Date of Award


Degree Type


Degree Name

Doctor of Philosophy


Comparative and Experimental Medicine

Major Professor

Michael D. Karlstad

Committee Members

Brian J. Daley, Patricia K. Tithof, Jay Whelan


Enteral nutrition with eicosapentaenoic acid (EPA; 20:5 n-3) and γ-linolenic acid (GLA; 18:3 n-6) decreased pulmonary inflammation by reducing neutrophil counts and chemotactic factors in bronchoalveolar lavage fluid during acute respiratory distress syndrome (ARDS). We hypothesize that the antiinflammatory effects of EPA and GLA may be due, in part, to induction of neutrophil apoptosis. Neutrophil apoptosis is an important physiological process in the resolution of pulmonary inflammation. Furthermore, pro-inflammatory eicosanoids formed from arachidonic acid (AA) by lipoxygenase (LO) and cyclooxygenase (COX) pathways have been shown to inhibit apoptosis in certain cell types. The purpose of this dissertation is to determine whether EPA and GLA, alone or in combination, trigger apoptotic cell death in the human promyelocytic leukemia HL-60 cell line. This dissertation will also determine whether inhibition of LO and COX increases apoptosis with AA, the combination of EPA and GLA, and EPA in HL-60 cells, in vitro. Finally, this dissertation will determine the apoptotic gene profile of EPA/GLA-treated HL-60 cells.

Study 1, HL-60 cells were incubated with 10, 20, 50, and 100 μM EPA, GLA or various combinations of EPA and GLA for 2, 4, 8, 12 and 24 hours. Oleic acid (18:1 n-9) was used as a fatty acid control. Flow cytometry using dualstaining with propidium iodide and annexin V-FITC assessed apoptosis, necrosis and viability. Apoptosis was verified by DNA fragmentation as assessed by agarose gel electrophoresis. EPA, GLA, and various combinations of EPA and GLA significantly induced apoptosis and reduced cell viability in HL-60 cells. vi Viability was significantly reduced to the same extent with the combination of 50 μM EPA\20 μM GLA as compared with 100 μM EPA. These data indicate that EPA and GLA, alone or in combination, reduce cell survival by induction of apoptosis. Thus, induction of apoptosis by select dietary n-3 (EPA) and n-6 (GLA) polyunsaturated fatty acids may be the mechanism of reduces pulmonary inflammation in ARDS.

Study 2 was conducted as follows: HL-60 cells were incubated with 50 μM AA and an enzyme inhibitor (1-10 μM) for either COX, LO, 12-LO, and 5-LO for 12 hours. Flow cytometry was used to assess viability, apoptosis, and necrosis. Apoptosis was further assessed using TUNEL and DNA fragmentation. The highest concentration of LO inhibitors, but not COX inhibitors, decreased viability and increased apoptosis and necrosis in the presence of exogenous AA. These results suggest that disruption of the metabolism of AA by LO, in particular 5-LO, decreases cell survival and increases apoptosis. Thus, downstream metabolic processing of AA by LO but not COX plays a critical role in the regulation of HL- 60 cell apoptosis.

Study 3 was conducted as follows: HL-60 cells were incubated with 50 μM EPA /20 μM GLA in the presence of an enzyme inhibitor (1-10 μM) for 12 hours. Compounds were used to inhibit COX (ibuprofen), 12-LO (baicalein), or 5-LO (AA-861). Flow cytometry assessed viability, apoptosis, and necrosis. 5- LO inhibition reduced cell viability and increased cell death (apoptosis+necrosis) in EPA/GLA-treated HL-60 cells. Inhibition of COX 1 and 2 and 12-LO had no significant effect on cellular viability and death in EPA/GLA-treated HL-60 cells. vii Adding leukotriene B4 counteracted the effect of 5-LO inhibition on apoptosis in EPA/GLA-treated HL-60 cells. These data suggest that the processing of EPA and GLA by 5-LO is critical to HL-60 cell survival.

Study 4 was conducted as follows: HL-60 cells were incubated with 50 μM EPA in the presence of an enzyme inhibitor (1-10 μM) for 12 hours. Compounds were used to inhibit either COX 1 and 2 (ibuprofen), 5-, 12-, 15-LO (NDGA), 12-LO (baicalein), 5-LO (AA-861), and 5-LO activating protein (MK- 886). Eicosanoid (0.001-1.0 μM) add back experiments were also conducted; LTB4 and 5-HETE with 5-LO inhibition and 12-HETE with 12-LO inhibition. Flow cytometry, TUNEL assay, and caspase-3 western blotting were used to assess apoptosis. Inhibition of COX 1 and 2 had no effect on apoptosis. Inhibition of 5- LO and 12-LO significantly increased apoptosis in EPA-treated HL-60 cells. Furthermore, addition of LTB4 reduced apoptosis to levels significantly lower than EPA-treated HL-60 cells alone. 5-HETE and 12-HETE also lowered apoptosis to control levels. These data indicate that inhibition of LO, particularly 5-LO, increased apoptosis in EPA-treated HL-60 cells. Furthermore, this study clearly demonstrated that the products of the LO enzymes, particularly LTB4, are critical in the regulation of apoptosis in EPA-treated HL-60 cells.

Study 5 was conducted as follows: HL-60 cells were incubated with 50 μM EPA and 20 μM GLA for 12 hours. Apoptotic gene profiles were assessed using GEArray original series human apoptosis-2 gene arrays. Total RNA was extracted and probes were synthesized with a primer mix unique for each array. Arrays were hybridized and developed in a non-isotopic manner. Arrays were viii visualized using a 16-bit CCD camera. The apoptotic gene profile for no treatment and EPA/GLA-treated HL-60 cells after 12-hours differed from each other. When comparing density values as a percent of control, all genes assayed were down-regulated. When comparing arrays’ relative intensities, several genes were different between control and treatment groups. For example, Bad, Bax, and Fas were down regulated and Bax, Bcl-2, and p53 were up regulated as compared with control.

This dissertation has provided a possible mechanism for the observed effect of EPA and GLA in human and animal studies of acute respiratory distress syndrome (ARDS): i.e., that enteral nutrition with specialized diets containing EPA and GLA reduced the number of neutrophils and reduced the levels of LTB4 in the bronchoalveolar lavage fluid, reduced pulmonary inflammation, and improved clinical outcomes in patients and animals with ARDS. This dissertation demonstrates that the observed clinical results may be due to increased neutrophil apoptosis induced by EPA and GLA. Induction of apoptosis likely occurs through a change in the eicosanoid profile as well a genetic change through the up-regulation and down-regulation of various pro- and anti-apoptotic genes. Thus, the ability to alter the inflammatory cell phospholipids profile to reduce the amount of AA available for metabolism to pro-inflammatory eicosanoids, in conjunction with 5-lipoxygenase inhibitory compounds, may lead to new therapies that will increase neutrophil apoptosis and provide a injurylimiting pathway in the resolution of pulmonary inflammation in ARDS.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."