The plasmids used in this experiment were laboratory-modified plasmids as described previously (Shang et al., 2022). The Cas9 gene was carried by pUF1 plasmid, and the guide RNA (5′-GGTTGTTGGGAAGAAGCAAA-3′) and homologous arm sequences were carried by pL6cs plasmid. Plasmid construction is described below. First, the guide RNA was cloned into the pL6cs construct between XhoI and AvrII restriction enzyme sites. Subsequently, the C-terminal homologous arm sequences of pfnsun1 with the ty1-glms tag were cloned into AflII and AscI restriction enzyme sites. The successfully constructed plasmids were transformed into Escherichia coli XL10 for amplification and purification. Primers used to amplify homologous sequences are listed below: pfnsun1-5′HR-F: GGA TAA TGC AAT GGA TAC AC; pfnsun1-5′HR-R:GTT CAA AAT GTA TGG CAA CAT C; pfnsun1-3′HR-F:ACA AAT GGT TCT GGA GGT GAA GCA AAA GGA AAA ATA ATA ATA GAT G; and pfnsun1-3′HR-R: ATC CTT TTT ACC AAG CAC TC.

100 μg purified pL6cs-nsun1-ty1-glmS plasmids and 100 μg pUF1-Cas9-BSD plasmids were electro-transfected into fresh RBCs under the condition of 310V and 950μF, then enriched schizont-stage parasites were added into RBCs. As soon as the parasitemia reached 5% after invasion, selection drugs (2.5 nM WR99210 and 2 μg/ml Blasticidin S deaminase (BSD)) were started to add until live parasites could be found in the Giemsa solution-stained thin blood smears. Then parasites were collected and genomic DNA was extracted for PCR identification and Sanger sequencing. Once the strain was successfully constructed, it was cloned out by limiting dilution cloning as described (Fan et al., 2020).

Plasmodium falciparum lines 3D7 and Pfnsun1-ty1-glmS were cultivated in vitro according to standard procedures. In brief, parasites were cultured in complete RPMI 1640 medium (Thermo Fisher Scientific, Carlsbad, CA, USA) containing 0.5% Albumax I (Thermo Fisher Scientific) and grown in media with type O+ erythrocytes at a 2% hematocrit. The incubator was set to a gas phase condition of 5% O₂, 5% CO₂, and 90% N₂ at 37°C. To obtain tightly synchronized parasites, parasites were purified with 70%/40% Percoll–sorbitol gradients at the schizont stage. After schizonts invaded fresh RBCs for about 5 h, the resulting rings were synchronized with 5% sorbitol treatment and used for subsequent research.

Pfnsun1-ty1-glmS strain was tightly synchronized to a 5-hour window as previously mentioned. In brief, parasites were collected by the Pecoll-sorbitol solution at schizont stage, subsequently invading new erythrocytes. 5-6 hours later, ring-stage synchronization was performed using the sorbitol solution. The strictly synchronized parasites were diluted to 0.1% parasitemia at ring stage and cultured in 6-well plate which were divided into 2 groups, i.e. with or without 5 mM glucosamine (GlcN). Parasites were cultured for four consecutive replicating cycles, and Giemsa-stained thin blood smears were collected from each cycle to count parasitemia. After three independent replications of the experiment, the data obtained were graphed using GraphPad Prism9 software.

The Pfnsun1-ty1-glmS strain was tightly synchronized, and ring-stage parasites were diluted to 1% parasitemia with 2% hematocrit in the presence or absence of 5 mM GlcN. Ring, trophozoite, and schizont stage samples were collected for Western blot analysis at the next life cycle. Briefly, infected RBCs were lysed by 0.15% saponin to release parasites, which were washed by PBS and resuspended in an equal volume of 2 × protein loading buffer, then heated at 100°C for 5 min. The samples were stored at – 80°C for later experiments. Proteins were separated by 8%–10% SDS-PAGE according to the molecular weight and transferred to the PVDF membrane. The primary antibodies were incubated for 1 h at RT. After three washes with PBST, membranes were incubated with the secondary antibody for 1 h at RT. After five washes, target protein signals were detected using ECL Western Blotting Kit (GE Healthcare, USA). The primary antibodies used in this study were mouse anti-ty1 (Sigma, Germany) at 1:1,000 and rabbit anti-aldolase (Abcam, England) at 1:2,000. The HRP-conjugated secondary antibodies were goat antimouse IgG (Abcam, England) and goat antirabbit IgG, which were diluted to 1:5,000.

