THE EFFECT OF CHANGES IN SODIUM CONCENTRATIONS ON MITOCHONDRIA By: Teresa Heavilin A Senior Honors Project Presented to the Honors College East Carolina University In Partial Fulfillment of the Requirements for Graduation with Honors By: Teresa Heavilin Greenville, NC May 2015 Approved by: Dr. Mary Farwell Department of Biology at East Carolina University Dr. Richard McCabe Department of Foundational Sciences at East Carolina University School of Dental Medicine 2 Table of Contents: Abstract………………………………………………………………...…………………..Page 3 Introduction......................................................................................................................….Page 4 Materials and Methods……………………………………………………….…………….Page 10 Results……………………………………………………………………………….……..Page 14 Discussion……………………………………………………………………………….....Page 23 Acknowledgements………………………………………………………………………...Page 25 References………………………………………………………………………………….Page 25 3 Abstract: Atherosclerosis, hardening of the arteries, is the major cause of stroke, peripheral artery disease and heart attack. . Overproduction of superoxide anions in mitochondria can give rise to atherosclerosis. Superoxide anions are produced naturally in mitochondria and the organelle has defenses to control the amount of superoxide anions in the mitochondria. However, an overproduction of superoxide anions overwhelms the organelle’s defenses. A low level of overproduction of superoxide anions can cause oxidation of fatty acids; which leads to atherosclerosis. A larger level of overproduction of superoxide anions can lead to malfunctions in mitochondria, oxidative stress, and eventually cell death. Previous studies have focused on free radical scavengers; like vitamins A & E. However, little research has been done to discover what environmental changes in the mitochondria organelles can cause an overproduction of superoxide anions. This is important because if we can determine what causes a significant increase in overproduction of superoxide anions we can prevent this condition instead of treating it after overproduction happens. Previous research indicates that sodium acts on the mitochondria via the sodium-calcium exchange. Based on this knowledge, we hypothesize that an increase in sodium concentration in the environment will result in an increase of superoxide anion production in mitochondria. We want to establish whether there is a specific sodium concentration that will lead to a significant increase in superoxide anions compared to the other sodium concentrations. To determine this, we isolate mitochondria from chicken livers and assay using a 96 micro-well plate with varying amounts of sodium concentrations. The potassium in the wells is used to prevent changing the ionic strength of the medium. The absorbance of the wells are measured to determine the amount of superoxide anions produced in each well. Our preliminary results were not valid due to an 4 error in the preparation of the buffer solutions. The results from the mitochondria assay without varying sodium shows a linear fit and support that the mitochondria are intact and capable of producing superoxide anions. Our final results of the mitochondria assay with varying sodium concentrations had slight but acceptable variance. The final results suggest that sodium might not be as important to superoxide anion production as previously thought. Introduction: Mitochondrial Origin: An overproduction in superoxide anion production inside mitochondria can lead to many health disorders so it is important to understand how the mitochondria functions, how superoxide anions are produced and how the cellular environment can effect production of superoxide anions. Mitochondria are subcellular organelles that are theorized to have a unique origin. Due to the fact that cells create a new mitochondrion by replicating an existing mitochondrion, the question arises: where did the first mitochondrion come from? Organisms that have cells with a nucleus are considered eukaryotes while simpler organisms that contain cells without a nucleus and one circular strand of DNA are called prokaryotes. Mitochondria share a lot of similar traits with prokaryotes. Mitochondria do not have a nucleus and contain a single circular DNA molecule. Due to the similarities between mitochondria and prokaryotes; it is believed that mitochondria were once bacteria. Scientists have developed a theory that describes how they believe a bacterium came to live inside a cell-called endosymbiotic theory.1 The endosymbiotic theory states that the following events were likely to have occurred. Early bacteria could not utilize oxygen until aerobic respiration began to evolve. Along the way 5 