A guide to selecting high-performing antibodies for human Midkine for use in Western blot and immunoprecipitation

Midkine is a secreted protein that acts as a growth factor or cytokine involved in cell survival and inflammatory processes. It accumulates in amyloid plaques, which are hallmarks of Alzheimer’s Disease (AD). The reproducibility of Midkine research would be enhanced if the community had access to well-characterized anti-Midkine antibodies. In this study, we characterized 8 commercial Midkine antibodies for Western blot and immunoprecipitation, using a standardized experimental protocol based on comparing read-outs in a knockout cell line and isogenic parental control. These studies are part of a larger, collaborative initiative seeking to address the antibody reproducibility issue by characterizing commercially available antibodies for human proteins and publishing the results openly as a resource for the scientific community. While use of antibodies and protocols vary between laboratories, we encourage readers to use this report as a guide to select the most appropriate antibodies for their specific needs.


Introduction
The neurotrophic and developmental factor Midkine is a secreted heparin-binding cytokine. 1[4] Midkine is involved in numerous processes that promote cell growth, and may contribute to the pathogenesis of inflammatory diseases. 1,46][7] Mechanistic studies would be greatly facilitated by the availability of high-quality antibodies for Midkine.][10] Here, we compared the performance of a range of commercially available antibodies for Midkine use in Western Blot and immunoprecipitation.This article serves as a valuable guide to help researchers select high-quality antibodies for their specific needs, facilitating the biochemical and cellular assessment of Midkine properties and function.

Results and discussion
2][13] The first step is to identify a human cell line(s) that expresses sufficient levels of a given protein to generate a measurable signal.To this end, we examined the DepMap transcriptomics database to identify all cell lines that express the target at levels greater than 2.5 log2 (transcripts per million "TPM"+1), which we have found to be a suitable cut-off (Cancer Dependency Map Portal, RRID: SCR_017655).Commercially available human HAP1 cells expressed the Midkine transcript at RNA levels above the average range of cancer cells analyzed. 14Parental and MDK knockout human HAP1 cells were obtained from Horizon Discovery (Table 1).
Midkine is predicted to be a secreted protein.Accordingly, we collected concentrated culture media from both wild-type and MDK KO cells and used the conditioned media to probe the performance of the antibodies (Table 2) side-by-side by Western blot and immunoprecipitation.The profiles of the tested antibodies are shown in Figures 1 and 2.
In conclusion, we screened eight Midkine commercial antibodies by Western blot and immunoprecipitation.Several high-quality antibodies that successfully detect Midkine under our standardized experimental conditions can be identified.In our effort to address the antibody reliability and reproducibility challenges in scientific research, the authors recommend the antibodies that demonstrated to be underperforming under our standard procedure be removed from the commercial antibody market.However, the authors do not engage in result analysis or offer explicit antibody recommendations.A limitation of this study is the use of universal protocols -any conclusions remain relevant within the confines of the experimental setup and cell line used in this study.Our primary aim is to deliver top-tier data to the scientific community, grounded in Open Science principles.This empowers experts to interpret the characterization data independently, enabling them to make informed choices regarding the most suitable antibodies for their specific experimental needs.REVISED Amendments from Version 3 In our efforts to ensure reproducibility, version 4 includes amendments to the Methods and Figure 1 legend to provide detailed information on how the HAP1 WT and MDK KO cells were prepared prior to Western blot evaluation.Antibodies GTX 116089 and NBP2-66948 were removed from the commercial antibody market and are no longer available.Antibody MA5-32538 has the same clone ID as NBP2-66948, and is still available on the market as an alternative.
Antibody screening by Western blot on culture medium HAP1 cells (WT and MDK KO) were grown in 20 ml complete medium to 80% confluence in 150 mm dishes.At 80% confluency, HAP1 cells were at a density of ~1.0 Â 10 6 /ml (~20.0Â 10 6 /20 ml).To prepare for protein extraction, cells were washed 3Â with PBS and starved for ~18 hrs.Culture media were collected and centrifuged for 10 min at 500 Â g to eliminate cells and larger contaminants, then for 10 min at 4500 Â 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 Â g for 15 min.The resulting 500 μl of concentrated media from each filter unit were centrifuged again at 4000 Â g for 15 min using Amicon Ultra-0.5 Centrifugal Filter Units (MilliporeSigma cat.number UFC5010) to 200 μl.The final protein concentration was 0.3 mg/ml and 100 μl (30 μg) of media were used for each blot.
Western blots were performed as described in our standard operating procedure. 13,15A Bradford reagent assay was performed to determine the protein concentration of the culture medium.The concentration was then adjusted to ensure equal amounts of protein were loaded in each gel.Western blots were performed with large 10-20% gradient polyacrylamide gels, loaded with 30 μg of protein in each lane and then transferred on nitrocellulose membranes.Proteins on the blots were visualized with Ponceau staining which is scanned to show together with individual Western blot.Blots were blocked with 5% milk for 1 hr, and antibodies were 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 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.Membranes were incubated with ECL from Pierce (cat.number 32106) prior to detection with HyBlot CL autoradiography films from Denville (cat.number 1159T41).
Antibody screening by immunoprecipitation on culture medium Immunoprecipitation was performed as described in our standard operating procedure. 13,16Antibody-bead conjugates were prepared by adding 1.0 μg of antibody to 500 μl of Pierce IP Lysis from Thermo Fisher Scientific (cat.number 87788) in a microcentrifuge tube, together with 30 μl of Dynabeads protein A -(for rabbit antibodies) or protein G -(for mouse and goat antibodies) from Thermo Fisher Scientific (cat.number 10002D and 10004D, respectively).Tubes were rocked overnight at 4°C followed by two washes with the Pierce IP Buffer to remove unbound antibodies.Pierce IP Lysis Buffer (25 mM Tris-HCl pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% NP-40 and 5% glycerol) was supplemented with 1Â Halt Protease and Phosphatase Inhibitor Cocktail from Thermo Fisher Scientific (cat.number 78446).
Starved HAP1 WT culture media were concentrated as described above.The concentrated culture media were diluted in Pierce IP Lysis Buffer, and 1ml aliquots at 0.3 mg/ml of lysate were incubated with an antibody-bead conjugate for ~2 hrs at 4°C.The unbound fractions were collected, and beads were subsequently washed three times with 1.0 ml of IP Buffer and processed for SDS-PAGE and immunoblot on 10-20% polyacrylamide gels.Prot-A: HRP was used as a secondary detection system (MilliporeSigma, cat.number P8651) at a dilution of 0.4 μg/ml for an experiment where a rabbit antibody was used for both immunoprecipitation and its corresponding Western Blot.The results of the publication are very helpful for the selection of suitable Midkine antibodies.
However, there are ambiguities in the experimental and methodological part: The protocol mentions that the cell culture supernatants were obtained from starving cultures and then concentrated (also shown in scheme A in Figure 1.
For the reproducibility of the preparation of these supernatants, there is no information on the cell density/ml of the cultivated HAP1 cells or the initial volume of the cell culture supernatant concentrated for analysis.
Furthermore, the description of Figure 1 B refers to cell lysates: " Lysates of HAP1 (WT and MDK KO) were prepared, and ~30 μg of protein was processed for Western blot."This is not explained in the experimental section.This should definitely be explained or corrected.

