Two-Photon Macromolecular Probe Based on a Quadrupolar Anthracenyl Scaffold for Sensitive Recognition of Serum Proteins under Simulated Physiological Conditions.

The binding interaction of a biocompatible water-soluble polycationic two-photon fluorophore (Ant-PIm) toward human serum albumin (HSA) was thoroughly investigated under simulated physiological conditions using a combination of steady-state, time-resolved, and two-photon excited fluorescence techniques. The emission properties of both Ant-PIm and the fluorescent amino acid residues in HSA undergo remarkable changes upon complexation allowing the thermodynamic profile associated with Ant-PIm-HSA complexation to be accurately established. The marked increase in Ant-PIm fluorescence intensity and quantum yield in the proteinous environment seems to be the outcome of the attenuation of radiationless decay pathways resulting from motional restriction imposed on the fluorophore. Fluorescence resonance energy transfer and site-marker competitive experiments provide conclusive evidence that the binding of Ant-PIm preferentially occurs within the subdomain IIA. The pronounced hypsochromic effect and increased fluorescence enhancement upon association with HSA, compared to that of bovine serum albumin (BSA) and other biological interferents, makes the polymeric Ant-PIm probe a valuable sensing agent in rather complex biological environments, allowing facile discrimination between the closely related HSA and BSA. Furthermore, the strong two-photon absorption (TPA) with a maximum located at 820 nm along with a TPA cross section σ2 > 800 GM, and the marked changes in the position and intensity of the band upon complexation definitely make Ant-PIm a promising probe for two-photon excited fluorescence-based discrimination of HSA from BSA.

Introduction

The archetypical human serum albumin (HSA) is the main extracellular protein of the circulatory system.[1−3] In particular, the ability of HSA to interact with a wide range of endogenous metabolites and exogenous drugs, which may modify their pharmacokinetic and pharmacodynamic properties and influence their distribution and availability toward the biological target, is well known.[4−6]

X-ray crystallographic analysis of HSA has revealed that this globular protein, a 585 amino acid residue monomer, consists of three homologous α-helical domains (I–III), each of which is subdivided into two subdomains A and B.[7] Competitive studies and crystal structure analysis enabled the identification of two specific ligand binding sites within the hydrophobic cavities of the protein template named either subdomains IIA and IIIA or Sudlow’s site I and II, respectively, in which molecules acting as HSA probes bind to this biomolecule with association constants ranging from 103 to 106 M–1.[8−10] It is known that either a low (hypoproteinemia) or high level (microalbuminaria) of HSA in the blood plasma is a characteristic feature of physical health issues such as cirrhosis, chronic hepatitis, diabetes, and hypertension.[11,12] In this context, the selectivity of a probe with respect to binding with HSA or other relevant reactive biological interferents (bovine serum albumin (BSA), myoglobin, lysozyme, chymotrypsin, chymotrypsinogen A, l-cysteine, l-glutathione, and l-arginine) is of great interest for laboratory biomedical analyses. In this article, we were particularly interested in controlling the HSA selectivity over BSA. BSA, which demonstrates 70% of the biological functions of HSA, is widely used as a HSA replacement due to its lower cost in many biochemical and pharmacological assays;[13−16] the ability to discriminate between these two similar proteins is therefore a challenge of great importance. Although several fluorescent probes for serum albumin detection have been reported, most of them have shown poor selectivity for HSA over BSA and their detection limits were found to be above 30 mg/L.[17−28] Additionally, their excitation and emission wavelengths often lie in the UV or visible region of the electromagnetic spectrum, which does not allow easy discrimination from biological autofluorescence. Thus, red emitting probes that can be excited by two-photon absorption (TPA) in the near infrared (NIR) range would constitute a significant improvement over existing systems.

Indeed, the development of new, bright, two-photon fluorescence probes, possessing a relatively high TPA cross section in the NIR region, has received much attention during the last decade due to their promising applications in both laboratory biological imaging and clinical diagnosis.[29−35] Advantages such as larger penetration depth of the exciting light, low tissue autofluorescence background, reduced photodamage, and photobleaching make two-photon based technologies an advantageous substitute over their one-photon counterparts.[29−35]

We recently developed a polymer engineering strategy for water-soluble polymer NIR probes with large two-photon excited fluorescence brightness for cellular and intravital two-photon laser scanning microscopy (TPLSM) imaging.[36−38] From a chemist’s point of view, to meet the criteria needed for the development of suitable nonlinear probes useful for in vitro or in vivo cell imaging in tissues, detection of chemical analytes, and monitoring of biological processes with excellent spatial and temporal resolution, requires to address several pivotal points, which were central in our rational molecular engineering approach:[38−40]

