Regulation of liver glutamate dehydrogenase from an anoxia-tolerant freshwater turtle

Background: Freshwater turtles, Trachemys scripta elegans, are one of the few vertebrate species capable of surviving prolonged periods without oxygen. Anoxic survival requires numerous physiological and biochemical changes, including a drastic reduction in metabolic rate, a cessation of oxygen-based metabolism, and a suppression of urea-synthesis. Given this state, the purpose of this study was to investigate the possible regulation of liver glutamate dehydrogenase (GDH), a key enzyme in both nitrogen and carbohydrate metabolism, when these freshwater turtles transition from normoxic to anoxic conditions. Methods: GDH was purified to electrophoretic homogeneity using a combination of blue-agarose and GTP-agarose chromatography. Subsequent kinetic analysis of GDH derived from the liver of both control (normoxic) and 20 h anoxic turtles was performed spectrophotometrically. ProQ Diamond phosphoprotein staining was used to determine if GDH was present in differently phosphorylated forms between normoxic and anoxic states, and in vitro incubations with alkaline and acid phosphatases were used to determine if changes in phosphorylation state resulted in kinetic changes. Results: Kinetic studies revealed that the anoxic form of GDH was significantly less active than the aerobic control, as well as more susceptible to pH-induced inactivation, and GTP inhibition. ProQ Diamond phosphoprotein staining indicated that anoxic liver GDH was significantly less phosphorylated than control GDH. Subsequent stimulation of exogenous alkaline and acid phosphatase activity significantly lowered the Km α-ketoglutarate for control GDH to a value similar to the value found for anoxic GDH in its native state. Conclusion: We conclude that effector molecules, tissue acidification and reversible phosphorylation act in concert to suppress GDH during anoxia in accordance with general metabolic rate depression and the overall shutdown of oxygen-based metabolism. © 2012 Bell et al; licensee Herbert Publications Ltd. This is an Open Access article distributed under the terms of Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0). This permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Correspondence: [email protected] 1 Department of Chemistry, Carleton University, 1125 Colonel By Drive, Ottawa, Ontario, Canada. Background Glutamate dehydrogenase (GDH; E.C. 1.4.1.3) is an enzyme present in the mitochondrial matrix that catalyzes the reversible NAD(P)+linked oxidative deamination of L-glutamate to α-ketoglutarate and ammonium ion. In the glutamate-oxidizing reaction (denoted the forward reaction) GDH is capable of shuttling the carbon skeletons of several amino acids (glutamate, glutamine, arginine, proline, histidine) into the Krebs cycle in the form of α-ketoglutarate where they can contribute to energy production or gluconeogenesis. Alternatively, in the glutamate-synthesizing reaction (denoted the reverse reaction) GDH is able to produce glutamate for use in protein synthesis or in transamination reactions to produce other amino acids. GDH’s role in such important cellular processes suggests the need for regulation, and indeed, this enzyme has been found to be regulated allosterically by common nucleotide cofactors (AMP, ADP, ATP, and GTP; reviewed in [1]) as well as by reversible phophorylation in bacteria, yeast, and most recently, in the liver of Richardson’s ground squirrels during winter hibernation [2-6]. Although known to be regulated, the control of amino acid synthesis/degradation through GDH has yet to be investigated during prolonged oxygen deprivation. Those organisms that can endure environments under low oxygen (hypoxia) or no oxygen (anoxia) conditions require extensive biochemical, behavioral and physiological changes. These changes typically include sluggish movements, decreased blood pressure, heart rate, and renal function [7]. Large scale metabolic changes are also needed to adjust to long term oxygen deprivation, and this could possibly include: (i) compiling large reserves of fermentable