Expression profiling of the schizont and trophozoite stages of Plasmodium falciparum with a long-oligonucleotide microarray

Abstract

Background: The worldwide persistence of drug-resistant Plasmodium falciparum, the most lethal variety of human malaria, is a global health concern. The P. falciparum sequencing project has brought new opportunities for identifying molecular targets for antimalarial drug and vaccine development.

Results: We developed a software package, ArrayOligoSelector, to design an open reading frame (ORF)-specific DNA microarray using the publicly available P. falciparum genome sequence. Each gene was represented by one or more long 70 mer oligonucleotides selected on the basis of uniqueness within the genome, exclusion of low-complexity sequence, balanced base composition and proximity to the 3' end. A first-generation microarray representing approximately 6,000 ORFs of the P. falciparum genome was constructed. Array performance was evaluated through the use of control oligonucleotide sets with increasing levels of introduced mutations, as well as traditional northern blotting. Using this array, we extensively characterized the gene-expression profile of the intraerythrocytic trophozoite and schizont stages of P. falciparum. The results revealed extensive transcriptional regulation of genes specialized for processes specific to these two stages.

Conclusions: DNA microarrays based on long oligonucleotides are powerful tools for the functional annotation and exploration of the P. falciparum genome. Expression profiling of trophozoites and schizonts revealed genes associated with stage-specific processes and may serve as the basis for future drug targets and vaccine development.

Figure 1

Schematic of the ArrayOligoSelector oligonucleotide-selection algorithm. Oligonucleotide selection begins with the collection of all possible 70 mer oligonucleotides from a given ORF. Four filters are executed in parallel: selection for uniqueness within the genome, an optional user-defined pattern filter, avoidance of significant secondary structure (self-binding), and avoidance of low-complexity sequence. The intersection of the set of all oligonucleotides passing these filters are then further selected for a desired base composition and then ranked by proximity to the 3' end of the ORF.

Figure 2

Example of sequence parameters measured by ArrayOligoSelector for a putative member of the var gene family (PlasmoDB v4.0 annotated gene ID PF08_0140). A schematic of the target gene is shown above plots for each filter indicating the positions of four Duffy binding-like domains (DBL) and a putative transmembrane domain (TM). For each filter, a 70 bp window was measured for all possible positions within the gene. The dashed line represents the average value obtained for each filter where 6,000 random 70 bp sequences were chosen from the total collection of P. falciparum predicted ORFs. The black circle denotes the position of the final candidate oligonucleotide (oligo ID F44871_2) chosen for the array.

Figure 3

Relationship of calculated binding energy to relative hybridization intensity. For each of the oligonucleotides shown in Figure 4, a binding energy (kcal/mol) to the perfect match sequence was calculated using ArrayOligoSelector and plotted against relative hybridization intensity. The Pearson correlation coefficient |r| = 0.91 indicates a strong correlation between the calculated binding energy and hybridization performance.

Figure 4

The hybridization performance of long oligonucleotides in relation to internal versus terminal mismatches. Ten putative genes were chosen arbitrarily. Each gene is represented by a series of ten oligonucleotides. The first (rightmost) oligonucleotide in each series represents a perfect match to the target sequence. Mismatches were introduced in increasing numbers (10% steps) into each subsequent oligonucleotide in each series. The pattern of matches and mismatches within each oligonucleotide is shown below each graph, where a black bar indicates a perfect match and a gray bar indicates a mismatch. (a-e) Mismatch positions were randomized ('distributed set'). (g-k) Mismatch positions were chosen such that a contiguous stretch of perfect-match sequence remained in the middle of the oligonucleotide ('anchored set'). The average hybridization performance, measured as the normalized total intensity for 16 independent experiments, is plotted for each oligonucleotide. (f) The average plot for a-e; (l) the average plot for g-k.

Figure 5

Comparison of trophozoite and schizont stages of P. falciparum. Hierarchical cluster analysis of six replicate microarray hybridizations is shown for the 854 genes that yielded at least a twofold expression difference in at least four experiments. In hybridizations 1-3, the schizont RNA was labeled with Cy3 (green signal) and trophozoite RNA was labeled with Cy5 (red signal) and in analyses 4-6, the fluorophore order was reversed. Examples of major functional gene groups enriched in either studied stage, number of corresponding ORFs and several previously characterized representatives of each group are indicated. All gene expression data are available at PlasmoDB and the DeRisi Lab website [23].

Figure 6

Hybridization performance of multiple oligonucleotides representing single ORFs. The average expression ratios were calculated for each oligonucleotide overlapping distinct regions of three ORFs: (a) PFD0985w encoding a predicted hypothetical protein; (b) PFE0040c encoding PfEMP2; and (c) PFA0110w encoding RESA1. The arrows indicate the location of oligonucleotide elements within the coding sequence (thick line), and the ratios express fold enrichment in either trophozoite stage versus schizont stage (t) or schizont stage versus trophozoite stage (s).

Figure 7

Northern blot validation of the microarray results. (a) Comparison of total RNA and poly(A)⁺ RNA for northern analysis. 12 μg total RNA and 2 μg poly(A)⁺ mRNA from both the schizont (s) and trophozoite (t) stages of P. falciparum were blotted and probed with a PCR fragment overlapping the genomic sequence corresponding to oligonucleotide ID M11919_1, which represents the 41 kD antigen (p41), fructose-bisphosphate aldolase (PlasmoDB v4.0 annotated gene ID PF14_0425). (b) Total RNA northern blots were probed with PCR probes overlapping the corresponding oligonucleotide elements ('gene specific') for each gene. The selected gene set included a probable calcium-transporting ATPase (PlasmoDB MAL13P1.61), plasmodial heat-shock protein (PlasmoDB PFI0875w), a member of a family of conserved hypothetical proteins (PlasmoDB PFI1445w) and an ORF with sequence similarity to a sodium-and chloride-dependent taurine transporter (PlasmoDB MAL13P1.130), lactate dehydrogenase (PlasmoDB PF13_0141), and a heat-shock protein homolog of Hsp70-3 (PlasmoDB PF11_0351). The northern blot membranes were stripped and reprobed with the fructose bisphosphate aldolase control probe. Northern blot expression levels were measured by a phosphorimager. For each gene-specific probe, the measured ratio of trophozoite to schizont was divided by the ratio measured for aldolase on the same membrane. The average ratio from six replicate hybridizations on the microarray, including fluorophore reversal, is shown in the lower panel. The ratios express fold enrichment in either the trophozoite stage versus schizont stage (t) or schizont stage versus trophozoite stage (s).