A diversity of alkylation/acylation products of uracil and its derivatives: Synthesis and a structural study

Article abstract of DOI:10.1039/c8ob02552e

tert-Butyl dicarbonate (Boc₂O) and ethyl iodide (EtI) reactions with uracil (U), thymine (T) and 6-methyluracil (6-MU) were performed following routine procedures in pyridine/DMF solvents and with DMAP as the catalyst. Among 20 synthesized compounds, a derivative of 6-methyluracil substituted by the Boc-pyridine moiety at the C5 position appeared unexpectedly. The NMR spectra confirmed the molecular structure of all uracil derivatives. Parallel quantum mechanical DFT calculations supported the experimental findings.

Full text of DOI:10.1039/c8ob02552e

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
A diversity of alkylation/acylation products of
uracil and its derivatives: synthesis and a structural
study†
Cite this: Org. Biomol. Chem., 2019,
17, 354
Olga Michalak, *a Piotr Cmoch,b Piotr Krzeczyński,a Marcin Cybulskia and
Andrzej Leśa,c
tert-Butyl dicarbonate (Boc2O) and ethyl iodide (EtI) reactions with uracil (U), thymine (T) and 6-methyl-
uracil (6-MU) were performed following routine procedures in pyridine/DMF solvents and with DMAP as
the catalyst. Among 20 synthesized compounds, a derivative of 6-methyluracil substituted by the Boc-
pyridine moiety at the C5 position appeared unexpectedly. The NMR spectra confirmed the molecular
structure of all uracil derivatives. Parallel quantum mechanical DFT calculations supported the experi-
mental findings.
Received 15th October 2018,
Accepted 10th December 2018
DOI: 10.1039/c8ob02552e
rsc.li/obc
N1,N3-di-Boc derivative. This reaction was carried out in the
presence of K2CO3 in dioxane with 31% yield. Bessières et al.13
Introduction
The uracil moiety is one of the most important structures developed the synthesis of the Boc derivative at the N3-position
encountered in life sciences. As nucleobases, uracil (U) and of thymine with a high yield, comprising two phases: complete
thymine (T) occur in the RNA and DNA of nucleic acids and pyrimidine protection by the microwave radiation, followed by
other natural products, i.e. willardiine, sparsomycin, and poly- the selective N1 deprotection by SiO2 in dimethoxymethane/
oxins (Fig. 1).1 Various derivatives of the uracil family serve as ethanol (9 : 1).
the building blocks in the synthesis of nucleoside analogues
with potential antiviral, anti-inflammatory or anticancer et al.11 according to the procedure developed for thymine,
N1-Boc protected uracil was synthesized by Jaime-Figueroa
properties.2
using 1.0 eq. Boc2O. Likewise, Lipani et al.14 obtained N1-Boc
For an effective synthetic strategy, protection of the N1- and/ uracil by the regioselective derivatization, using 1.1 eq. Boc2O
or N3-amine position of the uracil core by various groups, such and pyridine as the solvent.
as SEM,3 acyl,4 benzyl,5 benzoyl,6 benzyloxymethyl,7 tetra-
There are no available literature data for the Boc protection
hydropyranyl,8 can be achieved. While working on our recent of 6-methyluracil. However, Lee15 obtained 6-methyl-2,4-
research project, we selected the Boc-protecting group as the pyrimidyldiesters as a result of the 2,4-dihydroxy-6-methyl-
most efficient for the thymine and 6-methyluracil intermedi- pyrimidine reaction with the acyl chlorides in dichloro-
ates, as it was stable in comparison with most nucleophiles methane in the presence of TEA at room temperature. Active
and bases9 and could be removed under neutral conditions in esters, i.e. O2- and O4-acetylation products, were obtained with
a clean and selective manner.10
good yields and they were thermally stable.
Surprisingly, only three teams have so far described the syn-
Due to the incomplete and inconsistent data concerning
thesis of the Boc-protected thymine under basic conditions. the Boc protection of uracil, thymine and 6-methyluracil, we
Jaime-Figueroa et al.11 obtained an N1-Boc derivative using
Boc2O at a stoichiometric ratio in relation to thymine, while
Gothelf et al.12 obtained N3-Boc protected thymine by the
selective removal of N1-Boc from the previously synthesized
aPharmaceutical Research Institute, 8 Rydygiera Str., 01-793 Warsaw, Poland.
E-mail: o.michalak@ifarm.eu; Tel: +48 22 456 3925
bInstitute of Organic Chemistry Polish Academy of Sciences, 44/52 Kasprzaka street,
01-224 Warsaw, Poland
cUniversity of Warsaw, Faculty of Chemistry, 1 Pasteura Str., 02-093 Warsaw, Poland
†Electronic supplementary information (ESI) available: DFT data, comparison of
DFT and experimental results. See DOI: 10.1039/c8ob02552e
Fig. 1 Bioactive molecules based on the uracil moiety.
