Biocompatible well-defined chromophore–polymer conjugates for photodynamic therapy and two-photon imaging

Abstract

A versatile approach is introduced for the synthesis of well-defined, biocompatible conjugates combining two-photon chromophores and hydrophilic multifunctional polymers synthesized by RAFT controlled radical polymerization. As an illustration, two different classes of conjugates carrying multiple fluorophores (based on an anthracene moiety, Anth) or photosensitizers (based on a dibromobenzene moiety, DBB) along the polymer chain were elaborated for bioimaging and photodynamic therapy (PDT) applications, respectively. In both cases, the polymer greatly improved the solubility in biorelevant media as well as the cell uptake. Anth conjugates provided high quality fluorescence microscopy images using both one- and two-photon excitation. DBB conjugates potently induced the death of cancer cells upon photoactivation.

---

Introduction

Two-photon absorption (TPA) chromophores have been used in various photonic application fields¹ and have more recently raised increasing interest for bioimaging and therapy applications.^2–4 Further progress in fluorescence imaging and some photodynamic cancer therapy schemes requires improved fluorophores and photosensitizers (PS).⁵ Such improvement could arise from chromophores especially designed to exhibit high TPA.⁵ Indeed, two-photon excitation (TPE) is confined at the focal point of the laser (ca. 1 μm³) and allows high 3D resolution of the excitation that reduces out-of-focus damage and would improve the spatial targeting of tumors. Moreover, TPE is generally performed under near-infrared (NIR) irradiation in a wavelength range of 700–1100 nm where absorption, scattering of light and auto-fluorescence of the biological tissues are low, enabling a better penetration of the incident light and improving the signal-to-noise ratio.

The use of TPA chromophores in vitro or in vivo requires a fine control of their behavior under biorelevant conditions, i.e. solubility, absence of toxicity, concentration at the desired locus of action, and conjugation with bioactive molecules. Unfortunately, efficient TPA chromophores are inherently highly conjugated hydrophobic molecules and thus poorly soluble in aqueous media.^6,7 Therefore, a common strategy relies on their incorporation into biocompatible polymeric carriers. TPA chromophores can be encapsulated into micelles,^8–11 or inside nanoparticles obtained by microemulsion¹² or formulation processes.¹³ However, since chromophore leaking can progressively occur, covalent bonds may be preferred. This can be achieved either by (co)polymerization of chromophore-containing monomers¹⁴ (that should not interfere with the polymerization mechanism, nor be degraded) or by post-modification of functional polymers.¹⁵

In this context, chromophore–polymer conjugates of well-defined compositions are required to provide a reliable and reproducible effect. However, control over the polymer structure and the number of incorporated chromophores is often limited. Then, a strategy consists in grafting chromophores at the periphery of dendrimers,^16–18 leading to nano-sized conjugates that exhibit very high extinction coefficients and cross-section values (generally determined in organic solvents). Nevertheless, dendrimer synthesis is rather complex, and their use is likely to be restricted to very high value purposes.

To overcome most of the abovementioned limitations, our goal was to generate highly efficient TPA bioimaging probes and PDT photosensitizers by developing a straightforward synthesis of chromophore–polymer conjugates exhibiting a well-defined and tunable structure. We decided to use a polymer platform presenting the following characteristics: (i) being water-soluble and biocompatible; (ii) presenting multiple reactive sites for the binding of a controlled number of chromophores per polymer chain to get enhanced TPA cross-section and PS-associated phototoxicity; (iii) featuring a controlled molecular weight (MW), dispersity (Đ), composition and microstructure to provide very homogeneous conjugates; and (iv) easy to synthesize via a versatile polymerization process.

