Enhancing Gene Co-Expression Network Inference for the Malaria Parasite Plasmodium falciparum.

BACKGROUND: Malaria results in more than 550,000 deaths each year due to drug resistance in the most lethal Plasmodium (P.) species P. falciparum. A full P. falciparum genome was published in 2002, yet 44.6% of its genes have unknown functions. Improving the functional annotation of genes is important for identifying drug targets and understanding the evolution of drug resistance. RESULTS: Genes function by interacting with one another. So, analyzing gene co-expression networks can enhance functional annotations and prioritize genes for wet lab validation. Earlier efforts to build gene co-expression networks in P. falciparum have been limited to a single network inference method or gaining biological understanding for only a single gene and its interacting partners. Here, we explore multiple inference methods and aim to systematically predict functional annotations for all P. falciparum genes. We evaluate each inferred network based on how well it predicts existing gene-Gene Ontology (GO) term annotations using network clustering and leave-one-out crossvalidation. We assess overlaps of the different networks' edges (gene co-expression relationships), as well as predicted functional knowledge. The networks' edges are overall complementary: 47-85% of all edges are unique to each network. In terms of the accuracy of predicting gene functional annotations, all networks yielded relatively high precision (as high as 87% for the network inferred using mutual information), but the highest recall reached was below 15%. All networks having low recall means that none of them capture a large amount of all existing gene-GO term annotations. In fact, their annotation predictions are highly complementary, with the largest pairwise overlap of only 27%. We provide ranked lists of inferred gene-gene interactions and predicted gene-GO term annotations for future use and wet lab validation by the malaria community. CONCLUSIONS: The different networks seem to capture different aspects of the P. falciparum biology in terms of both inferred interactions and predicted gene functional annotations. Thus, relying on a single network inference method should be avoided when possible. SUPPLEMENTARY DATA: Attached.

A Plasmodium falciparum genetic cross reveals the contributions of pfcrt and plasmepsin II/III to piperaquine drug resistance.

Piperaquine (PPQ) is widely used in combination with dihydroartemisinin as a first-line treatment against malaria. Multiple genetic drivers of PPQ resistance have been reported, including mutations in the Plasmodium falciparum chloroquine resistance transporter (pfcrt) and increased copies of plasmepsin II/III (pm2/3). We generated a cross between a Cambodia-derived multidrug-resistant KEL1/PLA1 lineage isolate (KH004) and a drug-susceptible Malawian parasite (Mal31). Mal31 harbors a wild-type (3D7-like) pfcrt allele and a single copy of pm2/3, while KH004 has a chloroquine-resistant (Dd2-like) pfcrt allele with an additional G367C substitution and multiple copies of pm2/3. We recovered 104 unique recombinant parasites and examined a targeted set of progeny representing all possible combinations of variants at pfcrt and pm2/3. We performed a detailed analysis of competitive fitness and a range of PPQ susceptibility phenotypes with these progenies, including PPQ survival assay, area under the dose response curve, and a limited point IC50. We find that inheritance of the KH004 pfcrt allele is required for reduced PPQ sensitivity, whereas copy number variation in pm2/3 further decreases susceptibility but does not confer resistance in the absence of additional mutations in pfcrt. A deep investigation of genotype-phenotype relationships demonstrates that progeny clones from experimental crosses can be used to understand the relative contributions of pfcrt, pm2/3, and parasite genetic background to a range of PPQ-related traits. Additionally, we find that the resistance phenotype associated with parasites inheriting the G367C substitution in pfcrt is consistent with previously validated PPQ resistance mutations in this transporter.IMPORTANCEResistance to piperaquine, used in combination with dihydroartemisinin, has emerged in Cambodia and threatens to spread to other malaria-endemic regions. Understanding the causal mutations of drug resistance and their impact on parasite fitness is critical for surveillance and intervention and can also reveal new avenues to limiting the evolution and spread of drug resistance. An experimental genetic cross is a powerful tool for pinpointing the genetic determinants of key drug resistance and fitness phenotypes and has the distinct advantage of quantifying the effects of naturally evolved genetic variation. Our study was strengthened since the full range of copies of KH004 pm2/3 was inherited among the progeny clones, allowing us to directly test the role of the pm2/3 copy number on resistance-related phenotypes in the context of a unique pfcrt allele. Our multigene model suggests an important role for both loci in the evolution of this multidrug-resistant parasite lineage.

