Phagemid cloning and phage production

The M13 phage (M13K07 New England Biolabs, Ipswich, MA, USA) has been genetically modified to display a specific targeting peptide, CPDIERPMC, fused in frame with the pIII capsid protein [23]. The peptide coding sequence (CDS) was obtained by annealing oligonucleotides AD0186 (CATGGCCTGTCCAATTGAAGATAGGCCTATGTGTGGTGGCGGTG) and AD0187 (GATCCACCGCCACCACACATAGGCCTATCTTCAATTGGACAGGC) generating NcoI and BamHI overhangs. This insert was directionally cloned into the pSex81 vector (ProGen, Heidelberg, Germany) linearized with the same restriction enzymes (NEB, New England Biolabs). The resulting phagemid (pPK24), carrying the CPDIERPMC peptide fused in frame with the pIII CDS, preceded by a PelB leader sequence allowing for periplasmatic localization and virion incorporation of the fusion protein (Fig. 1I), was validated by Sanger sequencing and Immunoblotting for expression of the fusion construct. After transformation in E. coli TG1, positive colonies were selected by ampicillin resistance and superinfected with Hyperphage helper phage (ProGen), to achieve pentavalent type 3 display of the pIII peptide fusion. Infected bacteria were grown overnight in LB medium supplemented with ampicillin (100 mg/L), kanamycin (25 mg/L) and 0.4 mM isopropyl ß-D-1-thiogalactopyranoside (IPTG). IPTG was added to induce the expression of the CPDIERPMC-pIII fusion, ampicillin was added to select pPK24-positive bacteria, and kanamycin was added to select only Hyperphage-superinfected bacteria. After growth, culture was pelleted for 30 min at 6000g to remove bacteria, the supernatant with the M13CC virions was collected and supplemented with 4% w/v of polyethylene glycol (PEG) 8000 and 3% w/v NaCl. The solution was incubated for 90 min at 4° C and then centrifugated for 15 min at 15000g at 4 °C. Pelleted phages were resuspended in sterile phosphate buffered saline (PBS) 1×. Phage concentrations were calculated by measuring the absorbance at 269 nm in a UV–Vis spectrophotometer (Agilent Cary-60, Agilent, Santa Clara, CA, USA) using an extinction coefficient of ε = 3.84 cm2 mg−1. The wild type phages (M13) were produced starting from un-transformed TG1 colonies grown to OD600nm = 0.4 and infected with M13K07 Helper phage (NEB), culture was growth overnight in kanamycin supplemented LB for the selection of M13K07-infected bacteria. Purification was performed as described before.

Fig. 1

M13CC targets CC cells. A Phage modification scheme, The CPDIRIERPMC coding sequence was cloned in frame with pIII (orange) generating the pPK24 plasmid. Transformed bacteria superinfected with Hyperphage helper produce modified phages. Immunoblot of M13CC demonstrates incorporation of modified pIII in the purified virions. Immunohistochemical confocal microscopy of HT29 (B, D, F, H) and DLD1 (C, E, G, I) cell lines incubated with PBS (B, C), M13 (D, E), M13CC (F, G) and M13CC after preincubation with fibronectin (FN) (H, I). Nuclei are in cyan while the major coating protein pVIII of phage is in magenta. Scale bar = 50 µm. J Quantitative analysis performed on confocal images. Fluorescence intensity detected in confocal images were expressed as fold increase in comparison to the control PBS (dashed line) (n = 20). Statistical significance was calculated by one-way parametric ANOVA in comparison to the control (PBS); ****p < 0.0001

Immunoblotting

M13CC phages at a concentration of 10^10 pfu/µL were resolved on a 12% SDS–polyacrylamide gel. Following electrophoresis, the proteins were transferred onto a PVDF (polyvinylidene difluoride) membrane (Immobilion-P, Millipore, France). The membrane was then blocked with a Blocking Solution (1X PBS pH 7.4, 5% milk, and 0.05% Tween), and then incubated with Anti-M13 pIII Monoclonal Antibody (New England BioLabs, Ipswich, MA, USA) diluted 1:5000 in the blocking solution for 1 h at room temperature. After this incubation, the membrane was then washed three times with Washing Solution (1X PBS and 0.05% Tween), and further incubated with a horseradish peroxidase (HRP)-conjugated IgG anti-mouse secondary antibody (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) diluted 1:10,000 for 1 h at room temperature. The membrane was developed using an enhanced chemiluminescence (ECL) solution, comprising 1.25 mM luminol in 100 mM Tris (pH 8.8), 6.8 mM coumaric acid, and 30% hydrogen peroxide. Chemiluminescent signals were captured using the ChemiDoc™ Imaging System (Bio-Rad, Basel, Switzerland).

