Proteomic analysis of protein turnover by metabolic whole rodent pulse-chase isotopic labeling and shotgun mass spectrometry analysis

SUMMARY

The analysis of protein half-life and degradation dynamics has proven critically important to our understanding of a broad and diverse set of biological conditions ranging from cancer to neurodegeneration. Historically these protein turn over measures have been performed in cells by monitoring protein levels after “pulse” labeling of newly synthesized proteins and subsequent chase periods. Comparing the level of labeled protein remaining as a function of time to the initial level reveals the protein’s half-life. In this method we provide a detailed description of the work-flow required for the determination of protein turn-over rates on a whole proteome scale in vivo.

Our approach starts with the metabolic labeling of whole rodents by restricting all the nitrogen in their diet to exclusively nitrogen-15 in the form of spirulina algae. After near complete organismal labeling with nitrogen-15 the rodents are then switched to a normal nitrogen-14 rich diet for time periods of days to years. Tissues are harvested, the extracts are fractionated, and the proteins are digested to peptides. Peptides are separated by multi-dimensional liquid chromatography and analyzed by high resolution orbitrap mass spectrometry (MS). The nitrogen-15 containing proteins are then identified and measured by the bioinformatic proteome analysis tools Sequest, DTASelect2, and Census. Through these protein measures, our approach reveals in vivo protein decay rates proteome-wide.

1. INTRODUCTION

To determine the rate of protein decay, new and old versions of each protein must be discernable and ideally both be measurable. Typically cells are initially “pulsed” with a traceable molecular label (such as methionine enriched with sulfer-35 atoms) which are incorporated into newly synthesized proteins and subsequently “chased” with a normal containing methionine (sulfer-32 fraction of 95.02%) (1,2). By comparing the initial amount of a specific labeled protein to that remaining as a function of a time, a measure of protein half-life can be obtained (3). These analysis can also provide key information on the possibility that different pools of the same protein exist and have dissimilar decay kinetics. Recently, proteomic technologies have been applied to gain insight into the analysis of protein turnover dynamics on a proteome-wide scale. By combining the yeast whole genome tap-tag gene library, translation inhibition with cyclohexamide, and epitope tag western blot analysis it was determined that on average the life time of a yeast protein is about 43 minutes (4). In cultured HeLa and C2C12 cells, “pulse-only” stable isotope labeling of cells in culture (SILAC) for several durations with time course mass spectrometry (MS) analysis showed average protein half-life in mouse and human cells <2 days (5). In mice, by using MS analysis to measure the rate by which isotopes are metabolically incorporated into proteins and modeling it has been suggested that on average proteins in brain tissue have a lifetime of 9.0 days, liver 3.0 days, and blood 3.5 days (6,7).

We have developed a straight-forward systematic approach to monitor protein decay dynamics on a global scale in the most relevant biological context, in vivo. Our approach has verified previously reported rapid degradation dynamics for nearly all proteins. Unexpected we also find a limited number of intracellular extremely long lived proteins (ELLPs) which reside in the nucleus and cytoplasm of post mitotic neurons (8,9). Our approach also confirmed the existence ELLPs in the myelin sheath and eye lens (10–12)). Application of our approach to proteinopathy disease mouse models (such as Alzheimer’s, Parkinson’s, and Huntington’s disease) could provide new insight on pathogenic mechanisms by identifying disease specific long lived proteins.

2. MATERIALS

Buffers and solutions for MS analysis should be prepared with analytical chromatography grade solvents and for biochemical experiments we prepared buffers with ultra-pure water (Milli-Q® Water Purification Systems, 18-megohm-cm deionized water). Solutions should be stored at room temperature unless otherwise indicated. To minimize keratin contamination gloves should be worn during the preparation of all buffers and samples.

2.1. Nitrogen-15 or Nitrogen-14 spirulina rodent chow components

1. Nitrogen-15 enriched spirulina algae: Nitrogen-15 enriched (>94%) spirulina were purchased from Cambridge Isotopes (13), Cambridge, MA, USA or can be grown and prepared in house as previously described (14–16).

