Gallstones are a prevalent global disease, up to 20% of adults develop gallstones [1,2]. Various risk factors, such as age, pregnancy, lack of exercise, obesity, and overnutrition, contribute to the formation of gallstones [2]. Impaired gallbladder emptying leading to cholestasis is considered a fundamental dynamic factor in gallstone formation [3]. Furthermore, bile stasis caused by gallstones can lead to inflammatory changes in the bile ducts, ultimately progressing to Cholangiocarcinoma. Throughout this process, there is an exacerbation of iron-dependent oxidative stress, with ferroptosis playing a role [4,5]. Nitric oxide (NO), as an important gaseous signal mole, plays a vital role in living organisms. It is a relaxation factor [6,7], and is closely related to intercellular communication, neurotransmission, cardiovascular, gastrointestinal, urogenital system, respiratory system, anti-pathogenicity, anti-tumor response and oxidative stress, and so on [8]. Notably, NO is an important neurotransmitter, which is abundant in the sphincter of Oddi and regulates biliary movement [9,10]. This means that the course of gallstones might be accompanied by changes in NO levels. On the other hand, in patients with gallstones, the accumulation of lipid droplets (LDs) leads to adipocyte hypertrophy [11]. Therefore, the detection of NO in LDs, ferroptosis, and gallstone disease is of great significance.

Fluorescent probes, known for their high sensitivity, good biocompatibility, and ability to perform real-time non-invasive detection, have been widely employed in biological imaging [12-16]. Ratiometric fluorescence probes, capable of self-tuning between two fluorescence bands, effectively overcome issues related to probe concentration, pH value, instrument efficiency, and temperature, making them particularly useful for the detection of important substances in living organisms [17-22]. Meanwhile, organelle-targeted fluorescent probes provide a more precise imaging tool for understanding changes in important substances within cells [23-25]. So far, although many excellent NO probes have been reported [26-30], organelle-targeted ratiometric fluorescent probes for NO are rarely reported [31-33]. Particularly, the development of a lipid droplet-targetable fluorescent probe for ratiometric imaging of NO remains unexplored.

Here, we report a readily obtainable fluorescent probe (LR-NH, Scheme 1) for ratiometric imaging of NO in LDs. The probe uses an anthracimide as the fluorophore and a secondary amine as the NO reaction site [34]. The fluorophore contains a hydrophobic n-octyl carbon chain, giving the probe LR-NH a lipophilic structure and allowing it to target LDs. Our results show that LR-NH exhibits rapid, colorimetric, and ratiometric fluorescence changes for NO with high selectivity and sensitivity. Moreover, LR-NH can effectively target LDs and image both exogenous and endogenous NO in living cells. LR-NH also shows the ability to image the level changes of NO during cellular inflammation and ferroptosis. More importantly, LR-NH can target the gallbladder and image NO in gallstone disease models. To our knowledge, LR-NH is the first probe that enables ratiometrically imaging of NO in LDs and gallbladder (Table S1 in Supporting information).

LR-NH was synthesized according to Scheme S1 (Supporting information) and the characterization data were listed in Figs. S1–S6 (Supporting information). After obtaining LR-NH, its spectra before and after the addition of NO were explored (Fig. 1). LR-NH shows a broad strong absorption band around 536 nm and strong fluorescence at 620 nm. After the addition of NO, this absorption band significantly decreased. Meanwhile, the fluorescence at 620 nm decreased and a new fluorescence peak at 515 nm emerged, resulting in a ratiometric fluorescence change with a remarkable blue shift (105 nm). Concurrently, the purple probe solution turned nearly colorless and the emission color turned from red to green. A time-dependent analysis of the fluorescence intensity changes before and after adding NO was performed (Fig. S7 in Supporting information). It was observed that LR-NH exhibited good stability before the addition of NO. However, after adding NO, the fluorescence at 620 and 515 nm rapidly decreased and increased, respectively, and saturation was reached within 5 min, indicating that LR-NH rapidly and ratiometrically responds to NO. Altogether, LR-NH is a rapid, colorimetric, and ratiometric fluorescent probe for NO.

