close

Вход

Забыли?

вход по аккаунту

?

cctc.201700918

код для вставкиСкачать
DOI: 10.1002/cctc.201700918
Full Papers
Pd-Catalyzed versus Uncatalyzed, PhI(OAc)2-Mediated
Cyclization Reactions of N6-([1,1?-Biaryl]-2-yl)Adenine
Nucleosides
Sakilam Satishkumar,[a] Suresh Poudapally,[b] Prasanna K. Vuram,[a] Venkateshwarlu Gurram,[b]
Narender Pottabathini,[b] Dellamol Sebastian,[a, c] Lijia Yang,[a] Padmanava Pradhan,[a] and
Mahesh K. Lakshman*[a, c]
In this work we have assessed reactions of N6-([1,1?-biaryl]-2yl)adenine nucleosides with Pd(OAc)2 and PhI(OAc)2, via a PdII/
PdIV redox cycle. The substrates are readily obtained by Pd/
Xantphos-catalyzed reaction of adenine nucleosides with 2bromo-1,1?-biaryls. In PhMe, the N6-biarylyl nucleosides gave
C6-carbazolyl nucleoside analogues by C N bond formation
with the exocyclic N6 nitrogen atom. In the solvent screening
for the Pd-catalyzed reactions, an uncatalyzed process was
found to be operational. It was observed that the carbazolyl
products could also be obtained in the absence of a metal catalyst by reaction with PhI(OAc)2 in 1,1,1,3,3,3-hexafluoroisopropanol (HFIP). Thus, under Pd catalysis and in HFIP, reactions
proceed to provide carbazolyl nucleoside analogues, with
some differences. If reactions of N6-biarylyl nucleoside substrates were conducted in MeCN, formation of aryl benzimidazopurinyl nucleoside derivatives was observed in many cases
by C N bond formation with the N1 ring nitrogen atom of the
purine (carbazole and benzimidazole isomers are readily separated by chromatography). Whereas PdII/PdIV redox is responsible for carbazole formation under the metal-catalyzed conditions, in HFIP and MeCN radical cations and/or nitrenium ions
can be intermediates. An extensive set of radical inhibition
experiments was conducted and the data are presented.
Introduction
Nucleosides are an important class of biomolecules, modifications of which have the potential for manifold applications.
Most notably, modified nucleosides are central in the control
and treatment of viral diseases, and importantly, in the contemporary contexts of emerging viral diseases and bioterrorism
threats.[1] Modified nucleosides have also impacted cancer therapeutics, hematological malignancies, and solid tumors,[2, 3] and
are important biological probes.[4] For these reasons and as yet
undiscovered applications, studies of methods to rapidly
modify the nucleoside structural scaffold are anticipated to
remain a critical task. In one aspect of our studies, which also
aim at understanding new reactivities of nucleoside substrates,
we are interested in involving the purinyl nitrogen atoms in
chemical transformations. For example, we have studied metalcatalyzed N-directed C H bond activation procedures towards
novel nucleoside motifs,[5, 6] leading to compounds that are
otherwise not readily attainable.
Hypervalent iodine chemistry has become an important
field,[7?11] because reagents are generally readily accessible. For
purine and purine nucleoside modification, the combination of
PhI(OAc)2/Cu(OTf)2 in 1:1 AcOH/Ac2O, at 80 8C has been utilized
for the construction of C N bonds in N6-aryl adenine and adenosine triacetates, leading to benzimidazolyl purines and
purine nucleoside analogues.[12] Both precursors for this
chemistry are relatively stable to the acidic reaction conditions
at elevated temperature, and the acetate protecting groups of
the ribonucleosides further augment glycosidic stability. Thus,
the substrates are less prone to acidic deglycosylation than
their more sensitive 2?-deoxyribosyl counterparts. With regard
to this point, the t1/2 for the deglycosylation of adenosine is 11
days in 0.1 M HCl (37 8C), whereas that for 2?-deoxadenosine is
15 min under the same conditions.[13] In agreement with these
data, in recent work, we observed extensive degradation of a
3?,5?-di-O-silyl N6-phenyl 2?-deoxyribosyl substrate under the reported conditions.[14]
Whereas benzimidazolyl nucleoside analogues are available
via such C N bond-forming reactions with a purinyl ring nitrogen,[12, 14] formation of carbazolyl nucleoside analogues by invoking reaction with the exocyclic nitrogen atom is as yet un-
[a] Dr. S. Satishkumar, Dr. P. K. Vuram, D. Sebastian, Dr. L. Yang, Dr. P. Pradhan,
Prof. M. K. Lakshman
Department of Chemistry
The City College of New York
160 Convent Avenue, New York, NY 10031 (USA)
E-mail: mlakshman@ccny.cuny.edu
[b] S. Poudapally, Dr. V. Gurram, Dr. N. Pottabathini
Discovery Services, GVK Biosciences Pvt. Ltd.
28A IDA Nacharam, Hyderabad 500076, Telangana (India)
[c] D. Sebastian, Prof. M. K. Lakshman
The Ph.D. Program in Chemistry
The Graduate Center of the City University of New York
New York, NY 10016 (USA)
Supporting Information and the ORCID identification number(s) for the
author(s) of this article can be found under:
https://doi.org/10.1002/cctc.201700918.
ChemCatChem 2017, 9, 1 ? 13
These are not the final page numbers! оо
1
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers
known. This can be envisioned by use of N6-([1,1?-biaryl]-2-yl)adenine nucleosides as precursors. However, these substrates
present an interesting case study. Here two C N bond formations can be envisaged, involving either the exocyclic N6 or the
embedded purinyl N1 atoms (Scheme 1), leading either to
carbazole or to benzimidazole formation.
Table 1. Synthesis of precursors via Pd-mediated C N bond formation.[a]
Scheme 1. Two C N bond forming modes in N6-([1,1?-biaryl]-2-yl)adenine
nucleosides.
Relevant to this work, Pd- and/or Cu- as well as Pt-mediated
dehydrogenative cyclizations of o-aminobiaryls to carbazoles
have been accomplished.[16?18] In a complementary fashion,
metal-free, IIII reagent-mediated carbazole formation from 2acetamidobiaryls has also been reported.[19] Independently, cyclization of 2-sulfonamido and 2-acetamidobiaryls to carbazoles
under both Cu-catalyzed and metal-free reactions with PhI(OTFA)2 as oxidant were demonstrated.[20] One consideration for us
was the presence of a free secondary amino group in the precursors (Scheme 1). However, such amino groups have not deterred Pd-mediated carbazole formation with PhI(OAc)2 as oxidant,[16] and both primary and secondary amino groups have
not been problematic in PhI(OAc)2- and PhI(OTFA)2-mediated
cyclizations.[21] In the present work we describe reactions of N6biarylyl adenine nucleosides with hypervalent IIII reagents primarily leading to carbazolyl nucleoside analogues. These are
non-trivial substrates for such reactions because they generally
possess complex reactivity patterns and contain multiple coordinating Lewis basic, nitrogen and oxygen atoms, and a labile
glycosidic linkage.
Results and Discussion
A series of 2?-deoxyribosyl and ribosyl nucleoside analogues 3?
13 were synthesized by Pd-catalyzed C N bond formation between the O-t-BuMe2Si-protected adenine nucleosides and obromobiaryls (Table 1).[22] It should be noted that aryl amination reactions with nucleosides as the amine component are
not readily accomplished, and the present data further demonstrate the unique utility of the Pd/Xantphos system for this
transformation.
