Pyrazine Analogues of Dipyrrolylquinoxalines

Article abstract of DOI:10.1021/ol0355635

(Equation presented) The synthesis of novel pyrazine derivatives bearing pyrrolic substituents is reported; the ability of these systems to bind certain biologically relevant anions in dichloromethane is also detailed.

Full text of DOI:10.1021/ol0355635

ORGANIC
LETTERS
2003
Vol. 5, No. 22
4141-4144
Pyrazine Analogues of
Dipyrrolylquinoxalines
Jonathan L. Sessler,* G. Dan Pantos, Evgeny Katayev, and Vincent M. Lynch
Department of Chemistry and Biochemistry and Institute for Cellular and
Molecular Biology, 1 UniVersity Station A5300, UniVersity of Texas at Austin,
Austin, Texas 78712-0165
sessler@mail.utexas.edu
Received August 20, 2003
ABSTRACT
The synthesis of novel pyrazine derivatives bearing pyrrolic substituents is reported; the ability of these systems to bind certain biologically
relevant anions in dichloromethane is also detailed.
Stimulated in part by a growing awareness of the environ-
mental and clinical importance of anions,1 considerable effort
has been devoted recently to the design and synthesis of new
anion-binding agents. Nonetheless, despite impressive ad-
vances, there remains a need for anion receptors displaying
higher binding affinities and improved binding selectivities.
This need reflects not only the relative newness of the field
but also the fact that, in contrast to cations, anions are
characterized by a variety of geometries, including many that
are nonspherical, and a full range of charge-to-size ratios.2
The latter systems offer the advantage of being easy to
prepare and allowing for the direct, so-called naked eye
detection of fluoride and, in certain cases, phosphate-type
anions. Generalization of the dipyrrolylquinoxaline (DPQ)
strategy could lead to receptors with different intrinsic
binding properties. In this paper, we report the synthesis of
a new set of DPQ analogues, the pyrazine-oligopyrroles 2-7,
and show that certain members of this class (e.g., 7) bind
biologically relevant anions such as chloride, dihydrogen-
phosphate, and mono- and dicarboxylates in dichloromethane
with selectivities that differ from those of the corresponding
DPQ derivative (i.e., 8).6
Over the last 10 years or so, our group has developed a
variety of pyrrole-based anion receptors, including ones based
on calixpyrroles,3 sapphyrins,4 and dipyrrolylquinoxalines.5,6
The synthesis of the pyrazine oligopyrrole receptor 7
begins with 1,2-bis-(1H-pyrrol-2-yl)ethane-1,2-dione 1,5
which upon reaction with diaminomaleonitrile (DAMN)
yields 5,6-bis(1H-pyrrol-2-yl)pyrazine-2,3-dicarbonitrile 2 in
55% yield. Vilsmeier formylation of this simple DPQ
analogue using 1.2 equiv of the formylating agent produces
the monoformylated species 3 in 62% yield. The diformy-
lated derivative 4 is obtained, again via a Vilsmeyer
procedure, in 71% yield when 3 equiv of the formylating
agent is used (Scheme 1).
(1) Sorochinskii, V. V.; Kurganov, B. I. Appl. Biochem. Microbiol. 1997,
33, 579. Schreerder, J.; Engbersen, J. F. J.; Reinhoudt, D. N. Recl. TraV.
Chim. Pays-Bas 1996, 115, 307.
(2) Beer P. D.; Gale, P. A. Angew. Chem., Int. Ed. 2001, 40, 486.
(3) Gale, P. A.; Anzenbacher, P., Jr.; Sessler, J. L.Coord. Chem. ReV.
2001, 222, 57.
(4) Sessler, J. L.; Davis, J. M. Acc. Chem. Res. 2001, 34, 989.
(5) Black, C. B.; Andrioletti, B.; Try, A. C.; Ruiperez, C.; Sessler, J. L.
J. Am. Chem. Soc. 1999, 121, 10438. Mizuno, T.; Wei, W.-H.; Eller, L. R.;
Sessler, J. L. J. Am. Chem. Soc. 2002, 124, 1134. Kirkovits, G. J.;
Zimmerman, R. S.; Huggins, M. T.; Lynch, V. M.; Sessler, J. L. Eur. J.
Org. Chem. 2002, 3768.
In analogy to what proved true in the DPQ series,6 once
in hand, the formylated intermediates 3 and 4 could be
(6) Sessler, J. L.; Maeda, H.; Mizuno, T.; Lynch, V. M.; Furuta, H. Chem.
Commun. 2002, 862.
10.1021/ol0355635 CCC: $25.00 © 2003 American Chemical Society
Published on Web 10/01/2003

