Purification and properties of glyceraldehyde-3-phosphate dehydrogenase from the skeletal muscle of the hibernating ground squirrel, Ictidomys tridecemlineatus

Animals

Thirteen-lined ground squirrels (Ictidomys tridecemlineatus) were captured in the fall in Illinois and transported to an animal care facility on the National Institutes of Health campus in Bethesda, Maryland. Hibernation experiments were approved by the Animal Care and Use Committee of the National Institute of Neurological Disorders and Stroke (Permit Number ASP 1223-05) and conducted there as previously reported (Mamady & Storey, 2006). Individuals denoted as hibernators were sampled between December and February and only when animals had been in continuous torpor for at least 3 days, as evidenced by continuous low body temperature (∼6 °C) monitored by a subepidermal sensor (denoted as long torpor or LT). Animals that had not entered torpor after at least 3 days in the cold room and showed a continuous high body temperature (36–38 °C) were sampled as euthermic controls (denoted as EC). All animals were sacrificed by decapitation and tissues were quickly excised and frozen in liquid nitrogen. Tissues were then transported by air to Carleton University on dry ice where they were subsequently stored at −80 °C. Ground squirrel hind leg thigh muscle was the tissue taken for skeletal muscle studies.

Purification of GAPDH

Frozen skeletal muscle samples were homogenized 1:5 w:v in buffer A (25 mM imidazole, pH 7.0, 1.25 mM EDTA, 1.25 mM EGTA, 12.5 mM β-glycerophosphate, 5 mM β-mercaptoethanol, 5% v:v glycerol, and a few crystals of phenylmethylsulphonyl fluoride added just prior to homogenization) using a Polytron homogenizer. Homogenates were centrifuged at 13,500 g for 30 min at 5 °C, after which supernatants were decanted and held on ice until use in the first purification step.

A blue-agarose column was prepared with the dimensions of 1 × 3.5 cm (diameter × height) and equilibrated in 15 mL of buffer A. A 1.5 mL aliquot of crude muscle extract was then loaded on the column, which was then washed with 30 mL of buffer A. This was sufficient to reduce the absorbance of the fractions at 280 nm to that of the buffer alone. GAPDH was then eluted with a 0–1 M KCl gradient in buffer A and activity was assayed under V_max conditions. The top three activity fractions were pooled and applied to a phenyl-agarose column.

The phenyl-agarose column was prepared with the same dimensions as for the blue-agarose column, and equilibrated in 15 mL buffer B (buffer A containing 1 M KCl). Salt was also added to the pooled enzyme from the blue-agarose column to bring the concentration to approximately 1 M KCl, and then the enzyme was applied to the phenyl-agarose column. The column was washed with 30 mL of buffer B, which was sufficient to wash away any unbound protein. GAPDH was eluted with a linear 1–0 M KCl gradient (i.e., declining salt concentration) and assayed under V_max conditions. The three top activity fractions were pooled and held at 4 °C until use.

GAPDH assay

GAPDH was assayed spectrophotometrically at 340 nm using a Thermo Labsystems Multiskan spectrophotometer. Optimal assay conditions for the oxidation of D-glyceraldehyde-3-phosphate (GAP) were 2.5 mM GAP, 1.5 mM NAD⁺, 15 mM Na₂HAsO₄, and 50 mM NaH₂PO₄, pH 8.0. Optimal assay conditions for the reverse reaction were 2.5 mM 3-phosphoglycerate (3PG), 2 mM Mg-ATP, 0.1 mM NADH, 1 unit of phosphoglycerate kinase, and 50 mM Tris–HCl buffer, pH 8.0. Regardless of the assay direction, the enzyme reaction was initiated by the addition of 10 µL of purified GAPDH protein.

It is important to note that in assessing GAPDH activity in the 1,3-bisphosphoglycerate (1,3BPG)-reducing direction that commercial 3-phosphoglycerate kinase was used to rapidly convert 3-phosphoglycerate into 1,3BPG for use in the GAPDH reaction. Thus, kinetic parameters for 1,3BPG were measured indirectly in this study (although phosphoglycerate kinase activity was always in high excess in the assay).

Microplate Analysis (MPA) and Kinetics computer programs were used to analyze the data and determine Michaelis constants (K_m) for substrates and inhibitor concentrations that reduced enzyme activity by 50% (I₅₀) values (Brooks, 1992; Brooks, 1994). GAPDH activity is presented as units/mg soluble protein with 1 Unit is defined as the amount of enzyme necessary to produce 1 µmol of NADH or NAD⁺ per minute at room temperature. Soluble protein was measured using the Coomassie Brilliant Blue dye binding method with the Bio-Rad prepared reagent and a standard curve created using bovine serum albumin.

