CRISPR interference for high-throughput gene function analysis in Plasmodium knowlesi

Malaria is still a devastating disease that is caused by infection with unicellular parasites of the genus Plasmodium. In 2022, it was responsible for an estimated 249 million cases as well as an estimated 608 000 deaths worldwide. While the disease, especially cases of uncomplicated malaria, can be treated with therapeutic intervention, emergence of resistance against these antimalarial drugs has been a continuous challenge. Elucidating gene function in a systematic way has been integral for understanding the biology in other organisms in the genomic and post-genomic era. This approach has been impeded in Plasmodium, however, due to an extraordinarily high genomic AT-content and poor transfection efficiency in Plasmodium falciparum, the strain responsible for most global cases and fatalities. The latter has put large-scale genetic approaches out of reach until relatively recently, so that currently ca. 40% of the parasite’s genes only have putative functional annotation, while ca. 30% have no functional annotation at all. Genetic studies that have been performed in Plasmodium parasites so far greatly increased our knowledge about the parasite’s genome. However, due to the nature of gene knockout approaches, they could not provide functional information about genes that are essential during the asexual blood-stage of the parasite, which has been determined to be ca. half of the genome. In contrast, knockdown approaches have the capacity to provide a quantifiable phenotype for essential genes because gene function is not entirely abrogated. Therefore, I sought to establish gene knockdown by CRISPR interference in the parasite strain Plasmodium knowlesi in this doctoral thesis so that this technology may eventually be used for high-throughput gene function analysis. This strain was chosen because it had exhibited exceptionally high transfection efficiency in previous reports and, thus, harbored the potential to target a large part, if not all, of the parasite genome. To this end, I used a parasite line with stable expression of the catalytically inactive version of Cas9, dCas9, that had previously been generated by Ann-Kathrin Mehnert during her Master thesis. I performed initial experiments to assess this cell line’s transfection efficiency with a reporter construct, which remained inconclusive because only ca. half of transfected parasites that were resistant to the selection drug WR99210 expressed the reporter. Therefore, resistance against WR99210 was subsequently used as readout for successful transfection, revealing that parasite survival was very low and likely negatively impacted overall transfection efficiency, which I determined to be ca. 0.3%. For knockdown validation, I cloned and then transfected constructs for expression of sgRNAs targeting the essential VPS32 gene and the non-essential CEN4 and SAS6 genes into the dCas9-expressing parasite line and performed growth assays as well as transcript level analysis via RT-qPCR. Although I could not detect a significant downregulation of VPS32 transcript levels in the respective knockdown parasite line, I identified a significant growth defect of this line compared to the control line. As expected, targeting the non-essential CEN4 gene did not elicit a growth defect, however, I detected a significant downregulation in transcripts in one of the knockdown lines compared to a control line. Targeting the SAS6 gene did not result in transcriptional downregulation in any of the generated knockdown lines. In a small-scale proof-of-principle screen, I used a library comprising 20 different sgRNA constructs targeting four genes in total: VPS32, CEN4, SAS6 and CEN3. Monitoring the constructs’ relative abundance ratios over time in six conditions over two biological replicates revealed that I could not detect a specific and efficient knockdown effect in this screen with the CRISPR interference system that I employed. In conclusion, I was able to determine the transfection efficiency of the P. knowlesi dCas9 line used in this thesis, which was lower than published efficiencies for the P. knowlesi wildtype strain but sufficiently high to, in theory, facilitate screens of moderately large scale. Additionally, I was able to show a significant downregulation of CEN4 transcripts in one knockdown parasite line as well as a significant growth defect in the VPS32 knockdown line. Unfortunately, the results of two replicates of a small-scale proof-of-principle screen revealed that I was not able to detect specific and efficient knockdown via the CRISPR interference system used in this thesis. Further inquiry into how efficient CRISPR interference works in this parasite strain and potentially subsequent development of technology is thus necessary to validate it as a tool for high-throughput genetic screening approaches.