A Plasmodium falciparum genetic cross reveals the contributions of pfcrt and plasmepsin II/III to piperaquine drug resistance.

Piperaquine (PPQ) is widely used in combination with dihydroartemisinin (DHA) as a first-line treatment against malaria parasites. Multiple genetic drivers of PPQ resistance have been reported, including mutations in the Plasmodium falciparum chloroquine resistance transporter (pfcrt) and increased copies of plasmepsin II/III (pm2/3). We generated a cross between a Cambodia-derived multi-drug resistant KEL1/PLA1 lineage isolate (KH004) and a drug susceptible parasite isolated in Malawi (Mal31). Mal31 harbors a wild-type (3D7-like) pfcrt allele and a single copy of pm2/3, while KH004 has a chloroquine-resistant (Dd2-like) pfcrt allele with an additional G367C substitution and four copies of pm2/3. We recovered 104 unique recombinant progeny and examined a targeted set of progeny representing all possible combinations of variants at pfcrt and pm2/3 for detailed analysis of competitive fitness and a range of PPQ susceptibility phenotypes, including PPQ survival assay (PSA), area under the dose-response curve (AUC), and a limited point IC50 (LP-IC50). We find that inheritance of the KH004 pfcrt allele is required for PPQ resistance, whereas copy number variation in pm2/3 further enhances resistance but does not confer resistance in the absence of PPQ-R-associated mutations in pfcrt. Deeper investigation of genotype-phenotype relationships demonstrates that progeny clones from experimental crosses can be used to understand the relative contributions of pfcrt, pm2/3, and parasite genetic background, to a range of PPQ-related traits and confirm the critical role of the PfCRT G367C substitution in PPQ resistance.

Chloroquine resistance evolution in Plasmodium falciparum is mediated by the putative amino acid transporter AAT1.

Malaria parasites break down host haemoglobin into peptides and amino acids in the digestive vacuole for export to the parasite cytoplasm for growth: interrupting this process is central to the mode of action of several antimalarial drugs. Mutations in the chloroquine (CQ) resistance transporter, pfcrt, located in the digestive vacuole membrane, confer CQ resistance in Plasmodium falciparum, and typically also affect parasite fitness. However, the role of other parasite loci in the evolution of CQ resistance is unclear. Here we use a combination of population genomics, genetic crosses and gene editing to demonstrate that a second vacuolar transporter plays a key role in both resistance and compensatory evolution. Longitudinal genomic analyses of the Gambian parasites revealed temporal signatures of selection on a putative amino acid transporter (pfaat1) variant S258L, which increased from 0% to 97% in frequency between 1984 and 2014 in parallel with the pfcrt1 K76T variant. Parasite genetic crosses then identified a chromosome 6 quantitative trait locus containing pfaat1 that is selected by CQ treatment. Gene editing demonstrated that pfaat1 S258L potentiates CQ resistance but at a cost of reduced fitness, while pfaat1 F313S, a common southeast Asian polymorphism, reduces CQ resistance while restoring fitness. Our analyses reveal hidden complexity in CQ resistance evolution, suggesting that pfaat1 may underlie regional differences in the dynamics of resistance evolution, and modulate parasite resistance or fitness by manipulating the balance between both amino acid and drug transport.

A Malaria Parasite Cross Reveals Genetic Determinants of Plasmodium falciparum Growth in Different Culture Media.