an anaerobic eukaryote is hypothesized to have phagocytosed an aerobic bacterium that was not immediately destroyed upon being engulfed. Instead the aerobic bacterium continued to live and divide in the eukaryotic cell; completely safe from the outside world. The eukaryotic cell benefited from this relationship as well. The aerobic bacterium would take the oxygen (which was originally toxic to the anaerobic eukaryotes) and produce energy; resulting in rapid growth and division of eukaryotes with an aerobic prokaryote inside. Eventually, these eukaryotes outcompeted the eukaryotes that were without the aerobic bacteria until almost all the eukaryotes had a bacterium that was performing aerobic respiration for them. Scientists believe that mitochondria were once the free aerobic bacterium that got consumed by a eukaryotic cell; that over time became a unique organelle.1 Mitochondria Structure: Mitochondria can be found in every cell in the human body except in some components of the blood. Different cell types contain a different quantity of mitochondria but all mitochondria can be found in the cytoplasm of the cell. Mitochondria are a few hundred nanometers long and occupy a considerable amount of cell volume. A mitochondrion is mostly made up of proteins and lipids and is essentially a sack of water encased in a membrane. Unlike other organelles, mitochondria have two membranes: an outer membrane and an inner membrane.1 The outer membrane is not folded and has large pores lined with the protein porin which allows small molecules to pass through. The inner membrane is folded and has a lot of surface area. This inner membrane is impermeable to the majority of molecules, including charged ones. The molecules that do pass are under the control of carrier proteins that are embedded inside the 6 membrane wall and facilitate active transport into and out of the mitochondrial matrix. Along with the transmembrane proteins embedded in the inner membrane there are also transmembrane enzymes in the inner membrane. These transmembrane molecules form complexes along the inner membrane. These complexes as well as the tight control of molecules passing into and out of the mitochondrial matrix are central to mitochondrial function.1 The electron transport chain consists of four complexes: Complex 1 NADH-Q Oxidoreductase, Complex 2 Succinate-Q Reductase, Complex 3 Q-Cytochrome c Oxidoreductase, and Complex 4 Cytochrome c Reductase. Complex 1 NADH-Q regenerates NAD+ by removing the H and the electrons from NADH. The electrons enter Complex 1 and will be moved down the electron transport chain. The H+ molecule is pumped out of the complex into the intermembrane space. Complex 2 Succinate-Q Reductase takes the electrons and the H+ molecules from FADH2 to regenerate FAD. The electrons enter the electron transport chain but the H+ molecules are not pumped out of the mitochondrial matrix. Complex 2 is the only complex that does not pump protons out of mitochondrial matrix. Complex 3 Q-Cytochrome c Oxidoreductase accepts 2 electrons from coenzyme Q and pumps protons out into the intermembrane space. Complex 4 Cytochrome c Reductase accepts 4 electrons from Cytochrome c to reduce O2 into 2 molecules of H2O. The water will pass out of the membrane freely and the left of the H+ molecules are pumped out into the intermembrane space. All of the H+ molecules that were pumped out created a proton gradient. The mitochondria uses the proton motive force the proton gradient creates to synthesize ATP via mitochondrial ATPase1. Mitochondria and Superoxide Production: 7 When oxygen is reduced to water by the enzyme Cytochrome c Oxidase (Complex 4); some of the intermediate oxygen molecules are converted to superoxide anion radicals. Superoxide anion is a stable intermediate that is produced when oxygen is reduced 1 electron at a time and can be found on the inner side of the inner mitochondrial membrane. Superoxide anion mediates oxidative chain reactions and is the precursor to most reactive oxygen species such as hydroxyl radical or peroxynitrite. Superoxide is produced by enzyme sources at different redox sites along the respiratory chain.2 At complex 1the enzyme NADH dehydrogenase and at complex 2 the enzyme succinate dehydrogenase both produces superoxide anions. At complexes 3 and 4, the enzyme ubiquinol-cytochrome c reductase is the enzyme responsible for the production of superoxide anion 2.The majority of superoxide anions are produced by the mitochondrial respiratory chain; specifically Complex 1 and Complex 3 3, 4 .Complex 1 (NADH: ubiquinone oxidoreductase) releases superoxide anions into the mitochondrial matrix while Complex 3 (ubiquinol: cytochrome c