Is the rationale for creating the dataset(s) clearly described? Yes
Are the protocols appropriate and is the work technically sound?Partly

Are the datasets clearly presented in a useable and accessible format? Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: proteinchemisty I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.
research of this target associated to many pathological conditions.We hope our response to your comments (italicized) successfully answers any concerns you previously had.

Reviewer Comment:
However, there are ambiguities in the experimental and methodological part: The protocol mentions that the cell culture supernatants were obtained from starving cultures and then concentrated (also shown in scheme A in Figure 1.
For the reproducibility of the preparation of these supernatants, there is no information on the cell density/ml of the cultivated HAP1 cells or the initial volume of the cell culture supernatant concentrated for analysis.
cells and larger contaminants, then for 10 min at 4500 x 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 x g for 15min.The resulting 500 μl of concentrated media from each filter unit were centrifuged again at 4000 x g for 15 min using Amicon Ultra-0.5 Centrifugal Filter Units (MilliporeSigma cat.number UFC5010) to 200 μl.The final protein concentration was 0.3 mg/ml and 100 μl (30 μg) of media were used for each blot.
An updated version of the manuscript has been submitted and we have included this protocol for cell preparation in the Methods section.
This work reveals that among the antibodies tested, three performed well for Western blotting (WB), and while all showed positive signals for immunoprecipitation (IP), six completely depleted the supernatant of the culture from Midkine protein.Two antibodies performed well in both experiments.Although the authors do not provide a detailed commentary and interpretation of their results, the data presentation is clear enough for readers to identify potentially suitable antibodies for WB and IP studies.

Specific comments:
It would have been valuable if the authors had conducted a similar comparison for Immunofluorescence using WT and KO HAP1 cells.

