Synthetic modeling of the structure and function of the rare-earth dependent methanol dehydrogenase cofactor

Article abstract of DOI:10.1016/bs.mie.2021.01.037

Historically, rare-earth ions have been considered irrelevant to biology. Recently, the active sites of certain methanol dehydrogenase (MDH) enzymes have been shown to contain a redox-inactive, rare-earth (RE) cation coordinated by the redox-active pyrroloquinoline quinone (PQQ) cofactor. Importantly, it was demonstrated that rare earths were essential for the growth of certain methylotrophs that incorporated the XoxF-MDH. In this chapter, we summarize the optimized synthesis of a previously published rare-earth complex that serves as a model of the active site of this RE-containing MDH enzyme. The structure and reactivity of the metalated complex, [La(L_QQ)(NO₃)₃] are also discussed. [La(L_QQ)(NO₃)₃] catalytically oxidizes the test alcohol substrate, p-methylbenzyl alcohol, ^4MeBnOH, to p-methylbenzaldehyde, ^4MePhCHO, in the presence of a base (2,6-lutidine) and a terminal oxidant (ferrocenium hexafluorophosphate) with ~ 17 turnovers. By studying this synthetic model, we have developed a body of evidence about both the reactivity and the mechanism of dehydrogenation of alcohols as a molecular analogue to a native, rare-earth dependent enzyme.

Full text of DOI:10.1016/bs.mie.2021.01.037

CHAPTER TWO
Synthetic modeling of the
structure and function of the
rare-earth dependent methanol
dehydrogenase cofactor
Alison L. Knasin and Eric J. Schelter∗
P. Roy and Diana T. Vagelos Laboratories, Department of Chemistry, University of Pennsylvania, Philadelphia,
PA, United States
*
Corresponding author: e-mail address: schelter@sas.upenn.edu
Contents
1
2
3
4
5
6
. Introduction
. Ligand design and realization
. LQQ—Synthetic pathway and purification
. Structure and reactivity of [La(LQQ)(NO
. Summary
20
24
30
37
43
43
43
45
46
47
48
49
50
52
52
3 3
) ]
. Experimental section
6
6
6
6
6
6
6
.1 Preparation of compound 4
.2 Preparation of compound 5
.3 Preparation of compound 6
.4 Preparation of compound 11
.5 Preparation of compound 13
.6 Preparation of compound 14
.7 Preparation of compound 16
Acknowledgments
References
Abstract
Historically, rare-earth ions have been considered irrelevant to biology. Recently, the
active sites of certain methanol dehydrogenase (MDH) enzymes have been shown
to contain a redox-inactive, rare-earth (RE) cation coordinated by the redox-active
pyrroloquinoline quinone (PQQ) cofactor. Importantly, it was demonstrated that rare
earths were essential for the growth of certain methylotrophs that incorporated the
XoxF-MDH. In this chapter, we summarize the optimized synthesis of a previously pub-
lished rare-earth complex that serves as a model of the active site of this RE-containing
MDH enzyme. The structure and reactivity of the metalated complex, [La(LQQ)(NO
3 3
) ]
are also discussed. [La(LQQ)(NO ) ] catalytically oxidizes the test alcohol substrate,
3 3
Methods in Enzymology, Volume 650
ISSN 0076-6879
https://doi.org/10.1016/bs.mie.2021.01.037
Copyright # 2021 Elsevier Inc.
All rights reserved.
1
9

20
Alison L. Knasin and Eric J. Schelter
4Me
4
Me
p-methylbenzyl alcohol, BnOH, to p-methylbenzaldehyde, PhCHO, in the presence
of a base (2,6-lutidine) and a terminal oxidant (ferrocenium hexafluorophosphate)
with ꢀ17 turnovers. By studying this synthetic model, we have developed a body of evi-
dence about both the reactivity and the mechanism of dehydrogenation of alcohols as a
molecular analogue to a native, rare-earth dependent enzyme.
Abbreviations
AE
addition-elimination
DFT
HT
density functional theory
hydride transfer
MDH methanol dehydrogenase
PQQ
RE
pyrroloquinoline quinone
rare earth
1. Introduction
The 17 rare-earth (RE) elements comprise the group III elements
Sc, Y, and the lanthanide series (La-Lu). Although they are indicated as
rare,” these elements have natural abundances similar to that of d-block
“
metals, including copper, cobalt, and zinc (Haxel, Hedrick, & Orris, 2002;
Rudnick & Gao, 2003). The RE elements display similar chemical properties
under ambient conditions, with the 3+ oxidation state being the most chem-
ically stable. The RE elements are electropositive, kinetically labile, and
Lewis acidic. Along the lanthanide series, there is a decrease in the size of
the elements’ atomic and ionic radii, due to population of the 4f-electron
shell, an effect referred to as the lanthanide contraction. The 4f valence orbitals
exhibit almost entirely ionic bonding, which leads to complexes of the rare
earths exhibiting only small crystal field effects with typical coordination
numbers between 8 and 12 (Cheisson & Schelter, 2019). The contraction also
results in an increase in the Lewis acidity of these elements across the series;
3+
3+
La is the least Lewis acidic cation and Lu is the most Lewis acidic of the
lanthanide series. The RE elements have many uses in everyday life. Their
electronic characteristics are applied in electronic displays, magnets, and wind
turbines, among other applications. Previously, one of the only applications
for these elements within biology was use as contrast agents within medical
imaging tools (Cheisson & Schelter, 2019). However, it has now been shown
that these elements have a role in biology.
Microbiologists have demonstrated that the rare-earth elements are essen-
tial for some methanotrophic life (Pol et al., 2014). Through comparison of