fuels; (ii) utilizing strategies to buffer or excrete anaerobic metabolic end products that are typically acidic; (iii) utilizing alternative anaerobic routes of substrate fermentation that are linked to enhanced ATP output; (iv) developing good antioxidant defenses in preparation for the reintroduction of oxygen to the body; (v) up-regulating genes that aid anoxia survival, and most importantly, (vi) a strong reduction of metabolic rate [8,9]. Metabolic rate depression is essential due to the large reduction in ATP production by fermentative pathways in comparison to oxidative metabolism. With this reduction in ATP output during anoxia, there is typically a coordinated reduction in ATP consuming processes, such as protein synthesis, protein degradation, gluconeogenesis, urea synthesis, and ion motive ATPases [10]. Freshwater turtles from the Trachemys and Chrysemys genera are among the few vertebrate species that are capable of surviving extended periods of anoxia. The red-eared slider, Trachemys scripta elegans, is a major model for studies of anoxia tolerance, and is the animal investigated in this study. During the winter these turtles can remain submerged in cold water for 4-5 months to escape freezing air temperatures. While submerged, these red-eared sliders can absorb a sufficient amount of O2 in cold water to drive their metabolic needs [11], however, as the dissolved oxygen levels become reduced in HOAJ Biology ISSN 2050-0874 Bell et al. HOAJ Biology 2012, http://www.hoajonline.com/journals/pdf/2050-0874-1-3.pdf 2 doi: 10.7243/2050-0874-1-3 ice-locked lakes and rivers and anoxic conditions emerge, these turtles become facultative anaerobes. In order to survive without oxygen T. s. elegans typically suppress their metabolic rate to approximately 10-20% of the corresponding aerobic rate at the same temperature [12]. More specifically, hepatocytes isolated from Chrysemys picta bellii demonstrated a 90% reduction in metabolic rate under anoxic conditions when compared to hepatocytes in a normoxic state [13]. During anoxia these turtles rely completely on glycolysis for energy and have not developed alternative fermentative pathways for increased ATP production as is seen in many other anoxia tolerant species [14]. As a result, the turtles have developed mechanisms to buffer against severe acidosis, as tissue acidification is inevitable due to ATP hydrolysis outweighing glycolytic ATP production (i.e. there will be a net production of H+ in the cell)[15]. The inability of the Krebs cycle and the ETC to participate in energy production suggests that it may be necessary to regulate the enzymes that feed these processes during anoxia. With this in mind, it was hypothesized that liver GDH would be suppressed upon long-term exposure of freshwater turtles to an anoxic environment. The present study analyzes liver GDH purified from T. s. elegans comparing and contrasting glutamate-oxidizing and glutamate-synthesizing reaction kinetics for the enzyme from control and 20 h anoxic animals, as well as responses of each enzyme form to cellular metabolites, changes in pH, and reversible phosphorylation. Methods Animals Adult red-eared slider turtles, T. s. elegans, were obtained from Wards Natural Science, Mississauga, Ontario during the winter months and maintained in tanks of dechlorinated water at 7°C for three weeks prior to experimentation. Turtles had access to deep water, a small platform, and food in the form of trout pellets, lettuce and egg shells. Control (normoxic) turtles were sampled directly from the tanks, whereas anoxia was imposed by submerging turtles in sealed tanks filled with deoxygenated water (previously bubbled with 100% nitrogen gas) at 4°C. A wire mesh placed below the surface of the water prevented the turtles from surfacing. Although forced submergence is thought to cause stress for turtles at room temperature, turtles submerged in cold water (~3°C) are typically quiescent and seldom surface to breathe [11,16]. Thus, covering their containers at this point mimics what they would experience in winter environments where eventually ice forms and prevents resurfacing. Turtles used