Products in Table 1 were then utilized for the remaining investigations on the formation of C6 carbazolyl nucleosides. We
chose to explore the use of Pd(OAc)2/oxidant combinations for
the conversion of substrate 3 to carbazole 14. This was partially prompted by our observation that even catalytic amounts of
Cu(OTf)2 are detrimental to nucleosides.[14] These results are
summarized in Table 2. NaBO3З4 H2O, Cu(OAc)2, and t-BuOOH
&
ChemCatChem 2017, 9, 1 ? 13
www.chemcatchem.org
Entry
Substrate
Aryl bromide
Product: yield
1
1
3: 88 %
2
1
4: 81 %
3
1
5: 75 %
4
1
6: 69 %
5
1
7: 73 %
6
1
8: 72 %
7
1
9: 89 %
8
2
10: 67 %
9
2
11: 76 %
10
2
12: 73 %
11
2
13: 68 %
[a] Yields are of isolated, purified products.
proved to be generally ineffective for the conversion (entries 1?4). However, the use of PhI(OAc)2 was more promising
in PhMe (entries 5?7) and a good product yield of 69 % was attained at 55 8C (entry 5). Lowering the Pd(OAc)2 loading from
20 mol % to 5 or 10 mol % led to incomplete reactions (> 50 %
and ca. 15?20 % of starting material was present in these
cases, respectively) and a decrease in product yield (entries 7
and 8). Conducting the reaction at room temperature with
20 mol % of Pd(OAc)2 led to reasonable product formation, but
the reaction remained incomplete (entry 9). PhI(OTFA)2 and
2
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
нн These are not the final page numbers!
Full Papers
Table 2. Summary of optimization experiments for the dehydrogenative C N bond formation using Pd(OAc)2.[a]
Entry
Conditions
Observations
1
2
3[c]
4
5
6
7[f]
8[g]
9
10
11
NaBO3З4 H2O (4 equiv), MeCN, 60 8C, 24 h
NaBO3З4 H2O (4 equiv), PhMe, 100 8C, 24 h
Cu(OAc)2 (1.3 equiv), O2 (1 atm), PhMe, 100 8C, 24 h
t-BuOOH (1.3 equiv),[e] PhMe, 100 8C, 24 h
PhI(OAc)2 (1.3 equiv), PhMe, 55 8C, 3 h
PhI(OAc)2 (1.3 equiv), PhMe, 100 8C, 2 h
PhI(OAc)2 (1.3 equiv), PhMe, 55 8C, 3 h
PhI(OAc)2 (1.3 equiv), PhMe, 55 8C, 3 h
PhI(OAc)2 (1.3 equiv), PhMe, 25?30 8C, 36 h
PhI(OTFA)2 (1.3 equiv), PhMe, 55 8C, 1.5 h
PhI(OH)OTs (1.3 equiv), PhMe, 55 8C, 2 h
Trace of product observed[b]
Ca. 50 % 3 remained[b]
14: 13 % yield[d]
No conversion of 3[b]
14: 69 %[d]
14: 57 %[d] and a trace of 3 was remained[b]
14: 24 %,[d] < 50 % conversion[b]
14: 50 %[d] and ca. 15?20 % of 3 remained[b]
14: 47 %[d] and ca. 10 % of 3 was recovered[d]
Substrate 3 decomposed
Substrate 3 decomposed
[a] Optimization reactions were conducted at a 0.09 M concentration of 3 (0.047 mmol), using 20 mol % of Pd(OAc)2. [b] Observed by TLC. [c] Powdered 4
molecular sieves (1.33 times by weight of 3) was used. [d] Isolated yield. [e] A 70 % aqueous solution was used. [f] Reaction using 5 mol % Pd(OAc)2. [g] Reaction using 10 mol % Pd(OAc)2.
[hydroxy(tosyloxy)iodo]benzene (HTIB, Koser?s reagent), which
have been used for cyclization reactions of simpler systems,[20, 23?25] led only to degradation of substrate 3. This clearly
shows the narrow window of reagent compatibility with these
sensitive nucleoside substrates. The 1H NMR spectrum of compound 14 clearly showed that it was the carbazolyl derivative
on the basis of the anticipated two doublets (2H each) at d =
8.09 and 7.97 ppm, and two triplets (2H each) at d = 7.44 and
7.36 ppm in the aromatic region of the spectrum.
Other substrates were evaluated under the identified conditions and the products prepared are shown in Figure 1. Substrate 4, which can produce two isomeric carbazoles, gave the
linear benzo[b]carbazole derivative 15, and a better 41 % yield
was obtained at 100 8C. Nitro precursor 5 underwent successful
reaction, yielding carbazole 16 in good yield. Substrates 6 and
7 with a p-alkoxy substituent on the aryl ring remote from the
purine gave modest yields of products 17 and 18. If the methoxy group is on the aryl ring proximal to the purine (substrate 8), product 17 was obtained in 31 % yield from an incomplete reaction (35 % based on recovered precursor 8). In
addition, ca. 2 % of an acetoxylated product was obtained
(plausibly carbazole 17 with an OAc group flanked by nitrogen
and oxygen atoms). Precursor 9, with the p-cyano group, gave
a very low conversion to carbazole 19. In this reaction, another
byproduct was isolated in 3 % yield. This compound showed
an ABquartet in its 1H NMR spectrum (2 H, d = 7.75 and 7.72 ppm,
J = 8.2 Hz), a doublet (1 H, d = 6.93, J = 10.5 Hz), doublet (d =
6.82, J = 1.8 Hz), and a double doublet (d = 6.60, J = 2.0,
10.0 Hz). The HRMS spectrum for this compound showed a
[M + Na] + = 693.3061. On the basis of both the HRMS information and the 10 Hz J values observed in the 1H NMR specChemCatChem 2017, 9, 1 ? 13
www.chemcatchem.org
These are not the final page numbers! оо
trum, structure 19? is proposed. Such a product could be
formed through acetoxylation by PhI(OAc)2 and hydrolysis of
the ester during workup and/or chromatography. We have observed a similar product in our previously reported work.14
With ribosyl precursors 10 and 11, complete conversion was
observed with 10 mol % Pd(OAc)2. In contrast to the successful
reactions, substrates 12 and 13, with strongly electron-depleting groups, underwent only degradation under these conditions (compare with the low yield of 19 from a precursor with
an electron-withdrawing group).
The outcome of these reactions is reminiscent of carbazole
formation from 2-aryl anilines[16] but with some differences.
Yields of products from substrates with alkoxy groups in simpler systems were generally good in comparison to nucleosides. As with the deoxynucleosides, a higher (20 mol %) catalyst load was necessary for cyclization to a N-glycosyl carbazole.[16] We devoted substantial efforts to isolate reaction intermediates, such as carbopalladated species. Although we have
had success in this respect in prior C H bond activation studies of purines and purine nucleosides,[6] we were not able to
isolate any intermediate. Thus, we propose that the exocyclic
amino group, despite the presence of other coordinating
atoms, interacts with the Pd center. This is followed by a N-directed carbopalladation, oxidation of PdII to PdIV by PhI(OAc)2,
and reductive elimination resulting in the C N bond
(Scheme 2, other ligands on the PdII and PdIV are not speculated). This mechanism implies adequate electron density on the
N6 atom, discussed below.
In the course of these optimization reactions with Pd(OAc)2/
PhI(OAc)2, MeCN and THF were also evaluated as solvents.
Here, we obtained some unexpected results. In MeCN under
3
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers
Figure 1. Products and yields obtained by Pd(OAc)2/PhI(OAc)2 mediated C N bond formation, bonds formed are shown by a heavy line. PG = t-BuMe2Si.
Formation of benzimidazole 22 appeared to occur
from a secondary, uncatalyzed reaction. Thus, this
prompted an assessment of reactions in other solvents, under metal-free conditions as well. From the
results summarized in Table 3, certain trends
emerge. In PhMe, the reaction is distinctly Pd-catalyzed, because neither formation of carbazole 14 nor
Scheme 2. A plausible mechanism for the C H bond activation/cyclization leading to C6
benzimidazole 22 was observed in the absence of a
carbazolyl nucleoside analogues (bond formed is shown by a heavy line).
catalyst (entry 1). Although CH2Cl2 has been used for
reactions of N-(biphenyl)pyridine-2-amines with
otherwise similar conditions, a 66 % yield of carbazole 14 was
PhI(OAc)2 and PhI(OTFA)2,[24] both CH2Cl2 and iPrOH were clearobtained, a yield comparable to that obtained in PhMe, accomly unsuitable (entries 2 and 3). In our previous work, MeNO2
panied by 22 % of an isomeric product that by detailed characterization was identified as benzimidazole 22 (overall 88 % yield, Scheme 3).
