what is best strategy for water soluble fluorescence dyes

doi.org/10.26434/chemrxiv.12562310.v1

What Is Best Strategy for Water Soluble Fluorescence Dyes? – a CaseStudy Using Long Fluorescence Lifetime DAOTA DyesNiels Bisballe, Bo W. Laursen

Submitted date: 25/06/2020 • Posted date: 29/06/2020Licence: CC BY-NC-ND 4.0Citation information: Bisballe, Niels; Laursen, Bo W. (2020): What Is Best Strategy for Water SolubleFluorescence Dyes? – a Case Study Using Long Fluorescence Lifetime DAOTA Dyes. ChemRxiv. Preprint.https://doi.org/10.26434/chemrxiv.12562310.v1

The applications of organic fluorophores in biological sciences rely heavily on their properties in aqueoussolution. The lipophilic nature of virtually all such chromophores provides several challenges to adapt them tobiologically relevant conditions. In this work we investigate three different strategies for achievingwater-solubility of the diazaoxatriangulenium (DAOTA+) chromophore: hydrophilic counter ions, aromaticsulfonation of the chromophore core, and attachment of cationic or zwitterionic side chains. The longfluorescence lifetime (FLT, τf » 20 ns) of DAOTA+ makes it a sensitive probe for changes in the rate ofnon-radiative deactivation and for aggregation leading to multi exponential decay profiles. Direct sulfonation ofthe chromophore, as applied in several Alexa dyes, does indeed increase solubility drastically, but at the costof greatly reduced quantum yields (QY) due to enhanced non-radiative deactivation processes. Theintroduction of either cationic (4) or zwitterionic side chains (5), however, brings the FLT (τf = 18 ns) and QY(φf = 0.56) of the dye to the same level as the parent chromophore in acetonitrile. For these derivativestime-resolved fluorescence spectroscopy also reveals a high resistance to aggregation and non-specificbinding in a high loading of bovine serum albumin (BSA). The results clearly show that addition of chargedflexible side chains is preferable to direct sulfonation of the chromophore core.

File list (2)

download fileview on ChemRxivPreprint.pdf (603.42 KiB)

download fileview on ChemRxivsupplementary_information.pdf (4.66 MiB)

http://doi.org/10.26434/chemrxiv.12562310.v1

https://chemrxiv.org/authors/Niels_Bisballe/8996522

https://chemrxiv.org/authors/Bo_W_Laursen/7406852

https://chemrxiv.org/ndownloader/files/23420105

https://chemrxiv.org/articles/What_Is_Best_Strategy_for_Water_Soluble_Fluorescence_Dyes_a_Case_Study_Using_Long_Fluorescence_Lifetime_DAOTA_Dyes/12562310/1?file=23420105



What is best strategy for water soluble fluorescence dyes? –

A case study using long fluorescence lifetime DAOTA dyes

Niels Bisballea and Bo W. Laursen*a

Address:

a Nano-Science Center & Department of Chemistry, University of Copenhagen,

Universitetsparken 5, DK-2100, Copenhagen Ø, Denmark.

* Corresponding author e-mail:

[email protected]

2

Abstract

The applications of organic fluorophores in biological sciences rely heavily on their properties in

aqueous solution. The lipophilic nature of virtually all such chromophores provides several

challenges to adapt them to biologically relevant conditions. In this work we investigate three

different strategies for achieving water-solubility of the diazaoxatriangulenium (DAOTA+)

chromophore: hydrophilic counter ions, aromatic sulfonation of the chromophore core, and

attachment of cationic or zwitterionic side chains. The long fluorescence lifetime (FLT, τf 20 ns)

of DAOTA+ makes it a sensitive probe for changes in the rate of non-radiative deactivation and

for aggregation leading to multi exponential decay profiles. Direct sulfonation of the chromophore,

as applied in several Alexa dyes, does indeed increase solubility drastically, but at the cost of

greatly reduced quantum yields (QY) due to enhanced non-radiative deactivation processes. The

introduction of either cationic (4) or zwitterionic side chains (5), however, brings the FLT (τf = 18

ns) and QY (φf = 0.56) of the dye to the same level as the parent chromophore in acetonitrile. For

these derivatives time-resolved fluorescence spectroscopy also reveals a high resistance to

aggregation and non-specific binding in a high loading of bovine serum albumin (BSA). The

results clearly show that addition of charged flexible side chains is preferable to direct sulfonation

of the chromophore core.

3

Introduction

Molecular behavior of a solute in a given solvent is a critical property in several chemical sciences,

ranging from the distribution of drugs in the human body to the self-assembly of nanostructures.1,

2 This is indeed also the case in the field of bio-imaging and for assays relying on fluorescent dyes,

where enhanced water solubility of dyes leads to improved signal to background ratios and

elimination of artifacts.3-5

Organic chromophores in general consist of extended π-conjugated systems responsible for

absorption and emission of ultra-violet (UV) to near-infrared (NIR) electromagnetic radiation.

This inherently introduces a hydrophobic structural element to the fluorescent molecule, favoring

dissolution in organic solvents, while inducing aggregation due to poor solvation in aqueous

environments.6, 7 In bio-imaging the latter is seldom desirable, but notable exceptions include the

probing of membranes and staining lipophilic compartments.8, 9 Dye aggregation may lead to

unwanted changes in absorption and emission spectra, reduced fluorescence intensity and

lifetime.10-13 But even when aggregates are avoided e.g. by high dilution, moderate water-solubility

may still lead to problems arising from hydrophobic interactions with other amphiphilic solutes

such as biomacromolecules. This may lead to nonspecific binding and can in turn also alter optical

properties.14-17 Thus, for general application of fluorescence dyes, high water-solubility is a key

feature since it is expected to reduce both dye-dye interactions (aggregation and self-quenching)

as well as nonspecific interactions with biomolecules.

A testament to the importance of good solvation of fluorescent probes in aqueous media is the

pioneering work of Haugland leading to the highly successful commercial Alexa Fluor® dyes

(Thermo Fisher).18, 19 The Alexa Fluor® dyes are traditional organic chromophores, mainly

xanthenes and cyanines, modified with anionic sulfonate groups to drastically increase their water

solubility and thus performance in bioimaging. The same strategy has since been employed for

several other chromophore types and by other commercial brands.

While many different chromophore modifications have been applied for increasing water-

solubility, including anionic sulfonates and phosphonates, zwitterionic sulfobetaines and neutral

PEG chains, there is presently no general consensus about the preferred or optimal modifications

of fluorescence dyes to increase water-solubility and minimize artefacts in bio-imaging.

Rhodamine and borondipyrromethene (BODIPY) chromophores are popular in bioimaging and

4

enhancing their water-solubility is a desirable improvement. A key feature of these dyes is their

high brightness (ε ∙ φf). They also feature moderate fluorescence lifetimes (FLTs) in the range of

3-5 ns and have seen application in time resolved fluorescence spectroscopy.20-22 Inspired by the

Alexa Fluor® dyes, Kolmakov et al. (2010) have synthesized a red emitting, water-soluble

rhodamine dye by allylic sulfonation.23 They have successfully demonstrated the use of the dye in

advanced super-resolution fluorescence imaging techniques. Since, Kolmakov et al. (2012) have

also assessed the solubilizing effects of phosphonate and hydroxyl groups in the same position.24

Investigations into solubilizing the BODIPY chromophore are more varied. Early examples carry

sulfonate groups directly on the chromophore.25, 26 Later investigations have focused on

introducing water solubility through functionalized side chains to keep the electron distribution of

the chromophore intact. The solubilizing groups are typically comprised of sulfonates,

phosphonates and sulfobetaines.27-29

Another successful approach to water solubility has been post-synthetic functionalization of dyes

with a sulfonated peptide linker. The linker can be introduced through well-known couplings like

amidation and the click-reaction, which many commercially available dyes are already set up for.30-

32 This strategy, however, seems to work best for dyes that already contain hydrophilic elements.

The peptide linker may feature further functional groups suitable to generate bioconjugates.33

We have in recent years been working with synthesis and applications of various aza-/oxa-

triangulenium dyes (Chart 1), in particular the ADOTA+ and DAOTA+ derivatives. These

fluorophores display especially long intrinsic FLT in combination with good quantum yield (QY),

making them uniquely suited for time-gated imaging34, fluorescence lifetime imaging microscopy

(FLIM)35, 36 and fluorescence polarization assays37, 38. The long FLT arises from the combination

of moderate radiative rate (kf) and very low rates of non-radiative deactivation (knr).39 The major

challenge in designing and modifying long FLT dyes is to keep the rate of non-radiative

deactivation very low. which is achieved through exceptional structural rigidity.

In this work we investigate design strategies to obtain highly water soluble triangulenium dyes.

Firstly, such modifications have not been made before and would greatly increase the scope of

these unique long FLT dyes. Secondly, the long FLT of these dyes provide a highly sensitive test

case to compare various strategies for enhancing water solubility. This sensitivity arises from the

intrinsically lower rate of fluorescence, which ensures that any additional non-radiative

5

contributions to deactivation of the excited state will have a much greater impact than in standard

fluorophore systems with a shorter lifetime. We will also use fluorescence lifetime measurements

as a convenient tool to probe the degree of aggregation and/or association with bio molecules as

any of such events will lead to inhomogeneity in the population of dyes and thus deviations from

simple single exponential fluorescence decay rates.

The DAOTA+ chromophore (Chart 1) was chosen as the starting point for this study. It features a

long FLT (τf > 20 ns in MeCN and DCM) and a high quantum yield (φf = 0.58 in MeCN, 0.80 in

DCM).39

Chart 1. Structures of diazaoxatriangulenium (DAOTA+),

azadioxatriangulenium (ADOTA+), dimethoxyquinacridinium

(DMQA+), carbon-bridged diazatriangulenium (CDATA+) and benzo-

fused diazatriangulenium (BDATA+).

The DAOTA+ chromophore is promising due to high photo stability40, 41 and resistance to

quenching from amino acids.42 However, it suffers from low water solubility, which reduces its

brightness and FLT in aqueous solutions. The chromophore is closely structurally related to other

reported cationic triangulenium fluorophores such as: ADOTA+ 43, DMQA+ 44, CDATA+ 45 and

6

BDATA+ 46 dyes (Chart 1), but also to other much studied cationic dyes including helicenes47, 48,

acridines49, 50 and rhodamines51, 52. Thus, we expect the here derived guidelines to be generally

applicable to a large range of cationic fluorescent dyes.

Results and discussion

Strategies for chromophore modifications and synthesis

With a series of six new DAOTA+ dyes (Chart 2) we will evaluate various strategies for increased

water-solubility: 1) Introduction of hydrophilic counter ions. 2) Direct sulfonation of the

chromophore system. 3) Addition of charged side chains. Traditionally, triangulenium dyes have

been synthesized with the highly lipophilic tetrafluoroborate (BF4-) and hexafluorophosphate (PF6

-

) anions.

Chart 2. Structures of the six water-soluble DAOTA+ dyes 1-6, along with their QYs (φf) and FLTs (τf) in pure water. a Only the FLT corresponding to the freely solvated dye of a bi-exponential decay is shown (see Table 2 for details).

7

This strategy was developed with facile synthetic workup in mind, exploiting the lack of

dissociation of these salts in aqueous media.53 To obtain water-soluble derivatives, the anion of

Pr2DAOTA+ was exchanged for the hydrophilic and biologically prevalent chloride ion. This was

conveniently achieved using an anion exchange resin. This simple modification resulted in the

highly water-soluble (>10 mM) dye, 1 (Chart 2), from the practically insoluble BF4- salt in

quantitative yield (Scheme S1).

