Asymmetric zinc(ii) complexes as functional and structural models for phosphoesterases

Article abstract of DOI:10.1039/c3dt50514f

We report two asymmetric ligands for the generation of structural and functional dinuclear metal complexes as phosphoesterase mimics. Two zinc(ii) complexes, [Zn₂(CH₃L4)(CH₃CO₂) ₂]⁺ (CH₃HL4 = 2-(((2-methoxyethyl)(pyridin-2- ylmethyl)amino)methyl)-4-methyl-6-(((pyridin-2-ylmethyl)amino)methyl)phenol) and [Zn₂(CH₃L5)(CH₃CO₂) ₂]⁺ (CH₃HL5 = 2-(((2-methoxyethyl)(pyridine-2- ylmethyl)amino)methyl)-4-methyl-6-(((pyridin-2-ylmethyl)(4-vinylbenzyl)amino) methyl)phenol) were synthesized and characterized by X-ray crystallography. The structures showed that the ligands enforce a mixed 6,5-coordinate environment in the solid state. ¹H-, ¹³C- and ³¹P-NMR, mass spectrometry and infrared spectroscopy were used to further characterize the compounds in the solid state and in solution. The zinc(ii) complexes hydrolyzed the organophosphate substrate bis-(2,4-dinitrophenol)phosphate (BDNPP), the nucleophile proposed to be a terminal water molecule (pK_a 7.2). The ligand CH₃HL4 was immobilised on Merrifield resin and its zinc(ii) complex generated. Infrared spectroscopy, microanalysis and XPS measurements confirmed successful immobilisation, with a catalyst loading of ～1.45 mmol g^-1 resin. The resin bound complex was active towards BDNPP and displayed similar pH dependence to the complex in solution. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3dt50514f

Dalton
Transactions
View Article Online
View Journal
PAPER
Asymmetric zinc(II) complexes as functional and
structural models for phosphoesterases†
Cite this: DOI: 10.1039/c3dt50514f
Lena J. Daumann, Laurène Marty, Gerhard Schenk and Lawrence R. Gahan*
We report two asymmetric ligands for the generation of structural and functional dinuclear metal com-
plexes as phosphoesterase mimics. Two zinc(II) complexes, [Zn2(CH3L4)(CH3CO2)2]+ (CH3HL4 = 2-(((2-
methoxyethyl)(pyridin-2-ylmethyl)amino)methyl)-4-methyl-6-(((pyridin-2-ylmethyl)amino)methyl)-
phenol) and [Zn2(CH3L5)(CH3CO2)2]+ (CH3HL5
= 2-(((2-methoxyethyl)(pyridine-2-ylmethyl)amino)-
methyl)-4-methyl-6-(((pyridin-2-ylmethyl)(4-vinylbenzyl)amino)methyl)phenol) were synthesized and
characterized by X-ray crystallography. The structures showed that the ligands enforce a mixed 6,5-co-
ordinate environment in the solid state. 1H-, 13C- and 31P-NMR, mass spectrometry and infrared spectro-
scopy were used to further characterize the compounds in the solid state and in solution. The zinc(II)
complexes hydrolyzed the organophosphate substrate bis-(2,4-dinitrophenol)phosphate (BDNPP), the
nucleophile proposed to be a terminal water molecule (pKa 7.2). The ligand CH3HL4 was immobilised on
Merrifield resin and its zinc(II) complex generated. Infrared spectroscopy, microanalysis and XPS measure-
ments confirmed successful immobilisation, with a catalyst loading of ∼1.45 mmol g−1 resin. The resin
bound complex was active towards BDNPP and displayed similar pH dependence to the complex in
solution.
Received 26th February 2013,
Accepted 2nd May 2013
DOI: 10.1039/c3dt50514f
www.rsc.org/dalton
By careful design the asymmetric ligand can offer both
differentiated metal binding sites, and/or sites for further syn-
Introduction
In order to generate more structurally comparable biomimetics thetic elaboration. For example, an asymmetric Fe(III)Zn(II)
for dinuclear metallohydrolases much effort has been devoted purple acid phosphatase (PAP) model complex with potential
to the synthesis of asymmetric ligands. It has been proposed for immobilisation on 3-aminopropyl silica has been
that these asymmetric complexes are not only more appropri- reported.17 The immobilised complex exhibited reaction rates
ate functional models for the active site of phosphoesterase toward the hydrolysis of BDNPP similar to those found with
enzymes, but also that they exhibit enhanced catalytic rates the free complex; in addition, the immobilised catalyst was
compared with their symmetric counterparts.1–3 Ligands used reusable. In addition, a PAP model with pendant olefinic arms
to generate purple acid phosphatase,1,4–10 phosphoesterase,11 suitable for copolymerisation and thus incorporation into a
urease,12,13 catechol oxidase14 and manganese catalase bio- polymer has also been reported.18 Such biomimetics, and ulti-
mimetics15,16 have been reported. Some ligands engender both a mately metalloenzymes, immobilised on inorganic solid sup-
hard and a soft coordination site resulting in heterodinuclear ports, have the potential for organophosphate pesticide
complexes as, for example, models for purple acid phosphat- bioremediation. The immobilisation of complexes on in-
ase metalloenzymes.4,7 In other cases the ligands have a struc- organic or organic supports has been extensively explored and
tural variation in one arm4,6,7 whilst in others one donor arm modified silica, organic resins such as Tentagel or Merrifield
has been omitted,1,14–16 with the vacant coordination site resin and other polymers have served previously as sup-
often found to be occupied by water or solvent molecules in ports.19,20 Merrifield resin attached complexes have been
the complex.15,16
reported useful for various applications such as Ni(II) catalysed
cross coupling reactions,21 the oxidative DNA cleavage cata-
lysed by an immobilised cyclen Cu(II),20 Pd(II) catalysed
Suzuki–Miyaura cross-coupling reactions,19 salen Mn(II) com-
plexes for epoxidation reactions,22 and the Co(II) catalysed
polymerisation of vinyl acetate,23 to list a few examples. The
advantage of the immobilised system is that the complex can
be easily separated from the reaction mixture and the loaded
resin recycled.
School of Chemistry and Molecular Biosciences, The University of Queensland,
Brisbane, QLD 4072, Australia. E-mail: gahan@uq.edu.au
†Electronic supplementary information (ESI) available: Mass spectra, NMR titra-
tions, IR spectra and additional XPS data can be found in the ESI. CCDC 925903
and 925904. For ESI and crystallographic data in CIF or other electronic format
see DOI: 10.1039/c3dt50514f
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
We have been interested in the glycerophosphodiesterase single quantum correlation (HSQC) and heterobinuclear multi-
enzyme (GpdQ) from Enterobacter aerogenes. GpdQ is a highly ple bond connectivity (HMBC) experiments were used
promiscuous dinuclear metallohydrolase with respect to both to assign each signal in the spectra of the final ligands.
substrate specificity and metal ion composition; in addition, it 31P NMR spectra were recorded at room temperature
is particularly robust suggesting that it may be a promising with 85% phosphoric acid as external standard (δP = 0.00).
candidate for bioremedial applications.24–28 In order to fully Positive-ion electrospray mass spectrometry was carried out
study this metalloenzyme we have prepared a number of with a Q-Star time-of-flight mass spectrometer, and the data
mimics of its active site structure in order to investigate the were processed with Bruker Compass Data Analysis 4.0 soft-
mechanism of action with a view to preparing active model ware. FT-Infrared spectroscopy was carried out with a Perkin
systems.29
Elmer FT-IR Spectrometer SPECTRUM 2000 with a Smiths Dura-
In this work we describe the synthesis and characterisation of SamplIR II ATR diamond window. Elemental microanalyses
two asymmetric ligands 2-(((2-methoxyethyl)(pyridin-2-ylmethyl)- (C, H, N) were performed with a Carlo Erba Elemental Analy-
amino)methyl)-4-methyl-6-(((pyridin-2-ylmethyl)amino)methyl)- ser, model NA1500, by Mr George Blazak at the University of
phenol
(CH3HL4)
and
2-(((2-methoxyethyl)(pyridine-2- Queensland. X-ray photoelectron spectroscopy (XPS) was con-
ylmethyl)-amino)methyl)-4-methyl-6-(((pyridin-2-ylmethyl)(4-vinyl- ducted with Dr Barry Wood at the Centre for Microscopy and
benzyl)-amino)methyl)phenol (CH3HL5) (Chart 1) and their Microanalysis, University of Queensland, with a Kratos Axis
respective zinc(II) complexes as structural and functional Ultra photoelectron spectrometer which uses Al Kα (1253.6 eV)
models for the enzyme GpdQ. The immobilisation of the X-rays. The software Casa XPS was used for data processing.
complex with CH3HL4 on Merrifield resin and the activity
towards the substrate bis-(2,4-dinitrophenol)phosphate
(BDNPP) are reported.