2. Rodent chow: Rodent chow has been prepared by mixing nitrogen-15 or nitrogen-14 spirulina with protein-free diet mixture powder (Harlan TD 93328) in a 1 to 3 ratio. Pellets were prepared by adding ultra-pure water to the power mixture and worked the mixture into dough shaped into cylinders. Individual ~2-cm discs we cut from the cylinders and dried at 60 ºC for 2–4 hours and then 35 ºC overnight on screen trays in an Excalibur food dehydrator (17). Alternatively, nitrogen-15 spirulina containing chow can be purchased pre-prepared from CIL / Harlan Laboratories Inc. with 22% protein / 65% carbohydrate (carbon, hydrogen, oxygen as CHO), 13% fat composition [see Note 1].

2.2. Representative protein fractionation components

1. Tissue homogenization buffer: 0.32 M sucrose, 4 mM Hepes (pH 7.4), 1 mM MgCl2 and protease inhibitors (Sigma) [see Note 2]. 1 M Hepes, add 600mL water to a glass beaker, weigh and add 238.3 g of Hepes, add stir bar to dissolve on a stir plate. Determine pH and adjust with HCl or NaOH to pH 7.4 final. Transfer to a graduated cylinder and add water to 1L. Store at 4 °C. 1 M MgCl₂, add 500mL water to a glass beaker, weigh and add 203.3 g of MgCl₂ 6H2O, add stir bar to dissolve on a stir plate. Transfer to a graduated cylinder and add water to 1L. To a 250mL glass beaker, a stir bar, add 50mL of water, 5 mL of 1 M Hepes (pH 7.4), 0.1 mL of 1M MgCl₂, and 10.9 g of Sucrose. Transfer to a 100 mL graduated cylinder and add water to 100 mL (18).

2. Sucrose gradient buffers (0.85 M / 1.0 M / 1.2 M / 2.0 M): Weigh 28.9 g, 34.0 g, 40.9 g, 69.1 g and prepare 100 mL of buffer as described above except substitute the indicated amount of sucrose for each buffer.

3. 2,2,2-trichloroacetic acid (TCA) buffer: prepare a 100% (wt / vol) TCA solution with water.

2.3. Protein digestion components

1. Urea protein denaturation buffer: Dissolve 0.395 g of solid Ammonium bicarbonate in 100mL of water to prepare 50 mM (AMBC) adjust to pH 7.5 as described above; aliquot and store at −20 °C. Add 0.240 g of urea to 320 ul of AMBC buffer to prepare 8 M solution [see Note 3].

2. ProteaseMAX surfactant buffer: Dissolve solid ProteaseMAX in 500ul of AMBC to prepare 0.2% solution or 100ul for 1% [see Note 4].

3. Reduction buffer: Dissolve solid Tris (2-carboxyethyl) phosphine hydrochloride (TCEP) in AMBC to prepare 0.5 M solution [see Note 5].

4. Alkylation buffer: Dissolve solid Iodoacetamide in AMBC and prepare 1 M solution.

5. Trypsin buffer: Dissolve 20 ug vial of lyophilized trypsin (Promega) in 40 ul of buffer [see Note 6].

2.4. Liquid chromatography components

1. HPLC buffer A: Prepare HPLC buffer A with 95% water, 5% acetonitrile and 0.1% formic acid (vol / vol).

2. HPLC buffer B: Prepare HPLC buffer B with 20% water, 80% acetonitrile and 0.1% formic acid (vol / vol).

3. HPLC buffer C : Prepare HPLC buffer C with 500 mM ammonium acetate, 5% (vol / vol) acetonitrile and 0.1% (vol / vol) formic acid.

4. Make Kasil frit and prepare multidimensional protein identification (MudPIT) column by bomb packing strong cation exchange (SCX) / reversed phase resins as previously described (19–21).

2.5. Mass Spectrometer

1. Tune and calibrate electrospray high resolution orbitrap mass spectrometer (Thermo Scientific™ Orbitrap Velos Pro or Orbitrap Tribrid Fusion) per the manufacturer’s instructions with Pierce LTQ Velos ESI Positive Ion Calibration Solution [see Note 7].

2.7. Proteomic analysis software

1. IP2 (Integrated Proteomic Analysis environment is commercially available; http://integratedproteomics.com/).

2. RawExtractor (Spectra extraction tool is freely downloadable; http://fields.scripps.edu/researchtools.php).

3. Sequest / Prolucid (Protein database search algorithm is freely downloadable; http://fields.scripps.edu/researchtools.php).