What genes determine in vitro growth and nutrient utilization in asexual blood-stage malaria parasites? Competition experiments between NF54, clone 3D7, a lab-adapted African parasite, and a recently isolated Asian parasite (NHP4026) reveal contrasting outcomes in different media: 3D7 outcompetes NHP4026 in media containing human serum, while NHP4026 outcompetes 3D7 in media containing AlbuMAX, a commercial lipid-rich bovine serum formulation. To determine the basis for this polymorphism, we conducted parasite genetic crosses using humanized mice and compared genome-wide allele frequency changes in three independent progeny populations cultured in media containing human serum or AlbuMAX. This bulk segregant analysis detected three quantitative trait loci (QTL) regions [on chromosome (chr) 2 containing aspartate transaminase AST; chr 13 containing EBA-140; and chr 14 containing cysteine protease ATG4] linked with differential growth in serum or AlbuMAX in each of the three independent progeny pools. Selection driving differential growth was strong (s = 0.10 - 0.23 per 48-hour lifecycle). We conducted validation experiments for the strongest QTL on chr 13: competition experiments between ΔEBA-140 and 3D7 wildtype parasites showed fitness reversals in the two medium types as seen in the parental parasites, validating this locus as the causative gene. These results (i) demonstrate the effectiveness of bulk segregant analysis for dissecting fitness traits in P. falciparum genetic crosses, and (ii) reveal intimate links between red blood cell invasion and nutrient composition of growth media. Use of parasite crosses combined with bulk segregant analysis will allow systematic dissection of key nutrient acquisition/metabolism and red blood cell invasion pathways in P. falciparum.

Optimizing bulk segregant analysis of drug resistance using Plasmodium falciparum genetic crosses conducted in humanized mice.

Classical malaria parasite genetic crosses involve isolation, genotyping, and phenotyping of progeny parasites, which is time consuming and laborious. We tested a rapid alternative approach-bulk segregant analysis (BSA)-that utilizes sequencing of bulk progeny populations with and without drug selection for rapid identification of drug resistance loci. We used dihydroartemisinin (DHA) selection in two genetic crosses and investigated how synchronization, cryopreservation, and the drug selection regimen impacted BSA success. We detected a robust quantitative trait locus (QTL) at kelch13 in both crosses but did not detect QTLs at four other candidate loci. QTLs were detected using synchronized, but not unsynchronized progeny pools, consistent with the stage-specific action of DHA. We also successfully applied BSA to cryopreserved progeny pools, expanding the utility of this approach. We conclude that BSA provides a powerful approach for investigating the genetic architecture of drug resistance in Plasmodium falciparum.

Sources of transcription variation in Plasmodium falciparum.

Variation in transcript abundance can contribute to both short-term environmental response and long-term evolutionary adaptation. Most studies are designed to assess differences in mean transcription levels and do not consider other potentially important and confounding sources of transcriptional variation. Detailed quantification of variation sources will improve our ability to detect and identify the mechanisms that contribute to genome-wide transcription changes that underpin adaptive responses. To quantify innate levels of expression variation, we measured mRNA levels for more than 5000 genes in the malaria parasite, Plasmodium falciparum, among clones derived from two parasite strains across biologically and experimentally replicated batches. Using a mixed effects model, we partitioned the total variation among four sources-between strain, within strain, environmental batch effects, and stochastic noise. We found 646 genes with significant variation attributable to at least one of these sources. These genes were categorized by their predominant variation source and further examined using gene ontology enrichment analysis to associate function with each source of variation. Genes with environmental batch effect and within strain transcript variation may contribute to phenotypic plasticity, while genes with between strain variation may contribute to adaptive responses and processes that lead to parasite strain-specific survival under varied conditions.

The power and promise of genetic mapping from Plasmodium falciparum crosses utilizing human liver-chimeric mice.