Thus, we directed our choice to a copolymer consisting of water-soluble and biocompatible N-acryloylmorpholine (NAM) units copolymerized with reactive N-acryloxysuccinimide (NAS) units (chromophore binding sites). Our previous studies on the RAFT copolymerization of this monomer pair resulted in an excellent control over the architecture of the P(NAM-co-NAS) chains in terms of MW and dispersity but also composition and microstructure. Not only the microstructure is identical from one chain to another – which is inherent to a living copolymerization – but, for most NAM/NAS molar ratios, the intrachain compositional drift is very low. Moreover, at the so-called azeotropic composition (60/40 molar ratio for this monomer pair), there is no compositional drift, resulting in a very regular spacing of the NAS units along the polymer chains.¹⁹

Novel TPA chromophore–polymer conjugates were elaborated by a combination of those multifunctional copolymers with two different TPA chromophores (based on an anthracene (Anth) and a dibromobenzene (DBB) moiety) designed for two-photon imaging and photodynamic therapy, respectively, that had recently given encouraging results (after encapsulation in Pluronic).¹¹ For the present study, these chromophores were functionalized in order to carry (i) two short PEG chains to improve compatibility with aqueous media and (ii) a primary amino group for the covalent binding to the hydrophilic P(NAM-co-NAS) copolymer (Scheme 1). To the best of our knowledge, the resulting conjugates constitute the first example of TPA multichromophoric probes based on a polymer chain synthesized by controlled radical polymerization for both two-photon bio-imaging with excitation in the NIR regime and PDT.

Scheme 1 (A) Structures of the two TPA chromophores and the P(NAM-co-NAS) random copolymer used in this study. (B) General structure of the chromophore–polymer conjugates with the NAM units (blue circles), the NAS units after binding of the TPA chromophores (red circles), the hydrolyzed NAS units (green circles), the Anth or DBB chromophoric centers (red rectangles) and the oligo-ethyleneglycol side-chains (yellow triangles).

Herein, we report on their synthesis and spectroscopic characterization in various solvents, including water. In addition, we present a thorough biological evaluation of these conjugates in cellulo that demonstrates their ability to serve as TPA bioimaging probes (Anth conjugate) and PDT photosensitizers (DBB conjugate), and highlights the role of the polymer in their behavior and efficiency under biorelevant conditions.

Results and discussion

Synthesis of TPA chromophore–polymer conjugates

The chromophore–polymer conjugates were synthesized in high yield (80–90%) from the same copolymer backbone (M_n = 12 900 g mol⁻¹; Đ = 1.05) and the amino-functionalized chromophores (Scheme 1). After hydrolysis of the remaining activated ester functions and thorough purification, two anionic (carboxylated) conjugates were obtained with an average of 6.1 Anth (6Anth-H) and 4.4 DBB (4DBB-H) per polymer chain, respectively (Table 1).

Table 1 Physico-chemical characteristics of TPA chromophore–polymer conjugates^a

Conjugates	Y (%)	n c	n COO−	M n
a Y: coupling yield (%); nc: average number of chromophores per polymer chain determined by SEC/UV (nc = Y × n0 chromophores/n0 polymer chains, with n0 being the initial mole number of chromophores and polymer chains that were reacted). The error associated with nc, determined through this method was lower than 10%; nCOO−: average number of carboxylate charges per polymer chain (33.9 − nc); Mn: number average molecular weight of the chromophore–polymer conjugates (g mol^−1) calculated assuming that the conjugates were in their sodium carboxylate form after dialysis and lyophilization.
6Anth-H	90	6.1	27.8	14 800
4DBB-H	81	4.4	29.5	13 800

---

Spectroscopic characterization

In contrast to the free chromophores that were only soluble in organic solvents, all conjugates were soluble in aqueous media as well as in polar organic solvents. In all cases, neither macroscopic nor microscopic aggregation (using dynamic light scattering, DLS) was observed. Thanks to the very good water-solubility of the chromophore–polymer conjugates, their spectroscopic parameters could be determined in water (Table 2). Measurements were also performed in dioxane, a low polarity solvent known to exhibit physical characteristics close to biological micro-environments.²⁰ Moreover, dioxane enabled a comparison between the conjugates and the free chromophores.