oxidoreductase) releases superoxide anions into the intermembrane space of the mitochondria 5. The mitochondria can regulate levels of superoxide anion production with the help of antioxidant defenses, regulators and intracellular pH. However, an overproduction of superoxide anions by the respiratory chain can overwhelm the mitochondrial controls; which can lead to oxidative stress within the mitochondria. Superoxide overproduction can be extremely harmful to the cells and evidence supports the concept that superoxide anions can lead to disease such as atherosclerosis. The production of superoxide anions can be kept at a steady state by regulators, antioxidant defenses, signal transduction, and intracellular pH. The two known regulators that can control mitochondrial 8 superoxide are: thiol switching and uncoupling proteins. However, once superoxide anions and other reactive species are produced the two regulators cannot control the radicals.6 At lower levels of superoxide production, intra-mitochondrial antioxidant defenses can help control the amount of reactive species and also their location. . For example, superoxide dismutases controls the amount of superoxide is converted to hydrogen peroxide and the amount that is converted to hydroxyl anion.2 Some other antioxidant defenses include: glutathione peroxidase which eliminates mitochondrial H2O2, catalase, and membrane-linked succinatedependent reductase which can detoxify several radicals.7 In addition to antioxidant defenses, the mitochondria use DNA-repairing enzymes to repair damages done by oxidative stress.2 The cellular environment of the mitochondria can affect the production of superoxide anions. The concentrations of potassium and sodium that mitochondria is exposed to could affect the production of superoxide anions. While there has not been research on the effect of sodium on free radical formation in mitochondria; there is research data that supports that increased potassium concentration inhibits free radical formation in the cells. Dr. Richard McCabe and his colleagues performed a study to determine whether superoxide anion formation by vascular cells would be affected by changes to potassium concentrations. The researchers used two different methods in this study. The researchers had recently isolated white blood cells as well as cultured endothelial and macrophage cells that they used to assess the effects of potassium. The researchers assessed potassium effects on the reactive species in cultured cells. In addition, the researchers measured the rate of cytochrome c reduction in the cells with potassium present at different concentrations. Reducing the potassium resulted in an increase in the rate of cytochrome reduction but if the potassium was increased 9 then the rate of cytochrome reduction was reduced. These results led to the conclusion that physiological increases in the concentration of potassium inhibit the rate of superoxide anion formation and reactive oxygen species formation in the cells. This study established that increases in potassium can inhibit the rate of free radical formation8. While studies have been conducted on the relationship between potassium and the activity of mitochondria superoxide anions, no studies have been done on sodium and how this ion effects superoxide production in mitochondria. The sodium/potassium pump is the main way sodium is released from the cell. The sodium/potassium pump is an ATPase pump. The pump is made up of an alpha and beta subunit. Accompanying the phosphorylation of ADP, the ATPase binds the sodium molecules to these two subunits and the sodium molecules are then transported out of the cell while potassium molecules are pumped in. Once sodium is into the cell, mitochondria can utilize it and bring the sodium into its intermembrane space using mitochondrial sodium transport pathways. Mitochondrial sodium is lower than intracellular sodium but studies have shown that as intracellular sodium increases the mitochondrial sodium will as well.9 Mitochondria utilize the sodium/calcium exchange to bring intracellular sodium into the mitochondria. This pathway is sensitive to any changes in the concentration of intracellular sodium. This pathway is at half-maximal activity at 5-8mM sodium and large changes in the intracellular sodium can lead to substantial release of mitochondrial calcium. This is important because matrix calcium regulates mitochondrial energetics and cell death. Mitochondrial sodium is removed from the mitochondria via the sodium/hydrogen exchange pathway. The sodium that enters through the sodium/calcium exchange is then released out of the mitochondria, while a 10 hydrogen molecule takes its place. This