Murine macrophage-like RAW 264.7 cells (American Type Culture Collection TIB-71, Monassas, VA) were cultured under standard incubation conditions (37°C, 5% CO₂) and grown in RPMI supplemented with 10% FBS (Atlanta Biologicals, Atlanta, GA) and 1 μg/mL P/S (Cellgro MediaTech, Herndon, VA). Cells were plated at a density of 4 × 10⁶ cells/well in six well plates and incubated for 24 h prior to treatment.

Cells were washed once with Dulbecco's PBS (DPBS) and treated with 40 μM 15(S)-HETE. Immediately following treatment, LPS was added to all wells at a final concentration of 1 μg/mL. After 24 h, cells were washed three times with DPBS and scraped from the wells. Three biological replicates (composed of six pooled wells each) per sample were prepared. Total RNA was isolated using the Versagene RNA purification and DNase treatment kits, following the manufacturer's recommendations.

Microarray analysis was performed by the Vanderbilt Microarray Shared Resource. Three biological replicates of each treatment were analyzed for quality (Agilent 2100 Bioanalyzer, Agilent Technologies, Palo Alto, CA). One microgram of total RNA (30 ng mRNA) was used to generate first strand cDNA using the NanoAmp RT-IVT labeling kit according to the manufacturer's protocol. Following first strand synthesis, second strand synthesis was completed. The resulting cDNA was then purified using an ABI kit-provided column and the entire reaction was used in an IVT reaction to generate DIG-labeled cRNA. The cRNA was then purified using a kit-provided column and assessed for quality on an Agilent Bioanalyzer. All reactions meeting ABI criteria in terms of quantity and size of target produced were fragmented and then hybridized to an ABI mouse genome survey microarray for 16 h with agitation at 55°C per the manufacturer's protocol. Following the addition of the chemiluminescence reaction substrate, each array was immediately imaged on the 1700 Chemiluminescent Analyzer, and a primary analysis was completed by the AB1700 Expression Array System Software (v 1.1.1). Expression values were quantile normalized and filtered based on S/N (> 3) and flag value (< 5000). GeneSpring GX 7.3.1 software (Agilent Silicon Genetics, Redwood City, CA) was used to determine statistically significant differentially expressed genes. T-tests were performed on probes altered by ≥ 1.8-fold in 2 of 3 samples (0.025 p value cut-off, Benjamini-Hochberg multiple testing correction, parametric test, variances not assumed equal) in treated stimulated cells (experimental) relative to stimulated cells (control). Genes were classified according to genes ontology (GO) terms using GeneSpring. In accordance with MIAME procedure, microarray data have been submitted to the NCBI Gene Expression Omnibus and can be found under series number GSE15070. Nomenclature for genes and proteins is as described by the Mouse Genome Informatics (MGI) database guidelines.

Ingenuity Pathways Analysis (IPA) was used for gene expression analysis (Ingenuity Systems^®). A data set containing gene identifiers and corresponding expression values was uploaded into the application, and each identifier was mapped to its corresponding gene object in the Ingenuity knowledge base (IKB). A functional analysis was performed to determine biological functions that were most significant to the genes in the data set. A network analysis was also performed whereby focus genes were overlaid onto a global molecular network developed from information contained in the IKB. Networks of focus genes were then algorithmically generated based on their connectivity. A functional analysis of each network identified the biological functions that were most significant to the genes in the network, and canonical pathway analysis identified the pathways from the IPA library of canonical pathways that were most significant to the data set. Fischer's exact test was used to calculate a p value determining the probability that that each biological function assigned to a network or data set, or the association between the genes in the data set and the canonical pathway, are explained by chance alone.

