Synthesis of azide congeners of preQ1 as potential substrates for tRNA guanine transglycosylase

Article abstract of DOI:10.1002/jhet.4220

PreQ₁ (2) is a precursor of queuine (1) that in eubacteria is incorporated into transfer RNA (tRNA) by tRNA guanine transglycosylase (TGT) before being further elaborated into queuine. The queuine modification is unusual and occurs across all eukaryotes and eubacteria with few exceptions, but its function remains unclear. As the modified nucleotide occurs through incorporation of a specially synthesized nucleotide instead of via modification of a genetically encoded base, a study of the sites of modification by prepared probes is possible. We report the synthesis of two novel azide congeners (3,4) of preQ₁ for this purpose. The evaluation of their interaction with TGT shows that both probes act as weak competitive inhibitors of guanine exchange of guanine(34) tRNA^Tyr with a K_i of ~70 μM. However, we could not show that these are substrates for TGT-catalyzed incorporation into tRNA. We believe the reason for this is a marked loss of binding due to the azide functionality of 3 and 4 abrogating the possibility of two hydrogen bonds to the carbonyl group of Leu231 and Met260 of TGT, previously observed for the terminal methylene amine of preQ₁ by x-ray crystallography.

Full text of DOI:10.1002/jhet.4220

Received: 12 October 2020
DOI: 10.1002/jhet.4220
Revised: 17 December 2020
Accepted: 24 December 2020
N O T E S
Synthesis of azide congeners of preQ1 as potential
substrates for tRNA guanine transglycosylase
Allen F. Brooks | George A. Garcia | Hollis D. Showalter
Department of Medicinal Chemistry,
Abstract
University of Michigan, Ann Arbor,
Michigan
PreQ1 (2) is a precursor of queuine (1) that in eubacteria is incorporated into
transfer RNA (tRNA) by tRNA guanine transglycosylase (TGT) before being
further elaborated into queuine. The queuine modification is unusual and
occurs across all eukaryotes and eubacteria with few exceptions, but its func-
tion remains unclear. As the modified nucleotide occurs through incorporation
of a specially synthesized nucleotide instead of via modification of a genetically
encoded base, a study of the sites of modification by prepared probes is possi-
ble. We report the synthesis of two novel azide congeners (3,4) of preQ1 for this
purpose. The evaluation of their interaction with TGT shows that both probes
act as weak competitive inhibitors of guanine exchange of guanine(34) tRNATyr
with a Ki of ꢀ70 μM. However, we could not show that these are substrates for
TGT-catalyzed incorporation into tRNA. We believe the reason for this is a
marked loss of binding due to the azide functionality of 3 and 4 abrogating the
possibility of two hydrogen bonds to the carbonyl group of Leu231 and Met260
of TGT, previously observed for the terminal methylene amine of preQ1 by
x-ray crystallography.
Correspondence
Hollis D. Showalter, Department of
Medicinal Chemistry, University of
Michigan, Ann Arbor, MI 48109-1065,
USA.
Email: showalh@umich.edu
Funding information
University of Michigan College of
Pharmacy Upjohn Research Fund;
University of Michigan College of
Pharmacy Vahlteich Research Fund;
University of Michigan Rackham Regents
Graduate Fellowship; University of
Michigan Sheila Cresswell Graduate
Fellowship
1 | INTRODUCTION
fully elucidated with all the involved enzymes identified.
The queuine modification is known to be present in the
Queuine (1; Figure 1) makes up one of approximately
100 base-modified nucleotides found in RNA.[1] Since its
discovery many details of its incorporation into RNA
have been described and its potential role examined.
Unlike the majority of modified nucleotides that result
from modifications of the genetically encoded nucleo-
tides, queuine is incorporated into transfer RNA (tRNA)
by transglycosylation.[2] Eukaryae incorporates queuine
obtained through their diet, whereas eubacteria bio-
synthesize and incorporate the queuine precursor preQ1
(2; Figure 1) via multistep modifications from guanine
(34) tRNA to yield queuine(34) tRNA. As a result, the
enzyme responsible for the incorporation of the modifica-
tion into tRNA, tRNA guanine transglycosylase (TGT),
differs between eukaryal and eubacterial organisms.[3]
The pathway for the biosynthesis of queuine has been
wobble position of the anticodon loop of tRNATyr
,
tRNAHis, tRNAAsp, and tRNAAsn [4]
Investigations have
.
determined that its presence has an effect on codon rec-
ognition and promoting translational fidelity.
