Production, purification, and characterization of E. coli GMMA

GMMA was produced from a ΔtolR E. coli BL21(DE3) mutant strain, mutated in order to increase the rate of vesicle release through the deletion of tolR gene. Briefly, an overnight Luria-Bertani broth preinoculum was diluted 1:50 in HTMC medium to grow bacteria at 30 °C under shaking (180 rpm). After 8 h, bacteria were pelleted with a 45’ centrifugation to collect the supernatant containing GMMA vesicles. GMMA were then purified through tangential flow filtration (100 kDa Hydrosart membrane, Sartorius), followed by an ultracentrifugation step (50,000 rpm, 1 h, 4 °C), resuspending GMMA pellet in PBS. Purity and integrity of GMMA were verified by HPLC-SEC analysis using TSKgel 6000PW + G4000PW columns in series and by DLS, respectively54.

Synthesis of compound 5

To a solution of compound 3 (252 mg, 0.55 mmol) in DMF (5 mL), HOBt (335 mg, 2.2 mmol), EDAC (420 mg, 2.2 mmol), and NMM (600 µL, 3.3 mmol) were added. The suspension was stirred at rt for 10 min and then linker 3 (260 mg, 0.66 mmol) was added. The mixture was stirred at rt for 2 h until TLC (CH2Cl2:CH3OH = 9:1) showed the disappearance of the starting material. The suspension was diluted with EtOAc (50 mL), washed with citric acid 10% (w/w) (3 × 30 mL), then with NaHCO3 s.s. (1 × 30 mL) and eventually brine (1 × 30 mL). The organic layers were dried over Na2SO4 and concentrated under vacuum to obtain a crude product (426 mg) that was purified by flash column chromatography on silica gel (CH2Cl2:CH3OH = 95:5) to give compound 5 (317 mg, 0.48 mmol, 90% yield) as a white foam. For characterization data see Supplementary.

Synthesis of compound 7

Compound 5 (360 mg, 0.61 mmol) was dissolved in NH3 in CH3OH 4 M (5.5 mL). After stirring for 3 h, the reaction mixture was concentrated under vacuum to give derivative 7 (288 mg, >95% yield) as a light-yellow foam, which was used in the next step without any further purification. For characterization data see Supplementary.

Synthesis of compound 6

To a solution of compound 7 (166 mg, 0.36 mmol) in CH3OH (5 mL) and 5 drops of H2O, Pd/C (80 mg, 0.18 mmol) was added under the N2 atmosphere. The mixture was stirred at rt under H2 atmosphere for 18 h until TLC control (CH2Cl2:CH3OH = 8:2) showed no presence of the starting material. The mixture was filtered through a pad of Celite® and the solvent was evaporated under a vacuum to obtain pure compound 6 (140 mg, >95% yield) as a white foam. For characterization data see Supplementary.

Synthesis of compound 1

To a solution of crude compound 6 (105 mg, 0.24 mmol) and NMM (110 µL, 0.97 mmol) in dry DMF (2 mL), bis-succinimide adipate 8 (330 mg, 0.97 mmol) was added. After stirring the mixture for 16 h at rt the precipitate was discarded, and the solvent was evaporated under vacuum. The residue was suspended in EtOAc and the precipitate was purified by flash chromatography on silica gel (CH2Cl2:CH3OH = 85:15 to 70:30) to give pure compound 1 (55 mg, 40% over 3 steps). For characterization data see Supplementary.

Synthesis of compound 9

To a suspension of 7 (140 mg, 0.3 mmol) in acetone (3 mL), DMP (2 mL, 15 mmol) and pTsOH (until pH 3) were added and the solution was stirred at room temperature for 15 h. After complete conversion, the mixture was quenched with Et3N (pH 8), diluted with CH2Cl2 (50 mL), and washed with HCl 1 M (2 × 25 mL) and water (2 × 25 mL). The organic layer was dried over Na2SO4 and concentrated under a vacuum. The crude was purified by flash chromatography on silica gel (EtOAc:CH3OH = 95:5 to 90:10) to afford compound 9 as a white solid (115 mg, 65% yield). For characterization data see Supplementary.

