Convenient synthesis of regioisomerically pure 5- and 6-functionalized xanthene dyes via SNAr reaction and comparison of their reactivity towards click reaction

Article abstract of DOI:10.1016/j.tet.2020.131087

In this study, we present an efficient and general strategy for the individual preparation of both isomers of 5(6)-functionalized xanthene-based fluorophores. Spectroscopic analysis of the acid-base equilibrium of 5(6)-nitro xanthene dyes has shown that they exist predominantly in the colorless spirolactone form in certain aprotic dipolar solvents. In such solvents, regioisomerically selective ipso-substitution of the nitro group by sodium azide occurs at the 6- position but not at the 5- position due to the electron-withdrawing spirolactone moiety at the para-position, relative to the nitro group. This reaction allows the separation of the isomer with the 6-azide group and the intact 5-nitro isomer. The 5-nitro group of the latter was then reduced to an amino group and subsequently converted to an azide group. This strategy enables the preparation of both the 5- and 6-functionalized isomers individually from a mixture of precursors, which is otherwise unachievable. The azide isomers were then compared in reactivity by strain-promoted azide-alkyne cycloaddition (SPAAC) with bicycle[6.1.0]non-4-yene (BCN).

Full text of DOI:10.1016/j.tet.2020.131087

Tetrahedron xxx (xxxx) xxx
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Convenient synthesis of regioisomerically pure 5- and
6-functionalized xanthene dyes via SNAr reaction and comparison of
their reactivity towards click reaction
Masaya Mori, Yuuta Fujikawa*, Hideshi Inoue
School of Life Sciences, Tokyo University of Pharmacy and Life Sciences, 1432-1 Horinouchi, Hachioji, Tokyo, 192-0392, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
In this study, we present an efficient and general strategy for the individual preparation of both isomers
of 5(6)-functionalized xanthene-based fluorophores. Spectroscopic analysis of the acid-base equilibrium
of 5(6)-nitro xanthene dyes has shown that they exist predominantly in the colorless spirolactone form
in certain aprotic dipolar solvents. In such solvents, regioisomerically selective ipso-substitution of the
nitro group by sodium azide occurs at the 6- position but not at the 5- position due to the electron-
withdrawing spirolactone moiety at the para-position, relative to the nitro group. This reaction allows
the separation of the isomer with the 6-azide group and the intact 5-nitro isomer. The 5-nitro group of
the latter was then reduced to an amino group and subsequently converted to an azide group. This
strategy enables the preparation of both the 5- and 6-functionalized isomers individually from a mixture
of precursors, which is otherwise unachievable. The azide isomers were then compared in reactivity by
strain-promoted azide-alkyne cycloaddition (SPAAC) with bicycle[6.1.0]non-4-yene (BCN).
© 2020 Published by Elsevier Ltd.
Received 3 December 2019
Received in revised form
24 February 2020
Accepted 26 February 2020
Available online xxx
Keywords:
Fluorescent dye
Chemical equilibrium
SNAr reaction
SPAAC
1. Introduction
and have also been applied to the click reaction [11,12].
However, previously reported biological applications of
Xanthene-based dyes, such as fluoresceins and rhodamines, are
widely used in various bioassays as labeling agents and fluores-
cence probes, because they have a variety of visible light emissions
ranging from blue-green to red, a high molar extinction coefficient
and quantum efficacy, high water-solubility, and are easy to syn-
thesize and derivatize [1,2].
Since xanthene dyes are composed of a meso-phenyl ring and a
xanthene moiety, they can be derivatized to give various functions
by modifying the phenyl-ring moiety while retaining water-
solubility and resistance to self-aggregation. In this context, 5- or
6-functionalization of the meso-phenyl ring in the xanthene de-
rivatives is crucial for the synthesis of fluorogenic probes and for
the modification and bioconjugation of biopolymers [3]. For
example, 5(6)-aminofluorescein is a key intermediate to lead to the
respective isothiocyanate (ITC), iodo (-I), and azide (-N3) derivatives
[4]. These 5(6)-functionalized fluorescein derivatives have been
used as labeling agents of target proteins [5,6], synthetic in-
termediates for coupling reactions [7,8], fluorogenic probes [9,10],
different regioisomers have shown different results between the
isomers [13,14]. In addition, fluorogenic probes based on 5- and 6-
functionalized fluorescein derivatives were reported to show dif-
ferences in target selectivity [15] or enzymatic kinetic parameters
[16] between regioisomers. These examples suggest that sensing
applications should be performed with both isomers to avoid
misinterpretation based on the results obtained with only one
isomer. Further, to developing more sophisticated probes, it is
desirable to assess the sensing performance (e.g. selectivity and
reactivity) of fluorogenic probes for each regioisomer.
A typical synthetic procedure based on the acid catalyzed
condensation of asymmetric 4-functionalized phthalic anhydride
and resorcinol or aminophenol produces a regioisomeric mixture of
5(6)-functionalized xanthene-based dyes. To address this problem,
numerous efforts have been dedicated to separate the isomers of
xanthene derivatives by fractional recrystallization [4,17,18] or to
prepare a single isomer with the use of an asymmetric starting
material [19]. These approaches have often been applied to the
preparation of pure regioisomers of 5- and 6-carboxyfluorescein. In
contrast, synthetic procedures for the preparation of pure isomers
functionalized with an azide or amino group at the 5- or 6-position
* Corresponding author.
E-mail address: yfuji@toyaku.ac.jp (Y. Fujikawa).
https://doi.org/10.1016/j.tet.2020.131087
0040-4020/© 2020 Published by Elsevier Ltd.
