Solvent-Controlled, Site-Selective N-Alkylation Reactions of Azolo-Fused Ring Heterocycles at N1-, N2-, and N3-Positions, Including Pyrazolo[3,4- d]pyrimidines, Purines, [1,2,3]Triazolo[4,5]pyridines, and Related Deaza-Compounds

Article abstract of DOI:10.1021/acs.joc.8b00540

Alkylation of 4-methoxy-1H-pyrazolo[3,4-d]pyrimidine (1b) with iodomethane in THF using NaHMDS as base selectively provided N2-methyl product 4-methoxy-2-methyl-2H-pyrazolo[3,4-d]pyrimidine (3b) in an 8/1 ratio over N1-methyl product (2b). Interestingly, conducting the reaction in DMSO reversed selectivity to provide a 4/1 ratio of N1/N2 methylated products. Crystal structures of product 3b with N1 and N7 coordinated to sodium indicated a potential role for the latter reinforcing the N2-selectivity. Limits of selectivity were tested with 26 heterocycles which revealed that N7 was a controlling element directing alkylations to favor N2 for pyrazolo- and N3 for imidazo- and triazolo-fused ring heterocycles when conducted in THF. Use of ¹H-detected pulsed field gradient-stimulated echo (PFG-STE) NMR defined the molecular weights of ionic reactive complexes. This data and DFT charge distribution calculations suggest close ion pairs (CIPs) or tight ion pairs (TIPs) control alkylation selectivity in THF and solvent-separated ion pairs (SIPs) are the reactive species in DMSO.

Full text of DOI:10.1021/acs.joc.8b00540

Subscriber access provided by Washington University | Libraries
Article
Solvent Controlled, Site-Selective N-Alkylation Reactions
of Azolo-Fused Ring Heterocycles at N1-, N2- and N3-
Positions, Including Pyrazolo[3,4-d]pyrimidines, Purines,
1,2,3]Triazolo[4,5]pyridines and Related Deaza-Compounds
[
Brett Carder Bookser, Michael I. Weinhouse, Aaron C. Burns, Andrew N. Valiere, Lino J.
Valdez, Pawel Stanczak, Jim Na, Arnold L. Rheingold, Curtis E. Moore, and Brian Dyck
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.8b00540 • Publication Date (Web): 23 May 2018
Downloaded from http://pubs.acs.org on May 23, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 61
The Journal of Organic Chemistry
TITLE
1
2
Solvent Controlled, Site-Selective N-Alkylation Reactions of Azolo-Fused Ring Heterocycles at N1-, N2- and N3-
Positions, Including Pyrazolo[3,4-d]pyrimidines, Purines, [1,2,3]Triazolo[4,5]pyridines and Related Deaza-
Compounds
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
*, †
†
†
†
†
‡
1Brett C. Bookser , Michael I. Weinhouse , Aaron C. Burns , Andrew N. Valiere , Lino J. Valdez , Pawel Stanczak ,
2
3

§
§
†
Jim Na , Arnold L. Rheingold , Curtis E. Moore and Brian Dyck
4
5
6
†
‡

7Departments of Medicinal Chemistry, Structural Biology and Scientific Computing, Dart NeuroScience LLC, 12278
8
Scripps Summit Dr., San Diego, CA 92131.
4
5code 0358, 9500 Gilman Drive, La Jolla, CA 92093-0358.
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 61
TABLE OF CONTENTS GRAPHIC
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
ABSTRACT
4
5
6
7
Alkylation of 4-methoxy-1H-pyrazolo[3,4-d]pyrimidine (1b) with iodomethane in THF using NaHMDS as base
8
selectively provided N2-methyl product 4-methoxy-2-methyl-2H-pyrazolo[3,4-d]pyrimidine (3b) in 8/1 ratio over N1-
4
N1/N2 methylated products. Crystal structures of product 3b with N1 and N7 coordinated to sodium indicated a
5
6
potential role for the latter reinforcing the N2-selectivity. Limits of selectivity were tested with 26 heterocycles which
7
8
revealed that N7 was a controlling element directing alkylations to favor N2 for pyrazolo- and N3 for imidazo- and
9
0
1
1triazolo-fused ring heterocycles when conducted in THF. Use of H-detected pulsed field gradient-stimulated echo
2
3
(
PFG-STE) NMR defined the molecular weights of ionic reactive complexes. This data and DFT charge distribution
4
5
6
calculations suggest close ion pairs (CIPs) or tight ion pairs (TIPs) control alkylation selectivity in THF and solvent
7
8separated ion pairs (SIPs) are the reactive species in DMSO.
9
0
1
INTRODUCTION
2
3
4
1
Fused 5,6-ring systems are found in 52 of the 100 top selling drugs. In addition, nitrogen containing
5
6
2
7
8
heterocycles are contained in 59% of small-molecule drugs. Of these ring systems, the purine ring is the second most
9
0
2
3
common fused ring nitrogen heterocycle. The prevalence is so common in drug discovery that the purine ring and
1
2
4
similar 5,6-fused ring heterocycles have been described as "privileged structures". This may be a logical consequence
3
4of the importance of purine rings in a myriad of biological processes. Many drug discovery efforts towards purine
5
6
7
5
recognizing targets involve creative mimicking of the endogenous purine core. This is the case for kinases, purine
8
9
0
2
ACS Paragon Plus Environment

Page 3 of 61
The Journal of Organic Chemistry
6
7
receptors, phosphodiesterases and other important biological targets. Additionally, it has been observed that
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
incorporation of an N atom to replace an aromatic CH group may provide substantial gains in potency and physical
8
chemical properties in drug discovery lead compounds. Synthetic access to these types of compounds often involves
substitution reactions of commercially available N-containing heteroaryls. Perhaps the simplest of these reactions is N-
alkylation of a heteroaryl NH-position. Our understanding of how to perform this reaction in a site-selective manner on
0
1
9
purine and purine-like ring systems continues to grow. Yet, challenges remain, as the ambident nature of the azole ring
2
anion complicates the ability to achieve site-selective N-alkylation (see Table 1).
3
4
5
6
7
Table 1. Azolo-Fused Ring Heterocycle N-Alkylations Explored.
8
9
0
1
2
3
4
5
X2 X3 X5 X7
6ring name (1)
7
8
9
1
H-Indazole
N
N
CH CH CH
0
1
H-Pyrazolo[4,3-c]pyridine
H-Pyrazolo[3,4-b]pyridine
CH
N
CH
N
1
2
1
N
CH CH
3
4
51H-Pyrazolo[3,4-d]pyrimidine
N
CH
N
N
N
6
7
8
1
H-Benzo[d]imidazole
H-Imidazo[4,5-c]pyridine
H-Imidazo[4,5-b]pyridine
H-Purine
CH
CH
CH
CH
N
CH CH
9
1
N
N
CH
N
CH
N
0
1
3
N
2
3
9
N
N
4
5
61H-Benzo[d][1,2,3]triazole
N
CH CH
1
H-[1,2,3]Triazolo[4,5-c]pyridine N
H-[1,2,3]Triazolo[4,5-b]pyridine N
N
N
CH
N
N
CH
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 61
In medicinal chemistry these ring systems can become central core structural elements. In the example depicted
in Figure 1, combining N1-, N2-, and N3-alkylations with the three possible mono-substitutions of the six-member
ring, minimally nine different two-vector combinations are conceivable. Having site-control of the azole ring alkylation
provides the ability to design compound structures to have specific size, shape and electronic features to complement
protein binding pockets and achieve the highly potent, and specific binding interactions necessary to impact the
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
biological function desired for a pharmacological effect. On the contrary, compound library synthesis methods which
2
3
produce an equal distribution of alkylation products from each core heterocycle, i.e. all azole nitrogen sites being
4
5alkylated equally, may also be an advantage.
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
Figure 1. Possible Vector Substitution Patterns in a 6-5 Ring System.
4
5
6
This report presents our study of the alkylation of the ring systems in Table 1 and how the X-ray crystal
structure determination of one of the reaction products inspired a strategy to predict alkylation outcomes depending on
7
8
9
the nature of the azole, the solvent and the presence or absence of a N7 nitrogen atom.
0
1
2RESULTS AND DISCUSSION
3
4
5
6
The alkylation of 1H-pyrazolo[3,4-d]pyrimidines has typically been observed to be selective for N1 in various
7
8
1
0
11
12
conditions such as phase transfer catalysis in benzene, Cs
2
CO
3
in DMF, or Mitsunobu conditions in THF. One
9
0
1
exception to this trend showed N2-alkylation selectivity for the specific heterocycle, 4,6-bis(methylthio)-1H-
2
3
13
pyrazolo[3,4-d]pyrimidine, under conditions of NaH in THF. In our investigations of the one-pot S
N
Ar/alkylation
4
5
1
0
reaction of 1a, selective production of 3b was observed and gravimetrically determined to be produced in greater than
6
7
8
9
0
4
ACS Paragon Plus Environment

