Media and reagents

RPMI 1640 (Lonza) supplemented with 10% fetal bovine serum (FBS), 100 μg/mL normocin (InvivoGen), 50 μg/mL penicillin‒streptomycin, 1% nonessential amino acids, 1% MEM vitamins and 1 mM sodium pyruvate was used. AIM V medium (Gibco) was employed for the in vitro expansion of Tregs. Ham’s F12 medium (Lonza) supplemented with 10% FBS was used for PC3 and B16 cell line culture. Inhibitors for IDO (1-methyl tryptophan, 1-MT), glycolysis (2-deoxy-D-glucose, 2-DG), NF-κB (BAY117082) (Sigma‒Aldrich), Syk (piceatannol) (Enzo Life Sciences), MEK 1/2 (U0126), p38 (SB202190), JNK (SP600125), mTOR (rapamycin) (InvivoGen), calcium chelating agent (EDTA), heparin (Laboratorio Reig Jofré) and the corresponding vehicles were used. Blocking antibodies for IL-10 (clone JES3-9D7, Biolegend) and PD-L1 (clone 29E.2A3, Biolegend) and corresponding isotype controls were used for the blocking experiments. We used lipopolysaccharide (LPS) from Escherichia coli O127:B8 (Sigma‒Aldrich).

Ehrlich tumor (ET), PC3 and B16-F10 cell lines

ET cells, initially originating from hyperdiploid Ehrlich-Lettré mouse ascites tumor cells, are derived from a cloned cell variant selected for its high reactivity with a monoclonal A10 antibody. PC3 cells were kindly provided by Dr. Eduardo Martínez-Naves, Department of Immunology, Ophthalmology and ENT, School of Medicine, Complutense University of Madrid, Madrid, Spain. B16-F10 cells were provided by the Department of Immunology, Hospital Clínico San Carlos, Instituto de Medicina del Laboratorio (IML), Complutense University of Madrid, Madrid, Spain.

Isolation of Ca10 from ET cells

Ca10 was obtained from the supernatants of ET cells growing in vitro in serum-free medium for 24 h. Briefly, the pooled supernatants collected by centrifugation were subjected to tangential ultrafiltration using 300 kDa membranes (Pall Corporation) under high ionic strength (1 M NaCl). The Ca10-enriched fraction obtained from the retentate was subsequently thawed with distilled water on 300 kDa membranes and lyophilized, having a relative composition of 85 ± 5% carbohydrate and 10 ± 5% protein with no significant amounts of nucleic acids or lipids, as observed by NMR. In some experiments, Ca10 was obtained from ET cells cultured for 72 h in the presence or absence (control) of GAG biosynthesis inhibitors: per-O-acetyl-4-fluoro-4-deoxy-GlcNAc (125 μM) (Biosynth-Carbosynth), 4-fluoro-4-deoxy-GlcNAc (1 mM) (Biosynth-Carbosynth), sodium chlorate (30 mM) (Sigma‒Aldrich), and xyloside (1 mM) (4-methylumbelliferyl-beta-D-xylopyranoside; Apollo Scientific).

Monoclonal A10 antibody

The murine IgM monoclonal A10 antibody (mAb A10) was provided by Inmunotek S.L. and purified from the culture supernatant of a cloned A10-producing hybridoma. A10 was obtained by immunization with devitalized ET cells and selected by their reactivity against carbohydrates on the surface of murine ET cells and their ability to inhibit tumor growth in vivo [22].

