Validation of a yeast malate dehydrogenase 2 (Mdh2) antibody tested for use in western blots

Malate dehydrogenases (Mdhs) reversibly convert malate to oxaloacetate and serve as important enzymes in several metabolic pathways. In the yeast Saccharomyces cerevisiae there are three Mdh isozymes, localized to different compartments in the cell. In order to identify specifically the Mdh2 isozyme, GenScript USA produced three different antibodies that we further tested by western blot. All three antibodies recognized the S. cerevisiae Mdh2 with different background and specificity properties. One of the antibodies had a relatively low background and high specificity and thus can be used for specific identification of Mdh2 in various experimental settings.


Introduction
Malate dehydrogenases (Mdhs) catalyse the interconversion of malate and oxaloacetate using NAD + or NADH 1 . In the yeast Saccharomyces cerevisiae there are three known isozymes: Mdh1 is located in mitochondria, Mdh2 is mostly cytosolic, and Mdh3 is localized to peroxisomes. The residues of the three isozymes are between 43-50% identical 2 . It is thus very important for purposes of isolation and identification to utilize specific antibodies that will recognize only one specific isozyme. Using three Mdh2 peptides, which were specifically designed to unique regions in the Mdh2 protein, GenScript USA produced three different antibodies that should have high specificity for S. cerevisiae Mdh2 relative to Mdh1 and Mdh3. We then tested all three antibodies by western blotting and found one with specific binding. Due to its specificity, this antibody has the potential to also work in other experimental assays such as immunoprecipitation, immunohistochemistry and ELISA (Enzyme-Linked Immunosorbent Assay).
All three antibodies were raised in New Zealand Rabbit and are polyclonal. The immunogen for all three antibodies is conjugated KLH Peptide. To find specific Mdh2 peptides, the three S. cerevisiae malate dehydrogenases (Mdh1, Mdh2 and Mdh3) were aligned and analysed by GenScript, using their Antigen Design Tool. From this analysis, three peptides (corresponding to the three antibodies) were selected: #1: CHPQSRNSGIER-RIM; #2: CINIESGLTPRVNSM; #3: MPHSVTPSIEQDSLC. The cysteines in the N' terminus (peptides number 1 and 2) or C' terminus (peptide number 3) were added for KLH conjugation.

Yeast strains and strain construction
Yeast strains are all based on the BY4741 laboratory strain 3 . Manipulations were performed using a standard PEG/LiAC transformation protocol 4 . The GFP-Mdh1, GFP-Mdh2, GFP-Mdh3 and OE-mCherry-Mdh3 strains were picked from the SWAT N' GFP or N' mCherry yeast libraries that were recently prepared in our lab 5 . All Strains with a fluorescent tag, including the strains that were picked from the SWAT libraries were verified using PCR (one primer from within the endogenous Open Reading Frame (ORF) and one from within the tag sequence) as well as by fluorescent microscopy. Primers for creating strains with deletions or C' tagging of genes (Δmdh1, Δmdh2, Δmdh3 and Mdh2-mCherry) were designed using the Primers-4-Yeast web tool 6 . All deletions were verified using primers from within the endogenous ORF. All primers are summarized in Table 1.