the dialdehyde 4 to the corresponding dialcohol7 and then
reacting immediately with pyrrole in the presence of a
catalytic quantity of TFA; this yields the desired tetrapyr-
rolylpyrazine 6 in 28% yield (cf. Scheme 2).
Scheme 1a
The solid-state structure of the basic dipyrrolypyrazine
(DPP) core was determined by subjecting single crystals of
2 to X-ray diffraction analysis (Figure 1). The two pyrrole
a Key: (a) DAMN, BF3Et2O, CH2Cl2, rt, 12 h; (b) (i) POCl3,
DMF, 1,2-dichloroethane, reflux, 2 h; (ii) NaOAc, reflux, 1 h.
reacted with pyrrole in the presence of a catalytic amount
of TFA to give the tetra- and hexapyrrolic species 5 and 7
in yields of 50% and 84%, respectively (Scheme 2). The
symmetrical tetrapyrrolylpyrazine 6 is obtained by reducing
Scheme 2a
Figure 1. X-ray structure of 2 with all heavy atoms labeled. The
thermal ellipsoids were scaled to the 50% probability level.8
NHs are pointing away from each other, as observed in the
original dipyrrolylquinoxaline systems.
There are two crystallographically independent molecules
of 2 per asymmetric unit. These molecules alternate along
the a axis and are found in the form of an extended two-
dimensional array. This array and likely the conformations
of the individual DPP molecules are stabilized by comple-
mentary hydrogen bonds between the pyrrole NH (N1) and
cyano N (N18) moieties of a first molecule of 2 and
the corresponding pair (N20′, N13′, respectively) of a
second molecule. The geometry of these interactions are:
N1-H1‚‚‚N18 (related by 1 - x, 1 - y, 1 - z); N‚‚‚N 2.958-
(2) Å, H‚‚‚N 2.12(2) Å, N-H‚‚‚N 161(2)°; N1′-H1′‚‚‚N18′
(related by 2-x, 1-y, 1-z); N‚‚‚N 3.136(3) Å, H‚‚‚N 2.30(2)
Å, N-H‚‚‚N 165(2)°; N13′-H13′‚‚‚N20′ (related by 1 - x,
2 - y, 1 - z); N‚‚‚N 3.063(2) Å, H‚‚‚N 2.31(2) Å,
N-H‚‚‚N 152(2)°. (Dashed lines are indicative of a H-
bonding interaction.)
To probe the possible solution-state conformational prop-
erties of the DPPs, NOESY experiments were performed
using the unsymmetric derivative 5. No coupling was
observed between the four â-pyrrole hydrogens, indicating
that the conformation where the two internal pyrrole moieties
point away from each other, likely exists only in the solid
state. This lack of coupling can be ascribed to the ability of
the pyrroles in these “claw”-like systems to rotate freely in
a Key: (a) pyrrole, TFA, rt, 2 h; (b) LiH3Bpyrr, THF, rt, 6 h.11
(7) Collins, C. J.; Fischer, G. B.; Reem, A.; Goralski, C. T.; Singaram,
B. Tetrahedron Lett. 1997, 38, 529.
4142
Org. Lett., Vol. 5, No. 22, 2003

