A versatile route to red-emitting carbopyronine dyes for optical microscopy and nanoscopy

FULL PAPER

DOI: 10.1002/ejoc.201000343

A Versatile Route to Red-Emitting Carbopyronine Dyes for Optical

Microscopy and Nanoscopy

Kirill Kolmakov,^[a]Vladimir N. Belov,*^[a]Christian A. Wurm,^[a]Benjamin Harke,^[a]

Marcel Leutenegger,^[a]Christian Eggeling,^[a]and Stefan W. Hell*^[a]

Keywords: Fluorescence / Chromophores / Carbocycles / Fluorescent probes / Fluorescence spectroscopy

Biological microscopy favors photostable fluorescent markers

with large fluorescence quantum yields, low dark triplet state

population, good biocompatibility and absorption and emis-

sion maxima in the near-infrared, where cellular autofluo-

rescence is minimized. In the present study, carbopyronines

absorbing around 640 nm and emitting at around 660 nm,

with a low intersystem crossing rate (k_isc≈ 0.5ϫ 10⁶s^–1) and

excellent properties for cellular imaging were synthesized.

A general synthetic route to carbopyronines with functional

groups variable in the final steps of the synthesis or in the

resulting fluorescent dye is presented. Possessing two 2-me-

thoxyethyl groups, the parent dye is soluble in water and

most organic solvents. Demethylation of the dye or its precur-

sors is straightforward, clean, and furnishes compounds with

one or two 2-hydroxyethyl groups, which can be used for fur-

ther transformations. Modifications in the linker-containing

carboxy group are also possible. A multistep synthesis of the

dye starting from a simple precursor and utilizing a single

temporary protective group is described. The presented ap-

proach may be further applied to the design of caged carbo-

pyronines.

to supply a wide variety of fluorescent dyes, allowing the

selection of the marker with the most suitable characteris-

tics. Attempts to device and improve photostable red-emit-

ting dyes are being undertaken in a number of research

groups. Recent publications on this topic describe water-

soluble terrylenediimides,^[4a,4b]soluble quaterrylenedi-

imides^[4c]and bisanthenes,^[4d]new hydrophilic BODIPY de-

rivatives,^[4e–4g]squaraine dyes,^[4h]and dicyanomethylene di-

hydrofuranes.^[4i]However, some important pieces of data

on the photophysical properties of their bioconjugates and

microscopic applications of these compound classes are still

lacking.

Despite many attempts to design novel and improved

red-emitting dyes on the basis of different chemical classes,

the number of compounds that perform satisfactorily in

fluorescence-based microscopy is still limited. Here we re-

port on the synthesis of a structural scaffold that realizes a

whole range of novel red-emitting carbopyronine dyes with

improved solubility in water and variable chemical groups.

For the first time, an improved and detailed synthesis of a

carbopyronin scaffold is described that allows modifications

on the final product, i.e., a photostable dye with large fluo-

rescence quantum yield and the required absorption and

emission bands in the red.

Introduction

Fluorescent dyes that absorb in the far-red or near-infra-

red (IR) optical region are indispensable for a variety of

microscopic techniques in biology, physics and chemistry.

Irradiation at wavelengths above 600 nm is much less invas-

ive and minimizes the undesired background signal origi-

nating from cellular autofluorescence. Far-field optical

nanoscopy^[1]in its particular applications^[2]such as STED

(stimulated emission depletion) nanoscopy,^{[2a,2d,2h,2p]}

PALM (photoactivtion localization microscopy),^[2i,2s]

STORM (stochastic reconstruction microscopy)^[2j,2u,2v]or

GSDIM (ground state depletion with individual molecular

return)^[3]pose very strict and sometimes contradictory

requirements on the utilized fluorescent markers. Among

them, the most important are: large quantum yield of fluo-

rescence (greater than 0.5), low population of the dark trip-

let state, high photostability, good solubility in water, and a

reactive group with a linker for conjugation to biological

molecules such as proteins or nucleic acids. Finally, to make

the conjugation possible, the markers (usually dye active es-

ters) need to be stable enough in aqueous solutions. How-

ever, in practice, all of these requirements can seldom be

fulfilled by a single substance. Therefore, it is advantageous

[a] Department of NanoBiophotonics, Max Planck Institute for

Biophysical Chemistry,

Results and Discussion

Am Fassberg 11, 37077 Göttingen, Germany

Fax: +49-551-2012505

Choosing Carbopyronines

E-mail: vbelov@gwdg.de

shell @gwdg.de

Very recently we described novel red-emitting rhodamine

dyes of various polarities.^[4j]Due to large fluorescence

Supporting information for this article is available on the

WWW under http://dx.doi.org/10.1002/ejoc.201000343.

Eur. J. Org. Chem. 2010, 3593–3610

3593

V. N. Belov, S. W. Hell et al.

FULL PAPER

quantum yields, high photostabilities and variable hydro- ing a large bathochromic shift in the absorption and emis-

philicities, they compete well in microscopy with the com- sion bands of approximately 50 nm^[6]and probably chang-

mercially available near IR fluorophores such as ATTO ing the rate of triplet state formation. Due to the bathoch-

633, ATTO 647N, Alexa 633, and Alexa 647. The presented romic shift provided by the carbopyronine core, there is no

rhodamine dyes have maxima of absorption and emission need for four fluorine atoms in the o-substituted benzoic

at 630–640 nm and 660 nm, respectively. This allows the use acid residue, as in the previously synthesized rhodamine

of convenient excitation sources such as He–Ne, diode or dyes.^[4]

krypton ion lasers. However, a limit has been reached for

further red-shifting the absorption and emission maxima of because the fluorinated aromatic ring drastically decreases

the practically useful rhodamines.^[5]

the solubility of a dye in water and increases its lipophilicity.

In fact, the presence of fluorine atoms is a drawback,

In our search for novel IR fluorophores, we decided to As a consequence, a very hydrophilic group (or groups;

switch to a class of dyes other than rhodamines. Carbopy- such as two sulfonic acid residues) was necessary to com-

ronine dyes are chemically very similar to the rhodamines pensate these factors.^[4]Besides that, four fluorine atoms

(see Scheme 1 for structures). Compared to rhodamines, the increase the molecular mass, which is not desirable for a

oxygen atom at position 10 of the xanthene fragment is re- fluorescent label. The carbopyronines, we anticipated,

placed by the geminal dimethyl group [C(CH₃)₂], introduc- would not require such a polar group (SO₃H) to be water-

soluble.

Synthetic Pathway to Carbopyronines and Possibilities for

Dye Design

The synthesis of carbopyronines is a rather challenging

task. In all available publications, the syntheses of these

dyes are described either incompletely or for very simple

derivatives only.^[6a–6c]As can be seen in the patents, many

important details, particularly in regard to the preparation

of key intermediates, are not disclosed at all.^[6d,6e,7]Despite

the large number of compounds claimed as examples, it is

not clear whether it is easy or not to vary the residues in

carbopyronine-containing scaffolds so that their properties,

particularly polarity and/or water solubility, would be con-

trolled.

Double bonds in bridges and open chains are more prone

to photooxidation (as occurs in cyanine dyes) than the un-

saturated cyclic π-systems of the main chromophores in

rhodamines and carbopyronines. The exclusion of such

double bonds from the scaffold was thus expected to im-

prove the photostability. In addition, double bonds are

known to reduce the fluorescence quantum yield of some

carbopyronines.^[7]Other important requirements for poten-

tial candidates are the presence of a hydrophilic group (or

groups), the presence of a sterically unhindered reactive site

for conjugation, and a moderate molecular mass (the ab-

sence of excessive side chains or extra aromatic rings).

A literature survey, complimented by our own experience

in the synthesis of rhodamine derivatives,^[2n–2r]led us to a

particular structural scaffold of pyronine fluorescent dyes

(Scheme 1). The scaffold combines several features that, to-

gether, meet all the requirements of an optimized fluoro-

phore in the context of micro- and nanoscopic methods.

Scaffold 1, accompanied by its retrosynthetic analysis

(Scheme 1), illustrates our approach, which was intended

to be uniform for carbopyronines and capable of providing

derivatives with variable functional groups and variable

properties. Apart from the absence of double bonds, some

other structural features providing high fluorescence quan-

Scheme 1. General retrosynthetic scheme for a carbopyronine scaf-

fold (1) with variable substituents.

tum yields and chemical stability were introduced.^[8]

3594

www.eurjoc.org

Eur. J. Org. Chem. 2010, 3593–3610

Red-Emitting Carbopyronine Dyes

First, we decided to install methylene bridges that would block in the dye design, we utilized an N-methyl-β-alanine

“rigidize” the fluorophore molecule on both sides, thus in- bridge, which was used in our previous studies.^[2p,4]Esterifi-

creasing the fluorescence quantum yield (Rhodamine 101, cation or amidation of the carboxylic group in the o-substi-

whose Φ_fl.is 1.0, and a significant number of other com- tuted phenyl ring provides an additional redshift in absorp-

mercial dyes have these bridges). The chemical stability is tion and emission spectra (ca. 5–10 nm; compared to a

also increased, when the planarity of the π-system improves compound with the free carboxylate). Also importantly, the

the delocalization of the positive charge. Moreover, the two β-alanine residue has a sterically unhindered carboxyl

aromatic CH groups are blocked by the alkyl groups and group that is suitable for further conjugation reactions.

are thus no longer prone to oxidation. Despite the merits

of the julolidine fragment (which is present in Rhodamine

101 and carbopyronine fluorescent dye ATTO 647N),

where the nitrogen atom is incorporated into the junction

Description of the Synthesis

of three cycles, we decided not to use this structural feature

The first crucial step of the synthesis is the condensation

because it would have been impossible to attach two (vari- of building blocks B and C, both of which are prepared

ous) groups to the nitrogen atoms. In the synthetic ap- from the same precursor A. According to the patent literat-

proach that we suggest here, the modifications at these het- ure,^[6d–6e,7]the condensation is best assisted by BCl₃and

eroatoms were expected to play the key role (see the modi- followed by cyclization at elevated temperatures in mineral

fications of scaffold 1 outlined in Scheme 1).

acid media. For most amino-protective groups these condi-

The initial transformations in Scheme 1 are based on the tions are prohibitively severe. The requirement of smooth

approach first described by Frantzeskos for the simplest and clean deprotection, on the other hand, leaves only the

carbopyronine dyes – N,NЈ-dimethyl derivatives (without benzyl group (Bn) as a viable candidate. The whole se-

methylene bridges).^[6b]As explained above, our approach quence outlined in Scheme 1 is unfeasible without the tem-

was flexible in regard to the variability of substituents. R⁴porary protective groups (R¹) and their removal in a later

at the nitrogen atoms in the final steps of the synthesis or step. The compatibility of the protective groups (PGs) and

even in the final compound, i.e., the fluorescent dyes 1a–c. reaction conditions is critical due to the wide variety of rea-

It is noteworthy that this option has not been explored be- gents used. There are clearly no groups that could resist all

fore. From the details that are available in the patent literat- the reagents (see Schemes 1, 2 and 3) and, importantly, be

ure,^[6d–6e,7]it can be concluded that the substituents at the smoothly removed in the end. As regards benzyl as PG,

nitrogen atoms in carbopyronine dyes are attached in early we could not retain this group after the condensation step

steps of the synthesis, and thus cannot be easily modified. because the CH₂groups would have been oxidized. More-

In our approach, the whole pathway starts with one simple over, the debenzylation on Pd/C in the presence of hydrogen

precursor (compound A), which is transformed into two or its donors^[10](e.g., HCOOH or HCOONH₄)^[11]is very

building blocks (B and C) that are, in turn, utilized for the likely to also reduce the carbopyronine core, as long as the

subsequent condensation step. The building blocks B and third aromatic ring is attached. Due to these precautions,

C should bear temporary protective groups (not necessarily the benzyl group was used only for temporary protection

the same) that can withstand the drastic condensation con- (R¹in Scheme 1) and was removed after the condensation

ditions. These groups (R¹in Scheme 1) have to be changed step (Scheme 2).

in later steps to the required functional groups (R⁴) that

As will be shown below, the removal of the temporary

allow further chemical transformations. We purposely protective group (Bn) from dyes of this class went smoothly

planned scaffold 1 to be symmetrical because we expected (see Scheme 2), which already meant that some success in

to obtain maximum effect with two residues R⁴. The solu- carbopyronine dye design had been achieved. As long as

bility in water may be controlled by the hydrophilicity of this step is performed (Scheme 1), any residues (R⁴) that are

these groups. If only one of the two groups is modified, compatible with the reagents used in further steps can be

an additional coupling site can be provided. This might be attached. This, in turn, will increase diversity within Scaf-

important, especially when the carboxy group in the o-sub- fold 1.

stituted phenyl ring remains free for other kinds of modifi-

Both building blocks B and C originate from compound

cations (e.g., for caging, which involves the preparation of A. We considered 1,2,3,4-tetrahydroquinoline derivatives as

the colorless and photosensitive spiroamides for optical candidates for the starting compound A with methylene

nanoscopy^[2m–2r]and novel caged dyes^[9]).

bridges of a reasonable length. As regards building block

On the way to Scaffold 1, one last obstacle remained. B, there is one difficulty that had to be addressed. Whereas

Protective groups and residues (R¹and R⁴) that were suit- para-electrophilic substitution in A presents no difficulty,^[12]

able for the whole synthetic sequence starting from com- the meta-type substitution (directed by the protonated sec-

pound A (Scheme 1) needed to be found. Not only do these ondary aromatic amino group) does not proceed very easily

groups need to withstand the drastic conditions of some and cleanly (see ref.^[13]and Exp. Section for details).

reactions, but also need to be smoothly removed under con-

Thereafter, the best candidate for R⁴, the second and the

ditions that keep the other parts of the molecule un- final (if ever possible) PG in our synthesis, had to be selec-

changed. The latter requirement is especially important at tive. First, the dye (target compound) would be more practi-

the very end of the sequence. As an important building cally important if R⁴alone was polar (and/or hydrophilic).

Eur. J. Org. Chem. 2010, 3593–3610

www.eurjoc.org

3595

V. N. Belov, S. W. Hell et al.

FULL PAPER

According to a review,^[14]some “soft” demethylating

agents might be compatible with carboxymethyl and amido

groups. Particularly, the selective demethylation of a methyl

ether that also had a carboxymethyl (CO₂Me) function was

described.^[15]Once the reaction proceeds, the resulting 2-

hydroxyethyl group could be subjected to various transfor-

mations. The CO₂Me function, at the same time, would re-

main unchanged and could be hydrolyzed for conjugation

purposes in the very final stage. The methoxy group

(CH₃O) itself is far more polar than an alkyl group, and is

known to increase the solubility and reduce the crystallinity

of organic compounds. We had every reason to expect this

effect to be demonstrated in case of the 2-methoxyethyl sub-

stituent, thus increasing the polarity and hydrophilic prop-

erties of the final fluorescent dyes.

Thus, we developed a scheme for the preparation of car-

bopyronine dyes with variable substituents from one nitro-

gen-containing starting compound (A, Scheme 1). After the

scheme was put into practice, we found that, despite the

wide range of reagents utilized, the protective groups

needed to be switched only once or twice throughout the

whole synthesis. Schemes 2 and 3 depict the actual prepara-

tion of

a red-emitting fluorophore (compound 1c,

Scheme 3) that proved to be excellent for bioconjugation

and STED imaging, as expected. Except for the very first

steps (depicted in Scheme 1), the yields are average values

from 2–3 experiments (for details, see Exp. Section and Sup-

porting Information).