Please cite this article as: M. Mori et al., Convenient synthesis of regioisomerically pure 5- and 6-functionalized xanthene dyes via SNAr reaction
and comparison of their reactivity towards click reaction, Tetrahedron, https://doi.org/10.1016/j.tet.2020.131087

2
M. Mori et al. / Tetrahedron xxx (xxxx) xxx
have not been reported. This is partly because xanthene dyes with a
nitro group at the 5- or 6-position have similar physicochemical
properties to each other, which prevents fractional recrystalliza-
tion. Malakhov and Burgess succeeded in the synthesis of pure
regioisomeric 5-functionalized 2ʹ, 7ʹ-dichlorofluorescein (DCF) by
recrystallization [20]. Although the 5-isomer is useful by itself, the
6-isomer is not obtained by this approach, so that it is impossible to
compare the labeling kinetics and optical properties of the two
isomers.
Here we present a simple synthetic strategy based on the
regioisomer selective SNAr azidation that selectively provides in-
dividual azide- or amino-functionalized regioisomers. Optimiza-
tion of the aprotic dipolar solvent used shifts the equilibrium
between the open and closed form completely to a closed spi-
rolactone, allowing the ipso-substitution of the nitro group at 6-
position. Based on this, both regioisomers were prepared for two
fluorophores, DCF and tetramethylrhodamine (TMR), allowing for
comparison of their reactivity and changes in their spectroscopic
properties. Then 5-N3DCF was compared with the 6-isomer in
reactivity by strain-promoted azide-alkyne cycloaddition (SPAAC)
with bicycle[6.1.0]non-4-yene (BCN). Our new synthetic method
allows for the convenient regioselective preparation of function-
alized xanthene dyes and facilitates the development of new la-
beling dyes that can be used in the field of biochemistry and
chemical biology.
temperature, it was poured onto H2O (100 ml) on ice. The mixture
was basified with 10 N NaOH to pH 14 and stirred at room tem-
perature. The mixture was acidified with concentrated HCl to pH 2
and the precipitate was collected by filtration and washed three
times with ice-cold H2O. The residue was purified by silica gel
column chromatography (CHCl3/MeOH ¼ 10/1 to 2/1) to yield 5(6)-
NO2DCF as an orange solid (2.75 g, 6.16 mmol, yield 58%).
2.3.2. 6-Azido-2ʹ,7ʹ -dichlorofluorescein (6-N3DCF), 5-nitro-2ʹ,7ʹ-
dichlorofluorescein (6-NO2DCF)
To a suspension of 5(6)-NO2DCF (467 mg, 1.05 mmol, 5-isomer:
6-isomer ¼ 1:1) in HMPA (2 ml), sodium azide (211 mg, 3.25 mmol)
was added slowly and the reaction mixture was stirred at 80 ꢂC for
4 h. After it was cooled to room temperature, the reaction mixture
was diluted with saturated aqueous NaH2PO4 and the mixture was
extracted three times with AcOEt. The organic layer was washed
twice with saturated NaH2PO4, once with brine, dried over anhy-
drous MgSO4, filtered, and evaporated under reduced pressure. The
residue was purified by silica gel column chromatography (CHCl3/
MeOH ¼ 10/1 to 4/1), yielding 6-N3DCF (234 mg, 0.530 mmol, yield
50%) and 5-NO2DCF (155 mg, 0.348 mmol, yield 33%) as a red solid.
2.3.3. 6-N3DCF
1H NMR (400 MHz, DMSO‑d6)
d
11.08 (2H, br), 7.98 (1H, d,
J ¼ 8.2 Hz), 7.40 (1H, dd, J ¼ 8.2, 1.8 Hz), 7.13 (1H, d, J ¼ 1.8 Hz), 6.88
(2H, s), 6.71 (2H, s). 13C NMR (100 MHz, DMSO‑d6)
167.4, 155.1,
d
2. Materials and methods
153.7,150.0,147.4,128.3,126.8,122.2,121.9,116.2,114.3,110.2,103.6,
81.0. HRMS (ESI-TOF) m/z calculated: C20H8N3O5Cl2 [M ꢀ H]-:
439.9841, 441.9812, found: 439.9844 (þ0.3 mmu), 441.9820 (þ0.8
mmu).
2.1. General procedures
All reagents were purchased from commercial suppliers and
used without further purification. 1H and 13C NMR spectra were
obtained with the DRX-400 spectrometer (Bruker, Billerica, MA,
2.3.4. 5-NO2DCF
1H NMR (400 MHz, DMSO‑d6)
d
8.87 (1H, d, J ¼ 2.3 Hz), 8.38 (1H,
USA) at 400 and 100 MHz, respectively. Chemical shifts (
d) were
dd, J ¼ 8.5, 2.5 Hz), 7.54 (1H, d, J ¼ 8.2 Hz), 6.67 (2H, s), 6.21 (2H, s).
HRMS (ESI-TOF) m/z calculated: C20H10NO7Cl2 [MþH]þ: 445.9834,
447.9805, found:445.9835 (þ0.1 mmu), 447.9809 (þ0.4 mmu).
calibrated to the solvent peak (
d
2.49 and 39.5 for 1H and 13C NMR
in DMSO‑d6). A precoated thin layer chromatography (TLC) plate
(Merck, silica gel 60 F254, No. 1.05715.0001) was used for TLC
analysis. Column chromatography was performed with silica gel
(200e300 mesh) (Kanto Chem. Inc.). High resolution mass spectra
(HRMS) were obtained from a micro mass LCT spectrometer (Wa-
ters, Milford, MA, USA) under positive or negative electro spray
ionization (ESIþ or ESIꢀ) conditions. The absorption spectra of each
2.3.5. 6-Amino-2ʹ,7ʹ-dichlorofluorescein (6-NH2DCF)
A solution of 6-N3DCF (150 mg, 0.340 mmol) and Na2Se9H2O
(163 mg, 0.677 mml) in H2O (3 ml) was stirred at 50 ꢂC for 30 min.