Table 2 Spectroscopic parameters of the TPA chromophore–polymer conjugates^a

	Solvent	λ max (abs; em)	ε	ϕ
a λ max (abs; em): maximal absorption and emission wavelengths (nm); ε: molar extinction coefficient (cm^−1 M^−1); ϕ: fluorescence quantum yield. b Anth chromophore was analyzed in its amine-protected form.
Anth ^b	CHCl3	490; 557	33 000	0.57
	Dioxane	484; 548	36 000	0.61
6Anth-H	Dioxane	476; 554	108 000	0.08
	Water	476; 609	100 000	0.01
DBB	CHCl3	400; 460	58 000	0.23
	Dioxane	394; 448	60 000	0.18
4DBB-H	Dioxane	389; 460	202 000	0.10
	Water	420; 505	177 000	0.01

---

Absorption spectra of the conjugates were similar to those of the free chromophores, except for a slight broadening of the absorption bands in water (thus a concomitant decrease of the extinction coefficient). Yet, the maximum absorption wavelength remained almost unchanged (ESI, Fig. S2†). In contrast, the fluorescence emission of both conjugates was clearly red-shifted (by about 50 nm) in water compared to dioxane (increasing solvent polarity) resulting in larger Stokes shifts. Their fluorescence quantum yield (ϕ) was found to decrease significantly as the protic character of the solvent increased (ESI, Table S2†). Thus, ϕ values were much higher in dioxane than in water for both conjugates. This indicated that the nature of the solvent critically impacts the conformation of the conjugates and the probability of fluorescence self-quenching (due to chromophore–chromophore interactions) which is indeed a well-known phenomenon associated with a high local chromophore concentration.²¹ TPA cross-section (σ_TPA) values of 420 and 400 GM were determined in chloroform for Anth and DBB free chromophores, respectively. Concerning the conjugates, σ_TPA (as well as the molar extinction coefficient, ε, Table 2) could be measured in water which was not possible for the free chromophores. Values were as high as 790 GM (6Anth-H) and 1380 GM (4DBB-H) at 740 nm, corresponding to 130 and 310 GM per bound chromophore, respectively (lower σ_TPA values are generally observed in aqueous media compared to organic solvents²²). In summary, besides providing water-solubility, multifunctional hydrophilic polymers led to conjugates with high ε and σ_TPA values in water.

Biological evaluation

Biological evaluation of the conjugates was then conducted in cellulo. The objectives were to address the following questions: (i) Are the conjugates non-toxic in the absence of photoactivation? (ii) Is their cellular internalization favored compared to free chromophores? (iii) Can 6Anth-H conjugate be used as a probe for fluorescence microscopy under one- and two-photon excitation? (iv) Can 4DBB-H conjugate induce death of cancer cells upon photoactivation?

Two photon excited fluorescence microscopy using 6Anth-H probe

To assess the cytotoxicity and the uptake of 6Anth-H, we first used Baf3 cells since they are very sensitive to a wide spectrum of cell death inducing treatments.²³ Baf3 cells were incubated for various lengths of time with 6Anth-H (10⁻⁵ mol L⁻¹ chromophore) and their fluorescence was recorded by flow cytometry (ESI, Fig. S3†). At this concentration, conjugates were not cytotoxic over a 24-hour period (ESI, Fig. S4†). Increase of fluorescence intensity over time was indicative of a continuous uptake of 6Anth-H (ESI, Fig. S5†). To test whether cellular uptake of 6Anth-H could be detected by one- and two-photon microscopy, we used a B16-F10 melanoma cell line (adherent cells, easily examined by microscopy). Upon one-photon excitation at 488 nm, 6Anth-H emission was detected throughout the cytoplasm of the cells (not in the nucleus) and concentrated in the perinuclear region (Fig. 1A, top panels). 6Anth-H was also successfully visualized upon two-photon excitation with a maximal signal and a minimal auto-fluorescence at 750 nm excitation (NIR) (Fig. 1A, bottom panels).

Fig. 1 (A) One-photon (top panels) and two-photon (bottom panels) fluorescence images of 6Anth-H conjugate. Adherent B16-F10 melanoma cells were cultured for 24 h in the absence (left panel) or in the presence of 6Anth-H (right panels, 10⁻⁵ mol L⁻¹ chromophore). Plasma membrane (red) and nuclei (blue) were visualized in all cases after one-photon excitation using anti-CD44 antibody (405 nm) and DRAQ5 DNA dye (633 nm) respectively. 6Anth-H (green) was detected using one-photon (488 nm) or two-photon excitation (750 nm). Scale bar is 10 μm. (B) (Top graph) Fluorescence emission spectrum measured in cellulo after two-photon excitation at 750 nm in the absence or in the presence of 6Anth-H (image acquisition at multiple wavelengths). (Bottom graph) Spectrum of the same conjugate in dioxane and pure water (fluorescence spectroscopy).

