Imaging Acetaldehyde Formation During Ethanol Metabolism in Living Cells using a Hydrazinyl Naphthalimide Fluorescent Probe

Abstract

The metabolism of ethanol to acetaldehyde has been visualized in living lung epithelial cells using a hydrazinyl naphthalimide fluorescent probe. Utilizing a condensation reaction between carbonyls and a hydrazine moeity, we demonstrate that the fluorescent probe (Aldehydefluor-1) AF1 reacts with a range of reactive carbonyl species including formaldehyde, acetaldehyde, glyoxylic acid, and methyl glyoxal. With AF1, it is possible to directly visualize endogenous carbonyl metabolites. Here, we have applied it towards the visualization of acetaldehyde generated from alcohol dehydrogenase mediated ethanol metabolism, validating it as a useful tool to study the roles of alcohol in respiratory disease and other pathological mechanisms.

Acetaldehyde plays a notorious role as a toxic intermediate in the enzymatic metabolism of ethanol.^1,2,3 The undesired side effects of this metabolism – headache, nausea, dry mouth, vertigo, and sensitivity to light and sound – mark the well-known symptoms of a hangover. The lungs are particularly vulnerable to ethanol-mediated tissue damage due to repeated exposure to ethanol after ingestion.⁴ Mammalian lungs express class I alcohol dehydrogenase^5,6 and ethanol metabolism can potentially proceed via oxidative or non-oxidative mechanisms.⁷ Heavy ethanol consumption has been implicated as a risk factor in chronic obstructive pulmonary disease (COPD) and the term "whiskey bronchitis" has been used to describe this ethanol-mediated etiology.⁸ Interestingly, ethanol can both relieve⁹ and exacerbate¹⁰ acute and chronic airway obstruction, depending on factors such as the mode of delivery and genetics. Studies have demonstrated that administration of alcohol to patients with asthma results in significant bronchodilation.¹¹ However, it has also been demonstrated that acetaldehyde, the first product of ethanol metabolism, initiates bronchoconstriction.¹⁰ There exists a clear, but complicated, relationship between ethanol metabolism and respiratory ailments such as asthma.

Despite progress, studying lung epithelial ethanol metabolism remains challenging due to a lack of methods for the direct detection of metabolic products in living models. Measurements in mammalian lung tissue homogenates show a steep dependence on experimental conditions, particularly pH.^12,13 As a result, the precise biochemical events that lead to these pathologies remain unclear. This is mainly due to the furtive nature of acetaldehyde, necessitating careful sample processing and detection methods such as spectrophotometry, nuclear magnetic resonance imaging (NMR), or gas chromatography mass-spectrometry (GC-MS), which typically result in processing and destruction of the sample.¹⁴ The use of carbonyl-responsive fluorescent probes^15–21 could potentially circumvent these problems by allowing for the real-time, unobtrusive visualization of acetaldehyde as it is produced in living cells.

Inspired by a previous study,²² we anticipated that a hydrazinyl naphthalimide-based fluorescent probe would react with aldehydes to yield hydrazone products, providing a fluorescence response that would increase with higher levels of aldehydes. If combined with the proper controls, this probe could be useful for tracking enzymatically generated acetaldehyde in living cells. With this goal in mind, the fluorescent probe AF1 was synthesized from the commercially available starting material 4-bromo-1,8-naphthalimide, a fluorescent naphthalic scaffold (Scheme 1). Introduction of a hydrazine moiety using a nucleophilic aromatic substitution reaction provides AF1, which displays quenched fluorescence likely due to photoinduced electron transfer (PET) quenching. Reaction of AF1 with acetaldehyde results in the formation of the hydrazone structure, which was confirmed via ¹H NMR and high resolution mass spectrometry (Fig. S5).

Scheme 1.

Synthesis and reaction-based acetaldehyde detection of AF1.