Animal studies

For all of the experiments, the animals were maintained in the Experimental Medicine and Surgery Department, Hospital Clínico San Carlos (Madrid, Spain) under the provisions of R.D. 53/2013 and D.C. 86/609/CEE; RD 1201/2005. C57BL/6J mice (7 weeks old, Charles River) were inoculated with 105 ET cells by subcutaneous injection into the left groin. Tumor size was monitored weekly by calculating the tumor area (mm2, width × length) and tumor volume (mm3, (4π/3) × (width/2)2 × (length/2)). The tumor length and width were measured with a Vernier caliper. Mice were sacrificed on different days (22, 32, 42, and 58 days) of tumor development. After sacrifice, taking all tumors into consideration, the average tumor volume was determined to be 246 mm3, with mice bearing small (below the average, <246 mm3) and large (above the average, >246 mm3) tumors. For some experiments, C57BL/6J tumor-free mice were alternately administered (i.v./i.p.) isolated Ca10 (40 μg) in the presence or absence of the mAb A10 (80, 160 or 320 μg) for 7 days, and mice were sacrificed the day after the last inoculation. In all experiments, spleen and blood samples, collected by cardiac puncture, were obtained to quantify splenic Treg numbers and circulating Ca10 levels, respectively, as described below.

Quantification of Ca10 levels

Ca10 levels were measured in ET cell culture supernatants, mouse and human sera and isolated Ca10 preparations subjected to different treatments by a tailor-made mAb A10-based sandwich ELISA. In this assay, purified mAb A10 (IgMκ) from culture supernatants of the mAb A10-producing hybridoma was used as the capture antibody, and horseradish peroxidase-labeled mAb A10 (HRP-A10) from the same source was used as the detection antibody. Plates were coated overnight at 4 °C with 5 μg/mL mAb A10 in 0.05 M carbonate-bicarbonate buffer (pH 9.4). Then, they were washed with PBS 0.25% Tween-20 (PBS-Tw) and incubated for 2 h at room temperature (RT) with the samples diluted in PBS-Tw. After a wash step, a dilution of HRP-mAb A10 (1:1000 dilution) in PBS-Tw and 5% FBS was added and incubated for 2 h at RT. After a final wash, the peroxidase substrate (OPD, 0.63 mg/mL) (Thermo-Fischer) was added to 0.1 M citrate buffer with 0.03% H202 - pH 5.5. The enzymatic reaction was allowed to develop for 30 min and stopped by adding a 10% HCl solution. The Ca10 concentration was expressed as arbitrary units (AU) per mL by extrapolating the OD values (495 nm) of a reference curve established with Ca10. The detection and quantitation limits were established as 0.03 and 0.05 AU/mL, respectively, with a linear range between 0.03 and 1.56 AU/mL (R2 > 0,98). Sensitivity and specificity (% of recovery = 99.85) assays were performed to validate its use for Ca10 analysis. In some experiments, the samples to be tested for Ca10 were previously subjected to enzymatic treatment: (i) Bacteroides heparinase II plus heparinase III (New England Biolabs) prepared in heparinase reaction buffer at final concentrations of 85 AU/mL and 14.8 AU/mL, respectively, for 24 h at 30 °C; (ii) Streptomyces griseus Pronase (Roche) prepared in 10 mM CaCl2 at a final concentration of 0.2 mg/mL for 24 h at 40 °C; and (iii) Sialidase 33 A (Nzytech) prepared in reaction buffer at a final concentration of 0.375 μg/mL. After enzymatic treatment, enzymes were inactivated at 100 °C for 1 min. To monitor oxidized Ca10 (Ca10 OX) reactivity, Ca10 was previously oxidized with sodium periodate (50 mM) (Sigma‒Aldrich) for 70 min at RT in the dark. To remove the reagent, distilled water was added, and the sample was centrifuged for 10 min at 14,000 × g using a Nanosep 3 K Omega (Pall Corporation).

Generation of hmoDCs, purification of naive CD4+ T cells and isolation of DCs

Peripheral blood mononuclear cells (PBMCs) from buffy coats of healthy donors were obtained by using Ficoll density gradient centrifugation (800 × g, 20 min). Monocytes were isolated from total PBMCs using anti-CD14 microbeads and cultured for 6 days with RPMI medium containing 100 ng/mL IL-4 and GM-CSF (PeproTech) to generate immature human monocyte-derived dendritic cells (hmoDCs). To generate Ca10/hmoDCs, Ca10 (20 μg/mL) was added on Days 0 and 4 of differentiation. Peripheral blood naive CD4+ T cells and DCs were isolated using the “Naive CD4+ T-Cell Isolation Kit” and “Blood Dendritic Cell Isolation Kit II” (Miltenyi Biotec), respectively, according to the manufacturer’s protocol. In all cases, the purity and phenotype of cells were analyzed by flow cytometry with lineage-specific markers, and cell viability was determined by trypan blue exclusion with a light microscope.