Genetic crosses are most powerful for linkage analysis when progeny numbers are high, parental alleles segregate evenly and numbers of inbred progeny are minimized. We previously developed a novel genetic crossing platform for the human malaria parasite Plasmodium falciparum, an obligately sexual, hermaphroditic protozoan, using mice carrying human hepatocytes (the human liver-chimeric FRG NOD huHep mouse) as the vertebrate host. We report on two genetic crosses-(1) an allopatric cross between a laboratory-adapted parasite (NF54) of African origin and a recently patient-derived Asian parasite, and (2) a sympatric cross between two recently patient-derived Asian parasites. We generated 144 unique recombinant clones from the two crosses, doubling the number of unique recombinant progeny generated in the previous 30 years. The allopatric African/Asian cross has minimal levels of inbreeding and extreme segregation distortion, while in the sympatric Asian cross, inbred progeny predominate and parental alleles segregate evenly. Using simulations, we demonstrate that these progeny provide the power to map small-effect mutations and epistatic interactions. The segregation distortion in the allopatric cross slightly erodes power to detect linkage in several genome regions. We greatly increase the power and the precision to map biomedically important traits with these new large progeny panels.

The extended recovery ring-stage survival assay provides a superior association with patient clearance half-life and increases throughput.

BACKGROUND: Tracking and understanding artemisinin resistance is key for preventing global setbacks in malaria eradication efforts. The ring-stage survival assay (RSA) is the current gold standard for in vitro artemisinin resistance phenotyping. However, the RSA has several drawbacks: it is relatively low throughput, has high variance due to microscopy readout, and correlates poorly with the current benchmark for in vivo resistance, patient clearance half-life post-artemisinin treatment. Here a modified RSA is presented, the extended Recovery Ring-stage Survival Assay (eRRSA), using 15 cloned patient isolates from Southeast Asia with a range of patient clearance half-lives, including parasite isolates with and without kelch13 mutations. METHODS: Plasmodium falciparum cultures were synchronized with single layer Percoll during the schizont stage of the intraerythrocytic development cycle. Cultures were left to reinvade to early ring-stage and parasitaemia was quantified using flow cytometry. Cultures were diluted to 2% haematocrit and 0.5% parasitaemia in a 96-well plate to start the assay, allowing for increased throughput and decreased variability between biological replicates. Parasites were treated with 700 nM of dihydroartemisinin or 0.02% dimethyl sulfoxide (DMSO) for 6 h, washed three times in drug-free media, and incubated for 66 or 114 h, when samples were collected and frozen for PCR amplification. A SYBR Green-based quantitative PCR method was used to quantify the fold-change between treated and untreated samples. RESULTS: 15 cloned patient isolates from Southeast Asia with a range of patient clearance half-lives were assayed using the eRRSA. Due to the large number of pyknotic and dying parasites at 66 h post-exposure (72 h sample), parasites were grown for an additional cell cycle (114 h post-exposure, 120 h sample), which drastically improved correlation with patient clearance half-life compared to the 66 h post-exposure sample. A Spearman correlation of - 0.8393 between fold change and patient clearance half-life was identified in these 15 isolates from Southeast Asia, which is the strongest correlation reported to date. CONCLUSIONS: eRRSA drastically increases the efficiency and accuracy of in vitro artemisinin resistance phenotyping compared to the traditional RSA, which paves the way for extensive in vitro phenotyping of hundreds of artemisinin resistant parasites.

Altered expression of K13 disrupts DNA replication and repair in Plasmodium falciparum.