Base modification by transglycosylation is unusual
and represents an interesting entry point for study with
small molecule probes. Previous studies from our labora-
tory with tritium-labeled queuine and preQ1 have shown
differences between the eukaryotic and eubacterial ver-
sions of TGT with the prevalence of queuine extending
beyond the previously known four tRNAs.[3,5] To further
build on this finding, we turned our attention to finding
a means to identify the specific tRNA species that are
modified. Toward that end, we have designed two novel
azide probes (3 and 4), based on the heterocyclic core of
queuine and preQ1, for studies showing them to be
1192
© 2020 Wiley Periodicals LLC.
wileyonlinelibrary.com/journal/jhet
J Heterocyclic Chem. 2021;58:1192–1198.

BROOKS ET AL.
1193
substrates for TGT. These included bioorthogonal Stau-
dinger and/or click ligation chemistry to isolate modified
strands of tRNA. Structurally, 3 is more closely related to
preQ1, which is important for recognition by TGT for its
incorporation into tRNA. Since the benzylic disposition
of the azide function raised questions about its potential
stability, we decided to mitigate this concern by also
making a more stable one-carbon homolog 4. This paper
describes the synthesis of 3 and 4 followed by studies to
determine if these are substrates for TGT.
2 | RESULTS AND DISCUSSION
2.1 | Chemistry
The synthesis of benzylic azide probe 3 is shown in
Scheme 1 with aldehyde 9 as the key intermediate. Prior
synthesis of precursor 8 (preQ0) proceeded via in situ gener-
ation of a 2-chloro-3-oxopropanenitrile from the base-
catalyzed condensation of methyl formate (5) and chlo-
roacetonitrile followed by acidic workup. The resultant oil,
FIGURE 1 Queuine (1), precursor
PreQ1 (2) and target azide probes (3, 4)
SCHEME 1 Synthetic
pathway to azide probe 3

1194
BROOKS ET AL.
without isolation, was then further reacted with 2,6-dia-
minopyrimidin-4(3H)-one (7) to give pyrrolopyrimidinone
nitrile 8.[6] Since the yield of 8 was variable from run to run
(never exceeding 70%), we decided to try pre-isolation of
2-chloro-3-oxopropanenitrile as its sodium salt (6) prior to
reaction with 7. This was successfully accomplished on a
0.2 mole scale in 93% yield as a 7:1 mixture of E:Z isomers
by 1H NMR integration. The salt is a stable white solid and
with storage in a desiccator maintained its physical and
chemical properties for over a year, giving comparable
yields of 8 in subsequent runs. Condensation of 6 with
7 then provided 8 in 96% yield. Following our work, a patent
disclosed a similar preparation of 6 in 88% yield on a 0.265
scale.[7] We believe that 6 will serve as a valuable 3-carbon
synthon for the synthesis of related 4-oxo-4,7-dihydro-3H-
pyrrolo[2,3-d]pyrimidine-5-carbonitrile heterocycles. With
a reproducible synthesis of nitrile 8, it was then parlayed to
aldehyde 9 in 42% yield in two standard steps (trityl protec-
tion then DIBAL-H reduction) by our previously reported
procedure, each involving in situ silylation to provide solu-
tion solubility.[8]
There is only one report in the literature of the syn-
thesis of a benzylic azide of the 4-oxo-4,7-dihydro-3H-
pyrrolo[2,3-d]pyrimidine ring system found in 3.[9] This
proceeded via azide displacement of a benzylic bromide
on a heterocyclic framework requiring multiple protec-
tion/deprotection steps of each nitrogen. There are no
reports of the synthesis of an azide related to 4.
Our strategy to 3 was to employ a much more readily
available benzylic alcohol. Thus, reduction of aldehyde 9 to
alcohol 10 proceeded uneventfully with sodium borohy-
dride and set the stage for generation of target benzylic
azide 3. Standard conditions of mesylation/azide displace-
ment unsurprisingly failed. A Lewis acid-catalyzed method
(boron trifluoride etherate/trimethylsilyl azide)[10] provided
a small amount of 11 with the remainder of products as
decomposition materials. Several other Lewis and mineral
acids of this reaction were evaluated, but all fared poorly. It
is likely the above conditions generate a benzylic-type cat-
ion, which readily leads to polymerization.