Rare-earth dependent methanol dehydrogenase cofactor
21
the amounts of depleted rare earths to methane concentration associated with
the Deepwater Horizon oil spill, methanotrophs were found to uptake and use
REs during metabolism of methane. These results contributed to evidence
that implicates REs as biologically significant metal ions (Huang et al.,
2019; Shiller, Chan, Joung, Redmond, & Kessler, 2017). Recently,
Cotruvo and coworkers found that a newly discovered rare-earth sequestering
8
protein, lanmodulin, exhibits a ꢀ10 higher affinity toward rare-earth ions
3
+
2+
RE ) compared to calcium ions (Ca ), indicating that Nature has evolved
(
a distinct transport mechanism for rare-earth cations (Cotruvo, Featherson,
Mattocks, Ho, & Laremore, 2018). Collectively, these results show that bio-
logical systems affect rare-earth dependent reactivity, as well as potential
rare-earth sequestration. Recently, others have found prokaryotes that can
uptake and use rare-earth elements, which indicates previously unrecognized
biological significance (Bonificio & Clarke, 2016; Cotruvo, 2019; Daumann,
2019; Park, Brewer, Reed, Lammers, & Jiao, 2017). Whereas the biology of
RE-containing cofactors has been the focus of increased investigation, far less
research has focused on using molecular chemistry to model RE-based
biochemistry.
Methanol dehydrogenase (MDH) enzymes contain a redox-active
pyrroloquinoline quinone (PQQ) cofactor bound to a metal cation. They
oxidize methanol to formaldehyde, generating carbon sources used as
energy and as biomass. In 2011, researchers found some methanotrophs
to contain certain MDH enzymes (termed XoxF-type MDHs) which could
3+
uptake and use rare-earth cations (RE ), rather than typically observed cal-
2+
cium cations (Ca ), to catalyze alcohol oxidation (Hibi et al., 2011;
Nakagawa et al., 2012) (Fig. 1). The following year, Nakagawa and
3+
coworkers implicated a catalytic role for the La
cation within
XoxF-type MDH (Nakagawa et al., 2012). Subsequently, in 2014, certain
acidophilic, methanotrophic bacteria were shown to contain an XoxF-type
3+
MDH. This enzyme shows optimal activity with RE , whereas MxaF-
2+
MDH displays optimal activity with Ca ions (Huang, Yu, & Chistoserdova,
018; Keltjens, Pol, Reimann, & Op den Camp, 2014; Pol et al., 2014;
2
Shiller et al., 2017). The XoxF- and MxaF-type MDHs contain about a 50%
amino acid sequence homology (Deng, Ro, & Rosenzweig, 2018). The
inner-coordination sphere of these MDHs differs only by one aspartate residue
3+
that binds to the RE ions in the XoxF-type homodimers, presumably due
to differences in both size and charge of the redox-inactive metal center
(Fig. 2).

22
Alison L. Knasin and Eric J. Schelter
Fig. 1 (A) Enzymatic reactions for the oxidation of methane to formaldehyde,
B) Oxidation of substrates by PQQ, (C) Active site of a rare-earth dependent MDH with
(
ethanol as substrate. A portion of PQQ is highlighted to show the conserved motif within
the biomimetic ligand described in this work. Adapted with permission from the American
Chemical Society: McSkimming, A., Cheisson, T., Carroll, P. J., & Schelter, E. J. (2018). Functional
synthetic model for the lanthanide-dependent quinoid alcohol dehydrogenase active site.
Journal of the American Chemical Society, 140(4), 1223–1226. https:/doi.org/10.1021/
jacs.7b12318. Copyright 2018 American Chemical Society.
Fig. 2 Overview of MDH reactivity (top) and comparison of active sites of XoxF- and
3
+
2+
MxaF-type MDHs with ionic radii of La and Ca (Shannon, 1976). These two depic-
tions of complexes show an open coordination site reserved for binding of either a
solvent molecule or substrate (bottom).

Rare-earth dependent methanol dehydrogenase cofactor
23
Molecular model compounds have contributed to understanding systems
such as methane monooxygenases (both particulate and soluble methane
monooxygenases, pMMO and sMMO, respectively) and the oxygen-
evolving complex of Photosystem II (Cady, Crabtree, & Brudvig, 2008;
Culpepper & Rosenzweig, 2012; Kanady, Tsui, Day, & Agapie, 2011;
Tinberg & Lippard, 2011). These synthetic models give researchers the
opportunity to investigate the mechanistic details of enzymatic active sites
without requiring the specific conditions that biological systems often
require. Thus, it would be advantageous to use molecular chemistry to study
the active sites and the reactivity of RE-containing enzymes.
Only a few reports of PQQ model complexes have been published;
these reports do not necessarily include structural characterization data.
Additionally, there have been few reports of synthetic rare-earth complexes
of PQQ or its analogues (Itoh et al., 1995; Itoh, Kawakami, & Fukuzumi,
1997, 1998; Mitome, Ishizuka, Shiota, Yoshizawa, & Kojima, 2013, 2015;
Nakamura, Kohzuma, Kuma, & Suzuki, 1994; Suzuki, Sakurai, Itoh, &
Ohshiro, 1988; Wanner, Sixt, Klinkhammer, & Kaim, 1999). The challenges
associated with modeling the MDH active site include possible non-natural
binding motifs, such as binding through the two quinone oxygen atoms,
or the possibility of multiple quinoline quinone moieties binding to a single
metal center (Chen et al., 2012; Liu et al., 2009; Wei et al., 2015). Recently,
Daumann and coworkers isolated and crystallized Eu-PQQ outside of a pro-
tein environment for the first time (Lumpe et al., 2020). They found that the
3
+
complex crystallizes as a dimer, Eu PQQ , with Eu binding to the biolog-
2
2
ically relevant sites of PQQ.
Our group published the synthesis of a ligand framework, LQQ, to serve
as the model complex of PQQ (Fig. 3). The metalated complex with lan-
thanum, [La(LQQ)(NO ) ], was the first RE-containing model system. This
3 3
system can be used to study the mechanism of RE-dependent, catalytic alco-
hol oxidation. In this chapter, we discuss our route to synthesize the ligand
LQQ, and the optimization of its synthesis and purification. Additionally,
we discuss the metalation of LQQ with lanthanum and the reactivity of this
complex. We also describe the computational work performed using this
complex to study the mechanism of dehydrogenation of p-methylbenzyl
4
Me
4Me
alcohol ( BnOH) to p-methylbenzaldehyde ( PhCO). The develop-
ment of [La(LQQ)(NO ) ] allowed us to investigate reactivity and mecha-
3 3
nism, and to draw a connection with the native biological enzyme.