Pfnsun1-ty1-glmS parasites were tightly synchronized by Percoll and sorbitol to a 5-h window with or without 5 mM GlcN treatment as described above. Samples were collected in TRIzol at the ring (10–15 hpi), trophozoite (25–30 hpi), and schizont (40–45 hpi) stages during the next cycle, respectively. Total RNA was extracted with the kit (Zymo Research, USA) according to a standardized procedure. Library preparation for strand-specific RNA-seq was carried out by poly(A) selection with the KAPA mRNA Capture Beads (KAPA) and fragmentation to about 300–400 nucleotides (nt) in length according to the KAPA Stranded mRNA-Seq Kit (KK8421). Libraries were sequenced on an Illumina NovaSeq 6000 system to generate 150 bp pair-end reads.

RNA-seq reads were trimmed by trim-galore, then aligned to the PlasmoDB-45_Pfalciparum3D7_Genome using Hisat2. Read counts were obtained using FeatureCounts. Subsequently, DESeq2 quantified differentially expressed genes (DEGs) with the criteria of ≥2-fold alteration and p-adj < 0.05. The Gene Ontology (GO) enrichment analysis was performed on PlasmoDB (https://plasmodb.org/plasmo/).

RIP assays were performed as previously described (Fan et al., 2020). Briefly, about 5 × 10⁹ synchronized ring-stage parasites were collected and lysed by saponin. The resulting parasite pellet was lysed under nondenaturing conditions (50 mM Tris-Cl at pH 7.4, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 1% NP-40, 1× proteinase inhibitors, 0.2 U/μl RNase inhibitor) for 2 h at 4°C with rotation. The supernatant was collected and incubated with 10 µg mouse anti-ty1 antibodies for 3 h, then incubated overnight with protein G magnetic beads at 4°C. The complex was washed twice with IPP500 (500 mM NaCl, 10 mM Tris-Cl at pH 8.0, 0.05% NP-40, 1 × proteinase inhibitor, 0.2 U/µl RNase inhibitor) and once with PBS, then RNA was eluted by TRIzol reagent and extracted by the phenol-chloroform method. The RNA was directly used to prepare strand-specific RNA-seq libraries without poly(A) enrichment. Libraries were sequenced on an Illumina NovaSeq 6000 system using 150 bp pair-end reads.

For RIP-seq data analysis, reads were removed with trim-galore and aligned to the genome by Hisat2. Samtools was used to sort reads. Peaks (q-value cut-off < 0.05) were identified using the macs2 call peak command, by comparing the control and the group with the ty1 antibody, using default settings. Enrichment heatmaps and profile plots were generated using the deepTools computeMatrix and plotHeatmap tools. Peak annotation was carried out using ChIPseeker in RStudio.

Pfnsun1 (PF3D7_0704200) ORF contains 4,263 bases, encoding a protein of approximately 141.2 kDa in molecular weight and containing one NOP2 structural domain. In order to compare the homology within the NSUN family, we analyzed the amino acid sequences of PfNSUN1 to PfNSUN4, along with other common Plasmodium NSUN1 proteins. Our findings showed that sequences of PfNSUN1 had low similarity to PfNSUN2-4 but was conserved among different Plasmodium species ( Figure 1A ). To determine the function of pfnsun1, we attempted to disrupt the pfnsun1 gene. However, following three unsuccessful transfection attempts, it was concluded that the gene might be indispensable. Consequently, we opted to construct a knockdown strain. About 1 kb bases before and after the stop codon of the pfnsun1 gene were selected as homologous arm sequences, and three tandem ty1 tags and a glmS sequence were cloned into the plasmid. The constructed plasmids were electro-transferred into wild-type parasites, which were then cultured for about 3 weeks with WR and BSD drug selection until live parasites were observed by microscopy ( Figure 1B ). To confirm successful transfection, we designed a forward primer P1, upstream of the 5′ homologous arm, a reverse primer P2 within the glmS sequence, and another reverse primer P3 within the 3′ homologous arm sequence, respectively. The sequence lengths of wild-type and transgenic strains were verified using the P1+P2 and P1+P3 primers. Agarose gel electrophoresis results confirmed the successful generation of the pfnsun1-ty1-glmS parasite strain ( Figure 1C ).