Our synthesis started with 1,2,3,4-tetrahydroquinoline

(4), as building block A (Scheme 2). To convert this into

building block B, a halogen (Br or I) was introduced into

position 7 in the aromatic ring. The stepwise transforma-

tion of the aromatic bromide into the isopropenyl residue

could be achieved by lithiation followed by reaction with

acetone and dehydratation of the tertiary carbinol so

formed.^[16]Initially, we started with nitration, as the only

known preparative method for the selective meta-substitu-

tion in compound 4 (an aromatic amine).^[13b]The transfor-

mation of a nitro group into a halogen is a classical method

that involves a Sandmeier reaction on the corresponding

amine. In the case of substrate 4, the required sequence was

longer: the NH group first had to be temporary protected

Scheme 2. Synthesis of the key precursor 3a for carbopyronine

dyes. Reagents and conditions (isolated yields are represented by

average values obtained from 2–3 experiments): (a) 98% HNO₃/

96% H₂SO₄, 10 °C, 2 h; (b) Ac₂O, pyridine, 90 °C, 1 h; (c) H₂, Pd/

C (10% Pd), r.t., 16 h; (d) KNO₂/glacial HOAc, KI, 0 °C, 24 h;

(e) 5% NaOH in aq. MeOH, reflux, 30 h; (f) BnCl, KI, K₂CO₃,

DMF, r.t., 2 h; (g) BnCl, KOAc, HOAc, DMF, r.t., 20 h; (h) Br₂,

Ag₂SO₄in H₂SO₄; (i) nBuLi (2.5  in hexanes), –78 °C,1 h; by an acetyl (Ac) or other electron-withdrawing group.

(j) Me₂CO, –78 °C, 0.5 h; (k) POCl₃/DMF, ClCH₂CH₂Cl, 0 °C; re-

The synthesis of iodide 5-H,I from the corresponding ni-

flux, 1 h; (l) KHSO₄, PhCl, 130–140 °C, 15 min; (m) NaBH₄,

tro compound 5a was carried out analogously to 6-iodo-

EtOH, r.t., 1 h; (n) BCl₃in CH₂Cl₂, r.t., 16 h, then P₂O₅and

2,3-dihydro-1H-indole, the preparation of which is lengthy,

H₃PO₄, 110 °C, 2 h; (o) Pd/C (10% Pd), MeOH/Et₂O, 40–50 °C,

yet generally high-yielding.^[17]The synthesis of the bromo-

0.5 h; (p) Br(CH₂)₂OMe, K₂CO₃, DMF, 100 °C, 2.5 h; (q) air oxy-

gen.

substituted analogue via a Sandmeier reaction has not been

reported. The best yields of N-benzyl derivatives (5-Bn,I, 6,

and 7-Bn) were achieved by direct benzylation; the re-

ductive benzylation with benzaldehyde gave lower yields, es-

Secondly, a very important option would be the ability to pecially for 6. The conditions used for the dehydratation

remove or to modify this group (or a part of it) so that the of carbinol 8 were much improved compared to the single

resulting molecule becomes even more polar and hydro- literature report that describes its simplest analogue – 3-

philic. As a suitable candidate for R⁴, we chose the 2-meth- (isopropenyl)-N,NЈ-dimethylaniline.^[6a]Particularly, using

oxyethyl group (-CH₂CH₂OCH₃). This group can resist oxi- this approach there was is no need for vacuum distillation,

dants under mild conditions, and can also be smoothly de- and the reaction time was much shorter when a high-boiling

methylated.

solvent was used (see Exp. Section). Also importantly, we

3596

www.eurjoc.org

Eur. J. Org. Chem. 2010, 3593–3610

Red-Emitting Carbopyronine Dyes

and cleaner than expected, considering the relatively long

timescales and the yields reported in the literature for some

simple substrates.^[11a,11b]

Compounds 12 and 13 are readily oxidized by air oxy-

gen, which is typical for carbopyronines, as established by

Frantzeskos.^[6b]In our synthesis, the preparation and isola-

tion of the colored form 14 was unnecessary, but its forma-

tion was witnessed by the deep-blue color that rapidly ap-

pears upon exposure to air, and was also confirmed by mass

spectroscopic analysis. For compound 14, further alkylation

at both nitrogen atoms is impossible once their nucleophi-

licity is lost (due to the presence of the delocalized positive

charge). Therefore, in our pathway, the alkylation has to

precede the oxidation (Scheme 1 and Scheme 2), and had

to be performed in an inert atmosphere. Before the reaction

was complete, we could observe the formation of the par-

tially alkylated (unsymmetrical) product 3b, the structure

of which was confirmed by MS and NMR spectroscopic

analysis. As regards the oxidation of compound 3a to

ketone 2, which is another important step of the synthesis,

a modest yield was reported for a far less sophisticated car-

bopyronine-containing substrate by using permanganate as

oxidant.^[6b]Expecting even poorer yields with compound

3a, we first tried some different oxidants – selenium dioxide

and chromic anhydride in 1,4-dioxane and acetic acid,

respectively. Unfortunately, only trace amounts of 2 were

detected, even though CrO₃worked well for some similar

substrates,^[18]and SeO₂is known to be a “softer” oxidant.

In our experiments with KMnO₄, when the protocol men-

tioned above was reproduced,^[6b]the yield was poor (15%).

However, the yield was increased to 70% by lowering the

reaction temperature and controlling the rate of KMnO₄

addition. It is noteworthy that, in our experiments, neither

acetic acid nor base (NaOH or Na₂CO₃) were helpful for

this oxidation reaction. In the next step of the synthesis, the

third aromatic ring, which bears a protected (precon-

structed) carboxyl function, is attached. To this end, the

corresponding aryllithium reagent (reagent 15, Scheme 3)

Scheme 3. The actual synthesis of the carbopyronine dye 1c:

(a) KMnO₄, Me₂CO, 0 °C; (b) 15, THF, –78 to 0 °C; MeOH,

AcOH; (c) aq. NaOH, r.t. 1 h; (d) 20% aq. HCl, 80 °C, 16 h;

(e) MeI/DMF, K₂CO₃, r.t.; (f) aq. KOH, r.t.; (g) 2% aq. H₂SO₄/

AcOH, 80 °C, 8 h; (h) POCl₃, ClCH₂CH₂Cl, 70 °C, 3 h;

(i) MeNH(CH₂)₂CO₂Me, Et₃N, CH₂Cl₂, –10 °C; (j) HATU [2-(1H-

7-azabenzotriazol-1-yl)-1,1,3,3-tetramethyluronium

hexafluoro-

phosphate] Et₃N, MeCN, 0 °C.

explored a shorter path to carbinol 8 via direct bromination was preprepared, then reacted with the ketone and, finally,

of 4 to bromide 7-H. The bromination of indoline and some the carboxyl function was liberated. As seen in a number

similar substrates in the presence of Ag₂SO₄in sulfuric acid of examples,^[19]4,4-dimethyl-2-oxazoline is widely used as a

was studied by Miyake and Kikugawa.^[13a]

carboxyl synthon. The deprotection is normally performed

According to these authors, the separation of the iso- by heating in aqueous HCl. Drexhage and co-workers used

meric bromides by means of gas chromatography was prob- 2-(2-bromophenyl)-4,4-dimethyloxazoline (reagent 15,

lematic (in fact, the separation was easier for the corre- Scheme 3; for its preparation see ref.^[19d]) to synthesize some

sponding dihydroindoles in the next step of their synthesis). carbopyronine dyes; these syntheses always involved acid-

However, when we applied the bromination procedure to assisted demasking procedures.^[6b,7]

1,2,3,4-tetrahydroquinoline (4), it was possible to isolate the

It was not clear that the same approach would be appli-

pure 7-bromo derivative 7-H by means of standard column cable to our substrates (16a or 19) because they contain 2-

chromatography. Even though the isolated yield of the re- methoxyethyl groups (R⁴), which are sensitive to mineral

quired regioisomer was modest (although only slightly less acids. In fact, most alkyl ethers undergo dealkylation in

than that achieved in HF/SbF₅superacidic medium, see lat- strong acid media.^[14]On the other hand, tert-butyl esters

er),^[13c]the bromination protocol is straightforward and al- are known to be cleaved by acids under milder condi-

lows five steps to be omitted (see Scheme 2). The condensa- tions.^[20]However, attempts to lithiate tert-butyl 2-bromo-

tion of compounds 10 and 11, and the subsequent cycliza- benzoate (Scheme 1) and use it as a substitute for the Li-

tion, were executed in one pot with excellent yields, so the containing reagent 15 failed. Arylation of ketone substrate

purification of carbopyronine 13 was not required. The Pd- 2 with reagent 15, followed by aqueous workup, led to

assisted debenzylation to compound 12 proceeded far easier amino ester 16a, the oxazoline ring of which has already

Eur. J. Org. Chem. 2010, 3593–3610

www.eurjoc.org

3597

V. N. Belov, S. W. Hell et al.

FULL PAPER

been cleaved (see Scheme 3). There are two details about the carboxyl groups can be modified (or protected) in a

this reaction that are worth mentioning: First, during the stepwise fashion.

preparation of reagent 15 (which is used in large excess),

Apart from acid hydrolysis of oxazoline derivatives, an

the use of excess tBuLi should be avoided because the latter alternative method for demasking carboxyl functions for

readily reacts with ketone 2 to form a side-product. Sec- acid-sensitive substrates is known;^[22]this approach makes

ondly, as our subsequent experiments showed, compound use of methylation with methyl iodide to a quaternary salt,

16a is sensitive to bases (note the base-assisted cyclization which is subjected to alkaline hydrolysis. We were unsure

to 17, see Scheme 3 and Exp. Section for details). Therefore, about applying this protocol to compound 16a for the fol-

to avoid complications, the reaction mixture was neutral- lowing reasons: First, in compound 16a, the oxazoline ring

ized (or acidified) before being diluted with water. The best is already cleaved, which means that the substrate differs

procedure we found was to pour the reaction mixture into from that described by Meyers et al.^[22]Secondly, and more

a cool methanolic solution of acetic acid.

importantly, compound 16a could be isolated and stored

Under basic conditions, instead of the expected normal only as its salt, which means that a base is required to liber-

hydrolysis of amino ester 16a, a fast rearrangement to com- ate the amino group for the alkylation. As mentioned

pound 17, a spiroamide, was observed, which represents the above, we have established that bases, particularly K₂CO₃,

“closed” (colorless) form of the corresponding primary promote a fast rearrangement to spiroamide 17 (as shown

amide. Such cyclization of primary amides is typical for in Scheme 3). In order to inhibit this undesired process be-

rhodamines^[21]and, probably, also for carbopyronines. Its fore the alkaline hydrolysis, we applied acetic anhydride and

driving force is the formation of stable non-ionic spiro com- ethyl isocyanate as reagents to protect the amino group in

pounds, such as lactones 18a and 18b (see Scheme 3). Being 16a. Unfortunately, neither of these approaches were suc-

an amide, compound 17 is stable towards alkaline hydroly- cessful: in both cases, complex reaction mixtures were ob-

sis; at elevated temperatures it decomposes to a mixture of tained in which compound 17 was detected as the major

colorless products. In contrast, heating in glacial acetic acid component. However, we found that full methylation of

in the presence of H₂SO₄, or with POCl₃in 1,2-dichloro- amine 16a can be an option. A large excess of MeI in N,N-

ethane, causes a “recyclization” to the corresponding oxaz- dimethylformamide (DMF) cleanly methylates the amine to

oline derivative 19.

the quarternary salt 16b (see Scheme 3), which can be sa-

Initial experiments on the acid hydrolysis of 16a with ponified with dilute alkali in a one-pot fashion with very

aqueous hydrochloric acid were surprising because the reac- high yield in approximately two hours at 0 °C. The separa-

tion proceeded through compound 18 as a major intermedi- tion of 1a from partially demethylated product 18b – a side

ate. It required a long time (more than 10 h) to complete, product of acid hydrolysis – is avoided because the demeth-

as indicated by HPLC monitoring. The picture got even ylation does not occur under basic conditions. However,

more complicated when dilute (2–5%) hydrochloric acid due to its longer reaction time, the acid hydrolysis proved

was used: again, spiroamide 17 was formed, the hydrolysis to be the only feasible way to prepare an asymmetrical de-

of which required much longer and the reaction did not rivative such as 18a (an example of scaffold 1 bearing two

proceed cleanly. Oxazoline derivative 19, together with de- different residues R⁴at the nitrogen atoms). The reaction

methylated compounds 18a and 18b, was also detected in can be monitored analytically and stopped at the appropri-

the reaction mixture. In concentrated HCl the hydrolysis ate time, which, as mentioned above, might be advan-

proceeds faster, but the results of the experiments proved tageous for further dye design.

difficult to reproduce and, more importantly, the 2-methox-

In the next step of the synthesis, an N-methyl-β-alanine

yethyl groups were found to have undergone complete de- bridge was attached to the carboxyl group of compound

methylation to form compound 18b (R¹= R²= H). Fortu- 1a by amidation (see Schemes 1 and 2). This involved the

nately, we found that when the HCl concentration is moder- formation of an acid chloride, followed by a one-pot reac-

ate (20–22%) and the reaction temperature is lower (80 °C), tion with an amine. Taking some risk of demethylation, we

compound 1a is obtained with good yields in an acceptable used POCl₃as in our previous study.^[4]Despite the high

timescale (15–18 h). We also established that, in acidic me- reaction temperature (70 °C) and the relatively long reac-

dia, an equilibrium exists between 16a and 19 (see Scheme 3 tion time (2–3 h), the one-pot procedure was clean and high

and Exp. Section). A useful feature of this method is that yielding (see Scheme 3 and Exp. Section). The saponifica-

partial demethylation does occur, and the mono methylated tion of ester 1b to free acid 1c also proceeded smoothly to

product 18a (with RЈ = H) is always formed in considerable provide the required carbopyronine-containing dye. To

amounts (1:4–1:2), depending on the reaction time. In fact, avoid hydrolysis of the amide bond^[2p]and formation of the

all three possible reaction products, 1a, 18a, and 18b are colorless spiro lactones (analogues of compound 1a), the

normally detected in the reaction mixture (see Scheme 3 alkaline hydrolysis was performed in a dilute solution under

and Exp. Section for details) and can be separated by col- mild conditions. Dye 1a was obtained as a dark-blue, water-

umn chromatography. In nonpolar solvents, the three com- soluble (Ͼ5%) solid with a high extinction coefficient and

pounds exist predominantly in their lactone forms, as con- intense red fluorescence in solutions (see Table 1 for the

firmed by NMR analysis and witnessed by their very pale spectral properties). The compound is also readily soluble

colors (almost colorless). The partially demethylated prod- in most organic solvents (except alkanes). Furthermore, its

uct 18a is a valuable intermediate in which the hydroxyl and hydroxysuccinimide ester (1-NHS) – the amino reactive de-

3598

www.eurjoc.org

Eur. J. Org. Chem. 2010, 3593–3610

Red-Emitting Carbopyronine Dyes

rivative that can be used for conjugation – formed smoothly

and proved to be stable enough to be isolated by standard

column chromatography using a water-containing mobile

phase.

Table 1. Spectroscopic properties of the fluorescent dyes (for struc-

tures see Schemes 2–5 and Figure 1) and the conjugate of dye 1c

with sheep anti-mouse antibodies.^[a]

λ_max[nm] λ_max[nm] εϫ10^–5

Φ_fl[%]^[b]

(abs.)

(fl.)

[^–1cm^–1] H₂O

MeOH

16a

636

639

640

641

640

637

644

657

661

662

661

663

664

660

669

1.14

0.89

0.70

0.80

0.70

0.68

0.67

0.94

1.5

62

58

53

56

54

80

65

73

70

68

65

70

69

68

–

1b

1c

20a

20b

21a

21b

22

Scheme 4. Demethylation of bis-N-methoxyethylated carbopyron-

ines 1b and 1c. Reagents and conditions: (a) BBr₃, CH₂Cl₂, 0 °C;

(b) Me₃SiI, CH₂Cl₂; (c) BBr₃, NaI, 15-crown-5, CH₂Cl₂, –40 to

0 °C; (d) 0.05  aq. NaOH, THF, 10 °C.

