Profluorescent Fluoroquinolone-Nitroxides for Investigating Antibiotic–Bacterial Interactions

Fluorescent probes are widely used for imaging and measuring dynamic processes in living cells. Fluorescent antibiotics are valuable tools for examining antibiotic–bacterial interactions, antimicrobial resistance and elucidating antibiotic modes of action. Profluorescent nitroxides are ‘switch on’ fluorescent probes used to visualize and monitor intracellular free radical and redox processes in biological systems. Here, we have combined the inherent fluorescent and antimicrobial properties of the fluoroquinolone core structure with the fluorescence suppression capabilities of a nitroxide to produce the first example of a profluorescent fluoroquinolone-nitroxide probe. Fluoroquinolone-nitroxide (FN) 14 exhibited significant suppression of fluorescence (>36-fold), which could be restored via radical trapping (fluoroquinolone-methoxyamine 17) or reduction to the corresponding hydroxylamine 20. Importantly, FN 14 was able to enter both Gram-positive and Gram-negative bacterial cells, emitted a measurable fluorescence signal upon cell entry (switch on), and retained antibacterial activity. In conclusion, profluorescent nitroxide antibiotics offer a new powerful tool for visualizing antibiotic–bacterial interactions and researching intracellular chemical processes.


Introduction
Fluorescent antibiotics can provide valuable insight, often in real time, into interactions between antibiotics and bacterial/host cells. These innovative compounds have aided in elucidating antibiotic modes of action [1], assessing drug toxicity [2], facilitating diagnoses [3], and examining bacterial antimicrobial resistance [4,5]. The development of a fluorescent antibiotic probe generally utilizes one of two methodologies. A fluorophore, such as boron-dipyrromethene (BODIPY) [6], rhodamine [7] or dansyl [8], can be covalently linked to an existing antibiotic to produce a fluorophore-antibiotic conjugate ( Figure 1A). Most fluorescent antibiotic probes are generated via this method; however, this can alter antibiotic binding to cellular targets, reduce antibiotic potency and modify pharmacokinetics [9]. Thus, an alternative approach is to utilize intrinsically fluorescent antibiotics, such as the fluoroquinolones [10] or anthraquinone glycosides [11], as this method negates the challenges associated with tethering a fluorophore to an existing antibiotic. While both are valid methods, neither produce a final product capable of monitoring both antibiotic-bacterial interactions and chemical processes involving free radical and redox reactions. This is important as the induction of free radical and redox processes within bacterial cells following treatment with A PFN consists of a nitroxide moiety covalently attached to a fluorophore and exhibits a substantial suppression of fluorescence (nitroxides are efficient quenchers of excited states) [18][19][20]. Once the free radical of the nitroxide is removed, either by radical trapping or through redox chemistry, fluorescence is restored. PFNs have been successfully utilized for a variety of applications, including the identification of radical-based reaction intermediates [21,22], assessing the oxidative capacity of pollution [23][24][25], investigating polymer degradation [26][27][28], detecting reactive oxygen species in biological systems [29][30][31][32][33], and most recently as bacteriological probes to monitor free radical and redox processes [34].
In this work, we sought to combine the properties of a PFN with a fluorescent antibiotic, in order to create a probe capable of monitoring both antibiotic-bacterial interactions and free radical and redox processes within bacteria. Herein, we present the design, synthesis, photophysical and biological evaluation of profluorescent fluoroquinolone-nitroxides based on the inherent antibacterial and fluorescent properties of the fluoroquinolone core structure.
In this work, we sought to combine the properties of a PFN with a fluorescent antibiotic, in order to create a probe capable of monitoring both antibiotic-bacterial interactions and free radical and redox processes within bacteria. Herein, we present the design, synthesis, photophysical and biological evaluation of profluorescent fluoroquinolone-nitroxides based on the inherent antibacterial and fluorescent properties of the fluoroquinolone core structure.
