Investigation of liver alcohol dehydrogenase catalysis using an NADH biomimetic and comparison with a synthetic zinc model complex

Article abstract of DOI:10.1016/j.poly.2015.11.027

We have compared the catalytic activity of horse liver alcohol dehydrogenase (LADH) with a synthetic zinc model complex in the presence of N-benzyl-1,4-dihydronicotinamide (BNAH), a cofactor which serves as a biomimetic for the natural cofactor NADH. We have used five different substrates (benzaldehyde, p-anisaldehyde, 4-nitrobenzaldehyde, 2-pyridine carboxaldehyde, and 5-pyrimidine carboxaldehyde) in this study. These substrates vary in their substituent inductive effect, which is the ability to donate or withdraw electron density away from their carbonyl-functional group. Our results reveal that in the presence of NADH, geometric factors (induced fit of the substrate and cofactor in the enzyme active site) are vital. However, reactivity assays show that in the presence of BNAH, there is a strong correlation between substrate electronic environment and the observed catalytic rate, i.e. the more electron withdrawn the substrate, the greater the speed at which the reduction reaction occurs. NMR spectroscopy reveals that a synthetic zinc model complex catalyzes the reduction of substrates in a manner consistent with LADH enzyme.

Full text of DOI:10.1016/j.poly.2015.11.027

Accepted Manuscript
Investigation of liver alcohol dehydrogenase catalysis using an NADH biomi-
metic and comparison with a synthetic zinc model complex
James R. Sunderland, Xingjian Tao, Elizabeth E. Butrick, Lauren C. Keilich,
Christine E. Villa, John R. Miecznikowski, Swapan S. Jain
PII:
S0277-5387(15)00704-4
DOI:
http://dx.doi.org/10.1016/j.poly.2015.11.027
POLY 11672
Reference:
To appear in:
Polyhedron
Received Date:
Accepted Date:
28 September 2015
17 November 2015
Please cite this article as: J.R. Sunderland, X. Tao, E.E. Butrick, L.C. Keilich, C.E. Villa, J.R. Miecznikowski, S.S.
Jain, Investigation of liver alcohol dehydrogenase catalysis using an NADH biomimetic and comparison with a
synthetic zinc model complex, Polyhedron (2015), doi: http://dx.doi.org/10.1016/j.poly.2015.11.027
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Investigation of liver alcohol dehydrogenase catalysis using an NADH
biomimetic and comparison with a synthetic zinc model complex
1
James R. Sunderland , Xingjian Tao , Elizabeth E. Butrick , Lauren C. Keilich ,
1
2
2
2 2*
Christine E. Villa , John R. Miecznikowski , and Swapan S. Jain
1*
1
Department of Chemistry, Bard College, 30 Campus Road, Annandale-on-Hudson, NY 12504, USA
2
Department of Chemistry and Biochemistry, Fairfield University, 1073 North Benson Road, Fairfield, CT
6824, USA
0
*
Corresponding Authors: Tel: +1-845-752-2354. Fax: +1-845-752-2339. Email: sjain@bard.edu.
Tel: +1-203-254-4000 x 2125. Fax: +1-203-254-4034. Email: jmiecznikowski@fairfield.edu.
Abstract
We have compared the catalytic activity of horse liver alcohol dehydrogenase
LADH) with a synthetic zinc model complex in the presence of N-benzyl-1,4-
(
dihydronicotinamide (BNAH), a cofactor which serves as a biomimetic for the natural
cofactor NADH. We have used five different substrates (benzaldehyde, p-anisaldehyde,
4
-nitrobenzaldhyde, 2-pyridine carboxaldehyde, and 5-pyrimidine carboxaldehyde) in
this study. These substrates vary in their substituent inductive effect, which is the ability
to donate or withdraw electron density away from their carbonyl-functional group. Our
results reveal that in the presence of NADH, geometric factors (induced fit of the
substrate and cofactor in the enzyme active site) are vital. However, reactivity assays
show that in the presence of BNAH, there is a strong correlation between substrate
electronic environment and the observed catalytic rate, i.e. the more electron withdrawn
the substrate, the greater the speed at which the reduction reaction occurs. NMR
spectroscopy reveals that a synthetic zinc model complex catalyzes the reduction of
substrates in a manner consistent with LADH enzyme.
Jain et al.
Manuscript File
1