BACKGROUND: Plasmodium falciparum exhibits resistance to the artemisinin component of the frontline antimalarial treatment Artemisinin-based Combination Therapy in South East Asia. Millions of lives will be at risk if artemisinin resistance (ART-R) spreads to Africa. Single non-synonymous mutations in the propeller region of PF3D7_1343700,"K13" are implicated in resistance. In this work, we use transcriptional profiling to characterize a laboratory-generated k13 insertional mutant previously demonstrated to have increased sensitivity to artemisinins to explore the functional role of k13. RESULTS: A set of RNA-seq and microarray experiments confirmed that the expression profile of k13 is specifically altered during the early ring and early trophozoite stages of the mutant intraerythrocytic development cycle. The down-regulation of k13 transcripts in this mutant during the early ring stage is associated with a transcriptome advance towards a more trophozoite-like state. To discover the specific downstream effect of k13 dysregulation, we developed a new computational method to search for differential gene expression while accounting for the temporal sequence of transcription. We found that the strongest biological signature of the transcriptome shift is an up-regulation of DNA replication and repair genes during the early ring developmental stage and a down-regulation of DNA replication and repair genes during the early trophozoite stage; by contrast, the expressions of housekeeping genes are unchanged. This effect, due to k13 dysregulation, is antagonistic, such that k13 levels are negatively correlated with DNA replication and repair gene expression. CONCLUSION: Our results support a role for k13 as a stress response regulator consistent with the hypothesis that artemisinins mode of action is oxidative stress and k13 as a functional homolog of Keap1 which in humans regulates DNA replication and repair genes in response to oxidative stress.

Hackathons as a means of accelerating scientific discoveries and knowledge transfer.

Scientific research plays a key role in the advancement of human knowledge and pursuit of solutions to important societal challenges. Typically, research occurs within specific institutions where data are generated and subsequently analyzed. Although collaborative science bringing together multiple institutions is now common, in such collaborations the analytical processing of the data is often performed by individual researchers within the team, with only limited internal oversight and critical analysis of the workflow prior to publication. Here, we show how hackathons can be a means of enhancing collaborative science by enabling peer review before results of analyses are published by cross-validating the design of studies or underlying data sets and by driving reproducibility of scientific analyses. Traditionally, in data analysis processes, data generators and bioinformaticians are divided and do not collaborate on analyzing the data. Hackathons are a good strategy to build bridges over the traditional divide and are potentially a great agile extension to the more structured collaborations between multiple investigators and institutions.

Simultaneous genome-wide gene expression and transcript isoform profiling in the human malaria parasite.

Gene expression DNA microarrays have been vital for characterizing whole-genome transcriptional profiles. Nevertheless, their effectiveness relies heavily on the accuracy of genome sequences, the annotation of gene structures, and the sequence-dependent performance of individual probes. Currently available gene expression arrays for the malaria parasite Plasmodium falciparum rely on an average of 2 probes per gene, usually positioned near the 3' end of genes; consequently, existing designs are prone to measurement bias and cannot capture complexities such as the occurrence of transcript isoforms arising from alternative splicing or alternative start/ stop sites. Here, we describe two novel gene expression arrays with exon-focused probes designed with an average of 12 and 20 probes spanning each gene. This high probe density minimizes signal noise inherent in probe-to-probe sequence-dependent hybridization intensity. We demonstrate that these exon arrays accurately profile genome-wide expression, comparing favorably to currently available arrays and RNA-seq profiling, and can detect alternatively spliced transcript isoforms as well as non-coding RNAs (ncRNAs). Of the 964 candidate alternate splicing events from published RNA-seq studies, 162 are confirmed using the exon array. Furthermore, the exon array predicted 330 previously unidentified alternate splicing events. Gene expression microarrays continue to offer a cost-effective alternative to RNA-seq for the simultaneous monitoring of gene expression and alternative splicing events. Microarrays may even be preferred in some cases due to their affordability and the rapid turn-around of results when hundreds of samples are required for fine-scale systems biology investigations, including the monitoring of the networks of gene co-expression in the emergence of drug resistance.

An integrative analysis of small molecule transcriptional responses in the human malaria parasite Plasmodium falciparum.

