Abstract
Near-infrared (NIR) dyes are desirable for biological imaging applications including photoacoustic (PA) and fluorescence imaging. Nonetheless, current NIR dyes are often plagued by relatively large molecular weights, poor water solubility, and limited photostability. Herein, we provide the first examples of azaphosphinate dyes which display desirable properties such as low molecular weight, absorption/emission above 750 nm, and remarkable water solubility. In PA imaging, an azaphosphinate dye exhibited a 4.1-fold enhancement in intensity compared to commonly used standards, the ability to multiplex with existing dyes in whole blood, imaging depths of 2.75 cm in a tissue model, and contrast in mice. An improved derivative for fluorescence imaging displayed a > 10-fold reduction in photobleaching in water compared to the FDA-approved indocyanine green dye and could be visualized in mice. This new dye class provides a robust scaffold for the development of photoacoustic or NIR fluorescence imaging agents.
Keywords: Fluorescence, Fluorescent probes, Imaging agents, Near-Infrared dyes, Photoacoustic probes
Graphical Abstract
We present a new class of NIR dyes termed azaphosphinate NR dyes with desirable properties including low molecular weight, absorption/ emission above 750 nm, and remarkable water solubility. Derivatives of these dyes can be used for photoacoustic or fluorescence imaging, providing a new NIR scaffold for the design of imaging probes.

Introduction
Small molecule dyes with absorption and emission maxima in the 680–980 nm range of the NIR are highly desirable for bioimaging applications due to decreased background from endogenous chromophores and increased tissue penetration in this wavelength range.[1,2,3,4] Although several efforts have identified promising new NIR scaffolds,[5,6,7,8,9,10,11,12,13] current dyes within the 680–980 nm range often suffer from relatively high molecular weights, poor water solubility, and decreased chemical and photostability.[14,15] Indeed, the classic FDA-approved dyes within this wavelength range, indocyanine green (ICG) and methylene blue (MB, Figure S1), suffer from these issues. For example, sulfonation is required to increase the water solubility of ICG. However, this modification does not entirely suppress aggregation of this dye. Moreover, the flexible nature of the polymethine chromophore within this dye renders it susceptible to excited state photochemistry that limits dye photostability.[16] On the other hand, MB is a relatively low molecular weight thiazine dye scaffold with emission tailing into the NIR region. Although this dye has been widely used for imaging purposes,[17,18,19,20,21,22] its tendency to form H-type dimers in aqueous solution[23,24,25,26,27] limits its utility for imaging applications. Thus, there is a need for continued development of next generation NIR dyes with improved properties. The availability of such dyes will ultimately improve capabilities in emerging areas of biomedical imaging.
Photoacoustic (PA) and NIR fluorescence imaging represent two such emerging imaging techniques. In the case of PA imaging, pulsed NIR excitation of a dye results in the production of pulsed sound waves through the PA effect.[28] As opposed to fluorescence imaging, PA provides increased imaging depth due to reduced scattering of sound in tissues.[1,3,29] Although recent efforts have identified elegant approaches to enable turn-on (or acoustogentic) signal generation from small molecule PA probes,[30,31,32,33] PA imaging is still limited by the relative lack of diversity in robust scaffolds that can be used to design PA probes. Alternatively, NIR fluorescence imaging relies on monitoring the emission of NIR absorbing and emitting fluorophores, providing increased tissue penetration relative to dyes in the visible spectral region. Since both PA and NIR fluorescence imaging rely on the availability of robust NIR absorbing dyes, both techniques are limited by the previously mentioned issues of aggregation, relatively large molecular weight, and reduced chemical and photostability of NIR dyes.[14,15] Consequently, chemistry-focused efforts to expand the toolbox of low molecular weight, robust NIR dyes are needed to ultimately advance PA as well as fluorescence imaging.
While ICG and MB have been FDA-approved for over 50-years and extensively utilized in both PA and NIR fluorescence imaging, probe development has been hindered by the inherent photophysical and chemical limitations of these dyes. Notably, the structure of MB aggregates have been extensively studied and indicates that an intermolecular sulfur-nitrogen dipole-dipole interaction and π-π stacking of the chromophore leads to dimer formation (Figure 1a).[34,35,36] Given the low molecular weight and desirable chemical and photostability of MB, we were intrigued by the idea of suppressing aggregation of this scaffold while simultaneously red-shifting excitation and emission into the NIR range.
Figure 1.
a) A previously reported structure of MB dimers in aqueous solution. b) The core azaphosphinate dye from this work. Phosphinate substitution inhibits dye aggregation.