Structure of compound 22 was determined by
careful analysis of its 1H NMR spectrum (see Supporting Information for all detailed structural analyses). In
addition, two key gCOSY correlations, shown by the
blue arrows in Scheme 3, were observed. At a lower
10 mol % of Pd(OAc)2, the same two products were
observed. Here a lower 50 % yield of the carbazole 14 was obtained and the yield of benzimidazole 22 was 27 %. In contrast, in THF, carbazole 14
was obtained in 51 % yield and a trace of benzimida- Scheme 3. Formation of two isomeric products in MeCN under the Pd-catalyzed condizole 22 was observed by TLC.
tions (bonds formed are shown by a heavy line).
&
ChemCatChem 2017, 9, 1 ? 13
www.chemcatchem.org
4
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
нн These are not the final page numbers!
Full Papers
improved yield of 14 than via the Pd-catalyzed method. By
contrast to the metal-catalyzed method that gave the benzo[b]carbazole 15 from precursor 4 in a modest yield, the benzo[a]carbazole derivative 23 was obtained in a low yield in
HFIP. This outcome parallels electrophilic aromatic substitution
of naphthalene, which is not normally subject to steric hindrance by a peri hydrogen atom that could influence a Pd-catalyzed reaction. The yield of 23 improved slightly in the less
acidic 1,1,1-trifluoroethanol (TFE), but as could be expected
from data in Table 3, ca. 9 % of the benzimidazole derivative
was observed in this solvent (24 % total yield).
Substrates 5?7 reacted uneventfully to furnish carbazoles 16?18. The yield of product 16 from nitrated precursor 5
was comparable across the two approaches. However, cyclization yields from precursors 6 and 7 to yield 17 and 18, respectively, was much improved in the uncatalyzed approach. Cyclization of substrate 8, with a methoxy group on the proximal
ring, furnished only carbazole 17, but the reaction in HFIP led
to extensive precursor degradation. Use of TFE gave a good
68 % yield of product 17. The p-cyano precursor 9 also gave a
low product yield in HFIP, but the product formed was benzimidazole 24. Use of TFE as solvent, increased the yield of 24 to
47 %. From both reactions, iminoquinone 19? was isolated, but
it was formed to a much lower extent in TFE. Reactions of
these relatively sensitive substrates indicate that pKa modulation of the fluorinated solvent can possibly influence this reaction (pKa of HFIP = 9.3, pKa of TFE = 12.9),[27] and TFE, with a
reasonable radical-cation stabilizing power, is also suitable in
some instances, as in the reactions of 8 and 9. Formation of
carbazole 17 from substrate 8 is contrary to the substituentcontrolled bias observed previously in reactions of simpler systems.[24] The anticipation would have been formation of the
benzimidazole product as a result of the high electron density
in the proximal ring. This indicates the presence of additional
subtle regio-directing factors in reactions of nucleosides. As
with the 2?-deoxyribonucleosides, riboside precursors 10 and
11 reacted smoothly to furnish the expected carbazoles 20
and 21, respectively. The yield of 20 was superior to that in
the Pd-catalyzed method, whereas the yield of 21 was comparable between the two. As with the p-cyano precursor 9, presence of strongly electron-depleting groups on the distal aryl
ring in precursors 12 and 13 prevented carbazole formation,
and benzimidazoles 25 and 26 were obtained in good yields.
Notably, these precursors (12 and 13) underwent degradation
in the Pd-mediated approach.
Because, reactions in MeCN appeared to produce benzimidazolyl products in competition with the isomeric carbazolyl analogues, we also investigated reactions in MeCN. Results from
these experiments are shown in Table 4. In contrast to reactions in the fluorinated solvents, in which exclusive carbazole
formation occurred for the cases studied, with the exception
of the CN and CF3-containing substrates 9, 12, and 13, benzimidazole formation occurs with substrates 3?5, 7, 10, and 11.
If a methoxy group was present in either the aromatic ring,
proximal to the purine or remote from it, only carbazole formation was observed, although aryl ring electronics would have
predicted formation of the benzimidazole from substrate 8.
Table 3. Solvent effects in PhI(OAc)2 mediated C N bond-formation.[a]
Entry
Solvent
Temp [oC]
Time [h]
Yield of 14 and 22 [%]
1
2
3
4
5
6
7
8
9
10
PhMe
CH2Cl2
iPrOH
MeNO2
TFE
HFIP
MeCN
MeCN
n-PrCN
MeCN
55
55
55
55
55
55
55
r.t.
55
55
20
4
4
4
1
1
4
20
4
19
ND[b,c]
ND[b,d]
ND[b,d]
14: 21, 22: 18
14: 48, 22: 22
14: 78, 22: none
14: 15, 22: 45
14: 23, 22: 48[e]
14: 17, 22: 42[f]
ND[b,g]
[a] Reactions were conducted at a 0.09 M concentration of 3
(0.047 mmol), using 1.3 equiv of PhI(OAc)2. Where indicated, yields are of
isolated, purified products. [b] ND = Not determined. [c] Only traces of 14
and 22 were observed by TLC. [d] Multiple spots were observed by TLC.
[e] Reaction was incomplete and ca. 10-15 % of 3 was observed by TLC.
[f] A trace of 3 was observed by TLC. [g] Reaction was conducted with
1 equiv of Cs2CO3 but more than 50 % of 3 was observed by TLC.
promoted benzimidazopurine nucleoside formation from N6aryl adenosine in a modest yield.[14] But here, in MeNO2 both
14 and 22 were obtained in 1:1 ratio (combined 39 %,
entry 4). Also, in our previous work, TFE proved to be a reasonably good solvent for cyclization to the benzimidazole derivatives.[14] However, here a good overall yield of the two isomeric
products was obtained, with the ratio favoring carbazole 14
over benzimidazole 22 by 2:1 (entry 5). An interesting outcome emerged on use of 1,1,1,3,3,3-hexafluoroisopropanol
(HFIP) as solvent (entry 6). Here, carbazole 14 was observed as
the exclusive product. This is notable because, previously, N6monoaryl adenine nucleosides gave good to excellent yields of
benzimidazolyl derivatives in this solvent.[14] HFIP is known to
increase persistence of radical cations,[26] plausible intermediates in reactions involving hypervalent iodine reagents. Thus,
reactions in HFIP and by Pd catalysis essentially produce carbazoles.
In contrast to the results above, in MeCN at 55 8C, the product ratio favored benzimidazole 22 rather than carbazole 14 by
3:1 (entry 7). At room temperature, although the overall product yield was better and a slightly better yield of 22 was obtained, this reaction remained incomplete (entry 8). These
product ratios are opposite to those from the Pd-catalyzed reactions in Scheme 3, indicating that Pd catalysis favors carbazole formation. Use of n-PrCN in place of MeCN did not alter
the outcome substantially although a trace of substrate 3 remained unconsumed (entry 9). In an attempt to increase the
formation of 14 by deprotonation, a reaction was attempted
with Cs2CO3 (entry 10). However, even after 19 h, more than
50 % of substrate 3 remained and the formation of both isomers, 14 and 22, was observed.
On the basis of these observations, reactions of the remaining substrates were evaluated in HFIP, and these data are summarized in Figure 2. Comparisons of the results from the two
methods are relevant, as no comparable chemistry of nucleoside substrates is presently known. In HFIP, substrate 3 gave an
ChemCatChem 2017, 9, 1 ? 13
www.chemcatchem.org
These are not the final page numbers! оо
5
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers
Figure 2. Products obtained by reactions of N6-([1,1?-biaryl]-2-yl)adenine nucleosides with PhI(OAc)2 in HFIP or TFE, bonds formed are shown by a heavy line.
