of heparan sulfate proteoglycan (HSPG)-mediated
47
cellular accumulation of coordination compounds. HSPGs
act as receptors for highly charged PPCs, facilitating their
cellular accumulation by this unique mechanism of internal-
ization not shared with the mononuclear cisplatin and
ORCID
Anil Kumar Gorle: 0000-0001-7838-2625
Premraj Rajaratnam: 0000-0002-3788-5346
Chih-Wei Chang: 0000-0002-5009-0302
Mark von Itzstein: 0000-0001-6302-7524
Susan J. Berners-Price: 0000-0002-7129-0039
Nicholas P. Farrell: 0000-0001-7160-7182
4
7,48
oxaliplatin.
This was shown, first, by comparing the
cellular accumulation of PPCs in the parental Chinese Hamster
Ovarian cell line, CHO-K1, and GAG-deficient CHO-pgsA745
(
deficient in the GAG−biosynthetic xylosyltransferase 2
47
XYLT2 enzyme) cells. Decreased cellular accumulation of
Notes
PPCs in the GAG-deficient cell lines translated to significantly
47
The authors declare no competing financial interest.
decreased cytotoxicity and apoptotic events. The availability
of site-specific mutants allows the molecular details of the
accumulation mechanism to be examined. In this respect, the
pgsE-606 CHO mutant is 3−5-fold defective in N-
sulfotransferase activity, resulting in an approximate 2−3-fold
ACKNOWLEDGMENTS
■
The work was supported by grants from the Australian
Research Council (Grant DP150100308) to S.J.B.-P., M.v.I.,
and N.P.F. and, in part, from the National Institutes of Health
(Grant RO1CA78754) to N.P.F. The DFT study in this work
was undertaken with the assistance of resources and services
from the National Computational Infrastructure, which is
49
reduction of the extent of N-sulfation. Preliminary studies
show that the accumulation of Triplatin is diminished in the
CHO-pgsE-606 cell line, with concomitant reduction in
5
0
cytotoxicity. These results are at least consistent with the
H
DOI: 10.1021/acs.inorgchem.8b03035
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
supported by the Australian Government. We thank Dr. Robin
Thomson for careful reading of the manuscript.
reactions between the cis-diamminediaquaplatinum(II) cation and the
oxygen-donor ligands hydroxide, perchlorate, nitrate, sulfate,
phosphate, and acetate. Inorg. Chem. 1984, 23, 3514−3521.
(21) Centerwall, C. R.; Goodisman, J.; Kerwood, D. J.; Dabrowiak, J.
REFERENCES
■
C. Cisplatin carbonato complexes. Implications for uptake, antitumor
(
1) Hudak, J. E.; Bertozzi, C. R. Glycotherapy: New advances inspire
a reemergence of glycans in medicine. Chem. Biol. 2014, 21, 16−37.
properties, and toxicity. J. Am. Chem. Soc. 2005, 127, 12768−12769.
(22) Di Pasqua, A. J.; Goodisman, J.; Kerwood, D. J.; Toms, B. B.;
(
2) Cummings, R. D.; Pierce, J. M. The challenge and promise of
glycomics. Chem. Biol. 2014, 21, 1−15.
3) Codd, R. Metalloglycomics: a new perspective upon competitive
metal-carbohydrate binding using EPR spectroscopy. Chem. Commun.
004, 2653−2655.
4) Farrell, N. P.; Gorle, A. K.; Peterson, E. J.; Berners-Price, S. J. In
Dubowy, R. L.; Dabrowiak, J. C. Activation of carboplatin by
carbonate. Chem. Res. Toxicol. 2006, 19, 139−149.
(
(23) Di Pasqua, A. J.; Centerwall, C. R.; Kerwood, D. J.; Dabrowiak,
J. C. Formation of carbonato and hydroxo complexes in the reaction
of platinum anticancer drugs with carbonate. Inorg. Chem. 2009, 48,
2
(
1
(
192−1197.
Metal Ions in Life Sciences; Sigel, A., Sigel, H., Freisinger, E., Sigel, R. K.
O., Eds.; Walter de Gruyter GmbH: Berlin, Germany, 2018; Vol. 18,
pp 109−140.
24) Di Pasqua, A. J.; Goodisman, J.; Kerwood, D. J.; Toms, B. B.;
Dubowy, R. L.; Dabrowiak, J. C. Role of carbonate in the cytotoxicity
of carboplatin. Chem. Res. Toxicol. 2007, 20, 896−904.
