SNAr displacements with 6-(fluoro, chloro, bromo, iodo, and alkylsulfonyl)purine nucleosides: Synthesis, kinetics, and mechanism

Article abstract of DOI:10.1021/ja070021u

S_NAr reactions with 6-(fluoro, chloro, bromo, iodo, and alkylsulfonyl)purine nucleosides and nitrogen, oxygen, and sulfur nucleophiles were studied. Pseudo-first-order kinetics were measured with 6-halopurine compounds, and comparative reactivi

Full text of DOI:10.1021/ja070021u

Published on Web 04/17/2007
SNAr Displacements with 6-(Fluoro, Chloro, Bromo, Iodo, and
Alkylsulfonyl)purine Nucleosides: Synthesis, Kinetics, and
Mechanism1
Jiangqiong Liu and Morris J. Robins*
Contribution from the Department of Chemistry and Biochemistry, Brigham Young UniVersity,
ProVo, Utah 84602-5700
Received January 2, 2007; E-mail: morris_robins@byu.edu
Abstract: SNAr reactions with 6-(fluoro, chloro, bromo, iodo, and alkylsulfonyl)purine nucleosides and
nitrogen, oxygen, and sulfur nucleophiles were studied. Pseudo-first-order kinetics were measured with
6-halopurine compounds, and comparative reactivities were determined versus a 6-(alkylsulfonyl)purine
nucleoside. The displacement reactivity order was: F > Br > Cl > I (with BuNH2/MeCN), F > Cl ≈ Br >
I (with MeOH/1,8-diazabicyclo[5.4.0]undec-7-ene (DBU)/MeCN), and F > Br > I > Cl [with K+ -SCOCH3/
dimethyl sulfoxide (DMSO)]. The order of reactivity with a weakly basic arylamine (aniline) was: I > Br >
Cl . F (with 5 equiv of aniline in MeCN at 70 °C). However, those reactions with aniline were autocatalytic
and had significant induction periods (∼50 min for the iodo compound and ∼6 h for the fluoro analogue).
Addition of trifluoroacetic acid (TFA) eliminated the induction period, and the order then was F > I > Br >
Cl (with 5 equiv of aniline and 2 equiv of TFA in MeCN at 50 °C). The 6-(alkylsulfonyl)purine nucleoside
analogue was more reactive than the 6-fluoropurine compound with both MeOH/DBU/MeCN and iPentSH/
DBU/MeCN and was more reactive than the Cl, Br, and I compounds with BuNH2 and aniline/TFA. Titration
of the 6-halopurine nucleosides in CDCl3 with TFA showed progressive downfield 1H NMR chemical shifts
for H8 (larger) and H2 (smaller). The major site of protonation as N7 for both the 6-fluoro and 6-bromo
analogues was confirmed by large upfield shifts (∼16 ppm) of the 15N NMR signal for N7 upon addition of
TFA (1.6 equiv). Mechanistic considerations and resolution of prior conflicting results are presented.
and Sonogashira5 reactions. Many nitrogen-, oxygen-, and
sulfur-linked substituents have been introduced at C6 by
Introduction
Purine bases play central roles in a wide variety of biological
processes.2 Advances in the synthesis of purines that are
modified at C6 include the use of SNAr,3 Suzuki-Miyaura,4
nucleophilic aromatic substitution with 6-halopurine nucleosides.
It is well-known6 that the order of reactivity for SNAr reactions
with 1-halo-2,4-dinitrobenzenes is F > Cl g Br > I. However,
conflicting orders of reactivity have been noted with 6-halopu-
rine derivatives. Ve´liz and Beal3b reported that 6-bromopurine
nucleosides were more reactive than the corresponding 6-chlo-
ropurine compounds in SNAr reactions with a weakly nucleo-
philic arylamine, and our studies7 had shown that 6-iodopurine
nucleosides were more reactive than 6-chloro analogues in SN-
Ar reactions with aniline. Because 6-(alkylsulfonyl)purine
derivatives are known to be highly reactive substrates for SNAr
displacements,8 it was of interest to compare the reactivity of
such a sulfone with those of the four 6-halopurine nucleosides.
(1) Nucleic Acid Related Compounds. 144. For paper 143, see: Nowak, I.;
Robins, M. J. J. Org. Chem., published online Mar 27, 2007
http://dx.doi.org/10.1021/jo062614d.
(2) (a) Brathe, A.; Gundersen, L.-L.; Rise, F.; Eriken, A. B.; Vollsnes, A. V.;
Wang, L. Tetrahedron 1999, 55, 211-228. (b) Cocuzza, A. J.; Chidester,
D. R.; Culp, S.; Fitzgerald, L.; Gilligan, P. Bioorg. Med. Chem. Lett. 1999,
9, 1063-1066. (c) Verdugo, D. E.; Cancilla, M. T.; Ge, X.; Gray, N. S.;
Chang, Y.-T.; Schultz, P. G.; Negishi, M.; Leary, J. A.; Bertozzi, C. R. J.
Med. Chem. 2001, 44, 2683-2686. (d) Perez, O. D.; Chang, Y.-T.; Rosania,
G.; Sutherlin, D.; Schultz, P. G. Chem. Biol. 2002, 9, 475-483.
(3) (a) Kappler, F.; Hampton, A. J. Med. Chem. 1990, 33, 2545-2551. (b)
Ve´liz, E. A.; Beal, P. A. J. Org. Chem. 2001, 66, 8592-8598. (c) Pathak,
A. K.; Pathak, V.; Seitz, L. E.; Suling, W. J.; Reynolds, R. C. J. Med.
Chem. 2004, 47, 273-276. (d) Wanner, M. J.; Koch, M.; Koomen, G.-J.
