Glycosidase- and β-lactamase-like activity of dinuclear copper(II) patellamide complexes

Article abstract of DOI:10.1016/j.jinorgbio.2016.02.014

Prochloron, a blue-green algae belonging to ancient prokaryotes, produces, like other cyanobacteria, cyclic pseudo-peptides, which are also found in its obligate symbiont ascidiae (Lissoclinum patellum). Although research has focused for some time on the

Full text of DOI:10.1016/j.jinorgbio.2016.02.014

Journal of Inorganic Biochemistry 159 (2016) 70–75
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Glycosidase- and β-lactamase-like activity of dinuclear copper(II)
☆
patellamide complexes
⁎
Peter Comba , Annika Eisenschmidt, Nora Kipper, Jasmin Schießl
University of Heidelberg, Institute of Inorganic Chemistry and Interdisciplinary Center for Scientific Computing, Im Neuenheimer Feld 270, D-69120 Heidelberg, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Prochloron, a blue-green algae belonging to ancient prokaryotes, produces, like other cyanobacteria, cyclic
pseudo-peptides, which are also found in its obligate symbiont ascidiae (Lissoclinum patellum). Although research
has focused for some time on the putative metabolic function of these cyclic peptides, to date it is still not
understood. Their role might be connected to the increased concentrations of divalent metal ions, especially
CuII, found in ascidiae. Dinuclear copper(II) complexes of cyclic pseudo-peptides revealed a broad hydrolytic
capacity, including carboanhydrase and phosphatase activity. This study reports their β-lactamase as well
Received 20 November 2015
Received in revised form 27 January 2016
Accepted 10 February 2016
Available online 13 February 2016
Keywords:
as α- and β-glycosidase activity with kcat = (11.34
0.91)ˑ10−4 s−1 for β-lactamase, kcat = (1.55
Cyclic peptides
Copper(II) complexes
Ascidian
0.13)ˑ10−4 s−1 for α-glycosidase and kcat = (1.22 0.09)ˑ10−4 s−1 for β-glycosidase activity.
© 2016 Elsevier Inc. All rights reserved.
Prochloron
Glycosidase
Lactamase
1. Introduction
catabolic role, namely glycosidase-like activity and a potential antibiotic
role as a β-lactamase-like enzyme were chosen.
Patellamides like ascidiacyclamide or patellamide D (see Fig. 1), are
cyclic peptides, naturally produced via a ribosomal pathway by the
cyanobacterium Prochloron [1,2]. These cyclic pseudo-octapeptides are
found in high quantities in their obligate symbiont L. patellum that
moreover exhibits high concentrations of copper(II) compared to the
surrounding sea water [3]. The four azole-N as well as the four amide-
N atoms are possible donors for metal ion coordination. Due to the
low yield of biosynthesis and isolation from natural sources, routes
were developed to synthetically produce cyclic pseudo-peptides [4–6].
A number of different model peptides were produced, in the following
referred to as “pat”, differing foremost in the stereochemistry and
azole heterocycles (see Fig. 1), and have been extensively studied
towards their copper(II) complexing behavior, exhibiting a preference
for dinuclear complexes in the NImidazole-NAmide-NImidazole binding site
[3]. It could be shown that the dinuclear copper(II) complexes of the
synthetic peptides are catalytically active in the physiological pH
range as carboanhydrase- and phosphatase-like enzyme models [7,8].
In this study, the dinuclear copper(II) complexes of the cyclic pseu-
do-octapeptides H4pat1 and H4pat2 (Fig. 2), the solution structures of
which are known from a combination of EPR spectroscopy, spectra sim-
ulations and molecular modeling, [3,6,9] are investigated concerning
their catalytic activity at higher pH values. For that, a hypothetically
1.1. Glycoside hydrolases
Glycosyl transfer reactions are essential in biology and were
therefore targets for model catalyst research. However, for a long time
only substrates with an additional binding site for Lewis acids like
Al3+, Zn2+ and Fe3+ were investigated. Only recently, few studies on
dinuclear copper(II) complexes as enzyme models for glycosidase
appeared [10–14]. According to their structure, glycosidases are divided
in different subfamilies. Most of these have an organic active site re-
sponsible for the catalytic glycoside cleavage depicted in Scheme 1.