After synthesizing AF1, the purified crystalline product was aliquoted into individual Eppendorf tubes and kept at −20 °C until use. For initial experimentation, an aliquot of AF1 was removed from storage at −20 °C and allowed to thaw before addition of enough dimethyl sulphoxide (DMSO) to make a 2.5 mM solution. This dissolved aliquot was then refrozen until the next experiment, and this freeze-thaw cycle repeated until no aliquot remained. However, a significant decrease in fluorescent turn-on was observed after a single freeze-thaw cycle of a given dissolved AF1 aliquot (Fig. S6), indicating the importance of using a fresh aliquot for experiments. This reduction in signal over time may help explain differences in selectivity of fluorescent probes with similar chemical structures.¹⁹ All further experiments were run with AF1 aliquots dissolved in DMSO the same day of experimentation.

Upon reacting 10 µM AF1 with 200 µM acetaldehyde in a buffered aqueous system, a 13-fold increase in fluorescence emission at 551 nm was observed after 60 minutes (Fig. 1). Selectivity studies were performed with 10 µM AF1 and 100 µM of biologically relevant reactive carbonyls, reducing sugars, and selected sulphur, oxygen, and nitrogen species (Fig. 2). In addition to the fluorescence response towards acetaldehyde, we note that a significant turn-on was observed for formaldehyde, glyoxylic acid, and methyl glyoxal, indicating that this class of hydrazinyl-based aldehyde probes is not completely selective for a single reactive carbonyl species. The increased response for formaldehyde agrees with previous studies showing that formaldehyde is more effective at trapping hydrazine groups than more substituted carbonyls like acetone²³ and follows known reactivity trends of aldehydes and ketones with aryl hydrazine molecules.²⁴

Figure 1.

Fluorescence response after treating 10 µM AF1 with 200 µM CH₃CHO in 20 mM HEPES buffer (pH 7.4) containing 0.2% DMSO for 0, 1, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, and 60 minutes. λ_ex = 440 nm.

Figure 2.

Fluorescence response of 10 µM AF1 to biologically relevant reactive carbonyl species and selected reactive sulphur, oxygen, and nitrogen species in 20 mM HEPES buffer (pH 7.4) containing 0.2% DMSO. Bars represent fold turn-on at 550 nm acquired 1, 5, 10, 15, 20, 25, and 30 minutes after addition of the analyte. Data were acquired using λ_ex = 440 nm. Error bars are ± S.D. Legend (all 100 µM except glutathione at 5 mM): (1) blank; (2) acetaldehyde, (3) formaldehyde; (4) 4-hydroxynonenal; (5) glyoxal; (6) glyoxylic acid; (7) methyl glyoxal; (8) sodium pyruvate; (9) α-ketoglutaric acid; (10) glutathione, 5 mM; (11) DEA NONOate; (12) Angeli’s salt; (13) fructose; (14) glucose.

Having characterized the response and selectivity, the fluorescent probe AF1 was then characterized in living human lung epithelial cells (A549) using an EVOS-fl fluorescence microscope equipped with a GFP filter cube. First, the ability of AF1 to image exogenous acetaldehyde was examined. Only low fluorescent intensity was observed upon addition of AF1 to A549 cells (Fig. 3a), with a 2-fold increase in fluorescent response upon addition of 100 µM acetaldehyde (Fig. 3b) and a 5-fold increase upon addition of 1 mM acetaldehyde (Fig. 3c). This trend clearly indicates the ability of AF1 to report on exogenous acetaldehyde in a cellular environment.

Figure 3.

Fluorescence microscopy images of exogenous acetaldehyde detection in live A549 cells using AF1. (a) A549 cells incubated with 10 µM AF1 for 60 minutes at 37 °C. (b) A549 cells incubated with 10 µM AF1 for 60 minutes at 37 °C with 100 µM acetaldehyde added for the final 30 minutes. (c) A549 cells incubated with 10 µM AF1 for 60 minutes at 37 °C with 1 mM acetaldehyde added for the final 30 minutes. (d)–(f) Brightfield images of the fields in (a)–(c). (g) Quantification of cellular imaging experiments. Error bars are S.E. Values are the average of n = 16 fields of cells from 5 biological replicates.