The molecule should display a large TPA cross section and significant fluorescence quantum yield in the biological transparency window (BTW: 680–1300 nm);

The molecule should possess good two-photon brightness (φ × σ2), in the far-red or NIR regions, to be implemented as a microscopic tool in TPLSM procedures;

The molecule must be water soluble (the majority of nonlinear probes are not soluble in water media due to the extended aromatic surfaces needed to enhance the σ2 values), biocompatible, and furtive to the immune system to ensure a relatively long residence time within the organism that is compatible with imaging of biological processes;

The molecule must be nontoxic, affordable at a large scale, and exert good selectivity toward biological targets/compartments.

An overwhelming majority of the aforementioned points were addressed for the neutral macromolecular probes Ant-PHEA(36) and the quasi-quadrupolar Ant2-PHEA,[38] which are two of our best-suited reported probes in this regard.

We recently found that Ant-PHEA and its analogous cationic (Ant-PIm) anthracene polymeric probe, designed using this approach, presented relatively high affinity toward double-stranded DNA, even if the latter seemed to be particularly sensitive to the surroundings.[41,42] This overall framework prompted us to investigate the affinity of Ant-PIm (Figure ), whose synthesis has been reported elsewhere,[42] toward HSA and its selectivity over BSA.

Figure 1

Chemical structure of Ant-PIm (n = 4 and 5).

In this contribution, we report the characterization of the HSA–Ant-PIm binding mode using a combination of steady-state, time-resolved, and two-photon excited fluorescence techniques. In particular, we provide detailed insights into the affinity constant, binding sites, intermolecular distance, nonlinear optical parameters, and secondary structure changes determining the Ant-PIm–HSA association mechanism. Furthermore, the ability of Ant-PIm to selectively recognize and discriminate HSA from BSA and common biological interferents is also described, highlighting the outstanding properties of the probe to suitably detect HSA in rather complex biological environments. This is the first report in which a water-soluble nonlinear probe, which can be excited by TPA within the BTW, is used to selectively detect and discriminate serum proteins in rather complex biological environments under simulated physiological conditions.

Results and Discussion

Effect of Ant-PIm on HSA Fluorescence Spectra

HSA contains a single tryptophan, Trp-214, and 18 tyrosine (Tyr) residues, which are responsible for its intrinsic fluorescence.[43] Trp-214 is located in subdomain IIA within a hydrophobic pocket, whereas the Tyrs are distributed along the whole peptide chain.[44] Upon excitation at 280 nm, both the tryptophan and Tyr residues are readily excited but most of the fluorescence arises from Trp-214 due to the resonance energy transfer from Tyr to tryptophan.[43] However, the excitation wavelength of 293 nm allows only the tryptophan residue to emit fluorescence.[43] If Ant-PIm interacts with HSA, the fluorescence properties of HSA may be modified depending on the proximity of the ligand to the intrinsic protein fluorophores (Trp and/or Tyr). To determine whether both the tryptophan and Tyr residues are involved in the Ant-PIm–HSA association process the protein fluorescence emission spectra were recorded in the absence and in the presence of the anthracenyl derivative using excitation wavelengths of 280 and 293 nm (Figure ).

Figure 2

Fluorescence emission spectra of HSA (10.0 μM) treated with: 0.0, 2.5, 5.0, 7.5, 10.0, 12.5, and 15.0 μM (curves 1–7) Ant-PIm at 298 K at (A) λexc = 280 nm and (B) λexc = 293 nm.

Upon addition of increasing concentrations of Ant-PIm, a steady decrease in the HSA fluorescence intensity (≈91% at λexc = 280 nm and ≈86% at λexc = 293 nm), with a concomitant red shift (λmax (λ = 340 nm and λmax (λ = 346 nm) of the maximum emission wavelength of 8 and 4 nm, respectively, was observed, suggesting that almost quantitative energy transfer from the aa residues to Ant-PIm occurs. To provide further insights into the quenching nature of the HSA–Ant-PIm complex, the fluorescence data were analyzed using the Stern–Volmer equation assuming a static binding quenching phenomenon, as discussed in detail further on.[45,46]where F0 and F denote the steady-state fluorescence intensities in the absence and presence of Ant-PIm, respectively. KS is the Stern–Volmer constant used in the presence of a static quenching mechanism, [Q] is the Ant-PIm concentration, kq is the apparent quenching rate constant of the biomolecules, and τ0 is the average excited-state lifetime of HSA without a quencher and it is equal to 3.53 × 10–9 s.[47,48]

Due to Ant-PIm absorption at the excitation and emission wavelengths, an inner filter correction was applied before quantitatively analyzing the data using the following equation.[49]where Fcorr and Fobs are the corrected and observed fluorescence, respectively, and Aex and Aem are the absorbance values at the excitation and emission wavelengths, respectively.