Kinetic studies
Kinetic studies were conducted with bis-(2,4-dinitrophenol)-
phosphate (BDNPP) as substrate using a Varian Cary50 Bio UV/
Visible spectrophotometer with a Peltier temperature control-
Materials and methods
ler and 10 mm quartz cuvettes. The initial-rate method was
employed, and assays were measured such that the initial
linear portion of the data was used for analysis. Product for-
mation was determined at 25 °C by monitoring the formation
of 2,4-dinitrophenol. Throughout the pH range studied
(4.75–8.5), the extinction coefficient of this product at 400 nm
varies from 7180 at pH 4.5, 10 080 at 5.0; 11 400 at pH 5.5 to
12 000 at 6.0 and 12 100 at pH 6.5–8.5.30 All assays were
measured in 50 : 50 acetonitrile–buffer with the substrate and
complex initially dissolved in acetonitrile. The aqueous multi-
component buffer was comprised of 50 mM 2-(N-morpholino)-
General methods
1H NMR spectra were recorded at room temperature with a
300, 400 or 500 MHz Bruker AV 300/400/500 spectrometer.
Chemical shifts are reported in δ units relative to CHCl3 (δH
=
7.24), CD3CN (δH = 1.93). The following abbreviations were
used: s = singlet, d = doublet, t = triplet, dd = doublet of
doublet, dt = doublet of triplet, m = multiplet. The software
used for data processing was TOPSPIN 2.1. 13C NMR spectra
were recorded at room temperature with a 100 MHz Bruker AV
400/500 spectrometer. Chemical shifts are given in δ units rela-
ethanesulfonic
acid
(MES),
4-(2-hydroxyethyl)-1-piper-
tive to CDCl3 (central line of triplet: δC = 77.0), d4-MeOD (δC
=
azineethanesulfonic acid (HEPES), 2-(N-cyclohexylamino)ethane
sulfonic acid (CHES) and N-cyclohexyl-3-aminopropanesulf-
onic acid (CAPS) with controlled ionic strength (LiClO4)
250 mM. The pH values reported are those of the aqueous
component; it should however be noted that the pH of a solu-
tion of the buffer was the same within error as a 1 : 1 mixture
of buffer and acetonitrile.31 Assays for pH dependence were 40
µM in complex and 5 mM in BDNPP; for substrate dependence
they were 40 µM in complex and 1–11.5 mM in BDNPP. pH-
dependence data for monoprotic events were fit to eqn (1).32
49.0), CD3CN (δC = 1.30), D2O with 5% dioxane (δC = 67.2). Two-
dimensional correlation spectroscopy (COSY), heterobinuclear
vmax
1 þ ð½Hþꢀ=KaÞ
v0 ¼
ð1Þ
The data derived from substrate dependence were fit to the
Michaelis–Menten eqn (2).32
vmax½Sꢀ
v0 ¼
ð2Þ
KM þ ½Sꢀ
Here, v0 is the initial rate, vmax is the maximum rate, KM is
Chart 1 Ligands reported previously and the two new CH3HL4 and CH3HL5
ligands.
the Michaelis constant, and [S] is the substrate concentration.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Complex dependence was measured with a fixed substrate con- amino)methyl)-5-methylbenzaldehyde (1.69 g, 5.37 mmol) was
centration at 5 mM and complex concentrations ranging from dissolved in methanol (50 mL) and 2-aminomethylpyridine
20–120 µM. Background assays were conducted by measuring (0.58 g, 5.37 mmol) in methanol (25 mL) added drop wise at
the autohydrolysis and hydrolysis by two equivalents of zinc(II) room temperature. The resulting mixture was stirred at 50 °C
acetate and were subtracted from the data. For the kinetic for two hours, the mixture subsequently cooled to 0 °C and
investigation of the immobilised resin, an assay was estab- sodium borohydride (0.72 g, 19.0 mmol) added in small por-
lished which allowed determination of the amount of BDNPP tions. After heating to reflux for 3 hours the mixture was con-
hydrolysed by the resins. In order to obtain initial rates, the centrated in vacuo, taken up in acidified water (100 mL, pH 2)
absorbance of a solution of multicomponent buffer, aceto- and extracted with dichloromethane (3 × 35 mL). The com-
nitrile and BDNPP was determined (T = 0), and then 0.5 mg of bined organic extracts were washed with saturated NaHCO3
resin added. After 15 minutes 1 mL of the suspension was solution (3 × 50 mL) and dried over Na2SO4. After removal of
transferred to an Eppendorf tube and centrifuged for 10 the solvent in vacuo the crude ligand (1.53 g, 70%) was purified
seconds; the absorbance of the supernatant was measured with flash column chromatography (EtOAc until the first band
again (T = 15) and the Abs/min calculated. For every value this was eluted, MeOH–EtOAc 9 : 1, FeCl3 stain, Rf = 0.48 in EtOAc)
was done in triplicate, however, due to swelling properties of to yield 900 mg (41%) of the ligand as a yellow oil. 1H NMR
the resin and inhomogeneous catalyst loading on the Merri- (CDCl3, 500.13 MHz) δ 2.20 (s, 3H, arCCH3); 2.76 (t, 2H,
field beads the errors were larger than for the free complex.
NCH2CH2, J = 5.5 Hz); 3.26 (s, 3H, OCH3); 3.50 (t, 2H,
NCH2CH2, J = 5.5 Hz); 3.77 (s, 2H, arCCH2N); 3.85 (s, 2H,
arCCH2NH); 3.92 (s, 2H, NCH2py); 3.96 (s, 2H, NHCH2py); 6.78
Crystallographic measurements
Crystallographic data for the complexes were collected at (d, 1H, arCH, J = 1.8 Hz); 6.92 (d, 1H, arCH, J = 1.8 Hz); 7.13
293(2) K ([Zn2(CH3L4)(CH3CO2)2]PF6) and 150 K ([Zn2(CH3L5)- (qd, 2H, pyH, J = 7.4, 1.2 Hz); 7.34 (dd, 2H, pyH, J = 5.1, 1.5
(CH3CO2)2]BPh4) with an Oxford Diffraction Gemini Ultra dual Hz); 7.62 (tt, 2H, pyH, J = 7.6, 1.9 Hz); 8.51 (ddt, 2H, pyH, J =
source (Mo and Cu) CCD diffractometer, for the former using 6.6, 1.7, 0.9 Hz). 13C NMR (CDCl3, 100.62 MHz) δ 20.4
Mo (λKα = 0.71073 Å) and the latter using Cu (λKα = 1.5418 Å) (arCCH3); 49.2 (NCH2py); 52.6 (NCH2CH2); 53.7 (NHCH2py);
radiation. The structures were solved by direct methods 56.9 (arCCH2N); 58.6 (OCH3); 60.0 (arCCH2NH); 70.1
(SIR-92)33 and refined (SHELXL 97)34 by full matrix least (NCH2CH2); 122.0 (pyCH); 122.2 (pyCH); 122.3 (pyCH); 122.4
squares methods based on F2.35 These programs were accessed (arC); 123.4 (pyCH); 124.4 (arC); 127.8 (arCCH3); 129.2 (arCH);
through the WINGX 1.70.01 crystallographic collective 129.7 (arCH); 136.5 (pyCH); 136.6 (pyCH); 148.9 (pyCH); 149.0
package.36 All non-hydrogen atoms were refined anisotropi- (pyCH); 153.5 (arCOH); 158.0 (pyCCH2); 158.6 (pyCCH2). FT-IR
cally unless they were disordered. Hydrogen atoms were fixed spectroscopy (ν, cm−1) 3402.3 (m, O–H str); 2917.2, 2818.5 (m,
geometrically and were not refined. X-ray data of the published C–H str); 1108.4 (m, C–O str); 862.5 (w, ar-H); 755.2 (s, py-H).
structures were deposited with the Cambridge Crystallographic ESI mass spectrometry (methanol) m/z 407.23 [C24H30N4O2 + H]+.
Data Centre CCDC 925903 and 925904.