However, the 4th family was shown to be metal ion dependent [15].
As the use of these enzymes for glycosyl transfer reactions is limited
by their low availability [16], the design and evaluation of new glycosi-
dase mimics with increased selectivity and reaction yields is desirable
[10]. The development of model compounds that catalyze glycosyl
transfer reactions can be interesting for the synthesis of carbohydrates,
which are needed as food fillers and may also be of interest as antiviral
agents or new antibiotics.
1.2. β-lactam hydrolases
Six different subtypes of β-lactam antibiotics, penams, cephems,
monobactams, clavams, penems and carbapenems, are known [17]. Five
of them are readily hydrolyzed by metallo-β-lactamases, only mono-
bactams withstand the predominantly dinuclear zinc(II) complexes.
☆
In memory of Professor Graeme Hanson and his contributions to Bioinorganic
Chemistry and Electron Paramagnetic Resonance.
⁎
Corresponding author.
http://dx.doi.org/10.1016/j.jinorgbio.2016.02.014
0162-0134/© 2016 Elsevier Inc. All rights reserved.

P. Comba et al. / Journal of Inorganic Biochemistry 159 (2016) 70–75
71
Fig. 1. Ascidiacyclamide and synthetic model ligands H4pat1 and H4pat2; the color code in Ascidiacyclamide shows the amino acid building blocks of one half of the molecule, also relevant
in the biosynthesis and showing why these are best named cyclic pseudo-octapeptides.
Fig. 2. Computed structures of hydroxido-bridged [Cu2(H2pat2)(OH)(H2O)2]+ (1) (top) and [Cu2(H2pat2)(OH)(H2O)3]+ (2) (bottom); orange: CuII, red: O, blue: N.
Both zinc(II) ions bind water and, due to their Lewis acidity, decrease its
pKa, resulting in a hydroxido complex. The proposed mechanism of the
cleavage is shown in Scheme 2 [17].
2. Results and discussion
2.1. Syntheses and structures of the dinuclear complexes
Upon cleavage of the amide bond the formerly antibiotic compound
cannot bind irreversibly to PBP, which are responsible for the establish-
ment of peptide bonds in murein, a component of the bacterial cell wall.
Due to their broad substrate specificity profile and the high activity
of class B metallo-β-lactamases, especially towards carbapenems,
exhibiting the broadest spectrum of antibiotic activity, current research
focuses on understanding this class of enzymes. Profound comprehen-
sion of the mechanism and the structures of metallo-β-lactamases
might be the foundation for a potential application in bioremediation
as well as for drug design [18].
The ligands were obtained as described before [4–6,19]. The geometry
of the metal-free ligands was extensively discussed [3]. Analogous to the
zinc(II) complexes described above, a hydroxido-bridged conformation of
the dicopper(II) complexes of the patellamides was reported [20]. In ad-
dition to the μ-hydroxido-bridged coordination mode, a dicopper(II)
complex with terminal OH− donors is also possible and likely to be the
catalytically active species, as substrate coordination may be preferred
to this active site. These two structures have been computed (see Fig. 2,
structural data are given as Supporting Information), and analogous
Scheme 1. Schematic representation of the metal ion free glycoside hydrolysis.

72
P. Comba et al. / Journal of Inorganic Biochemistry 159 (2016) 70–75
Scheme 2. Proposed mechanism of the hydrolysis of β-lactam-antibiotics via NDM-1 (New Delhi Metallo-β-lactamase1) [17,18].
Scheme 3. Catalytic hydrolysis of 4-nitrophenyl-D-galactopyranoside [10].
Scheme 4. Schematic representation of β-lactam hydrolysis.
structures have been identified by EPR spectroscopy [3,6,9]. The major
difference between the bridged and unbridged structures 1 and 2 with
the same ligand H4pat2 is the angle between the two copper(II) centers.