KK 114

ATTO 647N

Conjugate

antibody

1c

–

642

665

–

26^[c]

–

[a] λ_max(abs.), λ_max(fl.), ε, and Φ_fl. are absorption maxima, emission

maxima, extinction coefficient, and fluorescence quantum yield,

respectively, in H₂O. [b] ATTO 633 was used as a reference. [c] In

aqueous PBS buffer with degree of labelling (DOL) 4.4, see Exp.

Section for details.

Apart from demethylation, we also used dye 1c to at-

tempt a peptide-type modification in the linker that con-

tains the carboxyl group (N-methyl-β-alanine bridge). The

reagent of choice was -cysteic acid, which is an easily avail-

able bifunctional building block that bears a sulfonic acid

residue (see Scheme 5).^[23a]Even though dye 1c is well-solu-

ble in water (Ͼ5%), the presence of a sulfonic acid group

To explore the possibility of further modifications of dye might be crucial for applications requiring very hydrophilic

1a, we attempted a demethylation, which was potentially compounds (e.g., for cell microinjections). The high po-

the most troublesome step of the whole sequence. In the larity provides compounds with good affinity to glass

course of the demethylation, it was important to keep the (which is important, for example, in non-linear optical nano-

carboxyl methyl ester group intact because successfully lithography) and negative affinity to lipohilic fragments in

achieving this would demonstrate the possibility of further biomolecules. Furthermore, in the long run, it would be

stepwise modifications on the dye, the carboxyl group of very interesting to establish whether or not the SO₃H group,

which is protected up to a certain synthetic step. To this as a part of a linker, has an influence on the fluorescence

end we used methyl ester 1b as a test substrate. First, we quantum yield, the imaging performance in STED, or the

attempted to reproduce the classical protocols that utilize chemical and photostability of the dye. In addition, one

trimethylsilyl iodide (TMS-I)^[14]or BBr₃in the presence of could expect that an extra amino acid fragment may elong-

NaI and 15-crown-5 (see Scheme 4).^[15]None of these ap- ate the linking bridge and thus reduce undesired inter-

proaches were successful, which was unexpected. Instead, ference between the dye core and the labeled site. -Cysteic

we found that simple treatment with a solution of BBr₃in acid methyl ester hydrochloride was prepared by a known

dichloromethane furnished the desired demethylated com- procedure.^[23b]Starting from compound 1c, the modified

pound 20a in good yield. The corresponding acid 20b was methyl ester 21a, the free amido acid 21b, and its active

also formed as a by-product in moderate yields. Another 2,3,5,6-tetrafluorophenyl ester 21-TFP were obtained.

possible by-product, the 2-bromoethyl derivative (where These kinds of active esters may be used for labeling pur-

OH is replaced with Br, as takes place in many other sub- poses as alternatives to NHS esters.^[24]Active ester 21-TFP

strates),^[14]was not detected at all, which demonstrates the proved to be very unstable due to the close proximity of the

high selectivity of the demethylation. Accordingly, under SO₃H group to the free carboxy group in the dye molecule.

the same reaction conditions, compound 1c (the free acid) As a result, this compound did not have a purity exceeding

was cleanly demethylated to 20b. This may be significant 70% even immediately after the chromatographic fractions

for preparative chemistry because, as our literature survey were pooled at low temperatures; in the course of freeze-

showed, no reports on the demethylation of substrates con- drying, the concentration of this active ester dropped even

taining a tertiary nitrogen connected to a methoxyalkyl further. Unlike compound 1c, we failed to obtain the corre-

group have been published. Thus, the opportunity to sponding NHS-ester in this case. The lower stability of 21-

smoothly vary substituents in the final dye (depicted by TFP resulted in a lower degree of labeling (by a factor of

Scaffold 1) has been demonstrated.

two) compared to that achieved with active ester 1-NHS

Eur. J. Org. Chem. 2010, 3593–3610

www.eurjoc.org

3599

V. N. Belov, S. W. Hell et al.

FULL PAPER

Scheme 5. Peptide-type modification of carbopyronine dye 1c with -cysteic acid. Reagents and conditions: (a) 2-(1H-benzotriazol-1-yl)-

1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU), Et₃N, MeCN, -cysteic acid methyl ester, 0 °C; (b) 0.05  aq. NaOH, THF,

10 °C; (c) HATU, Et₃N, 2,3,5,6-tetrafluorophenol, MeCN, 0 °C; (d) HBTU, Et₃N, -β-alanine methyl ester, MeCN, 0 °C; (e) HATU, Et₃N,

MeCN, 0 °C; * HPLC purity of the product immediately after isolation.

(which is far easier to manipulate). In addition, the conju-

Along with the general synthetic approach depicted in

gates obtained with 21-TFP were insufficiently bright under Scheme 1, we also tried an alternative pathway that utilized

the conditions of STED imaging. To improve the perform- a decorated benzhydrol derivative (structure 23, Scheme 6)

ance of the modified dye 21b, we extended the linking group as a building block in the cyclization step. As shown in

by an additional β-alanine fragment, which only slightly in- Scheme 6, compound 23 already has an extra benzene ring

creased the overall molecular weight. Dye 22, which was with a masked carboxylate function (in contrast to the pri-

spectrally identical to 21b (see below), and its active ester mary carbinol 10 in Scheme 2). In this approach, we at-

22-NHS were prepared by conventional methods. As ex- tempted to build a carbopyronine core with three aromatic

pected, ester 22-NHS demonstrated very good hydrolytic rings (24) in an early step to bypass the oxidation with

stability, which is helpful for the standard conjugation pro- KMnO₄(which was considered to be potentially problem-

tocol.

atic because of the moderate or poor yields reported for

Scheme 6. Attempted alternative synthesis of carbopyronine dye 24 (N,NЈ-dibenzyl analogue of compound 19 in Scheme 2, b) via benz-

hydrol derivative 23. Reagents and conditions: (a) THF, –78 °C; (b) BCl₃in CH₂Cl₂, r.t.; (c) P₂O₅and H₃PO₄, 110–150 °C; (d) Bu₄NIO₄.

3600

www.eurjoc.org

Eur. J. Org. Chem. 2010, 3593–3610

Red-Emitting Carbopyronine Dyes

simpler analogs).^[6b]Also, importantly, benzhydrol 23 itself ganic dyes in water.^[26a]In contrast, the intersystem-crossing

can be prepared from aldehyde 9, which is an early precur- rates differ by a factor of three, i.e., k_isc≈ 0.5ϫ10⁶s^–1for

sor of our synthesis (Scheme 2), by conventional methods compound 1c and ATTO 647N, and k_isc≈ 1–2ϫ10⁶s^–1for

(see the Supporting Information for details). Despite all compounds 21b, 22, and KK114, respectively. Low intersys-

attempts, the BCl₃-assisted cyclization did not proceed. tem-crossing rates are favorable for fluorescence microscopy

Longer exposure (several days) at ambient temperature re- because they minimize dark state population and thus max-

sulted in only very slight conversion. Several products were imize signal yield and minimize photobleaching yield.^[26a]

formed, however, none of them produced the required blue The increased intersystem-crossing rate of compounds 21b,

color upon oxidation by periodate. Use of polyphosporic 22, and KK114, compared to compound 1c, are very likely

acid (in a mixture with H₃PO₄) and P₂O₅as condensation due to the presence of the sulfonic acid groups (see Table 2).

agents, was also unhelpful. These results are in accordance Interestingly, the presence or absence of these groups seems

with previous reports,^[6b]however, as reported in one to be more important than the nature of the fluorophore

patent,^[7]cyclization of this type in the presence of P₂O₅core (rhodamine or carbopyronine).

was possible although details were not available.

Spectroscopic Properties of the Carbopyronine Dyes

Table 1 lists the maxima of the absorption and emission

bands, the maximum absorption coefficients (ε), and the

fluorescence quantum yields of the free acids 1c, 20b, 21b,

and 22, methyl esters (1b, 20a, and 21a), and amino ester

16a. These data highlight the influence of the polar (hydro-

philic) groups. For comparison, the table also details the

maximum absorption coefficients (ε) and the fluorescence

quantum yields (Φ_fl) of the known red-emitting dyes

KK114 [see ref.^[4j]for compound 6 and ref.^[25]for applica-

tions] and ATTO 647N (a widely used commercial dye pro-

vided by Atto-Tec GmbH, Siegen, Germany; for structures,

see Figure 1). The positions of the absorption (636–641 nm)

and emission (657–664 nm) maxima, the absorption coeffi-

cients (ε ≈ 0.7–0.8ϫ10⁵^–1cm^–1) and fluorescence quan-

tum yields (Φ_fl≈ 0.5–0.6) of the newly synthesized carbopy-

ronine dyes are very similar, regardless of the presence or

Figure 1. Structures of rhodamine KK114 and the carbopyronine

dye ATTO 647N.

absence of the sulfonic acid residues and groups attached

to the benzoic acid site (see Scheme 2). This is in contrast

to the previously obtained rhodamine dyes, the sulfonation

of which had been observed to markedly (up to more than

20%) increase the fluorescence quantum yield.^[4]One can

Table 2. Rates of intersystem crossing (k_isc) and triplet state depop-

assume that the two sulfonic groups in the allylic positions

of the rhodamines have a significant effect on the fluores-

cence. Furthermore, in the new carbopyronines, changing

the COOCH₃group to COOH, or OCH₃to OH does not

seem to have any noticeable influence on Φ_fl. The values of

ε are slightly below those of the reference dyes KK114 and

ATTO 647N (ε ≈ 1–1.5ϫ10⁵^–1cm^–1) and the values of Φ_fl

are in the same range as that of ATTO 647N but below

those of KK114 (Φ_fl≈ 0.8).^[4]The fluorescence quantum

yields are large enough to record fluorescence correlation

spectroscopy (FCS) data of, e.g., compounds 1c, 21b, and

22, allowing the determination of the rates k_iscand k_Tof

the intersystem-crossing to and from the dark triplet

state.^[26a]We recorded the FCS data and determined the

rates in the same way as described in our previous paper.^[4j]

These values, which were obtained from aqueous solutions,

ulation (k_T) as determined by FCS for the new carbopyronine and

reference dyes with and without sulfonic acid groups.

k_isc(ϫ 10^–6)

k_T(ϫ 10^–6)

HO₃S group

1c

0.6

1.1

1

2

0.5

0.3

–

+

–

21b

22

KK 114

ATTO 647N

Use of the New Carbopyronine Dyes in Fluorescence

Microscopy

To evaluate the performance of the described dyes in mi-

are listed in Table 2. The triplet decay time of k_T

=

croscopic and nanoscopic applications, three dyes (1c, 21b,

0.3ϫ10⁶s^–1proved to be the same for all the dyes and cor- and 22) were coupled to antibodies and applied in immuno-

relates well with the values of k_Tdetermined for other or- fluorescence labeling studies. In the conjugated state, the

Eur. J. Org. Chem. 2010, 3593–3610

www.eurjoc.org

3601

V. N. Belov, S. W. Hell et al.

FULL PAPER

fluorescence quantum yield for dye 1c was found to be cence signal of the same area in the immunolabeled cell

26%. The decrease in Φ_fl. after bio-conjugation is well- samples over the course of several scans. Note the unexpec-

known and was expected.^[2p]ATTO 647N and the spectrally ted behavior of ATTO 647N in the absence of the STED

similar rhodamine dye KK114 (compound 6 in ref.^[4]and beam (Figure 3, a) and at high excitation powers (see also

Figure 1), were used as references.

Figure S2 in the Supporting Information). This effect may

After immunofluorescence labeling of the tubulin cyto- be explained in terms of breaking the nonfluorescent aggre-

skeleton in PtK2 cells (a well-known model structure), the gates of the dye molecules, which start to emit after dissoci-

fixed cells were imaged using conventional confocal micro- ation. Without STED light, under confocal conditions, the

scopy and stimulated emission depletion (STED) nano- new dyes are more photostable than rhodamine KK114,

scopy (Figure 2).

but bleach faster than ATTO 647N. However, under STED

conditions, the bleaching rates of the new dyes are enhanced

as compared to the exceptionally photostable fluorescent

dyes ATTO 647N and KK114. The increased photoreactiv-

ity under STED conditions may result from the higher ab-

sorption of the STED light by the dye in its first excited

singlet or triplet states.^[27]Nevertheless, the resolution and

brightness of the STED images are excellent for all new

dyes, except 21b (see Figure 2 above and Figure S1 in the

Supporting Information).

Figure 2. Conventional confocal microscopy (left panels) and

STED nanoscopy images (right panels) of microtubule in fixed

PtK2 cells immunolabeled with compound 1c (a), compound 22

(b), ATTO 647N (c), and KK114 (d).

We found that all the new dyes tested here were suitable

for immunofluorescence labeling using convenient standard

protocols, and the conventional as well as high-resolution

STED images were of sufficient brightness and signal-to-

noise ratio. For example, excellent high resolution images

were obtained for dyes 1c and 22 (see Figure 2). However,

unexpectedly, dye 21b gave the unstable NHS ester and pro-

vided images with considerably lower brightness.

Figure 3. Photostabilities of the dyes under conventional confocal

(a) and STED conditions (b). Relative change in the total fluores-

cence signal of the same area of the microtubule-immunolabeled

cell samples in the course of the progressive scanning. Average exci-

tation power 1 µW at 640 nm and average STED power 118 mW

To evaluate and quantify the photostability of the newly

developed dyes, bleaching curves under confocal (without

STED, Figure 3, a) and STED conditions (Figure 3, b) were

recorded. For this purpose, we compared the total fluores- at 760 nm (in the focal spot of the objective lens).

3602 Eur. J. Org. Chem. 2010, 3593–3610

www.eurjoc.org

Red-Emitting Carbopyronine Dyes

ramethylsilane as an internal standard (δ = 0 ppm) using signals of

the residual protons of CHCl₃(δ = 7.26 ppm) in CDCl₃, CHD₂OD

(δ = 3.31 ppm) in CD₃OD, HOD (δ = 4.75 ppm) in D₂O,

[D₅]acetone (δ = 2.04 ppm) in [D₆]acetone or [D₅]DMSO (δ =

2.50 ppm) in [D₆]DMSO. Multiplicities in the ¹³C NMR spectra

were determined by Attached Proton Test (ATP) measurements.

Low-resolution mass spectra (electrospray ionization, ESI) were

obtained with LCQ and ESI-TOF mass spectrometers [MICRO-

TOF (focus), Bruker]. A MICROTOF spectrometer equipped with

an ESI ion source (Apollo) and direct injector with LC autosam-

pler (Agilent RR 1200) was used to obtain high-resolution mass

spectra (ESI-HRMS). ESI-HRMS were also obtained with an

APEX IV spectrometer (Bruker). HPLC system (Knauer):

Smartline pump 1000 (2ϫ), UV detector 2500, column thermostat

4000 (25 °C), mixing chamber, injection valve with 20 µL loop for

the analytical column; 6-port-3-channel switching valve; analytical

column: Eurospher-100 C18, 5 µm, 250ϫ4 mm, 1.1 mL/min; sol-

vent A: H₂O (HPLC grade) + 0.1% v/v trifluoroacetic acid (TFA);

solvent B: MeCN + 0.1% v/v TFA; detection at 636 nm (if not

stated otherwise). The reactions were monitored by TLC on

MERCK ready-to-use plates with silica gel 60 (F₂₅₄). Column

chromatography: MERCK silica gel, grade 60, 0.04–0.063 mm.

While using MeCN/H₂O mixtures as mobile phase, compressed air

(CAUTION: pressure!) was applied due to the high column resis-

tance. All compounds (including intermediates) were stored in a

refrigerator at about +5 °C, unless otherwise stated.