Next, we analysed the efficacy of AF1 as a tool in the measurement of endogenous acetaldehyde, specifically as a result of alcohol dehydrogenase mediated metabolism of ethanol.⁷ Primary lung epithelial cells display significant biological responses to ethanol upon incubation between 2 and 6 hours.^25–27 Therefore, we decided to utilize AF1 to monitor dose-dependent acetaldehyde formation with 2-hour incubations with increasing concentrations of ethanol. We observed an increase in fluorescence emission with increasing ethanol concentration, which reached a maximum and statistically significant (p < 0.01) value at 50 mM ethanol (Fig. 4). In order to confirm that the observed signal was due to the activity of alcohol dehydrogenase, the selective class I alcohol dehydrogenase inhibitor fomepizole (FPZ) was employed (Fig. 5).^28,29 A comparison of cells incubated with solely ethanol (Fig. 5b) to cells incubated with both ethanol and FPZ (Fig. 5c) showed a visible and significant (p < 0.05) decrease in overall fluorescent intensity in the cells allowed to co-incubate with FPZ. This key experiment reveals a decrease in the metabolism of ethanol to acetaldehyde, indicating that signal from AF1 under these conditions is predominately due to cellular acetaldehyde formation.

Figure 4.

Fluorescence microscopy images of acetaldehyde generated during ethanol metabolism in live A549 cells using AF1. (a) A549 cells incubated with 10 µM AF1 for 180 minutes at 37 °C. (b) A549 cells incubated with 10 µM AF1 for 180 minutes at 37 °C with 25 mM ethanol added for the final 150 minutes. (c) A549 cells incubated with 10 µM AF1 for 180 minutes at 37 °C with 50 mM ethanol added for the final 150 minutes. (d)–(f) Brightfield images of the fields in (a)–(c). (g) Quantification of cellular imaging experiments. Error bars are S.E. Control values are n = 34 fields of cells from 8 biological replicates, 50 mM ethanol values are averages from n = 34 fields of cells from 8 biological replicates, and other values are averages of n = 9–10 fields of cells from 3 biological replicates. Statistical significance was assessed using a two-tailed student's t-test. ** p < 0.01.

Figure 5.

Fluorescence microscopy images of the inhibition of acetaldehyde generated during ethanol metabolism in live A549 cells using AF1. (a) A549 cells incubated with 10 µM AF1 for 180 minutes at 37 °C. (b) A549 cells incubated with 10 µM AF1 for 180 minutes at 37 °C with 50 mM ethanol added for the final 150 minutes. (c) A549 cells incubated with 10 µM AF1 for 180 minutes at 37 °C with 50 mM ethanol and 1 mM fomepizole added for the final 150 minutes. (d)–(f) Brightfield images of the fields in (a)–(c). (g) Quantification of cellular imaging experiments. Error bars are S.E. Control values are n = 34 fields of cells from 8 biological replicates, 50 mM ethanol values are averages from n = 34 fields of cells from 8 biological replicates, 50 mM EtOH + FPZ values are the average of n = 21 fields of cells from 4 biological replicates. Statistical significance was assessed using a two-tailed student's t-test. ** p < 0.01, * p < 0.05.

Based on these results, the following conclusions can be made: (1) The hydrazinyl naphthalimide scaffold provides a fluorescence response to a range of aldehydes including acetaldehyde, formaldehyde, glyoxylic acid, and methyl glyoxal, (2) aliquots must be used shortly after preparation to maintain a reliable response, and (3) AF1 can be used to detect cellular generated acetaldehyde from ethanol metabolism in A549 cells. Similar hydrazinyl naphthalimide scaffolds have been used for detecting other reactive carbonyl species,^19,22 highlighting the need to carefully interpret results in light of appropriate controls. In this study, fomepizole was employed as an inhibitor for alcohol dehydrogenase, indicating that the metabolism of ethanol to acetaldehyde is responsible for the probe's response. While a fluorescent probe with higher selectivity for acetaldehyde would be preferred, this study provides a singular example of acetaldehyde imaging in living cells and should serve as a useful method for researchers in the field.