Phage vector conjugation with the RB photosensitizer

The RB carboxylic group was covalently conjugated to the free amine-containing residues on the capsid of M13 and M13CC via cross-coupling reaction using 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC)/N-hydroxysuccinimide (NHS), resulting in the formation of amide bonds. NHS and EDC were added under stirring to a 10 mM RB solution in DMSO, reaching final concentrations of 15 mM NHS and 10 mM EDC. The mixture was incubated in dark conditions for three hours at 25 °C under constant shaking at 700 rpm. Then, 50 μL of the activated RB solution were added dropwise to 1 mL of PBS solution containing 40 nM of phages (M13 or M13CC). After the addition, each solution was incubated overnight in dark conditions at 25 °C, under shaking at 700 rpm. To remove unreacted RB and other reaction byproducts, the bioconjugates were dialyzed (14 kDa cut-off regenerated cellulose membrane) versus 100 mM sodium carbonate buffer at pH 9. The last dialysis step was carried out versus PBS.

Cell cultures

Authenticated CC cell lines HT29 (mismatch repair [MMR] proficient cells), DLD1 and HCT116 (MMR deficient cells), as well as non-CC cell line A431 and BT474, were purchased from American Type Culture Collection (ATCC, Manassas, VA, USA). HT29 and HCT116 were cultured in McCoy’s 5A medium, while DLD1, A431 and BT474 in RPMI 1640. Both media were supplemented with 10% heat inactivated fetal bovine serum (FBS), 1% l-glutamine 200 mM and 1% penicillin/streptomycin solution 100 U/mL (all purchased from Euroclone, Pero, Italy). Cells were cultured at 37 °C in a 5% CO2 humidified incubator.

Spheroid generation

To reproducibly generate multicellular spheroids of uniform size, we cultured CC cells using the low adhesion multiwell plate. Spheroids were formed using 96U Bottom Plate (Nunclon Sphera, Thermo Fisher Scientific, Waltham, MA USA). Briefly, 500 cells were seeded as 100 μL per well by centrifugation. After seeding, plates were cultured at 37 °C with 5% CO2 for 5 days to allow cell assembly and the formation of the spheroids. Half of the culture medium was replaced every three days. All steps were performed using an automatic multichannel pipette at a flow rate of 10 μL/s (Gilson, Middleton, WI, USA). Following microtissue formation, spheroids were transferred in the GravityTrap recipient plate (Insphero AG, Schlieren, Switzerland) allowing monitoring of the spheroid growth, viability, and microscope analysis. Spheroids displayed a slow, exponential, and homogeneous growth that we monitored for 5 days. According to literature data [27], proliferating cells were mainly detectable in the ring of cells localized in the outer layer of the spheroid, while the inner core was characterized by live quiescent cells or dying cells.

Cell viability assay and analysis of cell death mechanisms

Cells were treated and incubated in complete medium with increasing PS concentrations (0.01 to 1 µM) of RB alone or M13CC-RB, corresponding to picomolar concentrations of bacteriophage, for 45 min. After incubation, cells were washed twice with PBS 1X, to remove the excess of bioconjugated phages.

Cells were irradiated for 30 min or shorter time according to experimental exigencies in PBS with a low irradiance white light LED (24 mW/cm2). These conditions were determined by comparing the efficacy of several irradiation times (from 10 to 30 min). No significant differences were observed at the highest tested concentrations of RB equivalents (1 µM). However, a significant difference was observed at 0.1 µM, where we recorded a higher percentage of cytotoxic effects for M13CC-RB after 30 min exposure compared to 10 min (data not shown). For this reason, 30 min light exposure was used as experimental condition for all the following experiments, unless differently reported. In parallel, cells were treated according to the before mentioned conditions and kept in the dark. After irradiation or dark incubation, cells were recovered for 24 h or shorter in complete medium. The analysis of cell viability was performed using the metabolic MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay (Merck, Darmstadt, Germany) for both irradiated and not irradiated cells, as previously described [22].