Substitution of the bridging oxygen atom in xanthene scaffolds with relatively electron-withdrawing functionalities such as silicon[37,38,39] and phosphorus[40,41,42] substituents results in a decrease in HOMO-LUMO energy gaps and a substantial red-shift in absorbance and emission maxima of the resulting dyes. Our lab has previously developed a series of phosphinate-containing xanthenes, termed Nebraska Red (NR) dyes.[40,41] These dyes display > 100 nm red-shifts in absorbance and emission compared to the parent dyes, and can be used for various applications in chemical biology including self-reporting cargo delivery, targeted toxicity in cancer cells, and imaging.[41,43,44] More recently, the Miller group has extended the approach of heteroatom substitution to oxazine dyes, which are structurally related to MB, yielding dimethylsilyl-containing azasiline dyes.[45] The resulting azasiline dyes are red-shifted by 57–83 nm relative to the parent oxazine. This seminal work clearly demonstrates the potential to red-shift thiazine-type scaffolds using the bridging atom replacement strategy. However, an increase in the quantum yield of these dyes in aqueous solutions was observed upon detergent addition, indicating that the dimethylsilyl group does not fully inhibit aggregation of this scaffold. Inspired by this work, we hypothesized that the sp3 hybridized and negatively charged phosphinate functionality may result in more potent disruption of dye aggregation in this family (Figure 1b). Herein, we show that phosphinate substitution in the MB scaffold virtually abolishes aggregation in aqueous samples. Furthermore, the resulting azaphosphinate dyes display absorbance and emission above 750 nm and improved photostability compared to ICG. Derivatives of this new scaffold provide either PA or fluorescence readouts in living cells and mice. Thus, this robust, low molecular weight scaffold provides a new template for construction of improved PA and fluorescence imaging probes.
Results and Discussion
Design and Synthesis of Azaphosphinate Dyes.
Work from the Miller group indicated that the inclusion of methyl groups at the 4 and 6 positions of azasiline dyes increased their stability.[45] Accordingly, we envisioned the construction of a tetramethyl azaphosphinate NR derivative (Scheme 1). Before embarking on a synthetic campaign, we utilized recently published trendlines from the calibration of computational estimations of absorbance and emission to experimental parameters for xanthene-based NR dyes.[46] This calibrated computational data enables rapid prediction of NIR dye absorbance and emission with mean percent errors of 2.2 and 2.8%, respectively. Using this approach, we predicted absorbance and emission maxima of 715 and 732 nm, respectively, for NR751 (see Supporting Information). Confident that phosphinate substitution would induce a substantial red-shift in excitation and emission, we began the synthesis of this new dye by reductive amination with 3-bromo-5-methylaniline yielding the dimethylaniline intermediate 1. Lithiation of dimethylaniline 1 and subsequent treatment with ethyl dichlorophosphate furnished the phosphinate ester 2. Lastly, nitrosation-cyclization[47] and simultaneous ester cleavage yielded the final dye.
Scheme 1.
Synthesis of NR751.
Photophysical Characterization of NR751.
Next, we measured the photophysical properties of our new NR dye. Gratifyingly, in PBS the dye showed an absorbance and emission maxima of 751 and 779 nm, which is 87 and 94 nm red-shifted compared to MB, respectively (Figure 2a and b). Moreover, these experimental values correspond to 4.8 and 6% errors in computationally predicted absorbance and emission, respectively, and indicate the ability to utilize our experimentally calibrated computational dataset[46] to predict absorbance and emission of structurally diverse dyes. Based on the absorbance maxima of this dye, we named it NR751 to be consistent with previous NR dyes.[40,41] Next, we measured the molar extinction coefficient (ε), quantum yield (Φ), and fluorescence lifetime (τ) of NR751 in PBS (Figure 2b). NR751 exhibited a low quantum yield of 0.4%, indicating significant non-radiative relaxation from the excited state. Since PA signal generation relies on non-radiative relaxation to induce local pressure increases through thermal expansion events,[30,33,48] we surmised that NR751 may be a useful PA agent. In support of this notion, we calculated the nonradiative decay rate (knr) for NR751 using the observed Φ and τ.[49] Indeed, NR751 exhibits a rapid nonradiative decay rate of 1.42×1010 s−1 which is 5.4-fold faster than MB (2.63×109 s−1). These results indicated that NR751 may show promise as a PA agent.
Figure 2.
a) Normalized absorption and emission spectra of NR751 in PBS. b) Photophysical properties of NR751 in PBS.
Water Solubility of NR751.
Prior to the analysis of PA signal generation by NR751, we investigated the ability of the phosphinate functionality to suppress aggregation in NR751. Gratifyingly, the absorbance of increasing concentrations of NR751 in PBS yielded a linear relationship up to at least 200 μM (Figure 3a). Alternatively, the absorbance of MB showed a clear deviation from linearity with increasing concentration (Figure 3a), a characteristic signature of aggregation in MB and other small molecule dyes.[23,24,50,51] Moreover, solutions containing increasing concentrations of MB displayed increased intensity of a blue-shifted absorbance band at 605 nm, which has been assigned to H-aggregates of MB (Figure 3b).[50] Addition of 1% Tween-20 suppressed the 605 nm absorbance peak of MB while the absorption spectra of NR751 remained virtually unchanged under identical conditions (Figure S2). These experiments demonstrate that the phosphinate functionality can red-shift the absorbance and emission of azaphosphinate NR dyes, and effectively abolish aggregation in relevant concentration regimes used for imaging.