The angle between the two sites in 1 is 107°, 2 adopts a different confor-
mation with a 132° angle between the two sites (twist of the imidazole-
amide-imidazole-copper planes). This leads to a larger distance between
the copper(II) centers (3.98 Å in 1, 5.23 Å in 2).
galactopyranoside as a glycosidase model substrate proved that the
hydrolysis product of these highly activated ethers are spectrophoto-
metrically easily detectable (Scheme 3) [10]. (See Scheme 4)
The optimal pH range for the maximum kinetic activity was deter-
mined by pH-dependent measurements with a constant catalyst
(40 μM) and substrate concentration (30 mM) in a 0.3: 1.3: 1 mixture
of MeCN: buffer: MeOH as solvent; the pH values were controlled by a
multicomponent buffer, see Experimental Section. The rate for the
uncatalyzed reaction was subtracted to yield the catalytic rate. The
pH-dependence of the catalysis rates were measured in steps of 0.5
pH units or smaller from pH = 8.0 to pH = 11.5 for the dicopper(II)
complexes 3 and 2 of the cyclic pseudo-octapeptides H4pat1 and
H4pat2. The pH profiles are shown in Fig. 3.
2.2. Glycosidase-like activity
The experimental investigations involved 4-nitrophenyl-α-D-
glucopyranoside and 4-nitrophenyl-β-D-glucopyranoside as glycosi-
dase model substrates. Previous experiments with 4-nitrophenyl-α-D-
Fig. 3. pH profiles of [CuII2(H4pat1)(OH)]+ (solid) (3), [CuII2(H4pat2)(OH)]+ (2) (dashed) catalyzed 4-nitrophenyl-α-D-glucopyranoside (left) and 4-nitrophenyl-β-D-glucopyranoside
(right) hydrolysis (c = 10.5 mM).

P. Comba et al. / Journal of Inorganic Biochemistry 159 (2016) 70–75
73
Table 1
Kinetic data from pH-profiles of [Cu2(H2pat1)(OH)]+ (3) and [Cu2(H2pat2)(OH)]+ (2) for glycosidase activity and corresponding Michaelis–Menten parameters.
Catalyst
pHmax
v0,max [× 10−9 M/s]
pKa1
9.86 0.03
9.73 0.09
10.19 0.04
9.21 0.02
pKa2
γ
kcat 10−4 [s−1
]
KM [mM]
kcat/KM 10−2 [M−1 s−1
]
α-glyc.
3
2
3
2
10.03
9.92
10.36
9.43
6.19 0.53
1.11 0.21
3.36 0.03
5.86 0.02
10.20 0.14
10.09 0.04
10.53 0.04
9.64 0.02
0.1954 0.0357
0.1748 0.0417
0.2378 0.0521
0.2435 0.0508
1.55 0.13
0.28 0.01
1.22 0.09
1.08 0.04
4.64 0.43
1.90 0.47
3.56 0.23
2.12 0.17
3.30 0.3
1.49 0.1
3.43 0.04
5.09 0.02
β-glyc.
All measured data were fitted by a non-linear regression to a
Michaelis–Menten model. The determined pH profiles were fitted
with Eq. (1), which is based on a model for a diprotic system [21].