4. DTASelect2 (protein dataset filtering tool is freely downloadable; http://fields.scripps.edu/researchtools.php).

5. Census (protein quantitation software is freely downloadable; http://fields.scripps.edu/researchtools.php).

3. METHODS

Perform all procedures at room temperature unless noted.

3.1. Metabolic labeling of whole rodents

1. Obtain two recently weaned female rats and allow to acclimate in the university approved animal facility for several days [see Note 8].

2. Replace standard rodent chow with nitrogen-15 containing spirulina chow and house for >10 weeks [see Note 9].

3. Introduce male breeder rat into breeding cages and monitor female rat for weight gain indicative of successful pregnancy [see Note 10].

4. Closely monitor cages for pups and document successful breeding. Continue feeding with exclusively nitrogen-15 containing spirulina chow while pups are nursing [see Note 11].

5. Once pups are weaned, feed with exclusively nitrogen-15 containing spirulina chow for additional 3–4 weeks.

6. Start chase period by switching to regular nitrogen-14 rodent chow [see Note 12].

3.2. Tissue harvest

1. Sacrifice time = 0 animal with CO₂ as the primary mechanism and secondarily by decapitation.

2. Harvest all tissues with standard dissection procedures and carefully label and freeze each tissue in a separate tube in liquid nitrogen and then store at −80 °C.

3. Sacrifice littermates at additional time points and repeat dissection and tissue harvesting as needed [see Note 13] [Fig. 1].

Fig. 1.

Metabolic pulse chase labeling of rats work flow to measure protein turnover dynamics in vivo. Freshly weaned female rat (first generation), is obtained and the diet is switched completely to nitrogen-15 containing food for 10 – 16 weeks. Male rat is introduced and female rat remains on nitrogen-15 diet while pregnant and during the nursing of her pups. Pups (second generation) are sacrificed at several time points including time = 0, before switching to nitrogen-14 chow. For the identification and analysis of extremely long live proteins we found 6 months and 12 month reliable chase durations. As a negative control we analyze an unlabeled pup after feeding regular nitrogen-14 chow. After the animals are sacrificed their tissues are dissected, proteins solubilized and then fractionated. The proteins are then digested to peptides prior to LC-MS and bioinformatic analysis.

3.3. Representative protein fractionation from brain tissue

1. Homogenize rat brain in 12 mL of tissue homogenization buffer on ice and centrifuge at 4 °C, 1500 g for 15min and the supernatant was collected (post nuclear supernatant).

2. Centrifuge supernatant at 4 °C, 18000 g for 20 min, collect the resulting supernatant (cytosol) and pellet (crude membrane).

3. Re-suspend pellet in homogenization buffer and load onto 0.85 M / 1.0 M / 1.2 M sucrose gradient and centrifuged at 4°C, 78000 g for 120 min, and collect the material focused at the 1.0 M/1.2 M interface was collected (synaptosomes).

4. Add Triton X-100 to 0.5% final concentration and extract at 4 °C, by end-over-end agitation for 20 min.

5. Centrifuge the extract at 4°C, 32000 g for 20 min, and collect the supernatant (soluble synaptosome).

6. Re-suspend pellet in homogenization buffer and load onto a 1.0 M / 1.5 M / 2.0 M sucrose gradient and centrifuged at 4°C, 170000 g for 120 min (18).