Page 5 of 61
The Journal of Organic Chemistry
1
13
1
00% yield (Equation 1, 2b was not isolated in this experiment). Various NMR experiments ( H, C, 2D NOESY and
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
D HMBC experiments in DMSO-d
6
, CDCl
3
, THF-d
8
, and methanol-d ) on the product produced identical data to that
4
1
0
observed for authentic compound 3b. However, X-ray crystallography (see Figure 2) and microanalysis revealed this
product to be 3b complexed with sodium iodide and water in the variable formula [3b]
corrected the observed yield to 65% based on the heavier of the two complexes, 5b.
x
[NaI]
y
[H
2
O] (5) which
z
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
Upon ethyl acetate/hexane recrystallization, two different crystal forms were identified with defined oligomer
complexes [3b] [NaI] [H O] (5a) and [3b] [NaI] [H O] (5b) (Figure 2). These structures represent the first reported
3
2
2
6
4
3
2
2
5
6
7pyrazolo[3,4-d]pyrimidine complexes with sodium ions. The exact nature of the complex seems to be dependent on the
8
9
0
availability of water with the higher hydrated complex 5a fully separating the sodium and iodide ions. The relative
1
2
geometry of N1 to N7 is nearly ideal to form neighboring apical ligand bonds both above and below two sodium ions in
3
4
1
4
C
2h symmetric orientation in 5a. Six water molecules occupy the equatorial binding sites of the side-by-side
5
6octahedral sodium binding complexes. The N1-Na1 and N7-Na1’ bond distances average 2.540 Ǻ which agrees with
7
8
9
15
16
previously published coordination complexes of sodium with pyrazolo or pyrimidinyl nitrogens. The Na-Na distance
0
1
in 5a, of 3.292 Ǻ is more compressed, enforced by the N1 and N7 bonds, compared to that of the simple water bridged
2
1
7
3NaI complex (NaI)
2
(H
2
O) which is 3.56 Ǻ. By comparison, complex 5b with only two water molecules, instead of
4
4
5
6
the six as in 5a, contains sodium atoms bridged by iodides and water molecules similar to the reference compound
7
8
9
1
7
(NaI)
2
(H
2
O) . Yet similar to complex 5a, the pyrazolopyrimidine N1 and N7 of 5b occupy apical ligand bonds with
4
0N1-Na1 and N7B-Na2 bond distances averaging 2.511 Ǻ. Interestingly, in order to accommodate bridging iodides, the
1
2
3
heterocycles turn 90 degrees on the apical axis relative to complex 5a. Also, in another striking difference, apical
4
5
bonding is N1-Na1-N1’ in 5b but is N1-Na1-N7’ in 5a. The Na1-Na2 intra-atomic distance is 3.951 Ǻ in 5b where the
6
7
sodium atoms are bridged simultaneously by water and iodide. This distance, 0.656 Ǻ longer than the Na1-Na1’ length
8
9
0
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 61
in 5a, is appropriate since it falls approximately in the middle of Na-Na distances bridged individually by water (3.56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Ǻ) or iodide (4.51 Ǻ) in the reference compound (NaI)
2
(H
2
O)
.17
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
Figure 2. X-ray Crystal Structure Forms of [3b]
x
[NaI]
y
[H
2
O] (5), ellipsoid contours for each at 50%
z
5
6
probability. Selected bond lengths (Ǻ), 5a: Na1-Na1’ 3.295(5), Na1-N1 2.534(6), Na1-N7’ 2.546(5); 5b: Na1-Na2
7
8
9
0
6
ACS Paragon Plus Environment

Page 7 of 61
The Journal of Organic Chemistry
3
.951(11), Na2-N7B 2.502 (2), Na1-N1 2.519(2). Selected bond angles (deg), 5a: O1S-Na1-O3S 170.03(19), O2S-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Na1-O4S 172.04(17), N1-Na1-N7’ 158.88(19), Na1’-O1S-Na1 89.9(3), Na1’-O2S-Na1 87.7(3); 5b: Na1-I1-Na2’
7
7.26(3), N1’-Na1-N1 165.45(11), Na2-O1S-Na1 93.96(7), O1S-Na2-I2 173.46(6), O1S’-Na1-I1’ 158.14(5).
These crystal structures indicate that N1 and N7 of compound 3b can have ligand bonding interactions with
sodium in the solid state. The association with sodium iodide is maintained through silica gel chromatography
0
indicating solution state association as well. These observations resulted in a hypothesis that the selective N2 alkylation
*
tight ion pair or ion pair dimer (see intermediate 6 for example, Equation 2), then N1-alkylation is limited and thus N2-
2
3and without N7.
In the optimization study, the alkylation of 1b with iodomethane was studied in detail with N1/N2 (2b/3b)
6
7Solvent choice proved to be the primary factor determining alkylation selectivity under a variety of conditions as
8
9
shown in Table 2. Less polar solvents like dioxane (entries 1-3) and THF (entries 4-5) favored production of the N2-
0
1
methyl product 3b with the N1/N2 (2b/3b) product ratios varying from 1/5 to 1/10. Toluene and acetonitrile were
2
3
4ineffective as solvents perhaps due to solubility limitations (entries 9-10). Polar solvents DMSO (entries 6-7), DMF
5
6
(
entry 11) and DMPU (entry 12) reversed the selectivity to favor production of the N1-product 2b with ratios of 2b/3b
7
8
9
0
7
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 61
of 2/1 to 5/1. Thus, for substrate 1b, N2-site-selectivity can be effected by choosing THF or dioxane to preferentially
provide 3b (69% yield, entry 4) consistent with a tight ion pair (similar to 6) favoring reactivity at N2. Use of DMSO
and other polar solvents selectively provided 2b (55% yield, entry 6) consistent with a solvent separated ion pair and
permitting the more reactive solvated nitrogen anion to be alkylated. Notably, conducting the reaction in 10:1
THF/DMSO produced a 1:1 distribution of 2b/3b (entry 8).
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
In this table and subsequent tables, most often 3 equivalents of iodomethane or other electrophile were used as
the alkylating reagent. The reaction also occurs with essentially the same product composition when 1.5 equivalents of
4
5
electrophile are used, compare entries 4 to 5 and 6 to 7, except more time is required for the reactions to go to
6
DMSO reaction was technically 3:1 DMSO/THF. Moreover, rigorously excluding THF by employing solid NaOt-Bu in
7
8
DMSO, resulted in no further N1 selectivity, (see entries 17 and 18).
9
0
1
2
3
4
Larger cations, such as those derived by reactions with bases DBU (entry 23), P
2
-Et (entries 24-26) and Cs
2
CO
3
(
entries 30-31) also did not provide selectivity for the N2 product 3b and instead favored N1 (2b) in THF and provided
5
6
7increased N1/N2 product ratios when conducted in DMSO (entries 26 and 31). This is consistent with the cation being
8
9
too large to form an effective tight ion-pair with N1.
0
1
2
Grignard reagents such as t-BuMgCl and MeMgCl were not effective bases (entries 20-21) providing for
3
4
5complete recovery of the heterocycle 1b. Compare this to the use of MeMgCl to provide selective N7-alkylpurine
6
7
8
9
products. Other weak bases such as Hünig’s base, Li
2
CO
3
, Na
2
CO
3
, and K
2
CO were ineffective in THF (entries 22,
3
9
2
7-29).
0
1
2
3
4
5
6
7
8
9
0
8
ACS Paragon Plus Environment