We then turned to an alternate method for substitut-
ing a benzylic-type alcohol with an azide under basic
conditions.[11] Hence, treatment of alcohol 10 with diph-
enylphosphoryl azide (DPPA) and 1,8-diazabicyclo[5.4.0]
undec-7-ene (DBU) resulted in a high-yield transforma-
tion to azide 11. The reaction likely proceeds through a
diphenylphosphoryl intermediate with the released azide
displacing the now activated alcohol. Azide 11 was then
converted at low temperature to probe 3 in 85% yield by
removal of the trityl protecting group with trifluoroacetic
acid in the presence of triethylsilane as a scavenger. This
scheme yielded probe 3 in an overall 32% yield from the
starting salt 6.
The synthesis of the second azide probe 4, which fea-
tures an additional methylene spacer between the hetero-
cyclic ring and the azide, is shown in Scheme 2. It begins
with the benzyl ether 12, which was prepared in a
straightforward four-step process in 24% overall yield as
reported by Hornillo-Aruajo et al.[12] Their removal of
the benzyl protecting group to give alcohol 13 utilized
boron trichloride. In our hands, we found this to be cum-
bersome due to the resulting boronates that made
extracting the product difficult upon workup. We deter-
mined that transfer hydrogenation with ammonium for-
mate and 10% palladium on carbon was cleaner and
much easier for protecting group removal, giving 13 in
59% yield.
Standard methods to generate probe 4 from alcohol
13 failed. These included mesylation/azide displacement
and Mitsunobu conditions. A successful method was to
SCHEME 2 Synthetic pathway to
azide probe 4

BROOKS ET AL.
1195
employ similar conditions used to synthesize azide 3. As
13 is a primary alkyl alcohol rather than a benzylic type,
modifications to the reaction conditions were necessary.
Thus, treatment of 13 with DPPA and DBU in DMF did
not give an expected azide, but cleanly yielded the phos-
phoryl ester (14) in 98% yield, which could be isolated
and purified. To displace the ester with azide, we utilized
a literature method that describes the transformation of
the 6-position alcohol of a sugar to an azide.[13] Thus, the
phosphoryl ester (14) was treated with sodium azide,
15-crown-5 ether, and tetrabutylammonium iodide to
provide the desired azide probe 4 in 63% yield and an
overall 36% yield from 12.
TGT and tRNA, but without incorporation into tRNA
(data not shown). Thus, cumulative data confirm that
our azide probes are not substrates for TGT-catalyzed
incorporation into tRNA.
3 | CONCLUSIONS
We have synthesized two novel azide congeners (3,4) of
preQ1 and evaluated their interaction with TGT. Both
compounds were made in good overall yield with the
sequence to azide 3 featuring a scalable synthesis of
2-chloro-2-cyanoethen-1-olate, sodium salt (6), a stable,
storable solid that promises to serve as a valuable
3-carbon synthon for the synthesis of related 4-oxo-4,-
7-dihydro-3H-pyrrolo[2,3-d]pyrimidine-5-carbonitrile het-
erocycles. Both probes act as weak competitive inhibitors
of guanine exchange of guanine(34) tRNA with a Ki of
ꢀ70 μM with additional experiments showing that these
are not substrates for TGT-catalyzed incorporation into
tRNA. We believe the reason for this is a marked loss of
binding due to the azide functionality of 3 and 4 abrogat-
ing the possibility of two hydrogen bonds to the car-
bonyl group of Leu231 and Met260 of TGT, previously
observed for the terminal methylene amine of preQ1 by
x-ray crystallography.[16] This hypothesis is supported
by recent publications from the Kelly lab on the synthe-
sis of a chain-extended congener of preQ1 that is a TGT-
specific substrate that rapidly reverses disease in murine
experimental autoimmune encephalomyelitis,[17] and
the Devaraj lab[18] on the synthesis of TGT-specific pro-
bes to label RNA. Each of these structures incorporates
the basic benzylic amine of preQ1 and is readily incorpo-
rated into RNA.
