Analysis of Chemical Equilibrium of Silicon ... - ACS Publications

Aug 3, 2015 - Analysis of Chemical Equilibrium of Silicon-Substituted Fluorescein and Its Application to Develop a Scaffold for Red Fluorescent Probes...
1 downloads 5 Views 1MB Size
Page 1 of 10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Analysis of Chemical Equilibrium of Si-substituted Fluorescein and its Application to Develop a Scaffold for Red Fluorescent Probes Kazuhisa Hirabayashi,† Kenjiro Hanaoka,*,† Toshio Takayanagi,‡ Yuko Toki,† Takahiro Egawa,† Mako Kamiya,§,ǂ Toru Komatsu,†, ǂ Tasuku Ueno,† Takuya Terai,† Kengo Yoshida,║ Masanobu Uchiyama,†,║ Tetsuo Nagano,┴ Yasuteru Urano*,†,§, ¶ †

Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan Department of Life System, Institute of Technology and Science, The University of Tokushima, 2-1 Minami-josanjima, Tokushima 770-8506, Japan ┴ Open Innovation Center for Drug Discovery, The University of Tokyo, Tokyo 113-0033, Japan § Graduate School of Medicine, The University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan ║ Elements Chemistry Laboratory, RIKEN, and Advanced Elements Chemistry Research Team, Riken Center for Sustainable Resource Science (CSRS), 2-1 Hirosawa, Wako-shi, Saitama 351-0198, Japan ǂ PREST and ¶CREST, JST, Saitama 332-0012, Japan ‡

ABSTRACT: Fluorescein is a representative green fluorophore that has been widely used as a scaffold of practically useful green fluorescent probes. Here, we report synthesis and characterization of a Si-substituted fluorescein, i.e., 2-COOH TokyoMagenta (2COOH TM), which is a fluorescein analogue in which the O atom at the 10’ position of the xanthene moiety of fluorescein is replaced with a Si atom. This fluorescein analogue forms a spirolactone ring via intramolecular nucleophilic attack of the carboxylic group in a pH-dependent manner. Consequently, 2-COOH TM exhibits characteristic large pH-dependent absorption and fluorescence spectral changes: (1) 2-COOH TM is colorless at acidic pH, whereas fluorescein retains observable absorption and fluorescence even at acidic pH, and the absorption maximum is also shifted; (2) the absorption spectral change occurs above pH 7.0 for 2COOH TM and below pH 7.0 for fluorescein; (3) 2-COOH TM shows a much sharper pH response than fluorescein because of its pKa inversion, i.e., pKa1 > pKa2. These features are also different from those of a compound without the carboxylic group, 2-Me TokyoMagenta (2-Me TM). Analysis of the chemical equilibrium between pH 3.0 and 11.0 disclosed that 2-COOH TM favors the colorless and non-fluorescent lactone form, compared with fluorescein. Substitution of Cl atoms at the 4’ and 5’ positions of the xanthene moiety of 2-COOH TM to obtain 2-COOH DCTM shifted the equilibrium so that the new derivative exists predominantly in the strongly fluorescent open form at physiological pH (pH 7.4). To demonstrate the practical utility of 2-COOH DCTM as a novel scaffold for red fluorescent probes, we employed it to develop a probe for β-galactosidase.

INTRODUCTION Fluorescein (Figure 1a) was first synthesized by Adolf von Baeyer in 1871, and is currently one of the most widely used fluorophores1 due to its high water solubility, high fluorescence quantum yield and high molar extinction coefficient (Φf = 0.85, ε490 = 8.8×104 cm−1M−1).2,3 For example, fluorescein derivatives have been used as fluorescent tags for biomolecules such as proteins and DNA in biological studies,4,5 as pH sensors based on their pH-dependent absorption and fluorescence change in analytical science,2,6 and diagnostic agents for fundoscopy in ophthalmology.7,8 Fluorescein is also used as a platform for many kinds of fluorescent probes, including Ca2+ indicators such as Fluo-3, Fluo-4, Calcium Green-1 and Oregon Green 488 BAPTA-1.9-21 The chemical structure of fluorescein consists of two parts, i.e., the benzene moiety and the xanthene moiety (the fluorophore), as shown in Figure 1a, and these two parts are orthog-

onal to each other.9 Interestingly, fluorescein shows a complex chemical equilibrium involving tautomerism in aqueous solutions between pH 3.0 and 11.0, as shown in Figure 1b,22,23 because the presence of the carboxylic group at the 2-position of the benzene moiety makes it possible for the molecule to form an intramolecular spirolactone.24 As a consequence of this equilibrium, large spectral (absorption and fluorescence) changes are observed in response to pH change. The apparent pKa values of this equilibrium are as follows: neutralmonoanion pKa1 = 4.2, and monoanion-dianion pKa2 = 6.4 (Figure 1b).23 Moreover, strongly fluorescent fluorescein becomes colorless and non-fluorescent after intramolecular spirolactone formation, owing to disruption of the π-conjugation system of the xanthene moiety (Figure 1b). This off/on fluorescence change has been utilized in the design of various fluorescent probes, including probes for esterase,11 βgalactosidase,12 alkaline phosphatase,12,13 hydrogen sulfide,14