To verify the knockdown effect of the pfnsun1-ty1-glmS transgenic strain, we performed a Western blot assay. The addition of 5 mM GlcN had no significant impact on the growth of wild-type P. falciparum parasites as confirmed by previous studies (Liu et al., 2020; Shang et al., 2021b). Thus, we collected the ring, trophozoite and schizont stage samples in the second growth cycle with or without GlcN to extract protein after synchronizing the parasites. The results showed insignificant changes in protein levels at ring and trophozoite stages but a significant decrease at the schizont stage ( Figure 2A ). The gray value of protein electrophoresis with GlcN treatment at the schizont stage decreased by approximately 34% compared to the untreated group ( Figure 2B ). To investigate the importance of PfNSUN1 protein in the intraerythrocytic developmental cycle, we performed growth curve experiment. We minimized the growth window of the transgenic parasite line to 5 h after rigorous synchronization. When parasites reached the early trophozoite stage, we diluted the parasitemia to 0.1% and continued cultivating for four growth cycles with or without GlcN and counted parasitemia. The results showed that at the fourth cycle, parasitemia of the group without GlcN treatment reached an average of 19.7%, while parasitemia of the group with drug treatment reduced to 12.9%, indicating an approximate 35% decrease in growth efficiency ( Figure 2C ). These results suggest that the pfnsun1 gene plays an important role in the intraerythrocytic developmental cycle.

To explore the function of PfNSUN1 protein as an epigenetic regulator, we performed strict synchronization and divided the samples into GlcN-added and non-GlcN-added groups. Transcriptome samples were collected in the subsequent growth cycle ( Figure 2D ). RNA-seq data revealed significant transcriptomic changes following pfnsun1 knockdown. Specifically, there were only 18, 32, and 114 upregulated genes, but 224, 278, and 556 downregulated DEGs at the ring, trophozoite, and schizont stages, respectively ( Supplementary Table S1 ). We highlighted 10 genes with the greatest changes at each stage and found that most of the genes were antigenic variant genes or genes with unknown functions. Notably, K2 was significantly downregulated at the ring stage, and hap2, resa, and resa3 were significantly downregulated at the schizont stage. In light of these findings, PfNSUN1 may be regarded as a positive regulator that facilitates gene expression ( Figures 2E, F ).

To clarify the genes affected by pfnsun1 knockdown, we took the intersection of down-regulated differentially expressed genes (DEGs) at ring, trophozoite and schizont stages. This analysis revealed 12 genes with consistently decreased expression across all stages ( Figure 3A ). These 12 genes included sir2a, dip13, alv7, and imc20. Sir2a is an important regulator in P. falciparum and is critical for maintaining a stable heterochromatin environment and exclusive expression of antigenic variant genes (Petter et al., 2011; Mancio-Silva et al., 2013) ( Figure 3B ). Given that the schizont stage exhibited the highest number of downregulated DEGs, we selected these genes for GO enrichment analysis. The analysis indicated that these genes impact multiple pathways, including host entry, movement within the host environment, biological process involved in symbiotic interaction, obsolete pathogenesis, actin cytoskeleton organization, protein phosphorylation, peptidyl−serine modification, etc. ( Figure 3C ). To ensure comprehensive pathway identification by GO analysis, we also performed KEGG enrichment analysis on these genes and found that these genes were involved in the regulation of amino sugar and nucleotide sugar metabolism, glycerophospholipid metabolism, glycerolipid metabolism, fatty acid biosynthesis, and other pathways ( Figure 3D ). To ascertain the impact of PfNSUN1 on parasite growth, we investigated the downregulation of eba, msp, rap, rh, and ron invasion genes. Our findings revealed that the transcripts of eba and rh family members exhibited a downregulation of over 4-fold ( Figure 3E ). It was plausible that PfNSUN1 influenced the expression of these invasion-related genes. In conclusion, these findings indicate that the reduction in parasitemia may be associated with conserved biological processes, such as invasion and metabolism.