pathway is driven by the proton gradient inside the mitochondria.9 Previous research has shown that potassium enters the cell by passing through the sodium-potassium ATPase pump and that an increase of potassium inside the cell leads to a reduction in superoxide anion production8. We also know that sodium acts on the mitochondria via the sodium-calcium exchange. Based on past results, we theorize that a change in the sodium concentration in the mitochondrial environment will have a direct impact on superoxide production. We hypothesize that an increase in sodium concentration in the environment will result in an increase of superoxide anion production in mitochondria. We want to establish if there is a specific sodium concentration will lead to a significant increase in superoxide anions compared to the other sodium concentrations. To determine this, we isolate intact mitochondria from chicken livers at the centrifuge speed of 3000 rpm. Next an assay is performed using a 96 micro-well plate. Each well has a different sodium/potassium ratio and 20µL of mitochondria. The potassium is used to prevent changing the ionic strength of the medium as the concentration of sodium is increased. The plate is then read at 450 nm and 492 nm in a Chromate microplate reader to measure the amount of superoxide anions produced in each well. Materials and Methods: Four stock solutions were made to use for all of the assays performed: 100 mL of 1 M Tris buffer at pH 7.44, 100 mL of 10 mM EGTA at a pH 8, 100 mL of 1 M MgCl2, and 100 ml of 10 mM CaCl2. Three buffers were made using these four stock solution. A 100mM Cherry buffer with a pH of 7.6 was used as a transport solution for when the livers were in route to the lab and during the isolation process. For the assay of the mitochondria, a 100mM Sodium buffer 11 with a pH of 7.6 and a 100mM Cherry buffer with a pH of 7.6 were prepared. 200 mM potassium succinate solution was made to use during the assays. Isolation protocol: Once all of the solutions were made, isolations were performed on the chicken livers to separate intact mitochondria from the rest of the cells. The chicken livers were provided from chickens on Dr. Farwell’s farm. The livers were transported into lab in the Cherry buffer transport solution. The liver was minced and homogenized to break up the tissue. We were unsure at what centrifuge speed would isolate intact mitochondria so we performed an isolation at several centrifuge speeds (Refer to Table 1 and Figure 1 below). The homogenized chicken liver was centrifuged at each speed for 10 minutes. The pellets were saved and washed with the Cherry buffer transport solution while the supernatant was discarded. The pellets were suspended in micro tubes with 10% DMSO and were frozen. Intact mitochondria are necessary for the assay. Intact mitochondria have membranes that are undamaged and the transport channels into and out of the mitochondria are still present. The sodium/calcium exchange is still functioning in intact mitochondria which is not the case in damaged mitochondria. In damaged mitochondria, the membranes are broken and there is no regulation of sodium intake into the cell. An assay was performed on the mitochondria at different centrifuge speeds to measure the activity of the mitochondria when added to Nitro Blue Tetrazolium and ATP; as seen in Figure 1. The mitochondria isolated at centrifuge speeds 2000 rpm, 3000 rpm, and 6500 rpm were the most active and the mitochondria isolated at 3000 rpm had the most activity. This indicates that intact mitochondria are isolated at the centrifuge speed 12 3000 rpm. Another assay was performed on the mitochondria isolated at centrifuge speeds 2000 rpm, 3000 rpm, and 6500 rpm and the results were replicated (Refer to Figure 2). Table 1: Centrifuge speeds performed during mitochondria isolation Well Centrifuge Speed (rpm) A B C D E 2,000 pellet 3,000 pellet 6,500 pellet 9,500 pellet 13,000 pellet 13,000 supernatant F Figure 1: Picture of activity of mitochondria isolated at centrifuge speeds mentioned in Table 1 Figure 2: Second Assay performed to confirm intact mitochondria are isolated at 3000 rpm 13 Assay Protocol: Each row of the micro well plate will have a desired sodium concentration and each row will increase in the concentration of sodium in the well. Potassium is added to each well to bring the total volume inside each micro well to 200 µl (Refer to Table 2). The purpose of the potassium in each well is to ensure that the ionic strength of the medium in each well does not vary among the micro wells. The purpose of the assay is to determine if changing the concentrations of sodium in the wells will affect the superoxide anion production