354 | Org. Biomol. Chem., 2019, 17, 354–362
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
have decided to validate the hypothesis of different chemical thymine with Boc2O in the presence of DMAP in acetonitrile
affinities of these structurally similar nucleobases to the acyla- leads to only one product: N1-Boc-thymine (2a) with a good
tion reagent. As a continuation of this effort, we have also yield,11 see Scheme 1.
tested the course of the alkylation reaction with EtI by compre-
We have decided to apply this procedure to obtain the
hensively analyzing the separated products. All the obtained N1-Boc protected analogue of other pyrimidine bases, i.e.
compounds were isolated and fully characterized by various 6-methyluracil, using 1.5 mol eq. Boc2O. The result was unex-
NMR and HRMS techniques. Numerous quantum mechanical pected. Our experiments afforded not one, but four different
calculations within the density functional theory (DFT/B3LYP) products, see Table 1.
were performed to support the experimental results and
discuss product structures.
The compounds were isolated by column chromatography
and their structures were characterized by the NMR data to
determine the acylation position. It turned out that one of the
unexpectedly identified products was 1,3-di-O-tert-butyloxy-
carbonyl-(1H,3H)-6-methylpyrimidine (1d). A careful analysis of
the NMR data for 1d, especially 15N NMR chemical shifts, led
to the confirmation of this structure. The comparison of 15N
NMR data for 1d with the data collected for the unprotected
6-methyluracil showed an essential change in the nitrogen
nuclei character of diester 1d. In 1, two nitrogen nuclei are of
the “amide” type, whereas in 1d both nitrogens are of the “pyr-
idine” type. This transformation of the nitrogen character was
related to the following changes: nitrogen/carbon atom hybrid-
ization and the aromatization of the ring, resulting in the
nitrogen/carbon shielding changes. Other reaction products,
identified on the basis of the analysis of the 1H-13C/15N HMBC
spectra, were as follows: N1/N3-di-Boc derivative (1c) and two
mono–Boc derivatives at the N1 or N3 positions (1a and 1b)
(Scheme 1). The comparison of the 13C and 15N NMR chemical
shifts between 1 and 1a/1b/1c did not reveal such spectacular
differences as in the case of 1/1d, but several significant effects
Results and discussion
Chemistry and NMR spectroscopic analysis
Acylation of uracil (U), thymine (T) and 6-methyluracil
(6-MU) with Boc2O. According to the literature, the reaction of
Scheme 1 Reagents and conditions: (a) Boc2O (3 eq.), DMAP, MeCN,
r.t., (b) Boc2O (3 eq.), pyridine, MeCN, temp. 55 °C.
Table 1 Distribution of the products in the alkylation and acylation reactions of the uracil family under different conditions
Acylation conditions
Alkylation conditionsc
aa
bb
Nucleobase substrate
Product
1.5 eq. Boc2O
3 eq. Boc2O
3 eq. Boc2O
Product
1.0 eq. EtI
1.5 eq. EtI
3 eq. EtI
1a
1b
1c
1d
1e
1g
5d
5
5
12
—
—
3
3
11
32
—
—
9
15
20
—
—
20
1h
1i
1j
1k
1l
—
—
5
75
—
3
—
—
99
—
—
17d
30
—
—
2a
2b
2c
33
—
—
59
—
39
—
4
89
2h
2i
2j
19
—
26
22
—
30
33
—
32
3a
3b
3c
3f
—
—
—
—
12
12
42
—
—
23
77
—
3h
3i
3j
—
—
—
—
12
1
31
1
16
1
29
1
3m
a a: Boc2O (3 eq.), DMAP, MeCN, r.t;. b b: Boc2O (3 eq.), pyridine, MeCN, temp. 55 °C. c EtI, K2CO3, TBABr, DMF, r.t. d Yields [%].
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 354–362 | 355

View Article Online
Paper
Organic & Biomolecular Chemistry
were also noticed. An important one refers to the change of
the 15N shielding after the replacement of hydrogen by the Boc
group at the nitrogen atom. When both N1 and N3 protons
were replaced with the Boc group (1c), the 15N shielding
decreased by ca. 30 ppm, respectively (Table 2).
After proton N3 or N1 substitution with Boc, a decrease in
the 15N shielding was still evident (ca. 30 ppm), while the
unsubstituted nitrogen nucleus was only insignificantly
shielded by ca. 2–3 ppm. Additionally, the above mentioned re-
placement caused small, but noteworthy 13C shielding effects.
ortho-Carbon nuclei relative to nitrogens always experienced a
shielding increase by ca. 2–3 ppm (Table 2).
These interesting results prompted us to investigate in
greater detail the regioselectivity of the acylation of uracil and
its isomeric methyl derivatives: thymine and 6-methyluracil.
We began by examining the effects of the Boc2O reagent excess
on the reaction product specificity. Then, the influence of
other reaction conditions, such as the pyridine addition or the
increase of the temperature on the reaction’s regioselectivity,
was studied.
In the first type of the experiment (path a, Scheme 1)
thymine, 6-methyluracil and uracil were reacted with 1.5 or
3 eq. of Boc2O in the presence of DMAP in acetonitrile. As a
result, two products were obtained and isolated for thymine.
Based on the analysis of the multinuclear NMR data, especially
the 1H/13C HMBC experiments, the structures of these pro-
ducts were confirmed. The first was Boc-di-substituted
thymine at the positions N1 and N3 (2c) and the second one
was mono-substituted thymine at the N3 position (2a) (see
Table 2). In the case of uracil, like for thymine, the di-substi-
tuted compound at the N1/N3 position (3c) was the main reac-
tion product. Moreover, apart from the Boc-mono-substituted
product at the N3 position (3a), a similar amount of the N1-Boc
1
product was formed (3b). The H, 13C and 15N NMR data for
the Boc-substituted-6-methyluracil are presented in Table 2.