As mentioned in the introduction direct sulfonation of the chromophore core has been successfully

applied in several cases, including Alexa 488 and Alexa 532 to name a few,16 as well as the

aforementioned BODIPY dyes. Functionalization directly on the DAOTA+ chromophore can

roughly be broken down to occur in three different regions: the positions neighboring the O-bridge

(positions 3 and 5, Chart 1), positions neighboring the N-bridges (positions 1, 7, 9 and 11), and

positions para to the carbenium center (positions 2, 6 and 10). Delgado et al. (2018) have shown

large effects of simple functional groups attached directly to the 9-position of DAOTA+ via

electrophilic aromatic substitution, with a clear connection between electron density

(donating/withdrawing groups) and spectral properties.54 This is in line with observations from the

DAOTA+ precursor, DMQA+.55 We suspect that substitution neighboring the O-bridge is

preferable to positions neighboring the N-bridge, since steric interference with the side-chain in

the latter case is expected to reduce planarity. Despite EWGs on the 6-position enhancing

fluorescence of the helical DMQA+ 55, in planar systems functionalization next to N-bridges has a

negative impact on fluorescence properties (QY and FLT) as seen for DAOTA+, and in our

previous observations with chlorination of triazatriangulenium salts. 54, 56

Selective core sulfonation on the 3- and 5-positions of DAOTA+ was achieved by overnight

reaction in concentrated sulfuric acid (Scheme S1). A trace amount of the mono-sulfonated product

(2) could also be obtained in this way, with the primary product containing sulfonates in both

positions (3, 88 % yield). The mono-sulfonated product is likely to be a result of desulfonation

during aqueous workup. Both products are soluble in water. The selectivity for the 3- and 5-

positions is result of the strongly acidic conditions which favors electrophilic substitution in

positions most remote to the nitrogen bridges, as reported by Duwald et al. (2017) for the DMQA+

system (Chart 1).57

8

Modifying the dye with solvating groups on the side chains is expected to have little influence on

the photophysics. We have several examples of DAOTA+ modified in these positions, where the

emissive properties intrinsic to the chromophore remain largely intact.35, 39, 42 The side-chain

modified dyes were synthesized via the traditional pathway for triangulenium dyes where side

chains are introduced by substitution with primary amines during formation of the nitrogen bridges

in the ring system.53 By this classical pathway a DAOTA+ dye carrying two 3-

dimethylaminopropyl side chains (9, Scheme S2) was obtained as a key intermediate. Methylation

of 9 with iodomethane gave tri-cationic 4 after anion exchange. Zwitterionic side chains were

conveniently obtained by alkylation of 9 with 1,3-propanesultone, as previously reported for the

CalFluor chromophores58, to give 5 after anion exchange.

Finally, we were interested in combining the above strategies. By sulfonating the PF6- salt of 4

under conditions similar to the synthesis of 3, 6 was obtained after anion exchange (Scheme S2).

Full experimental details and characterizations of all new compounds are given in Supporting

Information.

Photophysical properties of water-soluble triangulenium dyes

With this diverse set of water-soluble triangulenium dyes (1-6) in hand, we first investigated the

effects of the modifications on the spectral features. As expected for the simple anion exchange

absorption and emission are nearly identical for 1 and Pr2DAOTA+ (Table 1). Absorption spectra

of 2 and 3, containing one and two sulfonic acid groups respectively, were surprisingly similar.

These modifications introduce a redshift in both absorption (S0 → S1) and emission maxima

compared to those of 1 (Table 1, Figure 1). The (S0 → S2) absorption maximum on the other hand

is blueshifted. The ionic side chains of 4 and 5 resulted in a blueshift shift in both (S0 → S1)

absorption and emission maxima. The (S0 → S1) absorption band of both 4 and 5 are narrower than

observed for 1 and Pr2DAOTA+ (Figure 1). For comparison, the absorption spectrum of 5 is similar

in width and shape to the parent chromophore, Pr2DAOTA+ in MeCN, while 1 has less pronounced

features (Figure S13).

9

Table 1 Steady state absorption and emission values for 1-6 (Chart 1) in water and Pr2DAOTA+ in acetonitrile for

comparison. Full absorption and emission spectra as well as excitation specta for 1-6 can be found in the ESI

(Figures S1-S6).

Solvent λmax (abs)

[nm]

S0 → S1

λmax (abs)

[nm]

S0 → S2

λmax (em)

[nm]

S1 → S0

Stokes shift

[cm-1]

φfa

Pr2DAOTA+ b, c MeCN 557 449 590 1004 0.58

1 H2O 558 449 587 885 0.31

2 H2O 564 440 594 895 0.27

3 H2O 565 435 594 864 0.28

4 H2O 549 448 576 854 0.56

5 H2O 548 449 576 887 0.56

6 H2O 558 439 584 798 0.36

a Quantum yields are determined relative to Rhodamine 6G in ethanol (φf = 0.95). Details on their calculation can be found in the ESI. b Data for

Pr2DAOTA+ taken from Bogh et al. (2017)39. c Full absorption spectrum can be found in Figure S13.

Figure 1. (A) Normalized overlay of absorption spectra of the DAOTA+ chromophore without modifications (1),

with a sulfonated core (3, similar to 2), with ionic side chains (5, similar to 4) and both of the latter modifications

combined (6). (B) Normalized overlay of emission spectra, analogous to A.

A narrower absorption band is an indication of a more homogeneous solvation, validating the

strategy of placing the functional groups on the side chains. Combining the approaches of

sulfonating directly on the chromophore and having quaternary ammonium groups on the side

chains (6) results in spectra clearly relatable to each individual modification: the (S0 → S1)

absorption redshift of the sulfonic acid groups, as observed for 2 and 3, and blue shift of the

400 450 500 550 600

0,0

0,2

0,4

0,6

0,8

1,0

No

rma

lize

d a

bso

rban

ce

Wavelength [nm]

1

3

5

6

A

550 600 650 700 750

0,0

0,2

0,4

0,6

0,8

1,0

Norm

aliz

ed e

mis

sio

n

Wavelength [nm]

1

3

5

6

B

10

charged side chains, as observed for 4 and 5, counteract. The resulting absorption maximum is

similar to that of 1, which contains neither modification. The blue shift of the (S0 → S2) absorption

observed for 2 and 3, and attributed to the sulfonic acid groups, persists in 6 as this transition is

unaffected by the ammonium side chains.

The spectral shifts can be explained by the electronic effects of the modifications on the electronic

transitions (Figure 2). The (S0 → S1) transition is dominated by the N-bridges donating electron

density to the formal cation center. The positively charged ammonium groups of the side chains

in 4 and 5 poses an electrostatic influence on the N-bridges impeding this process, leading to a

blueshift of the transition. Similarly, the sulfonates neighboring the O-bridge reduce its donating

capabilities in the (S0 → S2) transition. The (S0

→ S1) redshift from the sulfonates is surprising

considering its position,59 and we speculate that it might be a secondary effect of the S2 blueshift

reducing the mixing of S1 and S2 and thus increasing their separation.

Figure 2. Substituent effects on the electronic

transition dipoles for the first (red) and second (blue)

excited state illustrated for 6.43 The electron donating

capabilities of the heteroatom bridges is reduced in

both cases, leading to more energetic transitions.

Despite the similar spectral properties of 1 and Pr2DAOTA+, the QY (Table 1) is almost halved

for the former in aqueous solution. This can in part be attributed to poorer solvation of the cation

in water compared to acetonitrile. A broadening of both absorption and emission bands were

observed with increased dye concentration when measuring the QY, suggesting aggregation

(Figures S23-S24).

11

However, an intrinsic quenching of fluorescence observed for dyes dissolved in water must also

be considered to be a significant contribution. The extent of this effect can be determined by

comparing the fluorescence in water and deuterium oxide and is discussed later in this work.

Direct sulfonation of the π-system lead to a decreased QY. When comparing the absorption and

emission spectra from the QY titration of 2, a slight broadening of the bands is seen with increased

concentration which could indicate aggregation (Figures S26-S27). No broadening is observed for

the double sulfonated 3 (Figures S29-S30), but the QY was still as low as 1.

We were pleased to find that the QYs of 4 and 5 in water (Table 1) are comparable to that of

Pr2DAOTA+ in MeCN. Introducing solubilizing groups on the side chains of DAOTA+ promises

to be a viable strategy in translating the desired photophysical properties to aqueous solution. The

QY of 6 lies between the values resulting from the individual modifications (3 and 4 respectively).

This confirms that introducing the sulfonic acid groups has a negative impact on the QY.

Time resolved fluorescence spectroscopy in water

In order to explain the differences in QY of the synthesized dyes and to develop insight into the

dyes’ molecular behavior in aqueous solution, we turned to time-resolved fluorescence

spectroscopy. Measuring the FLT and looking at the decay profiles allows us to distinguish well-

solvated dyes displaying a mono-exponential fluorescence decay due to the homogeneity of the

sample from inhomogeneous samples containing subpopulations of aggregated dyes and thus

displaying multi exponential decay profiles. We found that in pure water only the dyes 3-5 display

mono-exponential fluorescence decays (Table 2) and can be considered fully and homogeneously

solvated.

The bi-exponential decay of 1 in water indicates poor solvation and the formation of aggregates,

which show a longer lifetime component. The quantum yield of 1 is almost half (Table 1)

compared to the best dyes (4 and 5) while the dominant FLT component only is reduced by 20%.

This means that the observed aggregates must have very low quantum yields and/or non-emissive

aggregates are also present.

12

Table 2. Comparison of fluorescence lifetimes of 1-6 (Chart 1) in water, PBS and PBS with the addition of BSA (600

µM). Full decay profiles can be found in the ESI (Figures S14-S19).

H2O PBS PBS + BSA

τf1 [ns] τf2 [ns]a τavg

[ns]b

τf1 [ns] τf2 [ns]a τavg

[ns]b

τf1 [ns] τf2 [ns]a τf3 [ns]a, c τavg

[ns]b

1 15.2 22.6 (19 %) 16.6 14.9 21.6 (26 %) 16.6 10.5 18.3 (60 %) 1.3 (1 %) 15.0

2 13.5 24.1 (12 %) 14.8 11.7 16.5 (54 %) 14.3 12.3 23.2 (51 %) 1.6 (2 %) 17.7

3 12.5 - - 11.5 20.2 (7 %) 12.1 11.9 25.0 (59 %) 1.8 (1 %) 19.5

4 18.3 - - 18.1 - - 18.6 - 7.7 (3 %) 18.2

5 18.6 - - 18.3 - - 18.5 - 6.3 (2 %) 18.3

6 14.6 20.5 (7 %) 15.0 14.2 - - 14.1 - 2.4 (1 %) 14.1

a Secondary fluorescence lifetime components and their weighted intensities in parenthesis. b Intensity weighted average fluorescence lifetimes,

where multiexponential fluorescence decay is observed. c Interpreted as the fluorescence lifetime component of dye bound to BSA at quenching

sites.

Two-fold sulfonation of the chromophore (3) resulted in mono-exponential fluorescence decay

reveals a homogeneous solvation of the chromophore. However, a large simultaneous drop in both

FLT (12.5 ns) and QY (0.28) (Table 1) show that the sulfonation indeed accelerates non-radiative

deactivation. The negative impact on FLY and QY from core sulfonation is confirmed by the mono

sulfonated derivative 2, which furthermore exerts a bi-exponential decay, indicating remaining

problems with aggregation of this overall neutral species.

The mono-exponential decays of 4 and 5 both show comparable FLTs above 18 ns. Along with

the measured QYs (Table 1), these results indicate that the dyes are well-solvated in water as a

result of the charged side chains. Remarkably, both the FLTs and QYs of these dyes are

comparable to Pr2DAOTA+ in MeCN solution.39

The fluorescence decay of 6 features a small secondary component in pure water. We speculate

that the charge complementarity of the two positive and two negative groups in each end of the

molecule may lead to head-to-tail aggregates. The lower FLT compared to 1 must, as discussed for

2 and 3, be attributed to the introduction of sulfonic acids groups on the chromophore core.