Introduction
Liver alcohol dehydrogenase (LADH) is a dimeric enzyme where each subunit of
2
+
the enzyme contains a cofactor-binding domain and a Zn metal-containing catalytic
domain. [1] NADH is a natural cofactor that functions as a hydride donor in enzymatic
reactions where aldehydes and ketones are reduced to their respective alcohol moieties.
2
+
Binding of NADH occurs in close proximity to the Zn center of LADH enzyme. [2]
Crystal structure data shows that NADH binds very tightly within the coenzyme-binding
domain of liver alcohol dehydrogenase enzyme by utilizing direct electrostatic contacts
with eight amino acid residues. [3]
Figure 1. Chemical structure of NADH (left) and BNAH (right).
N-benzyl-1,4-dihydronicotinamide, abbreviated as BNAH, has been used as a
biomimetic in place of NADH for a variety of oxidoreductase enzymes including
monoxygenases and dehydrogenases. [4,5] BNAH is preferred in industrial applications
because unlike NADH, it does not suffer from the high cost of preparation and instability
under higher temperature and pressure [6]. In 2002, Lo and Fish showed that the
nicotinamide moiety is the only essential requirement for catalysis in case of horse
LADH enzyme. [4] Even though the nicotinamide moiety, which is crucial for catalysis,
Jain et al.
Manuscript File
2

is identical in case of both NADH and BNAH, the chemical structure of BNAH, shown
in Figure 1, lacks the ribose, pyrophosphate, and adenine groups. These substantial
changes in the chemical structure of BNAH likely exclude the formation of many of the
amino acid interactions that typically occur in case of bound NADH [7]. It is important to
note that the missing amino acids contacts in case of BNAH binding are only responsible
for cofactor binding and not catalysis.
Figure 2. Reaction depicting enzymatic reduction of benzaldehyde in the presence of
NADH cofactor. BNAH can serve as a cofactor in place of NADH in this reaction.
Our work compares the catalytic activity of alcohol dehydrogenase (EC 1.1.1.1),
isolated from horse liver, with a synthetic model zinc complex in the presence of BNAH
as a cofactor. A typical reduction reaction of benzaldehyde to benzyl alcohol is shown in
Figure 2. Model enzyme complexes are synthetic complexes that mimic the active site of
an enzyme in terms of chemical structure and redox properties. In case of liver alcohol
dehydrogenase enzyme, the active site contains a zinc metal, which is ligated to a labile
water molecule, two cysteine residues and one histidine residue (denoted as an ‘SNS’
ligand environment). [8] Synthetic model complexes are not only useful in industrial
applications but they are also instrumental in understanding the activity and function of
natural enzyme active sites. [9] Synthetic model complexes for liver alcohol
dehydrogenase have been reported previously. [10-23] However, comparative analysis of
Jain et al.
Manuscript File
3

an enzyme’s catalytic rate with that of a synthetic model active site complex needs
further investigation. Several structural models for LADH have been reported with the
same donor atoms as the metalloenzyme, but no reactivity data were reported for these
complexes. Reactivity has been reported for zinc LADH model complexes with donor
atoms that are different than the enzyme’s active site. [24]
We have modeled the structure and reactivity of the zinc active site using a new
family of robust pincer ligands coordinating the same S,N,S donor atoms as the enzyme.
[23, 25] We are not aware of any enzyme comparative studies where LADH enzyme is
compared to the activity of LADH model complexes.
In our comparative studies of LADH enzyme with a model active site complex, we
have varied the substrates of the chemical reaction. Substrates that are similar in their
geometric structure but vary in their electronic properties can affect reaction rates
especially those involving oxidation and reduction. The presence of electron-donating
groups (EDG) or electron-withdrawing groups (EWG) can affect the enzyme’s catalytic
activity. [26] We have utilized a total of five substrates in our work. Their chemical
structures are shown in Figure 3. We hypothesize that the reduction of an aldehyde
substrate should be directly proportional to its ability to withdraw electron density away
from the carbonyl carbon atom of the molecule. In comparison to benzaldehyde, p-
anisaldehyde is expected to have a lower enzymatic rate under identical experimental
conditions due to the presence of an –OCH substitution (EDG). Whereas 4-
3
nitrobenzaldhyde, 2-pyridine carboxaldehyde, and 5-pyrimidine carboxaldehyde are
expected to accelerate the reduction reaction due to the presence of EWG groups. In
previous work, Blomquist reasoned that the electron density around the carbonyl carbon
of an aldehyde substrate should have an effect on the kinetics of the LADH enzyme
Jain et al.
Manuscript File
4

catalyzed reaction. [27] It was concluded that the electron withdrawing capability of a
particular substrate tends to speed up the reaction rate in small increments. It was also
found that the lipophilic nature of the substrate could stabilize the protein-substrate
complex more so than polarized substrates, especially during the reaction when the
substrate is in an excited polarized state.
We have used UV-Vis spectroscopy and cyclic voltammetry to correlate the rates of
LADH enzyme catalyzed reactions for five substrates with their substituent effect
(Hammett constant) in the presence of NADH and BNAH cofactors. We have also used
NMR spectroscopy to determine the reactivity of the substrates using a synthetic zinc
model complex. Our work has implications about steric and electronic factors that govern
chemical reactions in case of dehydrogenases. Our data shows that LADH enzyme can
tolerate various substrates as well as an NADH cofactor mimic. We have shown that a
synthetic zinc active site model complex catalyzes aldehyde reductions in a manner
similar to LADH enzyme.
Jain et al.
Manuscript File
5