Synthesis
of
2-(((2-methoxyethyl)(pyridine-2-ylmethyl)-
amino)methyl)-4-methyl-6-(((pyridin-2-ylmethyl)(4-vinylbenzyl)-
amino)methyl)phenol (CH3HL5). CH3HL4 (500 mg, 1.2 mmol)
Ligand syntheses
3-(Chloromethyl)-2-hydroxy-5-methylbenzaldehyde and 2-meth- and vinylbenzyl chloride (0.174 mL, 1.1 mmol) were stirred
oxy-N-(pyridin-2-ylmethyl)aminoethanol
were
prepared with dry powdered potassium carbonate (0.256 g, 1.9 mmol) in
following previously published procedures.29,37 Merrifield acetonitrile (5 mL) at room temperature for two days. After this
resin (1% crosslinked, 3.5 mmol g−1 Cl) was obtained from time the mixture was filtered and concentrated in vacuo. The
Aldrich Chemical Company (microanalysis found C 80.18, H crude residue was purified by flash column chromatography
6.64%).
(EtOAc–MeOH 5 : 2) and CH3HL5 obtained as orange oil (yield:
Synthesis of 2-(((2-methoxyethyl)(pyridin-2-ylmethyl)amino)- 38%, 0.24 g). 1H NMR (CDCl3, 500.13 MHz) δ 2.24 (s, 3H,
methyl)-4-methyl-6-((( pyridin-2-ylmethyl)amino)methyl)- arCCH3); 2.77 (t, 2H, NCH2CH2, J = 5.7 Hz); 3.27 (s, 3H, OCH3);
phenol (CH3HL4). Triethylamine (2.6 mL) was added drop 3.52 (t, 2H, NCH2CH2, J = 5.7 Hz); 3.65 (s, 2H, CH2); 3.71 (s,
wise to a mixture of 3-(chloromethyl)-2-hydroxy-5-methylbenz- 2H, CH2); 3.77 (s, 2H, CH2); 3.79 (s, 2H, CH2); 3.86 (s, 2H,
aldehyde (1.0 g, 6.0 mmol) and 2-methoxy-N-(pyridin-2- CH2); 5.19 (dd, 1H, CHvCH2 cis, J = 10.9, 0.6 Hz); 5.69 (dd,
ylmethyl)ethanamine (1.0 g, 6.0 mmol) in tetrahydrofuran 1H, CHvCH2 trans, J = 17.6, 0.8 Hz); 6.67 (dd, 1H, CHvCH2,
(45 mL) at room temperature and the mixture stirred for J = 17.6, 10.9 Hz); 6.89 (d, 1H, arCH, J = 1.7 Hz); 7.03 (d, 1H,
24 hours. After filtration and concentration in vacuo the arCH, J = 1.8 Hz); 7.13 (m, 2H, pyCH, J = 6.4, 0.9 Hz); 7.34 (s,
residue was taken up in water (30 mL) and extracted with 4H, vinylbenzylH); 7.43 (d, 1H, pyCH, J = 7.8 Hz); 7.51 (d, 1H,
dichloromethane (3 × 30 mL). The combined organic extracts pyCH, J = 7.8 Hz); 7.62 (m, 2H, pyCH, J = 7.7, 1.8 Hz); 8.51 (m,
were dried over Na2SO4 and concentrated in vacuo to yield 2H, pyCH, J = 0.9 Hz). 13C NMR (CDCl3, 100.62 MHz) δ 20.6
1.69 g (89%) of crude 2-hydroxy-3-(((2-methoxyethyl)(pyridin-2- (arCCH3); 52.9 (NCH2CH2); 53.8 (CH2); 55.7 (CH2); 57.9 (CH2);
ylmethyl)amino)methyl)-5-methylbenzaldehyde as an orange 58.6 (OCH3); 59.6 (CH2); 60.4 (CH2); 70.4 (NCH2CH2); 113.4
oil which was used in the next step without further purifi- (CHvCH2); 121.8 (pyCH); 122.0 (pyCH); 122.9 (pyCH); 123.2
cation. The 2-hydroxy-3-(((2-methoxyethyl)(pyridin-2-ylmethyl)- (pyCH); 123.9 (arC-CHvCH2); 126.1 (vinylbenzyl arCH); 127.5
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
(arCCH3); 129.0 (arCH); 129.1 (vinylbenzyl arCH); 129.3 (arCH); Further analysis was done with the hexafluorophosphate
136.4 (pyCH); 136.5 (pyCH); 136.6 (CHvCH2); 138.5 (arC); derivative. ESI mass spectrometry (methanol) m/z = 751.1
148.8 (pyCH); 148.9 (pyCH); 153.5 (arCOH); 158.0 (pyCCH2); (100%), 749.1 (95%) calc. for [C35H47N4O6Zn2]+ m/z = 749.2
158.6 (pyCCH2). FT-IR spectroscopy (ν, cm−1) 2918.3, 2815.5 (100%), 751.2 (96.1%); (acetonitrile) m/z = 851.3, calc. 851.23
(m, C–H str); 1108.5 (m, C–O str); 860.9 (w, ar-H); 755.4 (s, py- (100%), 849.23 (85.5%) [C41H49N6O6Zn2]+; m/z = 769.1, calc.
H). ESI mass spectrometry (methanol) m/z 523.20 [C33H38N4O2 769.17 (100%), 767.18 (87%), [C37H43N4O6Zn2]+; m/z = 723.1,
+ H]+.
calc. 721.17 (100%), 732.17 (96.6%), [C33H43N4O6Zn2]+), m/z =
Synthesis of [Zn2(CH3L4)(CH3CO2)2](PF6). CH3HL4 (80 mg, 653.1, calc. m/z 653.11 (100%), 655.11 (96.2%),
0.2 mmol) and zinc(II) acetate dihydrate (86 mg, 0.4 mmol) [C28H35N4O6Zn2]+. FT-IR spectroscopy (ν, cm−1) 2926.6, 2856.4
were combined in methanol (5 mL) and refluxed for 30 min. (w, C–H str); 1597.8 (s, bridging acetate antisym. str); 1431.4 (s,
After the solution was cooled to room temperature sodium bridging acetate sym. str); 832.1 (s, P–F str); 764.6, 728.5 (m,
hexafluorophosphate (65 mg, 0.4 mmol) was added, the pale pyH def); 555.3 (m, P–F). Microanalysis Anal calc. for
yellow solution filtered and left to evaporate. The white crystals C38H48N4O7Zn2PF6 C 48.12, H 5.10, N 5.91; found: C 48.19, H
which formed after 3 days were collected and dried in air 4.71, N 5.36%. 1H NMR (CD3CN, 500.13 MHz) δ 1.94 (s, 6H,
(85 mg, 54.8%). ESI mass spectrometry (methanol) m/z: 655.11 CH3CO2−); 2.14 (s, 3H, arCH3); 2.50–2.56 (m, 2H, NCH2CH2);
[C28H35N4O6Zn2]+, 595.06 [C26H31N4O4Zn2]+; (acetonitrile) 2.84–2.90 (m, 2H, NCH2CH2); 3.03 (s, 3H, OCH3); 3.57 (d, 1H,
m/z
= 749.1, calc. m/z 749.17 (100%), 748.17 (79.2%), arCH2N, J = 12.6 Hz); 3.64 (d, 1H, NCH2py, J = 12.1 Hz);
[C60H88N10O18Zn4]2+. FT-IR spectroscopy (ν, cm−1) 3306.9 (w, 3.71–3.91 (m, 5H, CH2); 4.15 (d, 1H, arCH2N, J = 12.5);
N–H str); 2925.3 (w, C–H str); 1596.3 (s, bridging acetate 4.21–4.27 (m, 2H, CH2); 5.20 (d, 1H, CHvCH2, Jcis = 10.9 Hz);
antisym. str); 1421.6 (s, bridging acetate sym. str); 832.3 (s, P–F 5.25 (dd, 1H, CHvCH2, Jcis = 11.0, 0.5 Hz, second isomer); 5.69
str); 767.7, 663.1 (m, pyH def); 555.6 (m, P–F). Microanalysis (d, 1H, CHvCH2, Jtrans = 17.6 Hz); 5.81 (d, 1H, CHvCH2,
Anal calc. for C28H38N4O6Zn2PF6 C 41.29, H 4.24, N 7.13; Jtrans = 17.6 Hz, second isomer); 6.62 (dd, 1H, CHvCH2, J = 17.7,
found:
C 41.21, H 4.25, N
6.72%. 1H NMR (CD3CN, 10.9, 0.8 Hz); 6.71 (dd, 1H, CHvCH2, J = 17.7, 10.9 Hz, second
500.13 MHz); δ 1.97 (s, 6H, acetateCH3); 2.55 (m, 2H, isomer); 6.81 (d, 2H, arCH, J = 1.8 Hz); 6.89 (d, 2H, arCH, J =
NCH2CH2); 2.75 (m, 1H, NCH2CH2); 2.82 (d, 1H NCH2CH2, J = 1.8 Hz); 6.95 (s, 2H, arCH); 7.10–7.53 (m, 4H, pyCH); 7.91 (m,
11.1 Hz); 3.04 (s, 3H, OCH3); 3.65 (d, 1H, arCH2N, J = 12.2 Hz); 2H, pyCH, J = 7.8, 1.6 Hz); 8.70 (m, 2H, pyCH). 13C NMR
3.79 (d, 1H, NCH2py, J = 11.4 Hz); 3.85 (d, 1H, NCH2py, J = 15.0 (CD3CN, 100.62 MHz) δ 20.2 (arCCH3); 23.1 (CH3CO2−); 54.7
Hz); 3.99 (t, 1H, arCH2N, J = 12.8 Hz); 4.13 (t, 1H, NCH2py, J = (CH2); 58.6 (OCH3); 59.1, 59.5, 60.8, 61.2, 61.3, 69.3 (CH2);
11.6 Hz); 4.21 (dd, 1H, arCH2N, J = 12.1, 5.1 Hz); 4.31 (d, 1H, 114.9 (CHvCH2); 115.2 (CHvCH2, second isomer); 124.0,
NCH2py, J = 15.2 Hz); 4.35 (d, 1H, arCH2N, J = 12.3 Hz); 6.94 (s, 124.3, 124.7, 125.0, 125.1 (pyCH); 125.2, 125.3 (Cquart); 126.8,
1H, arH); 6.97 (s, 1H, arH); 7.40–7.35 (m, 4H, pyH); 7.98 (mc, 126.9 (CH); 127.0, 127.3, 131.6 (Cquart), 132.8, 133.0 (CH); 133.2
2H, pyH, J = 7.7, 0.8 Hz); 8.60 (d, 1H, pyH, J = 4.8 Hz); 8.77 (d, (CH); 133.6 (CH); 133.7 (Cquart); 137.1 (CHvCH2); 138.5 (arC);
1H, pyH, J = 4.7 Hz). 13C NMR (CD3CN, 100.62 MHz); 20.3 140.8, 141.3, 141.5, 148.2, 148.5, 149.8 (CH); 155.8, 156.5,
(CH3); 24.3 (acetateCH3); 51.5 (arCH2N); 52.8 (NCH2py); 55.1 160.8, 161.4 (Cquart), 179.0 (CH3CO2−).