J. Med. Chem. 2004, 47, 6875-6883. (e) Laufer, S. A.; Domeyer, D. M.;
Scior, T. R. F.; Albrecht, W.; Hauser, D. R. J. J. Med. Chem. 2005, 48,
710-722. (f) Wang, J.-Q.; Kreklau, E. L.; Bailey, B. J.; Erickson, L. C.;
Zheng, Q.-H. Bioorg. Med. Chem. 2005, 13, 5779-5786. (g) Maruyama,
T.; Moriwaka, N.; Demizu, Y.; Ohtsuka, M. Tetrahedron Lett. 2005, 46,
8225-8228. (h) Nair, V.; Uchil, V.; Neamati, N. Bioorg. Med. Chem. Lett.
2006, 16, 1920-1923.
(4) (a) Hocek, M.; Holy, A.; Votruba, I.; Dvorakova, H. J. Med. Chem. 2000,
43, 1817-1825. (b) Hoeck, M.; Holy, A.; Votruba, I.; Dvorakova, H.
Collect. Czech. Chem. Commun. 2000, 65, 1683-1697. (c) Hocek, M.;
Holy, A.; Votruba, I.; Dvorakova, H. Collect. Czech. Chem. Commun. 2001,
66, 483-499. (d) Yao, S.-W.; Lopes, V. H. C.; Fernandez, F.; Garcia-
Mera, X.; Morales, M.; Rodriguez-Borges, J. E.; Cordeiro, M. N. D. S.
Bioorg. Med. Chem. 2003, 11, 4999-5006. (e) Fernandez, F.; Garcia-Mera,
X.; Morales, M.; Vilarino, L.; Caamano, O.; De Clercq, E. Tetrahedron
2004, 60, 9245-9253. (f) Hocek, M.; Pohl, R. Synthesis 2004, 2869-
2876.
(5) (a) Matsuda, A.; Shinozaki, M.; Yamaguchi, T.; Homma, H.; Nomoto, R.;
Miyasaka, T.; Watanabe, Y.; Abiru, T. J. Med. Chem. 1992, 35, 241-252.
(b) Brathe, A.; Andresen, G.; Gundersen, L.-L.; Malterud, K. E.; Rise, F.
Bioorg. Med. Chem. 2002, 10, 1581-1586. (c) Brathe, A.; Gundersen, L.-
L.; Nissen-Meyer, J.; Rise, F.; Spilsberg, B. Bioorg. Med. Chem. Lett. 2003,
13, 877-880. (d) Naus, P.; Votruba, I.; Hocek, M. Collect. Czech. Chem.
Commun. 2004, 69, 1955-1970. (e) Turek, P.; Kotora, M.; Hocek, M.;
Votruba, I. Collect. Czech. Chem. Commun. 2005, 70, 339-349. (f) Berg,
T. C.; Gundersen, L.-L.; Eriksen, A. B.; Malterud, K. E. Eur. J. Org. Chem.
2005, 4988-4994.
(6) Bunnet, J. F.; Zahler, R. E. Chem. ReV. 1951, 49, 273-412.
(7) Liu, J.; Janeba, Z.; Robins, M. J. Org. Lett. 2004, 6, 2917-2919.
(8) Lin, X.; Robins, M. J. Org. Lett. 2000, 2, 3497-3499 and references cited
therein.
9
5962
J. AM. CHEM. SOC. 2007, 129, 5962-5968
10.1021/ja070021u CCC: $37.00 © 2007 American Chemical Society

SNAr Displacements with Purine Nucleosides
A R T I C L E S
Scheme 1
Scheme 2
niline/acetonitrile/∆) of 2′,3′,5′-tri-O-(2,4,6-trimethylbenzoyl)-
inosine [Ino(OMes)3] gave the crystalline 6-chloropurine nu-
cleoside 2.
We first employed conversion of adenosine derivatives into
6-bromopurine analogues by diazotive bromodeamination,15,16
but the 6-bromopurine nucleoside was contaminated with traces
of a reductively deaminated purine nucleoside that was hard to
remove. SNAr displacements of chloride from 2 with metal
bromides were not effective (<35% of 3 with NaBr/trifluoro-
acetic acid (TFA) and <70% with LiBr/TFA). Silicon-mediated
chloride/bromide exchange with chloropyridines showed that
the much stronger silicon-chlorine bond favors replacement
by bromide.17 That approach was successful, and conversion
of 2 to 3 proceeded cleanly with trimethylsilyl bromide
(TMSBr) in butanone (methyl ethyl ketone, MEK) at
e-40 °C [>98% conversion with no detected purine nucleoside
byproduct (1H NMR), 80% crystalline 3 isolated].
A Finkelstein-SNAr reaction (NaI/MEK at e-40 °C)7
efficiently converted 2 into the 6-iodopurine analogue 4.
Treatment of 2 with 3-methylbutane-1-thiol/1,8-diazabicyclo-
[5.4.0]undec-7-ene (DBU) gave the 6-(isopentylsulfanyl)purine
nucleoside 5, which was oxidized (oxone/pH 5 buffer/brine/
MeOH/H2O)8 to give the sulfone 6.
Kinetic Studies. All kinetic determinations (in triplicate) were
conducted under pseudo-first-order conditions:
A systematic comparison of 6-(fluoro-, chloro-, bromo, iodo,
and alkylsulfonyl)purine nucleosides in SNAr reactions was
lacking.
k1t ) -2.303 log(C/C0) + a
(1)
Results and Discussion
where C/C0 is the ratio of the concentration of 6-halopurine
nucleoside in the mixture at time t to the initial concentration
of 6-halopurine nucleoside. Values of the term [-log (C/C0)]
were plotted against [t (min) k (s-1) ) k1 (min-1)/60].
Synthesis of Suitable 6-(Substituted)purine Nucleoside
Substrates. Nucleoside derivatives with the sugar hydroxyl
groups protected with stable, lipophilic entities are much easier
to manipulate. We first examined 2′,3′,5′-tri-O-(4-methylben-
zoyl) derivatives, which are easily prepared and readily crystal-
lized.9 However, minor quantities of partially deprotected
byproducts were observed in the more basic nucleophilic
systems upon extended exposure with the p-toluyl esters.