The results obtained by fluorescence microscopy clearly demonstrate the potential of 6Anth-H conjugate to be used as a fluorescent probe. From a general perspective, ionic strength, pH and polarity can vary from one micro-environment to another. Important solvatochromism effects and ϕ variation may be observed depending on the localization of the probe inside the cell. As an illustration, we compared the 6Anth-H fluorescence emission spectrum in cellulo with the spectra recorded in pure water and in dioxane (Fig. 1B). In the absence of 6Anth-H, background auto-fluorescence was negligible. In the presence of 6Anth-H, the fluorescence emission spectrum in cellulo exhibited a maximum close to 550–560 nm and a very good signal-to-noise ratio. This emission spectrum differed from the one in pure water (maximum significantly blue-shifted compared to 609 nm in water), confirming that the conjugate experienced a different environment. Conversely, the observed emission profile exhibited a maximum close to that of the conjugate in dioxane solution, reflecting that the conjugates were mainly in interaction with low polar regions of the cells (e.g. lipid membranes, proteins). The fluorescence spectrum in cellulo together with higher fluorescence quantum yield of 6Anth-H in dioxane than in water probably explains the very good quality of the images recorded upon one- and two-photon excitation.

Photo-induced cell death using 4DBB-H photosensitizer

DBB derivatives are potential two-photon photosensitizers that were previously shown to induce cell death upon one-photon photoactivation.¹¹ However, internalization into cells was limited when the free PS was solubilized in DMSO (due to its hydrophobic nature).¹¹ Interestingly, when bound onto the hydrophilic polymer (4DBB-H conjugate), cellular uptake was greatly enhanced (Fig. 2A). Compared with PS encapsulated in Pluronic,¹¹ cellular uptake was faster and more efficient.

Fig. 2 (A) Kinetics of 4DBB-H uptake. Baf3 cells were incubated without any PS (black line) or with the free PS in DMSO (red line, 10⁻⁵ mol L⁻¹) or with 4DBB-H conjugate containing the equivalent of 10⁻⁵ mol L⁻¹DBB (blue line) for the indicated period of time. Uptake was evaluated by flow cytometry. Results are expressed as the average of the mean fluorescence intensity ± SD of 2 independent experiments. (B) Induction of cell death upon photoactivation. Baf3 cells and B16-F10 melanoma cells were incubated for 24 h without any PS (black) or with increasing amount of free PS in DMSO (red) or 4DBB-H conjugate (blue) (4 × 10⁻⁷, 2 × 10⁻⁶ and 10⁻⁵ mol L⁻¹DBB). Then, cells were submitted (+) or not (−) to irradiation (365 nm, 8 J cm⁻²). The percentage of cell mortality was assessed 5 h after irradiation by staining with propidium iodide followed by flow cytometry analysis. Results show the mean mortality ± SD of 2 to 4 independent experiments.

In the absence of light irradiation, 4DBB-H was not toxic. Hence, Baf3 or B16-F10 cells did not show signs of increased mortality when incubated with 4DBB-H for 24 h (Fig. 2B) or 48 h (data not shown). In contrast, irradiation of cells previously incubated for 24 h with 4DBB-H rapidly induced cell death in a dose-dependent manner. Induction of cell death upon photoactivation was very efficient for concentration as low as 2 × 10⁻⁶ mol L⁻¹DBB, especially in the case of B16-F10 melanoma cells. In contrast, the free PS in DMSO did not induce significant Baf3 cell death (toxicity at high concentration being mainly due to DMSO rather than photoactivation) and only induced B16-F10 cell death at the highest concentration (10⁻⁵ mol L⁻¹) upon irradiation.

Altogether, this comparison between 4DBB-H conjugate and free PS showed compelling evidence that the copolymer strongly enhances the photosensitizer cell uptake and its ability to induce cell death. It has to be stressed that such comparison was possible thanks to the controlled number of PS bound onto the conjugate. This is not always feasible for other systems. These encouraging results will lead to further investigations, particularly regarding cell death induction under two-photon irradiation.