All compounds were rigorously purified by flash chro-
matography or crystallization, and their structural assign-
ments were supported by diagnostic peaks in NMR
spectra and by mass spectrometry.
2.2 | Biology
All biological studies were conducted with E. coli TGT
and E. coli tRNATyr
.
Our initial goal was to determine if azide probe 3 is
an inactivator of TGT. This was based on concerns that
its benzylic-like connection could cause it to unravel to
an electrophilic trapping species. Following protocols by
Hoops et al.,[14] we determined that 3 is not an
inactivator of TGT, and thus we proceeded with incorpo-
ration experiments with this and homolog 4.
Studies with 14C-guanine (Figures S1 and S2, Supple-
mentary Information) followed by kinetic analysis indi-
cated that both azide probes act as weak competitive
inhibitors of guanine exchange by TGT with a Ki of
ꢀ70 μM compared to 0.35 μM for guanine and 0.05 μM
for its natural substrate preQ1.[3] In one approach, to test
for incorporation into tRNA as competitive substrates,
While disappointed that our azide probes are not sub-
strates for tRNA guanine transglycosylase, we feel that
these can serve as valuable intermediates for click reactions
with alkynes to provide a wide range of analogues incor-
porating substituted 1,2,3-triazines as replacements
for heterocycloalkyl or benzenoid moieties that are tethered
to the 5-alkyl-2-amino-3,7-dihydro-4H-pyrrolo[2,3-d]
pyrimidin-4-one scaffold. Potential applications to broad-
spectrum antiproliferative agents[19] or thymidylate synthase
inhibitors[20] exemplify these possibilities.
the azide probes at 200 μM (approximately 3× Ki,apparent
)
were incubated overnight with tRNA, and TGT was
added every 6 h. Two tRNAs were used as substrate: (i) a
mini-hairpin of tRNA, utilized in earlier work, to demon-
strate preQ1 incorporation by mass spectrometry,[15] and
(ii) the other is a full-length tRNA under the same condi-
tions. Mass spectrometry analysis of an untreated sample
of each tRNA was performed concurrently. The reactions
were quenched by ethanol precipitation and the samples
were desalted and submitted for electrospray ionization
mass spectrometry analysis. In neither experiment was
incorporation of either azide probe observed with the
intact mass of starting tRNA observed instead. Addition-
ally, preliminary experiments with the azide probes
showed them to react with Staudinger and copper-free
click ligation reagents bound to biotin in the presence of
4 | EXPERIMENTAL
4.1 | Chemistry
All reagents were obtained from commercial suppliers and
used as received. Toluene and pyridine were distilled over
calcium hydride. Glassware was oven-dried before use for

1196
BROOKS ET AL.
reactions run under anhydrous conditions. Melting points
were determined in open capillary tubes on a Laboratory
Devices Mel-Temp apparatus and are uncorrected. NMR
spectra were recorded on a Bruker instrument at 500 MHz
for 1H and 125 MHz for 13C spectra. Chemical shift values
are recorded in δ units (ppm). Mass spectra (MS) were
recorded on a Micromass TofSpec-2E Matrix-Assisted,
Laser-Desorption, Time-of-Flight Mass Spectrometer in the
positive ESI mode unless otherwise noted. High-resolution
mass spectrometry (HRMS) analysis was performed on an
Agilent Q-TOF system. TLC was performed on EM Science
aluminum baked silica gel (SiO2) plates. Visualization was
performed with 254 nm UV light and 2,4-dinitroph-
enylhydrazine (2,4-DNP) stain to identify aldehydes.
Extraction solutions were dried over magnesium sulfate or
sodium sulfate prior to concentration.
25 ꢁC. The rapidly stirring suspension was heated at 50 ꢁC
for 16 h and then at reflux for 1 h. The resulting suspension
was cooled to room temperature and collected on a fritted
funnel. The light brown solid was washed with portions of
water and then acetone. Vacuum drying provided 0.133 g
1
(96%) of 3: mp 360 ꢁC (dec); H NMR (DMSO-d6) δ 11.98
(br s, 1H) 10.74 (br s, 1H), 7.59 (s, 1H), 6.43 (s, 2H); 13C
NMR (DMSO-d6) δ 158.0, 154.3, 152.1, 128.2, 116.4,
99.2, 86.0.