Therefore, 2′,3′,5′-tri-O-(2,4,6-trimethylbenzoyl) (mesitoyl, Mes)
analogues were prepared7,10 and found to be stable11 in all of
our systems. Our fluorodeamination method12 [pyridine‚HF/tert-
butyl nitrite (TBN)] with 2′,3′,5′-tri-O-(2,4,6-trimethylbenzoyl)-
adenosine13 [Ado(OMes)3] (Scheme 1) gave the 6-fluoropurine
nucleoside 1. Chlorodeoxygenation14 (POCl3/N,N-dimethyla-
SNAr Reactions of 6-Halopurine Nucleosides with a
Primary Aliphatic Amine. The 6-halopurine nucleosides 1-4
(Scheme 2) were treated with butylamine (10 equiv) in aceto-
nitrile at 25 °C to give the 6-(butylamino)purine compound 7.
The rate constant for the reaction with the fluoropurine
nucleoside 1 was too fast to measure, even at 0 °C. The
reactivity order was F . Br > Cl > I (Figure 1, Table 1).
SNAr Reactions of 6-Halopurine Nucleosides with an
Aliphatic Alcohol/Base. Compounds 1-4 (Scheme 3) were
treated with DBU (∼5 equiv) in methanol/acetonitrile (1:1 v/v)
at 25 °C to give the 6-methoxypurine nucleoside 8. Again, the
rate constant for the reaction with the 6-fluoropurine nucleoside
1 was too fast to measure, even at 0 °C. The reactivity order
(9) Robins, M. J.; Barr, P. J.; Giziewicz, J. Can. J. Chem. 1982, 60, 554-557.
(10) Parish, R. C.; Stock, L. M. J. Org. Chem. 1965, 30, 927-929.
(11) Gassman, P. G.; Schenk, W. N. J. Org. Chem. 1977, 42, 918-920.
(12) (a) Robins, M. J.; Uznan´ski, B. Can. J. Chem. 1981, 59, 2608-2611. (b)
Secrist, J. A., III; Bennett, L. L., Jr.; Allan, P. W.; Rose, L. M.; Chang,
C.-H.; Montgomery, J. A. J. Med. Chem. 1986, 29, 2069-2074.
(13) Liu, J.; Robins, M. J. Org. Lett. 2005, 7, 1149-1151.
(14) (a) Gerster, J. F.; Jones, J. W.; Robins, R. K. J. Org. Chem. 1963, 28,
945-948. (b) Robins, M. J.; Uznan´ski, B. Can. J. Chem. 1981, 59, 2601-
2607.
(15) (a) Nair, V.; Richardson, S. G. Tetrahedron Lett. 1979, 1181-1184. (b)
Nair, V.; Richardson, S. G. J. Org. Chem. 1980, 45, 3969-3974.
(16) (a) Francom, P.; Janeba, Z.; Shibuya, S.; Robins, M. J. J. Org. Chem. 2002,
67, 6788-6796. (b) Francom, P.; Robins, M. J. J. Org. Chem. 2003, 68,
666-669.
(17) Schlosser, M.; Cottet, F. Eur. J. Org. Chem. 2002, 4181-4184.
9
J. AM. CHEM. SOC. VOL. 129, NO. 18, 2007 5963

A R T I C L E S
Liu and Robins
Figure 2. SNAr reactions of 1-4 with aniline.
Scheme 5
Figure 1. SNAr reactions of 2-4 with butylamine.
Table 1. Kinetic Data and SNAr Rate Constantsa
1
nucleophile
solvent
temp (
°
C)
entry
X
k (s-
)
Rb
nc
BuNH2
BuNH2
BuNH2
MeCN
MeCN
MeCN
25
25
25
25
25
25
30
30
30
30
50
50
50
50
1
2
3
4
5
6
7
8
9
Cl 4.6 × 10-5 0.997
5
Br 7.3 × 10-5 0.993
5
5
5
5
5
5
9
5
5
6
9
I
2.3 × 10-5 0.996
MeOH/DBU MeCN
MeOH/DBU MeCN
MeOH/DBU MeCN
K+ -SCOMe DMSO
K+ -SCOMe DMSO
K+ -SCOMe DMSO
K+ -SCOMe DMSO
PhNH2/TFA
PhNH2/TFA
PhNH2/TFA
PhNH2/TFA
Cl 1.0 × 10-4 0.990
Br 1.0 × 10-4 0.991
SNAr Reactions of 6-Halopurine Nucleosides with an
Aromatic Amine. Ve´liz and Beal3b reported that 6-bromopurine
nucleosides reacted smoothly with aromatic amines in methanol,
in contrast to the lack of reactivity they found with acetonitrile
as solvent. They also noted that the 6-bromo compounds were
more reactive than 6-chloro analogues, whereas Lakshman et
al.18 had resorted to palladium-catalyzed coupling of 6-chloro-
purine compounds with arylamines. Because the 6-iodopurine
compound 4 was not very soluble in methanol, we used
acetonitrile. Treatment of 1-4 with aniline (5 equiv) in
acetonitrile at 70 °C (Scheme 5) resulted in SNAr displacementss
but with significant lag times. The reactivity order was I > Br
> Cl > F with lag periods of ∼50 min (4), ∼1 h (3), ∼2 h (2),
and ∼6 h (1) (Figure 2).
I
F
4.6 × 10-5 0.989
1.6 × 10-3 0.999
Cl 2.2 × 10-4 0.999
Br 5.2 × 10-4 0.999
10
11
12
I
F
5.2 × 10-4 0.999
MeCN
MeCN
MeCN
MeCN
2.5 × 10-3 0.998
Cl 3.8 × 10-4 0.999
13 Br 6.9 × 10-4 0.999 10
14
I
1.7 × 10-3 0.999
7
a Experiments were repeated three times [differences among the three
experiments were within (k ( 0.1) × 10-n]. b Correlation coefficient.
c Number of points.