With the new FNs 14-16 and their corresponding FMs 17-19 in hand, we proceeded to evaluate their photophysical properties. All FNs 14-16 and their corresponding FMs 17-19 displayed absorbance spectra, fluorescence spectra and extinction coefficients (Table 1) characteristic of fluoroquinolones [44]. However, FNs 15 and 16, and FMs 18 and 19, which contained aromatic isoindoline-based functionality, exhibited substantially reduced quantum yields (ՓF) (>10-fold lower) when compared to the compounds FN 14 and FM 17 (piperidine-based functionality). This reduction in fluorescence potentially arises from a disruption to the delocalized π-electron system of the quinolone core, by the amine linked aromatic system of the isoindoline. Consequently, these results indicate that the addition of an aromatic ring via an amine linkage to the fluoroquinolone core at the C-7 position negatively impacts the fluorescent intensity of the fluorophore. In addition to the FNs 14-16, their corresponding fluoroquinolone-methoxyamine (FM) derivatives 17-19 were also synthesized, to examine the specific effect of the nitroxide moiety on fluoroquinolone fluorescence suppression. Furthermore, they could also serve as controls in assessing any antibacterial activity of the nitroxide moiety, and enable the intermediates to be well characterized by NMR spectroscopy (nitroxides are paramagnetic and typically display significantly broadened NMR spectroscopy signals). Utilizing Fenton conditions, the nitroxides 2-4 were reacted with hydrogen peroxide, iron (II) sulfate heptahydrate and DMSO to furnish methoxyamines 5-7 in moderate to excellent yield (77-95%). Subsequent amination, followed by base mediated deprotection, as described previously, afforded FMs 17-19 in high yield (81-87%) (Scheme 1).
With the new FNs 14-16 and their corresponding FMs 17-19 in hand, we proceeded to evaluate their photophysical properties. All FNs 14-16 and their corresponding FMs 17-19 displayed absorbance spectra, fluorescence spectra and extinction coefficients (Table 1) characteristic of fluoroquinolones [44]. However, FNs 15 and 16, and FMs 18 and 19, which contained aromatic isoindoline-based functionality, exhibited substantially reduced quantum yields (ΦF) (>10-fold lower) when compared to the compounds FN 14 and FM 17 (piperidine-based functionality). This reduction in fluorescence potentially arises from a disruption to the delocalized π-electron system of the quinolone core, by the amine linked aromatic system of the isoindoline. Consequently, these results indicate that the addition of an aromatic ring via an amine linkage to the fluoroquinolone core at the C-7 position negatively impacts the fluorescent intensity of the fluorophore. A comparison of the fluorescence arising from solutions of FNs 14-16 and their corresponding FMs 17-19 in chloroform identified a substantial fluorescence suppression in the presence of the nitroxide moieties (calculated from the ratio between the quantum yields of the corresponding methoxyamine and nitroxide conjugates) ( Table 1). Fluorescence suppression was greatest in FNs 15 and 16 (67.5-and 75.0-fold, respectively), both of which bear the isoindoline core. FN 14 also demonstrated significant fluorescence suppression (27.7-fold), albeit lower than the other two nitroxide conjugates (FNs 15 and 16). While these findings confirm that the physical properties of FNs 14-16 are suitable for profluorescent probe applications, these specific PFNs were designed for biological applications, and thus testing was repeated in an aqueous solution. Unfortunately, FNs 15-16 and their FM derivatives 18-19 displayed no measurable fluorescence in the aqueous solution, indicating water-induced fluorescence quenching. Thus, the photophysical properties of these compounds would not be optimal for aqueous biological applications (removal of the nitroxide would restore fluorescence, but no signal would be detectable due to water-induced fluorescence quenching). Conversely, FN 14 and FM 17 not only retained their fluorescence in the aqueous solution ( Figure 2), but FM 17 actually produced a higher fluorescence quantum yield, which resulted in an improved suppression ratio in the aqueous solution compared to chloroform (Table 1). Subsequently, FN 14 and FM 17 were identified as possessing optimal photophysical properties for antibiotic-bacterial interaction studies.  