After the solution was cooled to room temperature, saturated
aqueous NaH2PO4 was added to the mixture, and the mixture was
extracted with AcOEt. The organic layer was then washed twice
with saturated aqueous NaH2PO4, once with brine, then dried over
anhydrous MgSO4, filtered, and evaporated under reduced pres-
sure. The residue was purified by silica gel column chromatography
(CHCl3/MeOH ¼ 10/1 to 4/1) to give the title solid as an orange solid
(101 mg, 0.242 mmol, yield 71%). 1H NMR (400 MHz, DMSO‑d6)
compound (50 mM) were measured with a V-550 UV/VIS spectro-
photometer (JASCO Corp., Tokyo, Japan).
2.2. HPLC analysis
HPLC analysis was performed on a HPLC system composed of a
pump (LC-20AT, Shimadzu) and a photodiode array detector (SPD-
M20A, Shimadzu). Clarity software (DataApex, Czech Republic) was
used for operation of the system and data analysis. For the time
d
11.06 (2H, br), 7.60 (1H, d, J ¼ 8.2 Hz), 6.87 (2H, s), 6.78 (1H, dd,
J ¼ 8.7, 1.8 Hz), 6.68 (2H, s), 6.37 (2H, s), 6.12 (1H, d, J ¼ 1.4 Hz). 13
C
NMR (100 MHz, DMSO‑d6)
d 168.4, 155.8, 155.3, 154.9, 149.8, 128.0,
course analysis, 1.5
with in 1.5 ml of 10 mM ammonium acetate containing 200
DCF. The solution (25 L) was injected and subjected to the analysis.
Separation was achieved on Mightysil RP-18 GP column
m); Kanto Chemical, Tokyo, Japan) Eluent A:
m
L of the reaction mixture was taken and mixed
126.4, 116.1, 115.9, 111.5, 111.4, 105.3, 103.6, 79.3. HRMS (ESI-TOF) m/
z calculated: C20H10NO5Cl2 [M ꢀ H]-: 413.9936, 415.9907, found:
413.9940 (þ0.4 mmu), 415.9915 (þ0.8 mmu).
mM
m
a
(250 mm ꢁ 4.6 mm (5
m
2.3.6. 5-Amino-2ʹ,7ʹ-dichlorofluorescein (5-NH2DCF)
B ¼ 85: 15 / 50: 50 (20 min) / 50: 50 (30 min) at a flow rate of
A solution of 5-NO2DCF (101 mg, 0.227 mmol) and Na2Se9H2O
(206 mg, 0.857 mml) in H2O (2 ml) was stirred at 60 ꢂC for 2 h. After
the solution was cooled to room temperature, saturated aqueous
NaH2PO4 was added to the mixture, and the mixture was extracted
three times with AcOEt. The organic layer was then washed twice
with saturated aqueous NaH2PO4, once with brine, then dried over
anhydrous MgSO4, filtered, and evaporated under reduced pres-
sure. The residue was purified by silica gel column chromatography
(CHCl3/MeOH ¼ 4/1) to give the title compound as an orange solid
1.0 ml/min. A: 10 mM ammonium acetate, B: MeCN.
2.3. Organic synthesis
2.3.1. 5(6)-Nitro-2ʹ,7ʹ-dichlorofluorescein (5(6)-NO2DCF)
A suspension of 4-nitrophthalic anhydride (2.05 g, 10.6 mmol)
and 4-chlororesorcinol (4.63 g, 32.0 mmol) in CH3SO3H (5 ml) was
stirred at 100 ꢂC for 11 h. After the mixture cooled to room
Please cite this article as: M. Mori et al., Convenient synthesis of regioisomerically pure 5- and 6-functionalized xanthene dyes via SNAr reaction
and comparison of their reactivity towards click reaction, Tetrahedron, https://doi.org/10.1016/j.tet.2020.131087

M. Mori et al. / Tetrahedron xxx (xxxx) xxx
3
(40.1 mg, 0.0963 mmol, yield 42%). 1H NMR (400 MHz, DMSO‑d6)
11.08 (2H, br), 7.23e6.77 (5H, m), 6.65 (2H, s), 5.85 (2H, s). HRMS
(ESI-TOF) m/z calculated: 20H10NO5Cl2 [M
H]-: 413.9936,
415.9907, found: 413.9937 (þ0.1 mmu), 415.9911 (þ0.4 mmu).
2.3.12. 6-Amino-tetramethylrhodamine (6-NH2TMR)
d
A suspension of 6-N3TMR (34 mg, 0.078 mmol) and Na2Se9H2O
(88 mg, 0.37 mml) in MeOH/H2O (2 ml, 1:1) was stirred at 60 ꢂC for
1 h. A suspension of Na2Se9H2O (108 mg, 0.451 mml) in MeOH
(1 ml) was added to the reaction mixture and stirred at 70 ꢂC for 1 h.