The plot of F0/F versus [Ant-PIm] is shown in Figure A.

Figure 3

(A) Stern–Volmer plots for the quenching of HSA fluorescence by Ant-PIm at 298 K (λexc = 280 nm and 293 K). (B) Stern–Volmer plot for the quenching of HSA fluorescence by Ant-PIm at 298, 304, and 310 K (λexc = 280 nm).

A significant difference in magnitude of the KS values (KS (λ = (6.8 ± 0.2) × 104 M–1 and KS (λ = (4.8 ± 0.3) × 104 M–1) after excitation at these two wavelengths was observed indicating the involvement, to some extent, of the Tyr residues in the molecular interaction between HSA and Ant-PIm, and this is further confirmed by the Fourier transform infrared (FT-IR) studies discussed further on.

It is known that quenching of the fluorescence intensity can be ascribed to a wide variety of molecular interactions including ground-state complex formation, collisional quenching, excited-state reactions, molecular rearrangement, and energy transfer.[45] Such different mechanisms are usually collectively considered as either static or dynamic quenching. Static quenching arises from the formation of a ground-state dark complex between the fluorophore (F) and the quencher (Q), whereas dynamic quenching refers to the diffusive encounter between the fluorophore and the quencher during the lifetime of the excited state.

Static and dynamic quenching can be distinguished by their different dependence on temperature.[45] Higher temperatures lead to faster diffusion and thus larger amounts of collisional quenching, whereas in the case of static quenching, they will typically result in the dissociation of weakly bound complexes. The KS temperature-dependence analysis (Figure B and Table ) shows that less quenching occurs at higher temperatures strongly pointing toward formation of the Ant-PIm–HSA complex through a static mechanism. Further evidence was also provided by analysis of the kq values, which were found to be much higher than the maximum diffusion collisional quenching rate of various quenchers with biopolymers ≈ 2.0 × 1010 M–1 s–1, confirming the ground-state complex formation (Table ).

Table 1

Stern–Volmer (KS), Quenching Rate Constant (kq), Association Constants (Ka and Kb), and Number of Binding Sites (n) in the Interaction between Ant-PIm and HSA at Various Temperatures

T (K)	KS (M–1)	kq (M–1 s–1)	n	Ka (M–1)	Kb (M–1)
298	(6.8 ± 0.2) × 104	(1.9 ± 0.2) × 1013	1.0	(4.0 ± 0.3) × 104	(5.2 ± 0.3) × 104
304	(5.7 ± 0.2) × 104	(1.6 ± 0.2) × 1013	1.1	(4.5 ± 0.2) × 104	(5.4 ± 0.3) × 104
310	(4.9 ± 0.1) × 104	(1.4 ± 0.1) × 1013	1.2	(5.0 ± 0.2) × 104	(5.8 ± 0.1) × 104

Fluorescence Binding Data Analysis

By exploiting the fluorescence titration data and assuming a static quenching event, we evaluated the association constant (Ka) using the modified Stern–Volmer equation[45]where ΔF is the difference between the total fluorescence in the absence and in the presence of Ant-PIm and fa is the fraction of fluorescence that is accessible to the quencher and is equivalent to the number of binding sites (n).

The plot of F0/ΔF versus the inverse of the quencher concentration [Q] is linear, which suggests a single component donor quenching system that would be expected for one major binding mode (Figure A). The obtained data listed in Table clearly evidence the accessibility of the fluorophore to the quencher and a moderate binding affinity.

Figure 4

(A) Modified Stern–Volmer plots of the Ant-PIm–HSA association system at 298, 304, and 310 K. (B) Scatchard plots of the Ant-PIm–HSA association system at 298, 304, and 310 K.

When a guest molecule binds to a set of equivalent sites on a macromolecule, the equilibrium binding constant (Kb) can be further calculated according to the Scatchard equation[50−52]where r is the number of ligands attached to a single protein, Df is the molar concentration of the free ligand, and n refers to the binding site multiplicity per class of binding sites.

As reported in Table , the magnitudes of the binding constants (Kb) are in good agreement with those found using the modified Stern–Volmer plots (Ka) and agree well with literature data on a variety of therapeutic compounds targeting the protein template.[50,53−55] Furthermore, the linearity of the Scatchard plots (Figure B) corroborates well with the aforementioned hypothesis that only one class of binding sites is available to the quencher and it also highlights the noncooperativity of the process.