24
Alison L. Knasin and Eric J. Schelter
Fig. 3 Structural comparison of PQQ and LQQ
.
2. Ligand design and realization
Interest in XoxF-type MDHs was sparked when the X-ray structure
of this enzyme, isolated from Methylacidiphilum fumariolicum SolV (PDB:
4
MAE), was shown to contain a rare-earth cation within its active site that
was directly bound to the redox-active PQQ cofactor (Bogart, Lewis, &
Schelter, 2015; Huang et al., 2018; Keltjens et al., 2014; Pol et al., 2014;
Shiller et al., 2017). The RE-dependent MDH showed faster methanol oxi-
dation rates (Keltjens et al., 2014; Picone & Op den Camp, 2019). The
Michaelis constants (K ) for XoxF-type Ce-MDH from M. fumariolicum
M
SolV and MxaF-type Ca-MDH from Pseudomonas W1 were reported as
0
.8 and 15 μM, respectively (Keltjens et al., 2014). The k /KM values,
cat
also known as the specificity constant, for the oxidation of methanol to
formaldehyde for these two strains of MDHs were reported at 11,600 and
ꢁ ꢁ1
1
2
40s mM , respectively (Keltjens et al., 2014). This RE-dependent
MDH was shown to also oxidize formaldehyde to formic acid, which
was not observed for the Ca-containing analogue. These observations con-
tributed to earlier reports of rare-earth cations having a biological role and a
competitive advantage over Ca-dependent MDH. It was hypothesized
that rare-earth cations were more effective than calcium ions in their role
of substrate activation, due to increased Lewis acidity. Density functional
theory (DFT) was used to evaluate and compare the electronic structure
of the active sites of XoxF- and MxaF-type MDHs (Bogart et al., 2015).

Rare-earth dependent methanol dehydrogenase cofactor
25
Notably, two different mechanistic pathways have been suggested for
methanol dehydrogenation in Ca-MDHs, a hydride transfer (HT) or
addition-elimination (AE) pathway. The rate-limiting step of the hydride
transfer mechanism is a concomitant transfer of a hydride from the methyl
group to the electrophilic quinone carbon and the proton of the alcohol to
the carboxylates from pendant amino acid residues (Fig. 4). Alternatively, for
the addition-elimination step, the rate-limiting step is a nucleophilic addition
of the alcohol to the electrophilic carbon and concomitant proton transfer
to the acetal oxygen atom (Fig. 5). To rationalize the biological mechanism
of oxidizing methanol, a synthetic model can be used to investigate these
different mechanisms.
The reported solution electrochemistry of PQQ (and its requisite salts)
has been challenging to measure unambiguously, due to a strong pH depen-
dence. Our group began to study the electrochemical properties of the
quinone cofactor PQQ, by synthesizing a series of PQQ surrogates 3a–c
(Scheme 1) (Dorfner, Carroll, & Schelter, 2015). This approach allowed
for a predictive method of determining the first and second reduction poten-
tials, with the goal of predicting these properties of other PQQ analogues
not yet synthesized. To achieve this predictive power, the computed
ΔG values of various o- and p-quinones were plotted against their experi-
mental reduction potentials, which allowed for an experimental vs compu-
tational correlation to be established.
We expected to use this predictive method to study and synthetically mod-
ify the electronic properties of quinone cofactors. These species would then be
coordinated with rare-earth cations and examined to gain information about
XoxF-type MDH. However, once coordinated to rare earths, the resulting
complexes were found to be reactive with solvent and the coordination chem-
istry was difficult to control and study. Thus, we were motivated to synthesize a
different ligand that featured steric motifs enforced by XoxF-type MDH, to
ultimately control the binding motif of the rare-earth cation.
The ligand LQQ (Scheme 2) was synthesized to probe the mechanism for
a model complex of RE-dependent MDH (McSkimming, Cheisson,
Carroll, & Schelter, 2018). A 14-step synthesis was developed and optimized
(Scheme 3), with the goal to isolate large quantities of this ligand to allow for
exploratory synthetic and mechanistic experimentation. The ligand LQQ is
sterically saturated to replicate the structure and bonding of native MDH.
Additionally, the nitrate counterions serve as a “proton shuttle,” similarly
to the carboxylates in amino acids within the natural system.

Fig. 4 Dehydrogenation of methanol to formaldehyde by Ca-MDH through the hydride transfer mechanism.

Fig. 5 Dehydrogenation of methanol to formaldehyde by Ca-MDH through the addition-elimination mechanism.

Scheme 1 Initial attempt to synthesize a series of PQQ analogues.

Rare-earth dependent methanol dehydrogenase cofactor
29
Scheme 2 Synthesis of LQQ
.
Scheme 3 Modified, optimized synthesis of LQQ
.

30
Alison L. Knasin and Eric J. Schelter
3
. L —Synthetic pathway and purification
QQ
The reaction conditions for three of the steps of the synthesis of LQQ
were improved from the previously published synthetic pathway to repro-
ducibly provide higher yields and purities (McSkimming et al., 2018). This
convergent synthesis consists of two parallel pathways, one comprising qui-
none 11 and one comprising aniline 16, followed by a Schiff base conden-
sation coupling as the terminal step (Scheme 2). It should be noted that the
synthesis described was scalable to a 1.5mol scale for the amidation reaction
to yield compound 4. In performing the synthesis of compounds 4–7, a
heating mantle and a mechanical stirrer should be used in place of magnetic
stir bar and oil bath conditions. At each step, purity of isolated material was
1
confirmed through use of H NMR spectroscopy.
The quinone synthesis begins with an amidation of 3-hydroxy-2-
naphthoic acid using urea and phosphoric acid, giving primary amide 4
(
Scheme 4), a procedure that was adapted from a reported synthesis
Georg, 1953). For this reaction, 6equiv. urea were used as solvent and
(
reagent. Once the urea was molten, 3-hydroxy-2-naphthoic acid was added
as a solid and 0.1equiv. phosphoric acid were added dropwise to the reaction
mixture at 150°C. Note: the addition of acid to the reaction mixture is an
exothermic process, which could lead to effervescence. This operation
should be performed slowly and closely monitored over the course of
the addition of acid. After the addition, the reaction was stirred at 180°C
for 3h. Decomposition occurs when the temperature exceeds 180°C,
evidenced by the mixture changing from an orange to a red color.
Orange solid was precipitated from solution by addition of 6L water to
the reaction mixture, which was subsequently cooled to room temperature.
The reaction mixture was filtered and dried on the filter, giving 4 as an
orange solid.
Scheme 4 Amidation of 3-hydroxy-2-naphthoic acid with phosphoric acid and urea to
yield primary amide 4.

