Affinity purification of native ThC
The sponge specimen used here was collected in Chuuk State of Federated States of Micronesia in 2009 under permission of Department of Marine Resources, Chuuk State FSM, and was extracted as described previously12. A sponge aqueous extract was treated with acidic buffer (pH 3.0) to obtain a ThC-enriched extract. The extract was applied to a 1-mL Sepharose-fucose affinity gel (EY Laboratories, Inc.). The column was eluted first with 50 mM Tris-HCl buffer and then with fucose. The fucose eluent was dialyzed to yield purified protein.
Cell culture and proliferation assay
ThC cell proliferation assays were performed as described previously12. Briefly, the murine interleukin-3-dependent pro-B-cell line Ba/F3 expressing human MPL (Ba/F3-HuMpl cells)27 was precultured for 4 days and then harvested by centrifugation at 1,000 rpm for 3 min. After washing with PBS (-), the collected cells were resuspended in RPMI-1640 medium containing 10% FBS at a cell density of 6.0 x 104 cells/mL. A 90-µL aliquot of the cell resuspension was transferred to a 96-well plate. In the presence of various concentrations of ThC or recombinant ThCs (10 µL), the cells were cultivated at 37 °C under air with 5% CO2. PBS and TPO (PeproTech) were used as negative and positive controls, respectively. After 4 days of cultivation, cell proliferation was measured with a cell counting kit (Dojindo). A 10-µL aliquot of the cell counting kit was added to each well. After incubation for 2 h, the absorption at 450 nm (Abs450) was recorded with a microplate reader. For the 6-alkynyl-fucose assay, Ba/F3-HuMpl cells were pretreated with 6-Alk-Fuc for 6 h, ThC or TPO was added, and the cells were cultured for 4 days.
Preparation of samples for the cell proliferation assay
ConA (Sigma), hypnin (from Hypnea japonica), PA-IIL (Fujifilm-Wako), and BC2LC-CTD (recombinant) were dissolved in PBS (-). 6-Alkynyl fucose (Peptide Institute, Inc.) was dissolved at 100 mM in DMSO and diluted with PBS (-) to each concentration. Sugar solutions were prepared with PBS (-) except N-acetylneuraminic acid (NANA). NANA was suspended in PBS (-) and neutralized with aqueous NaOH to pH 7.0. All the reagents were filter-sterilized with a 0.2-mm filter prior to use.
Determination of the amino acid sequence
Draft amino acid Edman degradation was performed using ThC purified by SDS–PAGE. The gel was electroblotted on a PVDF membrane, and the band for ThC was cut out for the N-terminal amino acid sequence using an automated sequencer (Procise 492HT). The internal amino acid sequence was obtained by digesting purified ThC with either (Fujifilm-Wako), chymotrypsin (Fujifilm-Wako) or V8 protease (Fujifilm-Wako). Each digest was separated by HPLC using a reversed-phase column (VYDAC protein&peptide C18) with a gradient (0-50%) of 0.1% aqueous TFA and acetonitrile. Each of the peptide fragments was subjected to de novo sequence analysis using MALDI-TOF MS/MS and to Edman degradation. The deduced amino acid sequences were mapped to give a draft amino acid sequence of ThC. The entire amino acid sequence was then confirmed by mass spectrometry as follows. A drop (4.5 µL) of crystallization supernatant from the X-ray analysis experiment containing 7.0 µg of native ThC was mixed with 80 µL of acetone and then centrifuged at 19,000 x g and 4 °C for 15 min. The precipitate was resuspended in 80 µL of acetone. After centrifugation at 19,000 x g and 4 °C for 15 min, the precipitate was further washed as described above and then air-dried. The dried sample was resuspended in 40 µL of 1× phase transfer surfactant (PTS)28. A 20-µL aliquot of resuspended sample was incubated with 2 µL of 200 mM Bond-Breaker TCEP solution (Thermo Fisher Scientific) to cleave the disulfide bond for 30 min at 50 °C. The reduced thiol was alkylated by 2 µL of 375 mM 2-iodoacetamide for 30 min at room temperature in the dark. After alkylation, an excess amount of 2-iodoacetamide was reacted with 2 µL of 400 mM L-Cys for 10 min at room temperature in the dark. The alkylated sample was digested by either 200 ng of trypsin (Promega) with 200 ng of Lys-C (Fujifilm-Wako) at 37 °C overnight or 200 ng of chymotrypsin (Promega) with 10 mM CaCl2 at 25 °C overnight. A total of 30 µL of the digestion sample was precipitated by the addition of 45 µL of 1.7% trifluoroacetic acid (TFA). After centrifugation