solution at room temperature. An important consequence of
the resulting, inferred, low barrier to rotation is that it should
allow each pyrrole subunit present in receptors such as 7 to
adjust their spatial orientation individually so as to adopt
collectively a conformation that is conducive to anion
binding.
particular, it is characterized by a charged carboxylate “head”
that could fit between the “pyrrolic claws”, as well as by a
methyl “tail” that could function as a cap for the host-guest
ensemble by isolating the complex and its anionic “head”
from solvent or reducing interactions with the tetrabutylam-
monium countercation.
The interaction of 7 and 8, its DPQ analogue, with several
biologically relevant anions, including mono- and dicar-
boxylates, was studied in dichloromethane solution using
UV-vis spectroscopy. The relative host/guest stoichiometries
were determined via Job plots. In both cases, standard
protocols, as used previously by our group5,9 (and others),
were employed. In analogy to previous studies, the anions
were used in the form of their corresponding tetrabutylam-
monium salts. Because they were not available commercially,
the ditetrabutylammonium salts of oxalic, malonic, and
succinic acids were prepared in accord with literature
procedures.10
Although admittedly speculative, the above rationalizations
cannot be effectively evoked in the case of the other anions
considered in this study. Thus, a closer correspondence
between charge density and affinity is expected. In fact, with
the exception of acetate, the relative affinity of 7 decreases
with increasing anion size as follows: chloride/dihydrogen-
phosphate/oxalate/malonate/succinate ) 914:562:453:40:1.
Here, it is important to note that the binding constant for
the succinate anion used to determine this set of ratios was
taken as the square root of the product of Ka1 × Ka2 since
Job-plot analysis for the interaction of 7 with succinate anion
revealed a 2:1 binding mode. By contrast, the other anions
were all found to be bound in a 1:1 stoichiometry, as judged
from such analysis (Table 1). Needless to say, the malonate/
succinate selectivity is much lower if the Ka for malonate is
compared to effective equilibrium constant (Ka1) associated
with the binding of the first anionic equivalent of succinate
to receptor 7. Using these values, a malonate/succinate
selectivity of 1.1:1 is obtained.
As revealed by inspection of Table 1, receptor 7 displays
a higher affinity for acetate anion (Ka: 176 000 ( 11 000
Table 1. Affinity Constants (M-1) for Anion Binding
Receptors 7 and 8 As Determined from UV-vis Spectroscopic
Titrations in CH2Cl2. The Anions Were Studied in the Form of
Their Tetrabutylammonium Salts
To provide a basis for comparison between this new series
of receptors and the DPQs reported earlier, the anion
recognition behavior of receptor 8 was analyzed using the
same series of anions. Whereas the interactions between 8
-
and both Cl- and H2PO4 have been reported, its ability to
bind mono- and dicarboxylate anions was not previously
considered. As revealed by the findings summarized in Table
1, this functionalized DPQ derivative binds carboxylate-type
anions well. However, in contrast to 7, the affinities for this
class of anions are found to be nearly invariant to structure
and a 1:1 anion-to-receptor stoichiometry was observed.
Also, in further contrast to what is true for 7, it is phosphate,
not chloride, that is bound with highest affinity within the
series of studied anions.
Comparing the two receptors, 7 and 8, supports the
emerging notion that small changes in the overall molecular
architecture can have a significant influence on the binding
capabilities of a given type of anion binding motif. In the
present instance, the more rigid nature of the bridging
aromatic unit (quinoxaline vs pyrazine) present in 8 could
render differences in anion size less important (due to existing
preorganization imposed by the spacer).12 Likewise, the larger
nature of the overall host-guest ensemble could mitigate
the beneficial effects of the protective methyl “tail” present
in the acetate complex. In any case, it is important to
appreciate that “fine-tuning” of anion affinities and selectivi-
ties is possible in the context of dipyrrolylquinoxaline-type
receptors.
anion
Cl-
H2PO4
acetate
oxalate
malonate
succinate
Ka (7/guest)a
Ka(8/guest)a
48 000 ( 2700
30 000 ( 1500
175 000 ( 11 000
24 000 ( 1300
2100 ( 200
5800c
300 000c
46 000 ( 4600
30 000 ( 2100
21 000 ( 2800
53 000 ( 8000
-
2000 ( 150b
a The R2 values for the curve fits used to determine the affinity constants
range between 0.969 and 0.998. b In the case of succinate + 7, a 2:1 binding
stoichiometry is observed. The tabulated value is for the first binding
interaction, Ka1; the calculated equilibrium constant for the second binding
event is 1.4 ( 0.5 M-1 c From ref 6.
.
M-1) compared to the other anions subject to study (i.e.,
chloride, dihydrogenphosphate, oxalate, malonate, and suc-
cinate). While acetate anion is neither the smallest of the
anions analyzed nor the one with the highest charge density,
it does have a size and shape that differs from the others. In
(8) Crystallographic data for 2: C29 H18 Cl2 N12, Mw ) 605.45, triclinic,
P-1, a ) 7.3374(1) Å, R ) 79.514(1)°, b ) 12.8235(2) Å, â ) 80.208(1)°,
c ) 15.6794(3) Å,γ ) 84.107(1)°, V ) 1425.60(4) Å3, Dc ) 1.410 g cm-3
,
In summary, a new set of polypyrrole-based anion recep-
tors has been synthesized. While X-ray diffraction analysis
Z ) 2, R ) 0.0464, GOF ) 1.023 (I > 2σ(I)), Rw ) 0.1186 (all data),
CCDC 217630. See the Supporting Information for CIF files.
(9) Bucher, C.; Zimmerman, R. S.; Lynch, V.; Kral, V.; Sessler, J. L J.
Am. Chem. Soc. 2001, 123, 2099. Sessler, J. L.; Camiolo, S.; Gale, P. A.
Coord. Chem. ReV. 2003, 240, 17.
(11) LiH3Bpyrr ) lithium pyrrolidinoborohydride was synthesized ac-
cording to the literature; see: Thomas, L. S.; Collins, C. J.; Cuzens, J. R.;
Spiciarich, D.; Goralski, C. T.; Singaram, B. J. Org. Chem. 2001, 66, 1999.
(10) Cotton, F. A.; Lin, C.; Murillo, C. A. Inorg. Chem. 2001, 40, 478.
Org. Lett., Vol. 5, No. 22, 2003
4143

