Differential Gene Expression in Host Ubiquitination Processes in Childhood Malarial Anemia

Abstract

Background: Malaria remains one of the leading global causes of childhood morbidity and mortality. In holoendemic Plasmodium falciparum transmission regions, such as western Kenya, severe malarial anemia [SMA, hemoglobin (Hb) < 6.0 g/dl] is the primary form of severe disease. Ubiquitination is essential for regulating intracellular processes involved in innate and adaptive immunity. Although dysregulation in ubiquitin molecular processes is central to the pathogenesis of multiple human diseases, the expression patterns of ubiquitination genes in SMA remain unexplored. Methods: To examine the role of the ubiquitination processes in pathogenesis of SMA, differential gene expression profiles were determined in Kenyan children (n = 44, aged <48 mos) with either mild malarial anemia (MlMA; Hb ≥9.0 g/dl; n = 23) or SMA (Hb <6.0 g/dl; n = 21) using the Qiagen Human Ubiquitination Pathway RT² Profiler PCR Array containing a set of 84 human ubiquitination genes. Results: In children with SMA, 10 genes were down-regulated (BRCC3, FBXO3, MARCH5, RFWD2, SMURF2, UBA6, UBE2A, UBE2D1, UBE2L3, UBR1), and five genes were up-regulated (MDM2, PARK2, STUB1, UBE2E3, UBE2M). Enrichment analyses revealed Ubiquitin-Proteasomal Proteolysis as the top disrupted process, along with altered sub-networks involved in proteasomal, protein, and ubiquitin-dependent catabolic processes. Conclusion: Collectively, these novel results show that protein coding genes of the ubiquitination processes are involved in the pathogenesis of SMA.

Keywords: Plasmodium falciparum; differential gene expression; malarial anemia; ubiquitin proteasome system; ubiquitination.

FIGURE 1

Comparison of ubiquitination gene expression levels. Children (n = 44), with mild malarial anemia (MlMA; Hb ≥9.0 g/dl, n = 23) and severe malarial anemia (SMA; Hb <6.0 g/dl, n = 21) were enrolled into the study. Gene expression profiles were measured using the Human Ubiquitylation Pathway RT² Profiler PCR Array kit. Geometric mean was used as a normalization factor, and data standardized using five housekeeping genes [Actin, beta (ACTB), Beta-2-microglobulin (B2M), Glyceraldehyde-3-phosphate dehydrogenase (GAPDH), Hypoxanthine phosphoribosyltransferase 1 (HPRT1) and Ribosomal protein, large, P0 (RPLPO)]. Data were analyzed by the ΔΔ ^C_T method (2^−ΔΔC_T ) (Livak and Schmittgen, 2001), using the RT² Profiler PCR Array Data Analysis Webportal (Qiagen, United States). Fold regulation set at 1.5, and p ≤ 0.050. (A). The Volcano Plot shows gene expression changes that plots the log base 2 of each gene fold change value on the x-axis versus the negative log base 10 of each genes p-value on the y-axis. The center vertical line indicates unchanged gene expression, while the two outer vertical lines indicate the selected fold regulation threshold, with the data points right of the solid line indicating upregulated genes and those to the left representing downregulated genes. p-values were calculated using the student’s t-test of the triplicate raw C_T values. (B). Heat map showing the graphical and color-coded representation of fold regulation data between MlMA and SMA groups overlaid onto the PCR array plate layout. The yellow color represents the average magnitude of gene expression. The brightest red represents the smallest value, and the brightest green represents the highest value. (C). Cluster gram of non-supervised hierarchical clustering of the entire dataset showing a heat map with dendrograms indicating co-regulated genes across the clinical groups. The black color represents the average magnitude of gene expression. The brightest green represents the smallest value, and the brightest red represents the highest value. Similarities of genes across the PCR array was calculated using a correlation coefficient between 2 dimensional profiles.

FIGURE 2

Differentially expressed gene enrichment analysis of the top scored process networks. Relationship between differentially expressed ubiquitination genes (p ≤ 0.050) in the case (SMA; n = 21) and control (MlMA; n = 23) groups was determined using enrichment analysis to identify process networks on MetaCore™. Additional enrichment analysis for same differentially expressed genes (p ≤ 0.050) was done using canonical pathway modeling to map out associated subnetwork processes. (A). Ubiquitin-Proteasomal Proteolysis (FDR, p = 7.049 × 10⁻¹¹) process network that encompassed 8 of the 15 genes that were significantly dysregulated. The blue-shaded circles show down-regulated genes and the red-shaded circles are up-regulated genes, all from the ubiquitination panel. (B). The sub-network [CFTR, Proteasome (20S core), CHIP, RBP-J kappa (CBF1), c-Jun, p = 1.610 × 10⁻⁴¹] contains 11 seed nodes (genes with p < 0.050 for differential expression between SMA and MlMA) and 31 total nodes. (C). The sub-network (p53, NF-kB, UBE2E3, MDM2, SUMO-1, p = 1.460 × 10⁻³²) contains 8 seed nodes and 13 total nodes.