4.1.3 | 2-Amino-4,7-dihydro-4-oxo-3H-
pyrrolo[2,3-d]pyrimidine-5-carbonitrile (8;
preQ0 from free form of 6)
Reaction of 2,6-diaminopyrimidin-4-one (7; 13.32 g, 105.6
mmol) was carried out as previously described[6] to give a
68% yield of 8 with spectral data identical to the prepara-
tion using sodium salt 6.
4.1.1 | 2-Chloro-2-cyanoethen-1-olate,
sodium salt (6)
A dry 250 ml three-neck RB flask was equipped with a
mechanical stirrer and an addition funnel. The flask was
charged with a suspension of sodium methoxide (10.8 g,
200 mmol) in toluene (100 ml), which was cooled to
−5 ꢁC in a water ice-salt bath. Methyl formate (5;
37.0 ml, 600 mmol) was added slowly over 5 min and the
solution temperature was maintained for 15–20 min.
Chloroacetonitrile (12.66 ml, 200 mmol) was added
dropwise over ꢀ1 h from the addition funnel with each
droplet resulting in a local discharge of yellow color. The
mixture was kept at −5 ꢁC for 3 h and then allowed to
come to room temperature, becoming a thick buttery yel-
low suspension after 16 h. The suspension was filtered on
a glass Buchner fritted funnel and the collected white
solid was washed with ethyl acetate and dried under vac-
4.1.4 | 5-(Hydroxymethyl)-
2-(tritylamino)-3H-pyrrolo[2,3-d]pyrimidin-
4(7H)-one (10)
An ice-cold suspension of 4-oxo-2-(tritylamino)-4,7-di-
hydro-3H-pyrrolo[2,3-d]pyrimidine-5-carbaldehyde[8] (9;
0.10 g; 0.24 mmol) in methanol (1 ml) was treated with
sodium borohydride (0.027 g; 0.71 mmol) and the mix-
ture was stirred for 3 h. The mixture was quenched with
brine and extracted with dichloromethane (3x). The
combined extracts were dried and concentrated to leave
1
10 (0.1 g, 99%) as a white powder: Rf 0.33 (EtOAc); H
NMR (DMSO-d6) δ 10.69 (br s, 1H), 10.51 (br s, 1H), 7.41
(br s, 1H), 7.31–7.20 (m, 15H), 6.34 (s, 1H), 4.95 (br s,
1H), 4.41 (d, J = 5.6, 2H); MS m/z 445.1 (M + Na)+.
1
uum overnight to give 23.2 g (93%) of product: H NMR
(DMSO-d6) δ 8.50, 8.16 (integrates for 7:1 mixture of E:Z
isomers); 13C NMR (DMSO-d6) δ 168.23, 126.55, 67.78;
The material was stored in a desiccator at room tempera-
ture. The material maintained its physical and chemical
properties for over a year, giving comparable yields in
subsequent reactions.
4.1.5 | 5-(Azidomethyl)-2-(tritylamino)-
3H-pyrrolo[2,3-d]pyrimidin-4(7H)-one (11)
An ice-cold solution of alcohol 10 (0.05 g, 0.12 mmol),
diphenylphosphoryl azide (DPPA; 0.031 ml, 0.142 mmol),
and tetrahydrofuran (1 ml) was treated dropwise with
1,8-diazabicycloundec-7-ene (DBU) (0.021 ml, 0.142 mmol).
The mixture was stirred for 4 h followed by additional
equivalent charges of DPPA and DBU. After 18 h of total
reaction time, the solution was quenched with water and
then extracted with diethyl ether (3x). The combined
extracts were washed successively with water (2x) and
brine, dried, and concentrated to leave a solid that was
purified by flash silica gel chromatography (gradient elu-
tion from 8:2 to 1:1 hexanes/ethyl acetate). Product
4.1.2 | 2-Amino-4,7-dihydro-4-oxo-3H-
pyrrolo[2,3-d]pyrimidine-5-carbonitrile (8;
preQ0 from sodium salt 6)
An RB flask was charged with a stir bar, anhydrous sodium
acetate (0.065 g, 0.79 mmol), 2,6-diaminopyrimidin-4-one
(7; 0.10 g, 0.79 mmol), sodium salt 6 (0.25 g, 2 mmol), gla-
cial acetic acid (0.124 ml), and distilled water (3.2 ml) at

BROOKS ET AL.