Figure 3.
a) The relationship between concentration and absorbance for MB and NR751 in PBS. b) Normalized absorbance spectra for increasing concentrations of NR751 and MB in PBS.
Solvatochromic Nature of NR751.
To investigate the solvatochromic properties of NR751, we measured its photophysical properties in a panel of solvents (Figure S3). Interestingly, a solvent-dependent shift in absorbance and emission maxima was observed (Figure 4a and b). Peak absorbance and emission ranged from 727–751 nm and 743–772 nm, respectively, and correlated with the solvent polarity parameter ET(30)[52] (Figure 4c). The brightest fluorescence emission (defined as ε x Φ) was observed in dichloromethane (35-fold increase compared to water, Figure 4d and e). These results indicate the potential of NR751 as a solvatochromic indicator.
Figure 4.
Normalized absorbance (a) and emission (b) spectra of NR751 in the indicated solvents. c) The correlation between either NR751 absorbance (red) or emission (blue) maxima in wavenumber versus ET(30) for each solvent. d) Fluorescence brightness of NR751 in the indicated solvent. e) Photophysical properties of NR751 in the indicated solvent.
PA Signal Generation from NR751.
Having fully characterized the spectral and physical properties of NR751, we evaluated the ability of this new dye to produce PA signal. PA signal from PBS solutions containing increasing concentrations of NR751 in a tissue-mimicking phantom consisting of 5% agarose and 2.5% milk were linear to at least 50 μM (Figure S4). As anticipated, the PA signal from NR751 overlaid with the absorbance profile for the dye, producing maximal signal at 750 nm (Figure 5a and b). Next, we compared the PA signal from NR751 to PA-HD (Figure S5a), a hemicyanine based PA scaffold with similar absorbance profile to NR751,[53] in aqueous solutions containing BSA or FBS. NR751 displayed significantly increased PA intensity in PBS compared to PA-HD (4.8-fold, Figure S5b). This trend of increased PA intensity of NR751 compared to PA-HD was maintained in 5% BSA (1.7-fold, Figure S5c) and FBS (1.4-fold, Figure S5d), although the presence of protein slightly reduced PA signal from NR751 while enhancing signal from PA-HD. Additionally, NR751 displayed enhanced PA photostability in aqueous solutions containing protein (13.1-fold in PBS with 5% BSA and 3.5-fold in FBS) compared to PA-HD (Figure S6).
Figure 5.
a) PA images of PBS solutions containing 50 μM NR751 (left) or no dye (right) in tissue-mimicking phantoms consisting of 5% agarose and 2.5% milk. b) Overlay of normalized PA and absorbance spectra of NR751. c) PA spectra from the indicated dyes (50 μM each) in tissue-mimicking phantoms.
Although MB is a commonly used standard in PA imaging its absorbance maximum lies outside of the range of commercial instrumentation (Figures S1 and S2). PA signal intensity of equivalent concentrations of NR751 and MB demonstrated that NR751 produced a 4.1-fold higher PA signal compared to MB using commerical PA imaging instrumentation (Figure 5c). Additionally, we observed a 10% increase in PA signal from NR751 relative to ICG and a more defined PA spectra for NR751 (potentially due to aggregation of ICG, Figure 5c). Broad absorbance spectra can complicate spectral unmixing during multiplexed imaging. Taken together, these data demonstrate that the increased solubility and structural rigidity of azaphosphinate dyes can be leveraged to afford probes with improved PA signal output and photostability in aqueous media compared to conventional dyes.
Multiplexed PA Imaging with NR751.
Encouraged by the ability of NR751 to produce relatively intense PA signal with well-defined features, due to increased solubility, we asked whether MB and NR751 PA signals could be spectrally unmixed. As a proof-of-principle for multiplexed PA imaging with NR751 and MB, we prepared a two-channel tissue-mimicking phantom for simultaneously imaging samples of each dye. First, we investigated the ability to distinguish between NR751 or MB and PBS. Indeed, spectral unmixing yielded images in which the signal from each dye could be clearly distinguished (Figure S7a and b). To provide a more physiologically relevant test, we also asked whether solutions of NR751 or MB could be imaged in defibrinated sheep blood. Here too, the PA signal from each dye could be resolved from oxyhemoglobin using spectral unmixing (Figure S7c and d). Lastly, we investigated the ability to simultaneously distinguish between PA signal from NR751 or MB in defibrinated sheep blood. Indeed, we were able to resolve the PA signal from each dye (Figure 6). These experiments provide a proof-of-principle for multiplexed PA imaging with NR751 and MB.
