Substituent effects on thermal decolorization rates of bisbenzospiropyrans

Article abstract of DOI:10.1021/jo051449j

A novel application of photochromic molecules is to mimic physiological oscillatory calcium signals by reversibly binding and releasing calcium ions in response to light. Substituent changes on the largely unexplored photochromic bisbenzospiropyran scaffold led to significant changes in thermal fading rates in several organic solvents. Excellent correlations have been found between fading rates and empirical Hammett constants as well as calculated ground-state energies. These correlations can be used to improve scaffold design.

Full text of DOI:10.1021/jo051449j

Substituent Effects on Thermal
Decolorization Rates of
Bisbenzospiropyrans
Nina T. Lu, Vi N. Nguyen, Satish Kumar, and
Alison McCurdy*
FIGURE 1. Representation of a photochromic chelator. The
square represents a metal ion.
The Department of Chemistry and Biochemistry, California
State University, Los Angeles, 5151 State University Dr.,
Los Angeles, California 90032
amccurd@calstatela.edu
Received July 13, 2005
FIGURE 2. Photochromic scaffold bisbenzospiropyran 1 as
well as one of eight possible stereoisomeric open forms.
documented.3-5 However, these systems lack specificity
for calcium over other univalent and divalent cations, and
are poorly soluble in water. To date, there are no calcium-
selective binders that can be used to generate the wide
spectrum of physiological oscillatory calcium signals in
aqueous systems.
A novel application of photochromic molecules is to mimic
physiological oscillatory calcium signals by reversibly bind-
ing and releasing calcium ions in response to light. Sub-
stituent changes on the largely unexplored photochromic
bisbenzospiropyran scaffold led to significant changes in
thermal fading rates in several organic solvents. Excellent
correlations have been found between fading rates and
empirical Hammett constants as well as calculated ground-
state energies. These correlations can be used to improve
scaffold design.
This novel application requires a compound with both
suitable metal binding and appropriate photochemical
properties. Compound 1, a substituted 2,2′-spirobi[2H-
-
1-benzopyran] where R8,8′ ) -N(CH2CO2
) shown in
2
Figure 2, was designed to serve as a photochromic
calcium-selective chelator. The shape of this photochro-
mic scaffold uniquely allows multiple convergent chelat-
ing carboxylate ligands to be appended to it. The spiro-
pyran class of photochromic compounds undergoes
heterolytic cleavage from closed to open forms, and
photochromism is observable at a given temperature
unless the rate of thermal closure (k∆) is too fast.6,7 Prior
work by this group on 1 (R3,3′ ) -CH3; R8,8′ ) -N(CH2-
CO2-)2) showed moderately strong and selective binding
of calcium over magnesium in aqueous buffered systems,
but was not found to be photochromic under our experi-
mental conditions.8 Current efforts are directed toward
Calcium (Ca2+) is a second messenger in important
signal transduction systems that translate extracellular
signals into cellular response.1 Often, in both excitable
and nonexcitable cell types, the intracellular calcium
signal is oscillatory.2 Extensive experimental and theo-
retical work has been carried out to understand how the
oscillations arise, but much less is known about the
effects of Ca2+ oscillations on a molecular level. Under-
standing calcium signaling will lead to insights about
both normal and disease states in humans. Photorevers-
ible calcium-specific chelators appear to be an effective
approach to this area of research. Photochromic com-
pounds are well suited for controlled ion binding and
release because they undergo reversible structural changes
upon irradiation that can form or disrupt a binding
cavity. For reversible ion binding, the photochromic
molecule must interconvert between low affinity (B) and
high affinity (A) binding states. (Figure 1). The use of
light to control metal ion binding and release is well
(3) Winkler, J. D. In Biological Applications of Photochemical
Switches; Morrison, H., Ed; John Wiley and Sons: New York, 1993;
Chapter 3.