Quantitative real-time reverse transcription polymerase chain reaction (qRT-PCR) was used to validate the expression levels of genes identified as differentially expressed by microarray analysis. Quadruplicate measurements for n = 3 independent biological replicates per sample were performed. cDNA was reverse-transcribed from 0.5 μg of total RNA using random hexamer primers and Superscript II Reverse Transcriptase (Invitrogen). Reactions were purified using Qiagen's PCR Purification Kit following the manufacturer's protocol. Following RT, all assays were performed with Applied Biosystems TaqMan FAM labeled 20× probes: Arf3 (Taqman assay Mm00500194_m1), Cldn11 (Mm00500915_m1), Cxcl11 (Mm00444662_m1), Mapk14 (Mm00442497_m1), Prdx1 (Mm01621996_s1), Sdc1 (Mm00448918_m1), and Egr1 (Mm00656724_m1). Ywhaz was chosen as the endogenous control based on results obtained from an Applied Biosystems mouse endogenous control array. cDNA amplification was performed using TaqMan 2× Universal PCR Master Mix (Applied Biosystems), and standard Taqman cycling conditions were used as specified by the manufacturer. Cycling and data collection were performed using the Applied Biosystems 7900 HT instrument, and analysis was performed using SDS software to calculate Ct values for each detector. Ct values were processed based on the comparative Ct method where the relative transcript level of each target gene was calculated according to the equation 2^-ΔCt, where ΔCt is defined as Ct target gene – Ct Ywhaz.

LPS-stimulated macrophage-like RAW 264.7 cells were treated for 24 h with 40 μM 15(S)-HETE based on the estimate that trophozoites and Hz contained 33–39 μmol 15-HETE/L RBC 5. Statistically significant (p ≤ 0.025) changes in gene expression (fold change ≥ 1.8 relative to stimulated cells) were identified by microarray analysis. Given that this study aims to explore potential alterations in gene expression that are incurred by 15-HETE during haemozoin phagocytosis, differentially expressed mRNAs were controlled by comparison with a particulate latex bead challenge and BH treatment under the same conditions. Figure 1 illustrates that 15-HETE had a much greater effect on induction of gene expression than repression (293 transcripts versus 100 transcripts, respectively), but overall was very modest compared to either latex bead or BH treatment.

Ingenuity Pathway Analysis (IPA) software was used to examine biological relationships associated with 15-HETE-mediated expression changes. Identifiers and relative levels of altered genes comprising the data set were imported and mapped by IPA for comparison to molecules within the Ingenuity knowledge base (IKB). Two types of IPA analyses were performed. First, a network analysis was employed to reveal direct and indirect relationships that exist between specific genes in the data set. This analysis resulted in the generation of a network map which illustrates direct and indirect connections between focus genes. Second, a functional analysis was performed to identify the biological processes that are most relevant to the entire set of differentially expressed genes. This analysis resulted in a list of significant biological functions associated with the data set as a whole. Functional analyses were also used to find biological processes associated with individual networks. Focus genes, imported genes that are eligible for generating interaction networks based on incorporation in IKB, were used to identify relationships based on known interactions in the literature. Each network is associated with a score indicating the likelihood that the focus genes occur in the network by random chance. Networks scoring 10 or higher (score is defined as -log (p value)) are considered significant.

Among the transcripts modulated by 15-HETE treatment, 263 were eligible for network analysis based on IPA criteria, mapping to 11 relevant interaction networks. The most significant network (Figure 2a) had a score of 51 and associated 27 focus genes. Several transcriptional regulators were among the products encoded by these genes (Bclaf1, Med1, Noc2l, Rnf4, and Zfp36l1). This network also contained Il1b, Cyp3a4, Gnas, and Adfp. A functional analysis performed on this particular network indicated that the differentially expressed genes were associated with "lipid metabolism" and "small molecule biochemistry" (p = 1.27 × 10^-4).

Figure 2b shows the second most significant interaction network identified by IPA network analysis. Eighteen focus genes were incorporated into the network with a score of 29. Functional analysis of the network revealed that the genes were involved in "molecular transport" (p = 9.42 × 10^-7) and "cellular movement" (p = 9.77 × 10^-6). This network is enriched with focus genes encoding molecules associated with the plasma membrane such as Pkd2, Cd300a, Cldn11, Gypc, Klra4, peptidase Adam9 and transporters Atp1a2, Slc16a1, and Slc16a3. Consistent with these genes, the network predicted interactions with several other plasma membrane molecules (Tjp2, Bsg, Cdh1, Tspan3, Tspan4, Cd247, Itgb1, and Atp1b2) that were not present in the data file.