Scheme 3
Our observation of such lag times rationalizes the apparent
differences between our results and those of Ve´liz and Beal.3b
Accelerating rates of displacement of halide from 6-halopurine
compounds with aniline at longer reaction times is consistent
with autocatalysis by the generated hydrogen halides. Trifluo-
roacetic acid (TFA) had been used for identification of proto-
nation sites on purines in the 15N NMR studies of Roberts and
co-workers,19 and addition of TFA enhanced the reactivity of
purine derivatives in SNAr reactions.20 Our addition of TFA (2
equiv) to the reaction mixtures of Scheme 5 resulted in
disappearance of lag times, and the SNAr reactions proceeded
at accelerated rates (and at 50 °C, rather than 70 °C). The
Scheme 4
was F . Cl ≈ Br > I (kinetic plots are in the Supporting
Information, Table 1).
(18) (a) Lakshman, M. K.; Keeler, J. C.; Hilmer, J. H.; Martin, J. Q. J. Am.
Chem. Soc. 1999, 121, 6090-6091. (b) Lakshman, M. K.; Hilmer, J. H.;
Martin, J. Q.; Keeler, J. C.; Dinh, Y. Q. V.; Ngassa, F. N.; Russon, L. M.
J. Am. Chem. Soc. 2001, 123, 7779-7787. (c) Lakshman, M. K.; Gunda,
P. Org. Lett. 2003, 5, 39-42. (d) Lakshman, M. K.; Ngassa, F. N.; Bae,
S.; Buchanan, D. G.; Hahn, H.-T.; Mah, H. J. Org. Chem. 2003, 68, 6020-
6030.
(19) (a) Markowski, V.; Sullivan, G. R.; Roberts, J. D. J. Am. Chem. Soc. 1977,
99, 714-718. (b) Gonnella, N. C.; Nakanishi, H.; Holtwick, J. B.; Horowitz,
D. S.; Kanamori, K.; Leonard, N. J.; Roberts, J. D. J. Am. Chem. Soc.
1983, 105, 2050-2055.
(20) Whitfield, H. J.; Griffin, R. J.; Hardcastle, I. R.; Henderson, A.; Meneyrol,
J.; Mesguiche, V.; Sayle, K. L.; Golding, B. T. Chem. Commun. 2003,
2802-2803.
SNAr Reactions of 6-Halopurine Nucleosides with Thiolate
Nucleophiles. Compounds 1-4 were treated with potassium
thioacetate (5 equiv) in dimethyl sulfoxide (DMSO) at 30 °C
to give the 6-(acetylthio)purine nucleoside 9 (Scheme 4). The
reactivity order was F > Br ≈ I > Cl (kinetic plots are in the
Supporting Information, Table 1) [the same reactivity order (F
> Br ≈ I > Cl) was observed with 3-methylbutane-1-thiol/
DBU in acetonitrile at e-40 °C].
9
5964 J. AM. CHEM. SOC. VOL. 129, NO. 18, 2007

SNAr Displacements with Purine Nucleosides
A R T I C L E S
Figure 6. Thermodynamic protonation of N7.
Table 2. Changes of 15N NMR Chemical Shiftsa of 3 in CDCl3
with Addition of TFA
TFA (equiv)
N1
N3
N7
N9
0
102.1
102.6
103.9
105.1
107.7
137.2
137.1
136.9
136.8
136.6
140.4
143.0
148.3
151.6
156.2
217.7
217.3
216.5
216.1
215.2
0.2
0.6
1.0
1.6
a Shifts are given in parts per million (ppm) upfield from external
CH315NO2.
Figure 3. SNAr reactions of 1-4 with TFA/aniline.
TFA. The relative changes (∆δ) for H8 (Figure 4) are greater
than those for H2 (Figure 5) with added TFA.
The ∆δ changes of both the H2 and H8 signals for these
compounds are in the order 4 > 3 > 2 > 1. The larger relative
shifts for the H8 signals indicate that thermodynamic protonation
occurs at N7 > N1 (Figure 6). Greater stabilization of positive
charge at protonated N7 might result from peri interactions with
electrons on X [the largest effect observed with H8 of 4] rather
than from resonance delocalization of electron density directly
from X to protonated N1.
Although the larger ∆δ shifts for H8 relative to H2 suggested
preferential protonation at N7, 15N NMR is a superior method
to probe for protonation on nitrogen because 15N chemical shifts
are very sensitive to changes in the local environment.21 TFA
was used for identification of protonation sites on purines and
nucleosides with 15N NMR,19 and 15N NMR has been used to
evaluate metal binding to specific nitrogen atoms and for
comparison of binding potentials of different sites in ham-
merhead ribozymes.21
Figure 4. Effects of addition of TFA on H8 of 1-4.
Roberts and co-workers19a had observed upfield 15N NMR
∆δ shifts of ∼72 ppm for N1 of adenosine (1.6 equiv of TFA)
and ∼66 ppm for N7 of guanosine (1.86 equiv of TFA) in
DMSO. In contrast, we found an upfield ∆δ shift of 1.2 ppm
for N7 of 3 in DMSO (1.6 equiv of TFA). The ∆δ shifts for
N1, N3, and N9 were all <1 ppm. Adenosine (pKa 3.4) and
guanosine (pKa 1.6) are considerably more basic than DMSO
(pKa -1.8) and compete effectively for protons in TFA/DMSO.
The much less basic 3 was protonated to only a small extent in
TFA/DMSO [i.e., the ∆δ shifts for adenosine (N1) and
guanosine (N7) were much larger than that for N7 of the weakly
basic 3]. CDCl3 is essentially nonbasic, and the 15N NMR ∆δ
shift for N7 of 3 upon addition of TFA (1.6 equiv) is much
larger in CDCl3 (∼16 ppm) (Table 2) than in DMSO (1.2 ppm).