Synthesis of compound 11

To a suspension compound 9 (454 mg, 0.91 mmol) and 10 (1.92 g, 2.9 mmol) in a mixture of dry CH3CN:CH2Cl2 = 10:1 (v/v, 11 mL) cooled to –40 °C, NIS (1.0 g, 4.55 mmol) and TfOH (140 μL, 2.3 mmol) were added and the solution was stirred at –40 °C under N2 atmosphere. After 2 h, the reaction was quenched with Et3N until pH 8 and the mixture was diluted with CH2Cl2 (100 mL) and washed with Na2S2O3 1 M (4 × 50 mL). The organic layer was dried over Na2SO4 and concentrated under a vacuum to give the crude product. This latter was solubilized in AcOH 80% in H2O (v/v, 10 mL) and the reaction mixture was stirred overnight at 40 °C. The solvent was removed under vacuum and the product was purified by flash chromatography on silica gel (CH2Cl2:CH3OH = 96:4 to 90:10) to give 11 (541 mg, 65% yield calculated over 2 steps). For characterization data see Supplementary.

Synthesis of compound 12

To a solution of compound 11 (345 mg, 0.34 mmol) in EtOH (20 mL) Pd/C (200 mg, 1.0 mmol) was added under the N2 atmosphere. The mixture was stirred at rt under H2 atmosphere for 3 days then the mixture was filtered through a pad of Celite® and the solvent was evaporated under vacuum to obtain crude as a light-brown foam. The crude was dissolved in NH3 4 M in CH3OH (4 mL) and the reaction mixture was stirred at room temperature. After 2 days, the solvent was removed under vacuum and the product was purified by washing several times with Et2O to give pure 12 (170 mg, 70% yield over two steps) as light-brown foam. For characterization data see Supplementary.

Synthesis of compound 2

To a solution of compound 12 (40 mg, 0.06 mmol) in dry DMF (2 mL) and NMM (50 µL, 0.25 mmol), bis-succinimide adipate 8 (75 mg, 0.22 mmol) was added. After stirring the mixture for 16 h at rt the precipitate was filtered off, and the solvent was evaporated in vacuo. The residue was suspended in EtOAc and the precipitate was washed several times with EtOAc to give activated compound 2 (25 mg, 50% yield) as a light-brown foam which was used immediately. For characterization data see Supplementary.

Synthesis of compound 14

To a solution of compound 3 (253 mg, 0.46 mmol) in DMF (8 mL), HOBt (250 mg, 1.84 mmol), EDAC (350 mg, 1.84 mmol), and NMM (300 µL, 2.76 mmol) were added. The suspension was stirred at rt for 10 min and then 13 (228 mg, 0.69 mmol) was added. The mixture was stirred at rt for 1.5 h until TLC (CH2Cl2:CH3OH = 9:1) showed the disappearance of the starting material. The suspension was diluted with EtOAc (100 mL), and washed with H2O (6 × 70 mL). The organic layers were dried over Na2SO4 and concentrated under vacuum to afford pure compound 14 (247 mg, 92% yield) as a light-yellow foam. For characterization data see Supplementary.

Synthesis of compound 15

To a solution of compound 14 (211 mg, 0.36 mmol) in a mixture of THF:H2O = 4:1 (v/v, 10 mL) LiOH.H2O was added (75 mg, 1.8 mmol). The mixture was stirred at rt for 19 h until TLC (CH2Cl2:CH3OH = 8:2) showed the disappearance of the starting material. The mixture was evaporated under reduced pressure to give crude compound 15 (260 mg, >95% yield) as a white foam which was used without any further purification. For characterization data see Supplementary.

Synthesis of compound 16

To a dispersion of compound 15 (73 mg, 0.17 mmol) in dry DMF (5 mL), HBTU (120 mg, 0.32 mmol), and DIPEA (80 µL, 0.45 mmol) were added. The mixture was stirred at rt until dissolution and then NHS (33 mg, 0.29 mmol) was added. The mixture was stirred at rt and monitored via TLC (CH2Cl2:CH3OH = 8:2) for 16 h until the disappearance of the starting material. The dispersion was evaporated under reduced pressure to give crude (270 mg) which was dispersed in acetone. The precipitate was filtered and washed with acetone (3 × 2 mL) to give activated compound 16 (57 mg, 62%) as a light-brown cerous solid which was stored under an N2 atmosphere at −10 °C. For characterization data see Supplementary.