at 19,000 x g and 4 °C for 15 min, the supernatant was purified using a Stage-Tip as described previously29. Peptide fragments were eluted from the Stage-Tip using 70% acetonitrile and 0.1% TFA, and the elution was freeze-dried. Recombinant ThC (rThC Q25, 2.2 µg) dissolved in 20 µL of PTS was also digested according to the protocol described above. The peptide fragments were dissolved with 10 µL of 0.1% TFA, and the peptides of the native ThC and rThC Q25 were analyzed with a quadrupole Orbitrap benchtop mass spectrometer (Q-Exactive, Thermo Fisher Scientific) equipped with a Nanospace SI-2 HPLC system (Osaka Soda Co., Ltd). The column temperature was maintained at 45 °C. The flow rate of the mobile phase was 200 μL/min; mobile phase A consisted of 0.05% formic acid (FA), and mobile phase B consisted of 0.05% FA/90% acetonitrile. The mobile phase gradient was programmed as follows: 0% B (0-2 min), 0-35% B (2-12 min), 35-55% B (12-15 min), 55-80% B (15-16 min), 80% B (16-18 min), 80-0% B (18-18.5 min), and 0% B (18.5-20 min). MS1 spectra were collected in the scan range of 350−1,200 m/z at 70,000 resolution and 200 m/z to hit an AGC target of 1×106 with an injection time of 200 ms. The AGC target value for fragment spectra was set to 1×105, and the intensity threshold was maintained at 3.3×104. The isolation width was set to 2.4 m/z, and the 12 most intense ions were fragmented in a data-dependent mode by collision-induced dissociation with a normalized collision energy of 27.
The amino acid sequences of the draft sequence and rThC Q25 were added to the UniProt sequence database (release 31st July 2019, entry 557,016, all species, reviewed). The MS file was searched against the database using Proteome Discoverer 1.4 software (Thermo Fisher Scientific) and PEAKS Studio Version X (Bioinformatics Solutions). The setting parameters were as follows: enzyme, trypsin (semi) or chymotrypsin (semi); maximum missed cleavage sites, 2 (Proteome Discoverer) or 4 (PEAKS); precursor mass tolerance, 6 ppm; fragment mass tolerance, 0.02 Da; fixed modification, cysteine carbamidomethylation. The peptide identification was filtered to a false discovery rate of less than 1%.
Overexpression and purification of recombinant ThC (rThC), rThC mutants, and BC2LC-CTD
An expression vector of recombinant ThC (rThC) and BC2LC-CTD was constructed by cloning a synthesized DNA fragment corresponding to the amino acid sequence of ThC determined by mass spectrometry and the C-terminal domain of BC2LC into the NdeI/XhoI site of a modified pET26 and pET28 vector, respectively. The 6× His tag was attached to the N-terminus of ThC. The expression vector of rThC mutants was constructed by inverse PCR using the expression vector of rThC as a template.
Escherichia coli strain BL21(DE3) harboring the expression vector of the desired protein was cultivated in LB medium at 37 °C with shaking at 120 rpm. When the OD600 reached 0.6, isopropyl-b-D-thiogalactopyranoside (IPTG) was added to the medium at a final concentration of 0.2 mM to induce the expression of rThC, and the mixture was then incubated at 25 °C overnight.
Cells were harvested by centrifugation at 4,000 x g for 30 min. The collected cells were suspended in buffer A composed of 20 mM HEPES-NaOH (pH 8.0) and 200 mM NaCl and then disrupted with a UD-211 ultrasonic disruptor (TOMY SEIKO). After centrifugation at 40,000 x g for 30 min, the supernatant was loaded onto a 1-mL column of Ni Sepharose (GE Healthcare). After washing with sonication buffer, the bound protein was eluted using a concentration gradient of imidazole in the sonication buffer. Fractions containing purified rThC were further purified by size-exclusion chromatography using HiLoad 26/600 Superdex 75 pg (GE Healthcare) preequilibrated with buffer A. Fractions containing rThC were collected and used for further experiments. SeMet-substituted rThC was expressed and purified by the same method as rThC with the exception that SeMet-substituted M9 medium was used instead of LB medium.
Isothermal titration calorimetry (ITC)
ITC measurement was performed with an iTC200 (GE Healthcare). The cell was filled with approximately 100 µM rThC, 100 µM BC2LC-CTD, or 35 µM PA-IIL, and the syringe was filled with 1.5 µM fucose or mannose. The protein was injected 18 times in a portion of 2 µL over 120 s. The data were analyzed with the program ORIGIN.