Extensive protonation at N7 of 1 (Table 3) and 3 (Table 2) by
TFA should occur in nonbasic solvents such as CH3CN (pKa
-10) as well as in CDCl3. Our 15N NMR data show that
6-fluoropurine nucleoside 1 undergoes protonation at N7 (∼16
ppm upfield shift) (Table 3). Preferential protonation at N7 of
the 6-bromopurine nucleoside 3 is apparent (∼16 ppm upfield
shift), but protonation at N1 of 3 (∼5.5 ppm upfield shift) also
is significant (Table 2) in comparison with protonation at N1
of 1 (Table 3).
Figure 5. Effects of addition of TFA on H2 of 1-4.
reactivity order with aniline (5 equiv) plus TFA (2 equiv) was
F > I > Br > Cl (Figure 3). The kinetic data from Schemes
2-4 and from Scheme 5 (PhNH2 + TFA at 50 °C) are collected
in Table 1.
The dramatic effect of adding TFA to SNAr reactions with
aniline prompted our investigation of the site(s) of protonation
1
on the purine bases. Changes in the H NMR chemical shifts
of H2 and H8 of 1-4 [0.041 mmol in CDCl3 (0.6 mL) in
standard NMR tubes] were measured as a function of added
(21) Wang, G.; Gaffney, B. L.; Jones, R. A. J. Am. Chem. Soc. 2004, 126, 8908-
8909.
9
J. AM. CHEM. SOC. VOL. 129, NO. 18, 2007 5965

A R T I C L E S
Liu and Robins
Table 3. Changes of 15N NMR Chemical Shiftsa of 1 in CDCl3
with Addition of TFA
Table 4. Orders of Leaving Group Reactivities with Compounds
1-4 and 6
TFA (equiv)
N1
N3
N7
N9
nucleophile
reactivity
0
139.0
138.9
138.9
139.0
139.8
134.6
134.6
134.4
134.3
134.2
145.9
147.0
152.8
156.9
161.8
217.7
216.9
BuNH2
F > RSO2 > Br > Cl > I
RSO2 > F > Br ≈ Cl > I
RSO2 > F > Br ≈ I > Cl
I > Br > RSO2 > Cl > F
F > RSO2 > I > Br > Cl
0.2
0.6
1.0
1.6
MeOH/DBU
iPentSH/DBU
PhNH2
216.0
215.2
PhNH2/TFA
a Shifts are given in parts per million (ppm) upfield from external
CH315NO2.
Sulfone 6 was more reactive than the 6-fluoropurine analogue
1 with methanol/DBU and with 3-methylbutane-1-thiol/DBU
(overall ordering: 6 > 1 > 3 ≈ 2 > 4 with MeOH/DBU and
6 > 1 > 3 ≈ 4 > 2 with iPentSH/DBU). Sulfone 6 was second
to 1 with both butylamine and aniline plus TFA (overall
ordering: 1 > 6 > 3 > 2 > 4 with BuNH2 and 1 > 6 > 4 >
3 > 2 with PhNH2/TFA). The 6-iodopurine analogue 4 was
most reactive with aniline in the absence of TFA; and 6 was
third in reactivity, behind the bromopurine compound 3 (overall
ordering: 4 > 3 > 6 > 2 > 1). Combinations of C-X bond
stability and autocatalysis by displaced HX might be operative
in the latter series.
Our 1H and 15N NMR data are consistent in that larger
downfield 1H NMR ∆δ shifts for the H8 signals relative to those
for H2 coincide with preferential N7 protonation apparent in
1
15N NMR spectra with additions of TFA. Although H NMR
∆δ shift trends are parallel in spectra of 1-4, differences are
apparent (Figures 4 and 5). Protonation at N1 of 3 (Table 2)
relative to the ∆δ shifts for N1 in 15N NMR spectra of 1
(Table 3) suggest that different mechanistic combinations might
be operative in acid-catalyzed SNAr reactions with such purine
derivatives that have several basic sites. A mechanism proposed
for SNAr reactions of 1-fluoro-2,4-dinitrobenzene with weakly
nucleophilic anilines invoked reversible addition in the first step
followed by rate-limiting loss of fluoride.22 Our reactions of
aniline with 1-4 (Figure 2) are consistent with this hypothesis.
Loss of halide (and a proton) from the addition complex
generates HX for autocatalysis, and the order of reactivity is 4
> 3 > 2 . 1. Addition of TFA results in thermodynamic
protonation at N7 (and likely kinetic protonation at N1sand
possibly at N3), which assists the addition of aniline at C6 of
the CdN double bond. However, TFA also could protonate
(hydrogen bond) with fluoride (HsF-C6), which would
activate the F-C bond for departure of HF. Hydrogen bonding
would be much less significant with chloride and absent with
bromide and iodide. Halide-C6 bond breaking must be involved
because the ordering is 1 > 4 > 3 > 2 (Figure 3) with TFA
added. Multiple mechanistic effects include (1) thermodynamic
and kinetic protonation of the 6-halopurine substrates, as well
as the aniline nucleophile; (2) hydrogen bonding with halide
leaving groups; and (3) carbon-halogen bond strengths. Our
studies have clarified some parameters involved in choices of
synthetic approaches with 6-halopurine nucleosides as SNAr
substrates.
Conclusions
Our results demonstrate that the 6-fluoropurine nucleoside 1
is the most reactive substrate for SNAr reactions among the four
6-halopurine analogues with an aliphatic amine, an arylamine
plus TFA, and with oxygen and sulfur nucleophiles. However,
the 6-iodopurine analogue 4 is the best substrate for the aromatic
amine (autocatalysis by generated HI) in the absence of an
external acid catalyst. The 6-(alkylsulfonyl)purine nucleoside
6 is even more reactive than 1 with oxygen and sulfur
nucleophiles. An array of substrates and quantitative as well as
qualitative comparisons are now available to guide choices for
syntheses of 6-(substituted)purine derivatives that are readily
accessible by SNAr processes.
Experimental Section24
6-Fluoro-9-[2,3,5-tri-O-(2,4,6-trimethylbenzoyl)-â-D-ribofurano-
syl]purine (1). This compound13 was prepared as reported: mp 117-
120 °C. Anal. Calcd for C40H41FN4O7: C, 67.78; H, 5.83; N, 7.90.