Human cell cultures

Immature hmoDCs or human total blood DCs(106 cells per mL) were treated with Ca10 (20 µg/mL) for 18 h. Subsequently, the cells were collected and centrifuged. Flow cytometry and qPCR were used to analyzed the cell phenotypes, and ELISA was used to quantify the levels of TNF-α, IL-6, IL-1β and IL-10 in the cell-free supernatants. For inhibition experiments, hmoDCs were preincubated for 30 min with EDTA (0.5 mM), piceatannol (25 μM), U0126 (1 μM), SB202190 (1 μM), SP600125 (10 μM), BAY117082 (2.5 μM), 1-MT (1 mM), rapamycin (100 nM), or heparin (100 IU/ml) and for 1 h with 2-DG (10 mM) or anti-IL-10 (2.5 μg/ml) or the corresponding vehicle control or isotype control prior to treatment. Then, hmoDCs were treated with Ca10 for 18 h in the presence of the corresponding inhibitors. To analyze the contribution of the carbohydrate structure, hmoDCs were stimulated with 20 μg/mL Ca10 (without treatment), oxidized Ca10 (Ca10 OX), Ca10 treated with heparinase (HPSE treat.) or Ca10 treated with pronase (Ca10+Pronase) for 18 h. For blocking experiments with the mAb A10, Ca10 was previously incubated for 1 h with mAb A10 or the corresponding isotype control at a ratio of 1:10 or 1:5 (Ca10:A10) with continuous stirring, and then, it was added to hmoDCs for 18 h. Ca10/hmoDCs or conventional hmoDCs (106 cells per mL) were treated with medium (negative control) or LPS (100 ng/mL) for 18 h. Flow cytometry and qPCR were used to analyzed the cell phenotypes, and ELISA was used to quantify the levels of TNF-α, IL-6, IL-1β and IL-10 in the cell-free supernatants. In all cases, cell viability was analyzed by trypan blue exclusion under microscopy.

Migration (wound healing) and proliferation (MTT) assays

PC3 and B16-F10 cells were detached from the tissue culture plate using TripLETM Express (Gibco). Cells were centrifuged and resuspended in culture media. Cells were plated in a 48-well plate for 100% confluence in 24 h. A wound approximately 100 µm wide was generated using a pipette tip to measure wound healing (cell migration). Media and cell debris were removed, and culture media were added with or without Ca10 (20 µg/mL) for 48 h. The width of the scratch was measured at different time points and represented relative wound closure. Migration was recorded at 6 time points (0, 4, 8, 12, 24 and 48 h) using optical microscopy.

PC3 and B16-F10 cells were seeded at 5 × 104 cells per well in a 96-well plate for 24 h. Then, media and cell debris were removed, and culture media were added with or without Ca10 (20 µg/mL) for 48 h. Cell proliferation at 24 and 48 h was analyzed following the Cell Proliferation Kit MTT standard protocol (Roche).