Figure 6.
a) Spectral unmixing of 100 μM NR751 (green) or MB (magenta) PA signal in defibrinated sheep’s blood using a two-channel tissue mimicking phantom consisting of 5% agarose and 2.5% milk. b) Mean pixel intensity of NR751 and MB within the indicated region of interest (red circle) from panel a.
PA Imaging in Tissues.
Tissue imaging is an established approach to evaluate the performance of PA reagents since tissues simulate optical and acoustic parameters found in living animals (optical absorption and scattering, speed of sound, acoustic attenuation, acoustic backscatter coefficient, etc.).[54,55,56] To determine the maximal imaging depth of NR751 in tissues, solutions of 50 μM dye in PBS were placed at the center of chicken breast cylinders of increasing thickness (Figure S8). PA imaging of these samples resulted in a gradual decrease in PA signal with increasing tissue depth, yielding an imaging depth of 2.75 cm for NR751 under these conditions (Figure 7). This data demonstrates that azaphosphinate NR dyes provide a robust scaffold for the development of PA imaging reagents.
Figure 7.
a) PA images of PBS solutions with or without 50 μM NR751 in chicken breast cylinders of increasing thickness. b) Average PA signal intensity from 7 positions, scanned in triplicate within the indicated region of interest (red circle) from panel a. **** indicates p<0.0001 (two sample independent t-test). c) PA signal-to-noise ratios (SNR) at the indicated imaging depths.
PA Imaging in Living Mice.
Encouraged by the ability to image NR751 in the above tissue model, we investigated the ability to resolve signal from NR751 in living mice. First, we assessed the toxicity of NR751 in cell culture (Figure S9) as well as compared singlet oxygen genration from NR751 to MB (Figure S10). We did not observe cellualr toxicity at concentrations used in this study up to 24 hrs in cell culture nor did we observe evidence for singlet oxygen genration from NR751 under the conditions used herein. Accordingly, we administered subcutaneous injections of NR751 (1 mg/kg) in the right abdomen. The mice were then scanned from the chest to the abdomen. The PA signal from NR751 could be clearly distinguished from endogenous absorbers such as hemoglobin and oxyhemoglobin by employing a linear-regression, spectral unmixing approach (Figure 8a and b). Additionally, quantified PA signal from NR751 in the injection region, for n = 3 mice in each group, indicated a 93.3-fold increase in PA signal in mice injected with NR751 compared to saline alone (Figure 8c). These experiments demonstrate the ability to resolve PA signal from NR751 in living mice.
Figure 8.
Representative cross-sectional (a) or maximal intensity projections of cross-sectional images from the mouse chest (top) to abdomen (bottom, b) PA images of nude mice following subcutaneous injection of 100 μL saline without (left) or with 1 mg/kg NR751 (right). c) Average PA signal intensity for NR751 after spectral unmixing within the indicated region of interest (red circle) from panel a (n = 3). *** indicates p<0.001 (two-tailed independent t-test).
Enhancing Azaphosphinate Fluorescence.
Having demonstrated the utility of NR751 as a PA imaging probe, we turned our attention toward improving the fluorescence properties of this dye. Given the remarkable water solubility of NR751, we hypothesized that twisted intramolecular charge transfer (TICT) may be suppressing the Φ of this scaffold.[57,58,59,60]
Therefore, we set out to inhibit TICT using the well-established azetidine auxochrome.[60] An azetidine-containing derivative, termed NR751-Az, was synthesized by modifying published procedures for azasiline dyes (Scheme 2).[45] Briefly, the starting material 1-(3-bromo-5-methylphenyl) azetidine was prepared via Buchwald-Hartwig amination. Phosphinate amide 3 was furnished by lithiation followed by addition to diethylphosphoramidous dichloride and subsequent oxidation with hydrogen peroxide in situ. After dibromination of phosphinate amide 3 with 1,3-dibromo-5,5-dimethylhydantoin (DDH), the resulting intermediate 4 was subjected to lithiation followed by addition to isopentyl nitrite and iron (II) chloride catalyzed reduction in situ. Finally, NR751-Az was obtained by dehydrogenation of the precursor followed by phosphine amide hydrolysis under acidic conditions. We then measured the spectroscopic properties of this new dye. Similar to NR751, NR751-Az exhibited an absorption maximum of 751 nm and an emission maximum of 775 nm (Figure 9a and b). However, the Φ increased by 2-fold to 0.8%, leading to a 1.8-fold increase in fluorescence brightness compared to NR751 (Figures 2b and 9b). This increase in quantum yield resulted in a 2.8-fold decrease in PA signal from NR751-Az compared to NR751 (Figure S11), presumably due to decreased vibrational relaxation of NR751-Az.[30] Thus, in addition to acting as PA probes, azaphosphinate NR dyes also provide a novel scaffold for the development of robust NIR fluorescent imaging probes.
Scheme 2.
Synthesis of NR751-AZ.
Figure 9.
a) Normalized absorbance and emission spectra of NR751-Az in PBS. b) Photophysical properties of NR751-Az in PBS.
Chemical- and Photo-Stability of Azaphosphinate Dyes.