It is thought that Hz impairs cellular function through the generation and introduction of toxic species such as lipid peroxidation products into cells. Previously, the ability of HNE to stimulate a transcriptional response was examined in macrophage-like cells 12. It was observed that HNE significantly impacted a wide range of steady-state responses (e.g., macrophage activation, immune and inflammatory responses, NF-κB signal transduction, ECM degradation, and dyserythropoiesis). Comparison of the number of gene expression changes influenced by 15-HETE and HNE indicates that 15-HETE modulates a number of mRNA targets but is a much less potent agent than HNE (Figure 3).

IPA was also used to perform a functional analysis on genes within the entire data set. Comparison of the biological functions modulated by 15-HETE (Table 1) and HNE 12 revealed that 15-HETE affected a considerably smaller group of transcripts than HNE but mediated a comparable response in terms of the number of molecular and cellular functions and the specific categories affected. Both lipid peroxidation products altered "Cell Cycle", "Cell Morphology", "Cellular Assembly and Organization", "Cell Death", "Cellular Development", "Cell Growth and Proliferation", "Gene Expression", and "Small Molecule Biochemistry". 15-HETE affected several unique categories including "Carbohydrate Metabolism", "Drug Metabolism", "Lipid Metabolism", "Molecular Transport", "RNA Damage and Repair", and "RNA Post-Translational Modification".

Both IPA network and functional analyses identified a large group of "lipid metabolism" and "carbohydrate metabolism" expression changes. Given that Il1b acts upstream of Cyp3a4, Ugdh, Gnas, Gm2a, Psen1, and Il15, stimulated Il1b expression may be indirectly involved in the up-regulation of each of these genes in this study. Expression of several "small molecule biochemistry" transcriptional regulators (Bclaf1, Med1, Rnf4, Noc2l, and Zfp36l1) was also identified.

Differentially expressed genes were sorted into lists based on the direction of regulation, and corresponding Gene Ontology (GO) categories were identified. Gene expression alterations mediated by 15-HETE were compared to two groups of transcripts. The first group consisted of specific genes or gene products associated with human 16 or murine 1718 models of malarial infection or Hz exposure 19. Common transcripts were primarily associated with "cell-to-cell signaling and interaction" and "immune response" (e.g., Fcgrt, Cd86, C5ar1, Ccr4, Mapk14, Pik3ap1, Tapbp, and Tnfaip6). Enhanced expression of guanylate nucleotide binding proteins (Gbp) 1 and 3 observed in this study was consistent with expression changes reported in human and experimental murine malaria 16171820. The second group included genes classified under specific GO processes that are overexpressed in the Plasmodium yoelii model 20 and/or naturally acquired Plasmodium falciparum infections 21, including cell-cell signaling, defense response, immune response, inflammatory response, and signal transduction, among others. Differential expression mediated by 15-HETE treatment that correlated with either of the two groups described above is listed in Table 2. The relatively limited correlation reflects differences between 15-HETE-mediated expression changes in this model and expression changes observed during naturally acquired or experimental malaria. While RAW 264.7 cells have previously been shown to mimic monocyte/macrophage immuno-modulation in the presence of HNE 12, the findings presented herein suggest that 15-HETE is not a major contributor to the altered immune response observed in these cells types during infection.

qRT-PCR was used to confirm several genes susceptible to differential regulation by 15-HETE. The analysis focused on selected genes implicated in the host response to malaria. The results shown in Figure 4 are expressed as fold change relative to LPS-stimulated cells. In agreement with the microarray results in terms of magnitude and direction of change, 15-HETE stimulated the expression of Arf3 (ADP-ribosylation factor 3), Cldn11 (claudin 11), Cxcl11 (chemokine (C-X-C motif) ligand 11), Mapk14 (mitogen-activated protein kinase 14), Prdx1 (peroxiredoxin 1), and Sdc1 (syndecan 1) and repressed the expression of Egr1 (early growth response 1).