13

Solvation of dyes in buffers and protein solutions

To draw parallels to biological systems, time-resolved fluorescence of the dyes in phosphate

buffered saline (PBS) was also investigated (Table 2). The obtained decays were mostly similar to

the ones observed in pure water. For 3, a biexponential decay was observed. The increased salt

concentration of the solution could lower the solvation of the dye, thus inducing aggregation to

some extent.60 Inversely, a mono-exponential decay for 6 in PBS could be due to the breaking up

of the proposed head-to-tail aggregates due to the additional ionic strength stabilizing the solvation

of the highly charged dye. It should be noted, however, that for either dye these are relatively small

effects.

The fluorescence lifetime measurements were repeated with the addition of 600 µM BSA to

introduce sites for non-specific binding, relatable to biologically relevant media.61 For all dyes an

additional short-lived component was observed, leading to an overall decrease in the average FLT.

The short-lived component is more prevalent for the dyes that already display compromised

solvation in PBS. We tentatively assign the short lived component to dye bound to BSA at a

quenching site. Especially the core sulfonated anionic dye, 3, shows to be severely compromised

at high protein concentrations. This observation is noteworthy since sulfonation and overall

negative charge is the preferred modification for many commercial fluorescence dyes.

The well-solvated and bright chromophores, 4 and 5 are only slightly affected by BSA. The

population of freely dissolved chromophores in these cases does not seem to be affected by the

presence of BSA at retaining the long FLT of 18 ns, while a small fraction of the molecules bound

to BSA display a 6-8 ns lifetime. The interaction between dye molecules and BSA must therefore

be static on the nanosecond timescale. Any significant presence of one or more “dark” interactions

with the protein can be ruled out based on the relative fluorescence intensities, which show a

reduction of 6-10 % upon addition of BSA (Figures S10-S11).

Intrinsic water quenching

Aside from solubility and solvation, fluorescence of organic chromophores in water is further

complicated by a quenching effect intrinsic to the solvent. This has been illustrated several times

by enhanced fluorescence intensity and FLT in D2O.62-65 The exact mechanism by which this

14

quenching occurs and what parameters affect it is still to be fully understood and it is thus difficult

to predict the extent of the effect.

Measuring the FLT in D2O for the dyes 1-6 (Table 3) showed a significant increase compared to

H2O. A 20 % enhancement is seen for 4 and 5 already displaying long FLTs in pure water. This

effectively surpasses its parent’s performance in MeCN and brings the fluorophore on par with

Pr2DAOTA+ in DCM39, in which a similar solvent quenching effect is not expected to take place.

A 30 % enhancement is seen in the remaining cases. Where bi- exponential decays are observed,

both fluorescence decay components see an increase, suggesting that even dye aggregates are

affected by the quenching effect of water.

Table 3. Fluorescence lifetimes of 1-6 (Chart 1) in deuterium

oxide and the relative increase compared to water (Table 2).

τf1 [ns] τf2 [ns] a τavg [ns] b τ(D2O)/τ(H2O)

1 19.3 27.6 (22 %) 21.1 1.27c

2 16.8 23.6 (32 %) 19.0 1.28c

3 16.6 - - 1.33

4 21.9 - - 1.20

5 22.1 - - 1.19

6 19.2 - - 1.28c a Secondary fluorescence lifetime components and their weighted intensities

in parenthesis. b Amplitude weighted average fluorescence lifetimes, where

multiexponential fluorescence decay is observed. c Value calculated based

on the intensity weighted average fluorescence lifetime for bi-exponential

decays.

Rates of the excited state processes

With the steady state and time-resolved fluorescence results in hand we can now look at how the

structural differences of the dyes affect the associated excited state deactivation processes.

Calculating the rate constants (Table 4) from the measured FLTs and QYs reveals that 4 and 5

featuring the hydrophilic side chains have a significantly higher radiative rate kf than the remaining

derivatives. This effect we assign to the electrostatic impact of the positively charged ammonium

group on the nitrogen donor groups in the DAOTA+ chromophore, which also did result in a

blueshift and narrowing of the absorption and emission bands.

15

Most importantly, 4 and 5 display very low rates of non-radiative deactivation knr (Table 4),

showing that this design is ideal for solubilizing DAOTA+ in water. This becomes even more clear

when considering that almost 40 % of the non-radiative deactivation observed for these dyes in

water seems to be solvent specific quenching, as can be seen by comparing the rates in water and

D2O (Table 4). Interestingly, it seems that the aromatic sulfonates attached directly to the

chromophore core opens up additional paths of non-radiative deactivation increasing knr by a factor

of two. This is most clearly seen by comparison between the derivatives well solvated in D2O

where effects of aggregation and specific water quenching are not contributing. Here the core

sulfonated compounds 3 and 6 display knr twice that of 4 and 5.

Table 4. Radiative and non-radiative rate constants for the depopulation

of the excited state of 1-6 in water and deuterium oxide.

kf

[107 s-1]

knr (H2O)

[107 s-1]

knr (D2O) a

[107 s-1]

Δknr b

[107 s-1]

Pr2DAOTA+ c, d 2.6 2.1 - -

1e 2.0 4.5 3.1 1.4

2e 2.0 5.4 4.0 1.5

3 2.2 5.8 3.8 2.0

4 3.1 2.4 1.5 0.9

5 3.0 2.4 1.5 0.9

6e 2.5 4.4 2.7 1.6

a The rate constant of fluorescence is assumed to be the same in water and deuterium

oxide. b Effectively the quenching contribution of water, kQ·[Q]. c In acetonitrile. d Data

from Bogh et al. (2017).39 e Rate constants calculated from dominant component of

biexponential decay fits.

Conclusions

Six new water-soluble DAOTA+-derivatives were obtained through combinations of three

different strategies for modifying the chromophore. Exchanging the lipophilic anion for chloride

(1) lead to a substantial gain in water solubility, but solvation and aggregation of the cationic

chromophore remain unaltered. While twofold direct sulfonation of the DAOTA+ chromophore

(3) does enhance solvation in pure water, aggregation and non-specific binding are still a concern

in biologically relevant environments. Despite further effort to modify DAOTA+ in this regard (6),

16

core modifications remain problematic as the introduction of additional non-radiative deactivation

pathways compromises fluorescence lifetime and quantum yield. Excellent photophysical

properties in water resulted from introducing charged side chains in derivatives 4 and 5. This

design strategy resulted in dyes with long FLT of 18 ns and high quantum yields of 56 %, to our

knowledge, unmatched by any other water-soluble small molecule fluorophores emitting beyond

550 nm. We managed to utilize the FLT response as a tool to investigate how different solvation

strategies influence molecular events and interactions. These results help in finding the optimal

design for a bioconjugable and water-soluble FLT probe.

In first instance these results suggest that charged and zwiterionic side chains most likely also are

a preferable design strategy for organic fluorophores in general over traditional core sulfonation.

The potential usefulness of this strategy extends beyond the field of dyestuff. Tuning solvation

through different functionalizations is useful in several fields, e.g. in the science of nanostructures,

where controlled self-assembly in various solvents is of interest. Long FLT probes hold potential

as tools for investigating such solvation effects. We see the DAOTA+ chromophore as a good

candidate for further investigation of such properties, in particular due to the sensitivity and

resolution of time-resolved fluorescence spectroscopy as demonstrated. These dyes should also be

considered in the investigation of fluorescence quenching effects, since time-resolved

measurements are much more sensitive and convenient than corresponding intensity-based

experiments. Further, they provide detailed information on effects causing changes in fluorescence

response. Through combinations of these approaches, valuable information on the time scales of

the investigated events can be had, allowing for discrimination between static and dynamic

molecular events.

Conflicts of interest

The authors declare the following competing financial interests: Bo W. Laursen is

associated with the company KU-dyes, which produces and sells fluorescent dyes

(including triangulenium dyes).

17

Acknowledgements

The work was supported by the Danish Council of Independent Research (DFF-6111-00483).

References

1. G. L. Perlovich, A. O. Surov and A. Bauer-Brandl, J Pharm Biomed Anal, 2007, 45, 679-

687.

2. K. Ariga, M. Nishikawa, T. Mori, J. Takeya, L. K. Shrestha and J. P. Hill, Sci Technol Adv

Mater, 2019, 20, 51-95.

3. E. Oliveira, E. Bertolo, C. Nunez, V. Pilla, H. M. Santos, J. Fernandez-Lodeiro, A.

Fernandez-Lodeiro, J. Djafari, J. L. Capelo and C. Lodeiro, Chemistryopen, 2018, 7, 9-52.

4. A. N. Butkevich, V. N. Belov, K. Kolmakov, V. V. Sokolov, H. Shojaei, S. C. Sidenstein,

D. Kamin, J. Matthias, R. Vlijm, J. Engelhardt and S. W. Hell, Chemistry, 2017, 23, 12114-12119.

5. X. Wang, Y. Z. Hu, A. Chen, Y. Wu, R. Aggeler, Q. Low, H. C. Kang and K. R. Gee,

Chem Commun (Camb), 2016, 52, 4022-4024.

6. C. A. Hunter and J. K. M. Sanders, J Am Chem Soc, 1990, 112, 5525-5534.

7. D. Chandler, Nature, 2005, 437, 640-647.

8. J. Li, N. Kwon, Y. Jeong, S. Lee, G. Kim and J. Yoon, ACS Appl Mater Interfaces, 2018,

10, 12150-12154.

9. M. Collot, T. K. Fam, P. Ashokkumar, O. Faklaris, T. Galli, L. Danglot and A. S.

Klymchenko, J Am Chem Soc, 2018, 140, 5401-5411.

10. N. J. Hestand and F. C. Spano, Acc Chem Res, 2017, 50, 341-350.

11. K. Tani, C. Ito, Y. Hanawa, M. Uchida, K. Otaguro, H. Horiuchi and H. Hiratsuka, J Phys

Chem B, 2008, 112, 836-844.

12. A. B. Descalzo, P. Ashokkumar, Z. Shen and K. Rurack, Chemphotochem, 2019, 4, 120-

131.

18

13. R. L. Halterman, J. L. Moore and W. T. Yip, J Fluoresc, 2011, 21, 1467-1478.

14. L. C. Zanetti-Domingues, C. J. Tynan, D. J. Rolfe, D. T. Clarke and M. Martin-Fernandez,

Plos One, 2013, 8, e74200.

15. C. Sanchez-Rico, L. Voith von Voithenberg, L. Warner, D. C. Lamb and M. Sattler,

Chemistry, 2017, 23, 14267-14277.

16. L. D. Hughes, R. J. Rawle and S. G. Boxer, Plos One, 2014, 9, e87649.

17. Y. S. Zeng, R. C. Gao, T. W. Wu, C. Cho and K. T. Tan, Bioconjug Chem, 2016, 27, 1872-

1879.

18. N. Panchuk-Voloshina, R. P. Haugland, J. Bishop-Stewart, M. K. Bhalgat, P. J. Millard, F.

Mao, W. Y. Leung and R. P. Haugland, J Histochem Cytochem, 1999, 47, 1179-1188.

19. F. Mao, W. Y. Leung and R. P. Haugland, 1999, WO9915517.

20. W. Lin and T. Chen, Anal Biochem, 2013, 443, 252-260.

21. S. Daddi Oubekka, R. Briandet, M. P. Fontaine-Aupart and K. Steenkeste, Antimicrob

Agents Chemother, 2012, 56, 3349-3358.