magnitude for α- and β-glycosides. For α-glycosides, the optimal pH
is ca. 10 for 3 and 2, for β-glycosidase the optimal pH is 10.4 for 3
and 9.4 for 2. The uncatalyzed background rate constants for α- and
β-glycosidase are kuncat = 0.0135∙10−4 s−1 and kuncat = 0.0286∙
10−4 s−1, respectively, [22] and this corresponds to rate enhancements
kcat/kuncat for the pat1 and pat2 based catalysts of 115 and 21 for α- and
of 43 and 37 for β-glycosidase, respectively. That is, the catalyst with the
configuration of the natural product (pat1) generally is a more efficient
catalyst, and this observation was also made for the carboanhydrase and
phosphatase activities, i.e. the configuration of the side-chains of the
cyclic pseudo-octapeptides is of crucial importance [7,8]. Note that the
optimum pH value for the catalytic hydrolysis of glycosides of around
10 is outside the physiological range. It is interesting to note here that
this is not the case with the carboanhydrase and phosphatase activities
which have optimum performance around pH = 7 [7,8]. Glycosidase
activity with dicopper(II) complexes has only been studied with a
small range of other model systems [10–13]. Interestingly, these also
have optimum performance in the range of pH = 10. So far, it has
not been studied whether and how this is related to the substrate
used. Natural glycosidase activity is reported to be in the range of
KM = 4–14 mM and kcat = 0.14–14 s−1 [30] for α-glycosidase activity
(substrate phenyl-α-glycoside) and KM = 0.07 mM, kcat = 169 s−1 for
Agrobacterium β-Glucosidase (substrate: 4-nitrophenyl-D-glucoside)
[23].
Hþ
!
γKa
2
ꢀ
ꢁ
vmax 1 þ
Hþ
!
v0
¼
ð1Þ
ꢀ
ꢁ
Ka
2
ꢀ
ꢁ
1 þ
þ
Hþ
Ka
1
Here, v0 is the initial and vmax the maximum reaction rate, reached
under given conditions. The factor γ is related to the relative activity
of the two active species in equilibrium (EnS and En−1S); a value of
γ less than unity corresponds to a more active EnS adduct and a value
larger than one considers the deprotonated adduct En−1S as more
active. The rates for the two deprotonation steps pKa1 and pKa2 are
given in Table 1. The velocity v0,max is defined as the maximum initial
rate at given conditions at optimum pH. Measurements at the optimum
pH values in dependence of the substrate concentration were also fitted
according to a Michaelis–Menten model (see Fig. 4 and Table 1). Here,
the concentration of 4-nitrophenyl-α/β-D-glucopyranoside was varied
between 1 and 10 mM. The data were fitted to Eq. (2), providing the
Michaelis–Menten constant KM, which was used to determine kcat
(Eq. (3)), both summarized in Table 1.
½Sꢀ
KM þ0½Sꢀ
ꢂ
ꢃ
v ¼ vmax
∙
ð2Þ
ð3Þ
2.3. β-lactamase-like activity
0
vmax
½Kꢀ0
Similar to the glycosidase activity measurements, the optimal
pH range for maximum lactamase activity was determined by pH-
dependent measurements with a constant catalyst (5 μM) and substrate
concentration (25 μM) in a 1:1 MeCN: buffer mixture as solvent; the pH
values were controlled by a multicomponent buffer, see Experimental
Section. The pH dependence of the catalysis rates were measured in
steps of 0.5 pH units or smaller from pH = 9.0 to pH = 12.0 for the
dicopper(II) complexes of the cyclic pseudo-octapeptides H4pat1 and
H4pat2. For [Cu2(H2pat2)(OH)]+ (2) no activity could be observed. The
initial rates at each pH value were corrected for the corresponding
autohydrolysis rates. The pH profile for [Cu2(H2pat1)(OH)]+ (3) is
kcat
¼
The α-glycosidase-like model system shows a narrow pH range.
The difference of the determined pKa values is very small and the max-
imum activities are at pH = 10.03 ([CuII2(H4pat1)(OH)]+, 3) and 9.92
([CuII2(H4pat2)(OH)]+, 2). For the β-glycosidase data, the difference
between the determined pKa values is very small and the maximum
activities are at pH = 10.36 ([CuII2(H4pat1)(OH)]+, 3) and 9.43
([CuII2(H4pat2)(OH)]+, 2). The glycosidase-like activity of [Cu2(H2pat1)
(OH)]+ (3) and [Cu2(H2pat2)(OH)]+ (2) are of the same order of
Fig. 4. Michaelis–Menten diagrams of [Cu2(H2pat1)(OH)]+(3, solid), [Cu2(H2pat2)(OH)]+ (2, dashed) catalyzed 4-nitrophenyl-α-D-glucopyranoside hydrolysis (pH for the substrate
dependence: α-glycosidase: 10.03 [Cu2(H2pat1)(OH)]+, 9.92 [Cu2(H2pat2)(OH)]+ and β-glycosidase: 10.36 [Cu2(H2pat1)(OH)]+, 9.43 Cu2(H2pat2)(OH)]+).