Rare-earth dependent methanol dehydrogenase cofactor
31
In the next step, the naphthol of amide 4 was protected using tosyl chlo-
ride, to give tosylated 5 (Scheme 5). Once 4 was dissolved in dimethyl sulf-
oxide (DMSO) and 1.5equiv. triethyl amine (Et N) were added, the
3
reaction mixture was cooled until nearly frozen, as the addition of tosyl chlo-
ride was exothermic. 1.2equiv. tosyl chloride should be added slowly and
the reaction mixture stirred for 0.5h while still cold. A yellow solid precip-
itated from solution by adding water (three times the volume of DMSO
originally added). The yellow solid was filtered and washed with a copious
amount of water. This solid was resuspended in isopropanol, filtered and
dried on the filter, giving amide 5 as an off-white solid.
Scheme 5 Protection of compound 4 with tosyl chloride to yield compound 5.
Compound 6 was obtained through a Hofmann rearrangement and
deprotection (Scheme 6). A solution of 6.5M sodium hydroxide (NaOH)
and 1,4-dioxane was cooled to 10°C. While cold, 1equiv. bromine was
added dropwise using an addition funnel. Compound 5 was added to solu-
tion slowly and the reaction mixture was stirred for 0.5h while still cold. The
mixture was then refluxed at 100°C for 16h. After 1h, the light brown sus-
pension changed to a dark brown solution. The reaction mixture was cooled
to room temperature and acidified to pH 0 with 12M hydrochloric acid
(HCl). As a safety measure, ice should be added into the flask, along with
the HCl. The aqueous phase was separated using diethyl ether (Et O).
2
This aqueous layer was subsequently filtered. The solution was adjusted
to pH 3 with 3M NaOH and then to pH 8 with sodium carbonate
(Na CO ). The light brown solid that precipitated (6) was filtered and dried
2 3
on a B u€ chner funnel.
Scheme 6 Hofmann rearrangement and deprotection of compound 5, leading to
compound 6.

32
Alison L. Knasin and Eric J. Schelter
The following step is a Skraup reaction to produce 7 (Scheme 7). Aniline
was dissolved in a 3:1 solution of 6M HCl and toluene and heated to
00°C. 3equiv. crotonaldehyde were added using an addition funnel in a
:1 solution in toluene. This reaction mixture was stirred at 100°C for
h. Once cooled to room temperature, the organic layer was separated
6
1
1
3
using additional toluene and 6M HCl. The aqueous portion was adjusted
to pH 3 with 3M NaOH, and then to pH 8 with sodium bicarbonate
(
NaHCO ). The organic layer was separated with ethyl acetate, washed
3
with brine, and dried over sodium sulfate (Na SO ). The reaction mixture
4
2
was filtered and concentrated under reduced pressure to yield a brown oil.
This crude product was purified using flash chromatography (95:5 hexanes:
ethyl acetate eluent), as indicated by work from the Williams group, and
recrystallized twice from methanol to obtain 7 as a white solid (Bakewell
et al., 2012).
Scheme 7 Skraup reaction of compound 6 to yield compound 7.
The ensuing step consists of a tert-butyldimethylsilyl (TBS) protection of
7
to produce 8 as a white solid (Scheme 8). Compound 7 was dissolved in
0
N,N -dimethylformamide (DMF) and 2.5equiv. imidazole dissolved in
DMF were added. 1.3equiv. tert-butyldimethylsilyl chloride (TBSCl) were
added portion-wise and the reaction mixture was stirred at room tempera-
ture for 3h. A copious amount of distilled water was added slowly (200mL
over 10min) to precipitate 8 from solution. This solid was filtered and dried
on a medium porosity fritted filter.
Scheme 8 tert-Butyldimethylsilyl protection of compound 7 to obtain compound 8.

Rare-earth dependent methanol dehydrogenase cofactor
33
To obtain aldehyde 9, 1.3equiv. selenium dioxide (SeO ) were dissolved
2
in dioxane (Scheme 9). Compound 8 was dissolved in 1,4-dioxane and the
two solutions were mixed. This mixture was refluxed at 110°C for 16h.
Once cooled to room temperature, 250mL hexanes were added and the
crude mixture was passed through a pad of silica (ꢀ5cm in diameter and
ꢀ
10cm in height) using Et O as the eluent. The filtrate was concentrated
2
under reduced pressure and then 20mL of a 1:1 solution of dichloromethane
CH Cl ) and hexanes was added. This solution was passed through a second
(
2
2
pad of silica (ꢀ5cm in diameter and ꢀ10cm in height) using a 1:10 solution
of Et O and hexanes as the eluent. The filtrate was once again concentrated
2
under reduced pressure to yield 9 as a yellow solid. No further purification
was performed on the compound.
Scheme 9 Selenium dioxide oxidation of compound 8 to compound 9.
For the next step, a TBS deprotection of 9, led to aldehyde 10. Protected
aldehyde (8) was dissolved in a 1:3 mixture of tetrahydrofuran (THF) and
methanol (Scheme 10). Enough 12M HCl was added dropwise to solubilize
8
, and the solution was stirred at room temperature for 2h. A solid was pre-
cipitated from solution once 100mL saturated aqueous solution of
NaHCO were added. This yellow solid (10) was filtered, washed with hex-
3
anes, and dried on a medium porosity fritted filter.
Scheme 10 Deprotection of compound 9 with hydrochloric acid to yield compound 10.
In the final step of the quinone synthesis, the oxidation of phenol 10 to
quinone 11 was originally reported to employ IBX (o-iodoxybenzoic acid);
however, it was found that this gave quinone 11 together with o-iodobenzoic
acid as an impurity, which was difficult to remove. Changing the oxidant to

34
Alison L. Knasin and Eric J. Schelter
polymer-supported IBX yielded quinone 11 without the o-iodobenzoic acid
impurity (Scheme 11) (Dorfner et al., 2015). For this oxidation, compound 10
was dissolved in CH CN. 1.75equiv. polymer-supported IBX were added as
3
a solid and the reaction mixture was stirred at r.t. for 16h. The suspension was
filtered and the solid was washed with CH CN until the filtrate was colorless.
3
The filtrate was passed through a pad of Celite (ꢀ5cm in diameter and
ꢀ
10cm in height) using a 1:1 solution of Et O and CH Cl as the eluent.
2 2 2
The filtrate was concentrated under reduced pressure and compound 11
was precipitated out of solution with addition of 1,2-dichloroethane. This
orange solid was filtered and dried on a medium porosity fritted filter. No fur-
ther purification was required.
Scheme 11 Oxidation of compound 10 with polymer-supported o-iodoxybenzoic acid
to yield compound 11.
The chelator synthesis begins with a previously published peptide
coupling to give 12 as a brown solid (Scheme 12) ( Joshi et al., 2013). 1equiv.
dicyclohexylamine was dissolved in CH Cl and cooled to 10°C. To this
2
2
solution, 1.1equiv. chloroacetyl chloride were added as a solid and 2.5equiv.
Et N were added dropwise with stirring. The reaction mixture was stirred for
3
0
.5h at 10°C, allowed to warm to room temperature, and washed with
100mL saturated NaHCO3 and 100mL of brine. Once dried over
Na SO and filtered, the filtrate was concentrated under reduced pressure
4
2
to yield a brown solid. This crude solid required trituration with Et O to
2
remove an insoluble impurity. To do so, the crude solid was stirred in
1
00mL Et O and filtered. The filtrate contained the product, and the
2
filtered solid contained the insoluble impurity. The filtrate was once again
concentrated under reduced pressure to obtain 12 as a brown solid.
Scheme 12 Peptide coupling between dicyclohexylamine and chloroacetyl chloride to
yield compound 12.

