heating can dramatically accelerate reactions on planar
cellulose supports,6,7,13 we examined these conditions for
deazalumazine macroarray (5) generation. By spotting the
array with reagents and subjecting the array to MW heating
(400 W, 10 min) four times in succession,14 complete
conversions to deazalumazines 6 were obtained. Product
purities for 6 ranged from 70% to 90% (as determined post-
cleavage; Table 1) with only one exception: chalcones (1)
with p-dimethylamino substituents in the B ring (R2 )
4-NMe2) were found to be unreactive. These precursors could
not be converted to deazalumazines under a wide range of
MW-assisted conditions; we speculate that the increased
electron density of these chalcones (1) is inhibiting nucleo-
philic attack on the R,â-unsaturated system.
Prior to compound cleavage, we evaluated the fluorescent
properties of the cyanopyridines (2) and deazalumazines (5)
in the macroarray format. We found that simple visual
inspection of the arrays under irradiation from a handheld
UV lamp (at 254 nm)15 provided a straightforward primary
screen to determine if the compounds were fluorescent
(Figure 1). In some cases, this visual assay also facilitated
To facilitate more quantitative analysis of these fluoro-
phores, the array members were each punched-out of the
support, cleaved using trifluoroacetic acid (TFA) vapor
(Scheme 1), and evaluated in solution.11 We found that the
quantity of cyanopyridine 3 or deazalumazine 6 obtained
from a single spot (∼100 nmol) was sufficient to determine
excitation and emission spectra, as well as relative quantum
yields (Table 1). Further, the good to high compound purities
enabled the examination of these materials post-cleavage
without further purification steps.
Overall, our quantitative SAR study largely correlated with
that determined qualitatively under the UV lamp. The
deazalumazines (6) demonstrated markedly higher quantum
yields than the cyanopyridines (3) (Table 1). As we had
observed visually, the presence of a hydroxyl or other
electron donating group in the para position of the A ring
greatly enhanced fluorescence in both dye classes (3a-l and
6a-k), whereas the deazalumazines and cyanopyridines that
contained only a m-hydroxyl group (3m and 6s) had very
low quantum yields (<0.01). Incorporation of additional
electron-donating groups on the A ring of either compound
class produced higher wavelength emissions but concomitant
reductions in quantum yields.
Most alterations to the B ring on either 3 or 6 had little
effect on their emission spectra (Table 1). However, addition
of a p-dimethylamino substituent to the cyanopyridines (3)
had a profound effect on their spectral characteristics.
Fluorescence spectra of 3f and 3l displayed significant red
shifts (91 and 64 nm, respectively) compared to the other
cyanopyridines (3) studied. Again, these data correlated with
our initial visual screen of macroarray 2 (Figure 1A). Finally,
both the cyanopyridines (3) and deazalumazines (6) exhibited
uniformly large Stokes shifts (90-120 nm).16 Large Stokes
shifts are highly desirable in fluorescent dyes because they
minimize reabsorption of emitted light by the dye itself.17
This property suggests the potential utility of cyanopyridines
and deazalumazines as dye molecules.
For further analysis of these two dye classes, three
representative compounds (3e, 3f, and 6c) were synthesized
on a larger scale in solution.11 As expected, their spectral
characteristics closely matched those generated using the
macroarray format. We next investigated the pH dependence
of these spectral properties. Buffered ethanolic solutions of
3e, 3f, and 6c were studied over a wide pH range (1.6-10),
and the excitation and emission wavelengths were found to
be constant over this range. While the quantum yields of 3e
and 6c were nearly constant from low to neutral pH, the
quantum yield of 3f varied, albeit only slightly, increasing
from 0.06 at pH 1.6 to 0.09 at pH 7.6. At pH > 7.6, the
quantum yields of all three fluorophores began to diminish.
We reasoned that the reduction in quantum yield at higher
pH was due to deprotonation of the phenols in 3e, 3f, and
Figure 1. Subsections of (A) cyanopyridine macroarray 2 and (B)
deazalumazine macroarray 5 irradiated at 254 nm using a handheld
UV lamp.15 Graphics were obtained with a digital camera and edited
in AdobePhotoshop.11 Scale: white line ) 1 cm.
qualitative SAR to be derived across the entire macroarray.
Analysis of the irradiated cyanopyridine array 2 revealed
two important SAR trends (Figure 1A). First, the presence
of a m-hydroxyl group on the A ring (R1) of cyanopyridines
2 (middle row) caused a modest reduction in fluorescence
intensity relative to the para position (top row). Second, and
more strikingly, a p-dimethylamino substituent on the B ring
(R2) caused a shift from blue to bright yellow-green
fluorescence for the cyanopyridines 2 (far left column). For
the deazalumazine macroarray 5, we observed a similar
diminution in fluorescence intensity when the A ring (R1)
possessed a m-hydroxyl substituent (Figure 1B). Contribution
of the p-dimethylamino groups could not be assessed,
however, due to the inaccessibility of these deazalumazines
via our synthetic method (see above).
(13) MW heating is seeing increasing use in solid-phase combinatorial
synthesis: (a) Blackwell, H. E. Org. Biomol. Chem. 2003, 1, 1251-1255.
(b) Kappe, C. O.; Dallinger, D. Nat. ReV. Drug DiscoV. 2006, 5, 51-63.
(14) All MW-assisted reactions were performed in a Milestone Mi-
croSYNTH Labstation multimodal MW reactor using power (wattage)
control. Temperature control was not possible due to the low solvent
volumes used (i.e., microliter scale). See Supporting Information for full
details.
(16) Fluorescein, one of the most commonly used fluorophores, has a
Stokes shift of 19 nm. For a recent report of fluorescein derivatives, see:
Urano, Y.; Kamiya, M.; Kanda, K.; Ueno, T.; Hirose, K.; Nagano, T. J.
Am. Chem. Soc. 2005, 127, 4888-4894.
(17) (a) Jameson, D. M.; Croney, J. C.; Moens, P. D. Methods Enzymol.
2003, 360, 1-43. (b) Lacowicz, J. R. Principles of Fluorescence Spectros-
copy; Kluwer Academic/Plenum Publishers: New York, 1999.
(15) C. Entela Mineralight Lamp; model UVGL-58.
Org. Lett., Vol. 8, No. 8, 2006
1647
Bowman, Matthew D.