1 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

and sulfane sulfur.15 The chemical and photophysical properties of fluorescein are unique and have been studied by many researchers.2,6,22,23,25-30 We recently reported a novel red fluorophore, TokyoMagenta (TM), in which the O atom at the 10’ position of the xanthene moiety of fluorescein is replaced with a Si atom, causing the absorbance and fluorescence wavelengths to be shifted to 90 nm longer wavelength than those of fluorescein.31 We showed that TM is a useful red-emitting fluorophore that retains many of the advantages of fluorescein mentioned above, because of the similarity of its chemical structure to that of fluorescein. However, although fluorescein has a carboxylic group at the 2-position of the benzene moiety, only TM derivatives with a methyl group or a spirothiophene at the 2 position of the benzene moiety have been reported so far.31-33 Here, we describe the synthesis of 2-COOH TM, which has a carboxylic group at the 2-position of the benzene moiety (Figure 1a). We found it exhibits a unique prototropic equilibrium that is suitable for the development of red-florescent probes. As proof-of-principle, we designed, synthesized and evaluated a red-fluorescent probe for β-galactosidase.

Page 2 of 10

fluorescence intensity of 2-COOH TM decreased as the pH was lowered and the solution was completely colorless and non-fluorescent below pH 7.0 (Figure 2a and b), whereas a solution of fluorescein showed strong absorption in the visible region and fluorescence throughout the pH range of 3.0 to 7.0 (Figure 2a and c). Moreover, 2-COOH TM showed a sharper pH response than did fluorescein (Figure 2d). These properties, while different from those of fluorescein, are similar to those of phenolphthalein34 and carbofluorescein35, though the reason for this is unclear. Table 1. Photophysical Properties of 2-COOH TM and fluorescein

ε

λabs (nm)

λem (nm)

(M−1cm−1)

2-COOH TM a

580

598

1.1 × 105

0.38 c

Fluorescein

492

511

8.8 × 104 b

0.85 d

2-Me TM

582

598

1.1 × 105 e

0.42 e

Φfl

Measured in 100 mM sodium phosphate buffer at pH 9.0 in the presence of 1% DMSO as a cosolvent. b Ref. 2. c The fluorescence quantum yield (Φfl) of 2-COOH TM was calculated using that of TokyoMagenta (Φfl = 0.42) in 100 mM sodium phosphate buffer at pH 9.0 as a standard.31 d Ref. 3. e Ref. 31. a

Figure 1. (a) Chemical structures of fluorescein and 2-COOH TM. (b) Chemical equilibrium of fluorescein between pH 3.0 and 11.0 according to ref. 23.

RESULTS AND DISCUSSION Photochemical Properties of 2-COOH TM. 2-COOH TM was synthesized using a modification of the procedure described in our previous report (Scheme S1).31 We examined its photophysical properties in 100 mM sodium phosphate buffer at pH 9.0 and compared them with those of fluorescein (Table 1). 2-COOH TM emitted red fluorescence (λem = 598 nm), and also showed sufficiently large values of molar absorption coefficient and fluorescence quantum yield for practical use in fluorescence imaging applications. As shown in Figure 2a, the absorption and fluorescence intensity of 2-COOH TM showed a strong pH dependency, but interestingly, its behavior was different from that of fluorescein; both the absorption and

Figure 2. Comparison of the pH-dependencies of 2-COOH TM and fluorescein. (a) Photographs of solutions of 2-COOH TM and fluorescein under white light or under irradiation with a handy UV lamp (λem = 365 nm) at various pH values between 3.0 and 11.0. (b) Absorption and emission spectra of 1 µM 2-COOH TM (λex = 580 nm) at various pH values. (c) pH-Dependency of absorption spectra of 1 µM fluorescein. (d) pH Plot of normalized absorbance at 580 nm for 2-COOH TM (red) and at 494 nm for fluorescein (green). Measurements were made at 25°C in 100 mM

2 ACS Paragon Plus Environment

Page 3 of 10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

sodium phosphate buffer at various pH values in the presence of 1% DMSO as a cosolvent.

