A Far-Red Fluorescent Probe to Visualize Gram-Positive Bacteria in Patient Samples

Gram-positive bacteria, in particular Staphylococcus aureus (S. aureus), are the leading bacterial cause of death in high-income countries and can cause invasive infections at various body sites. These infections are associated with prolonged hospital stays, a large economic burden, considerable treatment failure, and high mortality rates. So far, there is only limited knowledge about the specific locations where S. aureus resides in the human body during various infections. Hence, the visualization of S. aureus holds significant importance in microbiological research. Herein, we report the development and validation of a far-red fluorescent probe to detect Gram-positive bacteria, with a focus on staphylococci, in human biopsies from deep-seated infections. This probe displays strong fluorescence and low background in human tissues, outperforming current tools for S. aureus detection. Several applications are demonstrated, including fixed- and live-cell imaging, flow cytometry, and super-resolution bacterial imaging.

Staphylococcus aureus (S. aureus) stands as the primary bacterial contributor to death in high-income countries. 1S. aureus commonly infects the skin and soft tissues but can also spread directly or via the bloodstream to virtually any site of the body including bone and joints, lungs, heart valves, or the central nervous system. 2,3These infections are associated with prolonged hospital stays, a large economic burden, 20−30% treatment failure, and mortality rates of up to 20−40%. 4,5S. aureus infections are often resilient to antibiotic treatment, even if the causative S. aureus strain appears to be susceptible to the employed antibiotics in clinical microbiology laboratories.These infections often need mechanical removal of the primary site of infection by surgery or drainage for a successful treatment.To date, local conditions and mechanisms in infected human tissues remain poorly characterized.Thus, a better understanding of the pathological mechanisms underlying the persistence of S. aureus in human tissue is necessary.
Fluorescent tools have been developed to detect Grampositive bacteria, including S. aureus, in the context of mammalian cells and tissues.Immunostaining is frequently used to label bacteria of interest in tissue sections.An antibody against S. aureus surface proteins (anti-SA) is commonly applied to detect S. aureus, 6 but exhibits staining heterogeneity due to varying levels of protein expression.An antibody against wall teichoic acids (anti-WTA) detects Gram-positive bacteria more comprehensively, but scaling up experiments using this expensive antibody is rather costly. 7,8Alternatively, antibiotics conjugated with fluorophores can be used as bacterial detection tools. 9These small-molecule probes can have significant advantages over antibodies, including better tissue penetration, higher chemical stability, compatibility with antibody-based detection of other biomolecules, and lower cost.This kind of probe has been developed to image Grampositive bacteria, including S. aureus, by leveraging the binding of the glycopeptide vancomycin to D-alanyl-D-alanine moieties of lipid II (a precursor of peptidoglycan) and the mature peptidoglycan layer of bacteria. 10Vancomycin labeled with the green fluorophore boron dipyrromethene (Van-BODIPY), 11 the orange fluorophore rhodamine B (Van-Rh) 12 and (Rho-FF-Van), 13 a red-emitting merocyanine (Merocy-Van) 14 or with a near-infrared dye (Vanco-800CW) have been reported (Figure S1). 15 These probes, however, exhibit drawbacks such as spectral overlap with host autofluorescence, which is particularly high in inflamed tissues (Van-BODIPY), 11,16 low solubility in aqueous buffer and very low brightness (Merocy-Van), or an excitation wavelength that is not commonly available in fluorescence microscopes or flow cytometers (Vanco-800CW).
We posited that detection of Gram-positive bacteria, and in particular of S. aureus, in mammalian cells and human tissues, including patient biopsies, would benefit from a different fluorophore that is bright and photostable, has minimal fluorescence background regardless of the sample heterogeneity or preservation method, is simple to use, and can be analyzed by common instruments found in research laboratories.Considering these challenges, we selected the JF 669 dye as the fluorescent reporter. 17This fluorophore is bright and photostable at a long wavelength, yet it can be excited with a common laser used in microscopes and flow cytometers.Moreover, compared to the popular JF 646 dye, 18 JF 669 is more photostable and is more fluorogenic, 17 i.e., it produces a brighter signal upon binding to an intended macromolecular target.
Probe Van-JF 669 was conveniently prepared in one step from commercially available vancomycin and JF 669 -NHS ester (Figure 1A).Structural analysis by high-resolution tandem mass spectrometry revealed that conjugation of the dye occurs selectively at the primary amine on the sugar moiety of vancomycin (Figure S2), thus this synthesis produces a homogeneous probe with a well-defined structure.Moreover, the primary amine on the disaccharide does not have specific interactions with D-alanyl-D-alanine, 19 thus installation of the fluorophore at this position does not prevent binding to peptidoglycan.
The excitation and emission spectra of Van-JF 669 displayed essentially the same wavelengths as those of the free dye (Figure S3), and the quantum yield also remained unchanged (ϕ JF669 = 0.37, 17 ϕ Van-JF669 = 0.37).Van-JF 669 preserved the polarity-dependent fluorogenicity of the parent dye, displaying low fluorescence in an apolar medium (Figure S4).This behavior contrasts with those of traditional dyes, such as BODIPY, which have stronger fluorescence in media of low polarity (Figure S4) and thus display large background fluorescence due to binding to hydrophobic substances, such as membranes.Moreover, the photophysical properties and photostability of fluorophores reported as fluorescent vancomycin are compared (Table S1, Figures S5 and S6).BODIPY FL and Rhodamine B exhibit lower photostability compared with JF669 and the merocyanine dye.However, merocyanine displayed poor solubility and a small quantum yield and molar extinction coefficient in an aqueous buffer, making it significantly less useful for advanced imaging experiments.