Analysis of Protein Secondary Structure

Changes in the protein secondary structure upon ligand interaction were studied by exploiting the circular dichroism (CD) technique. The HSA CD spectrum consists of one negative dichroic band located in the far UV spectral region, which is characteristic of the α-helical structure of the protein.[56]Ant-PIm, which is achiral, is thus CD inactive.[42] Upon addition of Ant-PIm to the HSA solution, a steady decrease in the negative ellipticities at 208 and 222 nm was observed indicating a decrease in the intrinsic HSA α-helix content (Figure ).

Figure 5

CD spectra of HSA (1.0 μM) treated with: 0.0 (black line), 1.25 (red line), 2.5 (blue line), and 5.0 (green line) μM Ant-PIm at 298 K.

To provide quantitative information on the HSA secondary structure, we calculated the mean residue ellipticity (MRE) value at 208 nm and the α-helix % content using the following equations[56]where Θ208 is the observed CD value (in millidegrees), Cp is the concentration of the protein, l is the path length of the cuvette, and n is the number of amino acid residues of the protein (585); 4000 is the MRE of the β-sheet and random coil conformation at 208 nm and 33 000 is the MRE of a pure α-helix at 208 nm.

According to the above equations, the percentages of α-helix at different HSA/Ant-PIm molar ratios were calculated. It was found that the HSA α-helix content decreases from 56.8 to 46.2, 33.8, and 24.1% with increasing Ant-PIm concentration (r = 0.8, 0.4, and 0.2, respectively). The CD data categorically established that the presence of the ligand perturbs the protein secondary structure, even if the lack of shift in CDmax and the similarity of the intrinsic dichroic protein bands in the absence and in the presence of the ligand suggest that HSA retained its original conformation excluding a denaturation process. To further support our hypothesis, infrared spectroscopic studies were carried out for the free and complexed protein (Figure ).

Figure 6

FT-IR spectra of the free HSA and Ant-PIm and their relative difference FT-IR spectra at a Ant-PIm/HSA molar ratio equal to 1.

As one can easily see, no major spectral shift for the protein amide I band at 1654 cm–1 (mainly C=O stretch) and amide II band at 1539 cm–1 (N–H bending coupled with C–N stretching mode) could be observed upon ligand interaction. Nevertheless, a decrease in intensity of both the amide bands was detected suggesting a weak overall perturbation of the protein secondary structure due to the reduction in α-helix content, which correlates well with the CD results discussed above.[55,57−59] It is worth noting that the weak band at 1454 cm–1 of the free HSA, assigned to the Tyr side chain vibration, shifted to 1448 cm–1 upon ligand addition, which confirms the involvement to some extent of the Tyr residues in the molecular interaction between HSA and Ant-PIm.[60,61] Further insights into the Ant-PIm–HSA adduct were also provided by analyzing the intrinsic vibrational bands of the anthracenyl derivative. Upon complexation, the bands of the free fluorophore at 1729 and 1167 cm–1 (assigned to C=O and C–O stretching, respectively) shifted to 1734 and 1170 cm–1, respectively, presumably due to the occurrence of external binding interactions, electrostatic in nature, between the hydrogen bond donor (Tyr 411 and Ser 489) and acceptor sites of the protein and ligand, respectively. Moreover, the free Ant-PIm bands at 1602, 1390, and 767 cm–1 ascribed to the C–C stretching and C–H bending of the aromatic rings shifted to 1621, 1386, and 760 cm–1, respectively, highlighting the presence of hydrophobic contacts through the anthracenyl polymer aliphatic chain and the surrounding hydrophobic environment in which the tryptophan residue is located, which is in good agreement with the thermodynamic studies reported within the Supporting Information.

Intermolecular Energy Transfer

The steady-state emission data obtained (vide supra) upon excitation of Trp-214 (λexc = 280 nm) suggest the existence of energy transfer from Trp-214 to Ant-PIm. Figure shows the overlap between the HSA emission and Ant-PIm absorption spectra, respectively.

Figure 7

Spectral overlap between the HSA emission and Ant-PIm absorption spectra. [HSA]/[Ant-PIm] = 1.