For 6-methyluracil, the main product proved to be di-substi-
tuted at the O2/O4 di-O-tert-butyloxycarbonyl-pyrimidine
derivative (1d).
Additionally, after the reaction of Boc2O with 6-methyluracil
three other products were found: di-N1/N3-Boc (1c) and small
quantities of the mono-substituted N1-Boc and N3-Boc (1a, 1b),
see Table 2.
While changing the conditions of the acylation by replacing
DMAP with pyridine and increasing the temperature to 55 °C
(path b), practically only one di-N1/N3-Boc product (2c) was
obtained for thymine. This result was consistent with literature
data;12 however, a small amount of an N3-substituted product
(2b) was also isolated. For uracil, two products: di-N1/N3-Boc
(3c, yield 77%) and N3-Boc (3b, yield 23%) were obtained. In
the case of 6-methyluracil, three previously described products
were obtained again: 1a/1b/1c. In contrast to the reaction with
DMAP (path a), no O-acylated derivatives of 6-methyluracil
(at the O2 and O4 positions) were observed under these
conditions.
Unexpected product of the 6-methyluracil (6-MU) acylation
with Boc2O and the mechanism of its formation. During the
356 | Org. Biomol. Chem., 2019, 17, 354–362
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
column chromatography purification of the 6-methyluracil mechanism. The NMR studies, including theoretical DFT cal-
reaction mixture, a new entity was isolated. The NMR spectra culations, suggested the chemical structure of an unknown 1g
of this compound (1g) did not correspond to those of any pre- compound as shown in Table 3.
viously identified Boc-6-methyluracil derivatives. Very broad
Results of the modelling of the 1g formation with the DFT
signals at δ ca. 4.5 and 4.7 ppm were observed at ambient B3LYP/6-31G(d,p) calculations are given in the ESI.†
1
temperature in the H NMR spectrum. Lowering the tempera-
Although the details of the mechanism of the 1g formation
ture to −20 °C sharpened the 1H NMR signals and allowed full were unknown, it is reasonable to assume that it may be analo-
analysis of the NMR spectra and assignment of the structure gous to the known reactions of the structurally similar com-
of this new compound. In the 1H NMR spectrum of 1g pounds. It is possible that the reaction follows a route compar-
recorded at low temperature there was no H5 signal at δ ca. able to the coupling of an indole with Boc-pyridine with the
5.7 ppm which had been typically detected for uracils consecutive proton elimination from the indole ring.16
(Table 2). Moreover, the correlation spot in the 2D 1H–13C plausible pathway for the 1g formation is presented in
A
HSQC spectrum for this C5–H5̲ ̲ pair was not observed in the Scheme 2. First, the pyridine reaction with Boc2O afforded the
low temperature spectrum. Instead, five “aromatic/double salt 5. This compound was coupled with N3-Boc-6-methyluracil
bond” proton signals, which are mutually coupled in the COSY (1b), previously formed in the reaction environment.
spectrum, were noticed. The 1H NMR signal at δ = 4.77 ppm in
This process led to the intermediate complex 6 which was
the 2D 1H–13C HMBC spectrum correlates with five carbon transformed to 1g through the removal of the C5 proton from
atoms at δ = 161.1, 151.7, 123.4, 122.9 and 113.8 ppm, respect- the pyrimidine ring by the Boc anionic form. The modeling of
ively. As it was proved by analyzing the 1H–13C HMBC corre- the 1g formation with the DFT B3LYP/6-31G(d,p) calculations
lation for the methyl group at δ = 2.24 ppm, the 13C NMR predicts the final output of the reaction free energy of about
chemical shifts (161.1, 151.7 and 113.8 ppm) are related to the 14 kcal mol−1 (Table S9, ESI†).
6-methyluracil ring and this is why two remaining δC values at
The molecular model of 1g is presented in Fig. 2.
123.4 and 122.9 ppm are related to the pyridine ring.
A rough estimation of the top of the energy barrier corres-
Additionally, the 2D H–15N HMBC experiment revealed corre- ponding to the transition structure 6 is about 63 kcal mol−1
lation spots related to three 15N NMR chemical shifts of pyri- (the B3LYP/6-31G(d) calculations). This result was obtained
midine (δ = −243.3 and −194.4 ppm) and pyridine rings. The under the assumption that (Boc)2O acts as an acylation
additional nitrogen signal at δ = −254.2 ppm comes from the
1
pyridine ring as evidenced by the 1H–15N correlations of the
pyridine protons at δ = 4.64, 4.74, 6.76 and 6.91 ppm (Table 3).
This allowed us to state that the compound 1g was not the
product of a simple N- or O-acylation, but of another reaction
Table 3 The NMR chemical shifts for compound 1g
Scheme 2 A plausible pathway for the formation of 1g.