Figure 3. Chemical structures of benzaldehyde-based substrates.
Experimental Section
Synthesis of BNAH and Determination of its Extinction Coefficient
BNAH (N-benzyl-1,4-dihydronicotinamide) was synthesized and purified in our
laboratory according to previously published protocols. [4] Three samples of solid BNAH
(2.0 mg, 5.0 mg, and 10.0 mg) were weighed out and dissolved in 1.0 mL of acetone.
Using each of these stock solutions, triplicate samples were prepared by diluting stock
BNAH in 10 mM potassium phosphate buffer, pH 7.10. UV-Vis spectra were recorded
and the absorbance value at 360 nm was used to calculate the extinction coefficient (4890
-
1
-1
75 L.mol .cm ) of BNAH.
±
Alcohol Dehydrogenase Enzyme Assays
Disodium NADH salt, equine liver alcohol dehydrogenase enzyme, and the
aldehyde substrates shown in Figure 3 (benzaldehyde, 4-nitrobenzaldhyde, p-
Jain et al.
Manuscript File
6

anisaldehyde, 2-pyridine carboxaldehyde, 5-pyrimidine carboxaldehyde) were purchased
from Sigma (St. Louis, MO). Enzyme and NADH stocks were prepared in potassium
o
phosphate buffer (10 mM, pH 7.10) and stored at -20 C. Aldehyde substrates were
prepared in potassium phosphate buffer (10 mM, pH 7.10) containing 1% acetone.
BNAH stock was prepared in acetone and subsequently diluted with phosphate buffer
containing 1% acetone. Enzyme kinetic reactions (2.5 mL final volume) were carried out
o
in a quartz cuvette at 25 C in 10 mM potassium phosphate buffer (pH 7.10) with a final
concentration of 150 µM for each of the substrates and NADH. LADH enzyme (0.1 unit)
was added directly to the cuvette containing the reaction mixture and absorbance spectra
were recorded as a function of time. Change in absorbance at 340 nm was monitored as a
function of time to determine the rate of the reaction in the presence of NADH. Reactions
were carried out in triplicate. To determine the utilization of BNAH as a potential
cofactor, identical reactions (in triplicate) were carried out except that NADH in the
reaction mixture was replaced with BNAH (150 µM) and absorbance change was
measured at 360 nm.
Reactivity with Synthetic Model Zinc Complexes
Model zinc complex 3c was synthesized in our laboratory according to previously
published protocols. [6, 23, 25] In a typical reaction, 0.1 mmol of an aldehyde substrate,
0
.2 mmol of BNAH, and 0.1 mmol of the zinc complex or 0.1 mmol ZnCl were
2
dissolved in 3 mL of CDCl . The reaction was heated at reflux. Aliquots of the reaction
3
1
were taken at certain times and analyzed using H NMR spectroscopy. All data are
averages of at least two runs. All of the reagents and solvents were purchased from Acros
Organics except for 2-pyridine carboxaldehyde, which was purchased from Alfa Aesar.
Jain et al.
Manuscript File
7

Cyclic Voltammetry
Substrates used in the cyclic voltammetry (CV) studies were benzaldehyde, 4-
nitrobenzaldhyde, p-anisaldehyde, 2-pyridine carboxaldehyde, and 5-pyrimidine
carboxaldehyde. The supporting electrolyte solution was 0.1M tertbutylammonium
perchlorate prepared in acetonitrile. Substrates were dissolved at a final concentration of
2
mM in the supporting electrolyte. CV was carried out using a gold electrode an Epsilon
Potentiostat with a BASi-C-3 cell stand. Briefly, 10 mL of the individual substrate
solution was added to the sample chamber and warmed to room temperature. The initial
and final potential was set at 0 mV. The switching potential was set to -1500 mV with a
scan rate of 50 mV/s. Oxidation and reduction peaks (in mV) were used to calculate
Hammett constants for each substrate where benzaldehyde was used as a reference.
Detailed calculations are shown in the supplementary information section and the data is
shown as Table S2.
Results & Discussion
We first determined whether the LADH enzyme under our experimental conditions
could tolerate BNAH cofactor. The catalytic efficiency of LADH enzyme was measured
under physiological conditions using BNAH as a biomimetic for NADH. Our work was
carried out at pH 7.1 because it has been previously shown that the association of NADH
to alcohol dehydrogenase enzyme is maximal below pH 9. [27] The natural cofactor
NADH has a UV-Vis absorption band centered at 340 nm whereas BNAH has a UV-Vis
absorption band centered at 360 nm (Figure 4). In our enzyme kinetic assay, a decline in
Jain et al.
Manuscript File
8

