BSA-AuNCs characterization
The as-prepared BSA-AuNCs exhibit both in aqueous dispersion (inset – Figure 1A) and powder sample (inset – Figure 1B) a light brown color under ambiental (amb.) light excitation, while under UV light excitation turns reddish. This is the first indicator of the intrinsic PL of the obtained BSA-AuNCs. Also, the absence of a localized surface plasmon resonance band in their absorption spectrum (Figure 1A), suggests the formation of cluster-size nanoparticles. This phenomenon (terminologie), takes place usually when the nanostructure’s dimension approaches the Fermi wavelength of metals (~2 nm), transforming the continuous energy level band into discrete ones and giving NCs molecule-like properties. In consequence, NCs show intrinsic PL originating from sp-sp (intraband) and sp-d (interband) transitions 24. As shown in Figure 1B, BSA-AuNCs exhibit both in aqueous solution and as lyophilized powder a strong emission at 655 nm under 405 nm excitation. This proves their great stability and conservation of fluorescence properties in different states of matter, which is a great desiderate for various fluorescence-based application.
It was previously demonstrated that the BSA-AuNCs are photoluminescent under a large range of excitation wavelengths19. In this work, we have selected to perform our steady-state fluorescence investigations under continuous 405 nm laser excitation to match the excitation wavelength of the pulsed laser used in our future time-resolved spectroscopy and microscopy experiments. The absolute quantum yield of the BSA-AuNCs in solution under an excitation of 405 nm was calculated to be 6%. Even though the quantum yield of BSA-AuNCs was relatively calculated to be around this value in other works12,25,26, to our knowledge, this is the first time the absolute value is reported. Beside their intrinsic PL, NCs are also known for their high photostability27. As shown in Figure 1C, under continuous excitation at 405 nm for 15 min, the BSA-AuNCs retain in solution 86.8% of their original PL, while the powder sample reaches a value of 99.5%, which represents an extremely important advantage for future bioimaging applications. The morphology and structural characteristics of the BSA-AuNCs were rigorously detailed in our previous work as obtained from TEM, XPS and DLS measurements19. Briefly, the BSA-AuNCs are formed of AuNCs with an average size of 2-3 nm embedded in a BSA corona of 25 ± 12 nm. The aforementioned results prove the successful formation of BSA-AuNCs with an intrinsic red PL.
OPE and TPE FLIM assays on BSA-AuNCs powder
As it was previously demonstrated, FLIM is a very attractive fluorescence microscopy, overall superior to conventional fluorescence microscopy due to the enhanced contrast provided by sensitivity of fluorescence lifetime to local environment18. Due to the lack of studies on the PL of the BSA-AuNCs in a solid-state, the next part of the study was focused on the fluorescence lifetime imaging of BSA-AuNCs powder in order to expose the spectral and time-resolved response of the NCs in experimental conditions close to the biological ones Therefore, we recorded the OPE FLIM images of the same BSA-AuNCs microflake under 405 nm excitation, using different laser powers on the probe ranging between 0.07 and 0.67 µW. The collected OPE FLIM images and the corresponding optical image of the investigated microflake are presented in Figure 2.
BSA-AuNCs in powder state exhibit uniform and intense PL under the used excitation wavelength, while the intensity of the emission seems to be strongly correlated with the excitation laser power. Notably, after repeated excitations of the same microflake with different laser powers, the BSA-AuNCs show no trace of photobleaching, outcome that might not be able to achieve with a normal fluorophore28. The overall intensity of each FLIM image at different laser powers was extracted and presented in Figure 3A.
The PL of BSA-AuNCs microflake increases linearly with the increase of laser power, an expected phenomenon taking in consideration that at OPE the fluorescence emission intensity linear dependency can de described by the following equation29:
where I is the fluorescence intensity, K’ is a constant that depends on the fluorescence quantum yield, the molecule’s geometry and other factors, ɛ is the molar absorptivity, b is the laser path length inside the molecule, c is the concentration of the molecule, and Po is the excitation laser power.
Moreover, a representative fluorescence lifetime decay curve was extracted from the recorded OPE FLM images, using the same 405 nm excitation, in order to investigate the lifetime of the BSA-AuNCs powder microflake’s PL. The decay curve extracted from the microflake, presented in Figure 3A along with the fitting curve, reveals a two-exponential behavior. Each calculated lifetime component, obtained by fitting the lifetime decay curve (Figure 3B) and presented in Table 1 along with the average obtained value, was successfully assigned to a specific transition between the energy levels of BSA-AuNCs, under a pulse laser excitation.
Succinctly, when BSA-AuNCs absorb light, electrons will transition to an excited singlet level, either first (S1) or second. After an extremely short period of time, electrons relax to the ground vibrational state of S1 where two processes can occur: intersystem crossing (ISC) or prompt fluorescence emission (PF). If the ISC process takes place, the electron’s spin multiplicity changes to the first excited triplet level. At this point they can bounce back to S1 as a result of a thermal energy absorption, process called reverse-ISC. In the end, the electrons will relax through a temperature activated radiative process, also known as delayed fluorescence (DF)30. Consequently, the fast lifetime (τ1) can be assigned to ISC, while the longer component (τ2) corresponds to PF process. Unfortunately, the DF lifetime31 is over the limit value that our equipment can measure. These results are in a good agreement with our previous reported data obtained in solution19.
