Validation and in vivo characterization of research antibodies for Moesin, CD44, Midkine, and sFRP-1.

Mouse brain collection

For terminal tissue collection, mice were anesthetized with isoflurane to the surgical plane of anesthesia, and brains were collected following decapitation. The brain was extracted and rinsed in ice-cold PBS, cerebellum was removed, and the cortex was bisected into left and right hemispheres then snap frozen and stored at -80 °C until use. Each hemibrain was weighed and then immersed in 1 ml/100 mg tissue homogenization buffer (THB; 2 mM Tris (pH 7.4), 250 mM sucrose 0.5 mM EDTA 0.5 mM EGTA) supplemented with 1X Pierce Protease Inhibitor (Thermo Scientific #A32953) and 1x Phosphatase Inhibitor Cocktail 2 (Sigma-Aldrich #P5726, St. Louis, MO). The tissue was homogenized for 20 sec on ice with a Benchmark D1000 hand-held homogenizer beginning at medium and increasing to high speed. Total protein concentration of the resulting homogenate was measured using a Bradford assay (Bradford 1976). Briefly, 5 μl of homogenate diluted 10× in PBS was added to a 96-well plate. Pierce™ Coomassie Plus (Bradford) Assay Reagent (Life Technologies, Chicago, IL, USA; 250 μl) was added to each well and the plate was read at 595 nm using a SprectraMax i3x (Molecular devices) with SoftMax Pro V7.0.2. The protein concentration was calculated by comparing to a standard curve generated by adding varying known concentrations of Albumin Standard (0, 125, 250, 500, 750, 1000, 1500 and 2000 μl/ml; (Thermo Scientific #23210, Rockford, IL, USA) in duplicate to wells.

Western blots

Cell cultures: Wild type (WT) and MDK knockout (KO) HAP1 cells were obtained from Horizon Discovery. Cell lines were prepared by first washing 3× with PBS then starved for ~18 hrs. Culture media were collected and centrifuged for 10 min at 500 × g to eliminate cells and larger contaminants, then for 10 min at 4500 × g to eliminate smaller contaminants. Culture media were initially concentrated using Amicon Ultra-15 Centrifugal Filter Units (MilliporeSigma cat. number UFC9010) by centrifuging at 4000 × g for 15 min. The resulting 500 μl of the concentrated media were centrifuged again at 4000 × g for 15 min using Amicon Ultra- 0.5 Centrifugal Filter Units (MilliporeSigma cat. number UFC5010) to 200 μl. ~30 μg of protein was loaded and processed for Western blot using the Midkine antibody (AF7769). Western blot was performed using a large 10-20% gradient polyacrylamide gel and transferred on a nitrocellulose membrane. Proteins on the blot were visualized with Ponceau staining which is scanned to show together with the Western blot. The blot was blocked with 5% milk for 1 hr, and the Midkine antibody (AF7769) was incubated overnight at 4°C with 5% bovine serum albumin in TBS with 0.1% Tween 20 (TBST). Following three washes with TBST, the peroxidase conjugated secondary antibody was incubated on the membrane at a dilution of ~0.2 μg/ml in TBST with 5% milk for 1 hr at room temperature followed by three more washes with TBST. The membrane was incubated with ECL from Pierce (cat. number 32106) prior to detection with HyBlot CL autoradiography films from Denville (cat. number 1159T41). The MDK KO cell line can be confirmed based on the results of the Western blot, as the Midkine signal appears in the lane loaded with the WT cells but disappears in lane loaded with the MDK KO cells. Predicted band size: ~16 kDa.

Brain Homogenates: Homogenate from N=2 male and n=2 female 9 month aged 5XFAD subjects and homogenate from N=4 age-matched individual male C57BL/6J subjects, which served as WT controls, were included on each gel. Equal amounts of protein (25 μg) samples were prepared in 1x loading buffer (4x: 4 ml 100% glycerol, 2.4 ml 1M Tris/HCl (pH 6.8), 0.8 g SDS, 4 mg bromophenol blue, 3.1 ml H2O, 0.5 ml beta-mercaptoethanol) and incubated at 95 °C for 5 min. Samples were separated with SDS-polyacrylamide gel electrophoresis (4–15% Mini-PROTEAN^® TGX™ Precast Protein Gels, 10-well, 30 μl, Bio-rad #4561083) using 120 V for 60 min, then transferred onto a nitrocellulose membrane using a Trans-Blot SD Semi-Dry Electrophoretic transfer cell (Bio-rad #170-3940) at 25 V for 30 min. Non-specific binding was blocked using EveryBlot Blocking Buffer (Bio-rad #12010020) for 5 min at room temperature with gentle rocking. Primary antibodies were prepared in EveryBlot Blocking buffer at the dilutions indicated in Table 1. Blots were immersed in primary antibody solutions overnight at 4 °C with gentle rocking. Membranes were then washed with 1X TBST (3 × 5 min) and immersed in 1:1000 fluorescent secondary antibody (Moesin, CD44 and sFRP-1, StarBright blue 700 goat anti-rabbit IgG, Bio-rad #12004161; Midkine, IRDye 800CW Donkey anti-Goat IgG, Licor, Lincoln, NE, USA) and 1:2000 hFAB™ Rhodamine Anti-GAPDH Antibody (Bio-rad #12004167) for 1 h at room temperature. Membranes were then washed (2 × 5 min in 1X TBST; 2 × 5 min in 1X TBS) then scanned with a Bio-rad ChemiDoc MP Imaging System. Pixel intensity was quantified using ImageJ, the public domain NIH Image program (available at http://rsb.info.nih.gov/nih-image/). Statistical analyses and graphing were performed using an unpaired T-test GraphPad Prism Version 9.3.1 (San Diego, CA).