in the present study were sampled after 20 h of anoxic submergence. All turtles were killed by decapitation, via a protocol approved by the university Animal Care Committee and meeting the guidelines of the Canadian Council on Animal Care. Organs were quickly dissected out, frozen in liquid nitrogen and transferred to a -80°C freezer for storage. Preparation of tissue extracts and GDH purification Samples of frozen liver were homogenized 1:5 w:v in a homogenization buffer containing 50 mM Tris-HCl (pH 8.0), 10 mM 2-mercaptoethanol, 10% glycerol and inhibitors of protein kinases (2.5 mM EDTA, 2.5 mM EGTA) and protein phosphatases (25 mM β-GP) with a few crystals of the serine protease inhibitor phenylmethylsulphonyl fluoride (PMSF) added just prior to homogenization. Initial tests showed that 25 mM NaF, another protein phosphatase inhibitor, decreased GDH activity over time and was not used in any assays. Homogenates were centrifuged for 30 minutes at 13,500 x g at 5°C and the supernatant was decanted and held on ice until purification columns could be prepared. Purification of GDH began with the application of a 1.5 mL aliquot of tissue extract to a Blue-agarose column (2.5 x 1.8 cm h x d equilibrated in homogenization buffer). The column was then rinsed with 10 mL of homogenization buffer to remove unbound material. Turtle liver GDH was then eluted by a linear salt gradient of 0-0.5 M KCl. Fractions of ~750 μL were collected by a Gilson FC203B Fraction Collector. GDH activity was determined by assaying each fraction under optimal conditions (as determined for the crude extract in the forward direction). The five most active fractions were then pooled and held on ice until a GTP-agarose column was prepared. The GTP-agarose column (2.8 x 1.2 cm h x d) was equilibrated in homogenization buffer and a 2 mL aliquot of the pooled eluant from the Blue-agarose column was then applied. The column was washed with 10 mL of homogenization buffer to remove any unbound material, and then GDH was eluted with a 0-1 M linear KCl gradient. Fractions were assayed using the optimal conditions for GDH, and the five most active fractions were pooled and held on ice until use in kinetic assays. GDH Assay GDH activity was assayed spectrophotometrically at 340 nm using a Thermo Labsystems Multiskan spectrophotometer. All assays were performed at room temperature and the optimal assay conditions for purified turtle liver GDH in the forward direction were 50 mM L-glutamate, 1.5 mM NAD+, 0.5 mM Mg-ADP, and 50 mM Tris-HCl buffer, pH 8.0 in a total volume of 200 μL with 25 μL of purified extract used per assay. However, without Mg-ADP, the optimum concentration of L-glutamate was 15 mM. In the reverse reaction, the optimal concentrations of substrates were 1 mM α-ketoglutarate, 100 mM NH4Cl, 0.1 mM NADH, 0.5 mM Mg-ADP, 50 mM HEPES buffer, pH 7.2 with 10 μL of purified liver extract used in each assay. Enzyme activity was measured in U/mg of soluble protein. Km values for substrates were determined at optimal co-substrate concentrations. Analysis of the effects of activators and inhibitors was carried out at sub-optimal substrate concentrations (for the forward reaction 2 mM L-glutamate, and 0.5 mM NAD+; for the reverse reaction 0.2 mM α-ketoglutarate, 0.05 mM NADH, and 40 mM NH4Cl). ProQ Diamond phosphoprotein staining The phosphorylation state of purified liver GDH from control and 20 h anoxic turtles was assessed by ProQ Diamond phosphoprotein staining. The five fractions, originating from the blue-agarose column, with the most activity were pooled and soluble protein content was Bell et al. HOAJ Biology 2012, http://www.hoajonline.com/journals/pdf/2050-0874-1-3.pdf 3 doi: 10.7243/2050-0874-1-3 quantified using the Coomassie blue dye-binding method. Aliquots of the pooled fractions were then mixed 1:1 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, 10% v/v 2-mercaptoethanol) and then boiled for 5 minutes, cooled on ice and frozen at -20°C until use. Aliquots containing 0.5 μg of protein were added to the wells of a 10% SDS-PAGE gel. The gel was run at 180 V for 45 minutes in running buffer containing 25 