Conclusion and Outlook

The new red-emitting carbopyronine dyes can be used as

fluorescent labels in various microscopic and nanoscopic

applications. Although less photostable under STED condi-

tions than the spectrally similar dyes KK114 and ATTO

647N, they are excellent for STED nanoscopy and possess

certain valuable features that fluorophores KK114 and

ATTO 647N do not have. The flexible synthetic approach

presented here affords intermediate 18a with a free hydroxyl

group in the side chain and a carboxyl group in the benzene

ring. The hydroxyl group can be protected, and then the

carboxyl group may be transformed into new derivatives

with remarkable properties. For example, photosensitive

spiroamides^{[2m–2o,2r,21]}or novel caged (masked) carbopyr-

onines^[9b]may be obtained. Deprotection of the hydroxyl

group followed by its transformation into an amino or thiol

reactive site will provide the derivatives required for conju-

gation with biomolecules. Such options are impossible for

KK114 and ATTO 647N in which additional anchoring

sites cannot be incorporated unless the synthetic schemes

are changed completely.

Analogously to rhodamine spiroamides,^{[2m–2o,2r,21]}car-

boryronine amido derivatives must exist predominantly in

the “closed” colorless form (e.g., spiroamide 17). Some of

the spiroamides can be photoactivated and transformed

into the colored and fluorescent (zwitter)ionic state. Ther-

mal relaxation reaction of this “open” fluorescent form is

very likely to restore the initial nonfluorescent compound

so that the whole ring-opening and ring-closing sequence

may be repeated several times. Therefore, we expect carbop-

yronine spiroamides to be a valuable addition to the

multicolor toolbox of photochromic spiroamides. On the

other hand, the novel caging groups keep the initial cationic

dyes nonfluorescent and provide an irreversible transforma-

tion into the colored and highly fluorescent derivatives by

irradiation with violet or blue (visible) light.^[9b]

The real potential of the photochromic and caged red-

emitting dyes may be revealed in applications where track-

ing of dynamic processes is required or when these com-

pounds are used together with the spectrally similar conven-

tional fluorophores (e.g., KK114 or ATTO 647N). Thus,

the present work opens the way to novel photochromic and/

or caged carbopyronines, the utilization of which will help

to push the frontiers of optical microscopy and nanoscopy.

Fluorescence Microscopy, Sample Preparation, Materials and Meth-

ods: For immunolabeling, cultured PtK2 cells originating from the

marsupial (kidney epithelia), Potorous tridactylus, and cells of the

human osteosarcoma cell line U2OS, were grown on coverslips

overnight and fixed with absol. methanol (–20 °C). After washing

in PBS (137 m NaCl, 3 m KCl, 8 m Na₂HPO₄, 1.5 m

KH₂PO₄, pH 7) and blocking with 5% (w/v) bovine serum albumin

in PBS, the cells were incubated with mouse monoclonal antibodies

against α-Tubulin (Sigma–Aldrich, St. Louis, MO, USA). The pri-

mary antibodies were detected with secondary antibodies (sheep

anti-mouse; Jackson ImmunoResearch, West Grove, PA, USA) cus-

tom-labeled with ATTO 647N (AttoTec, Siegen, Germany),

KK114, 1c, 21b, and 22. After immunolabeling, the samples were

mounted in MOWIOL containing 0.1% (w/v) 1,4-diazabicy-

clo[2.2.2]octane (DABCO; Sigma–Aldrich). The STED setup used

for the high-resolution measurements was described before.^[28]The

pulsed excitation was implemented by a 640 nm laser diode (Pico-

quant GmbH, Germany), which was triggered by the photo diode

signal of the STED laser (MIRA 900, Coherent, USA) running at

760 nm. The donut-shaped intensity distribution required for the

STED beam was achieved by adding a phase plate with helical

phase retardation (RPC photonics, Rochester, NY, USA) to the

STED beam. Scanning was realized by a two-axis beam scanner

(Yanus IV, Till Photonics, Germany). Detection of the fluorescence

photons was performed with four avalanche photodetectors (APD,

Perkin–Elmer), coupled to a multi-mode fiber splitter acting as the

confocal pinhole (0.7 times Airy discs), due to the high photon

flux.

Experimental Section

General: UV/Vis absorption spectra were recorded with a Varian

Cary 4000 UV/Vis spectrophotometer, and fluorescence spectra

with a Varian Cary Eclipse fluorescence spectrophotometer. Reac-

tions were carried out with magnetic stirring in Schlenk flasks

equipped with septa or reflux condensers with bubble-counters un-

der argon, using a standard manifold with vacuum and argon lines.

N-Benzyl-7-iodo-1,2,3,4-tetrahydroquinoline

(5-Bn,I):

7-Iodo-

1,2,3,4-tetrahydroquinoline (5-H, I; for the preparation see the Sup-

porting Information) was benzylated with benzyl chloride (BnCl),

and the reaction product (5-Bn,I) was isolated as a hydrochloride.

Routine NMR spectra were recorded with a Varian MERCURY- In a typical experiment, 5-H,I (hydrochloride; 1.66 g, 5.6 mol) was

300 spectrometer operating at 300.5 (¹H) and 75.5 (¹³C and APT)

MHz. ¹H and ¹³C NMR spectra were also recorded with Varian

INOVA 600 (600 MHz) and Varian INOVA 500 (125.7 MHz) in-

struments, respectively. All NMR spectra are referenced to tet-

stirred overnight with finely powdered K₂CO₃(1.70 g, 12 mmol),

KI (1.66 g, 10 mmol), and BnCl (1.26 g, 10 mmol) in DMF

(10 mL). The reaction mixture was diluted with H₂O (30 mL), ex-

tracted with CH₂Cl₂(3ϫ30 mL), and the combined organic solu-

Eur. J. Org. Chem. 2010, 3593–3610

www.eurjoc.org

3603

V. N. Belov, S. W. Hell et al.

FULL PAPER

tions were dried (Na₂SO₄) and the solvents evaporated. The residue

was dissolved in Et₂O (50 mL), and the solution was decolorized

with silica gel (0.5 g) upon stirring (r.t., 15 min). The mixture was

filtered and commercial 5  HCl solution in 2-propanol (3 mL,

ACROS Organics) was added. After 30 min, the fine precipitate

was filtered, washed with hexane, and dried in air to furnish 5-H,I

(1.35 g, 69%, M = 385) as a salt. To obtain the free amine (5-H,I),

umn chromatography (260 g of SiO₂; hexane/CH₂Cl₂, 4:1). 5-

Bromo-1,2,3,4-tetrahydroquinoline and the starting material 4,

both of which are more polar then the title compound, were also

collected in subsequent fractions. The yield of compound 7-H (col-

orless solid, m.p. 68–69 °C) was 2.52 g (26%). The expected posi-

tion of bromine atom was confirmed by NMR analysis, and the

purity of compound 7-H was assessed by TLC (hexane/CHCl₃, 3:1;

1

the product was shaken with sat. aq. NaHCO₃(15 mL) and CH₂Cl₂R_f= 0.15). H NMR (300 MHz, CDCl₃): δ = 1.91 (m, 2 H, CH₂),

(20 mL), the aqueous layer was extracted with CH₂Cl₂(2ϫ15 mL)

and the combined organic solutions were dried (Na₂SO₄), decolo-

rized with silica gel (0.5 g) upon stirring (r.t., 15 min), and filtered.

The solvents were evaporated in vacuo at temperatures not ex-

ceeding 35 °C. In the course of the evaporation, hexane (2ϫ10 mL)

was added to the residue to completely remove the CH₂Cl₂(which

reacts with BuLi in the next step). Benzylated amine (5-Bn,I; 1.25 g,

64% from 5-H,I) was obtained as beige crystals with m.p. 61–62 °C;

2.64 (m, 2 H, CH₂), 3.23 (m, 2 H, CH₂), 3.90 (br. s, 1 H, NH), 6.68

(d, J = 9 Hz, 1 H), 6.73 (d, J = 9 Hz, 1 H), 6.78 (d, J = 9 Hz, 1 H)

ppm. ¹³C NMR (75.5 MHz, CD₃CN): δ = 21.7 (CH₂), 26.6 (CH₂),

41.6 (CH₂), 116.3, 119.4, 120.1, 130.7, 145.9 ppm.

N-Benzyl-7-bromo-1,2,3,4-tetrahydroquinoline (7-Bn): Compound

7-H (17 mmol, 3.70 g) was benzylated with benzyl chloride as de-

scribed for the iodo-substituted analogue 5-H,I (see above). The

reaction was monitored by TLC (hexane/CH₂Cl₂, 3:1), and the

product was isolated as a pale-brown oil in 70% yield (4.01 g) after

column chromatography (200 g of SiO₂; hexane/CH₂Cl₂, 4:1). ¹H

NMR (300 MHz, CDCl₃): δ = 1.96 (quint, J = 6 Hz, 2 H, CH₂),

2.64 (t, J = 6 Hz, 2 H, CH₂), 3.23 (t, J = 6 Hz, 2 H, CH₂), 4.39 (s,

2 H, CH₂Ar), 6.57 (m, 2 H), 6.78 (d, J = 8 Hz, 1 H), 7.16–7.40 (m,

5 H, Ph) ppm. MS (ESI+): m/z = 302 [M + H]⁺. HRMS: calcd.

for C₁₆H₁₆BrN [M + H] 302.0466; found 302.0463.

1

(CAUTION: the product is photosensitive). H NMR (300 MHz,

CDCl₃): δ = 1.98 (quint, J = 6 Hz, 2 H), 2.78 (t, J = 6 Hz, 2 H),

3.26 (t, J = 6 Hz, 2 H), 4.22 (s, 2 H, CH₂Ph), 6.64 (d, J = 8 Hz, 1

H), 6.82 (s, 1 H), 6.85 (dd, J = 8 Hz, 1 H), 7.20–7.40 (m, 5 H, Ph)

ppm. ¹³C NMR (75.5 MHz, CDCl₃): δ = 21.9 (CH₂), 27.9 (CH₂),

49.4 (CH₂), 54.8 (CH₂Ph), 92.2 (CI), 119.1, 121.8, 124.6, 126.6

(2ϫ), 127.0, 128.7 (2ϫ), 130.4, 138.0, 146.8 ppm. MS (ESI+): m/z

= 350 [M + H]⁺. HRMS: calcd. for C₁₆H₁₆IN [M + H] 350.0400;

found 350.0397.

The lithiation of 7-Bn (see Scheme 2, a) was performed as for 5-

Bn,I (the iodo-substituted analog, see above) to afford carbinol 8,

2-(N-Benzyl-1,2,3,4-tetrahydroquinolin-7-yl)propan-2-ol (8): In

a

1

which was identical to the previously obtained samples (TLC, H

typical experiment, commercial nBuLi (2.5 , 4.0 mL, 10 mmol)

was injected into a cooled 50 mL Schlenk Flask (external dry ice

bath, –78 °C) that had been flushed with argon and charged with

anhydrous THF (6 mL). A solution of 5-Bn,I (1.04 g, 2.9 mmol) in

anhydrous THF (12 mL) was introduced over a period of 6 min

with stirring. After 1 h stirring at this temperature, acetone (3 mL)

was injected in one portion and, after 30 min, the mixture was

poured into ice-cold 20% aq. NH₄Cl (50 mL). The aqueous layer

was separated and extracted with EtOAc (3ϫ20 mL), the com-

bined organic extracts dried (Na₂SO₄), and the solvents evaporated.

Carbinol 8 was isolated as a colorless oil in 75% yield (0.61 g)

by means of flash chromatography (30 g of SiO₂; hexane/EtOAc,

8:1Ǟ1:1). ¹H NMR (300 MHz, CDCl₃): δ = 1.26 (s, 6 H, CH₃),

1.82 (br. s, 1 H, OH), 2.04 (quint, J = 6 Hz, 2 H), 2.83 (t, J = 6 Hz,

2 H), 3.42 (t, J = 6 Hz, 2 H), 4.58 (s, 2 H, CH₂Ph), 6.76–6.80 (m,

2 H), 6.82 (s, 1 H), 7.02 (d, J = 8 Hz, 1 H), 7.32–7.40 (m, 5 H, Ph)

ppm. ¹³C NMR (75.5 MHz, CDCl₃): δ = 22.7 (CH₂), 28.1 (CH₂),

31.9 (CH₃), 50.2 (CH₂), 55.7 (CH₂Ph), 72.8 (C-OH), 107.7, 112.3,

121.1, 127.1 (2ϫ), 128.5, 128.9 (2ϫ), 129.1, 139.3, 145.6,

148.6 ppm. MS (ESI+): m/z (%) = 304 (100) [M + Na]⁺. HRMS:

calcd for C₁₉H₂₃NO [M + H] 282.1852; found 282.1854.

NMR), in 88% yield.

N-Benzyl-7-isopropenyl-1,2,3,4-tetrahydroquinoline (11): In a typical

experiment, carbinol 8 (281 mg, 1.00 mmol) in chlorobenzene

(1 mL) was placed in a screw-cap test tube containing KHSO₄

(136 mg, 1 mmol) and a magnetic stirring bar. The test tube was

flushed with argon, sealed, and vigorously stirred for 15 min in a

preheated oil bath at 140 °C (CAUTION: slight internal pressure!).

The reaction mixture was cooled, diluted with hexane (2 mL), and

transferred by means of a Pasteur pipette (decanting from the inor-

ganic precipitate) straight into a column charged with SiO₂(12 g),

which was eluted with hexane/CH₂Cl₂(6:1) to afford 218 mg (83%)

of a “TLC-pure” alkene 11 as a colorless photosensitive oil. ¹H

NMR (300 MHz, CDCl₃): δ = 2.06 (m, 2 H), 2.12 (s, 6 H, CH₃),

2.85 (t, J = 6 Hz, 2 H), 3.42 (t, J = 6 Hz, 2 H), 4.60 (s, 2 H, CH₂Ph),

5.00, 5.22 (2ϫs, 2 H, C=CH₂), 6.76 (s, 1 H), 6.80 (dd, J = 8 Hz, 1

H), 7.02 (d, J = 8 Hz, 1 H), 7.32–7.40 (m, 5 H, Ph) ppm. ¹³C NMR

(75.5 MHz, CDCl₃): δ = 21.9 (CH₃), 22.3 (CH₂), 27.9 (CH₂), 49.9

(CH₂), 55.3 (CH₂N), 108.4 (CH), 111.2 (CH₂), 113.4 (CH), 121.8

(C), 126.6 (2ϫCH), 126.7 (CH), 128.5 (2ϫCH), 128.7 (CH), 139.0

(C), 140.3 (C), 143.9 (C), 145.3 (C) ppm. MS (ESI+): m/z (%) =

264 (100) [M + H]⁺. HRMS: calcd. for C₁₉H₂₁N [M + H] 264.1747;

found 264.1749.

7-Bromo-1,2,3,4-tetrahydroquinoline (7-H): (Direct bromination of

1,2,3,4-tetrahydroquinoline). The protocol for indoline bromina-

tion^[13a]was applied to 1,2,3,4-tetrahydroquinoline (4). The sub-

strate 4 (6.03 g, 45 mmol) was added with vigorous stirring to a

cooled solution of Ag₂SO₄(7.67 g, 25 mmol) in concd. H₂SO₄

(70 mL) at 0 °C (ice bath) in a flask equipped with a reflux con-

denser, and Br₂(8.5 g, 53 mmol) was added within 10 min at 0 °C.

The mixture was then stirred until most of bromine had reacted

(2.5 h). An extra portion of Br₂(1.0 g, 6.3 mmol) was added and

stirring was continued for an additional 4 h. The reaction mixture

was poured onto crushed ice (300 g) and the solution was filtered.