74
P. Comba et al. / Journal of Inorganic Biochemistry 159 (2016) 70–75
4. Experimental section
All solvents and reagents (absolute, p.a. grade and purum grade) were
obtained commercially and used without further purification. Dry
solvents were kept above molecular sieves. MilliQ water (R N 18MΩ)
was used for the kinetic assays.
4.1. Ligands and complexes
The cyclic pseudo-peptides were prepared according to [4,6,19].
4.2. Hydrolase assays
4.2.1. Preparation of the multicomponent buffer solutions
Fig. 5. pH profile of the [CuII2(H4pat1)(OH)]+ (3) nitrocefin hydrolysis (c = 25 μM; the
solid line is a basis spline function and not fit to a kinetic model).
The aqueous buffer consisted of CAPS, N-cyclohexyl-3-amino-
propanesulfonic acid, pKa = 10.40, CHES, 2-(N-cyclohexylamino)
ethanesulfonic acid, pKa = 9.30, HEPES, 4-(2-hydroxyethyl)-1-
piperazinylethanesulfonic acid, pKa = 7.55 and MES, 2-(N-morpholino)
ethanesulfonic acid, pKa = 6.15. Lithium perchlorate was added to
achieve a constant ionic strength of μ = 0.45. Each component was
dissolved in Milli-Q water. A standard solution with 55.56 mM of the
buffer components and 277.8 mM lithium perchlorate was prepared.
Aliquotes of 45 mL of the standard buffer were adjusted to the desired
pH value by addition of 2 M NaOH. A Metrom 713 pH meter which is
equipped with a KCl electrode was used to adjust pH values at 25 °C.
The pH-meter was calibrated with pH standard solutions at pH values
of 4, 7 and 9. Subsequently the aliquotes were filled up to 50 mL leading
to a final buffer concentration of 50 mM and 250 mM lithium perchlorate.
Metal ions were removed by stirring adjusted buffers over night with
Chelex 100, which was afterwards filtered off by using 45 μm syringe
filters. Finally all buffers were degassed by flushing N2 through the
solution in an ultrasonic bath for 3 h in order to remove potentially dis-
solved CO2.
Table 2
Kinetic data from pH-profiles of [Cu2(H2pat1)(OH)]+ (3) for the hydrolysis of nitrocefin
and the corresponding Michaelis–Menten parameters determined at pH 11.5.
Catalyst pHmax v0,max ∙10−9 pKa1
kcat × 10−4
[s−1
KM
[μM]
kcat/KM
[M/s]
]
[M−1 s−1
]
3
11.50
3.82 0.03
≈11.3 11.34 ( 0.91) 22.47 50.47
shown in Fig. 5 and the resulting kinetic parameters and pKa values are
summarized in Table 2 (see Supporting Information for a plot of the sub-
strate concentration dependent kinetic data). β-lactamase activity
is observed at approx. pH 11 with resulting kcat/KM = 50.5 M−1 s−1
,
and this is consistent with similar dinuclear model complexes [7].
The uncatalyzed background rate constant for β-lactam hydrolysis is
kuncat = 0.025ˑ10−4 s−1 [24], and the rate enhancement therefore is
kcat/kuncat = 454. The kinetic Michealis Menten data reported for natural
β-lactamase activity is reported to be in the range of KM = 16–100 μM
and kcat = 0.3–200 s−1 for Bacteroides fragilis and Aeromonas hydrophila
respctively (substrate nitrocefin) [25].