PG = t-BuMe2Si.
Precursor 6 gave a better yield of product 17 than its isomer 8,
possibly owing to reaction at the more electron-rich ring.
In contrast to the reaction of substrate 6, if a phenoxy group
was present on the remote ring in analogue 7, some benzimidazole 29 was observed. This may be owing to a reduced electron density in the remote ring by virtue of an aryl ring on the
ether oxygen atom. Nitro derivative 5 gave a 1.6:1 ratio of carbazole 15 and benzimidazole 28. Here again, formation of a
substantial amount of benzimidazole 28 is contrary to the reported cyclization bias.[24] Reactivity trends of ribosides 10 and
11 paralleled those of the deoxyribose analogues, and the CF3substituted precursors 12 and 13 gave benzimidazoles 25 and
26 exclusively, in yields comparable to those in HFIP.
Mechanistically, there are multiple possible pathways that
can result in the carbazole and/or benzimidazole products.
Among the obvious, two involve radical intermediates and one
a nitrenium ion. As shown in Scheme 4, and as proposed by
Kita et al.,[28] formation of p-complex I, if aryl ring electronics
permit, can lead to radical cation II that can cyclize to the carbazole via III by the subsequent loss of a hydrogen radical. Nitrogen-centered radical pathways[29] to products are also possible in which species IV can fragment by a radical process,
eventually cyclizing to the carbazole via III. Alternatively, nitre-
&
ChemCatChem 2017, 9, 1 ? 13
www.chemcatchem.org
nium ion pathways are also possible in reactions mediated by
PhI(OAc)2 and PhI(OTFA)2.[19, 30] As shown in Scheme 4, in addition to a nitrogen-centered radical, IV can lead to nitrenium
ion V. Consideration of resonance pairs A/B and C/D indicates
that both carbazole and benzimidazole formation can occur by
the same mechanistic manifolds, in which final cyclization
occurs either with ring A or ring B.
In contrast to simpler molecules, nucleosides contain four nitrogen atoms, all of which could interact with Pd and the IIII reagent. It is remarkable, therefore, that these complex molecules respond quite well to the transformations. To gain some
insight into the electron densities on the nitrogen atoms of
the purine bases (N1, N3, N7, N9 and N6), we evaluated the
NBO analyzed natural charges in compounds 32?41 (Table 5).
To save computational time, the sugar unprotected structures
were chosen for DFT analysis using the B3LYP/6?311G**(d,p)
basis set. Of these the N6 and N1 atoms are involved in the
chemistry described here. Generally, the order of electron density is N6 > N1 > N3 > N7 > N9. This indicates that the exocyclic
nitrogen atom is most competent for interaction with both PdII
and PhI(OAc)2, and the same time N1 has adequate electron
density to participate in cyclization with electron-deficient
centers.
6
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
нн These are not the final page numbers!
Full Papers
yielding reaction of 9 and the failure of substrates 12 and 13 in the Pd-catalyzed reaction may
be a result of the extreme depletion of electron density in the CN and CF3-containing rings, which could
deter C Pd s-bond formation. But on a different
mechanistic basis, in the uncatalyzed reactions, cyclization can occur via the N1 atom to the less electron-depleted proximal ring, producing benzimidazoles 24?26. The atomic charges in 37 and 38 are
closely similar and, therefore, one could expect comparable reactivity. This appears to be the case under
all conditions (except that substrate 8 seems to be
more sensitive to the acidity of the fluorinated sol1
2
3
Entry Substrate
R, R , R , R
Yield of
Yield of
vent). The most intriguing appears to be the reactiviBenzimidazole [%] Carbazole [%]
ty of nitro-substituted precursors 5 and 11 in MeCN.
1
2
3
22: 45
14: 15
1
3: X = H
R=R =R =R =H
Here are cases for which one would not have ex27: 18
23: 50
2
4: X = H
R = R3 = H,
R1 = R2 =
pected reactivity in the ring containing that substituent. We do not know the underlying reasons for this,
but it may be related to the higher electron density
3
5: X = H
R = NO2, R1 = R2 = R3 = H
28: 30
16: 47
at the N6 atom, which can influence reactivity at the
17: 70
4
6: X = H
R = R1 = R3 = H, R2 = OMe ?
ortho-position. However, these two substrates under18: 66
5
7: X = H
R = R1 = R3 = H, R2 = OPh 29: 7
went smooth reactions under Pd catalysis, indicating
1
2
3
6
8: X = H
R = OMe, R = R = R = H ?
17: 53
that the nitro group does not deter C Pd s-bond
?
7
9: X = H
R = H, R1 = R3 = H, R2 = CN 24: 64
30: 58
20: 16
8
10: X = OTBDMS R = R1 = R2 = R3 = H
formation with the remote ring. Comparison of the
31: 37
21: 35
9
11: X = OTBDMS R = NO2, R1 = R2 = R3 = H
NBO analyzed natural charges on the N1 and N6
25: 72
?
10
12: X = OTBDMS R = R1 = R2 = H, R3 = CF3
atoms of the nucleosides to the comparable atoms
26: 69
?
11
13: X = OTBDMS R = R2 = H, R1 = R3 = CF3
in N-(biphenyl)pyridin-2-amines (Figure 3), shows a
[a] Reactions times were 4 h except in the case of 5 for which the reaction time was
lower variation in the latter. Thus, this may be anoth3 h. [b] Yields reported are of chromatographically separated, purified products.
er indicator that extrapolation of the reactivities of
simpler compounds to those of more complex
nucleosides is not straightforward.
In previous studies involving simpler substrates, reactions
What is interesting is that in going from the N6-monoaryl to
mediated by hypervalent iodine reagents have been proposed
N6-biaryl derivatives, the electron density increases on the N6
to proceed via radical,[20] radical cation,[24] and nitrenium ion[19]
atom, and is greater in substrates bearing electron-depleting
substituents (it is greatest in the case of 39, the NBO analyzed
intermediates. Also, PhI(OTFA)2 that could not be used here,
natural charge on the N atom of the CN group is 0.321). This
has been proposed to yield radical cations, but not
can be a factor in initiating reactions at the N6 atom. The lowPhI(OAc)2.[24] Therefore, we decided to evaluate the outcomes
Table 4. Reactions of N6-([1,1?-biaryl]-2-yl)adenine nucleosides with PhI(OAc)2 in
MeCN.[a,b]
Scheme 4. A mechanistic manifold that can lead to the carbazole and benzimidazole products via radical and/or cationic intermediates.
ChemCatChem 2017, 9, 1 ? 13
www.chemcatchem.org
These are not the final page numbers! оо
7
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers
Table 6. Radical inhibition experiments with several N6-([1,1?-biaryl]-2-yl)
adenine nucleosides.[a]
Table 5. NBO analyzed natural charges on the nitrogen atoms of the
purine rings.