After the solution was cooled to room temperature, saturated
aqueous NaH2PO4 was added to the mixture, and the mixture was
extracted three times with AcOEt. The organic layer was washed
with brine, then dried over anhydrous MgSO4, filtered, and evap-
orated under reduced pressure. The residue was purified by silica
gel column chromatography (CHCl3/MeOH/MeCN ¼ 4/1/1) to give
the title solid as a brown solid (19 mg, 0.048 mmol, yield 62%). 1H
C
ꢀ
2.3.7. 5-Azido-2ʹ,7ʹ-dichlorofluorescein (5-N3DCF)
To a solution of 5-NH2DCF (24.1 mg, 0.0579 mmol) in AcOH/H2O
(1.2 ml, 1:1), sodium nitrite (8.1 mg, 0.020 mmol) was added on ice
and stirred on ice for 1 h. Sodium azide (11.7 mg, 0.180 mmol) was
added to the reaction mixture and stirred on ice for 1 h. Then, the
reaction mixture was stirred at room temperature for 1.5 h. After
saturated NaH2PO4 was added to the mixture, the solution was
extracted three times with AcOEt. The organic layer was washed
twice with saturated aqueous NaH2PO4, once with brine, dried over
anhydrous Na2SO4, filtered, and evaporated under reduced pres-
sure. The residue was purified by silica gel column chromatography
(CHCl3/MeOH ¼ 4/1) to give the title solid as an orange solid
NMR (400 MHz, DMSO‑d6)
J ¼ 8.5, 1.6 Hz), 6.65e6.36 (6H, m), 6.27e6.00 (3H, m), 2.93 (12H, s),
13C NMR (100 MHz, DMSO‑d6)
169.1, 155.4, 151.9, 151.8, 128.4,
d
7.55 (1H, d, J ¼ 8.2 Hz), 6.72 (1H, dd,
d
125.8, 115.2, 112.6, 109.0, 107.4, 105.7, 97.9, 79.2, 39.9. HRMS (ESI-
TOF) m/z calculated:
402.1811 (ꢀ0.7 mmu).
C
24H24N3O3 [MþH]þ: 402.1818, found:
(20.5 mg, 0.0464 mmol, yield 80%). 1H NMR (400 MHz, DMSO‑d6)
d:
11.12 (2H, br), 7.63 (1H, d, J ¼ 1.8 Hz), 7.52 (1H, dd, J ¼ 8.2, 2.3 Hz),
7.36 (1H, d, J ¼ 8.2), 6.88 (2H, s), 6.73 (2H, s). HRMS (ESI-TOF) m/z
calculated: C20H8N3O5Cl2 [M ꢀ H]-: 439.9841, 441.9812, found:
439.9845 (þ0.4 mmu), 441.9819 (þ0.7 mmu).
2.3.13. 5-Azide-tetramethylrhodamine (5-N3TMR)
A suspension of 5-NO2TMR (38.4 mg, 0.0891 mmol) and
Na2Se9H2O (101 mg, 0.422 mml) in H2O/DMSO (2 ml, 1:1) was
stirred at 60 ꢂC for 30 min. After the solution was cooled to room
temperature, saturated aqueous NaH2PO4 was added to the
mixture, and the mixture was extracted three times with AcOEt.
The organic layer was washed with brine, then dried over anhy-
drous MgSO4, filtered, and evaporated under reduced pressure. The
crude extract, obtained by column separation with silica gel (CHCl3/
CH3OH ¼ 4/1 to 1/1), was used for the next reaction without further
purification. To a solution of crude extract in H2O/AcOH (1.2 ml,
1:2), sodium nitrite (18.9 mg, 0.274 mmol) was added and stirred
on ice for 1 h. Then, sodium azide (21.0 mg, 0.323 mmol) was added
and stirred on ice for 1 h and then at room temperature for 1.5 h.
The solution was diluted with saturated NaH2PO4 and extracted
three times with AcOEt. The organic layer was washed twice with
saturated aqueous NaH2PO4, once with brine, dried over anhydrous
Na2SO4, filtered, and evaporated under reduced pressure. The res-
idue was purified by silica gel column chromatography (CHCl3/
MeOH ¼ 4/1) to give the title compound as a dark purple solid
(11.4 mg, 0.0267 mmol, yield 30% in two steps). 1H NMR (400 MHz,
2.3.8. 5(6)-Nitro-2ʹ,7ʹ-tetramethylrhodamine (5(6)-NO2TMR)
A suspension of 4-nitrophthalic anhydride (1.03 g, 5.33 mmol)
and 3-(dimethylamino)phenol (1.65 g, 12.0 mmol) and poly-
phosphoric acid (105 mg) in DMF (30 ml) was stirred at 120 ꢂC for
12 h. After cooling to room temperature, the solution was evapo-
rated under reduced pressure. The residue was dissolved in
methanol (5 ml) and the methanol solution was poured onto H2O
(100 ml) on ice. The precipitate was collected by filtration and
washed three times with ice-cold water. The residue was purified
by silica gel column chromatography (CHCl3/MeOH ¼ 8/1 to 1/1) to
give 5(6)-NO2TMR as a red solid (131 mg, 0.304 mmol, yield 6%).
2.3.9. 6-Azido-tetramethylrhodamine (6-N3TMR), 5-azido-
tetramethylrhodamine (5-N3TMR)
To a suspension of 5(6)-NO2TMR (104 mg, 0.242 mmol/isomer,
5-isomer:6-isomer ¼ 1:1) in NMP (2 ml), sodium azide (55.2 mg,
0.846 mmol) was added slowly and the reaction mixture was stir-
red at 80 ꢂC for 3 h. After cooling to room temperature, the reaction
mixture was diluted with saturated aqueous NaHCO3. The precip-
itate was collected by filtration and washed three times with ice-
cold water. The crude extract was purified by silica gel column
chromatography twice (CHCl3/MeOH ¼ 10:1 and 1:1, respectively),
yielding 6-N3TMR (13.9 mg, 0.0325 mmol, yield 13%) and 5-
NO2DCF (44.1 mg, 0.102 mmol, yield 42%) as red solids.