Based on the structural rigidity of the azaphosphinate NR scaffold and the ability of azetidine auxochromes to increase photostability,[60] we hypothesized that both NR751 and, to a greater extent, NR751-Az would display significantly enhanced chemical and photostability. Therefore, we first evaluated the chemical stability of azaphosphinate NR dyes. Both dyes exhibited virtually identical absorbance spectra across a range of biologically relevant pH values (4–9) over 4 hours (Figure S12). To investigate the stability of these dyes in the presence of biologically relevant reactive oxygen/nitrogen species (ROS/RNS), we conducted stability assays in the presence of HOCl, H2O2, TBHP, ONOO−, ·OH, · OtBu, and NO. The results revealed that NR751 and NR751-Az maintained their absorbance for a minimum of 12 hours in the presence of the majority of ROS/RNS in our panel (Figure S13). However, in the presence of ONOO−, we observed a 22 and 58% decrease in absorbance for NR751 and NR751-Az, respectively (Figure S13). Next, we examined the stability of NR751 and NR751-Az in commonly used cell culture components. Within 4 hours of exposure to DMEM or FBS, NR751 displayed a 54 or 32% decrease in absorbance in the respective media (Figure S14). Conversely, NR751-Az displayed relatively enhanced stability in both DMEM and FBS. Finally, we also investigated the photostability of azaphosphinate NR dyes. As a comparison, we chose the FDA-approved dye ICG. Indeed, NR751 displayed a 10.0-fold longer photobleaching half-life compared to ICG in PBS, while virtually no fluorescence decay was observed for NR751-Az (Figure S15). However, we observed that the half-life of NR751 photobleaching was accelerated in samples containing protein (FBS or BSA) presumably due to protein binding (Figure S15b and c). The azetidine auxochromes in NR751-Az virtually abolished photobleaching in BSA and increased the photobleaching half-life relative to NR751 by 2.8-fold in FBS (Figure S15b and c). Overall, these experiments demonstrate that azaphosphinate NR dyes are well suited for biological imaging applications.
NIR Fluorescence Microscopy with NR751-Az.
As a preliminary assessment of the utility of azaphosphinate NR dyes for fluorescence imaging, we investigated the ability to stain cellular organelles using confocal microscopy. For this purpose, we chose NR751-Az due to its optimized fluorescence brightness. Previous work from our lab has shown that phosphinate dyes are typically membrane impermeable,[40,41,43] likely due to the negatively charged phosphinate. Indeed, treatment of HeLa cells with NR751-Az resulted in no observable cellular fluorescence from NR751-Az (Figure 10a), further highlighting the membrane impermeable nature of phosphinate dyes. We have previously shown that phosphinate esterification can dramatically increase the membrane permeability of NR dyes,[40,41,43,44] however initial studies of a phosphinate ethyl ester-containing derivative of NR751 demonstrated virtually instantaneous phosphinate ester hydrolysis in water (Figure S16). We are currently investigating approaches to further stabilize phosphinate esters in azaphosphinate NR dyes using structural modifications, in a similar fashion to stabilized esters in rhodamine-based NR dyes.[41,44] As an alternative approach, we were inspired by the formation of complexes between SDS and MB,[61] and hypothesized that a cationic lipid such as cetyltrimethylammonium bromide (CTAB) could form a complex with NR751-Az similarly to other dyes.[62,63,64] We envisioned that the resulting CTAB/dye complex would mask the negative charge on NR751-Az, thereby increasing its permeability to cellular membranes. Indeed, the NR751-Az/CTAB complex, termed NR751-Az-CTAB, significantly enhanced the membrane permeability of NR751-Az (Figure 10b). Importantly, the concentration of CTAB used in this study was not toxic to HeLa cells (Figure S17). Co-treatment of HeLa cells with NR751-Az-CTAB and organelle trackers indicated that NR751-Az fluorescence was mainly localized in mitochondria (Figure 10c and Figure S18). While the structure and mechanism of cellular uptake of the CTAB/dye complex is currently under investigation, these data demonstrate the ability to deliver phosphinate-containing dyes to living cells and the potential of NR751-Az for organelle labeling and NIR fluorescence imaging applications.
Figure 10.
Confocal microscopy images of HeLa cells incubated with 20 μM NR751-Az (a) or NR751-Az-CTAB (b). c) Confocal microscopy images of the indicated organelle tracker and 50 μM NR751-Az-CTAB. Merged images show predominant localization to the mitochondria. Scale bars = 20 μm.
NIR Fluorescence Imaging with NR751-Az in Living Mice.
To demonstrate the ability to visualize NR751-Az fluorescence in a more complex system, we injected saline solutions without or with 50 μM NR751-Az intramuscularly in the hindlimbs of live mice and imaged immediately following injection. Indeed, a clear 5.0-fold increase in fluorescence from hindlimb sites injected with NR751-Az was observed compared to saline (Figure 11 and Figure S19). These results indicate that the quantum yield enhancement in NR751-Az can be leveraged for fluorescence imaging in mice.
