Table 1.
Tissue sources and numbers of tissue and tissue sections screened by immunofluorescence localization of antibodies.
Fig 1.
Identification of the target of the G8 mAb.
The G8 mAb was used to screen clones of HEK-239T cells expressing a single human membrane protein. G8 binding was detected by flow cytometry. A. Binding values were normalized and transformed to give a single numerical value for binding of the G8 mAb against each target protein (normalized target binding). Non-specific fluorescence was determined to be any value below three standard deviations above noise. G8 binding above background occurred with the clone expressing BAI1. B. Validation of binding of the G8 mAb to BAI1 or vector alone was carried out with HEK-293T cells transfected with the plasmid construct expressing target using a serial dilution of mAb. C. Binding of different concentrations of the G8 mAb was calculated as the signal to background ratio of fluorescence from the BAI1 expressing clone to empty vector clone.
Fig 2.
The G8 mAb binds to purified BAI1 in ELISA.
Plates were coated with recombinant human BAI1 protein produced by CHO cells. Binding of the G8 IgM mAb and R&D BAI1 IgG mAb (a-BAI) was detected with anti-mouse IgM and IgG affinity purified F(ab’)2 secondary antibodies, respectively, conjugated with horseradish peroxidase. Both mAbs bound to BAI1 protein.
Fig 3.
Identification of epitope of the G8 mAb.
An alanine-scan library of BAI1 was constructed. The G8 and reference 5D6 mAbs were screened for binding to each individual BAI1 variant. A. Identification of candidate critical clones for G8 mAb binding was determined in duplicate by high throughput flow cytometry. For each point, background fluorescence was subtracted from the raw data that were then normalized to G8 reactivity with wild type BAI1. For each mutant clone, the mean binding value was plotted as a function of expression represented by control antibody reactivity. Candidate critical clones (red circles) were identified with a threshold (dashed line) of <40% binding of the G8 mAb and >50% of binding of the control 5D6 mAb to wild type BAI1. B. Identification of validated critical residues for G8 mAb binding. Candidate critical residues for G8 were rescreened in quadruplicate and the mean binding reactivities obtained. Validated critical residues for G8 binding (outlined in red) were residues whose mutations that resulted in <20% of wild type binding but positive (>70% binding) for binding of the 5D6 control mAb. Additional validated secondary residues (outlined in blue) were identified that did not meet the threshold guidelines but whose decreased binding activity and proximity to critical residues suggested that they may be part of the antibody epitope. C. Visualization of validated critical residues for G8 mAb binding. Critical residues (red spheres) and secondary residues (blue spheres) which may also contribute to binding were visualized on two Phyre-generated model structures that were combined to model BAI1 residues 262–934. The G8 critical residues were visualized on a model of the BAI1 thrombospondin repeat domains based on the structure of properdin (PDB ID# 1WOR; [15].
Fig 4.
Co-localization of the G8 and R&D BAI1 mAbs in the skin and eyes.
Tissue sections of human tattooed skin (A-C), human anterior lens tissue (D-F), rabbit eyes (G-I) and rat retina (J-L) were double labeled with the G8 (red) and BAI1 (green) mAbs. Nuclei were stained with Hoechst dye (blue). Unmerged images are shown in A, B, D, E and the insets in G-I and K and L. Overlap of red and green appear yellow in triple merged images in C, F, G-I, K and L. A merge of fluorescence and DIC is shown in the inset of C. The G8 and BAI1 mAbs labeled the same cells in the skin, human lens, rabbit lens (G), ciliary body (H) and cornea (I), and the inner plexiform and inner nuclear layers of the mouse retina (K and L). Minimal background fluorescence was observed in the anti-IgM and anti-IgG secondary antibody controls for the skin (M), human lens tissue (N), rabbit lens (O) and mouse retina P and Q. Bar = 9 μM in A-I and K-Q and 54 μM in J.
Fig 5.
Co-localization of the R&D BAI1 mAb with antibodies to Noggin and MyoD in the skin and eyes.
Tissue sections of human tattooed skin (A and B), human anterior lens tissue (C) and rabbit anterior cavity (D-I) were double labeled with the R&D BAI1 mAb and antibodies to noggin or MyoD. The colors of the fluorescent secondary antibodies are indicated in each photograph. Overlap of red and green appears yellow in merged images. Nuclei were stained with Hoechst dye. Double labeled cells were present in the skin (A and B), anterior human lens tissue (C) and the rabbit lens (D and G), ciliary body (E and H) and cornea (F and I). Bar = 9 μM.
Table 2.
Localization of antibodies to BAI1, Noggin and MyoD in the skin, eyes and brain.
Human lens tissue and tissue sections of human tattooed skin, rabbit anterior cavity, rat retina and mouse brain were double labeled with the G8 and R&D BAI1 mAbs and antibodies to Noggin and MyoD. The total numbers of double labeled cells were counted in a subset of tissue sections. The total numbers of single labeled cells were quantified in all sections with the exception of MyoD-/R&D BAI1 mAb+ and MyoD-/G8- cells that were counted in a subset of skin and anterior cavity sections. The numbers of tissues and sections scored are indicated in parenthesis. The results are the mean ± standard deviation. The R&D BAI1 mAb co-localized with G8 and Noggin with rare exception. BAI1+ cells without MyoD were present in the skin and anterior cavity.
Fig 6.
Localization of BAI1, Noggin, Iba1, NeuN and GFAP in the mouse brain.
Tissue sections from the day-10 mouse brain were stained with hematoxylin and eosin (sagittal sections A and D; coronal sections B and C) or double labeled with the G8 and the R&D BAI1 mAbs, or G8 and antibodies to Iba1, NeuN or GFAP. The areas within the boxes of the H&E stained sections are shown at high magnification in the fluorescence photomicrographs. The colors of the fluorescent secondary antibodies are indicated in the unmerged photographs (E, F, H and I). Nuclei were stained with Hoechst dye. Overlap of red and green, when present, appears yellow in merged images (G, J, K, L and M). The G8 and BAI1 mAbs labeled the same subpopulation of cells in the hippocampal formation (box in A; E-G). The Noggin and BAI1 antibodies also bound to the same cells in the glomerular layer of the olfactory bulb (box in B; H-J). G8 did not co-localize with Iba1 (K, from box in A), NeuN (L from lower box in C) or GFAP (M from box in D). Minimal fluorescence was observed with the anti-IgM and anti-IgG (inset in G) or the anti-goat and anti-IgG (inset in J) secondary antibodies. Bar = 270 μM in A-D and 9 μM in E-M.