A comparison of the fluorescence arising from solutions of FNs 14-16 and their corresponding FMs 17-19 in chloroform identified a substantial fluorescence suppression in the presence of the nitroxide moieties (calculated from the ratio between the quantum yields of the corresponding methoxyamine and nitroxide conjugates) ( Table 1). Fluorescence suppression was greatest in FNs 15 and 16 (67.5-and 75.0-fold, respectively), both of which bear the isoindoline core. FN 14 also demonstrated significant fluorescence suppression (27.7-fold), albeit lower than the other two nitroxide conjugates (FNs 15 and 16). While these findings confirm that the physical properties of FNs 14-16 are suitable for profluorescent probe applications, these specific PFNs were designed for biological applications, and thus testing was repeated in an aqueous solution. Unfortunately, FNs 15-16 and their FM derivatives 18-19 displayed no measurable fluorescence in the aqueous solution, indicating water-induced fluorescence quenching. Thus, the photophysical properties of these compounds would not be optimal for aqueous biological applications (removal of the nitroxide would restore fluorescence, but no signal would be detectable due to water-induced fluorescence quenching). Conversely, FN 14 and FM 17 not only retained their fluorescence in the aqueous solution ( Figure 2), but FM 17 actually produced a higher fluorescence quantum yield, which resulted in an improved suppression ratio in the aqueous solution compared to chloroform (Table 1). Subsequently, FN 14 and FM 17 were identified as possessing optimal photophysical properties for antibioticbacterial interaction studies. To assess the dynamic emission range of FN 14, the reduction of FN 14 with an excess (1000 equivalents) of sodium ascorbate (Scheme 2), chosen due to its aqueous solubility, was conducted  To assess the dynamic emission range of FN 14, the reduction of FN 14 with an excess (1000 equivalents) of sodium ascorbate (Scheme 2), chosen due to its aqueous solubility, was conducted and monitored via fluorescence spectroscopy. Following treatment of FN 14 with sodium ascorbate in aqueous solution (distilled water, pH 7), reduction of the nitroxide (quenched species) to the corresponding hydroxylamine (fluorescent species), resulted in a steady increase in the fluorescence emission over time (measured every minute for 30 minutes) ( Figure 3). Importantly, complete restoration of fluorescence was achieved (>99% switch on) compared to methoxyamine derivative 17 at the same concentration after approximately 30 minutes. This result suggests that the fluorescence quantum yield of FM 17 and the corresponding hydroxylamine derivative 20 of FN 14 are similar, and thus FM 17 is a true and accurate fluorescent control for FN 14. Furthermore, the fact that the fluorescence of FN 14 was completely restored by removal of the free radical nitroxide highlights its profluorescent nature and demonstrates its potential value as a biological probe for visualizing free radical and redox processes. Furthermore, the fact that the fluorescence of FN 14 was completely restored by removal of the free radical nitroxide highlights its profluorescent nature and demonstrates its potential value as a biological probe for visualizing free radical and redox processes.  With the intended use of FN 14 and FM 17 in biological systems, we pondered the effect of pH on the fluorescence emission of FN 14 and FM 17. Thus, we decided to examine the effect of pH on fluorescence intensity for FM 17 between the pH range of 0-14 ( Figure 4). FM 17 demonstrated fluorescence intensity stability between pH 6-12, supporting its suitability as a bacteriological probe. Interestingly, the fluorescence intensity of FM 17 significantly increased (>2.5-fold) at lower pH (<6), indicating that protonation of the heteroatoms within the conjugated system of the fluorophore drastically increases the fluorescence output of the compound. Conversely, at higher pH (>12) the fluorescence of FM 17 was considerably reduced and eventually completely quenched (pH 14). While compounds FN 14 and FM 17 were not specifically designed as fluorescent pH probes, their pHdependent fluorescence could certainly be exploited for this purpose in suitable applications.  Furthermore, the fact that the fluorescence of FN 14 was completely restored by removal of the free radical nitroxide highlights its profluorescent nature and demonstrates its potential value as a biological probe for visualizing free radical and redox processes.  