1197
fractions were combined and concentrated to leave 11
(0.05 g, 94%) as a white powder: Rf 0.52 (1:1 hexanes/
ethyl acetate); H NMR (DMSO-d6) δ 10.94 (br s, 1H),
10.36 (br s, 1H), 7.42 (br s, 1H), 7.31–7.20 (m, 15H), 6.58
(s, 1H), 4.31 (br s, 2H); MS m/z 459.1 (M + Na)+, 895.3
(2 M + H)+.
(0.350 g, 98%) as a white solid: Rf 0.76 (4:1 ethyl acetate/
methanol); H NMR (DMSO-d6) 10.80 (br s, 1H), 10.25
(br s, 1H), 7.39 (t, J = 7.7, 3H), 7.24 (t, J = 7.7, 2H), 7.17
(d, J = 7.7, 3H), 6.46 (s, 1H), 6.04 (br s, 2H), 4.52 (q,
J = 6.8, 2H), 2.96 (t, J = 6.8, 2H); MS m/z 449.0 (M
+ Na)+, 875.0 (2 M + Na)+.
1
1
4.1.6 | 2-Amino-5-(azidomethyl)-3H-
pyrrolo[2,3-d]pyrimidin-4(7H)-one (3)
4.1.9 | 2-Amino-5-(2-azidoethyl)-3H-
pyrrolo[2,3-d]pyrimidin-4(7H)-one (4)
A stirring solution of azide 11 (0.10 g, 0.022 mmol) and
triethylsilane (3.57 μl, 0.022 mmol) in dichloromethane
(0.5 ml) was cooled to −78 ꢁC, and treated with trifluo-
roacetic acid (3.32 μl, 0.045 mmol). The mixture was
maintained at −78 ꢁC for 3 h, and then allowed to slowly
warm to room temperature. A formed precipitate was col-
lected and dried to leave 3 (0.039 g, 85%) as an off-white
A mixture of 14 (0.186 g, 0.436 mmol), sodium azide
(0.142 g, 2.181 mmol), tetra-n-butylammonium iodide
(0.016 g, 0.044 mmol), 15-crown-5 ether (10 μl, 0.044 mmol),
and N,N-dimethylacetamide (3.5 ml) was stirred at 80 ꢁC for
18 h. The mixture was concentrated in vacuo to give a brown
solid that was dissolved in methanol and adsorbed onto a
small amount of flash silica gel for dry loading onto a
chromatography column. The product was then purified
by flash silica gel chromatography using gradient elu-
tion (ethyl acetate to 4:1 ethyl acetate/methanol). Prod-
uct fractions were pooled and concentrated to leave
4 (0.060 g, 63%) as a yellow-white powder: Rf 0.63 (4:1,
1
powder: H NMR (DMSO-d6) δ 11.11 (br s, 1H), 10.44
(br s, 1H), 6.76 (s, 1H), 6.20 (br s, 2H), 4.38 (s, 2H); 13C
NMR (DMSO-d6) δ 158.69, 157.95, 152.39, 116.95, 111.07,
98.32, 45.65; MS m/z 228.0 (M + Na)+, 433.1 (2 M
+ Na)+; HRMS m/z (M + Na)+: calcd for C7H9N5ONa+,
228.0610; found, 228.0611.
1
ethyl acetate/methanol); H NMR (DMSO-d6) δ 10.77
(br s, 1H), 10.22 (br s, 1H), 6.48 (s, 1H), 6.03 (s, 2H), 3.56
(t, J = 6.6, 2H), 2.82 (t, J = 6.6, 2H); 13C NMR (DMSO-
d6) δ 159.69, 152.74, 151.99, 115.20, 114.52, 100.00,
51.53, 26.42; MS m/z 220.1 (M + H)+; HRMS m/z (M
+ H)+: calcd for C8H10N7O+, 220.0947; found, 220.0949.