Elucidation of Chemical Equilibrium of 2-COOH TM. To investigate the reason for the difference in pH dependency between 2-COOH TM and fluorescein, we examined the pHdependency of the absorbance change of 2-COOH TM at 580 nm (Figure 2d). Initially, to determine the apparent aciddissociation constant, the plot was fitted to a monophasic equation; however, based on the χ2 values, we found that biphasic dissociation (χ2 = 0.0021) was more appropriate than monophasic dissociation (χ2 = 0.036) to fit the sharp pH response of 2-COOH TM (Figure S1). Interestingly, 2-COOH TM showed a larger value of pKa1 (8.2) than pKa2 (7.9). pKa1 is usually smaller than pKa2 for successive acid-dissociation processes, because removal of the second proton from the monoanion form is harder than that of the first proton from the neutral form, as exemplified by succinic acid (pKa1 = 4.21 < pKa2 = 5.64), adipic acid (pKa1 = 4.42 < pKa2 = 5.41), and phthalic acid (pKa1 =2.95 < pKa2 = 5.41).36 Such a reversal of pKa values has not been observed in previous research on fluorescein derivatives,2,6,22,23,25-30 or even phenolphthalein derivatives34,37, with only a few exceptions38,39. In principle, the closer the two dissociation constants are, the sharper the pH response is; in other words, this reversal of pKa values accounts for the sharpness of the pH response of 2-COOH TM. To confirm this interpretation, we examined the absorbance change in detail. There was no absorption spectral change throughout the pH range of 3.0 to 7.0 (Figure 2a and b), and this indicates that two dissociation steps above pH 7.0 should be considered. We also determined the acid dissociation constants by means of an equilibrium analysis method utilizing capillary zone electrophoresis (Figure S2).37,40 The obtained values were pKa1 = 8.4 and pKa2 = 7.9, and these results supported the putative reversal of pKa values, despite of the slight difference from the pKa values obtained from curve fitting of the absorbance change; this may be due to the difference in solvent conditions. Moreover, the normalized electrophoretic mobility was almost 0 at pH 7.0. This indicates that 2-COOH TM exists in neutral form with no net charge below pH 7.0, and that the two apparent dissociation constants above pH 7.0 are neutral-monoanion pKa1 and monoanion-dianion pKa2. We next investigated the nature of the above three species (neutral form, monoanion, and dianion) present in aqueous solutions between pH 3.0 and 11.0. Two or three potential tautomeric forms (lactone, carboxylate anion, and carboxylic acid) based on the form of the carboxy group at the 2-position of the benzene moiety can be drawn for each species of 2COOH TM, as in the case of fluorescein derivatives (Figure 3a).2,6,22,23,25-30 We first synthesized 2-Me TM,31 which has the same xanthene moiety as 2-COOH TM but cannot form an intramolecular spirolactone ring, and examined its absorption spectra. The absorption of 2-Me TM also showed pH dependency, but its behavior was quite different from that of 2COOH TM: a solution of 2-Me TM showed a typical monophasic absorption change with pKa = 6.8, and the neutral or anion form showed the absorption maximum at 472 or 582 nm, respectively, as previously reported by our group (Figure 3b).31 As shown in Figure 3c, the pH response of 2-COOH TM was also sharper than that of 2-Me TM. Thus, the reason why 2COOH TM shows no absorption in the visible region below pH 7.0 appears to be that 2-COOH TM exists predominantly in the neutral form as the intramolecular spirolactone form NI

between pH 3.0 and 7.0 (Figure 3a). Interestingly, no difference in the sharpness of the pH response was observed between fluorescein and 2-Me TokyoGreen,9 which is

Figure 3. (a) Possible prototropic forms of 2-COOH TM. (b) Chemical equilibrium of 2-Me TM and pH-dependency of absorp-

3 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

tion spectra of 1 µM 2-Me TM at 25°C in 100 mM sodium phosphate buffer containing 1% DMSO as a cosolvent. (c) pH Plot of normalized absorbance at 580 nm for 2-COOH TM (red) and at 582 nm for 2-Me TM (orange). Measurements were made at 25°C in 100 mM sodium phosphate buffer at various pH values containing 1% DMSO as a cosolvent. (d) Theoretical curves of molar fractions of each form of 2-COOH TM assuming pKa1 = 8.2 and pKa2 = 7.9; points (●) indicate observed pH-dependent change of normalized absorbance at 580 nm of 2-COOH TM. (e) Estimated chemical equilibrium of 2-COOH TM Me ether. (f) Absorbance spectra of 1 µM 2-COOH TM Me ether at various pH values (25°C) in 100 mM sodium phosphate buffer containing 1% DMSO as a cosolvent. (g) Estimated chemical equilibrium of 2COOH TM. K = [MII] / [MI].

a fluorescein analogue with a methyl group at the 2-position of the benzene moiety. The pH dependency of the absorption spectrum of 2-Me TokyoGreen with pKa = 6.3 (Figure S3) is similar to that of fluorescein with pKa2 = 6.4 (Figure 1b and 2c). We also measured the 1H NMR spectrum of 2-COOH TM in DCl/D2O at pD = 3.4 to confirm that this compound was not decomposed under acidic conditions. The sample used for this NMR measurement was colorless and the 1H NMR signals of the xanthene moiety were symmetrical and could be assigned to the NI neutral form (Figure S4). Next, we considered the form of 2-COOH TM above pH 10.0. As shown in Figure 3a, two dianion forms, DI and DII, can exist in this pH range. The absorption maximum of 2COOH TM above pH 10.0 was 580 nm, similar to that of 2Me TM (582 nm), and the ɛ value of 2-COOH TM above pH 10.0 was 110,000 (580 nm), which is almost the same as that of 2-Me TM.31 We further measured the 1H NMR spectrum of 2-COOH TM in NaOD/D2O at pD = 12.1 (Figure S5). The sample for this NMR measurement was red-colored, as was the solution of 2-Me TM. Examination of the 1H NMR spectrum of 2-COOH TM indicated that the signals of its xanthene moiety were similar to those of 2-Me TM. Therefore, we concluded that above pH 10.0, the dianion exists almost wholly in DI form. Finally, we examined the form of 2-COOH TM between pH 7.0 and 10.0. Three monoanion forms, MI, MII and MIII, can be considered, shown in Figure 3a. However, MIII was assumed to represent a very minor species because the sitespecific pKa value2,41 of the hydroxyl group of the xanthene moiety of TM derivatives (pKa = 6.8 for 2-Me TM31) is estimated to be larger than that of carboxylic group at the 2position of the benzene moiety (pKa = 4.2 for benzoic acid36 and pKa = 3.1-3.4 for fluorescein2) by about 3 pH units, indicating that the MIII species should represent about 0.1% of the monoanion quinoid species (MII and MIII). This assumption is also based on the fact that a carboxy group at the 2-position of the benzene moiety of 2-COOH TM has little effect on the site-specific pKa value of the hydroxyl group of the xanthene moiety of TM derivatives. Indeed, we synthesized 2-MeO TM (Scheme S4) and 2-NO2 TM (Scheme S5, Figure S7a), and determined their pKa values (Figure S7b-d) as 6.9 and 6.8, respectively (almost the same as that of 2-Me TM: pKa = 6.8). Indeed, when we simulated the existence ratio of each species (neutral, monoanion and dianion forms) by using pKa1 = 8.2 and pKa2 = 7.9 (determined from the absorbance spectral change as described above), the absorbance change of 2COOH TM at 580 nm corresponded to the existence ratio of the dianion form (Figure 3d). Although the existence ratio of