The filtrate was poured onto crushed ice (500 g) and KOH (150 g),

and the filtration was repeated. The precipitate and the solution

were extracted with CH₂Cl₂(4ϫ80 mL) and the combined extracts

were washed with brine, dried (Na₂SO₄), and the solvents evapo-

rated. The title compound was isolated as a colorless solid by col-

Carbopyronines 12, 13, and 14: The condensation was carried out

as follows: A 100 mL-Schlenk flask fitted with a septum was

flushed with argon and charged with a CH₂Cl₂solution (40 mL)

of compounds 11 (618 mg, 2.35 mmol) and 10 [600 mg, 2.37 mmol;

prepared by routine methods from 1,2,3,4-tetrahydroquinoline (4),

as described in the Supporting Information], then with BCl₃(1 

in CH₂Cl₂, 2.70 mL, 2.70 mmol) at 0 °C, and the reaction mixture

was stirred at r.t. overnight. Polyphosphoric acid (VWR Inter-

national, 20 g) was mixed with 85% aq. phosphoric acid and heated

to 80–100 °C with manual stirring; the melt was allowed to cool to

r.t. and poured into the reaction flask. Through a thick cannula as

an outlet, the CH₂Cl₂was slowly evaporated in an argon purge

with stirring and slight heating. The temperature was raised to

110 °C and the mixture was maintained at this temperature for 2 h

3604

www.eurjoc.org

Eur. J. Org. Chem. 2010, 3593–3610

Red-Emitting Carbopyronine Dyes

with stirring under a slow argon flow (HCl gas evolved and the

viscous material completely dissolved within 15 min). The reaction

mixture was allowed to cool and then poured onto ice-cold H₂O

(120 mL) and CH₂Cl₂(100 mL). The viscous material remaining in

the flask was thoroughly washed out with MeOH (2ϫ10 mL), the

combined mixture was well stirred, and the organic phase was sepa-

rated. The aqueous layer was extracted with CH₂Cl₂(3ϫ50 mL)

and the combined organic extracts were washed with brine (30 mL)

and dried with anhydrous K₂CO₃. The solvent was completely re-

moved in vacuo to afford 1.10 g of residue, which consisted of com-

pound 13 and a very small amount of its oxidized (colored) form

14 (TLC; EtOAc/hexane, 1:1). In initial experiments, compound 13

was isolated in 85% yield by flash chromatography (hexane/EtOAc/

CH₂Cl₂, 2:1:1) as a pale-yellow amorphous material that rapidly

oxidized in air to a dark-blue dye with intense red florescence; the

mass spectrum of the oxidation product was in agreement with

structure 14: MS (ESI+): m/z (%) = 497 (100) [M]⁺. HRMS: calcd.

NMR 3b (300 MHz, CD₃CN): δ = 1.43 (s, 6 H, 2ϫCH₃), 1.82

(quint, J = 6 Hz, 4 H, 2ϫCH₂), 2.63 (t, J = 6 Hz, 4 H, 2ϫCH₂),

3.20 (t, J = 6 Hz, 2 H, CH₂), 3.28 (t, J = 6 Hz, 2 H, CH₂), 3.30 (s,

3 H, OCH₃), 3.45 (t, J = 6 Hz, 2 H, CH₂N), 3.56 (t, J = 6 Hz, 2

H, CH₂O), 3.75 (s, 2 H, ArCH₂Ar), 4.22 (br. s, 1 H, NH), 6.60 (s,

2 H), 6.68–7.00 (m, 1 H) ppm. MS (ESI+): m/z (%) = 399 [M

+ Na]⁺. HRMS: calcd for C₂₅H₃₂N₂O [M + H] 377.2587; found

377.2582.

The reaction mixture was diluted with H₂O (10 mL), extracted with

CH₂Cl₂(3ϫ20 mL), and the combined organic extract was washed

with H₂O (2ϫ 10 mL), dried (Na₂SO₄) and evaporated to furnish

3a (820 mg, 98%) as an amorphous grey solid, which was air-sensi-

tive, slightly soluble in hexane, and well-soluble in chlorinated sol-

1

vents. H NMR (300 MHz, CD₃CN): δ = 1.43 (s, 6 H, 2ϫCH₃),

1.82 (quint, J = 6 Hz, 4 H, 2ϫCH₂), 2.63 (t, J = 6 Hz, 4 H,

2ϫCH₂), 3.28 (t, J = 6 Hz, 4 H, 2ϫCH₂), 3.31 (s, 6 H, 2ϫOCH₃),

3.45 (t, J = 6 Hz, 4 H, 2ϫCH₂N), 3.56 (t, J = 6 Hz, 4 H,

2ϫCH₂O), 3.75 (s, 2 H, ArCH₂Ar), 6.76–6.78 (br. s, 4 H) ppm.

¹³C NMR (75.5 MHz, CD₃CN): δ = 23.3 (CH₂), 28.2 (CH₂), 29.7

(CH₃), 33.6 (CH₂), 50.9 (C), 52.0 (CH₂), 59.0 (OCH₃), 70.5 (CH₂),

108.2, 121.0, 124.0, 129.0 (2ϫ), 144.2, 144.5 ppm. HRMS: calcd.

for C₂₈H₃₈N₂O₂[M + Na] 457.2825; found 457.2829.

1

for C₃₆H₃₇N₂[M]⁺498.2951; found 497.2943. H NMR (for com-

pound 13, N-benzyl derivative, 300 MHz, CD₃CN): δ = 1.12 (s, 6

H, 2ϫCH₃), 1.82 (m, 4 H, 2ϫCH₂), 2.70 (t, J = 6 Hz, 4 H,

2ϫCH₂), 3.36 (t, J = 6 Hz, 4 H, 2ϫCH₂), 3.63 (s, 2 H, ArCH₂Ar),

4.42 (s, 4 H, CH₂Ph), 6.52 (s, 2 H), 6.76 (s, 2 H), 7.20–7.40 (m, 5

H, Ph) ppm. ¹³C NMR (75.5 MHz, CD₃CN): δ = 23.4 (CH₂), 28.2

(CH₂), 29.6 (CH₃), 33.6 (CH₂), 39.5 (C_q), 51.3 (CH₂), 56.2 (CH₂),

108.9 (CH), 121.2 (C), 124.2 (C), 127.5 (CH), 127.6 (2ϫCH), 127.8

(CH), 129.0 (CH), 129.3 (CH), 129.4 (2ϫC), 140.9 (C), 144.0 (C),

144.6 (C) ppm. MS (ESI+): m/z (%) = 497 (100) [M⁺for the oxid-

ized form 14]. See above for the HRMS of compound 14.

Ketone 2: To a solution of the crude compound 3a (820 mg,

1.89 mmol) from the previous step in acetone (20 mL), maintained

at –12 to –15 °C (external ice/salt bath), finely powdered KMnO₄

(633 mg, 4.00 mmoL) was added with vigorous stirring in small

portions (8ϫ80 mg, 15 min between portions) over a period of 2 h.

The solution was stirred for an additional 15 min, diluted with

CH₂Cl₂(800 mL), and filtered through a paper filter. The brown

precipitate was washed with additional CH₂Cl₂(2ϫ40 mL) and

the solvents were evaporated. Ketone 2 was isolated by chromatog-

raphy on SiO₂(90 g; hexane/EtOAc/CH₂Cl₂, 1:1:5) to remove the

less and more polar impurities. The main fraction afforded 2

(530 mg, 62% from compound 12) as a bright-yellow solid (m.p.

186–188 °C), which was photosensitive and gave intense green fluo-

rescence in solution. The reaction was monitored by TLC (EtOAc/

CHCl₃, 1:6). ¹H NMR (300 MHz, CD₃CN): δ = 1.61 (s, 6 H,

2ϫCH₃), 1.82 (quint, J = 6 Hz, 4 H, 2ϫCH₂), 2.76 (t, J = 6 Hz,

4 H, 2ϫCH₂), 3.33 (s, 6 H, 2ϫOCH₃), 3.42 (t, J = 6 Hz, 4 H,

2ϫCH₂), 3.60 (m, 8 H, 2ϫCH₂N, 2 ϫCH₂O), 6.78 (s, 2 H), 7.72

(s, 2 H) ppm. ¹³C NMR (75.5 MHz, CD₃CN): δ = 22.6 (CH₂), 28.3

(CH₂), 33.6 (CH₂), 38.3 (CH₂), 51.0 (C), 51.6 (CH₂), 59.1 (OCH₃),

70.6 (CH₂), 107.8, 119.8, 122.0, 127.8 (2ϫC), 149.6, 151.5, 180.5

(C=O) ppm. HRMS: calcd. for C₂₈H₂₆N₂O₃[M + H] 449.2799;

found 449.2792.

The removal of the benzyl protective group was carried out as fol-

lows: In an argon atmosphere, upon slight heating and with stir-

ring, the crude compound 13 was dissolved in a mixture of MeOH

(100 mL) and Et₂O (40 mL) in a flask equipped with a gas inlet, a

reflux condenser, and a bubble counter. Upon cooling to r.t., Pd/C

(10% Pd, oxidized form, VWR International; 1.50 g) and ammo-

nium formate (NH₄OOCH; 2.0 g, 32 mmol) were added, the flask

was flushed with Ar, and the reaction mixture was stirred for

50 min at 50 °C and at r.t. for 1 h. The mixture was diluted with

CH₂Cl₂(100 mL), filtered through Celite, and the solvents were

evaporated. (CAUTION: the Pd/C catalyst may cause ignition!). In

the course of the evaporation n-heptane (2ϫ20 mL) was added to

remove MeOH. The residue (air-sensitive!) was subjected to col-

umn chromatography (40 g of SiO₂; hexane/EtOAc/CH₂Cl₂, 10:5:1)

to furnish 615 mg (82%) of 12 as a pale-green crystalline powder

(m.p. 157–158 °C). ¹H NMR (300 MHz, CD₃CN): δ = 1.41 (s, 6

H, 2ϫCH₃), 1.82 (quint, J = 6 Hz, 4 H, 2ϫCH₂), 2.62 (t, J =

6 Hz, 4 H, 2ϫCH₂), 3.20 (t, J = 6 Hz, 4 H, 2ϫCH₂), 3.64 (s, 2 H,

ArCH₂Ar), 4.22 (br. s, 2 H, NH), 6.60 (s, 2 H), 6.74 (s, 2 H) ppm.

¹³C NMR (75.5 MHz, CD₃CN): δ = 23.3 (CH₂), 27.2 (CH₂), 29.2

(CH₃), 31.7 (CH₂), 34.3 (C), 42.6 (CH₂), 110.9 (CH), 119.6 (C),

125.4 (C), 129.2 (CH), 144.4 (C), 144.5 (C) ppm. MS (ESI+): m/z

(%) = 319 (100) [M + H]⁺, 317 (90) [M⁺for the oxidized form]⁺.

HRMS: calcd. for C₂₂H₂₆N₂[M + H] 319.2169; found 319.2168.

Amino Ester 16a: In an argon-flushed Schlenk flask (50 mL) fitted

with

a

septum, 2-(2-bromophenyl)-4,4-dimethyl-2-oxazoline^[19d]

(635 mg, 2.5 mmol) in anhydrous THF (12 mL) was lithiated with

tBuLi (1.5  in pentane, 1.75 mL, 2.63 mol) at –78 °C, and the mix-

ture was stirred for 2 h at this temperature to form the reagent 15.

A solution of ketone 2 (224 mg, 0.500 mmol) in anhydrous THF

(6 mL) was added and stirring was continued for an additional 6 h

at –78 °C, then overnight at 0 °C (ice bath). The reaction mixture

was poured into a stirred ice-cold solution of glacial HOAc (1 mL)

in MeOH (15 mL), and the resulting dark-blue solution was evapo-

rated in vacuo. The residue was separated on SiO₂(30 g; MeCN/

H₂O, 10:1) to separate the yellow and blue impurities, then eluted

with a MeCN/H₂O (3:1) mixture containing 0.2% (vol.) TFA. The

N-(2-Methoxyethyl) Carbopyronine Derivatives 3a and 3b: Com-

pound 12 (0.61 g, 1.92 mmol) in DMF (7 mL) was placed in a

screw-cap test tube containing 2-methoxyethyl bromide (3.90 g,

28 mmol) and finely powdered K₂CO₃(0.99 g, 7.2 mmol). The test

tube was flushed with argon, sealed, and the mixture was magneti-

cally stirred at 100 °C (CAUTION: slight internal pressure!) for

2.5 h, until the alkylation was complete [TLC monitoring (EtOAc/

hexane, 1:2)]. In the course of the alkylation the partially alkylated dark-blue fluorescent eluate was collected and the solvents were

product 3b (RЈ = H, Scheme 3, a) was detected: in the preliminary

experiments this compound was isolated by chromatography (hex-

ane/EtOAc, 7:1) and identified in ¹H NMR and MS spectra. ¹H

evaporated at a temperature below 38 °C in vacuo to a volume

of 10 mL. The residue was shaken with CH₂Cl₂(80 mL) and sat.

NaHCO₃(20 mL), and the organic layer was separated, dried

Eur. J. Org. Chem. 2010, 3593–3610

www.eurjoc.org

3605

V. N. Belov, S. W. Hell et al.

FULL PAPER

(Na₂SO₄), and evaporated to afford 16a (380 mg, 91%, bis-trifluo-

roacetate) as a dark-blue amorphous solid. HPLC: t_R= 11.2 min,

A/B 70:30–0:100 in 25 min. ¹H NMR (300 MHz, CD₃CN): δ = 1.32

1

18a: H NMR (300 MHz, CDCl₃): δ = 1.62 (s, 3 H, CH₃), 1.68 (s,

3 H, CH₃), 1.80 (m, 4 H, 2ϫCH₂), 2.50 (m, 4 H, 2ϫCH₂), 3.08

(m, 4 H, 2ϫCH₂), 3.38 (s, 6 H, 2ϫOCH₃), 3.46 (m, 4 H,

(s, 6 H, 2ϫCH₃), 1.76 (s, 3 H, CH₃), 1.80 (s, 3 H, CH₃), 1.88 (m, 2ϫCH₂N), 3.60 (m, 2 H, CH₂OMe), 3.82 (t, J = 6 Hz, 2 H,

4 H, 2ϫCH₂), 2.50 (t, J = 6 Hz, 4 H, 2ϫCH₂), 3.40 (s, 6 H, CH₂OH), 6.22 (d, J = 9 Hz, 2 H), 6.76 (s, 2 H), 6.82 (s, 2 H), 7.06

2ϫOCH₃), 3.64 (t, J = 6 Hz, 4 H, 2ϫCH₂), 3.78 (t, J = 6 Hz, 4 (d, J = 9 Hz, 1 H), 7.54 (m, 2 H), 7.98 (d, J = 9 Hz, 1 H) ppm.

H, 2ϫCH₂N), 3.92 (t, J = 6 Hz, 4 H, 2ϫCH₂O), 4.20 (s, 2 H,

CH₂O), 6.70 (s, 2 H), 7.28 (s, 2 H), 7.40 (d, J = 9 Hz, 1 H), 7.76–

7.90 (m, 2 H), 8.40 (br. s, 1 H, NH₃⁺), 8.40 (d, J = 9 Hz, 1 H) ppm.