Found: C, 67.68; H, 5.70; N, 7.83.
6-Chloro-9-[2,3,5-tri-O-(2,4,6-trimethylbenzoyl)-â-D-ribofurano-
syl]purine (2). This compound7 was prepared as reported: mp 123-
126 °C. Anal. Calcd for C40H41ClN4O7: C, 66.25; H, 5.70; N, 7.73.
Found: C, 66.50; H, 5.70; N, 7.88.
Comparisons of a 6-(Alkylsulfonyl)purine Nucleoside with
6-Halopurine Analogues. Oxidation of easily accessible 6-(alkyl-
sulfanyl)purine derivatives provides the 6-(alkylsulfonyl)purine
counterparts. Wetzel and Eckstein23 had noted that 6-(methyl-
sulfonyl)-9-(â-D-ribofuranosyl)purine underwent SNAr displace-
ments readily. Reactivity competitions between equimolar
quantities of 9-[(2,3,5-tri-O-(2,4,6-trimethylbenzoyl)-â-D-ribo-
furanosyl]-6-(3-methylbutylsulfonyl)purine13 (6) and each of the
four 6-halopurine analogues with butylamine, methanol, 3-me-
thylbutane-1-thiol, and aniline (with and without 2 equiv of
TFA) under the same conditions described for the kinetic studies
are summarized in Table 4. Because sulfone 6 was more reactive
6-Bromo-9-[2,3,5-tri-O-(2,4,6-trimethylbenzoyl)-â-D-ribofurano-
syl]purine (3). A solution of 2 (200 mg, 0.27 mmol) and TMSBr (0.4
mL, 0.3 g, 3 mmol) in butanone (5 mL) was stirred at -40 °C for 2 h.
The reaction mixture was poured into saturated NaHCO3/H2O and
extracted (CH2Cl2). The organic layer was washed (brine) and dried
(Na2SO4). Volatiles were evaporated, and the residue was recrystallized
1
(EtOH) to give 3 (170 mg, 80%): mp 121-124 °C; H NMR δ 2.07,
2.19, 2.24 (3 s, 3 × 6H), 2.25, 2.28, 2.32 (3 s, 3 × 3H), 4.70-4.81
(m, 3H), 6.14 (t, J ) 5.0 Hz, 1H), 6.34 (d, J ) 5.0 Hz, 1H), 6.38 (t,
J ) 5.5 Hz, 1H), 6.78, 6.83, 6.86 (3 s, 3 × 2H) 8.22, 8.59 (2 s, 2 ×
1H); 13C NMR δ 20.06, 20.10, 20.2, 21.4, 21.5, 63.4, 71.5, 73.9, 81.2,
87.7, 128.6, 128.85, 128.88, 128.94, 129.2, 129.9, 135.0, 135.4, 136.0,
136.1, 140.2, 140.5, 140.7, 143.88, 143.93, 150.1, 152.4, 168.5, 168.8,
169.7; HRMS m/z 791.2037 [M + Na+ (C40H4179BrN4O7Na) )
791.2056]. Anal. Calcd for C40H41BrN4O7: C, 62.31; H, 5.21; N, 7.17.
Found: C, 62.42; H, 5.37; N, 7.28.
i
than the fluoride 1 with MeOH/DBU and PentSH/DBU, and
the reactivity with BuNH2 also was too fast under comparable
conditions (as noted above for the 6-fluoro analogue 1), kinetic
plots with sulfone 6 are not included.
(22) (a) Forlani, L.; Tortelli, V. J. Chem. Res., Synop. 1982, 62-63. (b) Forlani,
L. J. Chem. Res., Synop. 1984, 260-261.
(23) Wetzel, R.; Eckstein, F. J. Org. Chem. 1975, 40, 658-660.
(24) Experimental details are in the Supporting Information.
9
5966 J. AM. CHEM. SOC. VOL. 129, NO. 18, 2007

SNAr Displacements with Purine Nucleosides
A R T I C L E S
Table 5. 15N-1H and 15N-19F Coupling Constants for 1 and 3
6-Iodo-9-[2,3,5-tri-O-(2,4,6-trimethylbenzoyl)-â-D-ribofuranosyl]-
purine (4). This compound7 was prepared as reported: mp 143-145
°C. Anal. Calcd for C40H41IN4O7: C, 58.83; H, 5.06; N, 6.86. Found:
C, 58.60; H, 4.92; N, 6.72.
6-(3-Methylbutylsulfanyl)-9-[2,3,5-tri-O-(2,4,6-trimethylbenzoyl)-
â-D-ribofuranosyl]purine (5). This compound13 was prepared and
characterized as reported.
6-(3-Methylbutylsulfonyl)-9-[2,3,5-tri-O-(2,4,6-trimethylbenzoyl)-
â-D-ribofuranosyl]purine (6). This compound13 was prepared and
characterized as reported.
compd
J(N1 H2), Hz
-
J(N3 H2), Hz
-
J(N7 H8), Hz
-
J(N1 F), Hz
-
J(N3 F), Hz
-
J(N7 F), Hz
-
1
3
15.4
15.9
14.8
14.6
12.1
11.6
47.0
7.3
4.7
73.9, 81.2, 87.7, 128.6, 128.87, 128.93, 129.3, 130.1, 135.5, 136.0,
136.2, 136.5, 140.1, 140.4, 140.6, 141.8, 143.7, 144.8, 168.4, 168.8,
169.8, 176.9; HRMS m/z 745.2664 [M + Na+ (C40H42N4O7SNa) )
745.2672].