the absorbance at these wavelengths is indicative of the oxidation of the cofactor (NADH
+
+
to NAD or BNAH to BNA ) with the concomitant reduction of an aldehyde substrate to
an alcohol. Least squares fit of the data (absorbance vs. time) was used to calculate the
initial rate of the benzaldehyde reduction reaction in the presence of NADH and BNAH
(inset, Figure 4). Results in Figure 4 show that LADH enzyme is able to effectively
catalyze the reduction of benzaldehyde (rate of 5.0 ± 0.3 nmoles/min) in the presence of
BNAH albeit at a lower rate in comparison to NADH (10.5 ± 0.3 nmoles/min). It is
+
important to note that the oxidized species (BNA ) does not absorb at 360 nm, which
allows us to monitor the enzymatic reaction progress at this particular wavelength
without interference from the oxidized molecule. Enzymatic reactions tend to slow down
over time due to the depletion of substrate as well as due to product inhibition. [28] In all
of our reactions, we observed an initial linear increase in product formation that plateaued
over time. Initial rate data (first 2 minutes of each reaction) as well as total product
formation over a longer duration are shown in the supplementary information section
(Table S1). Control reactions of NADH and BNAH with benzaldehyde alone (minus
enzyme) or with enzyme alone (minus benzaldehyde) show no change in UV-Vis
absorbance over a 30-minute time period (Figure S4, Supplementary Information).
Jain et al.
Manuscript File
9

a) 1
.50#
.00#
.50#
1
0
0
.00"
.90"
.80"
1
0
0
y"="$0.0336x"+"0.9838"
"
0
"
1"
2"
3"
time (min)
.00#
3
00#
340#
380#
420#
460#
500#
wavelength #( nm)#
b)
1
.50#
.00#
.50#
1
1
1
.50"
.40"
.30"
y"="$0.0176x"+"1.4716"
"
1
0
0
0
"
1"
2"
3"
time (min)
.00#
3
00#
340#
380#
420#
460#
500#
wavelength #( nm)#
Figure 4. (a) UV-Vis spectra of the enzymatic assay with benzaldehyde substrate
illustrating the depletion of (a) NADH and (b) BNAH. Insets in each graph depict the
change in absorbance maximum as a function of time for each cofactor. Linear equations
obtained from the least squares fitting were used to determine the initial enzymatic rate in
units of nmoles/min of substrate converted to product.
Jain et al.
Manuscript File
10

The two-fold decline in the reaction rate with BNAH, in comparison to NADH, was
not unexpected as it is likely that BNAH binds weakly to the cofactor-binding domain of
LADH enzyme. The low affinity binding of BNAH to the enzyme is likely a consequence
of the significant differences in the chemical structure between NADH and BNAH.
Although the chemical reaction involving the hydride transfer is the same in both
cofactors, previous work has shown that BNAH (K ~ 3.7 mM) binds to monooxygenase
M
enzymes with a 130-fold lower affinity than NADH (K ~ 27 µM). [7] In an effort of
M
further understand changes in reaction rates in the presence of NADH vs. BNAH, we
conducted a substrate study to determine whether reactions rates are governed by the
inductive effect imparted by each substrate. Hammett constants (σ) have been widely
used in inorganic, organic and biophysical studies to gauge the changes in equilibrium
constant of a substrate resulting from the substitution of a hydrogen atom on the aromatic
ring by other functional groups. [29] We hypothesized that the reduction rate of an
aldehyde substrate should be directly proportional to its ability to withdraw electron
density away from the carbonyl carbon atom of the molecule. Furthermore, the
correlation between enzymatic catalytic rates and Hammett constants should be
independent of the cofactor (NADH or BNAH) utilized in the reaction. Cyclic
voltammetry was used to measure the redox potentials for five substrates (benzaldehyde,
4
-nitrobenzaldhyde, p-anisaldehyde, 2-pyridine carboxaldehyde, and 5-pyrimidine
carboxaldehyde) used in this study. Cyclic voltammograms of all five substrates are
presented in the supplementary information section (Figure S3).
Log K values were determined for each of the five substrates using redox potential
data from cyclic voltammetry studies. Hammett constants have previously been reported
Jain et al.
Manuscript File
11

for benzaldehyde, p-anisaldehyde, and 4-nitrobenzaldehyde. [29] In Figure 5a, log (K/Ko)
values were plotted as a function of known Hammett constant values for benzaldehyde,
p-anisaldehyde, and 4-nitrobenzaldehyde where K represents the reference compound
o
benzaldehyde with a log (K/K ) value of 0.0. The data was fit to a linear least squares fit
o
function and the equation was used to determine the unknown Hammett values for 2-
pyridine carboxaldehyde and 5-pyrimidine carboxaldehyde (calculations are shown in
supplementary information and data is presented in Table S2). After obtaining the
Hammett values for each of the substrates, we plotted initial enzymatic rates (nmoles of
substrate converted to product per minute) on the y-axis with the corresponding Hammett
constant values on the x-axis (Figure 5b and 5c).
2
In case of the natural cofactor NADH, we found a poor correlation (R = 0.544)
between the inductive effect exerted by the substrates (Hammett constant value) and their
corresponding enzymatic rate (Figure 5b). This was contrary to our expectations because
we hypothesized that the catalytic activity of the LADH enzyme, in the absence of major
structural differences between the five substrates, should be directly proportional to the
electronic environment of each substrate. Upon further reflection, reactions at the active
site of an enzyme generally occur according to a lock-and-key mechanism. The substrates
and the necessary cofactors fit in a very specific geometric conformation, which is in
accordance with the geometric shape of the active site. [30] It is plausible that the
NADH-bound enzyme complex only tolerates substrates with specific chemical structure
and orientation. If that is the case, then the poor correlation between the substrate
electronic environment and the observed catalytic rate is not surprising. Nevertheless, we
observe that the three substrates (4-nitrobenzaldehyde, 2-pyridine carboxaldehyde, and 5-
Jain et al.
Manuscript File
12