With the intended use of FN 14 and FM 17 in biological systems, we pondered the effect of pH on the fluorescence emission of FN 14 and FM 17. Thus, we decided to examine the effect of pH on fluorescence intensity for FM 17 between the pH range of 0-14 ( Figure 4). FM 17 demonstrated fluorescence intensity stability between pH 6-12, supporting its suitability as a bacteriological probe. Interestingly, the fluorescence intensity of FM 17 significantly increased (>2.5-fold) at lower pH (<6), indicating that protonation of the heteroatoms within the conjugated system of the fluorophore drastically increases the fluorescence output of the compound. Conversely, at higher pH (>12) the fluorescence of FM 17 was considerably reduced and eventually completely quenched (pH 14). While compounds FN 14 and FM 17 were not specifically designed as fluorescent pH probes, their pHdependent fluorescence could certainly be exploited for this purpose in suitable applications.  With the intended use of FN 14 and FM 17 in biological systems, we pondered the effect of pH on the fluorescence emission of FN 14 and FM 17. Thus, we decided to examine the effect of pH on fluorescence intensity for FM 17 between the pH range of 0-14 ( Figure 4). FM 17 demonstrated fluorescence intensity stability between pH 6-12, supporting its suitability as a bacteriological probe. Interestingly, the fluorescence intensity of FM 17 significantly increased (>2.5-fold) at lower pH (<6), indicating that protonation of the heteroatoms within the conjugated system of the fluorophore drastically increases the fluorescence output of the compound. Conversely, at higher pH (>12) the fluorescence of FM 17 was considerably reduced and eventually completely quenched (pH 14). While compounds FN 14 and FM 17 were not specifically designed as fluorescent pH probes, their pH-dependent fluorescence could certainly be exploited for this purpose in suitable applications.  Following examination of the photophysical properties of FNs 14-16 and FMs 17-19, we proceeded to determine whether these compounds retained antimicrobial activity. Our biological investigations were initiated with the screening of FNs 14-16 and FMs 17-19 in minimum inhibitory concentration (MIC) assays against the common Gram-negative pathogens P. aeruginosa and E. coli, and Gram-positive pathogens S. aureus and E. faecalis (Table 2). FNs 14 and 16 exhibited the highest activity against S. aureus (MIC ≤ 20 µM). Interestingly, their corresponding FMs 17 and 19 both demonstrated no activity against this species (MIC > 1200 µM), suggesting that the presence of the free radical nitroxide may mediate S. aureus antibacterial activity. This same trend was also observed against E. faecalis with FN 16 (MIC ≤ 310 µM) and the corresponding FM 19 (MIC > 600 µM). As this trend was not observed for the Gram-negative species tested (P. aeruginosa and E. coli), it suggests that this activity may be specific against Gram-positive bacteria.
Furthermore, considering that nitroxides possess no inherent antibacterial activity (Supplementary Material, Table S1), the difference between the nitroxide-containing conjugate and its corresponding methoxyamine derivative against Gram-positive species was surprising. While the activity of FNs 14 and 16 was highest against S. aureus, these two conjugates also exhibited Gramnegative antimicrobial activity, with FN 14 being most active against E. coli (MIC ≤ 100 µM) and FN 16 against P. aeruginosa (MIC ≤ 160 µM). The aqueous fluorescent properties of FN 14 combined with its antibacterial activity, suggesting this compound would be useful for monitoring antibioticbacterial interactions in both Gram-positive and Gram-negative bacteria.
[a] All MICs were determined via broth microdilution method in accordance with CLSI standard; [b] Highest concentration tested.
As FN 14 and the corresponding FM 17 demonstrated optimal photophysical and biological properties, these compounds were subsequently evaluated for use as bacteriological probes.   14 ≤770 ≤100 ≤20 [a] All MICs were determined via broth microdilution method in accordance with CLSI standard; [b] Highest concentration tested.
Furthermore, considering that nitroxides possess no inherent antibacterial activity (Supplementary Material, Table S1), the difference between the nitroxide-containing conjugate and its corresponding methoxyamine derivative against Gram-positive species was surprising. While the activity of FNs 14 and 16 was highest against S. aureus, these two conjugates also exhibited Gram-negative antimicrobial activity, with FN 14 being most active against E. coli (MIC ≤ 100 µM) and FN 16 against P. aeruginosa (MIC ≤ 160 µM). The aqueous fluorescent properties of FN 14 combined with its antibacterial activity, suggesting this compound would be useful for monitoring antibiotic-bacterial interactions in both Gram-positive and Gram-negative bacteria.