Van-JF 669 exhibited a minimum inhibitory concentration (MIC) of 5 mg L −1 (2.3 μM) against lab strain methicillinresistant S. aureus (MRSA, ATCC 43300), lab strain methicillin-sensitive S. aureus (MSSA, ATCC 29213), and clinical isolate S. aureus (MSSA, PROSA28), which is significantly higher than vancomycin (MIC = 0.31 mg L −1 − 0.2 μM) in the same strains (Figure S7).This MIC difference indicates that the attached fluorophore diminished vancomycin's affinity for its target D-ala-D-ala, similar to other conjugates attached at the primary amine. 10,19The reduced toxicity suggests that our probe might operate differently from the original vancomycin, potentially offering an advantage by enabling live cell imaging without significantly affecting bacterial viability.JF 669 at 10 nM specifically stained S. epidermidis cells but not human cells (Figure 1C and Figure S8).This high selectivity permitted imaging without a necessity to wash away the staining solution.Van-BODIPY also labeled the bacterial cell wall (Figure S9), but the signal-to-background ratio was significantly lower than that of Van-JF 669 (Figure 1D).
Van-JF 699 also selectively stained S. aureus in cell-culture infections of human monocyte-like THP-1 cells (ATCC TIB-202) with S. aureus (Cowan).Imaging flow cytometry revealed the selective staining of S. aureus.The signal-to-background ratio was higher compared to an alternative labeling method based on cell-wall incorporation of a tetramethyl rhodaminefunctionalized D-amino acid (RADA) 20 (Figure 2A−C).Van-JF 669 also stained the clinical isolate S. aureus (PROSA25) in primary monocyte-derived macrophages isolated from healthy human donors (Figure 2D).
To enhance staining of live S. aureus, we mixed 130 nM Van-JF 669 with 170 nM unlabeled vancomycin, which promotes the formation of heterodimers of these compounds with increased affinity for bacterial cell walls, particularly for live S. aureus (Supporting Video 1). 19,21S. aureus was able to divide under these conditions, and the no-wash nature of the staining enabled efficient and sustained labeling of newly formed daughter cells.
To evaluate the labeling efficiency of Van-JF 669 in patient tissues, we costained a biopsy section from a patient with osteomyelitis using Van-JF 669 and an antibody to S. aureus protein A (anti-SA).We observed optimal staining of the S. aureus cell wall in these samples at slightly higher concentrations (130 nM), probably because of low-level adsorption to the tissue.Whereas anti-SA stained only the outer surface proteins of S. aureus, Van-JF 669 stained also the septum between two emerging daughter cells, which is an area of the cell wall with particularly active de novo peptidoglycan synthesis (Figure 3A−C). 22These data confirm the identity of Van-JF 669 -positive particles as S. aureus.
In a formalin-fixed, paraffin-embedded (FFPE) section of a biopsy from a prosthetic joint infection, staining with 200 nM Van-BODIPY led to substantial background fluorescence (Figure 3D), whereas staining of the same tissue with 130 nM Van-JF 669 produced less background emission (Figure 3E).Moreover, Van-JF 669 is compatible with antibody labeling, making it appropriate for multicolor experiments in which more than one target is visualized simultaneously (Figure 3F).These experiments confirm the suitability of Van-JF 669 for detecting S. aureus in patient-derived biopsies.
Finally, we hypothesized that the spirocyclization equilibrium that drives the fluorogenicity of Van-JF 669 could lead to fluorescence blinking, 23 which could be leveraged to perform single-molecule localization microscopy (SMLM). 24We first confirmed that single-molecule blinking could be detected by total-internal reflection fluorescence (TIRF) microscopy (Supporting Video 2).This blinking likely arises from a combination of spontaneous interconversion between the spirocyclic (dark) and zwitterionic (fluorescent) isomers of the JF 669 dye and the transient binding of vancomycin to the peptidoglycan layer of bacteria.The spontaneous blinking of probe Van-JF 669 allowed us to perform live-cell SMLM by simply incubating S. epidermidis with Van-JF 669 (0.1 nM) without the need for cell fixation or the use of special blinking buffers.SMLM of live S. epidermidis (Figure 4A) produced images that had nearly 1 order of magnitude better resolution than confocal images.For example, the cell wall of S. epidermidis in confocal imaging appeared to have a width of ∼260 nm, whereas SMLM using Van-JF 669 gave images in which the cell wall has a width of ∼40 nm (Figure 4b), which is consistent with direct observations for Gram-positive bacteria by cryo-electron microscopy. 25These results demonstrate that Van-JF 669 can be used for live-cell super-resolution imaging of the cell wall of Gram-positive bacteria.
In summary, we developed a far-red fluorescent probe to detect Gram-positive bacteria in in vitro bacterial cultures, cellculture infections, and human tissues from patients with deepseated infections.This probe is easy to synthesize, can be used to stain complex samples, does not rely on genetic manipulation of the specimen, has virtually no background fluorescence, and is compatible with flow cytometry, fluorescence microscopy, and live-cell, super-resolution microscopy.Thus, Van-JF 669 is a convenient and reliable tool to visualize Gram-positive bacteria in complex samples.