Page 9 of 61
The Journal of Organic Chemistry
Table 2. Regioselective N-Alkylation Optimizationa
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
b
c
b
c
1bb 2b/3bd
entry
base
solvent
temp/time
2b (yield) 3b (yield)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
1
2
NaOMe
NaHMDS
NaHMDS
NaHMDS
NaHMDS
NaHMDS
NaHMDS
dioxane
dioxane
dioxane
THF
80 ˚C/24h
80 ˚C/0.5h
20 ˚C/7h
20 ˚C/1h
20 ˚C/24h
20 ˚C/1h
20 ˚C/48h
20 ˚C/1h
20 ˚C/3h
20 ˚C/3h
20 ˚C/1h
20 ˚C/1h
20 ˚C/7h
20 ˚C/4h
20 ˚C/5h
20 ˚C/0.5h
20 ˚C/1h
20 ˚C/1h
20 ˚C/5h
20 ˚C/6h
50 ˚C/24h
66 ˚C/7h
66 ˚C/5h
20 ˚C/5min
8
8
72 (60)
82 (48)
77 (45)
89 (69)
84 (68)
22 (13)
29 (18)
55 (40)
0
5
10
5
1/9
1/10
1/7
1/8
1/5
3/1
2/1
1/1
--
3
11 (8)
11 (8)
16 (0)
73 (55)
65 (52)
44 (38)
0
4
0
5e
6
THF
1 (2)
5 (6)
6 (9)
0
DMSO
DMSO
7e
8
NaHMDS 10-1 THF/DMSO
8 9
NaHMDS
NaHMDS
NaHMDS
NaHMDS
LiHMDS
KHMDS
LiOt-Bu
toluene
acetonitrile
DMF
DMPU
THF
100
89
4
9
0 10
1
2 11
0
11
--
69
27
3/1
5/1
1/6
1/5
1/6
1/10
3
1
1
1
1
1
1
1
1
2
3
4
5
6
7
8
9
63
14
0
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
15
85
0
THF
17 (9)
14 (6)
9 (8)
62 (29)
70 (37)
11 (11)
0
81 (54)
86 (47)
89 (58)
2
THF
0
NaOt-Bu
NaOt-Bu
NaOt-Buf
KOt-Bu
THF
2
DMSO
DMSO
THF
25 (15) 12 (15) 2/1
30 (18)
0
3
2/1
1/8
--
84 (48)
9 20
t-BuMgCl
MeMgCl
THF
1
0
99
100
97
8
0
1 21
THF
0
--
2
3 22 Hünig’s base
THF
1
1
--
4
2
2
3
4
DBU
-Etg
THF
61
31
20
2/1
4/1
5
6
7
8
9
0
P
2
THF
80 (55)
9
0
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 61
2
2
2
2
2
3
3
5
6
7
8
9
0
1
P
P
2
2
-Etg
-Etg
THF
DMSO
THF
-78 ˚C/0.5h
60 (27)
85 (63)
0.4
36 (2)
4
0
2/1
6/1
--
20 ˚C/0.5h
65 ˚C/24h
66 ˚C/5h
66 ˚C/5h
66 ˚C/1h
20 ˚C/3h
15 (16)
0.6
6
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Li
2
CO
CO
CO
3
99
92
32
3
Na
2
3
THF
2
--
K
2
3
THF
13
53
1/4
1/1
4/1
Cs
Cs
2
CO
CO
3
3
THF
40
57
0
1
2
3
2
DMSO
77
18
5
4a
5
Standard reaction conditions: compound 1b (100 mg, 0.67 mmol) was dissolved or suspended in 3.3 mL of solvent at 0
C (except DMSO reactions were started at 15 ˚C), base (1 mmol) as a solution or solid (MHMDS and MOt-Bu with M
Li, Na, K; and t-BuMgCl were 1 or 2 M solutions in THF unless noted otherwise) was added followed by
˚
6
7
=
8iodomethane (0.12 mL, 2 mmol) and the mixture was stirred at the designated temperature for the designated time. Then
9
the mixture was diluted with CH
2
Cl
(3×). The combined organic extracts were dried (MgSO
residue. HPLC % composition determined from the crude product residue at 220 nm. Isolated % yield after subjecting
2
and aqueous 1 M NH
4
Cl, the organic layer separated and the aqueous layer
0
1
extracted with CH
2
Cl
2
4
) and evaporated to crude product
b
c
2
3the crude product residue to SiO
2
chromatography through a 12 g SiO column eluted with a gradient of 10-100% ethyl
2
4
d
e
f
acetate in heptane. Ratio of HPLC peak areas. Performed with 1.5 instead of 3 equivalents of MeI (1 mmol). Solid
5
g
NaOt-Bu was used. N'''-[bis(dimethylamino)(ethylimino)phosphoranyl]-N,N,N',N',N'',N''-hexamethyl-phosphorimidic
6
triamide.
7
8
9
0
1
2
3
With the systems NaHMDS/THF and NaHMDS/DMSO established to selectively produce either 2b or 3b upon
reaction of 1b with iodomethane, the scope of the alkylation of 1b was investigated with other electrophiles (Table 3).18
4
5
6
As expected, in THF, the alkylations generally occurred selectively on the N2 position with isolated yield ratios of
7
8N1/N2 (2/3) ranging from 1:2 (entry 9) to 1:17 (entry 2). Some THF reactions with hindered alkyl halide electrophiles,
9
0
entries 3, 4, 6, 7, and 10, were slow to react at room temperature (20 ˚C) and required higher temperatures (microwave
1
2
heating at 160 ˚C/0.75 h) to reach completion. Isolated yields of the major N2-products (3) from reactions conducted in
3
4
5
THF were mostly > 50%. With very reactive electrophiles, entries 9 and 11, N2 selectivity became weaker. In the
6
7
glycosylation reaction in THF (entry 11), no selectivity was observed and yet a higher yield of N2-product was
8
9
0
1
2
3
4
obtained (1:1 N1/N2 ratio) compared to previous reports.19
By comparison, reactions in DMSO, with all but two electrophiles (entries 6 and 10) showed N1-selectivity
(
N1/N2, 2/3 ratios of 2:1 up to 19:1, see Table 3). All DMSO reactions were performed at 20 ˚C. In general, yields of
5
6
the major products, 2, were satisfactory except for the challenging 4-iodotetrahydro-2H-pyran electrophile, entry 10.
7
8
9
0
10
ACS Paragon Plus Environment