(4) (a) Kimura, K. Coord. Chem. Rev. 1996, 148, 41-61. (b) Des-
vergne, J.-P.; Bouas-Laurent, H.; Perez-Inestrosa, E.; Marsau, P.;
Cotrait, M. Coord. Chem. Rev. 1999, 185-186, 357-371. (c) Tanaka,
M.; Kamada, K.; Ando, H.; Kitagaki, T.; Shibutani, Y.; Kimura, K. J.
Org. Chem. 2000, 65, 4342-4347. (d) Itagaki, H.; Masuda, W.;
Hirayanagi, Y. Chem. Phys. Lett. 1999, 309, 402-406. (e) Takeshita,
M.; Irie, M. J. Org. Chem. 1998, 63, 6643-6649. (f) Stauffer, M. T.;
Knowles, D. B.; Brennan, C.; Funderburk, L.; Lin, F.-T.; Weber, S. G.
Chem. Commun. 1997, 287-288.
(5) (a) Filley, J.; Ibrahim, M. A.; Nimlos, M. R.; Watt, A. S.; Blake,
D. M. J. Photochem. Photobiol. A 1998, 117, 193-198. (b) Collins, G.
E.; Choi, L.-S.; Ewing, K. J.; Michelet, V.; Bowen, C.; Winkler, J. D.
Chem. Commun. 1999, 321-322. (c) Stauffer, M. T.; Weber, S. G. Anal.
Chem. 1999, 71, 1146-1151. (d) Wojtyk, J. T. C.; Kazmaier, P. M.;
Buncel, E. Chem. Commun. 1998, 1703-1704.
* Address correspondence to this author. Phone: 323-343-2362.
Fax: 323-343-6490.
(1) (a) Berridge, M. J.; Cobbold, P. H.; Cuthbertson, K. S. R. Philos.
Trans. R. Soc. London, Ser. B 1988, 320, 325-343. (b) Calcium as a
Cellular Regulator; Carafoli, E., Klee, C., Eds.; Oxford University
Press: New York, 1999. (c) Clapham, D. E. Cell 1995, 80, 259-268.
(2) Bootman, M. D.; Lipp, P.; Berridge, M. J. J. Cell Sci. 2001, 114,
2213-2222.
(6) Minkin, V. I. Chem. Rev. 2004, 104, 2751-2776.
(7) Bertelson, R. C. In Photochromism; Brown, G. H., Ed.; Tech-
niques of Chemistry, Vol. III.; Wiley-Interscience: New York, 1971;
pp 45-431.
(8) McCurdy, A.; Kawaoka, A. M.; Thai, H.; Yoon, S. Tetrahedron
Lett. 2001, 42, 7763-7766.
10.1021/jo051449j CCC: $30.25 © 2005 American Chemical Society
Published on Web 09/22/2005
J. Org. Chem. 2005, 70, 9067-9070
9067

TABLE 1. Substituted Bisbenzospiropyrans (1)
Synthesized
TABLE 2. Thermal Fade Rate Constants at -70 °C for
Selected Compounds 1 in Various Solvents
1a
1b
1c
1d
1e
1f
1g
1h
k∆ (min-1
toluene
)
R3,3′
R6,6′
R8,8′
H
OMe
H
H
Me
H
H
H
H
H
F
H
H
Cl
H
H
NO2
H
Me
OMe
H
Me
Me
H
MeCy
THF
iPrOH
6-OMe 7.75 × 10-3 4.20 × 10-3 2.19 × 10-3 1.65 × 10-3
6-Me
6-H
4.94 × 10-2 1.37 × 10-2 8.77 × 10-3 2.09 × 10-3
1.12 × 10-1 1.32 × 10-2 1.09 × 10-2 1.50 × 10-3
7.86 × 10-2 1.64 × 10-2 8.91 × 10-3 2.65 × 10-3
1.44 × 10-1 1.61 × 10-2 8.52 × 10-3 7.24 × 10-3
N/A
1i
1j
1k
1l
1m
1n
1o
6-F
R3,3′
R6,6′
R8,8′
Me
H
H
Me
NO2
H
Ph
Me
H
Ph
H
H
H
H
Me
H
H
F
6-Cl
1f-l
H
8-OMe 8.32 × 10-2
H
OMe
8-Me
8-F
1.88 × 10-2
9.27 × 10-2
showed related compounds substituted at R3,3′ were in
fact photochromic at room temperature in toluene.11 The
different intensity of the light sources may explain these
different observations. Also, the presence of a nitro group
(1f) caused an immediate change in the visible spectrum
upon irradiation, suggesting very efficient photocolora-
tion, but almost as quickly the signal disappeared, even
while under irradiation. This is consistent with the
observation that nitro groups improve quantum yields,
but also can increase k∆ and side reactions.12
More polar solvents lower the rate of thermal closure
up to 100-fold (graphical representation of these data is
in the Supporting Information). This effect is consistent
with more polar solvents better stabilizing the more polar
open form of the molecules. In addition, more polar
solvents decrease the sensitivity of k∆ to substituents.