As FN 14 and the corresponding FM 17 demonstrated optimal photophysical and biological properties, these compounds were subsequently evaluated for use as bacteriological probes.  Figure S1); however, when the concentration of FN 14 was increased to 600 µM, nearly all bacterial cells fluoresced (~90%) ( Figure 5B). A similar pattern was observed with E. coli cells treated with FN 14 ( Figure 5C). Interestingly, this concentration-dependent fluorescence output was not observed in Gram-positive bacteria (S. aureus and E. faecalis). In fact, when either S. aureus or E. faecalis were treated with FN 14 (150 µM), almost every bacterial cell emitted measurable fluorescence (~99%) ( Figure 5D,E, respectively), suggesting that the process by which fluorescence is activated for FN 14 occurs more readily and/or more frequently in Gram-positive species.
aureus, and E. faecalis cells for 90 minutes, then visualized via fluorescence microscopy. FN 14 emitted bright fluorescence upon cell entry in all species tested ( Figure 5B-E). However, FN 14 bacterial cell entry and fluorescence was found to be both concentration and species specific. When FN 14 was administered to P. aeruginosa at 150 µM, very few (~10%) cells fluoresced (Supplementary Material, Figure S1); however, when the concentration of FN 14 was increased to 600 µM, nearly all bacterial cells fluoresced (~90%) ( Figure 5B). A similar pattern was observed with E. coli cells treated with FN 14 ( Figure 5C). Interestingly, this concentration-dependent fluorescence output was not observed in Gram-positive bacteria (S. aureus and E. faecalis). In fact, when either S. aureus or E. faecalis were treated with FN 14 (150 µM), almost every bacterial cell emitted measurable fluorescence (~99%) ( Figures 5 D and E, respectively), suggesting that the process by which fluorescence is activated for FN 14 occurs more readily and/or more frequently in Gram-positive species.
Intriguingly, despite FM 17 being a fluorescence activated derivative of FN 14, and hence always fluorescent, it did not emit a measurable cell-associated fluorescence signal. Instead, fluorescence was only detected in the liquid medium, where FM 17 formed fluorescent aggregates (Figures 5G-J). The lack of any bacterial-associated fluorescence for FM 17 suggests that it either does not enter bacterial cells or its fluorescence is quenched intracellularly. The inability of FM 17-19 to translocate through the bacterial cell envelope could potentially explains their lack of antimicrobial activity against the Gram-positive pathogens S. aureus and E. faecalis (Table 1)  To test the hypothesis that fluorescence activation of FN 14 occurs intracellularly, we examined the fluorescence properties of FN 14 and FM 17 in medium only (no bacterial cells present). Here, we treated the medium with either 150 or 600 µM of FN 14 or FM 17 for 90 minutes. Our findings indicated that FN 14 in medium emits no measurable fluorescence ( Figure 5A) (FN 14 also emitted no measurable fluorescence in PBS, LB, and MH). However, when bacterial cells were present in a medium containing FN 14, they became highly fluorescent while the surrounding medium still exhibited no fluorescence ( Figure 5B-E). Interestingly, in similar assays, FM 17 became immediately fluorescent in medium alone ( Figure 5F) and could be seen to form small aggregates and crystals (Figures 5F-J) that were highly fluorescent. As FN 14 was not visible in medium but clearly visible inside bacterial cells, while FM 17 was only visible in medium, we can conclude that FN 14's  (Table 1). However, this possibility would not explain their activity against Gram-negative pathogens P. aeruginosa and E. coli, where the potency of both the FNs 14-16 and their corresponding FMs 17-19 derivatives is conserved.
