New carbohydrate binding domains identified by phage display based functional metagenomic screens of human gut microbiota

Purification of T7Select 10-3b phages

The T7Select 10-3b phage vector (Novagen, Merck) was used in this study. Non-recombinant/recombinant T7Select 10-3b phages were purified from the lysate of infected E. coli strain BL5403 by PEG-8000 precipitation as per the method described in T7 select system manual. The pellet was suspended in sterile water, to which equal volume of chloroform was added, vortexed to mix well, and centrifuged for 5 min at 5000 rpm. The aqueous layer containing phages was pipetted out avoiding the interface, and passed through a 0.22 μm filter in a sterile tube to obtain purified phages. Finally, phage DNA was purified by phenol-chloroform extraction and ethanol precipitation.

Preparation of packaging extract

We made the packaging extract with the lysate of replication-deficient T7 Δ9-10B, D104, Δ38 phage, following the protocol as kindly provided by Prof. F. William Studier, Brookhaven National Laboratory and Dr. Tatjana Heinrich, Institute for Child Health Research, Western Australia.

For the propagation of replication defective phage, an overnight culture of E. coli BL21/pAR3924,5453 was set up by inoculation of a single colony in 5 ml of LB/G medium (LB + 0.3% filter sterile glucose with 50 µg/ml Carbenicillin and 30 µg/ml Chloramphenicol), and the culture was grown with shaking at 200 rpm at 37 °C. A secondary culture was set up by diluting the overnight culture 1:200 into 50 ml of LB/G medium and grown with shaking at 200 rpm and 37 °C until the cell density corresponded to an OD600 of 0.5. Then, 5 ml of this culture was transferred to a tube and grown for another 30 min and kept at 4 °C till required. The remaining 45 ml culture was infected with 10 µl of replication defective phage (PEG-precipitated, titer: 1–2  × 1011 pfu/ml) and kept in an incubator shaker at 37 °C for 3–4 h until lysis was observed. Lysis was determined by comparing the O.D. of culture with and without phage addition. Following lysis, the cell debris was pelleted down by centrifugation at 12,000 rpm for 30 min at 4 °C and the supernatant was transferred into a fresh tube. For long-term storage, the clarified lysate was kept at −80 °C without addition of glycerol. For the generation of packaging extract, phage-containing lysate was PEG-precipitated by the addition of 0.1 volume of 5 M NaCl and 0.166 volume of 50% PEG-8000, mixing well on a vortex, and incubation at 4 °C overnight. Subsequently, the mixture was centrifuged for 20 min at 4 °C at 12,000 rpm. The supernatant was removed carefully, and the pellet was briefly air-dried and resuspended in PBS (one-tenth the volume of original phage-containing lysate). The titer was checked with E. coli BL21/pAR3924,5453 culture as described in T7Select System (Novagen) manual.

For the preparation of replication defective phage for T7 packaging extract, an overnight culture of BL24/pAR3924,5453 host strain was set up in medium (LB + 0.3% filter sterile glucose with 50 µg/ml Carbenicillin and 30 µg/ml Chloramphenicol) by inoculation of one colony into 5 ml of broth and the culture was grown with shaking overnight at 37 °C. Then, 200 ml of a secondary culture was set up by diluting the overnight culture 1:200 in a 1 liter flask containing LB/G medium and culturing with continuous shaking at 200 rpm and 30 °C until the cell density corresponded to an OD600 of 1. At OD600 of 1.0, the bacterial culture was infected with replication defective phage at multiplicity of infection (MOI) of 7 and grown by shaking (200 rpm) at 30 °C. Exactly after 25 min of infection, the culture was chilled by placing the flask in ice water for ~5 min. Then, the culture was transferred into a cold centrifuge bottle and cells were pelleted for 2 min at 8000 rpm at 4 °C, and the remaining procedure was performed at 4 °C. The supernatant was decanted and removed, and the pellet was resuspended in 25 ml of cold extract buffer (100 mM NaCl, 20 mM Tris-Cl pH 8.0, 6 mM MgSO4). Cells were pelleted again at 8000 rpm for 2 min and the supernatant was removed. The pellet was resuspended in 400 µl of extract buffer using a pipette tip, and transferred to a cold microcentrifuge tube kept on ice. The cells were lysed by freezing in liquid nitrogen for 5 min and thawed to 30 °C quickly in water bath. The freeze-thaw was repeated, and the tube placed on ice for 5 min, and the lysate cleared by centrifugation at 12,000 rpm at 4 °C for 10 min. The supernatant (clarified extract) obtained was saved. To make packaging extract, 25 µl of 50% dextran, 3.33 µl each of MgSO4, ATP, and Spermidine, and 2 µl of β-mercaptoethanol (diluted 1:100 volume/volume in water) were added per 100 µl of clarified extract. Then, the packaging extract was aliquoted and stored at −80 °C.