(CH2CH2); 59.1 (OCH3); 61.9 (arCH2N); 69.6 (CH2CH2); 124.2
Immobilisation
of
CH3HL4
on
Merrifield
resin
(pyCH); 125.0 (pyCH); 125.1 (pyCH); 125.3 (pyCH); 125.8 (M-CH3HL4). To Merrifield resin (1% crosslinked, 3.5 mmol
(arC); 126.4 (arC); 126.4 (arCCH3); 132.2 (arCH); 132.8 (arCH); g−1 Cl, 500 mg) in acetonitrile (10 mL) was added CH3HL4
140.8 (pyCH); 141.0 (pyCH); 148.6 (pyCH); 148.7 (pyCH); 155.8 (320 mg, 0.78 mmol) and potassium carbonate (250 mg) and
(pyC); 161.1 (arCO); 168.4 (CO2−).
the mixture stirred for 10 days. After this time the orange
Synthesis of [Zn2(CH3L5)(CH3CO2)2](PF6)·MeOH. CH3HL5 beads of the resin were dried on a sintered glass funnel (the
(50 mg, 0.09 mmol) and zinc(II) acetate dihydrate (42 mg, filtrate was collected to recover any unbound ligand) and
0.19 mmol) were combined in methanol (5 mL) and refluxed washed with water (20 mL) and methanol (20 mL). The resin
for 30 min. After cooling to room temperature, sodium hexa- was further dried under high vacuum to yield 646 mg of
fluorophosphate (48 mg, 0.29 mmol) was added, the pale M-CH3HL4. 26 mg of ligand were recovered from the filtrate.
yellow solution filtered and left to evaporate. After 3 days the Total loading 1.45 mmol ligand g−1 resin. FT-IR spectroscopy
resulting white powder was collected and dried in air (29 mg, (ν, cm−1) 3352.3 (b, O–H str); 3024.5, 2921.7 (w, C–H str);
37%). Crystals suitable for X-ray structure analysis were 1110.0 (m, C–O–C str); 814.4, 756.4, 699.4 (m, Ar-H/Py-H str).
obtained as follows: CH3HL5 (30 mg, 0.05 mmol) and zinc(II) Microanalysis found C 77.39, H 7.22, N 4.86%.
acetate dihydrate (25 mg, 0.11 mmol) were combined in
Synthesis of the immobilised zinc(II) complex M-[Zn2-
methanol (3 mL) and refluxed for 30 min, and after cooling to (CH3L4)(CH3COO)2]. To M-CH3HL4 (400 mg) in methanol
room temperature sodium tetraphenylborate (59 mg, (10 mL) was added zinc(II) acetate dihydrate (400 mg) and the
0.17 mmol) was added and the pale yellow solution filtered mixture refluxed for 30 minutes and then stirred for 24 h. Sub-
and left to evaporate. The white powder was taken up in sequently the resin was dried and washed with water (10 mL)
acetone/isopropanol and layered with hexane which yielded and methanol (20 mL) to yield 505 mg of the immobilised
colourless crystals, desiccating readily upon removal from the complex M-[Zn2(CH3L4)(CH3COO)2]. FT-IR spectroscopy (ν,
solvent. X-ray structure analysis was thus conducted at 150 K. cm−1) 3410.1 (b, O–H str); 3025.4, 2923.5 (w, C–H str); 1596.1
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
(s, bridging acetate antisym. str); 1420.5 (s, bridging acetate
sym. str); 808.9, 760.9, 700.1, 664.3 (m, Ar-H/Py-H str). Micro-
analysis found C 64.42, H 6.18, N 3.58%. The resin was further
characterized with XPS.
Results and discussion
Ligand nomenclature
The nomenclature employed for these types of ligands
(CH3H3L1, CH3HL2 and CH3HL3, Chart 1) has been described
previously;29 the designations L4 and L5 follow this previous
nomenclature. In general, the nomenclature employed for the
ligands denotes the number of labile protons upon complexa-
tion and the substituent on the para-position of the phen-
oxide. Hence, CH3HL4 and CH3HL5 have the same phenolate
backbone with a methyl group in para position of the phenolic
oxygen and have one potential site for deprotonation, the
phenolic –OH; complexation as CH3L4− and CH3L5− implies a
single deprotonation.
Ligand and complex synthesis
CH3HL4 ligand synthesis was achieved using a protocol
similar to that previously published.38 Addition of 2-methoxy-
N-(pyridin-2-ylmethyl)aminoethanol to the 3-(chloromethyl)-2-
hydroxy-5-methylbenzaldehyde precursor, subsequent Schiff
base condensation with 2-aminomethylpyridine and reduction
with sodium borohydride gave the desired ligand CH3HL4 as a
yellow oil in moderate yield. Treatment of CH3HL4 with vinyl
benzyl chloride in the presence of potassium carbonate
resulted in the ligand CH3HL5. For the dizinc(II) complex with
CH3HL4, crystals suitable for X-ray crystallography were
obtained after multiple crystallisations using ether diffusion
into a methanolic solution of the complex. Crystals of
[Zn2(CH3L5)(CH3COO)2]BPh4 were obtained after slow
diffusion (3 weeks) of hexane into an acetone solution of the
complex. The crystals desiccated upon removal from the
mother liquor, requiring low temperature during X-ray struc-
ture analysis. The remaining analysis of this complex was
undertaken with the [Zn2(CH3L5)(CH3COO)2]PF6 derivative.
Fig. 1 ORTEP plots (25%) of the two asymmetric complexes. Top: the two
molecules in the asymmetric unit of [Zn2(CH3L4)-(CH3COO)2]PF6, with the pyri-
dine and methyl-ether arm positions of the minor isomer represented in dashed
bonds. Bottom: [Zn2(CH3L5)-(CH3COO)2]BPh4. Counter ions, solvent molecules
and hydrogen atom are omitted for clarity.
complex crystallized with one Zn(II) ion six coordinate and the
other in a five coordinate environment (Fig. 1). The vinylbenzyl
group is arranged in such a way that it shields one site on the
second Zn(II) ion leaving this site almost square pyramidal.