4.2.2. Glycosidase-like activity
The glycosidase-like activity was determined by measuring the hy-
drolysis of 4-nitrophenyl-α-D-glucopyranoside and 4-nitrophenyl-β-
D-glucopyranoside hydrolysis. The hydrolysis product 4-nitrophenole
was produced which could be detected by monitoring the increase of
an strong absorbance at 410 nm (ε is pH dependent).
3. Conclusions
The patellamide-dicopper(II) complexes are among the few exam-
ples of dinuclear copper(II) complexes acting as glycosidase-like
model compounds. Taking into account the broad reactivity pattern of
these compounds with respect to hydrolysis reactions, namely phos-
phatase, carboanhydrase, glycosidases as well as β-lactamase, several
questions emerge. First and probably foremost, it is uncertain whether
the catalytic activities observed for the dinuclear copper(II) complexes
can be associated with the function of the cyclic peptides in Prochloron
or Lissoclinum patellum. Currently, we are therefore carrying out in
vivo studies towards an understanding of the stability of dinuclear
copper(II) complexes with these naturally occurring ligands. Second
and not less important is the question, whether all of the hydrolyses
are of importance for the symbiosis partners. Is each of the reactions
happening in a separate compartment in the cyanobacterium, providing
the optimum pH value? Or is none of these reactions observed in the
test tube actually metabolically relevant for the ascidians?
These questions are in the focus of ongoing research. If glycosidase as
well as carboanhydrase activity in the cell were mediated by the same
dinuclear copper(II) patellamide complexes, this would indicate that
the same enzyme is capable of fixing carbon from CO2 as well as to
catabolize the products from the assimilation (Calvin cycle). Here,
research has to be carried out as to where which reaction takes place.
Prochloron might be responsible for the fixation, whereas the ascidian
might require the catabolic glycosidase activity.
Initially, the extinction coefficients of the product formation of
4-nitrophenolate in a multicomponent buffer were calculated by cali-
bration curves for each pH value (pH = 5.5–11.5) using Lambert Beer's
law (Table 1). The extinction coefficients of pH = 8.7–11.5 are similar to
those determined for 4-nitrophenolate in CAPS buffer (ε = 16,190 M-
1 cm-1) [10]. The obtained extinction coefficients (see Supporting Infor-
mation) were used to convert the absorbance of the reaction product
into molar amounts.
UV–Vis-spectra data were recorded on a Jasco V-570 spectropho-
tometer equipped with a Jasco ETC-505 T cryostat at 25 °C and in a
0.3: 1.3: 1 MeCN:buffer:MeOH solution. Using time-course measure-
ments at fixed wavelengths (λmax = 410 nm), spectrophotometric
titrations were performed with a cyclic pseudo-peptide concentra-
tion of 1 mM (MeOHdry). CuII(CF3SO3)2 (25 mM, MeOHdry) and
base (n-Bu4N)(OMe) (25 mM, MeOHdry) were added to different
pH-samples. The final concentrations of dicopper(II) complexes
of H4pat1 and H4pat2 in the cuvette were 40 μM. All solutions were
degassed and kept under argon at 7 °C. Blank tests, the auto-
hydrolysis of 4-nitrophenyl-α-D-glucopyranoside or 4-nitrophenyl-β-
D-glucopyranoside respectively, were subtracted from the reported
kinetic rates. For the determination of pH dependent reaction velocities
the substrate concentration of 30 mM (MeOHdry) was chosen. For
substrate dependency measurements at constant pH the substrate
concentration was varied between 1 and 50 mM.

P. Comba et al. / Journal of Inorganic Biochemistry 159 (2016) 70–75
75
4.2.3. β-Lactamase-like activity
Heidelberg Graduate School of Computational Methods for the Sciences
(HGS) and the University of Heidelberg are gratefully acknowledged.