LR-NH was exposed to a wide range of reactive species (NO, NO₃⁻, NO₂⁻, O₂^∙−, HNO, ONOO⁻ ClO⁻, ROO^∙OH, H₂O₂), amino acids (Arg, Ala, Phe, Ser, Val, Trp, Thr, Lys), and other species (F⁻, Cl⁻, Br⁻, I⁻, HCO₃⁻, HSO₃⁻, SO₄²⁻, K⁺, Ca²⁺, Zn²⁺, Mg²⁺, Fe³⁺, Fe²⁺, ascorbic acid (AA), dehydroascorbic acid (DHA), methylglyoxal (MGO), glutathione (GSH), homocysteine (Hcy), and cysteine (Cys)) to assess its selectivity. As depicted in Fig. 2A, only NO leads to significant changes in LR-NH fluorescence. As a result, only NO induced obvious ratiometric fluorescence changes at 515/620 nm (Fig. 2B). Moreover, even with other analytes present, LR-NH showed significant ratiometric fluorescence changes for NO (Fig. S8 in Supporting information). These findings indicate that LR-NH can detect NO with high selectivity.

We also evaluated the sensitivity of LR-NH to NO. As depicted in Fig. 2C, when the concentration of NO gradually increases, the fluorescence of LR-NH at 620 nm and 515 nm progressively decreases and increases, respectively. The fluorescence ratio of F₅₁₅/F₆₂₀ was directly correlated to the concentration of NO, and a maximum enhancement was found to be ~235-fold (from 0.03 to 7.04, see the inset of Fig. 2C) after the addition of 35 µmol/L of NO. Importantly, there is a linear relationship between the ratio of F₅₁₅/F₆₂₀ and the NO concentration of NO (0–0.9 µmol/L) (Fig. 2D), which gives a low detection limit of 0.49 nmol/L (3σ/k method), indicating a high sensitivity of LR-NH for NO. Based on these results and relevant references [34,35], a corresponding reaction mechanism was proposed in Scheme S2 (Supporting information) to explain the fluorescence changes of LR-NO for NO. The product LR-NO was confirmed by high resolution mass spectrometry (HR-MS) (Fig. S9 in Supporting information), which supports this mechanism.

The influence of pH on the detection of NO was also investigated. As depicted in Fig. S10 (Supporting information), within the pH range of 3.0–11, the F₅₁₅/F₆₂₀ of LR-NH remained stable in the absence of NO. However, after adding NO, the F₅₁₅/F₆₂₀ exhibited a significant increase over a wide pH range (3.0–11) including the physiological pH range. This demonstrates that LR-NH can be utilized under physiological pH conditions.

To assess the cytotoxicity of LR-NH, the MTT method was employed on two cell lines (HeLa and RAW264.7 cells). The results indicated that even if the concentration of LR-NH was used up to 50 µmol/L, the cell survival rate after 24 h incubation was greater than 85% (Fig. S11 in Supporting information), suggesting that LR-NH exhibits low cytotoxicity.

We next evaluated the subcellular targeting ability of LR-NH. As Fig. 3 shows, living cells stained with LR-NH displayed many strong fluorescent red dots, which highly overlap with the fluorescence of the commercial LDs dye Lipid-tracker Green (Pearson’s co-localization coefficient: 0.98). However, it was found that the fluorescence of LR-NH overlapped relatively poorly with other commercial organelle dyes, such as Lyso-tracker Green and Mito-tracker Green (Fig. S12 in Supporting information). These results suggest that LR-NH primarily targets lipid droplets.

Next, LR-NH was explored for NO imaging in living cells. The imaging of exogenous NO in HeLa cells was first studied. As depicted in Fig. 4, the cells treated with LR-NH exhibited strong red fluorescence but very weak green fluorescence (Fig. 4A). However, when the cells were added with the NO donor DEA·NONOate (200 and 500 µmol/L, respectively), the red fluorescence in cells decreased, meanwhile, the green fluorescence increased (Figs. 4B and C). Consequently, the fluorescence ratio of F_green/F_red showed a significant enhancement (from 0.13 to 1.45, Fig. 4D), indicating that LR-NH can effectively detect exogenous NO in living cells.

Then, the imaging of endogenous NO in RAW264.7 cells was conducted. In these experiments, endogenous NO production was stimulated by lipopolysaccharide (LPS) (and LPS + L-Arg) and inhibited by NG-monomethyl-L-arginine (L-NMMA). As shown in Fig. 5, the cells treated with LR-NH only showed strong red fluorescence (control group). However, using LPS or LPS + L-Arg to treat cells and then staining with LR-NH, the red fluorescence inside the cells reduced significantly, while the green fluorescence enhanced significantly. In the L-NMMA treated group, as expected, there was no significant decrease in intracellular red fluorescence and no significant enhancement in green fluorescence. This indicates that endogenous NO can be imaged by LR-NH. With these results, we hence can conclude that LR-NH is a promising new probe that can be used for ratiometric fluorescent detecting endogenous and exogenous NO in LDs.