A balance between removal of Plasmodium from circulation and sequestration inside host cells is crucial for parasite survival during infection. Sequestration is mediated by cytoadherence, specifically, the adherence of parasitized RBCs (PRBCs) and leukocytes to capillary and post-capillary venular endothelial cells (EC). This cytoadherence reduces blood flow and causes metabolic dysfunction 28 and is thought to be a major factor associated with cerebral malaria (CM). The mechanism(s) used for adhesion and migration involve the expression of constitutive ligands (i.e., adhesion molecules) and receptors on PRBCs or leukocytes and EC. Cell-cell and cell-matrix interactions are also mediated by the secretion of microbial products or cytokines, which enhance the expression of inducible adhesion molecules.

Investigation of potential arachidonic acid metabolite involvement in cytoadherence identified 15-HETE as an agent capable of stimulating basal adhesion of erythrocytes 15 and monocytes to EC 2930. In this study, 15-HETE induced the expression of several transcripts involved in integrin signaling (e.g., Crkl, Rap2b, Arf3). The expression of genes encoding Pkd2 and Sdc1, which are involved in cell-cell and cell-matrix interactions, and Ptpn14, which has alleged involvement in cell adhesion, were also induced by 15-HETE.

The inflammatory response to malaria, both acute and chronic, follows a predictable sequence of events. Initial vascular changes precede increases in permeability, which ultimately causes oedema. Enhanced cytoadherence results in the accumulation, adherence, and migration of leukocytes through vascular endothelium. Molecular mediators are subsequently released and contribute to both the immune response and recruitment/activation of effector cells. Overwhelming evidence demonstrates that the pathophysiology of malaria involves both systemic and local cytokine release. The recruitment of phagocytes around cerebral capillaries has been observed in CM and likely explains increased chemotaxis and chemokinesis 31. CM is a severe complication of P. falciparum infection that is characterized by cytoadherence in cerebral microvasculature. Accumulation of Hz-loaded monocytes has been observed in brains of CM victims 32 and may contribute to the disruption of endothelial basement membrane and subsequent extravasation of blood cells 26. Importantly, blood brain barrier (BBB) destruction and enhanced vascular permeability/oedema are major factors associated with CM 3334.

A potential contribution of 15-HETE toward increased vascular permeability has been examined in the lung. Administration of this hydroxylated fatty acid was shown to increase respiratory oedema fluid production 35, suggesting a role as an inflammatory mediator. The current analysis identified the "Leukocyte Extravasation Signaling" pathway as being significantly (p = 0.015) affected by 15-HETE. Specifically, the steady-state expression of Arhgap12, Cldn11, Crkl, Mapk14, and Ptk2b was up-regulated. Although 15-HETE is generally considered to have anti-inflammatory properties, activation of a large group of genes encoding inflammatory response molecules was observed (Table 2).

15-HETE was recently shown to enhance IL1β expression and MMP9 activity in human monocytes 26. The current study identified a different response to 15-HETE. While Il1b mRNA was elevated in 15-HETE treated LPS-stimulated RAW 264.7 cells, Mmp9 mRNA was down-regulated (-4.8-fold by qRT-PCR). Mmp9 expression can be regulated through a variety of signaling cascades including NF-κB, p38 MAPK, and ERK1/2 pathways 36. It was proposed that enhanced regulation of IL1B and MMP9 by 15-HETE in human monocytes may be associated with NF-κB signaling 26 based on reports demonstrating NF-κB-mediated MMP9 expression in LPS-stimulated RAW 264.7 cells 36. This mechanism seems unlikely given that 15-HETE has been shown to impair NF-κB-mediated expression of iNOS in LPS-stimulated RAW 264.7 cells 13. Furthermore, PPARγ ligands have been shown to repress NF-κB signal transduction 3738 and inhibit MMP9 expression, secretion, and activity in macrophages and vascular smooth muscle cells 394041, in accord with the results of this study.