The Förster equation for nonradiative energy transfer (FRET) can be used to probe the proximity and relative angular orientation between the excited molecule (donor) and its neighbor (acceptor).[62−64] It is well established that FRET occurs only when all of the following three requirements are fulfilled: (i) the donor (D) is a fluorescence emitter; (ii) the fluorescence emission spectrum of the donor and the absorption spectrum of the acceptor (A) at least partially overlap; and (iii) the distance between D and A is lower than 10 nm.[40] Under such circumstances, the distance r between HSA (Trp-214) and Ant-PIm can be calculated using the following equation[45]where E and r denote the efficiency of transfer and the average distance between D and A, respectively; R0 is the critical distance when the energy transfer is 50% and is expressed by the following equation.[45]where K2 is the orientation factor related to the geometry of the donor and acceptor dipoles in random orientation and its value is equal to 2/3 (which is the approximate case considered here, as we cannot get more precise information); n is the average refractive index of the media; Φ is the fluorescence quantum yield of the donor; and J is the overlap integral between the emission spectrum of the donor and the absorption spectrum of the acceptor and it is defined by the equation given below.[45]where F(λ) is the corrected fluorescence intensity of the donor in the wavelength range from λ to λ + Δλ and ε(λ) is the extinction coefficient of the acceptor at each λ.

By applying eqs –9 and knowing that n = 1.36 and Φ = 0.118,[45,65] we calculated the following parameters: J = 2.70 × 10–14 cm3 M–1; E = 0.33; R0 = 2.89 nm; r = 3.25 nm. As the average distance between donor and acceptor is within the 2–10 nm range and 0.5R0 < r < 1.5R0, these results indicate that the energy transfer from HSA to Ant-PIm occurs with high probability and provides further evidence for Ant-PIm–HSA complexation through a static quenching mechanism.[6] The average distance r was found to be in good agreement with those reported for a variety of compounds targeting Sudlow’s sites.[6,50,54,64]

According to the Förster theory, the rate constant of energy transfer (kET) from Trp-214 to Ant-PIm can be estimated by the following equation.[45]where τHSA = 3.53 × 10–9 s.[47,48] By implementing the parameters calculated above (eqs –9) into the equation, we obtain kET = 1.40 × 108 s–1, which indicates that the energy transfer rate between Trp-214 and Ant-PIm is fast enough to compete with the radiative deactivation of the tryptophan residue.[47,48]

Effect of HSA on Ant-PIm Fluorescence Spectra

Ant-PIm is a strong luminophore and shows an emission band centered at 570 nm when excited with 516 nm wavelength radiation (Figure ). Changes in the emission spectra of Ant-PIm upon incremental addition of HSA were thus investigated.

Figure 8

Fluorescence emission spectra of Ant-PIm (2.5 μM) at 298 K treated with: 0, 5, 10, 15, 20, 25, 30, 35, 40, 45, and 50 μM (curves 1–11) HSA. Inset: plot of 1/(I – I0) against 1/[HSA] for the Ant-PIm–HSA system at 298 K (λexc = 516 nm).

Upon incrementally increasing concentrations of HSA, an enhancement in the fluorescence intensity of Ant-PIm along with a 14 nm blueshift of the maximum emission wavelength was detected. The enhancement in the fluorescence intensity and change in maximal wavelength can both be ascribed to either (i) the reduction in polarity of the environment around Ant-PIm that arises from its interaction with the hydrophobic pocket of the protein or (ii) the accommodation of the Ant’s chromophore into the HSA template; this may simultaneously impose a specific conformation on its long conjugated carbon skeleton, causing modification in the ground and excited-state geometries of the chromophore, and reducing the freedom of rotation of the fluorophore, which in turn, limits the possibility of vibrational deactivation pathways.

To provide quantitative information on the strength of the HSA–Ant-PIm interaction, the modified Benesi–Hildebrand equation was used[66]where I0 and I are the emission intensities of Ant-PIm in the absence and presence of the protein, respectively, and I∞ is the fluorescence intensity at saturated interaction, which is reached when the excess of added protein in the medium is so large that Ant-PIm is exclusively present in its bound form. The plot of 1/(I – I0) versus 1/CHSA is linear over the whole concentration range studied and the binding constant (Kf) can be calculated from the ratio of the intercept and the slope (inset Figure ) providing a value equal to (8.5 ± 0.1) × 104 M–1, which confirms a relatively high binding affinity.

Moreover, by exploiting the linear dependence of the Ant-PIm emission changes as a function of HSA concentration, we determined a limit of detection (LOD) (=3σ/slope) = 7.5 × 10–3 mg/mL, which makes Ant-PIm an ideal probe to estimate HSA in normal and albuminuria urine samples.[26]

Time-Resolved Fluorescence Decay Measurements

To elucidate the origin of the Ant-PIm fluorescence enhancement in complex with HSA, the changes in the fluorescence decay functions upon binding to the protein were determined, and are shown in Figure .