Cell viability after PDT with M13CC-RB in presence of ROS inhibitor was performed as described above with few modifications. During incubation of cells with M13CC-RB, irradiation and recovery after treatment, media were supplemented with either N-acetyl-l-cysteine (NAC) or vitamin E at the final concentration of 10 mM and 100 µM, respectively.

To investigate the mechanism of cell death induced by M13CC-RB, after 3, 6 or 24 h from treatment, cells were washed, gently trypsinized using TrypLE (Gibco, Thermo Fisher) and incubated with 100 µL of Guava Nexin Reagent (Luminex, Austin, Texas, USA), containing 7-aminoactinomycin (AAD) and annexin V-phycoerythrin (PE), for 20 min at room temperature in the dark. After incubation, cells were analyzed using the flow cytometer Guava EasyCyte 6-2L (Luminex). Considering that at the highest concentration of M13CC-RB tested we recorded a signal spill over into the PE fluorescence channel, to confirm that the autofluorescence spill over does not create an artifact, we performed an additional test, the supravital 7-AAD analysis. We incubated cells with 7-AAD (1 h, room temperature) and recorded 7-AAD fluorescence and forward scatter (FSC). The long incubation permits a discrimination of viable cells which exclude the dye totally, from apoptotic cells that fluoresce dimly, and necrotic and late apoptotic cells that fluoresce brightly and are characterized by a decrease in FSC [28, 29].

To measure caspase activation, the Caspase-Glo® 3/7 luminescent assay (Promega) was used on cells incubated with different concentrations of M13CC-RB, 3 and 24 h post-irradiation. Control samples were kept in the dark. This single "add-mix-measure" reagent, was added to the photosensitized cell cultures resulting in cell lysis, followed by caspase cleavage of the substrate and generation of a “glow-type” luminescent signal recorded in an EnSpire (PerkinElmer) multimode plate reader, all according to the manufacturer instructions.

To monitor spheroids growth and homogeneity and to preliminary assess the eventual cytotoxic effect of M13CC-RB on multicellular tumor spheroids, CyQUANT® cell proliferation assay (Invitrogen, Thermo Fisher) was used, according to manufacturer’s instructions. Briefly, after treatment with M13CC-RB (45 min incubation) and irradiation (30 min) spheroids were incubated with the dye CyQUANT® GR properly diluted (400-fold) into the 1X cell-lysis buffer purchased in the kit and 20-fold diluted in distilled water prior to use. Solutions have been kept in the dark and used within a few hours of their preparation. Spheroids were then analyzed using the Nikon Eclipse Ti equipped with Digital Sight camera DS U3 (Nikon, Tokyo, Japan) or Nikon A1R confocal microscope (Nikon).

Spheroid viability was quantitatively assessed after treatment with phage M13CC-RB or RB alone at the same concentrations of RB equivalents (0.01 to 3 µM). Spheroids were incubated with phage M13CC-RB or RB for 45 min and irradiated for 30 min. After 24 h, the luminescent assay 3D CellTiter-Glo® (Promega, Madison, WI, USA) was employed to quantify the decrease in cell viability. Briefly, 100 µL of 3D CellTiter-Glo® were added to the complete medium 24 h after the treatment, solution was pipetted about 50 times for each well to disaggregate the spheroids. Luminescence was recorded using EnSpire multimode plate reader (PerkinElmer, Walthman, MA, USA). Data were normalized on the viability of untreated spheroids.