DMSO‑d6)
d
7.62 (1H, d, J ¼ 1.8 Hz), 7.48 (1H, dd, J ¼ 8.2, 2.3 Hz), 7.23
(1H, d, J ¼ 8.2 Hz), 6.55e6.44 (6H, m), 2.93 (12H, s). 13C NMR
(100 MHz, DMSO‑d6)
d 168.0, 152.2, 152.0, 148.8, 141.5, 128.41,
128.37, 126.6, 125.6, 114.3, 109.0, 105.8, 98.0, 85.0, 39.8. HRMS (ESI-
TOF) m/z calculated:
428.1711 (ꢀ1.2 mmu).
C
24H22N5O3 [MþH]þ: 428.1723, found:
3. Results
3.1. Rationale and spectroscopic observation of DCF
2.3.10. 6-N3TMR
1H NMR (400 MHz, DMSO‑d6 plus 30% DClaq
) d 8.18 (1H, d,
The synthetic utility of nitro groups as leaving groups for direct
ipso-substitution with nucleophiles (e.g. azides, thiols, alkoxides) in
dipolar aprotic solvents has been demonstrated in many types of
reactions [21]. The structural requirement of ipso-substitution of
the nitro group in nitroarenes is the presence of a functional group
that stabilizes the transitional Meisenheimer complex at the ortho-
or para-position.
J ¼ 8.7 Hz), 7.47 (1H, dd, J ¼ 8.7, 2.3 Hz), 7.19 (1H, d, J ¼ 2.3 Hz), 7.05
(2H, dd, J ¼ 9.1, 2.3 Hz), 7.01 (2H, d, J ¼ 9.6), 6.89 (2H, d, J ¼ 2.3), 3.21
(12H, s). 13C NMR (100 MHz, DMSO‑d6 plus 30% DClaq
) d 165.7, 157.2,
157.1, 144.6, 135.8, 133.4, 131.0, 127.1, 121.1, 115.1, 113.3, 96.5, 40.9.
HRMS (ESI-TOF) m/z calculated: C24H22N5O3 [MþH]þ: 428.1723,
found: 428.1718 (ꢀ0.5 mmu).
A xanthene dye with a carboxylic group in the 3-position of the
meso-phenyl ring forms a spirolactone ring via intramolecular
nucleophilic attack of the carboxylate, which results in an equilib-
rium between the open (colored) and closed spirolactone (color-
less) forms (Scheme 1). The position of equilibrium between the
two xanthene isomers depends on the temperature, pH, hydrogen
bonding ability, and self-organization of solvent, where protic
2.3.11. 5-NO2TMR
1H NMR (400 MHz, DMSO‑d6)
d
8.64 (1H, d, J ¼ 1.8 Hz), 8.53 (1H,
dd, J ¼ 8.5, 2.1 Hz), 7.49 (1H, d, J ¼ 8.7 Hz), 6.59 (2H, d, J ¼ 8.7 Hz),
6.51 (2H, d, J ¼ 2.3 Hz), 6.48 (2H, dd, J ¼ 8.7, 2.7 Hz), 2.94 (12H, s).
HRMS (ESI-TOF) m/z calculated: C24H21N3O5Na [MþNa]þ: 454.1379,
found: 454.1371 (ꢀ0.8 mmu).
Please cite this article as: M. Mori et al., Convenient synthesis of regioisomerically pure 5- and 6-functionalized xanthene dyes via SNAr reaction
and comparison of their reactivity towards click reaction, Tetrahedron, https://doi.org/10.1016/j.tet.2020.131087

4
M. Mori et al. / Tetrahedron xxx (xxxx) xxx
Scheme 1. Chemical equilibrium of 5(6)-nitro-2ʹ,7ʹ-dichlorofluorescein (DCF).
solvents favor the open form and aprotic solvents favor the closed
spirolactone form [22,23]. In the closed spirolactone form, the nitro
group of 5- or 6-NO2DCF is in the meta- or para-position relative to
the strong electron-withdrawing spirocyclic ester, respectively.
Therefore, it is expected that the spirocyclic ester causes a differ-
ence in the reactivity of the ipso-carbon between the 5- and 6-
NO2DCF regioisomers and promotes the selective nucleophilic
attack on the ipso-carbon of the 6-nitro group rather than the 5-
nitro group (Scheme 2). To find a solvent that shifts the equilib-
rium to increase the proportion of the closed spirolactone form, the
5(6)-NO2DCF regioisomeric mixture was dissolved in several
aprotic dipolar solvents and the absorption spectra were measured
(Fig. 1). While the 5(6)-NO2DCF mixture shows high absorbance in
H2O or N, N-dimethylformamide (DMF), it shows a dramatically
lower absorbance in N-methylpyrolidone (NMP) and hexamethyl-
phosphoric triamide (HMPA). Especially, the visible light absorption
completely disappeared in HMPA. This result suggests that the
equilibrium of 5(6)-NO2DCF is shifted significantly to the closed
spirolactone form in HMPA.