Table 1. The PL lifetime decay parameters obtained for a BSA-AuNCs microflake powder under OPE and TPE.
Probe
BSA-AuNCs
|
τ1 (ns)
|
A1 (%)
|
τ2 (ns)
|
A2 (%)
|
τ av (ns)
|
χ2
|
Powder under OPE
|
13.1
|
33.1
|
0.69
|
66.9
|
12.1
|
1.27
|
Powder under TPE
|
6.8
|
41.6
|
0.61
|
58.4
|
6.1
|
1.12
|
Agarose-Phantom under TPE
|
6.7
|
27.2
|
0.65
|
72.8
|
5.5
|
1.02
|
Compared to OPE methods, the TPE fluorescence imaging microscopy presents numerous advantages such as highly localized excitation providing background free information, elimination of out-of-focus absorption, reduced photobleaching and phototoxicity, the possibility to eliminate the pinhole, flexible detection geometries and more efficient photon detection, while maintaining a high image contrast20. Nonetheless, the TPE method requires higher wavelength photons, around the NIR region, giving the technique increased imaging depth and therefore the possibility to be applied in biomedical applications such as live cell and tissue imaging. Despite its superiority, the TPE method had only been recently exploited in the investigation of NCs’ PL21–23, but up to our knowledge none of them on NCs in solid-state. Therefore, we performed here TPE FLIM assays on the same BSA-AuNCs microflake under an 810 nm excitation using different excitation power (ranging between 5 and 20 mW at the laser cavity exit). The obtained TPE FLIM images together with the corresponding optical image of the microflake are presented in Figure 4.
Remarkably, the BSA-AuNCs microflake exhibits an intense and relatively uniform fluorescence emission under TPE at 810 nm. As it is well known that under NIR excitation, water and chromophores exhibit minimal scattering and absorption, these results prove the ability of BSA-AuNCs to be explored as potential fluorescence contrast agents in deep tissue imaging. Moreover, the quadratic dependence of their PL’s intensity (overall intensity of each FLIM image) on the excitation laser power (Figure 5A) certainly demonstrates that the obtained emission originates from a TPE process.
The fitting procedures performed on the lifetime decay curve extracted from the BSA-AuNCs microflake (Figure 5B) exposes the two previously described processes: PF (τ2) and ISC (τ1), demonstrating that the PL properties of the AuNCs preserved under TPE. Noteworthy, the BSA-AuNCs PL lifetime components do not overlap with the lifetime of endogenous intracellular autofluorescence, typically between 2 and 3 ns. Finally, we investigated the photostability BSA-AuNCs microflake under continuous pulsed irradiation at 810 nm for 1 min. Figure 5C shows that, BSA-AuNCs kept more than 99.5% of their initial PL, an extremely important advantage when it comes to bioimaging applications which proves the feasibility of using them as reliable stable contrast agents for non-invasive imaging assays of biological samples using NIR irradiation.
TPE FLIM assays on Agarose-Phantom@BSA-AuNCs
Bionanomedicine is among the most exploited research topic, while its main goal is to come with solutions to medical problems using nanometric devices. However, multiple promising studies stop at the in vitro level. For example, in our previous work, we demonstrated the ability of BSA-AuNCs to perform as fluorescence contrast agents for the visualization of ovarian cancer cells under visible light excitation. However, it is well known that NIR excitation provides not only the possibility of deep tissue imaging but also non-invasive ability. Therefore, we internalized the BSA-AuNCs inside a cancer tissue mimicking agarose-phantom (Agarose-Phantom@BSA-AuNCs) and tested their efficiency as contrast agents under non-invasive NIR TPE in simulated ex vivo environment, as a novel proof of concept of AuNCs-based NIR tissue imaging approach. The TPE FLIM image of Agarose-Phantom@BSA-AuNCs obtained under 810 nm excitation (Figure 6A) reveals an intense and homogenous fluorescence signal originating from the entrapped BSA-AuNCs. We point out that reference Agarose-Phantom exhibits unsignificant signal under the same conditions. In the obtained image, we are clearly able to differentiate between areas with BSA-AuNCs (green) and without (black), which demonstrates the ability of BSA-AuNCs to emit a strong PL under NIR excitation, when embedded in a cancer tissue-mimicking solid phantom. In addition, the spectral information together with PL lifetime values obtained for the Agarose-Phantom@BSA-AuNCs are in good agreement with the results obtained on lyophilized powder, proving that the emissive properties of BSA-AuNCs maintained even inside the tissue mimicking agarose phantom. Specifically, the PL spectrum extracted from Agarose-Phantom@BSA-AuNCs, under TPE at 810 nm, matches the PL of BSA-AuNCs in both solution phase (Figure 6B).
Additionally, the PL lifetime components obtained after the fitting of the decay curve from Figure 6C and presented in Table 1, are similar to the values obtained for BSA-AuNCs powder under TPE, with a small variation of contributions, most probably due to different environmental parameters. Therefore, the aforementioned results confirm the uniform distribution and the excellent staining ability of BSA-AuNCs inside tissue-like phantoms, under TPE, demonstrating, as a proof of concept, the ability of BSA-AuNCs to perform as reliable fluorescent contrast agents in future ex vivo or even in vivo non-invasive NIR FLIM-based imaging applications.