7. Collected material at the 1.5 M / 2.0 M interface (post synaptic density, PSD).

8. Add 0.5% Triton X-100 and detergent soluble material extracted at 4°C, by end-over-end agitation for 10 min.

9. Centrifuge extract at 4°C, 100000 g for 20 min and re-suspended the pellet in homogenization buffer (purified PSD).

3.4. Protein digestion and peptide preparation

• 1.To each fraction (100 ug) add TCA to 20% (vol / vol) final concentration, vortex, incubate on ice at 4 °C for 4 hours to overnight [see Note 14].
• 2.Centrifuge at 14,000 g for 45 min at 4 °C.
• 3.Discard supernatant and wash the pellet with 1 ml of ice-cold acetone.
• 4.Centrifuge the tube at 14,000 g for 10 min at 4 °C.
• 5.Remove the acetone and wash the pellet with 1 ml of ice-cold acetone (two washes in total).
• 6.Centrifuge the tube at 14,000 g for 10 min at 4 °C.
• 6.Remove supernatant and air-dry the pellet at room temperature.
• 7.Add 50 μl of urea buffer and re-suspend dry protein pellet and vortex for at least 1 h.
• 8.Add 50 μl of 0.2% (wt / vol) ProteaseMAX and vortex for at least 1h.
• 9.Add 1 μl of TCEP buffer and vortex the mixture for at least 1 additional hour.
• 10.Add 2 μl of IAA buffer, mix well and incubate in the dark for 20 min.
• 11.Squelch alkylation reaction by adding 5 μl of TCEP buffer.
• 12.Add 150 μl of AMBC and mix well [see Note 15].
• 13.Add 2.5 μl of 1% (wt / vol) proteaseMAX and briefly vortex.
• 14.Add 2 – 4 μg of sequencing-grade trypsin and incubate the mixture overnight at 37 °C with shaking.
• 15.Recover the samples and store them at −80 °C [see Note 16].

3.5. Loading the peptides on the column

1. Thaw peptides and acidify to a 5% (vol / vol) final concentration with formic acid.

2. Centrifuge the tube at 14,000 g for 15 minutes at room temperature and transfer supernatant to a new tube.

3. Directly Load peptide sample onto SCX / RP column with a bomb at a pressure of 500–1,000 p.s.i.(22) [see Note 17].

4. Wash column with buffer A for 30 minutes on bomb.

5. Pull a 15-cm tip of 100-μm glass capillary and use bomb to pack RP resin.

6. Flow buffer B for 15 minutes to wash the analytical tip.

7. Flow buffer A for 15 minutes to equilibrate the analytical tip.

3.5. Liquid chromatography / mass spectrometry

1. Connect the MudPIT column (frit connected to the analytical tip with an IDEX union) to the HPLC pump and start buffer A to ensure stable flow rates and pressure without leaks.

2. Generate eleven-step LC and MS methods with Xcaliber software.

3. Start the eleven-step LC/MS sequence with the Xcaliber software on the MS computer [see Note 18]. The analysis will be performed over a 22–24 hour time period per sample analysis.

3.6. Bioinformatic data analysis

1. Process acquired .RAW files by first extracting them to .MS1 and .MS2 format with RawExtractor software on the mass spectrometer’s PC (23).

2. Upload all files (33 total, 11 .RAW, 11 .MS1, and 11 .MS2) into the IP2 software.

3. Perform Prolucid heavy and light database search with the rat (species matched) protein database and parameters such as a fixed modification of 57.02146 on cysteine, possessing at least one tryptic terminus, and with unlimited missed cleavages (24) [see Note 19] [Fig. 2].

4. Filter and control false-discovery rate for each dataset individually with DTASelect with target-decoy strategy (concatenated forward-reverse amino acid sequence protein database) to ensure a 0–1% false discovery rate at the protein level(25).

5. To view the proteins which are identified (based on matched MS scan) only in the heavy search, run “heavy only” DTASelect analysis.

6. Perform peptide quantitation and enrichment analysis with Census software within IP2 (26–28) [see Note 20] [Fig. 3].

Fig. 2.

Bioinformatic spectral analysis paradigm. Theoretical representation of a zoomed MS1 spectral scan, starred peaks are selected for MS2 and indicate identification of both the abundant nitrogen-14 light (starred peak) and the low abundance nitrogen-15 heavy (starred peak) isotopic peaks. MS1 ion abundance are analyzed as reconstructed chromatograms based on the identification of the light or heavy peak (grey bar). To determine the peptide abundancies, the area under each curve is calculated and compared to determine the relative abundancies of the light “new” and heavy “old” peptides.

Fig. 3.

Incorporation of MS1 isotopic envelope shape measurement into protein turnover protein analysis workflow increases confidence and shows system wide protein degradation dynamics. (A) Theoretical MS1 isotopic spectral envelope after 0 and or 30 day nitrogen-14 chase periods both with heavy identifications (100% of nitrogen atoms are nitrogen-15). The corresponding “light” isotopic envelope enrichment is determined by comparing to the acquired m / z isotopic envelope shape to a broad range of predicted enrichment peak patterns to determine the percentage of nitrogen-15 atoms. (B) Binned peptide nitrogen-15 enrichment distribution from synaptosome extracts after 0, 2, 7, 30, or 180 days of nitrogen-14 chase.