Confined photo-release of nitric oxide with simultaneous two-photon fluorescence tracking in a cellular system

Nitric oxide (NO) is a key signaling molecule in biological systems. New tools are required to therapeutically modulate NO levels with confined precision. This study explores the photoactivatable properties of an NO releasing compound (CPA), based on cupferron O-alkylated with an anthracene derivative. Upon light stimulation, CPA uncages two species: cupferron, which liberates NO, and an anthrylmethyl carbocation, which evolves into a fluorescent reporter. Proof-of-principle is demonstrated using one- and two-photon excitation (1PE and 2PE) in a cellular system (A431 cells). It was found that 1PE induces cell toxicity, while 2PE does not. Since 1PE using UV light is more likely to generate cellular photodamage, the cell toxicity observed using 1PE is most likely a combinatory effect of NO release and other UV-induced damage, which should be subject to further investigation. On the other hand, absence of phototoxicity using 2PE suggests that NO alone is not cytotoxic. This leads to the conclusion that the concept of 2PE photorelease of NO from CPA enable opportunities for biological studies of NO signaling with confined precision of NO release with minimal cytotoxicity.


Supplemental data Cell toxicity screening of DMSO as drug diluent for A431 cells
Since the CPA compound was only found to be soluble in DMSA, an analysis on what suitable concentration range of DMSO was performed. A431 cells were seeded at 1x10 4 per well tissue culture treated 96 well plates (Nunc cell culture plates, ThermoFisher Scientific) and allowed to incubate at 37 degrees in 5% CO2 for 24 hours before experiment. Four hours prior to irradiation, full growth media was replaced with varying percentage (1%, 2.5%, 5%, 7.5%, 10%, 12.5%, 15%, and 20%) DMSO in FBS-free media with the exception of one column replaced with 100% full growth media for control. Toxicity of varying concentrations of DMSO was evaluated using the MTT viability assay (Sigma-Aldrich, Stockholm, Sweden). As shown by Supplemental Fig. S1, the cell toxicity was found to be strongly increased for DMSO concentration above 2.5%. Thus the DMSO concentration was kept at or below 2.5 % in in the experiments. Figure S1. Cell toxicity of A431 cells in the presence of DMSO at varying concentrations in FBS-free media (0 -30% DMSO in media by volume). Cells were incubated with DMSO concentrations for 4 h prior to evaluation of cell viability by the MTT assay. Replicates n = 4, error bars showing standard deviation.

Cell toxicity, 1PE photoactivation using MTT assay
In addition to measuring cell toxicity following photoactivation treatment with the AlamarBlue assay; cell viability was monitored using MTT-assay using protocol as suggested by manufacturer (Sigma-Aldrich, Stockholm, Sweden). Briefly, 20 µL of MTT (5 mg/mL) in cell media was added to each well of the 96well plates. After 4 h of incubation, media and MTT solution was replaced with 100 µL HCl 0.01 M in DMSO. Absorbance was measured at 550 nm using a SpectraMax M2 Multi-mode microplate reader (Molecular Devices, Berkshire, UK). Experiments were performed in replicates of 6. Statistical analysis was performed using excel data analysis toolbox student's t-test. P values are shown as annotations to cell viability graph and standard error of mean across 6 replicates.

Supplemental figure S2
. Cell toxicity of A431 cells using MTT assay after one-photon induced photoactivation at CPA concentration at 1.25% and 2.5% compared to control with cell media only. Photoactivation obtained using a broad spectrum UV-lamp at light doses of 0, 5 J/cm 2 . Statistical analysis performed from replicates of n=6. Error bars represent standard error of mean.

Scavenging experiment
A scavenging experiment was performed, adopting a protocol by others 1 . Cells were cultured and seeded as earlier. CPA at 2.5%. was added together with either superoxide dismutase (Sigma-Aldrich, Stockholm, Sweden) as superoxide scavenger, uric acid (Sigma-Aldrich, Stockholm, Sweden) as peroxynitrite scavenger, or catalase (Merck Chemicals and Life Science, Solna, Sweden) as hydrogen peroxide scavenger, at concentrations of 100 ug/mL (PBS, pH 7.4), at the same time as addition of CPA, and incubated for 4 hours. Irradiation using UV was performed as earlier, but light dosage was adjusted to 10 J/cm 2 . Immediately following irradiation, all solutions were removed and replaced with fresh media and allowed to recover overnight and cell toxicity assessed. The data is presented in Figure S3 and controls in Figure S4. No significant difference in cytotoxic effect upon irradiation compared to control cells was found when uric acid, known as peroxynitrite scavenger 1,2 , was added together with the CPA. Addition of superoxide dismutase and catalase, i.e., superoxide and hydrogen peroxide scavengers,respectively, showed no significant difference in cell toxicity compared to CPA only after irradiation. This experiment indicates that generation of peroxynitrite upon irradiation might be involved for causing the cell toxicity following 1PE irradiation of CPA; however to fully elucidate the mechanism further experiments should be undertaken.

Theoretical analysis
In order to compare the likelihood for a CPA molecule undergoing photo-induced decomposition through either 1PE or 2PE using the conditions given in the study a theoretical calculation was performed. The value for molar extinction is approximately ε = 10 000 M cm -1 according to Vittorino et al. 3 , corresponding to an absorption crossection σ1PA = 3.8 × 10 -8 cm 2 (using σ1PA = ln(10) × ε / NA). The two photon absorption crossection is approximated to be similar to anthracene in solution δ2PA = 1 × 10 -53 cm 4 s -1 , determined by Webman and Jortner 4 .
Given the experimental conditions in the 1PE experiments using a wavelength at λ1PE = 370 nm and intensity I1PE = 0.2 W/cm 2 , yields a photon flux approximately Φ1PE = 3 × 10 17 cm -2 s -1 . For 2PE excitation, pulsed NIR light is utilized. Thus the photon flux in the 2PE case needs to be calculated for the laser pulse at the focal spot, given that the repetition rate is 80 MHz, pulse duration approximately 100 fs, lambda λ1PE = 750 nm, numerical aperture of objective lens NA = 1, and laser power at sample about 20 mW, giving Φ2PE = 1 × 10 30 cm -2 s -1 .
The probability for a molecule undergoing photodecomposition and subsequent generation of NO will be proportional to the probability of undergoing excitation, which for the two different scenarios can be estimated to stated in the units as s -1 . Given the numbers above the P2PE ≈ 2 × 10 7 s -1 >> P1PE ≈ 1 × 10 3 s -1 , which means that the probability for a specific molecule is much more likely to undergo 2PE rather than 1PE given the conditions above. So as long as the experiments are designed to cover similar total excitation volumes and total light dosages (in our study 5 J/cm 2 ), 2PE should in fact be much more effective to decompose CPA in order to release NO. Thus the lack of cell toxicity observed using 2PE in our experiments cannot simply be explained by lower probability of 2PE and confinement effect, but alternative explanations should be considered.