Page 11 of 61
The Journal of Organic Chemistry
a
Table 3. Alkylation of 4-Methoxy-1H-pyrazolo[3,4-d]pyrimidine (1b)
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
b
THF/DMSO
in THFb
in DMSO ______
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7a
8
entry
R-X
Me-I
temp (˚C) time (h) N1 (%) N2 (%) N1/N2
N1 (%) N2 (%) N1/N2
1
2
3
4c
5
6
7
8
9
20/20
20/20
1/1
2b (8) 3b (69)
1/9
2b (55) 3b (13)
2c (67) 3c (13)
2d (41) 3d (9)
2e (37) 3e (11)
4/1
5/1
5/1
3/1
5/1
1/1
19/1
Et-I
24/0.5
2c (4) 3c (66) 1/17
i-Pr-I
160/20 0.75/1.5 2d (6) 3d (32)
160/20 0.75/144 2e (7) 3e (57)
48/144 2f (10) 3f (50)
1/5
1/8
1/5
1-iodo-2-methylpropane
(bromomethyl)cyclopropane
benzyl chloride
65/20
2f (42)
3f (9)
160/20 0.75/1
2g (4) 3g (56) 1/14
2g (40) 3g (41)
2h (57) 3h (3)
1-bromo-2-methoxyethane
3,3-dimethylallyl bromide
ethyl 2-chloroacetate
4-iodotetrahydro-2H-pyran
compound 7
160/20 0.75/20 2h (18) 3h (48)
1/3
1/14
1/2
20/20
20/20
2/3
2i (4)
3i (56)
2i (50)
2j (59) 3j (24)
2k (6) 3k (4)
3i (0) >10/1
16/3
2j (28) 3j (57)
2/1
1/1
2/1
1
0
160/20 0.75/19 2k (0) 3k (10) <1/10
20/20 4/1 2l (32) 3l (28) 1/1
ll (2)
1
1d
2l (24) 3l (12)
2ll (7)
2
b
See standard reaction conditions as described in Table 1 substituting R-X for methyl iodide. Chromatographically
isolated yields. Experiment in THF conducted with 3 mmol 1b. Only 1.5 equivalents of compound 7 was used.
c
d
9
0
Structures:
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
The N7-atom contributes key ligand-bonding interactions with sodium in the crystal structures 5a and 5b
6(Figure 2). To gauge the role of N7 in N2-alkylation site-selectivity in THF, 15 pyrazolo-fused rings with either CH or
7
8
9
0
11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 61
1
8
N at position 7 were compared (see Table 4). Heterocycles were chosen with additional substituents that would be of
interest in medicinal chemistry for exploring a variety of substitution vectors (see Figure 1). In entries 1-4, heterocycles
with CH instead of N in position 7 (1m and 1n), showed no N2-selectivity in THF. Introduction of N7 in similar
heterocycles 1o and 1p, entries 5-8, in THF showed 1:10 and 1:5 N1/N2 ratios, respectively; while in DMSO the
selectivity reversed to 15:1 and 9:1 N1/N2, respectively. Heterocycles with a 4-methoxy substituent, entries 9-16,
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
showed similar N1/N2 selectivity effects when comparing THF to DMSO which was dependent on the presence of N7.
2
3
Heterocycles having CH instead of N in position 7, 1q and 1r, provided N1-selective alkylation in both solvents. But
4
5heterocycles containing N7, 1s and 1b (vida supra) provided for N2-selectivity in THF, which again reversed to N1
6
7
when conducted in DMSO. Heterocycles containing a 4-chloro substituent had the same selectivity profile (entries 17-
8
9
2
2). Finally, the influence of steric hindrance from a 3-bromo substituent did little to redirect the THF alkylation away
0
1
2from N2 when N7 was present (entries 25 and 29) giving 1:2 N1/N2 alkylation selectivity while heterocycles having
3
4
5
CH in position 7 showed good N1-selectivity (4:1 to >10:1) in both THF and DMSO (entries 23, 24, 27 and 28). This
6
7
study emphasized the importance of N7 to achieving N2-selective alkylation in THF. When CH was present instead of
8
9
N in position 7, alkylation almost always favored N1 in both solvents. Moreover, all of these heterocycles favored N1-
0
1alkylation in DMSO.
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
12
ACS Paragon Plus Environment

Page 13 of 61
The Journal of Organic Chemistry
a
Table 4. Application to Other Pyrazolo-fused Ring Heterocycles
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
N1 (%)b
N2 (%)b
N1/N2
0 entry heterocycle time (h) solvent
1
2
3
4
5
6
1
2
1m
1m
96
1
THF
2m (49)
2m (45)
3m (31)
3m (17)
2/1
3/1
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
DMSO
3c
4
1n
1n
5
4
THF
2n (44)
2n (48)
3n (42)
3n (15)
1/1
3/1
DMSO
5
6
1o
1o
2
4
THF
2o (7)
3o (69)
3o (3)
1/10
15/1
DMSO
2o (46)
7
8
1p
1p
4
4
THF
2p (8)
3p (38)
3p (3)
1/5
9/1
DMSO
2p (27)
9
1q
1q
2
2
THF
2q (57)
2q (65)
3q (25)
3q (23)
2/1
3/1
2 10
DMSO
3
4
5
6
7
8
9
0
13
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 61
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
1
1
2
1r
1r
2
2
THF
2r (50)
2r (46)
3r (35)
3/2
3/1
DMSO
3r (18)
0
1
2
3
4
5
1
3
1s
1s
2
2
THF
2s (5)
3s (30)
3s (5)
1/6
6 14
DMSO
2s (67)
13/1
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
1
1
5
6
1b
1b
1
1
THF
2b (8)
3b (69)
3b (13)
1/8
4/1
DMSO
2b (55)
3 17
1t
1t
16
1
THF
2t (43)
2t (52)
3t (39)
3t (27)
1/1
2/1
4
5 18
DMSO
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
2
9
0
1u
1u
4
1
THF
2u (11)
2u (44)
3u (81)
3u (0)
1/8
DMSO
>10/1
2 21
1a
1a
1.5
1.5
THF
2a (6)
3a (36)
3a (11)
1/6
4/1
3
2
2
DMSO
2a (47)
4
5
6
7
8
9
0
14
ACS Paragon Plus Environment