Atom
δppm experimental
δppm DFT
H1 (N1)
C2
10.81 (−243.3)
147.6
6.69 (−258.4)
154.9
N3
C4
−194.4
161.1
−199.9
168.5
C5
C6
113.8
151.7
118.2
159.0
H(CH3) at C6 (C7)
C8
H9 (C9)
H10 (C10)
H11 (C11)
N12
2.24 (16.7)
150.0
4.77 (27.8)
4.74 (106.3)*
6.91 (122.9)**
−254.2
2.19 (21.6)
159.9
4.28 (39.7)
4.91 (114.6)
6.93 (130.9)
−259.5
H13 (C13)
H14 (C14)
C15
CIV(N12)/CIV(N3)
CH3(CIVN12)/CH3(CIVN3)
6.76 (123.4)**
4.64 (105.6)*
149.9
82.4/87.0
27.8/28.0
6.86 (131.5)
4.78 (113.7)
157.3
92.2/97.0
28.3/28.9
Fig. 2 The molecular model of 1g obtained with the theoretical DFT/
6-311G(d) calculations.
*, ** – 1H/13C NMR signals can be assigned inversely.
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 354–362 | 357

View Article Online
Paper
Organic & Biomolecular Chemistry
reagent and simultaneously as the medium, i.e. Boc−, abstract-
ing the hydrogen atom from the pyrimidine ring.
The molecular mechanism of the acylation reactions for
other uracil derivatives under our complex and multi-com-
ponent conditions (DMAP, MeCN, pyridine, variable concen-
tration of (Boc)2O, temperature r.t. and 55 °C) demands an
extended study which is considered as the next step of our
investigation. As long as the details of the molecular mecha-
nism are not determined, it is hard to predict the corresponding
kinetic barriers and provide a reasonable explanation of a diver-
sity in the yield of the uracil derivatives obtained here.
Alkylation of uracil, thymine and 6-methyluracil with EtI. A
variety of derivatives obtained in the course of pyrimidine acyl-
ation with Boc2O encouraged us to test the regioselectivity of
the ethylation reaction with EtI. The ethylation reactions of
thymine, uracil and 6-methyluracil were carried out in a polar
aprotic solvent (DMF) in the presence of K2CO3 and catalytic
amounts of tetrabutylammonium bromide (TBABr). All pro-
ducts were identified in the course of 2D NMR experiments
1
1
including H–13C and H–15N HSQC and HMBC correlations.
The use of primary halogenoalkane resulted in the monoalky-
lation of thymine and uracil at the position N1 (2h/3h) and di-
alkylation at N1/N3 atoms (2j/3j),17,18 see Scheme 3.
In the case of uracil we also observed the formation of
small quantities of two products: N1/O4- (3m) and N3- (3i). This
was consistent with the observations of Gambacorta et al.19
who had also described the formation of the N1/O4-uracil
product while studying the ethylation reaction of the lithium
or potassium salts of 4-methoxy-2(1H)-pyrimidinone conju-
gated bases. However, these N1/O4-derivatives had not been
completely characterized in the quoted paper.
For 6-methyluracil, a small amount of the O2- and N3-di-
alkylation product 1l was identified in addition to the N3/N1-di-
alkylation (1j) and N3-monoalkylation (1i) products. In contrast
to the observations of Gambacorta et al.,19 who had identified
small quantities of the N1/O4-diethyl derivative of thymine
after the ethylation procedure, we did not observe any thymine
O-alkylation in our experiments.
The NMR studies showed that the replacement of the
proton with the ethyl group in compounds 1–3 caused minor
changes in the 15N shielding in comparison with the effect
observed for the corresponding Boc analogues. The difference
was smaller than 30 ppm and depended on the position of the
substitution (Table 4). In the dialkylated compounds, the 15N
Scheme 3 Reagents and conditions: EtI, K2CO3, TBABr, DMF, r.t.
358 | Org. Biomol. Chem., 2019, 17, 354–362
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
shielding decrease by ca. 8 and 11 ppm for N1 and N3 nuclei (ESI†). These tables show that the theoretically predicted
was observed. Likewise, we noticed a similar effect in the com- chemical shifts and the experimental data of the 1H, 13C or
pounds with monosubstituted nitrogen nuclei (ca. 8 ppm for 15N nuclei are highly correlated with R2 in the range 0.87–0.99
N1; ca. 12 ppm for N3). However, for the unsubstituted nitro- (Table S3†). Based on the calculated regression lines, it was
gens an inverse effect for the 15N shielding was observed (ca. possible to estimate the lacking chemical shift of the N3 nitro-
2–3 ppm, Table 4) when compared with the respective shield- gen in the 2c and 3m compounds. Also, the regression lines
ings in substrates 1–3. For the carbon atoms, these changes supported the signal assignment of the NMR chemical shifts
were smaller and less representative. The position of the sub- to certain nuclei in the investigated molecules.