Synthesis of Tn_mime[28]BSA

BSA (19.8 mg, 0.3 µmol) was dissolved in NAPi Phosphonate Buffer (2 mL, 150 mM, pH 7.4) and slowly treated with a solution of 16 (57 mg, 46 µmol) dissolved in dry DMF (150 µL). The mixture was stirred at 4 °C for 24 h, the success of the reaction was proved by an SDS–PAGE, and then the conjugate was centrifuged to remove the brown solid precipitate. The water layer was purified by HPLC-SEC and concentrated to 100 µM. using a 35 kDa MWCO membrane centrifugal device (Millipore). The number of functionalized lysine was confirmed by MALDI/TOF-TOF UltraFlex III analysis: about 28 glycans were linked to BSA (Tn_mime[28]BSA) (see Fig. S1).

Synthesis of STn_mime[8]BSA

BSA (10 mg, 0.15 µmol) was dissolved in NAPi Phosphonate Buffer (2 mL, 150 mM, pH 7.4) and slowly treated with a solution of 2 (25 mg, 26 µmol) dissolved in dry DMF (150 µL). The mixture was stirred at 4 °C for 24 h, the success of the reaction was proved by an SDS-PAGE, then the conjugate was centrifuged to remove the brown solid precipitate. The water layer was purified by HPLC-SEC and concentrated to 100 µM. using a 35 kDa MWCO membrane centrifugal device (Millipore). The number of functionalized lysine was confirmed by MALDI/TOF-TOF UltraFlex III analysis: 8 glycans were linked to BSA (STn_mime[8]BSA) (see Fig. S2).

Synthesis of derivative 18

To a solution of compound 173 (126 mg, 0.27 mmol) in dry DMF (7.5 mL) and NMM (50 µL, 0.25 mmol), bis-succinimide adipate 8 (342 mg, 1.0 mmol) was added. After stirring the mixture for 18 h at rt the precipitate was filtered off, and the solvent was evaporated in vacuo. The residue was suspended in EtOAc and the precipitate was washed several times with EtOAc to give activated compound 18 (93 mg, 50% yield) as a yellowish foam which was used immediately. For characterization data see Supplementary.

Synthesis of native Tn[22]BSA

BSA (20.0 mg, 0.3 µmol) was dissolved in NAPi Phosphonate Buffer (2 mL, 150 mM, pH 7.4) and slowly treated with a solution of derivative 18 (31 mg, 45 µmol) dissolved in dry DMF (150 µL). The mixture was stirred at 4 °C for 24 h, the success of the reaction was proved by an SDS-PAGE, then the conjugate was centrifuged to remove the dark yellow solid precipitate. The water layer was purified by HPLC-SEC and concentrated to 100 µM. using a 35 kDa MWCO membrane centrifugal device (Millipore). The number of functionalized lysine was confirmed by MALDI/TOF-TOF UltraFlex III analysis: about 22 glycans were linked to BSA (Native Tn[22]BSA) (see Fig. S3).

Synthesis and characterization of GMMA conjugates

After concentration of GMMA at ~10 mg/mL in 50 mM phosphate buffer pH 7.2, conjugation of each functionalized mimetic was performed with a GMMA/mimetic w/w ratio of 1:1, keeping reaction mixtures under agitation with a rotating wheel at room temperature overnight. Conjugates were purified through centrifugal ultrafiltration using Amicon Ultra device with a membrane cut-off of 100 kDa (Merck, Darmstadt, Germany), and removal of unconjugated mimetic was verified by HPLC-SEC following absorbance at 214 nm. The purified conjugates were analyzed by SDS–PAGE analysis to verify conjugate formation, as previously described (Vaccines. Vaccines 2020, 8, 540). Total protein recovery was quantified by Micro BCA, and the amount of conjugated mimetic was indirectly estimated by quantifying the remaining free amino groups on GMMA with 2,4,6-trinitrobenzenesulfonic acid (TNBS) assay55. Conjugates’ size distribution was determined by DLS.