Figure 11.
a) A representative fluorescence image of a nude mice with intramuscular injection sites for saline (left hindlimb) or 50 μM NR751-Az in saline containing 0.5% DMSO (right hindlimb) indicated. b) Average fluorescence intensity within the indicated region of interest (red circle) from panel a (n = 3 mice in each group). Mice were imaged immediately following injection. *** indicates p<0.001 (two-tailed independent t-test).
Conclusions
We have rationally designed a low molecular weight dye with NIR absorption and emission in the 750 nm range. Our design strategy, which employed the incorporation of a phosphinate functionality into the oxazine scaffold, not only red-shifted the absorbance and emission of the resulting dye but also dramatically suppressed aggregation in aqueous solutions. To the best of our knowledge, NR751 is the lowest molecular weight, water soluble dye with absorbance and emission maxima in this spectral range.
Leveraging the improved water solubility of NR751, we demonstrated that this dye scaffold can produce a robust PA signal. Indeed, NR751 produces a 4.1- and 1.1-fold stronger PA signal on commercial instrumentation than currently used standards such as MB and ICG. Given the well-defined PA signal profile of this dye, multiplexing with currently available PA probes, such as MB, in blood samples is possible. Moreover, in its current form, NR751 can achieve 2.75 cm imaging depths in tissue samples and can be visualized in live mice. Our lab is currently pursuing the use of NR751 as a scaffold for developing in vivo PA imaging probes, activatable PA probes, and more intense PA reagents.
Lastly, we demonstrated the ability to increase the fluorescence brightness of NR751 through the incorporation of azetidine auxochromes, yielding NR751-Az. This improved fluorescent dye displays reduced photobleaching compared to ICG in water. We further developed a protocol for the delivery of NR751-Az into living cells for fluorescence microscopy applications and demonstrated the ability to resolve this dye in live mice. Our laboratory is currently pursuing targeted NR751-Az probes for animal imaging, activatable probes for analyte sensing, and approaches to further increase the fluorescence brightness of this scaffold.
In summary, we report a new class of azaphosphinate NR dyes with desirable properties that can be leveraged for either PA or NIR fluorescence imaging. Ultimately, the choice of a water soluble dye, such as the azaphosphinates described herein, or a more hydrophobic dye such as ICG will depend on the imaging application. In certain cases, the enhanced fluorescence and photostability of ICG[65,66] upon protein binding may be useful for imaging structures that are readily accessible in the blood stream. However, protein binding may complicate the design of probes for intracellular imaging applications, limit visualization of spatially restricted areas, or hinder reaction with handles for analyte sensing or in vivo labeling. We anticipate that the availability of this new, highly water soluble dye class will stimulate the design of robust imaging probes to illuminate biology. We suspect that other dye scaffolds may also benefit from the coarse tuning and solubility-enhancing effects of the phosphinate functionality.
Supplementary Material
Acknowledgements
F. B. was supported by a Marry Anderson Harrison Jefferson Fellowship from the Jefferson Scholars Foundation. We thank the Biomolecular Magnetic Resonance and Hi-Resolution Mass Spectrometer Facility for assistance with compound characterization. We acknowledge Xinqi Zhou from the Department of Chemistry, University of California, Berkeley for assistance with manuscript editing. We acknowledge the Keck Center for Cellular Imaging for the use of the Leica SP5X microscopy system (NIH, S10RR025616). Photoacoustic imaging studies were performed in the Bioimaging and Applied Research Core at Virginia Commonwealth University. We acknowledge financial support from the NIH (P30CA16059 and UM1CA186644, SG; P30CA044579 and R01CA223767, JH; and R35GM148221, CIS) and the University of Virginia. The content of this work is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Footnotes
Conflict of Interests
R.Y., F.B., Y.F., and C.I.S. have filed a patent application on phosphinate-containing dyes.