1
stitution was determined by analyzing the H/13C NMR chemi-
The Gibbs free energy outputs for the model reactions of
cal shifts for methylene of ethyl groups. In the case of the N1- Boc2O and EtI with uracil derivatives were estimated using the
ethyl derivatives, the 1H/13C chemical shifts were as follows: B3LYP/6-311G(d) method (Tables S4 and S5 in the ESI†). In the
ca. 3.80/45 ppm, whereas for N3-ethyl uracils: ca. 4.00/36 ppm.
majority of cases, the theoretical prediction of the Gibbs free
The NMR parameters, similarly to the 1/1d pair of the com- energy of the model reactions leading to the mono/di-substi-
pounds, reflected the changes in the bond location/hybridiz- tutions of Boc/Et of uracil, thymine and 6-methyluracil corre-
ation in the O2- and N3-dialkylation products. The substitution sponded qualitatively to the experimental reaction yield
of the protons at O2 and N3 in 1 caused the 15N shielding (Tables S6 and S7†). For example, for a synthesized substi-
decrease of the nitrogen nuclei by ca. 60 (N1) and 10 ppm (N3) tution product the corresponding Gibbs free energy of the
in compound 1l. The same trend (shielding decrease) was model reactions was predicted to be negative (for 1b
noticeable in the13C NMR spectrum for the C2/C5/C6 and CH3 (−10.49 kcal mol−1), 1c (−18.86), and 1d (−6.71)) except for
carbon atoms, and was 4, 6, 10 and 5 ppm, respectively. N1-Boc-6-methyluracil, where Gibbs free energy was slightly
Likewise, the replacement of the protons in compound 3 at the positive (1a (+2.11 kcal mol−1)).
N1- and O4- positions leading to 3m resulted in the minor
13C/15N shielding changes. For the nuclei N1/C2/C4 and C6, usually obtained with the lowest yield.
The experimental results also showed that product 1a is
these effects were as follows: ca. 17, 5, 7 and 6 ppm, respectively.
An interesting feature of the theoretical calculations was
A bigger change should have been noticed for the N3 atom but the predicted formation of many more substituted derivatives
unfortunately the correlation spot for this nucleus was not than those found experimentally. We can only speculate that
1
observed in the H–15N HMBC experiment. Only based on the in the theoretical modeling some unknown details of the reac-
calculated regression line could we determine the 15N chemical tion mechanism had been omitted or perhaps the kinetic
shift of the N3 nucleus in the 3m compound (ca. −161 ppm). barrier was too high, preventing the detection of the supposed
The positions of the ethyl groups in 1l and 3m strongly influ- derivatives in the course of the experiment. A newly found
1
enced the H/13C NMR chemical shifts of these alkyl moieties. compound 1g and intermediate structures of its (hypothetical)
In particular, the atoms of the methylene groups exhibited synthesis were studied using the DFT B3LYP method with a
1
characteristic H/13C NMR chemical shifts (Table 4) depending smaller 6-31G(d) basis set. The molecular structure of 1g is
on whether they adjoined nitrogen or oxygen. The carbon nuclei given in Table S8,† while free energies of the (hypothetical)
of the methylene groups at the nitrogen atoms N1/N3 are more reaction components are given in Table S9.† It is predicted
shielded than the respective carbons connected to the oxygen that the free energy of the 1g formation is about 14 kcal mol−1
.
atom by ca. 20–30 ppm. Moreover, the 13C shielding for the In the course of the reaction, a transition structure 6 is postu-
methylene carbon depends strongly on the position of the lated to have been formed (Table S10†).
methylene group (N1- or N3-). For the N1-derivatives, carbon
Another interesting theoretical result was the predicted sus-
nuclei are less shielded by ca. 8–10 ppm than in the N3-substi- ceptibility of the C5 and C6 positions of uracil derivatives to
tuted products. Thus, this effect allowed easy determination of the electrophilic attack. It is worth mentioning that such a
the alkylation position in the uracil derivatives, which made the property corresponded well with the C5 and C6 gas theoretical
entire identification possible and faster.
phase acidities which were larger than the N1 or N3 acidities
for uracil and 6-methyluracil, following calculations by means
of the B3LYP/6-31+G(d) method by Kurinovich et al.21 Another
The quantum mechanical DFT calculations
Such a diversity of the substituted reaction products turned argument can be raised based on the reactivity index in the
out to be quite unexpected. Therefore, theoretical calculations form of the Fukui’s cumulative function22 for the electrophilic
were performed to estimate the molecular structure and NMR attack, f−, calculated for the C5/C6 positions (Table S11†).
chemical shifts as well as the thermodynamic conditions of Considerably negative f− values for 6-methyluracil suggested
these reactions. The molecular structures and their molecular that the C5 position should be one of the possible regions for
energies were calculated by using the quantum mechanical the electrophilic attack, as it occurred with the reaction of 1b
DFT B3LYP method with the 6-31G(d) and 6-311G(d) basis sets with Boc-pyridine 5 to give 1g. However, most of the predicted
and simulating the solvent effects (acetonitrile or DMF) with C5- or C6-substituted derivatives were not found in the present
the SMD model of the solvent.20
experiments. One cannot exclude a possibility of synthesizing
The predicted chemical shifts and a comparison of the them via alternative reaction routes, other than those proposed
empirical and theoretical values are given in Tables S1 and S2 here. We have also observed that the pyridine and DMAP sol-
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 354–362 | 359

View Article Online
Paper
Organic & Biomolecular Chemistry
vents used in the reaction of Boc2O with thymine and
Derivatives of 6-methyluracil. 1a: (22 mg, 3%), white solid;
6-methyluracil affected the reaction yield of the N1,N3-deriva- Mp 296.4 °C dec. (hexane–AcOEt); HRMS (ESI) m/z: [M + Na]+
tives. The theoretical calculations showed a minor effect on calcd for C10H14N2O6Na: 249.0843, found: 249.0851.
the reaction ΔG caused by the hydrogen bonds between pyri-
1b: (21 mg, 3%), solid; Mp 249.8 °C dec. (hexane–AcOEt);
dine molecules and mono/di-Boc derivatives of 6-methyluracil HRMS (ESI) m/z: [M + Na]+ calcd for C10H14N2O6Na: 249.0843,
(Table S12†). A more pronounced effect on ΔG can be seen for found: 249.0851.