The β-lactamase-like activity was determined by measuring the
hydrolysis of nitrocefin. The hydrolysis product was produced which
could be detected by monitoring the decrease of a strong absorbance
of nitrocefin at 390 nm monitored over a period of 420 s, whereas dur-
ing the initial 120 s the equilibrium was reached, which is why these
data have not been used. As nitrocefin, as well as its hydrolysis product
show absorption at 390 nm, a corrected extinction coefficient was used
13,415 M−1 cm−1. UV–Vis-spectra data were recorded on a Jasco V-570
spectrophotometer equipped with a Jasco ETC-505 T cryostat at 37 °C
and in a 1: 1 MeCN:buffer solution. Using time-course measurements
at fixed wavelengths (λmax = 390 nm), spectrophotometric titrations
were performed with a cyclic pseudo-peptide concentration of 1 mM
(MeCNdry). CuII(CF3SO3)2 (25 mM, MeCNdry) and base (n-Bu4N)(OMe)
(25 mM, MeCNdry) were added to different pH-samples. The final con-
centrations of dicopper(II) complexes of H4pat1 and H4pat2 in the cuvette
were 5 μM. All solutions were degassed and kept under argon. Blank tests,
the autohydrolysis of nitrocefin, were subtracted from the reported ki-
netic rates. For the determination of pH dependent reaction velocities
the substrate concentration of 25 μM was chosen.
Appendix A. Supplementary data
The Supporting Information includes the cartesian coordinates, a list
of selected bond lengths and angles of the structures discussed (1, 2)
and details on the kinetic studies. Supplementary data associated with
this article can be found in the online version, at http://dx.doi.org/10.
1016/j.jinorgbio.2016.02.014.
References
[1] W.E. Houssen, J. Koehnke, D. Zollman, J. Vendome, A. Raab, M.C.M. Smith, J.H.
Naismith, M. Jaspars, Chembiochem 13 (2012) 2683–2689.
[2] J. Koehnke, A. Bent, W.E. Houssen, D. Zollman, F. Morawitz, S. Shirran, J. Vendome,
A.F. Nneoyiegbe, L. Trembleau, C.H. Botting, M.C.M. Smith, M. Jaspars, J.H.
Naismith, Nat. Struct. Mol. Biol. 19 (2012) 767–773.
[3] P. Comba, N. Dovalil, L.R. Gahan, G.R. Hanson, M. Westphal, Dalton Trans. (Dalton
Perspective) 43 (2014) 1935–1956.
[4] G. Haberhauer, F. Rominger, Eur. J. Org. Chem. (2003) 3209–3218.
[5] G. Haberhauer, Tetrahedron Lett. 49 (2008) 2421.
[6] P. Comba, N. Dovalil, L.R. Gahan, G. Haberhauer, G.R. Hanson, C.J. Noble, B. Seibold, P.
Analogous to glycosidase activity measurements, initially the opti-
mum pH range for lactamase activity was determined. The pH-
dependence of the catalysis rates were measured in steps of 0.5 pH
units or smaller from pH = 9.0 to pH = 12.0. Subsequently Michaelis–
Menten measurements were carried out at the optimum pH value
of 11.5. For substrate dependency measurements at constant pH the sub-
strate concentration was varied between 10 and 50 μM.
Vadivelu, Chem. Eur. J. 18 (2012) 2578–2590.
[7] P. Comba, L.R. Gahan, G.R. Hanson, M. Westphal, Chem. Commun. 48 (2012)
9364–9366.
[8] P. Comba, L.R. Gahan, G.R. Hanson, M. Maeder, M. Westphal, Dalton 43 (2014)
3144–3152.
[9] P.V. Bernhardt, P. Comba, T.W. Hambley, S.S. Massoud, S. Stebler, Inorg. Chem. 31
(1992) 2644–2651.
[10] S. Striegler, N.A. Dunaway, M.G. Gichinga, J.D. Barnett, A.G.D. Nelson, Inorg. Chem. 49
(2010) 2639–2648.
Time dependent UV/Vis spectra monitoring β-lactamase-like
activity were recorded on a J&M Tidas II with the program Spectralysis
2.0 using an external cryostat Q-Blue Wireless Temperature Control
in the range between 300 and 850 nm. These spectra were recorded
at 37 °C at pH 11.5 with a nitrocefin concentration of 40 μM and a com-
plex concentration of 5 μM for 20 min at time intervals of Δt = 46.7 s
(see Supporting Information). The resulting spectra were corrected by
subtraction the obtained spectrum of the complex.