Conclusions

In conclusion, a versatile and controlled approach for the synthesis of well-defined TPA chromophore–polymer conjugates is described. Our results demonstrate that highly water-soluble polymer conjugates bearing multiple chromophores along the chain could be elaborated and provided valuable imaging probes for one- and two-photon excitation fluorescence microscopy as well as novel photosensitizers with significantly improved efficiency for photodynamic cancer therapy. In particular, cellular uptake was considerably increased when chromophores were covalently bound onto the polymer. We thus believe that this modular system could be an attractive candidate for numerous applications in biology and medicine ranging from diagnostics to therapeutics. We are currently working on extending the strategy to other systems thanks to the ability to finely tune the nature of the polymer and the chromophores as well as the architecture and the functionality of the conjugates.

Experimental section

Materials

Detailed synthesis of TPA chromophores is described in the ESI.† Synthesis of the free PS, solubilized in DMSO for the phototoxicity assays, was previously described in the literature.¹¹ All chemicals and solvents were purchased from Sigma Aldrich or Acros at the highest purity available and used without further purification.

Analytical methods

SEC/UV. Coupling yields of the reaction between P(NAM-co-NAS) and chromophores were determined (before hydrolysis) by size exclusion chromatography (SEC) using UV detection at 510 and 400 nm for Anth and DBB conjugates, respectively. The SEC apparatus consisted of a Varian ProStar Dynamax 800 system fitted with a Styragel HR4E linear column (7.8 × 300 mm²). Eluent was dimethylformamide (DMF) with LiBr (0.05 mol L⁻¹). Samples were prepared by diluting 50 μL of reaction medium in 1 mL of eluent and injected via a 30 μL sample loop. Data acquisition and treatment was performed using Galaxie software.

UV-Visible absorption. Spectra were recorded on a Jasco V-670 spectrophotometer at ambient temperature using 1 cm quartz cells. Extinction coefficients ε were calculated using Beer–Lambert law at 470 nm and 390 nm for Anth and DBB compounds, respectively.

Fluorescence spectroscopy. Fluorescence emission spectra were recorded on a Horiba-Jobin Yvon Fluorolog-3 spectrofluorometer, at ambient temperature using a 1 cm quartz cell. Excitation wavelengths were 470 nm and 390 nm for Anth- and DBB-related compounds, respectively.
Fluorescence quantum yield. Fluorescence quantum yields, ϕ, were calculated according to eqn (1) for diluted solutions having an optical density lower than 0.1, where A is the absorbance at the excitation wavelength λ, n is the refractive index and D is the integrated luminescence intensity. “r” and “x” stand for reference and sample. References were rubrene (ϕ = 0.27 in methanol) and Coumarin 153 (ϕ = 0.45 in methanol) for Anth and DBB compounds, respectively.

Two-Photon Absorption (TPA) spectroscopy. Two-Photon Absorption. (TPA) cross-section spectra were obtained between 700 and 950 nm by up-conversion fluorescence using a Ti:sapphire femtosecond laser (Coherent Verdi-V5 pump coupled with a Femtosecond MIRA 900 cavity) with an intensity power fixed at 50 mW and a S2000 Ocean Optics Spectrometer. The excitation beam was collimated over the cell (10 mm) with a pulse frequency of 80 MHz. Incident beam intensity was adjusted to 50 mW in order to ensure an intensity-squared dependence of the fluorescence over the whole spectral range. Detector integration time was set at 1 s. Measurements were performed at room temperature in chloroform or water and at a concentration of 10⁻⁴ mol L⁻¹ in chromophore. Calibration was performed by comparison with the published two-photon absorption spectrum of Coumarin-307 (ϕ = 0.56 in ethanol) between 700 and 900 nm.²⁴

I^TwoPhoton_Fl ∝ σ_TPA × I_laser² × ϕ × c × K_laser (2)

TPA cross-section. TPA cross-section values (σ_TPA) were obtained using eqn (2), where I_Fl represents the two-photon fluorescence intensity; I_laser the LASER intensity at 50 mW, K_laser a constant depending on the measurement set-up; ϕ the fluorescence quantum yield and c the concentration of chromophore (c = 10⁻⁴ mol L⁻¹). The measurements were conducted in an intensity regime where the fluorescence signal showed a quadratic dependence on the intensity of the excitation beam, as checked for all samples. The uncertainty in the measured cross-sections is about 15% in CHCl₃ and 20% in water.