22. K. Suhling, J. Levitt and P. H. Chung, Methods Mol Biol, 2014, 1076, 503-519.

23. K. Kolmakov, V. N. Belov, J. Bierwagen, C. Ringemann, V. Muller, C. Eggeling and S.

W. Hell, Chemistry, 2010, 16, 158-166.

24. K. Kolmakov, C. A. Wurm, R. Hennig, E. Rapp, S. Jakobs, V. N. Belov and S. W. Hell,

Chemistry, 2012, 18, 12986-12998.

25. H. J. Wories, J. H. Koek, G. Lodder, J. Lugtenburg, R. Fokkens, O. Driessen and G. R.

Mohn, Recl Trav Chim Pay B, 1985, 104, 288-291.

26. L. Li, J. Han, B. Nguyen and K. Burgess, J Org Chem, 2008, 73, 1963-1970.

27. S. L. Niu, G. Ulrich, R. Ziessel, A. Kiss, P. Y. Renard and A. Romieu, Org Lett, 2009, 11,

2049-2052.

28. T. Bura and R. Ziessel, Org Lett, 2011, 13, 3072-3075.

19

29. S. L. Niu, C. Massif, G. Ulrich, P. Y. Renard, A. Romieu and R. Ziessel, Chemistry, 2012,

18, 7229-7242.

30. A. Romieu, D. Tavernier-Lohr, S. Pellet-Rostaing, M. Lemaire and P.-Y. Renard,

Tetrahedron Letters, 2010, 51, 3304-3308.

31. A. Romieu, T. Bruckdorfer, G. Clave, V. Grandclaude, C. Massif and P. Y. Renard, Org

Biomol Chem, 2011, 9, 5337-5342.

32. C. Massif, S. Dautrey, A. Haefele, R. Ziessel, P. Y. Renard and A. Romieu, Org Biomol

Chem, 2012, 10, 4330-4336.

33. K. Kolmakov, V. N. Belov, C. A. Wurm, B. Harke, M. Leutenegger, C. Eggeling and S.

W. Hell, European Journal of Organic Chemistry, 2010, 2010, 3593-3610.

34. R. M. Rich, D. L. Stankowska, B. P. Maliwal, T. J. Sorensen, B. W. Laursen, R. R.

Krishnamoorthy, Z. Gryczynski, J. Borejdo, I. Gryczynski and R. Fudala, Anal Bioanal Chem,

2013, 405, 2065-2075.

35. B. P. Maliwal, R. Fudala, S. Raut, R. Kokate, T. J. Sorensen, B. W. Laursen, Z. Gryczynski

and I. Gryczynski, Plos One, 2013, 8, e63043.

36. A. Shivalingam, M. A. Izquierdo, A. L. Marois, A. Vysniauskas, K. Suhling, M. K.

Kuimova and R. Vilar, Nat Commun, 2015, 6, 8178.

37. T. J. Sørensen, E. Thyrhaug, M. Szabelski, R. Luchowski, I. Gryczynski, Z. Gryczynski

and B. W. Laursen, Methods Appl Fluores, 2013, 1, 025001.

38. S. A. Bogh, I. Bora, M. Rosenberg, E. Thyrhaug, B. W. Laursen and T. J. Sorensen,

Methods Appl Fluoresc, 2015, 3, 045001.

39. S. A. Bogh, M. Simmermacher, M. Westberg, M. Bregnhoj, M. Rosenberg, L. De Vico,

M. Veiga, B. W. Laursen, P. R. Ogilby, S. P. A. Sauer and T. J. Sorensen, Acs Omega, 2017, 2,

193-203.

40. T. J. Sørensen, M. Rosenberg, C. G. Frankær and B. W. Laursen, Advanced Materials

Technologies, 2018, 4, 1800561.

20

41. I. Dalfen, R. I. Dmitriev, G. Holst, I. Klimant and S. M. Borisov, Anal Chem, 2019, 91,

808-816.

42. I. Bora, S. A. Bogh, M. Rosenberg, M. Santella, T. J. Sorensen and B. W. Laursen, Org

Biomol Chem, 2016, 14, 1091-1101.

43. E. Thyrhaug, T. J. Sorensen, I. Gryczynski, Z. Gryczynski and B. W. Laursen, J Phys Chem

A, 2013, 117, 2160-2168.

44. C. Herse, D. Bas, F. C. Krebs, T. Burgi, J. Weber, T. Wesolowski, B. W. Laursen and J.

Lacour, Angew Chem Int Ed Engl, 2003, 42, 3162-3166.

45. M. Rosenberg, K. R. Rostgaard, Z. Liao, A. O. Madsen, K. L. Martinez, T. Vosch and B.

W. Laursen, Chem Sci, 2018, 9, 3122-3130.

46. M. Rosenberg, M. Santella, S. A. Bogh, A. V. Munoz, H. O. B. Andersen, O. Hammerich,

I. Bora, K. Lincke and B. W. Laursen, J Org Chem, 2019, 84, 2556-2567.

47. O. Kel, P. Sherin, N. Mehanna, B. Laleu, J. Lacour and E. Vauthey, Photochem Photobiol

Sci, 2012, 11, 623-631.

48. J. Guin, C. Besnard and J. Lacour, Org Lett, 2010, 12, 1748-1751.

49. Y. K. Yang and J. Tae, Org Lett, 2006, 8, 5721-5723.

50. A. Joshi-Pangu, F. Levesque, H. G. Roth, S. F. Oliver, L. C. Campeau, D. Nicewicz and

D. A. DiRocco, J Org Chem, 2016, 81, 7244-7249.

51. J. B. Grimm, A. K. Muthusamy, Y. Liang, T. A. Brown, W. C. Lemon, R. Patel, R. Lu, J.

J. Macklin, P. J. Keller, N. Ji and L. D. Lavis, Nat Methods, 2017, 14, 987-994.

52. M. Beija, C. A. Afonso and J. M. Martinho, Chem Soc Rev, 2009, 38, 2410-2433.

53. B. W. Laursen and F. C. Krebs, Chem-Eur J, 2001, 7, 1773-1783.

54. I. H. Delgado, S. Pascal, C. Besnard, S. Voci, L. Bouffier, N. Sojic and J. Lacour,

Chemistry, 2018, 24, 10186-10195.

55. I. H. Delgado, S. Pascal, A. Wallabregue, R. Duwald, C. Besnard, L. Guenee, C. Nancoz,

E. Vauthey, R. C. Tovar, J. L. Lunkley, G. Muller and J. Lacour, Chem Sci, 2016, 7, 4685-4693.

21

56. X.-M. Hu, Q. Chen, Z.-Y. Sui, Z.-Q. Zhao, N. Bovet, B. W. Laursen and B.-H. Han, Rsc

Adv, 2015, 5, 90135-90143.

57. R. Duwald, S. Pascal, J. Bosson, S. Grass, C. Besnard, T. Burgi and J. Lacour, Chemistry,

2017, 23, 13596-13601.

58. P. Shieh, V. T. Dien, B. J. Beahm, J. M. Castellano, T. Wyss-Coray and C. R. Bertozzi, J

Am Chem Soc, 2015, 137, 7145-7151.

59. E. Thyrhaug, T. J. Sorensen, I. Gryczynski, Z. Gryczynski and B. W. Laursen, J Phys Chem

A, 2013, 117, 2160-2168.

60. A. K. Chibisov, H. Görner and T. D. Slavnova, Chem Phys Lett, 2004, 390, 240-245.

61. G. L. Trainor, Expert Opin Drug Discov, 2007, 2, 51-64.

62. P. Taborsky, J. Kucera, J. Jurica and O. Pes, J Chromatogr B Analyt Technol Biomed Life

Sci, 2018, 1092, 7-14.

63. K. Klehs, C. Spahn, U. Endesfelder, S. F. Lee, A. Furstenberg and M. Heilemann,

Chemphyschem, 2014, 15, 637-641.

64. J. Olmsted, 3rd and D. R. Kearns, Biochemistry, 1977, 16, 3647-3654.

65. J. Kucera, P. Lubal, S. Lis and P. Taborsky, Talanta, 2018, 184, 364-368.

download fileview on ChemRxivPreprint.pdf (603.42 KiB)



S1

Electronic Supplementary Information

What is best strategy for water soluble fluorescence dyes? – A

case study using long fluorescence lifetime DAOTA dyes

Niels Bisballea, Bo W. Laursena*

a Nano-Science Center & Department of Chemistry, University of Copenhagen, Universitetsparken 5, DK-

2100, Copenhagen Ø, Denmark.

*[email protected]

Table of Contents Experimental details ..................................................................................................................................... S2

Synthetic procedures .................................................................................................................................... S3

Steady state spectra of 1 to 6 ..................................................................................................................... S11

Time resolved fluorescence data for 1 to 6 ................................................................................................ S18

Determination of quantum yields for 1 to 6 .............................................................................................. S24

NMR spectra of 1 to 6 and 8 to 12 ............................................................................................................. S41

References .................................................................................................................................................. S59

S2

Experimental details Spectroscopy. All measurements were performed on dye solutions in a 10 mm quartz cuvette. Fluorescence

spectra were recorded of solutions with an absorbance below 0.1 in respect to the S0 ->S1 transition. UV/Vis

absorption spectra were recorded on an Agilent Cary 300 UV-Vis Spectrophotometer. Absorption was

measured using a single beam setup and subtracting a spectrum of the pure solvent as the background. Data

points were acquired for each 1 nm. Emission and excitation spectra were recorded on an Agilent Cary Eclipse

Fluorescence Spectrophotometer. Fluorescence lifetimes were measured by Time-correlated single photon

counting (TCSPC) on a FluoTime 300 instrument (PicoQuant, Berlin, Germany) using a PMA-182 detector

(PicoQuant) with a spectral range of 185-820 nm. Samples were excited using pulsed laser excitation at 507

nm (LDH-P-C-510, PicoQuant) and the emission monochromator was set to 590 nm. All decays were recorded

at a sample temperature of 22°C. Decay data was fitted to an exponential function (mono-, bi- or tri-

exponential) using the FluoFit software package (PicoQuant) by reconvolution with a resulting χ2 < 1.1. The

instrument response function was recorded using a dilute Ludox solution with the emission monochromator

set to the excitation wavelength.

Quantum yields of fluorescence were measured relative to Rhodamine 6G in ethanol (φ = 0.95 ± 0.015)1

according to standard protocol2 and IUPAC recommendations3. Absorption spectra were recorded as

described above with a spectral bandwidth of 2.0 nm. Emission spectra were recorded on a FluoTime 300

instrument with a sample holder temperature of 22°C. Samples were excited using a continuous-wave xenon

arc lamp (PicoQuant) at 500 nm with a spectral bandwidth of 2.0 nm. Emission was recorded using a PMA-

182 detector from 505-800 nm with a spectral bandwidth of 1.5 nm. Data points were collected with a 1 nm

interval. Emission spectra were corrected for detector sensitivity using the manufacturer-supplied correction

file and for excitation power by recording lamp intensity with a photodiode for each point. A background

spectrum recorded using pure solvent was subtracted. For each compound a linear fit was obtained by

plotting the integrated emission spectra as a function of the fraction of absorbed light, f, at 500 nm (eq. 12).

𝑓 = 1 − 10−𝐴(𝜆𝑒𝑥)

Quantum yields of fluorescence were calculated from the using eq. 2 (modified from ref. 2) by using the slope

of the linear fit obtained above (Δ), where n is the refractive index of the solvent. Values for the refractive

indices were used for the sodium D-line at 298.15K (nD, ethanol = 1.3593, nD, water = 1.3326).4

Φ𝑓 = Φ𝑓,𝑟𝑒𝑓 ×𝑛2

𝑛𝑟𝑒𝑓2 ×

Δ

Δ𝑟𝑒𝑓

Dyes and solvents. A commercially available source of Rhodamine 6G was used without further purification.