This trend has been observed for other photochromic
spiropyrans,13 with a notable exception. One group has
reported that substituent effects are negligible in non-
polar solvents, but strongly present in polar solvents.14
They cited a change in ground-state structure and
mechanism due to solvation effects. In nonpolar solvents,
a quinoid resonance form was dominant, following the
substituent-insensitive electrocyclic ring closure pathway.
In polar solvents, the zwitterionic resonance form was a
major contributor, with large substituent effects on
closure rates. The data in Table 2 do not suggest a
mechanism change due to either substituent or solvent
effects.
FIGURE 3. UV-vis spectra showing the photochemical
opening of 1b in methylcyclohexane at -70 °C with UV light
(left, during 6 min of irradiation) and thermal fading (right,
during 200 min of darkness).
a systematic investigation of the tunability of photochro-
mic properties of bisbenzospiropyran, given the limited
structure-activity data published for this photochromic
scaffold. In this report, the effect of changing substituents
on k∆ is examined. These efforts will aid in assessing the
viability of this photochromic scaffold for use in a
calcium-signaling mimic.
The compounds listed in Table 1 were prepared ac-
cording to literature methods9 as described in the Sup-
porting Information.
Representative photochromic behavior is illustrated by
Figure 3. After reaching a photostationary state upon UV
irradiation at -70 °C, (thermal) fading was monitored
in darkness at -70 °C. This temperature was chosen so
that kinetics of the fastest reaction monitored were slow
relative to the instrument scan rate. Thermal fading
follows first-order kinetics at shorter, but not at longer
times. This complex kinetic behavior is also characteristic
of many indolino-spiropyrans.7,10 Table 2 lists first-order
rate constants analyzed at short times (first 10% of
experiment). As discussed in more detail below, these
data show both solvent dependence and substituent
effects on thermal fading rates. Qualitatively, the more
polar solvent and the more electron donating substituent
slow the rate of thermal fading. Higher temperatures
increase the rate of thermal closure (data not shown) so
that photocoloration is not observed at room temperature
for these compounds.
The symmetric substitution of the bisbenzospiropyran
decreases rates of thermal closure up to 20-fold. These
effects, while small, are significant. The same substituent
at R6 and R6′ should have both stabilizing and destabiliz-
ing effects on the zwitterionic open form, which might
be expected to cancel out completely. Interestingly, they
do not.
Substituent effects have been studied on related
spirobenzopyrans with indoline and benzothiazoline het-
erocyclic systems.7,11,15 In some cases, structural modifi-
(11) Appriou, P.; Trebaul, C.; Brelivet, J.; Garnier, F.; Guglielmetti,
R. Bull. Soc. Chim. Fr. 1976, 2039-2045.
(12) Malatesta, V. In Organic Photochromic and Thermochromic
Compounds; Crano, J. C., Guglielmetti, R. J., Eds; Plenum Press: New
York, 1999; Vol. 2, Chapter 2.