Analysis of ROS generation in culture medium and intracellularly

To determine the M13CC–RB ability to produce peroxides upon visible light irradiation, Amplex® Red (AR) assay was used in a cell-free system. AR is a nonfluorescent and colorless molecule that reacts with peroxides, catalyzed by horseradish peroxidase (HRP), producing a fluorescent dye with an absorption maximum at 563 nm and an emission maximum at 587 nm, named resorufin. The number of generated peroxides was determined as the difference between resorufin produced by irradiated and not irradiated samples. The working solution (WS), containing AR and HRP, was prepared by mixing 1 mL of 50 mM PBS at pH 7.4 with 10 μL of AR 50 mM solution dissolved in DMSO and 10 μL of PBS solution of HRP 0.4 mg/mL. Sample solutions at different concentrations (0, 0.25, 0.5, 1, 2, or 4 μM) of RB and M13CC–RB were prepared in PBS 50 mM. Two identical 96 multiwell plates were prepared with the sample solutions (90 μL), using three technical replicates for each concentration. One plate was irradiated for 30 min with a white LED lamp (Valex cold white LED, irradiance on the plate surface = 24 mW/cm2, measured with the photo-radiometer Delta Ohm LP 471 RAD), while the second was kept in dark, as control. 10 μL of WS was then added to each well on both the plates. After the addition, the plates were incubated in the dark for 30 min, then the emission intensity was measured at 590 nm (λex = 530 nm). Standard solutions of H2O2 were used to generate a calibration curve for converting the fluorescence intensity signal into the concentration of peroxides. Fluorescence measurements were carried out using a PerkinElmer EnSpire® Multimode Plate Reader.

To quantify the singlet oxygen (1O2) quantum yield (ΦΔ) of photoirradiated M13CC–RB, 9,10-anthracenediyl-bis(methylene) dimalonic acid (ABMDMA) was used as 1O2 detector in a cell-free system. The singlet oxygen produced in the solution upon visible light irradiation reacts with ABMDMA giving an endoperoxide and resulting in the bleaching of ABMDMA. The photoinduced generation of singlet oxygen was evaluated from the decrease in the absorbance at 400 nm. Briefly, 400 μL of iso-absorbing RB and M13CC–RB sample solutions were prepared in PBS, dissolved in D2O, at 15 μM of RB. 2 μL of a 5 mM ABMDMA stock solution in DMSO were added to each solution under investigation. The samples were then irradiated with a white LED lamp (irradiance on the cuvette surface = 2.4 mW/cm2) while maintained under stirring. ΦΔ of M13CC–RB was determined using the following equation ΦΔS = kS/kR × ΦΔR where k is the slope of the photodegradation rate of ABMDMA, S is the sample under investigation (M13CC–RB), R is the reference (RB), and ΦΔR is the 1O2 quantum yield of the reference (RB) which is known to be ΦΔR = 0.76 [30].

To assess the intracellular ROS production 20,000 HT29 cells were seeded in a 96 well-plate. Cells were treated and incubated in complete medium with increasing concentration of M13CC-RB (0.3–1 µM) at picomolar concentrations of bacteriophage or RB alone for 45 min. After incubation, cells were washed twice with PBS, to remove the excess of bioconjugated phage or PS. Cells were irradiated for 30 min in PBS. In parallel, to check eventual ROS generation in dark condition, cells were treated according to the before mentioned conditions and kept in the dark. After irradiation, 100 µL of ROSGlo (Promega) were added to each well and cells were incubated for 20 min at room temperature. Luminescence was read using EnSpire® Multimode plate reader (PerkinElmer).

Microscopic analysis

The modified tropism of M13CC toward CC cells was demonstrated by immunohistochemical confocal microscopy. 50,000 cells were seeded on round coverslips and grown overnight in the incubator. The cells were then incubated with or without 0.1 mg fibronectin (FN) (Superfibronectin, Sigma-Aldrich) for 30 min, followed by the addition of 10^12 phages (M13/M13CC) and incubation for 45 min. Unbound or excess phages were removed by washing twice with PBS, and the cells were fixed with 4% paraformaldehyde (PFA) in PBS for 15 min at room temperature. The cells were then washed with PBS + Tween 20 0.05% (Washing Buffer -WB) and permeabilized for 15 min with Triton 0.1%. Cells were incubated for 45 min with blocking solution (2% milk in PBS), washed with WB and incubated for 1 h in WB supplemented with mouse monoclonal anti-M13/fd/F1 filamentous phage (Progen) diluted 1:500. The cells were then washed three times with WB and incubated for 1 h in the presence of AlexaFluor568™ goat anti-mouse (Invitrogen). The excess of secondary antibody was removed by three washes with WB. Cells were stained for 15 min with Hoechst 33342 (1 µg/µL). Round coverslips were then washed and placed in an Attofluor cell chamber (Invitrogen) with 1 mL of PBS. Images were acquired with the Nikon A1R confocal microscope, and the laser settings were kept fixed for all images acquired.