Rare-earth dependent methanol dehydrogenase cofactor
35
The next step involved sequential S 2 reactions with benzylamine
N
(Scheme 13). Compound 12 was dissolved in acetonitrile (CH CN) with
3
0.5equiv. benzylamine, 2equiv. KI, and 2.6equiv. K CO . This reaction
2 3
mixture was stirred for 16h at 80°C. Once allowed to cool to room tem-
perature, 300mL distilled water were added, and the reaction mixture was
extracted three times with 50mL of a 3:1 solution of hexanes:ethyl acetate.
It should be noted that copious amounts of water were required to avoid a
three-phase mixture. Additional water increases the miscibility between the
acetonitrile and water layers, which ensured a biphasic work-up with aque-
ous and organic layers. After extraction, the organic layer was washed with
brine, dried over Na SO , and filtered. The filtrate was concentrated under
2
4
reduced pressure and the crude solid was recrystallized from concentrated
ethanol to obtain 13 as a white solid.
N
Scheme 13 S 2 reaction of compound 12 with benzylamine to give compound 13.
Benzylamine 13 was isolated as an ammonium salt from the above reac-
tion as evidenced by a tertiary ammonium N-H stretch observed in the IR
ꢁ
1
spectrum (2663cm ). Attempted hydrogenations with the ammonium salt
to produce 14 were unsuccessful, but the hydrogenation was successfully
performed in the presence of an exogenous base (Et N) (Scheme 14).
3
Once compound 13 was dissolved in 150mL ethanol, 1equiv. Et N was
3
added and the solution was stirred at room temperature for 15min. 9equiv.
ammonium formate and 10mol% activated Pd/C were added as solids and
the reaction mixture was refluxed at 80°C for 16h. Once allowed to cool to
room temperature, the crude mixture was filtered over Celite and washed
with ethanol until the eluent was colorless. The filtrate was concentrated
under reduced pressure to yield 14 as a white solid.
Scheme 14 Hydrogenation of compound 13 with palladium over carbon in the pres-
ence of an exogenous base to yield compound 14.

36
Alison L. Knasin and Eric J. Schelter
The next step involves a nucleophilic aromatic substitution reaction
(Scheme 15). A neat reaction of 14, 0.8equiv. 1-fluoro-2-nitrobenzene,
and 1.5equiv. K CO were stirred at 110°C for 16h. The reaction was
2
3
cooled to room temperature and the orange solid was dissolved in
50mL Et O and filtered over Celite. To precipitate unreacted starting
2
2
material, 150mL of 2M HCl were added and the resulting slurry was filtered
over Celite. The organic layer was separated, washed with brine, dried over
Na SO , and filtered. The filtrate was concentrated under reduced pressure
2
4
and the crude material was triturated with hexanes.
Scheme 15 Nucleophilic aromatic substitution reaction between compound 14 and
1-fluoro-2-nitrobenzene to yield compound 15.
The final step of the chelator synthesis, reduction of 15 to 16 was
reported using NiCl ꢂ2(H O) and NaBH (McSkimming et al., 2018).
2
2
4
However, we recently noted that SnCl ꢂ2(H O) in HCl gave clean conver-
2
2
sion to the aniline (Scheme 16). Compound 15 was dissolved in 20mL eth-
anol with minimal 12M HCl and 3.6equiv. SnCl ꢂ2(H O). The reaction
2
2
mixture was stirred at 60°C for 32h. Once allowed to cool to room tem-
perature, 75mL water were added, along with 3M NaOH until the solution
was pH 14, which led to the formation of a white precipitate. A saturated
solution of NaHCO was added and the reaction mixture was stirred at r.
3
t. for 15min. The slurry was filtered over Celite. The eluent was extracted
three times with 50mL of CH Cl , washed with 50mL of brine, dried over
2
2
Na SO , and filtered. The filtrate was concentrated under reduced pressure,
2
4
which afforded compound 16 as a brown solid.
Scheme 16 Tin chloride reduction of compound 15 to give compound 16.
The terminal step of the convergent synthesis of LQQ was a Schiff base
condensation of compounds 11 and 16 (Scheme 17). 1equiv. 11 and
1
.4equiv. 16 were dissolved at r.t. in CH Cl and stirred for 1h. The
2 2

Rare-earth dependent methanol dehydrogenase cofactor
37
reaction mixture was concentrated to about 5mL and hexanes was added to
precipitate LQQ as a red powder. This solid was collected by filtration,
washed three times with 25mL hexanes, and dried on a medium porosity
fritted filter. Prior to transfer into a glovebox, the ligand LQQ was dried
under dynamic vacuum overnight.
Scheme 17 Schiff base condensation between compounds 11 and 16 to give LQQ
.
4
. Structure and reactivity of [La(L )(NO ) ]
QQ 3 3
[La(LQQ)(NO ) ] was isolated from metalation of [La(NO ) (THF) ]
3 3 3 3 4
with LQQ as a moisture-sensitive solid. In a nitrogen-filled glovebox, 1equiv.
LQQ was suspended in CH CN and 1.1equiv. [La(NO ) (THF) ] were dis-
3
3 3
4
solved in CH CN. The [La(NO ) (THF) ] solution was added to the solution
3
3 3
4
of LQQ and the reaction mixture was stirred for 1h. An orange solid precip-
itated from the solution which was shown to be the complex: [La(LQQ
NO ) ]. The product was isolated by filtration, washed with CH CN, and
)
(
3
3
3
dried under reduced pressure (Fig. 6). Crystals of [La(LQQ)(NO ) ] were
3 3
obtained through slow diffusion of hexanes into dichloromethane. Analysis
of the structure by X-ray crystallography showed a binding motif with strong
similarities to the reported coordination environment of the XoxF-MDH
enzymatic active site (Fig. 7) (McSkimming et al., 2018; Pol et al., 2014).
2
Notably, the nitrate counteranions are bound in a κ -mode, analogous to
the carboxylate groups of the amino acids within the active site in the natural
system. Confirmation of the quinone moiety was determined by analysis of the
bond length of C(2)-C(3). A typical C-C bond length for quinone,
semiquinone, and catechol species are 1.530(5), 1.42(1)–1.448(9), and 1.348
˚
(
10)–1.415(6) A, respectively (Pierpont & Buchanan, 1981). This C(2)-C(3)
bond length within [La(LQQ)(NO ) ], as determined through analysis of the
3 3
˚
crystal structure, was 1.542(5) A, which confirmed the quinone assignment.
Solution electrochemical experiments, using cyclic voltammetry,
n
performed on LQQ in CH Cl with 0.1M [ Pr N][BAr ] showed a revers-
F
4
2
2
4
+
ible couple at E1/2 ¼ꢁ0.95V vs Fc /Fc, which was assigned as the