pyrimidine carboxaldehyde) containing EWGs show a higher reaction rate whereas
benzaldehyde (reference compound) and electron donating p-anisaldehyde show a lower
reaction rate (Figure 5b). These results indicate that the LADH active site is partially
flexible. Observation of catalytic activity in the presence of varying substrates is also
governed by an induced-fit mechanism where the substrate binding play an important role
in determining the final shape of the enzyme active site. Even though the electron-
withdrawn substrates show a higher catalytic rate, the absence of a strong correlation
illustrates that reactions by the enzyme LADH in the presence of its natural cofactor
NADH are primarily governed by steric and geometric factors and to a lesser extent by
the inductive effect exerted by the substrate.
Jain et al.
Manuscript File
13

4
3Nitrobenzaldehyde$
1
0
0
0
0
.00$
.80$
.60$
.40$
.20$
a)
b)
c)
y $= $1 .1641x $3 $0 .0096$
R² $= $0 .99982$
Benzaldehyde$
.00$
0.00$
0.20$
0
3
0.40$
30.20$
0.20$
0.40$
0.60$
0.80$
3
30.40$
p3Anisaldehyde$
30.60$
Hamme- $c onstant $( σ)$
p
p
Figure 5: (a) Measured values of log (K/K ) vs. known Hammett constant values.
o
Hammett constants for 2-pyridine carboxaldehyde and 5-pyrimidine carboxaldehyde
were extrapolated using the equation in the figure. (b) Correlation plot of initial
enzymatic rate vs. Hammett constant for all five substrates in the presence of NADH
cofactor. (c) BNAH correlation plot of initial enzymatic rate vs. Hammett constants.
Jain et al.
Manuscript File
14

Interestingly, reactions conducted in the presence of BNAH cofactor show an excellent
2
correlation (R = 0.972) between the inductive effect exerted by the substrates and the
observed reaction rates (Figure 5c). Our enzyme kinetic assays show that 4-
nitrobenzaldehyde has the highest reactivity (9.0 ± 0.2 nmoles/min) amongst the five
substrates tested and the reaction rate correlates well with its large Hammett constant
value of 0.78. Similarly, p-anisaldehyde, which is electron donating in nature due to the
presence of the –OCH group, shows the lowest reactivity (2.3 ± 0.1 nmoles/min) and the
3
smallest Hammett constant value (-0.27).
These results suggest that the reduction reactions catalyzed by the LADH enzyme in
the presence of BNAH cofactor are primarily governed by the inductive effect exerted by
the substrate molecules even though steric factors such as the presence of BNAH and
substrate in a particular orientation within the active site of the enzyme are important.
Previous work has shown that there is some plasticity in the alcohol dehydrogenase
active site during reactions with substrates that vary in structure. [31] Furthermore,
BNAH (although similar to NADH in its hydride donor reaction) is structurally distinct
from NADH and likely binds weakly to the enzyme active site. In the absence of a lock-
and-key induced fit mechanism that is observed in a naturally occurring active site of the
LADH enzyme with bound NADH, the electronic environment of the substrates is the
primary driver of the reaction rates in the presence of BNAH.
Jain et al.
Manuscript File
15

Comparison with zinc model complexes
Recently, we prepared a series of tridentate SNS zinc(II) pincer complexes that model
the zinc(II) active site of the LADH enzyme. [6, 23] The tridentate pincer SNS ligand
precursors incorporate bis-thione substituted based imidazole and triazole moieties. The
nitrogen donor atom comes from a nitrogen atom in a pyridine ring. [25] We wanted to
determine whether the conclusions reached in our LADH trials with different substrates
could be extended to a synthetic model enzyme complex. Synthetic model complexes
have enormous potential especially in industrial applications because higher product
yields can be achieved at elevated temperature and pressure which necessitates the use of
synthetic model enzyme active sites as well as a cofactor mimetic like BNAH. For
example, Schrock and co-workers have reduced dinitrogen to ammonia using a
molybdenum catalyst. [32] The tridentate ligands used in some of these systems were
relatively rigid as the pyridine and the imidazoles were directly bound to each other. By
using the starting material 2,6-(dibromomethyl) pyridine, we introduced a methylene
linker into the pincer ligand, thereby allowing us to examine the effect of ligand
flexibility on substrate binding. We modified the electronic environment imparted by our
ligand set through the use of imidazole- and triazole-based precursors. [33,34] In order to
understand the presence and sterics of ancillary alkyl groups on the ability of these model
systems to reduce aldehydes, the ligands were prepared with imidazolyl or triazolyl rings
having R groups of various sizes as shown in Figure 6.
Jain et al.
Manuscript File
16