Immunohistochemistry of Vibratome Sections of
Human Biopsies.60 μm vibratome sections from PROSA329 (stored in cryoprotectant buffer at −20 °C) were washed three times for 5 min with Tris-buffered saline, pH 7.4 (TBS), at room temperature. 26For antigen retrieval, sections were incubated in 1 mL of prewarmed 10 mM sodium citrate, pH 8.5, at 37 °C for 30 min.Sections were washed three times with TBST (0.1% Triton X-100 in TBS), blocked for 30 min with 1% BSA-fraction V, 2% human serum in TBST, and washed again three times with TBST.Endogenous biotin sites were blocked using a Streptavidin/Biotin blocking kit (Vector Laboratories, VC-SP-2002) following manufacturer's protocols.Sections were incubated in 10 μg mL −1 anti-Staphylococcus aureus antibody (Abcam, ab20920) in a blocking buffer (with 1% BSA-fraction V, 2% human serum in TBST) at 4 °C overnight, washed three times with TBST, and stained with 5 μg mL −1 Alexa Fluor 568 donkey antirabbit IgG (Invitrogen, A10042) for 1 h at room temperature.Sections were washed three times with TBS and incubated with Van-JF 669 (130 nM) and DAPI (3 μM) for 10 min.Sections were washed three times with TBS, mounted in 40 μL of mounting medium (Agilent DAKO, Cat No. S3023) and examined using an Olympus SpinSR Spinning disc microscope.
Flow Cytometry.Cocultures were performed with a multiplicity of infection (MOI) of 2. A clinical isolate of S. aureus (PROSA25) (10 6 cells from an overnight culture resuspendend in RPMI + 10% FBS) and THP1 cells (5 × 10 5 cells in RPMI + 10% FBS) were cocultured in a 96 well-plate at 37 °C with 5% CO 2 for 2 h.Afterward, the plates were centrifuged for 5 min at 350 × G.Each well was blocked using 50 μL of FACS buffer (PBS + 5% FBS) + 2.5 μL of Fc receptor (BioLegend, Cat No. 422302) block per well, and then 150 μL of FACS buffer was added and spun down.The sample was fixed by the addition of 5 μL of 4% paraformaldehyde for 20 min in the dark at room temperature.After the addition of 150 μL of permeabilization buffer (BioLegend, Cat.No. 421002), the sample was centrifuged.Subsequently, Van-JF 669 (1.2 μM) and DAPI (3 μM) in permeabilization buffer were stained for 15 min in the dark at room temperature.150 μL of the FACS buffer was added.After spinning down and resuspending in 200 μL of PBS, samples were analyzed by flow cytometry (Cytoflex) and image stream.
Staining of FFPE Sections.Formalin-fixed paraffinembedded patient biopsies were cut into 4 μm-thick sections by a microtome (Leica, SM2010R).Samples were deparaffinized by washing twice with xylene for 10 min.Rehydration was carried out by incubating twice with 100% ethanol for 5 min, 96% ethanol for 2 min, 70% ethanol for 2 min, 50% ethanol for 2 min, and distilled water for 2 min.Sections were incubated with EDTA based BOND Epitope Retrieval solution, pH = 9.0 (Biosystems, AR9640) at 95 °C for 20 min for antigen retrieval.After the samples were cooled for 15 min, sections were transferred in PBST for 5 min.The slide was dried with Kimwipes and circle was drawn around the sections with a hydrophobic pen.300 μL of blocking buffer (5% goat serum in PBS + 0.1% Tween 20 (PBST) was added to slides, and the slide was incubated in a dark humidity chamber for 1 h at room temperature.After removal of the blocking buffer, the primary rabbit anti human anti-Rab5 antibody (Abcam, ab218624) was incubated overnight at 4 °C in a dark humidity chamber.Next, the primary antibody was removed and rinsed three times with PBST for 5 min.Samples were incubated with a secondary antibody at room temperature in a dark humidifier chamber for 1 h.Subsequently, the secondary antibody was removed and rinsed three times with PBST for 5 min.Van-JF 669 (130 nM) or Van-BODIPY (200 nM) and DAPI (3 μM) were incubated at room temperature in a dark humidity chamber for 10 min.After being rinsed three times with PBS for 5 min, the slide was mounted and visualized by confocal microscopy.Tissue Collection.Fresh tissue samples were collected intraoperatively by an orthopedic surgeon at the site of clinical apparent infection from deep-seated S. aureus major bone and joint infections.The patients were included and operated between November 2020 and November 2022 at the University Hospital of Basel, Switzerland by orthopedic surgeons specialized in musculoskeletal infections.Fresh samples were either directly used for flow cytometry or directly fixed in 4% formaldehyde until further use.The study was approved by the local Ethical Review Board (Ethikkommission Nordwestschweiz, Project-ID 2020-02588) and performed in compliance with all relevant ethical regulations.