In vitro T7 phage packaging was done according to the T7Select System (Novagen) manual. The number of recombinants generated was determined by performing a plaque assay as described in the manual.

Preparation of non-recombinant T7Select 10-3b phages

To generate non-recombinant phages (henceforth referred to as T7Select 10-3b phage), 9 µl packaging extract, 1 µl T7Select 10-3b DNA (100–500 ng of T7 DNA), and 1 µl T7 DNA polymerase (NEB; 10 U/µl) diluted 1:10 in extract buffer were mixed and incubated at 22 °C for 2 h. The reaction was stopped by the addition of 90 µl of LB. The titer was checked as per instructions in the Novagen manual.

Construction of recombinant fucose-binding T7Select 10-3b phage (SrNaFLD-T7Select 10-3b phage)

A 759 bp fragment corresponding to the Na domain and the fucose binding F-type lectin domain (FLD) of a Streptosporangium roseum gene,SrFucNaFLD (GenBank accession no: ACZ87343.1) was PCR amplified from an SrFucNaFLD-pET-28a(+) construct (Bishnoi, R. et al.) using Taq DNA polymerase, forward primer with an EcoRI site: (5′-CCG GAA TTC TGT CCG TTC ACT GCT GCG CAC C-3′) and reverse primer with a HindIII site: (5′-CCC AAG CTT GCC ACG CAC TTG GAC TTC TGC CA−3′).

The EcoRI and HindIII digested gene fragment was ligated to the EcoRI and HindIII digested T7 vector arms (T7Select System, Novagen) according to the user manual. Briefly, for sticky end ligation, 0.5 µl (0.5 µg/µl) of EcoRI and HindIII digested vector arms and EcoRI and HindIII digested insert in a 1:3 (vector:insert) ratio were assembled together with 0.5 µl 10× T4 ligase buffer, 0.5 µl 10 mM ATP, 0.5 µl 100 mM DTT, 1 µl T4 DNA ligase and water in a total reaction volume of 5 µl and incubated at 16 °C for 16 h.

The ligation mixture was packaged into the packaging extract, the phages were amplified and plated. Plaque plugs were picked by micropipette tips and placed in 30 µl of Phage Extraction Buffer (20 mM Tris-HCl, pH 8.0, 100 mM NaCl, 6 mM MgSO4). They were screened by PCR amplification using T7SelectUP (5ʹ-GGAGCTGTCGTATTCCAGTC-3ʹ) and T7SelectDOWN (5ʹ-AACCCCTCAAGACCCGTTTA-3ʹ) primers for the presence of insert. Recombinant phages (referred to as SrNaFLD-T7 phages) from single plaques obtained from agar plugs were amplified in fresh culture to rule out the possibility of the presence of non-recombinant phages in the stock.

Construction of metagenomic phage display library

This study was approved by Council of Scientific and Industrial Research (CSIR)- Institute of Microbial Technology Institutional Ethics Committee (Human) (Project number 11 IEC/1/9-2014) and Council of Scientific and Industrial Research (CSIR)- Institute of Microbial Technology Institutional Biosafety Committee (Project number IBSC/2012-2/21). We received written informed consent to release the information obtained and publish the study following the subject’s participation without disclosing the subject’s identity. Subjects were self-reported healthy individuals with local dietary intake from Chandigarh (30.7333˚N, 76.7794˚E), Ladakh (34.425960˚N 76.824421˚E), Khargone (22.226704˚N, 75.863329˚E), and Jaisalmer (26.36539˚N, 70.42584˚E) in the age group of 18–58 years. Eleven of the 50 samples were from female subjects, and the mean age was 32. Samples from Ladakh, Khargone, and Jaisalmer were collected on 27 October 2013, 25–26 November 2014, and 9 November 2014, respectively. Fecal samples were collected in sterile containers, transported on dry ice, and stored at −80 °C until further processing.