(5) Codd, R.; Irwin, J. A.; Lay, P. A. Sialoglycoprotein and
(25) Centerwall, C. R.; Kerwood, D. J.; Goodisman, J.; Toms, B. B.;
carbohydrate complexes in chromium toxicity. Curr. Opin. Chem. Biol.
003, 7, 213−219.
6) Codd, R.; Lay, P. A. Chromium(V)-sialic (neuraminic) acid
Dabrowiak, J. C. New extracellular resistance mechanism for cisplatin.
2
(
J. Inorg. Biochem. 2008, 102, 1044−1049.
(26) Mauldin, S. K.; Plescia, M.; Richard, F. A.; Wyrick, S. D.;
species are formed from mixtures of chromium(VI) and saliva. J. Am.
Voyksner, R. D.; Chaney, S. G. Displacement of the bidentate
malonate ligand from (d,l-trans-1,2-diaminocyclohexane)-
malonatoplatinum(II) by physiologically important compounds in
vitro. Biochem. Pharmacol. 1988, 37, 3321−3333.
Chem. Soc. 2001, 123, 11799−11800.
(7) Waller, L. N.; Fox, N.; Fox, K. F.; Fox, A.; Price, R. L.
Ruthenium red staining for ultrastructural visualization of a
glycoprotein layer surrounding the spore of Bacillus anthracis and
Bacillus subtilis. J. Microbiol. Methods 2004, 58, 23−30.
(27) Mauldin, S. K.; Gibbons, G.; Wyrick, S. D.; Chaney, S. G.
Intracellular biotransformation of platinum compounds with the 1,2-
diaminocyclohexane carrier ligand in the L1210 cell line. Cancer Res.
(8) Rozenberg, G. I.; Espada, J.; de Cidre, L. L.; Eijan, A. M.; Calvo,
́
J. C.; Bertolesi, G. E. Heparan sulfate, heparin, and heparinase activity
detection on polyacrylamide gel electrophoresis using the fluoro-
chrome tris(2,2’-bipyridine) ruthenium (II). Electrophoresis 2001, 22,
1
(
988, 48, 5136−5144.
28) Mauldin, S. K.; Husain, I.; Sancar, A.; Chaney, S. G. Effects of
the bidentate malonate ligand on the utilization and cytotoxicity of
platinum compounds in the L1210 cell line. Cancer Res. 1986, 46,
3
(
−11.
9) Cheng, T.-T.; Yao, J.-L.; Gao, X.; Sun, W.; Shi, S.; Yao, T.-M. A
2
(
876−2882.
new fluorescence ‘switch on’ assay for heparin detection by using a
functional ruthenium polypyridyl complex. Analyst 2013, 138, 3483−
29) Appleton, T. G.; Hall, J. R.; Ralph, S. F.; Thompson, C. S. M.
Reactions of platinum(II) aqua complexes. 2. Platinum-195 NMR
study of reactions between the tetraaquaplatinum(II) cation and
chloride, hydroxide, perchlorate, nitrate, sulfate, phosphate, and
acetate. Inorg. Chem. 1984, 23, 3521−3525.
3
(
489.
10) Szelke, H.; Harenberg, J.; Kramer, R. Detection and
̈
neutralisation of heparin by a fluorescent ruthenium compound.
Thromb. Haemostasis 2009, 102, 859−864.
(30) Ruhayel, R. A.; Corry, B.; Braun, C.; Thomas, D. S.; Berners-
(11) Vouras, M.; Schubert, M. The outer sphere association of
Price, S. J.; Farrell, N. P. Determination of the kinetic profile of a
chondroitin sulfate with polyvalent complex cations. J. Am. Chem. Soc.
957, 79, 792−795.
12) Farber, S. J.; Schubert, M. The binding of cations by
chondroitin sulfate. J. Clin. Invest. 1957, 36, 1715−1722.
13) Graham, H. D.; Williams, J. L. Quantitative aspects of the
dinuclear platinum anticancer complex in the presence of sulfate:
1
(
1
15
Introducing a new tool for the expedited analysis of 2D [ H, N]
HSQC NMR spectra. Inorg. Chem. 2010, 49, 10815−10819.