Page 4 of 10

the monoanion form was calculated to be low in any pH region, such a well correspondence suggests that MIII appears not to exist to any significant degree in this pH region, because MIII should possess absorption at 580 nm, like 2-Me TM. Thus, we focused on the other two forms, MI and MII, as potential forms of the monoanion. We designed and synthesized 2-COOH TM Me ether to examine the chemical equilibrium between the neutral form and the monoanion form, with reference to a previous report on the chemical equilibrium of fluorescein,22 and we measured its absorption spectra at various pH values between pH 3.0 and 11.0. 2-COOH TM Me ether showed no absorbance in the visible region below pH 8.0. On the other hand, a small absorbance change at 475 nm, which could be fitted to a monophasic equation, was observed between pH 8.0 and 11.0 (Figure 3e, f, and Figure S6a). The chemical equilibrium of 2-COOH TM Me ether was also analyzed by means of capillary zone electrophoresis. The electrophoretic mobility change of 2-COOH TM in response to pH change of the migrating solution was almost double that of 2COOH TM Me ether, which supports the idea that 2-COOH TM Me ether exists as a monoanion above pH 10.0, while 2COOH TM exists as a dianion above pH 10.0 (Figure S6b). Moreover, in comparison to the neutral form of 2-Me TM (ɛ = 29,000),31 ɛ of 2-COOH TM Me ether was 5,200 even above pH 11.0 (Figure 3f). This indicates that 2-COOH TM Me ether monoanion exists in form I and form II, as shown in Figure 3e, and the monoanion of 2-COOH TM is also likely to have tautomeric forms. On the basis of the above results, we considered that the chemical equilibrium of 2-COOH TM might be as shown in Figure 3g.

Figure 4. Extended chemical equilibrium of 2-COOH TM between monoanion and dianion forms for calculation of the K value (= [MII] / [MI]) from the apparent pKa (Figure S8) and sitespecific pKa values.

For determination of the equilibrium constant K (= [MII] / [MI]) (Figure 3g), we used the relationship between apparent pKa values and site-specific or microscopic pKa values (Figure S8).2,41 When tautomers with site-specific pKa values are in chemical equilibrium (Figure S8a), the apparent pKa (Figure S8b) can be calculated from site-specific pKa values by employing the equation shown in Figure S8c. The extended chemical equilibrium model for 2-COOH TM derivatives (Figure S8) can be written as shown in Figure 4. We also determined the apparent pKa2 as 7.9 (Figure 3g) by curve fitting with the biphasic equation. Then, we calculated K (= [MII] / [MI]) (Figure 3g, Figure 4) from the following equation, which corresponds to Figure S8c (DI is thought to be the only dianion form of 2-COOH TM (Figure 3g), i.e., [DI] / ([DI] + [DII]) ≃ 1 in Figure 4):

4 ACS Paragon Plus Environment

Page 5 of 10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

pKa2 = pK22a2 − log K/(1+K) where pKa2 = 7.9; pK22a2 is the site-specific pKa value between the MII and DI forms. We also assumed that pK22a2 is equal to the pKa value of 2-Me TM (pKa = 6.8). Calculation of the equilibrium constant K (= [MII] / [MI]) (Figure 3g) gave a value of 0.09, and it was concluded that MI exists in 11-fold excess over MII, i.e., most of the monoanion form takes the lactone structure, which has weak absorption in the visible region. Thus, in summary, the neutral form of fluorescein exists as multiple tautomers, whereas the neutral form of 2-COOH TM exists predominantly as the colorless lactone structure, while the monoanion form exists as multiple tautomers. The reason for this is probably that 2-COOH TM forms an intramolecular spirolactone more readily than fluorescein. In the TM structure, the O atom at the 10’ position of the xanthene chromophore of fluorescein is replaced with a Si atom, and it is considered that the redshift of TM fluorophores compared with fluorescein is due to σ*-π* conjugation involving the Si atom, similar to that in Si-containing pyronine.42,43 Indeed, DFT (B3LYP/6-31+G*) calculations strongly support this interpretation, i.e., (1) the HOMO-LUMO gap of the TM scaffold is narrower than that of fluorescein, and (2) the LUMO energy level of the TM scaffold (largely localized on the xanthene moiety) is wellstabilized compared with that of fluorescein (Figure S9). The low-lying LUMO orbital, in which the 9’ position of the xanthene moiety has the highest orbital coefficient, may account for the easy spirolactone formation of 2-COOH TM. These results are consistent with the idea that 2-COOH TM predominantly exists as the colorless lactone in the neutral and the monoanion forms, and exhibits tautomerization of the monoanion form. These properties are probably mainly related to the relationship between pKa1 and pKa2, i.e., pKa1 > pKa2. Such pKa inversion has been reported for fluorescein in the presence of organic solvent (for example, water/DMSO = 1 : 9).38,39 Similarly, the two pKa values of phenolphthalein are very close each other under aqueous conditions (pKa1 = 9.1 and pKa2 = 9.5), though pKa2 is slightly larger than pKa1.34,37 Phenolphthalein tends to form a colorless spirolactone ring in the neutral and monoanion forms, and shows tautomerization of the monoanion form, like 2-COOH TM (Figure S10). Moreover, the dramatic pH-dependent fluorescence increase of 2-COOH TM due to the pKa inversion suggests that 2COOH TM might be useful as a pH indicator, like pH-sensors based on pH sensitive polymer44 or pH-induced micellization45. Indeed, 2-COOH TM showed a 10-fold absorption increase when the pH was changed from pH 7.5 to pH 8.5, whereas 2Me TM showed 5-fold absorption increase when the pH was changed from pH 6.0 to pH 7.0. Development of a Fluorescent Probe for β -Galactosidase. We have previously reported a design strategy for fluorescent probes operating in the red wavelength region based on the large absorption spectral change of 2-Me TM upon deprotonation of the hydroxyl group of the xanthene moiety, and we developed a fluorescent probe for β-galactosidase, 2-Me TM βgal.31 Probes based on 2-Me TM, however, have a large absorption at around 450 nm and their application for dual-color imaging is problematic because of the inner-filter effect. However, 2-COOH TM has a greater preference for the colorless and non-fluorescent intramolecular spirolactone form compared with fluorescein, and might not suffer from this limitation. Therefore, we set out to develop off/on fluorescent probes by utilizing this characteristic of 2-COOH TM (Figure