Metagenomic DNA isolation was done as follows. DNA was isolated from 50 fecal samples provided by 45 volunteers using ZR Fecal DNA MiniPrep™ (for 18 samples), MoBio PowerFecal DNA Isolation Kit (for 27 samples), QIAamp DNA stool minikit (for 4 samples), and MDI Stool Genomic DNA Miniprep Kit (for 1 sample) following the respective manufacturers’ instructions. The procedure was slightly modified for isolation of DNA using ZR Fecal DNA MiniPrep™. Lysis buffer was added to (up to) 150 mg of fecal sample and the sample was subjected to three rounds of homogenization (speed 6) for 40 s each (with cooling on ice for 1 minute after each round of homogenization) using FastPrep 120. The remaining steps of the protocol were performed as per the manufacturer’s instructions. Isolated DNA was subjected to RNAse treatment followed by phenol-chloroform extraction and ethanol precipitation. The quality of the isolated DNA was assessed by visualization following agarose gel electrophoresis and the DNA was quantified by using NanoDrop 1000 spectrophotometer.

Metagenomic DNA fragmentation and repair was done as follows. The DNA isolated from different fecal samples was fragmented by sonication to generate maximum fragments of 500–3000 bp. Fragmented DNA of 500–3000 bp was resolved by electrophoresis on a 1% agarose gel, excised, and purified by QIAquick Gel Extraction Kit. Equal amounts of purified DNA from all the samples were pooled. This pooled DNA was end repaired using NEBNext End Repair Module (New England BioLabs).

T7 phage vector digestion and dephosphorylation was done as follows. T7 phage vector arms were prepared by restriction digestion with SmaI (New England Labs) at 25 °C for 4 h to generate blunt ends, followed by heat inactivation of SmaI at 65 °C for 20 min, and dephosphorylation by the addition of Calf Intestinal Alkaline Phosphatase (New England Labs) and incubation at 37 °C for 30 min. Vector arms were purified by phenol extraction.

For ligation of T7 phage vector arms and metagenomic DNA fragments, the following procedure was used. A blunt end ligation reaction was set up by assembling 1 µl (1 µg/µl) SmaI digested vector arms, and end repaired 500–3000 bp metagenomic DNA in a 1:6 (vector arms: insert) ratio (considering 1500 bp as average size), 1 µl 10x T4 ligase buffer, 1 µl 100 mM ATP, 1 µl 100 mM DTT, 2 µl 50% PEG-4000, 2 µl of T4 ligase and TE buffer in a total reaction volume of 10 µl. The ligation mixture was then incubated at 16 °C for 24 h in a water bath.

For packaging and amplification of the metagenomic phage display library, the following procedure was used. The ligation mixture was packaged into the packaging extract to obtain a human fecal metagenomic phage display library (of 6 ml volume) with a titer of 2 × 106 pfu/ml and a total of 1.2 × 107 phages. The phages were amplified and the titer of the amplified metagenomic phage display library was 3 × 1011 pfu/ml. The amplified metagenomic phage display library therefore contained ~3 × 104 copies of each recombinant phage clone per ml of metagenomic phage display library. The amplified metagenomic phage display library was plated, and recombinant phage plaques screened by PCR amplification using T7SelectUP (5ʹ-GGAGCTGTCGTATTCCAGTC-3ʹ) and T7SelectDOWN (5ʹ-AACCCCTCAAGACCCGTTTA-3ʹ) primers for the presence of insert.

Preparation of 96-well plates coated with carbohydrate ligands for biopanning

Mucin from porcine stomach (Sigma) dissolved in sodium carbonate-bicarbonate buffer (pH 9.2) was applied (100 µl of 100 µg/ml each well) to 96-well plates (Nunc MaxiSorp ELISA plates) overnight at 4 °C. Each well was washed twice with 200 µl phosphate-buffered saline (8 mM Na2HPO4, 150 mM NaCl, 2 mM KH2PO4, 3 mM KCl, pH 7.4) containing 0.05% Tween 20 (PBS-T) and once with 200 µl PBS. Then, the plates were blocked with blocking agents for 4 h at room temperature. Bovine Serum Albumin (BSA) (300 µl of 3% in PBS) was used for blocking in the first, second, fifth, and sixth cycles of phage display selection whereas skim milk powder (300 µl of 5% in PBS) was used for the third, fourth, seventh, and eighth selection cycles. The plates were washed thrice with PBST and twice with PBS. Finally, 100 µl of water was pipetted to each well, sealed with parafilm and stored at 4 °C.