Figure 9

Representative time-resolved fluorescence decay profiles (λexc = 515 nm) of Ant-PIm at different HSA/Ant-PIm molar ratios. [Ant-PIm] = 2.5 μM.

The free Ant-PIm was found to exhibit a mono-exponential decay pattern that comprised a slow component of 3.27 ns. However, as reported in Table , the time-resolved fluorescence decay of Ant-PIm bound to HSA could be satisfactorily described by a bi-exponential pattern with two distinct lifetimes.

Table 2

Time-Resolved Fluorescence Decay Parameters, Fluorescence Quantum Yield, and Kinetic Parameters of Ant-PIm with Increasing Concentrations of HSA

[HSA] (μM)	α1 (%)	α2 (%)	τ1 (ns)	τ2 (ns)	⟨τ⟩ (ns)	Φf	kr (×107 s–1)	knr (×107 s–1)
0	100		3.27		3.27	0.3	9.17	21.41
5	38	62	2.19	4.16	3.68
10	40	60	2.31	4.41	3.86
15	48	52	2.70	4.77	4.06	0.53	13.05	11.58

Upon addition of HSA, the component τ1, which corresponds to the free Ant-PIm molecule, shows a sudden decrease in magnitude followed by the simultaneous appearance of a new component τ2, which provides the largest contribution. The direct outcome of the binding phenomenon between the Ant’s chromophore and the protein template can be confirmed by the progressive increase in the average lifetime ⟨τ⟩ with increasing protein concentration.

A more quantitative data analysis was undertaken to demarcate the contribution of the radiative (kr) and nonradiative (knr) decay constants of Ant-PIm in an aqueous buffer solution and protein environment according to the following equations[45]where Φf denotes the fluorescence quantum yield of Ant-PIm. On the basis of the calculated results shown in Table , we infer that the longest chromophore lifetime is the direct outcome of the attenuation of the radiationless pathway via the motional restriction imposed on the Ant-PIm moiety.

Unraveling the Selective Ant-PIm Binding to Biological Targets

These results prompted us to to investigate whether the revealed emission enhancement upon HSA interaction could be taken advantage of in the conception of a biocompatible fluorescent probe that is able to discriminate HSA from BSA. As shown in Figure , an enhancement in the intrinsic Ant-PIm fluorescence intensity with a concomitant 4 nm blueshift of λmax was observed upon BSA addition. Consequently, the reduced hyperchromic and hypsochromic effect observed for the Ant-PIm–BSA complex (Kf  = (3.0 ± 0.3) × 104 M–1) as compared to that of the Ant-PIm–HSA adduct allows easy discrimination of HSA from BSA, either by the monitoring of their emission spectra or by naked-eye detection, as shown in Figure .

Figure 10

Fluorescence emission spectra of Ant-PIm (2.5 μM) at 298 K treated with: 0, 5, 10, 15, 20, 25, 35, and 50 μM (curves 1–8) of BSA. Inset: plot of 1/(I – I0) against 1/[BSA] for the Ant-PIm-BSA system at 298 K.

Figure 11

(Left panel) Sensitive discrimination of HSA from BSA by Ant-PIm after a 515 nm irradiation. From left to right: free Ant-PIm (orangish), Ant-PIm–BSA (goldish), and Ant-PIm–HSA (yellow-greenish). [Ant-PIm] = 2.5 μM and [BSA] = [HSA] = 15 μM. (Right panel) Normalized fluorescence intensity responses of Ant-PIm in the absence and presence of serum proteins.

Further insights were also provided by testing the Ant-PIm-sensing response toward common biological interferents by fluorescence spectroscopy experiments. As shown in Figure , no, or very weak, changes to λmax occurred in the intrinsic Ant-PIm emission upon binding to common biomolecules, which highlights the probe’s excellent ability to suitably detect HSA in rather complex biological environments.

Figure 12

(Left panel A) (A) HSA; (B) Myoglobin; (C) l-glutathione; (D) free Ant-PIm; (E) l-arginine; (F) lysozyme; (G) l-cysteine; (H) chymotrypsin; (I) chymotrypsinogen A; (J) BSA. [Ant-PIm] = 2.5 μM and [biological interferents] = 15 μM. (Right panel B) Normalized fluorescence intensity responses of Ant-PIm (2.5 μM) to HSA, BSA, and other biological interferents (15 μM).

