Rational Design of Phe‐BODIPY Amino Acids as Fluorogenic Building Blocks for Peptide‐Based Detection of Urinary Tract Candida Infections

Abstract Fungal infections caused by Candida species are among the most prevalent in hospitalized patients. However, current methods for the detection of Candida fungal cells in clinical samples rely on time‐consuming assays that hamper rapid and reliable diagnosis. Herein, we describe the rational development of new Phe‐BODIPY amino acids as small fluorogenic building blocks and their application to generate fluorescent antimicrobial peptides for rapid labelling of Candida cells in urine. We have used computational methods to analyse the fluorogenic behaviour of BODIPY‐substituted aromatic amino acids and performed bioactivity and confocal microscopy experiments in different strains to confirm the utility and versatility of peptides incorporating Phe‐BODIPYs. Finally, we have designed a simple and sensitive fluorescence‐based assay for the detection of Candida albicans in human urine samples.

Coupling was carried out using Fmoc-AA-OH (4 eq.), coupling reagent (4 eq.), OxymaPure (4 eq.) and DIPEA (8 eq.) in DMF for 1 h. The resin was then washed with DMF (5 × 1 min), DCM (5 × 1 min) and filtered. The completion of the coupling step was confirmed using Kaiser Test. Before the next coupling cycle, Fmoc group is removed as described above. Cleavage from resin for compounds 12 and 14-17: The peptide was cleaved from the resin using 2% TFA, 2.5% TIS in DCM (5 × 1 min) (12, 15 and 17) or 2% TFA/DCM (14 and 16) and washed with DCM (2 × 1 min). The combined filtrates were collected into a round bottom flask containing DCM (10 mL) and concentrated under reduced pressure. Cleavage from resin for compound 13: The peptide was cleaved from the resin using 95% TFA, 2.5% TIS in DCM (1h) and washed with DCM (4 × 1 min). The combined filtrates were collected into a round bottom flask and concentrated under reduced pressure.
After cleavage as described above, the crude peptide was precipitated by adding cold Et 2O (dropwise) and the resulting precipitate was decanted and dried (x2), obtaining 29 mg of benzyloxycarbonyl (Z) lysine-protected peptide. The crude peptide (24 mg, 0.013 mmol) was S7 deprotected by means of hydrogenation. Peptide was dissolved in HCOOH/DMF/MeOH (0.05:3.3:1) (1.9 mL), followed by addition of 20% Pd(OH)2/C (6.1 mg, ca. quarter of peptide mass). Then, the reaction flask was flushed with N2/vacuum cycles (×3) and filled with H2. The reaction mixture was stirred under H 2 at r.t. for 1 h (monitored by HPLC-MS). The catalyst was removed by filtration under Celite and washed with MeOH. Filtrate was collected in a round bottom flask and solvent was removed under reduced pressure. The residue was dissolved in CH 3CN:H2O and lyophilised. Purification was conducted by semi-Preparative HPLC using a 0-50% gradient over 25 min, with detection at 220 and 280 nm. Pure fractions were collected and lyophilised to afford pure peptide 12 as a white solid (2.3 mg, 10% yield).
Fmoc-Pro-OH, Fmoc-Leu-OH, Fmoc-His(Trt)-OH, Fmoc-Ile-OH, Fmoc-Ser(Trt)-OH, Fmoc-Lys(Boc)-OH and Fmoc-Phe-OH were used as building blocks. After cleavage as described above, the peptide crude was precipitated by adding cold Et 2O (dropwise) and the resulting precipitate was decanted and dried to afford 36 mg of a white solid corresponding to the protected peptide. Then, 16 mg of cleaved peptide (1.0 eq.), PyOxim (1.5 eq.) and OxymaPure (1.5 eq.) were dissolved in DMF:ACN (1:1, 0.001 M). After setting the cocktail at -10 ˚C using a salted ice bath, DIPEA (3.0 eq.) was added, and the mixture stirred overnight at r.t. After S8 solvent removal under reduced pressure, the crude peptide was dissolved in TFA:TIS:H2O (95:2.5:2.5) for 1h to remove the side-chain protecting groups. Then, the crude was concentrated under reduced pressure followed by precipitation in cold Et 2O (dropwise).
Purification was conducted by semi-Preparative HPLC using a 0-60% gradient over 25 min, with detection at 220 and 260 nm. Pure fractions were collected and lyophilised to afford pure peptide 14 as a white solid (3.0 mg, 26% yield from cyclization step).
Fmoc-Lys(Boc)-OH, Fmoc-Ile-OH, Fmoc-Phe-OH and Fmoc-Trp-OH were used as building blocks. After cleavage as described above, the peptide crude was precipitated by adding cold Et 2O (dropwise) and the resulting precipitate was decanted and dried to afford 36 mg of a white solid corresponding to the protected peptide. Then, 16 mg of cleaved peptide (1.0 eq.), PyOxim (1.5 eq.) and OxymaPure (1.5 eq.) were dissolved in DMF:ACN (1:1, 0.001 M). After setting the cocktail at -10˚C using a salted ice bath, DIPEA (3.0 eq.) was added and the mixture stirred for 2 days at r.t. After solvent removal under reduced pressure, the crude peptide was dissolved in TFA:H 2O:DCM (30:2.5:67.5) for 40 min to remove the side-chain protecting groups. Then, the crude was concentrated under reduced pressure followed by precipitation in cold Et2O (dropwise). Purification was conducted by semi-Preparative HPLC using a 0-50% gradient S9 over 25 min, with detection at 220 and 280 nm. Pure fractions were collected and lyophilised to afford pure peptide 16 as a white solid (1.9 mg, 16% yield from cyclization step).
Pure fractions were collected and lyophilised to afford pure peptide 18 as an orange solid (3.0 mg, 27% yield).