For Biotin-PAA-sugars, NeutrAvidin-coated plates were made by incubating 50 µl (4 mg/ml in deionized, sterile water) of NeutrAvidin in 96-well plates (Nunc MaxiSorp ELISA plates) overnight at 4 °C. The remaining steps were similar to that in mucin plate preparation except for the composition of the buffers used – TBS (20 mM Tris, 150 mM NaCl, pH 7.4) and TBS-T (TBS + Tween-20 0.1%). Before incubation with phages (initiating biopanning) NeutrAvidin coated plates were incubated with biotin-PAA-sugars (Glycotech; 50 µl of 10 µg/ml) (listed in Table S1) for 1 h at 37 °C. Washing was done thrice with TBS-T and twice with TBS.

Water soluble polysaccharides and peptidoglycans (Sigma; listed in Table S1) were applied (100 µl of 10 µg/ml in deionized, sterile water) to 96-well plates (Costar 3598 plates) followed by drying to the well surface by evaporation overnight at 37 °C24. Water insoluble polysaccharides were dissolved in DMSO (1 mg/ml) and 1 µl of the solution was applied to each well followed by 99 µl of deionized water to Costar 3598 plates. Then, it was left for drying by evaporation overnight at 37 °C. The remaining steps of biopanning were as mentioned for mucin and NeutrAvidin. The buffers used were TBS and TBS-T.

Biopanning

Ligand-coated plates were incubated with 100 µl of phages (overnight at 4 °C in first selection cycle, for 2 h at 37 °C in selection cycles 2–5, and for 2 h at 37 °C with shaking at 100 rpm in selection cycles 6–8), washed 6 times with PBST (in case of mucin) or TBST, and thrice with PBS (in case of mucin) or TBS, and then incubated with the appropriate elution reagent (as listed in Table S1 for other ligands) for 1 h at 37 °C to elute bound phages. After 1 h of incubation at 37 °C, the elution reagent containing phages was pipetted out into sterile microcentrifuge tubes. Next, 10 µl of the elution reagent was used for checking titer and the remaining solution was used for amplification in host cell culture. Following lysis of host cells, the debris was pelleted by centrifugation at 12,000 rpm for 15 min at 4 °C. The supernatant was used for the next cycle of phage display selection. Eight cycles of phage display selection were performed.

For biopanning of the metagenomic phage display library, we calculated the multiplicity of screening to be ~3000, considering that we used 100 µl of amplified metagenomic phage display library with a titer of 3 × 1011 pfu/ml (which contained ~3 × 104 copies of each recombinant phage clone per ml of metagenomic phage display library).

Sequencing and sequence/structure analysis

Following selection, randomly picked phage plaques were screened by PCR amplification using T7SelectUP (5ʹ-GGAGCTGTCGTATTCCAGTC-3ʹ) and T7SelectDOWN (5ʹ-AACCCCTCAAGACCCGTTTA-3ʹ) primers for the presence of insert. Phage plaques with amplicon size >144 bp (which is the size of the amplicon expected for a non-recombinant phage amplified by the T7UP and T7DOWN primers) were considered to be recombinant. Enriched recombinant phages were further subjected to Sanger sequencing in-house (at IMTECH, using 16-capillary 3130xl Genetic Analyzer, Applied Biosystems). DNA sequences obtained were translated using ExPASy80 and visualized using FinchTV version 1.4.0. Sequences with more than 30 continuous amino acids (without any stop codon) were considered for further analysis. These metagenomic sequences were subjected to searches against both the protein sequence database, ‘UniRef100’81, and the profile databases, ‘pfamA-full’ and ‘pfamB’31, and ‘dbCAN2’33 using mmseqs282 at a very high sensitivity (-s 8.5). In addition, an all-against-all sequence search of the initial 33 query proteins was also performed. The general command run for these searches was ‘mmseqs easy-search input_protein_seqs search_database output.result_file tmp_directory -s 8.5 –format-mode 2‘. In the search against UniRef100, for each query protein sequence, the top five returned hits were retained and were then further searched for pfamA-full domains as above using mmseqs2. The fields ‘representativeMember_organismName‘ and ‘cluster_representative_name‘ (in Table S5) were obtained using uniprot API81. In the search result against pfamA-full, the field ‘pfam-family_description‘ (in Table S7) was obtained using github.com/AlbertoBoldrini/python-pfam (a python interface for pfamA). Three-dimensional structure predictions were performed for all amino acid sequences using a local server installed AlphaFold2, and structures were visualized using reverse rainbow coloring based on pLDDT scores in PyMOL (Schrodinger) (Supplementary Data 2).

