Comparative Proteomic Analysis of Whole Kidney, Medulla, and Cortical Tubules in Diabetic Pathogenesis of Kidney Injury in Mice

We're interested in defining the molecular changes in the kidney using mass spectrometry. So we're particularly interested in defining those changes when they're related to disease processes such as diabetes or diabetic kidney disease.

Defining disease-associated proteomic changes in specific kidney regions and structures in cells is challenging when analyzing whole kidney extracts where medium and low abundant proteins are masked by high abundant proteins.

[Timothy] We're interested in deep tissue characterization using proteomics in mass spectrometry and defining the molecular events that change in the kidney, and in the glomerulus particularly, in response to disease.

Our proteomic approaches allow proteomic analysis of separated kidney regions and structures without requirements for dissecting microscopes or advanced hands-on dissection skills or specific sample protein extraction protocols.

The next logical progression in our research program will be to define the cellular events that occur in these tissues. So we will utilize a single cell proteomics to define effects on the different cell components in the kidney, in the glomerulus particularly, to determine what effects happen to cells in response to these disease contexts.

[Narrator] To begin, dissect out the kidneys from the anesthetized mouse and place the kidneys on a glass plate or Petri dish set on ice. Using a razor blade, make several transverse cuts through the kidneys to obtain three to four slices between the superior and inferior poles. Reserve one to two middle slices and the two poles from each kidney to serve as whole kidney samples. Place a portion into a 1.5 milliliter centrifuge tube containing ice cold homogenization buffer. Keeping the middle transverse slices flat, carefully slice away the cortex using a razor blade or scalpel. Keep the cortex separated from the medulla regions for the cortical tubule isolation protocol. Place a portion of the dissected medulla into a 1.5 milliliter centrifuge tube placed on ice and add ice cold homogenization buffer. Using a razor blade, mince the dissected cortex into one to three millimeter pieces on a glass plate or Petri dish forming a paste. Digest the minced cortex in one milliliter of type IA collagenase for 30 minutes at 37 degrees Celsius in a rocking water bath. Now place a 100-micrometer cell strainer on top of a 50-milliliter conical tube set on ice. Transfer the digested cortical suspension onto the 100-micrometer cell strainer, and gently press through the strainer using the plunger of a 10-milliliter syringe. Wash the top of the strainer with one milliliter of PBS. Pass the filtrate through a 100-micrometer strainer followed by a 70-micrometer strainer. Spin the final filtrate at 120 g for two minutes at four degrees Celsius. After discarding excess PBS, check the purity of the cortical tubule pellet under a microscope using a 10x objective. If more than 10% of the fraction contains glomeruli, resuspend the pellet with one milliliter of PBS and pass through a clean 70-micrometer cell strainer without additional washing. Now homogenize the samples in 1.5 milliliter microcentrifuge tubes using a plastic pestle. After 15 minutes, sonicate the sample for five minutes in a bath sonicator with water at 25 degrees Celsius. Now leave the samples on ice for another 10 minutes. Briefly vortex the homogenate and tritrate 20 times with a pipette. Leave the sample on ice for another 15 minutes. After tritrating the sample several times, centrifuge at 13,000 g for 20 minutes at four degrees Celsius and transfer the cleared protein extracts to clean tubes. Dilute approximately 40 to 50 micrograms of sample into sample lysis buffer to a final volume of 46 microliters and incubate the sample at 65 degrees Celsius for 30 minutes. Then add four microliters of 0.5 molar iodoacetamide in LC/MS-grade water and incubate at room temperature in the dark for 30 minutes. Now, add five microliters of 12% weight-by-weight phosphoric acid in LC/MS-grade water followed by 350 microliters of suspension trap binding buffer. Add the sample to the suspension trap column and centrifuge at 4,000 g for 30 seconds at room temperature in a fixed angle rotor to pass each volume through the column. Wash the column four times with 400 microliter volumes of suspension trap binding buffer. And centrifuge at 4,000 g for 30 seconds at room temperature in a fixed rotor after each wash. Then centrifuge the column at 4,000 g for one minute at room temperature to completely remove binding buffer. Now transfer the suspension trap column to a clean collection tube and add 2.5 to 5 micrograms of trypsin in 125 microliters of 50 millimolar TEABC at pH 8.5 in LC/MS-grade water. Inject an equal mass of peptides onto one-dimensional nano LC and fractionate on reversed phase C18 columns. Elute the peptides directly into the mass spectrometer at a spray voltage of 1.8 kilovolts with the ion transfer tube maintained at 250 degrees Celsius. Acquire spectra in data-dependent ion mode where the most intense tandem mass spectrometry fragment is removed from the analysis queue. Following spectral acquisition, submit raw files for spectral assignment and protein and peptide identification. Use LC-MS/MS peak's data analysis software to search and filter spectra using strict 1% false discovery rate criteria against the mouse-reviewed FASTA database. Specify carbamidomethylation of cysteine as a fixed modification and oxidation of methionine as a variable modification and submit to peak software for spectral assignment and peptide and protein identification. Convert protein lists to CSV format and assign group or sample labels to each column and row. Open the MetaboAnalyst software. Submit the CSV file and impute missing values using the 1/5 minimum value rule for missing variables. Filter data to exclude highly variable values using the interquartile range applying a 40% variance cutoff. Normalize values to median intensity, log 10 transform the data and mean center for scaling. Generate volcano plots for fold change of 1.5 and p-values below 0.05. Generate heat maps and perform partial least squares discriminate analysis, or PLS-DA, and statistical analysis of the sample groups using variable importance in projection. Identify potential follow-up candidates based on criteria of fold change exceeding 1.5 and p-value less than 0.05, or variable importance in projection greater than one in PLS-DA analysis. Global proteomic changes in the OVE26 diabetic mouse kidney were assessed. Significant upregulation of GAPDH and ALDH6A1 indicated metabolic responses, while MME and CTSA were robustly downregulated. Notably, many of the protein VIPs were found in the volcano plot analyses. Hierarchical heat maps with group and feature clustering further illustrated differences in proteomes at the group and individual protein levels between OVE26 and FVB kidneys. Cortical tubule proteomics analysis identified Txn2 and Ppef2 as significantly upregulated, while Lrba and Nhef2 were downregulated in OVE26 mice. Medullary proteomics analysis showed upregulation of Phf8 and Skp1 and downregulation of Rhoa and Prx in OVE26 kidneys. A Venn diagram comparison of proteomes from whole kidney, tubules and medulla demonstrated minimal overlap between these data sets with either zero or two shared candidates using statistical thresholds.