5). Molecular design of fluorescent probes based on fluorescein is based on the idea that the hydroxyl groups at the 3’ and 6’ positions can be modified with enzyme substrates to afford the colorless and non-fluorescent lactone form, as shown in Figure 5a, and then enzymatic reaction releases a strongly fluorescent product. However, a two-step reaction is required to generate strong fluorescence, and the intermediate in which only one of the substrate moieties is cleaved by the enzyme is still weakly fluorescent.11-13 Therefore, such probes tend to have a slow reaction rate and their fluorescence is not quantitatively related to the enzymatic activity. On the other hand, we expected that 2-COOH TM would take the colorless and non-fluorescent lactone form even when a single hydroxyl group of the xanthene moiety was substituted, due to the ease of intramolecular spirolactone formation compared with fluorescein (Figure 5b).

Figure 5. (a) Reaction scheme of a fluorescent probe based on fluorescein. (b) Reaction scheme of a fluorescent probe based on 2-COOH TM.

This molecular design approach would not be directly applicable to 2-COOH TM, which is present predominantly in the colorless and non-fluorescent lactone form under physiological conditions (aqueous solution at pH 7.4). Therefore, we set out to obtain a derivative of 2-COOH TM that would show strong fluorescence under physiological conditions by lowering the pKa values (pKa1 and pKa2) (Figure 3g). In the fluorescein scaffold, it is well-known that the pKa values can be lowered by the introduction of electron-withdrawing groups such as chlorine and fluorine into the xanthene moiety. For example, the pKa2 of fluorescein in Figure 1b (6.4) was shifted to the acidic region by the introduction of two chlorines (pKa2 = 4.5) or two fluorines (pKa2 = 4.8) into the xanthene moiety (Figure 6a; left).16,46 Thus, we introduced chlorines or fluorines into the xanthene moiety of 2-COOH TM (X = Cl: 2-COOH DCTM; X = F: 2-COOH DFTM) (Figure 6a; right), and examined the pH-dependency of the absorption and emission spectra of the resulting compounds (Figure 6b,c). Both compounds, 2-COOH DCTM and 2-COOH DFTM, showed a sharp pHdependent absorption change similar to that of 2-COOH TM, and exhibited strong absorption and fluorescence even at pH 7.4 (Figure 6d, Table 2). The pKa values (pKa1 and pKa2) were determined by biphasic fitting as described for 2-COOH TM. The values for 2-COOH DCTM and 2-COOH DFTM were pKa1 = 7.0, pKa2 = 6.2 and pKa1 = 7.1, pKa2 = 6.5, respectively

5 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(Table 3). The large pKa inversion of 2-COOH DCTM resulted in 13-fold absorption increment from pH 6.0 to pH 7.0. We also determined the acid dissociation constants by the equilibrium analysis method utilizing capillary zone electrophoresis, and the calculated values were similar to the above values (Table 3). Thus, 2-COOH DCTM and 2-COOH DFTM showed similar photophysical properties and reversed pKa values to those of 2-COOH TM. They are expected to exhibit similar chemical equilibria to 2-COOH TM, except for the pKa values.