Compound
32
33
34
35
36
37
38
39
40
41
N1
0.552
0.550
0.552
0.553
0.548
0.552
0.553
0.556
0.548
0.556
N3
N7
0.536
0.533
0.538
0.524
0.532
0.537
0.536
0.535
0.533
0.525
0.504
0.490
0.485
0.483
0.492
0.485
0.486
0.516
0.479
0.488
N9
0.446
0.460
0.453
0.454
0.447
0.453
0.452
0.450
0.455
0.431
N6
0.563
0.577
0.607
0.608
0.642
0.607
0.608
0.804
0.638
0.627
Entry
Substrate
Inhibitor
Solvent
Result
1
3
BHT
HFIP
2
3
DPE
HFIP
3
3
BHT
MeCN
4
3
DPE
MeCN
5
5
BHT
HFIP
6
5
DPE
HFIP
7
5
BHT
MeCN
8
5
DPE
MeCN
9
6
BHT
HFIP
10
6
DPE
HFIP
11
6
BHT
MeCN
12
6
DPE
MeCN
13
10
BHT
HFIP
14
10
DPE
HFIP
15
10
BHT
MeCN
16
10
DPE
MeCN
17
13
BHT
HFIP
18
13
DPE
HFIP
19
13
BHT
MeCN
20
13
DPE
MeCN
14: 17 %
3: 59 % recovered
14: none
3: 87 % recovered
14: 10 %, 22: 35 %
3: 12 % recovered
14: 8 %, 22: 37 %
3: none
16: 18 %
5: 44 %
16: 39 %
5: 53 %
16: 18 %, 28: 6 %
5: 68 %
16: 39 %, 28: 21 %
5: 21 %
17: none
6: 59 % recovered[b]
17: none
6: 91 % recovered
17: 52 %
6: 8 % recovered
17: 61 %
6: none
20: 16 %
10: 71 %
20: none
10: 79 %
20: 11 %, 30: 37 %
10: none
20: 8 %, 30: 39 %
10: none
26: 33 %
13: 44 % recovered
26: none
13: 89 % recovered
26: 20 %
13: 30 % recovered
26: 31 %
13: none
[a] Reactions were conducted on 0.05 mmol of the N6-([1,1?-biaryl]-2-yl)adenine nucleosides. [b] A second, impure fraction of 6 (8.0 mg) was also
obtained by chromatography.
proportion of products, indicating another significant non-radical mechanistic pathway(s). Perhaps the most telling information comes from precursor 6, which is well suited to form a
radical cation in the methoxy-substituted ring. This substrate
yields the same carbazole product 17 in both HFIP and MeCN.
In HFIP, product formation was completely suppressed with
both inhibitors. In contrast, substantial formation of carbazole 17 was observed in MeCN with both inhibitors. Therefore,
such outcomes can be accounted for by considering competing radical cation/radical and nitrenium ions that can come
into play depending on the nature of the biarylyl purine
system and solvent. Thus, an expanded substrate scope is necessary to be able to comment on the mechanisms that may be
involved.
To further test the influence of reaction conditions on the
formation of possible reactive intermediates and, hence, the
Figure 3. NBO analyzed natural charges on the nitrogen atoms in N-(biphenyl)pyridin-2-amines using the B3LYP/6?311G**(d,p) basis set.
of reactions with several substrates using the exact same radical inhibitors that have been used in previous studies, but for
which only single substrates were tested.[20, 24] These data are
shown in Table 6.
In almost all cases products were formed albeit with suppression. Notably, use of DPE in HFIP fully prevented product
formation from 3 and 10. Thus, radical cations/radicals can be
major players in HFIP. In MeCN, product yields are altered in
comparison to reactions shown in Table 4. An assessment
based on one substrate only could have led to a generalization
that radical cations/radicals are likely the only species involved.
Clearly, in MeCN, these substrates yield substantially greater
&
ChemCatChem 2017, 9, 1 ? 13
www.chemcatchem.org
8
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
нн These are not the final page numbers!
Full Papers
product distribution, we modified the reaction medium. Thus,
we conducted two reactions of substrate 3 in MeCN with increasing amounts of HFIP and one with HFIP and DPE. The results are graphically represented in Figure 4. At 5 equiv of HFIP,
the extent of benzimidazole formation increased whereas carbazole formation decreased (for reference, the first bar graph
Notably though, it was possible to recover nearly 100 % of
DPE, indicating that the inhibitor does not react with the IIII reagent. These observations appear to support our proposal on
the solvent-dependence of the intermediates formed as well.
Conclusion
Nucleosides generally possess unique reactivity properties, different to those of simpler heterocyclic systems, and structural
modifications of nucleosides are challenging because of the
multiple basic nitrogen atoms and the labile glycosidic linkage.
Herein, we have studied intramolecular C N bond formation
of N6-([1,1?-biaryl]-2-yl)adenine nucleosides promoted by
PhI(OAc)2 under both Pd-mediated and uncatalyzed reaction
conditions. Two different C N bonds can be formed with these
substrates. Among those studied herein, Pd-mediated reactions produce carbazole derivatives as do uncatalyzed reactions in HFIP. Only if the aryl ring remote to the purine is electron depleted, does benzimidazole formation occur. In MeCN,
on the other hand, significant amounts of isomeric benzimidazole products can be obtained in several cases, again depending upon the electronic nature of the biaryl moiety. Studies
reveal that competing reactive intermediates can be formed
from any substrate, depending on the biarylyl purine electronics and the reaction solvent. This is exemplified in one case in
which an identical product is formed in HFIP and MeCN, but
radical inhibition experiments indicate radical cation/radical intermediates in the former and a possible nitrenium ion in the
latter. By contrast, radical inhibition experiments with N6-monoaryl adenine analogues conform to a possible single reaction
manifold, involving radicals in HFIP. Furthermore, at least in the
one example evaluated, modulation of reactivity in MeCN by
addition of a small amount of HFIP appears to improve formation of the benzimidazole isomer. Notably, this chemistry
allows the development of significant complexity on the sensitive nucleoside scaffolds through relatively simple operations.
Figure 4. Results obtained in the cyclization reactions of 3 in MeCN under
varying conditions.
pair is the result from the reaction in just MeCN). However,
with 10 equiv of HFIP, the extent of carbazole formation increased as may be anticipated from the reactivity of precursor
3 in HFIP, whereas that of the benzimidazole decreased. The
overall yields in these two cases were similar, so it appears that
there may be a balance between the carbazole and benzimidazole formation that is subtly dependent on conditions. On addition of 1 equiv of DPE (the inhibitor that prevented reaction
in HFIP) to a reaction mixture that contained 5 equiv of HFIP,
the result was closer to the reaction in just MeCN. Thus, any
radical intermediates stemming from the presence of HFIP
may be suppressed by the inhibitor (for comparison, the fifth
bar graph pair is the result from the reaction in MeCN in the
presence of DPE). These data also support the idea that no
single mechanism may be broadly ascribable, with possibly,
overlapping pathways involving radicals and/or nitrenium ions
that are sensitive to multiple factors.
In contrast to these data on the N6-biaryl adenine nucleosides, reactions of N6-monoaryl analogues in HFIP are fully suppressed in the presence of 1 equiv of these radical inhibitors,
no matter what substituent was present on the aryl ring.[14] To
compare with this result, we also assessed the reaction of 3?,5?di-O-t-BuMe2Si N6-phenyl-2?-deoxyadenosine with 1.3 equiv of
PhI(OAc)2, in MeCN in the presence of 1.0 equiv of DPE, at
55 8C. This reaction gave 55 % of the cyclized benzimidazopurine nucleoside product and 22 % of recovered starting material.