Figure 6: Zinc-based SNS model complexes previously prepared by Miecznikowski et
al. [6, 23, 25]
Based on the data we previously published, the choice of the alkyl group is
important as zinc complex 3c (Figure 6) gave the highest percent conversion for the
conversion of 4-nitrobenzaldehyde to 4-nitrobenzyl alcohol in the presence of BNAH in
refluxing CDCl . [23].
3
In order to correlate the biochemical assay results and the cyclic voltammetry
results with our model system reactivity data, we screened zinc complex 3c for activity
through the reduction of a variety of aldehydes in the presence of BNAH (eq. 1 and 2).
Jain et al.
Manuscript File
17

Table 1 illustrates the reactivity data for the most active zinc complex, 3c, as well as
for ZnCl (eq. 1 and 2). For all reactivity experimentation, 0.1 mmol of aldehyde, 0.1
2
mmol zinc precursor or 0.2 mmol of ZnCl , and 0.2 mmol BNAH were used. Product
2
1
formation was detected by H NMR by comparison with authentic material.
1
For the conversion of 4-nitrobenzaldehyde to 4-nitrobenzyl alcohol, H NMR was
used to follow the disappearance of the aldehyde proton (δ 10.2 ppm) and the shifting of
the aromatic C-H protons in the alcohol product (δ 8.24 ppm). For the reduction of 2-
1
pyridine carboxaldehyde, H NMR spectroscopy was used to follow the disappearance of
the aldehyde proton (δ 10.1 ppm) and the appearance of the aromatic C-H protons in the
1
alcohol product (δ 8.0 ppm). For the conversion of benzaldehyde to benzyl alcohol, H
NMR spectroscopy was used to follow the disappearance of the aldehyde proton (δ 10.0
ppm) and the appearance of the methylene protons in the alcohol product (δ 4.7 ppm).
1
Finally, for the reduction of p-anisaldehyde, H NMR spectroscopy was used to follow
the disappearance of the aldehyde proton (δ 9.9 ppm) and the appearance of the
methylene protons in the alcohol product (δ 4.8 ppm). For all of the substrates, the
aromatic proton resonances of the C-H protons in the alcohol product were
spectroscopically distinct from the other products or the starting material’s resonances
1
and no overlap of H NMR resonances was observed (Figure S2, Supplementary
Information).
As shown in Table 1, the SNS zinc(II) complex enhances the rate of the reaction for
the conversion of 4-nitrobenzaldehyde to 4-nitrobenzyl alcohol when compared to ZnCl2.
In no case was there any indication of reduction of nitro substituents. Use of the SNS
zinc(II) complex results in quantitative conversion of 2-pyridine carboxaldehyde to
Jain et al.
Manuscript File
18

pyridin-2-ylmethanol. 2-pyridine carboxaldehyde may be a good substrate for reduction
by the zinc complex since there may be faster hydrogen transfer between the co-factor
and the substrate, upon coordination to the zinc active site. Mechanistically, others have
proposed a hydrogen atom transfer between the cofactor and the substrate upon
coordination to the zinc active site based upon a previously reported mechanism for
LADH enzyme offered by Makowska-Grzyska and co-workers. [35] A proposed
mechanism for the reduction of aldehydes by zinc(II) containing LADH is given below in
figure 7.
Figure 7: Plausible mechanism for the reduction of aldehydes by LADH.
Jain et al.
Manuscript File
19

Table 1. Stoichiometric reactivity data for the reduction of aldehydes by a SNS zinc(II)
pincer complex.
Conditions: 0.1 mmol aldehyde, 0.2 mmol BNAH, 0.1 mmol SNS Zn Complex or 0.2
mmol ZnCl , and 3 mL CDCl . All reactions were run for 20 hrs.
2
3
The more electron rich aldehydes (Table 1, entries 7 – 12) were not reduced when
the aldehydes were reacted with BNAH and the zinc complex. For these examples, the
substrate itself may poison the zinc complex by coordinating to the zinc complex during
the reaction. For the substrates where reduction was observed, the alcohol product may
inhibit the reaction. As the alcohol is formed, it may coordinate to the zinc metal center
as the reaction progresses, and thereby hinder the reaction.
Entries 2 and 5 in Table 1 indicate that ZnCl reacts stoichiometrically with
2
electron-withdrawing aldehydes such as 4-nitrobenzaldehyde or 2-pyridine
carboxaldehyde to yield the alcohol product to a small extent (ca. 18-19 % conversion).
Jain et al.
Manuscript File
20