BACKGROUND: Transcriptional responses to small molecules can provide insights into drug mode of action (MOA). The capacity of the human malaria parasite, Plasmodium falciparum, to respond specifically to transcriptional perturbations has been unclear based on past approaches. Here, we present the most extensive profiling to date of the parasite's transcriptional responsiveness to thirty-one chemically and functionally diverse small molecules. METHODS: We exposed two laboratory strains of the human malaria parasite P. falciparum to brief treatments of thirty-one chemically and functionally diverse small molecules associated with biological effects across multiple pathways based on various levels of evidence. We investigated the impact of chemical composition and MOA on gene expression similarities that arise between perturbations by various compounds. To determine the target biological pathways for each small molecule, we developed a novel framework for encoding small molecule effects on a spectra of biological processes or GO functions that are enriched in the differentially expressed genes of a given small molecule perturbation. RESULTS: We find that small molecules associated with similar transcriptional responses contain similar chemical features, and/ or have a shared MOA. The approach also revealed complex relationships between drugs and biological pathways that are missed by most exisiting approaches. For example, the approach was able to partition small molecule responses into drug-specific effects versus non-specific effects. CONCLUSIONS: Our work provides a new framework for linking transcriptional responses to drug MOA in P. falciparum and can be generalized for the same purpose in other organisms.

Chemogenomic profiling of Plasmodium falciparum as a tool to aid antimalarial drug discovery.

The spread of Plasmodium falciparum multidrug resistance highlights the urgency to discover new targets and chemical scaffolds. Unfortunately, lack of experimentally validated functional information about most P. falciparum genes remains a strategic hurdle. Chemogenomic profiling is an established tool for classification of drugs with similar mechanisms of action by comparing drug fitness profiles in a collection of mutants. Inferences of drug mechanisms of action and targets can be obtained by associations between shifts in drug fitness and specific genetic changes in the mutants. In this screen, P. falciparum, piggyBac single insertion mutants were profiled for altered responses to antimalarial drugs and metabolic inhibitors to create chemogenomic profiles. Drugs targeting the same pathway shared similar response profiles and multiple pairwise correlations of the chemogenomic profiles revealed novel insights into drugs' mechanisms of action. A mutant of the artemisinin resistance candidate gene - "K13-propeller" gene (PF3D7_1343700) exhibited increased susceptibility to artemisinin drugs and identified a cluster of 7 mutants based on similar enhanced responses to the drugs tested. Our approach of chemogenomic profiling reveals artemisinin functional activity, linked by the unexpected drug-gene relationships of these mutants, to signal transduction and cell cycle regulation pathways.

Predicting functional and regulatory divergence of a drug resistance transporter gene in the human malaria parasite.

BACKGROUND: The paradigm of resistance evolution to chemotherapeutic agents is that a key coding mutation in a specific gene drives resistance to a particular drug. In the case of resistance to the anti-malarial drug chloroquine (CQ), a specific mutation in the transporter pfcrt is associated with resistance. Here, we apply a series of analytical steps to gene expression data from our lab and leverage 3 independent datasets to identify pfcrt-interacting genes. Resulting networks provide insights into pfcrt's biological functions and regulation, as well as the divergent phenotypic effects of its allelic variants in different genetic backgrounds. RESULTS: To identify pfcrt-interacting genes, we analyze pfcrt co-expression networks in 2 phenotypic states - CQ-resistant (CQR) and CQ-sensitive (CQS) recombinant progeny clones - using a computational approach that prioritizes gene interactions into functional and regulatory relationships. For both phenotypic states, pfcrt co-expressed gene sets are associated with hemoglobin metabolism, consistent with CQ's expected mode of action. To predict the drivers of co-expression divergence, we integrate topological relationships in the co-expression networks with available high confidence protein-protein interaction data. This analysis identifies 3 transcriptional regulators from the ApiAP2 family and histone acetylation as potential mediators of these divergences. We validate the predicted divergences in DNA mismatch repair and histone acetylation by measuring the effects of small molecule inhibitors in recombinant progeny clones combined with quantitative trait locus (QTL) mapping. CONCLUSIONS: This work demonstrates the utility of differential co-expression viewed in a network framework to uncover functional and regulatory divergence in phenotypically distinct parasites. pfcrt-associated co-expression in the CQ resistant progeny highlights CQR-specific gene relationships and possible targeted intervention strategies. The approaches outlined here can be readily generalized to other parasite populations and drug resistances.