The determination of k∆ was not possible for certain
compounds (1f-l). For example, substituents at R3,3′
prohibited photocoloration under these conditions. In
contrast, prior work by Appriou et al. with a flashlamp
(13) (a) Samat, A.; Metzger, J.; Mentienne, F.; Garnier, F.; Duboius,
J. E.; Guglielmetti, R. J. Phys. Chem. 1972, 76, 3554-3558. (b)
Vandewyer, P. H.; Hoefnagels, J. H.; Smets, G. Tetrahedron 1969, 25,
3251-3266. (c) Keum, S.; Hur, M.; Kazmaier, P. M.; Buncel, E. Can.
J. Chem. 1991, 69, 1940-1947. (d) Wojtyk, J. T. C.; Wasey, A.;
Kazmaier, P. M.; Hoz, S.; Buncel, E. J. Phys. Chem. A 2000, 104, 9046-
9055.
(9) (a) Bergmann, E. D.; Weizmann, A.; Fischer, E. J. Am. Chem.
Soc. 1950, 72, 5009-5012. (b) Inbasekaran, M. Heterocycles 1993, 35,
1121-1123.
(10) Delbaere, S.; Micheau, J.; Vermeersch, G. J. Org. Chem. 2003,
68, 8968-8973.
(14) Keum, S.; Lee, K. Bull. Korean Chem. Soc. 1993, 14, 16-18.
9068 J. Org. Chem., Vol. 70, No. 22, 2005

TABLE 3. Relative Energies (kcal/mole) Calculated for
Closed and Open Forms of Bisbenzospiropyrans
1a
1b
1c
1d
1e
closed
ctc
0
0
0
0
0
4.85
5.93
6.41
7.18
12.63
13.58
14.64
14.47
6.09
7.51
8.08
8.15
14.45
15.38
15.71
15.74
7.16
8.68
9.08
9.24
15.51
16.54
16.66
16.70
7.27
8.35
8.89
9.37
14.89
16.37
16.81
16.51
7.37
8.82
9.37
9.48
15.22
16.79
16.80
16.57
ctt
FIGURE 4. Substituent effects in positions A, B, and C on
k∆. The heterocyclic ring is indoline or benzothiazoline.
ttt
ttc
cct
tct
tcc
ccc
rich ring. This is consistent with the observation that
even symmetric substitution influences k∆. When sub-
stituents at R8,8′ are included (1m-o), the best fit
requires the ES (Taft) substituent parameter for position
R
8,8′, and the overall fit decreases (R ) 0.968). The best
equation is log k ) (2.155) × (σp+ or ES) - (1.452) × (σp-
or ES) - (0.812).
A correlation between initial rates of thermal closure
of 1a-e in methylcyclohexane and calculated quantities
was also attempted. Density functional calculations
(B3LYP/6-31G*) were performed to obtain ground-state
energies of both closed spiro and open planar, conjugated
forms (Table 3). Closed forms were found to be more
stable than open forms by approximately 5-7 kcal/mol.
The open forms exist as eight possible stereoisomers.
Each of the three bonds of the chain linking the aromatic
rings may be cisoid (“c”) or transoid (“t”). For compounds
1a-e, the lowest energy open stereoisomer is “ctc”, shown
in Figure 2. Illustrations of all eight stereoisomers
including stereochemical designations are included in the
Supporting Information.
Unlike other reports,15b,c no correlation was found here
between log k∆ and any calculated bond length, bond
length alternation in the open form, HOMO-LUMO gap
of closed or open forms, or dipole change. However, there
was excellent correlation between (Eclsosed - Eopen) and
log k∆ (R ) 0.996-0.953 for the four most stable transoid
isomers of each compound). These plots are included in
the Supporting Information. As substituents stabilize the
open form relative to the less polar transition state, the
kinetic barrier to thermal closure increases.