Western blot analysis
Yeast proteins were extracted by the NaOH protocol as previously described 7 and resolved on polyacrylamide gels with the following modifications: 1. Liquid cultures were grown over night at 30°C in glucose containing media, and then were diluted and incubated for an additional 4-6 hours to get to mid-log phase. Cells that were grown in a different carbon source than glucose were grown in synthetic media with 0.1% glucose (6.7 g/l yeast nitrogen base and 0.1% glucose) for the overnight incubation, then diluted in synthetic media containing galactose (6.7 g/l yeast nitrogen base and 2% galactose) or oleic acid (6.7 g/l yeast nitrogen base, 0.2% oleic acid and 0.1% Tween-80) and incubated for an additional 4-6 hours. Cells that were grown in glucose were grown in yeast extract peptone dextrose (YPD) rich medium (1% Yeast Extract, 2% peptone, 2% glucose) for the whole process.
2. The extractions were incubated for 10 minutes at 70°C and 15 µl supernatant was loaded per lane of the gel.
The gels were transferred to nitrocellulose membrane blots, blocked for 1 hour in SEA block diluted in TBS-T (1:5) at room temperature (RT), and probed with primary rabbit/mouse antibodies against Mdh2, Histone H3, Actin and mCherry ( Table 2) for 1 hour at RT. Final concentrations of the primary antibodies were: 1 µg/ml for anti-Mdh2 peptides #1 and #3, 0.5 µg/ml for anti-Mdh2 peptide #2, anti-mCherry and anti-Actin and 0.2 µg/ml for anti-Histone H3. The membranes were washed 3 times without incubation in TBS-T, followed by 3x3 minute washes in TBS-T. The membranes were then probed with a secondary goat anti-rabbit/mouse antibody conjugated to IRDye800 or to IRDye680 (Table 2) for 30 minutes at RT and washed as before. Final concentrations of the secondary antibodies were 0.1 µg/ml for IRDye 800CW Goat anti-Rabbit IgG and IRDye 680RD Goat anti-mouse IgG, and 0.2 µg/ml for Goat antimouse HRP conjugated. Membranes were scanned for infrared signal using the Odyssey Imaging System (Li-Cor). For detecting the anti-Actin primary antibody (Table 2), a goat anti-mouse HRP conjugated secondary antibody ( Table 2) was used. The membrane was then washed with ECL reagents and scanned using imageQuant LAS 4000 system (GE Healthcare). Images were transferred to ImageJ 1.51s for slight adjustments of contrast and brightness. For further information about the western blot reagents see Table 3.

Controls
Several controls were used in this study. Loading controls for the total amount of protein were performed using anti-Histone H3 or anti-Actin antibodies. A Δmdh2 strain, alongside

Amendments from Version 1
Raw images of all western blots were added in a PDF format as well as TIFFs, to enable all readers to view the raw data files. The data itself is unchanged. Δmdh1 and Δmdh3 strains, were used to verify the specificity of the Mdh2 antibodies. An anti-mCherry antibody was used as a control for the bands detected in Mdh2-mCherry or Tef2-mCherry-Mdh3 (over expression) compared to the bands detected by the anti-Mdh2 antibodies. An anti-GFP antibody was used as a control for the bands detected in GFP-Mdh1, GFP-Mdh2 and GFP-Mdh3 with anti Mdh2 antibody peptide #3.

Results
The three GenScript antibodies detect Mdh2 from cells grown on oleic acid For first examination of the three antibodies, we grew control cells in three different carbon sources: glucose, galactose and oleic acid. Since transcription of S. cerevisiae Mdh2 is repressed by glucose 8 , we hypothesized that the antibodies will not detect Mdh2 in extracts of cells grown in glucose, but will detect it in the galactose and oleic acid samples. We then performed protein extraction using the NaOH approach, and examined the three GenScript anti-Mdh2 antibodies.
Indeed, all three antibodies could not detect a specific band at the correct size of Mdh2 (41 KDa) in protein extractions from cells grown in glucose (Figure 1, Glu). Although the general protein levels of the control strain grown in oleic acid were much lower than from the levels of controls grown in glucose, a prominent band at the correct size for Mdh2 was detected by all three antibodies. In galactose there was no prominent band at the correct size, although this could be due to the general low levels of protein in this condition, as can be seen by the Histone H3 loading control. Endogenous Mdh2 was also recognized in protein extractions from cells grown on other carbon sources such as EtOH (not shown). This suggests  Histone H3 was used as a loading control. In total we detected seven times a band at the size of Mdh2 when cells were grown on Oleate: One time using antibody1, three times using antibody 2 and three times using antibody 3. that the upregulation of Mdh2 in cells grown in oleic acid is due to the removal of the glucose repression and not because of an oleic acid-specific up-regulation.