Page 15 of 61
The Journal of Organic Chemistry
1
2
3
4
2
2
3
4
1v
1v
4
4
THF
2v (63)
2v (56)
3v (15)
4/1
8/1
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
DMSO
3v (7)
0
1
2
3
4
5
2
5
1w
1w
24
1
THF
2w (27)
2w (80)
3w (47)
3w (7)
1/2
6 26
DMSO
11/1
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
2
2
7
8
1x
1x
16
4
THF
2x (61)
2x (67)
3x (10)
3x (0)
6/1
DMSO
>10/1
3 29
1y
1y
16
16
THF
2y (25)
2y (66)
3y (43)
3y (0)
1/2
4
5 30
DMSO
>10/1
6
7
8a
b
See standard reaction conditions as described in Table 1. Chromatographically isolated yields.
9
c
0 See ref. 20a for methylation of 1n in THF.
1
2
3
9
4
5
6
In a recent report on selective N3-alkylation of imidazo-fused heterocycles using MeMgCl and THF, it was
observed that a steric component was necessary to drive the selectivity (see Scheme 1). For example, 4-chloro substrate
7
8
9
0
1
z exhibited complete N3-selectivity to product 4z while 4-H substrate 1a' was alkylated with no selectivity to products
12a' and 4a' (see Equations 3 and 4). Thus, it was of interest to evaluate whether N1 vs N3 alkylation site selectivity
2
3
4
could be influenced in imidazo-fused ring heterocycles when a proximal nitrogen atom (N7) was present, and
5
6
especially test whether selectivity would be independent of steric influence.
7
8
9
0
15
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 61
a
Scheme 1. Literature N3-Selective Imidazo-Fused Ring Alkylation
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
_
_____________________________________________________________________________
9
aSee ref. 9.
0
1
2
3
4
1
8
Table 5 shows the application of our method to imidazo-fused ring heterocycles. Entries 1-6 confirmed the
5N7-effect also applies to this ring type with heterocycles unsubstituted at C4. Alkylation of N3 is preferred with N7-
6
7
8
containing heterocycles in THF where heterocycles 1c' and 1d' provided 1:4 and 1:3 N1/N3-alkylation ratios,
9
0
respectively, with major products 4c' (54%) and 4d' (66%) thus adding a N3-selective alkylation to this structure type.
1
2
The N7 effect was further confirmed with 4-chloro substituted ring systems in entries 7-15. Heterocycles 1f' and 1h'
3
4
selectively provided 4f' (56%) and 4h' (55%) in THF reactions while heterocycle 1g' provided the lowest N1 to N3
5
6
selectivity switch going from DMSO (3:1) to THF (1:1). Of note, heterocycles 1e', 1f', 1g' and 1h' have been
7
8
9
0
1
2
3
demonstrated to selectively alkylate N3 with the MeMgCl/THF method.9
The reaction in DMSO favored the N1 products with heterocycles 1c', 1f', 1g' and 1h'. Also, similar to entry 8
2
1
in Table 2, with heterocycles 1e' and 1f', a 10:1 THF/DMSO solvent ratio provided a 1:1 N1/N3 ratio for product
4
5
6
pairs 2e'/4e' and 2f'/4f'.
7
8
9
0
1
2
3
4
5
6
7
8
9
0
16
ACS Paragon Plus Environment

Page 17 of 61
The Journal of Organic Chemistry
Table 5. Application to Imidazo-fused Ring Heterocyclesa
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0 entry
heterocycle
time (h) solvent
N1 (%)b
N3 (%)b
N1/N3
1
2
3
4
5
6
1
1b'
1b'
2
2
THF
2b' (52)
2b' (33)
4b' (52)
4b' (33)
1/1
1/1
7
8
2
9
DMSO
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
3
4
1c'
1c'
2
2
THF
2c' (14)
2c' (37)
4c' (54)
4c' (6)
1/4
6/1
DMSO
5
6
1d'
1d'
1.5
1.5
THF
2d' (19)
2d' (46)
4d' (66)
4d' (37)
1/3
1/1
DMSO
7c
8
1e'
1e'
3
3
THF/DMSO
DMSO
2e' (42)
2e' (28)
4e' (32)
4e' (24)
1/1
1/1
9
1f'
1f'
1f'
2
THF
2f' (18)
2f' (33)
2f' (34)
4f' (56)
4f' (29)
4f' (14)
1/3
1/1
2/1
1
0c
1.5 THF/DMSO
96 DMSO
6 11
7
8
9
0
17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 61
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
1
2
3
1g'
1g'
5
5
THF
2g' (33)
2g' (40)
4g' (42)
1/1
3/1
DMSO
4g' (15)
0
1
2
3
4
5
1
4
1h'
1h'
1.5
1.5
THF
2h' (25)
2h' (41)
4h' (55)
4h' (23)
1/2
2/1
6 15
DMSO
7
8
9
a
b
See standard reaction conditions as described in Table 1. Chromatographically isolated yields.
1 Conducted in 10:1 THF/DMSO.
0
c
2
3
4
5
6
7
2
2
Finally, very little study has been made of the alkylation of triazolo-fused ring heterocycles. To conclude this
8study several triazolo-fused ring alkylations were examined in THF and DMSO and a small effect was observed
9
0
1
18
favoring N3-alkylation in THF when an N7 nitrogen atom was present (Table 6). Alkylation of heterocycles 1j' and
2
3
4
1
l' in THF favorably produced N3-alkyl products 4j' (43%) and 4l' (24%), respectively, with 1:1:2 N1/N2/N3
5alkylation ratios in both cases (entries 3 and 6). In DMSO, those heterocycles gave statistically distributed 1:1:1
6
7
8
9
0
N1/N2/N3 product ratios (entries 4 and 7). Triazolo-fused ring heterocycles having CH instead of the N7 atom gave
different product distributions (entries 1, 2 and 5).
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
18
ACS Paragon Plus Environment

Page 19 of 61
The Journal of Organic Chemistry
Table 6. Application to Triazolo-fused Ring Heterocyclesa
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
entry heterocycle
time (h)
solvent
N1 (%)b
N2 (%)b
N3 (%)b
N1/N2/N3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7a
1c
2
1i'
1i'
24
2
THF
2i' (45)
2i' (29)
3i' (17)
3i' (25)
4i' (18)
4i' (15)
3/1/1
2/2/1
DMSO
3c
4
1j'
1j'
48
4
THF
2j' (19)
2j' (24)
3j' (17)
3j' (25)
4j' (43)
4j' (20)
1/1/2
1/1/1
DMSO
5d
1k'
4
THF
2k' (25)
3k' (26)
4k' (25)
1/1/1
6
7
1l'
1l'
4
THF
2l' (11)
2l' (19)
3l' (14)
3l' (12)
4l' (24)
4l' (14)
1/1/2
1/1/1
24
DMSO
b
c
See standard reaction conditions as described in Table 1. Chromatographically isolated yields. Product structures
confirmed by X-ray crystallography, see Supporting Information. See ref. 20a.
8
9
d
0
1
2
3
4
5
6
7
8
9
0
MECHANISM
19
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 61
Solvent polarity is known to have a dramatic effect on the regioselectivity of reactions in which the mechanism
involves close ion pairs (CIPs) or solvent separated ion pairs (SIP).23, 24, 25 In order to probe whether ion pairs play a
role in the selectivities observed for alkylation reactions of azolo-fused ring heterocycles, especially those containing
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
*
*
N7 instead of CH at that position, computational analysis of anion charge density was performed on 1b and 1r (Table
7). Comparative analysis of calculated charge density using density functional theory (DFT) calculations at the
0
1
**
**
B3LYP/6-31G**(d) level of theory for these anions (as well as tautomer anions 1b and 1r ) show charge density is
2
3
2
6
focused more on N1 than on N2 by about 0.2 charge units for each. Importantly, the DFT calculations indicate that
4
5substitution of N for CH at position 7 (variable Y in Table 7) does not influence charge density concentration at N1 in
6
both compounds, thus the N1-anion should be more reactive in each compound especially in a polar ion solvating
1
2methylation reactions of 1b and 1r conducted in DMSO with N1/N2 ratios of 4:1 and 3:1 (Table 4, entries 16 and 12,
2
3
6
nonselective for 1r (Table 4, entries 15 and 11, respectively), which indicates a separate THF solvent effect that is
7
8
much more enhanced for molecules containing N7.
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
20
ACS Paragon Plus Environment