38
Alison L. Knasin and Eric J. Schelter
3 3 4 3 3
Fig. 6 Metalation of LQQ with [La(NO ) (THF) ] to give [La(LQQ)(NO ) ]. Ellipsoids are
shown at the 30% probability level. Hydrogen atoms and cyclohexyl groups have been
removed for clarity. Adapted with permission from the American Chemical Society:
McSkimming, A., Cheisson, T., Carroll, P. J., & Schelter, E. J. (2018). Functional synthetic
model for the lanthanide-dependent quinoid alcohol dehydrogenase active site. Journal
of the American Chemical Society, 140(4), 1223–1226. https: // doi.org/10.1021/jacs.
7
b12318. Copyright 2018 American Chemical Society.
3 3
Fig. 7 Comparison of XoxF-type MDH active site with [La(LQQ)(NO ) ].
quinone-semiquinone couple. There was also an irreversible process at
+
E ¼ꢁ1.76V vs Fc /Fc, due to the formation of the catecholate dianion.
pc
The CV experiments performed on [La(L )(NO ) ] under the same
QQ 3 3
conditions showed an anodic shift of the reversible quinone-semiquinone
+
couple to ꢁ0.34V vs Fc /Fc (Fig. 8), indicating an increased stability of this
reduction when the lanthanum cation was coordinated. These electrochem-
ical data were complicated and two quasi-reversible reduction events at
+
ꢁ
0.68 and ꢁ0.77V vs Fc /Fc were not definitively assigned. They are
possibly assignable to the loss of nitrate ligand(s) during the scan, giving
another species upon reduction. However, the anodic shifting of the
quinone-semiquinone couple by ꢀ600mV from free ligand to [La(LQQ)
(NO ) ] provided direct evidence that the semiquinone species was
3 3
stabilized by the bound metal ion. Thus, it was hypothesized that
La(LQQ)(NO ) ] would promote oxidation of alcohol, compared to free
[
3 3
ligand, LQQ
.

Rare-earth dependent methanol dehydrogenase cofactor
39
Fig. 8 Cyclic voltammograms of [La(LQQ)(NO
3
)
3
] (solid line) and LQQ (dashed line) in a
0
.1M solution of tetrapropylammonium tetrakis[3,5-bis(trifluoromethyl)phenyl]borate
n
F
ꢁ1
(
[ Pr
American Chemical Society: McSkimming, A., Cheisson, T., Carroll, P. J., & Schelter, E. J.
2018). Functional synthetic model for the lanthanide-dependent quinoid alcohol dehydro-
4 4 2 2
N][BAr ]) in CH Cl at a 100mVs scan rate. Adapted with permission from the
(
genase active site. Journal of the American Chemical Society, 140(4), 1223–1226. https://
doi.org/10.1021/jacs.7b12318. Copyright 2018 American Chemical Society.
Subjecting free LQQ in a CD Cl solution to fourfold excess of
2
2
4Me
BnOH showed no reaction under anaerobic conditions, as determined
4Me 1
PhCHO, using H NMR
by the absence of detectable aldehyde product,
spectroscopy, after 72h (Gnanaprakasam, Zhang, & Milstein, 2010;
Kawahara, Fujita, & Yamaguchi, 2012). Itoh and Fukuzumi demonstrated
that addition of base promoted the dehydrogenation of methanol to form-
aldehyde (Itoh et al., 1997, 1998). However, when 2.2equiv. of the base
DBU (DBU¼ 1,8-diazabicyclo(5.4.0)undec-7-ene) was added to solution
4
Me
4Me
with L
and
BnOH, only trace
PhCHO was formed: <3% yield
QQ
1
after 30min, 13% yield after 4h by H NMR spectroscopy. Conversely,
[
La(L )(NO ) ] in a CD Cl solution demonstrated the ability to produce
QQ
3 3
2
2
4
Me
PhCHO both with and without a base present, albeit more efficiently in
the presence of a base (Table 1).
4
Me
When performed stoichiometrically, the dehydrogenation of
BnOH
4Me
using [La(LQQ)(NO ) ] gave the oxidized product,
3 3
1
characterized by H NMR spectroscopy, as well as [La(LQQ )(NO3)]2
PhCHO, which was
2ꢁ
(
Fig. 9). This dimer, where LQQ was in the catechol redox state, was isolated
independently by performing a chemical reduction with addition of 2equiv.
Cp Co] and characterized by both X-ray crystallography and infrared spec-
[
2
troscopy. The reduction of a quinone to a catechol is a two-electron redox
process, as is the oxidation of an alcohol to an aldehyde. The product of

40
Alison L. Knasin and Eric J. Schelter
Table 1 Yields of p-methylbenzaldehyde from catalytic
oxidation of p-methylbenzyl alcohol.
Catalyst
None
LQQ
Time (h)
24
Yield (%)
<1
14
84
96
5
21
[
La(LQQ)(NO3)3]
21
44
[
[
La(NO ) (THF) ]
21
24
3
3
4
a
La(LQQ)(NO3)3]
30
67
67
72
b
[
La(LQQ)(NO3)3]
<10
a
2
.5equiv. alcohol in CD
2
Cl
2
.
b
Same conditions+2.2equiv. DBU.
this stoichiometric reaction was a dimer in the catechol redox state, which is
reminiscent of the enzymatic cofactor that cycles between quinone and
catechol states.
It was hypothesized that addition of a terminal oxidant and base had the
4
Me
potential to catalytically dehydrogenate
BnOH using [La(L )(NO ) ]
QQ 3 3
4Me
(
Table 1). To perform the catalytic oxidation, 1equiv.
BnOH, 2equiv.
2
,6-lutidine, and about 0.006mmol (SiMe ) O (as an internal standard)
2
3
were dissolved in a 1:1 mixture of CD Cl and CD CN (0.4mL total).
2
2
3
The 2,6-lutidine was added last, due to its volatility. It is important to note
that the pK of 2,6-lutidine is ꢀ7, which is closer to physiological pH (pH
a
ꢀ
7.4) when compared to the pK of DBU, at ꢀ12 ( Jia, Ramstad, & Zhong,
a
2
001; Kaupmees, Trumall, & Leito, 2014). The weaker base was chosen for
these studies due to its more physiologically relevant pK . Once these three
a
materials were dissolved, the reaction mixture was added to a J-Young tube,
1
and an initial H NMR spectrum was collected as a blank. After collecting
the blank, the contents of the tube were transferred back into a vial and 5mol
%
catalyst in minimal CD Cl and 2equiv. [Fc][PF ] were added. Note that
2 2 6
[
Fc][PF ] is poorly soluble. The solution within the vial was mixed for
6
1
1 and 44h, at which points, H NMR spectra were taken (by dissolving
2