Fig. 1. Absorption spectra of 5(6)-nitro-DCF in various polar solvents. The inset shows
the color of 5(6)-NO2DCF (50
light.
mM) in each solvent photographed under incandescent
to 6-NO2DCF gradually decreased, a new peak was observed to
increase (Fig. 2A). After 2 h, the 6-NO2DCF peak completely dis-
appeared and the product peak reached its maximum. LC-MS
analysis demonstrated that the product peak showed a molecular
mass number corresponding to the expected reaction product,
azide-substituted DCF. In contrast, no change in the 5-NO2DCF peak
was observed under these conditions (Fig. 2).
The same reaction was also carried out in NMP and DMF for
comparison with HMPA. RP-HPLC analysis clearly revealed that the
reaction proceeded in all the solvents tested, but the formation of
the SNAr product differed greatly; with most formation in HMPA,
less in NMP, and the least in DMF, which is the reverse order of the
magnitude of the dipole moment (Fig. S1). Therefore, the results
suggest that the formation of spirolactone promotes the SNAr re-
action selectively toward 6-NO2DCF.
3.2. Regioisomer selective SNAr reaction on 5(6)-NO2DCF mixture
Next, we investigated the isomer-selective SNAr reaction of 5(6)-
NO2DCF in HMPA. For this purpose, sodium azide was used as a
nucleophile, because azidated dyes are often used as synthetic in-
termediates and can be used for labeling experiments [11,12]. 5(6)-
NO2DCF was mixed with three equimolar of sodium azide and
reacted in HMPA. Preliminary experiments showed no reaction at
room temperature (data not shown), so the reaction was carried out
at 80 ꢂC. Reversed phase (RP)-HPLC analysis was performed to
monitor the progression of the reaction. As the peak corresponding
After the regioselective SNAr reaction of 5(6)-NO2DCF, 6-N3DCF
and 5-NO2DCF were isolated from the reaction mixture through
column chromatography. On the one hand, 6-N3DCF was assigned
by conventional spectroscopic measurements and 1He13C HMBC
NMR spectrum analysis (Fig. 3), then converted to 6-NH2DCF
(Scheme 3). On the other hand, 5-NO2DCF was reduced to corre-
sponding 5-NH2DCF with sodium sulfide, and then followed by
diazotization and subsequent treatment with sodium azide to yield
Scheme 2. Plausible mechanism of regioselective SNAr reaction of 6-NO2DCF. (A) 6-NO2DCF undergoes SNAr substitution by stabilizing the Meisenheimer complex with an electron-
withdrawing spirocyclic ester in the para position. (B) 5- NO2DCF does not react with sodium azide because there is no stabilizing factor for the Meisenheimer complex.
Please cite this article as: M. Mori et al., Convenient synthesis of regioisomerically pure 5- and 6-functionalized xanthene dyes via SNAr reaction
and comparison of their reactivity towards click reaction, Tetrahedron, https://doi.org/10.1016/j.tet.2020.131087

M. Mori et al. / Tetrahedron xxx (xxxx) xxx
5
Fig. 2. (A) Time course of the RP-HPLC elution profile of the reaction mixture. Asterisk represents the peak of DCF as an internal control. At each time, an aliquot of the reaction
mixture was mixed with DCF solution and subjected to RP-HPLC analysis. Absorbance at 505 nm was monitored for detection. (B) Time course of the content of each compound.
Relative intensity represents corresponding peak relative to that of the DCF peak. 6-N3DCF (Red triangles), 6-NO2DCF (Blue circles), and 5-NO2DCF (Black squares).
Fig. 3. 1He13C HMBC NMR spectrum of 6-N3DCF.
5-N3DCF in 80% yield (Scheme 3). Taken together, this new syn-
thetic strategy can be used to prepare each regioisomer of 5- and 6-
functionalized DCF.
spirolactone form in these solvents. Therefore, 5(6)-NO2TMR was
subjected to a reaction with sodium azide in NMP, resulting in the
formation of 6-N3TMR in 13% yield. 6-N3TMR was reduced to 6-
NH2TMR in 62% yield, which can be a key intermediate for
further synthesis. The remaining 5-NO2TMR was also isolated and
reduced to 5-NH2TMR, followed by diazotization and subsequent
treatment with sodium azide to provide 5-N3TMR in 30% yield in
two steps (Scheme S1). The isolated pure regioisomers were both
assigned by conventional spectroscopic measurements. These re-
sults show that this regioselective SNAr reaction can be applied to
different fluorophores and that both regioisomers of 5(6)-func-
tionalized TMR can be easily synthesized individually.
3.3. Regioisomer-selective SNAr reaction on the 5(6)-NO2TMR
mixture
Having successfully synthesized both the 5- and 6-regioisomers
for each of NH2- and N3-DCF, we next asked whether this approach
could be applied to rhodamine fluorophores. To examine this, tet-
ramethylrhodamine (TMR) was chosen as the fluorophore because
TMR is robustly used as the labeling reagent in studies of chemical
biology [24,25] and biochemical applications [26,27]. 5(6)-
NO2TMR, prepared according to a previously reported procedure
[17], completely lost visible light absorption in NMP and DMF
among the aprotic dipolar solvents we tested (Fig. S2). This suggests
that the equilibrium shifted from the open form to the closed
3.4. Comparison of the reactivity of regioisomers on click reactions
It is important to acquire information on the changes in the
spectroscopic properties and kinetics of the regioisomers during
Please cite this article as: M. Mori et al., Convenient synthesis of regioisomerically pure 5- and 6-functionalized xanthene dyes via SNAr reaction
and comparison of their reactivity towards click reaction, Tetrahedron, https://doi.org/10.1016/j.tet.2020.131087

6
M. Mori et al. / Tetrahedron xxx (xxxx) xxx
Scheme 3. The synthesis route for 5- and 6-substituted DCF.