Animals

Five-week-old female CD1 and BALB/c mice were purchased from Charles River (Italy). Animals were housed 4 per cage under a light-dark (12 h/12 h) cycle with ad libitum access to water and food. Experimental manipulations were performed under inhalation anesthesia induced by 4% and maintained by a 2.3% mixture of isofluorane in oxygen administered through facial masks. At the end of the study, all animals were anasthetized with gas anesthesia (2.5% isofluorane), to perform final blood collection and then sacrificed by cervical dislocation. Experiments were carried out in compliance with the bioethical principles adopted by Directive 2010/63/EU on the protection of animals used for scientific purposes and Italian Legislative Decree 4 March 2014, no. 26. Implementation of Directive 2010/63/EU on the protection of animals used for scientific purposes. The experimental protocol was approved by the Local Ethical Committee (Toscana Life Sciences Foundation’s Animal Welfare Body) and authorized by the Italian Ministry of Health (authorization nos. 924/2020-PR and 924/2020-PR).

Mice immunizations

For immunogenicity studies, 5-week-old female CD1 (8 mice/group) were subcutaneously immunized with two doses of different antigen formulations, given at 4 weeks intervals (day 0, T0; day 28, T28): (Group 1) GMMA-Tn (7 µg antigen/dose, conjugated to 21.4 µg of GMMA), (Group 2) a combination of GMMA and unconjugated Tn mimetic (21.4 µg GMMA + 7 µg Tn mimetic), (Group 3) GMMA-STn (7 µg antigen/dose conjugated to 20.5 µg GMMA), (Group 4) a combination of GMMA and unconjugated STn mimetic (20.5 µg GMMA + 7 µg STn mimetic), and (Group 5) CRM-Tn (7 µg antigen/dose conjugated to 41.7 µg CRM) used as benchmark control to compare the antibody responses. Sera were collected for immunogenicity analysis at day 27 (T27, post 1) and 42 from baseline (T42, 14 days after the second dose, post 2).

For efficacy studies, 5-week-old female BALB/c mice (8 mice/group) before tumor cells implantation were subcutaneously immunized with 6 doses, with the first and the second given at 4-week-interval (day 0, T0; day 28, T28) and the remaining 4 doses given at 1-week interval (day 36, T36; day 42, T42; day 56, T56; day 63, T63). Mice received the same antigenic formulations used for CD1 immunization; one additional mice group was included and treated with raw GMMA as negative control, in detail: (Group 1) raw GMMA (21.4 µg), (Group 2) a combination of GMMA and unconjugated Tn mimetic, (Group 3) GMMA-Tn, (Group 4) a combination of GMMA and unconjugated STn mimetic, (Group 5) GMMA-STn and (Group 6) CRM-Tn. Sera were collected for immunogenicity analysis before the first immunization (pre-immune), at day 27 (T27, 7 days after the first immunization, post 1) at day 35 (T35, 7 days after the second immunization, post 2), and at day 70 (T70, 7 days after the last immunization, post 6).

Cell lines

Murine 4T1-Luc2, T47D, and MCF7 were purchased from ATTC. Cells were cultured and passaged as recommended by the manufacturers. 4T1-Luc2 is a luciferase-expressing cell line derived from parental line CRL-2539 by transduction with a lentiviral vector encoding firefly luciferase gene (luc2) under the control of EF-1 alpha promoter.

Analysis of antigen-specific antibodies response by ELISA

The antibody response induced by immunizations against the Tn- and STn-mimetics or native Tn antigen, was measured by multiplex Enzyme-Linked Assay (ELISA). 384-well Maxisorp plates (Nunc, Thermo Fisher Scientific) were coated with: Tn mimetic or STn mimetic or native Tn, by adding 10 mL/well of a 5 mg/mL antigen-BSA conjugates’ solution in PBS buffer, followed by incubation overnight at 4 °C. After three washings with PBS buffer containing 0.05% Tween 20 (Sigma) (TPBS), a blocking step was performed by adding 35 mL of BSA solution at 5% in TPBS and incubating the plates for 1 h at 37 °C. After washing as above, up to eight two-fold serial dilutions of pre-immune, in TPBs were then performed along each column, plates were incubated for 1 h at 37 °C and washed 3×. Twenty mL/well of secondary antibody coupled to HRP, appropriately diluted in TPBS (goat anti-mouse IgG, Invitrogen) were added and the plates were incubated for 1 h at 37 °C. After four more washes, 20 mL/well of 3,3´,5,5´-tetramethylbenzidine (TMB) solution was added. After 10 min of incubation at room temperature, plates were read at 450 nm.