Carbohydrate binding assay
The carbohydrate binding specificity of rThC was analyzed with a carbohydrate Gel Kit#1 (EY Laboratories). The binding to fucose, mannose, lactose, N-acetylglucosamine, and N-acetylgalactosamine was evaluated with resin in which each carbohydrate was immobilized. A 0.5-mL aliquot of 0.1 mg/mL purified rThC was loaded on 0.1 mL of resin immobilizing each carbohydrate. After washing with 0.5 mL of buffer, the bound rThC was competitively eluted by elution buffer containing 0.2 M carbohydrate immobilized on the resin. The specific binding of rThC to the carbohydrate was evaluated by SDS–PAGE.
Crystallization, X-ray diffraction data collection, and structure determination
For crystallization, purified proteins concentrated up to approx. 6 mg/mL were used. Crystals of nThC were grown from a buffer composed of 0.1 M Tris-HCl (pH 8.5), 0.2 M MgCl2, 30% (w/v) PEG 4000. The diffraction dataset of the nThC was collected at Advanced Photon Source (IL, USA). The diffraction data of nThC were processed with the program HKL200030.
For phasing of the nThC, the crystal structure of SeMet-substituted rThC (SeMet-rThC) was determined. Crystals of SeMet-rThC were grown from a buffer composed of 100 mM sodium acetate (pH 3.3~5.5), 20% PEG3350~6000, and 20% PEG400. rThC Q25K was crystallized in the presence of 5 mM CaCl2 because a biochemical analysis revealed that rThC requires Ca2+ ions for its carbohydrate binding activity. Crystals of rThC Q25K in the presence of 5 mM CaCl2 were grown from a buffer composed of 100 mM sodium acetate (pH 3.3~5.5), 20% PEG3350~6000, and 20% PEG400. Crystals of rThC in complex with fucose or mannose were obtained by cocrystallization, in which 5 mM fucose or mannose was added to the purified rThC solution. X-ray diffraction experiments were conducted in Photon Factory (Tsukuba, Japan) and SPring-8 (Harima, Japan). Diffraction data of SeMet-substituted rThC, rThC in the presence of CaCl2, rThC in complex with fucose, and rThC in complex with mannose were collected in Photon Factory. The diffraction data of rThC were processed with the program XDS31. The statistics of data collection are summarized in Supplementary Table 2.
The crystal structure of SeMet-rThC was determined by the Se-SAD method.
The sites of Se were determined using the program HKL2MAP32. Phasing and model building were performed using phenix.autosol33. The crystal structure of nThC was determined by the molecular replacement method using the program phenix.mr34 with the structure of SeMet-rThC as the search probe. The crystal structures of rThC and its complexes with Ca2+ ions, mannose, and fucose were determined by a molecular replacement method with the structure of nThC as the search probe. An anomalous difference Fourier map was calculated using phenix.maps. Structure refinement was performed using phenix.refine 35.
STAT5 reporter assay
For the expression of MPL with an amino acid substitution from asparagine (N) to glutamine (Q) on a potential N-glycosylation site at N117, 178, 298, and 358, cDNAs were created by PCR mutagenesis (primers listed in Supplementary Table 2) and subcloned into the pcDNA3.1 vector (Life Technologies). V5-tagged mutant MPL and untagged wild-type MPL were used for the reporter assay. All plasmids constructed were verified by sequencing before use. The reporter assay was performed as described previously21 with the following modifications. The reporter activity was measured 24 h after transfection and 19 h after the addition of ThC or TPO.
Immunoblot
All immunoblot analyses were performed as described previously10. The following primary antibodies purchased from Cell Signaling were additionally used in this study: anti-STAT5(#94205).
Measurement the levels of MPL on cell surface
Cell surface MPL was isolated and detected as described previously7 with following modifications. Ba/F3-HuMpl cells were pre-cultured with the medium in the absence of agonist overnight, and then incubated with the media in the presence or absence of ThC or TPO at 37°C in a humidified incubator with 5% CO2. For the examination of synergistic effect of ThC and TPO, the pre-culture was performed with the media in the absence of agonist overnight and in the presence of ThC for 2 hours before the addition of TPO.
Statistical analysis
Data were obtained more than triplicate at each datapoint representing mean value and error bars ±SD. Histograms were analyzed using Dunnett’s multiple comparison test. Concentration response curves are generated using GraphPad Prism 8.0.3. EC50 values with 95% confident intervals were obtained by dose response curve fittings with nonlinear regression curve fit with four parameters.