With the above results, we explored LR-NH to detect the level changes of NO during cell ferroptosis. As indicated in Fig. 6A, the control group displayed significant fluorescence in the red channel. Upon stimulation of cells with erastin to induce ferroptosis, green fluorescence increased while red fluorescence weakened, leading to a significant increase in the F_green/F_red ratio (Fig. 6B). When Fer-1 was added to inhibit the ferroptosis process, the F_green/F_red ratio fell between that of the control group and the erastin group (Fig. 6C). These findings indicate an increase in NO levels during the erastin-induced ferroptosis.

The excellent fluorescence imaging performance of LR-NH in living cells prompted us to further explore its applications in vivo. All the following mice experiments were approved by the Animal Ethical Experimentation Committee of Central China Normal University. Interestingly, when LR-NH was tail vein injected into mice, we found that the liver area showed significant fluorescence enhancement signals within 30 min (Figs. S13A and B in Supporting information). However, when the mice were dissected and their organs were removed for imaging, we found that the fluorescence came from the gallbladder rather than the liver (Figs. S13C and D in Supporting information). This indicates that LR-NH targets the gallbladder. It is worth noting that there are almost no reports in the literature on fluorescent probes that can locate the gallbladder. Therefore, LR-NH is a unique probe that can locate the gallbladder, providing the possibility for imaging the gallbladder and its related diseases. Considering that gallstone disease is very common, we established mouse gallstone models [36] and used LR-NH for imaging. As shown in Fig. S14 (Supporting information), unlike the normal group, obvious gallstone crystallizations were observed in the bile of the gallstone mouse model group, indicating that the gallstone model was successfully established. Then, the normal mice and gallstone disease mice were imaged with LR-NH. As depicted in Fig. S15 (Supporting information), when LR-NH was injected into the mice, both the normal (as control) and the gallstone mice showed obviously increased fluorescence at 620 nm in the gallbladder area, however, the gallstone control group displayed weaker fluorescence compared to the control group. Considering that LR-NH is sensitive to NO, this observation might suggest a higher concentration of NO in the gallstone mice group. However, due to the tissue penetration limitation of short-wavelength green fluorescence, we could not directly observe the changes in green fluorescence in vivo. So, we dissected the mice for imaging at two channels. As shown in Figs. 7A and B, the gallstone group showed weaker fluorescence at 620 nm but stronger fluorescence at 520 nm than the control group. These findings indicate that the level of NO increases in gallstone disease. The main organs were then isolated from the control group and the gallstone group for imaging at both green (520 nm) and red (620 nm) channels. As shown in Figs. 7C and D, indeed, fluorescence is concentrated in the gallbladder. Moreover, compared to the control group, the gallbladder in the gallstone group showed stronger fluorescence at 520 nm and weaker fluorescence at 620 nm. This indicates that LR-NH is a promising probe that can selectively image the gallbladder and detect the changes in NO concentration in the gallbladder.

In conclusion, we have successfully developed LR-NH, an LD-targetable fluorescent probe for ratiometric imaging of NO. LR-NH exhibits high selectivity and sensitivity for NO with remarkable ratiometric fluorescence changes at 515/620 nm and high LD-targeting ability. LR-NH also can detect both endogenous and exogenous NO in LDs and provides a promising new probe to ratiometrically image the level changes of NO during cellular inflammation and ferroptosis. Importantly, LR-NH can target the gallbladder to detect the level changes of NO in mouse gallstone models. It is worth noting that LR-NH is the first fluorescent probe that enables ratiometrically imaging of NO in LDs and gallbladder. Overall, this study provides an unprecedented tool for NO imaging in LDs and gallbladder-related diseases.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Zhoupeng Zheng: Writing – original draft, Methodology, Investigation, Formal analysis, Data curation. Shengyi Gong: Writing – review & editing, Formal analysis, Data curation. Qianhua Li: Writing – review & editing, Investigation, Formal analysis. Shiya Zhang: Writing – review & editing, Formal analysis. Guoqiang Feng: Writing – review & editing, Supervision, Project administration, Funding acquisition, Conceptualization.

This work was supported by the National Natural Science Foundation of China (No. 22077044) and the Natural Science Foundation of Hubei Province (No. 2022CFA033).

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110191.