Page 6 of 10

Table 2. Photophysical Properties of 2-COOH TM Derivatives λabs

λem

ε

(nm) a

(nm) a

(M−1cm−1) a

2-COOH TM

580

598

1.1 × 105

0.38

2-COOH DCTM

591

607

1.2 × 105

0.48

2-COOH DFTM

581

596

1.2 × 105

0.54

Φfl b

Measured in 100 mM sodium phosphate buffer, pH 10.0, containing 1% DMSO as a cosolvent. b Fluorescence quantum yields were calculated using that of TokyoMagenta (Φfl = 0.42) in 100 mM sodium phosphate buffer at pH 9.0 as a standard.31 a

Table 3. Parameters of the Chemical Equilibria of 2-COOH TM Derivatives pKa1 from abs. a

pKa2 from abs. a

pKa1 from CE b

pKa2 from CE b

Kc

2-COOH TM

8.2

7.9

8.4

7.9

0.086

2-COOH DCTM

7.0

6.2

7.0

6.4

0.11

2-COOH DFTM

7.1

6.5

7.3

6.6

0.067

Determined by biphasic fitting to the absorbance change. b Determined by the equilibrium analysis method utilizing capillary zone electrophoresis. c The calculated equilibrium constant K (= [MII] / [MI] or [form II] / [form I]) as shown in Figure 3g or Figure 7b, respectively. a

Figure 6. (a) Chemical structures of the halogenated derivatives of fluorescein and 2-COOH TM. (b,c) pH-Dependency of absorption (left) and emission (right) spectra of 2-COOH DCTM (X = Cl; λex = 591 nm) (b) and 2-COOH DFTM (X = F; λex = 581 nm) (c). Measurements were made at 25°C in 100 mM sodium phosphate buffer at various pH, containing 1% DMSO as a cosolvent. (d) pH Plots of normalized absorbance at 580 nm for 2-COOH TM, 591 nm for 2-COOH DCTM and 581 nm for 2-COOH DFTM.

Figure 7. (a) The pKa values of 2-Me TM (X = H), 2-Me DCTM (X = Cl) and 2-Me DFTM (X = F). (b) Putative chemical equilibrium of 2-COOH DCTM.

We further synthesized 2-Me DCTM32 and 2-Me DFTM, and determined their pKa values to obtain the equilibrium constant K (= [form II] / [form I]) values of 2-COOH DCTM and 2-COOH DFTM (Figure 3g, Figure 7). The pKa values of 2Me DCTM and 2-Me DFTM were 5.2 32 and 5.3 (Figure S11), and the equilibrium constant K values of 2-COOH DCTM and

6 ACS Paragon Plus Environment

Page 7 of 10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

2-COOH DFTM were 0.11 and 0.067, respectively (Table 3). These values are similar to that of 2-COOH TM. Scheme 1. Synthesis of 2-COOH DCTM β gal

Table 4. Photophysical Properties of 2-COOH DCTM βgal

2-COOH DCTM

βgal

λabs

λem

ε

(nm) a

(nm) a

(M−1cm−1) a

460

555

1.2 × 103

Φfl a,b 0.09

Measured in 100 mM sodium phosphate buffer at pH 7.4, containing 1% DMSO as a cosolvent. b Fluorescence quantum yield was calculated using that of fluorescein (Φfl = 0.85) in 0.1 N NaOH aq. as a standard.3 a

To validate our proposed design strategy based on the 2COOH DCTM scaffold, we designed and synthesized 2COOH DCTM βgal as a red fluorescent probe for βgalactosidase, which is widely used as a reporter enzyme in biological studies (Scheme 1).47-50 First, we measured the absorption and emission spectra, and ɛ and Φfl values of 2COOH DCTM βgal at pH 7.4 (Table 4). The molar extinction coefficient was very small (ɛ = 1,200), and it was considered that 2-COOH DCTM βgal formed an intramolecular spirolactone ring, as expected. Second, we examined the pHdependency of the absorption spectrum of 2-COOH DCTM βgal, and calculated its pKa value (Figure S12). The pKa value of 2-COOH DCTM βgal was 7.4 and the probe showed a very small molar extinction coefficient (ɛ = 2,600) even above pH 9.0, where essentially all the probe molecules should exist as monoanion (form I and form II), as shown in Figure 8a, based on the pKa value. The neutral form of 2-Me TM and 2-Me DCTM under acidic conditions showed ɛ = 29,000, so the ratio of [form II] / [form I] was calculated to be 0.099. These results indicate that less than 5% of 2-COOH DCTM βgal exists as colored monoanion form II at pH 7.4 (see Figure 8a). In other words, the substitution of a single hydroxyl group of the xanthene moiety of 2-COOH DCTM makes the molecule colorless (and non-fluorescent), as we had expected.

Figure 8. (a) Estimated chemical equilibrium of 2-COOH DCTM βgal. (b) Absorption (left) and emission (right) spectra of 1 µM 2COOH DCTM βgal after addition of β-galactosidase (0.3 unit). All experiments were performed at 37°C in 3.0 ml total volume of 100 mM sodium phosphate buffer at pH 7.4, containing 0.1% DMSO as a cosolvent. Excitation wavelength of 2-COOH DCTM βgal was 591 nm. The reaction scheme of 2-COOH DCTM βgal and photographs of the sample solutions before and after the enzymatic reaction are also shown.

Next, we examined the enzymatic reaction of 2-COOH DCTM βgal with β-galactosidase in cuvette. 2-COOH DCTM βgal was almost colorless and non-fluorescent before the enzymatic reaction, and became red-colored and strongly fluorescent after the enzymatic reaction, showing >1,000 fluorescence increase (Figure 8b). Finally, the enzymatic parameters, kcat and Km, of 2-COOH DCTM βgal were determined and compared with those of other reported fluorescent probes for β-galactosidase, 2-Me TM βgal, TG-βGal and FDG (Figure S13, S14). It is noteworthy that 2-COOH DCTM βgal showed the highest reactivity for β-galactosidase among these fluorescent probes. CONCLUSION In this study, we focused on the difference of the pHdependent absorption change between fluorescein and the Sisubstituted fluorescein, 2-COOH TM, which differ only in the atom at the 10’ position of the xanthene moiety. Noteworthy features are: (1) 2-COOH TM is colorless at acidic pH, though fluorescein exhibits absorption even at acidic pH; (2) 2-COOH TM shows an absorption spectral change above pH 7.0, whereas that of fluorescein is below pH 7.0; (3) 2-COOH TM