click reactions. To this aim, we performed a comparative study of
the regioisomers of N3DCF in reactivity with an aliphatic cyclo-
octyne, BCN (Fig. 4A). BCN has unique properties among various
cycloalkynes used for SPAAC and is more reactive with electron
deficient azides [28]. We investigated whether the rate of the
SPAAC reaction of BCN for the 5- or 6-isomer differs in different
solvents. To examine the reactivity, each isomer of N3DCF was
incubated with BCN in phosphate buffered saline (PBS, pH 7.4) or
several aprotic solvents for 30 min at room temperature. The re-
action mixture was subjected to HPLC analysis to analyze the
conversion of substrate to product. The data are summarized in
Fig. 4B and C. When the SPAAC reaction was carried out in PBS,
comparable product formation was observed between the two
isomers. In contrast, in HMPA, the SPAAC reaction of N3DCF yielded
1.5-fold more product derived from the 6-isomer than the 5-isomer
(Fig. 4B). These results indicate that in HMPA, the 6-azide group is
more reactive with BCN than the 5-azide group, consistent with the
results obtained from the synthesis of 6-N3DCF. Absorption and
fluorescence spectra of each isomer of DCF were analyzed before
and after the SPAAC reaction (Fig. 4C and D). During the click re-
action, the change in maximum absorption or emission wavelength
of 5-N3DCF was only about 1 nm, whereas the 6-isomer showed a
slightly larger bathochromic shift in both absorption and emission
maxima, about 3 and 4 nm, respectively. In line with this obser-
vation, the quantum yield changed greatly from 0.727 to 0.744 for
the 6-isomer, but only from 0.702 to 0.706 for the 5-isomer. These
Fig. 4. Reactivity on SPAAC and change of spectroscopic properties of 5- or 6-N3DCF. (A) Reaction of 5- and 6-N3DCF with a BCN derivative. (B) Reactivity of 5- (blue column) and 6-
N3DCF (red column) with the BCN derivative in the SPAAC reaction. HPLC analysis of the reaction mixture gave peak areas corresponding to each compound (absorbance at 505 nm).
The reaction was carried out in the indicated solvent containing either 5-N3DCF or 6-N3DCF and the BCN derivative for 1 h at 37 ꢂC. Conversion was defined as the percentage of the
formation of click product from the corresponding N3DCF. (C, D) The normalized absorption and fluorescence spectra of 5-N3DCF (C) and 6-N3DCF (D). “N3DCF Click” represents the
reaction product of the SPAAC reaction.
Please cite this article as: M. Mori et al., Convenient synthesis of regioisomerically pure 5- and 6-functionalized xanthene dyes via SNAr reaction
and comparison of their reactivity towards click reaction, Tetrahedron, https://doi.org/10.1016/j.tet.2020.131087

M. Mori et al. / Tetrahedron xxx (xxxx) xxx
7
[5] A.H. Coons, J. Hugh, C.R. Norman, B. Ernst, The demonstration of pneumo-
coccal antigen in tissues by the use of fluorescent antibody, J. Immunol. 45
(1942) 159e170.
[6] M.L. Smith, T.R. Carski, C.W. Griffin, Modification of fluorescent-antibody
procedures employing crystalline tetramethylrhodamine isothiocyanate,
J. Bacteriol. 83 (1962) 1358e1359.
[7] M. Abo, et al., Development of a highly sensitive fluorescence probe for
hydrogen peroxide, J. Am. Chem. Soc. 133 (2011) 10629e10637.
[8] A. Wieczorek, P. Werther, J. Euchner, R. Wombacher, Green- to far-red-
emitting fluorogenic tetrazine probes-synthetic access and no-wash protein
imaging inside living cells, Chem. Sci. 8 (2017) 1506e1510.
[9] F.J. Huo, J. Kang, C. Yin, J. Chao, Y. Zhang, Highly selective fluorescent and
colorimetric probe for live-cell monitoring of sulphide based on bioorthogonal
reaction, Sci. Rep. 5 (2015) 2e6.
[10] A. Rotman, J. Heldman, Intracellular viscosity changes during activation of
blood platelets: studies by fluorescence polarization, Biochemistry 20 (1981)
5995e5999.
[11] A. Salic, T.J. Mitchison, A chemical method for fast and sensitive detection of
DNA synthesis in vivo, Proc. Natl. Acad. Sci. U. S. A 105 (2008) 2415e2420.
[12] C. Büll, et al., Steering siglecesialic acid interactions on living cells using
bioorthogonal chemistry, Angew. Chem. Int. Ed. 56 (2017) 3309e3313.
[13] K. Ajtai, et al., Stereospecific reaction of muscle fiber proteins with the 5’ or 6’
isomer of (iodoacetamido) tetramethylrhodamine, Biochemistry 31 (1992)
12431e12440.
[14] F. Stagge, G.Y. Mitronova, V.N. Belov, C.A. Wurm, S. Jakobs, Snap-, CLIP- and
halo-tag labelling of budding yeast cells, PloS One 8 (2013) 1e9.
[15] T. Hirano, K. Kikuchi, Y. Urano, T. Higuchi, T. Nagano, Highly zinc-selective
fluorescent sensor molecules suitable for biological applications, J. Am.
Chem. Soc. 122 (2000) 12399e12400.