[11] S. Striegler, M. Dittel, R. Kanso, N.A. Alonso, E.C. Duin, Inorg. Chem. 50 (2011) 8869.
[12] S. Striegler, J.D. Barnett, N.A. Dunaway, ACS Catal. 2 (2012) 50.
[13] J.D. Barnett, S. Striegler, Top. Catal. 55 (2012) 460.
[14] S. Haldar, A.K. Patra, M. Bera, RSC Adv. 4 (2014) 62851–62861.
[15] C.S. Rye, S.G. Withers, Curr. Opin. Chem. Biol. 4 (5) (2000) 573–580.
[16] S.H. Khan, R.A. O'Neill, Modern Methods in Carbohydrate Synthesis, Harwood
Academic Publishers, Amsterdam, the Netherlands, 1996.
[17] C. Bebrone, Biochem. Pharmacol. 74 (2007) 1686.
[18] L. Daumann, G. Schenk, L.R. Gahan, Eur. J. Inorg. Chem. (2014) 2869–2885.
[19] G. Haberhauer, A. Pinter, T. Oeser, F. Rominger, Eur. J. Org. Chem. (2007) 1779–1792.
[20] P. Comba, N. Dovalil, G.R. Hanson, G. Linti, Inorg. Chem. 50 (2011) 5165–5174.
[21] I.H. Segel, Enzyme Kinetics - Behavior and Analysis of Rapid Equilibrium and Steady-
State Enzyme Systems, Wiley-VCH, New York, 1975.
4.3. DFT Calculations
[22] F. Ortega-Caballero, C. Rousseau, B. Christensen, T.E. Petersen, M. Bols, J. Am. Chem.
Soc. 127 (2005) 3238–3239.
[23] J.B. Kempton, S.G. Withers, Biochemistry 31 (1992) 9961–9969.
[24] W.J. Song, F.A. Tezcan, Science 346 (2014) 1525–1528.
[25] A. Felici, G. Amicosante, A. Oratore, R. Strom, P. Ledent, B. Joris, L. Fanuel, J.M. Frere,
Biochem. J. 291 (1993) 151–155.
[26] A.D. Becke, J. Chem. Phys. B 98 (1993) 5648.
[27] A. Schäfer, H. Horn, R. Ahlrichs, J. Chem. Phys. 97 (1992) 2571–2577.
[28] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G.
Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M.
Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro, M. Bearpark,
J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, T. Keith, R. Kobayashi, J. Normand,
K. Raghavachari, A. Rendell, J.C. Burant, S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N.J.
Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L.
Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S.
Dapprich, A.D. Daniels, O. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox,
Gaussian, Gaussian, Inc., Wallingford CT, 2013.
Fully optimized molecular structures were obtained through geom-
etry optimizations employing the B3LYP [26] functional in conjunction
with the def2-tzvp basis set [27]. All calculations were performed with
Gaussian 09 [28]. Solvation was approximated by the polarizable con-
tinuum model [29] with the permittivity set to methanol. Single point
energy and frequency calculations were carried out on the optimized
structures. All pictures of computed structures were produced with
Avogadro [30].
Acknowledgements
We are grateful for the computational resources provided by
the bwForCluster JUSTUS, funded by the Ministry of Science, Research
and Arts and the Universities of the State of Baden-Württemberg,
Germany, within the framework program bwHPC-C5 as well as for the
bwGRID Cluster, funded by the Ministry of Education and Research
BMBF. Financial support by the German Science Foundation (DFG), the
[29] G. Scalmani, M.J. Frisch, J. Chem. Phys. 132 (2010) 114110.
[30] M.D. Hanwell, D.E. Curtis, D.C. Lonie, T. Vandermeersch, E. Zurek, G.R. Hutchinson,
J. Chem. Inf. 4 (2012) 17.