4.1.7 | 2-Amino-5-(2-hydroxyethyl)-3H-
pyrrolo[2,3-d]pyrimidin-4(7H)-one (13)
A mixture of benzyl ether 12[12] (0.5 g, 1.76 mmol),
ammonium formate (1.11 g, 17.6 mmol), and methanol
(7 ml) was heated to reflux followed by the addition of
10% palladium on carbon (0.187 g, 0.176 mmol). After
8 h at reflux, the mixture was filtered through a pad of
Celite followed by a methanol washing. The concentra-
tion of the combined filtrates left a solid that was tritu-
rated in hot ethyl acetate to give 13 (0.2 g, 59%) as a
white powder: Rf 0.40 (4:1 ethyl acetate/methanol). The
1H NMR was the same as previously described.[12]
4.2 | Biology
All biological studies were conducted with E. coli TGT and
E. coli tRNATyr. Biotin-bound reagents for Staudinger and
copper-free click ligation chemistry were synthesized or
commercially available. These were CAS 608514-42-7[21]
and 1255942-07-4,[22] respectively.
4.2.1 | Inactivation study of azide 3 with
4.1.8 | 2-(2-Amino-4-oxo-4,7-dihydro-3H-
pyrrolo[2,3-d]pyrimidin-5-yl)ethyl diphenyl
phosphate (14)
TGT 14C-guanine exchange assay
Experimental protocols by Hoops et al. were followed.[14]
Azide 3 at 150 μM was incubated with 5 μM TGT in the
presence of tRNA at 20 μM in bicine buffer for 15, 30,
60, and 120 min. A control preincubation of TGT without
any azide 3 was also performed. Following incubation,
TGT was diluted into a 14C-guanine exchange assay with
the following conditions in a bicine buffer: 20 μM tRNA,
0.1 μM TGT (from preincubation), 20 μM 14C-guanine.
Experiments were run in duplicate and time points were
taken at 2, 4, 6, 8, and 10 min.
An ice-cold solution of alcohol 13 (0.162 g, 0.834 mmol),
diphenyl phosphoryl azide (0.539 ml, 2.5 mmol), and
N,N-dimethylformamide (6.7 ml) was treated with DBU
(0.374 ml, 2.5 mmol). The mixture was stirred for 16 h at
room temperature, diluted with water, and extracted with
ethyl acetate (3×). The combined organic extracts were
washed with brine, dried, and concentrated to a solid that
was triturated in ꢀ9:1 ethyl acetate/methanol to give 14

1198
BROOKS ET AL.
4.2.2 | TGT [14C] guanine exchange assay
of azide probes 3 and 4
[3] Y.-C. Chen, A. F. Brooks, D. M. Goodenough-Lashua, J. D.
Kittendorf, H. D. Showalter, G. A. Garcia, Nucleic Acids Res.
2011, 39, 2834.
[4] a) F. Harada, S. Nishimura, Biochemistry 1972, 11, 301. b)
R. C. Morris, M. S. Elliott, Mol. Genet. Metab. 2001, 74, 147.
[5] A. F. Brooks, C. S. Velez-Martinez, H. D. H. Showalter, G. A.
Garcia, Biochem. Biophys. Res. Commun. 2012, 425, 83.
[6] M. T. Migawa, J. M. Hinkley, G. C. Hoops, L. B. Townsend,
Synth. Commun. 1996, 26, 3317.
[7] F. Higuchi, U. S. Patent Application 2019/0048023 A1, 2019.
[8] A. F. Brooks, G. A. Garcia, H. D. H. Showalter, Tetrahedron
Lett. 2010, 51, 4163.
Assays were performed with the azide probes under the
following conditions: 0.1 μM TGT; 10 μM tRNA; 0.75 μM,
1.50 μM, or 5.00 μM 14C-guanine at six concentrations of
azide probe. These included 5, 50, 100, 250, and 500 μM
for 3 and 10, 100, 175, 250, 500, and 1000 μM for 4. The
controls (without azide) were also run for each concen-
tration. Experiments were performed in duplicate in
bicine buffer. Data were obtained at time points for 2, 4,
6, 8, and 10 min by quenching and precipitating 70 μl of
reaction mixture in 5% TCA. The resulting precipitate
was collected by filtration. Liquid scintillation counting
provided raw data for analysis. Data are shown in
Figures S1 and S2 (Supplementary Information).
[9] T. Ohgi, T. Kondo, T. Goto, Chem. Lett. 1979, (10), 1283.
[10] V. Terrasson, S. Marque, M. Georgy, J.-M. Campagne, D.