Absolute ethanol (≥ 99.98%) was acquired from VWR International. Water was purified to have a conductivity

below 0.056 µS cm-1. Phosphate buffered saline (PBS) was prepared by dissolving NaCl (4.00 g), KCl (101 mg),

Na2HPO4 (710 mg) and KH2PO4 (120 mg) in approximately 400 mL of purified water. pH was measured to be

7.36 and the solution was diluted further to 500 mL in a volumetric flask with purified water. The solution of

bovine serum albumin (BSA) in PBS was prepared by gentle dispersion of BSA (801 mg, ≥ 96%, lyophilized,

Sigma Aldrich) in about 15 mL of PBS in a 20 mL volumetric flask. The dispersion was left standing overnight

to allow foaming to cease, after which more PBS was added for a total volume of 20 mL. Fluorescence lifetime

measurements in BSA-PBS were performed on the day the solution had been prepared.

S3

Synthetic procedures

Scheme S1. Synthesis of 1-3 from PrDAOTA+ BF4-. (a) Amberlite© IRA-400(Cl) ion exchange resin,

MeCN/H2O; (b) H2SO4, rt, 1 h; (c) H2SO4, rt, 16 h.

S4

Scheme S2. Synthesis of 4-6 starting from the common triangulenium precursor 7. (a) i. 3-(Dimethylamino)-

1-propylamin, NMP, rt, 30 min; ii. 110°C, 20 min; iii. KPF6 (aq); (b) Pyridine hydrochloride, 200°C, 1.5 h; (c)

1,3-propanesultone, DIPEA, MeCN, rt, 16 h, darkness; (d) Iodomethane, K2CO3, MeCN, rt, 16 h; (e) H2SO4, rt,

16 h; (f) Amberlite© IRA-400(Cl) ion exchange resin, MeCN/H2O.

S5

General remarks. Compound 7 and Pr2DAOTA+ BF4- were prepared according to literature procedures.5, 6

Other starting materials and reagents were obtained through commercial suppliers and used as received

without further purification, unless otherwise noted. Acetonitrile was dried over 4 Å molecular sieves for at

least 48 hours prior to use, where the dry solvent was required. C18 RP silica with a pore size of 53-80 Å and

a particle size of 35-70 µm was obtained from Carl Roth Gmbh (Karlsruhe, Germany). Anion exchange was

performed using Amberlite© IRA-400(Cl) resin (CAS: 9002-24-8) obtained from Alfa Aesar Gmbh (Karlsruhe,

Germany). 1H-NMR, 13C-NMR and 19F-NMR spectra were acquired on 500 MHz instruments by Bruker.

Chemical shifts for 1H-NMR and 13C-NMR spectra are reported relative to TMS, referenced to the solvent

residual peaks; CDCl3 (1H = 7.26, 13C = 77.16), CD3CN (1H = 1.94, 13C = 1.32), CD3OD (1H = 3.31, 13C = 49.00),

D2O (1H = 4.79), except for 13C-NMR in D2O, in which signals are referenced to the (CH3)3Si- signal of the

internal DSS standard. 19F-NMR chemical shifts are referenced to neat trifluoroacetic acid as an external

reference (δ = -77.87 ppm relative to CFCl3)7 in a separate flame-sealed lock tube placed coaxially inside the

NMR tube during sample acquisition. For 19F-NMR spectra the FID was enhanced by exponential

multiplication with a line broadening factor of 8.0 Hz to reduce noise caused by FID truncation. Peak

separation (705 – 708 Hz) remains well above spectral resolution (4.23 Hz). HRMS was recorded on an ESP-

MALDI-FT-ICR instrument equipped with a 7 T magnet (the instrument was calibrated using sodium

trifluoroacetate cluster ions prior to acquiring the spectra) or a MicrOTOF-QII-system using ESP (calibrated

using formic acid). Elemental analysis was done at the University of Copenhagen, Department of Chemistry,

Elemental Analysis Laboratory, Universitetsparken 5, Copenhagen DK-2100, Denmark.

General procedure 1 – Anion exchange of triangulenium dyes: A column of Amberlite IRA-400 (Cl) ion

exchange resin (60 mm in height x 10 mm in diameter) was rinsed thoroughly with water (300 mL), before

being equilibrated with the specified eluent (100 mL). The hexafluorophosphate salt of the triangulenium

dye was dissolved in a minimal amount of the eluent and passed through column. The colored eluate was

collected and the solvents were removed under reduced pressure, followed by drying under vacuum (<

1mbar) for at least 12 h to give the corresponding chloride salt.

General procedure 2 – Aromatic sulfonation of triangulenium dyes: A small vial was charged with the

corresponding triangulenium dye and a magnetic stir bar. Concentrated sulfuric acid (98 %) was added and

gentle fuming and bright orange luminescence was briefly observed. The vial was sealed with a screw cap

and was kept at room temperature for the indicated amount of time under moderate stirring. The deep red

reaction solution was quenched by adding it to a slurry of 15 g of ice in 5 mL of water. The magenta red

solution was neutralized by careful addition of 1 M K2CO3 (18 times the volume of sulfuric acid used), resulting

in CO2 (g) formation. The solvent was carefully evaporated under reduced pressure to leave the dye on a base

of potassium sulfate. A plug of C18-functionalized silica gel (40 mm in height x 40 mm in diameter) in

water/acetonitrile 19:1 was prepared and at no point allowed to run dry of eluent. The mixture of dye and

inorganic salts was dissolved in a minimal amount of the eluent and gently loaded onto the plug without

disturbing the surface. The salts were eluted through the plug with 100 mL of the loading eluent. The eluent

was switched to water/acetonitrile 1:1 to elute the product off the plug. The eluate containing the dye was

concentrated under reduced pressure, and the product was purified by reversed-phase chromatography on

C18-functionalized silica (water-acetonitrile-formic acid) to give the pure dye.

S6

8,12-Dipropyl-8,12-dihydro-3a2H-benzo[ij]xantheno[1,9,8-cdef][2,7]naphthyridin-3a2-ylium chloride (1):

The product was obtained through the general procedure 1, starting from Pr2DAOTA+ BF4- (50 mg, 0.11

mmol). Water-acetonitrile 1:1 was used as the eluent and gave the pure product as a dark red solid (45 mg,

>99 %). 1H-NMR (500 MHz, CD3OD): δ 8.27 (t, J = 8.6 Hz, 1H), 8.08 (t, J = 8.5 Hz, 2H), 7.61 (d, J = 8.8 Hz, 2H),

7.52 (d, J = 8.6 Hz, 2H), 7.29 (d, J = 8.2 Hz, 2H), 4.49 (m, 4H), 1.97 (m[h], J = 7.5 Hz, 4H), 1.21 (t, J = 7.4 Hz, 6H). 13C-NMR (126 MHz, CD3OD): δ 154.0 (-), 142.3 (-), 141.1 (-), 141.1 (-), 140.9 (+), 139.7 (+), 112.9 (-), 110.3 (+),

109.6 (+), 108.9 (-), 107.0 (+), 50.3 (-), 20.2 (-), 11.1 (+). HRMS (ESP): Calcd for C25H23N2O+ [M+]: 367.1810.

Found: 367.1804. Elem. Anal. (Found: C, 73.25; H, 6.4; N, 6.5. C25H23ClN2O requires: C, 74.5; H, 5.75; N, 6.95

%). Rf (Water:Acetonitrile:Formic acid, 60:40:0.1) = 0.07.

8,12-Dipropyl-8,12-dihydrobenzo[ij]xantheno[1,9,8-cdef][2,7]naphthyridin-3a2-ylium-3-sulfonate (2): The

product was obtained through the general procedure 2, reacting Pr2DAOTA+ BF4- (52 mg, 0.11 mmol) in 500

µL of sulfuric acid for 1 h. Purification by reversed-phase column chromatography was achieved with an

eluent of water-acetonitrile-formic acid in a 60:40:0.1 ratio to give the product as a magenta solid (1.4 mg, 3

%). A substantial impurity could still be detected by NMR, giving rise to a single at δ 8.48, presumably being

a formate salt. Due to the low quantities of product and the insignificance of the salt in fluorescence

spectroscopy, no further effort was made to remove it. 1H-NMR (500 MHz, D2O): δ 8.30 (d, J = 9.1 Hz, 1H),

7.86 (t, J = 8.4 Hz, 2H), 7.38 (d, J = 9.2 Hz, 1H), 7.30 (d, J = 8.1 Hz, 1H), 7.19 (d, J = 8.8 Hz, 1H), 7.10 (d, J = 8.6

Hz, 1H), 7.02 (d, J = 8.7 Hz, 1H), 4.14-4.03 (m, 2H), 3.98-3.86 (m, 2H), 1.77-1.67 (m, 2H), 1.59-1.48 (m, 2H),

1.05 (t, J = 7.3 Hz, 3H), 0.98 (t, J = 7.4 Hz, 3H). Rf (Water:Acetonitrile:Formic acid, 60:40:0.1) = 0.53.

Potassium 8,12-dipropyl-8,12-dihydrobenzo[ij]xantheno[1,9,8-cdef][2,7]naphthyridin-3a2-ylium-3,5-

disulfonate (3): The product was obtained through the general procedure 2, reacting Pr2DAOTA+ BF4- (51 mg,

S7

0.11 mmol) in 500 µL of sulfuric acid for 16 h. Purification by reversed-phase chromatography was achieved

with an eluent of water-acetonitrile-formic acid in a 80:20:0.1 ratio to give the product as a magenta red solid

(56 mg, 88%). 1H-NMR (500 MHz, D2O): δ 8.45 (d, J = 9.1 Hz, 2H), 7.89 (t, J = 8.5 Hz, 1H), 7.45 (d, J = 9.3 Hz,

2H), 7.11 (d, J = 8.6 Hz, 2H), 4.20 (s br, 4H), 1.76-1.63 (m, 4H), 1.02 (t, J = 7.3 Hz, 6H). 13C-NMR (126 MHz,

D2O): δ 150.7 (-), 143.0 (-), 142.8 (+), 142.0 (-), 141.4 (-), 140.4 (+), 126.3 (-), 113.6 (-), 112.0 (+), 109.2 (-),

108.9 (+), 51.7 (-), 21.2 (-), 12.7 (+). HRMS (ESP): Calcd for C25H21N2O7S2- [M-]: 525.0790. Found: 525.0811.

Elem. Anal. (Found: C, 52.15; H, 3.6; N, 4.7; S, 10.35. C25H21KN2O7S2 requires: C, 53.2; H, 3.75; N, 4.95; S, 11.35

%). Rf (Water:Acetonitrile:Formic acid, 60:40:0.1) = 0.28.

8,12-Bis(3-(trimethylammonio)propyl)-8,12-dihydro-3a2H-benzo[ij]xantheno[1,9,8-

cdef][2,7]naphthyridin-3a2-ylium chloride (4): The product was obtained through the general procedure 1,

starting from 11 (50 mg, 54 µmol). Water-acetonitrile 1:1 was used as the eluent and gave the pure product

as a dark red solid (30 mg, 93 %). 1H-NMR (500 MHz, CD3OD): δ 8.38 (t, J = 8.6 Hz, 1H), 8.20 (t, J = 8.5 Hz, 2H),

7.81 (d, J = 8.8 Hz, 2H), 7.76 (d, J = 8.6 Hz, 2H), 7.42 (d, J = 8.2 Hz, 2H), 4.71 (t, J = 7.9, 4H), 3.82 (m, 4H), 3.22

(s, 18H), 2.46 (p, J = 8.9 Hz, 4H). 13C-NMR (126 MHz, CD3OD): δ 154.3 (-), 142.4 (-), 141.8 (-), 141.5 (+), 141.1

(-), 140.3 (+), 113.0 (-), 110.3 (+), 110.1 (+), 109.2 (-), 107.4 (+), 64.1 (-), 53.9 (+), 53.9 (+), 53.8 (+), 45.5 (-),

20.9 (-). HRMS (ESP): Calcd for C31H39Cl2N4O+ [M3++2Cl-]: 553.2501. Found: 553.2511. Elem. Anal. (Found: C,

46.5; H, 6.4; N, 6.55. C31H39Cl3N4O requires: C, 63.1; H, 6.65; N, 9.5 %)i. Rf (Water:Acetonitrile:Formic acid,

60:40:0.1) = 0.22.