7 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

shows a much sharper pH response than fluorescein because of its pKa inversion, i.e., pKa1 > pKa2. These features are also different from those of a related compound lacking the carboxylic group, 2-Me TM. 2-COOH TM exists predominantly in the colorless and non-fluorescent lactone form, compared with fluorescein. Therefore, it may find application as a pH indicator with a dramatic, pH-dependent fluorescence increase. Modification of 2-COOH TM by dichlorination afforded 2COOH DCTM, which exists mainly in the fluorescent open form at physiological pH, and is therefore a candidate scaffold for red fluorescent probes. To validate this idea, we used 2COOH DCTM to synthesize a red florescent probe for βgalactosidase. These results illustrate the concept that precise control of the conformational change of small molecules, in the present case by utilizing the ease of intramolecular lactone ring formation, can provide a basis for designing novel chemical indicators for a wide range of molecules of interest in the life sciences.

AUTHOR INFORMATION Corresponding Author [email protected] [email protected]

ACKNOWLEDGMENT This work was supported in parts by MEXT (Specially Promoted Research Grant Nos. 22000006 to T.N., Grant Nos. 24689003, 24659042 and 26104509 to K. Hanaoka, and Grant Nos. 24655147 to T.K.), and SENTAN, JST to K. Hanaoka. K. Hanaoka was also supported by grants from Mochida Memorial Foundation for Medical and Pharmaceutical Research, The Naito Foundation, The Asahi Glass Foundation, Takeda Science Foundation and The Cosmetology Research Foundation. K. Hirabayashi was supported by a Grant-in-Aid for JSPS Fellows.

ASSOCIATED CONTENT Supporting Information Available. Synthesis, experimental details and characterization of compounds, quantum chemical calculation, in vitro assays, and HPLC analysis. These materials are available free of charge via the internet at http://pubs.acs.org.

REFERENCES (1) Baeyer, A. Ber. Dtsch. Chem. Ges. 1871, 4, 555–558. (2) Klonis, N.; Sawyer, W. H. J. Fluoresc. 1996, 6, 147–157. (3) Paeker, C. A.; Rees, W. T. Analyst 1960, 85, 587–600. (4) Maeda, H.; Ishida, N.; Kawauchi, H.; Tuzimura, K. J. Biochem. 1969, 65, 777–783. (5) Souchier, C.; French, M.; Benchaib, M.; Catallo, R.; Bryon, P. A. Cytometry 1995, 20, 203–209. (6) Sjöback, R.; Nygren, J.; Kubista, M. Spectrochim. Acta. Part A 1995, 51, L7–L21. (7) Maurice, D. M. Invest. Ophthalmol 1967, 6, 464–477. (8) Nanjiani, M. Fluorescein Angiography Technique, Interpretation and Application, Oxford University Press, 1991. (9) Urano, Y.; Kamiya, M.; Kanda, K.; Ueno, T.; Hirose, K.; Nagano, T. J. Am. Chem. Soc. 2005, 127, 4888–4894. (10) Ueno, T.; Urano, Y.; Setsukinai, K.; Takakusa, H; Kojima, H.; Kikuchi, K; Ohkubo, K.; Fukuzumi, S.; Nagano, T. J. Am. Chem. Soc. 2004, 126, 14079–14085. (11) Rotman, B.; Papermaster, D. W. Proc. Nat. Acad. Sci. U. S. A. 1966, 55, 134–141.