Flow cytometry

The following anti-human mAbs were used for flow cytometry: fluorescein isothiocyanate (FITC)-conjugated anti-HLA-DR; phycoerythrin (PE)-conjugated anti-CD86; allophycocyanin (APC)-conjugated anti-HLA-DR; FITC-conjugated anti-CD1c; and PE-conjugated anti-CD303 (Miltenyi Biotec); APC-conjugated anti-CD83; FITC-conjugated anti-PD-L1; phycoerythrin-Cy7 (PE/Cy7)-conjugated anti-CD19; peridinin-chlorophyll-protein (PerCP)-conjugated anti-CD4; Alexa Fluor 488-conjugated anti-FOXP3; PE-conjugated anti-CD127; and APC-conjugated anti-CD25 (BioLegend); and PE-conjugated anti-ICOSL (BD Pharmingen). Cells were washed with PBS 2 mM EDTA and 0.5% BSA and stained for 15 min at RT in the dark. For analysis of FOXP3 expression in human T cells cocultured with DCs, the cells were first subjected to surface staining with anti-human PE-CD127, PerCP-CD4, and APC-CD25 antibodies. After fixation and permeabilization, the cells were stained with Alexa Fluor 488-FOXP3 according to the manufacturer’s recommendations. The same protocol was carried out for the characterization of CD4+CD25highFOXP3+ Tregs in mouse splenocytes using the following anti-mouse mAbs: PerCP-conjugated anti-CD4, PE-conjugated anti-CD25, and Alexa Fluor 488-conjugated anti-FOXP3 (BioLegend). The viability dye eFluor 660 (eBioscience) was used to assess viability by flow cytometry, and dead cells were excluded from the analysis. For each staining, the corresponding isotype controls (IgG2A-FITC, IgG1-Alexa Fluor 488, IgG1-PE, IgG2A-PerCP, or IgG1-APC) were also assayed. The following antibodies were used for the flow cytometry analysis of Ehrlich tumor cells: A10 and anti-mouse IgM (μ-chain specific)-FITC produced in goat (Sigma‒Aldrich). Flow cytometry analysis was performed using a FACSCalibur in the Cytometry and Fluorescence Microscopy Unit at Complutense University of Madrid.

Cytokine quantification

The levels of TNF-α, IL-6, IL-1β, IL-10, IFN-γ and IL-5 in cell-free supernatants were quantified by sandwich ELISA kits (BD Biosciences) following the manufacturer’s instructions.

RNA isolation, cDNA synthesis, and quantitative real-time RT‒PCR

RNA was isolated from harvested cells using an RNeasy mini kit (Qiagen), and cDNA was generated using a PrimeScript RT reagent Kit (Takara) according to the manufacturers’ instructions. Real-time quantitative PCR was performed on cDNA using FastStart Universal SYBR Green Master (Rox) (Roche). The sequences of the primer pairs used were as follows: IL6 (forward, GGTACATCCTCGACGGCATCT; reverse GTGCCTCTTTGCTGCTTTCAC), IL1B (forward, TTTTTGCTGTGAGTCCCGGAG; reverse TTCGACACATGGGATAACGAGG), IL10 (forward, GTGATGCCCCAAGCTGAGA; reverse CACGGCCTTGCTCTTGTTTT), PDL1 (forward, AAGATGAGGATATTTGCTGTCTTTATATTC; reverse, GTCCTTGGGAACCGTGACAGT), indoleamine 2,3-dioxygenase (IDO) (forward, AGAAGTGGGCTTTGCTCTGC; reverse, TGGCAAGACCTTACGGACATCTC), suppressor of cytokine signaling 1 (SOCS1) (forward, CCCTGGTTGTTGTAGCAGCTT; reverse, CAACCCCTGGTTTGTGCAA), SOCS3 (forward, CCTCAGCATCTCTGTCGGAAGA; reverse, GCATCGTACTGGTCCAGGAACT), retinaldehyde dehydrogenase 1 (RALDH1) (forward, CTGCCGGGAAAAGCAATCT; reverse, AAATTCAACAGCATTGTCCAAGTC), RALDH2 (forward, AGGGCAGTTCTTGCAACCAT; reverse, GCGTAATATCGAAAGGT). Samples were run on a real-time PCR system (ABI Prism 7900 HT; Applied Biosystems). Data were normalized to EF1A and displayed as arbitrary units calculated as 2−ΔCT values multiplied by 104. ΔCT was defined as the difference between the cycle threshold value for the gene of interest and EF1A.