Cloning, expression, and purification of recombinant metagenomic insert sequences

The PCR amplicons of the clones, MG1 and MN3, amplified with the primers, forward: 5ʹ-GATCCGAATTCTCTCCTGCAGGGATATC-3ʹ and reverse: 5ʹ-AAC CCC TCA AGA CCC GTT TAG AGG CC-3ʹ, were digested with the restriction enzymes, EcoRI and HindIII. Similarly, the PCR amplicons of the clones, MU1 and MU3, amplified with the primers, forward: 5ʹ-CAGCCATATG GGAGCTGTCGTATTCCAGTCA-3ʹ and reverse: 5ʹ-AAC CCC TCA AGA CCC GTT TAG AGG CC-3ʹ, were digested with the restriction enzymes, NdeI and XhoI. These PCR amplicons were ligated into the expression vector, pET-28a(+), digested with corresponding enzymes and treated with CIP. E. coli TOP 10 cells were transformed with the ligation mixture and the transformants were grown on LB agar plates containing kanamycin (50 µg/ml) at 37 °C for 14-16 h. Transformant colonies were screened by restriction digestion and confirmed by DNA sequencing.

For expression of recombinant proteins, a primary overnight culture of E. coli BL21(DE3) cells transformed with the pET-28a(+) plasmid clone was used to inoculate LB medium containing kanamycin (50 µg/ml). The cultures were grown with continuous shaking at 200 rpm to an OD600 of 0.6–0.8 at 37 °C, whereupon recombinant protein expression was induced as follows. We used 0.1 mM IPTG for 3 h 30 min at 37 °C with 200 rpm shaking for the expression of MG1, MN3, MU1, and MU3 protein (subsequently used for glycan array analysis). We used 0.1 mM IPTG for 10 h at 22 °C with 180 rpm shaking for the expression of MG1 (subsequently used for CD spectroscopy and ITC assays). We used 1 mM IPTG for 3 h 30 min at 37 °C with 200 rpm shaking for the expression of MN3, MU1, and MU3 (subsequently used for CD spectroscopy, ELLA and ITC assays). We used 1 mM IPTG for 3 h at 37 °C with 200 rpm shaking for the expression of Lev3, St1, and St5.

For protein purification, cells were harvested by centrifugation at 4000 × g for 7–10 min. Recombinant proteins with N-terminal hexahistidine tags were purified by metal ion affinity chromatography. The cell pellet was resuspended in lysis buffer, the composition of which was as follows. We used 150 mM or 300 mM NaCl, 25 mM imidazole, and 20 mM Tris, pH 7.4 for MG1, MN3, MU1, and MU3. For MG1, where sonication was used to lyse the cells, the lysis buffer also included 20 mU/mL DNase I (Roche), 100 µg/mL lysozyme (Sigma) and 1 mM PMSF (Sigma). We used 150 mM NaCl, 20 mM Tris, pH 7.5 for Lev-Lev5. We used 150 mM NaCl, 0.5% N-lauryl sarcosine and 20 mM Tris, pH 7.5 for St-Glc1 and St-Glc5v2.