Similarly, to assess specific retargeting of M13CC-RB towards CC cells, HT29 and DLD1 were seeded on round coverslips and grown overnight in the incubator. Next, cells were incubated with an equivalent phage concentration of 1 µM of RB in complete cell medium for 45 min at 37° C, washed and stained with Hoechst 33342 (1 µg/µL), used as nucleic dye, for 10 min. Round coverslips were then washed and fitted into an Attofluor cell chamber (Invitrogen) with 1 mL of 10% FBS and DMEM without phenol red supplemented with 1% penicillin/streptomycin. Images were acquired with Nikon A1R confocal microscope. Fluorescence quantification was performed on images acquired with Fiji free software.

Spheroid images were taken in bright field using Zeiss Axiovert 40 CFL microscope (Zeiss, Oberkochen, Germany) to evaluate the structure of each spheroid and select the most similar ones for further analyses. For the penetration evaluation, spheroids were incubated for 45 min in a single drop of complete McCoy cell medium added with M13CC-RB conjugate phage or RB alone at a concentration of 3 µM RB equivalents. Calcein and Hoechst 33342 were added to the spheroids at concentrations of 100 nM and 1 µg/µL, respectively.

Evaluation of penetration on fixed and optically cleared spheroids was performed as previously described by Nürnberg and colleagues with some modifications [31]. 3D cell cultures of HT29, DLD1 and HCT116 were generated, incubated with M13CC-RB for 45 min, washed thrice with PBS to remove unbound phages and then fixed with 4% PFA (Merck) for 1 h. Spheroids were then washed with PBS supplemented with FBS 1%, permeabilized with a solution of PBS and Triton 1% for 2 h with gentle shaking, and washed again with PBS supplemented with FBS 1%. Spheroids were then quenched for 1 h with 0.5 M glycine in PBS and then with a solution of 0.3 M glycine, 20% DMSO and 0.2% Tryton X-100 in PBS for 30 min. Samples were washed twice with PBS supplemented with FBS 1% and subjected to optical clearing with 88% glycerol for 18 h at room temperature. Spheroids were stained with 1 µg/µL of Hoechst 33342 and then observed with Nikon A1R confocal microscope using ND-acquisition z-stack.

For the conformational change evaluation to support the spheroid viability assay, samples were incubated with M13CC-RB conjugate phage or RB alone and Hoechst 33342 at the final concentration of 1 µg/mL. Samples were treated with low irradiance white light for 30 min as described before or kept in the dark. After 24 h, the spheroids’ structure was investigated in brightfield and with Nikon A1R confocal microscope using ND-acquisition z-stack to evaluate the penetration of the conjugated phage and the structural modifications.

Flow cytometry

The tropism of M13CC-RB or free-RB to HT29 and DLD1 cell lines was evaluated by flow cytometry. Free-RB or bioconjugated phage were incubated in complete media, at the final RB concentration of 1 µM, with 500,000 adherent cells for 45 min. Unbound phages/sensitizer were removed by washing thrice with PBS and cells were detached using trypsin 1x. After trypsin inactivation with complete media, cells were washed with PBS, resuspended in 0.5 mL of PBS and analyzed with CytoFLEX S (Beckam Coulter). The fluorescence of at least 10,000 events was evaluated in the PE channel. Data analysis was achieved with CytExpert (Beckam Coulter) and FlowJo™ 10.0.7r2 version (Becton Dickinson, Franklin Lakes, NJ, USA).

Statistical analyses

The results are expressed as mean ± SEM of at least three independent experiments. One or two-ways parametric Anova, or t-test were used for the comparison of the results and Dunnett or Tukei were used as post-tests. The statistical software GraphPad InStat 8.0 version (GraphPad Prism, San Diego, CA) was used and p < 0.05 was considered significant.