FIGURE 5. For compounds 1a-e, experimental log k∆ (closed
circles) compared to predicted values (open circles) obtained
from a dual parameter Hammett plot (multiple R ) 0.999
when σp+ and σp- parameters were used).
cations have resulted in dramatic changes of thermal
fading rate constants of 3-4 orders of magnitude in both
nonpolar and polar solvents. Generally, as the open form
is stabilized electronically by substituents, k∆ decreases.
A summary of specific substituent effects on k∆ is
illustrated with Figure 4. Electron-donating substituents
decrease k∆ in positions A and B, while electron-
withdrawing substituents decrease k∆ in position C.
These substituent effects were found to be additive. A
dual parameter Hammett plot for kinetic data of a large
number of indolinospirobenzopyrans yielded the following
equation: log k ) (1.727) × (Σσindoline) - (1.703) × (Σσpyran
)
- 1.372 (R ) 0.937).7
For purposes of comparison, an additive empirical
Hammett-type plot of the form a × σ + b × σ + c was
made here for bisbenzospiropyran. The slope a should
be positive for the electron-poor ring and the slope b
should be negative for the electron-rich ring. By using
σp+ and σp-, an excellent correlation (R ) 0.999) was
obtained with the equation log k ) (1.73) × (σp+) - (0.653)
× (σp-) - (0.914). Figure 5 shows predicted and observed
values for each compound tested. Additional plots of
predicted and observed log k∆ versus each independent
variable (σpara+ and σpara-) are provided in the Supporting
Information. A comparison of the slopes of the dual
parameter Hammett plots reveals that the bisbenzospiro-
pyran system studied here appears to have a larger
difference in sensitivity to substituents between the two
rings than the indolinospirobenzopyran system has. For
1a-e, the substituent effect on the electron-deficient ring
is significantly more important than that for the electron-
This work demonstrates that both solvent and substit-
uents can be used to tune the photochromism of 2,2′-
spirobi[1H-1-benzopyran]. To mimic biological calcium
signals at physiological temperatures, the original de-
signed chelator 1 where R3,3′ ) CH3; R6,6′ ) H; R8,8′
)
N(CH2CO2-)2 must be modified significantly. The present
study also demonstrates that the synthetically simple
symmetric substitutions used here are not sufficient.
However, we have shown that either computational or
empirical Hammett correlations can be used reliably to
design a more appropriate scaffold. For example, the
Hammett equation predicts that if R6 ) NMe2 and R6′ )
CN, the molecule’s thermal closure rates will be ap-
proximately 500-fold slower than the slowest rate re-
ported here. If water solvation does not overwhelm
substituent effects and if photocoloration remains ef-
ficient, these new scaffolds should allow the observation
of photochromism under physiological conditions. The
(15) (a) Bertelson, R. C. In Organic Photochromic and Thermochro-
mic Compounds; Crano, J. C., Guglielmetti, R. J., Eds; Plenum Press:
New York, 1999; Vol. 1, Chapter 1. (b) Sheng, Y.; Leszczynski, J.;
Garcia, A.; Rosario, R.; Gust, D.; Springer, J. J. Phys. Chem. B 2004,
108, 16235-16243. (c) Kawanishi, Y.; Seki, K.; Tamaki, T.; Sakuragi,
M.; Suzuki, Y. J. Photochem. Photobiol. A 1997, 109, 237-242.
J. Org. Chem, Vol. 70, No. 22, 2005 9069

rated under reduced pressure. The resulting crude product was
subjected to flash column chromatography on silica gel. The
product was eluted with 15% diethyl ether in hexane. Yield
17.10%.
optimized photochromic chelator will be used to gain
insight into complex calcium signaling on a molecular
level.
Acid-catalyzed route: 5-Methoxysalicyladehyde (0.5 mL, 4.0
mmol), acetone (0.30 mL, 4.04 mmol), and a stir bar were added
to a round-bottom flask containing anhydrous ethanol (10 mL).