Flow cytometry analysis

For binding assays, T47D and 4T1 cells were grown to sub-confluency, harvested using enzyme-free cell dissociation buffer (Gibco), and centrifuged, counted, and resuspended in MACS buffer (Miltenyi Biotec) at a concentration of 105 cells/ml. Sera from different groups of mice (final dilution 1:100) were added to 50 μL of cells and incubated for 30 min at 4 °C. After two washes with MACS buffer, the cells were incubated with APC conjugated donkey anti-mouse IgG (Thermo Fisher Scientific) as a secondary antibody at a 1:500 dilution for 30 min at room temperature and in the dark to detect the bound antibody. Cells were washed and then stained with Live/Dead® Fixable Aqua Dead Cell Stain (Thermo Fisher Scientific) in order to eliminate dead cells contaminating the data. Cells were washed again before data acquisition on a BD FACSAria II Flow Cytometer (BD Biosciences) equipped with FACS DiVa software and quantified by FLOWJO software version 10 (LLC). Cells stained with sera from mice immunized with GMMA were designated as threshold.

To evaluate the effect of the presence of anti-GMMA antibodies on the specific binding of anti-Tn antibodies on the surface of T-47D cells, sera were depleted of anti-GMMA antibodies. Briefly, mouse sera were diluted 1:10 in 100 μL (two wells for each serum sample) of MACS buffer and incubated for one hour at 37 °C with GMMA precoated 96-well plates (Nunc MaxiSorp® flat-bottom). Sera incubated in the absence of GMMA antigen were used as a control. After incubation, the remaining depleted samples were collected and tested by flow cytometry analysis following the protocol described previously. To evaluate the content of specific anti-Tn antibodies in mouse sera, 10 μl of Tn mimetic or Tn native at 20 μg/ml in MACS buffer were mixed with 10 μl of sera for 1 h at 37 °C. After incubation 50 × 103 T47D cells were suspended in 80 μl of MACS buffer and added for 30 min at 4 °C. After two washes with MACS buffer, the cells were incubated with a secondary antibody following the protocol described previously.

Cytokines quantification

A ProcartaPlexTM Mouse and Rat Mix & Match Panels (Invitrogen, Cat. Number PPX-15-MXNKUN7) were used to measure a panel of cytokines and receptors. Mouse serum was diluted 1:10 for cytokines analysis. Sample preparation, together with kit standards, detection antibody, and streptavidin-PE, was performed according to the manufacturer’s magnetic immunoassay protocol. Standards and samples were run in duplicate. Cytokine fluorescence intensity was detected using the BioPlex 200 System machine with Multiplex Bio-Assay Analyzer. The standard data points (concentration vs. measurement) were plotted on semi-log axes and a five-parameter logistic (5PL) was made through the points. The concentrations of the samples were determined from the fit with specified dilution factors applied with BioPlex ManagerTM software version 5.0.

Implantation of 4T1-Luc2 TNBC cells in Syngeneic Mice and Tumor growth analysis

One week after the last immunization only mice groups 1, 3, and 5 were orthotopically implanted with a suspension of 8 × 104 4T1-Luc2 in 50 μl of PBS into the fat pad of the third pair of mammary glands using insulin syringes with 32 G needles. Tumor growth was assessed by bioluminescence imaging (BLI) (Ivis Lumina X-5; Perkin Elmer) at days 7 (W1), 14 (W2), and 21 (W3). At day 28 (W4) tumor volume was manually measured with a caliper. For BLI measurement, BALB/c mice bearing 4T1-LUC syngenic orthotopic tumors were injected intraperitoneally with 150 mg/kg of d-Luciferin (Caliper Life Sciences). Ten minutes post-injection, mice were anesthetized with 2% isoflurane and placed in a supine position with their abdomen facing the camera. Images were acquired for 10 seconds using the IVIS Lumina X5 Imaging System (PerkinElmer) following the manufacturer’s recommendations. Regions of interest (ROIs) were drawn around the tumor, and the total radiant efficiency (in photons/s/cm2/Sr) of the tumor was measured using Living Image 2.5 software.