(31) Davies, M. S.; Thomas, D. S.; Hegmans, A.; Berners-Price, S. J.;
(
Farrell, N. Kinetic and equilibria studies of the aquation of the
interaction of carrageenan and other hydrocolloids with polyvalent
trinuclear platinum phase II anticancer agent [{trans-PtCl-
cobalt complexes. J. Food Sci. 1966, 31, 362−372.
4
+
(
NH ) } {μ-trans-Pt(NH ) (NH (CH ) NH ) }] (BBR3464).
(14) Mangrum, J. B.; Engelmann, B. J.; Peterson, E. J.; Ryan, J. J.;
3 2 2 3 2 2 2 6 2 2
Inorg. Chem. 2002, 41, 1101−1109.
Berners-Price, S. J.; Farrell, N. P. A new approach to glycan targeting:
enzyme inhibition by oligosaccharide metalloshielding. Chem.
Commun. 2014, 50, 4056−4058.
(32) Zhang, J.; Thomas, D. S.; Davies, M. S.; Berners-Price, S. J.;
Farrell, N. Effects of geometric isomerism in dinuclear platinum
antitumor complexes on aquation reactions in the presence of
perchlorate, acetate and phosphate. JBIC, J. Biol. Inorg. Chem. 2005,
10, 652−666.
(33) Cai, C.; Dickinson, D. M.; Li, L.; Masuko, S.; Suflita, M.;
Schultz, V.; Nelson, S. D.; Bhaskar, U.; Liu, J.; Linhardt, R. J.
Fluorous-assisted chemoenzymatic synthesis of heparan sulfate
oligosaccharides. Org. Lett. 2014, 16, 2240−2243.
(34) Hu, Y.-P.; Lin, S.-Y.; Huang, C.-Y.; Zulueta, M. M. L.; Liu, J.-Y.;
Chang, W.; Hung, S.-C. Synthesis of 3-O-sulfonated heparan sulfate
octasaccharides that inhibit the herpes simplex virus type 1 host−cell
interaction. Nat. Chem. 2011, 3, 557−563.
(35) Inoue, Y.; Nagasawa, K. An improved method for the
preparation of crystalline sodium salts of 2-deoxy-2-sulfoamino-d-
glucose and methyl 2-deoxy-2-sulfoamino-α-d-glucopyranoside. Car-
bohydr. Res. 1979, 69, 297−300.
(36) Appleton, T. G. Donor atom preferences in complexes of
platinum and palladium with amino acids and related molecules.
Coord. Chem. Rev. 1997, 166, 313−359.
(37) Hollis, L. S.; Lippard, S. J. Synthesis, structure, and platinum-
195 NMR studies of binuclear complexes of cis-diammineplatinum-
(15) Berners-Price, S. J.; Ronconi, L.; Sadler, P. J. Insights into the
mechanism of action of platinum anticancer drugs from multinuclear
NMR spectroscopy. Prog. Nucl. Magn. Reson. Spectrosc. 2006, 49, 65−
9
(
8.
16) Peterson, E. J.; Daniel, A. G.; Katner, S. J.; Bohlmann, L.;
Chang, C.-W.; Bezos, A.; Parish, C. R.; von Itzstein, M.; Berners-Price,
S. J.; Farrell, N. P. Antiangiogenic platinum through glycan targeting.
Chem. Sci. 2017, 8, 241−252.
(17) Petitou, M.; van Boeckel, C. A. A. A synthetic antithrombin III
binding pentasaccharide is now a drug! What comes next? Angew.
Chem., Int. Ed. 2004, 43, 3118−3133.
(18) Chiodelli, P.; Bugatti, A.; Urbinati, C.; Rusnati, M. Heparin/
heparan sulfate proteoglycans glycomic interactome in angiogenesis:
Biological implications and therapeutical use. Molecules 2015, 20,
6
(
342−6388.
19) Weiss, R. J.; Esko, J. D.; Tor, Y. Targeting heparin and heparan
sulfate protein interactions. Org. Biomol. Chem. 2017, 15, 5656−5668.
20) Appleton, T. G.; Berry, R. D.; Davis, C. A.; Hall, J. R.; Kimlin,
H. A. Reactions of platinum(II) aqua complexes. 1. Multinuclear
platinum-195, nitrogen-15, and phosphorus-31) NMR study of
(
(
I
DOI: 10.1021/acs.inorgchem.8b03035
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
1
II) with bridging.alpha.-pyridonate ligands. J. Am. Chem. Soc. 1983,
05, 3494−3503.