ChemCatChem 2017, 9, 1 ? 13
www.chemcatchem.org
These are not the final page numbers! оо
Experimental Section
Reactions were conducted in screw-cap glass vials with Teflonlined caps. Thin layer chromatography was performed on 200 mm
aluminum-foil-backed silica plates for 2?-deoxynucleosides and on
Merck 60F254 aluminum-backed TLC plates for the ribose analogues. Column chromatography was performed using 100?200
mesh silica gel in all cases unless otherwise mentioned. Toluene
was distilled over Na prior to use and acetonitrile was distilled over
CaH2. Hexanes and EtOAc were distilled over CaSO4 prior to use for
column chromatography. Xantphos, Pd(OAc)2, Pd2(dba)3, Cs2CO3,
hexafluoroisopropanol (HFIP), 2,2,2-trifluoroethanol (TFE), and all
other reagents were obtained from commercial suppliers and used
as received. 1H NMR spectra were obtained at 500 MHz or at
400 MHz in the solvents indicated under the individual compound
headings, and are referenced to residual protonated solvent resonances. 13C NMR spectra were obtained at 125 MHz or at 100 MHz
in the solvents indicated under the individual compound headings,
and are referenced to the solvent resonances. 19F NMR spectra
were obtained at 376 MHz in the solvents indicated under the appropriate compound headings. Chemical shifts (d) are reported in
parts per million (ppm) and coupling constants (J) are in Hertz
9
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers
(Hz). Standard abbreviations are used to designate resonance multiplicities. The carbon atoms of the saccharide portion of the nucleoside are numbered 1? through 5? starting at the anomeric
carbon atom and proceeding through the carbon chain to the primary carbinol carbon atom. All 2-bromobiaryls were prepared according to the literature procedures.[32?34]
Representative example: carbazolyl purine 2?-deoxyribonucleoside 14
Synthesized from compound 3 (63.2 mg, 0.1 mmol). Chromatography with 5 % EtOAc in hexanes gave carbazole 14 (43.4 mg, 69 %
yield) as a colorless, thick gum. Rf (SiO2/20 % EtOAc in hexanes) =
0.65. 1H NMR (500 MHz, CDCl3): d = 9.00 (s, 1 H), 8.42 (s, 1 H), 8.09
(d, J = 7.3 Hz, 2 H), 7.97 (d, J = 8.3 Hz, 2 H), 7.44 (t, J = 7.8 Hz, 2 H),
7.36 (t, J = 7.6 Hz, 2 H), 6.62 (t, J = 6.6 Hz, 1 H), 4.71?4.68 (m, 1 H),
4.12?4.09 (m, 1 H), 3.91 (dd, J = 4.4, 11.2 Hz, 1 H), 3.82 (dd, J = 2.9,
11.2 Hz, 1 H), 2.79 (app quint, Japp 6.5 Hz, 1 H), 2.55 (ddd, J = 3.9,
6.1, 13.2 Hz, 1 H), 0.95 and 0.92 (2s,
18 H), 0.15, 0.14, 0.10, and 0.09 ppm (4s,
12 H). 13C NMR (125 MHz, CDCl3): d =
153.6, 152.5, 149.9, 142.2, 139.6, 126.8,
126.3, 125.7, 122.2, 120.1, 113.7, 88.4,
85.0, 72.3, 63.0, 41.4, 26.2, 26.0, 18.6,
18.3, 4.4, 4.5, 5.1, 5.2 ppm. HRMS
(ESI/TOF) calcd for C34H48N5O3Si2 [M +
H] + : 630.3290, found 630.3270.
General procedure for the amination of TBDMS-protected
adenosine or 2?-deoxyadenosine with 2-bromobiaryls
In a clean, dry, 15 mL vial equipped with a stirring bar were placed
3?,5?-di-O-TBDMS-protected 2?-deoxyadenosine 1 (1.05 equiv) or
2?,3?,5-tri-O-TBDMS-protected adenosine 2 (1.0 equiv), the 2-bromobiaryl (1.0 equiv), Xantphos (0.15 equiv), and Cs2CO3 (1.5 equiv) in
PhMe (the mixtures were 0.15 M in the nucleoside). The mixture
was flushed with nitrogen gas and stirred at room temperature for
2 min. Then Pd2(dba)3 (0.1 equiv) was added, the vial was flushed
with nitrogen gas, capped, and stirred at 100 8C. Reaction progress
was monitored by TLC. Upon completion, the solvent was removed
under reduced pressure, and the crude material was purified by
chromatography on a silica gel column (see specific compound
headings for details).
General procedure for metal-free C
H activation and C N bond formation in HFIP (or TFE)
In a clean, dry, 8 mL vial equipped with a stirring bar were placed
the biarylyl nucleoside precursor (0.1 mmol, 1 equiv), HFIP (or TFE)
(1.0 mL), and PhI(OAc)2 (41.9 mg, 0.13 mmol). The vial was flushed
with nitrogen gas, sealed, and stirred at 55 8C for 1 h (for biarylyl
nucleoside analogues of 2?-deoxyadenosine) or 0.5 h (for biarylyl
nucleoside analogues of adenosine). Reaction progress was monitored by TLC. Upon completion, the mixture was evaporated
under reduced pressure, and purified by chromatography on a
silica gel column (see specific compound headings for details).
Representative example: N6-([1,1?-biphenyl]-2-yl)-3?,5?-di-O(tert-butyldimethylsilyl)-2?-deoxyadenosine (3)
Synthesized from compound 1 (0.302 g, 0.63 mmol), 2-bromobiphenyl (0.140 g, 0.60 mmol), Xantphos (52.1 mg, 0.09 mmol),
Cs2CO3 (0.293 g, 0.90 mmol), and Pd2(dba)3 (54.9 mg, 0.06 mmol), in
PhMe (4 mL), over 24 h at 100 8C. Chromatography by sequential
elution with 5 % and 10 % EtOAc in hexanes gave compound 3
(0.335 g, 88 % yield) as a pale-yellow foam. Rf (SiO2/10 % EtOAc in
1
H NMR
(500 MHz,
hexanes) = 0.39.
CDCl3): d = 8.49 (s, 1 H), 8.38 (d, J =
7.7 Hz, 1 H), 8.03 (s, 1 H), 7.69 (br s, 1 H),
7.49?7.42 (m, 5 H), 7.39?7.34 (m, 2 H),
7.23 (t, J = 7.7 Hz, 1 H), 6.44 (t, J = 6.3 Hz,
1 H), 4.62?4.60 (m, 1 H), 4.03?4.01 (m,
1 H), 3.84 (dd, J = 4.4, 11.0 Hz, 1 H), 3.76
(dd, J = 3.3, 11.5 Hz, 1 H), 2.67 (app quint,
Japp 6.3 Hz, 1 H), 2.45?2.40 (m, 1 H), 0.92
and 0.90 (2s, 18 H), 0.11, 0.07, and
0.06 ppm (3s, 12 H). 13C NMR (125 MHz,
CDCl3): d = 152.9, 152.6, 149.5, 139.3,
138.6, 135.3, 133.8, 130.7, 129.5, 129.1, 128.3, 128.0, 124.4, 123.0,
121.2, 88.1, 84.5, 72.2, 63.0, 41.2, 26.1, 25.9, 18.6, 18.2, 4.5, 4.6,
5.2, 5.3 ppm. HRMS (ESI/TOF) calcd for C34H50N5O3Si2 [M + H] + :
632.3447, found 632.3443.
Note. Reactions with precursors 3, 5?7, 10 and 11 gave the carbazoles 14, 16?18, 20, and 21, respectively, as were obtained in reactions with Pd(OAc)2/PhI(OAc)2. Precursors 4, 8 and 9 largely underwent decomposition in HFIP. Therefore, the reactions of these
three substrates were conducted in the less acidic TFE. Carbazoles 23 and 17 were obtained from precursors 4 and 8, respectively, whereas benzimidazole 24 was obtained from 9. Benzimidazoles 25 and 26 were obtained from precursors 12 and 13, respectively. Refer to Figure 2 for product yields.
General procedure for metal free C H activation and C N
bond formation in MeCN
In a clean, dry, 8 mL vial equipped with a stirring bar were placed
the biarylyl nucleoside precursor (0.1 mmol, 1 equiv), MeCN
(1.0 mL), and PhI(OAc)2 (41.9 mg, 0.13 mmol). The vial was flushed
with nitrogen gas sealed, and stirred at 55 8C for 4 h (the exception
is substrate 5 for which the reaction time was 3 h). Reaction progress was monitored by TLC. Upon completion, the reaction mixture
was evaporated under reduced pressure, and purified by chromatography on a silica gel column (see specific compound headings
for details).