Prim, Adv. Synth. Catal. 2006, 348, 2063.
[11] A. S. Thompson, G. R. Humphrey, A. M. DeMarco, D. J.
Mathre, E. J. J. Grabowski, J. Org. Chem. 1993, 58, 5886.
[12] A. R. Hornillo-Araujo, A. J. M. Burrell, M. K. Aiertza, T.
Shibata, D. M. Hammond, D. Edmont, H. Adams, G. P.
Margison, D. M. Williams, Org. Biomol. Chem. 2006, 4, 1723.
[13] F. Liu, D. J. Austin, Tetrahedron Lett. 2001, 42, 3153.
[14] G. C. Hoops, L. B. Townsend, G. A. Garcia, Biochemistry 1995,
34, 15539.
4.2.3 | Azide incorporation into tRNA
TGT mediated incorporation of azide probes 3 and 4 into
tRNA was attempted with the following conditions:
80 μM tRNA, 200 μM azide probe, and 1.4 μM TGT in
bicine buffer. An additional aliquot of TGT was added at
6 h and 12 h, and the reaction was quenched by ethanol
precipitation at 18 h. The resulting tRNA was pelleted by
centrifugation at 13 k rpm for 10 min in a tabletop
Eppendorf centrifuge. The pellet was washed twice with
70% ethanol and air-dried 10 min before being suspended
in solvent for analysis by mass spectrometry.
[15] G. A. Garcia, S. M. Chervin, J. D. Kittendorf, Biochemistry
2009, 48, 11243.
[16] W. Xie, X. Liu, R. H. Huang, Nat. Struct. Biol. 2003, 10, 781.
[17] S. Varghese, M. Cotter, F. Chevot, C. Fergus, C. Cunningham,
K. H. Mills, S. J. Connon, J. M. Southern, V. P. Kelly, Nucleic
Acids Res. 2017, 45, 2029.
[18] a) S. C. Alexander, K. N. Busby, C. M. Cole, C. Y. Zhou, N. K.
Devaraj, J. Am. Chem. Soc. 2015, 137, 12756. b) F. Ehret, C. Y.
Zhou, S. C. Alexander, D. Zhang, N. K. Devaraj, Mol. Pharma-
ceutics 2018, 15, 737.
[19] J.-H. Lee, A. K. El-Damasy, S. H. Seo, C. G. Gadhe, A. N. Pae, N.
Jeong, S.-S. Hong, G. Keum, Bioorg. Med. Chem. 2018, 26, 5596.
[20] D. J. Czyzyk, M. Valhondo, L. Deiana, J. Tirado-Rives, W. L.
Jorgensen, K. S. Anderson, Eur. J. Med. Chem. 2019, 183,
111673.
ACKNOWLEDGMENTS
This work was supported by the University of Michigan
College of Pharmacy Vahlteich and Upjohn Research
Funds (GAG, HDS), and Rackham Regents and Sheila
Cresswell Graduate Fellowships (AFB).
[21] S. R. Rajski, L. R. Comstock, R. L. Weller, U. S. Patent Applica-
tion 2007/0161007 A1, 2007.
[22] A. Kuzmin, A. Poloukhtine, M. A. Wolfert, V. V. Popik,
Bioconjugate Chem. 2010, 21, 2076.
DATA AVAILABILITY STATEMENT
The data that supports the findings of this study are avail-
able in the supplementary material of this article.
SUPPORTING INFORMATION
Additional supporting information may be found online
in the Supporting Information section at the end of this
article.
ORCID
Hollis D. Showalter https://orcid.org/0000-0003-2863-
0170
How to cite this article: Brooks AF, Garcia GA,
Showalter HD. Synthesis of azide congeners of
preQ1 as potential substrates for tRNA guanine
transglycosylase. J Heterocyclic Chem. 2021;58:
1192–1198. https://doi.org/10.1002/jhet.4220
REFERENCES
[1] W. A. Cantara, P. F. Crain, J. Rozenski, J. A. McCloskey, K. A.
Harris, X. Zhang, F. A. P. Vendeix, D. Fabris, P. F. Agris,
Nucleic Acids Res. 2011, 39, D195.
[2] G. A. Garcia, J. D. Kittendorf, Bioorg. Chem. 2005, 33, 229.