Table 1.
Oligonucleotide primers used for the construction of the immunized rabbit phage library.
Fig 1.
Schematic outline of the procedure for the construction of the immunized rabbit scFv library.
Once the antibody titer of immunized rabbit, as determined by agglutination method, was higher than 1:1600, the total RNA was extracted from spleen (A). Then, cDNA was synthesized by using a mix of random hexamers and oligo-dT18 (B). The locations of all PCR primers on the two variable regions genes are indicated (C). The list of all primers used for the construction of the library is given in Table 1. The first PCR step comprised 29 PCR reactions for amplification of VH and VL gene repertoire. Then, equal amounts of VH and VLκ, and VH and VLλ were assembled together via a (G4S)3 linker sequence by overlap extension, followed by pull-through PCR steps (D). The amplified full-length scFv repertoire fragments were then cloned into a pMOD1.0 vector (see full-size map in Supporting information S2 Fig), between Sfi I and Not I sites, before being transformed into E. coli TG1 (E). The agarose gel (inset) illustrates an example of DNA fragments from various steps; lane 1: 100 bp DNA ladder (NEB, USA); lane 2: VLλ; lane 3: VLκ; lane 4: VH; and lane 5: assembled scFv fragments (F). Please note that even the PCR products of VH obtained from the PCR reactions were quite low as shown as a faint band in the lane 4 of the gel, we were able to use them to generate assemble products of scFv genes by pull-through PCR as shown in lane 5.
Table 2.
Selective enrichment of scFv antibodies during the biopanning process.
Fig 2.
Specific binding of selected phage-displayed scFv clones.
Phage ELISA results of the binding of scFv antibodies against DOA9 and other antigens in pure culture (A) and plant nodules (B) are shown. The two clones of phage-displayed rabbit scFv, i.e., RB9 and RG8, could bind specifically to only strain DOA9 but not to other Bradyrhizobium strains and Bacillus subtilis 168. Phage-displayed human 3C1 scFv antibody was used as a negative control in this assay. The average OD405 nm values and standard errors from triplicate wells are shown. Note that in panel B, there was no bacillus nodule because it can’t form nodule in plant roots.
Fig 3.
Amino acid sequence and 3D structure of isolated anti-DOA9 rabbit scFv antibody.
The primary amino acid sequence of the clone bDOA9rb8 is shown at the bottom. The flexible linker (G4S)3 that joins VH and VL segments is indicated. The three complementarity-determining regions (CDRs) are underlined. The 3D structure was done by Phyre2 software, using the structure of an scFv antibody in complex with an analogue of the main immunogenic region of the acetylcholine receptor (PDB code: 1F3R) as templates. The 3D structures in both space-filling and ribbon models with CDRs of VH and VL domains are indicated.
Fig 4.
Antibody characterization via dilution series.
(A) A checkerboard titration experiment of phage-displayed scFv clone bDOA9rb8. Various amount of phage particles were prepared by 10-fold dilution series starting from 1013, 1012, 1011, 1010, and 109 pfu/well. Each dilution of phage antibody was used to detect different amount of boiled DOA9 antigen at the protein concentration of 10, 5.0, 2.5, 1.3, 0.6 and 0.3 μg/well of total protein, determined by Bradford assay. Each reaction was done in triplicate. The phages were purified by PEG precipitation and re-suspended in PBS buffer at a concentration of 1011 pfu/μl. Bound phage was detected by anti-M13 HRP, using ABTS as color reagent. The Y-axis indicated the OD value and the SD measured from triplicate wells. (B) Detection limit of phage displayed scFv clone bDOA9rb8 against boiled DOA9. The 5-fold dilution series of boiled DOA9 protein antigen were prepared in sodium carbonate buffer starting from the protein concentration of 10, 2, 0.4, 0.08, 0.02, 0.003 and 0.00 μg/well. The ELISA was performed by using the optimum concentration of phage antibody (1012 pfu/well), determined from panel A. The average OD405 nm values and standard errors from triplicate wells are shown.