To test the hypothesis that fluorescence activation of FN 14 occurs intracellularly, we examined the fluorescence properties of FN 14 and FM 17 in medium only (no bacterial cells present). Here, we treated the medium with either 150 or 600 µM of FN 14 or FM 17 for 90 minutes. Our findings indicated that FN 14 in medium emits no measurable fluorescence ( Figure 5A) (FN 14 also emitted no measurable fluorescence in PBS, LB, and MH). However, when bacterial cells were present in a medium containing FN 14, they became highly fluorescent while the surrounding medium still exhibited no fluorescence ( Figure 5B-E). Interestingly, in similar assays, FM 17 became immediately fluorescent in medium alone ( Figure 5F) and could be seen to form small aggregates and crystals ( Figure 5F-J) that were highly fluorescent. As FN 14 was not visible in medium but clearly visible inside bacterial cells, while FM 17 was only visible in medium, we can conclude that FN 14's fluorescence is activated via an intracellular process, and thus, FN 14 can function as a true intracellular bacteriological probe with the potential to simultaneously monitor antibiotic-bacterial interactions and intracellular free radical and redox processes.
Importantly, FN 14 exhibited a fluorescence signal which did not require background subtraction or correction for bacteria autofluorescence (bacteria autofluorescence was not detected under these conditions). Furthermore, while the experiments reported here utilized an excitation wavelength of 365 nm, FN 14 was also efficiently excited by a 405 nm laser or a multiphoton laser set at 720 nm. Taken together these results demonstrate the utility of FN 14 and support its use as a potential live-cell imaging probe.

General Procedure for the Synthesis of Fluoroquinolone-Nitroxides 14-16 and Fluoroquinolone-Methoxyamines 17-19
Cesium carbonate (3 equiv), palladium acetate (6 mol %), BINAP (10 mol %), Q-Ester 1 (2 equiv) and the specific primary amine (1 equiv) were added to a Schlenk vessel under an atmosphere of argon. THF (60 mL), which had been degassed with argon, was then added. The vessel was sealed and heated at 65 • C for 72 hours. The reaction was allowed to cool to room temperature, and the solvent was removed via rotary evaporation. The resulting residue was washed three times with aqueous hydrochloric acid (2 M, 3 × 20 mL) and the combined filtrates were extracted with diethyl ether (3 × 10 mL). The aqueous phase was neutralized with saturated sodium carbonate and extracted with dichloromethane (3 × 50 mL). The combined extracts were dried over anhydrous sodium sulfate and the solvent removed in vacuo. Purification was achieved via column chromatography (SiO 2 , chloroform 98%, methanol 2%).

General Procedure for the Synthesis of FNs 14-19 via Base Mediated Ester Hydrolysis
Aqueous sodium hydroxide (2 M, 7 equiv) was added to a solution of the specific ethyl ester (1 equiv) in HPLC grade methanol (50 mL), and the resulting solution was stirred at 50 • C for 5 hours.
The reaction mixture was cooled to room temperature and diluted with deionized water (50 mL). The pH was adjusted to approximately 6 using aqueous hydrochloric acid (2 M) and the mixture extracted with dichloromethane (3 × 20 mL). The combined organic extracts were dried over anhydrous sodium sulfate, and the solvent was removed in vacuo. Purification was achieved via column chromatography (SiO 2 , chloroform 98%, methanol 2%).

Fluorescence Quantum Yield and Extinction Coefficient Calculations
Quantum yield efficiencies of fluorescence for compounds 14-19 were obtained from measurements at five different concentrations in water, ethanol, or chloroform using the following equation: where A and F denote the absorbance and fluorescence intensity, respectively, Σ[F] denotes the peak area of the fluorescence spectra, calculated by summation of the fluorescence intensity, and η denotes the refractive index of the solvent (chloroform = 1.444, ethanol = 1.362, water = 1.000). Anthracene (Φ F = 0.27 in ethanol) was used as the standard. Extinction coefficients were calculated from the obtained absorbance spectra. This is a standardised method and our values are consistent with the values reported for other profluorescent nitroxides using the same method [17,30,48].