mM Tris-base, 250 mM glycine, and 0.1% SDS. The gel was then washed twice in fixing solution (50% v:v methanol, 10% v:v acetic acid) for ten minutes each time, and then left overnight in the fixing solution at 4 ̊C. The following day the gel was washed with ddH2O three times for 10 minutes each time, and the ProQ Diamond Phosphoprotein stain (Invitrogen, Eugene, OR) was poured over the gel and allowed to sit with continuous motion for 90 minutes. During staining the container holding the gel was covered with tin foil to prevent light from interacting with the lightsensitive stain. The gel remained covered for the remaining steps. Following staining, the gel was washed three times with ddH2O for 5 minutes each. The ChemiGenius Bioimaging System (Syngene, Frederick, MD) was used to visualize the fluorescence intensity of the bands on the gel and the associated GeneTools software was used for quantification. An identical gel was run in parallel with the aforementioned gel, and after electrophoresis proteins were stained for 20 min with Coomassie blue (25% w/v Coomassie Brilliant Blue R in 50% v/v, 7.5% v/v acetic acid) and destained for 10 min with destaining solution (60% v/v methanol, 20% v/v acetic acid in ddH2O). GDH band intensities from ProQ Diamond chemiluminescence were normalized against the corresponding Coomassie brilliant blue stained band to normalize for any variations in sample loading. The band corresponding to GDH was identified using the commercially purified bovine liver GDH (Sigma) that was run on the gel with the control and anoxic turtle samples. Effect of alkaline and acid phosphtase activities on GDH activity GDH purified from control liver was incubated with commercial alkaline or acid phosphatases (Sigma) in an attempt to stimulate dephosphorylation of GDH. Purified tissue extracts were incubated for ~24 hours at 4°C in a 1:2 ratio with the incubation solution, designed so that following dilution, the final concentrations were as indicated below. All incubations contained a basic incubation buffer (50 mM HEPES, 10% v:v glycerol, 10 mM β-mercaptoethanol, pH, 7.2) with additions as follows. Control Incubations (also denoted as STOP): incubation buffer plus 2.5 mM EDTA, 2.5 mM EGTA and 25 mM β-GP. Alkaline or Acid Phosphatases: incubation buffer plus 7.5 mM MgCl2, 3.75 mM EDTA with either 30 U of alkaline phosphatase or 0.6 U of acid phosphatase. Following incubation, the Km α-ketoglutarate (in the absence of ADP) under optimum assay conditions was determined. Effect of pH on the Vmax ratio of purified GDH Purified GDH was assayed in both the forward and reverse directions (under optimal assay conditions) at pH 6.6 and pH 7.4. All of these reactions were done with and without 0.5 mM ADP. Vmax ratios were calculated by dividing the activity of the forward reaction by the activity of the reverse reaction. GDH structural stability To assess the possible structural differences between GDH from control and anoxic states, crude extracts were exposed to various levels of urea (0 4 M), guanidine hydrochloride (GnHCl; 0 4 M), or KCl (0 – 2.5 M). Crude turtle liver extracts were prepared as described above (prior to any GDH purification procedures), and 30 μL of extract were incubated with the various concentrations of the aforementioned molecules in a 200 μL final volume. Incubations were performed at room temperature (22°C) for 1 h prior to assay under optimal conditions for GDH in the forward direction. C50 (the concentration of denaturant that reduced activity by 50%) and Ka values were subsequently determined. Data, statistics and protein determination Enzyme activity was analyzed with a Microplate Analysis Program [17] and the kinetic parameters were determined using the Kinetics v.3.5.1 program [18]. Data are expressed as mean ± SEM from multiple independent determinations on separate preparations of enzyme. The statistical testing used in this study was the unpaired Student’s t-test (P < 0.05). Protein concentration in extracts was determined by the Coomassie blue dye-binding method using the BioRad prepared reagent and bovine serum albumin as the standard. Chemicals