2
We therefore propose that Zn acts as a Lewis acid catalyst in the reaction where ZnCl
+
2
is utilized for these substrates. Very little, if any, reduction was seen when ZnCl was
2
reacted stoichiometrically with the other aldehyde substrates listed in table 1 (entries 8,
and 11) and shown in equation 1. Results obtained with our zinc model complex 3c
parallel those obtained with natural enzyme because electron withdrawn substrates, 4-
nitrobenzaldehyde and 2-pyridine carboxaldehyde, show substantially higher reactivity in
both the studies. Furthermore electron rich aldehydes, benzaldehyde and p-anisaldehyde
show minimal reactivity, which is also in agreement with our results in both the
enzymatic and model complex systems.
Conclusions
Under physiological conditions, BNAH was used as a biomimetic for the natural
cofactor NADH in reduction reactions catalyzed by liver alcohol dehydrogenase enzyme.
In order to better understand the changes in reaction rates in the presence of NADH and
BNAH, we conducted a substrate study and found that natural cofactor NADH had a poor
2
correlation (R = 0.544) between the inductive effect exerted by the substrates and their
corresponding enzymatic rate, while reactions conducted in the presence of BNAH
2
cofactor showed an excellent correlation (R = 0.972) between the inductive effect
exerted by the substrates and the observed reaction rates. These results demonstrate that
catalysis in the presence of NADH is largely dependent upon steric factors, whereas
catalysis in the presence of BNAH is primarily influenced by the substrate’s inductive
effect. We also compared the catalytic activity of LADH enzyme with a synthetic model
enzyme complex, tridentate SNS zinc(II) pincer compound that models the zinc(II) active
Jain et al.
Manuscript File
21

site of the LADH enzyme in terms of chemical structure and redox properties. Current
work in our laboratory is focused on the modifications of BNAH at the 1 and 6 positions
of the nicotinamide ring with electron donating and electron withdrawing functional
groups. Our aim is to determine whether the inductive effect argument and its
subsequent effect on reaction rates can be extended to cofactors and not just substrates
alone.
Acknowledgements
This work is supported by Research Corporation grant # 21054 (SSJ) and a Dreyfus
Corporation Undergraduate Research and Lectureship award (SSJ). Generous support is
also acknowledged from The Bard Summer Research Institute (SSJ, JS, XT), The
Fairfield University Science Institute (JRM, CEV, EEB, and LCK), Bard College (SSJ)
and Fairfield University (JRM) start-up funding, Fairfield University Research Grants
(
(
JRM), and the Fairfield University Chemistry Department Summer Research Kuck Fund
CEV, EEB, and LCK).
Jain et al.
Manuscript File
22

References
[
[
[
1] H. Eklund, C.I. Brändén, J. Biol. Chem. 254 (1979) 3458–3461.
2] Ramaswamy, D.H. Park, B.V. Plapp, Biochemistry. 38 (1999) 13951–13959.
3] K.K. Kannan, B. Notstrand, K. Fridborg, S. Loevgren, A. Ohlsson, M. Petef. Proc
Natl Acad Sci USA. 72 (1975) 51–55.
[4] H.C. Lo, R.H. Fish, Angew Chem Int Ed. 41 (2002) 478–481.
[5] R. Meijers, R.J. Morris, H.W. Adolph, A. Merli, V.S. Lamzin, E.S. Cedergren-
Zeppezauer. J. Biol. Chem. 276 (2001) 9316-9321.
[6] J.R. Miecznikowski, W. Lo, M.A. Lynn, S. Jain, L.C. Keilich, L.C.; N.F. Kloczko,
B.E. O’Loughlin, A.P. DiMarzio, K.M. Foley, G.P. Lisi, D.J. Kwiecien, E.E. Butrick, E.
Powers, R. Al-Abbasee, R. Inorg Chim Acta, 387 (2012) 25-36.
[7] J. Lutz, F. Hollmann, T.V. Ho, A Schnyder, R.H. Fish, A. Schmid. J Organomet
Chem. 689 (2004) 4783–4790.
[8] K.K. Kannan, B. Nostrand, B. K. Fridborg, S. Lovgren, A. Ohlsson, M Petef,
Proc. Natl. Acad. Sci, USA, 72 (1975) 51.
[9] J.A. Ibers, R.H. Holm, Science. 209 (1980) 223–235.
[10] L. Cronin, P.H. Walton, Chem Commun. 20 (2003) 2638.
[
11] A. Dołęga, A. Pladzyk, K. Baranowska, J. Jezierska, Inorg. Chim. Acta. 362 (2009)
5
085–5096.
[
[
[
[
[
12] A. Dołęga. Wiadomosci Chemiczne, 64 (2010) 389-411.
13] A. Kropidowski, J. Chojnacki, B. Becker. J Inorg Biochem, 101 (2007) 578-584.
14] F. Effenberger, J. Roos, C. Kobier. Angew. Chem Int. Ed, 41 (2002) 1876-1879.
15] O. Seneque, O. Reinaud, M. Giorgi. Chem Commun (2011) 984-985.
16] L. M. Berreau, M.M. Makowska-Grzska, A.M. Arif. Inorg Chem, 40 (2001) 2212-
2
213.
[
2
17] C. Kimblin, B.M. Bridgewater, D.G. Churchill, G. Parkin, Chem. Commun, (1999)
301-2302.
[18] C. Kimblin, T. Hascall, G. Parkin, Inorg. Chem, 36 (1997), 5680-5681.
Jain et al.
Manuscript File
23