in the mitochondria. For each assay, 3 replicates were performed and 10 mL of Sodium buffer and 10 mL of Cherry buffer was used. To each buffer, approximately 20 mg of NBT, 2.8 mg of ATP and 500 µL of potassium succinate were added. In the micro-well plate, a different amount of sodium was added to each row and then potassium was added so that the total volume of each micro-well was 200 µL. This was replicated three times to see if results would be similar among the assays (Refer to Table 2). Once the sodium and potassium are added to the wells in the micro-plate, the micro-plate is incubated at 37oC for 30 minutes. During this time, the mitochondria isolated at 3000 rpm is removed from the freezer and thawed. After the 30 minutes, 20 µL of mitochondria are added to each well and then incubated at 37oC for 1 hour. After the 1 hour, the micro-well plate is then read in a Chromate microplate reader at 450 nm and 492 nm. If the mitochondria is clumped at the bottom of the well, add 10% DMSO to the well before reading the plate. 14 Table 2: Amount of sodium and potassium added to each row in micro-well plate for all assays Row K+ concentration Desired Na+ concentration µl Potassium µl Sodium Total µl A 100 mM 0 mM 200 0 200 B 95 mM 5 mM 190 10 200 C 90 mM 10 mM 180 20 200 D 85 mM 15 mM 170 30 200 E 80 mM 20 mM 160 40 200 F 70 mM 30 mM 140 60 200 G 60 mM 40 mM 120 80 200 F 50 mM 50 mM 100 100 200 Results: Preliminary Data: The first assays and the replicates of those assays had very strange results. The replicates should not be as varied as they appear and the expected did not occur. We expected that as the amount of sodium increased in the wells that the activity and the production of superoxide anions in the mitochondria would increase as well. However, the data shows that activity and superoxide production would spike at different sodium concentrations among the replications. The data suggests that an error occurred sometime along the assay process. We determined that the Sodium and Cherry buffers contained a significantly larger amount of calcium than what would be found naturally in the cell. This is problematic because we are experimenting with 15 sodium concentrations that are transported into the mitochondria via the sodium/calcium exchange. The large amount of calcium present in the buffers we were using for the assay would overwhelm the sodium/calcium exchange pathway; resulting in skewed data. (Refer to Figure 3, Table 3, and Graphs 1 and 2). Figure 3: Preliminary Assay 16 Table 3: Preliminary Data from Mitochondria Assay Mitochondria Assay Abs 450 nm Sodium mM 0 mM Trial 1 1.532 Trial 2 2.849 Trial 3 Trial 4 Average 1.756 2.798 2.234 Std. Dev Std. Dev % 0.687 68.7 5 mM 1.882 2.958 3.184 2.038 2.516 0.651 65.1 10 mM 1.685 1.201 1.854 1.825 1.641 0.303 30.3 15 mM 1.139 2.784 1.998 2.247 2.042 0.686 68.6 20 mM 2.567 1.934 2.571 1.529 2.15 0.511 51.1 30 mM 1.244 2.721 2.201 2.691 2.214 0.689 68.9 40 mM 1.147 0.831 1.69 2.329 1.499 0.657 65.7 50 mM 1.729 2.507 2.535 2.302 2.268 0.374 37.4 Graph 1: Absorbance readings for preliminary mitochondrial assay and replicates 17 Graph 2: Comparison of the mitochondrial assay and replicates Linear Assay: New Sodium and Cherry buffers were prepared with a calcium concentration that could be found in the intracellular environment of a healthy mitochondria. As a precaution, a separate assay was performed to ensure that the mitochondria was not a source of error in the preliminary assay. The assay contained 100 µl of sodium and 100 µl of potassium in each of the wells. The rows on the micro well plate had varying amounts of mitochondria and a total of 3 replicates were performed. The assay was performed to ensure that as the amount of mitochondria increased that the absorbance increased linearly. If you increase the number of mitochondria present in a well then there is more mitochondria present to produce more superoxide anions. In theory, the increase in the absorbance reading of the well should be linear to the amount of 18 mitochondria. Our results from this assay shows a linear relationship between the amount of mitochondria and the production of superoxide anions. The standard deviations of the sodium concentrations at 450 nm ranges from 1-9%. The standard deviations of the sodium concentrations at 492 nm ranges from 1-6%. The standard deviation among the trials were less than 3 % except for the 20 µl of mitochondria trials. The trials at this mitochondrial amount had a higher standard deviation than the rest of the mitochondria amounts. This jump in standard deviation could be the result of a pipet error. Considering the errors that can occur when pipetting, these standard deviations can