Western blot analysis

HmoDCs (106 cells per mL) were treated with Ca10 for 30 min at 37 °C. Then, cells were harvested and lysed with RIPA buffer (Thermo Fisher Scientific) in the presence of protease/phosphatase inhibitor cocktail (Cell Signaling) for 30 min at 4 °C with vortexing every 10 min. Lysates were clarified by centrifugation at 10,000 × g for 15 min at 4 °C. Protein quantification was performed with a Micro BCA Protein Assay Kit (Pierce), and samples with equal amounts of total protein were resolved by 10% SDS-polyacrylamide gel electrophoresis (SDS–PAGE). Proteins were then transferred to nitrocellulose membranes (Bio-Rad). The membrane was incubated with the following primary antibodies: phospho-ERK1/2 (Thr202/Tyr204), ERK1/2, phospho-SAPK/JNK (Thr183/Tyr185), SAPK/JNK, phospho-p38 MAPK (Thr180/Tyr182), p38 MAPK, phospho-IκBα (Ser32/35) or IκBα (1:1000, Cell Signaling), and β-actin (1:15000, Sigma‒Aldrich) and goat anti-rabbit (1:4000, Bio-Rad) or goat anti-mouse (1:2500, Pierce) conjugated with horseradish peroxidase as a secondary antibody. The signals were visualized with Clarity Western ECL Substrate (Bio-Rad) and detected in a Fujifilm LAS-3000 developer.

Metabolic studies

HmoDCs were stimulated with Ca10 for 18 h. For real-time metabolic characterization, the extracellular acidification rate (ECAR, mpH/min), mitochondrial oxygen consumption rate (OCR, pmol/min), and glycolytic proton efflux rate (glycoPER, pmol/min) were analyzed using a Seahorse XF HS Mini Analyzer (Agilent). HmoDCs after Ca10 stimulation and fresh Ca10/hmoDCs were harvested, washed and resuspended in DMEM supplemented with 10 mM glucose, 1 mM pyruvate and 2 mM glutamine. Cells (25 × 103/well) were plated in poly-D-lysine-coated 8-well miniplates and incubated in a non-CO2 incubator for 1 h at 37 °C. ECAR and OCR were analyzed using a glycolysis rate assay with final concentrations of 1 μM rotenone, 1 μM antimycin A, and 50 mM 2-deoxyglucose (2-DG). A complete glycoPER study was performed in two consecutive stages: basal glycolysis (without drugs) and electron transportation chain inhibition (rotenone and antimycin A). The OCR was analyzed using a Cell Mito Stress Test with final concentrations of 1.5 μM oligomycin, 1 μM carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP), 0.5 μM rotenone and 0.5 μM antimycin A. A complete OCR study was performed in four consecutive stages: basal respiration (without drugs), mitochondrial complex V inhibition (oligomycin), maximal respiration induction (FCCP), and electron transportation chain inhibition (rotenone and antimycin A). The glucose concentration in cell-free supernatants was determined by using the Glucose (GO) Assay Kit (Sigma‒Aldrich). The metabolic rate was calculated as the percentage of the medium without hmoDCs (2 mg/mL). The Warburg effect was determined by quantifying the optical density (OD) at 570 nm and calculated as 1/OD570. Lactate production in cell-free supernatants was determined by using a colorimetric L-lactate assay kit (Sigma‒Aldrich).

Coculture experiments and Treg suppression assay

Immature hmoDCs or total DCs were stimulated with Ca10 for 18 h as well as Ca10/hmoDCs and conventionally treated with medium for 18 h. Next, they were washed and cocultured with purified allogeneic naive CD4+ T cells (1:5 DC/T cells) for 5 days. The levels of IFN-γ, IL-5 and IL-10 were quantified in cell-free supernatants by ELISA, and the expression of FOXP3 was analyzed by flow cytometry. Inhibition assays were performed by adding anti-human IL-10 (2.5 μg/mL) or anti-human PD-L1 (10 μg/mL) to coculture. The induced CD4+CD127-CD25highFOXP3+ Tregs by allogeneic Ca10-treated hmoDCs were purified by sorting CD4+CD127-CD25high cells and mixed with carboxyfluorescein succinimidyl ester (CFSE)-labeled autologous PBMCs (responder cells) at different ratios and stimulated with plate-bound anti-human CD3 antibody (1 μg/mL, clone OKT3; eBioscience) and soluble anti-human CD28 (1 μg/mL, clone CD28.6; eBioscience) for 5 days. For control purposes, CFSE-labeled PBMCs were cultured alone with or without stimulation, and CD4+CD127+CD25- cells (non-Tregs) were also mixed with CFSE-labeled PBMCs. The proliferation of CD4+ T cells was determined by CFSE dilution and flow cytometry.