Crystallographic data and selected bond lengths and angles
are displayed in Tables 1 and 2, respectively.
The comparison of the bond lengths to the symmetrical
29
counterpart [Zn2(CH3L2)(CH3COO)2]PF6
show that the
O
Phenol–Zn bonds are similar with a mean of 2.024 Å for
Crystal structures
the asymmetric and 2.015 Å for the symmetric counterpart.
The methyl ether distance is shorter (2.320 Å) than in the sym-
Both dizinc(II) complexes crystallised in the triclinic space
−
ˉ
group P1. The zinc complex with CH3L4 crystallised as an metric complex (mean 2.432 Å). The Ntert–Zn (2.124 Å) and
asymmetric complex with one six-coordinate Zn(II) and one Npy–Zn (2.150 Å) distances are similar to the corresponding
distances in [Zn2(CH3L2)(CH3COO)2]PF6 (2.186 and 2.1355 Å).
Zn(II) in a five-coordinate (distorted trigonal bipyramidal)
environment (Fig. 1). Two μ2-CH3COO-κ2O:O′ anions but no The Oacetate–Zn bond lengths in [Zn2(CH3L4)(CH3COO)2]PF6
additional water or solvent molecules are bound to the vary from 1.971–2.058 Å while those from [Zn2(CH3L2)-
(CH3COO)2]PF6 encompass a range from 1.992 to 2.076 Å. The
complex in the solid state. The crystal structure shows the pres-
ence of two isomers, one present as a minor component in Zn–OPhenol–Zn angle is reduced to 108.17° in the asymmetric
one of the two asymmetric units. In the major isomer, the pyri- complex (109.70° with the CH3HL2 ligand). Zn–Zn distances
are as well slightly reduced in the asymmetric complex (3.222,
dine nitrogens are trans (anti) to each other while in the minor
isomer the pyridines are cis (syn) in respect of the phenolic 3.271 Å), as opposed to 3.296 Å in the symmetric counterpart.
oxygen.39 The ORTEP40 representation of the asymmetric unit
which contains two [Zn2(CH3L4)(CH3COO)2]PF6 complex enti-
The ligands create a geometry which make the zinc com-
plexes appropriate structural mimics of phosphoesterase
ties is shown (Fig. 1). The [Zn2(CH3L5)(CH3COO)2]BPh4 enzymes such as GpdQ.
A
close up view of the first
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 1 Crystallographic data for [Zn2(CH3L4)(CH3COO)2]PF6 and [Zn2(CH3L5)-
(CH3COO)2]BPh4
[Zn2(CH3L4)-
(CH3COO)2]PF6
[Zn2(CH3L5)-
(CH3COO)2]BPh4
Empirical formula
C28H35N4O6Zn2,
P1F6
799.31
C37H39N4O6Zn2,
C24H20B, C3H8O
1145.77
Formula weight
Temperature [K]
Wavelength [Å]
Crystal system
Space group
a [Å]
b [Å]
c [Å]
α [°]
β [°]
293(2)
150(2)
0.71073 (MoKα)
Triclinic
ˉ
P1
13.137(5)
13.669(6)
20.822(8)
85.047(3)
85.468(3)
66.511(4)
3412(2)
1.5418 (CuKα)
Triclinic
ˉ
P1
13.3209(6)
15.0129(7)
16.4893(6)
91.810(3)
108.350(4)
110.363(4)
2897.2(2)
2
Fig. 2 First coordination sphere of the two Zn(II) ions in the [Zn2(CH3L4)-
(CH3COO)2]PF6 complex (top left), [Zn2(CH3L5)(CH3COO)2]BPh4 (top right) and
in GpdQ (below).
γ [°]
Volume [Å3]
Z
4
ρ [mg m−3
]
1.556
1.530
1632
1.313
1.466
1199.8
νsym, respectively)41 and for the PF6 counter ion (832,
−
μ [mm−1
F(000)
]
555 cm−1) in addition to the C–O stretching vibration of the
methyl ether at 1022 cm−1 and the N–H stretch at 3307 cm−1
.
Θ range for data collection [°]
1.63 to 24.14
33 208
10 796 (0.0290)
1.032
R1 = 0.0661,
wR2 = 0.1831
R1 = 0.0779,
wR2 = 0.1940
925903
2.86–61.63
26 492
8953 (0.0277)
1.035
R1 = 0.0480,
wR2 = 0.1321
R1 = 0.0557,
wR2 = 0.1401
925904
Reflections collected
For the [Zn2(CH3L5)(CH3COO)2]PF6 complex the stretches at
1594 and 1429 cm−1 also suggest the presence of bidentate
bridging acetate.41 Moreover, C–H (2925, 2842 cm−1), C–O
(1022 cm−1) and bands typical for aromatic functionalities
(766, 662 cm−1) are similar to those observed for the complex
[Zn2(CH3L4)(CH3COO)2]PF6 (C–H 2925 cm−1, aromatic 768 and
663 cm−1). It can thus be assumed that a similar coordination
environment with two bridging acetates for [Zn2(CH3L5)-
Independent reflections (Rint
)
GOOF on F2
Final R indices
[I > 2σ(I)]
R indices (all data)
CCDC number
Table 2 Selected bond lengths [Å] and angles [°] for [Zn2(CH3L4)(CH3COO)2]- (CH3COO)2]PF6 is found in the solid state.
PF6 and [Zn2(CH3L5)(CH3COO)2]BPh4
Mass spectrometry
[[Zn2(CH3L4)(CH3COO)2]-
PF6
[Zn2(CH3L5)(CH3COO)2]-
BPh4
a
The electrospray mass spectra of the two zinc complexes were
recorded in MeOH and MeCN. The spectrum of [Zn2(CH3L4)-
(CH3COO)2]PF6 measured in MeOH is displayed in Fig. S1a†
and shows a peak at m/z 655.1 (100%) (calc. m/z = 653.11
(100%), 655.11 (96.2%), [C28H35N4O6Zn2]+) corresponding to
one negatively charged ligand, two Zn(II) ions and two acetates.
The major ion peak at m/z 641.1 is assigned to [Zn2(CH3L4)-
(CH3COO)(HCOO)]+ (calc. m/z = 641.1 (100%), 639.1 (98.7%),
637.1 (81.8%), [C27H33N4O6Zn2]+).
The spectrum recorded in MeCN showed multiple doubly
charged species, the major ion peak at m/z 749.1 is assigned to
a complex with two CH3L4−, four Zn(II), four acetates, six water
molecules and two MeCN molecules (calc. m/z 749.17 (100%),
748.17 (79.2%), [C60H88N10O18Zn4]2+, Fig. S1b†).
N1–Zn1
N2–Zn1
N3–Zn2
N4–Zn2
O1–Zn1
O1–Zn2
O2–Zn2
O3–Zn1
O4–Zn2
O5–Zn1
O6–Zn2
Zn1–O1–Zn2
Zn1–Zn2
2.133(5)
2.121(4)
2.182(4)
2.121(4)
2.026(4)
2.014(4)
2.320(4)
1.984(4)
2.112(4)
1.998 (4)
2.015(4)
108.17(16)
3.2714(12)
2.192(3)
2.108(3)
2.184(3)
2.131(3)
2.016(2)
2.014(2)
2.370(2)
2.020(2)
2.029(2)
1.986(2)
2.072(2)
106.98(9)
3.2391(6)
a Only the bond lengths and angles of the zinc complex labeled with
‘a’ (e.g. Zn1a) are displayed.
The major ion of the [Zn2(CH3L5)(CH3COO)2]PF6 complex
measured in MeOH corresponds to
a
[Zn2(CH3L5)-
coordination sphere of the two Zn(II) ions in the complexes is
displayed in Fig. 2 along with the first coordination sphere
donor atoms in GpdQ.27,28 While the acetate ligands do not
accurately mimic the bridging water molecules, the general
structure, geometry and donor-atoms of GpdQ are reproduced
well in the crystal structures.
(MeO)2(H2O)2]+ species (calc. for C35H47N4O6Zn2 m/z = 749.2
(100%), 751.2 (96.1%), found m/z = 751.1 (100%), 749.1 (95%),
Fig. S2a†). Higher molecular weight species with m/z > 800 are
also found and are most likely due to coordination of
additional solvent and anionic components. A large mass peak
due to free ligand is also found at m/z = 523.2 (CH3HL5 + H).