2ꢁ
4Me
Fig. 9 Production of [La(LQQ )(NO
3
)]
2
both stoichiometrically with DBU and
BnOH and independently using a chemical reduction with
[
Cp Co]. Adapted with permission from the American Chemical Society: McSkimming, A., Cheisson, T., Carroll, P. J., & Schelter, E. J. (2018).
2
Functional synthetic model for the lanthanide-dependent quinoid alcohol dehydrogenase active site. Journal of the American Chemical
Society, 140(4), 1223–1226. https: // doi.org/10.1021/jacs.7b12318. Copyright 2018 American Chemical Society.

42
Alison L. Knasin and Eric J. Schelter
an aliquot of the vial in minimal deuterated solvent). Yields were determined
through a comparison to the internal standard added.
Having established that [La(LQQ)(NO ) ] was a competent catalyst for
3 3
alcohol dehydrogenation, this biomimetic system allowed the possible
mechanistic pathways of the natural system to be explored. However, it is
important to note the differences between the biomimetic system and the
natural enzyme. For example, in [La(LQQ)(NO ) ], nitrates are present as
3 3
the counteranion, analogous to the carboxylate groups of amino acids within
4Me
the natural system. For our system,
BnOH was used as a test substrate for
the dehydrogenation, as compared to methanol as the substrate in Nature.
Finally, DBU was used in our experiments, but this base is not biologically
relevant, as was previously described.
Despite the limitations, to better understand our system and gain infor-
mation about the natural system, a computational study was performed in
which reaction coordinates were determined for two possible mechanistic
pathways (addition-elimination and hydride transfer) both in the presence
and absence of a base (Fig. 10). These results indicated that, in the absence
of a base, the hydride transfer mechanism was the most energetically feasible,
albeit the reaction barrier is such that reactivity is slow at room temperature.
In the presence of 2,6-lutidine, the hydride transfer was also favored. These
4Me
Fig. 10 Reaction coordinates for the dehydrogenation of
BnOH by [La(LQQ)(NO3)3],
as calculated by DFT in CH Cl . Comparison of addition-elimination and hydride transfer
2
2
mechanisms in the presence and absence of bases. Adapted with permission from the
American Chemical Society: McSkimming, A., Cheisson, T., Carroll, P. J., & Schelter, E. J.
(
2018). Functional synthetic model for the lanthanide-dependent quinoid alcohol dehydro-
genase active site. Journal of the American Chemical Society, 140(4), 1223–1226. https://
doi.org/10.1021/jacs.7b12318. Copyright 2018 American Chemical Society.

Rare-earth dependent methanol dehydrogenase cofactor
43
results were in agreement with the generally accepted mechanism for bio-
logical MDH enzymes (Idupulapati & Mainardi, 2010; Kay, Mennenga,
G o€ risch, & Bittl, 2006; Zhang, Reddy, & Bruice, 2007; Zheng, Xia,
Chen, Mathews, & Bruice, 2001).
5. Summary
In this chapter, we have detailed the synthesis and subsequent reactivity
study of a synthetic model complex of a rare-earth containing enzyme. The
ligand LQQ is synthesized through a 14-step pathway. Once metalated with
lanthanum, this system acts as a biomimetic surrogate for XoxF-type MDH,
a rare-earth dependent enzyme that catalyzes the dehydrogenation of methanol
to formaldehyde. The experimental findings for [La(L )(NO ) ] indicated
QQ
3 3
4Me
that both stoichiometric and catalytic oxidization of
BnOH were achieved,
showing that this complex serves as a biomimetic system. DFT computational
work performed using the model suggested a hydride transfer mechanism as the
4Me
lowest energy route to dehydrogenate
BnOH, which supports the hypoth-
esis for reactivity for the biological system.
6. Experimental section
This section details alterations made to the previously reported synthe-
sis of LQQ (McSkimming et al., 2018). These changes, where noted, pro-
vided improvements in yields and/or reproducibility of the individual
steps. Steps not explicitly included within this section were completed as
described in the original work (McSkimming et al., 2018). All large-scale
reactions were performed in a three-neck 12L round bottom flask. These
reactions were stirred with an overhead stirrer, heated with a heating mantle,
and cooled by immersion of the flask in a 5L bucket. A three-neck flask was
chosen so reaction mixtures could be stirred with the overhead stirrer while
reagents and solvents were added at the same time.
6
.1 Preparation of compound 4
Note—This synthesis was adapted from conditions reported in a patent
Georg, 1953).
(
a. 600g of urea were heated to 150°C until melted.
b. 300g of 3-hydroxy-2-naphthoic acid were added to yield a brown
mixture.
c. While still heating, 9mL phosphoric acid were added dropwise, and the
heat was increased to 180°C. Safety note—addition of phosphoric acid
leads to bubbling within the reaction. This addition is exothermic and
can effervesce if added too quickly.

44
Alison L. Knasin and Eric J. Schelter
d. The solution was stirred while heating at 180°C for 2h, after which the
reaction mixture was cooled to 150°C.
e. Orange solid was precipitated by addition of 6L H O to the reaction
2
mixture. This solution was allowed to cool to room temperature, filtered
and dried on a B u€ chner funnel.
f. The orange solid was dissolved in 1.5L DMSO while heating to 110°C.
g. Once fully dissolved, the solution was removed from heat and 4L H O
2
were added to precipitate 4 as a yellow-orange solid, which was filtered
and dried on a B u€ chner funnel to give the compound in 90% yield.
1
H NMR (400MHz, d -DMSO): δ 12.45 (s, 1H), 8.60 (s, 1H), 8.53
6
(
(
(
d, J ¼3.8Hz, 1H), 8.02 (s, 1H), 7.81 (dd, J ¼8.6, 4.0Hz, 1H), 7.73
dd, J ¼8.6, 4.1Hz, 1H), 7.50 (m, 1H), 7.34 (m, 1H), 7.48
d, J ¼4.0Hz, 1H) ppm (Fig. 11, Scheme 18).
1
Fig. 11 H NMR of 4 recorded at 400MHz in d
6
-DMSO. Peaks at δ 3.35 and 2.50ppm are
residual H O and DMSO, respectively.
2
Scheme 18 Preparation of compound 4.