Synthesis of reactive copolymers using RAFT polymerization

Poly(N-acryloylmorpholine-co-N-acryloxysuccinimide), P(NAM-co-NAS), with a 60/40 NAM/NAS molar ratio (M_n = 12 900 g mol⁻¹; Đ = 1.05; average number of activated ester units per chain = 33.9) (Scheme 1) was synthesized according to the previously reported RAFT polymerization procedure.¹⁹

TPA chromophore–polymer conjugates synthesis

Coupling of Anth onto the copolymer. 50 mg of P(NAM-co-NAS) copolymer were dissolved in 0.5 mL of chloroform (7.75 mM) in a 25 mL round bottom flask equipped with a magnetic stirrer. 20.3 mg of Anth in chloroform (55.2 mM, raw from NH-tBoc deprotection (ESI,† compound 2)) were then added together with 2 equivalents of triethylamine (TEA). Polymer concentration was then adjusted with chloroform at 10 mg mL⁻¹. Coupling reaction was carried out at 40 °C in the dark under stirring for 48 hours.

After coupling, the reaction medium was analyzed by SEC/UV in order to determine the coupling yield (Table 1). The conjugate was then precipitated in a large volume of diethyl ether and isolated from the supernatant by centrifugation. The procedure was repeated until complete discoloration of the supernatant, ensuring the absence of free unreacted chromophore. Purified orange conjugates were finally dried under vacuum up to constant weight.

Coupling of DBB onto the copolymer. The same procedure was conducted with 40 mg of P(NAM-co-NAS) copolymer and 14.3 mg of DBB to obtain the corresponding yellow-colored conjugate.

Hydrolysis of the remaining activated ester functions. After coupling of the chromophores onto the polymer backbone, conjugates were redispersed in 45 mL of a borate buffer (50 mM, pH = 9). Hydrolysis of the unreacted NAS activated ester groups into carboxylate units proceeded at room temperature under magnetic stirring for 3 days. After hydrolysis, conjugates were purified by dialysis against deionized water and dried by lyophilization.

Determination of the average number of chromophores per polymer chain, n_c

Coupling yields, Y, and the average number of chromophores per polymer chain, n_c, were determined before hydrolysis from SEC/UV chromatograms. Using UV detection (at the absorption maximum of the chromophores), two different peaks could be distinguished (Fig. 3): (A) chromophores bound to the copolymer chains (lower elution volume), (B) free unbound chromophores. A and B being the areas under each of these peaks, the coupling yield (Y) was determined using eqn (3) assuming that chromophore absorption properties remained unchanged upon coupling onto the copolymer.

Y = A/(A + B) (3)

Fig. 3 Example of a SEC/UV chromatogram of a sample withdrawn from the reaction medium during coupling of the chromophore onto the P(NAM-co-NAS) copolymer.

The latter assumption was validated by determination of the extinction coefficients (ε) in DMF (SEC eluent) of (i) the free chromophores and (ii) the chromophores bound onto the polymer. Similar values were obtained within 10% error. For DBB compounds in DMF at λ = 400 nm, ε_DBB = 55 000 cm⁻¹ M⁻¹ and ε_4DBB-H = 224 000 cm⁻¹ M⁻¹, i.e. ε = 51 000 cm⁻¹ M⁻¹ per bound chromophore (4.4 chromophores per chain).

n _c values were further confirmed from UV absorption spectra of the final conjugates by comparison of the total integrated absorption strength for the free chromophore and the corresponding conjugate. Using this alternative method, n_c values were 6.3 and 4.4 for Anth and DBB conjugates respectively (compared to 6.1 and 4.4 from the SEC/UV method). This determination was associated with a 10% error.

Biological evaluation

Cell cultures. Murine IL-3 dependent Baf3 cells and B16-F10 melanoma cells were cultivated in DMEM medium (Invitrogen Life Technologies) supplemented with 6% heat-inactivated Fetal Bovin Serum (Lonza, Belgium) and 10 mg L⁻¹ gentamicin (Invitrogen). Baf3 cells are non-adherent cells and were grown in suspension with 5% WEHI 3B cell-conditioned medium as a source of IL-3. B16-F10 cells are adherent cells and were grown as a sub-confluent monolayer. Cells were cultured at 37 °C and 7% CO₂ in a humidified atmosphere.