Two-Photon Fluorescence Response of the Free Ant-PIm and Ant-PIm–Protein Systems

Recent developments in TPA applications have focused on identifying extrinsic biological targeting agents that are easily excitable by a short pulse light source because they can avoid conventional one-photon drawbacks such as shallow penetration depth, a relatively high autofluorescence background from other naturally occurring emissive compounds, and the tissue photodamage inherent to conventional UV–visible (UV–vis) sources.[29−35] Consequently, NIR TPA excitable probes, with a high TPA cross section (σ2), are a much sought-after class of dyes for biomedical imaging, analysis, and diagnosis.[29−35]

As shown in Figure (top left panel), a relatively high TPA is observed in the spectral region between 780 and 860 nm with the maximum located at 820 nm for both the free and complexed Ant-PIm. It is worth noting that the cross section values vary significantly between the unbound (σ2 = 813 GM at 820 nm) and bound dye (σ2 = 469 GM and σ2 = 443 GM at 820 nm); therefore, the lack of a common scaling factor for the free and bound Ant-PIm molecule states makes the straightforward comparison of σ2 values very difficult, as the environmental conditions in which the fluorophore is placed can dramatically modify the transition dipole moment of the dye, affecting the two-photon cross section magnitude. However, if we compare the Ant-PIm performance in terms of molecular brightness (σ2 × Φf), we find two similar values for the free fluorophore (σ2( × Φf( = 244 GM at 820 nm) and its HSA-bound state (σ2( × Φf( = 248 GM at 820 nm) (Figure top right panel). These overall findings definitively make the anthracenyl derivative a promising sensing probe for serum protein detection. This was illustrated by irradiating the same cuvettes, as shown in Figure , with a Ti-Sapphire laser working at 820 nm: a detectable two-photon excited fluorescence was observed for all three mixtures, underlining even further the potential of Ant-PIm for use as a selective probe for the discrimination of HSA versus BSA even in complicated, auto-fluorescent biological mixtures (Figure bottom central panel).

Figure 13

(Top left panel) TPA cross section of Ant-PIm in the absence (black curve) and in the presence of HSA (green curve) and BSA (red curve). [Ant-PIm] = 2.5 μM and [HSA] = [BSA] = 15 μM. (Top right panel) Molecular brightness (σ2 × φf) plot of Ant-PIm in the absence (black curve) and presence of HSA (green curve). [Ant-PIm] = 2.5 μM and [HSA] = 15 μM. (Bottom central panel) Sensitive discrimination of HSA from BSA by Ant-PIm after a 820 nm irradiation. From left to right: free Ant-PIm (orangish), Ant-PIm–BSA (goldish), and Ant-PIm–HSA (yellow-greenish). [Ant-PIm] = 2.5 μM and [BSA] = [HSA] = 15 μM.

Conclusions

In summary, the HSA-binding properties of a water-soluble fluorophore have been comprehensively investigated under simulated physiological conditions. Steady-state fluorescence data, obtained as a function of temperature, confirmed that hydrophobic forces regulate the Ant-PIm–HSA complex formation. Displacement and energy transfer studies provide clear evidence for the binding of Ant-PIm in the protein subdomain IIA. CD and FT-IR structural analysis showed a decrease in the protein α-helix content upon ligand complexation even if no, or weak, overall changes in the HSA secondary structure were observed. Enhancement of the intrinsic Ant-PIm fluorescence intensity and changes in the fluorophore’s lifetime may be considered to be the direct outcome of the attenuation of the radiationless pathway via the motional restriction imposed on the Ant’s moiety from the protein template. The different Ant-PIm emission response to HSA and BSA categorically established the excellent selectivity of the probe to suitably target the two similar proteins even in rather complex biosystems, as proven by the lack of response toward common biological interferents. Moreover, the high TPA cross section (σ2 > 800 GM) maximum located at 820 nm makes the anthracenyl derivative a promising probe for serum protein discrimination. This is the first report in which a water-soluble two-photon fluorophore has been used for selective albumin detection at physiological pH. We speculate that future works in this area will open up new avenues to develop smart multi-photon anthracenyl-based sensors, which have the potential to emerge as specific bio-markers for diagnostic applications.

Materials and Methods

Synthesis of Ant-PIm

The synthetic route and the physical properties of the anthracenyl derivative Ant-PIm are reported in our previous papers.[42]

Reagents

All chemicals used throughout the experiments were purchased from commercial suppliers and used without further purification. HSA and BSA, purchased from Sigma-Aldrich, were diluted in Milli-Q water to a final concentration of 0.2 mM. Stock solutions of Myoglobin, l-glutathione, l-arginine, lysozyme, l-cysteine, chymotrypsin, and chymotrypsinogen A were prepared by dissolving the samples in Milli-Q water until the desired concentration was reached. Sodium cacodylate trihydrate (0.05 M), supplied by Sigma-Aldrich, was used to control the pH of the solutions (pH 7.2).