Rare-earth dependent methanol dehydrogenase cofactor
45
6.2 Preparation of compound 5
a. 30g of 4 were dissolved in 300mL DMSO.
b. 33.8mL Et N were added to the reaction mixture.
3
c. The reaction mixture was cooled in an ice bath (ꢀ10°C) until nearly
frozen.
d. 37.01g of TsCl were added slowly and the reaction mixture was stirred
while in the ice bath for 0.5h.
e. The yellow-white solid was precipitated from solution by addition of 2L
H O, filtered and dried on a B u€ chner funnel.
2
f. The solid was resuspended in 2L isopropanol, filtered, and dried on a
B u€ chner funnel to obtain 5 as a white solid in a 96% yield.
1
H NMR (300MHz, d -DMSO): δ 8.14 (s, 1H), 8.02 (m, 1H), 7.94 (m, 1H),
6
7
.77 (d, J ¼8.3Hz, 2H), 7.76 (s, 1H), 7.67 (s, 1H), 7.60 (m, 2H), 7.50 (s, 1H),
7
.45 (d, J ¼8.0Hz, 2H), 2.42 (s, 3H) ppm (Fig. 12, Scheme 19).
1
Fig. 12 H NMR of 5 recorded at 300MHz in d -DMSO. Peaks at δ 3.35 and 2.50ppm are
6
2
residual H O and DMSO, respectively.
Scheme 19 Preparation of compound 5.

46
Alison L. Knasin and Eric J. Schelter
6
.3 Preparation of compound 6
a. 1L of 6.5M NaOH and 250mL dioxane were cooled to 10°C in an
ice bath.
b. 11.5mL Br were added dropwise using an addition funnel, which
2
produced a yellow solution.
c. 5 was added and the reaction mixture was stirred while in an ice bath
for 0.5h.
d. The reaction mixture was transferred to an oil bath and refluxed (with
the use of a condenser) at 100°C for 16h, which produced a dark brown
solution.
e. The reaction mixture was allowed to cool to room temperature and
acidified to pH 0 with 12M HCl.
f. 1L Et O was added and the aqueous layer was separated and filtered.
2
g. The filtrate was treated with base to reach pH 3 using 3M NaOH and
then to pH 8 using a saturated solution of Na CO .
3
2
h. The resulting precipitate was filtered and dried on a B u€ chner funnel to
obtain 6 as a light brown solid in a 65% yield.
1
H NMR (400MHz, d -DMSO): δ 9.73 (s, 1H), 7.45 (d, J ¼7.9Hz, 1H),
6
7
6
.41 (d, J ¼7.8Hz, 1H), 7.09 (t, J ¼6.3Hz, 1H), 7.03 (t, J ¼6.6Hz, 1H),
.99 (s, 1H), 6.85 (s, 1H), 5.00 (s, 2H) ppm (Fig. 13, Scheme 20).
1
Fig. 13 H NMR of 6 recorded at 400MHz in d
6
-DMSO. Peaks at δ 3.57, 3.35 and
2
.50ppm are 1,4-dioxane, residual H
2
O, and DMSO, respectively.

Rare-earth dependent methanol dehydrogenase cofactor
47
Scheme 20 Preparation of compound 6.
6.4 Preparation of compound 11
a. 0.5g of 10 were dissolved in 100mL CH CN.
3
b. 3.2g of polymer-supported IBX were added as a solid.
c. The reaction mixture was stirred at room temperature for 16h.
d. The yellow-orangesuspensionwas filteredon a mediumporosityfrit andthe
solid was washed with 3ꢃ 15mL CH CN until the filtrate was colorless.
3
e. The filtrate was passed through a pad of Celite (ꢀ5cm in diameter and
ꢀ
10cm in height) using about 50mL of a 1:1 solution of Et O and
2
CH Cl as eluent.
2
2
f. The combined filtrate was concentrated under reduced pressure.
g. Orange solid was precipitated out of solution through addition of 15mL
1,2-dichloroethane.
h. This solid was filtered and dried on a medium porosity fritted filter to
afford 11 as an orange solid in a 30% yield.
1
H NMR data were consistent with the reported spectrum for this com-
pound (Fig. 14) (McSkimming et al., 2018).
1
Fig. 14 H NMR of 11 recorded at 400MHz in CD CN. Peaks at δ 3.83, 2.15, and 1.96ppm
3
2
are 1,2-dichloroethane, residual H O, and acetonitrile, respectively.

48
Alison L. Knasin and Eric J. Schelter
1
H NMR (400MHz, CD CN): δ 10.09 (s, 1H), 8.80 (d, J ¼8.4Hz,
3
1
7
H), 8.28 (d, J ¼7.9Hz, 1H), 8.23 (d, J ¼8.5Hz, 1H), 8.22 (d, J ¼7.8),
.89 (t, J ¼7.7Hz, 1H), 7.69 (t, J ¼7.5Hz, 1H) ppm (Fig. 14, Scheme 21).
Scheme 21 Preparation of compound 11.
6
a. 31.8g of 12, 6.6g of BnNH , 57.3g of KI, 44.3g of K CO , and 650mL
.5 Preparation of compound 13
2
2
3
CH CN were added to a flask and refluxed at 80°C for 16h.
3
b. The reaction mixture was cooled to room temperature and 400mL H O
2
were added in one portion to produce a brown, sticky precipitate.
c. The mixture was extracted twice with a solution of 3:1 hexanes:EtOAc,
washed with brine, and dried over Na SO .
2
4
d. The mixture was filtered on a medium porosity frit and the filtrate was
concentrated under reduced pressure to afford a sticky solid, which was
recrystallized in ethanol to give 13 as a white solid in a 65% yield.
1
H NMR (400MHz, CDCl ): δ 7.41 (d, J ¼8.6Hz, 2H), 7.29 (m, 3H),
3
3
.95 (s, 2H), 3.46 (s, 4H), 2.48 (s, 4H), 1.74–1.02 (m, 40H) ppm
Fig. 15, Scheme 22).
(
1
Fig. 15 H NMR of 13 recorded at 400MHz in CDCl
3
. Peaks at δ 7.26 and 1.56ppm are
2
chloroform and residual H O, respectively.