Page 21 of 61
The Journal of Organic Chemistry
a
Table 7. Density functional theory (DFT) charge distribution (vacuum).
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
charge
_
______________________
entry
anion
N1
N2
N1-N2
_
______________________________________
7
8
9
0
1
2
3
4
1
2
3
4
1b*
1b**
1r*
-0.642
-0.571
-0.588
-0.587
-0.418
-0.395
-0.394
-0.394
-0.224
-0.176
-0.194
-0.193
1r**
_
______________________________________
5
6a
7
8
9
0
Calculations at the B3LYP/6-31G**(d) level of theory with the net charge set to -1.
To better understand the specific role of THF in reversing N1/N2 alkylation selectivities with azolo-fused ring
1
1
2heterocycles containing N7, the H-detected pulsed field gradient-stimulated echo (PFG-STE) NMR method was
3
4
applied to examine the solution N-sodio species of these two compounds and their neutral counterparts in THF-d and
8
5
6
DMSO-d
6
.27, 28, 29, 30 The method efficiently measures the diffusion coefficient D
t
of molecules in solution. Since LogD
t
7
8
9relates directly to LogMW, this MW information can be used to interpret aggregation and solvation states of
0
1
2
31
molecules which in turn can be applied to mechanistic reasoning for the reactivity difference. Example applications
3
4
on N-lithio-complexes29, 31b, 31d indicate its potential for providing insight into how solution aggregates relate to reactive
5
6intermediates.
7
8
Each PFG-STE NMR experiment was conducted in THF-d
8
or DMSO-d solutions containing naphthalene
6
9
0
1analogs 7a, 7b and 7c as internal MW standards. The naphthalenes, because of their planar structural similarity to 1b
2
3
4
29
and 1r and their chemical inertness, served as suitable standards to form internal calibration curves (ICCs). For each
5
compound examined, the MW of its solution species aggregate (MWdet) as a NMR time average form28c was determined
6
7
8
9
0
21
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 61
by plotting LogDt,obs to respective ICCs, (see Figure 3 for an example analysis). Table 8 lists the results of this analysis
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with proposed time average solution species having calculated MW (MWcalc) values that agree to within 10% or better
(
MWerr) with the MWdet of the test compound in solution.31
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
Figure 3. H-detected pulsed field gradient-stimulated echo (PFG-STE) NMR MWdet analysis of N-sodio form of
1
2
3
*
heterocycle 1b in THF-d based on internal calibration curve (ICC) with standards 7a, 7b and 7c. ICC: y = -
8
4
2
*
*
50.3800x – 7.8670; R = 1.0000. See Supporting Information for all PFG-STE analysis of 1b, 1r, 1b and 1r in THF-d
8
6
7
8
and DMSO-d .
6
9
0
1
2
3
4
5
6
7
8
9
0
22
ACS Paragon Plus Environment

Page 23 of 61
The Journal of Organic Chemistry
Table 8. Solution species MWdet from PFG-STE NMR determined diffusion coefficients Dt, obs of 1b, 1r, 1b*
*
and DMSO-d
.a
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
and 1r measured in THF-d
8
6
0
1
2
3
4
D
t,obs
-9
MWdet
time average
sol'n species
MWcalc MWerr
(g/mol) (%)
b
2
5 entry heterocycle solvent (10 m /s) (g/mol)
6
7
_
___________________________________________________________________________________
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3a
4
5
1
2
3
4
5
6
7
8
1b
1r
THF
THF
1.8510
2.4405
187
195
246
239
209
404
1b(THF)0.5
1r(THF)0.5
1b(DMSO)
1r(DMSO)
186
185
228
227
208
387
-1
+5
+10
+5
+0.5
+4
0
1b
DMSO
DMSO
THF
0.36733
0.38243
1.7840
1r
1b*
1r*
1b*
1r*
1b Na(THF)0.5
*
*
THF
1.5633
1r Na(THF)
3
*
DMSO
DMSO
0.36170
0.37283
278 1b (Na)0.5(DMSO)1.5 278
*
260 1r (Na)0.5(DMSO)1.5
277
-7
1
Values Dt,obs for compounds are the average of at least 2 determinations from separate resolvable H resonances for
4% of samples; 89% of Dt, obs SD values were < 5%. See Supporting Information for all curves and data. For Dt,obs
based MWdet calculations, the protio-form of the NMR solvents are used since it was previously determined that
7deuterium containing solvents have MWdet values that agree with their protio-form MWs, see ref. 31d.
b
9
6
8
9
0
1
2
3
PFG-STE NMR analysis of the neutral compounds 1b and 1r revealed similar solution solvation states for the
4compound pair in THF and DMSO. In THF 1b and 1r are in equilibrium between desolvated and mono-THF solvated
5
6
7
8
9
0
1
2
3
molecular aggregates (Table 8, entries 1-2). While in DMSO, the compounds are each associated with one molecule of
solvent (entries 3-4).32
*
*
Remarkably, in THF the ionic compounds 1b and 1r exhibit starkly different aggregate forms with the former
having MWdet = 209 g/mol and the latter with MWdet = 404 (Table 8, entries 5-6). These values agree with time average
4
5
*
*
.33, 34 These proposed aggregates support a mechanistic explanation
23
6solution species 1b Na0.5(THF)0.5 and 1r Na(THF)
3
7
8
9
ACS Paragon Plus Environment
0

The Journal of Organic Chemistry
Page 24 of 61
*
*
for why 1b in THF is more selective for N2-alkylation (1:8 N1/N2) compared to 1r (3:2 N1/N2) and by extension
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
how other N7 containing azolo-fused ring heterocycles are selective for N2 or N3 (Tables 4-6). As described in Scheme
*
2
, Equation 5, the N7 containing heterocycle 1b in THF exists as an aggregate with MWdet = 209 representing the 1:1
*
*
equilibrium of the close ion pair (CIP) 1b Na(THF) (MW = 244) with the desolvated tight ion pair (TIP) 1b Na (MW =
3
5
+
172)
.
This lack of solvent to coordinate the Na ion in the TIP enforces a tighter N1-Na bond because of additional
0
1
through-space ligand bonding of N7 to Na and/or simply because N7 contributes, within the heterocycle bonding
2
3
network, a through-bond electronic force to stabilize the N1-Na interaction. The resulting TIP N1-Na bond does not
4
5separate enough to allow N1 alkylation and instead N2 is alkylated. Additionally, invoking a N7-Na ligand bond allows
6
7
8
rationalization for the formation of complex 5 containing the same bonding as observed by X-ray crystallography of the
9
0
*
isolated product (see 5a in Figure 2). The contrasting alkylation behavior of 1r is understandable considering the much
1
*
.34 The extra coordinating solvent likely contributes to a small degree of ion
2different aggregate in THF, 1r Na(THF)
3
3
4
5
*
pair separation and increased reactivity of the resulting SIP form 1r (THF)
n
, providing little discrimination between
6
alkylation of N1 vs. N2 (3:2 N1/N2).
7
8
9
0
1
*
*
PFG-STE NMR of the ionic species 1b and 1r in DMSO-d
6
produced similar MWdet values of 278 and 260
g/mol (Table 8, entries 7-8), respectively. These values are consistent with the same mechanism being in effect in
2
3
*
4DMSO regardless of the presence of N7. The MWdet values describe time-averaged ion pairs 1b (Na)0.5(DMSO)1.5 and
5
6
7
1
r*(Na)0.5(DMSO)1.5 that represent 1:1 equilibrium mixtures of CIP/SIP forms (Scheme 2, Equations 7b and 7r). The
8
9
*
*
stable CIP forms 1b Na(DMSO) and 1r Na(DMSO) each accept another molecule of DMSO which aids in the
0
1
+
*
*
formation of SIPs containing solvated Na ions separate from 1b (DMSO)
2
and 1r (DMSO)
2
, respectively. Then as
2
2
3, 24
3predicted by DFT calculations (Table 7) and in agreement with ion pair theory
the SIP anions alkylate primarily at
4
5
6
the nitrogen with the most negative charge, N1, producing products 2b and 2r (4:1 2b/3b and 3:1 2r/3r).
7
8
9
0
1
2
3
4
5
6
7
8
9
0
24
ACS Paragon Plus Environment