Pfnsun1 is an RNA methylation factor that influences gene expression by modifying RNA. Therefore, we collected a ring stage sample of the pfnsun1-ty1-glmS strain to identify its direct-binding RNAs by RIP-seq experiments. The results showed that RNAs bound by the PfNSUN1 protein were distributed across all 14 chromosomes, as illustrated in the peak diagram ( Figures 4A, B ). There were 154 target genes confirmed after aligning to the genome ( Supplementary Table S2 ). In combination with RNA-seq data, our findings indicated that these genes were predominantly down-regulated, thereby substantiating the hypothesis that PfNSUN1 exerted a positive regulatory effect on gene expression ( Figure 4C ). GO enrichment analysis showed that these genes were involved in pathways including cytoadherence to the microvasculature, biological adhesion, response to xenobiotic stimulus, modulation by symbiont of host process, evasion of host immune response, response to host immune response, response to host defenses, response to host, response to external biotic stimulus, and response to defenses of other organisms ( Figure 4D ). These results showed that PfNSUN1 is related in the response between parasite and host.

To validate target genes on which PfNSUN1 protein has a significant effect, we took intersections of target genes obtained from RIP-seq data with down-regulated DEGs at different stages. The down-regulated DEGs at ring, trophozoite, and schizont stages had 14, 12, and 11 same genes with PfNSUN1 protein target genes, respectively ( Figures 5A–C ). At ring stage, pfnsun1 mainly regulates cct, eg5, irp, suv3, m712, emp3, ap2-g5, p52, etc. The functions of suv3 and m712 are unknown. PfCCT has a key regulatory function in the second and rate-limiting step of the de novo phosphatidylcholine biosynthesis, which is essential for parasite survival (Izrael et al., 2020). PfEG5 is a molecular motor that cross-links microtubules, similar to other organisms (Cook et al., 2021). PfIRP regulates parasite environment homeostasis by binding iron response elements (Hodges et al., 2005). Pfemp3 encodes erythrocyte membrane protein 3, which appears on the cytoplasmic surface of the host cell membrane in the later stages and is associated with membrane skeleton and cytoadherence (Waterkeyn et al., 2000). An important factor affected by pfnsun1 is pfap2-g5, which belongs to the largest ApiAP2 transcription factor family in P. falciparum. A previous study revealed that pfap2-g5 is a transcriptional repressor that co-regulates gametocyte development with pfap2-g (Kafsack et al., 2014; Poran et al., 2017; Shang et al., 2021a) ( Figures 5D, E ). Additionally, we found that other ApiAP2 transcription factors showed a decrease in transcript level after pfnsun1 knockdown, such as ap2-o, PF3D7_0420300, PF3D7_1107800, etc. ( Supplementary Figure S1 ). PfP52 is a member of the 6-cysteine family, which can ensure that sporozoites enter hepatocytes (Arredondo and Kappe, 2017). Target genes such as xl1 were downregulated at the schizont stage ( Figure 5D ). PfXL1 is possibly associated with adhesion between parasites and erythrocytes (Spillman et al., 2016). Noteworthy, PfNSUN1 knockdown caused some ncRNA down-regulated at trophozoite stage, most of them were 28S rRNA, such as PF3D7_0112700, PF3D7_0532000, and PF3D7_0726000 ( Figures 5D, E ). The results indicated that PfNSUN1 is involved in ribosome biogenesis, which is consistent with human NSUN1 protein (Liao et al., 2022).