Effect of temperature on GAPDH activity

To assess the effect of temperature on GAPDH activity, various kinetic parameters were determined at low (5 ± 1 °C) and high (36 ± 1 °C) temperature. To establish the low temperature condition, the Thermo Labsystems Multiskan spectrophotometer and the microplates containing all solutions except for purified GAPDH (being cooled separately) were placed inside a temperature-controlled incubator set to 5 °C. To establish the high temperature conditions, the microplate chamber of the Thermo Labsystems Multiskan spectrophotometer was set to warm to 36 °C and the microplates (containing all solutions except for the purified GAPDH) were warmed on a Echo Thermomicroplate heater (Torrey Pines Scientific). It is important to note that for assays at 36 °C the enzyme was not allowed to equilibrate at this high temperature due to the lack of stability of GAPDH when heated (experimentally identified in this study). Prior to assay, the well temperature was monitored by a digital telethermometer placed inside the well. The final temperature of the well was also determined after the assay to ensure a stable reaction temperature.

The thermal stability of GAPDH was assessed in a manner similar to that described by Iddar et al. (2003). Briefly, aliquots of purified GAPDH were incubated in a water bath at a given temperature for 15 min, after which they were immediately chilled on ice and assayed under optimal conditions at 25 °C.

The effects of temperature on GAPDH were also analyzed using Arrhenius plots to determine GAPDH activation energy (E_a). The maximal velocities (V_max) of GAPDH in the 1,3BPG-reducing reaction were measured at 5 °C intervals from 5–35 °C. Assays at temperatures below or above 25 °C (room temperature) were conducted as described above for the 5 °C and 36 °C assays, respectively. Well temperature was again confirmed using a telethermometer at the end of each assay.

Differential scanning fluorimetry

Differential scanning fluorimetry (DSF) is a high throughput method that monitors the thermal unfolding of proteins in the presence of a fluorescent dye (Niesen, Berglund & Vedadi, 2007). Prior to DSF, purified EC and LT GAPDH were spun through a G50 Sephadex column, pre-equilibrated in DSF buffer (100 mM potassium phosphate, pH 7.0, 150 mM NaCl), at 2500 rpm for 1 min. Protein samples were then aliquoted to a concentration of 0.05 µg/µL/well into the wells of a 48-well, thin-walled PCR plate along with the dye SYPRO Orange (5X final concentration of the dye). PCR plates were then sealed with sealing tape and placed into a BioRad iCycler5 PCR instrument. SYPRO Orange (Invitrogen) fluorescence was monitored as described by Biggar and colleagues (2012). Briefly, SYPRO Orange was excited through the transmission of light through the FAM filter (485 ± 30 nm), with the subsequent emission of light through the ROX filter (625 ± 30 nm). Measurements were taken every 30 s in 1 °C increments from 25 °C to 97 °C. Subsequent analysis of the fluorescent data using OriginPro 8.5 and the Boltzmann distribution curve yielded the midpoint temperature of the protein-unfolding transition, known as the protein melting temperature (T_m), for GAPDH.

GAPDH digestion and mass spectrometry

Purified GAPDH samples were digested in 50 mM ammonium bicarbonate containing sequencing grade porcine trypsin at 0.013 µg/µL (Promega). Extracted peptides were dried down in a vacuum apparatus overnight followed by resuspension in 30 µL of 0.1% formic acid for use in subsequent mass spectrometry. Mass analysis was carried out on an ESI-TOF QUAD-MS/MS mass spectrometer. Samples were loaded onto a 5 µm × 50 mm column packed in-house with 5 µm C18 beads (100 Å pore size) (Michrom Bioresources, CA) at a rate of 300 µL/min using a Dinoex 3000 ultimate HPLC system (Thermo Fisher Scientific, IL). Peptides were eluted using a 0–60% (v/v) gradient of acetonitrile with 0.1% formic acid over 30 min. Mass spectra were analyzed using Mascot (Matrix Science). Mascot MS/MS ion search was conducted with the following parameters: allowance of two missed cleavages by trypsin, fixed modification of carbamidomethyl, peptide tolerance of 100 ppm, fragment mass tolerance of 100 ppm, and a peptide charge of 2+, 3+, or 4+. Peptides were only selected if the associated fragmentation patterns displayed three consecutive y or b ions.