[16] Y. Fujikawa, Y. Urano, T. Komatsu, K. Hanaoka, H. Kojima, T. Terai, H. Inoue,
T. Nagano, Design and synthesis of highly sensitive fluorogenic substrates for
glutathione S-transferase and application for activity imaging in living cells,
J. Am. Chem. Soc. 130 (2008) 14533e14543.
results suggest that the ground electronic state of the 6-N3DCF is
more susceptible to derivatization by the SPAAC reaction than that
of the corresponding 5-isomer.
4. Conclusion
Here we report a simple procedure for the preparation of each
isomer of 5(6)-functionalized DCF and TMR via regioselective SNAr
reaction. In addition to those described here, this synthetic strategy
is also applicable to other xanthene-based fluorophores such as
rhodol and X-rhodamine, as long as optimal conditions such as
solvents are determined. Therefore, this strategy is very versatile
and simple for the preparation of labeling reagents or synthetic
intermediates based on different regioisomers. Comparison of the
labeling kinetics and the spectral and photophysical properties for
each isomer highlighted the importance of preparing and
comparing two different regioisomers. In conclusion, the synthetic
strategy presented here helps expand the lineup of dyes that enable
protein bioconjugation in chemical biology and biomaterials.
Declaration of competing interest
The authors have no competing interests to declare.
Acknowledgements
[17] H. Yu, Y. Xiao, H. Guo, From spirolactam mixtures to regioisomerically pure 5-
and 6-rhodamines: a chemodosimeter-inspired strategy, Org. Lett. 14 (2012)
2014e2017.
[18] G.Y. Mitronova, et al., New fluorinated rhodamines for optical microscopy and
nanoscopy, Chem. Eur J. 16 (2010) 4477e4488.
We appreciate Dr. Haruhiko Fukaya (Tokyo University of Phar-
macy and Life Science) for help with HRMS measurements and Dr.
Yasuteru Urano, Ryosuke Kojima (The University of Tokyo) for help
with absolute quantum yield measurements. This research was
supported by the Private University Research Branding Project of
the Japanese Ministry of Education, Culture, Sports, Science and
Technology; by MEXT/JSPS KAKENHI (18K05362) to Y.F.
[19] S.J. Dwight, S. Levin, Scalable regioselective synthesis of rhodamine dyes, Org.
Lett. 18 (2016) 5316e5319.
[20] J. Castro, A. Malakhov, K. Burgess, Synthesis of regioisomerically pure 5-
functionalized 20,70-Dichlorofluoresceins, Synthesis (Stuttg)
1224e1226.
7
(2009)
[21] J.R. Beck, Nucleophilic displacement of aromatic nitro groups, Tetrahedron 34
(1978) 2057e2068.
[22] H. Leonhardt, L. Gordon, R. Livingston, Acid-base equilibriums of fluorescein
and 2’,7’-dichlorofluorescein in their ground and fluorescent states, J. Phys.
Chem. 75 (1971) 245e249.
[23] D. Magde, G.E. Rojas, P.G. Seybold, Solvent dependence of the fluorescence
lifetimes of xanthene dyes, Photochem. Photobiol. 70 (1999) 737e744.
[24] A.E. Speers, G.C. Adam, B.F. Cravatt, Activity-based protein profiling in vivo
using a copper(I)-catalyzed azide-alkyne [3 þ 2] cycloaddition, J. Am. Chem.
Soc. 125 (2003) 4686e4687.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.tet.2020.131087.
References
[25] R.N. Hannoush, N. Arenas-Ramirez, Imaging the lipidome:
u-alkynyl fatty
[1] R.P. Haugland, M.T.Z. Spence, I.D. Johnson, A. Basey, in: The Handbook: A
Guide to Fluorescent Probes and Labeling Technologies, tenth ed., Molecular
Probes, 2005. Molecular Probes.
[2] L.D. Lavis, Teaching old dyes new tricks: biological probes built from fluo-
resceins and rhodamines, Annu. Rev. Biochem. 86 (2017) annurev-biochem-
061516-044839.
[3] G.Y. Mitronova, et al., Functionalization of the meso-phenyl ring of rhodamine
dyes through SNAr with sulfur nucleophiles: synthesis, biophysical charac-
terizations, and comprehensive NMR analysis, Eur. J. Org Chem. (2015)
337e349 (2015).
[4] G. Jiao, J.W. Han, K. Burgess, Syntheses of regioisomerically pure 5- or 6-
halogenated fluoresceins, J. Org. Chem. 68 (2003) 8264e8267.
acids for detection and cellular visualization of lipid-modified proteins, ACS
Chem. Biol. 4 (2009) 581e587.
[26] K. Garai, C. Frieden, Quantitative analysis of the time course of Ab oligomer-
ization and subsequent growth steps using tetramethylrhodamine-labeled Ab,
Proc. Natl. Acad. Sci. U. S. A 110 (2013) 3321e3326.
[27] X. Li, T.M. Kapoor, Approach to profile proteins that recognize post-
translationally modified histone ‘tails’, J. Am. Chem. Soc. 132 (2010)
2504e2505.
[28] J. Dommerholt, et al., Highly accelerated inverse electron-demand cycload-
dition of electron-deficient azides with aliphatic cyclooctynes, Nat. Commun.
5 (2014) 1e7.
Please cite this article as: M. Mori et al., Convenient synthesis of regioisomerically pure 5- and 6-functionalized xanthene dyes via SNAr reaction
and comparison of their reactivity towards click reaction, Tetrahedron, https://doi.org/10.1016/j.tet.2020.131087