Tumor volumes were calculated according to the formula: Tumor volume = (l × w²)× 0.5, where l = largest diameter and w = width (perpendicular diameter) of the tumor (in mm). Tumor growth was monitored based on the current applicable Directive 2010/63/EU and national law and mice were sacrificed if the tumor size reached 1000 mm3.

Immunofluorescence staining

Primary tumors were fixed in 4% paraformaldehyde (PFA) and embedded in paraffin. Tissue sections (7 μm-thick) on slides were heated overnight at 70° and were deparaffinized in Xylene Substitute (A5597; Sigma) for 30 min at RT and rehydrated in sequential incubation over ethanol scale (100%–90%–70%–50% and 30%) than washed with deionized water and phosphate-buffered saline PBS 1X. For antigen retrieval, slides were placed in preheated Tris–EDTA-based pH 9 buffer, enriched with NP40 0.5% and Tween 20 0.5% and incubated at 90 °C for 30 min, then let cool down at room temperature for 30 min in the same buffer, prior to immunofluorescence staining procedure. As to decrease the non-specific background fluorescence, tissue sections were washed with PBS 1X and then blocked with a blocking solution containing 5% normal goat serum blocking solution (S-1000-20; Vector Laboratories) in PBS for 1 h at room temperature. Primary antibodies staining was performed overnight at 4 °C: anti-rabbit CD3 (78588; Cell Signaling, 1:300), CD11c (97585; Cell Signaling, 1:300), F4-80 (70076; Cell Signaling, 1:300), CD19 (90176; Cell Signaling, 1:300). Tissue sections were washed in PBS 1X and were incubated with secondary antibody, anti-rabbit Alexa Fluor 647 (A-21245; Thermo Fisher, 1:500). Nuclei were stained with DAPI (62248; Thermo Fisher, 1:1000), washed with PBS 1X and mounted using Prolong Diamond mounting reagent (P36970; Thermo Fisher). Images were acquired using spinning disk confocal microscopy on a Nikon instrument (Ti; Nikon Europe) equipped with X-Light-V2 + VCS spinning disk head (CrestOptics) and EM-CCD Camera (iXon DU888; ANdor Technologies) with laser excitation (LDI Ltd), employing ×5, ×10, and ×20 objectives (Nikon Europe).

Ultrasound analyses of tumor growth in animal control group (saline)

A control mouse group treated with saline solution was included in the trial. However, in this group, we observed a very rapid outgrowth of orthotopic tumors compared to the groups treated with GMMA formulations, which did not allow a fair quantitative comparison by bioluminescence imaging analysis. Ultrasound tridimensional imaging was thus used. In Table S3 data are reported.

Animals and tumor model

Immunocompetent female BALB/c mice aged 6–8 weeks were used. The orthotopic breast tumor was generated by injecting 4T1-Luc mammary tumor cells into the fourth mammary gland. The cells were suspended in PBS and inoculated in a volume of 50 µL. Animals were monitored for tumor growth and overall well-being. All animal experiments were conducted in accordance with institutional guidelines and approved by the animal experimentation ethics committee.

Ultrasound system

Ultrasound analyses were performed using a high-frequency ultrasound system (Vevo 2100, VisualSonics) with a 40 MHz probe for high-resolution tumor imaging.

Animal preparation for ultrasound

Mice were anesthetized with isoflurane (2–3% oxygen) and positioned on a heated platform to maintain body temperature. Hair in the mammary region was removed using depilatory cream to improve image quality. Ultrasound gel was applied to prevent air bubbles between the probe and the skin.

Image acquisition

Ultrasound images were acquired in B-mode for tumor morphology assessment. The tumor area was measured using an integrated function for the quantification of ultrasound images in the Vevo 2100 software

Statistical analyses

Statistical analyses of serological data, flow cytometry, and bioluminescence were performed using GraphPad Prism (version 9.4.1). Statistical significance between groups was determined using the Mann–Whitney test to compare ranks or using two-way ANOVA Tukey’s multiple comparison test. P-values < 0.05 were considered significant. The correlation matrix was represented as a correlogram between tumor inhibition values, titers of immunoglobulin subsets, and serum cytokine levels and was performed using the Pearson correlation coefficient.