be considered acceptable variance. These results indicate that the mitochondria are not the cause of the skewed results from the preliminary assay (Refer to Tables 4 and 5and refer to Graphs 3 and 4). Table 4: Dose-response mitochondria assay at 450 nm Linear Mitochondria Assay Abs 450 nm Avg. Std. Dev. Std. Dev. % 0.483 0.49 0.022 2.2 0.648 0.644 0.634 0.021 2.1 0.694 0.745 0.703 0.714 0.027 2.7 20 0.807 0.955 0.940 0.901 0.081 8.1 30 1.021 1.040 1.004 1.022 0.018 1.8 40 1.150 1.132 1.172 1.151 0.02 2 Mitochondria µl Trial 1 Trial 2 0 0.472 0.515 5 0.610 10 Trial 3 19 Graph 3: Comparison of assay and replicates of linear assay at 450 nm Table 5: Dose-response mitochondria assay at 492 nm Linear Mitochondria Assay Abs 492 nm Trial 2 Avg. Std. Dev. Std. Dev. % Mitochondria µl Trial 1 Trial 3 0 0.288 0.304 0.287 0.293 0.01 1 5 0.395 0.421 0.402 0.406 0.013 1.3 10 0.469 0.493 0.458 0.473 0.018 1.8 20 0.571 0.674 0.653 0.633 0.054 5.4 30 0.761 0.799 0.766 0.775 0.021 2.1 40 0.908 0.920 0.873 0.9 0.024 2.4 20 Graph 4: Comparison of assay and replicates of linear assay at 492 nm Final Results: Using the new solutions with the appropriate calcium concentrations and the correct dilution of mitochondria to yield a linear assay, the final assay and replicates with the sodium concentration showed slight variation in the results (Refer to Tables 6 and 7 and Graphs 5 and 6). The standard deviations of the sodium concentrations at 450 nm ranges from 2-11%. The standard deviations of the sodium concentrations at 492 nm ranges from 2-8%. The purpose of the assays was to determine if an increase in sodium concentration in the environment surrounding the mitochondria would result in an increase in superoxide production. However, our results show slight discrepancies amongst the replicates. Considering a 3% error caused by pipetting techniques, this range of variance can be considered acceptable. In the assay, there was no sodium concentration that all 4 replicates increased or decreased in superoxide anions. These 21 results suggest that sodium might not be an important factor in the production of superoxide anions as was originally hypothesized. Table 6: Results of mitochondria assay with varying sodium at 450 nm Mitochondria Assay Abs 450 nm Sodium mM 0 mM 5 mM 10 mM 15 mM 20 mM 30 mM 40 mM 50 mM Trial 1 0.781 0.874 0.850 0.787 0.810 0.808 0.911 0.954 Trial 2 0.904 0.860 0.827 0.828 0.896 0.813 0.886 0.944 Trial 3 0.819 0.909 1.070 0.822 0.856 0.893 0.969 0.901 Avg. Std. Trial 4 Dev. 0.896 0.85 0.060 0.905 0.887 0.024 0.895 0.911 0.11 0.950 0.847 0.071 0.938 0.875 0.055 0.949 0.866 0.068 0.900 0.917 0.036 0.980 0.945 0.033 Std. Dev. % 6.0 2.4 11 7.1 5.5 6.8 3.6 3.3 Graph 5: Comparison of mitochondria assay and replicates with varying sodium at 450 nm 22 Table 7: Results of mitochondria assay with varying sodium at 492 nm Mitochondria Assay Abs 492 nm Sodium mM 0 mM Trial 1 Trial 2 Trial 3 0.588 0.639 0.632 5 mM 0.647 0.621 0.671 Avg. Std. Dev. Std. Dev. Trial 4 % 0.692 0.030 3.0 0.658 0.653 0.024 2.4 0.672 10 mM 0.615 0.605 0.789 0.670 0.670 0.084 8.4 15 mM 0.573 0.629 0.620 0.671 0.623 0.040 4.0 20 mM 0.600 0.701 0.683 0.687 0.668 0.046 4.6 30 mM 0.562 0.613 0.657 0.699 0.633 0.059 5.9 40 mM 0.677 0.654 0.712 0.633 0.669 0.034 3.4 50 mM 0.748 0.702 0.663 0.674 0.697 0.038 3.8 Graph 6: Comparison of mitochondria assay and replicates with varying sodium at 492 nm 23 Discussion: The assays and the replicates from the preliminary experiment showed significant variation. The range of variance between trials for a sodium concentration was as high as 69%; which is unacceptable. From the results we received, we immediately recognized that a major error occurred somewhere in the protocol. We discovered the buffers had too much calcium which we hypothesized overwhelmed the sodium/calcium exchange in the mitochondria. New buffers with appropriate levels of calcium were made. In addition, a linear assay on the mitochondria solely was performed; to ensure that the mitochondria were not a source of error. The assays and the replicates from the final experiment showed slight variance. The readings at 492 nm showed smaller variance than the readings at 450 nm and were a range of about 1-10%. Through practicing pipetting skills, I pipet at about a 3% error. Considering the error that could have occurred while pipetting into the wells, this range of variance can be considered acceptable. Other sources of errors could be the location in the tube the mitochondria was pipetted from. If the mitochondria settled to the bottom of the tube but 20 µl were pipetted from the top of the tube for one well and then 20 µl were pipetted from the bottom of the tube for one well, a large variation can occur between the two well readings. Another source of error could be the temperature the buffer solutions and mitochondria were at the time of pipetting. The micro pipettes draw in more cool liquid that the amount it is set for. At the beginning of the experiment, the first wells could have more of the buffers and mitochondria then we set the pipette at. The last wells pipetted probably has the accurate amount of buffers and mitochondria since as the experiment went on the buffers and mitochondria warmed. 