Supporting information for this article is available on the WWW under https://doi.org/10.1002/chem.202303331
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- [1].Wang LV, Hu S, Science 2012, 335, 1458–1462. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Taruttis A, Ntziachristos V, Nat. Photonics 2015, 9, 219–227. [Google Scholar]
- [3].Escobedo JO, Rusin O, Lim S, Strongin RM, Curr. Opin. Chem. Biol 2010, 14, 64–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Kobayashi H, Ogawa M, Alford R, Choyke PL, Urano Y, Chem. Rev 2010, 110, 2620–2640. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Cosco ED, Caram JR, Bruns OT, Franke D, Day RA, Farr EP, Bawendi MG, Sletten EM, Angew. Chem. Int. Ed 2017, 56,13126–13129. [DOI] [PubMed] [Google Scholar]
- [6].Cosco ED, Arus BA, Spearman AL, Atallah TL, Lim I, Leland OS, Caram JR, Bischof TS, Bruns OT, Sletten EM, J. Am. Chem. Soc 2021, 143, 6836–6846. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Daly HC, Matikonda SS, Steffens HC, Ruehle B, Resch-Genger U, Ivanic J, Schnermann MJ, Photochem. Photobiol 2022, 98, 325–333. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Matikonda SS, Ivanic J, Gomez M, Hammersley G, Schnermann MJ, Chem. Sci 2020, 11, 7302–7307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Michie MS, Gotz R, Franke C, Bowler M, Kumari N, Magidson V, Levitus M, Loncarek J, Sauer M, Schnermann MJ, J. Am. Chem. Soc 2017, 139, 12406–12409. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Gayton J, Autry SA, Meador W, Parkin SR, Hill GA Jr., Hammer NI, Delcamp JH, J. Org. Chem 2019, 84, 687–697. [DOI] [PubMed] [Google Scholar]
- [11].Rathnamalala CSL, Gayton JN, Dorris AL, Autry SA, Meador W, Hammer NI, Delcamp JH, Scott CN, J. Org. Chem 2019, 84, 13186–13193. [DOI] [PubMed] [Google Scholar]
- [12].Rathnamalala CSL, Pino NW, Herring BS, Hooper M, Gwaltney SR, Chan J, Scott CN, Org. Lett 2021, 23, 7640–7644. [DOI] [PubMed] [Google Scholar]
- [13].Zhou X, Fang Y, Wimalasiri V, Stains CI, Miller EW, Chem. Commun 2022, 58, 11941–11944. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Weber J, Beard PC, Bohndiek SE, Nat. Methods 2016, 13, 639–650. [DOI] [PubMed] [Google Scholar]
- [15].Borg RE, Rochford J, Photochem. Photobiol 2018, 94, 1175–1209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Matikonda SS, Helmerich DA, Meub M, Beliu G, Kollmannsberger P, Greer A, Sauer M, Schnermann MJ, ACS Cent. Sci 2021, 7, 1144–1155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Garcia-Uribe A, Erpelding TN, Krumholz A, Ke H, Maslov K, Appleton C, Margenthaler JA, Wang LV, Sci. Rep 2015, 5, 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Erpelding TN, Kim C, Pramanik M, Jankovic L, Maslov K, Guo Z, Margenthaler JA, Pashley MD, Wang LV, Radiology 2010, 256, 102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [19].Song KH, Stein EW, Margenthaler JA, Wang LV, J. Biomed. Opt 2008, 13, 054033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].Kim C, Erpelding TN, Jankovic L, Wang LV, Philos. Trans. R. Soc. London Ser. A 2011, 369, 4644–1650. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Tummers QR, Boonstra MC, Frangioni JV, van de Velde CJ, Vahrmeijer AL, Bonsing BA, Int. J. Surg. Case Rep 2015, 6, 150–153. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].van Manen L, Handgraaf HJ, Diana M, Dijkstra J, Ishizawa T, Vahrmeijer AL, Mieog JSD, J. Surg. Oncol 2018, 118, 283–300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Golz EK, Vander Griend DA, Anal. Chem 2013, 85, 1240–1246. [DOI] [PubMed] [Google Scholar]
- [24].Shahinyan GA, Amirbekyan AY, Markarian SA, Acta A Mol. Biomol. Spectrosc 2019, 217, 170–175. [DOI] [PubMed] [Google Scholar]
- [25].Braswell E, J. Phys. Chem 1968, 72, 2477–2483. [Google Scholar]
- [26].Mukerjee P, Ghosh AK, J. Am. Chem. Soc 1970, 92, 6408–6412. [Google Scholar]
- [27].Murakami K, Dyes Pigm. 2002, 53, 31–43. [Google Scholar]
- [28].Bell AG, Am. J. Sci 1880, s3-20, 305–324. [Google Scholar]
- [29].Wang LV, Yao J, Nat. Methods 2016, 13, 627–638. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [30].Knox HJ, Chan J, Acc. Chem. Res 2018, 51, 2897–2905. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [31].Reinhardt CJ, Chan J, Biochemistry 2018, 57, 194–199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Yadav AK, Hernandez S, Su S, Chan J, Curr. Opin. Chem. Biol 2020, 57, 114–121. [DOI] [PubMed] [Google Scholar]