3,3'-((Benzo[ij]xantheno[1,9,8-cdef][2,7]naphthyridine-3a2-ylium-8,12-diylbis(propane-3,1-

diyl))bis(dimethylammoniumdiyl))bis(propane-1-sulfonate) chloride (5): The product was obtained through

the general procedure 1, starting from 10 (50 mg, 59 µmol). Water-acetonitrile 1:1 was used as the eluent

and gave the pure product as a dark red solid (42 mg, 97 %). 1H-NMR (500 MHz, D2O): δ 8.25 (t, J = 8.5 Hz,

1H), 8.07 (t, J = 8.4 Hz, 2H), 7.41 (d, J = 8.8 Hz, 2H), 7.36 (d, J = 8.7 Hz, 2H), 7.15 (d, J = 8.2 Hz, 2H), 4.36 (s br,

4H), 3.57 (s br, 4H), 3.45 (t, J = 8.6 Hz, 4H), 3.11 (s, 12H), 2.84 (t, J = 7.0 Hz, 4H), 2.29 (s br, 4H), 2.08 (s br, 4H). 13C-NMR (126 MHz, D2O): δ 154.7 (-), 143.0 (+), 142.6 (-), 142.0 (+), 141.9 (-), 141.3 (-), 113.1 (-), 111.9 (+),

i There appears to be an issue with incomplete combustion of 4, presumably due to the tricationic character. The same relative deviation from the calculated values is observed for the corresponding PF6

- salt (11).

S8

111.1 (+), 109.3 (-), 108.5 (+), 64.4 (-), 62.3 (-), 53.9 (+), 49.6 (-), 46.1 (-), 21.7 (-), 20.9 (-). HRMS (ESP): Calcd

for C35H45N4O7S2+ [M+]: 697.2730. Found: 697.2739. Elem. Anal. (Found: C, 55.05; H, 6.45; N, 7.15; S, 7.7.

C35H45ClN4O7S2 requires: C, 57.35; H, 6.2; N, 7.65; S, 8.75 %). Rf (Water:Acetonitrile:Formic acid, 60:40:0.1) =

0.36.

8,12-Bis(3-(trimethylammonio)propyl)-8,12-dihydrobenzo[ij]xantheno[1,9,8-cdef][2,7]naphthyridin-3a2-

ylium-3,5-disulfonate chloride (6): The product was obtained through the general procedure 1, starting from

12 (12 mg, 15 µmol). Water-acetonitrile 1:1 was used as the eluent and gave 8 mg of the pure product as a

dark red solid (77 %, 12 µmol). 1H-NMR (500 MHz, D2O): δ 8.49 (d, J = 9.0 Hz, 2H), 8.02 (t, J = 8.5 Hz, 1H), 7.46

(d, J = 9.2 Hz, 2H), 7.26 (d, J = 8.7 Hz, 2H), 4.61 (s br, 4H), 3.71 – 3.61 (m, 4H), 3.15 (s, 18H), 2.27 (m, 4H).

HRMS (ESP): Calcd for C31H37N4O7S2+ [M+]: 641.2104. Found: 641.0811. Rf (Water:Acetonitrile:Formic acid,

60:40:0.1) = 0.21.

5,9-Bis(3-(dimethylamino)propyl)-1,13-dimethoxy-5,9-dihydro-13bH-quinolino[2,3,4-kl]acridin-13b-ylium

hexafluorophosphate (8): A solution of 7 (2.00 g, 3.92 mmol) in NMP (10 mL) was added dropwise to a

solution of 3-(dimethylamino)-1-propylamine (1.75 mL, 13.9 mmol) in NMP (5 mL) over 15 min. After another

15 min the resulting red solution was heated to 110°C for 20 min. The dark green solution was allowed to

cool to rt and was crashed out in 500 mL of a 9:1 mixture of 0.2 M KPF6 and 1 M NaOH. The formed precipitate

was isolated by filtration. The product was dissolved in a minimal amount of acetonitrile and poured into

Et2O (750 mL) under vigorous stirring. The precipitate was isolated by filtration and washed with Et2O (3 x 50

mL). The solid was recrystallized from a hot mixture of 2-propanol and acetonitrile (9:1, 100 mL/g). The

product was obtained as 710 mg of dark crystals (56 %). 1H-NMR (500 MHz, CD3CN): δ 8.20 (t, J = 8.5 Hz, 1H),

7.91 (dd, J = 8.7, 8.0 Hz, 2H), 7.66 (d, J = 8.6 Hz, 2H), 7.59 (d, J = 8.9 Hz, 2H), 6.92 (d, J = 8.0 Hz, 2H), 4.81-4.73

(m, 2H), 4.62-4.52 (m, 2H), 3.73 (s, 6H), 2.50 (t, J = 6.3 Hz, 4H), 2.28 (s, 12H), 2.21-2.07 (m, 4H). 13C-NMR (126

MHz, CD3CN): δ 160.6 (-), 143.5 (-), 143.2 (-), 140.0 (-), 137.9 (+), 137.4 (+), 120.4 (-), 113.9 (-), 108.5 (+), 105.9

(+), 103.9 (+), 57.2 (-), 56.6 (+), 45.9 (+), 25.2 (-). 19F-NMR (470 MHz, CD3CN): δ -72.26 (d, 1JPF = 705 Hz, PF6-).

HRMS (ESP): Calcd for C31H39N4O2+ [M+]: 499.3073. Found: 499.3066. Elem. Anal. (Found: C, 57.15; H, 6.2; N,

8.55. C31H39F6N4O2P requires: C, 57.75; H, 6.1; N, 8.7 %).

S9

8,12-Bis(3-(dimethylamino)propyl)-8,12-dihydro-3a2H-benzo[ij]xantheno[1,9,8-cdef][2,7]naphthyridin-

3a2-ylium hexafluorophosphate (9): 8 (1.20 g, 1.86 mmol) was added to molten pyridine hydrochloride (20

g) at 200°C and stirred gently for 1.5 h, while the dark green solution turned magenta. The reaction mixture

was allowed to cool to room temperature, before 0.2 M KPF6 (100 mL) was added to dissolve the pyridine

hydrochloride and precipitate the product. The magenta precipitate was isolated by filtration, before being

reprecipitated in 0.2 M KPF6 (100 mL, basified with 10 mL 1M NaOH) from a minimal amount of acetonitrile

and filtered again. The isolated magenta solid was reprecipitated in Et2O (250 mL) from acetonitrile and

isolated by filtration. The solid was washed with Et2O (3 x 50 mL). The product was recrystallized from hot

MeOH (25 mL/g) and washed with cold MeOH (10 mL) and Et2O (3 x 25 mL) to give 583 mg (52 %, 0.974

mmol) of red crystals. 1H-NMR (500 MHz, CD3CN): δ 8.23 (t, J = 8.6 Hz, 1H), 8.05 (t, J = 8.5 Hz, 2H), 7.64 (d, J

= 8.8 Hz, 2H), 7.55 (d, J = 8.7 Hz, 2H), 7.27 (d, J = 8.1 Hz, 2H), 4.53 (t, J = 7.7 Hz, 4H), 2.46 (t, J = 6.3 Hz, 4H),

2.28 (s, 12H), 2.03 (m, 4H). 13C-NMR (126 MHz, CD3CN): δ 153.6 (-), 142.1 (-), 141.0 (-), 140.8 (-), 140.6 (+),

139.4 (+), 112.7 (-), 110.3 (+), 109.3 (+), 108.7 (-), 106.8 (+), 56.8 (-), 47.2 (-), 45.8 (+), 24.6 (-). 19F-NMR (470

MHz, CD3CN): δ -72.91 (d, 1JPF = 707 Hz, PF6-). HRMS (ESP): Calcd for C29H33N4O+ [M+]: 453.2654. Found:

453.2650. Elem. Anal. (Found: C, 57.45; H, 5.45; N, 9.1. C29H33F6N4OP requires: C, 58.2; H, 5.55; N, 9.35 %).

3,3'-((Benzo[ij]xantheno[1,9,8-cdef][2,7]naphthyridine-3a2-ylium-8,12-diylbis(propane-3,1-

diyl))bis(dimethylammoniumdiyl))bis(propane-1-sulfonate) hexafluorophosphate (10): 9 (99 mg, 0.166

mmol) and 1,3-propanesultone (203 mg, 1.66 mmol) were dissolved in dry acetonitrile (1.7 mL) under argon

in a dry round-bottomed flask. Diisopropylethylamine (0.29 mL, 1.66 mmol) was added and the flask was

wrapped in aluminium foil to shield the reaction from light. The dark red reaction was stirred at room

temperature for 17 hours, before being poured into Et2O (100 mL) under vigorous stirring. The magenta

precipitate was isolated by filtration and washed with Et2O (3 x 25 mL). The product was recrystallized from

hot water (6 mL). The crystals were dried to give 104 mg (74 %, 0.123 mmol). 1H-NMR (500 MHz, D2O): δ 8.25

(t, J = 8.5 Hz, 1H), 8.08 (t, J = 8.4 Hz, 2H), 7.43 (d, J = 8.8 Hz, 2H), 7.36 (d, J = 8.7 Hz, 2H), 7.17 (d, J = 8.2 Hz,

2H), 4.39 (s br, 4H), 3.56 (s br, 4H), 3.44 (t, J = 8.5 Hz, 4H), 3.10 (s, 12 H), 2.84 (t, J = 7.0 Hz, 4H), 2.30 (s br,

4H), 2.08 (s br, 4H). 13C-NMR (126 MHz, D2O): δ 154.7 (-), 143.0 (+), 142.9 (-), 142.0 (+), 142.0 (-), 141.3 (-),

113.2 (-), 111.9 (+), 111.1 (+), 109.4 (-), 108.5 (+), 64.5 (-), 62.3 (-), 53.9 (+), 49.6 (-), 46.1 (-), 21.7 (-), 20.9 (-). 19F-NMR (470 MHz, D2O): δ -72.14 (d, 1JPF = 708 Hz, PF6

-). HRMS (ESP): Calcd for C35H45N4O7S2+ [M+]: 697.2730.

S10

Found: 697.2746. Elem. Anal. (Found: C, 47.45; H, 5.65; N, 6.2; S, 6.9. C35H45F6N4O7PS2 requires: C, 49.9; H,

5.4; N, 6.65; S, 7.6 %). Rf (Water:Acetonitrile:Formic acid, 60:40:0.1) = 0.31.