General procedure for C H activation and C N bond formation with Pd(OAc)2/PhI(OAc)2
In a clean, dry, 8 mL vial equipped with a stirring bar were placed
the biarylyl nucleoside precursor (0.1 mmol, 1 equiv), PhMe
(1.0 mL), Pd(OAc)2 (4.5 mg, 0.02 mmol for precursors 3?9, and
2.25 mg, 0.01 mmol for precursors 10?13), and PhI(OAc)2 (41.9 mg,
0.13 mmol). The vial was flushed with nitrogen gas, sealed, and
stirred at 55 8C. Reaction progress was monitored by TLC, refer to
Figure 1 for reaction times. Upon completion, the mixture was
evaporated under reduced pressure, and purified by chromatography on a silica gel column (see specific compound headings for
details).
&
ChemCatChem 2017, 9, 1 ? 13
www.chemcatchem.org
Representative example: 10-phenylbenzimidazolyl purine 2?deoxyribonucleoside 22
Synthesized from compound 3 (63.2 mg, 0.1 mmol). Chromatography with 5 % EtOAc in hexanes gave carbazole 14 (9.5 mg, 15 %
10
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
нн These are not the final page numbers!
Full Papers
yield) as a colorless gum. Subsequent elution with 30 % EtOAc in
hexanes gave benzimidazole 22 (28.4 mg, 45 % yield) as a yellow
foam. Rf (SiO2/20 % EtOAc in hexanes) of 14 = 0.65 and 22 = 0.38.
1
H NMR of 22 (500 MHz, CDCl3): d =
9.11 (s, 1 H), 8.31 (s, 1 H), 8.20 (d, J =
7.7 Hz, 2 H), 7.92 (d, J = 8.2 Hz, 1 H),
7.75 (d, J = 7.7 Hz, 1 H), 7.53?7.47 (m,
3 H), 7.39 (t, J = 7.4 Hz, 1 H), 6.58 (t,
J = 6.3 Hz, 1 H), 4.70?4.63 (m, 1 H),
4.11?4.05 (m, 1 H), 3.88 (dd, J = 4.1,
11.2 Hz, 1 H), 3.81 (dd, J = 3.0,
11.3 Hz, 1 H), 2.65 (app quint, Japp
6.4 Hz, 1 H), 2.52 (ddd, J = 3.9, 6.0,
13.2 Hz, 1 H), 0.94 and 0.92 (2s, 18 H),
0.13, 0.10, and 0.09 ppm (3s, 12 H).
13
C NMR (125 MHz, CDCl3): d = 144.2,
142.9, 140.8, 139.0, 137.9, 135.4,
133.2, 129.8, 128.7, 128.4, 127.8, 125.7, 123.5, 122.6, 109.2, 88.3,
84.9, 72.1, 63.0, 41.9, 26.2, 26.0, 18.6, 18.2, 4.4, 4.5, 5.1,
5.3 ppm. HRMS (ESI/TOF) calcd for C34H48N5O3Si2 [M + H] + :
630.3290, found 630.3283. For analysis data of 14, see the
Pd(OAc)2/PhI(OAc)2 reaction conditions above.
[5]
[6]
[7]
[8]
[9]
[10]
Acknowledgements
[11]
[12]
Support of this work by NSF award CHE-1265687 to M. K. L. and
PSC CUNY awards to P. P. and M. K. L. is gratefully acknowledged.
Infrastructural support at CCNY was provided by NIH grant
G12MD007603 from the National Institute on Minority Health
and Health Disparities. Computational results were obtained
through the City University of New York High Performance Computing Center at the College of Staten Island, which was partially
supported by NSF awards CNS-0958379, CNS-0855217 and ACI1126113. S. P. thanks Dr. Vidya Katangoor (JNTUH, Karimnagar,
Telangana) for her support.
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
Conflict of interest
[22]
The authors declare no conflict of interest.
Keywords: benzimidazole З
hypervalent iodine З palladium
carbazole
[23]
З
cyclization
[24]
З
[25]
[26]
[1] a) Modified Nucleosides: In Biochemistry Biotechnology and Medicine (Ed.:
P. Herdewijn), Wiley-VCH, Weinheim, 2008; b) Antiviral Nucleosides:
Chiral Synthesis and Chemotherapy (Ed.: C. K. Chu), Elsevier, Amsterdam,
2003; c) Antiviral Drug Discovery for Emerging Diseases and Bioterrorism
Threats (Ed.: P. F. Torrence), Wiley, Hoboken, 2005.
[2] Deoxynucleoside Analogs in Cancer Therapy (Ed.: G. J. Peters), Humana,
Totowa, 2003.
[3] a) L. P. Jordheim, D. Durantel, F. Zoulim, C. Dumontet, Nat. Rev. Drug Discovery 2013, 12, 447 ? 464; b) V. L. Damaraju, S. Damaraju, J. D. Young,
S. A. Baldwin, J. Mackey, M. B. Sawyer, C. E. Cass, Oncogene 2003, 22,
7524 ? 7536; c) C. M. Galmarini, J. R. Mackey, C. Dumontet, Lancet Oncol.
2002, 3, 415 ? 424.
[4] a) A. Matarazzo, R. H. E. Hudson, Tetrahedron 2015, 71, 1627 ? 1657; b) X.
Su, X. Xiao, C. Zhang, M. Zhao, Appl. Spectrosc. 2012, 66, 1249 ? 1261;
c) A. A. Tanpure, M. G. Pawar, S. G. Srivatsan, Isr. J. Chem. 2013, 53, 366 ?
378; d) R. W. Sinkeldam, N. J. Greco, Y. Tor, Chem. Rev. 2010, 110, 2579 ?
2619; e) J. N. Wilson, E. T. Kool, Org. Biomol. Chem. 2006, 4, 4265 ? 4274;
f) L. I. Wiebe, Braz. Arch. Biol. Technol. 2007, 50, 445 ? 459; g) M. E. Haw-
ChemCatChem 2017, 9, 1 ? 13
www.chemcatchem.org
These are not the final page numbers! оо
[27]
[28]
[29]
[30]
11
kins, in Topics in Fluorescence Spectroscopy, Vol. 7: DNA Technology (Ed.:
J. R. Lakowicz), Kluwer Academic/Plenum Publishers, New York, 2003,
pp. 151 ? 175.
M. K. Lakshman, A. C. Deb, R. R. Chamala, P. Pradhan, R. Pratap, Angew.
Chem. Int. Ed. 2011, 50, 11400 ? 11404; Angew. Chem. 2011, 123, 11602 ?
11606.
R. R. Chamala, D. Parrish, P. Pradhan, M. K. Lakshman, J. Org. Chem.
2013, 78, 7423 ? 7435.
V. V. Zhdankin, Hypervalent Iodine Chemistry: Preparation, Structure and
Synthetic Applications of Polyvalent Iodine Compounds, John Wiley &
Sons, Ltd., Chichester, UK, 2014.
a) Hypervalent Iodine Chemistry, Vol. 373 of Top. Curr. Chem. (Ed.: T.
Wirth), Springer, Cham, 2016; b) Hypervalent Iodine Chemistry: Modern
Developments in Organic Synthesis, Vol. 224 of Top. Curr. Chem. (Ed. T.
Wirth), Springer, Berlin, Heidelberg, 2003; c) A. Varvoglis, Hypervalent
Iodine in Organic Synthesis, Academic Press, San Diego, 1997.
For selected recent reviews, please see: a) R. Narayan, S. Manna, A. P.
Antonchick, Synlett 2015, 26, 1785 ? 1803; b) F. V. Singh, T. Wirth, Chem.
Asian J. 2014, 9, 950 ? 971; c) A. Parra, S. Reboredo, Chem. Eur. J. 2013,
19, 17244 ? 17260; d) P. Finkbeiner, B. J. Nachtsheim, Synthesis 2013, 45,
979 ? 999; e) L. Pouysgu, D. Deffieux, S. Quideau, Tetrahedron 2010, 66,
2235 ? 2261; f) T. Dohi, Y. Kita, Chem. Commun. 2009, 2073 ? 2085; g) V. V.