- [33].Zhao Z, Swartchick CB, Chan J, Chem. Soc. Rev 2022, 51, 829–868. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [34].Ovchinnikov OV, Evtukhova AV, Kondratenko TS, Smirnov MS, Khokhlov VY, Erina OV, Vib. Spectrosc 2016, 86, 181–189. [Google Scholar]
- [35].Fernández-Pérez A, Valdés-Solís T, Marbán G, Dyes Pigm. 2019, 161, 448–456. [Google Scholar]
- [36].Canossa S, Predieri G, Graiff C, Acta Crystallogr. E: Crystallogr. Commun 2018, 74, 587–593. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [37].Koide Y, Urano Y, Hanaoka K, Terai T, Nagano T, J. Am. Chem. Soc 2011, 133, 5680–5682. [DOI] [PubMed] [Google Scholar]
- [38].Yin R, Fang Y, Zhou X, Stains CI, Methods Enzymol. 2020, 639, 23–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [39].Zhou X, Lesiak L, Lai R, Beck JR, Zhao J, Elowsky CG, Li H, Stains CI, Angew. Chem. Int. Ed 2017, 56, 4197–4200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [40].Fang Y, Good GN, Zhou X, Stains CI, Chem. Commun 2019, 55, 5962–5965. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [41].Zhou X, Lai R, Beck JR, Li H, Stains CI, Chem. Commun 2016, 52, 12290–12293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [42].Chai X, Cui X, Wang B, Yang F, Cai Y, Wu Q, Wang T, Chem. Eur. J 2015,21, 16754–16758. [DOI] [PubMed] [Google Scholar]
- [43].Lesiak L, Zhou X, Fang Y, Zhao J, Beck JR, Stains CI, Org. Biomol. Chem 2020, 18, 2459–2467. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [44].Zhou X, Fang Y, Lesiak L, Stains CI, ChemBioChem. 2019, 20, 1712–1716. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [45].Choi A, Miller SC, Org. Lett 2018, 20, 4482–4485. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [46].DiMeglio D, Zhou X, Wirth T, Brøndsted F, Lesiak L, Fang Y, Shadmehr M, Stains C, Chem. Eur. J 2023,29, e202202861. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [47].Wang L, Liu J, Zhao S, Zhang H, Sun Y, Wei A, Guo W, Chem. Commun 2020, 56, 7718–7721. [DOI] [PubMed] [Google Scholar]
- [48].Gardner SH, Brady CJ, Keeton C, Yadav AK, Mallojjala SC, Lucero MY, Su S, Yu Z, Hirschi JS, Mirica LM, Angew. Chem. Int. Ed 2021, 133, 19008–19014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [49].Hatamimoslehabadi M, Bellinger S, La J, Ahmad E, Frenette M, Yelleswarapu C, Rochford J, J. Phys. Chem. C 2017, 121, 24168–24178. [Google Scholar]
- [50].Patil K, Pawar R, Talap P, Phys. Chem. Chem. Phys 2000, 2, 4313–4317. [Google Scholar]
- [51].Fernandez-Perez A, Marban G, ACS Omega 2020, 5, 29801–29815. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [52].Reichardt C, Chem. Rev 1994, 94, 2319–2358. [Google Scholar]
- [53].Lucero MY, Gardner SH, Yadav AK, Borri A, Zhao Z, Chan J, Angew. Chem. Int. Ed 2022, 61, e202211774. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [54].Kim C, Qin R, Xu JS, Wang LV, Xu RX, J. Biomed. Opt 2010, 15, 010510. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [55].Cook JR, Bouchard RR, Emelianov SY, Biomed. Opt. Express 2011, 2, 3193–3206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [56].Aziz A, Medina-Sánchez M, Claussen J, Schmidt OG, Nano Lett. 2019, 19, 6612–6620. [DOI] [PubMed] [Google Scholar]
- [57].Song X, Johnson A, Foley J, J. Am. Chem. Soc 2008, 130, 17652–17653. [DOI] [PubMed] [Google Scholar]
- [58].Grimm JB, Muthusamy AK, Liang Y, Brown TA, Lemon WC, Patel R, Lu R, Macklin JJ, Keller PJ, Ji N, Lavis LD, Nat. Methods 2017, 14, 987–994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [59].Lv X, Gao C, Han T, Shi H, Guo W, Chem. Commun 2020, 56, 715–718. [DOI] [PubMed] [Google Scholar]
- [60].Grimm JB, English BP, Chen J, Slaughter JP, Zhang Z, Revyakin A, Patel R, Macklin JJ, Normanno D, Singer RH, Lionnet T, Lavis LD, Nat. Methods 2015, 12, 244–250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [61].Wang J, Lin C-Y, Moore C, Jhunjhunwala A, Jokerst JV, Langmuir 2018, 34, 359–365. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [62].Tian H, Qian J, Sun Q, Jiang C, Zhang R, Zhang W, Analyst 2014, 139, 3373–3377. [DOI] [PubMed] [Google Scholar]
- [63].Mohd Yusof NS, Khan MN, Ashokkumar M, J. Phys. Chem. C 2012, 116, 15019–15027. [Google Scholar]
- [64].Manna A, Chakravorti S, J. Mol. Liq 2012, 168, 94–101. [Google Scholar]
- [65].Jung BS, Vullev VI, Anvari B, IEEE J. Sel. Top. Quantum Electron 2014, 19, 149–157. [Google Scholar]
- [66].Mordon S, Devoisselle JM, Soulie-Begu S, Desmettre T, Microvasc. Res 1998, 55, 146–152. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.