The cells were disrupted using a French Press (using 300 mM NaCl; in case of MG1, MN3, MU1 and MU3 proteins subsequently used in glycan array analysis) or a probe type ultrasonicator (Materials and Sonics INC) for 30 min (amplitude 25%, pulse 10 s on and 10 s off) (using 150 mM NaCl; for proteins expressed for all other biochemical assays), and the lysate was clarified by centrifugation at 16,000 × g for 20–60 min at 4 °C. The supernatant was loaded on to a His-bind (Pierce metal affinity Ni-NTA resin) column (pre-equilibrated with lysis buffer) and incubated for 2–8 h at 4 °C with end-over-end rotation on a Rotospin (Tarsons). The column was then washed with wash buffer (40 mM imidazole in lysis buffer), and the protein eluted with elution buffer (250 mM imidazole in lysis buffer). The protein was subsequently dialyzed extensively against TBS, and the purity of the preparation evaluated by SDS-PAGE.

Where required (for MG1, MN3, MU1, and MU3), size exclusion chromatography was done using HiPrepTM 26/60 HR Sephacryl S-200 (GE Healthcare life sciences) column on a GE healthcare Akta Purifier Fast protein liquid chromatography (FPLC). For size exclusion chromatography, the duly washed column (with degassed sterile water) was equilibrated with 150 mM NaCl and 20 mM Tris, pH 7.5 at a suitable flow rate mostly (1.0 ml/min). The absorbance of the sample was measured at 280 nm and 260 nm. Peak fractions were collected and the concentrated sample was assessed for purity by SDS-PAGE. Protein concentration was estimated by measuring the absorbance at 280 nm using Nano-Drop spectrophotometer.

Western analysis was performed using mouse anti-polyHistidine antibody (H1029, Sigma) and mouse anti-C-terminal-His antibody (R930-25, Invitrogen) followed by HRP-conjugated donkey anti-mouse IgG (715-035-150, Jackson Immunoresearch), mass spectrometry for intact mass analysis was performed on an Agilent i6550 Quadrupole Time Of Flight instrument with electrospray ionization (CSIR-IMTECH mass spectrometry facility) and in-gel trypsin digestion and MS/MS analysis was performed on a Thermo Fisher Scientific LTQ Orbitrap Velos Pro ion-trap mass spectrometer (Taplin Mass Spectrometry Facility, Harvard Medical School, Boston, Massachusetts, USA).

Glycan micro-array analysis

For analysis of glycan binding specificities, glycan microarray analysis of the purified recombinant proteins, MG1, MN3, MU1, and MU3 (in TBS with 10 mM CaCl2) was performed at the National Center for Functional Glycomics (NCFG) at Beth Israel Deaconess Medical Center, Harvard Medical School using CFG glycan microarray version 5.3. Protein concentrations used were 5 μg/ml, 50 μg/ml, and 400 μg/ml for MG1, and 5 μg/ml and 50 μg/ml for MN3, MU1 and MU3. Briefly, the proteins were allowed to bind to the glycans in the microarray, the slide washed to remove non-specifically bound protein, and the bound protein detected with mouse anti-C-terminal 6xHis antibody and Alexa488-conjugated secondary anti-mouse antibody, followed by fluorescence scanning on a Perkin Elmer Scan Array Scanner. The spots with highest and lowest fluorescence intensity (of the eight spots for each glycan) were removed, and the remaining glycan array data analyzed by MotifFinder version 2.2.536,37. We employed the automated model building function and default model building settings of MotifFinder (motif complexity parameter: 0.0, model complexity parameter: 0.01, minimum split size: 3, rebuild Fold-N times: 0, cross validation Fold N: 5, optimize top N leads per split: 2, test top N relations per split: 1, guide glycan selection seed: 0). With these settings, we used the glycan array data from multiple protein concentrations to build one model each for MG1, MN3, MU1, and MU3 (Fig. S4). Subsequently, we also used all the glycan array data of MG1, MN3, MU1, and MU3 together, and built a single automated model, altering the settings (motif complexity parameter: 0, model complexity parameter: 0.05, minimum split size: 5) to achieve a final list of binding motifs that was representative of the binding motifs of each of these proteins. Using this output, we then compiled a custom motif list, and used the manual model building function to build one model each for MG1, MN3, MU1, and MU3, using the glycan array data of all protein concentrations (Fig. 3).