Flow cytometry. To measure the chromophore uptake, 2 × 10⁵ Baf3 cells mL⁻¹ were incubated at 37 °C with 6Anth-H, 4DBB-H, or free PS in DMSO (10⁻⁵ mol L⁻¹ chromophore) for the indicated period of time. Cells were analyzed using a LSR or a Canto II flow cytometer (Becton Dickinson Bioscience) with a 488 nm laser and a 525 nm emission filter or a 405 nm laser and a 450 nm emission filter to detect incorporation of Anth and DBB derivatives, respectively. At least 5000 cells per condition were recorded and dot plots were analyzed using FlowJo Software (Tree Star, Inc., USA) (ESI, Fig. S3†).

To measure cell mortality, cells were collected and incubated with propidium iodide (PI, Sigma, 2 mg L⁻¹) a standard flow cytometry viability probe and at least 5000 cells were analyzed immediately by flow cytometry. This allows us to distinguish between viable cells with the intact plasma membrane that exclude PI and non-viable cells that are permeable to PI. Baf3 cells are non-adherent and were collected by pipetting. Adherent cells loosen their attachment or detach from the substratum during apoptosis. Therefore cell mortality among B16-F10 cells was analyzed by pelleting floating cells and adherent cells harvested by trypsinization. Cells were then stained with PI and analyzed by flow cytometry.

Phototoxicity. To assay the phototoxicity of 4DBB-H in cell culture systems, 2 × 10⁵ Baf3 cells mL⁻¹ were seeded into 96-well plates and cultured for 24 h in the presence of the free PS in DMSO or 4DBB-H at the indicated concentration. B16-F10 were seeded at 6 × 10⁴ cells mL⁻¹ into 24-well plates, grown overnight and incubated for 24 h with the PS. Cells were irradiated at 8 J cm⁻² (Baf3 cells) or 16 J cm⁻² (B16-F10 cells) at 365 nm (Bio-Sun, Vilber Loumart, Marne-la-vallée, France). Cell viability was assessed 5 h later by flow cytometry.

Cell fluorescence imaging. B16-F10 cells were plated on glass coverslips (2 × 10⁴ cells per coverslip) and allowed to adhere overnight. Cells were then incubated for 24 h with 6Anth-H (10⁻⁵ mol L⁻¹Anth). DRAQ5 (10 μM, Alexis Biochemicals) was added for the last 3 minutes of culture to stain the nuclei. Cells were washed with PBS and fixed with 4% paraformaldehyde (Sigma-Aldrich). The cell surface was stained using anti-CD44-eFluor450 (2 μg mL⁻¹, Ebiosciences) in PBS for 2 h at room temperature. Specimens were mounted (Antifade Kit, Molecular Probes) and examined using an inverted LSM 710 Zeiss confocal microscope (Carl Zeiss, Germany) with a 63×/1.4 objective. Zeiss 710 uses a variable spectral detection system. 405 nm and 633 nm lasers were used for anti-CD44-eFluor450 and DRAQ5 excitation respectively. Imaging of anthracene conjugate was performed using one-photon excitation (488 nm) or two-photon excitation (750 nm) with tunable infrared laser Ti:Sapphire Chameleon Ultra II (690 nm–1040 nm). Images were processed with ImageJ software, a public domain image processing and analysis program.

Acknowledgements

We acknowledge the contribution of the imaging facility (PLATIM) and the cytometry platforms (Biosciences Gerland – Lyon Sud, UMS3444, US8), Jean Bernard for technical support in two-photon absorption measurements and Christophe Place for fruitful discussion about this manuscript. This work was supported by grants from INSERM, the ARC, the Ligue du Rhône, the Région Rhône-Alpes, Université Lyon 1. We also thank the ANR for PHL post-doctoral grant and financial support (NanoPDT, no. ANR-09-NANO-027) as well as LyonBiopole. CC acknowledges a PhD grant from the French Ministry of Research and Education.