Page 8 of 10

(12) Rotman, B.; Zderic, J. A.; Edelstein, M. Proc. Nat. Acad. Sci. U. S. A. 1963, 50, 1–6. (13) Huang, Z.; Wang, Q.; Ly, H. D.; Gorvindarajan, A.; Scheigetz, J.; Zamboni, R.; Desmarais, S.; Ramachandran, C. J. Biomol. Screen. 1999, 4, 327–334. (14) Liu, C.; Pan, J.; Li, S.; Zhao, Y.; Wu, L.Y.; Berkman, C. E.; Whorton, A. R. and Xian, M. Angew. Chem. 2011, 123, 10511–10513. (15) Chen, W.; Liu, C.; Peng, B.; Zhao, Y.; Pacheco, A.; Xian, M. Chem. Sci. 2013, 4, 2892–2896. (16) Minta, A.; Kao, J. P. Y.; Tsien, R. Y. J. Biol. Chem. 1989, 264, 8171–8178. (17) Kao, J. P. Y., Harootunian, A. T.; Tsien, R. Y. J. Biol. Chem. 1989, 264, 8179–8184. (18) Kojima, H.; Nakatsubo, N.; Kikuchi, K.; Kawahara, S.; Kirino, Y.; Nagoshi, H.; Hirata, Y.; Nagano, T. Anal. Chem. 1998, 70, 2446– 2453. (19) Hirano, T.; Kikuchi, K.; Urano, Y.; Higuchi, T.; Nagano, T. J. Am. Chem. Soc. 2000, 122, 12399–12400. (20) Setsukinai, K.; Urano, Y.,; Kakinuma, K.; Majima, H. J.; Nagano, T. J. Biol. Chem. 2003, 278, 3170–3175. (21) Johnson, I.; Spence, M. T. Z., Ed. The Molecular Probes® Handbook: A Guide to Fluorescent Probes and Labeling Technologies, 11th Ed. Molecular Probes, Inc. 2010. (22) Chen, S.-C.; Nakamura, H.; Tamura, Z. Chem. Pharm. Bull. 1979, 27, 475–479. (23) Tamura Z.; Morioka T.; Maeda M.; Tsuji A. Bunseki Kagaku, 1994, 43, 339–346. (24) Polyakova, I. N.; Starikova, Z. A.; Parusnikov, B. V.; Krasavin, I. A.; Dobryakova, G. M.; Zhadanov, B. V. J. Struct. Chem. 1984, 25, 752–757. (25) Mchedlov-Petrossyan, N. O.; Vodolazkaya, N. A.; Gurina, Y. A.; Sun, W. C.; Gee, K. R. J. Phys. Chem. B 2010, 114, 4551–4564. (26) Mchedlov-Petrosyan, N. O.; Vodolazkaya, N. A.; Martynova, V. P.; Samoilov, D. V.; El’tsovK, A. V. Russ. J. Gen. Chem. 2002, 72, 785–792. (27) Schröder, C. R.; Weidgans, B. M.; Klimant, I. Analyst 2005, 130, 907–916. (28) Crovetto, L.; Paredes, J. M.; Rios, R.; Talavera, E. M.; Alvarez-Pez, J. M. J. Phys. Chem. A 2007, 111, 13311–13320. (29) Lavis, L. D.; Rutkovski, T. J.; Raines, R. T. Anal. Chem. 2007, 79, 6775–6782. (30) Zhang, X.; Liu, Q.; Son, A.; Zhang, Q.; Zhang, F.; Zhao, F. Photochem. Photobiol. Sci. 2008, 7, 1079–1084. (31) Egawa, T.; Koide, Y.; Hanaoka, K.; Komatsu, T.; Terai, T.; Nagano, T. Chem. Commun. 2011, 14, 4162–4164. (32) Egawa, T.; Hirabayashi, K; Koide, Y.; Kobayashi, C.; Takahashi, N.; Mineno, T.; Terai, T.; Ueno, T.; Komatsu, T.; Ikegaya, Y.; Matsuki, N.; Nagano, T.; Hanaoka, K. Angew. Chem. Int. Ed. 2013, 52, 3874–3877. (33) Best, Q. A.; Sattenapally, N; Dyer, D. J.; Scott, C. N.; McCarroll, M. E. J. Am. Chem. Soc. 2013, 135, 13365–13370. (34) Tamura, A.; Abe, S.; Ito, K.; Maeda, M. Anal. Sci. 1996, 12, 927–930. (35) Grimm, J. B.; Sung, A. J.; Legant, W. R.; Hulamm, H.; Matlosz, S. M.; Betzig, E.; Lavis, L. D. ACS Chem. Biol. 2013, 8, 1303– 1310. (36) Dean, J. A.; Lange's Handbook of Chemistry, 13th Ed. 1985, McGraw Hill, New York, Table 8.8. (37) Takayanagi, T.; Motomizu, S. Chem. Lett. 2001, 1, 14–15. (38) Mchedlov-Petrosyan, N. O.; Mayorga, R. S. J. Chem. Soc. Faraday Trans. 1992, 88, 3025–3032. (39) Mchedlov-Petrosyan, N. O.; Salamanova, N. V.; Vodolazkaya, N. A.; Gurina, Y. A.; Borodenko, V.I. J. Phys. Org. Chem. 2006, 19, 365–375. (40) Beckers, J. L.; Everaerts, F. M.; Ackermans, M. T. J. Chromatogr. 1991, 537, 407–428. . (41) Jang, Y. H.; William, A. Goddard III; Katherine, T. N.; Sowers, L. C.; Hwang, S.; Chung, D. S. Chem. Res. Toxicol. 2002, 15, 1023–1035. (42) Fu, M.; Xiao, Y.; Qian, X.; Zhao, D.; Xu, Y. Chem. Commun., 2008, 15, 1780–1782.

8 ACS Paragon Plus Environment

Page 9 of 10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

(43) Yamaguchi, S.; Tamao, K. J. Chem. Soc. Dalton Trans. 1998, 3693–3702. (44) Uchiyama, S.; Makino, Y. Chem. Commun., 2009, 19, 26462648. (45) Zhou, K.; Liu, H; Zhang, S.; Huang, X.; Wang, Y.; Huang, G. ; Sumer, B. D.; Gao, J. J. Am. Chem. Soc. 2012, 134, 7803-7811. (46) Mottram, L.; Boonyarattanakalin, S.; Kovel, R. E.; Peterson, B. R. Org. Lett. 2006, 8, 581–584.

(47) Naylor, L.H. Biochem. Pharmacol. 1999, 58, 749–757. (48) Spergel, D. J.; Krüth, U.; Shimshek, D. R.; Sprengel, R.; Seeburg, P. H. Prog. Neurobiol. 2001, 63, 673–686. (49) Alam, J.; Cook, J. L. Anal. Biochem. 1990, 188, 245–254. (50) Pietri, T.; Eder, O.; Blanche, M.; Thiery, J. P.; Dufour, S. Dev. Biol. 2003, 259, 176–187.

9 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 10

For TOC only

10 ACS Paragon Plus Environment