24 There was no sodium concentration in which an assay and it replicates all increased in activity. The increases in activity and production of superoxide anions in the mitochondria appear to be independent of the changes in sodium concentration. The results from the preliminary vary significantly and were the result in an error in the protocol. The results from the final experiment does not support our hypothesis that an increase in sodium concentration will directly result in an increase in superoxide anion production. Our results suggest that the sodium concentration might not be important in the production of superoxide anions. There may not be a direct correlation between an increase in sodium concentrations and an increase in superoxide anion production. Sodium might be necessary in the environment for mitochondria to produce superoxide anions but the amount of sodium might not be a cause of the overproduction of superoxide anions. Another possibility is that sodium must be coupled with another ion in order to effect the production of superoxide anions in the mitochondria. Future research could involve performing these isolation and assay protocols on other animal tissue and compare the results to see if sodium effects mitochondria organelles in a similar manner and if the range of variance is improved. This would allow us to see if changes in sodium concentration could have more of an impact on superoxide anion production in other organisms in comparison to the chicken. Another item to consider is if a co-activator or secondary ion must be present with sodium in order to impact the production of superoxide anions. More research is needed to be done that investigates the function of the sodium/calcium exchange and how the exchange is effected by the cellular environment. 25 Acknowledgements: Dr. Richard McCabe, PhD and professor at ECU School of Dental Medicine for his guidance and direction as mentor Dr. Mary Farwell, PhD for her guidance, providing chicken livers, and use of her lab. Staci Love for assistance with calculations. References: 1) De Grey, Aubrey D. The Mitochondrial Free Radical Theory of Aging. Georgetown, TX: R.G Landes Company, 1999.Landes Bioscience, Inc.. Web. 12 Oct. 2014. <https://www.landesbioscience.com/iu/DeGrey_9781587064425.pdf>. 2) Turrens, Julio. "Mitochondrial Formation of Reactive Oxygen Species." J Physiol 552.2 (2003): 335-344. Web. 12 Oct. 2014. <http://jp.physoc.org/content/552/2/335.full>. 3) Lanciano, P., Khalfaoui-Hassani, B., Selamoglu, N., Ghelli, A., Rugolo, M., & Daldal, F. (2013). Molecular mechanisms of superoxide production by complex iii: A bacterial versus human mitochondrial comparative case study. Biochimica Et Biophysica Acta (BBA) - Bioenergetics, 1827(11-12), 1332-1339. Retrieved from http://www.sciencedirect.com/science/article/pii/S0005272813000686 4) 4. Lenaz, G. (2001). The mitochondrial production of reactive oxygen species: Mechanisms and implications in human pathology. IUBMB Lif, 52(3-5), 159-164. Retrieved from http://onlinelibrary.wiley.com/doi/10.1080/15216540152845957/pdf 26 5) Drose, S., & Brandt, U. (2012). Molecular mechanisms of superoxide production by the mitochondrial respiratory chain. Adv Exp Med Biol., 748, 145-169. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/22729857 6) Gutierrez, Jessica, Scott W. Ballinger, Victor M. Darley-Usmar, and Aimee Lander. "Free Radicals, Mitochondria, and Oxidized Lipids the Emerging Role in Signal Transduction in Vascular Cells." American Heart Association 99(2006): 924-932. Web. 12 Oct. 2014. <http://circres.ahajournals.org/content/99/9/924.full>. 7) Turrens, Julio F. "Superoxide Production by the Mitochondrial Respiratory Chain." Bioscience Reports 17.1 (1997): 3-8. Web. 18 Oct. 2014. <http://www-04.allportland.net/bsr/017/0003/0170003.pdf>. 8) McCabe R.D., Bakarich M.A., Srivastava K., Young D.B.: Potassium inhibits free radical formation. Hypertension 1994;24:77-82. 9) Bers, Donald M., William H. Barry, and Sanda Despa. "Intracellular Na+ regulation in cardiac myocytes.” Cardiovascular Research 57(2003): 897-912. Web. 12 Oct. 2014. http://cardiovascres.oxfordjournals.org/content/57/4/897.full