DMAP H-bonded to the reagents, though essentially it does
1c: (111 mg, 11%), white solid; Mp 121.5 °C (hexane–
not change the order of ΔG predicted with the model where no AcOEt); HRMS (ESI) m/z: [M + Na]+ calcd for C15H22N2O6Na:
explicit solvent was used.
349.1376, found: 349.1367.
1d: (327 mg, 32%), solid; Mp 68.7 °C (hexane–AcOEt);
HRMS (ESI) m/z: [M + Na]+ calcd for C15H22N2O6Na: 349.1373,
found: 349.1376.
1g: (257 mg, 20%), foam; HRMS (ESI) m/z: [M − H] calcd for
C20H26N3O6: 404.1822, found: 404.1818.
Experimental
Chemicals and analytical grade solvents were purchased from
commercial suppliers and used without further purification
unless stated otherwise. Flash column chromatography was
performed on silica gels (200–300 mesh).
Thymines. 2a: (423 mg, 59%), solid; Mp 324.4 °C dec.
(hexane–AcOEt); HRMS (ESI) m/z: [M
+
Na]+ calcd for
C10H14N2O6Na: 249.0846, found: 249.0851.
2b: (28 mg, 4%), white solid; (lit.12) Mp 309.8 °C dec.
The melting points were determined by using a Melting
Point System (Mettler Toledo MP70).
(hexane–AcOEt); HRMS (ESI) m/z: [M
− H] calcd for
1
The H NMR, 13C NMR and 15N NMR (as 2D experiments)
C10H13N2O4: 225.0875, found: 225.0878.
2c: (403 mg, 39%), solid; (lit.12) Mp 145.6 °C (hexane–
AcOEt); HRMS (ESI) m/z: [M + Na]+ calcd for C15H22N2O6Na:
349.1376, found: 349.1365.
spectra were recorded in CDCl3 and CDCl3/DMSO-d6 solutions
with
a Varian-NMR-vnmrs600 spectrometer (at 298 K)
equipped with a 600 MHz PFG Auto XID (1H/15N–31P 5 mm)
Uracils. 3a: (84 mg, 12%), solid; (lit.14) Mp 290.1 °C dec.
indirect probehead. Standard experimental conditions and
standard Varian programs (ChemPack 4.1) were used. The H
1
(hexane–AcOEt); HRMS (ESI) m/z: [M
− H] calcd for
and 13C NMR chemical shifts are given relative to the TMS
signal at δ = 0.0 ppm, whereas neat nitromethane at δ =
0.0 ppm was used as a standard for the 15N NMR chemical
shifts. The concentration of the solutions used for the
measurements was about 10–20 mg of the compounds in
0.6 cm3 of solvents. In the case of uracil and its methyl deriva-
tives (thymine and 6-methyluracil), the appropriate com-
pounds were measured in saturated solution in CDCl3/DMSO
(10 : 1).
C9H11N2O4: 211.0719, found: 211.0717.
3b: (83 mg, 12%), solid; Mp 322.4 °C dec. (hexane–AcOEt);
HRMS (ESI) m/z: [M − H] calcd for C9H11N2O4: 211.0719,
found, 211.0722.
3c: (464 mg, 42%), white solid; Mp 98.8 °C (hexane–AcOEt);
HRMS (ESI) m/z: [M + Na]+ calcd for C14H20N2O6Na: 335.1219,
found, 335.1209.
Alkylation of the pirymidine base (uracil, 6-methyluracil and
thymine) with EtI
The mass spectra were recorded on a MaldiSYNAPT G2-S
HDMS (Waters) Spectrometer via 25 electrospray ionization
(ESI-MS).
Procedure. EtI (144 µl, 1.79 mmol) was added to the stirred
solution of thymine (150 mg, 1.19 mmol), K2CO3 (329 mg,
2.38 mmol) and TBABr (48 mg, 0.148 mM) in anhydrous DMF
(2.5 ml). The reaction mixture was stirred for 48 h at room temp-
erature. Water (15 ml) and CH2Cl2 were added. The organic layer
Alkylation of the pirymidine base (uracil, 6-methyluracil,
thymine) with Boc2O
Procedure A. 4-DMAP (8 mg, 0.065 mmol) was added to the was separated, and the aqueous phase was extracted twice with
solution of Boc2O (2.08 g, 9.52 mmol) and 6-methyluracil CH2Cl2 (15 ml). The combined organic layers were dried over
(400 mg, 3.17 mmol) in MeCN (20 ml). The reaction mixture MgSO4 and filtered. Then, the solvent was removed by evapor-
was stirred for 48 h at room temperature. The solvent was evap- ation in vacuo. The residue was purified by flash chromatography
orated in vacuo and the residue was purified by flash chrom- on silica gel (eluent: hexane–AcOEt 7 : 3 than hexane–AcOEt 1 : 1).
atography on silica gel (eluent: hexane–AcOEt, 8 : 2, 7 : 3, 1 : 1).