SDS polyacrylamide gel electrophoresis, immunoblotting, and ProQ diamond phosphoprotein staining

Purified EC and LT GAPDH samples were mixed 2:1 (v:v) with SDS loading buffer (100 mM Tris buffer, pH 6.8, 4% w:v SDS, 20% v:v glycerol, 0.2% w:v bromophenol blue, and 10% v:v 2-mercaptoethanol) and boiled for 5 min, cooled on ice and frozen at −20 °C until use.

SDS resolving gels (15% v/v acrylamide, 400 mM Tris, pH 8.8, 0.1% w/v SDS, 0.2% w/v ammonium persulfate [APS], 0.04% v/v TEMED) were prepared with a 5% stacking gel (5% acrylamide, 190 mM Tris, pH 6.8, 0.1% w/v SDS, 0.15% w/v APS, 0.1% v/v TEMED). Purified EC and LT GAPDH were loaded onto these gels and separated electrophoretically in SDS-PAGE running buffer (25 mM Tris-base, 190 mM glycine, and 0.1% w/v SDS) at 180 V for 45 min. A 3 µL aliquot of Spectra™ Multicolor Broad Range Protein Ladder was added to one lane of every gel to provide molecular weight markers. Commercial pure rabbit muscle GAPDH (Boehringer Mannheim) was also loaded onto the gel to confirm the correct location of GAPDH subunits. Following electrophoresis, proteins were either electroblotted onto polyvinylidiene difluoride (PVDF) membranes (Millipore) by wet transfer and used for immunoblotting or fixed and stained with ProQ diamond phosphoprotein stain.

With regards to phosphoprotein staining, the SDS-gel containing purified EC and LT GAPDH was washed three times in fixing solution (50% v:v methanol, 10% v:v acetic acid) and subsequently left overnight at 4 °C in this solution. The following day the gel was washed three times for ten minutes each time in ddH₂O, after which the gel was immersed in ProQ Diamond phosphoprotein stain (Invitrogen, Eugene, OR) for 30 min. Once the stain was placed onto the gel, the gel remained covered in aluminum foil to protect the light sensitive stain. After staining, the gel was washed three times for 10 min each time with ddH₂O before being viewed in a ChemiGenius Bioimaging System (Syngene, Fredrick, MD). The fluorescent bands were then quantified using GeneTools software. Subsequently, the SDS-gel was stained with Coomassie blue (0.25% w:v Coomassie Brilliant Blue R in 50% v/v methanol, 7.5% v:v acetic acid). The GAPDH ProQ band intensities were subsequently standardized against the Coomassie blue-stained GAPDH band intensities to negate errors caused by sample loading.

Immunoblotting began with the wet transfer of proteins from 15% SDS gels (run as above) onto PVDF membranes previously soaked in methanol. Electroblotting was performed at room temperature for 1.5 h at 160 mA under transfer buffer (25 mM Tris, pH 8.5, 192 mM glycine, and 20% v/v methanol). Following protein transfer, PVDF membranes were incubated overnight at 4 °C with one of the following primary antibodies that detect specific posttranslational modifications to the enzyme: (1) Pan-acetyl (Santa Cruz Biotechnology), (2) Phospho-serine (Calbiochem), (3) Phospho-threonine (Invitrogen), (4) Phospho-tyrosine (Invitrogen), or (5) Methyl-lysine (StressMarq Bosciences Inc.). All primary antibodies were diluted 1:1000 v:v in Tris-buffered saline with Tween-20 (TBST; 20 mM Tris-base, 140 mM NaCl, 0.05% Tween-20) with a small amount of sodium azide added. After the overnight incubation, membranes were washed with TBST three times for 5 min each, followed by incubation with the appropriate secondary antibody conjugated with horseradish peroxidase (Bioshop Canada) at a dilution of 1:4000 v:v in TBST. Membranes were incubated at room temperature for 1.5 h, and then washed three times for 5 min each time with ddH₂O. Signal was then detected using enhanced chemiluminescence (ECL), initiated by the addition of 600 µL of hydrogen peroxide and 600 µL of luminol reagent to the membrane’s surface for several seconds. The mixture was then poured off and the chemiluminescence was detected with the ChemiGenius Bioimaging System (Syngene, MD, USA). Membranes were then stained with Coomassie blue (0.25% w/v Coomassie Brilliant Blue R in 50% methanol, 7.5% acetic acid) and destained with destaining solution (60% v/v methanol, 20% v/v acetic acid in ddH₂O) until bands were clearly seen. Band densities were analyzed using the associated GeneTools software, and were normalized against with the corresponding Coomassie blue stained bands.