Notes and references

1. G. S. He, L.-S. Tan, Q. Zheng and P. N. Prasad, Chem. Rev., 2008, 108, 1245.
2. W. R. Zipfel, R. M. Williams and W. W. Webb, Nat. Biotechnol., 2003, 21, 1369.
3. A. Diaspro, G. Chirico and M. Collini, Q. Rev. Biophys., 2005, 38, 97.
4. S. Brown, Nat. Photonics, 2008, 2, 394.
5. J. P. Celli, B. Q. Spring, I. Rizvi, C. L. Evans, K. S. Samkoe, S. Verma, B. W. Pogue and T. Hasan, Chem. Rev., 2010, 110, 2795.
6. M. Pawlicki, H. A. Collins, R. G. Denning and H. L. Anderson, Angew. Chem., Int. Ed., 2009, 48, 3244.
7. M. Rumi, J. E. Ehrlich, A. A. Heikal, J. W. Perry, S. Barlow, Z. Hu, D. McCord-Maughon, T. C. Parker, H. Röckel, S. Thayumanavan, S. R. Marder, D. Beljonne and J.-L. Brédas, J. Am. Chem. Soc., 2010, 122, 9500.
8. Q. Zheng, T. Y. Ohulchanskyy, Y. Sahoo and P. N. Prasad, J. Phys. Chem. C, 2007, 111, 16846.
9. M. Maurin, L. Vurth, J.-C. Vial, P. L. Baldeck, S. R. Marder, B. Van der Sanden and O. Stéphan, Nanotechnology, 2009, 20, 235102.
10. C. D. Andrade, C. O. Yanez, M. A. Qaddoura, X. Wang, C. L. Arnett, S. A. Coombs, J. Yu, R. Bassiouni, M. V. Bondar and K. D. Belfield, J. Fluoresc., 2011, 21, 1223.
11. T. Gallavardin, M. Maurin, S. Marotte, T. Simon, A.-M. Gabudean, Y. Bretonnière, M. Lindgren, F. Lerouge, P. L. Baldeck, O. Stéphan, Y. Leverrier, J. Marvel, S. Parola, O. Maury and C. Andraud, Photochem. Photobiol. Sci., 2011, 10, 1216.
12. D. Gao, R. R. Agayan, H. Xu, M. A. Philbert and R. Kopelman, Nano Lett., 2006, 6, 2383.
13. X. Shen, F. He, J. Wu, G. Q. Xu, S. Q. Yao and Q.-H. Xu, Langmuir, 2011, 27, 1739.
14. J. Pecher, J. Huber, M. Winterhalder, A. Zumbusch and S. Mecking, Biomacromolecules, 2010, 11, 2776.
15. S. Biswas, X. Wang, A. R. Morales, H.-Y. Ahn and K. D. Belfield, Biomacromolecules, 2011, 12, 441.
16. A. Adronov, J. M. J. Fréchet, G. S. He, K.-S. Kim, S.-J. Chung, J. Swiatkiewicz and P. N. Prasad, Chem. Mater., 2000, 12, 2838.
17. M. A. Oar, J. M. Serin, W. R. Dichtel, J. M. J. Fréchet, T. Y. Ohulchanskyy and P. N. Prasad, Chem. Mater., 2005, 17, 2267.
18. O. Mongin, R. K. Thatavarathy, M. H. V. Werts, A.-M. Caminade, J.-P. Majoral and M. Blanchard-Desce, Chem. Commun., 2006, 915.
19. A. Favier, F. D'Agosto, M.-T. Charreyre and C. Pichot, Polymer, 2004, 45, 7821.
20. N. I. Zahid, O. K. Abou-Zied, R. Hashim and T. Heidelberg, J. Phys. Chem. C, 2011, 115, 19805.
21. J. Baumann and M. D. Fayer, J. Chem. Phys., 1986, 85, 4087.
22. H. Y. Woo, B. Liu, B. Kohler, D. Korystov, A. Mikhailovsky and G. C. Bazan, J. Am. Chem. Soc., 2005, 127, 14721.
23. J. Thomas, Y. Leverrier and J. Marvel, Oncogene, 1998, 16, 1399.
24. C. Xu and W. W. Webb, J. Opt. Soc. Am. B, 1996, 13, 481.

---

Footnotes

† Electronic supplementary information (ESI) available: Detailed synthesis of both TPA chromophores, complementary spectroscopic data, flow cytometry analyses, kinetics of cell uptake and cytotoxicity evaluation for 6Anth-H. See DOI: 10.1039/c2py20565c

‡ These authors contributed equally.

---