The mass spectrum for [Zn2(CH3L5)(CH3COO)2]PF6 in MeCN
shows at least four major species (Fig. S2b†): (i) observed m/z =
+
Infrared spectroscopy
The [Zn2(CH3L4)(CH3COO)2]PF6 complex exhibited stretches 851.3 assigned to [Zn2(CH3L5)(CH3COO)2(MeCN)2]+ (calc.
typical for bridging acetate (1596 and 1422 cm−1, νasym and m/z = 851.23 (100%), 849.23 (85.5%), [C41H49N6O6Zn2]+); (ii)
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
observed m/z
Paper
=
769.1 corresponds to
a
[Zn2(CH3L5)- two phosphate substrate molecules readily and show that the
(CH3COO)2]+ species (calc. 769.17 (100%), 767.18 (87%), acetates are not inhibiting substrate binding.
[C37H43N4O6Zn2]+; (iii) a peak at m/z = 723.1 is due to the loss
NMR spectroscopy
of acetate and coordination of water (calc. m/z = 721.17
(100%), 732.17 (96.6%), [C33H43N4O6Zn2]+), and (iv) the peak
NMR spectra were recorded for both complexes in a range of
at 653.1 can be attributed to the loss of the vinylbenzyl group
(calc. m/z 653.11 (100%), 655.11 (96.2%), [C28H35N4O6Zn2]+).
To investigate relevant species occurring during phosphate
ester hydrolysis, the mass spectra were measured in MeCN in
the presence of diphenyl phosphate (DPP). The [Zn2(CH3L4)-
(CH3COO)2]PF6 complex loses acetate ligands under mass
spectral conditions and forms, even in the presence of only
one equivalent DPP (Fig. S3a†), a complex with two DPP bound
(100%), illustrating the high affinity of this complex for phos-
phoester substrates. In addition, a small peak (∼25%) arising
from complex with one DPP, one hydroxide and an acetonitrile
solvent molecule is observed.
For [Zn2(CH3L5)(CH3COO)2]PF6 the mass spectrum in the
presence of one equivalent DPP is different (Fig. S3b†). The
base peak arises from a species with one DPP, a hydroxide and
MeCN; two peaks complexes arising from intact complex with
two acetates (29%) or two DPP molecules (47%) bound are
present. It is proposed that under the conditions of the mass
spectral measurements [Zn2(CH3L4)(CH3COO)2]PF6 has a high
affinity for diphenylphosphate (Fig. 3a), while [Zn2(CH3L5)-
(CH3COO)2]PF6, due to the bulky pendant vinylbenzyl group,
needs an excess of substrate to form species with one or two
DPP bound. Fig. 3b illustrates how a large excess of DPP
changes the species distribution for [Zn2(CH3L5)(CH3COO)2]-
PF6, the main species now being [Zn2(CH3L5)(DPP)2]+. These
experiments demonstrate that the complexes can bind one or
solvents. Fig. S4 and S5† show a comparison between the free
CH3HL4 ligand and its zinc(II) complex. A low intensity struc-
ture, not stemming from free ligand, is present (marked with *).
It is possible that these resonances represent either the
minor isomer observed in the X-ray crystal structure, or are
due to the ether arm not coordinated; a combination of both
is possible.
The aliphatic region is shown in Fig. S5† and demonstrates
how the protons of the CH2-groups in the free ligand become
inequivalent upon zinc binding, in accordance with the lack of
a C2-symmetry axis or other symmetric features in the
complex. The ratio of low intensity isomer (* in Fig. S4†) to
the major isomer depended strongly on the water content in
the mixture.
In an experiment where aliquots of D2O were titrated into a
solution of the complex in CD3CN, the proportion of the low
intensity species first reduced and then another set of low
intensity resonances appeared at higher field (Fig. S6†), high-
lighting the importance of the medium on species formation
and distribution.
To monitor how well CH3HL5 would assemble a catalyti-
cally competent dinuclear Zn(II) centre a 1H-NMR titration
experiment was conducted. Fig. S7† shows the spectral
changes after successive addition of one and two equivalents
of zinc(II) acetate to a solution of CH3HL5 in CD3CN. After
each addition the solvent in the NMR tube was brought to
Fig. 3 (a) Mass spectrum of [Zn2(CH3L4)(CH3COO)2]PF6 in the presence of excess (>25 eq.) diphenylphosphate in MeCN. Final concentration of complex and sub-
strate were 10 and 250 µM, respectively. (b) Mass spectrum of [Zn2(CH3L5)(CH3COO)2]PF6 in the presence of excess (>25 eq.) DPP in MeCN. Final concentration of
complex and substrate were 10 and 250 µM, respectively.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
reflux for 5 minutes prior to spectra recording. After addition species such as monodentately and bidentately bridged ones,
of one equivalent of zinc(II) acetate the resonances of the vinyl as well as free DPP in the mixture (see Gaussian deconvolution
protons had experienced negligible shifts suggesting that the in Fig. S10a†). Upon addition of a further equivalent of DPP to
ligand binds zinc(II) ions in a stepwise manner, occupying first the mixture the resonance arising from free DPP becomes
the less sterically hindered site. It is also apparent that after more clearly resolved (Fig. S10b†). This resonance becomes the
the addition of two equivalents zinc(II) acetate, the solution major species upon addition of excess DPP (see spectrum c in
still contains partially uncomplexed ligand. Addition of Fig. S10†). It should be noted that initially all NMR-titration
another equivalent of zinc(II) acetate resulted in a set of two experiments were conducted in the same solvent mixture as
independent resonances for the vinyl-protons in the spectrum used for the kinetic studies (1 : 1 water–acetonitrile), however,
of the fully assembled dinuclear complex, suggesting two iso- the spectra were in general noisier than in pure acetonitrile
meric forms in solution. Substrate binding was investigated which is attributed to the poor solubility of the ligands in the
with 31P NMR spectroscopy. The addition of one equivalent water–acetonitrile mixture. An example titration of zinc acetate
DPP to a solution of [Zn2(CH3L4)(CH3COO)2]PF6 in CD3CN and subsequently diphenyl phosphate to the ligand CH3HL5
shows multiple phosphate ester species but no resonance from in D2O–CD3CN, is included in the ESI (Fig. S11†).
free DPP (expected ∼−12 ppm) in accordance with the electro-
Immobilisation of [Zn2(CH3L4)(CH3COO)2]+ on Merrifield
resin
spray mass spectra experiments which showed high affinity for
this substrate (Fig. S8a†). In accordance with the mass spec-
trum of complex having one or two species bound, the
different signals observed in the 31P-spectrum are assigned to
a mix of species with one DPP monodentately or bidentately
bound (−8.0 and −8.3 ppm), and two DPP molecules also
mono- or bidentately bound (−9.1 and −9.5 ppm). Addition of
a second equivalent of DPP leads to an increase in intensity of
the two resonances at −9.5 and −9.1 ppm (Fig. S8b†), support-
ing the proposal of two DPP bound to the zinc complex in
either a either mono- or bidentate fashion. Further addition of
DPP gives rise to a signal at −12.2 ppm attributed to free DPP
(Fig. S8c†). In the 1H NMR spectrum of [Zn2(CH3L5)-
(CH3OO)2]+ with one and two equivalents of substrate
(Fig. S9c, d†) the signals are broadened suggesting a bridging
DPP conformation.42 One of the two sets of resonances attribu-
ted to the vinyl proton loses intensity upon the addition of one
or two DPP molecules to the NMR mixture (Fig. S9c–e†). Given
the bulky nature of the vinyl benzyl moiety these results
support the notion that in solution the vinyl benzyl group
exists in two conformations one directed towards the Zn(II)
centre and one pointing in the opposite direction, the latter
not being influenced by substrate binding. Moreover, the vinyl
benzyl group displays, after the addition of five equivalents
M-CH3HL4 was prepared by reacting the secondary amine
group of CH3HL4 with the chloromethylated Merrifield resin
(MR). The dizinc(II) complex was generated subsequently by
adding excess zinc acetate to the ligand substituted resin. The
IR spectrum (Fig. S12†) of M-[Zn2(CH3HL4)(CH3COO)2]+
clearly shows the appearance of the two asymmetric and sym-
metric acetate stretches (marked with *) when compared to the
spectrum of unsubstituted MR. Broad absorption bands at
3300 cm−1 indicate the presence of OH moieties probably in
the form of Zn(II) bound water or methanol. IR bands assigned
to symmetric and asymmetric acetate vibrations were not
observed when unsubstituted resin was treated with zinc
acetate. The XPS of commercially available MR showed, in
addition to the expected peaks from carbon and oxygen, the
presence of tin oxide;‡ for M-CH3HL4 peaks attributed to Si 2p
electrons were attributed to silicone oil contamination.44 The
XPS of the MR resin treated with Zn(OAc)2 showed only a small
amount of zinc (Fig. S13†) and no indication of the presence
of nitrogen. The XPS spectra of M-CH3HL4 showed nitrogen
peaks in addition to the peaks from MR (Fig. S13 and S14†).