Circular dichroism spectroscopy

Circular dichroism spectroscopic measurement for proteins MG1, MN3, MU1 or MU3 was done using a Jasco J815-1505 Spectropolarimeter (Jasco International co. ltd.). Spectra of protein solutions (0.2 mg/ml) in 20 mM phosphate buffer pH 7.4 were recorded at far UV (250-195 nm) region at a scan speed of 50 nm/min, a slit width of 1 nm, a data pitch of 0.1 nm, with three accumulations, and a path length of 1 mm. The mean residue molecular mass and number of amino acids in the protein sequence were used to calculate mean residue ellipticity, θ. The spectra plotted are moving averages of mean residue ellipticity calculated over a window of 10 readings.

Isothermal calorimetry

Isothermal calorimetry experiments were performed on a VP-ITC microcalorimeter (Malvern). The purified proteins MG1, MN3, MU1, and MU3, were extensively dialysed against suitable buffer, which was mostly 20 mM HEPES (4-(2-hydroxyethyl)−1-piperazineethanesulfonic acid), 150 mM NaCl, pH 7.5. For titrations of MG1, MN3, MU1, and MU3 against Lewis A tetraose, Lewis B tetraose, Lewis X tetraose, Lacto-N-difucohexaose II, and H antigen type-2 heptaose azido ethyl, and for titrations of MG1 and MU3 against H antigen type-2 pentaose β-N-Acetyl Propargyl, 20 mM phosphate buffer, pH 7.5 with 150 mM NaCl was used. The ligand solutions were prepared in the dialysate buffer. The experiments were performed by placing the protein in the sample cell and titrating ligand in a fixed volume at intervals of 180 s for 20 or 35 injections. To minimize any artifact associated with the loading of the ligand filled syringe, the first injection was performed with the volume of 0.2/0.4 µl and the corresponding data point was removed before final curve fitting. The heat changes due to heat of dilution were determined in two ways, one by titrating protein with buffer and another by titrating buffer with ligand. The experimental curve data was subtracted from one of these curves prior to data analysis. The subtracted data was fitted to a model using Microcal origin 7 analysis software and the thermodynamic parameters K, n and ∆H along with chi square value were obtained.

Enzyme-linked lectin assays

For MN3 binding assay, purified recombinant protein MN3 (100 µl of 50 µg/ml stock in TBS) was coated on a Maxisorp 96-well flat-bottom plate (Nunc) overnight at 4 oC. Subsequently, the wells were washed twice with 200 µl TBST and once with TBS, blocked with 3% BSA for 4 h at room temperature, washed, incubated with 1 µg biotin-PAA-β-D-galactose, 1 µg biotin-PAA-α-D-galactose, 1 µg biotin-PAA-α-L-fucose, 1 µg biotin-PAA-β-D-GlcNac, 1 µg biotin-PAA-α-NeuAc, 1 µg biotin-PAA-Lewisy, 1 µg biotin-PAA-H antigen type-2 triose, 1 µg biotin-PAA-α-D-Mannose, 1 µg biotin-PAA-β-D-Mannose or 1 µg biotin-PAA for 1.5 hours at 37 oC, washed, further incubated with 50 µl of HRP-conjugated streptavidin at room temperature, washed, and developed with 50 µl of TMB-ELISA substrate. The reaction was stopped by the addition of 2 N H2SO4, and the resulting color developed was measured at 450 nm in a Synergy H1 plate reader (Bio-Tek).

For Lev-Lev5 binding assay, levan, inulin and glucan (Sigma) were immobilized in the wells of a MaxiSorp 96-well flat-bottom Nunc plate (2 µg per well) overnight at 37 °C. Subsequently, the wells were washed twice with 200 µl TBST and once with TBS, blocked with 3% BSA for 4 h at room temperature, and washed. For binding assays, two-fold dilutions of Lev-Lev5 starting from 0.4 mg/mL were prepared and added to the wells (50 µl per well) in duplicate. Buffer was added instead of Lev-Lev5 in control wells. After 2 h of incubation at room temperature, wells were washed (thrice with TBST and twice with TBS), incubated with mouse anti-C-terminal-His antibody (R930-25, Invitrogen) (50 µl per well) for an hour followed by washing and incubation with HRP-conjugated donkey anti-mouse IgG (715-035-150, Jackson Immunoresearch) (50 µl per well) for an hour at room temperature. TMB-ELISA substrate was added (50 µl per well), the reaction was stopped by adding 2 N H2SO4 (50 µl per well), and the resulting color developed in each well was measured at 450 nm in a Synergy H1 plate reader (Biotek). Non-linear curve fitting of the data was performed using Sigma plot software.