Apparatus and Methods

Absorption spectra were recorded on a PerkinElmer Lambda 20 UV–vis spectrometer. Emission spectra were obtained with a Hitachi F-4500 spectrofluorometer equipped with a xenon lamp. Fluorescence lifetimes were determined with an Edinburgh Instruments FLS 980 spectrophotometer via time-correlated single-photon counting (TCSPC), with excitation from a 516 nm picosecond laser diode. Two-photon excited fluorescence (TPEF) was achieved with a coherent chameleon laser that delivered a train of ≈100 fs pulses with 80 MHz repetition rate. The TPEF spectra were recorded by an Ocean Optics 2000 fiber spectrometer. CD experiments were conducted on a Jasco J-815 spectropolarimeter (Jasco Inc). The infrared spectra were collected on a diamond crystal surface under vacuum (<1 hPa) using a Bruker Vertex70v FT-IR spectrometer. Quartz cells with a 1 cm path length were used throughout the measurements.

Titration experiments were recorded keeping the final volume of each solution constant to reduce dilution issues and thus achieve better reproducibility of the recorded data.

Steady-State Fluorescent Measurements

Fluorescence quenching studies were recorded at different temperatures (298, 304, and 310 K) keeping the HSA concentration constant (10 μM; λexc = 280 nm or λexc = 293 nm) and varying that of the anthracenyl derivative Ant-PIm until saturation was achieved.

The intrinsic Ant-PIm emission response toward biomolecules was observed at a fixed Ant-PIm concentration (2.5 μM; λexc = 516 nm) and varying that of the biological targets.

The resulting changes in the emission profiles were used to calculate the biophysical and thermodynamic parameters as well as the strength of binding.

Moreover, the linear dependence of the fluorescence emission of Ant-PIm in the presence of increasing concentrations of HSA allowed the LOD to be estimated, which was defined as: LOD = 3σ/k, where σ is the standard deviation of the blank measurement and k is the slope of the linear plot obtained by casting the changes in the fluorescence intensity of the Ant-PIm–HSA complex as a function of HSA concentration.[26,67,68]

Fluorescent Displacement Assay

Site-marker competition experiments were performed using the well known serum protein binders Ibuprofen and Warfarin purchased from Sigma-Aldrich. An equimolar concentration of both drugs and HSA was mixed and the relative product was scanned in the wavelength range 285–500 nm (λexc = 280). The resulting signal was normalized to 100% of the emission profile. Subsequently, various concentrations of Ant-PIm were added to the mixture and the fluorescence was measured after an incubation time of 10 min. The addition of Ant-PIm was continued until the fluorescence signal reached saturation. The study, which allowed us to conclude that Ant-PIm could efficiently replace both binders, is included as Supplementary Material.

Time-Resolved Fluorescence

Fluorescence decay traces of both the free Ant-PIm and its bound sate were recorded via TCSPC. The fluorescence was collected at a 90° geometry after passing through a polarizer set at the magic angle. Calculations of Ant-PIm excited-state lifetimes (τ) and the corresponding amplitude (A) were made using OriginPro 8 software. The amplitude-weighted fluorescence lifetime was calculated using the following formulawhere n denotes the number of decay components in the total function.

TPEF

TPEF measurements were performed at HSA or BSA/Ant-PIm molar ratios r = 0 and 6 using fluorescein as the reference. The relative concentration was adjusted so that the linear absorbance at 2hν was kept below 0.1 over the whole wavelength range studied (780–920 nm).

The TPA cross section was determined according to the following equationwhere σ2 is the TPA cross section, c and n are the concentration and refractive index, respectively, and F is the integrated area obtained from the TPEF spectrum. The subscript r refers to the reference solution.

CD Spectroscopy

CD spectra were recorded as an average of five scans within the wavelength range 200–500 nm. The Ant-PIm/HSA molar ratios were varied from 0 to 5 and the buffer contribution subtracted from each recorded spectrum. Quantitative analysis of the protein secondary structure was performed according to previous reports.[69]

FT-IR Spectroscopic Measurements

IR measurements were carried out at a fixed HSA/Ant-PIm ratio (r = 1) and the relative signal of either the free HSA or Ant-PIm was subtracted from the HSA–Ant-PIm complex spectrum to achieve only the contribution of the interaction.