Computational details
DFT and TD-DFT calculations were performed with the M06-2X hybrid exchange-correlation functional [3] and the 6-311+G(2d,p) Pople basis set as implemented in the Gaussian 09 [4] package. This choice is supported by previous benchmarks performed on aza-BODIPY and BODIPY dyes, [5] which demonstrate that this level of theory provides good consistency with experimental trends for optical spectra, yet a systematic overshooting of the transition energies (by c.a. 0.4 eV). However, this systematic error is not a concern for the present study as we are not interested in theoretically reproducing the experimental spectra, rather than comparing the different molecules studied (e.g., transition state barriers) on the same footing. Numerical frequency calculations were used to ascertain the nature of the stationary points and the same increased integration grid (i.e., ultrafine (99,590)) respect to the default setting was used in all computations as this is recommended for describing correctly very low frequency modes. S11

Experimental protocols for spectroscopical and biological assays
Fluorescence spectra and intensity acquisition.
Absorbance and emission spectra were determined in the range of 400-700 nm (every 2 nm) at the indicated concentrations on 96 or 384-well plates using a BioTek Cytation 3 spectrophotometer. Environmental sensitivity was measured by comparing the fluorescence emission in MeOH vs glycerol (compounds 4-7) and in presence of liposome suspensions in PBS or PBS alone (compound 17).

Measurement of extinction coefficients.
For extinction coefficient measurements, the absorbance of each sample at the maximum excitation wavelength was recorded and the extinction coefficient was then determined by fitting the data to Beer's law.

Measurement of quantum yields.
Quantum yields were determined by measuring the integrated emission area of the fluorescence spectra in PBS or in the presence of phosphatidylcholine: cholesterol liposome suspensions in PBS and comparing it to the area measured for a reference compound rhodamine 101 in MeOH as the reference compound. [6] Different working solutions of compounds and reference ranging

Stability tests of peptide 17 in urine samples.
To assess photo-and urine stability, probe 17 was added into diluted urine samples (urine: water, 1:6) and incubated in 384-well plates at 37°C. For photostability determination, the fluorescence emission (644 nm) of probe 17 (30 µM) was monitored using a BioTek Cytation 3 spectrophotometer. Values were obtained as means from three independent experiments with n=3. To determine chemical integrity, the absorbance (570 nm) of probe 17 (100 µM) was monitored by HPLC-MS at different timepoints.

Culture of fungal strains.
All strains used in this experiment were grown on SAB agar at 37 °C for 3 days. Cells were harvest using sterile inoculation loop by taking a single colony and resuspending in PBS supplemented with 0.1% tween-20 (PBST). The concentration of cells was then quantified with a haemocytometer. For the determination of the minimum inhibitory concentration, cell density was adjusted to 10 6 cells mL -1 with 20% liquid Vogel's medium.

Culture of E. coli.
E. coli was grown on Lysogeny Broth (LB) agar at 37 °C for 1 day and harvested using sterile inoculation loop by taking a single colony and resuspending in PBST. The number of cells was quantified using a haemocytometer. For the determination of the minimum inhibitory concentration, cell density was adjusted to 10 6 cells mL -1 with 20% liquid LB medium.

In vitro measurements of minimum inhibitory concentrations.
Minimum inhibition concentration (MIC) measurements were performed as described previously with minor changes. [7] Each compound was dissolved in DMSO at concentration of 100 mM, this was used as the stock solution. For testing the MIC, the stock solution was further diluted in water to reach a concentration of 1 mM and was added in to a 96-well plate cell culture plate. A serial dilution was then performed within the 96-well plate, and the compound solutions at different concentrations were then mixed with conidia suspended in 20% Vogel's medium to reach a final volume of 100 µL per well. The final conidia concentration was 5 × 10 5 cells mL -1 in 20% Vogel's medium, the highest tested concentration of each compound was at 50 mg mL -1 . After 48 h incubation at 37°C, MIC was determined by brightfield microscopy from three independent experiments (n = 3). For testing the MIC against E. coli, the same protocol was followed apart from using liquid LB as the medium at a final concentration of 10%.

Confocal live-cell microscopy.
Probe 17 was mixed with Candida spp. to reach a final concentration of 10 µM and a final cell concentration of 5 × 10 5 cells/mL in PBS. The cells combined with the peptide were dispensed into the wells of Ibidi μ-slide 8 well (Ibidi GmbH, Germany) and incubated for 10 min at r.t.
Live cell imaging of the germinated spores was performed using a Leica TCS SP8 confocal laser scanning microscope equipped with photomultiplier tubes, hybrid GaAsP detectors and a 63× water immersion objective and white light laser (575 nm was used for excitation S13 wavelength and 600-650 nm was used for emission). Images were taken at 15, 30 and 60 min timepoints. Images were processed using Imaris software 8.0 developed by Bitplane (Zurich, Switzerland).