Zhdankin, ARKIVOC 2009, 1 ? 62; h) V. V. Zhdankin, P. J. Stang, Chem. Rev.
2008, 108, 5299 ? 5358; i) R. M. Moriarty, J. Org. Chem. 2005, 70, 2893 ?
2903; j) V. V. Zhdankin, P. J. Stang, Chem. Rev. 2002, 102, 2523 ? 2584.
For recent reviews see: a) S. Murarka, A. P. Antonchick, Top. Curr. Chem.
2016, 373, 75 ? 104; b) Z. S. Zheng, D. Zhang-Negrerie, Y. F. Du, K. Zhao,
Sci. China Chem. 2014, 57, 189 ? 214.
A. Yoshimura, V. V. Zhdankin, Chem. Rev. 2016, 116, 3328 ? 3435.
G.-R. Qu, L. Liang, H.-Y. Niu, W.-H. Rao, H.-M. Guo, J. S. Fossey, Org. Lett.
2012, 14, 4494 ? 4497.
A. C. Rios, H. T. Yu, Y. Tor, J. Phys. Org. Chem. 2015, 28, 173 ? 180.
S. Satishkumar, M. K. Lakshman, Chem. Commun. 2017, 53, 2226 ? 2229.
W. C. P. Tsang, N. Zheng, S. L. Buchwald, J. Am. Chem. Soc. 2005, 127,
14560 ? 14561.
J. A. Jordan-Hore, C. C. C. Johansson, M. Gulias, E. M. Beck, M. J. Gaunt,
J. Am. Chem. Soc. 2008, 130, 16184 ? 16186.
S. W. Youn, J. H. Bihn, B. S. Kim, Org. Lett. 2011, 13, 3738 ? 3741.
M. Yamamoto, S. Matsubara, Chem. Lett. 2007, 36, 172 ? 173.
A. P. Antonchick, R. Samanta, K. Kulikov, J. Lategahn, Angew. Chem. Int.
Ed. 2011, 50, 8605 ? 8608; Angew. Chem. 2011, 123, 8764 ? 8767.
S. H. Cho, J. Yoon, S. Chang, J. Am. Chem. Soc. 2011, 133, 5996 ? 6005.
a) J. Sun, D. Zhang-Negrerie, Y. Du, K. Zhao, J. Org. Chem. 2015, 80,
1200 ? 1206; b) X. Ban, Y. Pan, Y. Lin, S. Wang, Y. Du, K. Zhao, Org.
Biomol. Chem. 2012, 10, 3606 ? 3609.
F. N. Ngassa, K. A. DeKorver, T. S. Melistas, E. A. H. Yeh, M. K. Lakshman,
Org. Lett. 2006, 8, 4613 ? 4616.
D. N. Rao, S. Rasheed, R. A. Vishwakarma, P. Das, RSC Adv. 2014, 4,
25600 ? 25604.
J.-H. Chu, W.-T. Hsu, Y.-H. Wu, M.-F. Chiang, N.-H. Hsu, H.-P. Huang, M.-J.
Wu, J. Org. Chem. 2014, 79, 11395 ? 11408.
G. Qian, B. Liu, Q. Tan, S. Zhang, B. Xu, Eur. J. Org. Chem. 2014, 4837 ?
4843.
L. Eberson, M. P. Hartshorn, O. Persson, J. Chem. Soc. Perkin Trans. 2
1995, 1735 ? 1744.
I. A. Shuklov, N. V. Dubrovina, A. Boerner, Synthesis 2007, 2925 ? 2943.
a) Y. Kita, T. Dohi, Chem. Rec. 2015, 15, 886 ? 906; b) Y. Kita, T. Takada, S.
Mihara, B. A. Whelan, H. Tohma, J. Org. Chem. 1995, 60, 7144 ? 7148;
c) Y. Kita, H. Tohma, K. Hatanaka, T. Takada, S. Fujita, S. Mitoh, H. Sakurai,
S. Oka, J. Am. Chem. Soc. 1994, 116, 3684 ? 3691.
a) H. Togo, Y. Harada, M. Yokoyama, J. Org. Chem. 2000, 65, 926 ? 929;
b) H. Togo, Y. Hoshina, T. Muraki, H. Nakayama, M. Yokoyama, J. Org.
Chem. 1998, 63, 5193 ? 5200.
a) R. Samanta, J. O. Bauer, C. Strohmann, A. P. Antonchick, Org. Lett.
2012, 14, 5518 ? 5521; b) H. Liu, X. Wang, Y. Gu, Org. Biomol. Chem.
2011, 9, 1614 ? 1620; c) I. Tellitu, E. Domnguez, Trends Heterocycl. Chem.
2011, 15, 23 ? 32; d) Y. Amano, S. Nishiyama, Tetrahedron Lett. 2006, 47,
6505 ? 6507; e) D. J. Wardrop, M. S. Burge, J. Org. Chem. 2005, 70,
10271 ? 10284; f) Y. Misu, H. Togo, Org. Biomol. Chem. 2003, 1, 1342 ?
1346; g) D. J. Wardrop, W. Zhang, C. L. Landrie, Tetrahedron Lett. 2004,
45, 4229 ? 4231; h) D. J. Wardrop, M. S. Burge, Chem. Commun. 2004,
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers
1230 ? 1231; i) Y. Kikugawa, A. Nagashima, T. Sakamoto, E. Miyazawa, M.
Shiiya, J. Org. Chem. 2003, 68, 6739 ? 6744; j) N. Itoh, T. Sakamoto, E.
Miyazawa, Y. Kikugawa, J. Org. Chem. 2002, 67, 7424 ? 7428; k) Y. Kikugawa, M. Kawase, Chem. Lett. 1990, 19, 581 ? 582.
[31] F. Weinhold, C. R. Landis, Discovering Chemistry With Natural Bond Orbitals, John Wiley & Sons, Hoboken, New Jersey, 2012, pp. 132 ? 133.
[32] C.-W. Li, C.-I. Wang, H.-Y. Liao, R. Chaudhuri, R.-S. Liu, J. Org. Chem.
2007, 72, 9203 ? 9207.
[33] H.-Y. Zhang, H.-M. Yi, G.-W. Wang, B. Yang, S.-D. Yang, Org. Lett. 2013,
15, 6186 ? 6189.
&
ChemCatChem 2017, 9, 1 ? 13
www.chemcatchem.org
[34] G. Karig, M.-T. Moon, N. Thasana, T. Gallagher, Org. Lett. 2002, 4, 3115 ?
3118.
Manuscript received: June 3, 2017
Revised manuscript received: August 19, 2017
Version of record online: && &&, 0000
12
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
нн These are not the final page numbers!
FULL PAPERS
S. Satishkumar, S. Poudapally,
P. K. Vuram, V. Gurram, N. Pottabathini,
D. Sebastian, L. Yang, P. Pradhan,
M. K. Lakshman*
&& ? &&
Pd-Catalyzed versus Uncatalyzed,
PhI(OAc)2-Mediated Cyclization
Reactions of N6-([1,1?-Biaryl]-2yl)Adenine Nucleosides
To catalyze or not to catalyze? The selectivity of C N bond formation of N6([1,1?-biaryl]-2-yl)adenine nucleosides
promoted by PhI(OAc)2 under both Pdmediated and uncatalyzed reaction conditions can be modulated to produce
either carbazolyl or aryl benzimidazo-
ChemCatChem 2017, 9, 1 ? 13
purinyl nucleoside derivatives. Whereas
PdII/PdIV redox is responsible for carbazole formation under the metal-catalyzed conditions, in HFIP and MeCN radical cations and/or nitrenium ions can
be intermediates.
www.chemcatchem.org
These are not the final page numbers! оо
13
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Документ
Категория
Без категории
Просмотров
3
Размер файла
1 589 Кб
Теги
cctc, 201700918
1/--страниц
Пожаловаться на содержимое документа