Obtaining, purifying and quantifying GVs

Initially, the GVs were collected from ANA cultures, obtained from Culture Collection of Algae and Protozoa, Scottish Marine Institute, cat. no. Culture Collection of Algae and Protozoa 1403/13F, following the Lakshmanan et al. (2017) protocol [ ]. Briefly, ANA is a well-established organism used for production of purified GVs [ , ]. ANA cells were cultured in BG-11 medium (Sigma Aldrich, St. Louis, MO), under a 12 h light–dark cycle, 100 μM photons (m-2 s-1), for 2 weeks.

The next step involved transferring the culture to a separating funnel, where cells were let to sit for 1–2 days, which allowed buoyant cells to rise to the top of the liquid. Cells were collected from the superficial layer and treated with a lysis solution (0.5% Tween, 500 mmol/L sorbitol, 10 mmol/L Tris pH 7.5) in a 1:1 ratio for 2 h at 4°C to release the GVs. Finally, GVs were retrieved by centrifugation of the lysed sample (200 g, 24 h, 4°C). The subnatant was removed and the remaining volume was suspended in phosphate-buffered saline (PBS). This process was repeated three times, and the final sample was stored at 4°C. GVs concentration was determined by measuring the suspension optical density (OD), at 500 nm (OD500), using a Synergy H4 hybrid microplate reader (Agilent Biotek, Santa Clara, CA).

GvpC engineering and purification of recombinant GvpC

The modification of the ANA GvpC protein-coding sequence was performed using the plasmid WT C-His ANA GvpC pET28a (Addgene, Watertown, MA), developed to express ANA GvpC in E. coli [ , ]. To obtain a GvpC able to interact with the CTSB protease, a five-codon sequence corresponding to the amino acid sequence Gly–Leu–Phe–Gly–Lys (cathepsin cleavage site) [ ] was added in the gvpC gene at a location corresponding with the GvpC second repeat [ , , ]. The synthetic version of the gene including this modification was constructed by the company GenOne Biotech (Rio do Janeiro, RJ, Brazil).

To obtain the recombinant proteins, the same procedure was performed for the native ANA GvpC and the engineered ANA GvpC, each produced from their respective vectors. The plasmids were transformed into competent E. coli BL21(DE3) cells (Invitrogen, Waltham, MA) by electroporation. Clones were selected from colonies grown on Luria Broth agar plates with 50 μg mL ⁻¹ of kanamycin, after incubation at 37°C for 18 h. Thereafter, cultures started inoculating one colony in 2 mL of Luria Broth medium (50 μg·mL ⁻¹ of kanamycin) and incubating overnight at 200 rpm at 30°C. To obtain a culture in the exponential growth phase, the overnight culture was diluted 1:75 in 50 mL of Luria Broth (50 μ·gmL ⁻¹ kanamycin) and grown during 3 h at 200 rpm at 30°C or until the OD600 reached 0.4–0.6. Protein expression was induced by adding 400 μM of isopropyl-ß-D-thiogalactopyranoside to the culture and incubating for 12 h at 200 rpm at 30°C. Cells were collected by centrifugation (15,000 g × 10 min, 4°C). The cell pellet was resuspended in 4 mL solulyse with 4 μL DNase I (10 μg·mL ⁻¹ ), kept on ice for 10 min and centrifuged (15,000 g , 15 min, 4°C). The cell pellet was resuspended in 4 mL solulyse with 40 μL lysozyme (0.25 mg·mL ⁻¹ ), kept on ice for 10 min and centrifuged as before. The final pellet was resuspended in a solubilization buffer (20 mmol/L Tris-HCl, 500 mmol/L NaCl, 6 mol/L urea, pH 8.0) and centrifuged (20,000 g , 20 min, 4°C). The supernatant was added to Ni-NTA resin (QIAGEN, Germantown, MD) and incubated for 1 h, 30 rpm, 4°C. The resin was placed in a column and washed with solubilization buffer with 20 mmol/L imidazole. The recombinant protein was eluted by adding solubilization buffer with 250 mmol/L imidazole. The purification of GvpC was confirmed by SDS-polyacrylamide gel electrophoresis (PAGE) analysis.