[
5
19] S.C. Shoner, K.J. Humphreys, D. Barnhart, J.A. Kovacs, Inorg Chem, 34 (1995)
933-5934.
[
20] D.T. Corwin, Jr., R. Fikar, S.A. Koch, Inorg Chem, 26 (1987), 3079-3080.
21] A.K.H. MacGibbon, S.C. Koerber, K. Pease, M.F. Dunn, Biochemistry 26 (1987)
[
3
058-3067.
[22] R.G. Khalifah, Inorg Chim Acta, 32 (1979), L53-L55.
[
23] J.R. Miecznikowski, W. Lo, M.A. Lynn, B.E. O’Loughlin, A.P. DiMarzio, A M.
Martinez, L. Lampe, K.M. Foley, L.C. Keilich, G.P. Lisi, D.J. Kwiecien, C.M.
Pires, W.J. Kelly, N.F. Kloczko, K.N. Morio. Inorg Chim Acta, 2011, 376, 515-
5
24.
[24] a) R. Walz, H. Vahrenkamp, Inorg. Chim. Acta 314 (2001) 58-62. (b) Y.H. Zhang,
H. Vahrenkamp, Inorg. Chim. Acta, 351(2003) 201-206.
[25] J.R. Miecznikowski, J.P. Jasinski, M.A. Lynn, S. Jain, E.E. Butrick, A.R.
Drozdoski, K.A. Archer, J.T. Panarra. Inorg Chim Acta, 394 (2013) 310-321.
[
[
[
[
[
26] L.P. Hammett, J Am. Chem Soc. 59 (1937) 96–103
27] C.H. Blomquist, Acta Chem Scand. 20 (1966)1747–1757.
28] R.F. Boyer, Concepts in Biochemistry, first ed., Brooks/Cole Pub Co, pp.147–162.
29] W.K. Seok, T.J. Meyer, Inorg Chem. 44 (2005) 3931–3941.
30] W.P. Jencks, Catalysis in Chemistry and Enzymology, McGraw-Hill Book Company,
New York, 2000.
[
[
[
[
[
31] S. Ramaswamy, H. Eklund, B.V. Plapp, Biochemistry. 33 (1994) 5230–5237.
32] D.V. Yandulov, R.R. Schrock, Science 2003, 301, 76-78.
33] J.R. Miecznikowski, R.H. Crabtree, Organomet 23 (2004) 629–631.
34] J.R. Miecznikowski, R.H. Crabtree, Polyhedron. 23 (2004) 2857–2872.
35] M.M. Makowska-Grzyska, P.C. Jeppson, R.A. Allred, A.M. Arif, L.M. Berreau.
Inorg Chem. 41 (2002) 4872–4887.
Jain et al.
Manuscript File
24

Graphical Abstract
Synopsis for Graphical Abstract (50 words)
We have compared the catalytic activity of liver alcohol dehydrogenase enzyme to a
synthetic model active site zinc complex. Substrates that vary in their ability to
donate or withdraw electron density from the reaction center were used to
differentiate between geometric and electronic factors that govern the observed
reaction rates.
Jain et al.
Manuscript File
25

Highlights
•
•
Liver alcohol dehydrogenase enzyme can catalyze reduction reactions in the
presence of BNAH (a biomimetic of NADH cofactor).
Catalytic activity of the enzyme was compared to a synthetic model zinc
active site complex.
•
•
Substrates were varied to determine the role of steric and electronic factors.
In the presence of NADH, steric factors govern reaction rates whereas in the
presence of BNAH, the electronic environment of the substrate determines
the reaction rates.
Jain et al.
Manuscript File
26

Table 1. Stoichiometric reactivity data for the reduction of aldehydes by a SNS zinc(II)
pincer complex.
Entry
Substrate
[{SNS}Zn
Complex] or
ZnCl2
Conversion
(%)
1
2
3
4-nitrobenzaldehyde
4-nitrobenzaldehyde
4-nitrobenzaldehyde
None
ZnCl2
SNS
< 5
18
59
4
5
6
2-Pyridine carboxaldehyde
2-Pyridine carboxaldehyde
2-Pyridine carboxaldehyde
None
ZnCl2
SNS
< 5
19
Quantitative
7
8
9
Benzaldehyde
Benzaldehyde
Benzaldehyde
None
ZnCl2
SNS
< 5
< 5
< 5
1
1
1
0
1
2
p-anisaldehyde
p-anisaldehyde
p-anisaldehyde
None
ZnCl2
SNS
< 5
< 5
< 5
Conditions: 0.1 mmol aldehyde, 0.2 mmol BNAH, 0.1 mmol SNS Zn Complex or 0.2
mmol ZnCl , and 3 mL CDCl . All reactions were run for 20 hrs.
2
3
Jain et al.
Manuscript File
27