The XPS spectra of M-[Zn2(CH3L4)(CH3COO)2]+ showed the
two peaks typical of Zn 2p electrons (Fig. 4), confirming the
uptake of zinc ions. Only one type of nitrogen is resolved, its
binding energy typical for tertiary nitrogen (pyridine, tertiary
amine).45,46 A small peak for chlorine (Cl 2p) was found in all
spectra (not labelled) and a detailed analysis suggested the
presence of both ionic chloride and carbon-bound chloride
(Fig. 4). Based on microanalytical data for M-CH3HL4 and
comparison with data for the commercially available resin a
catalyst loading of 1.45 mmol g−1 was determined.
1
DPP (Fig. S9e†), H-chemical shifts similar to those found in
the spectrum of the free ligand (Fig. S9a†). This implies that
the vinyl benzyl group points away (and is not further influ-
enced) from the zinc(II) ions when substrate is bound.
When comparing the 13C NMR spectra of free complex to
that in the presence of five equivalents of DPP it is observed
that the former spectrum has two sets of vinyl resonances at
114.9 and 115.2 ppm while the latter exhibits only one at
115.2 ppm. Moreover, the carbonylOAc resonance shifts upon
addition of DPP to the complex from 179.0 ppm (bridging
acetate) to 173.1 ppm (free acetate).43 The methylOAc resonance
also exhibits a shift (from 23.1 to 20.8 ppm), suggesting that
the acetates remain associated with the complex in solution,
but are readily displaced by phosphoester substrates.
Phosphoesterase Activity
The pH dependence of BDNPP hydrolysis was measured from
pH 4.75–10 (Fig. 5a). The activity for both [Zn2(CH3L4)-
(CH3CO2)2]+ and [Zn2(CH3L5)(CH3CO2)2]+ was dramatically
The 31P NMR spectrum of the [Zn2(CH3L5)(CH3OO)2]+
complex in the presence of one equivalent DPP shows one
‡The supplier (Sigma-Aldrich) did not disclose any information whether tin
oxide was used in the manufacturing process of this MR so the origin of this
broad resonance. This resonance possibly arises from multiple contamination is currently unknown.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Fig. 4 (a) XPS survey spectrum of M-[Zn2(CH3L4)(CH3COO)2]+, Gaussian deconvoluted (b) Cl 2p, (c) N 1s, (d) C 1s, (e) Zn 2p and (f) O 1s spectra of M-[Zn2(CH3L4)(CH3COO)2]+.
was observed. The leaching of zinc out of the complex at pH
values higher than 8 had been previously reported by Meyer
et al.47 The kinetic pKa values determined for [Zn2(CH3L4)-
(CH3COO)2]+ and [Zn2(CH3L5)(CH3COO)2]+ (7.39
0.19 and
0.19, respectively) are typical for a zinc-bound water
7.50
molecule.29,48 The dependence of BDNPP concentration on the
rate of hydrolysis for both complexes (conducted at optimal
pH 7.66) followed Michaelis–Menten kinetics (Fig. 5b). The
activity of [Zn2(CH3L5)(CH3COO)2]+ (kcat = 0.97 0.21 × 10−3
s−1, KM 7.01 2.57 mM; kcat/KM = 0.14 s−1 mM−1) is lower than
for the less sterically hindered complex [Zn2(CH3L4)-
(CH3COO)2]+ (kcat = 2.45
1.74 mM; kcat/KM = 0.26 s−1 M−1). Compared to Zn2-GpdQ
(kcat/KM = 50 s−1 M−1 28
the complexes are ∼200 and 360 times
0.27 × 10−3 −1, KM = 9.48
s
)
less efficient than the enzyme, respectively. The substrate affinity
for [Zn2(CH3L4)(CH3COO)2]+ is slightly higher than for the
vinyl benzyl complex, although the error for KM values is quite
large. The phosphoesterase activity of [Zn2(CH3L4)(CH3CO2)2]+
appeared typical of that reported for similar types of com-
plexes,29 suggesting that in this case the asymmetric nature of
the ligand has no influence on activity.1–3 It is clear, however,
that the steric bulk engendered by CH3HL5 does influence the
activity.
The phosphatase-like activity of M-[Zn2(CH3HL4)-
(CH3COO)2]+ was measured at pH values ranging from 4.75 to
10.5 using BDNPP. The pH dependence of BDNPP hydrolysis
for M-[Zn2(CH3L4)(CH3COO)2]+ is shown (Fig. 6a). The data
were consistent with the presence of one protonation equili-
brium (pKa = 7.61 0.71) and were thus fit to an equation
derived for a monoprotic system. Above pH 8 the resin leached
Fig. 5 (a) pH dependence of the rate of BDNPP hydrolysis by [Zn2(CH3L4)-
+
(CH3COO)2]+
(
) and [Zn2(CH3L5)(CH3COO)2]
(
○
●
), and (auto)hydrolysis by free
◊
Zn(II) ( ). (b) Substrate concentration dependence on the catalytic rate for
+
[Zn2(CH3L4)(CH3COO)2]+
(
(
○
) and [Zn2(CH3L5)(CH3COO)2]
●
) and (auto)hydro-
◊
lysis by free Zn(II) ( ).
reduced above pH 8 and the data became irreproducible; in zinc as was apparent by the presence of a fine white precipi-
some cases a white precipitate, presumed to be zinc hydroxide, tate.47 Due to the resulting irreproducibility the data above pH
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
herein suggest that incorporation of these types of complexes
onto an insoluble supports, in this case Merrifield resin,
results in a system exhibiting phosphoesterase activity. The
challenge now is two-fold, (i) to enhance the activity of the bio-
mimetic complexes towards environmentally relevant (organo-
phosphate pesticide) substrates, and (ii) to pursue the
immobilisation of active and robust metallohydrolases biomi-
metics for potential application in bioremediation of environ-
ments contaminated with phosphate ester pesticides.
Acknowledgements
This work was funded by a grant from the Australian Research
Council (DP0986613). LJD was funded by IPRS and UQ gradu-
ate school scholarships.
Notes and references
1 H. Carlsson, E. Nordlander and M. Jarenmark, C. R. Chim.,
2007, 10, 433–462.
2 M. Jarenmark, S. Kappen, M. Haukka and E. Nordlander,
Dalton Trans., 2008, 993–996.
3 M. Jarenmark, E. Csapo, J. Singh, S. Wockel, E. Farkas,
F. Meyer, M. Haukka and E. Nordlander, Dalton Trans.,
2010, 39, 8183–8194.
4 E. Lambert, B. Chabut, S. Chardon-Noblat, A. Deronzier,
G. Chottard, A. Bousseksou, J.-P. Tuchagues, J. Laugier,
M. Bardet and J.-M. Latour, J. Am. Chem. Soc., 1997, 119,
9424–9437.
Fig. 6 (a) pH dependence of BDNPP hydrolysis by M-[Zn2(CH3L4)(CH3COO)2]+
▲
(
), (b) Michaelis–Menten profile for BDNPP hydrolysis by M-[Zn2(CH3L4)-
(CH3COO)2]+ at pH 8.
8 are not displayed. A Michaelis–Menten profile was recorded
for this complex at pH 8 with BDNPP, following the same pro-
5 M. Lanznaster, A. Neves, A. J. Bortoluzzi, B. Szpoganicz and
E. Schwingel, Inorg. Chem., 2002, 41, 5641–5643.
6 C. Belle, I. Gautier-Luneau, L. Karmazin, J.-L. Pierre,
S. Albedyhl, B. Krebs and M. Bonin, Eur. J. Inorg. Chem.,
2002, 2002, 3087–3090.
cedure as used for the pH dependence (Fig. 6b). Here, Vmax
=
8
4 × 10−9 M s−1 with KM = 2 1 mM and, based on an esti-
mated catalyst loading of 1.45 mmol g−1, kcat = 11 6 × 10−6 s−1
.
7 C. Belle, I. Gautier-Luneau, J. L. Pierre, C. Scheer and
E. SaintAman, Inorg. Chem., 1996, 35, 3706–3708.
8 S. Albedyhl, M. T. Averbuch-Pouchot, C. Belle, B. Krebs,
J. L. Pierre, E. Saint-Aman and S. Torelli, Eur. J. Inorg.
Chem., 2001, 2001, 1457–1464.