The reaction mixture was cooled with the help of an ice bath
and hydrochloric acid gas was bubbled through the reaction for
2 h. The reaction mixture was allowed to come to room
temperature and stirred for 24 h. After 24 h, 5-methoxysalicy-
ladehyde (0.5 mL, 4.0 mmol) was added to the reaction mixture,
cooled with an ice bath, and HCl (g) was bubbled through the
reaction for another 2 h. The following day, the reaction was
extracted with methylene chloride several times. The organic
layer was dried over MgSO4. The crude material was subjected
to flash column chromatography on silica gel. The product was
eluted with 10% ethyl acetate in hexane. Yield 31%.
Experimental Section
Photochemistry: Light from a 200-W Mercury Xenon lamp,
filtered by water and a 320 nm band-pass filter, was directed
via a 3-ft quartz fiber optic bundle into the top of a reaction
vessel. The irradiated, stirred samples were cooled, using a
thermostated cryobath, to various temperatures. UV-vis obser-
vations were made with a Hellma 10-mm immersion probe.
Computation: The program Spartan ‘04 was used. Closed
forms were subjected to geometry optimization (B3LYP/6-31G*)
and the resulting geometry was used to determine energy. For
open forms, two isomers (“ttt” and “tcc”) were used as starting
points to generate a conformer distribution (four isomers in each
distribution). A figure defining the structures of the open forms
is included in the Supporting Information. For compounds with
methoxy substituents, AM1 was used to generate the conformer
distribution. The conformers were subjected to geometry opimi-
zation (B3LYP/6-31G*) and then single point energy determi-
nation. Cartesian coordinates and energies are listed after
synthetic and spectral data of the Supporting Information.
6,6′-Dimethoxy-2,2′-spirobi[2H-1-benzopyran] (1a). The
following are representative procedures for syntheses of the
molecules studied here.
1H NMR (300 MHz, CDCl3): δ 6.83-6.75 (m, 8H), 6.01-5.98
(d, J ) 9 Hz, 2H), 3.79 (s, 6H). 13C NMR (75 MHz, CDCl3): δ
154.6, 145.0, 126.8, 122.6, 120.4, 117.6, 115.9, 111.9, 95.3, 56.0.
HRMS (DEI) (m/z): calcd M+ 308.1150; found 308.1043 (M+).
Acknowledgment. This work was supported by a
generous grant from the National Institutes of Health
(NIGMS MBRS S06 GM08101).
Base-catalyzed route: 5-Methoxysalicyladehyde (1.22 g, 8.02
mmol), acetone (0.3 mL, 4.04 mmol), 4 N NaOH (4 mL, 16 mmol),
and ethanol (5 mL) were stirred in a 25-mL round-bottom flask
at room temperature for 12 h. The resulting reaction mixture
was added to water and neutralized with 5% HCl (25 mL). The
precipitate was extracted into ethyl acetate, which was dried
over anhydrous MgSO4 and evaporated under reduced pressure.
The residue was dried under vacuum and added to ethoxyethyl
acetate (10 mL) in a 25-mL round-bottom flask. The resulting
solution was refluxed for 1 h. The reaction was monitored by
TLC, which indicated completion of reaction after 1 h. The
reaction mixture was cooled and poured in water and extracted
into diethyl ether. The organic layer was washed twice with 1
M NaOH. The organic layer was dried over MgSO4 and evapo-
Supporting Information Available: General informa-
tion, analytical and spectral characterization data, graph of
log k∆ vs ET, calculated (Espiro - Eopen) versus log k∆ in
methylcyclohexane at -70 °C for compounds 1a-e, alternate
representations of dual parameter Hammett plots, nomencla-
ture guide for open forms of bisbenzospiropyrans, Cartesian
coordinates of calculated lowest energy geometries of all open
forms of bisbenzospiropyrans 1a-e, proton NMR spectra for
compounds whose rate data are reported in this work. This
material is available free of charge via the Internet at
http://pubs.acs.org.
JO051449J
9070 J. Org. Chem., Vol. 70, No. 22, 2005