In vitro incubations that stimulate protein kinases or protein phosphatases

Crude muscle extracts were prepared as described above, and then subsequently filtered through a small G50 Sephadex column that had been pre-equilibrated in buffer B (25 mM imidazole, 5% v:v glycerol, 5 mM 2-mercaptoethanol, pH 7.0). Aliquots of the filtered supernatants were incubated overnight at 4 °C with specific inhibitors and stimulators of protein kinases or protein phosphatases, as described in MacDonald & Storey (1999). Each aliquot was mixed 1:2 v:v with buffer B plus the following additions:

1. Stop conditions: 1.25 mM EGTA, 1.25 mM EDTA, and 12.5 mM ß-glycerophosphate.

2. Protein Tyrosine Phosphatases (PTP)—30 mM NaF

3. Alkaline Phosphatase (Alk PPase)—10 mM MgCl₂ + 5 mM EDTA + 1 U calf intestine alkaline phosphatase per incubation tube.

4. Protein Kinase C (PKC)—1.3 mM CaCl₂ + 7 µg/mL phorbol 12-myristate 13-acetate (PMA) + 5 mM ATP + 5 mM MgCl₂ + 30 mM NaF + 5 mM Na₃V O₄

5. AMP-activated Protein Kinase (AMPK)—1 mM AMP + 5 mM ATP + 10 mM MgCl₂ + 30 mM NaF + 5 mM Na₃V O₄.

6. Calcium/calmodulin Activated Protein Kinase (CaMK)—1 U of calmodulin/incubation tube + 1.3 mM CaCl₂ + 5 mM ATP + 10 mM MgCl₂ + 30 mM NaF + 5 mM Na₃V O₄.

Samples were spun at 2500 RPM for 1 min through a G50 Sephadex column that had been pre-equilibrated in buffer A. This removed small molecular weight molecules and allowed GAPDH activity to be reassessed in the GAP-oxidizing direction under optimal conditions.

GAPDH bioinformatics

The nucleotide and amino acid sequence of 13-lined ground squirrel GAPDH was deduced using multiple GAPDH sequences from a variety of animals. As the basis for the construction, the rat (Rattus norgevicus) GAPDH coding nucleotide sequence (NCBI Reference Sequence: NM_017008.3) was used in the Basic Local Alignment Search Tool (BLAST) to mine whole genome shotgun sequences from the 13-lined ground squirrel genome, and the sequence with the highest coverage was taken (Accession Number: AAQQ01785922.1). This sequence was then compared to three other whole genome shotgun sequences of 13-lined ground squirrel GAPDH as well as a low-coverage 1.9X 13-lined ground squirrel GAPDH sequence produced by the Broad Institute (http://www.ensembl.org/Spermophilus_tridecemlineatus/Info/Index). When discrepancies existed between the whole genome shotgun reads and the Ensemble sequence, typically, the whole genome shotgun sequence was taken as true. However, if there were discrepancies between whole genome shotgun reads themselves as well as the ensemble sequence, then GAPDH sequences from the rat, mouse, pig and cow were used to help decipher the most likely nucleotide sequence. Sequences were compared using DNAMAN Version 4.0 (Lynnon BioSoft).

The above amino acid sequence and Scansite 2.0 (Obenauer, Cantley & Yaffe, 2003) were used to assess potential phosphorylation sites and specific modifiers of GAPDH phosphorylation state. The predictions were made under medium stringency.

Purification of GAPDH

GAPDH from skeletal muscle of 13-lined ground was purified using a Blue-agarose affinity column followed by a hydrophobic column, phenyl-agarose. The overall yield of the final EC GAPDH preparation was 45% with a fold purification of 460 and a final specific activity of 146 U/mg (Table 1A). The overall yield of LT GAPDH was 22% with a fold purification of 230 and a specific activity of 100 U/mg. This purification procedure yielded electrophoretically pure preparations that were used in all subsequent experiments (Fig. 1).