8,12-Bis(3-(trimethylammonio)propyl)-8,12-dihydro-3a2H-benzo[ij]xantheno[1,9,8-

cdef][2,7]naphthyridin-3a2-ylium hexafluorophosphate (11): Iodomethane (20 µL, 321 µmol) was added to

a dark red solution of 9 (100 mg, 0.167 mmol) and K2CO3 (46 mg, 0.334 mmol) in dry acetonitrile (1.7 mL)

under argon. After 16.5 hours of mild stirring at room temperature, the bright red reaction solution was

crashed out in 0.2 M KPF6 (100 mL). The red precipitate was isolated by filtration and washed with 0.2 M KPF6

(3 x 25 mL). The precipitate was reprecipitated in Et2O (200 mL) from a minimal volume of acetonitrile under

vigorous stirring. The red precipitate was isolated by filtration and washed with Et2O (3 x 25 mL). This gave

140 mg (91 %, 0.152 mmol) of an amorphous red solid. 1H-NMR (500 MHz, CD3CN): δ 8.32 (t, J = 8.6 Hz, 1H),

8.17 (t, J = 8.5 Hz, 2H), 7.67 (d, J = 8.8 Hz, 2H), 7.61 (d, J = 8.6 Hz, 2H), 7.42 (d, J = 8.2 Hz, 2H), 4.59 (t, J = 7.9

Hz, 4H), 3.64 (m, 4H), 3.09 (s, 18H), 2.37 (p, J = 8.0 Hz, 4H). 13C-NMR (126 MHz, CD3CN): δ 153.9 (-), 142.0 (-),

141.8 (-), 141.0 (+), 140.8 (-), 139.9 (+), 112.8 (-), 110.2 (+), 109.9 (+), 109.0 (-), 107.2 (+), 63.8 (-), 54.3 (+),

54.3 (+), 54.2 (+), 45.3 (-), 20.7 (-). 19F-NMR (470 MHz, CD3CN): δ -72.19 (d, 1JPF = 706 Hz, PF6-). HRMS (ESP):

Calcd for C31H39F12N4OP2+ [M3++2PF6

-]: 773.2407. Found: 773.2411. Elem. Anal. (Found: C, 29.15; H, 3.75; N,

4.05. C31H39F18N4OP3 requires: C, 40.55; H, 4.3; N, 6.1 %).

8,12-Bis(3-(trimethylammonio)propyl)-8,12-dihydrobenzo[ij]xantheno[1,9,8-cdef][2,7]naphthyridin-3a2-

ylium-3,5-disulfonate hexafluorophosphate (12): The product was obtained through the general procedure

2, reacting 11 (24 mg, 26 µmol) in 250 µL of sulfuric acid for 17 hours. Purification by reversed-phase

chromatography was achieved with an eluent of water-acetonitrile-formic acid in an 80:20:0.1 ratio to give

product as a magenta red solid (18 mg, 88 %). 1H-NMR (500 MHz, D2O): δ 8.53 (d, J = 9.1 Hz, 2H), 8.11-8.04

(m, 1H), 7.52 (d, J = 9.3 Hz, 2H), 7.33 (d, J = 8.5 Hz, 2H), 4.67-4.60 (, 4H), 3.73 – 3.66 (m, 4H), 3.16 (s, J = 1.8

Hz, 18H), 2.35-2.26 (s, 4H). 19F-NMR (470 MHz, D2O): δ -72.24 (d, 1JPF = 705 Hz, PF6-). HRMS (ESP): Calcd for

C31H37N4O7S2+ [M+]: 641.2104. Found: 641. 0811. Rf (Water:Acetonitrile:Formic acid, 60:40:0.1) = 0.19.

S11

Steady state spectra of 1 to 6

Figure S1. Normalized absorption, emission and excitation spectra for 1 (5 µM) in water. Wavelengths in

legend give the excitation wavelength for emission spectra and emission wavelength for excitation spectra.



300 400 500 600 700

0,0

0,2

0,4

0,6

0,8

1,0

No

rma

lize

d a

bso

rban

ce,

em

issio

n

Wavelength [nm]

Absorption

Ems 530nm

Ems 507nm

Ems 485nm

Exc 587nm

Exc 650nm

300 400 500 600 700

0,0

0,2

0,4

0,6

0,8

1,0

No

rma

lize

d a

bso

rban

ce,

em

issio

n

Wavelength [nm]

Absorption

Ems 530nm

Ems 507nm

Ems 485nm

Exc 594nm

Exc 650nm

S12





300 400 500 600 700

0,0

0,2

0,4

0,6

0,8

1,0

Norm

aliz

ed a

bsorb

ance, em

issio

n

Wavelength [nm]

Absorption

Ems 530nm

Ems 507nm

Ems 485nm

Exc 594nm

Exc 650nm

300 400 500 600 700

0,0

0,2

0,4

0,6

0,8

1,0

No

rma

lize

d a

bso

rban

ce,

em

issio

n

Wavelength [nm]

Absorption

Ems 530nm

Ems 507nm

Ems 485nm

Exc 577nm

Exc 650nm

S13





300 400 500 600 700

0,0

0,2

0,4

0,6

0,8

1,0

Norm

aliz

ed a

bsorb

ance, em

issio

n

Wavelength [nm]

Absorption

Ems 530nm

Ems 507nm

Ems 485nm

Exc 577nm

Exc 650nm

300 400 500 600 700

0,0

0,2

0,4

0,6

0,8

1,0

No

rma

lize

d a

bso

rban

ce,

em

issio

n

Wavlength [nm]

Absorption

Ems 530nm

Ems 507nm

Ems 485nm

Exc 586nm

Exc 650nm

S14

Figure S7. Normalized absorption and emission spectra for 1 (5 µM) in PBS and with the addition of 600 µM

BSA. Emission spectra were recorded with excitation at 520 nm. The normalization of the emission

spectrum with the addition of BSA (red dash) is corrected for intensity to show the effect of BSA.




400 500 600 700 800

0,0

0,2

0,4

0,6

0,8

1,0

No

rma

lize

d a

bso

rban

ce,

em

issio

n

Wavelength (nm)

Abs PBS

Abs PBS+BSA

Ems PBS

Ems PBS+BSA

400 500 600 700 800

0,0

0,2

0,4

0,6

0,8

1,0

No

rma

lize

d a

bso

rban

ce,

em

issio

n

Wavelength (nm)

Abs PBS

Abs PBS+BSA

Ems PBS

Ems PBS+BSA

S15




Figure S10. Normalized absorption and emission spectra for 4 (5 µM) in PBS and with the addition of 600

µM BSA. Emission spectra were recorded with excitation at 520 nm. The normalization of the emission


400 500 600 700 800

0,0

0,2

0,4

0,6

0,8

1,0

No

rma

lize

d a

bso

rban

ce,

em

issio

n

Wavelength (nm)

Abs PBS

Abs PBS+BSA

Ems PBS

Ems PBS+BSA

400 500 600 700 800

0,0

0,2

0,4

0,6

0,8

1,0

No

rma

lize

d a

bso

rban

ce,

em

issio

n

Wavelength (nm)

Abs PBS

Abs PBS+BSA

Ems PBS

Ems PBS+BSA

S16







400 500 600 700 800

0,0

0,2

0,4

0,6

0,8

1,0

No

rma

lize

d a

bso

rban

ce,

em

issio

n

Wavelength (nm)

Abs PBS

Abs PBS+BSA

Ems PBS

Ems PBS+BSA

400 500 600 700 800

0,0

0,2

0,4

0,6

0,8

1,0

No

rma

lize

d a

bso

rban

ce,

em

issio

n

Wavelength (nm)

Abs PBS

Abs PBS+BSA

Ems PBS

Ems PBS+BSA

S17

Figure S13. Normalized absorption spectra of 1 and 5 in water and Pr2DAOTA+ BF4- in MeCN. The two latter

spectra appear similar in broadness, whereas 1 appears to be broader, which is also indicated in the less

pronounced shoulder of 1 around 520 nm.

400 450 500 550 600

0,0

0,2

0,4

0,6

0,8

1,0

No

rma

lize

d a

bso

rban

ce

Wavelength (nm)

1 (H2O)

5 (H2O)

Pr2DAOTA+ (MeCN)

S18

Time resolved fluorescence data for 1 to 6

Figure S14. Fluorescence decay profiles of 1 in water (top left), PBS (top right), PBS with 600 µM BSA

(bottom left) and D2O (bottom right).

S19



S20



S21



S22



S23



S24

Determination of quantum yields for 1 to 6

Fluorescence quantum yields were measured in two separate sessions, with each their determination of

the reference. Results are summarized in the tables below. Full spectra can be found on the following

pages.

Table S1. Summarized results of fluorescence quantum yield determination of 1-3. Figures S20-S31.

Dye Solvent Slope R2 Relative slope Φf

Rhodamine 6G Ethanol 2,89E+08 0,99627 1 0,950

1 Water 9,82E+07 0,9996 0,340 0,311

2 Water 8,53E+07 0,99865 0,296 0,270

3 Water 8,78E+07 0,99986 0,304 0,278

Table S2. Summarized results of fluorescence quantum yield determination of 4-6. Figures S32-S43.

Dye Solvent Slope R2 Relative slope Φf

Rhodamine 6G Ethanol 2,85E+08 0,99996 1 0,950

4 Water 1,74E+08 0,99992 0,611 0,558

5 Water 1,75E+08 0,99401 0,615 0,561

6 Water 1,14E+08 0,99703 0,398 0,364

S25

Quantum yield titration of Rhodamine 6G in absolute ethanol (as reference for 1-3)

Figure S20. Absolute (left) and normalized (right) absorption spectra for the quantum yield titration of

Rhodamine 6G in absolute ethanol as a reference for determining quantum yields of 1-3. The vertical

dashed line denotes the excitation wavelength used for measuring the emission spectra presented in figure

S14.

Figure S21. Absolute (left) and normalized (right) emission spectra for the quantum yield titration of

Rhodamine 6G in absolute ethanol as a reference for determining quantum yields of 1-3. Excitation at 500

nm.

400 450 500 550 600

0,00

0,02

0,04

0,06

0,08

Absorb

ance

Wavelength (nm)

R6G_06µL

R6G_12µL

R6G_18µL

R6G_24µL

R6G_30µL

500

400 450 500 550 600

0,0

0,2

0,4

0,6

0,8

1,0

Absorb

ance, norm

aliz

ed

Wavelength (nm)

R6G_06µL

R6G_12µL

R6G_18µL

R6G_24µL

R6G_30µL

550 600 650 700 750 800

0

40000

80000

120000

160000

200000

240000

Em

issio

n (

counts

)

Wavelength (nm)

R6G_06µL

R6G_12µL

R6G_18µL

R6G_24µL

R6G_30µL

550 600 650 700 750 800

0,0

0,2

0,4

0,6

0,8

1,0

Em

issio

n, norm

aliz

ed

Wavelength (nm)

R6G_06µL

R6G_12µL

R6G_18µL

R6G_24µL

R6G_30µL

S26

Figure S22. Linear fit of the integrated emission spectra as a function of sample absorbance for the spectra

of Rhodamine 6G in figures S13 and S14.

0,01 0,02 0,03 0,04 0,05 0,06

2000000

4000000

6000000

8000000

10000000

12000000

14000000

16000000 Power_corr

AU

C

f(A)

R-Square = 0,99627

Intercept = -594310,36719, Slope = 2,88622E8

X Intercept = 0,00206

S27

Quantum yield titration of 1 in water

Figure S23. Absolute (left) and normalized (right) absorption spectra for the quantum yield titration of 1 in

water. The vertical dashed line denotes the excitation wavelength used for measuring the emission spectra

presented in figure S17.

Figure S24. Absolute (left) and normalized (right) emission spectra for the quantum yield titration of 1 in

water. Excitation at 500 nm.

400 450 500 550 600

0,00

0,02

0,04

0,06

Absorb

ance

Wavelength (nm)

NBI-514_06µL

NBI-514_12µL

NBI-514_18µL

NBI-514_24µL

NBI-514_30µL

500

400 450 500 550 600

0,0

0,2

0,4

0,6

0,8

1,0

Absorb

ance, norm

aliz

ed

Wavelength (nm)

NBI-514_06µL

NBI-514_12µL

NBI-514_18µL

NBI-514_24µL

NBI-514_30µL

550 600 650 700 750 800

0

5000

10000

15000

20000

25000

30000

35000

Em

issio

n (

counts

)

Wavelength (nm)

NBI-514_06µL

NBI-514_12µL

NBI-514_18µL

NBI-514_24µL

NBI-514_30µL

550 600 650 700 750 800

0,0

0,2

0,4

0,6

0,8

1,0

Em

issio

n, norm

aliz

ed

Wavelength (nm)

NBI-514_06µL

NBI-514_12µL

NBI-514_18µL

NBI-514_24µL

NBI-514_30µL

S28


of 1 in figures S16 and S17.