Conclusion
The crystal structures of [Zn2(CH3L4)(CH3COO)2]PF6 and
[Zn2(CH3L5)(CH3COO)2]PF6 show one vacant site for coordi-
nation of nucleophiles and solvent molecules however, the
asymmetric ligand sphere did not appear to have a beneficial
9 S. Albedyhl, D. Schnieders, A. Jancso, T. Gajda and
B. Krebs, Eur. J. Inorg. Chem., 2002, 2002, 1400–1409.
effect on catalytic activity. The complex [Zn2(CH3L5)- 10 J. H. Satcher, M. W. Droege, M. M. Olmstead and
(CH3COO)2]PF6 with the more bulky ligand has an impaired A. L. Balch, Inorg. Chem., 2001, 40, 1454–1458.
substrate affinity as shown by mass spectrometry and 31P NMR 11 R. R. Buchholz, M. E. Etienne, A. Dorgelo, R. E. Mirams,
and the diminished catalytic activity is attributed to steric
effects of the vinylbenzyl group. Nevertheless, both complexes
are competent mimics of the phosphoesterase enzyme GpdQ.
S. J. Smith, S. Y. Chow, L. R. Hanton, G. B. Jameson,
G. Schenk and L. R. Gahan, Dalton Trans., 2008,
6045–6054.
In addition, the complexes are also good structural models for 12 F. Meyer, E. Kaifer, P. Kircher, K. Heinze and H. Pritzkow,
this enzyme. [Zn2(CH3L4)(CH3COO)2]PF6 was immobilised on Chem.–Eur. J., 1999, 5, 1617–1630.
Merrifield (polystyrene) resin and showed activity towards 13 H. Adams, S. Clunas, D. E. Fenton and D. N. Towers,
organophosphoester substrate. [Zn2(CH3L5)(CH3COO)2]PF6 J. Chem. Soc., Dalton Trans., 2002, 3933–3935.
with the vinyl benzyl arm can be described as the ‘monomer’ 14 I. A. Koval, D. Pursche, A. F. Stassen, P. Gamez, B. Krebs
of the immobilised complex. Moreover, the results presented
and J. Reedijk, Eur. J. Inorg. Chem., 2003, 2003, 1669–1674.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
15 L. Dubois, R. Caspar, L. Jacquamet, P.-E. Petit, 31 N. V. Kaminskaia, B. Spingler and S. J. Lippard, J. Am.
M.-F. Charlot, C. Baffert, M.-N. Collomb, A. Deronzier and
J.-M. Latour, Inorg. Chem., 2003, 42, 4817–4827.
16 L. Dubois, D.-F. Xiang, X.-S. Tan, J. Pécaut, P. Jones,
S. Baudron, L. Le Pape, J.-M. Latour, C. Baffert, S. Chardon-
Chem. Soc., 2000, 122, 6411–6422.
32 I. H. Segel, Enzyme Kinetics: Behavior and analysis of rapid
equilibrium and steady state enzyme systems, Wiley-Inter-
science, New York, 1975.
Noblat, M.-N. Collomb and A. Deronzier, Inorg. Chem., 33 A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi,
2003, 42, 750–760.
17 C. Piovezan, R. Jovito, A. J. Bortoluzzi, H. n. Terenzi,
M. C. Burla, G. Polidori and M. Camalli, J. Appl. Crystal-
logr., 1994, 27, 435.
F. L. Fischer, P. C. Severino, C. T. Pich, G. G. Azzolini, 34 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystal-
R. A. Peralta, L. M. Rossi and A. Neves, Inorg. Chem., 2010,
49, 2580–2582.
18 Y. L. M. Zee, L. R. Gahan and G. Schenk, Aust. J. Chem.,
2011, 64, 258–264.
19 J. Yang, P. Li and L. Wang, Synthesis, 2011, 1295–1301.
20 K. Li, J. Zhang, Z.-W. Zhang, Y.-Z. Xiang, H.-H. Lin and
X.-Q. Yu, J. Appl. Polym. Sci., 2009, 111, 2485–2492.
logr., 2008, 64, 112–122.
35 G. M. Sheldrick, SHELXL97: Program for the refinement of
crystal structures, University of Gottingen, Germany, 1997.
36 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837–838.
37 G.-C. Sun, Z.-H. He, Z.-J. Li, X.-D. Yuan, Z.-J. Yang,
G.-X. Wang, L.-F. Wang and C.-R. Liu, Molecules, 2001, 6,
1001–1005.
21 P. Styring, C. Grindon and C. M. Fisher, Catal. Lett., 2001, 38 A. K. Boudalis, R. E. Aston, S. J. Smith, R. E. Mirams,
77, 219–225.
M. J. Riley, G. Schenk, A. G. Blackman, L. R. Hanton and
L. R. Gahan, Dalton Trans., 2007, 5132–5139.
39 P. Comba, G. N. DeIuliis, G. A. Lawrance, S. M. Luther,
M. Maeder, A. L. Nolan, M. J. Robertson and P. Turner,
Dalton Trans., 2003, 2188–2193.
22 Q. H. Xia, H. Q. Ge, C. P. Ye, Z. M. Liu and K. X. Su, Chem.
Rev., 2005, 105, 1603–1662.
23 V. Sciannamea, A. Debuigne, Y. Piette, R. Jerome and
C. Detrembleur, Chem. Commun., 2006, 40, 4180–4182.
24 E. Ghanem, Y. Li, C. Xu and F. M. Raushel, Biochemistry, 40 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
2007, 46, 9032–9040.
25 K. S. Hadler, L. R. Gahan, D. L. Ollis and G. Schenk,
J. Inorg. Biochem., 2010, 104, 211–213.
26 F. Ely, K. S. Hadler, L. R. Gahan, L. W. Guddat, D. L. Ollis
and G. Schenk, Biochem. J., 2010, 432, 565–573.
41 K. Nakamoto, Infrared and Raman spectra of inorganic and
coordination compounds, Wiley, NY, 1978.
42 K. Selmeczi, C. Michel, A. Milet, I. Gautier-Luneau,
C. Philouze, J.-L. Pierre, D. Schnieders, A. Rompel and
C. Belle, Chem.–Eur. J., 2007, 13, 9093–9106.
27 K. S. Hadler, N. Mitić, F. Ely, G. R. Hanson, L. R. Gahan, 43 H. E. Gottlieb, V. Kotlyar and A. Nudelman, J. Org. Chem.,
J. A. Larrabee, D. L. Ollis and G. Schenk, J. Am. Chem. Soc.,
2009, 131, 11900–11908.
28 C. J. Jackson, K. S. Hadler, P. D. Carr, A. J. Oakley, S. Yip,
1997, 62, 7512–7515.
44 G. L. Weissler and R. W. Carlson, Vacuum physics and tech-
nology, Academic Press, New York, 1979.
G. Schenk and D. L. Ollis, Acta Crystallogr., Sect. F: Struct. 45 NIST, X-ray Photoelectron Spectroscopy Database, Version
Biol. Cryst. Commun., 2008, 64, 681–685. 4.1, http://srdata.nist.gov/xps/ Accessed 20.2.13.
29 L. J. Daumann, K. E. Dalle, G. Schenk, R. P. McGeary, 46 C. D. Wanger, W. M. Riggs, L. E. Davis, J. F. Moulder and
P. V. Bernhardt, D. L. Ollis and L. R. Gahan, Dalton Trans.,
2012, 41, 1695–1708.
30 R. A. Peralta, A. J. Bortoluzzi, B. de Souza, R. Jovito,
G. E. Muilenberg, Handbook of X-ray Photoelectron Spec-
troscopy, Perkin-Elmer Corp., Physical Electronics Division,
Minnesota, USA, 1979.
F. R. Xavier, R. A. A. Couto, A. Casellato, F. Nome, A. Dick, 47 B. Bauer-Siebenlist, F. Meyer, E. Farkas, D. Vidovic,
L. R. Gahan, G. Schenk, G. R. Hanson, F. C. S. de Paula,
E. C. Pereira-Maia, S. d. P. Machado, P. C. Severino, C. Pich,
J. A. Cuesta-Seijo, R. Herbst-Irmer and H. Pritzkow, Inorg.
Chem., 2004, 43, 4189–4202.
T. Bortolotto, H. Terenzi, E. E. Castellano, A. Neves and 48 J. Burgess, Metal ions in solution, Halsted Press, Chichester,
M. J. Riley, Inorg. Chem., 2010, 49, 11421–11438.
1978.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.