Page 25 of 61
The Journal of Organic Chemistry
Scheme 2. Ion Pair Theory Mechanisms for Solvent Controlled Site-Selective Pyrazolo-Fused Ring Heterocycle
Alkylationsa
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7________________________________________________________________________________________________
8
9
aMolecular weights (MW) are expressed in units of g/mol.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
25
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 61
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
CONCLUSION
Understanding and controlling azolo-heterocycle alkylation site-selectivity is critical to many medicinal
chemistry programs. This study has demonstrated that in THF, pyrazolo-, imidazo- and triazolo-fused ring heterocycles
that contain N7 will preferentially alkylate at N2 (for pyrazolo-) or N3 (for imidazo- and triazolo-) in good to excellent
0
selectivity. When conducted in DMSO, high selective alkylation of N1 is observed with pyrazolo- and imidazo-fused
5
6N1-selectivity in DMSO. However, imidazo-fused ring heterocycles lacking N7 show no selectivity in DMSO. In
7
DMSO, alkylation of triazolo-fused ring heterocycles is not selective with or without N7. Finally, equal alkylation
2
3THF/DMSO solvent mixtures.
4
5
6
X-ray crystal structures of sodium iodide complexes 5a and 5b and DFT calculations coupled with PFG-STE
7
8NMR MW analysis of the ionic reactive intermediates in solution support ion pair mechanistic interpretations.
9
0
1
*
Interestingly, these data suggest that in THF, N7 contributes to a tight N1-Na bond in the intermediate TIP 1b (Na)
(
Equation 5); which, as applied to pyrazolo-, imidazo- and triazolo-fused heterocycles, blocks N1 alkylation and forces
2
3
6
7
8
solvent separated anions with sodium cations (SIPs) make up less than 1% of the ionic species when compared to polar
9
0
solvents such as HMPA or DMSO and that CIPs or TIPs predominate in THF. In further agreement with ion-pair
1
2
theory, the inability to form SIPs in THF led to decreased reactivity as witnessed in alkylations with bulky electrophiles
3
4that required higher temperatures to go to completion as compared to DMSO (Table 3).
5
6
7
8
9
This work has uncovered a new principle of alkylation site control in azolo-fused ring heterocycles containing
the N7 atom. Reactions performed in THF will predictably alkylate preferentially at N2 or N3 and this is expected to
0
1
provide new opportunities for drug discovery projects to examine substituent effects into vector directions not
2
3
previously recognized as being accessible from such a simple reaction.
4
5
6
7
8
9
0
26
ACS Paragon Plus Environment