reveals that the two pyrroles in the basic dipyrrolypyrazine
core are oriented in opposite directions, NOESY experiments
provide support for the notion that pyrrolic subunits are free
to rotate in solution and hence interact effectively with
anionic species. In accord with such a conclusion, solution-
phase anion-binding studies (dichloromethane) reveal that
the hexapyrrole pyrazine derivative 7 is an effective anion
receptor. It binds oxalate with high selectivity relative to
malonate and succinate (Ka ratio of 450:40:1 in CH2Cl2).
However, it binds the simple monocarboxylate anion, acetate,
even more effectively. By contrast, the corresponding
hexapyrrolic quinoxaline derivative, 8, binds acetate no more
strongly than oxalate, malonate, or succinate (again in
dichloromethane solution). This disparate set of findings
supports the conclusion that small changes in structure can
lead to oligopyrrolic receptors with different anion binding
properties and sets the stage for further design and synthesis
efforts in the pyrrole-based anion recognition area.
Acknowledgment. This work has been supported by the
National Institutes of Health (Grant No. GM 58907 to J.L.S.).
Mr. Won-Seob Cho and Mr. Hiromitsu Maeda are warmly
thanked for many helpful discussions.
(12) We performed molecular modeling calculations, using HyperChem,
v7.1 in order to calculate the rotational barrier for the C2-C6 and C11-
C12 bonds (numbering of the atoms is according with the X-ray structure).
We used the di-tert-butyl analogues of both 7 and 8 so as to keep the
calculations tractable. The PM3 semiempirical method in conjunction with
the Conformation Search module was used to determine the structure with
the minimum energy. The criteria for convergence was a difference in energy
between two sequential structures below 0.01 kcal/(A mol). The results of
these calculation show that the rotational barrier is 37% higher in the case
of 8 vs 7. This is consistent with preorganization and conformation affects
being significant aryl bridged dipyrrole anion receptors, and more so in the
case of the larger dipyrrolylquinoxalines than in the smaller dipyrrolylpyra-
zines.
Supporting Information Available: Detailed descrip-
tions of experimental procedures; detailed anion binding data
and X-ray experimental data. This material is available free
of charge via the Internet at http://pubs.acs.org.
OL0355635
4144
Org. Lett., Vol. 5, No. 22, 2003