38) Hegmans, A.; Berners-Price, S. J.; Davies, M. S.; Thomas, D. S.;
(
Humphreys, A. S.; Farrell, N. Long range 1, 4 and 1, 6-interstrand
cross-links formed by a trinuclear platinum complex. Minor groove
preassociation affects kinetics and mechanism of cross-link formation
as well as adduct structure. J. Am. Chem. Soc. 2004, 126, 2166−2180.
(39) Ruhayel, R. A.; Berners-Price, S. J.; Farrell, N. P. Competitive
formation of both long-range 5′−5′ and short-range antiparallel 3′−3′
DNA interstrand cross-links by a trinuclear platinum complex on
binding to a 10-mer duplex. Dalton Trans. 2013, 42, 3181−3187.
(40) Ruhayel, R. A.; Moniodis, J. J.; Yang, X.; Kasparkova, J.; Brabec,
V.; Berners-Price, S. J.; Farrell, N. P. Factors affecting DNA−DNA
interstrand cross-links in the antiparallel 3′−3′ sense: A comparison
with the 5′−5′ directional isomer. Chem. - Eur. J. 2009, 15, 9365−
9
(
374.
41) Davies, M. S.; Berners-Price, S. J.; Cox, J. W.; Farrell, N. The
nature of the DNA template (single-versus double-stranded) affects
the rate of aquation of a dinuclear Pt anticancer drug. Chem. Commun.
2
(
003, 122−123.
42) Gorle, A. K.; Katner, S. J.; Johnson, W. E.; Lee, D. E.; Daniel, A.
G.; Ginsburg, E. P.; von Itzstein, M.; Berners-Price, S. J.; Farrell, N. P.
Substitution-inert polynuclear platinum complexes as metalloshielding
agents for heparan sulfate. Chem. - Eur. J. 2018, 24, 6606−6616.
(
43) El Masri, R.; Seffouh, A.; Lortat-Jacob, H.; Vives, R. R. The “in
̀
and out” of glucosamine 6-O-sulfation: the 6th sense of heparan
sulfate. Glycoconjugate J. 2017, 34, 285−298.
(
44) Faham, S.; Hileman, R. E.; Fromm, J. R.; Linhardt, R. J.; Rees,
D. C. Heparin structure and interactions with basic fibroblast growth
factor. Science 1996, 271, 1116−1120.
(45) Wesche, J.; Haglund, K.; Haugsten, E. M. Fibroblast growth
factors and their receptors in cancer. Biochem. J. 2011, 437, 199−213.
46) Fuster, M. M.; Esko, J. D. The sweet and sour of cancer:
glycans as novel therapeutic targets. Nat. Rev. Cancer 2005, 5, 526−
42.
47) Silva, H.; Frez
(
5
(
́
ard, F.; Peterson, E. J.; Kabolizadeh, P.; Ryan, J.
J.; Farrell, N. P. Heparan sulfate proteoglycan-mediated entry pathway
for charged tri-platinum compounds: differential cellular accumulation
mechanisms for platinum. Mol. Pharmaceutics 2012, 9, 1795−1802.
(48) Hall, M. D.; Okabe, M.; Shen, D.; Liang, X.; Gottesman, M. M.
The role of cellular accumulation in determining sensitivity to
platinum-based chemotherapy. Annu. Rev. Pharmacol. Toxicol. 2008,
4
(
8, 495−535.
49) Bame, K. J.; Esko, J. D. Undersulfated heparan sulfate in a
Chinese hamster ovary cell mutant defective in heparan sulfate N-
sulfotransferase. J. Biol. Chem. 1989, 264, 8059−8065.
(50) Katner, S. J.; Hampton, J. D.; Peterson, E. J.; Katsuta, E.;
Sayyad, M. R.; Takabe, K.; Koblinski, J.; Farrell, N. P. Abstract 3941:
Heparan sulfate, a new target for platinum in metastatic TNBC.
Cancer Res. 2018, 78, 3941−3941.
(51) Vallen, M. J. E.; van der Steen, S. C. H. A.; van Tilborg, A. A.
G.; Massuger, L. F. A. G.; van Kuppevelt, T. H. Sulfated sugars in the
extracellular matrix orchestrate ovarian cancer development: ‘When
sweet turns sour’. Gynecol. Oncol. 2014, 135, 371−381.
J
DOI: 10.1021/acs.inorgchem.8b03035
Inorg. Chem. XXXX, XXX, XXX−XXX