Page 27 of 61
EXPERIMENTAL SECTION
The Journal of Organic Chemistry
1
2
General Information. All starting materials and reagents were purchased from commercial sources and used as
received unless otherwise noted. Microwave reactions were performed with a single mode operating Biotage Initiator+
in sealed reaction vials at the indicated fixed temperature held constant for the designated reaction time. The
temperature within the microwave reaction vial was monitored by an IR sensor focused on a point on the reaction vial
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
glass. Melting points were obtained on a Electrothermal IA-9100 Melting Point Apparatus and are uncorrected. The
2
3
4
5
instrument was validated by melting a sample of 1-naphthoic acid, mp 160-161 ˚C (vendor reported mp 157-160 ˚C).
1
13
6
7
NMR spectra were recorded on 400 MHz and 600 MHz instruments. All H and C NMR data were referenced to the
8
9
internal deuterated solvent relative to TMS at 0 ppm. Analytical HPLC was performed with a C18 column (3 m, 3 × 50
0
1mm) and the solvent system A: pH 4.5 aqueous buffer (20 mM KH
2
PO ) and B: acetonitrile; eluting 0% B (2 min), 0-
4
2
3
4
2
0% B (2.5 min), 20-75% B (3.5 min), 75% B (1 min) with a flow rate of 1.5 mL/min. Analytical LC/MS was
5
6
performed on a UPLC-MS instrument equipped with a C18 column (1.7 m, 2.1 × 50 mm) and the solvent system A:
7
80.1% HCOOH in H
2
O and B: 0.1% HCOOH in acetonitrile, eluting 5-95% solvent B (0.75 min) with a flow rate of 1
9
0
1
mL/min at a column temperature of 45 ºC. Flash column chromatography (FCC) was performed on prepacked silica
2
3
4
(
SiO
2
) columns eluted with the indicated solvents. Preparative HPLC Method 1: on a 30 mm × 100 mm × 2.5 µm
(
particle size) C18 column with a gradient of acetonitrile (B) in water (A) (15 min) and 0.05% trifluoroacetic acid added
5
6
7
8
as a modifier to both phases with a flow rate of 60 mL/min. Elution profiles were monitored by UV at 254 and 220 nm.
Preparative HPLC Method 2: on a 30 mm × 100 mm × 5 µm (particle size) C18 column and the solvent system with a 9
3
4flow rate of 60 mL/min. Fractions collected from preparative HPLC were diluted with aqueous 1 M Na
2
CO and
3
5
6
7
extracted with CH
2
Cl
2
(3 × 50 mL); organic extract dried (MgSO ) and evaporated to product residue. The standard
4
8
9
crystallization method to prepare samples for mp determination, microanalysis or X-ray crystallographic structure
0
1determination, unless specified otherwise, was to dissolve 50 mg of sample in 0.5 mL THF in a 4 mL vial, place this
2
3
4
inside a 20 mL vial containing 5 mL pentane, cap the outer vial and after standing at rt for 24-72 h, the resulting
5
6
crystals were collected. C, H, I, N, and Na microanalysis were performed at a vendor where %I and %Cl were
7
8
9
0
27
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 61
determined by ion chromatography and %Na was determined with ICP-OES. Single crystal X-ray diffraction studies
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
were carried out on a Bruker Kappa APEX-II CCD diffractometer equipped with Mo K radiation (λ = 0.71073 Å).
α
4-Methoxy-1-methyl-1H-pyrazolo[3,4-d]pyrimidine (2b) and 4-Methoxy-2-methyl-2H-pyrazolo[3,4-
d]pyrimidine (3b). General Method 1. Methylation of Heterocycles in THF and DMSO. To 4-methoxy-1H-
0
1
2
3
pyrazolo[3,4-d]pyrimidine (1b) (100 mg, 0.67 mmol) suspended or dissolved, respectively, in 3.3 mL of THF (at 0 ˚C)
4
5or DMSO (at 15 ˚C), was added 1 mL of a 1 M solution of NaHMDS in THF (1 mmol) followed by iodomethane (0.12
6
7
mL, 2 mmol) and each mixture stirred at 20 ºC for 1h. Then the two reaction mixtures were separately diluted with
8
9
CH
2
Cl
2
and aqueous 1 M NH
4
Cl, the organic layers separated and the aqueous layers extracted with CH
2
Cl
2
(3×15
0
1
2mL). The combined organic extracts from each experiment were dried (MgSO
4
) and concentrated to crude product
3
4
5
residues. Purifications (FCC, SiO , 10-100% EtOAc/heptane) first afforded compound 2b as a white solid (THF: 9 mg,
2
6
1
13
) were identical to that reported;10 1H NMR (400
8
% yield; DMSO: 60 mg, 55% yield): H and C NMR (DMSO-d
6
7
8
MHz, CDCl
3
) δ 8.54 (s, 1H), 8.00 (s, 1H), 4.14 (s, 3H), 4.08 (s, 3H); 2D NOESY NMR (400 MHz, CDCl ) showed no
3
9
0
1relevant crosspeaks. X-ray crystallography confirmed the structure; see CCDC 1825997.
2
3
4
Continued elution provided compound 3b as a white solid (THF: 75 mg, 69% yield; DMSO: 14 mg, 13% yield):
5
67 mp 177-180 ˚C (lit. mp 166-169 ˚C);10 1H and C NMR (DMSO-d
) were identical to that reported; HSQC (DMSO-
C-H); 4.07 to 53.9 (OCH
); H NMR (400 MHz, CDCl ) δ 8.54
) δ 165.0, 161.9, 155.6, 123.6, 103.5, 54.0,
) showed a crosspeak from N2-CH (δ 4.05) to H3 (δ 7.94); 1H NMR (400
) δ 8.43 (s, 1H), 8.20 (s, 1H), 4.14 (s, 3H), 4.06 (s, 3H); 13C NMR (101MHz, THF-d ) δ 165.9, 163.1,
55.6, 124.5, 104.2, 54.1, 40.9; HSQC (THF-d ) cross peaks of δ 8.25-123 (H3-C3); 8.48 to 154.5 (H6-C6); 4.18 to
13
10
6
8
9
0
d
6
) showed cross peaks of δ 8.61-125.1 (H3-C3); 8.52 to 154.6 (H6-C6); 4.15 to 40.5 (NCH
3
3
1
2
1
C-H); NOESY (DMSO-d
6
) showed cross peaks from δ 8.61 to 4.15 (H3 to NCH
3
3
3
4
5
1
3
(
s, 1H), 7.94 (s, 1H), 4.14 (s, 3H), 4.05 (s, 3H); C NMR (101 MHz, CDCl
3
640.9; 2D NOESY NMR (400 MHz, CDCl
3
3
7
8
MHz, THF-d
8
8
9
0
1
2
1
8
1
339.5 (NCH
3
C-H); 4.10 to 53.0 (OCH
3
C-H); NOESY (THF-d
8
) cross peaks of δ 4.18 to 8.24 (NCH to H3); H NMR
3
4
5
6
7
8
9
0
28
ACS Paragon Plus Environment

Page 29 of 61
The Journal of Organic Chemistry
) δ 8.51 (s, 1H), 8.39 (s, 1H), 4.20 (s, 3H), 4.15 (s, 3H); X-ray crystallography confirmed the
(
400 MHz, methanol-d
4
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
structure; see CCDC 1825993.
4
-Chloro-1-methyl-1H-pyrazolo[3,4-d]pyrimidine (2a) and 4-Chloro-2-methyl-2H-pyrazolo[3,4-
d]pyrimidine (3a). Prepared from 4-chloro-1H-pyrazolo[3,4-d]pyrimidine (1a) (100 mg, 0.64 mmol) and iodomethane
as described in General Method 1. THF and DMSO conditions: 20 ºC/1.5 h. Purification (FCC, SiO , 10-100%
2
0
1
2
1
EtOAc/heptane) first afforded compound 2a as a white solid (THF: 7 mg, 6% yield; DMSO: 51 mg, 47% yield): H and
13C NMR (DMSO-d
) were identical to that reported.36
6
6
3
) δ 8.85 (s, 1H), 8.16 (s, 1H), 4.31 (s, 3H);
1
3
3
) showed a crosspeak from N2-CH (δ 4.31) to H3 (δ 8.16); C NMR (101 MHz,
3
3
4
5
CDCl ) δ 160.0, 156.2, 154.8, 124.9, 113.4, 41.6.
3
6
7
8
9
1
-Ethyl-4-methoxy-1H-pyrazolo[3,4-d]pyrimidine (2c) and 2-Ethyl-4-methoxy -2H-pyrazolo[3,4-
0d]pyrimidine (3c). Prepared from 4-methoxy-1H-pyrazolo[3,4-d]pyrimidine (1b) and 2-iodoethane (0.16 mL, 2 mmol)
1
as described in General Method 1. THF conditions: 20 ºC/24 h. DMSO conditions: 20 ºC/0.5 h. Purification (FCC,
2
, 10-100% EtOAc/heptane) first afforded first afforded compound 2c as a white solid (THF: 5 mg, 4% yield;
6
1
7DMSO: 80 mg, 67% yield): H NMR (400 MHz, CDCl
3
) δ 1.50 (t, J=7.21 Hz, 3 H) 4.13 (s, 3 H) 4.48 (q, J=7.34 Hz, 2
) δ 164.0, 154.9, 154.1, 130.9, 102.8, 54.1,
O: C, 53.92; H, 5.66; N, 31.44. Found: C, 53.99; H, 5.35; N, 31.53.
8
9
0
1
3
H) 7.28 (s, 1 H) 8.00 (s, 1 H) 8.53 (s, 1 H); C NMR (101 MHz, CDCl
3
1
2
4
2.5, 14.9. Anal. calcd for C
8
H
10
N
4
3
4
5
6
71
Continued elution provided compound 3c as a white solid (THF: 78 mg, 66% yield; DMSO: 16 mg, 13% yield):
H NMR (400 MHz, CDCl ) δ 1.65 (t, J=7.34 Hz, 3 H) 4.14 (s, 3 H) 4.47 (q, J=7.34 Hz, 2 H) 5.31 (s, 1 H) 7.29 (s, 1
) showed a crosspeak from N2-CH (δ 4.47) to H3
) δ 165.0, 161.5, 155.3, 122.0, 103.0, 53.9, 49.0, 15.3. Anal. calcd for C O:
3
8
9
0
H) 8.01 (s, 1 H) 8.63 (s, 1 H); 2D NOESY NMR (400 MHz, CDCl
3
2
1
1
3
(
δ 8.01); C NMR (101 MHz, CDCl
3
8
H
10
N
4
2
3
4
5
C, 53.92; H, 5.66; N, 31.44. Found: C, 54.22; H, 5.61; N, 31.14.
6
7
8
9
0
29
ACS Paragon Plus Environment