Fig 5.
SDS-PAGE and Westernblot analysis of purified soluble scFv antibody.
(A) The soluble scFv antibody against Bradyrhizobium sp. strain DOA9 clone bDOA9rb8 was purified from culture supernatant by IMAC. Lane M: protein molecular weight marker; lane S: culture supernatant input; lane FT: flow-through fraction; lanes W1, W2, and W3 indicate the three wash fractions; lanes E1, E2, and E3 are the three elution fractions. The soluble scFv antibody of approx. 30 kDa can be found in elution fractions 1 and 2. This scFv antibody was used to study the binding specificity in the next step. (B) SDS-PAGE and Western blot analysis of free scFv clone bDOA9rb8 obtained from periplasmic extracts.
Fig 6.
Binding property of free scFv against strain DOA9.
ELISA results of soluble scFv antibodies against bacterial targets in both free-living (A) and bacteroid form (B) in plant nodule are illustrated. Rabbit polyclonal antibody and Phage-display scFv were also used in the assay for comparison. Values are the mean of triplicate wells. Error bars show the standard deviation for each set of data. Panel C illustrates binding properties of free scFv against DOA9 in a checkerboard titration experiment. Serial dilutions of soluble scFv antibody ranging from 50–0 μg were add into wells of ELISA plated immobilized with various concentration of boiled DOA9, ranging from 10 to 0 μg of total protein. The Y-axis indicated the OD value and the SD measured from triplicate wells.
Fig 7.
Confocal microscopy images of immunofluorescence staining for Bradyrhizobium sp. prepared from pure culture and plant nodules.
Immunofluorescence staining of various bacterial targets in both pure culture and nodule forms are shown. The bacteria tested include Bradyrhizobium sp. strains DOA9 (A), SUTN9-2 (B), SUTN1-12 (C), and USDA110 (D). The bacterial samples were stained with various antibodies, i.e., rabbit polyclonal antibody, two clones of phage-displayed rabbit scFv, i.e., RB9 and RG8, which contained identical DNA sequence. Phage-displayed scFv antibody clone 3C1 that bind aflatoxin B1 was used as a negative control. Green spots indicated the green fluorescent staining of phage display scFv or polyclonal antibody, using secondary antibody conjugated to FITC. Plant cell walls were stained with a blue fluorophore (Calcofluor white M2R) and emitted blue color. The bacteroid are shown as green spots inside a blue plant cavity. Scale bar is 10 μm at 80× magnification, and 100 μm at 10× magnification for pure culture and nodule samples, respectively.
Fig 8.
Investigation of the morphology of symbiotic nodule using phage-displayed scFv clone bDOA9rb8.
Bright field and immunofluorescence staining images from confocal microscopy of bacteriod form of Bradyrhizobium sp. DOA9 in plant nodules at different magnifications are shown. Sample in A was detected with negative phage clone 3C1; B rabbit polyclonal antibody, or C the phage-displayed scFv clone bDOA9rb8 from this study. Blue color indicated the plate cell wall that was stained with blue fluorophore. Scale bar is 100 μm at 10X magnification and 50 μm at 40X magnifications.
Fig 9.
Determination of limit of detection (LOD) of whole cell DOA9 by phage ELISA using phage-displayed scFv bDOA9rb8.
The 5-fold dilution series of whole cell DOA9 was prepared in sodium carbonate buffer starting from 1.3×107, 2.6×106, 5.2×105, 1×105, 2×104, 4×103, 8×102, 1.6×102, and 0 cell/well. Phage-displayed scFv bDOA9rb8 was added at 1012 pfu/μl/well and detected by Phage ELISA as described. The lowest number of cells that gave the good signal from ELISA (indicated by arrow) was the detection limit of Bradyrhizobium sp. DOA9 in form of whole cell antigen.