Differentiation and expansion of human Tregs

For the in vitro generation of Tregs, peripheral blood naive CD4+ T cells were isolated using the “Naïve CD4+ T-Cell Isolation Kit” (Miltenyi Biotec). Then, the cells (106 cells per mL) were stimulated in 48-well plates in AIM V medium at 37 °C with 100 U/mL IL-2, 10 ng/mL TGF-β (PeproTech), 25 µL/mL ImmunoCult™ Human CD3/CD28/CD2 (CDmix) T-Cell Activator (STEMCELL Technologies), 5 µg/mL anti-IL-12/IL-23, 5 µg/mL anti-IL-4 and 5 µg/mL anti-IFN-γ (Biolegend) in the presence or absence of Ca10 (20 µg/mL). On Day 2, cell cultures were supplemented with 100 U/ml IL-2 in cRPMI medium. On Day 4, the cells were split, and the cultures were supplemented with 100 U/ml IL-2 in cRPMI medium. On Day 5, the cells were collected and centrifuged. For expansion experiments, the obtained Tregs (106 cells per mL) in the absence of Ca10 were restimulated in 48-well plates in cRPMI medium at 37 °C with 100 U/mL IL-2 and 25 µL/mL ImmunoCult™ Human CD3/CD28/CD2 (CDmix) T-Cell Activator in the presence or absence of Ca10 (20 µg/mL). On Day 1, cell cultures were supplemented with 100 U/ml IL-2 in cRPMI medium. On Day 2, the cells were split, and the cultures were supplemented with 100 U/ml IL-2 in cRPMI medium every one or two days. On Day 7, the cells were collected and centrifuged. For suppression assays, generated CD4+CD25+CD127-FOXP3+ Tregs were mixed with CFSE-labeled autologous PBMCs (responder cells) at different ratios and stimulated with 1 μg/mL anti-CD3, 1 μg/mL anti-CD28 mAb and 100 U/mL IL-2 for 5 days as described above.

NMR experiments

All NMR spectra were acquired using a Bruker AVANCE 600 MHz spectrometer equipped with a triple resonance TXI cryogenic probe and processed with TOPSIN 3.0 software (Bruker SA). NMR samples were prepared in deuterium oxide (D2O). One-dimensional proton (1D–1H) spectra, two-dimensional (2D) diffusion-ordered spectroscopy (DOSY), 2D heteronuclear 1H–13C single quantum coherence (1H–13C HSQC) and 2D homonuclear 1H–1H total correlation spectroscopy (TOCSY) spectra were acquired using standard pulse sequences included in TOPSPIN software to characterize the structure (TOCSY and HSQC) and hydrodynamic behavior (DOSY) of Ca10 samples.

To monitor periodate oxidation, Ca10 was dissolved in 50 mM sodium periodate (NaIO4) in D2O at a final concentration of 5 mg/mL in an NMR tube and introduced into the spectrometer probe adjusted at 298 K. One-dimensional 1D–1H and 2D DOSY spectra were acquired at sequential times to follow the oxidation reaction over 8 h. DOSY spectra were obtained using the standard Bruker pulse sequence (ledbpgp2s), acquiring 16 gradient points, with 128 scans each, between 2 and 95% gradient intensity using a diffusion time delay of 0.25 s and 2500 μs to achieve a wide pulse gradient.