¹³C NMR (75.5 MHz, CD₃CN): δ = 21.6 (CH₂), 26.3 (CH₃), 27.6

(CH₂), 28.2 (CH₂), 33.7 (CH₃), 33.9 (CH₃), 41.9 (C), 50.9 (CH₂),

¹³C NMR (75.5 MHz, CD₃CN): δ = 22.0 (CH₂), 27.6 (CH₂), 33.3

(CH₃), 35.1 (CH₃), 37.9 (CH₂), 50.3 (CH₂), 50.3 (CH₂), 51.3 (CH₂),

59.1 (OCH₃), 69.9 (C), 107.4, 108.1, 118.3, 121.6, 121.9, 124.1,

125.0, 127.2, 128.4, 128.5, 128.6, 134.3, 145.0, 145.3, 145.9, 146.5,

154.9, 171.0 (C=O) ppm. MS (ESI+): m/z (%) = 539 (100) [M +

52.0 (CH₂), 52.6 (CH₂), 59.2 (OCH₃), 70.7 (CH₂), 111.5, 121.2, H]⁺. HRMS: calcd. for C₃₄H₃₈N₂O₄[M + H] 539.2904; found

124.5 (CF₃), 127.8, 130.1, 131.2, 131.8, 133.3, 134.5, 138.4, 154.3,

156.1, 164.5 (C=O), 165.8 (CF₃C=O) ppm. MS (ESI+): m/z (%) =

624 (100) [M]⁺. HRMS: calcd. for C₃₉H₅₀N₃O₄[M]⁺624.3796;

found 624.3799.

539.2903.

18b: ¹H NMR (300 MHz, CD₃CN): δ = 1.60 (s, 3 H, CH₃), 1.78

(s, 3 H, CH₃), 1.80 (m, 4 H, 2ϫCH₂), 2.04 (br. s, 1 H, OH), 2.50

(m, 4 H, 2ϫCH₂), 3.30 (m, 4 H, 2ϫCH₂), 3.40–3.60 (m, 4 H,

2ϫCH₂N), 3.70–3.90 (m, 4 H, 2ϫCH₂O), 6.60 (m, 2 H), 7.06 (m,

2 H), 7.54 (m, 1 H), 7.60–7.80 (m, 3 H) ppm. MS (ESI+): m/z (%)

= 525 (100) [M]⁺. HRMS: calcd. for C₃₃H₃₇N₂O₄[M]⁺525.2748;

found 525.2743.

Carbopyronine Derivatives 1a, 18a, and 18b (Acidic Hydrolysis of

16a): In a typical experiment, the hydrolysis was carried out as

follows: Compound 16a (380 mg, 0.45 mmol) was dissolved in a

mixture of concd. HCl (20 mL, 0.24 mol) and H₂O (10 mL) and

heated with stirring in a flask fitted with a reflux condenser for 18 h

at 80 °C. The yellow solution was diluted with an equal volume of

H₂O, then CH₂Cl₂(40 mL) and EtOAc (60 mL) were added, and

the mixture was neutralized with solid NaHCO₃(22 g, 0.26 mol)

under vigorous stirring in a 600-mL beaker. The liquid was de-

canted from the small amount of solid and the aqueous layer was

separated and extracted with CH₂Cl₂(5ϫ40 mL), until the extract

became colorless. The combined organic layer was dried (Na₂SO₄),

evaporated in vacuo, and the residue was dissolved in a mixture of

MeCN (20 mL) and CH₂Cl₂(15 mL) and separated on SiO₂(80 g;

MeCN/H₂O, 20:1Ǟ5:1). The homogeneous fractions were pooled,

stepwise filtered through Rotilabo^®syringe filters (0.80 and

0.22 µm), and the solvents were evaporated in vacuo at a tempera-

ture not exceeding 38 °C. The following HPLC-pure compounds

were isolated (in order of elution): 1a (155 mg, 62%), 18b (66 mg,

27%), and 18c (7 mg, 3%), see Table 3. All the compounds were

pale-blue solids with decomposition temperatures of 176–178 °C;

they were almost colorless in non-polar solvents, dark-blue in

MeOH and MeCN, and turned blue on silica gel, particularly on

TLC plates. The reaction was monitored by HPLC (A/B 70:30–

0:100 in 25 min); t_R= 15.8, 13.2, and 10.4 min, respectively. The

yields values were in a good agreement [Ϯ(10–15)%] with the

HPLC areas.

Carbopyronine 1a (Alkaline Hydrolysis of 16b): Amino ester 16a

(20.0 mg, 23 µmol) was quaternized with MeI (0.10 mL,

1.50 mmol) in the presence of K₂CO₃(28 mg, 0.20 mmol) with vig-

orous stirring at r.t. for 2–4 h. The solvent was evaporated to dry-

ness at 0.2–0.4 mbar, and a solution of aq. NaOH (1 , 0.30 mL,

0.30 mmol) in a mixture of H₂O (1.5 mL) and THF (2 mL) was

added to the residue (containing the quaternary salt 16b; not iso-

lated, one spot on TLC). The resulting homogeneous solution was

left overnight at 4 °C, then partly neutralized with HOAc (20 µL,

0.36 mmol), evaporated in vacuo, and shaken with CH₂Cl₂(10 mL)

and H₂O (6 mL). The aqueous layer was separated, extracted with

an equal volume of CH₂Cl₂, and the organic extracts were washed

with aq. HCl (0.5 , 2 mL), sat. aq. NaHCO₃(2 mL), dried

(Na₂SO₄) and evaporated to furnish 10.3 mg (81%) of a compound

that was identical to the previously obtained samples of 1a (HPLC,

1

TLC, and H NMR analysis; see above).

Carbopyronine 1b (One-Pot Amidation of 1a): A 100-mL Schlenk

flask fitted with a septum was flushed with argon and charged with

1a (100 mg, 0.180 mmol) in 1,2-dichloroethane (18 mL) and POCl₃

(0.60 mL, 6 mmol). The dark-blue solution was stirred for 2.5 h at

70–72 °C (the pressure was relieved through a cannula in the first

15 min of heating). The solvent was then removed in vacuo, and

the residue was kept for an additional 1 h at 0.4–0.6 mbar and dis-

solved in CH₂Cl₂(15 mL) under argon. The flask was then cooled

to –10 °C (external dry-ice bath), and a solution containing methyl

3-(methylamino)propionate hydrochloride^[2p](92 mg, 0.60 mmol)

and Et₃N (0.13 mL, 0.9 mmol) in anhydrous MeCN (HPLC grade;

3 mL) was added in one portion with stirring. The solution was

warmed to r.t., stirred for an additional 15 min, and washed with

H₂O (10 mL) containing HOAc (0.60 mL, 10 mmol). The aqueous

layer was extracted with CH₂Cl₂(15 mL) and the combined organic

Table 3. Influence of reaction time on yields.

Reaction time [h]

Product yields (isolated) [%]

1a

18a

18b

14

18

22

65

62

52

13

27

26

–

3

5

1

1a: H NMR (300 MHz, CDCl₃): δ = 1.70 (s, 3 H, CH₃), 1.78 (s, layer was dried (Na₂SO₄) and the solvents evaporated. The residue

3 H, CH₃), 1.80 (m, 4 H, 2ϫCH₂), 2.48 (m, 4 H, 2ϫCH₂), 3.30 was purified by chromatography over SiO₂(35 g; MeCN/H₂O,

(m, 4 H, 2ϫCH₂), 3.38 (s, 6 H, 2ϫOCH₃), 3.46 (m, 4 H,

5:1Ǟ2:1). The main fraction was evaporated in vacuo at tempera-

2ϫCH₂N), 3.60 (m, 4 H, CH₂O), 6.20 (s, 2 H), 6.76 (s, 2 H), 7.06 tures below 37 °C to a volume of 10 mL, stepwise filtered through

(d, J = 9 Hz, 1 H), 7.52 (m, 2 H), 7.98 (d, J = 9 Hz, 1 H) ppm. Rotilabo^®syringe filters (0.80 and 0.22 µm), and freeze-dried to

¹³C NMR (75.5 MHz, CD₃CN): δ = 22.0 (CH₂), 27.6 (CH₂), 33.3

furnish the pure (HPLC, TLC) title compound (101 mg, 82%, hy-

(CH₃), 35.1 (CH₃), 37.7 (CH₂), 50.3 (CH₂), 51.3 (CH₂), 59.1 drochloride) as a bulky dark-blue solid. TLC: R_f= 0.30 (MeCN/

1

(OCH₃), 69.9 (C), 107.4, 118.4, 121.5, 123.9, 124.7, 127.1, 128.1,

128.5, 134.3, 144.7, 145.6, 155.6, 171.0 (C=O) ppm. MS (ESI+):

H₂O, 5:1); HPLC: t_R= 18.2 min (A/B 70:30–0:100 in 25 min). H

NMR (300 MHz, CDCl₃; signals of the major invertomer are

m/z (%) = 553 (100) [M + H]⁺. HRMS: calcd. for C₃₅H₄₀N₂O₄[M marked with an asterisk *): δ = 1.61 (s, 3 H, CH₃), 1.78 (s, 3 H,

+ H] 553.3061; found 553.3059.

CH₃), 1.88 (m, 4 H, 2ϫCH₂), 2.21 (t, J = 6 Hz, 2 H, CH₂β-Ala),

3606

www.eurjoc.org

Eur. J. Org. Chem. 2010, 3593–3610

Red-Emitting Carbopyronine Dyes

2.56 (t, J = 6 Hz, 2 H, CH₂β-Ala), 2.62 (m, 4 H, 2ϫCH₂), 2.82 calcd. for C₄₃H₅₁N₄O₇[M]⁺735.3752; found 735.3748. The mate-

(s, 3 H, NCH₃), 3.38 (s, 6 H, 2ϫOCH₃), 3.58 (s, 3 H, CO₂CH₃),

3.60 (m, 4 H, 2ϫCH₂N), 3.72 (m, 4 H, CH₂O), 3.42/3.82* (m, Σ

= 4 H, 2ϫCH₂), 6.70 (s, 2 H), 7.18 (s, 2 H), 7.24 (m, 1 H), 7.43

(m, 1 H), 7.58 (m, 2 H) ppm. ¹H NMR (300 MHz, CD₃CN; signals

rial was stored under argon at –20 °C for use in labeling studies.

Dyes 20a and 20b (Demethylation Products): To an ice-cold solution

of 1b (20 mg, 29 µmol) in CH₂Cl₂(10 mL), a commercial BBr₃

solution (1  in CH₂Cl₂, 0.17 mL, 0.17 mmol) was added and the

mixture was stirred for 2 h at 0 °C with TLC monitoring (MeCN/

H₂O, 7:1). The yellow solution was thoroughly washed with sat.

aq. NaHCO₃(6 mL), the aqueous layer was extracted with CH₂Cl₂

(3ϫ10 mL), the combined organic layer dried (Na₂SO₄), and the

solvents evaporated. The residue was purified by chromatography

over SiO₂(12 g; MeCN/H₂O, 6:1Ǟ2:1). The main fraction was

evaporated, filtered and freeze-dried, as described for 1b (see

above), to afford ester 20a (11 mg, 60%) as an amorphous solid.

The corresponding acid 20b was also isolated as a by-product in

15% yield. Under the same conditions and with the same workup,

acid 1c was demethylated to 20b (see below for the spectroscopic

data) with 74% isolated yield.

of

the

major

invertomer

are

marked

with

an

asterisk *): δ = 1.58/1.60* (s, Σ = 3 H, CH₃), 1.74*/1.80 (s, Σ = 3

H, CH₃), 1.85 (m, 4 H, 2ϫCH₂), 2.04 (t, J = 6 Hz, 2 H, CH₂β-

Ala), 2.50–2.60 (overlapped: m, 2 H, CH₂β-Ala, m, 4 H, 2ϫCH₂),

2.84*/3.06 (s, Σ = 3 H, NCH₃), 3.34 (s, 6 H, 2ϫOCH₃), 3.52 (s, 3

H, CO₂CH₃), 3.56 (m, 4 H, 2ϫCH₂N), 3.72 (m, 4 H, CH₂O), 3.38/

3.80* (m, Σ = 4 H, 2ϫCH₂Ar), 6.76 (s, 2 H), 7.18 (s, 2 H), 7.36

(m, 1 H), 7.45 (m, 1 H), 7.62 (m, 2 H) ppm. ¹³C NMR (75.5 MHz,

CD₃CN): δ = 21.7 (CH₂), 27.6 (CH₂), 30.3 (CH₂), 31.8 (CH₃), 32.3

(CH₂), 35.5/38.3 (CH₃), 42.0 (CH₂), 43.9 (CH₂), 50.9 (CH₂), 52.8

(CO₂CH₃), 59.3 (OCH₃), 70.8 (C), 70.9 (C), 111.7, 121.7, 124.2,

127.7, 129.7, 131.2, 124.9, 125.5, 131.8, 132.5, 135.6, 154.5, 156.3,

162.0, 169.2 (C=O), 172.8 (C=O) ppm. MS (ESI+): m/z (%) = 652

(100) [M]⁺. HRMS: calcd. for C₄₀H₅₀N₃O₅[M]⁺652.3745; found

652.3742.

The alkaline hydrolysis of 20a was carried out as for 1b (see above):

To a solution of 20a (11 mg, 17 µmol) in a mixture of THF (3 mL)

and H₂O (2 mL), aq. NaOH (0.1 , 0.6 mL, 0.06 mmol) was added

and the mixture was kept for 3 h at 5–10 °C, acidified with HOAc

(0.06 mL, 0.10 mmol), evaporated and separated over SiO₂(8 g;

MeCN/H₂O, 5:1Ǟ2:1). Solvent removal (as described above for

1b) afforded acid 20b (8.4 mg, 81%). HPLC: t_R(20a) = 11.5 min, t_R

(20b) = 9.7 min (A/B: 70:30–0:100 in 25 min). ¹H NMR (300 MHz,

CD₃CN; 20a: signals of the major invertomer are marked with an

asterisk *): δ = 1.58/1.61* (s, Σ = 3 H, CH₃), 1.66*/1.80 (s, Σ = 3

H, CH₃), 1.82 (m, 4 H, 2ϫCH₂), 2.02 (t, J = 6 Hz, 2 H, CH₂β-

Ala), 2.50–2.60 (overlapped: m, 2 H, CH₂β-Ala, m, 4 H, 2ϫCH₂),

2.82 (s, 3 H, NCH₃), 3.30–3.45 (br. m, 4 H, 2ϫCH₂N), 3.50 (s, 3

H, CO₂CH₃), 3.52–3.80 (br. m, 4 H, CH₂O), 3.75 (m, 4 H,

2ϫCH₂), 4.80–5.01 (br. m, 2 H, 2ϫOH), 6.68 (s, 2 H), 7.20 (s, 2

H), 7.33 (m, 1 H), 7.48 (m, 2 H), 7.61 (m, 1 H) ppm. MS (ESI+):

m/z (%) = 624 (100) [M]⁺. HRMS: calcd. for C₃₈H₄₆N₃O₅[M]⁺

624.3432; found 624.3430. 20b: ¹H NMR (300 MHz, CD₃CN): δ =

1.58/1.61* (s, Σ = 3 H, CH₃), 1.66*/1.80 (s, Σ = 3 H, CH₃), 1.82

(m, 4 H, 2ϫCH₂), 2.02 (t, J = 6 Hz, 2 H, CH₂β-Ala), 2.44 (t, J

= 6 Hz, 2 H, CH₂β-Ala), 2.58 (m, 4 H, 2ϫCH₂), 2.90 (s, 3 H,

NCH₃), 3.33–3.60 (br. m, 4 H, 2ϫCH₂N), 3.52–3.80 (br. m, 4 H,

2ϫCH₂O), 3.75 (m, 4 H, 2ϫCH₂), 4.80–5.01 (br. m, 2 H, 2ϫOH),

6.68 (s, 2 H), 7.20 (s, 2 H), 7.33 (m, 1 H), 7.48 (m, 2 H), 7.61 (m,

1 H) ppm. MS (ESI+): m/z (%) = 632 (100) [M + Na – H]⁺, 610

(30) [M]⁺. HRMS: calcd. for C₃₇H₄₄N₃O₅[M]⁺610.3275; found

610.3271.

Dye 1c (Free Acid): The amidation of 1a (0.180 mmol) was carried

out as described above. The crude amido ester 1b in CH₂Cl₂(30–

40 mL) was washed with sat. NaHCO₃(10 mL), dried (Na₂SO₄)

and the solvents evaporated. The residue was dissolved in aq.