General Procedure A: Reactions of 1-4 with Butylamine. BuNH2
(40 µL, 30 mg, 0.41 mmol) was added to a stirred solution of 2 (29.7
mg, 0.0410 mmol) in CH3CN (3 mL) at 25 °C. Aliquots were removed
after specified times and quenched into aqueous buffer (pHydrion buffer
6.00 ( 0.02). The mixtures were extracted (CH2Cl2), and volatiles were
evaporated from the organic layers. The residues were dissolved (CDCl3,
0.55 mL), and C/C0 ratios were measured by integration of H NMR
acquisitions. The same procedure was used for 3 and 4 (the reaction
with 1 was too fast to measure).
General Procedure D: Reactions of 1-4 with Aniline. PhNH2
(9.5 µL, 9.7 mg, 0.10 mmol) was added to a solution of 2 (15 mg,
0.021 mmol) in CD3CN (0.6 mL) in an NMR tube. The reaction mixture
was heated at 70 °C in the NMR spectrometer, and C/C0 ratios were
measured by integration of 1H NMR acquisition data. The same
procedure was used for 1, 3, and 4.
1
General Procedure E: Reactions of 1-4 with Aniline/TFA.
PhNH2 (9.5 µL, 9.7 mg, 0.10 mmol) and TFA (3.2 µL, 4.7 mg, 0.042
mmol) were added to a solution of 2 (15 mg, 0.021 mmol) in CD3CN
(0.6 mL) in an NMR tube. The reaction mixture was heated at 50 °C
in the NMR spectrometer, and C/C0 ratios were measured by integration
6-N-Butyl-2′,3′,5′-tri-O-(2,4,6-trimethylbenzoyl)adenosine (7). Treat-
ment of 2 (50 mg, 0.069 mmol) by general procedure A gave 7 (46
mg, 87%): 1H NMR δ 0.98 (t, J ) 7.3 Hz, 3H), 1.44-1.49 (m, 2H),
1.65-1.70 (m, 2H), 2.05, 2.18, 2.29 (3 s, 3 × 6H), 2.24, 2.27, 2.29 (3
s, 3 × 3H), 3.66 (br s, 2H), 4.68-4.71 (m, 1H), 4.73-4.77 (m, 1H),
4.80-4.84 (m, 1H), 5.70 (br s, 1H), 6.09-6.11 (m, 1H), 6.30 (d, J )
5.4 Hz, 1H), 6.34-6.37 (m, 1H), 6.76, 6.81, 6.86 (3 s, 3 × 2H), 7.82,
8.35 (2 s, 2 × 1H); 13C NMR δ 14.0, 20.1, 20.3, 21.3, 21.4, 32.0, 40.7,
64.0, 71.9, 73.7, 80.9, 86.6, 120.4, 128.78, 128.81, 128.84, 129.5, 130.2,
135.6, 135.9, 136.2, 136.3, 138.3, 139.9, 140.2, 140.4, 149.0, 153.8,
155.2, 168.5, 168.9, 169.8; HRMS m/z 762.3871 [M + H+ (C44H52N5O7)
) 762.3867].
General Procedure B: Reactions of 1-4 with Methanol/DBU.
DBU (31 µL, 31 mg, 0.21 mmol) was added to a stirred solution of 2
(29.7 mg, 0.0410 mmol) in CH3OH/CH3CN (1/1 v/v; 3 mL) at 25 °C.
Aliquots were removed after specified times and quenched into aqueous
buffer (pHydrion buffer 6.00 ( 0.02). The mixtures were extracted
(CH2Cl2), and volatiles were evaporated from the organic layers. The
residues were dissolved (CDCl3, 0.55 mL), and C/C0 ratios were
measured by integration of 1H NMR acquisitions. The same procedure
was used for 3 and 4 (the reaction with 1 was too fast to measure).
6-Methoxy-9-[2,3,5-tri-O-(2,4,6-trimethylbenzoyl)-â-D-ribofura-
nosyl]purine (8). Treatment of 2 (50 mg, 0.069 mmol) by general
procedure B gave 8 (44 mg, 88%): 1H NMR δ 2.05, 2.19 (2 s, 2 ×
6H), 2.24, 2.30 (2 s, 2 × 3H), 2.28 (s, 9H), 4.20 (s, 3H), 4.71-4.84
(m, 3H), 6.11-6.13 (m, 1H), 6.36-6.39 (m, 2H), 6.76, 6.82, 6.86 (3
s, 3 × 2H), 8.00, 8.50 (2 s, 2 × 1H); 13C NMR δ 20.07, 20.11, 21.35,
21.37, 21.43, 54.6, 63.8, 71.8, 73.7, 81.1, 87.0, 122.2, 128.6, 128.82,
128.85, 128.88, 129.4, 130.1, 135.5, 135.9, 136.2, 140.0, 140.4, 140.5,
140.9, 151.8, 152.8, 161.4, 168.4, 168.9, 169.8; HRMS m/z 743.3060
[M + Na+ (C41H44N4O8Na) ) 743.3057].
General Procedure C: Reactions of 1-4 with Potassium Thio-
acetate. KSAc (11.9 mg, 0.105 mmol) was added to a solution of 2
(14.8 mg, 0.0205 mmol) in DMSO-d6 (0.6 mL) in an NMR tube. The
reaction mixture was warmed at 30 °C in the NMR spectrometer, and
C/C0 ratios were measured by integration of 1H NMR acquisitions. The
same procedure was used for 1, 3, and 4. [A similar procedure was
used with iPentSH/DBU and 1-4 with cooling to ∼-40 °C in the NMR
spectrometer, and the same order of reactivity was observed.]
2′,3′,5′-Tri-O-(2,4,6-trimethylbenzoyl)-6-thioinosine. Treatment of
2 (50 mg, 0.069 mmol) by general procedure C (with CH3CN as solvent
instead of DMSO, and workup as in general procedure B) resulted in
S-deacetylation during workup to give the title compound (43 mg,
87%): 1H NMR δ 2.08, 2.18, 2.26 (3 s, 3 × 6H), 2.24, 2.27, 2.30 (3
s, 3 × 3H), 4.71-4.85 (m, 3H), 6.11 (t, J ) 5.0 Hz, 1H), 6.26 (d, J )
5.0 Hz, 1H), 6.33 (t, J ) 5.0 Hz, 1H), 6.77, 6.82, 6.87 (3 s, 3 × 2H),
8.12, 8.23 (2 s, 2 × 1H); 13C NMR δ 20.1, 20.2, 21.4, 21.5, 63.4, 71.5,
1
of H NMR adquisition data. The same procedure was used for 1, 3,
and 4.