GAPDH kinetics

The activity of GAPDH in both directions was investigated at two temperatures, one that corresponded to approximately euthermia (36 °C) and one typical of torpor (5 °C). Several significant differences in GAPDH kinetics were observed between EC and LT GAPDH under high and low temperatures. At 5 °C, the V_max values in the GAP-oxidizing direction and 1,3BPG-reducing directions were ∼63% and 44% less, respectively, for LT GAPDH as compared to EC GAPDH (Table 2). At the same temperature, the K_m NAD⁺ for LT GAPDH was 60% lower than that of EC GAPDH. At 36 °C, GAPDH V_max in the GAP-oxidizing direction was significantly different between EC and LT purified samples, LT GAPDH V_max being ∼67% less as compared to EC GAPDH. Similar to the situation at 5 °C, the K_m NAD⁺ was 60% lower for LT GAPDH as compared to the euthermic value (Table 2). K_m GAP and K_m 1,3BPG did not differ between GAPDH from EC and LT conditions at either temperature.

The maximal activity of GAPDH was also monitored over a range of temperatures between 5 °C and 35 °C to determine the activation energy (E_a) for the 1,3BPG-reducing reaction of GAPDH. The calculated E_a for LT GAPDH was 2-fold higher than the corresponding value for EC GAPDH (Table 3).

GAPDH stability and effectors

To assess the thermal stability of each enzyme form, the purified extracts were subjected to short (15 min) incubations at different temperatures, after which GAPDH was assayed at room temperature. Figure 2 indicates that LT GAPDH appears to be significantly less susceptible to thermal denaturation as compared to EC GAPDH, with 45 or 55 °C temperatures being required to cause a significant decrease in LT GAPDH enzyme activity as compared to 35 and 45 °C for EC GAPDH. Furthermore, LT GAPDH displayed a significantly higher thermal melting point of 56.1 ± 0.2 °C as compared to the euthermic GAPDH T_m of 54.7 ± 0.1 °C (Table 3).

GAPDH posttranslational modification

GAPDH posttranslational modification was primarily assessed through Western blots and phosphoprotein staining. Western blotting with antibodies detecting phospho-serine or phospho-threonine indicated no significant difference in these PTMs between EC and LT GAPDH (Fig. 3). This coincided with the assessment of the overall phosphorylation state of GAPDH by ProQ Diamond phosphoprotein staining which indicated no significant difference in total protein-bound phosphate between EC and LT conditions. However, GAPDH from the two conditions differed significantly in phospho-tyrosine content, with LT GAPDH showing 35% less phosphorylation as compared to the euthermic enzyme (Fig. 3).

Western blots looking at the overall acetylation level of purified GAPDH showed no significant difference between control and torpid enzyme forms (Fig. 4). The same results were seen when probing purified GAPDH with anti-methyl-lysine antibody (i.e., no significant differences between EC and LT conditions). Interestingly, analysis of tryptic peptides of purified GAPDH by ESI-QUAD TOF MS/MS revealed several peptides that have been lysine-methylated. A GAPDH peptide from the LT condition displayed a dimethylated lysine residue (K213). The same peptide from the EC condition also displayed this same dimethylation modification, but was also detected in a monomethylation form (Table 4). In addition to lysine-methylation a single peptide from GAPDH from the EC condition contained a methylated aspartate residue (D242). This peptide was not detected in the GAPDH from the LT condition. It should be noted that the sequence coverage for LT GAPDH was substantially lower than that for the control enzyme; approximately 11% for LT GAPDH as compared to 35% for EC GAPDH (Appendix S1A and S1B).

GAPDH regulation by reversible phosphorylation

Crude skeletal muscle samples were incubated in solutions that either stimulated the activities of selected endogenous protein kinases/phosphatases or in a solution that activated commercial alkaline phosphatase, and then GAPDH activity was reassessed at room temperature. Incubations that aimed to stimulate endogenous tyrosine phosphatases (PTP) did not have any significant effect on GAPDH activity while the action of commercial alkaline phosphatase (Alk Ppase) caused a 50% increase in EC GAPDH activity when compared to EC GAPDH held in STOP buffer (endogenous kinases and phosphatases both inhibited) (Fig. 5). Conversely, incubations that stimulated activities of protein kinase C (PKC), AMP-dependent protein kinase (AMPK), or Calcium-calmodulin protein kinase (CaMK) lead to ∼80% decreases in EC GAPDH activity as compared to the activity under STOP condition. Only incubation under conditions that stimulated CaMK altered the activity of LT GAPDH, the relative activity decreasing by ∼60% compared with STOP conditions.