0,005 0,010 0,015 0,020 0,025 0,030 0,035

500000

1000000

1500000

2000000

2500000

3000000

3500000 Power_corr

AU

C

f(A)

R-Square = 0,9996

Intercept = 162795,92029, Slope = 9,815E7

X Intercept = -0,00166

S29







400 500 600

0,00

0,02

0,04

0,06

0,08

Ab

so

rban

ce

Wavelength (nm)

NBI-504_20µL

NBI-504_40µL

NBI-504_60µL

NBI-504_80µL

NBI-504_100µL

500

400 500 600

0,0

0,2

0,4

0,6

0,8

1,0

Ab

so

rban

ce,

no

rmaliz

ed

Wavelength (nm)

NBI-504_20µL

NBI-504_40µL

NBI-504_60µL

NBI-504_80µL

NBI-504_100µL

550 600 650 700 750 800

0

10000

20000

30000

40000

50000

Em

issio

n (

counts

)

Wavelength (nm)

NBI-504_20µL

NBI-504_40µL

NBI-504_60µL

NBI-504_80µL

NBI-504_100µL

550 600 650 700 750 800

0,0

0,2

0,4

0,6

0,8

1,0

Em

issio

n, norm

aliz

ed

Wavelength (nm)

NBI-504_20µL

NBI-504_40µL

NBI-504_60µL

NBI-504_80µL

NBI-504_100µL

S30



0,005 0,010 0,015 0,020 0,025 0,030 0,035 0,040

1000000

1500000

2000000

2500000

3000000

3500000

4000000 AUC

AU

C

f(A)

R-Square = 0,99865



S31







400 450 500 550 600

0,00

0,02

0,04

0,06

0,08

Absorb

ance

Wavelength (nm)

NBI-503_06µL

NBI-503_12µL

NBI-503_18µL

NBI-503_24µL

NBI-503_30µL

500

400 450 500 550 600

0,0

0,2

0,4

0,6

0,8

1,0

Absorb

ance, norm

aliz

ed

Wavelength (nm)

NBI-503_06µL

NBI-503_12µL

NBI-503_18µL

NBI-503_24µL

NBI-503_30µL

550 600 650 700 750 800

0

8000

16000

24000

32000

40000

Em

issio

n (

counts

)

Wavelength (nm)

NBI-503_06µL

NBI-503_12µL

NBI-503_18µL

NBI-503_24µL

NBI-503_30µL

550 600 650 700 750 800

0,0

0,2

0,4

0,6

0,8

1,0

Em

issio

n n

orm

aliz

ed

Wavelength (nm)

NBI-503_06µL

NBI-503_12µL

NBI-503_18µL

NBI-503_24µL

NBI-503_30µL

S32



0,010 0,015 0,020 0,025 0,030 0,035 0,040 0,045 0,050

500000

1000000

1500000

2000000

2500000

3000000

3500000

4000000 AUC

AU

C

f(A)

R-Square = 0,99986


X Intercept = 1,92664E-4

S33

Quantum yield titration of Rhodamine 6G in absolute ethanol (as reference for 4-6)

Figure S32. Absolute (left) and normalized (right) absorption spectra for the quantum yield titration of

Rhodamine 6G in absolute ethanol as a reference for determining quantum yields of 4-6. The vertical

dashed line denotes the excitation wavelength used for measuring the emission spectra presented in figure

S26.

Figure S33. Absolute (left) and normalized (right) emission spectra for the quantum yield titration of

Rhodamine 6G in absolute ethanol as a reference for determining quantum yields of 4-6. Excitation at 500

nm.

400 450 500 550 600

0,00

0,02

0,04

0,06

0,08

0,10

Absorb

ance

Wavelength [nm]

R6G_06µL

R6G_12µL

R6G_18µL

R6G_24µL

R6G_30µL

500

400 450 500 550 600

0,0

0,2

0,4

0,6

0,8

1,0

Ab

so

rban

ce,

no

rmaliz

ed

Wavelength [nm]

R6G_06µL

R6G_12µL

R6G_18µL

R6G_24µL

R6G_30µL

550 600 650 700 750 800

0

80000

160000

240000

320000

400000

Em

issio

n (

co

un

ts)

Wavelength [nm]

R6G_06µL

R6G_12µL

R6G_18µL

R6G_24µL

R6G_30µL

550 600 650 700 750 800

0,0

0,2

0,4

0,6

0,8

1,0

Em

issio

n, n

orm

aliz

ed

Wavelength [nm]

R6G_06µL

R6G_12µL

R6G_18µL

R6G_24µL

R6G_30µL

S34


of Rhodamine 6G in figures S25 and S26.

0,01 0,02 0,03 0,04 0,05 0,06 0,07

2000000

4000000

6000000

8000000

10000000

12000000

14000000

16000000

18000000

20000000 AUC

AU

C

f(A)

R-Square = 0,99996


X Intercept = 4,27758E-5

S35







400 450 500 550 600

0,00

0,02

0,04

0,06

0,08

Ab

so

rban

ce

Wavelength [nm]

NBI-327_06µL

NBI-327_12µL

NBI-327_18µL

NBI-327_24µL

NBI-327_30µL

500

400 450 500 550 600

0,0

0,2

0,4

0,6

0,8

1,0

Ab

so

rban

ce,

no

rmaliz

ed

Wavelength [nm]

NBI-327_06µL

NBI-327_12µL

NBI-327_18µL

NBI-327_24µL

NBI-327_30µL

550 600 650 700 750 800

0

20000

40000

60000

80000

100000

120000

140000

160000

Em

issio

n, counts

Wavelength [nm]

NBI-327_06µL

NBI-327_12µL

NBI-327_18µL

NBI-327_24µL

NBI-327_30µL

550 600 650 700 750 800

0,0

0,2

0,4

0,6

0,8

1,0

Em

issio

n, norm

aliz

ed

Wavelength [nm]

NBI-327_06µL

NBI-327_12µL

NBI-327_18µL

NBI-327_24µL

NBI-327_30µL

S36



0,01 0,02 0,03 0,04 0,05 0,06

2000000

4000000

6000000

8000000

10000000

AUC

AU

C

f(A)

R-Square = 0,99992


X Intercept = -4,85366E-4

S37







400 450 500 550 600

0,00

0,02

0,04

0,06

Ab

so

rban

ce

Wavelength [nm]

NBI-507_06µL

NBI-507_12µL

NBI-507_18µL

NBI-507_24µL

NBI-507_30µL

500

400 450 500 550 600

0,0

0,2

0,4

0,6

0,8

1,0

Ab

so

rban

ce,

no

rmaliz

ed

Wavelength [nm]

NBI-507_06µL

NBI-507_12µL

NBI-507_18µL

NBI-507_24µL

NBI-507_30µL

550 600 650 700 750 800

0

20000

40000

60000

80000

100000

Em

issio

n (

counts

)

Wavelength [nm]

NBI-507_06µL

NBI-507_12µL

NBI-507_18µL

NBI-507_24µL

NBI-507_30µL

550 600 650 700 750 800

0,0

0,2

0,4

0,6

0,8

1,0

Em

issio

n, norm

aliz

ed

Wavelength [nm]

NBI-507_06µL

NBI-507_12µL

NBI-507_18µL

NBI-507_24µL

NBI-507_30µL

S38



0,005 0,010 0,015 0,020 0,025 0,030 0,035 0,040

1000000

2000000

3000000

4000000

5000000

6000000

7000000

8000000 AUC

AU

C

f(A)

R-Square = 0,99401


X Intercept = -9,84649E-4

S39







400 450 500 550 600

0,00

0,02

0,04

0,06

0,08

Ab

so

rban

ce

Wavelenght [nm]

NBI-506_08µL

NBI-506_16µL

NBI-506_24µL

NBI-506_32µL

NBI-506_40µL

500

400 450 500 550 600

0,0

0,2

0,4

0,6

0,8

1,0

Ab

so

rban

ce,

no

rmaliz

ed

Wavelength [nm]

NBI-506_08µL

NBI-506_16µL

NBI-506_24µL

NBI-506_32µL

NBI-506_40µL

550 600 650 700 750 800

0

20000

40000

60000

80000

100000

Em

issio

n (

counts

)

Wavelength (nm)

NBI-506_08µL

NBI-506_16µL

NBI-506_24µL

NBI-506_32µL

NBI-506_40µL

550 600 650 700 750 800

0,0

0,2

0,4

0,6

0,8

1,0

Em

issio

n, norm

aliz

ed

Wavelength [nm]

NBI-506_08µL

NBI-506_16µL

NBI-506_24µL

NBI-506_32µL

NBI-506_40µL

S40



0,01 0,02 0,03 0,04 0,05 0,06

1000000

2000000

3000000

4000000

5000000

6000000

7000000 AUC

AU

C

f(A)

R-Square = 0,99703



S41

NMR spectra of 1 to 6 and 8 to 12

Figure S44.1H-NMR of 1 in CD3OD at 500 MHz.

S42

Figure S45.13C-NMR-APT of 1 in CD3OD at 126 MHz. Spectrum is phased to show nuclei with an uneven

number of protons as positive.

Figure S46.19F-NMR of 1 in CD3OD at 470 MHz. The peak at -77.87 is the external reference of neat TFA.

S43

Figure S47. 1H-NMR of 2 in D2O at 500 MHz.

S44

Figure S48.1H-NMR of 3 in D2O at 500 MHz.

Figure S49.13C-NMR-APT of 3 in D2O at 126 MHz. Spectrum is phased to show nuclei with an uneven

number of protons as positive. Peaks at 57.1 (-), 21.8 (-), 17.7 (-) and 0.0 (+) are the internal DSS reference.

S45

Figure S50.1H-NMR of 4 in CD3OD at 500 MHz.

Figure S51.13C-NMR-APT of 4 in CD3OD at 126 MHz. Spectrum is phased to show nuclei with an uneven


S46

Figure S52.19F-NMR of 4 in CD3OD at 470 MHz. The peak at -77.87 is the external reference of neat TFA.

S47




S48

Figure S55.19F-NMR of 5 in D2O at 470 MHz. The peak at -77.87 is the external reference of neat TFA.

S49



S50

Figure S58.1H-NMR of 8 in CD3CN at 500 MHz.

Figure S59.13C-NMR-APT of 8 in CD3CN at 126 MHz. Spectrum is phased to show nuclei with an uneven


S51

Figure S60.19F-NMR of 8 in CD3CN at 470 MHz. The peak at -77.87 is the external reference of neat TFA.

S52




S53


S54




S55


S56




S57


S58



S59

References 1. A. M. Brouwer, Pure and Applied Chemistry, 2011, 83, 2213-2228. 2. C. Wurth, M. Grabolle, J. Pauli, M. Spieles and U. Resch-Genger, Nat Protoc, 2013, 8, 1535-1550. 3. U. Resch-Genger and K. Rurack, Pure and Applied Chemistry, 2013, 85, 2005-2026. 4. J. V. Herraez and R. Belda, J Solution Chem, 2006, 35, 1315-1328. 5. J. C. Martin and R. G. Smith, Journal of the American Chemical Society, 1964, 86, 2252-&. 6. B. W. Laursen and F. C. Krebs, Chem-Eur J, 2001, 7, 1773-1783. 7. M. G. Barlow, M. Green, R. N. Haszeldine and H. G. Higson, Journal of the Chemical Society B:

Physical Organic, 1966, DOI: 10.1039/j29660001025.

download fileview on ChemRxivsupplementary_information.pdf (4.66 MiB)



what is best strategy for water soluble fluorescence dyes

Documents