Thymines. 2h: (41 mg, 22%), white solid; (lit.18,25) Mp
Procedure B. Thymine (400 mg, 3.17 mmol), Boc2O (2.08 g, 227.6 °C (hexane–AcOEt); HRMS (EI) m/z: [M] calcd for
9.52 mmol), pyridine (4 ml), and MeCN (20 ml) were stirred C7H10N2O2: 154.0742, found: 154.0740.
together for 4 h at 55 °C. The reaction mixture was concen-
trated in vacuo and the residue was partitioned between (EI) m/z: [M] calcd for C9H14N2O2: 182.1055, found: 182.1055.
CH2Cl2 (25 ml) and water (25 ml). The organic layer was separ- 6-Methyluracils. 1i: (9 mg, 5%), white solid; Mp 203.6 °C
2j: (59 mg, 30%), solid; Mp 63.2 °C (hexane–AcOEt); HRMS
ated and the aqueous phase was extracted twice with CH2Cl2 (hexane–AcOEt); HRMS (ESI) m/z: [M] calcd for C7H9N2O2:
(15 ml). The combined organic layers were dried over MgSO4 153.0664, found: 153.0659.
and filtered. The solvent was removed by evaporation in vacuo
and the residue was purified by flash chromatography on silica HRMS (EI) m/z: [M] calcd for C9H14N2O2: 182.1055, found:
1j: (162 mg, 75%), solid; (lit.26) Mp 59.7 °C (hexane–AcOEt);
gel (eluent: hexane–AcOEt, 7 : 3).
182.1057.
360 | Org. Biomol. Chem., 2019, 17, 354–362
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
1l: (6 mg, 3%), oil; HRMS (EI) m/z: [M] calcd for C9H14N2O2: the manuscript; Piotr Cmoch: conceived and designed experi-
182.1055, found: 182.1051. ments, performed the NMR experiments, analyzed the data
Uracils (1.5 eq. of EtI). 3h: (93 mg, 12%), white solid; and wrote part of the paper; Piotr Krzeczyński: conceived and
(lit.5,27); Mp 152.7 °C (hexane–AcOEt); HRMS (EI) m/z: [M] designed experiments, performed the experiments, analyzed
calcd for C6H8N2O2: 140.0586, found: 140.0584.
3i: (5 mg, 1%), oil; HRMS (ESI) m/z: [M] calcd for ceived and designed experiments, analyzed the data, and
C6H7N2O2: 139.0508, found: 139.0509. wrote part of the paper; Andrzej Leś: conceived and designed
3j: (282 mg, 31%), oil (lit.5); HRMS (EI) m/z: [M] calcd for theoretical modeling, performed the calculations, analyzed the
C8H12N2O2: 168.0899, found: 168.0895. data, and wrote part of the paper. All authors consulted their
3m: (5 mg, 1%), oil; HRMS (ESI) m/z: [M + Na]+ calcd for results, have read, critically reviewed and agreed to the final
the data, and wrote part of the paper; Marcin Cybulski: con-
C8H12N2O2Na: 191.0796, found: 191.0791.
version of the manuscript.
Quantum mechanical DFT calculations
The theoretical calculations have been performed with the
Gaussian G09 suite of programs [Gaussian]. The molecular
geometries, harmonic frequencies and isotropic nuclear
shieldings (GIAO) were calculated following standard settings
within the G09 code.
Conflicts of interest
The authors declare no competing financial interest.
Acknowledgements
This work was supported under the framework of the statutory
project funded by the Polish Ministry of Science and Higher
Education. The calculations were performed at the
Interdisciplinary Centre for Mathematical and Computational
Modeling of the University of Warsaw (ICM UW, the computer
grant G18-6) which is kindly acknowledged for allocating facili-
ties and computer time.
Conclusions
In general, alkylation/acylation of pyrimidine bases afforded
mainly N1-mono-substituted and N1,N3-di-substituted deriva-
tives than N3- or acid labile O-alkyl derivatives.23 The regio-
selectivity of these reactions was determined by the acidity of
the ionisable protons of the heterocyclic ring. As mentioned in
the literature,24 the acidity of the N1–H proton (pKa = 9.43 for
U and 9.86 for thymine) is higher than that of N3–H (pKa > 13
for U and 13.96 for thymine). Despite a big difference in the
pKa of the protons it is difficult to selectively obtain the N3-
mono alkylation product.
In conclusion, the outcome of the alkylation of 6-methyl-
uracil, uracil and thymine with Boc2O depends strongly on the
reaction conditions. The regioselectivity of the Boc/Et substi-
tution of uracil, thymine and 6-methyluracil can be steered by a
reasonable choice of the experimental setup. We have studied
the following parameters controlling regioselectivity: the molar
concentration ratio of the substrate and alkylation agent, the
presence/absence of the catalyst (DMAP in this case), and the
temperature of the reaction (ambient, elevated). We have iso-
lated different acylation/alkylation products of three pyrimidine
bases: uracil, 6-methyluracil and thymine and discovered a new
product containing the N-Boc-pyridinium moiety at the C5-posi-
tion of 6-methyluracil. All the products were fully characterized
by using multinuclear NMR data.
Notes and references
1 (a) I. M. Lagoja, Chem. Biodiversity, 2005, 2, 1–50;
(b) J. I. Bardagía and R. A. Rossi, Org. Prep. Proced. Int.,
2009, 41, 479–514.