Fluorescence Spectroscopy Measurements: Reduction of FN 14 with Sodium Ascorbate
Sodium ascorbate (200 µM solution in water, 0.5 mL), was added to a solution of FN 14 (2 µM solution in water, 0.5 mL) in a 4-sided quartz cuvette, equipped with a magnetic stirrer bar. The resulting solution (1 µM of FN 14 1 equiv and 1000 µM sodium ascorbate 1000 equiv) was placed in the fluorescence spectrophotometer, equipped with a magnetic stirrer, and measurements were recorded every minute for 30 mins. The measurement of the blank sample (time = 0 min) was conducted similarly by adding water (0.5 mL) to a solution of FN 14 (2 µM solution in water, 0.5 mL).

Evaluating the Effect of pH on the Fluorescence Intensity of FM 17
Fifteen aqueous solutions (ranging from pH 0 to pH 14) containing the same concentration (500 µM) of FM 17 were prepared. Each respective solution was then analysed via fluorescence spectrophotometery (λ ex = 340 nm), and the total fluorescence area was calculated.

MIC Susceptibility Assays for Compounds 14-19
The MIC for each fluoroquinolone-based adduct 14-19 were determined by the broth microdilution method, in accordance with the 2015 (M07-A10) Clinical and Laboratory Standards Institute (CLSI). In a 96-well plate, twelve two-fold serial dilutions of each compound were prepared to a final volume of 100 µL in MH medium. At the initial time of inoculation, each well was inoculated with 5 × 10 5 bacterial cells, which had been prepared from fresh overnight cultures in MH. The MIC for a compound was defined as the lowest concentration of an agent that prevented visible bacterial growth after 18 hours of static incubation at 37 • C (MIC values were also confirmed by spectrophotometric analysis at OD 600 ). Compounds 14-19 were tested between the concentration ranges of 1200 to 0.6 µM. Working solutions of compounds 14-19 were prepared in MH medium that had been inoculated with bacteria at approximately 5 × 10 6 CFU mL −1 . Negative controls containing DMSO at the highest concentration required to produce a 1200 µM final concentration for compounds 14-19 were also prepared and serially diluted (12 dilutions total) in the same method as the antimicrobial agents. MIC values for compounds 14-16 were obtained from at least two independent experiments, each consisting of at least three biological replicates and at least two technical replicates of each biological replicate.

Fluorescence Microscopy of Bacterial Cells Treated with FN 14 or FM 17
Overnight bacterial cultures in LB (10 mL) were concentrated by centrifugation at 3000× g for 5 minutes. Cell pellets were washed twice in 10 mL saline (0.9%) and resuspended to approximately 10 9 CFU mL −1 in saline (0.9%). Cell suspensions were treated with FN 14 or FM 17 (150 or 600 µM) for 1.5 hour at 37 • C. Wet mounts (5 µL) of treated cell suspensions were prepared and immediately analyzed by fluorescence microscopy.
FNs 14-16 all exhibited substantially suppressed fluorescence in the presence of the nitroxide moiety (FN 14 36.7-fold, FN 15 67.5-fold, and FN 16 75-fold) when compared to their corresponding FMs 17-19. However, the photophysical properties of FN 14 were determined to be optimal for biological probe applications. We showed that FN 14 permeated several different bacterial species (both Gram-positive and Gram-negative) and fluoresced brightly upon bacterial cell entry exemplifying its potential as an intracellular bacteriological probe.
The experiments presented here have demonstrated that profluorescent antibiotic nitroxides such as FN 14 possess both desirable fluorescent antibiotic and profluorescent nitroxide probe capabilities. Furthermore, we have successfully produced a novel tool for simultaneously monitoring antibiotic-bacterial interactions and intracellular free radical and redox processes.
Supplementary Materials: The following are available online at http://www.mdpi.com/2079-6382/8/1/19/s1. Table S1: Measured MIC values for TEMPO, TMIO, and TEIO against Gram-positive P. aeruginosa and E. coli, and Gram-negative S. aureus and E. Faecalis. Figure S1: Fluorescent and brightfield overlay micrographs images of bacterial cells treated with FN 14 or FM 17, and 1 H NMR spectra and 13 C NMR spectra, and LC-MS chromatograms and HRMS spectra are provided for all novel compounds.