6-N-Phenyl-2′,3′,5′-tri-O-(2,4,6-trimethylbenzoyl)adenosine (10).
Treatment of 2 (50 mg, 0.069 mmol) by general procedure D (with
CH3CN instead of CD3CN) gave 10 (44 mg, 82%): 1H NMR δ 2.09,
2.21, 2.31 (3 s, 3 × 6H), 2.26, 2.29, 2.30 (3 s, 3 × 3H), 4.73-4.86
(m, 3H), 6.16 (t, J ) 4.9 Hz, 1H), 6.37 (d, J ) 4.9 Hz, 1H), 6.42 (t,
J ) 5.3 Hz, 1H), 6.78, 6.84, 6.87 (3 s, 3 × 2H), 7.15 (t, J ) 7.3 Hz,
1H), 7.41 (t, J ) 7.3 Hz, 2H), 7.81 (d, J ) 7.8 Hz, 2H), 7.85 (br s,
1H), 7.97, 8.50 (2 s, 2 × 1H); 13C NMR δ 20.1, 20.2, 21.38, 21.41,
21.44, 63.9, 71.8, 73.9, 81.0, 86.9, 120.8, 120.9, 124.0, 128.82, 128.86,
128.91, 129.3, 129.4 130.2, 135.5, 135.9, 136.2, 138.6, 139.5, 140.0,
140.4, 140.5, 149.6, 152.5, 153.4, 168.6, 168.9, 169.8; HRMS m/z
804.3368 [M + Na+ (C46H47N5O7Na) ) 804.3373].
Relative Reactivity Comparisons with 1-4 and 6. Solutions of
equimolar quantities of (1 and 6), (2 and 6), (3 and 6), and (4 and 6)
in five series of four separate NMR tubes were subjected to treatment
under the same conditions with the same molar ratios of BuNH2, MeOD/
i
DBU, PentSH/DBU, PhNH2, and PhNH2/TFA used in the kinetic
experiments. Integration of 1H NMR acquisition data allowed measure-
ment of the relative quantities of products formed.
Procedure for Titration with TFA (1H NMR). A solution of 1
(29.4 mg, 0.0410 mmol) in CDCl3 (0.6 mL) was added to an NMR
tube. Precise quantities of TFA were added, and the 1H NMR chemical
shifts of H2 and H8 were measured accurately. The same procedure
was used with equimolar quantities of 2, 3, and 4.
Procedure for Titration with TFA (15N NMR). A solution of 1
(200 mg, 0.282 mmol) in CDCl3 (0.5 mL) was added to an NMR tube.
Precise quantities of TFA were added, and 15N NMR chemical shifts
were measured. The 15N NMR acquisitions required 12 h for each set
of data (the signal/noise ratio precluded measurement of the shift for
N9 after addition of 0.6 equiv of TFA). The same procedure was used
for 3 (217 mg, 0.282 mmol).
15N Signal Assignments. The 15N-1H and 15N-19F coupling
constants (Table 5), as well as reported 15N NMR studies of related
compounds,25 were used to make resonance peak assignments for
2
specific nitrogen atoms. The J(N1-F) ) 47.0 Hz coupling constant for
the signal centered at 139 ppm in the spectrum of 1 is much larger
than the N3-F and N7-F couplings of 7.3 and 4.7 Hz. The resonance
peak for N7 at 145.9 ppm was identified by its N7-H8 coupling
constant (12.1 Hz), which is smaller than the N3-H2 coupling (14.8
Hz) at 134.6 ppm in the six-membered ring. The peak at 217.7 ppm
was assigned to N9 by analogy with adenosine (N9 at 205 ppm with
(25) Sibi, M. P.; Lichter, R. L. Org. Magn. Reson. 1980, 14, 494-496.
9
J. AM. CHEM. SOC. VOL. 129, NO. 18, 2007 5967

A R T I C L E S
Liu and Robins
HNO3 as external standard, 223 ppm with CH3NO2). Assignments of
resonance peaks to the nitrogen atoms in 3 were made on the basis of
15N-1H coupling constants and 15N studies of related compounds. The
N7 resonance was identified by its N7-H8 coupling constant (11.6
Hz), which is smaller than the N1-H2 (15.9 Hz) and N3-H2 (14.6
Hz) couplings in the six-membered ring. The reported coupling constant
for N1-H (15.9 Hz)26 of [1-15N]-6-bromo-9-(2′,3′,5′-tri-O-acetyl-â-D-
ribofuranosyl)purine is identical to that for N1-H (15.9 Hz) of 3. 1H-
15N HMQC spectra of 1 and 3 confirmed the 15N assignments.
Young University Roland K. Robins Graduate Research Fel-
lowships (J.L.) for support, and we thank Dr. Du Li for extensive
acquisitions of 15N NMR spectral data.
Supporting Information Available: General experimental
details, enlarged pseudo-first-order plots (containing all data
points) for SNAr reactions of 1-4 with nitrogen, oxygen, and
sulfur nucleophiles, and enlarged titration plots of H8 and H2
resonance shifts with additions of TFA. This material is available
free of charge via the Internet at http://pubs.acs.org.
Acknowledgment. We gratefully acknowledge pharmaceuti-
cal company unrestricted gift funds (M.J.R.) and Brigham
(26) Terrazas, M.; Ariza, X.; Farrax, J.; Guisado-Yang, J. M.; Vilarrasa, J. J.
Org. Chem. 2004, 69, 5473-5475.
JA070021U
9
5968 J. AM. CHEM. SOC. VOL. 129, NO. 18, 2007