2 A. Pałasz and D. Cież, Eur. J. Med. Chem., 2015, 97, 582–
611.
3 L. Arias, A. Guzman, S. Jaime–Figueroa, F. J. Lopez,
D. J. Morgans Jr., F. Padilla, A. Perez–Medrano,
C. Quintero, M. Romero and L. Sandoval, Synlett, 1997,
1233–1234.
4 A. L. Pogolotti Jr., D. Failla and D. V. Santi, J. Heterocycl.
Chem., 1972, 9, 1423.
5 I. Ishikawa, T. Itoh, H. Takayanagi, J. Oshima,
N. Kawahara, Y. Mizuno and H. Ogura, Chem. Pharm. Bull.,
1991, 39, 1922–1930.
6 M. Frieden, M. Giraud, C. B. Reese and Q. Song, J. Chem.
Soc., Perkin Trans. 1, 1998, 2827–2832.
7 N. G. Kundu and S. G. Khatri, Synthesis, 1985, 323–326.
8 S. Manfredine, P. G. Baraldi, R. Bazzanini, M. Guarneri and
D. Simoni, J. Chem. Soc., Chem. Commun., 1994, 583–584.
9 (a) A. Sarkar, S. R. Roy, N. Parikh and A. K. Chakraborti,
J. Org. Chem., 2011, 76, 7132–7140; (b) E. A. Englund,
H. N. Gopi and D. H. Appella, Org. Lett., 2004, 6, 213–215;
(c) R. Moumne, S. Lavielle and P. Karoyan, J. Org. Chem.,
2006, 71, 3332–3334; (d) D. M. Shendage, R. Froehlich and
G. Haufe, Org. Lett., 2004, 6, 3675–3678.
The experimental findings were supported by the quantum
mechanical DFT calculations of the molecular energies and
theoretically predicted chemical shifts.
Author contributions
Olga Michalak: conceived the idea, structure and design of the
paper, conceived and designed the synthesis and performed
the experiments, analyzed the data, and wrote the majority of
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 354–362 | 361

View Article Online
Paper
Organic & Biomolecular Chemistry
10 J. R. Hwu, M. L. Jain, S.-C. Tsay and G. H. Hakimelahi,
Tetrahedron Lett., 1996, 37, 2035–2038.
11 S. Jaime–Figueroa, A. Zamilpa, A. Guzmán and
D. J. Morgans, Synth. Commun., 2001, 31, 3739–3746.
12 M. F. Jacobsen, M. M. Knudsen and K. V. Gothelf, J. Org.
Chem., 2006, 71, 9183–9190.
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi,
J. Normand, K. Raghavachari, A. Rendell, J. C. Burant,
S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam,
M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo,
J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski,
R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth,
P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels,
O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and
D. J. Fox, Gaussian 09, Revision D.01, Gaussian, Inc.,
Wallingford CT, 2013.
13 M. Bessières, V. Roy and L. A. Agrofoglio, RSC Adv., 2014, 4,
59747–59749.
14 L. Lipani, D. Odadzic, L. Weizel, J.-S. Schwed, B. Sadek and
H. Stark, Eur. J. Med. Chem., 2014, 86, 578–588.
15 J.-In Lee, Bull. Korean Chem. Soc., 2010, 31, 749–752.
16 W.-B. Qin, J.-Yi Zhu, Y.-B. Kong, Y.-H. Bao, Z.-W. Chen and
L.-X. Liu, Org. Biomol. Chem., 2014, 12, 4252–4259.
17 Z. Paryżek and B. Tabaczka, Org. Prep. Proced. Int., 2001, 33,
400–405.
18 T. Tanabe, K. Yamauchi and M. Kinoshita, Bull. Chem. Soc.
Jpn., 1977, 50, 3021–3025.
19 A. Gambacorta, D. Tofani, M. A. Loreto, T. Gasperi and
R. Bernini, Tetrahedron, 2006, 62, 6848–6854.
21 M. A. Kurinovich and J. K. Lee, J. Am. Soc. Mass Spectrom.,
2002, 13, 985–995.
22 (a) K. Fukui, Science, 1982, 218, 747–754; (b) W. Yang and
W. J. Mortier, J. Am. Chem. Soc., 1986, 108, 5708–5711.
23 G. Shaw, in Comprehensive Heterocyclic Chemistry, ed.
A. R. Katritzky and C. W. Rees, Pergamon, Oxford, 1985,
p. 499.
20 M. J. Frisch, G. W. Trucks, G. E. Schlegel, H. B. Scuseria, 24 B. Nawrot, O. Michalak, S. Olejniczak, M. W. Wieczorek,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, T. Lis and W. J. Stec, Tetrahedron, 2001, 57, 3979–3985.
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, 25 Q.-Q. Wu and Q.-H. Song, J. Phys. Chem. B, 2010, 114,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, 9827–9832.
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, 26 H. Egg and I. Volgger, Synthesis, 1982, 1071–1073.
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, 27 A. Robin, K. Julienne, J.-C. Meslin and D. Deniaud,
H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta,
Eur. J. Org. Chem., 2006, 634–643.
362 | Org. Biomol. Chem., 2019, 17, 354–362
This journal is © The Royal Society of Chemistry 2019