For Lev-Lev5 competitive inhibition assay, eleven polysaccharides (levan, inulin, xylan, amylopectin, pectin, arabinogalactan, glucan, dextran, laminarin, amylose and starch; 10 µg/mL concentration) were coated on MaxiSorp flat-bottom 96 well Nunc plate overnight at 37 °C. Buffer was added instead of polysaccharides in control wells. Subsequently, the wells were washed twice with 200 µl TBST and once with TBS, blocked with 3% BSA for 4 h at room temperature, and washed. Lev-Lev5 (5 µg/mL) was pre-incubated for an hour with buffer or with 10 mM sugars (corresponding to the constituent monosaccharide sugars of the polysaccharides coated in the well), and then added into the wells with coated polysaccharides and incubated at room temperature for 2 h. Wells were washed (thrice with TBST and twice with TBS), incubated with mouse anti-C-terminal-His antibody (R930-25, Invitrogen) (50 µl per well) for an hour followed by washing and incubation with HRP-conjugated donkey anti-mouse IgG (715-035-150, Jackson Immunoresearch) (50 µl per well) for an hour at room temperature. TMB-ELISA substrate was added (50 µl per well), the reaction was stopped by adding 2 N H2SO4 (50 µl per well), and the resulting color developed in each well was measured at 450 nm in a Synergy H1 plate reader (Biotek).

For St-Glc1 and St-Glc5v2 binding assay with starch, amylose and amylopectin, the wells of a Costar 3598 96-well flat bottom plate were coated with 1 μg of starch, amylose or amylopectin (by the addition of 100 μl of aqueous 10 µg/ml stocks made from 1 mg/ml starch in DMSO, 1 mg/ml amylose in DMSO, and 1 mg/ml amylopectin in water), and dried overnight at 37 °C. Subsequently, the wells were washed twice with 200 µl TBST and once with TBS, blocked with 3% BSA for 4 h at room temperature, and washed. For binding assays, two-fold dilutions of St-Glc1 (starting from 0.3 mg/ml) and St-Glc5v2 (starting from 0.56 mg/ml) were added to the wells. After 2 h of incubation at room temperature, wells were washed (thrice with TBST and twice with TBS), incubated with mouse anti-polyHistidine antibody (H1029, Sigma) (50 µl per well) for an hour followed by washing and incubation with HRP-conjugated donkey anti-mouse IgG (715-035-150, Jackson Immunoresearch) (50 µl per well) for an hour at room temperature, followed by washing. TMB-ELISA substrate was added (50 µl per well), the reaction was stopped by adding 2 N H2SO4 (50 µl per well), and the resulting color developed in each well was measured at 450 nm in a Synergy H1 plate reader (Biotek). Non-linear curve fitting of the data was performed using Sigma plot software.

For St-Glc1 and St-Glc5v2 competitive inhibition assays with starch, amylose, and amylopectin, the wells of a Costar 3598 96-well flat bottom plate were coated with 1 μg of starch, amylose or amylopectin and blocked with 3% BSA as described above. Pre-incubation of St-Glc1 (0.06 mg/ml) and St-Glc5v2 (0.11 mg/ml) with 100 μg/ml starch, amylose, amylopectin and glucose was performed at 4 °C for 30 min. Fivefold dilutions of the proteins (with or without prior incubation with starch, amylose, or amylopectin) were added to the wells in triplicate and incubated for 2 h at room temperature. Wells were then washed, and bound protein detected by incubation with mouse anti-polyHistidine antibody (H1029, Sigma) followed by HRP-conjugated donkey anti-mouse IgG (715-035-150, Jackson Immunoresearch) as described above.

Statistics and reproducibility

Biochemical assays were replicated, means (and standard deviations where relevant) are plotted and details of number of replicates are mentioned in the figure legends. Screening with phage display library was performed only once because our goal was to find the carbohydrate binding sequences, and later confirm the binding by performing biochemical assays.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Read more here: Source link