Antibody 3 has high specificity for Mdh2
After the first comparison, we focused on antibodies 2 and 3 as they displayed the lowest background signal. We verified the affinity and specificity of the antibodies using several strains, in addition to the control strain: Δmdh1, Δmdh2, Δmdh3, as well as strains that had a mCherry tag fused to either Mdh2 or Mdh3 or GFP tag fused to Mdh1, Mdh2 or Mdh3. Both antibody 2 and antibody 3 detected endogenous Mdh2 in wild type cells and in Δmdh3 cells grown in oleic acid (Figure 2 and Figure 3), but were specific to Mdh2 as they did not cross react in the Δmdh2 strain. Another way to demonstrate specificity is to tag Mdh2 with mCherry, thus shifting only this specific isoform in size. Indeed,

Ab1
Ab2 Ab3 Histone H3 under these conditions both antibodies 2 and 3 detected a higher protein form only (Mdh2 tagged with mCherry = ~70 kDa), though with lower affinity. Reciprocally, neither antibody detected over expressed mCherry-Mdh3 (expressed under a Tef2 promotor), although it was highly expressed as could be verified by an anti-mCherry antibody. In addition, antibody 3 detected endogenous Mdh2 in Δmdh1 cells grown in oleic acid, as well as a bigger band size only in GFP-Mdh2 tagged cells (Mdh2 tagged with GFP = ~68 kDa) (Figure 4). Although both antibodies fit the requirement of specific identification of the yeast Mdh2, Figure 3. Antibody 3 has the best specificity to Mdh2. Western blot analysis was done on protein extractions from different strains grown on oleic acid (Ole) or glucose (Glu) (OE-mCherry-Mdh3 was grown in glucose) using antibody 3 and anti-mCherry. Similarly to antibody 2, antibody 3 does not recognize Mdh2 in Δmdh2 strain or in cells grown on glucose. This antibody has a good specificity to Mdh2, as seen by the low background signal. Actin was used as a loading control. Yellow arrows indicate bands corresponding to Mdh2. (OE = over expression). We saw that antibody 3 specifically recognizes Mdh2 in three experiments.  we have decided to use antibody 3, as it has a better specificity as seen from the lower background signal (Figure 3). As an antibody validation study the manuscript presented is complete, sufficient controls are included demonstrating the specificity of the antibodies for the Mdh2-protein with Ab3 sticking out as the best one because of low background. In figures 3 and 4 an unspecific band at the same size as the fusion proteins can be seen in many lanes, which is unfortunate raising doubts whether the Mdh2-mCherry or the GFP-Mdh2 are really specifically recognized. On the other hand, the native Mdh2-band at 41kD is missing in these lanes as expected increasing confidence into the specificity of the antibodies.

Dataset 1. Raw images of all western blots included in figures
Since Mdh2p is the isoform involved in the glyoxylate cycle it can be expected that this protein will be well expressed under conditions requiring this metabolic pathway, e.g. on ethanol, acetate or oleate. From the data presented in this manuscript it is not possible to conclude that "the upregulation of Mdh2 in cells grown on oleic acid is due to removal of the glucose repression and not because of an oleic acid-specific up-regulation". For such a statement one would need to compare and present quantitative expression data from cells grown on non-repressing carbon sources such as ethanol and on oleate. It may well be that the expression of Mdh2 is derepressed on ethanol and further induced when grown on oleate. Since this is neither essential nor the primary subject of the antibody validation study I would suggest removing this sentence from the final version of the manuscript.
I did have troubles in assessing the original data of the Westernblots, but this might be due to the old version of photoshop I'm using on my Computer.

Are sufficient details of materials, methods and analysis provided to allow replication by others? Yes
Are all the source data underlying the results available to ensure full reproducibility? Partly

Are the conclusions drawn adequately supported by the results? Yes
No competing interests were disclosed.

Competing Interests:
Referee Expertise: Organelle biogenesis, glyoxylate cycle, yeast I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
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