Selection of antibodies

We have characterized most commercially available antibodies against Moesin (Alshafie et al. 2023), Midkine (Ayoubi et al. 2023a), sFRP-1 (Ayoubi et al. 2023b) and CD44 (Ayoubi et al. 2021) using human WT and knockout cell lines for three research applications, including Western blot. For this study, we selected antibodies based on the following criteria: i) high antibody selectivity by WB based on our previous antibody characterization studies, ii) predicted reactivity with mouse based on the internal validation from the manufacturer, iii) band at the expected size by Western blot using mouse brain lysate (Figure 2) and iv) antibody clonality as we prioritize renewable reagents such as mouse monoclonal or rabbit recombinant antibodies. We selected the recombinant anti-Moesin ab52490, the recombinant anti-CD44 ab189524 and the recombinant anti-sFRP-1 ab267466 antibodies as they met our criteria. However, none of the previously characterized Midkine antibodies was expected to react with mouse Midkine. By searching on antibody manufacturers’ website, we identified the polyclonal antibody AF7769 from Bio-techne generated to specifically detect mouse Midkine. We validated AF7769 as we have done previously (Ayoubi et al. 2023a) (Figure 1). AF7769 met our selection criteria and was further selected for the downstream experiments. The underlying data for this study has been uploaded to Zenodo (Doolen & Rizzo 2023a).

Figure 1. Midkine antibody screening by Western Blot on culture media.

Lysates of HAP1 (WT and MDK KO) were prepared, and ~30 μg of protein was processed for Western blot with the Midkine antibody. The Ponceau stained transfers of each blot are presented to show equal loading of WT and KO media and protein transfer efficiency from the acrylamide gels to the nitrocellulose membrane.

Protein expression is increased in mouse models of AD

We next used the selected antibodies on protein extracts from cortex of male and female 9-month aged 5xFAD mouse brain homogenates and compared to age-matched C57BL/6J controls. Our immunoblotting results depicting their reactivity in mouse brain homogenate are shown in Figure 2. The antibody source information and the fold change in protein levels in 5xFAD are in Table 1. Anti-Moesin ab52490 reacted in mouse brain homogenate with a predicted molecular weight of 68 kDa and its expression was 2.8 times higher in 5xFAD compared to WT mouse brain (Figure 2A). Anti-CD44 ab189524 reacted with a band at the predicted size of 82 kDa and its expression was 1.9 times higher in 5xFAD compared to WT mouse brain (Figure 2B). Anti-Midkine AF7769 reacted with a band ~16 kDa and a 17.8 times greater expression in 5xFAD compared to WT mouse brain (Figure 2C). Anti-sFRP-1 ab267466 reacted with a band at 35 kDa as predicted and its expression was 11.9 times greater in 5xFAD compared to WT mouse brain (Figure 2D). Interestingly, both Midkine and sFRP-1 are secreted proteins and could still be identified in crude protein extracts from homogenates, prepared as described above. For analysis, an unpaired T-test was performed using Graphpad Prism Version 9.3.1 for Windows (www.graphpad.com).

Figure 2. Analysis of Moesin, CD44, Midkine and sFRP-1 protein level in 5XFAD AD mouse model.

Representative Western blots (left) and quantification (right). (A) Anti-Moesin; (Abcam #ab52490) reacted in mouse brain homogenate with a predicted molecular weight of 68 kDa and 2.8 times greater expression in 5XFAD relative to WT. (B) Anti-CD44 (Abcam #ab189524) reacted with a band at the predicted size of 82 kDa and 1.9 times greater expression in 5XFAD relative to WT. (C) Anti-Midkine (R&D Systems #AF7769) reacted with a band ~16 kDa with 17.8 times greater expression in 5XFAD relative to WT. (D) Anti-sFRP-1 (Abcam #ab267466) reacted with a band at 35 kDa as predicted with 11.9 times greater expression in 5xFAD relative to WT. Brightness and contrast were adjusted in figure for visualization purposes and were applied equally to entire blot. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 by unpaired t-test.