Figure 1 .
Figure 1.Synthesis and in vitro validation of Van-JF 669 .(A) Synthesis of Van-JF 669 .(B) Micrograph of S. epidermidis treated with 2 nM of Van-JF 669 .(C) Micrograph of S. epidermidis cocultured with HeLa cells treated with 1.5 μM Hoechst (nuclei = blue), 2 μM ER-Tracker Green (endoplasmic reticulum = green) and 10 nM Van-JF 669 (S. epidermidis = magenta).(D) Signal-to-background ratio of S. epidermidis cocultured with HeLa cells labeled Van-BODIPY or Van-JF 669 .N = 20 independent cells per sample were examined.Statistical significance was evaluated by a paired t-test.The horizontal line inside the box indicates the median, the box indicates the mean ± standard deviation, and whiskers indicate the minimum and maximum values.

Figure 2 .
Figure 2. Flow cytometry image of in vitro coculture of THP1 cells with clinical S. aureus (PROSA25) labeled with (A) 1.2 μM Van-JF 669 and (B) 4.84 μM RADA treated with 3 μM DAPI (nuclei = blue).(C) Signal-to-background ratio of S. aureus in THP1 cells labeled with RADA and Van-JF 669 .N = 38 and 48 independent cells per time point were examined (from left to right).Statistical significance was evaluated by a paired t test.The horizontal line inside the box indicates the median, the box indicates the mean ± standard deviation, and whiskers indicate the minimum and maximum values.Outliers were removed using the Robust Regression and Outlier Removal (ROUT) algorithm with Q = 1%.(D) Fluorescent images of clinical S. aureus (PROSA25)-primary macrophages cocultured.The clinical S. aureus was phagocyted by primary macrophages followed by treatment of 240 nM Van-JF 669 (S. aureus = magenta), antihuman Rab5 (human cells = green), and 3 μM DAPI (nuclei = blue).The inset displays a magnified view of the area marked with a white rectangle in the main image.

Figure 3 .
Figure 3. Fluorescence microscopy images of patient samples treated with fluorescent probes or antibodies.(A) 200 nM anti-SA-AF568, followed by (B) 130 nM Van-JF 669 and 3 μM DAPI.(C) Merge of panels A and B (anti-SA-AF569 = orange, Van-JF 669 = magenta, and nuclei = blue).Fluorescent images of an FFPE-preserved patient sample with the S. aureus prosthetic joint infection stained with (D) 200 nM Van-BODIPY and (E) 130 nM Van-JF 669 .Color bars indicate pixel values.(F) Multicolor fluorescent image of a biopsy from a patient with S. aureus prosthetic joint infection preserved by FFPE.The sample was treated with 130 nM Van-JF 669 (S. aureus = white), antihuman Rab5 (human intracellular protein = green), and 3 μM DAPI (nuclei = blue).

Figure 4 .
Figure 4. Super-resolution imaging of S. epidermidis cell wall.(A) Confocal micrograph (left) and SMLM (right) images of S. epidermidis treated with 0.1 nM Van-JF 669 and imaged in ultrapure water.(B) Pixel values and Lorentzian fitting of these values along the line displayed in panel (A).