NaOH (1 , 0.70 mL, 0.70 mmol) in a mixture of H₂O (10 mL)

and THF (15 mL). The saponification came to completion in 3 h

at 5–10 °C [TLC: R_f= 0.15 (1a; MeCN/H₂O, 5:1)]. The solution

was acidified with HOAc (0.6 mL, 1 mmol), evaporated to dryness

in vacuo at temperatures below 37 °C, and the residue was sepa-

rated on SiO₂(50 g; MeCN/H₂O, 5:1Ǟ2:1). The main fraction was

evaporated, filtered, and freeze-dried as described for 1b (see above)

to afford the title compound (88 mg, 76%) as a dark-blue solid.

HPLC: t_R= 15.5 min (A/B 70:30–0:100 in 25 min). ¹H NMR

(300 MHz, CDCl₃, signals of the major invertomer are marked

with an asterisk *): δ = 1.51*/1.60 (s, Σ = 3 H, CH₃), 1.66/1.72 (s,

3 H, CH₃), 1.88 (m, 4 H, 2ϫCH₂), 1.98*/2.22 (m, Σ = 2 H, CH₂

β-Ala), 2.44 (m, 4 H, 2ϫCH₂), 2.62–2.78 (m, 2 H, CH₂β-Ala),

2.67*/2.98 (s, Σ = 3 H, NCH₃), 3.36 (s, 6 H, 2ϫOCH₃), 3.60 (m,

4 H, 2ϫCH₂N), 3.72 (m, 4 H, CH₂O), 3.28/3.96* (m, Σ = 4 H,

2ϫArCH₂), 6.72 (s, 1 H), 6.80 (s, 1 H), 7.04 (d, J = 8 Hz, 2 H),

7.18–7.24 (m, 1 H), 7.42–7.60 (br. m, 3 H) ppm. ¹³C NMR

(75.5 MHz, CD₃CN): δ = 21.7 (CH₂), 27.6 (CH₂), 30.3 (CH₃), 31.8/

32.3* (CH₂), 36.8/38.0 (CH₃), 41.9 (CH₂), 43.0 (CH₂), 52.10 (C),

52.7 (CH₂), 59.3 (OCH₃), 70.8 (C), 111.5, 121.5, 121.6, 124.1,

124.2, 127.7, 127.9, 129.5, 129.6, 129.8, 131.2, 134.9, 135.5, 136.0,

154.4, 156.1, 156.2, 168.6/169.3* (C=O), 175.0/175.8* (C=O) ppm.

MS (ESI+): m/z (%) = 638 (100) [M + H]⁺. HRMS: calcd. for

C₃₉H₄₇N₃O₅[M + H] 638.3588; found 638.3604.

Dye 21a (Peptide-Type Coupling with L-Cysteic Acid): In a typical

experiment, compound 1c (4.0 mg, 6.3 µmol) in a screw-cap vial

was sonicated for 3 min in anhydrous MeCN (HPLC grade;

0.5 mL) and reacted with methyl cysteate hydrochloride^[23b]

(1.55 mg, 7.0 µmol) in the presence of HBTU (2.5 mg, 6.6 µmol)

and Et₃N (6.0 µL, 42 µmol) with stirring at 0 °C for 1 h. The reac-

tion solution was diluted with MeCN (3 mL) and charged straight

onto a column of silica gel (5 g). Elution with MeCN/H₂O (10:1)

followed by evaporation, filtration (Rotilabo^®syringe filters,

0.22 µm) and freeze-drying, afforded 21a (4.2 mg, 83%) as a dark-

blue amorphous solid, which was well-soluble in H₂O, MeOH,

Dye 1-NHS (The Active Ester): In a typical experiment, dye 1c

(5 mg, 8 µmol) in anhydrous MeCN (HPLC grade; 3–4 mL) was

treated with N-hydroxysuccinimide (14 mg, 0.120 mmol) in the

presence of HATU (12 mg, 30 µmol) and Et₃N (20 µL,

0.140 mmol) in a screw-cap vial at room temp. A conversion of ca.

90% was detected in 2 h, as shown by HPLC (t_R= 17.7 min; A/B

70:30–0:100 in 25 min) and by TLC (R_f= 0.50; silica plates; MeCN/

H₂O, 5:1). The reaction solution was loaded straight into a short MeCN, and CHCl₃. TLC: R_f= 0.30 (MeCN/H₂O, 10:1); HPLC:

1

column with SiO₂(5 g). The main fraction (MeCN/H₂O, 10:1) was

collected and quickly evaporated at r.t. in vacuo to a volume of ca.

10 mL, filtered through a Rotilabo^®syringe filter (0.22 µm), and

freeze-dried to afford 6–8 mg of a blue solid containing ca. 10–20%

of the starting acid (t_R= 15.5 min) and variable amounts of N-

hydroxysuccinimide. MS (ESI+): m/z (%) = 735 [M]⁺. HRMS:

t_R= 12.3 min (A/B: 70:30–0:100 in 25 min). H NMR (300 MHz,

CD₃CN): δ = 1.56/1.62* (s, Σ = 3 H, CH₃), 1.72*/1.82 (s, Σ = 3 H,

CH₃), 1.90 (m, 4 H, 2ϫCH₂), 2.20–2.30 (m, 2 H, CH₂β-Ala),

2.54–2.60 (overlapped: m, 2 H, CH₂β-Ala, m, 4 H, 2ϫCH₂),

2.82*/3.60 (s, Σ = 3 H, NCH₃), 3.36 (s, 6 H, 2ϫOCH₃), 3.58 (m,

4 H, 2ϫCH₂N), 3.60 (s, 3 H, CO₂CH₃), 3.68 (t, J = 6 Hz, 2 H,

Eur. J. Org. Chem. 2010, 3593–3610

www.eurjoc.org

3607

V. N. Belov, S. W. Hell et al.

FULL PAPER

CH₂SO₃), 3.50–3.80 (m,

4

H, CH₂O), 3.75/4.01 (m,

4

H,

(2 mL) at 4–5 °C overnight. The solution was neutralized with

2ϫCH₂Ar), 4.42 (br. s, 1 H, NH), 5.32 (m, 1 H, CH Cys), 6.78 (s, HOAc (60 µmol), evaporated to dryness in vacuo, and separated

2 H), 7.20 (d, J = 6 Hz, 2 H), 7.36 (m, 1 H), 7.62 (m, 2 H), 7.75

on SiO₂(7 g; MeCN/H₂O, 5:1Ǟ3:1). Filtration (Rotilabo^®syringe

filters, 0.22 µm), centrifugation, and freeze-drying, afforded the free

(d, J = 6 Hz, 1 H) ppm. ¹³C NMR (75.5 MHz, CD₃CN): δ = 21.7

(CH₂), 27.6 (CH₂), 31.8 (CH₃), 32.3 (CH₂), 35.5/38.1 (CH₃), 42.0 acid 22 (11 mg, 65%) as a dark-blue, fine crystalline powder that

(CH₂), 44.1 (CH₂), 51.0 (CO₂Me), 52.0 (CH₂), 52.5 (CH₂), 52.7 was well-soluble in H₂O and MeOH, very slightly soluble in

(CH), 59.2 (OMe), 70.8 (C), 70.9 (C), 111.5, 111.6, 121.7, 124.0,

127.8, 129.5, 131.0, 135.1, 135.2, 135.4, 137.3, 154.3, 156.0, 161.7*/

168.8 (C=O), 170.8/171.8* (C=O) ppm. MS (ESI+): m/z (%) = 825

(100) [M + Na]⁺. HRMS: calcd. for C₄₃H₅₄N₄O₉S [M + Na]

825.3504; found 825.3505.

MeCN, and insoluble in CHCl₃. TLC: R_f= 0.10 (MeCN/H₂O, 5:1);

HPLC: t_R= 9.2 min (A/B: 70:30–0:100 in 25 min). MS (ESI+): m/z

=

882 [M +

Na]⁺. HRMS (negative mode): calcd. for

C₄₅H₅₇N₅O₁₀S [M – H] 858.3753; found 858.3783. ¹H NMR

(300 MHz, [D₄]MeOH; signals of the major invertomer are marked

with an asterisk *): δ = 1.58/1.63* (s, Σ = 3 H, CH₃), 1.76*/1.80 (s,

Σ = 3 H, CH₃), 1.80 (m, 4 H, 2ϫCH₂), 2.20–2.30 (m, 4 H, 2ϫCH₂,

β-Ala), 2.36–2.40 (m, 4 H, 2ϫCH₂, β-Ala), 2.60 (m, 4 H, 2ϫCH₂),

2.65/2.91* (s, Σ = 3 H, NCH₃), 3.32 (m, 4 H, 2ϫCH₂N), 3.38 (s,

6 H, 2ϫOCH₃), 3.70 (t, J = 6 Hz, 2 H, CH₂SO₃), 3.75 (m, 4 H,

2ϫCH₂), 3.96 (m, 4 H, CH₂O), 4.40–4.60 (br. s, 2 H, 2ϫNH),

4.82 (overlapped: m, 1 H, CH Cys + HDO), 6.80 (d, J = 7 Hz, 2

H), 7.25 (d, J = 7 Hz, 2 H), 7.40 (br. m, 1 H), 7.65–7.70 (m, 3 H)

ppm. ¹³C NMR (75.5 MHz, CD₃CN): δ = 22.2 (CH₂), 24.0 (CH₂),

28.2 (CH₂), 32.0 (CH₃), 34.2 (CH₂), 36.1/36.8* (CH₃), 42.4 (CH₂),

45.1 (CH₂), 49.4 (CH₂), 49.6 (CH₂), 52.5 (CH₂), 52.6 (CH₂), 53.0

(CH), 59.5 (OCH₃), 71.3 (C), 112.0, 121.8, 121.9, 124.5, 128.5,

130.1, 130.1, 131.5, 135.6, 135.7, 135.8, 137.2, 154.9, 156.7, 161.5*/

170.5 (C=O), 172.0*/173.0 (C=O), 179.9*/180.1 (C=O) ppm.

Dye 21b (Free Acid): To a solution of 21a (74 mg, 0.92 mmol) in a

mixture of THF (15 mL) and H₂O (10 mL), aq. NaOH (1 ,

0.30 mL, 0.30 mmol) was added, and the mixture was kept for 2 h

at 0 °C, acidified with HOAc (0.10 mL, 1.80 mmol), evaporated

and purified by chromatography over SiO₂(36 g; MeCN/H₂O,

5:1Ǟ3:1). Solvent removal (as described for 1b, with filtration and

centrifugation) afforded acid 21b (66 mg, 91%) as a dark-blue fine

crystalline powder that was well-soluble in H₂O and MeOH, spar-

ingly soluble in MeCN, and insoluble in CHCl₃. TLC: R_f= 0.12

(MeCN/H₂O, 5:1); HPLC: t_R= 10.4 min (A/B: 70:30–0:100 in

1

25 min). H NMR (300 MHz, [D₄]MeOH; signals of the major in-

vertomer are marked with an asterisk *): δ = 1.59/1.62* (s, Σ = 3

H, CH₃), 1.78/1.92* (s, Σ = 3 H, CH₃), 1.82 (m, 4 H, 2ϫCH₂),

2.08–2.12 (m, 2 H, CH₂β-Ala), 2.36–2.40 (m, 2 H, CH₂β-Ala),

2.60 (m, 4 H, 2ϫCH₂), 2.66/2.92* (s, Σ = 3 H, NCH₃), 3.32 (m, 4

H, 2ϫCH₂N), 3.38 (s, 6 H, 2ϫOCH₃), 3.68 (t, J = 6 Hz, 2 H,

CH₂SO₃), 3.72 (m, 4 H, 2CH₂), 3.96 (m, 4 H, CH₂O), 4.45 (br. s,

1 H, NH), 4.82 (overlapped: m, 1 H, CH Cys + HOD), 6.78 (d, J

= 7 Hz, 2 H), 7.24 (d, J = 7 Hz, 2 H), 7.40 (br. m, 1 H), 7.64–7.66

(m, 3 H) ppm. ¹³C NMR (75.5 MHz, CD₃CN): δ = 21.7 (CH₂),

27.6 (CH₂), 31.8 (CH₃), 32.2 (CH₂), 35.3/38.4* (CH₃), 41.6 (CH₂),

44.0 (CH₂), 52.1 (CH₂), 52.6 (CH₂), 53.0 (CH), 59.2 (OCH₃), 70.9

(C), 71.0 (C), 111.4, 111.6, 121.9, 124.0, 128.1, 129.6, 131.1, 135.4,

135.0, 135.4, 137.3, 154.3, 156.2, 162.1*/169.0 (C=O), 171.0/171.9*

(C=O) ppm. MS (ESI+): m/z (%) = 833 [M + 2Na – H]⁺; MS (ESI–

): m/z (%) = 787 [M – H]^–. HRMS: calcd. for C₄₂H₅₂N₄O₉S [M +

H] 789.3528; found 789.3526.

Active Ester 22-NHS: Dye 22 (5 mg, 6 µmol) in anhydrous MeCN

(4 mL, HPLC grade) was treated with N-hydroxysuccinimide

(7 mg, 35 µmol) in the presence of HATU (17 mg, 45 µmol) and

Et₃N (7 µL, 49 µmol) in a screw-cap vial at room temp. The maxi-

mum conversion (82%) was reached in 2 h, as shown by HPLC (t_R

= 10.9 min; A/B: 70:30–0:100 in 25 min) and TLC (R_f= 0.20;

MeCN/H₂O, 5:1) analyses. The reaction solution was loaded di-

rectly into a short column with SiO₂(4 g; MeCN/H₂O, 10:1) and

the first colored fraction was collected, quickly evaporated in vacuo

at room temp. to the volume of about 10 mL, filtered through a

Rotilabo^®syringe filter (0.22 µm), and freeze-dried to afford 6 mg

of a blue solid material containing few percent of the starting acid

(t_R= 9.2 min) and N-hydroxysuccinimide (HPLC analysis with de-

tection at 636 nm showed a 93% area for the peak of 22-NHS). MS

(ESI+): m/z = 979 [M + Na]⁺. HRMS: calcd. for C₄₉H₆₀N₆O₁₂S [M

+ Na] 979.3882; found 979.3897. The material was stored under

argon at –20 °C for use in labeling studies.

Active Ester 21-TFP (2,3,5,6-Tetrafluorophenyl Ester of 21b): Dye

21b (3.6 mg, 4.6 µmol) in anhydrous MeCN (HPLC grade; 1 mL)

was treated with 2,3,5,6-tetrafluorophenol (15 mg, 100 µmol) in

CH₂Cl₂(1.5 mL) in the presence of HATU (17 mg, 45 µmol), and

Et₃N (10vol.-%. in MeCN, 40 µL, 28 µmol) under an argon atmo-

sphere at 4–5 °C. The dye dissolved completely in a few minutes

upon stirring, and the solution was kept for 1 h at this temperature

and loaded straight into a column with SiO₂(1.8 g). Elution with

cold (10 °C) MeCN/H₂O (10:1) followed by filtration (Rotilabo^®

syringe filters, 0.22 µm) and freeze-drying, afforded 5 mg of a solid

Supporting Information (see also the footnote on the first page of

this article): Syntheses and properties of some early precursors and

side-products, additional STED images with dyes 22 and KK114,

additional bleaching curves for dyes ATTO 647N, KK114, 1c, 21b,

and 22; full absorption and emission (UV/Vis) spectra of fluores-

cent carbopyronine dye 1c.

blue material containing 50–70% of active ester 22. HPLC: t_R

=

17 min (A/B: 70:30–0:100 in 25 min). The material (which was very

unstable in aqueous solutions) was stored under argon at –20 °C

for use in labeling studies.