and Biochemicals All biochemicals were from Sigma Chemical Company with a few exceptions. Urea and glycerol were obtained from Fisher Biotech Company; NH4Cl was from Mallinckrodt Co.; NAD+ was from Boehringer Mannheim; CaCl2 was purchased from J.T. Baker Chemical Company; okadaic acid was from CalBiochem; NaF was from Fisher Scientific, and β-glycerol phosphate was purchased from BIOSHOP. Results GDH purification The purification scheme for liver GDH from control turtles is depicted in Table 1. Using a combination of two affinity columns (Blue-agarose and GTP-agarose) GDH was purified 17-fold with an overall yield of 41%. The specific activity of the final purified control enzyme was 0.157 U/mg (forward direction). The effectiveness of each step in the purification process was assessed by SDS-polyacrylamide gel electrophoresis with Coomassie blue staining (Figure 1). After purification of GDH by GTP-agarose a single band was seen on the gel, which indicates that the enzyme was purified to homogeneity. The same purification procedure was used to purify GDH from liver of 20 h anoxia-exposed turtles. GDH substrate kinetics The optimum pH for purified GDH varied depending on the direction Bell et al. HOAJ Biology 2012, http://www.hoajonline.com/journals/pdf/2050-0874-1-3.pdf 4 doi: 10.7243/2050-0874-1-3 of the reaction. The forward reaction displayed a broad activity peak with ~90% of the activity being retained from pH 8-10. Activity dropped below 50% of the maximum activity below pH 7 and above pH 10.5. For the reverse reaction, the activity peak was also very broad, and ~90% of the maximal activity was retained from pH 7.09.5. Above pH 10.5, GDH activity dropped below 50% of the maximal activity (data not shown). GDH was purified from liver of both aerobic control and anoxic turtles and kinetic properties of the purified enzyme were assessed in both the forward (glutamate-utilizing) and reverse (glutamatesynthesizing) directions. In the forward direction, several kinetic parameters changed significantly (p<0.05) between control and 20 h anoxic turtles. For instance, the Km glutamate was ~69% higher and Vmax 39% lower for the anoxic enzyme when compared to control GDH. Similarly, in the reverse direction, the Km α-ketoglutarate of purified anoxic GDH was 45% lower and Vmax 70% lower (p<0.05) in comparison to the same value for aerobic control GDH (Table 2). Effects of cellular metabolites on GDH activity The effect of various cellular metabolites on GDH activity (at suboptimal substrate concentrations) was determined for both the forward and reverse reactions of GDH. In the forward direction, ADP acted as an activator of control and 20 h anoxic GDH; increasing activity by 70% and 63%, respectively. However, both the Ka and the fold activation of the enzyme were not significantly different between the two conditions (Table 3). That being said, ADP was able to alter other kinetic parameters that were assessed in its presence. For instance, for both control and 20 h anoxic GDH the Km glutamate increased significantly in the presence of ADP as contrasted with Km glutamate without ADP. Also, Km NAD + for anoxic GDH decreased by 54% (p<0.05) in the presence of ADP as compared to assays without ADP (Table 2). Although the kinetic values changed when assayed with and without ADP, these changes were not uniform for both control and anoxic GDH. For instance, Km glutamate in the presence of 0.5 mM ADP was ~45% higher for anoxic, compared with control GDH. Alternatively, the Km NAD + for anoxic GDH in the presence of ADP was significantly lower (p<0.05) as compared to the normoxic value. Also, the anoxic GDH maximal activity with ADP was 42% lower than the control Vmax (Table 2). Unlike the situation for the forward reaction, control and anoxic GDH displayed different sensitivities to activation by ADP while assayed Purification Step Total Protein (mg) Total Activity (U) Specific Activity (U/mg) Fold Purification % Yield Supernatant 0.30 0.0028 0.0093 100 Blue-Agarose 0.0074 0.0011 0.147 16 78 GTP-Agarose 0.0017 0.00061 0.360 39 41 Table 1. Purification of GDH from the liver of control T .s. elegans