To monitor heparinase digestion, 5 μL of Bacteroides heparinases II and III (400 UA/mL and 700 UA/mL, respectively) were added to the NMR tube containing 5 mg/mL Ca10 sample in heparinase buffer, and the tube was introduced in the spectrometer probe adjusted at 303 K. To evaluate the enzymatic reaction, 1D-1H and 2D DOSY spectra were acquired at sequential times for 24 h. The 1D-1H spectra of 128 scans of 32 K size were recorded by applying the TOPSIN zgesgp pulse program, which includes a gradient sculpting water suppression procedure. The 2D DOSY experiments were performed using the pulse sequence ledbpgp2s acquiring 32 gradient points, with 128 scans each between 2 and 95% gradient intensity. The diffusion time delay and gradient duration were 600 ms and 2500 μs, respectively, before the start of the reaction and 170 ms and 1700 μs at the end point of the reaction.

To monitor pronase digestion, Ca10 dissolved in deuterated buffer at 5 mg/mL was treated with pronase as indicated above. DOSY spectra were acquired at 298 K before and after pronase treatment with the same parameters used for the periodate oxidation sample.

Enzyme digestions

A collection of glycosidases was obtained: N-glycosidase (PNGase F), O-glycanase (endo-galactosaminidase), endo-β-acetylglucosaminidase, exoglycosidases (α- and β- galactosidases), α-mannosidase, glucosidases, glucosaminidase, α-fucosidase, sialidase (NZYTECH), chitosanase 8B, heparinases (New England Biolabs) and chondroitinase ABC. Working conditions with each enzyme were determined using model glycoproteins such as α1-acid glycoprotein, fetuin, asialofetuin, ribonuclease B and commercial oligosaccharides. Enzymatic digestions were analyzed by NMR, polyacrylamide gel electrophoresis (SDS‒PAGE) and Ca10-epitope sandwich ELISA evaluation.

Heparinase products and commercial heparan sulfate disaccharides

The sample of heparinase-digested Ca10 was filtered using Vivaspin™ 50 K Centrifugal Concentrators (Sartorius), and the filtrate was lyophilized and resuspended in D2O for NMR analysis. 1D–1H, 2D TOCSY, 1H–13C HSQC and DOSY spectra were acquired. For comparison, a collection of heparan sulfate disaccharide standards were also analyzed by NMR: ∆UA,2S – GlcNAc,6S (I-A) ∆UA – GlcNS,6S (II-S), ∆UA – GlcNS (IV-S), ∆UA – GlcNAc (IV-A), ∆UA,2S – GlcNAc (III-A) (Iduron, UK) and ∆UA – GlcN (IV-H) (Santa Cruz Biotechnology). Samples of each of them at a 1 mM concentration in D2O were prepared, and 1D–1H, 2D 1H–13C HSQC and DOSY spectra were acquired for comparison.

Human serum samples

Sera from patients diagnosed with prostate adenocarcinoma (n = 248), colorectal adenocarcinoma (n = 66) and other types of cancer (n = 71) at any stage of the disease were obtained from the Central Laboratory of Hospital Clínico San Carlos (Madrid, Spain) during routine follow-up. The levels of Ca10H in the serum samples of prostate cancer patients with (n = 42) or without (n = 30) bone metastases from the Urology Department were measured, as described above for Ca10, and alkaline phosphatase levels were measured by spectrophotometry (AU5800 Clinical Chemistry Analyzer, Beckman Coulter). The serum samples from the healthy donors (n = 131) used as controls were obtained from the Blood Donor Unit of the Hospital Clínico San Carlos.

Ethics approval

All patients and controls signed informed consent documents, and the data were treated according to recommended criteria of confidentiality, following the ethical guidelines of the 1975 Declaration of Helsinki. The study was approved by the local Ethics Committee (Hospital Clínico San Carlos, Madrid, CEIM 13/098).

Statistics

All the data are expressed as the mean ± SEM of the indicated parameters. Spearman’s correlation, paired or unpaired Student’s t test, the Wilcoxon test and the Mann‒Whitney U test were used for statistical analysis performed by GraphPad Prism software, version 8.0. Significance was defined as *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001.