Protein concentrations were measured using the Synergy H4 hybrid microplate reader (Agilent Biotek), corresponding with 0.5 and 0.6 mg·mL ⁻¹ of ANA GvpC and engineered ANA GvpC, respectively.

GV modification to incorporate the engineered GvpC

ANA GV (GVwt) was obtained as described previously. To remove the original GvpC ( Fig. 1 c), a GVwt preparation (OD500 = 10) was treated with 6 mol/L urea in 100 mmol/L Tris-HCl (pH 8.0) [ , ]. This included three rounds, each one consisting of an agitation step (50 rpm) during 1 h at 4°C, followed by centrifugation (200 g , 24 h, 4°C). The supernatant was removed at the end of the first two rounds, and a 6 mol/L urea solution was added to the remaining GV fraction. Thereafter, PBS was added to the GV fraction. The resulting GV preparation, from which GvpC was removed, was named ΔGV and was stored at 4°C.

The obtained GV*, the ΔGV preparation was incubated with purified recombinant GvpC: either ANA GvpC (wt) or the engineered ANA GvpC. The re-addition process consisted of three sequential dialysis steps, against 6 mol/L urea, 3 mol/L urea, and PBS, each for 24 h at 4°C, to gradually refold the protein onto the surface of the stripped GVs. The proteins were maintained in a dialysis tube with a 6–8 kDa molecular weight cutoff. The amount of GVs and GvpC used was based on the work of Lakshmanan et al. (2017) [ ]. Briefly, the amount of engineered GvpC (in nanomoles) to be added to stripped GVs was calculated as 2 x OD (GV) × 198 nm × volume of GVs (in litres) according to a 1:25 binding ratio of GvpC:GvpA. In this case, the GvpC protein was used at a final concentration of 0.45 mg·mL ⁻¹ in a volume of 4 mL of ΔGV at an OD500 of 4. After the final dialysis step, samples were centrifuged (200 g , 4 h, 4°C) to reduce the concentration of unbounded GvpC, and the GVs were then stored at 4°C. The preparations of ΔGV that received either the ANA GvpCwt or the engineered ANA GvpC, are herein referred to as GVwt* and GV*, respectively. The ΔGV, GVwt* and GV* concentrations were determined by OD500 using a Synergy H4 (Agilent Biotek).

CTSB reaction in solution

To analyse the GV* interaction with CTSB, an in vitro reaction was performed using human CTSB (SRP0289; Sigma Aldrich). Initially, 10 μL of CTSB (0.1 μg·mL ⁻¹ ) was diluted in 490 μL PBS (pH 6.0) and incubated at 37°C, 200 rpm for 15 min, to stabilize the enzyme. The concentration of CTSB was based on prior studies and had the intention of reproducing the concentration found in tumours [ , , , , , , ]. An aliquot (500 μL) of the GV* preparation (OD500 = 2) was added and mixed. An initial sample (100 μL) was immediately removed (instant T0) and stored at 4°C. After 24 h (T24) of incubation at 37°C, 200 rpm, another sample was obtained and stored at 4°C. The duration of the reaction (24 h) was set to enhance the interaction between GVs and CTSB. The same procedure was carried out for GVwt*, to act as a negative control for CTSB cleavage. Three replicates ( n = 3) were included in each CTSB reaction with the GV samples: GVwt* and GV*.

GV* incubation with the extracellular fraction of a culture of colon tumour cells

A preparation of GV* was incubated with the extracellular fraction retrieved from a colon tumour cell culture known for producing CTSB. Human HCT 116 cell line (91091005; Sigma Aldrich) was cultured in DMEM high glucose medium (12800017, Thermo Fisher Scientific, Waltham, MA), enriched with 10% foetal bovine serum (F2442 Sigma Aldrich) and 1% penicillin-streptomycin (15070063, Thermo Fisher Scientific). The cells were maintained in culture for 48 h until they achieved 90% confluence in a T-75 flask [ ]. At this point, the culture medium was pipetted and used as the tumour cell line extracellular fraction (TCEF).

TCEF (500 μL) was mixed with GV* (500 μL) and incubated for 24 h at 37°C, constituting of (TCEF + GV*). A similar procedure was followed for GVwt (resulting in TCEF + GVwt). In each case, six replicates were included for the reactions with TCEF ( n = 6).