Acknowledgments

Dye 22 (Linker Extension via Peptide-Type Coupling): Dye 21b

(16 mg, 20 µmol) in MeCN (3 mL) was stirred overnight at r.t. with

β-alanine methyl ester hydrochloride (6.4 mg, 46 µmol), HBTU

(23 mg, 60 µmol) and Et₃N (15 µL, 103 µmol). The reaction solu-

tion was diluted with MeCN (3 mL) and loaded directly into a

column of SiO₂(10 g). Upon elution with MeCN/H₂O (10:1), frac-

tions containing compounds far less polar than the starting mate-

rial were collected, filtered and the solvents evaporated. The dark-

blue amorphous solid (Methyl ester of 22) was saponified with aq.

NaOH (1 , 60 µL, 60 µmol) in a mixture of H₂O (3 mL) and THF

This work was supported by the Leibniz Prize fund of the Deutsche

Forschungsgemeinschaft (DFG) (to S. W. H.) and the Max-Planck-

Gesellschaft. The authors are grateful to Donald Ouw, who re-

corded optical spectra and determined the fluorescence quantum

yields, Reinhard Machinek, Holm Frauendorf and their co-workers

for recording numerous NMR and mass spectra at the Institut für

Organische und Biomolekulare Chemie (Georg-August-Universität

Göttingen), as well as to Jay Jethwa for critical reading of the ma-

nuscript. The authors thank E. Rothermel for excellent technical

assistance.

3608

www.eurjoc.org

Eur. J. Org. Chem. 2010, 3593–3610

Red-Emitting Carbopyronine Dyes

Shaker, Aachen, 1993; b) for the synthesis and properties of

red-emitting rhodamines with extended π-conjugation systems,

see: J. Liu, Z. Diwu, W.-Y. Leung, Y. Lu, B. Patch, R. P. Haug-

land, Tetrahedron Lett. 2003, 44, 4355–4359.

a) C. Aaron, C. C. Baker, J. Chem. Soc. B 1971, 319–324; b) J.

Frantzeskos, Dissertation (Neue langwellige Fluoresz-

enzfarbstoffe zur Markirung von Biomolekülen), Shaker Verlag

GmbH, Aachen, Germany, 2001; c) J. Arden-Jacob, J.

Frantzeskos, N. U. Kemnitzer, A. Zilles, K.-H. Drexhage, Spec-

trochim. Acta Sect. A 2001, 57, 2271–2283; d) R. O’Neill, P. V.

Fischer (Guava Tech. Inc.), WO 2004/003510 (08. Jan. 2004);

e) A. Zilles, J. Arden-Jacob, K.-H. Drexhage, N. U. Kemnitzer,

M. Hammers-Schneider (Atto-tec GmbH), WO 2005/003086

(13. Jan. 2005).

[1]

[2]

S. W. Hell, Far-Field Optical Nanoscopy, in: Single Molecule

Spectroscopy in Chemistry, Physics and Biology (Eds.: A. Graes-

lund, R. Rigler, J. Widengren), Springer, Berlin, 2009, pp. 365–

398.

[6]

For reviews, see: a) S. W. Hell, Science 2007, 316, 1153–1158;

b) M. Fernández-Suárez, A. Y. Ting, Nat. Rev. Mol. Cell Biol.

2008, 9, 929–943; c) D. Evanko, Nat. Methods 2009, 6, 19–20;

d) S. W. Hell, Nat. Methods 2009, 6, 24–32; e) X. Zhuang, Nat.

Photonics 2009, 3, 365–367; f) J. Lippincott-Schwartz, S. Man-

ley, Nat. Methods 2009, 6, 21–23; g) B. Huang, M. Bates, X.

Zhuang, Annu. Rev. Biochem. 2009, 78, 993–1016; for recent

reports, see: h) S. W. Hell, J. Wichmann, Opt. Lett. 1994, 19,

780–782; i) E. Betzig, G. H. Patterson, R. Sougrat, O. W. Lind-

wasser, S. Olenych, J. S. Bonifacino, M. W. Davidson, J. Lippin-

cott-Schwartz, H. F. Hess, Science 2006, 313, 1642–1645; j)

M. J. Rust, M. Bates, X. Zhuang, Nat. Methods 2006, 3, 793–

795; k) M. Bates, B. Huang, G. T. Dempsey, X. Zhuang, Sci-

ence 2007, 317, 1749–1753; l) S. T. Hess, T. P. K. Girirajan,

M. D. Mason, Biophys. J. 2006, 91, 4248–4272; m) J. Fölling,

V. Belov, R. Kunetsky, R. Medda, A. Schönle, A. Egner, C.

Eggeling, M. Bossi, S. W. Hell, Angew. Chem. 2007, 119, 6382–

6386; Angew. Chem. Int. Ed. 2007, 46, 6266–6270; n) J. Fölling,

V. Belov, D. Riedel, A. Schönle, A. Egner, C. Eggeling, M.

Bossi, S. W. Hell, ChemPhysChem 2008, 9, 321–326; o) M.

Bossi, J. Fölling, V. N. Belov, V. P. Boyarskiy, R. Medda, A.

Egner, C. Eggeling, A. Schönle, S. W. Hell, Nano Lett. 2008, 8,

2463–2468; p) V. P. Boyarskiy, V. N. Belov, R. Medda, B. Hein,

M. Bossi, S. W. Hell, Chem. Eur. J. 2008, 14, 1784–1792; q)

S. M. Polyakova, V. N. Belov, C. Eggeling, S. F. Yan, C. Ringe-

mann, G. Schwarzmann, A. de Meijere, S. W. Hell, Eur. J. Org.

Chem. 2009, 5162–5177; r) V. N. Belov, M. L. Bossi, J. Fölling,

V. P. Boyarskiy, S. W. Hell, Chem. Eur. J. 2009, 15, 10762–

10776; s) H. Schroff, C. G. Galbraith, J. A. Galbraith, H.

White, J. Gilette, S. Olenych, M. W. Davidson, E. Betzig, Proc.

Natl. Acad. Sci. USA 2007, 104, 20308–20313; t) M. Bates, B.

Huang, G. T. Dempsey, X. Zhuang, Science 2007, 317, 1749–

1753; u) B. Huang, W. Wang, M. Bates, X. Zhuang, Science

2008, 319, 810–813; v) B. Huang, S. A. Jones, B. Brandenburg,

X. Zhuang, Nat. Methods 2008, 5, 1047–1052.

[7]

[8]

J. Arden-Jacob, J. Frantzeskos, A. Zilles (K.-H. Drexhage), WO

00/64986 (02. Nov. 2000).

For a general description and spectral range of rhodamines as

fluorescent dyes, see: a) M. Beija, C. A. M. Afonso, J. M. G.

Martinho, Chem. Soc. Rev. 2009, 38, 2410–2433; b) M. S. T.

Gonçalves, Chem. Rev. 2009, 109, 190–212; c) R. P. Haugland,

A Guide to Fluorescent Probes and Labelling Technologies, Invi-

trogen, Carlsbad, 2005, pp. 11–37; d) for the use of rhodamines

in bioconjugation techniques, see: G. T. Hermanson, Biocon-

jugate Techniques, Academic Press, Amsterdam, 2008, pp. 415–

430; e) for laser dyes’ properties of rhodamines, see: F. P.

Schäfer, in: Dye Lasers; Topics in Applied Physics, v.1, (Ed.:

F. P. Schäfer), Springer, Berlin-Heidelberg, 1990, pp. 3–4; f) K.-

H. Drexhage, Dye Lasers; Topics in Applied Physics, v.1 (Ed.:

F. P. Schäfer), Springer, Berlin, Heidelberg, 1990, pp. 175–180;

g) K.-H. Drexhage, Top. Appl. Phys. 1973, 1, 155–200.

a) S. W. Hell, V. N. Belov, V. P. Boyarskiy, C. A. Wurm, S.

Jakobs, C. Geisler (Max Planck Innovation GmbH), PCT/EP/

2009/006578 (10.09.2009); b) V. N. Belov, V. P. Boyarskiy, C. A.

Wurm, S. Jakobs, S. W. Hell, Angew. Chem. 2010, 122, 3598–

3602; Angew. Chem. Int. Ed. 2010, 49, 3520–3523.

[9]

[10]

[11]

P. G. M. Wuts, Th. W. Green, Green’s Protective Groups in Or-

ganic Synthesis, John Wiley & Sons, New York, 2007, pp. 812–

818.

a) S. Ram, L. D. Spicer, Tetrahedron Lett. 1987, 28, 515–516;

b) T. Kiguchi, N. Kuninobu, Y. Takahashi, Y. Yoshida, T.

Naito, I. Ninomiya, Synthesis 1989, 778–781.

[3]

[4]

J. Fölling, M. Bossi, H. Bock, R. Medda, C. A. Wurm, B. Hein,

S. Jakobs, C. Eggeling, S. W. Hell, Nat. Methods 2008, 5, 943–

945.

[12]

[13]

J. F. Eggler, G. F. Holland, M. R. Johnson, R. A. Volkmann

(Pfizer Inc.) US Pat. 4738972 (19.04.1988).

a) K. Peneva, G. Mihov, F. Nolde, S. Rocha, J. Hotta, K.

Braeckmans, J. Hofkens, H. Uji-i, A. Hermann, K. Müllen,

Angew. Chem. 2008, 120, 3420–3423; Angew. Chem. Int. Ed.

2008, 47, 3372–3375; b) C. Jung, B. K. Müller, D. C. Lamb, F.

Nolde, K. Müllen, C. Bräuchle, J. Am. Chem. Soc. 2006, 128,

5283–5291; c) C. Jiao, K.-W. Huang, J. Luo, K. Zhang, C. Chi,

J. Wu, Org. Lett. 2009, 11, 4508–4511; d) J. Li, K. Zhang, X.

Zhang, K.-W. Huang, C. Chi, J. Wu, J. Org. Chem. 2010, 75,

856–863; e) S. L. Niu, G. Ulrich, R. Ziessel, A. Kiss, P.-Y. Re-

nard, A. Romieu, Org. Lett. 2009, 11, 2049–2052; for reviews

on BODIPY fluorescent dyes, see: f) G. Ulrich, R. Ziessel, A.

Harriman, Angew. Chem. Int. Ed. 2008, 47, 1184–1201; g) A.

Loudet, K. Burgess, Chem. Rev. 2007, 107, 4891–4932; h) J. J.

Gassensmith, E. Arunkumar, L. Barr, J. M. Baumes, K. M. Di-

Vittorio, J. R. Johnson, B. C. Noll, B. D. Smith, J. Am. Chem.

Soc. 2007, 129, 15054–15059; i) Z. Lu, N. Liu, S. J. Lord, S. D.

Bunge, W. E. Moerner, R. J. Twieg, Chem. Mater. 2009, 21,

797–810; j) K. Kolmakov, V. N. Belov, J. Bierwagen, C. Ringe-

mann, V. Müller, C. Eggeling, S. W. Hell, Chem. Eur. J. 2010,

16, 158–166.

a) Y. Miyake, Y. Kikugawa, J. Heterocycl. Chem. 1983, 20, 349–

352; b) G. Field, P. R. Hammond (Dept. of Energy, USA), US

Pat. 5283336 (01. Feb. 1994); c) C. Berrier, J. C. Jacquesy, M. P.

Jouannetaud, A. Renoux, New J. Chem. 1987, 11, 605–609.

B. C. Ranu, S. Bhar, Org. Prep. Proc. Int. 1996, 28, 371–409.

M. E. Kuehne, J. B. Pitner, J. Org. Chem. 1989, 54, 4553–4569.

a) J.-L. Luche, J.-C. Damiano, J. Am. Chem. Soc. 1980, 102,

7926–7927; b) S. Sabetay, Bull. Soc. Chim. Fr. 1930, 47, 614–

620.

Y. Yamada, A. Akiba, S. Arima, C. Okada, K. Yoshida, F.

Itou, T. Kai, T. Satou, K. Takeda, Y. Harigaya, Chem. Pharm.

Bull. 2005, 53, 1277–1290.

a) M. A. Davis, S. O. Winthrop, R. A. Thomas, F. Herr, M.-P.

Charest, R. Gaudry, J. Med. Chem. 1964, 7, 88–94; b) L. F.

Fieser, E. B. Hershberg, J. Am. Chem. Soc. 1937, 59, 2331–

2335.

[14]

[15]

[16]

[17]

[18]

[19]

a) A. I. Meyers, D. L. Temple, J. Am. Chem. Soc. 1970, 92,

6646–6647; b) T. D. Nelson, A. I. Meyers, J. Org. Chem. 1994,

59, 2577–2580; c) J. Sedelmeier, T. Hammerer, C. Bolm, Org.

Lett. 2008, 10, 917–920.

[5]

a) The presence of the very strong electron-acceptor groups –

e.g., CF₃and CN – in position 9 of the xanthene core shifts

the absorption and emission maxima of the rhodamine dyes

far more to the red, but all these dyes are hydrolytically un-

stable, due to the addition of water and other nucleophiles to

C-9: J. Arden-Jacob, Neue langwellige Xanthen-Farbstoffe für

Fluoreszenzsonden und Farbstofflaser (Dissertation), Verlag

[10]

[20]

[21]

See ref., pp. 582–588.

a) K. H. Knauer, R. Gleiter, Angew. Chem. 1977, 89, 116–117;

Angew. Chem. Int. Ed. Engl. 1977, 16, 113–113; b) H. Willwohl,

J. Wolfrum, R. Gleiter, Laser Chem. 1989, 10, 63–72.

A. I. Meyers, D. L. Temple, R. L. Nolen, E. D. Mihelich, J.

Org. Chem. 1974, 39, 2778–2783.

[22]

Eur. J. Org. Chem. 2010, 3593–3610

www.eurjoc.org

3609

V. N. Belov, S. W. Hell et al.

FULL PAPER

[23] a) A. Romieu, D. Tavernier-Lohr, S. Pellet-Rostaing, M. Le-

Schaffer, S. Felekyan, C. A. M. Seidel, J. Phys. Chem. A 2006,

110, 2979–2995.

maire, P.-Y. Renard, Tetrahedron Lett., DOI: 10.1016/j.tet-

let.2010.04.080; b) A. Rosowsky, R. A. Forsch, J. H. Freisheim, [27] J. Hotta, E. Fron, P. Dedecker, D. P. F. Janssen, C. Li, K.

R. G. Moran, M. Wick, J. Med. Chem. 1984, 27, 600–604.

[24] See Technical Tip (31. May 2007): http://ww w .probes.invitrogen.

com/media/pis/x30052.pdf.

[25] R. Schmidt, C. A. Wurm, A. Punge, A. Egner, S. Jakobs, S. W.

Hell, Nano Lett. 2009, 9, 2508–2510.

[26] a) J. Widengren, U. Mets, R. Rigler, J. Phys. Chem. 1995, 99,

13369–13379; b) C. Eggeling, J. Widengren, L. Brand, J.

Müllen, B. Harke, J. Bückers, S. W. Hell, J. Hofkens, J. Am.

Chem. Soc. 2010, 132, 5021–5023.

[28] a) B. Harke, C. K. Ullal, J. Keller, S. W. Hell, Nano Lett. 2008,

8, 1309–1313; b) B. Harke, J. Keller, C. K. Ullal, V. Westphal,

A. Schönle, S. W. Hell, Opt. Expr. 2008, 16, 4154–4162.

Received: March 12, 2010

Published Online: May 28, 2010

3610

www.eurjoc.org

Eur. J. Org. Chem. 2010, 3593–3610

Article Doi

A versatile route to red-emitting carbopyronine dyes for optical microscopy and nanoscopy

DOI: 10.1002/ejoc.201000343

Source and publish data:

Authors:

Article abstract of DOI:10.1002/ejoc.201000343

Full text of DOI:10.1002/ejoc.201000343

Products guided by the article

R&D Labs maybe for 1241912-44-6

Relevant to this article

Hot Product