1.
For greater clarity, the authors should specify in the title that their study pertains to human Midkine and mention that HAP1 is a human cell line.

2.
The term "high performing" used in the title may be misleading since the study primarily assesses antibody specificity rather than overall performance.If the authors want to use 3.
"High-performance" in the title, they should consider providing some interpretation of their data in the discussion.
In the introduction, the authors mentioned that Midkine (MK) is a potential biomarker and therapeutic target for Alzheimer's disease (AD).Two of the antibodies featured in this study, NBP2-66948 and GTX116089, are discontinued.The authors should mention this.

5.
Have the authors optimized the working concentration used?If so, they should explain the optimization process.

6.
The authors should also provide a rationale for selecting the antibodies tested.Many of them are not commonly cited in the literature.An explanation for their choice of antibodies would enhance the manuscript's comprehensibility.

7.
Finally, could the authors clarify why it is necessary to deplete the cells?Is it to promote Midkine secretion or to avoid interference from IgG in subsequent experiments?Further clarification on this point would be beneficial.

Are the datasets clearly presented in a useable and accessible format? Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Neuroscience, antibody generation and use.
I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.
now indicated with a superscript a.

6.
Have the authors optimized the working concentration used?If so, they should explain the optimization process.This study for Midkine is part of a much larger collaborative initiative, seeking to characterize antibodies for all human proteins to address the antibody liability crisis.It would be very difficult to optimize every working parameter when in the process of trying to characterize antibodies for 20,000 proteins in the human proteome.That being said, we have developed universal protocols that has been successful for characterizing antibodies for 85 proteins thus far.In our experiments, a major parameter that we can control to optimize the process is the level of endogeneous protein in the selected cell type.We try to cell lines, using DepMap portal, with adequate transcript levels of the protein target.
7.The authors should also provide a rationale for selecting the antibodies tested.Many of them are not commonly cited in the literature.An explanation for their choice of antibodies would enhance the manuscript's comprehensibility.The antibodies to be characterization under our standard procedure are donated by YCharOS partners (https://ycharos.com/partners/).These industry partners cover nearly 28% of all antibodies listed on the Antibody registry (antibodyregistry.org).When the industry partners are informed of upcoming targets, they donate the antibodies they have the capacity to supply at the time the study commences.9.Some references (10-13) are not complete.
The DOI for citations pertaining to our prior work, available on Zenodo, have been included in the new version submitted.While these references are not found on PubMed, and may seem "incomplete", they are crucial in helping readers comprehend the protocols used and access the underlying data.
Competing Interests: No competing interests were disclosed.
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias • You can publish traditional articles, null/negative results, case reports, data notes and more • The peer review process is transparent and collaborative • Your article is indexed in PubMed after passing peer review • Dedicated customer support at every stage • For pre-submission enquiries, contact research@f1000.com blot; IF = immunofluorescence; IP = immunoprecipitation.*Monoclonal antibody.**Recombinant antibody.a Antibodies that have been discontinued subsequent to the release of this study.

Figure 1 .
Figure 1.Midkine antibody screening by Western blot on culture media.A) Schematic of the protocol used to concentrate the culture media.B) Concentrated media of HAP1 (WT and MDK KO) were prepared, and ~30 μg of protein was processed for Western blot with the indicated Midkine antibodies.The Ponceau stained transfers of each blot are presented to show equal loading of WT and KO lysates and protein transfer efficiency from the acrylamide gels to the nitrocellulose membrane.Antibody dilutions were chosen according to the recommendations of the antibody supplier.An exceptions was given to antibody GTX116089, which was titrated to 1/1000, as the signal was too weak when following the supplier's recommendations.When suppliers did not recommend dilutions, we tested the antibodies at 1/200.Antibody dilution used: GTX108439 at 1/1000; GTX116089 at 1/100; AF-258-PB at 1/100; MAB2582** at 1/200; MAB2583* at 1/200; NBP2-66948** at 1/500; MA5-32538** at 1/500; ab52637** at 1/500.Predicted band size: 15 kDa.*Monoclonal antibody, **Recombinant antibody.

8 .
Finally, could the authors clarify why it is necessary to deplete the cells?Is it to promote Midkine secretion or to avoid interference from IgG in subsequent experiments?Further clarification on this point would be beneficial.To help us provide a complete response to this comment, could you specify what you mean by "deplete the cells"?

Table 1 .
Summary of the cell lines used.

Table 2 .
Summary of the Midkine antibodies tested.